Treatment Technologies for Removal of Methyl Tertiary Butyl Ether (MTBE) from Drinking Water:

Transcription

1 Treatment Technologies for Removal of Methyl Tertiary Butyl Ether (MTBE) from Drinking Water: Air Stripping Advanced Oxidation Processes Granular Activated Carbon Synthetic Resin Sorbents SECOND EDITION A A Report Publication Written of: for: The California MTBE Research Partnership Edited by: Edited by: Gina Melin, National Water Research Institute Gina Melin, National Water Research Institute December February i

2 Published by the Center for Groundwater Restoration and Protection National Water Research Institute NWRI Ellis Avenue P.O. Box Fountain Valley, California (714) Fax: (714) NWRI ii

3 Limitations This document is intended for use by members of the MTBE Research Partnership (Partnership) pursuant to the Research Partnership Agreement. The Research Partnership does not warrant, guarantee, or attest to the accuracy or completeness of the data, interpretations, and conclusions contained herein. Use of this document, or reliance on any information contained herein, by any party or entity other than members of the Research Partnership is at the risk of such parties or entities. i

4 California MTBE Research Partnership The California MTBE Research Partnership is comprised of the following individuals who shared their technical expertise by defining and reviewing this document and providing comments and suggestions: Steering Committee Mark Beuhler, Chair Metropolitan Water District of Southern California Ned Griffith, Lyondell Chemical Company Tracy Hemmeter, Santa Clara Valley Water District Dave Smith, ARCO Products Company Research Advisory Committee Walter Weber, Jr., Chair University of Michigan at Ann Arbor Rey Rodriguez, Chair, Source Water Protection Subcommittee H 2 O R 2 Consultants, Inc. Scott Tenney, Chair, Treatability Subcommittee Mobil Oil Corporation Lisa Anderson, Metropolitan Water District of Southern California Rich Atwater, Inland Empire Utilities Agency Bruce Bauman, American Petroleum Institute Ivo Bergsohn, South Tahoe Public Utilities District Tim Buscheck, Chevron Products Company David Camille, Tosco Robert Cheng, Long Beach Water Department Chi-Su Chou, Consultant Krista Clark, Association of California Water Agencies Mike Cooper, Placer Co. Daniel Creek, Alpine Environmental, Inc. Jim Crowley, Santa Clara Valley Water District James Davidson, Alpine Environmental, Inc. Marshall Davis, Metropolitan Water District of Southern California Shahla Farahnak, State Water Resources Control Board Amparo Flores, Malcolm Pirnie, Inc. Jack Fraim, Chevron John Gaston, CH2M Hill Don Gilson, Chevron Products Co. John Gustafson, Equilon, Inc. Elliot Heide, McClintock, Weston, Bens Roy Hodgen, Lyondell Chemical Company Rick Hydrick, South Tahoe Public Utilities District Steven Inn, Alameda County Water District Michael Kavanaugh, Malcolm Pirnie, Inc. Charles Kish, Riverview Water District John Kneiss, Oxygenated Fuel Association Sun Liang, Metropolitan Water District of Southern California Ronald Linsky, National Water Research Institute Dave McKinney, Shell Ralph Moran, ARCO Products Company Stephen Morse, Regional Water Quality Control Board, San Francisco Region David Pierce, Chevron Research and Technology Company Roger Pierno, Santa Clara Valley Water District Dave Ramsden, GZA, GeoEnvironmental, Inc. Melinda Rho, Los Angeles Department of Water and Power Martin Rigby, Orange County Water District Richard Sakaji, California Department of Health Services Dick Sloan, Lyondell Chemical Company Curt Stanley, Equilon, Inc. Andrew Stocking, Malcolm Pirnie, Inc. Paul Sun, Equilon, Inc. Martin Varga, Kern County Water Agency Mike Wang, Western States Petroleum Association Christine White, Equiva Services, LLC. Ken Williams, Regional Water Quality Control Board, Santa Ana Region Jeff Wilson, Western States Petroleum Association Dick Woodward, Sierra Environmental Services, Inc. ii

5 Acknowledgements The California MTBE Research Partnership wishes to acknowledge the extraordinary efforts of the many dedicated individuals who revised this document. Foremost, the Partnership would like to thank the Treatability Subcomittee, which is chaired by Scott Tenney of Mobil Oil Corporation. Members include: Todd Anderson, Santa Clara Valley Water District; Rich Atwater, Inland Empire Utilities Agency; Bruce Bauman, American Petroleum Institute; Robert Cheng, Long Beach Water Department; Chi-Su Chou, Consultant; Dan Creek, Alpine Environmental, Inc.; Jim Davidson, Alpine Environmental, Inc.; Marshall Davis, Metropolitan Water District of Southern California; Amparo Flores, Malcolm Pirnie, Inc.; Charles Kish, Riverview Water District; Michael Kavanaugh, Malcolm Pirnie, Inc.; Sun Liang, Metropolitan Water District of Southern of Southern California; David Pierce, Chevron Research and Technology Company; Roger Pierno, Santa Clara Valley Water District; David Ramsden, GZA, GeoEnvironmental, Inc.; Rey Rodriguez, H 2 O R 2 Consultants, Inc.; Richard Sakaji, State of California Department of Health Services; Dick Sloan, Lyondell Chemical Company; Curt Stanley, Equilon Enterprises LLC; Andrew Stocking, Malcolm Pirnie, Inc.; Paul T. Sun, Equilon Enterprises LLC; and Dick Woodward, Sierra Environmental Services The Partnership extends special thanks to Revision Project Manager Rey Rodriguez of H 2 OR 2 Consultants, Inc. The Partnership also wants to recognize the efforts of Gina Melin of the National Water Research Institute and Tim Hogan of Tim Hogan Design, who edited and designed this book. The Partnership is indebted to the authors of this book. They include: Chapter One: Introduction Rodriguez, R.A. 1 ; Davidson, J.M. 2 ; Creek, D.N. 2 ; Flores, A.E. 3 ; Stocking, A.J. 3 ; Kavanaugh, M.C. 3 Chapter Two: Air Stripping Stocking, A. J. 3 ; Eylers, H. 3 ; Wooden, M. 3 ; Herson, T. 3 ; Kavanaugh, M.C. 3 Chapter Three: Advanced Oxidation Processes Literature Review: Kommineni, S. 3 ; Zoeckler, J. 3, Stocking, A.J. 3 ; Liang, S. 4 ; Flores, A.E. 3 ; Kavanaugh, M.C. 3 Technology Cost Estimates: Rodriguez, R.A. 1 ; Browne, T. 5 ; Roberts, R. 5 ; Brown, A. 5 ; Stocking, A.J. 3 Chapter Four: Granular Activated Carbon Creek, D.N. 2 ; Davidson, J.M. 2 iii

6 Chapter Five: Synthetic Resin Sorbents Flores, A. E. 3 ; Stocking, A. J. 3 ; Kavanaugh, M.C. 3 Chapter Six: Summary of Findings and Conclusions Flores, A.E. 3 ; Stocking, A.J. 3 ; Creek, D.N. 2 ; Davidson, J.M. 2 ; Rodriguez, R.A. 1 ; Kavanaugh, M.C. 3 The Partnership acknowledges and appreciates the efforts of numerous peer reviewers for both the first and second editions, including: David Bacharowski, Los Angeles Regional Water Quality Control Board; Paul Gilbert-Snyder, State of California Department of Health Services; Sue Kaiser, California Air Resources Board; Steven Linder, United States Environmental Protection Agency; Ronald Linsky, National Water Research Institute; Gordon Schremp, California Energy Commission; and Gary Yamamoto, State of California Department of Health Services. Special thanks are given to the following reviewers for detailed insights and critiques that served to improve the quality of the revised report: Daniel Chang, University of California, Davis; Rick Sakaji, California Department of Health Services; Edward Schroeder, University of California, Davis; Neal Megonnell, Calgon Carbon Corporation; Terry Applebury, Applied Process Technologies; Barry Clark, Northeast Environmental Products; Ron White, Carbonair; Steve Cater, Calgon Carbon Corporation; Laura Schmidt, Rohm and Haas Company; and Phil Hodge, American Purification, Inc. And, finally, the Partnership would like to thank the numerous technology vendors who provided useful and up-to-date information to complete this report. Specific vendors are referenced in each chapter. 1 H2 O R 2 Consultants Inc.; 653 E. Michelle St.; West Covina, California 91790; Phone (626) Alpine Environmental, Inc.; 203 W. Myrtle St. Suite C.; Ft. Collins, Colorado 80521; Phone (970) Malcolm Pirnie, Inc.; 180 Grand Ave. # 1000; Oakland, California, 94612; Phone (510) Metropolitan Water District of Southern California; 700 Moreno Ave.; LaVerne, California 91750; Phone (909) Komex H2O Science Inc.; 550 Bolsa Ave., Suite 105; Huntington Beach, California 92649; (714) iv

7 Table of Contents 1.0 Introduction Background and Objectives Document Overview The California MTBE Research Partnership Preliminary Evaluation or Selection...5 of Treatment Technologies 1.2 History of MTBE Use MTBE as an Oxygenate Physical and Chemical Characteristics Impact of MTBE on Water Supplies Drinking Water Regulations Federal Standards California Drinking Water Standards Analytical Methods Integration of Technologies Conclusion References Air Stripping Background Air Stripping Application for MTBE Removal...25 from Drinking Water Objectives of the Evaluation Description of Technology - Air Stripping Background Process Principles Aeration Technologies Comparative Discussion of Air Strippers Permitting Flow Rate Removal Efficiency and Flow Rate Impact of Water Quality...51 v

8 2.3.5 Other Factors By-products Cost Effectiveness Off-gas Treatment Vapor Phase GAC Adsorption Thermal and Catalytic Oxidation Biological Treatment Evaluation and Screening of Off-gas Treatment Technologies Permitting Flow Rate Removal Efficiency Other Characteristics By-products Cost Effectiveness Optimization of Air Stripping Technologies Introduction and Design Equations Design Parameters Operating Parameters Summary Conclusions and Recommendations for Future Research Recommended Technologies Recommendations for Future Research References Advanced Oxidation Processes Background and Objectives of Evaluation Process Definition General Process Principles and Implentability Issues Water Quality Impacts General Advantages General Disadvantages AOP Technologies Established Technologies vi

9 3.4.1 Hydrogen Peroxide/Ozone (H 2 O 2 /O 3 ) UV Systems Emerging Technologies E-beam Treatment Cavitation TiO 2 -catalyzed UV Oxidation (TiO 2 /UV) Fenton s Reaction Comparative Discussion of AOPs Permitting Flow Rate Removal Efficiency Other Factors Cost Evaluation Overall Costs of AOP Systems Evaluation of Cost Estimates for Specific AOP Technologies Sensitivity Analysis Conclusions and Recommendations for Future Research Recommended Technologies Recommendations for Future Research References Granular Activated Carbon Background Nature of Problem Objectives General Evaluation of GAC Technology Description of Technology Benefits and Limitations Key Variables and Design Parameters Carbon Vendors Level of Development of Technology Technical Implementability Permitting vii

10 4.2.8 Current Usage of Technology Summary of Ongoing Research Detailed Evaluation of GAC Computer Modeling Flow Rate Removal Efficiency Other Characteristics Cost Estimates Sensitivity Analyses Conclusions and Research Recommendations Conclusions Recommendations for Future Research References Synthetic Resin Sorbents Introduction Background Objectives of the Evaluation Process Principles Equilibrium Sorption Models Sorption Rates Application of Synthetic Resins to Water Treatment Resin Production Physical and Chemical Properties of Resins Results of Bench-scale Studies Key Variables in the Design of Synthetic Resin Sorbent Systems Type of Synthetic Resin Background Water Quality Process Flow Configuration Regeneration Operating Parameters Manufacturers and Resin Unit Costs Pilot-scale/Field Studies viii

11 5.5.1 Major Oil Refinery in Bakersfield, California World Oil Service Station BP Oil Company Gas Station in Bellingham, MA Charnock Well Field in Santa Monica, CA Economic Analysis Objective and Background Cost Scenarios Assumptions Results Conclusions and Limitations Recommendations for Future Work Acknowledgements References Conclusions and Recommendations Summary of Key Findings and Conclusions Air Stripping AOPs GAC Synthetic Resin Sorbents Comparative Discussion of Different Technologies Permitting Reliability Flexibility Adaptability Potential for Modifications Cost Effectiveness Recommendations for Future Research Air Stripping AOPs GAC Resins ix

12 Appendices Appendix 2A-1 Cost Assumptions for Packed Tower Aeration Appendix 2A-2 Cost Assumptions for Low Profile Aeration Appendix 2B 9/1/98 La Crosse, Kansas Trip Report Appendix 2C Air Stripping Equipment Vendors Appendix 2D Off-gas Treatment Equipment Vendors Appendix 3A Assumptions for AOP Economic Analysis Appendix 4A Cost Estimates for GAC Appendix 5A Filtrasorb 600 Isotherm Appendix 5B Assumptions for Synthetic Resin Sorbents Cost Estimates x

13 List of Tables Table 1-1 Physical and Chemical Characteristics...8 of BTEX Compounds and MTBE Table 2-1 Description of Air Stripping Treatment Technologies...30 and Their Advantages and Disadvantages Table 2-2 Packed Tower Design Variables Table 2-3a Air Stripping Permitting Requirements...48 Table 2-3b Off-gas Treatment Permitting Requirements Table 2-4 Typical Maximum Hydraulic Capacities...50 for Commercially Available Systems Table 2-5 Problematic Iron and Hardness Concentrations...51 for Air Stripping Technologies Table 2-6 Comparison of Air Stripping Technologies Table 2-7 Sample Calculation of Capital, Annual, and Unit Treatment Costs for a 600 gpm Packed Tower, 2,000 to 20 ug/l MTBE Table 2-8 Initial Capital Expenses for Air Stripping Systems...58 Table 2-9 Annual O&M Costs for Air Stripping Systems...58 Table 2-10 Estimated Labor Hours for Maintenance and Sampling...59 of Air Stripping Systems in Hours per Week Table 2-11 Sampling Requirements for Air Stripping Systems...59 and Off-gas Treatment: No. of Samples Collected Weekly Table 2-12 Total Amortized Operating Costs ($/1,000 Gallons Treated)...61 for Air Stripping Systems Table 2-13 Expense Summary for Air Stripping Systems - Spray Tower Table 2-14 Expense Summary for Air Stripping Systems - Packed Tower...63 Table 2-15 Expense Summary for Air Stripping Systems - Low Profile...64 Table 2-16 Expense Summary for Air Stripping Systems - Bubble Aeration...65 Table 2-17 Expense Summary for Air Stripping Systems - Aspiration...66 Table 2-18 Sensitivity Analysis for Air Stripping Systems - Packed Tower Table 2-19 Sensitivity Analysis for Air Stripping Systems - Low Profile...68 Table 2-20 MTBE Air Stripper System Off-gas Removal Rates...70 Required to Meet 1 lb/day Discharge Limit Table 2-21 Description of Off-gas Treatment Technologies...71 Table 2-22 GAC Design Variables...73 xi

14 Table 2-23 Initial Capital Expenses for Off-gas Treatment Systems Table 2-24 Annual O&M Costs for Off-gas Treatment Systems...84 Table 2-25 Total Amortized Operating Costs ($/1000 Gallons Treated) for Off-gas Treatment Table 2-26 Air Stripper Exit Concentrations Initiating Off-gas Treatment...86 Table 2-27 Comparison of Off-gas Treatment Technologies...87 Table 2-28 Cost Summary for Off-gas Treatment Technologies at 5 ppmv MTBE..90 Table 2-29 Cost Summary for Off-gas Treatment Technologies at 0.5 ppmv MTBE..91 Table 2-30 Design and Operating Parameters...94 Table 2-31 Air Stripping and Off-gas Technology Combination Recommendations..104 Table 2-32 Cost for Technology (Air Stripping and Off-gas Treatment) Table 3-1 Hydroxyl Radical Rate Constants, k, for MTBE and Its By-products..120 Table 3-2 Brief Descriptions, System Components, Advantages, and Disadvantages of Established and Emerging AOP Technologies Table 3-3 Reactions, By-products, Interfering Compounds, and Oxidant Hierarchy Table 3-4 Summary of AOP Pilot and Field Studies Table 3-5 Summary of Vendor Information Table 3-6 Characteristics of Typical Low Pressure (LP), Medium Pressure (MP), and Pulsed-UV (P-UV) Lamps Table 3-7 Range of AOP Reactor Capacities and MTBE Removal Efficiencies Table 3-8 Comparative Analysis of Various AOPs Table 3-9 Cost of Hydrogen Peroxide Removal and Oxidation By-product Removal..184 Table 3-10 Costs of H 2 O 2 /O 3 System for MTBE Removal Table 3-11 Capital Costs of AOPs Table 3-12 Annual O&M Costs of AOPs Table 3-13 Total Amortized Operating Costs (per 1,000 Gallons Treated) for AOPs Table 3-14 Estimated By-product Formations and Residual Oxidant Concentrations for H 2 O 2 /MP-UV System Table 3-15 Effects of TOC on AOP Treatment Costs Table 3-16 Effects of Additional BTEX Contamination on AOP Treatment Costs Table 3-17 Effects of Design Life on AOP Treatment Costs Table 4-1 Summary of MTBE Isotherm Data for Activated Carbons xii

15 Table 4-2 Vendor Information Table 4-3 Estimated Carbon Usage Rates for Full-scale Systems Table 4-4 Results of AdDesignS Modeling Table 4-5 Results of AdDesignS Modeling for Sensitivity Analyses Table 4-6 Predicted Carbon Usage Rates and Breakthrough Times Using In-series Operation Table 4-7 Summary of Cost Estimates Table 4-8 Cost Estimates for Sensitivity Analyses Table 5-1 Steps in the Process of Sorption Table 5-2 Largest Manufacturers of Polymeric and Carbonaceous Resins Used for the Removal of Organic Compounds Table 5-3a Physical and Chemical Properties of Polymeric Sorbents Table 5-3b Physical and Chemical Properties of GAC and Carbonaceous Resins Table 5-4 MTBE Isotherm Studies and Their Associated Experimental Conditions and Freundlich Parameters Table 5-5 Chemical Properties of MTBE, m-xylene, and TBA Table 5-6 Manufacturer Information and Unit Costs Table 5-7 Option 1 (Series Operation): Capital, O&M, Regeneration, and Total Costs Table 5-8 Option 2 (Carousel Operation): Capital, O&M, Regeneration, and Total Costs Table 5-9a Regeneration Capital Costs for Option 1 (Series Operation) Table 5-9b Regeneration Capital Costs for Option 2 (Carousel Operation) Table 5-10a Regeneration O&M Costs for Option 1 (Series Operation) Table 5-10b Regeneration O&M Costs for Option 2 (Carousel Operation) Table 5-11 Regeneration Totals ($/1000 Gallons) Table 5-12 Time to Column Exhaustion for MTBE vs. TBA Table 5-13 Sensitivity Analysis of Cost Estimates Table 6-1 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Air Stripping Technologies Table 6-2 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Off-gas Treatment Technologies Table 6-3 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended AOP Technologies xiii

16 Table 6-4 Summary of Total Amortized Costs ($/1,000 Gallons;) for Recommended GAC Systems Table 6-5 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Synthetic Resin Systems Table 6-6 Summary of Total Amortized Costs ($/1,000 Gallons) for Various Recommended Treatment Systems xiv

17 List of Figures Figure 2-1 Estimated vs. experimental data for MTBE s Henry s constant...28 as a function of temperature. Figure 2-2 An illustration of a packed tower system manufactured by...31 Carbonair (1995). Figure 2-3 An illustration of a low profile system manufactured by Carbonair (1995). Figure 2-4 Components of an aspiration system (Maxi-Strip )...44 maufactured by Hazelton Environmental (1998). Figure 2-5 Cost of off-gas treatment technologies as a function...69 of air flow rate. Figure 2-6 Effects of increasing the removal efficiency on...95 the packing volume as a function of inlet water temperature. Figure 2-7 Effects of increasing the AWR on packing volume...95 as a function of removal efficiency. Figure 2-8 Effects of changing AWR and packing media...96 on packing volume for a given removal efficiency. Figure 2-9 Effects of changing AWR and packing media...96 on pressure drop for fixed tower dimensions. Figure 2-10 Effects of changing pressure drop on total brake power...97 and the packing volume as a function of packing. Figure 2-11 Effects of changing the AWR on pressure drop for a...98 given tower cross-sectional area. Figure 2-12 Demonstration of air stripping flexibility...98 (ability to handle a variety of flow rates) for a given design. Figure 2-13 Domains for cost-effectiveness of air strippers at varying removal efficiencies. Figure 3-1a A schematic of a conventional (continuously stirred tank reactor)..135 H 2 O 2 /O 3 system equipped with UV lamps. Figure 3-1b A schematic of a plug flow H 2 O 2 /O 3 system manufactured by Applied Process Technology, Inc. Figure 3-2 A schematic of a high energy electron beam system Figure 3-3 A schematic of a cavitation system (HYDROX) with supplemental chemical oxidants (e.g., H 2 O 2 ). Figure 3-4 A schematic of a fluidized bed TiO 2 /UV system Figure 3-5 A schematic of a system utilizing Fenton s Reaction Figure 4-1 Schematic for GAC using series operation xv

18 Figure 4-2 MTBE isotherms for GAC Figure 4-3 Predicted carbon usage rate vs. influent MTBE concentration Figure 4-4 Estimated unit treatment costs vs. system flow capacity for in-series operation. Figure 4-5 Estimated unit treatment costs vs. influent MTBE concentration for in-series operation. Figure 5-1 Comparison of Freundlich model and Dubinin-Astakov (DA) Model fits to experimental data for a carbonaceous resin (Ambersorb 572). Figure 5-2 Various matrices for polymeric resins Figure 5-3 MTBE isotherms Figure 5-4a & b Sorption capacities of various synthetic resins and GAC at aqueous concentrations of 100 and 1,000 µg/l MTBE. Figure 5-5a & b Effects of background humic substances in Santa Monica water on the sorption capacities of coconut-based GAC (GRC-22) and Ambersorb 572. Figure 5-6 Effects of m-xylene (Co = 43.2 mg/l) on the MTBE sorption capacities of Filtrasorb 400, Ambersorb 572, and Ambersorb 563. Figure 5-7 TBA isotherms Figure 5-8a & b Effects of TBA (100 µg/l) on the equilibrium sorption capacities of coconut-based GRC-22 and Ambersorb 572 resin. Figure 5-9 MTBE vs. TBA isotherms for Ambersorb and Dowex Optipore L-493 resins. Figure 5-10 Typical process flow configuration for a resin system Figure 5-11 Hypothetical breakthrough curve and its MTZ at one point in time. Figure 5-12 Process flow configurations for in-series operation vs. carousel operation of resin systems. Figure 5-13 Four operational models for two vessels in series xvi

19 List of Appendix Tables and Figures Table 2A-1 Packed Tower Aeration Assumptions Table 2A-2 Low Profile Aeration Assumptions Table 3A-1 Replacement Part Costs for AOPs Table 3A-2 Labor Costs for AOPs Table 3A-3 Analytical Costs for AOPs Table 3A-4 Chemical Costs for AOPs Table 3A-5 Electrical Costs for AOPs Table 3A-6 Estimated Labor Costs for H 2 O 2 /MP-UV System Table 3A-7 Estimated Labor Costs for H 2 O 2 /O 3 System Table 3A-8 Estimated Labor Costs for Ultrasound/H 2 O 2 Systems Table 3A-9 Estimated Labor Costs for TiO 2 /MP-UV System Table 4A-1 Estimated Labor Costs for GAC Systems Table 4A-2 Estimated Labor Costs for Sensitivity Analysis Table 4A-3 Estimated Costs for Carbon Adsorption Table 4A-4 Estimated Costs for Sensitivity Analysis Figure 5A-1 MTBE isotherms for Filtrasorb 600 and PCB, a coconut-based GAC. Table 5B-1 MTBE Sorption Modeling Results Table 5B-2 MTBE Assumptions Used in AdDesignS Table 5B-3 Calculation of Capital, Annual, and Unit Treatment Costs Table 5B-4 Labor and Analytical Assumptions xvii

20 xviii

21 List of Acronyms ACT ACWA AOP AQMD ASTM AWR BACT BAT BTEX BV CARB CB Cl CO COD D/DBP DBP DCA DCB DCE DHS DLR DO DOC E-beam EBCT EBRF EE/O EPA ETV FRP GAC GET gpm H H 2 O 2 HAA HAP HVEA kwh LP-UV hυ MCL accelerated column test Association of California Water Agencies advanced oxidation process Air Quality Management District American Society for Testing and Materials air/water ratio best available control technologies (as defined by EPA) best available treatment (technology) benzene, toluene, ethylbenzene, xylenes (o-, m-, p-xylene) bed volume; volume of a resin or GAC vessel California Air Resources Board conduction band chlorine radical carbon monoxide chemical oxygen demand disinfection/disinfection by-product disinfection by-products dichloroethane dichlorobenzene dichloroethylene (California) Department of Health Services detection limit for the purpose of reporting dissolved oxygen dissolved organic carbon high energy electron beam empty bed contact time Electron Beam Research Facility Electrical Energy per Order of Removal United States Environmental Protection Agency Environmental Technology Verification fiberglass reinforced plastic granular activated carbon Guardian Environmental Technologies gallons per minute hydrogen atom hydrogen peroxide haloacetic acid hazardous air pollutant High Voltage Environmental Applications kilowatt-hours continuous wave low pressure mercury vapor lamps UV radiation Primary Maximum Contaminant Level xix

22 MEK methyl ethyl ketone MP-UV continuous wave medium pressure mercury vapor lamps MTBE methyl tertiary butyl ether MTZ mass transfer zone NaOCl sodium hypochlorite NDMA N-nitroso dimethyl amine NHDES New Hampshire Department of Environmental Services NO X oxides of nitrogen NOM natural organic matter NPDES National Pollutant Discharge Elimination System OH hydroxyl radicals O 2 superoxide radical O 3 ozone O 3 /UV ozone/uv O&M operation and maintenance O&P overhead and profit OFA Oxygenated Fuels Association ORC Oxygen Releasing Compound P-UV Pulsed UV PCB polychlorinated biphenyls PCE perchloroethylene (also known as tetrachloroethylene) PEE Process Equipment and Engineering PHG Public Health Goal POC particulate organic carbon POE point-of-entry PSDM Pore Surface Diffusion Model RFG Reformulated Gasoline RSSCT rapid small-scale column test SDWA Safe Drinking Water Act SMCL Secondary Maximum Contaminant Level SO X oxides of sulfur SOC synthetic organic chemical TBA tertiary-butyl alcohol TBF tertiary-butyl formate TCE trichloroethylene THM trihalomethanes TiO 2 titanium dioxide TOC total organic carbon TPH total petroleum hydrocarbons UCLA University of California, Los Angeles UST underground storage tank UV ultraviolet VOC volatile organic compound VTM volume treated per carbon mass WSPA Western States Petroleum Association xx

23 1.0 Introduction Rey Rodriguez James Davidson, P.G. Daniel Creek, P.E. Amparo Flores Andrew Stocking, P.E. Michael Kavanaugh, Ph.D., P.E. 1

24 2

25 1.1 Background and Objectives This document presents the results of feasibility and economic analyses for methyl tertiary butyl ether (MTBE) removal from drinking water. The study was conducted using the most promising and/or widely accepted technologies for removing volatile organic compounds from drinking water namely, air stripping, advanced oxidation processes (AOPs), granular activated carbon (GAC), and synthetic resin sorbents. These technologies are evaluated as they apply specifically for removal of MTBE. The first edition of this document was published in December of This second edition is a significant improvement from the first edition. The most noted changes are the addition of a new chapter on synthetic resin sorbents (Chapter 5), refinement and update of costs for all technologies, significant revisions to the AOPs section, a new introductory chapter (Chapter 1), and a new chapter (Chapter 6) with overall conclusions and recommendations. An executive summary will also be published as a stand-alone summary of this report Document Overview Chapter 1 provides an introductory discussion on the history of the California MTBE Research Partnership (Partnership), background information on MTBE use, discussion of the physical and chemical properties of MTBE, maximum allowable contaminant levels in drinking water, drinking water regulations, and requirements for permitting MTBE treatment systems in California. Although this document is focused on drinking water systems with capacities from 60 to 6,000 gallons per minute (gpm), the information provided in this report should also benefit remediation applications where lower flow rate systems are typically required. Chapter 2 evaluates air stripping options with and without off-gas treatment for removal of MTBE from drinking water. Air stripping is a proven technology that has successfully removed MTBE in drinking water applications (e.g., La Crosse, Kansas and Rockaway Township, New Jersey). The ability to meet drinking water standards, technical implementability, reliability, flexibility, adaptability, and the potential for modifications are addressed for this technology. Chapter 2 also presents capital and annual operation and maintenance (O&M) costs for various air-stripping and off-gas treatment technologies at flow rates of 60, 600, and 6,000 gpm under influent MTBE concentrations of 20, 200, and 2,000 micrograms per liter (µg/l). Please note that these flow rates and influent concentrations were also utilized in the other technology evaluations. Chapter 3 evaluates advanced oxidation processes, which include use of ozone (O 3 ), hydrogen peroxide (H 2 O 2 ), ultraviolet (UV) light irradiation, high energy electron beam (E-beam) irradiation, ultrasonic and hydrodynamic cavitation, titanium dioxide (TiO 2 ) catalysis, and Fenton s Reaction for their ability to remove MTBE from water. While several laboratory and pilot-scale studies have been completed to investigate the ability of AOPs primarily H 2 O 2 /O 3 and H 2 O 2 /MP-UV to destroy organic compounds in drinking water, 3

26 there are a limited number of full-scale applications for drinking water treatment. For example, an H 2 O 2 /MP-UV system was installed in Salt Lake City, Utah for perchloroethylene (also known as tetrachloroethene, or PCE) removal from drinking water. The primary objective of Chapter 3 is to evaluate the feasibility of using AOPs for the removal of MTBE from drinking water. This feasibility evaluation includes a review of the chemical and physical principles behind AOPs, a discussion of the various established and emerging AOPs that have potential for MTBE removal, an analysis of the effect of water quality on the effectiveness of these AOPs, and a cost analysis based on information gathered from manufacturers, vendors, and actual pilot tests. Chapter 3 concludes with overall recommendations for implementation of AOPs for MTBE removal and recommendations for future work. Chapter 4 presents results of technical evaluations, computer modeling, and cost estimates for the removal of MTBE from drinking water using GAC adsorption. Carbon adsorption is a well understood process that is widely used for the removal of synthetic organic chemicals (SOCs) from water. This chapter provides a detailed feasibility analysis regarding the use of GAC specifically to remove MTBE from drinking water. This feasibility analysis includes determining the conditions (e.g., MTBE influent concentrations, background water quality) under which GAC is most likely to cost-effectively remove MTBE. Chapter 4 also includes a focused literature review and a compilation of vendor information to determine the benefits and limitations of GAC, key variables and design parameters, current usage of GAC for MTBE removal, etc. The detailed evaluation includes computer modeling results and cost estimates prepared to determine the impacts of influent MTBE concentrations, background organic matter, and the presence of other organic compounds upon the removal of MTBE using GAC. Chapter 5 evaluates the use of synthetic resin sorbents, which had not been addressed in the first edition of this report. Certain developments suggest that synthetic resins may be economically competitive with the treatment technologies previously investigated. First, preliminary findings suggest that synthetic resins may have sufficiently better sorption capacities for tertiary-butyl alcohol (TBA) compared to GAC and, therefore, may present a practical alternative treatment technology for TBA-contaminated sites. Second, improvements in resin regeneration processes may make the life-cycle cost of a resin system competitive or, perhaps, more economical than other options. Third, resins (unlike AOPs) do not produce oxidation by-products. The evaluation of synthetic resin sorbents included a literature review involving published literature and information provided by manufacturers, vendors, and major researchers; an economic analysis using the available literature and AdDesignS (a computer modeling software); and the identification of field sites. Chapter 5 concludes with recommendations for implementation of synthetic resins and recommendations for future work. 4

27 Chapter 6 presents a summary of the key findings and conclusions from the evaluations of individual technologies, provides a comparative discussion of the different technologies, and makes overall recommendations for future work The California MTBE Research Partnership In 1996, the City of Santa Monica, California lost 50 percent of its drinking water supply because of MTBE contamination. As a result, the city pursued a cooperative effort with the petrochemical industry and the Association of California Water Agencies (ACWA) to examine MTBE treatment alternatives. In October 1997, the Western States Petroleum Association (WSPA), the Oxygenated Fuels Association (OFA), and ACWA formed the California MTBE Research Partnership to develop a statewide research program concerned with MTBE treatment technology and sourceprotection issues. Since then, the Partnership has invested nearly a million dollars to support a comprehensive research program, which is managed by the National Water Research Institute, to address MTBE treatment and protection needs. The mission of the Partnership is to identify, prioritize, plan, and sponsor practical research projects to protect, treat, or remove MTBE contamination from drinking-water supplies. To fulfill this mission, the Partnership has created two focus-area subcommittees to investigate and develop Partnership projects: the Treatability Committee and the Source Water Protection Committee. Each group is comprised of engineers and scientists who represent water utilities, regulatory agencies, the petrochemical industry, academia, and the environmental consulting field Preliminary Evaluation or Selection of Treatment Technologies The purpose of the Treatability Committee is: 1) to evaluate existing and emerging technologies for removal of MTBE and other oxygenates from water; 2) to identify research needs to improve the efficiency of existing technologies; 3) to develop alternative technologies for MTBE removal; and, 4) to coordinate the funding and implementation of identified research projects. The first meeting of the Treatability Work Group in October 1997 identified research needs for the following five technologies: Air Stripping AOPs Adsorption Membranes Biological Treatment 5

28 Criteria for the selection and evaluation of alternative technologies included the following: Ability to consistently meet drinking water criteria (e.g., 5 µg/l) Cost-effectiveness By-product formation Residuals and their impact Impact of other constituents in water (e.g., iron, manganese, total organic carbon [TOC]) and feed pretreatment requirements Operability (e.g., ability to run unattended) Compatibility with existing technologies Robustness (e.g., ability to deal with high and low flows and concentrations) Simplicity of operation A second meeting of the Treatability Committee in January 1998 ranked the specific research needs identified during the first meeting and drafted research plan write-ups designed to meet these needs. The time scale for these tasks covered the fiscal years 1998 and Initially, the Treatability Committee decided that white papers and bench-scale studies would be conducted for each of the technologies. Subsequently, full-scale treatment evaluations may be considered for those technologies that are identified as having the greatest potential for cost competitive and effective removal of MTBE. At present, the state of knowledge on the treatment of MTBE in drinking water is limited due to the few instances of full-scale drinking water systems where MTBE treatment has been required. However, since the completion of the first edition of this document, numerous pilot and/or field studies have been conducted and/or identified and now provide a significant increase in information that was previously unavailable. The results of this technology evaluation will be of great interest to all water utilities, regulatory agencies, the petrochemical industry, the environmental community, and the public. These evaluations and this report can serve as guidance for federal and state regulators, water utilities, and the petrochemical industry when evaluating the treatment options for MTBE in drinking water. 6

29 1.2 History of MTBE Use MTBE as an Oxygenate MTBE is an oxygenate compound added to gasoline to enhance octane level and meet the oxygen requirements mandated in the Clean Air Acts Amendments. There are several types of oxygenated fuel Oxyfuel, Reformulated Gasoline (RFG), and California Air Resources Board (CARB) phase 2 fuel each with a different specification for oxygen content. Oxyfuel is mandated in carbon monoxide non-attainment areas and requires 2.7 percent by weight oxygen, which corresponds to 15 percent by volume MTBE. RFG is mandated in ozone non-attainment areas and requires 2.0 percent by weight oxygen, corresponding to 11 percent by volume MTBE. Finally, CARB fuel is a California-only fuel based on a predictive model that requires 1.8 to 2.2 percent by weight oxygen. When gasoline is released into the environment, a variety of chemical compounds, including MTBE, can be transferred into the air or water. Because of MTBE s particular physio-chemical properties (such as high solubility, low soil-to-water partition coefficient, and low Henry s constant), it is somewhat difficult to remove from water. Overall, MTBE treatment presents specific challenges, including the need to implement a readily available removal technology in a timely manner. Because of this, the Partnership s Treatability Committee investigated only the most promising and widely accepted technologies. The group chose to evaluate air stripping, advanced oxidation, carbon adsorption, and synthetic resin sorbents for two reasons: these technologies are proven effective in removing organic compounds; and, California regulators find these technologies generally acceptable for treating drinking water. Membranes, biological treatment, and other alternative technologies may be considered at a later date. Treatment costs depend on existing system conditions: specifically, system flow rates, MTBE influent concentrations, and the level of treatment needed for MTBE to reach the desired effluent goal. Based on these variables, the Treatability Committee chose specific conditions to evaluate priority technologies. The water-supply flow rates were considered at 6,000, 600, and 60 gallons per minute. At each of these flow rates, the technologies were analyzed for MTBE influent concentrations of 2,000, 200, and 20 µg/l. The MTBE effluent levels were considered at 20 µg/l, 5 µg/l, and non-detect (represented by 0.5 mg/l) Physical and Chemical Characteristics MTBE is a polar organic compound that has a chemical formula of CH 3 OC(CH 3 ) 3. At room temperature, it is a volatile, flammable, colorless liquid with a terpene-like odor (Squillace et al., 1997). The taste and odor thresholds for MTBE in water range from 2.5 µg/l to 680 µg/l, and 2 µg/l to 190 µg/l, respectively, depending on the individual being tested, level of chlorination, temperature of the water, and other factors (API, 1994; ARCO, 1993; Young et al., 1996; Shen et al., 1997; Malcolm Pirnie, 1998; Dale et al., 1998). 7

30 It has been suggested that the treatment of MTBE in drinking water using conventional treatment processes (i.e., air stripping and GAC) poses challenges relative to other organic contaminants, particularly benzene, toluene, ethylbenzene, and xylenes (BTEX) compounds, because of its unique physical and chemical properties. A summary of the physical and chemical characteristics of MTBE relative to BTEX compounds is presented in Table 1-1. As this table shows, MTBE s Henry s constant is approximately an order of magnitude lower than those of BTEX compounds. Since the air stripping potential of a compound is primarily determined by its Henry s constant (H), air stripping of MTBE will tend to be more difficult and more costly relative to these other compounds. Furthermore, the relatively low organic carbon partitioning coefficient (K OC ) and the high water solubility of MTBE make GAC adsorption less effective for MTBE relative to BTEX compounds. Table 1-1 Physical and Chemical Characteristics of BTEX Compounds and MTBE Physical and Chemical Properties Molecular weight [g/mole] Vapor 1 atm; 10 C Specific 25 C Boiling point [ C] Water solubility [mg/l] Vapor pressure [mm Hg] (@25 C) [kpa] (@100 F) Henry s Law Constant [ ] Log K OC Log K OW Diffusivity (m 2 /s) Liquid Gas Benzene Toluene Ethylbenzene o-xylene a a a a a MTBE 80.1 a a a a a 1730 a a 161 a 175 a 43,000-54,300 50,000 a 76, a 28.4 a 9.53 a 6.6 a a 1.26 b 3.79 b 1.23 b b 3.33 a 0.23 b b b b 20 C a a a 1.091, a a 1.035, a 2.36 b 2.73 b 3.24 b 3.10 b 1.20 a 9.95 x x x x x x 10-6 b 8.35 x 10-6 b 7.59 x 10-6 b 7.56 x 10-6 b 8.45 x 10-6 b a OSTP Report, 1997 b Crittenden et al.,

31 MTBE has been found to be recalcitrant relative to BTEX compounds; however, its biodegradation is actively being investigated by several researchers (e.g., Hanson et al., 1999; Salanitro et al., 1999) Impact of MTBE on Water Supplies The state of California is the third largest gasoline consumer in the world, with more than 13.7 billion gallons of gasoline used per year. Because of air pollution problems, CARB and the United States Environmental Protection Agency (EPA) mandated the use of oxygenated fuels in California to improve vehicle octane ratings and reduce tailpipe emissions. Although MTBE is credited with significantly reducing emission, it has also contaminated water supplies through accidental gasoline releases at dispensing sites, leaking product pipelines, and via leaks from underground storage tanks (USTs). According to California monitoring data (November 19, 1999), MTBE concentrations greater than the detection level have impacted 62 of the 6,409 drinking-water sources sampled. However, contamination is not limited to California. Nationwide, there are numerous reports of MTBE leaks and spills from petroleum facilities, including refineries, terminals, pipelines, and service stations. Consequently, MTBE has received widespread attention from municipalities and other drinking water providers, particularly in California where two major drinking water supplies have been significantly impacted by MTBE releases into the environment. 9

32 10

33 1.3 Drinking Water Regulations Drinking water quality requirements in California are governed by both the EPA and the California Department of Health Services (DHS) Office of Drinking Water Regulations. Suppliers of domestic water to the public must comply with the regulations adopted by the EPA under the United States Safe Drinking Water Act and with the DHS under the California Safe Drinking Water Act (SDWA). Because California has been granted primacy for enforcement of federal regulations, California must promulgate regulations that are no less stringent that the federal regulations. Historically, California drinking water regulations have been more stringent than the federal regulations. Since SDWA was passed in 1974 by the United States Congress, it has been amended several times with significant changes in the water quality regulations in the last 10 years. The passage of the SDWA authorized the EPA to develop national standards and regulations for drinking water. The regulations controlled the contaminants in drinking water that may adversely affect the overall aesthetics (taste, color, odor, or other conditions) of drinking water and those which would have an adverse effect on health. Historically, the primacy of drinking water regulations on both the state and federal level has been to mandate drinking water standards on health-related contaminants and recommend goals on aesthetic-related parameters. The health-based standards are referred to as Primary Maximum Contaminant Levels (MCLs) and aesthetic-based standards are referred to as Secondary Maximum Contaminant Levels (SMCLs). Public awareness has increased significantly due to extensive pesticide and organic solvent groundwater contamination throughout the United States and, more recently, due to the detections of MTBE in a limited number of drinking water supplies. As the knowledge of health effects of groundwater contaminants increases, and the analytical detection limits of contaminants decreases, new or amended regulations are promulgated. The SDWA amendments of 1986 and 1996 have enacted numerous revisions, including new drinking water standards and monitoring requirements for existing and new contaminants. California has adopted more stringent standards for a number of inorganic and volatile organic compounds than federal requirements. Furthermore, California has regulations where the California Office of Environmental Health Hazard Assessment (OEHHA) has established Public Health Goals (PHGs) that, in some cases, are more stringent than the Primary and Secondary Drinking Water Standards under federal drinking water standards. MTBE drinking water standards throughout the United States vary considerably. The concentrations range from 5 µg/l in California to over 240 µg/l in Michigan. In addition, approximately 17 states have not adopted standards and are relying on the EPA Drinking Water Advisory or waiting for the EPA to adopt an MCL. There is also a diversity of MTBE standards that have been developed for cleanup levels for contaminated sites. These standards vary significantly and many are adjusted based on the intended use of the type of water impacted. 11

34 1.3.1 Federal Standards In December of 1996, the EPA issued a draft health advisory of 70 µg/l based on the presumed health effects of MTBE. The advisory level was based on kidney and liver effects observed in laboratory animal experiments (Office of Science and Technology, Office of the EPA). In December 1997, the EPA Office of Water issued a Drinking Water Advisory on MTBE (Drinking Water Advisory: Consumer Acceptability Advice and Heath Effects on Methyl Tertiary-Butyl Ether). The advisory is not a mandatory standard for action, but discusses the limitations of estimating a risk level for MTBE in drinking water and characterizes the hazards associated with this route of exposure. The Advisory provides guidance to communities that have been exposed to water contaminated with MTBE. The Drinking Water Advisory concluded that keeping the concentrations in the range of 20 µg/l to 40 µg/l or below will likely avert unpleasant taste and odor effects California Drinking Water Standards In 1997, two California bills (SB1189 by Hayden and AB592 by Kuehl) were signed into law requiring the DHS to develop Primary and Secondary Drinking Water Standards for MTBE. The bills required that the DHS establish the secondary drinking water standard by July 1, 1998 and the primary drinking water standard by July 1, At the time the bills were signed by the Governor, an action level had already been established in California in 1991 along with a draft federal health advisory level. Action Levels In 1991, the Pesticide and Environmental Toxicology Section which is now OEHHA developed an action level of 35 µg/l for MTBE. Action levels provide drinking water guidance for chemicals without drinking water standards. The 35 µg/l is based on non-carcinogenic effects with a large uncertainty factor to provide a margin of safety to exposure in drinking water. In February 1997, DHS added MTBE to the unregulated (i.e., without an MCL or SMCL) chemical list for which monitoring is required. This monitoring requirement has created documentation on the extent of MTBE occurrences in groundwater and surface water sources. One important aspect in regards to monitoring for MTBE is the requirement that laboratories report MTBE analytical results at or above the detection limit for the purpose of reporting (DLR). The DLR established in August 1998 is 3 µg/l. Public Health Goals In March 1999, OEHHA adopted a PHG of 13 µg/l for MTBE. PHGs are developed for chemical contaminants based on the best available toxicological data in the scientific literature. PHGs provide estimates on the levels of contaminants in drinking water that would 12

35 pose no significant health risk to individuals consuming the water on a daily basis over a lifetime of exposure. The PHG for MTBE is based on carcinogenic effects observed in animals. They are strictly health-based standards set at levels that OEHHA has determined do not pose any significant risk to health. The PHG of 13 µg/l will be used as an advisory action level by DHS until a primary drinking water standard is established for MTBE. The action level will be used to protect against health risks associated with exposures to MTBE in drinking water. Secondary Drinking Water Standards On January 7, 1999, DHS adopted a secondary drinking water standard of 5 µg/l for MTBE. This enforceable SMCL was based on the aesthetic concerns associated with taste and odor detections at low concentrations. This level was established to protect the public from exposure to MTBE in drinking water at levels that can be smelled or tasted. If a drinking water source for a public water system is discovered to contain MTBE in excess of 5 µg/l, the water system is required to notify the governing body of the local agency in which users of the drinking water reside within 30 days of the discovery. The DHS will advise the utility to remove the contaminated source from service, but the utility may elect to continue using the contaminated water. The utility may choose to continue to deliver water above the secondary standard if it has no other sources to meet water system demands. However, the utility will have to provide public notification to each of its customers. Primary Drinking Water Standards A primary drinking water standard of 13 µg/l has been proposed by DHS and is expected to be adopted in early to mid Primary standards include consideration of health risks, technical feasibility of meeting the proposed level, and costs associated with compliance monitoring. Primary drinking water standards are not to be exceeded in water supplied to the public. Best Available Technology (BAT) In California, as part of the establishment of a primary drinking water standard, a BAT for treating drinking water contaminated with MTBE must be identified. Selection of the BAT has to take into consideration the costs and benefits of the best available technology that has proven effective under full-scale field applications (California Health and Safety Code, Section ). The September 7, 1999 proposed rulemaking for establishment of the Primary Drinking Water Standard, identified air stripping as the BAT for treating water contaminated with MTBE. Because there were only a limited number of sites which had full-scale field applications and because those sites identified were only air stripping systems, the proposal of air stripping as the BAT relied on these field applications. There was 13

36 not sufficient data in regards to other technologies, including AOPs, GAC, biological treatment, resins, and membrane separation. Air stripping is expected to be approved as the BAT by DHS for MTBE. However, it is also likely that, due to the significant amount of data compiled on the use of GAC for removal of MTBE, GAC may also be included as a BAT. Permitting Requirements in California In California, domestic water supply permits are issued by the DHS Office of Drinking Water. The Field Operations District Offices are responsible for the inspection and regulatory oversight of approximately 8,500 public water systems to assure delivery of safe drinking water to all California consumers. The oversight includes issuing permits, performing inspections of existing systems, reviewing and approving plans for new facilities, issuing administrative orders and citations for violations against systems in non-compliance with laws and regulations, and ensuring that water quality monitoring is conducted in accordance with the law. The installation of a treatment system to treat municipal wells impacted by groundwater contamination requires the approval of DHS. Water systems considering the installation of a treatment system to remove contaminants from groundwater have to undergo a formal process to obtain a permit or amend an existing water supply permit to permit the installation of the treatment system. Typically, the permit would require the applicant to complete an application and prepare a technical report providing sufficient detail to show that the proposed treatment system can produce a continuous supply of pure, wholesome, healthy, and potable water. The technical report must provide sufficient detail to allow DHS to prepare an engineering report that documents the findings of the DHS investigation and provides recommendations for issuance of a domestic water supply permit to operate the proposed treatment system. The DHS engineering report will provide information as to the source of the groundwater, historical and current extent of the contamination, potential impacts of the hydrogeology, potential contaminant migration, and effects of the underlying aquifers. The report will also address the sanitary hazards and safeguards that protect the well. Recommendations are then developed for the issuance of the water supply permit to include a detailed description of the treatment train, design flow rates, and operational constraints. Specific monitoring is defined in accordance with the DHS Vulnerability Assessment and Monitoring Frequency Guidelines and a monitoring plan is established along with the effluent standards that must meet all MCLs and Action Levels established by DHS. Daily operational records, such as flow rate, volumes treated, chlorine residual measurements, bacteriological sample testing, and operation changes, must be maintained. A specific operations plan must also be developed by the permittee. 14

37 Policy Memo Policy Guidance for Direct Domestic Use of Extremely Impaired Sources In November 1997, DHS prepared policy guidance Memorandum No for the direct domestic use of extremely impaired sources. The document provides the philosophy behind the policy, its purpose, and the specific elements of the evaluation process by which DHS approves the use of extremely impaired sources. The document sets forth the guidelines by which DHS approves the use of extremely impaired sources for drinking water use and the establishment of permit conditions. The drinking water source is considered the production well(s) and the aquifers(s) from which they draw. This guidance document outlines the elements of the evaluation process, including the 12-step process that would have to be conducted to receive a permit to use an extremely impaired drinking water source for domestic supply. Because the requirements would impact the selection of treatment for MTBE, relevant requirements of the policy are discussed in this document. As stated in Memorandum , DHS subscribes to the basic principle that only the best quality sources of water should be used for drinking. The memorandum goes on to say that these sources should be protected against contamination. DHS recognizes that the use of contaminated water as a drinking water source always poses a greater health risk and the use of an extremely impaired source should not be approved unless the additional health risks are known, minimized, and considered acceptable. In addition, DHS clearly states in the memorandum that drinking water quality and public health shall be given greater consideration than costs. This process requires the completion of 12 tasks in order to restore the use of water from the contaminated well field for domestic drinking water purposes. The permit application requires completion of the first eight tasks by the applicant. The 12 tasks identified in DHS Policy are as follows: 1. Perform Source Water Assessment 2. Perform Raw Water Quality Characterization 3. Develop Source Protection Program 4. Develop Effective Monitoring and Treatment 5. Develop Human Health Risks Associated with Failure of Proposed Treatment 6. Identification of Alternative Sources and Comparison of the Potential Health Risks to those of the Project Potential Health Risk 7. Completion of California Environmental Quality Act (CEQA) Review 8. Complete Permit Application 9. Public Hearing 15

38 10. DHS Evaluation 11. Requirements for DHS Approval 12. Issuance or denial of Permit After DHS reviews the permit application packet (Task 10) and other relevant information submitted, DHS will then request additional information or proceed to the Public Hearing (Task 9). Following the public hearing, DHS will either issue the permit with requirements, deny the permit (Tasks 11 and 12), or request additional information. Task 4 of the DHS Policy will be discussed here due to its importance for understanding these requirements when selecting treatment for water contaminated with MTBE. The treatment process for the use of extremely impaired sources must be at a minimum, the best available treatment technology as defined for the contaminant by the EPA or established by DHS. The treatment process must also be sufficiently reliable to the type and degree of contamination and must address all contaminants. Because the treatment process must also be commensurate to the degree of risk associated with the contaminants present and since a BAT for MTBE has not been established, the treatment process would have to be able to reliably produce water that meets drinking water standards. The use of multi-barrier or independent treatment systems operated in series could provide effective treatment as long as each treatment system is designed to treat all contaminants to meet drinking water standards. In accordance with Policy , it can be construed that the use of multi-barrier treatment applications are expected to be required for sources containing MTBE that are extremely impaired under the following conditions: The Primary Treatment is not sufficiently reliable. The Primary Treatment is of uncertain effectiveness. There is no direct way to measure the contaminant. The health effect of the contaminant is acute. Very large reductions in contaminant concentration are required. The treatment processes employed to treat sources contaminated with MTBE must protect the public from exposure to water-borne contaminants and must comply with all MCLs and action levels. The treatment processes must demonstrate the ability to meet varying source water quality while maintaining acceptable discharge criteria. The treatment process employed must include multiple levels of redundancy, must accommodate fluctuations in influent contaminant concentrations, and must be reliable to produce the purest wholesome water. Multi-barrier treatment systems are expected to be required under most sites impacted by MTBE in California. As such, these sites would require the installation of a redundant 16

39 treatment system. In most cases, this redundant system will be GAC, due to its proven ability to remove MTBE to levels lower than the secondary MCL of 5 ug/l. In addition, GAC is the least expensive treatment process to install as a redundant process Analytical Methods For drinking water, there are two most commonly used EPA analytical methods under existing regulations for detection of MTBE: EPA methods and These methods are used to quantify a significant number of regulated contaminants by most drinking water systems. Analogous to these two methods are EPA methods 8020 and 8260, which are used for nearly all types of water samples, but are most commonly used for water samples collected from contaminated groundwater sites. EPA method 8020 and EPA method 502 both use a Purge and Trap Capillary Column Gas Chromatography with Photoionization. EPA method 8260 is similar to EPA method in that they both use Capillary Column Gas Chromatography with Mass Spectrometry Data System. A discussion on the two drinking water methods and their applicability is presented below. EPA Method 502: Volatile Organic Compounds in Water by Purge and Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic Conductivity Detector in Series This method covers over 60 volatile organic compounds that contain halogen atoms and/or that are aromatic. It is primarily used for detection and quantification of contaminants in drinking water and other source waters. In this method, inert gas is bubbled through a 25 ml or 5 ml water sample. The quantity is dependent on the expected concentrations of analytes in the water sample. Purged sample components are trapped in a tube of sorbent material, which is then heated and backflushed with helium to desorb the trapped sample onto a capillary gas chromatography column. The analytes are then detected using a photoionization detector and a hall electrolytic conductivity placed in series. The photoionization detector is used for the aromatic compounds and the hall electrolytic coductivity is used for the halogenated compounds. The method detection limits are dependent on the characteristics of the gas chromatographic equipment used, but can be as low as 0.1 µg/l for MTBE. EPA Method 524.2: Measurement of Purgeable Organic Compounds in Water by Capillary Column Gas Chromatography with Mass Spectrometry Data System This method covers over 60 volatile organic compounds. It is primarily used for detection and quantification of contaminants in drinking water. In this method, inert gas is bubbled through a 25 ml or 5 ml water sample. The quantity is dependent on the expected concentrations and analytes in the water sample. Purged sample components are trapped in a tube of sorbent material, which is then heated and backflushed with helium to desorb the trapped sample 17

40 onto a capillary gas chromatography column. The analytes are then detected using the mass spectrometry data system. The method detection limits are dependent on the characteristics of the gas chromatographic mass spectrometry data system equipment used, but are commonly as low as 0.1 µg/l for MTBE. EPA methods 502 and are both methods acceptable for determining MTBE concentrations in groundwater; however, a significant number of false positive detections are suspected and have been reported using method 502. False positive detections occur when EPA method 502 detects MTBE while a confirmation analysis using EPA method does not detect MTBE. The misidentification using EPA method 502 may be attributed to the presence of other contaminants (hydrocarbons) that may coelute with MTBE and then are incorrectly identified by the photoionization detector. False positive detections may be expected in other EPA methods (such as 8020), which also use Purge and Trap Capillary Column Gas Chromatography with Photoionization. As laboratories gain further experience analyzing MTBE, it is less likely that false positive results will be obtained. However, detections of MTBE using EPA method 502 should always be confirmed using EPA method

41 1.4 Integration of Technologies This report presents an evaluation of air stripping, AOPs, GAC, and synthetic resin sorbents for removal of MTBE from drinking water. During the formation of the Partnership s Treatability Subcommittee, it was recognized that other emerging technologies are also applicable for MTBE drinking water treatment, including membranes and biological processes. The Research Partnership continues to recognize these and other alternative treatment processes as representing significant potential; however, they were not reviewed in this report due to their limited pilot and field testing and emerging nature. As additional field and pilot-scale research is completed by other entities, the Research Partnership or other MTBE research organizations may choose to pursue these alternative treatment options in the future. Despite the relatively proven nature of the treatment processes selected for review in this report, many treatment and remediation scenarios may require a combination of treatment processes to meet effluent criteria. For example, as noted above, all drinking water treatment process in California will likely require redundant treatment (i.e., a treatment process that serves as a barrier between the consumer and the contamination). In many cases, GAC will be used as the redundant process. Similarly, AOPs will often employ additional treatment processes to remove oxidation by-products or residual oxidants, as discussed in Chapter 3. Alternatively, some treatment processes may not be able to reach drinking-water effluent criteria alone, especially in cases where the influent concentrations are high (e.g., 2,000 µg/l) and the effluent criteria is non-detect (e.g., 0.5 µg/l). In these circumstances, two treatment processes may be required the first process would be used to lower concentrations by 90 to 99 percent and the second process would be used as a polishing step to remove the remaining 1 to 10 percent. Throughout this report, the economic evaluations for each treatment process were completed such that this process integration analysis is possible. For example, it is possible to estimate air stripping or AOP costs for removal of MTBE from 2,000 µg/l to 200 or 20 µg/l and GAC costs for removal of MTBE from 200 or 20 µg/l to 0.5 µg/l. The costs presented in this report have relied on similar assumptions for mark-up and water quality and, thus, the accuracy of the cost estimate for MTBE removal should not be affected by combining individual process costs Conclusion In summary, this report presents a detailed literature review of the state-of-knowledge regarding the use of air stripping, AOPs, GAC, and synthetic resin sorbents for removal of MTBE from drinking water. In addition, each chapter presents a cost analysis, which can be used to estimate drinking water treatment costs as a function of flow rate, influent concentration, and removal efficiency. However, this cost analysis does not present costs as a function of water quality and should not be used in place of a site-specific engineering cost estimates. The Partnership is currently developing scopes of work or implementing projects to fill some of the research gaps identified within this document for each technology. 19

42 20

43 1.5 References American Petroleum Institute (API) Odor Threshold Studies Performed with Gasoline and Gasoline Combined with MTBE, ETBE, and TAME. TRC Environmental Corporation. Conducted by TRC Environmental Corp. ARCO Chemical Company The Odor and Taste Threshold Studies Performed with Methyl Tertiary-Butyl Ether (MTBE) and Ethyl Tertiary- Butyl Ether (ETBE). TRC Environmental Corporation. Conducted by TRC Environmental Corp. Crittenden, Dave, Dave Hand, et al Environmental Technologies Design Option Tools (ETDOT) for The Clean Process Advisory Systems (CPAs) Adsorption, Aeration and Physical Properties Software. National Center for Clean Industrial and Treatment Technologies. Dale, M.S., Moylan, M.S., Koch, B., Davis, M.K MTBE: Taste-and-Odor Threshold Determinations Using the Flavor Profile Method. Metropolitan Water District of Southern California. Hanson, J.R., Ackerman, C.E., and Scow, K. M., J. Nov Biodegradation of Methyl tert- Butyl Ether by a Bacterial Pure Culture. Applied and Environmental Microbiology. Vol. 65, No. 11, pp Malcolm Pirnie, Inc. June, Taste and Odor Properties of Methyl Tertiary-Butyl Ether and Implications for Setting a Secondary Maximum Contaminant Level. Prepared for the Oxygenated Fuels Association, Inc. Office of Science and Technology Policy (OSTP). June Executive Office of the President. National Science and Technology Council. Committee on Environmental and Natural Resources. Interagency Assessment of Oxygenated Fuels. Washington, DC. Salanitro, J. P., G. E. Spinnler, C. C. Neaville, P. M. Maner, S. M. Stearns, P. C. Johnson, and C. Bruce Demonstration of the Enhanced MTBE Bioremediation (EMB) In Situ Process. The Fifth International In Situ On-Site Bioremediation Symposium. Squillace P. J., Pankow, J. F., Korte, N. E. and Zogorski, J. S Review of the Environmental Behavior and Fate of Methyl Tert-butyl Ether. Environmental Toxicology and Chemistry 16 (9): Shen, Y.F., Yoo, L.J., Fitzsimmons, S.R., Yamamoto, M.K., Threshold Odor Concentrations of MTBE and Other Fuel Oxygenates. Orange County Water District. Young, W.F., Horth, H., Crane, R., Ogden, T., Arnott, M Taste and Odour Threshold Concentrations of Potential Potable Water Contaminants. Vol. 30, No. 2 pp

44 22

45 2.0 Air Stripping Andrew Stocking, P.E. Hinrich Eylers, Ph.D. Michael Wooden Terri Herson Michael Kavanaugh, Ph.D., P.E. 23

46 24

47 2.1.1 Air Stripping Application for MTBE Removal from Drinking Water 2.1 Background Although MTBE has been commercially used in the United States since 1979, there are relatively few cases where drinking water supplies have been contaminated with MTBE at concentrations requiring treatment. Based on a limited review of such cases, only two major public water supplies have been identified where MTBE treatment has been implemented and the effluent from the treatment system is used as drinking water: Rockaway Township, New Jersey and LaCrosse, Kansas. In Rockaway Township, the groundwater was contaminated with several volatile organic compounds (VOCs), including MTBE. Initially, GAC was used, but costs were excessive (McKinnon and Dyksen, 1984). Subsequent modifications to the treatment system included using a packed tower air stripper prior to GAC polishing. The combined treatment process of air stripping followed by liquid phase GAC reduced initial MTBE concentrations at approximately 96 µg/l to below detection limits (approximately 5 µg/l). The volumetric air/water ratio used in the air stripping system was 200. In LaCrosse, Kansas, influent MTBE concentrations were as high as 900 µg/l. The treatment system, which went into operation in 1997, consists of two packed tower air strippers operated in series with an air/water ratio of 175 in each tower. The first air stripping tower typically reduces MTBE concentrations by approximately 90 percent and the second tower has consistently removed any remaining MTBE to below the treatment goal of 10 µg/l and commonly much further (detection limit is 0.2 µg/l). Appendix 2B provides a detailed description of the LaCrosse, Kansas facility as well as some operating data. In these two cases, MTBE is being successfully removed from drinking water using a packed tower air stripper, although the air/water ratios (based on a volumetric basis) are relatively high. For comparison, the removal of trichloroethylene (TCE) in drinking water applications typically requires a much lower air/water ratio of less than 30. Air stripping technologies are widely used for removing halogenated VOCs from drinking water supplies prior to distribution and use of the water for public consumption. Packed tower aeration is the most common air stripping technology for drinking water treatment. Packed tower aeration is a well-understood and proven technology (Roberts et al., 1985; Kavanaugh and Trussel, 1980), and there are many equipment vendors and packing manufacturers who provide the external and internal components for packed tower systems (Lamarre and Shearhouse, 1996). Other air stripping technologies have been used, but primarily at low flow rates (<100 gpm) or in a remediation context. Other air stripping technologies include spray towers, bubble aerators, low profile aerators, surface aerators, and aspiration or centrifugal aeration devices. 25

48 2.1.2 Objectives of the Evaluation Although air stripping using a packed tower is a widely used technology for VOCs, its application to MTBE removal from drinking water has been limited. In addition, this technology has several potential disadvantages that may limit its widespread use in drinking water applications. These limitations include: increased costs due to the potential need for off-gas treatment; delay due to permitting; aesthetic constraints due to tower heights; and some concerns over mechanical reliability. Also, the use of packed tower air strippers brings the groundwater in contact with the atmosphere, which may add other contaminants to the water. Consequently, the objectives of this chapter are to review air stripping and off-gas treatment technologies commonly used in drinking water applications and determine the most costeffective and reliable air stripping system option for MTBE removal from drinking water. This chapter will review the technologies available for air stripping and off-gas treatment and discuss each technology in terms of cost-effectiveness, reliability, and ease of implementation. The chapter will then conclude with recommendations for air stripping and off-gas treatment combinations for a variety of water treatment scenarios. 26

49 2.2 Description of Technology - Air Stripping Background As previously mentioned in Chapter 1, MTBE is a polar chemical and, as a result, is relatively soluble in water (48,000 mg/l at 20 C). MTBE has a relatively low Henry s constant and, thus, requires a higher air/water ratio in air stripping towers compared to the required air/water ratio for benzene or TCE. However, air stripping has been effectively used to remove MTBE from water, particularly from dilute solutions (i.e., less than 1 mg/l) that are typically associated with groundwater contamination from leaking underground fuel tanks (LUFTs) (Creek and Davidson, 1998) Process Principles The effectiveness of air stripping technologies to remove organic contaminants from water depends upon the volatility of the compound from water and the physical design of the air stripping technology. Air stripping relies on an equilibrium phase transfer process where the contaminant partitions between the aqueous phase and the air phase. The equilibrium partitioning coefficient is called the Henry s constant which, in dilute solutions, is determined by Raoult s law using the vapor pressure of the pure compound and its water solubility. In general, the higher the Henry s constant for a contaminant, the more effective air stripping will be for that contaminant. The Henry s constant for MTBE was recently reported to range from to (dimensionless) at 20 C (OSTP, 1997); however, the Henry s constant is typically thought to be closer to the lower end of this range. Figure 2-1 reports various Henry s constants for MTBE as a function of temperature, as compiled by Paul Sun of Equilon Enterprises, L.L.C. (Sun, 1998). At 20 C, MTBE s Henry s constant is approximately This value is several times lower than Henry s constants for common organic compounds found in groundwater such as TCE, PCE, or benzene. Thus, air stripping of MTBE is generally more difficult and more costly than removal of these other compounds. The effectiveness of air stripping organic compounds from water can be improved by raising the water temperature which, in turn, increases the Henry s constant. However, unless the heat is free waste heat from another process (e.g., thermal off-gas treatment), heating the water is expected to be cost-prohibitive. The low Henry s constant of MTBE may require a larger air stripping system to achieve the desired removal efficiency; however, air stripping may still be cost-effective relative to other technologies for MTBE treatment. One significant issue is the design of the contacting system between the contaminated water and the air used to strip out the organic compounds. In general, the goal is to maximize the extent of contact (maximum rate of mixing, highest specific surface area) while minimizing energy costs associated with the equipment design. This provides the highest rate of mass transfer from water to air at the lowest operating cost. The most common mass transfer design for air stripping systems is the use of randomly 27

50 packed towers. Other options include spray towers, low profile units, bubble diffusers, aspirators, and surface aerators. Selection of the appropriate technology is often site-specific. Henry's Law Constant for MTBE, Atm M 3 Mol H = *(293T)* *(1/293-1T) T = Temperature in K Shell Estimate Calgon Data WTC-Brutcher WTC-Tang WTC-Rodden IT data WTC-Wilcox McKay Data Temperature in C K eq = Mol/L-gas / Mol/L-Liquid) = H c / /(273 + C) Figure 2-1. Estimated vs. experimental data for MTBE s Henry s constant as a function of temperature. The following sections review the major established and emerging air stripping technologies applicable to MTBE treatment in drinking water applications. For a review of design issues, the reader is referred to standard textbooks on air stripping systems (Montgomery, 1985) Aeration Technologies Several types of air stripper technologies are currently available for removal of MTBE from water. This evaluation focuses on the capabilities and limitations of several major established and emerging air stripper technologies that are potentially applicable for MTBE removal. Applicability is evaluated based on the performance reported in engineering literature, vendor information, and professional experience with the equipment. Air stripper technologies evaluated in this report include: Packed Tower Aeration Low Profile Aeration Bubble Diffusion Aeration 28

51 Spray Towers Aspiration A brief description of each technology, including its advantages and disadvantages, is presented in Table 2-1. A detailed discussion of each technology follows. Packed Tower Aeration System Description In a packed tower, contaminated water flows downward by gravity through a circular or rectangular column that is filled with either randomly packed or structured packing material. Air is introduced into the tower below the packed bed and flows upward through the column countercurrent to the flow of water. The air can either be dispersed into the bottom of the column (forced draft system) or drawn out of the top of the column (induced draft system). The packing material is designed to maximize available specific surface area for contact between the contaminated water and the process air, thereby providing the maximum specific surface area possible for volatile contaminants to move from the liquid phase to the gas phase. Initial distribution of the influent water and process air over the entire cross section of the column is usually accomplished by vented orifice trays, influent troughs, or a spray nozzle header system. Of the three options, an orifice tray distributes air and water most effectively across the entire cross-sectional area of the column. The mathematics of packed tower aeration have been studied extensively and can be described relatively accurately by various correlations (e.g., Onda equations). While not presented in this text, full mathematical explanations are available elsewhere (Perry et al., 1984). An illustration of a packed tower system manufactured by Carbonair Environmental Systems, Inc. (New Hope, MN) is shown in Figure

52 30 Air Stripping Technology Packed Tower Low Profile Bubble Aeration Spray Tower Aspiration Brief Description Water trickles downward over packing media, creating a thin film. These thin films of water are met by a counter flow of air blowing in from the bottom of the tower. A series of stacked perforated trays with countercurrent air stream suspends water and volatilizes contaminants. Rising bubbles and turbulence provide the air/water interface needed for stripping without the need for packing or media. Water is sprayed downward through nozzles into a collection sump. Air is blown upward counter currently to the water and exits through a demister at the top of the column. A Venturi stripper uses highly turbulent jets of water to shear and accelerate fluid films within an open bore. System Components Supply Pumps Air Blower Influent and Discharge Pipes Supply Pumps Air Blower Influent and Discharge Pipes Supply Pumps Air Blower Influent and Discharge Pipes Supply Pumps Air Blower Influent and Discharge Pipes Supply Pumps Air Blower Influent and Discharge Pipes Advantages High flow capacity. Removes difficult to strip compounds. Low liquid pressure drop. Proven technology. Compact (low profile). Easily installed and maintained. Proven technology. High liquid and air turndown ratio. Simple device, low maintenance. Short set-up time. Low potential for fouling. Low pressure drop for gas (low blower cost). Simple operation (no mechanical parts). Compact - low profile. Compact and low profile. No significant problems with misting, freeze up or slime growth. Installation and operation can be easily staged; units can be installed or removed. Removal efficiency independent of air temperature. Low off-gas volume. Disadvantages Fouling results in loss of efficiency, and increased pressure drop. High gas pressure drop. Transportation/set-up more complex than low-profile systems. Channeling of water through packing may short-circuit treatment. Highly visible (profile) Multiple units typically required for flows >100 gpm. Scale formation dramatically decreases treatment efficiency. High gas-pressure drop requires high horsepower blower. High removal efficiencies require multiple units. High pressure drop for liquids. Packing may be required for MTBE removal (subject to fouling). Fouling of nozzle may reduce treatment efficiency. Low turndown ratio (unless nozzle is changed). May be high profile. Increased operating cost due to higher energy demands. Large footprint for high (> 90%) removal efficiency and moderate flows (> 100 gpm). Table 2-1 Description of Air Stripping Treatment Technologies and Their Advantages and Disadvantages

53 ANSI FLANGE (INFLUENT) OFF-GAS MANWAY INFLUENT DISTRIBUTOR PACKING MEDIA OPTIONAL INFLUENT RISER PIPE PACKING SUPPORT BLOWER SIGHT GLASS ANSI FLANGE (EFFLUENT) Figure 2-2. An illustration of a packed tower system manufactured by Carbonair (1995). 31

54 Advantages/Disadvantages Table 2-1 describes the primary advantages and disadvantages of various air stripping systems. Additional advantages of packed tower systems are listed as follows (Lenzo, 1985; Lenzo, 1994): Packed towers are among the most widely implemented VOC removal systems available today. They are commonly custom manufactured to meet the specific requirements of each application, although it is also possible to buy an off-the-shelf system. Packed towers have been used successfully to remove MTBE from water in drinking water applications. Computer models are available to design and optimize packed tower air strippers. Manufacturers often provide proprietary design programs based on a database of empirical performance data that take into account non-ideal flow and the impacts of water quality on cost. Thus, packed tower air stripping can be designed to achieve a high degree of process reliability. One disadvantage, however, is that operation and maintenance of packed towers can be impacted by characteristics of the influent water unrelated to the contaminants of concern. Four common operational problems and their mitigation measures are as follows: Corrosion. Contact between the water and the aluminum or steel tower can lead to corrosion, which weakens the tower frame and necessitates tower replacement. This can be avoided by coating the aluminum or steel with epoxy or by using stainless steel or fiberglass reinforced plastic (FRP) instead. However, FRP is less durable and becomes more prone to biological fouling as the ultraviolet-inhibiting materials within the FRP degrade. Scaling. The contact between contaminated water and the air stream in a packed tower typically results in ph increases during treatment. In cases of high influent water hardness, scaling (i.e., precipitation of calcium carbonate or calcium sulfate onto the packing media, column internal structures, and effluent piping) can occur. Scaling can be reduced or prevented by lowering the influent water ph using acid feed or by injecting anti-scaling chemicals into the influent. Periodic cleaning (e.g. with acid solutions) of the packing media, internal structures, and piping and/or replacement of the packing material may be required if scale prevention measures are not adequately employed in normal operation (Snoeyink and Jenkins, 1980; Lenzo, 1994). Iron Fouling. Groundwater is often low in dissolved oxygen and, therefore, contains iron mainly in the Fe 2 + (ferrous iron) oxidation state. When ferrous iron comes in contact with oxygen during the aeration process, it is oxidized to ferric iron (Fe 3 +) which forms an 32

55 insoluble precipitate that will lead to fouling. Prevention of iron fouling in packed columns is more difficult than prevention of scaling. In the most severe situations, periodic packing replacement may be required. Biological Fouling (Biofouling). If the packed tower internals are exposed to light and the influent water contains sufficient organic matter to sustain microbial growth, packed columns may be subject to biological fouling due to bio-growth or algae formation. Biofouling can be prevented by injection of a disinfectant (e.g., sodium hypochlorite) to the column influent stream, but this is limited by the need to minimize disinfection by-products such as trihalomethanes (THMs). It is important to note that the operational problems associated with corrosion, scaling, iron fouling, and biofouling also apply to the other air stripper technologies that are described below. However, the extent to which individual technologies are vulnerable to these factors may vary. Other disadvantages of a packed tower include: Short-circuiting due to poor water or air distribution, which can limit the system s maximum removal efficiency. Aesthetic concerns due to high visibility of packed towers. Key Variables/Design Parameters The removal efficiency of organic contaminants by packed towers is a function of many parameters (see Table 2-2). Manufacturers typically provide cylindrical towers with a limited selection of diameters. Economic considerations determine the trade-off between tower volume and air/water ratio as a function of standard air pressure drop and a given packing media. Because the tower volume directly affects capital costs, design optimization involves minimizing tower volume at a pressure drop that minimizes energy requirements. In any given application, the optimal liquid loading rate, packing height, and air/water ratio will be functions of site-specific characteristics of influent water quality, required VOC removal efficiencies, operational considerations, and economics as well as aesthetic concerns (see Section 2.6 for further discussion of system optimization). In addition, Table 2-2 shows the effects of increasing various parameters on the removal efficiency and cost (assuming a fixed tower volume, height, and packing) and the design of the packed tower (assuming a fixed removal efficiency). For example, in a groundwater treatment application, for a given tower design (fixed packing type, diameter, and height), increasing the water pumping rate to meet water demands will increase liquid loading. This causes a decrease in the air/water ratio, resulting in a decrease in removal efficiency and an increase in operating costs due to the greater volume of air required to meet the target removal efficiency. Similarly, while raising the influent water temperature will decrease the required tower volume for a given removal efficiency, it will also 33

56 Table 2-2 Packed Tower Design Variables Parameter Liquid Loading Rate Air/Water Ratio (AWR) Water Temperature Henry s Constant Packing Type and Size Pressure Drop / Depth Effect of Increasing ( ) Parameter on Operations and Cost, Assuming no Change in Tower Design Removal Efficiency Cost Removal Efficiency Cost Removal Efficiency Heating Cost Henry s Constant Removal Efficiency the Size Removal Efficiency Removal Efficiency Pump/Blower Cost Effect of Increasing ( ) Parameter on Tower Design, Assuming Removal Efficiency is Maintained Tower Height (HTU) Packing Volume Packing Volume Packing Volume AWR the Size Packing Volume Pressure Drop AWR increase operating costs substantially. In the case of tower design, the higher the design loading rate, the greater the tower height needed to achieve design removal efficiencies. System Installations and Manufacturers Packed towers, operated in parallel or in series, remove a wide range of volatile organic contaminants at many water treatment facilities in the United States at flow rates ranging from less than 1 million gallons per day (694 gpm) up to 20 million gallons per day (approximately 13,900 gpm). As mentioned in Section 2.1.1, the effluent water from packed tower air strippers has been used as drinking water in many treatment cases; notably, the MTBE removal sites in LaCrosse, Kansas and Rockaway Township, New Jersey (see Appendix 2B for a detailed description of the LaCrosse, Kansas packed tower air stripping operation). Packed towers with diameters up to 15 feet are manufactured from a number of materials, including plastic, fiberglass, aluminum, and steel, and are being used to remove a variety of contaminants. The largest manufacturer of pre-engineered systems is Layne Christensen Company (Bridgewater, NJ) with over 400 installations nationwide (approximately 280 installations in drinking water facilities). Layne Christensen Company has installed approximately 10 packed towers specifically designed for MTBE removal; these are installed at remediation sites and not at drinking water facilities. Tonka Equipment Company (Plymouth, MN) is the second largest manufacturer with approximately 100 installations, mostly in the Midwest. Carbonair (New Hope, MN) also has approximately 150 packed tower drinking 34

57 water installations. Appendix 2C lists the addresses of Layne Christensen Company, Tonka Equipment Company, Carbonair, and other air stripper equipment vendors referenced in this chapter. A variety of plastic packing media is available. Both structured packing media and randomly packed media are commercially available in various types of plastics and in a wide variety of sizes and geometric configurations. Most of these media, especially the newer generation of high efficiency packing, are proprietary products. Technical Implementability A packed tower air stripper requires supply pumps (sized only to provide the static head required to move the influent water to the top of the unit), air blowers, and a single influent and discharge pipe. There are no moving parts (blowers only) and, therefore, there are minimal daily maintenance requirements. If fouling is a concern, a chemical feed system is required, necessitating additional maintenance. Construction of a packed tower requires a relatively heavy foundation to allow the installation of seismic braces and wind supports. Tower heights can be reduced by placing two or more towers in series, which mitigates tower susceptibility to strong winds or light seismic activity. Low Profile Aeration System Description In a low profile aeration system, contaminated water is pumped to the top of the stripper where it flows over an inlet weir onto a baffled aeration tray. Air is forced up through perforations in the tray bottom by either a forced-draft or an induced-draft blower, creating highly turbulent conditions to maximize the contact of water and air. Multiple trays may be vertically stacked, with water flowing from upper trays to lower trays via overflow weirs. The contact time necessary to achieve the desired VOC removal efficiency is provided by selection of the tray size, liquid flow rate, and number of stacked trays. The air is discharged at the top of the aeration unit. A minimum air flow is required to prevent water from entering the aeration holes. An illustration of a low profile system manufactured by Carbonair Environmental Systems, Inc. (New Hope, MN) is shown in Figure

58 LEVEL CONTROL SIGHT GLASS AIR FLOW EXHAUST AIR FLOW METER WATER TEMPERATURE AIR TEMPERATURE BLOWER MUFFLER SAMPLE TAP PUMP DOWN Figure 2-3. An illustration of a low profile system manufactured by Carbonair (1995). 36

59 Advantages/Disadvantages See Table 2-1 for a summary of the advantages and disadvantages of low profile aerators. A more detailed listing of the advantages of low profile air strippers includes: Skid-mounted configuration that allows the units to be placed on a concrete pad or level floor surface by forklift. Also, the unit can be placed in a heated building for cold-weather operation. Limited mechanical connections blower inlet, stack gas discharge, and process water influent and discharge lines. Pre-packaged power and control panel for remote monitoring, and operation of pumps and blowers as well as operation of strippers in batch or continuous mode. Easy maintenance and cleaning of fouled low profile systems by removal of interlocking trays followed by pressure washing with or without chemical (e.g., acid) cleaning. Availability of vendor computer models based on empirical field-scale and bench-scale data that model non-ideal flow and water quality scenarios for design optimization. Disadvantages of the low-profile air stripper include those listed previously in this section for packed towers (i.e., corrosion, scaling, iron fouling, and biological fouling) in addition to (Malcolm Pirnie, Inc., 1992): Performance drop off with scale formation, as described for packed towers. If scaling causes the perforated trays to seal, performance can drop off much more rapidly than packed towers. Minimal air turndown capacity due to the extreme performance drop off if process water is no longer suspended on the tray surface and instead falls through the tray perforations. Not always weatherproof and, therefore, may require an enclosure for protection. Key Variables/Design Parameters Similar to a packed tower aeration system, the removal efficiency of a low profile aeration system is a function of many parameters, including water temperature, air/water flow ratio or contact time, number of trays, contact time, and the volatility of the contaminant. The effects of these parameters on the removal efficiency of a low profile system is similar to the case of a packed tower system (Table 2-2). The optimal system configuration will depend on sitespecific characteristics of the influent water, required VOC removal efficiencies, operational considerations, and economics. 37

60 System Installations and Manufacturers Low profile aeration technology is a widely proven air stripping process for drinking water and non-drinking water applications. The two primary vendors of low profile air strippers are North East Environmental Products, Inc., (West Lebanon, NH), manufacturer of the ShallowTray stripper, and Carbonair Environmental Systems, Inc. (New Hope, MN), manufacturer of Carbonair STAT. North East Environmental Products is the largest distributor with more than 4,000 units in operation across the country for removal of a variety of contaminants. Three hundred of these units are used for small-community and municipal water treatment applications, and approximately 1,000 units are used as household point-ofentry treatment units. Individual low profile units are capable of removing MTBE from water for flow rates up to 1,100 gpm, and parallel units are used for flow rates above this level in systems with capacities up to several million gallons per day. However, the flow rates are usually much lower (<100 gpm) (Shearhouse, 1998). There are over 200 low-profile installations in use for MTBE removal in groundwater remediation applications and at least five low-profile installations where the treated water is potable (there was no specific information identified for potable water installations, but these are likely to be small systems). See Appendix 2B for a description of the low profile system initially used at LaCrosse, Kansas. Technical Implementability As with packed towers, low profile air strippers require supply pumps, air blowers, and a single influent and discharge pipe. Aside from the blowers and pumps, there are no other moving parts; therefore, maintenance requirements are minimal. A discharge pump may be required to move the effluent water to the next stage of the process train (Lenzo, 1994). Low profile units are generally easy to dismantle and clean. The units are typically skid-mounted (resulting in easy installation) and require a small footprint or foundation. Bubble Diffusion Aeration System Description In a diffused bubble aeration system, air is released through fine bubble diffusers at the bottom of a water-filled tank, which is usually divided into baffled stages. Rising bubbles create turbulent mixing, which provide the air/water contact area necessary for contaminant stripping. If longer residence time is need to increase contaminant stripping, bubble aeration systems can be designed with baffles to create multiple chambers in a single unit or multiple units to increase the total water depth. MTBE removal efficiencies greater than 90 percent are expected to require the use of multiple units in series. Design variables requiring optimization include the basin depth, number of aeration stages, air flow rate, and water flow rate. 38

61 Advantages/Disadvantages See Table 2-1 for a summary of the advantages and disadvantages of this technology. A more detailed listing of the advantages of bubble aeration systems includes: Considerably lower profiles (less than 6-feet high) than packed towers. Skid-mounted configuration that allows the units to be placed on a concrete pad or level floor surface by fork-lift. Also, the unit can be placed in a heated building for cold-weather operation. High liquid and air turndown ratio (i.e., the ability to lower the liquid or air flow rates without radically decreasing performance). Limited mechanical connections blower inlet, stack gas discharge, and process water influent and discharge lines. Pre-packaged power and control panel for remote monitoring, and operation of pumps and blowers as well as operation of strippers in batch or continuous mode. Easy maintenance and cleaning of fouled baffles by removal of interlocking trays followed by pressure washing or pressure washing in combination with chemical (e.g., acid) cleaning. Disadvantages, as listed in Table 2-1, of the bubble aeration system include those listed previously in this section for packed towers (i.e., corrosion, scaling, iron fouling, and biological fouling) in addition to: High removal efficiencies (>90 percent) for high flow (>100 gpm) require several units in series. No known drinking water applications for MTBE removal. Smaller air/water contact area than low profile systems; less efficient for chemicals with a low Henry s constant such as MTBE. Significant performance drop off with scale formation on the diffuser, as described for packed towers. High gas-pressure drop requirements across the diffuser requiring a high-horsepower blower. Not weatherproof and, therefore, may require an enclosure for protection. 39

62 Key Variables/Design Parameters The key variables and design parameters required to design a diffused bubble aeration system include the anticipated water flow rate, influent contaminant concentrations, air and water temperature, required air/water ratio, water quality, and target effluent concentration. Removal of an organic compound by a diffused bubble aeration system is described by the following equation (Montgomery, 1985): where Q G is the gas (air) flow rate, H is the compound s Henry s constant, C L and C Lo are the effluent and influent liquid concentrations of the organic compound, K L a is the overall mass transfer constant, V L is the total liquid volume, Q L is the liquid flow rate, and Q G /Q L is the air/water ratio. According to this equation, as the air/water ratio increases, C L /C Lo decreases and the removal efficiency increases, as expected. The relationship shows that for MTBE, bubble aeration (assuming H = 0.03) requires either very high air/water ratios (>300 vol/vol) to achieve high removal efficiencies (>90 percent) or multiple units operated in series. Thus, this technology is not likely to be useful for MTBE removal if packed towers or low profile air strippers are available. System Installations and Manufacturers Diffused bubble aeration systems are manufactured by Lowry Engineering, Inc. of Unity, ME (The Stripper ), Aeromix Systems, Inc. of Minneapolis, MN (BREEZE ), and Carbtrol Corporation of Westport, CT. Major vendors for bubble aeration stripping systems are listed in Appendix 2C. In general, a simple bubble aeration unit may be effective for 90 percent MTBE removal at low flow rates (<60 gpm) and high air/water ratios (>300). For higher flow rates, multiple units are required, thereby increasing capital costs. According to vendors, there are a few diffused bubble aeration systems installed in the United States for MTBE removal; however, none are in drinking water applications. Technical Implementability CL CLO = 1 1+ Q H 1-exp K G [ LaV L QL HQ L ( )] On the basis of effectiveness and implementability, this technology is similar to low profile air stripping systems. The skid-mounted configuration allows the units to be placed on a concrete pad or level floor surface by forklift, with mechanical connections limited to blower inlet, stack gas discharge, and process water influent/discharge lines. As with low profile units, pre-packaged power and control panels are available to operate pumps and blowers. Maintenance includes routine diffuser cleaning and typical pump and blower maintenance. Cleaning to remove scale build-up a maintenance requirement common to nearly all air strippers involves disconnecting the diffusers followed by acid cleaning/ soaking or pressure washing. 40

63 Spray Towers System Description In a spray tower aeration system, contaminated water is passed through one or more nozzles and sprayed into a collection basin or tank. Spray aeration systems are typically used for degassing applications, although they have also been used to remove VOCs from water. In general, there are three types of spray towers: cocurrent, countercurrent, and cyclone systems. Cocurrent. In a cocurrent spray tower, the water-feeding nozzle sprays water in the same direction as the air flow. This configuration is typically less efficient than the other types of spray tower and, consequently, is rarely used. Cocurrent towers will not be discussed further. Countercurrent. In countercurrent spray towers, there are water-feeding nozzles at the top of the tower and a collection sump at the bottom. Air enters the bottom and is blown in an upward direction. The gas exits through a demister at the top. Cyclone. Cyclone spray towers have a tangential air inlet on the side along the unit s base. Air travels in a spiraling motion up the column and exits at the top. Water, sprayed uniformly across the column from a manifold in the upper half of the column, collects at the bottom and exits out a drain. Advantages/Disadvantages See Table 2-1 for a summary of the advantages and disadvantages of spray towers. Major advantages of the spray tower system include: Low pressure drop required for air loading resulting in a smaller blower and less power used compared to packed towers or low profile aerators. This could result in significant cost savings for off-gas treatment. However, if the spray tower is filled with packing, as is required for MTBE and other low volatility VOCs, the air flow rate and pressure drop is the same as for a packed tower. Simple operation, since there are few moving parts and few parameters to monitor. Short setup time for smaller pre-fabricated units, although larger custom systems require extensive transport and installation efforts. Easily adaptable tank and equipment designs for varying flow rates. 41

64 The disadvantages of spray tower systems include: High-pressure drop required for liquid loading due to the need to spray a fine mist of water to achieve the necessary air/water ratio for contaminant stripping, compared to packed tower or low profile aerators. Dramatic removal efficiency drop at higher flow rates. Packing is required to achieve >90 percent removal efficiency for MTBE and other low volatility compounds, which makes this essentially a packed tower. Possibility of nozzle fouling, resulting in increased pressure drop, and internal packing fouling, as discussed for packed tower aerators. Low liquid turndown ratio (unless nozzles are changed). No known MTBE drinking water applications. Key Variables/Design Parameters The mass transfer rate in a spray tower system is a function of the influent water droplet size, turbulence in the column, and distribution of influent water droplets in the tower. Droplet Size. As the droplet size decreases, the surface area to volume ratio for a given volume of water increases, resulting in greater mass transfer efficiency. Mass transfer is proportional to surface area for a given volume of water. The droplet size is a function of the nozzle design and water pressure. Turbulence. The column will operate most efficiently at maximum nozzle flow (i.e., highest turbulence). Because turndown of the liquid flow is not recommended, the best way to adjust the system for different influent water flow rates is to change the size of the feed nozzles. This will allow the operator to maintain a maximum nozzle flow rate and, thus, maximize turbulence inside the column. Distribution. The spray must be distributed evenly throughout the tower, with a minimum of spray striking the walls. The nozzle must also maintain proper distribution of droplet sizes at the maximum flow rate. The maximum total liquid loading should be in the range of 1 to 3 gpm/ft 2. The maximum gas flow rate is approximately 800 lb/hr ft 2 and is limited by liquid droplet entrainment (i.e., flooding) (Fleming, 1989). System Installations and Manufacturers Spray towers have been used to remove low concentrations of MTBE from water in only a few remediation cases and with limited success. Generally, packing material must be added 42

65 to the column to achieve good (90 to 99 percent) removal efficiencies. If packing is added, spray towers are essentially packed towers and, therefore, have the same fouling challenges as packed towers. Manufacturers of spray towers are listed in Appendix 2C. While there are currently no known drinking water applications of spray towers for MTBE removal, this technology has been used to remove other VOCs in drinking water situations. Technical Implementability The countercurrent spray tower is the most common design selected for spray towers because it allows for more than one stage and, thus, greater removal efficiencies than the cyclone tower, which is limited to only one stage. Furthermore, a countercurrent spray tower requires less maintenance and a smaller footprint than a cyclone spray tower for the same flow rate. Like the other air stripper technologies, spray towers require supply pumps, air blowers, and a single influent and discharge pipe. If there is no packing, the only piece of equipment that is susceptible to scaling is the spray nozzle, which can be easily replaced. Aspiration System Description Aspiration or centrifugal stripping involves injection of the contaminated water into a cocurrent, tangential-flow aspirator. Untreated and/or recirculated water is pumped into a collar and then through multiple orifices into the throat of the aspirator. As the water passes through the orifices, the orifices act like turbulent jets, which create a large water surface area and enhance the rate of mass transfer of the VOCs from the water to the air. The configuration of the collar and the type and number of orifices in the aspirator are designed to create a low air/water ratio, which ranges from 5:1 to 30:1 for each water pass. For high removal efficiencies, the treated water must be recirculated many times, thus creating a treatment system with an overall air/water ratio greater than 100:1 (Dempsey and Ackerman, 1989). Figure 2-4 shows the various components of an aspiration system manufactured by Hazleton Environmental. Advantages/Disadvantages See Table 2-1 for a summary of the advantages and disadvantages of aspiration. Additional advantages of the aspirator stripper include: Scaling does not occur, or occurs only to a limited extent due to the high turbulence within the unit. 43

66 The cocurrent air flow rate is induced by the turbulent water jets; this creates a low volume of off-gas requiring treatment. Misting, freezing, and decreased atmospheric temperature do not generally impact aspirator strippers. Aspirator strippers are easily installed and constructed with self-contained modular quickconnect units. They are also relatively non-intrusive with low visual and noise impact, unless high removal efficiencies are required. Disadvantages of the aspirator stripper include: Relatively high operating costs due to high energy demands from the high water pressure drop in aspirators (similar to spray towers) compared to low profile aeration systems. Limited maximum removal efficiency with one system at any flow rate; higher flow rates and higher removal efficiencies require multiple systems in parallel with significant water recirculation and high costs. Large footprint for high removal efficiency. Primary Air Inlet Deflector Plate Water Inlet 1 Deflector Feed Nozzles Secondary Air Inlet Diverging Nozzles Stripping Chamber Water Inlet 2 Containment Nozzles Coalescing Chamber Discharge Figure 2-4. Components of an aspiration system (Maxi-Strip ) manufactured by Hazleton Environmental (1998). 44

67 Key Variables/Design Parameters The key variables and design parameters required to design an aspirator system include influent liquid flow rate, influent contaminant concentrations, water temperature, water quality, and the target final effluent concentration. As with spray towers, the rate of mass transfer is a function of the droplet size, turbulence, and liquid distribution. The jets minimize the droplet size and maximize air/water turbulence while the collar distributes the water evenly across the diameter of the aspirator. A single aspirator stripper can support up to a flow of 500 gpm; larger flows will require several aspirators operating in parallel (Hazleton Environmental, 2000). The number of modules is determined by the flow through the system and the removal efficiency required. A single aspirator can only achieve approximately 18 percent MTBE removal efficiency for a single pass; therefore, water must be recirculated to achieve higher removal efficiencies or units must be designed in series. This will increase the number of aspirators and the costs required to meet treatment objectives. System Installations and Manufacturers There are over 100 aspirator systems in place to remove VOCs (Hazleton Environmental, 1998). Although some of these systems are being used for drinking water applications in the United States, there are no known installations specifically designed for MTBE removal. A limited number of manufacturers produce aspirator systems (see Appendix 2C). Of these, the Maxi-Strip hydraulic venturi stripper manufactured by Hazleton Environmental (Hazelton, PA) and distributed by Onion Enterprises (Walnut Creek, CA) appears to be the most well-established. Technical Implementability The Maxi-Strip system can be trailer mounted for easy transport to and from multiple locations. Minimal site preparation is required for its installation. A quick connect system is available, requiring only influent supply and discharge piping. Multiple units can be easily installed or removed as influent concentrations and effluent criteria change. The only moving parts are centrifugal pumps that require no special equipment or training. The cocurrent air flow design reduces off-gas volume and associated air treatment equipment. For ease of operation, system controls and logic are also available. 45

68 46

69 2.3 Comparative Discussion of Air Strippers Permitting In general, all of the air stripper systems described in Section 2.2 require state and municipality-specific permits for construction, air emissions, and process water discharges (see Table 2-3a). Construction permits may not be necessary for smaller low-profile installations (i.e., low profile, bubble diffusion, and aspirator units) if they can be installed in an existing enclosed treatment facility. Air emissions permits are generally required for any discharge of organic compounds to the atmosphere (see Table 2-3b). The air emissions standards will vary from region to region, but for purposes of comparison, this chapter has used the South Coast Air Quality Management District (South Coast AQMD) standard of 1 lb. VOC/day as the maximum allowable air emission rate. Due to this stringent requirement, ease of permitting will be defined by the volume of the off-gas stream and the concentration of VOCs in that gas stream for the various air stripping technologies. Process water discharges from any of the air stripping systems described above will be subject to nearly identical discharge permitting issues Flow Rate Flow rate is a governing factor in determining the removal efficiency from a given air stripper design. A summary of typical maximum hydraulic capacities for each of the technologies described in Section 2.2 is presented in Table 2-4. Design of larger systems may be feasible if a custom-designed unit is constructed. Low profile, bubble diffusers, and aspirators are typically bought as pre-designed units, whereas spray towers and packed towers are usually custom designed for the needed application. As indicated, packed towers are generally able to accept the highest flows before multiple, parallel units are required. 47

70 Technology Construction Requirements Operational and Maintenance Requirements Air Stripping Technology Packed Tower Free standing or guy wired. May require pretreatment to reduce fouling potential. Requires concrete pad. Cleaning of packing likely necessary for fouling. May be installed outdoors. Routine blower and pump maintenance required. Spray Tower Free standing or guy wired. May require pretreatment to reduce fouling. 48 Bubble Aeration Low Profile May require concrete pad. May be installed outdoors Typically requires process enclosure. Needs self-supporting stack. Typically requires process enclosure. Requires level surface. Larger units need self-supporting stack. Cleaning of packing (for MTBE) and nozzles may be required if fouling occurs. Routine blower and pump maintenance required. May require pretreatment to reduce fouling potential. Cleaning of bubble diffusers and tank may be required if fouling occurs. Routine blower and pump maintenance required. Needs enclosed heating if ambient temperature below freezing. May require pretreatment to reduce fouling potential. Cleaning of trays is required if fouling occurs. Routine blower and pump maintenance required. Needs enclosed heating if ambient temperature below freezing. Table 2-3a Air Stripping Permitting Requirements Aspiration May be installed in an enclosure or outdoors Routine pump maintenance required. May require concrete pad. 3 phase, 230/460 volt, 60 hertz, power typically needed. Larger units need self-supporting stack. Emissions limit of organic compounds to the atmosphere (range: Height of stack above ground level 1 to 10 lb/day). Interference from other buildings MTBE treated as an organic. Monitoring requirements: for new operation, daily monitoring; for operation up to two years, monthly monitoring.

71 Technology Construction Requirements Operational and Maintenance Requirements 49 Off Gas Treatment Technologies Vapor Phase GAC Minimum distance to outer boundary of a school: 1000 feet. Adsorption Minimum system design: two beds to provide standby device in case of carbon breakthrough. Thermal Oxidation Minimum distance to outer boundary of a school: 1000 feet. Safety device required for system shutdown during periods of low temperature. Catalytic Oxidation Minimum distance to outer boundary of a school: 1000 feet. Safety device required for system shutdown during periods of low temperature. Biofilter Monitoring requirements: for new operation, daily monitoring; for operation up to two years, monthly monitoring. Temperature requirement: 1500 F at exit point of combustion chamber. Monitor VOCs at outlet. Temperature requirement: 550 F at inlet. Degradation: required to demonstrate. By-products monitored Table 2-3b Off-gas Treatment Permitting Requirements

72 Table 2-4 Typical Maximum Hydraulic Capacities for Commercially Available Systems* Typical One-Pass MTBE Removal Efficiency (%) (w/max. AWR) Technology Max. Hydraulic Capacity for MTBE Removal Max. Hydraulic Capacity for Other VOCs Packed Tower 1,000 gpm Custom designs Low Profile 1,100 gpm gpm Bubble Diffusion 50 gpm gpm Spray Tower All custom designs All custom designs Aspiration 500 gpm gpm *All information supplied by vendors; information is for a single unit system with maximum air to water ratio (AWR) for the given system Removal Efficiency and Flow Rate All of the air stripping technologies are capable of meeting stringent removal efficiency requirements (>99 percent). High removal efficiencies are dependent on the process design of the system and the number of air stripping units included in the system. For a single unit, however, each of the technologies is limited to a maximum removal efficiency that can be practically achieved (see Table 2-4). For a single packed tower, the maximum design efficiency is usually less than or equal to 99 percent. Removals greater than this are difficult to maintain because of possible non-ideal flow through the tower, resulting in short-circuiting. The advantage of packed towers is that a single tower can achieve high removal efficiencies at very high flow rates whereas the other systems require several treatment units in series or are not practical for high flow rates. Low profile aeration systems are also capable of achieving removal efficiencies as high as 99 percent, but typically only for lower flow rates (<100 gpm). Simple unit bubble diffusion aerators and spray towers without packing are capable of achieving only 80 to 90 percent removal because of mass transfer limitations inherent to the technology. Both of these systems will likely not be selected for removal of MTBE because high removal efficiencies would require several units or towers in series, which make these systems not economically feasible. A spray tower can achieve higher removal efficiencies only if packing is used, which essentially makes the spray tower into a packed tower. A single aspirator unit can only achieve 60 to 70 percent removal efficiency; however, these units are designed to allow for significant water recirculation to achieve higher removal 50

73 efficiencies. To treat higher flow rates, aspirators must be placed in parallel. Recirculating the effluent water and placing multiple aspirators in parallel can result in a large footprint and high capital, operation, and maintenance costs. For this reason, aspirators are only practical for low flow rates (<60 gpm) Impact of Water Quality Water quality has a significant impact on all air stripping systems due to biofouling or scaling from iron precipitation and magnesium and calcium carbonate precipitation. Biofouling can be mitigated by pretreatment with a disinfectant. The overall impact of scaling on treatment can be mitigated by cleaning or scale removal, or by adding chelating agents to the air stripper influent. Packed tower and low profile aeration systems are generally the most sensitive to scale formation, followed by bubble aeration and spray towers, which exhibit lower scale formation due to a higher degree of turbulence. Aspirator systems are the least sensitive to fouling due to the high level of turbulence, which does not allow scaling to form. Table 2-5 below summarizes the range of iron and hardness levels that could lead to scaling. The low profile aeration system may be more susceptible to fouling than a packed tower because the air ports used to distribute air are small. However, the low profile stripper is more accessible for cleaning. Other water quality parameters that impact the operations of an air stripping system include manganese and chloride concentrations, ph, alkalinity, temperature, and oil and grease. Table 2-5 Problematic Iron and Hardness Concentrations for Air Stripping Technologies Technology Iron Concentration (mg/l) Hardness (mg/l) Packed Tower Low Profile Bubble Aeration ,000 Spray Tower Dependent on Configuration Dependent on Configuration Aspiration *All information obtained from vendors Other Factors 1,200 2,000 A comparative discussion of each of the air stripper technologies relative to effectiveness and implementability issues is presented below. Specifically, each of the technologies is discussed relative to its reliability, flexibility, adaptability, and potential for modification. Table 2-6 presents a comparative summary of the technologies with respect to each of these criteria. Construction, operation, and maintenance issues are summarized in Table 2-3a. 51

74 Reliability Reliability includes both process and mechanical reliability of the technology to meet treated water requirements consistently. In general, all of the technologies discussed above are mechanically reliable. There are few moving parts or mechanical equipment in any of the systems, limiting the need for change-outs or replacements. Based on this criterion alone, all of the systems would receive a MEDIUM or better rating for mechanical reliability. Process reliability for MTBE removal from water in a single unit varies for each technology (see Table 2-4). Combining mechanical and process reliability, bubble diffusion aeration and spray towers were given a LOW rating because a single unit cannot achieve removal efficiencies greater than 90 percent (i.e., these technologies are not reliable for MTBE removal). For the remaining technologies, their ability to treat MTBE is generally a function of flow rate. Packed tower air strippers can achieve greater than 95 percent MTBE removal at higher flows (i.e., 600 to 6,000 gpm), whereas low profile and aspiration air strippers are capable of high removal efficiencies for MTBE only at low flows (<100 gpm). To achieve the same degree of reliability for each air stripping system, multiple units may be required. Flexibility Flexibility is defined as the ability of the technology to handle a wide range of flow rates. The ability to handle a wide range of flows is not necessarily tied to overall hydraulic capacity but, rather, the ability of the unit to function as desired if flows drop significantly below, or increase significantly above, process design values. With the exception of the aspirator stripper and spray tower, the other air stripping technologies are able to handle varying liquid flow rates without requiring significant design changes (i.e., these technologies have a high liquid turndown ratio). Due to the reliance of the aspirator and spray tower strippers on high pressure liquid streams to effect mass transfer, these units are not wellsuited to handle changing flow rates. Adaptability Adaptability is defined as the ability of a technology to handle fluctuating contaminant influent concentrations and other water quality parameters of interest, such as scale formation and fouling. All of the technologies evaluated have specific removal efficiencies under given influent flows that are relatively independent of the influent concentration. If influent concentrations increase and effluent concentration goals remain unchanged, it may be necessary to increase the air/water ratio or recirculate the water to meet treated water goals. 52

75 Reliability Flexibility Adaptability Potential for Modifications Definition Proven technology for MTBE. Low mechanical failure potential. Able to handle a wide range of flows. Able to operate in batch or continuous mode. Able to handle changing influent concentrations. Able to handle changing water quality. Easily cleaned. Readily supplemented with additional or larger components (blower, tank, etc.) if influent conditions change. Readily combined with pre or post-treatment equipment. Can be turned down. Air Stripping Technology 53 Packed Tower HIGH Oldest design (widely used) % removal of MTBE at wide flow range. Low mechanical failure potential. Towers are weatherproof. Orific plate distribution system ensures even distribution of both water and air over entire cross-section; this results in maximum AWR interface and yields consistent results. Minimum instrumentation. HIGH Orifice plate distribution system allows a single unit to handle wide range of flows without affecting performance. Not well-suited to batch flow. Towers can be operated as standalone units or in parallel or series. Use of variable frequency drives on blower motor maintains constant AWR over a wide range of water flowrates, resulting in significant energy savings at lower water flows. HIGH/MEDIUM Can handle a wide range of daily or seasonal fluctuations in contaminant loading. Subject to scaling if hardness/ metals levels increase. Access to packing for cleaning is difficult and limited. HIGH/MEDIUM May be modified with larger blower. 30 to 40% turndown rate for air (can only be throttled back 60-70% of capacity before efficiency decreased. Height and weight considerations will limit amount of additional packing. Difficult to relocate tower. Diameter cannot be increased to handle more than maximum design flowrate. Table 2-6 Comparison of Air Stripping Technologies Low Profile HIGH/MEDIUM Proven technology for MTBE at low-medium flows. Few moving parts means minimal chance for mechanical failure. Minimal instrumentation. Not weatherproof, thus requiring construction of protective shelter. HIGH Able to handle wide range of flows. Can operate in continuous or batch mode. MEDIUM Able to handle daily or seasonal fluctuations in contaminant loading at low-medium flows. Subject to scaling if hardness/ metals levels increase. Readily cleaned. MEDIUM Flow/loadings above design conditions typically requires new unit. Blower cannot be turned down. Easily relocated due to size. Size of unit cannot be modified to handle water flow rates in excess of maximum design flow rate.

76 Reliability Flexibility Adaptability Potential for Modifications 54 Bubble Aeration Spray Tower Aspiration LOW Technology is proven to be effective in certain applications. Minimal effectiveness for MTBE. Low mechanical failure potential. Minimal instrumentation. Not weatherproof, thus requiring construction of protective shelter. LOW Technology is proven to be effective in certain applications. Low removal efficiency for MTBE unless packing added. Low mechanical failure potential. MEDIUM Successful Superfund applications. No specific MTBE applications identified; however, computer modeling has been developed from extensive testing to size an application and predict the performance of the operating system for MTBE removal. Good removals predicted at low to high flows. Only moving parts are centrifugal pumps (no blower) which are simple to maintain. HIGH Able to handle wide range of flows. Can operate in continuous or batch mode. LOW Water flow rate should be maintained constant for effective treatment. Not well-suited to batch flow if packing used. MEDIUM Not able to handle wide range of flows. Capable of batch and continuous operation without modification. MEDIUM Able to handle wide range of contaminant loadings at low flows. Resistant to scaling due to hardness/dissolved metals. Readily cleaned. MEDIUM Able to handle wide range of contaminant loadings at low flows. Subject to scaling if hardness/ metals levels increase. Easily maintained unless packing is required. HIGH Able to handle wide range of contaminant loadings at lowmedium flows. Resistant to scaling due to high degree of turbulence. Readily cleaned. MEDIUM Flow loadings above design conditions require new or additional unit. Blower turn down is feasible. Easily relocated due to size. Size of unit cannot be modified to handle water flow rates in excess of maximum design flow rate. MEDIUM Readily tied into other larger components. Low turn down ratio (nozzle must be changed to adjust flow rates). HIGH Additional modules can be added for difficult applications. Modules can be turned off if influent concentrations decrease. Table 2-6 (Continued) Comparison of Air Stripping Technologies

77 The efficiency of air strippers is not affected by other VOCs (including BTEX) in the waste stream due to the relatively high air/water ratio required for removal of MTBE. However, increased ph, iron, calcium, or magnesium in the influent stream could lead to increased scale formation. Ease of cleaning in the event that scaling occurs is, therefore, a related adaptability criterion. Packed columns and spray aeration systems fitted with packing (as is typically necessary for MTBE removal) are most difficult to clean due to the general inaccessibility of the packing material. In addition, the packing material provides a medium for scale and iron deposits. Alternatively, low profile systems have significant contact between the aerated process water and the tray surface, resulting in more rapid scale formation; however, access to trays and cleaning is easier for low profile systems. Aspirator and bubble aeration technologies do not generally rely on media surfaces to create mass transfer in the water column. Therefore, fouling is limited to the nozzles in these systems, which can be easily replaced. Potential for Modification The potential to implement equipment modifications is defined as the capability to change equipment due to a change in design conditions (e.g., metals precipitation pre-treatment equipment, activated carbon post-treatment). Turndown of system air flows to minimize energy costs if influent concentrations decrease is another example of system modifications. All the technologies considered have equal capability of being combined with pre- or posttreatment equipment, as the systems can be built or modified with influent feed tanks or discharge tanks. Similarly, each technology has difficulty handling flows significantly above maximum hydraulic design capacity, although the modular layout of the low profile and aspirator system may facilitate supplementing the process with additional units if higher flows are required. If off-gas treatment is required, it is advantageous to have an air flow rate as low as possible. Assuming a variable speed blower, the ability to turn down air flow rate is feasible for packed towers, bubble aeration systems, and spray towers. However, unpacked spray towers do not rely heavily on counter-current air flow to effect removal efficiency, thus resulting in low energy savings potential if influent concentrations decrease. Low profile units have almost no air turndown capability because the water must be kept from falling through tray perforations. Aspirator units do not rely on any forced air source to effect volatilization and, therefore, are not subject to this criterion By-products Aside from scale formation, by-products are not generally a concern with air stripping technologies. However, there is a potential for biological fouling that may produce biological solids in the treated water. Disinfection of the air stripper influent or effluent water could be implemented to prevent biological fouling. In addition to biological fouling, out-gassing of carbon dioxide will cause a rise in the ph, which may necessitate implementation of miti- 55

78 gation measures. In summary, none of the air stripping technologies demonstrates any particular advantage or disadvantage for minimizing by-product formation, although some of the air stripping technologies are more amenable to by-product removal (e.g., low profile aerators for scaling and fouling removal) Cost Effectiveness To compare capital costs for each of the air stripping systems evaluated, suppliers were provided with a number of treatment scenarios and asked to provide capital costs and motor horsepower requirements (as part of O&M costs) to meet the treatment requirements. At least one major manufacturer of each air stripping technology evaluated in Section 2.2 was contacted and asked to provide model selection, number of units required in parallel or series, and capital costs for the following potential MTBE treatment scenarios: Influent flows of 60, 600, and 6,000 gpm. Influent MTBE concentrations of 20, 200, and 2,000 µg/l. Effluent MTBE discharge requirements of 20, 5, or 0.5 µg/l. Table 2-7 provides a sample calculation of total capital costs, total annual costs, and unit treatment costs. As indicated on Table 2-8, capital costs for spray towers, packed towers, and low-profile systems are similar at low flow rates (60 gpm), with bubble diffusion and aspiration systems requiring significantly higher capital costs at higher influent concentrations due to the need for multiple units to achieve equivalent removal efficiency. At higher flows and removal rates, packed columns and spray towers become significantly more cost-effective than the low profile technologies, as predicted, while bubble diffusion and aspiration systems remain significantly higher in capital costs. To estimate operating costs for each of the systems, manufacturer-recommended horsepower sizing for the stripping system blowers and pumps were requested for each of the three influent flow rates and a required MTBE removal efficiency of 97.5 percent that corresponded to both the 20 µg/l to 0.5 µg/l and 200 µg/l to 5 µg/l influent/effluent scenarios. For the purpose of comparison, centrifugal feed and discharge pumps were not considered because this equipment is common to all of the systems. Supplemental feed pumps capable of delivering the required flows at high pressure drops were, however, considered for the spray tower and aspiration systems as this equipment is integral to their stripping operation. Table 2-9 presents a summary of annual O&M costs for each of the air stripping units under the scenario described above. These cost estimates include electrical costs based on an average electric rate of $0.08/kWhr, maintenance and sampling labor costs at $80/hr, and analytical costs at $200 per sample. The estimated sampling and maintenance labor hours are 56

79 presented in Table The sampling requirements used to estimate analytical costs are presented in Table It is important to note that the frequency of sampling will likely be much higher for the first few years of operation while the system receives regulatory approval. As indicated in Table 2-9, spray towers, packed towers, and low profile systems have competitive annual O&M costs at the low flow rate of 60 gpm while bubble diffusion and aspiration systems require considerably higher O&M costs (approximately three times higher) due to high electrical, labor, and analytical costs. At medium flow rates (600 gpm) and high flow rates (6,000 gpm), spray towers and low profile systems diverge considerably in costs from packed towers due to higher electrical costs for the spray towers and higher electrical, labor, and analytical costs for the low profile systems. In general, O&M costs increase with the number of units required for a given flow rate since more units require more power, maintenance, and sampling. Table 2-7 Sample Calculation of Capital, Annual, and Unit Treatment Costs for a 600 gpm Packed Tower, 2,000 to 20 µg/l MTBE Line Item 57 Cost Treatment Unit $125,000 Piping, Valves, Electrical (30%) $37,500 Site Work (10%) $12,500 SUBTOTAL $175,000 Contractor O&P (15%) $26,250 SUBTOTAL $201,250 Engineering (15%) $30,188 SUBTOTAL $231,438 Contingency (20%) $46,288 TOTAL CAPITAL $277,725 Amortized Annual Capital $22,381 Annual O&M $91,684 TOTAL ANNUAL COST $114,065 Annual Flow Treated (kgal) 315,000 UNIT TREATMENT COST ($/kgal)* $0.36 *To convert unit treatment cost to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr 600 gpm: ~12 hrs/week at $80/hr = $50,000/yr 6000 gpm: ~31 hrs/week at $80/hr = $130,000/yr 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for a packed tower packed tower.

80 Table 2-8 Initial Capital Expenses for Air Stripping Systems Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Spray Tower Packed Tower Low Profile Bubble Aeration Aspiration % $55,545 $66,654 ND $69,827 $51, % $66,654 $88,872 $95,537 $174,567 $127, % $55,545 $66,654 $45,880 $104,740 $77, % $66,654 $88,872 $71,149 $174,567 $127, % $88,872 $111,090 $88,845 $244,394 $204, % $77,763 $99,981 $52,443 $209,480 $215, % $88,872 $111,090 $58,955 $244,394 $204, % ND ND ND ND $333, % $177,744 $222,180 $259,871 $628,441 $355, % $211,071 $288,834 $519,741 ND $944, % $177,744 $233,289 $337,543 $1,152,141 $555, % $211,071 $288,834 $675,085 ND $944, % $333,270 $299,943 $776,172 ND ND % $266,616 $277,725 $675,085 ND ND % $333,270 $299,943 $776,172 ND ND % ND ND ND ND ND % $1,555,260 $1,999,620 $2,598,706 ND ND % $1,799,658 $2,221,800 $5,197,412 ND ND % $1,666,350 $2,021,838 $3,375,425 ND ND % $1,799,658 $2,221,800 $5,197,412 ND ND % $2,288,454 $2,788,359 $7,761,725 ND ND % $1,999,620 $2,532,852 ND ND ND % $2,288,454 $2,788,359 ND ND ND ND = no data available, system may require custom design Capital Expenses include: Equipment: Piping, valves, electrical (30%); site work (10%); contractor O&P (15%); engineering (15%); contingency (20%). Table 2-9 Annual O&M Costs for Air Stripping Systems Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Spray Tower Packed Tower Low Profile Bubble Aeration Aspiration % $48,933 $46,844 ND $151,557 $176, % $48,933 $47,888 $51,021 $151,557 $176, % $48,933 $46,844 $49,977 $151,557 $176, % $48,933 $47,888 $51,021 $151,557 $176, % $48,933 $48,410 $52,587 $151,557 $176, % $48,933 $48,410 $55,720 $151,557 $176, % $48,933 $48,410 $58,852 $151,557 $176, % ND $51,021 ND ND $176, % $102,126 $77,587 $226,294 $1,121,972 $370, % $102,126 $83,852 $249,789 ND $370, % $102,126 $81,242 $241,957 $1,121,972 $370, % $102,126 $83,852 $249,789 ND $370, % $102,126 $91,684 $281,114 ND ND % $102,126 $91,684 $249,789 ND ND % $102,126 $91,684 $281,114 ND ND % ND ND ND ND ND % $489,953 $257,620 $857,343 ND ND % $489,953 $312,440 $1,092,286 ND ND % $489,953 $296,777 $1,013,972 ND ND % $489,953 $312,440 $1,092,286 ND ND % $489,953 $312,440 $1,405,544 ND ND % $489,953 $328,103 ND ND ND % $489,953 $328,103 ND ND ND ND = no data available, system may require custom design O&M Costs include: 1. Power costs at $0.08/kWhr. 2. Labor costs estimated at $80/hr; see Tables 2-13 to 2-17 for a breakdown of estimated annual labor hours required for maintenance and sampling. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 58

81 Table 2-10 Estimated Labor Hours for Maintenance and Sampling of Air Stripping Systems* in Hours per Week Air Stripper 60 gpm 600 gpm 6,000 gpm Packed Tower Low Profile Bubble Diffusion N.E. Spray Tower Aspiration N.E. *Maintenance & sampling labor hours associated with selected off-gas treatment system *Maintenance & sampling labor hours associated with selected off-gas are included in these estimates. treatment system are included in these estimates. Notes: N.E. = not evaluated 1. N.E. = not evaluated. 2. Labor hours are dependent upon the number of units required. Table 2-11 Sampling Requirements for Air Stripping Systems and Off-gas Treatment: No. of Samples Collected Weekly Treatment System 60 gpm 600 gpm 6,000 gpm Air Stripper Packed Tower 2 water samples 2 7 Low Profile Bubble Diffusion 6 51 N.E. Spray Tower Aspiration 7 12 N.E. Off-Gas Treatment GAC 1 gas samples 1 7 Recuperative Thermal Oxidation Recuperative Flameless Thermal Oxidation Recuperative Catalytic Oxidation Non-Recuperative Catalytic Oxidation N.E. 1 N.E. Notes: Notes: For the the cost cost estimates, estimates, both both water water and and gas sample gas sample analyses analyses were priced were at priced approximately at $200 per approximately sample. $200 per sample. 2. N.E. = not evaluated Sampling N.E. = not requirements evaluated. are dependent upon the number of units required. 59

82 Amortized annual capital costs and annual O&M costs were combined to determine the total amortized operating cost for each system per 1,000 gallons of treated water (see Tables 2-12 through 2-17). The equipment was amortized at a discount rate of seven percent over a 30- year period. At the low flow rate (60 gpm), packed towers, spray towers, and low profile systems are the least expensive options. As the flow rate increases, unit costs drop at different rates with spray towers and packed towers emerging as the least expensive options. Although spray towers are cost competitive with packed towers and low-profile systems for the 60 gpm scenario, they have not been considered because, when they are filled with packing (as is required for MTBE removal), they become nearly identical to packed towers. Therefore, based on this cost comparison, packed towers represent the least expensive option for all drinking water applications. However, low profile strippers are competitive with packed towers at low flow rates (60 gpm) and are expected to be used preferentially to packed towers due to their ease of use and implementability. The results of this cost comparison indicate that bubble diffusers and aspiration strippers are the most expensive air stripping options and are, thus, not likely to be optimal for drinking water applications where MTBE removal is required. A sensitivity analysis was performed for the costs of the two most promising technologies, packed tower systems (Table 2-18) and low profile systems (Table 2-19), at a design flow rate of 600 gpm, an influent concentration of 200 µg/l MTBE, and an effluent concentration of 5 µg/l MTBE. For this analysis, we assumed that influent water with high potential for fouling (i.e., high natural organic material [NOM] concentrations) should be pre-treated with sodium hypochlorite or a similar disinfectant (approximately 5 mg/l free chlorine) prior to being fed into the air strippers. The additional costs associated with disinfection are the capital cost for a chemical feed system and annual chemical consumption. High potential for fouling results in a cost increase of approximately $0.13/1,000 gallons of treated water. As a result of the high air/water ratio required for MTBE removal, BTEX loads of up to 200 µg/l for each compound are expected not to affect the performance of either the packed tower or low profile systems. BTEX compounds have higher Henry s constants than MTBE and, therefore, are more easily removed by air stripping. Thus, capital, annual O&M, and unit costs are unaffected by BTEX loadings that are similar or lower than MTBE loadings on a mass basis. As can be expected, shortening the design life of these systems is expected to result in higher unit costs. Reducing the design life from 30 years to 2 years, while maintaining a seven percent discount rate, results in an approximate doubling of the unit costs for the packed tower system ($0.34 to $0.77/1,000 gallons) and the low profile system ($0.96 to $1.98/1,000 gallons). 60

83 61 Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal (%) Spray Tower Packed Tower Low Profile Bubble Aeration Aspiration % $1.69 $1.66 ND $4.98 $ % $1.72 $1.75 $1.86 $5.25 $ % $1.69 $1.66 $1.70 $5.07 $ % $1.72 $1.75 $1.80 $5.25 $ % $1.78 $1.82 $1.89 $5.43 $ % $1.75 $1.79 $1.90 $5.34 $ % $1.78 $1.82 $2.02 $5.43 $ % ND ND ND ND $ % $0.37 $0.30 $0.78 $3.72 $ % $0.38 $0.34 $0.92 ND $ % $0.37 $0.32 $0.85 $3.85 $ % $0.38 $0.34 $0.96 ND $ % $0.41 $0.37 $1.09 ND ND % $0.39 $0.36 $0.96 ND ND % $0.41 $0.37 $1.09 ND ND % ND ND ND ND ND % $0.20 $0.13 $0.34 ND ND % $0.20 $0.16 $0.48 ND ND % $0.20 $0.15 $0.41 ND ND % $0.20 $0.16 $0.48 ND ND % $0.21 $0.17 $0.64 ND ND % $0.21 $0.17 ND ND ND % $0.21 $0.18 ND ND ND ND = no data available, system may require custom design. *To convert costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr. 2. Labor costs estimated at $80/hr; see Tables 2-13 to 2-17 for a breakdown of estimated annual labor hours required for maintenance and sampling. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. Table 2-12 Total Amortized Operating Costs ($/1,000 Gallons Treated)* for Air Stripping Systems

84 Table 2-13 Expense Summary for Air Stripping Systems Spray Tower Flow (gpm) System Configuration Influent (µg/l) Effluent (µg/l) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* 60 single tower % $55,545 $48,933 $ single tower % $66,654 $48,933 $ single tower % $55,545 $48,933 $ single tower % $66,654 $48,933 $ single tower % $88,872 $48,933 $ single tower % $77,763 $48,933 $ single tower % $88,872 $48,933 $ ND % ND ND ND 600 single tower % $177,744 $102,126 $ single tower % $211,071 $102,126 $ single tower % $177,744 $102,126 $ single tower % $211,071 $102,126 $ single tower % $333,270 $102,126 $ single tower % $266,616 $102,126 $ single tower % $333,270 $102,126 $ ND % ND ND ND 6000 parallel towers % $1,555,260 $489,953 $ parallel towers % $1,799,658 $489,953 $ parallel towers % $1,666,350 $489,953 $ parallel towers % $1,799,658 $489,953 $ parallel towers % $2,288,454 $489,953 $ parallel towers % $1,999,620 $489,953 $ parallel towers % $2,288,454 $489,953 $0.21 ND = no data available, system may require custom design. *To convert unit treatment costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr. 600 gpm: ~12 hrs/week at $80/hr = $50,000/yr gpm: ~31 hrs/week at $80/hr = $130,000/yr. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 62

85 Table 2-14 Expense Summary for Air Stripping Systems Packed Tower Flow (gpm) System Configuration Influent (µg/l) Effluent (µg/l) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* dia. tower % $66,654 $46,844 $ dia. tower % $88,872 $47,888 $ dia. tower % $66,654 $46,844 $ dia. tower % $88,872 $47,888 $ dia. tower % $111,090 $48,410 $ dia. tower % $99,981 $48,410 $ dia. tower % $111,090 $48,410 $ ND % ND ND ND dia. tower % $222,180 $77,587 $ dia. tower % $288,834 $83,852 $ dia. tower % $233,289 $81,242 $ dia. tower % $288,834 $83,852 $ dia. tower % $299,943 $91,684 $ dia. tower % $277,725 $91,684 $ dia. tower % $299,943 $91,684 $ ND % ND ND ND x 11.5 dia. parallel towers % $1,999,620 $257,620 $ x 11.5 dia. parallel towers % $2,221,800 $312,440 $ x 11.5 dia. parallel towers % $2,021,838 $296,777 $ x 11.5 dia. parallel towers % $2,221,800 $312,440 $ x 11.5 dia. parallel towers % $2,788,359 $312,440 $ x 11.5 dia. parallel towers % $2,532,852 $328,103 $ x 11.5 dia. parallel towers % $2,788,359 $328,103 $0.18 ND = no data available, system may require custom design *To convert unit treatment costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr 600 gpm: ~12 hrs/week at $80/hr = $50,000/yr 6000 gpm: ~31 hrs/week at $80/hr = $130,000/yr 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 63

86 Table 2-15 Expense Summary for Air Stripping Systems Low Profile Flow (gpm) System Configuration Influent (µg/l) Effluent (µg/l) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* 60 Single unit % ND ND ND 60 Single unit % $95,537 $51,021 $ Single unit % $45,880 $49,977 $ Single unit % $71,149 $51,021 $ Single unit % $88,845 $52,587 $ Single unit % $52,443 $55,720 $ Single unit % $58,955 $58,852 $ ND % ND ND ND in parallel % $259,871 $226,294 $ in parallel % $519,741 $249,789 $ in parallel % $337,543 $241,957 $ in parallel % $675,085 $249,789 $ in parallel % $776,172 $281,114 $ in parallel % $675,085 $249,789 $ in parallel % $776,172 $281,114 $ ND % ND ND ND in parallel % $2,598,706 $857,343 $ in parallel % $5,197,412 $1,092,286 $ in parallel % $3,375,425 $1,013,972 $ in parallel % $5,197,412 $1,092,286 $ in parallel % $7,761,725 $1,405,544 $ ND % ND ND ND 6000 ND % ND ND ND ND = no data available, system may require custom design. *To convert unit treatment costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr. 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr. 600 gpm: ~31 hrs/week at $80/hr = $130,000/yr gpm: ~72 hrs/week at $80/hr = $300,000/yr. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 64

87 Table 2-16 Expense Summary for Air Stripping Systems Bubble Aeration Flow (gpm) System Configuration Influent (µg/l) Effluent (µg/l) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* 60 5 in parallel % $69,827 $151,557 $ in parallel % $174,567 $151,557 $ in parallel % $104,740 $151,557 $ in parallel % $174,567 $151,557 $ in parallel % $244,394 $151,557 $ in parallel % $209,480 $151,557 $ in parallel % $244,394 $151,557 $ in parallel % ND ND ND in parallel % $628,441 $1,121,972 $ ND % ND ND ND in parallel % $1,152,141 $1,121,972 $ ND % ND ND ND 600 ND % ND ND ND 600 ND % ND ND ND 600 ND % ND ND ND 600 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND ND = no data available, system may require custom design. *To convert unit treatment costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr. 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr. 600 gpm: ~31 hrs/week at $80/hr = $130,000/yr gpm: ~72 hrs/week at $80/hr = $300,000/yr. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 65

88 Table 2-17 Expense Summary for Air Stripping Systems Aspiration Flow (gpm) System Configuration Influent (µg/l) Effluent (µg/l) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal) 60 6 units in series % $51,101 $176,294 $ units in series % $127,754 $176,294 $ units in series % $77,763 $176,294 $ units in series % $127,754 $176,294 $ units in series % $204,406 $176,294 $ units in series % $215,515 $176,294 $ units in series % $204,406 $176,294 $ units in series % $333,270 $176,294 $ units in series % $355,488 $370,946 $ units in series % $944,265 $370,946 $ units in series % $555,450 $370,946 $ units in series % $944,265 $370,946 $ ND % ND ND ND 600 ND % ND ND ND 600 ND % ND ND ND 600 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND 6000 ND % ND ND ND ND = no data available, system may require custom design. *To convert unit treatment costs to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.08/kWhr. 2. Labor costs: Due to the lack of field data, annual labor costs for maintenance and sampling have been estimated as follows: 60 gpm: ~6 hrs/week at $80/hr = $25,000/yr. 600 gpm: ~12 hrs/week at $80/hr = $50,000/yr gpm: ~31 hrs/week at $80/hr = $130,000/yr. 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 66

89 Table 2-18 Sensitivity Analysis for Air Stripping Systems Packed Tower 600 gpm system, influent = 200 µg/l, effluent = 5 µg/l MTBE Sensitivity Parameter Capital Cost ($) Annual O&M ($) Total Annual Cost ($) Unit Cost ($/1000 gal) NOM Fouling Low Fouling $288,834 $83,852 $107,129 $0.34 Moderate Fouling $288,834 $83,852 $107,129 $0.34 High Fouling 1 $344,379 $118,865 $146,618 $0.46 BTEX Load No BTEX present BTEX at 20 µg/l each BTEX at 200 µg/l each $288,834 $83,852 $107,129 $0.34 Literature review indicates that packed tower performance should be unaffected by BTEX. 2 Literature review indicates that packed tower performance should be unaffected by BTEX. 2 Design Life 3 2 years $288,834 $83,852 $243,604 $ years $288,834 $83,852 $124,976 $ years $288,834 $83,852 $107,129 $ Increased costs include a NaOCl feed system (chemical storage tank, feed pump, controls, piping, etc.) for a dosage of 5 mg/l free chlorine. 12.5% NaOCl solution estimated at $80/30 gallon totes. 2 BTEX compounds are more easily removed through air stripping due to their higher Henry s constants. 3 At a 7% discount rate. 67

90 Table 2-19 Sensitivity Analysis for Air Stripping Systems Low Profile 600 gpm system, influent = 200 µg/l, effluent = 5 µg/l MTBE Sensitivity Parameter Capital Cost ($) Annual O&M ($) Total Annual Cost ($) Unit Cost ($/1000 gal) NOM Fouling Low Fouling Moderate Fouling $675,085 $249,789 $304,191 $0.96 $675,085 $249,789 $304,191 $0.96 High Fouling 1 $730,630 $284,802 $343,681 $1.09 BTEX Load No BTEX present $675,085 $249,789 $304,191 $0.96 BTEX at 20 µg/l each BTEX at 200 µg/l each Literature review indicates that low profile performance should be unaffected by BTEX. 2 Literature review indicates that packed tower performance should be unaffected by BTEX. 2 Design Life 3 2 years $675,085 $249,789 $623,173 $ years $675,085 $249,789 $345,906 $ years $675,085 $249,789 $304,191 $ Increased costs include a NaOCl feed system (chemical storage tank, feed pump, controls, piping, etc.) for a dosage of 5 mg/l free chlorine. 12.5% NaOCl solution estimated at $80/30 gallon totes. 2 BTEX compounds are more easily removed through air stripping due to their higher Henry s constants. 3 At a 7% discount rate. 68

91 2.4 Off-gas Treatment Off-gas treatment of the MTBE-contaminated air stream from an air stripping system is often required prior to discharge to the atmosphere. Treatment alternatives for removing MTBE in the air discharge include: carbon adsorption, thermal and catalytic oxidation, biological treatment, and gas-phase chemical oxidation. This section will evaluate alternative technologies for off-gas treatment for air streams containing low concentrations of MTBE (less than 10 ppmv or parts per million on a volume basis). Selection of off-gas treatment technologies for MTBE is a key factor in cost estimates for the overall air stripping technology evaluation (Figure 2-5). The cost of off-gas treatment is generally proportional to the volume of gas (air) being treated. Thus, the higher the air/water ratio in an air stripping system, the higher the costs. Typical volumetric air/water ratios for MTBE removal range from 100 to 200. This results in large gas volumes with very low concentrations of MTBE. For example, given a medium sized groundwater treatment system (600 gpm) with an influent water concentration of 200 µg/l MTBE, an air stripper operating at 90 percent efficiency emits air containing 0.3 ppmv MTBE. 6,000,000 5,000,000 Total Costs ($/yr) for Off-gas Treatment Total Costs ($/yr) for Off-Gas Treatment of of 5 ppmv MTBE 5 ppmv MTBE GAC Recuperative Thermal Oxidation Recuperative Flameless Thermal Oxidation Recuperative Catalytic Oxidation 4,000,000 $/yr 3,000,000 2,000,000 1,000, ,000 40,000 60,000 80, , , ,000 air flow rate (cfm) Figure 2-5. Cost of off-gas treatment technologies as a function of air flow rate. The regulations that define the maximum emissions limit of MTBE into the atmosphere vary from state to state and district to district. Under the Clean Air Act, MTBE is categorized as a hazardous air pollutant (HAP). In most states, agencies require off-gas treatment in conjunction with an air stripper. In the South Coast AQMD, for example, the emission level 69

92 at which control is required is 1 lb/day of VOC release. Other districts may require different emission standards based on criteria including location of the site and regional ambient air quality (attainment or nonattainment status). In this chapter, the 1 lb/day emissions limit is used for illustrative purposes and it has been assumed that MTBE is the only VOC in the off-gas stream. In Table 2-20 below, mass emissions of MTBE produced by an air stripper are shown as a function of the flow rate and influent concentration. In the case of the highest flow rate scenario (6,000 gpm), off-gas treatment must be employed for groundwater containing 200 µg/l and 2,000 µg/l MTBE with respective required removal efficiencies of 93 and 99.3 percent. In the case of the lowest flow rate scenario (60 gpm), only groundwater systems containing a concentration of 2,000 µg/l MTBE or higher will require off-gas control. A brief description of each off-gas treatment technology, including its advantages and disadvantages, is presented in Table Table 2-20 MTBE Air Stripper System Off-gas Removal Rates Required to Meet 1 lb/day Discharge Limit Groundwater Flow Rates C WATER µg/l C AIR ppmv (MTBE) lb/day Emission 60 GPM 600 GPM 6,000 GPM Removal Required for Off-Gas Control lb/day Emission Removal Required for Off-Gas Control lb/day Emission Removal Required for Off-Gas Control 2, % % % N/R % % N/R 0.11 N/R % *All information obtained from vendors. 1. Based on an AWR ratio of 160 and a groundwater effluent target concentration of 5 µg/l; assumes a 1 lb/day control limit. 2. N/R = off-gas control not required under the scenarios stated above. 3. A higher AWR ratio will result in lower CAIR but the same lb/day emission assuming the same removal efficiencies. 4. The presence of other volatile organic compounds will result in higher lb/day emissions. 70

93 Off-Gas Treatment Technology Brief Description System Components Advantages Disadvantages GAC Removal of pollutants by means of physical adsorption onto activated carbon grains. Fan Pretreat heater Adsorber vessels GAC Monitoring Instrumentation 100% removal rates attained. Regenerative carbon beds allow material recovery. Spent carbon transport and disposal may require hazardous waste handling permits. Humid air streams require heating for MTBE control in the presence of competing water vapor. 71 Thermal Oxidation Catalytic Oxidation Biofiltration Destruction of pollutants by thermal oxidation. Alternative oxidation process incorporating a reduced temperature burner and catalyst bed. Adsorption onto natural or inert media where microorganisms degrade and oxidize the pollutant. Fan Heater Combustion chamber Fuel Stack Fan Heater Catalyst Fuel Stack Blower Humidifier Biofilter bed Efficient at removing a many contaminants from gas streams in which concentrations and flow are apt to vary. Catalyzed reactions proceed at lower temperatures and allow lower energy requirements. Environmentally safe and create no secondary pollution. Necessity for high temperature operation and supplementary fuel Nitrogen oxides generated by combustion process. Potential catalyst poisoning caused by dust and heavy metals Increase in VOC content may cause temperature rise and destroy the catalyst. Limited examples of MTBE oxidation. Limited by the compound s inherent biodegradability. Can require large areas due to slow biodegradation rates. Microbial activity is very sensitive to biofilter conditions. Table 2-21 Description of Off-gas Treatment Technologies Advanced Oxidation Process Oxidation of organic compounds using reagents such as ozone, hydrogen peroxide, titanium oxide and UV. UV Photolytic reactor Ozone generator Reaction Vessel Isothermal technique requiring no fuel. Oxidation process without the production of nitrogen oxides. Gas phase application of AOP is limited compared to the treatment of aqueous systems.

94 2.4.1 Vapor Phase GAC Adsorption System Description Vapor phase GAC adsorption is a well-known technology used to remove a wide range of organic compounds from air streams. Removal of organic compounds from air occurs by means of physical adsorption on activated carbon. The air passes through a fixed bed of activated carbon until the capacity of the carbon is nearly exhausted. Activated carbons are derived from a variety of carbonaceous materials, including hard wood, coal, petroleum coke, fruit pits, and coconut shells. Gas phase carbon adsorbers are designed as either single-pass or regenerative beds, depending on the mass of chemicals in the feed stream. For detailed information on GAC treatment systems in liquid-phase applications, see Chapter 4. Advantages/Disadvantages See Table 2-21 for a summary list of advantages and disadvantages for this technology. The most important advantages of GAC include the following: GAC operates effectively in cases of low MTBE concentrations (less than 100 ppmv); whereas the effectiveness of alternative off-gas treatment technologies, including thermal and catalytic oxidation, is not well documented for very dilute air streams. GAC is readily available from local suppliers. GAC adsorbers are simple to install and easy to operate compared to technologies involving thermal or catalytic processes. GAC is not operated at high temperatures. The key disadvantages of GAC include: MTBE adsorbs poorly compared to other BTEX compounds on GAC. The capacity of GAC for MTBE removal is reduced by competing adsorbates (e.g., other organics, water vapor). Off-gas streams produced by air stripping are near 100 percent humidity, which will decrease the effectiveness of GAC and, thus, require heating or dehumidifying of the air stream. Raising the temperature of the off-gas to 85 F reduces the relative humidity to 50 percent. However, high air temperatures (above 80 F) will in turn reduce the adsorptive capacity of GAC. If regenerative beds are not used on-site, the carbon needs to be regenerated off-site. In comparison, thermal processes or biofiltration do not generate secondary waste streams. 72

95 Key Variables/Design Parameters The two key design factors for GAC systems include: (1) identification of the most effective carbon adsorbent for the removal of MTBE, and (2) selection of the bed length and crosssectional area. The key design parameters and their effect on regeneration frequency are listed in Table For example, increasing relative humidity (greater than 50 percent) causes premature breakthrough resulting in increased regeneration frequencies. As previously noted, the air streams from air strippers will contain 100 percent relative humidity. Heating of the air to reduce its relative humidity or dehumidifying of the air stream is often required to increase GAC capacity for the organic compound of concern. Table 2-22 GAC Design Variables Design Parameter Air Temperature Effect of Increasing Design Parameter on Regeneration Frequency Regeneration Frequency Effect of Increasing Design Parameter on Costs Annual Costs Humidity >50% Regeneration Frequency Annual Costs Capital Costs Influent Concentration Regeneration Frequency Annual Costs Superficial Velocity Regeneration Frequency Annual Costs Bed Length Regeneration Frequency Annual Costs Capital Costs Isotherm Constant (K) Regeneration Frequency Annual Costs System Installation and Manufacturers There are no known full-scale installations of GAC beds designed to remove MTBE from the vapor phase. In general, carbon adsorption has been used for at least four decades to provide solvent recovery and odor control, and treatment of gases containing other VOCs. Many commercial grades of activated carbon are produced throughout the United States by more than 20 vendors. Representative manufacturers include Calgon Carbon Corporation (Pittsburgh, PA), U.S. Filter/Westates (Los Angeles, CA), Carbochem, Inc. (Ardmore, PA), Tigg Corporation (Bridgeville, PA), and Nucon International Inc. (Columbus, OH). Adsorber beds are produced by many manufacturers, including Indusco Environmental (Atlanta, GA), Norit Americas Inc. (Atlanta, GA), and Tigg Corporation (Bridgeville, PA). A variety of services for supplying and periodically replacing GAC are available (see Chapter 4 for more information). 73

96 Based on MTBE adsorption equilibrium data from manufacturers, the air phase adsorption capacity of GAC ranges from 6 to 10 percent (g MTBE/100 g carbon) at 1 atm and 60 F for concentrations up to 100 ppmv (Calgon Carbon Corporation, 1998). However, an air stripper designed for MTBE removal employs a relatively high air/water ratio, diluting the MTBE air stream concentrations to much less than 10 ppmv (see Table 2-20). The more dilute the MTBE concentration, the lower the carbon adsorption capacity. In addition, humid off-gases from an air stripper reduce the capacity of GAC for MTBE adsorption. Therefore, GAC for MTBE likely ranges from two to five percent by weight. Technical Implementability GAC adsorbers are either packaged or custom built according to site conditions. Components can be separately installed, connected, and then filled with carbon. In addition to the vessel, the system includes auxiliary valves, gauges, and bed sampling ports. The exhausted carbon may be disposed of or regenerated at an off-site facility Thermal and Catalytic Oxidation System Description Other applicable technologies to treat MTBE off-gases generated from air stripping are thermal and catalytic oxidation. Thermal and catalytic oxidation processes can reliably and safely destroy up to 99 percent of the MTBE emissions. Based on manufacturer data, MTBE will readily burn in an oxidizer and will be converted to carbon dioxide and water as easily as other low molecular weight hydrocarbons (Thermatrix, Inc., 1998). However, a full-scale application of thermal processes to treat an air stream containing MTBE has not been reported. Thermal oxidizers are devices in which the air stream containing organic compounds is passed over or through a burner flame or other pre-heat device into a chamber where the organic compounds are oxidized. The units are typically single-chamber, refractory-lined oxidizers equipped with a propane or natural gas burner and a stack (Thermatrix, Inc., 1998). Catalytic oxidizers incorporate a bed of catalytic surfaces to initiate and promote oxidation at lower temperatures (600 to 800 F) than would be used in thermal oxidation (1,350 to 1,800 F). Typical catalysts are composed of a ceramic or metal substrate with a high surface area to volume ratio. Covering the ceramic or metal substrate is a thin layer of catalytic material. The most common catalytic materials used in the environmental industry are noble and base metals, such as platinum, palladium, and vanadium, dispersed on substrates including aluminum oxide, silicon oxide, titanium oxide, or crystalline alumina. In most cases, thermal or catalytic oxidation process costs can be lowered using heat recovery equipment. Oxidizers incorporate heat exchangers as recuperative or regenerative systems. In recuperative oxidizers, an air-to-air heat exchanger transfers the energy of the exhaust to 74

97 the incoming process gas stream to reduce auxiliary fuel costs; up to 70 percent of the heat of the exhaust gases can be recovered. In regenerative oxidizers, beds of ceramic materials are used as a medium to transfer the heat to the incoming process stream; a maximum of 95 percent heat recovery can be obtained. Typically, the capital costs for regenerative oxidizers are higher than recuperative systems (Advanced Environmental Systems, 1998). Advantages/Disadvantages See Table 2-21 for a summary list of the typical advantages and disadvantages of thermal and catalytic oxidation systems. The most important advantages of thermal oxidation include the following: Upon startup, destruction efficiencies consistently reach 99.9 percent for a variety of organic compounds. Heat recovery allows for a recuperation of heating costs. This is an advantage over carbon adsorption, where heating of humid air is required and the costs are not recovered. Thermal oxidation is highly flexible and performs reliably under changing conditions (when concentrations and flow vary). Thermal oxidization is also an advantage when the air stream contains catalyst inhibitors (e.g., sulfur). The key disadvantages of thermal oxidizers include: Compared to catalyst beds, thermal oxidation requires high temperature operation and relatively large amounts of supplementary fuel unless heat recovery is efficiently utilized. This translates into higher annual costs. Thermal oxidizers employ specialty materials required to withstand extreme temperatures (1,800 F). The cost of construction materials can be greater than the associated capital cost for a catalytic oxidizer. There is limited data on the efficiency and costs of treating high volume air streams with low MTBE concentrations. Generation of combustion by-products are possible (e.g., carbon monoxide [CO], oxides of nitrogen [NO x ], oxides of sulfur [SO x ]). The key advantages of catalytic oxidizers include: Catalytic oxidizers operate at lower temperatures compared to thermal oxidation, allowing lower energy requirements and directly translating into improved economics for fuel use. Catalytic oxidizers do not require temperature-resistant construction materials. 75

98 The most significant disadvantages of catalytic oxidizers include: If the composition of the gas stream is known to vary, a sudden increase in VOC content can cause temperatures to reach levels that deactivate or destroy catalysts. For MTBE treatment applications, periods of high MTBE loadings in the air stripping off-gas are not expected and, therefore, high temperature surges are not likely to occur within the catalyst bed for MTBE treatment. Catalyst beds are more easily clogged and destroyed by dust compared to thermal oxidizers. There is limited data on the efficiency and costs of treating high volume air streams with low MTBE concentrations. Generation of combustion by-products is possible (e.g., CO, NO x, SO x ). Key Variable/Design Parameters Thermal Oxidizers. The key design parameters for thermal oxidizers are gas flow rate, temperature and pressure of the inlet gas, concentration of the contaminant, fuel requirements and residence time in the reaction chamber. The following design conditions are typically specified: residence time in the combustion chamber ranges from 0.5 to 1.0 seconds, and operating temperatures are within the limits of 1,350 to 1,800 F (730 to 980 C). Gas flow rates can range from 300 to 30,000 standard cubic feet per minute. The efficiency of thermal oxidation for some VOCs is 90 percent at lower temperatures (1,350 F), but reaches up to 99.9 percent above 1,500 F for most organic compounds. Catalytic Oxidizers. Similar to thermal oxidization, design variables for catalytic oxidizers are gas flow rate, temperature and pressure of the inlet gas, concentration of the contaminant, fuel requirements, residence time in the reaction chamber, and space velocity in the catalyst bed. The space velocity is defined as the volumetric gas rate at operating conditions divided by the volume of the catalyst chamber. Catalytic oxidizers meet stringent air emissions requirements and demonstrate removal efficiencies up to 99 percent. System Installations and Manufacturers There are no reported pilot-scale or full-scale applications of thermal processes to treat gas streams containing MTBE; however, effective treatment is expected. Thermal oxidation is effective for treating most mixed hydrocarbon-laden fumes and deodorizing foul-smelling gases. Catalytic oxidation has also been widely used and is now a mature technology used to control air streams from wastewater, groundwater, and soil remediation projects. Catalytic oxidation technology continues to evolve with new heat recovery systems and catalyst materials to improve efficiency. There are over 40 vendors of thermal oxidizers nationwide (see Appendix 2D) and several proprietary variations are available. Catalytic oxidation systems specifically designed for remediation are manufactured by more than 20 companies. 76

99 ABB Preheater Inc. (Wellsville, Inc.) patented Combu-Changer regenerative thermal oxidizer system, Cor-Pak thermal oxidizer, and Cor-Pak catalytic oxidizer. Flameless Thermal Oxidizer is offered by Thermatrix, Inc. (San Jose, CA) and Advanced Environmental Systems Inc. (Elkton, MD). Other manufacturers include Megtec Corporation (De Pere, WI), North American Manufacturing Co. (Cleveland, OH), Catalytic Products International (Lake Zurich, IL), and CVM Corporation (Wilmington, DE). Technical Implementability Oxidizers are constructed either as package or field-erected units. The only moving part is the blower that will likely be connected to the influent air stream of the air stripper. Monitoring of the oxidizer is augmented by instrumentation that allows the operator to monitor in real time the working temperature, flow rate, and oxidation efficiency; maintenance requirements are minimized using these controls. Installation includes site preparation and construction of ductwork and foundations. One disadvantage of full-scale applications where air volumes are high is that the oxidizer footprint can be large greater than 30 feet in length and 20 feet in height (Advanced Environmental Systems, 1998). Smaller units can be installed without any restrictions Biological Treatment System Description Another option for off-gas control of organic compounds is the use of gas-phase biological aerobic oxidation. Studies have shown that MTBE can be degraded to carbon dioxide and water in biofilters, provided that sufficient residence time is available for growth of the MTBE biodegraders. Lag times ranging from 3 weeks to 12 months have been reported (Eweis et al., 1998). Biotreatment technologies employ either a biofilm on inert media (e.g., GAC) or a compost matrix. Advantages/Disadvantages See Table 2-21 for a summary list of the advantages and disadvantages of biological oxidation systems. The most important advantages of biofiltration include: Oxidation of organic chemicals is mediated by microbes in a natural media, without thermal processes. Biological filters contain environmentally safe components that create no secondary pollution. Operating costs are likely to be low. 77

100 The key disadvantages of biofiltration include the following: The biofiltration process is limited by the contaminant s inherent rate of biodegradability. Process may be highly sensitive to changing influent conditions. Performance may be low during the initial period of microorganism acclimation. There may be a slow transient response of biological systems to variations in the inlet concentration and flow. MTBE degradation appears to be slow and may not be sustainable due to low cell yields. However, research is on-going and biodegradation of MTBE may prove to be more feasible in the future. Key Variable/Design Parameters The key design parameters for biofilters are gas velocity across the bed face, length and volume of the bed, contaminant loading, and inlet gas flow rate. Volumetric production capacity, typically expressed as biomass per unit volume per unit time, tends to increase with contaminant concentration. Controlling variables are temperature, moisture content, ph, porosity, and nutrient concentrations. The optimum operating ranges for these parameters depends on the compound. Humidity must be carefully monitored in biofilters to avoid water clogging of the compost matrix pores. System Installations and Manufacturers Small biofilters are used extensively in the United States in compost piles and wastewater leach fields, removing odorous gases. This technology has also been implemented in Europe and Japan where at least 500 permanent biofilters are in operation (Bohn Biofilter Corporation, 1998). The technology has recently received more attention in the United States. Ametek Rotron Biofiltration Products (Saugerties, NY) has installed full-scale Biocube Aerobic Filters to treat gasoline hydrocarbon gases from soil vapor extraction systems. Bohn Biofilter Corporation (Tucson, AZ) has been involved in more than 20 permanent biofilter beds in the United States. Monsanto Enviro-Chem (St. Louis, MO) has also recently become a major supplier of biofilters. Specific strains of bacteria may be introduced into the filter to preferentially degrade MTBE. Bench-scale reactors used in laboratory studies have maintained stable activity for MTBE degradation. However, without enrichment and recycle, cell growth rates and cell yields are low (0.05/d and 0.1 to 0.3 g/g-mtbe, respectively). MTBE biofiltration research is currently on-going at several universities and industrial laboratories, including the University of California, Davis and Equilon Enterprises, LLC. (Eweis et al., 1998; Salanitro et al., 1999). While isolated strains of microorganisms are capable of degrading MTBE, field tests are required to determine persistence, stability, and metabolic activity in engineered processes under actual processing conditions. 78

101 Technical Implementability In compost biofilters, the medium is sometimes turned at intervals and must be replaced every 2 to 4 years. Systems using inert material are operated without any replacement schedule. The system components include a humidifier to saturate the vapor, blower, biofilter bed, distribution ducts beneath the bed, and a drainage pipe for recycling water to the humidifier. However, in order to treat the off-gas from an air stripping system, a humidifier and blower are not likely needed. Most biofilters also include a direct irrigation system consisting of sprinklers above the medium or soaker hoses within the medium. These can be used to add water when the humidified gas stream is not sufficient to meet the needs of the microbes and nutrients to maintain peak biological activity. Some systems incorporate interlocking trays to design for expandable modules. 79

102 80

103 2.5 Evaluation and Screening of Off-gas Treatment Technologies Permitting Among the off-gas treatment technologies, thermal oxidation, catalytic oxidation, or GAC adsorption systems would likely be accepted most easily by regulatory agencies as an effective means of treating vapor emissions contaminated with MTBE. When used in conjunction with the selected air stripping technology, both GAC adsorption and thermal oxidation are listed as best available control technologies (BACT) and, therefore, are expected to meet stringent requirements for removal efficiency. With respect to biofiltration, the compliance issues focus primarily on the need to demonstrate reliable and consistent biodegradation of MTBE. Periods of failure and low microorganism performance have been reported in laboratory studies (Salanitro et al., 1994). Regulators will likely inspect the site and review biofilter operations on the basis of stable performance, complete destruction of MTBE, and by-product formation. Permitting is highly site-specific depending on local demographics, site location, and the regional ambient air quality Flow Rate All of the chosen off-gas treatment technologies can handle a wide range of air stream flow rates. Carbon adsorption, biofiltration, and thermal and catalytic oxidation can all operate between several hundred cubic feet per minute (cfm) to over 100,000 cfm. However, capital investment and operating costs increase proportionally to the flow rate. Capital expenses, annual O&M costs, and total amortized operating costs are listed in Tables 2-23, 2-24, and 2-25, respectively. Biofiltration is not addressed in these tables because no vendor estimates were provided, as discussed at the end of this section Removal Efficiency Some California air districts (e.g., South Coast AQMD) enforce an emission control standard of 1 pound VOC/day from an air stripper. For this analysis, off-gas treatment has been implemented when MTBE concentrations at the air stripper exit result in emissions in excess of the 1 lb/day mass limit as shown in Table An off-gas treatment unit would be required to destroy enough MTBE to fall below the 1 lb/day discharge limit. At the low MTBE air concentrations expected in the off-gas from air strippers, GAC, and thermal and catalytic oxidizers are expected to achieve greater than 99 percent removal of MTBE. However, due to the low affinity for MTBE adsorption onto vapor phase GAC, field verification of a steady, continuous removal efficiency is required. Typical destruction efficiencies of thermal and catalytic oxidation are 99 percent and 97 percent, respectively, for concentrations of most organic compounds greater than 5,000 ppmv 81

104 (Schen et al., 1993). However, potential concentrations of MTBE generated from air stripping are three orders of magnitude lower than 5,000 ppmv. There is little or no inherent heating value in the concentrations of MTBE emitted from an air stripper. Costs for MTBE destruction will, therefore, be governed by the amount of energy required to operate the offgas treatment system, and depend very little on the influent MTBE gas concentration. Biofiltration has demonstrated between 90 to 98 percent removal efficiencies of other organic compounds at concentrations between 20 ppmv and 1,000 ppmv. Consistent removal efficiencies have not been demonstrated for concentrations of VOCs below 20 ppmv or for any concentrations of MTBE. Thus, it is expected that a biofiltration off-gas treatment unit for MTBE removal will require significant field verification prior to regulatory approval. 82

105 83 Flow (gpm) Influent MTBE (ppmv) GAC Recuperative Thermal Oxidation Recuperative Flameless Thermal Oxidation Recuperative Catalytic Oxidation Non-Recuperative Catalytic Oxidation $11,100 $199,800 $732,600 $166,500 ND $111,000 $666,000 $1,443,000 $510,600 $333, $1,010,100 $3,889,440 $5,550,000 $3,152,400 ND 60 5 $11,100 $199,800 $732,600 $166,500 ND $111,000 $666,000 $1,443,000 $510,600 $333, $1,010,100 $3,889,440 $5,550,000 $3,152,400 ND ND = no data available. Capital Expenses include: Equipment Piping, valves, electrical (30%) Site work (10%) Contractor O&P (15%) Engineering (15%) Contingency (20%) Data is based on vendor information provided for the following scenarios: MTBE Removal: Influent 5 ppmv, Effluent 0.03 ppmv. Air Temperature: 65 F. Flow Rate: 60 gpm system: 1,200 cfm (AWR 150:1). 600 gpm system: 12,000 cfm (AWR 150:1) gpm system: 120,000 cfm (AWR 150:1). Table 2-23 Initial Capital Expenses for Off-gas Treatment Systems

106 Flow (gpm) Influent MTBE (ppmv) GAC Recuperative Thermal Oxidation Recuperative Flameless Thermal Oxidation* Recuperative Catalytic Oxidation Non-Recuperative Catalytic Oxidation $16,076 $21,083 $50,006 $24,332 ND $67,165 $117,229 $61,920 $150,063 $382, $640,448 $1,078,688 $354,786 $1,407,032 ND 60 5 $57,704 $21,083 $50,006 $24,332 ND $483,440 $117,229 $61,920 $150,063 $382, $4,803,200 $1,078,688 $354,786 $1,407,032 ND ND = no data available O&M Costs include: 1. Power costs (at $0.08/kWhr). 2. Fuel costs (at $5.00/ MBtu). 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 4. GAC includes GAC changeout at $1.50/ lb, use rate of 0.05 lb/1000 ft 3 air at 5 ppmv MTBE and lb/1000 ft 3 at 0.5 ppmv MTBE. 5. Recuperative Flameless Thermal Oxidizer (at 600 gpm) incorporates a rotor concentrator. Data is based on vendor information provided for the following scenarios: 1. MTBE Removal: Influent 5 ppmv, Effluent 0.03 ppmv: 99.4% removal. Air Temperature: 65 F (except for GAC - air is heated to 85 F). 2. Flow Rate: 60 gpm system: 1,200 cfm (AWR 150:1). 600 gpm system: 12,000 cfm (AWR 150:1) gpm system: 120,000 cfm (AWR 150:1). Table 2-24 Annual O&M Costs for Off-gas Treatment Systems Cost estimate data provided by: Carbon Adsorption: Calgon Carbon (Pittsburgh, PA). Recuperative Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Recuperative Flameless Thermal Oxidation: Thermatrix (San Jose, CA). Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Non-Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD).

107 85 Flow (gpm) Influent MTBE (ppmv) GAC Recuperative Thermal Oxidation Recuperative Flameless Thermal Oxidation Recuperative Catalytic Oxidation Non-Recuperative Catalytic Oxidation $0.54 $1.18 $3.46 $1.20 ND $0.24 $0.54 $0.57 $0.61 $ $0.23 $0.44 $0.25 $0.53 ND 60 5 $1.86 $1.18 $3.46 $1.20 ND $1.56 $0.54 $0.57 $0.61 $ $1.55 $0.44 $0.25 $0.53 ND N/D = No data available. Amortization based on a 30-year period at a 7% discount rate. *To convert costs to $/acre-ft, multiply by 326. O&M Costs include: 1. Power costs (at $0.08/kWhr) 2. Fuel costs (at $5.00/ MBtu) 3. Analytical costs estimated at $200 per sample; see Table 2-11 for the number of samples required for each system. 4. GAC includes GAC changeout at $1.50/ lb, use rate of 0.05 lb/1000 ft 3 air at 5 ppmv MTBE and lb/1000 ft 3 at 0.5 ppmv MTBE. 5. Recuperative Flameless Thermal Oxidizer (at 600 gpm) incorporates a rotor concentrator. Data is based on vendor information provided for the following scenarios: 1. MTBE Removal: Influent 5 ppmv, Effluent 0.03 ppmv: 99.4% removal. Air Temperature: 65 F (except for GAC - air is heated to 85 F). 2. Flow Rate: 60 gpm system: 1,200 cfm (AWR 150:1). 600 gpm system: 12,000 cfm (AWR 150:1) gpm system: 120,000 cfm (AWR 150:1). Cost estimate data provided by: Carbon Adsorption: Calgon Carbon (Pittsburgh, PA). Recuperative Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Recuperative Flameless Thermal Oxidation: Thermatrix (San Jose, CA). Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Non-Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Table 2-25 Total Amortized Operating Costs ($/1,000 Gallons Treated)* for Off-gas Treatment

108 Table 2-26 Air Stripper Exit Concentrations Initiating Off-gas Treatment Effluent Air Stripper Concentration of MTBE initiating the need for off-gas treatment (emissions limit of 1 lb/day MTBE) Air/Water Ratio Air Stripper Flow Chart 60 GPM 600 GPM 6,000 GPM ppmv 0.4 ppmv 0.04 ppmv ppmv 0.3 ppmv 0.03 ppmv ppmv 0.2 ppmv 0.02 ppmv Notes: 1. Based on an air temperature of 65 F. 2. The presence of other volatile organic compounds in the air stream will result in a higher total lb/day emission Other Characteristics Table 2-27 contains a summary of the characteristics of off-gas technologies with respect to reliability, flexibility, adaptability, and potential for modification. Reliability Off-gas treatment is not well documented for the removal of MTBE, although experience with other VOCs indicates that GAC and thermal and catalytic oxidation technologies meet reliability criteria. A fixed bed carbon adsorber has demonstrated a high degree of system reliability due to process simplicity; however, reliability for MTBE removal will likely require field demonstration. Thermal oxidizers have few moving parts and are not known to mechanically fail and are, thus, expected to be highly reliable. The problems associated with recuperative or regenerative systems are likely to be related to heat exchanger fouling. Catalytic oxidation systems demonstrate only moderate reliability since feed streams containing dust and particulates can plug the catalyst section. Biofilters demonstrate low reliability because the activity of the microorganisms for MTBE destruction is unknown and could periodically drop to zero if influent concentrations are low. Also, there is less experience relative to the other off-gas treatment technologies with biofilter maintenance and operation. 86

109 Off-Gas Treatment Technology Reliability Flexibility Adaptability Potential for Modifications 87 GAC Thermal Oxidation Catalytic Oxidation HIGH Carbon adsorption cycle can be relied upon provided that operating procedures are followed. HIGH Oxidizers can be relied upon provided that operating procedures are followed. MEDIUM Failure of catalysts caused by certain heavy metals, dust, sulfur and chlorine can occur. MEDIUM Air flow rate design can range from 300 to 100,000 cfm. Molecular weight of target compounds varies from 45 and 130. Any compound will be adsorbed in proportion to its gas phase concentrations, but retention on the bed depends on volatility and nature of the compound. HIGH Air flow rate design can range from 100 to 100,000 cfm. Most organic compounds and hydrocarbons will be completely destroyed. HIGH Air flow rate design can range from 100 to 100,000 cfm. Most organic compounds and hydrocarbons will be completely destroyed. MEDIUM Variations in air quality can reduce the number of adsorption sites available to the contaminant of concern. Particles clog the adsorbent bed. HIGH Thermal oxidizers handle variations in influent conditions and air quality, continually removing VOCs to desirable levels. LOW Variations in air quality and influent conditions affect the VOC-air mixture and may cause a temperature rise resulting in decreased performance. MEDIUM New units can be easily added in modular fashion. LOW Thermal oxidizers can be modified by raising the operating temperature, but it leads to higher fuel requirements. LOW Raising temperatures in excess of the maximum design deactivates catalysts. Replacement of the catalyst section can be cost prohibitive. Table 2-27 Comparison of Off-gas Treatment Technologies Biofiltration LOW Process is not subject to significant mechanical failure, but biological activity can be unreliable. LOW Changing air flow rates may reduce removal efficiency. Vapor phase MTBE degradation not well demonstrated. Lag time delays onset of MTBE oxidation. LOW Variation in air quality and influent conditions can decrease or eliminate biological activity. LOW Biofilters can be enhanced by seeding differently, adding cosubstrates, etc. but response time to modified conditions is slow.

110 Flexibility GAC is moderately flexible, referring to its effectiveness over a range of air flow rates, target compound properties, and target compound concentrations. All organic compounds will be adsorbed to the GAC in proportion to their gas phase concentration, but retention on the carbon bed depends on the physical-chemical nature of the compound (see Chapter 4 for more information). Thermal treatment exhibits high flexibility and can be designed for both low and high flow rates (100 to 170,000 cfm). Biofiltration is currently considered to be minimally flexible (subject to change as a result of on-going research) due to the long lag time for initial MTBE oxidation. Adaptability Thermal oxidization is the most adaptable off-gas treatment technology. The performance of a thermal oxidizer is not affected by changing influent gas concentrations. Catalytic beds and biofilters are much less adaptable. Catalytic beds are temperature sensitive; a rise in temperature from changing VOC-air mixtures may result in decreased performance or catalyst failure. In biofiltration, fluctuating loadings of MTBE can be toxic to microbes, or cause microbes to preferentially metabolize other carbon substrates, leading to MTBE breakthrough. Potential for Modification GAC systems can be modified easily by addition of parallel units. The other off-gas treatment technologies are not readily suited for modifications. The potential to enhance removal in a thermal oxidizer primarily rests on increasing the temperature but this requires higher fuel use. Catalyst replacement is cost prohibitive and system modifications to the catalyst section are technically limited. Modifications to operating biofilters are also limited By-products It is not likely that large amounts of NO x, SO x, or CO will be generated by thermal oxidizers because advances in the design and operation have significantly decreased or eliminated the production of these oxides. Catalytic systems operate at relatively low temperatures, requiring less fuel than thermal oxidizers and normally have lower oxide emissions. By product formation during biofiltration is not well documented. GAC treatment of off-gasses is expect to produce no by-products. 88

111 2.5.6 Cost Effectiveness GAC The capital costs associated with the installation of an on-site GAC adsorption system are significantly less than thermal and catalytic oxidation systems, but higher than biofiltration. Based on a 5 ppmv influent MTBE concentration, O&M costs are expected to be high due to a relatively high frequency of regeneration or replacement. O&M cost estimates for GAC included carbon changeouts and analytical costs (see Table 2-24) while those of the oxidation systems included fuel costs, electric costs, and analytical costs. Estimates of carbon usage rates range from 50 to 150 lb GAC/day at 1,200 cfm (air flow rate from a 60 gpm water system using an air/water ratio of 150). For the highest assumed flow rate of 6,000 gpm, the carbon usage rate increases to approximately 8,600 lb GAC/day based on a usage rate of 0.05 lb GAC per 1000 cubic feet of air treated. Results of this cost analysis based on vendor data are shown in Tables 2-23, 2-24, 2-25, 2-28 and The unit costs ($/1,000 gallons treated water) of GAC gas-phase adsorption at 60, 600, and 6,000 gpm range from $1.86 to $1.55 for a 5 ppmv influent MTBE concentration and $0.54 to $0.23 at 0.5 ppmv influent MTBE concentration (see Table 2-25). At higher MTBE influent gas concentrations (i.e., 5 ppmv), these costs are generally not competitive with thermal and catalytic oxidation systems, as would be predicted from the high carbon usage rate (8,600 pounds GAC/day). At lower influent MTBE gas phase concentrations of 0.5 ppmv, carbon usage rates and annual O&M costs decrease (see Table 2-24) while thermal and catalytic oxidation costs are unaffected. This trend causes GAC to become the most cost-effective off-gas treatment technology at the lower influent concentrations. The other factor that could cause GAC to become more cost-effective than thermal or catalytic oxidation is the projected number of years of operation. Because carbon capital costs are substantially lower than oxidation capital costs, a shorter amortization period could cause GAC to become more cost-effective than oxidation. Consequently, if a drinking water treatment technology is only needed for a few years (10 years for a 60 gpm system, 2.5 years for a 600 gpm system, and 1 year for a 6,000 gpm system), rather than the 30 years assumed for this evaluation, it is more cost effective to install a GAC system. Thermal Treatment Compared to other off-gas treatment technologies, the capital costs associated with thermal treatment are the most expensive. The system has high capital and installation costs and variable O&M costs. Estimated capital and O&M costs for oxidation systems are presented in Tables 2-23, 2-24, 2-25, 2-28 and

112 Flow (gpm) System Name and Configuration Influent (ppmv) Effluent (ppmv) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* RTO: Burner, Combustion Chamber, Stack 60 Flameless RTO: Burner, Combustion Chamber, Stack % $199,800 $21,083 $ % $732,600 $50,006 $ RCO: Burner, Catalyst Bed, Stack % $166,500 $24,332 $ GAC (1,000-lb bed) % $11,100 $57,704 $ RTO: Burner, Combustion Chamber, Stack 600 Flameless RTO: Burner, Combustion Chamber, Stack % $666,000 $117,229 $ % $1,443,000 $61,920 $ RCO: Burner, Catalyst Bed, Stack % $510,600 $150,063 $ CO (No heat recovery): Burner, Catalyst Bed, Stack 600 GAC (dual bed unit with 16,000 lb total GAC) 6000 RTO: Burner, Combustion Chamber, Stack 6000 Flameless RTO: Burner, Combustion Chamber, Stack % $333,000 $1,407,032 $ % $111,000 $483,440 $ % $3,889,440 $1,078,688 $ % $5,550,000 $354,786 $ RCO: Burner, Catalyst Bed, Stack % $3,152,400 $1,407,032 $ GAC (7 dual-bed units, each with 22,000 lb total GAC) *To convert unit treatment costs to $/acre-ft, multiply by % $1,010,100 $4,803,200 $1.55 Cost estimate data provided by: Carbon Adsorption: Calgon Carbon (Pittsburgh, PA). Recuperative Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Recuperative Flameless Thermal Oxidation: Thermatrix (San Jose, CA). Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Non-Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Table 2-28 Cost Summary for Off-gas Treatment Technologies at 5 ppmv MTBE

113 91 Flow (gpm) System Name and Configuration Influent (ppmv) Effluent (ppmv) Removal (%) Capital Cost ($) Annual O&M ($) Unit Cost ($/1000 gal)* 60 RTO: Burner, Combustion % $199,800 $21,083 $1.18 Chamber, Stack 60 Flameless RTO: Burner, % $732,600 $50,006 $3.46 Combustion Chamber, Stack 60 RCO: Burner, Catalyst Bed, Stack % $166,500 $24,332 $ GAC (1,000-lb bed) % $11,100 $16,076 $ RTO: Burner, Combustion % $666,000 $117,229 $0.54 Chamber, Stack 600 Flameless RTO: Burner, % $1,443,000 $61,920 $0.57 Combustion Chamber, Stack 600 RCO: Burner, Catalyst Bed, Stack % $510,600 $150,063 $ CO (No heat recovery): Burner, % $333,000 $1,407,032 $1.30 Catalyst Bed, Stack 600 GAC (dual bed unit with 16,000 lb total GAC) % $111,000 $67,165 $ RTO: Burner, Combustion % $3,889,440 $1,078,688 $0.44 Chamber, Stack 6000 Flameless RTO: Burner, % $5,550,000 $354,786 $0.25 Combustion Chamber, Stack 6000 RCO: Burner, Catalyst Bed, Stack % $3,152,400 $1,407,032 $ GAC (7 dual-bed units, each with 22,000 lb total GAC) % $1,010,100 $640,448 $0.23 *To convert unit treatment costs to $/acre-ft, multiply by 326. Cost estimate data provided by: Carbon Adsorption: Calgon Carbon (Pittsburgh, PA). Recuperative Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Recuperative Flameless Thermal Oxidation: Thermatrix (San Jose, CA). Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Non-Recuperative Catalytic Thermal Oxidation: Advanced Environmental Systems (Elkton, MD). Table 2-29 Cost Summary for Off-gas Treatment Technologies at 0.5 ppmv MTBE

114 The primary factors that affect the cost of oxidation systems include air flow rates, desired level of destruction, and desired energy efficiency (Van der Vaart et al., 1991). The latter parameter desired energy efficiency becomes more important with increasing air flow rates and longer hours of operation. Based on vendor data, recuperative (i.e., with heat recovery) thermal oxidizers are more cost-effective than non-recuperative systems. The higher capital costs of recuperative oxidizers are offset by lower fuel usage costs and, thus, the increased initial capital costs pay for themselves within a short period of time. For example, in the case of a catalytic oxidation system designed for a water flow rate of 600 gpm, the total amortized operating cost of a non-recuperative system can be over twice as much as a recuperative system ($1.30 vs. $0.61/1,000 gallons treated; see Table 2-25). Biofiltration No engineering cost estimates were provided by vendors for biofiltration. In general, capital costs associated with biofiltration are similar to low compared to other technologies; however, the major cost savings typically result from operating costs (Dharmavaram, 1991). Operating costs are estimated to be less than $0.09 per 1,000 gallons of water treated. For MTBE, no full-scale vapor phase biofilters have been reported at drinking water facilities. Therefore, performance and costs of biofilters for MTBE treatment under actual process conditions are unknown at this time. However, a substantial amount of research is currently ongoing to investigate the feasibility of using biofilters for removal of MTBE from drinking water (e.g., University of California at Davis, University of California, Riverside, and Rutgers). Consequently, biofilters may prove to be an effective treatment technique in the future, but they are not yet proven to be reliable for off-gas treatment field operation. 92

115 2.6.1 Introduction and Design Equations 2.6 Optimization of Air Stripping Technologies One of the objectives of this document is to determine an appropriate range of design and operating parameters to demonstrate the cost-effectiveness of MTBE removal in air stripping drinking water applications. As explained previously, air stripping theory for packed towers is well developed and can be applied to process design, using few empirical correlations (Roberts et al., 1985; Kavanaugh et al., 1980; Ball et al., 1984; Perry et al., 1984; Hand et al., 1986). Using the defining equations, the purpose of this section is to explain how changing the initial operating assumptions will change the packed tower design, and how the operating parameters with a given packing diameter, height, and media will affect performance of a packed tower. This section is intended to illustrate the relationship between air stripping variables and design parameters and should not be used as an exact basis for air stripping design. For this reason, the safety factor, which would typically be divided by the height of the transfer unit (and set at approximately 0.8), has been set at unity. In addition, this section has relied upon a Henry s constant for MTBE of at 20 C. The actual Henry s constant for MTBE may be lower (see Figure 2-1), which will affect the size of the air stripping unit. Furthermore, the results presented in this section are not explicitly tied to the cost estimates previously provided, as all costs presented in this paper are based on assumptions and vendor estimates presented in Appendix 2A. For a packed tower, the design equations are as follows (Montgomery, 1985): NTU = Z = (HTU)(NTU) L HTU = K L a ( C in S C out ) (S 1)+1 ln S-1 S S = H * Where Z is the packing height (ft); HTU is the height of a transfer unit (ft); NTU is the number of transfer units; L is the liquid loading rate (ft 3 /min/ft 2 ); K L a is the overall mass transfer constant (1/min); C in is the influent concentration (µg/l); C out is the effluent concentration (µg/l); S is the stripping factor (dimensionless); H is the Henry s constant (dimensionless); G/L is the volumetric air/water ratio (dimensionless); and G is the air loading rate (ft 3 /min/ft 2 ). Using the above equations, in conjunction with empirical data on air pressure drop as a function of gas flow rate and packing, it is possible to evaluate the effect of the various design parameters on overall tower design. Table 2-30 below summarizes the design drivers that are G L 93

116 inherent to the air stripping design for treating contaminated water, the design parameters that must be determined prior to tower construction, and the design and operating parameters, some of which can be altered during tower operation. Figures 2-6 through 2-12 were developed to visualize the effect graphically of these design and operating parameters on system performance and cost. Table 2-30 Design and Operating Parameters Design Drivers Design Parameters Design & Operating Parameters Influent Concentration Packing Size and Type Influent Flow Temperature Removal Efficiency Tower Height Air / Water Ratio Chemical Properties Tower Diameter Liquid Loading Rate Water Quality Air Loading Rate Pressure Drop Design Parameters In order to design a packed tower air stripper, an air/water ratio, liquid loading, pressure drop, and air loading rate must be selected (design and operating parameters). This will define the tower height, diameter, and packing type (the design parameters) for given contaminants and removal efficiencies (design drivers). The design of an optimized packed tower is a complex function of the design and operating parameters. Figures 2-6 through 2-12 were developed by varying one parameter and calculating the effect on a second parameter, while keeping other parameters constant. A remediation technology design software package developed at Michigan Technical University was used for this purpose (Crittenden et al., 1998). Packing Volume Packing volume is a function of tower diameter and height. Each unit of height achieves the same removal efficiency, as shown in the equation for HTU. The diameter of the tower is a function of the stripping factor and liquid loading. If the air/water ratio is too large for a given diameter, the tower will flood. Figure 2-6 demonstrates that an increase in the removal efficiency requires an increase in the packing volume. As shown in Figure 2-7 there is an optimal stripping factor for a given removal efficiency to minimize the packing volume. Although Figure 2-7 shows that for stripping factors greater than two, an increase in the stripping factor slightly lowers removal efficiency, this is an artifact of setting the pressure drop and liquid flow rate constant. If these parameters are held constant, an increase in the air/water ratio increases the diameter of the tower, which increases the packing volume. 94

117 140 Required Packing Volume (m3) Required Packing Volume (m 3 ) o C 10 o 20 o Assumptions: S=5 (AWR= 161) 600 gpm Tripacks No. 2 Pressure Drop = 50 Pa/m Removal Efficiency Figure 2-6. Effects of increasing the removal efficiency on the packing volume as a function of inlet water temperature. Packing Volume (m 3 ) Packing Volume (m3) Increasing Removal Efficiency Assumptions: Pressure Drop Drop = 50 = 50 Pa/m Pa/m T=10 T = 10 C o C Influent Concentration = 200 = µg/l 200 µg/l 600 gpm flow rate rate 99% Removal 95% Removal 90% Removal Stripping Factor, S S = = H*(AWR) Figure 2-7. Effects of increasing the AWR on the packing volume as a function of removal efficiency. 95

118 160 Packing Volume (m 3 ) Packing Volume (m3) Tripacks No. 1 (2 Inch) Tripacks No. 2 (3.5 Inch) Assumptions: 95% 95% Removal Efficiency gpm gpm flow flow rate Pressure Pressure Drop Drop = = Pa/m Pa/m 200 µg/l influent 200µg/L influent 20 Tripacks No. 1/2 (1 Inch) Stripping Factor, S S=H*(AWR) = Figure 2-8. Effects of changing AWR and packing media on packing volume for a given removal efficiency. 700 Pressure Drop (Pa/m) Pressure Drop (Pa/m) Assumptions: Tower Assumptions: Dimensions = 8 diameter; Tower Dimensions 30 tall = 600 gpm 8' diameter; 30' tall Temp = 10 o C 600 gpm Temp = 10 C Tripacks No. 1/2 (1 Inch) Tripacks No. 1 (2 Inch) 100 Tripacks No. 2 (3.5 Inch) Stripping Factor, S S=H*(AWR) = Figure 2-9. Effects of changing AWR and packing media on pressure drop for fixed tower dimensions. 96

119 40 70 Packing Volume (m3) (m 3 ) Total Brake Power (kw) Total Brake Power (kw) Pressure Drop (Pa/m) Packing Volume: TriPacks No. 1/2 (1 (I inch) Packing Volume: TriPacks No. 2 (3.5 inches) Power: TriPacks No. 1 (2 inches) Packing Volume: TriPacks No. 1 (2 inches) Packing Power: TriPacks Volume: No. TriPacks 1/2 (I No. inch) 1/2 (1 inch) Power: TriPacks No. 2 (3.5 inches) Assumptions: Assumptions: Temperature Temperature = 10 C = 10 C S = S 5 = (AWR 5 (AWR = 161) = 161) gpm gpm 95% 95% Removal Removal Efficiency Efficiency Figure Effects of changing pressure drop on total brake power and the packing volume as a function of packing. 97

120 Assumption: 600 gpm Pressure Drop [Pa/m] ' Diameter 12' Diameter 21' Diameter Stripping Factor, S=H*(AWR) Figure Effects of changing the AWR on pressure drop for a given tower cross-sectional area. 105% Removal Efficiency Removal Efficiency 100% 95% 90% 85% 80% 60 gpm 600 gpm 1000 gpm Assumptions: 200 Assumptions: µg/l influent Air 200 Stripper µg/ldesign influent for 95% removal Air Stripper at design for 60095% gpmdesign removal flow at (S=5) 600 gpm design flow Tripacks No. 2 Temperature (S = 5) = 10 o C Tripacks No. 2 Temperature = 10 C 75% 70% Stripping Factor, S=H*(AWR) S = H*(AWR) Figure Demonstration of air stripping flexibility (ability to handle a variety of flow rates) for a given design. 98

121 Packing Media There are many different types and sizes of packing materials available, with rings, saddles, and spheres being the most prevalent packing shapes. Figures 2-8, 2-9, and 2-10 illustrate the effect of various sized Tripack media on required packing volume, pressure drop, and total power, respectively. As is seen in Figure 2-8, the largest packing size requires the most packing volume for a given removal efficiency. However, packing volume appears to be independent of size for Tripack media less than 2 inches. Figure 2-9 shows that as the stripping factor increases, packing size plays a more important role in determining the pressure drop. As expected, a smaller packing size causes a larger pressure drop for a given air/water ratio. Figure 2-9 also demonstrates that if a larger packing is chosen (2 inch or 3.5 inch), the stripping factor can be changed dramatically without significant changes in the pressure drop, relative to the 1-inch packing. The third variable to consider when choosing a packing media is the total amount of brake horsepower required to achieve the desired removal efficiency. Brake horsepower (the ideal power plus the frictional power requirements) represents the largest O&M cost for a packed tower air stripper and packing volume represents the capital costs. Figure 2-10 shows the trade-off between capital costs and O&M costs for a packed tower. As the pressure drop increases, the brake power will increase and the required packing volume will decrease for a given packing media. For a given removal efficiency, stripping factor, and water flow rate, the height of the packed tower increases as the size of the media increases. While this larger media increases the total brake power (see Figure 2-10), it does not significantly increase the packing volume. This is due to the trade-offs mentioned above. Finally, Figure 2-10 shows that, for a constant packing volume, increases in packing size require an increase in pressure drop to maintain a constant removal efficiency. As the packing size increases, turbulent mixing (i.e., contaminant removal) decreases, requiring greater packing depths to achieve the same removal efficiency Operating Parameters Temperature A simple, yet expensive, way to increase air stripping removal efficiency (see Figure 2-6) is to increase water temperature, either by adding heat to the influent water, or by using steam as the stripping fluid in place of air. As the water temperature increases, the Henry s constant and overall mass transfer coefficient will increase, thus reducing air/water ratios and the volume of packing required. In typical air stripping operations, the influent water temperature is not altered from the natural temperature of the groundwater. Figure 2-6 shows that if a higher inlet water temperature is used in the design, packing volume and capital costs can be reduced substantially. However, these capital cost savings are offset by an increase in O&M costs. For large-scale drinking water applications, use of steam stripping or heating of the water is not cost-effective unless there is a low cost source of heat available from another process (i.e., waste heat from a thermal oxidizer or other unrelated process). 99

122 Stripping Factor and Pressure Drop Increases in the air/water ratio for a fixed packing and flow rate will increase the air pressure drop. Figure 2-11 shows that, for a given tower diameter, an increase in the stripping factor will cause an increase in the pressure drop for a constant liquid flow rate. The magnitude of the pressure drop decreases as the tower diameter increases. Flow Rate and Removal Efficiency Figure 2-12 shows the effect of changing the water flow rate on the removal efficiency for a given tower design. As the water flow rate increases, the stripping factor must increase accordingly to achieve the same removal efficiency. As the water flow rate decreases, the stripping factor can be decreased without decreasing the removal efficiency. In practice, Figure 2-12 illustrates that a packed tower air stripper should be designed for the maximum flow rate that may need treatment. This will guarantee that the air stripper will be able to achieve the desired removal efficiency for all flow rates Summary The purpose of this analysis, besides demonstrating the relationship between the design driver variables, design parameters, and operating parameters, is to determine the range of optimum design parameters for an air stripper. Based on the above analysis, the following range of design and operating parameters for a packed tower stripper is recommended: Packing Size. Medium-sized packing (2 inch) offers the optimal trade-off between pressure drop, packing volume, and total brake power. As noted, 2-inch packing requires a larger pressure drop than 3-inch packing, but requires less total brake power (lower O&M costs) to achieve the same removal efficiency. Alternatively, 2-inch packing requires a lower pressure drop than 1-inch packing, but not a larger packing volume for a given removal efficiency. Stripping Factor. As explained previously, as the stripping factor increases, the O&M costs also increase. However, the higher the stripping factor, the higher the removal efficiency for a given packed tower volume. For this reason, stripping factors between four and seven represent an appropriate testing range. A field study should be designed to allow testing of this range of stripping factors. Performance data should be obtained over this range because the lower the acceptable stripping factor, the lower the off-gas capital and O&M costs. Other Factors. Liquid loading rate and possibly influent temperature should also be tested. As the liquid loading rates increase for a given tower volume, the stripping factor decreases. Changing the flow rate can significantly increase the O&M costs while augmenting the removal efficiency of MTBE from water. 100

123 2.7 Conclusions and Recommendations for Future Research Recommended Technologies Air stripping is a mature technology that has been widely used to produce potable water from sources contaminated with volatile organic compounds. As noted in this evaluation, however, there are only a few examples where air stripping has been used to remove MTBE or other fuel oxygenates from potable water sources. Thus, a pilot- or full-scale testing program is warranted to demonstrate the capabilities and possible limitations of this technology for producing potable water from sources contaminated with MTBE. In addition to testing the air stripping system, a pilot test should be performed to evaluate a compatible off-gas treatment technology. MTBE water treatment scenarios are expected to vary from site to site. Two primary scenarios are envisioned: 1) a remediation scenario with high initial MTBE concentrations (>1,000 µg/l) and low (<100 gpm) flow requirements; and 2) a drinking water scenario with low initial concentrations (<1,000 µg/l), but generally higher capacity requirements (>100 gpm). The recommendations discussed below will focus primarily on the application of air stripping systems in the second scenario: drinking water applications with lower MTBE influent concentrations. Recommended Stripping Technologies Packed tower aeration is superior to other air stripping technologies from a cost perspective, regardless of hydraulic capacity, removal efficiency requirements, or initial MTBE concentrations (see Table 2-12). At higher flow rates (>600 gpm) and removal efficiencies (>95 percent), packed towers are not only less expensive but, often, the only technology capable of achieving the treatment goal (see Figure 2-13). However, for lower flow rates (<100 gpm), low profile air strippers become cost competitive with packed towers ($1.80- $1.86 vs. $1.75 per 1,000 gallons treated, respectively, for 97.5 percent MTBE removal). Low profile air strippers are generally easier to install, maintain, and adapt to changing flow and water quality conditions than packed towers. Thus, for drinking water scenarios requiring hydraulic capacities less than 100 gpm, which may include either a remediation scenario or a drinking water scenario, a low profile air stripper is recommended. For hydraulic capacities greater than 100 gpm, the packed tower aeration technology is recommended. 101

124 Figure Domains for cost-effectiveness of air strippers at varying removal efficiencies. 102

125 Off-gas Treatment Technology Recommendation For this evaluation, it is assumed that off-gas treatment is required when MTBE gas phase loadings exceed 1 lb/day (see Table 2-12). If off-gas treatment is required and MTBE influent concentrations are low (<200 µg/l), GAC is the most cost-effective off-gas technology because carbon usage rates are low (as a result of the very dilute MTBE stream) and, thus, O&M costs remain low. If MTBE influent concentrations are much higher (e.g., the 2,000 µg/l scenario), oxidation is the recommended technology for an air stream from a packed tower system. The results of the cost analysis show that there is a small difference in costs between catalytic and thermal oxidation. Thus, thermal oxidation is the technology recommended for evaluation in conjunction with the selected aeration technology. Thermal oxidation using heat recovery and GAC are proven technologies. Because both technologies demonstrate an equally high level of reliability, flexibility, and removal efficiencies, costeffectiveness becomes the determining factor in the choice of an off-gas technology. Recommended Combined System Based on the above evaluation, the recommended combined technologies are a packed tower air stripper for high flows (>100 gpm) and the low profile system for low flows (<100 gpm). If necessary, both systems can be combined with the off-gas technology identified above for the respective flow rates and MTBE influent concentrations. The primary bases for this recommendation are reliability and economics. Both recommended aeration technologies have a long track record of successful operation in drinking water situations for removal of VOCs other than MTBE. As noted, unit costs for this combination of technologies are lower than costs for competing technologies in most cases. The recommended technology combinations are listed in Table The associated costs are listed in Table Recommendations for Future Research As stated previously, air stripping is a well understood technology with many installations across the country. However, besides the packed towers at LaCrosse, Kansas and Rockaway Township, New Jersey, there appears to be a lack of published data for air stripper applications used to remove MTBE in a drinking water context. Consequently, collection of cost and operational data from a variety of air stripping sites would better demonstrate the applicability and cost-effectiveness of air strippers in MTBE treatment scenarios. Cost data should include both real capital and operational and maintenance costs. Operational data should include influent concentrations, removal efficiencies, air and water flow rates, and air concentrations. 103

126 Table 2-31 Air Stripping and Off-gas Technology Combination Recommendations Flow Rates Influent MTBE Concentration = 2000 µg/l Influent MTBE Concentration = 200 µg/l Influent MTBE Concentration = 20 µg/l Effluent Concentration (µg/l) gpm PT + GAC PT + GAC PT + GAC PT + GAC PT + GAC 600 gpm PT + TO PT + TO PT + GAC PT + GAC PT + GAC PT PT 60 gpm LP + TO LP + TO LP LP LP LP LP PT = Packed Tower; LP = Low profile; GAC = Activated Carbon Off-gas; TO = Thermal Oxidation; CO = Catalytic Oxidation; = not evaluated; ( ) = close to discharge limit, may or may not be required And the corresponding costs are: Table 2-32 Cost for Technology (Air Stripping and Off-gas Treatment) Flow Rates Influent MTBE Concentration = 2000 µg/l Influent MTBE Concentration = 200 µg/l Influent MTBE Concentration = 20 µg/l Effluent Concentration (µg/l) gpm $0.38 $0.39 $0.40 $0.13 ($0.36) $0.16 ($0.39) 600 gpm $0.90 $0.91 $0.57 $0.59 $0.62 $0.30 $ gpm $3.08 $3.20 $1.07 $1.80 $1.89 $1.66 $

127 2.8 References Advanced Environmental Systems. Budget Cost, Thermal Oxidizer Oldfield Point Road. Elkton, MD June 1, Aeromix Systems Inc. Budget Cost and Horsepower Estimates and Product Literature North Second Street, Minneapolis, MN 55411, June 4, Ball, William, Jones, Monica D., and Kavanaugh, Michael C.. Mass Transfer of Volatile Organic Compounds in Packed Tower Aeration. Journal Water Pollution Control Federation. February, 1984: Barresi, Antonello, Mazzarino, Italo, et. Al.. Gas Phase Complete Catalytic Oxidation of Aromatic Hydrocarbon Mixtures, The Canadian Journal of Chemical Engineering. April, 1992: Bethea, Robert M. Air Pollution Control Technology. Van Nostrand Reinhold. New York: Gilbert, Bill. Personal Communication. Branch Environmental Corporation, P.O. Box 5265, 3461 Route 22 East, Somerville, NJ June 5, Bohn Biofilter Corporation. P.O. Box 44235, Tucson, Arizona June 5, Branch Environmental Corporation. Budget Cost and Horsepower Estimates. P.O. Box 5265, 3461 Route 22 East, Somerville, NJ 08876, June 2, Brauer, Heinz and Varma, Yalamanchili. Air Pollution Control Equipment. Springer-Verlag. Berlin: Calgon Carbon Corporation. Budget Cost, GAC Adsorber. Pittsburgh, PA June 1, Chang, Karl, Cheng, Shan, et. al.. Removal and Destruction of Benzene, Toluene, and Xylene from Wastewater by Air Stripping and Catalytic Oxidation. Industrial Engineering and Chemistry Research, 1992: Creek, D.N. and Davidson, J.M.. The Performance and Cost of MTBE Remediation Technologies, Proceedings of the Petroleum Hydrocarbons and Organic Chemicals In Ground Water; Prevention, Detection, and Remediation Conference, November, 1998, pp Crittenden, Dave, Hand, Dave, et al.. Environmental Technologies Design Option Tools (ETDOT) for The Clean Process Advisory Systems (CPAs) Adsorption, Aeration and Physical Properties Software. National Center for Clean Industrial and Treatment Technologies

128 Dawson, David. Biological Treatment of Gaseous Emissions, Water Environment Research. 1993: Davidson, Jim. Public Drinking Water Systems Impacted By MTBE Contamination. Alpine Environmental. February 17, Dempsey, Brian A. and Ackerman, John. Removal of Volatile Contaminants from Water by Aspiration Stripping Dharmavaram, S. Biofiltration: A Lean Emissions Abatement Technology in Air Pollution Control: Equipment, Inspection and Maintenance, and Fuels. Papers from the 84th Annual Meeting of the Air and Waste Management Association. 1991: Dvorak, Bruce, Herbeck, Christopher, et. al.. Selection Among Aqueous and Off-Gas Treatment Technologies for Synthetic Organic Chemicals. Journal of Environmental Engineering. June, 1996: EPA, Granular Activated Carbon Treatment, Engineering Bulletin, EPA, OERR, Washington DC, EPA 540/2-91/024. Eweis, Juana, Schroeder, Edward, et. al.. Biodegradation of MTBE in a Pilot Scale Biofilter, Pre-print, Batelle Conference Proceedings Fang, C.S. and Khor, Sok. Reduction of Volatile Organic Compounds in Aqueous Solutions Through Air Stripping and Gas Phase Carbon Adsorption. Environmental Progress. November 1989: Fleming, J.L.. Volatilization Technologies for Removing Organics from Water., Noyles Data Corporation, Geotech. Cost and Equipment Information. Jim Butler East 40th Ave, Denver, CO. May Girod, J. and Leclerc, J.P.. Removing a Small Quantity of THT from Gas Storage Groundwater through Air Stripping and Gas Phase Carbon Adsorption. Environmental Progress. Winter 1996: Hand, David, Crittenden, John, et. al.. Design and Evaluation of an Air Stripping Tower for Removing VOCs from Groundwater. Journal American Water Works Association. September, 1986: Hazleton Environmental. Budget Cost and Horsepower Estimates and Product Literature. 125 Butler Drive, Hazelton, PA June 3, Hazelton Environmental. Personal communication with Bill Everett. 125 Butler Drive, Hazelton, PA January

129 Heck, Ronald and Farrauto, Robert. Catalytic Air Pollution Control. International Thomas Publishing Inc John M. Balla, Senior Project Engineer, Layne Christensen Company (formerly Hydro Group, Inc.), Bridgewater, NJ. Personal Conversation, June Kavanaugh, M.C. and Trussel, R.R.. Design of Aeration Towers to Strip Volatile Contaminants From Drinking Water. Journal American Water Works Association, Vol 72:12, 1980: 684. Lamarre, Bruce and Shearhouse, D.. Stripping Organics from Groundwater and Wastewater, EEW, March Layne Christensen Company. Budget Cost and Horsepower Estimates. Bridgewater, NJ, June 2, Lenzo, Frank C., Air Stripping of VOCs from Groundwater: Decontaminating Polluted Water. For the 49th Annual Conference of the Indiana Water Pollution Control Association, August 19-21, Lenzo, Frank C., Aeration for the Removal of Radon from Ground Water. Malcolm Pirnie, Inc. Work Plan for Interim Remedial Measures (IRM) for Mercury Aircraft, Inc., Dresden, NY Facility., August McKinnon, R.J. and Dyksen, J.E.. Removing Organics From Groundwater Through Aeration Plus GAC, Journal AWWA, May 1984, p Michael Wooden. Personal Communication. Malcolm Pirnie, Inc. One International Blvd. Mahwah, New Jersey Montgomery, James M. Consulting Engineers Water Treatment Principles and Design. John Wiley & Sons, Inc. New York. North East Environmental Products, Inc. Budget Cost and Horsepower Estimates. 17 Technology Drive, West Lebanon, NY 03784, June 2, Office of Science and Technology Policy (OSTP). Executive Office of the President. National Science and Technology Council. Committee on Environmental and Natural Resources. Interagency Assessment of Oxygenated Fuels. Washington, DC. June Proceedings, Innovative Technologies for Site Remediation and Hazardous Waste Management, National Conference. American Society of Civil Engineers. New York: Perry, Robert H., Green, Don W., and Maloney, James O.. Perry s Chemical Engineer s Handbook. McGraw-Hill Chemical Engineering Series, Six Edition

130 Roberts, Paul V., Hopkins, Gary, et. al.. Evaluating Two-Resistance Models for Air Stripping of Volatile Contaminants in a Countercurrent, Packed Column. Environmental Science & Technology. Vol. 19 No. 2, 1985: Salanitro, Joseph, Diaz, L.A., et. al.. Isolation of a Bacterial Culture that Degrades Methyl tert-butyl Ether Applied and Environmental Microbiology. July Salanitro, J.P., Spinnler, G.E., Neaville, C.C., Maner, P.M., Stearns, S.M., Johnson, P.C., and Bruce, C.. Demonstration of the Enhanced MTBE Bioremediation (EMB) In-Situ Process. Presented at the In-Situ and On-Site Bioremediation Conference. April 19, Schen, Thomas, Schmidt, Charles E., et al.. Assessment and Control of VOC Emissions from Waste Treatment and Disposal Facilities. Van Nostrand Reinhold, International Thomas Publishing, New York, NY, Schwarzenbach, Rene, Gschwend, Philip, et. al.. Environmental Organic Chemistry. John Wiley and Sons. New York, NY: Shearhouse, Don. North East Environmental Products, Inc., 17 Technology Drive, West Lebanon, NY May Snoeyink, V.L. and Jenkins, D., Water Chemistry. John Wiley & Sons, New York, NY, Sun, P.T. Henry s Law Constant of MTBE at Various Temperatures. Equilon Enterprises LLC, September Thermatrix, Inc. Budget Cost, Thermal Oxidizer. 101 Metro Dr. Suite 248. San Jose, CA. June 2, Tigg Corporation. Budget Cost, GAC Adsorber. Pittsburgh, PA. June 3, Van der Vaart, D. and Vatavuk, W., The Cost Estimation of Thermal and Catalytic Incinerators for the Control of VOCs. Journal of Air and Waste Management Association. April 1991:

131 3.0 Advanced Oxidation Processes Literature Review Sunil Kommineni, Ph.D. Jeffrey Zoeckler Andrew Stocking, P.E. Sun Liang, Ph.D. Amparo Flores Michael Kavanaugh, Ph.D., P.E. Technology Cost Estimates Rey Rodriguez Tom Browne, Ph.D., P.E. Ruth Roberts, Ph.D. Anthony Brown Andrew Stocking, P.E. 109

132 110

133 3.1 Background and Objectives of Evaluation Organic compounds, including MTBE, have been treated in drinking water using AOPs at several sites across the United States over the past five years. The largest system (3,000 gpm) is a medium pressure H 2 O 2 /MP-UV system installed in Salt Lake City, Utah to remove up to 10 µg/l PCE from drinking water (Crawford, 1999). Other systems are installed in Canada for the removal of N-nitroso dimethyl amine (NDMA) and several are planned for installation at Suburban Water Company (Los Angeles, CA) and La Puente (Los Angeles, CA) (Crawford, 1999). AOPs represent an alternative drinking water treatment option to air stripping (see Chapter 2), GAC adsorption (see Chapter 4), and resin sorption (see Chapter 5) processes, which may be inefficient for MTBE removal in certain cases due to MTBE s relatively low Henry s constant and high solubility. Unlike air stripping and adsorption, which are phase-transfer processes, AOPs are destructive processes. AOPs destroy MTBE and other organic contaminants directly in the water through chemical transformation, as opposed to simply transferring them from the liquid phase into a gas phase (in the case of air stripping) or solid phase (in the case of GAC and resins). However, despite this advantage, there are significant limitations and challenges in the full-scale application of AOPs. In general, AOPs are much less well understood than air stripping and sorption due to the complex chemical and physical processes involved in oxidation reactions. The implementation of AOPs and the determination of their effectiveness are difficult for several reasons. As with all treatment technologies, the effectiveness of AOPs will be largely determined by the specific water quality matrix of the contaminated water. However, in the case of AOPs, the effects of background water quality on contaminant removal are much less well understood than for other technologies. For example, the presence of high bromide concentrations or NOM can result in the formation of regulated oxidation by-products that may cause water quality to deteriorate beyond its initial state of contamination. Similarly, the presence of nitrates, NOM, and carbonates can interfere with the destruction of the target contaminant(s) and ultimately reduce the effectiveness of the selected AOP. In general, most of the technical difficulties associated with AOPs stem from the fact that oxidation processes are non-selective with the potential for significant interference. To compensate for these limitations, more energy or higher chemical dosages may be required, potentially resulting in higher costs. The primary objective of this chapter is to evaluate the feasibility of using AOPs for the removal of MTBE from drinking water. This feasibility evaluation will include a review of the chemical and physical principles behind AOPs, a discussion of the various established and emerging AOP technologies that have potential for MTBE removal, an analysis of the effects of water quality on the effectiveness of these AOPs, and a cost analysis based on information gathered from manufacturers, vendors, and actual pilot tests. Based on the findings of the review and results of the cost analysis, this chapter will conclude with overall recommendations for implementation of AOPs for MTBE removal and recommendations for future work. 111

134 112

135 3.2 Process Definition Oxidation is defined as the transfer of one or more electrons from an electron donor (reductant) to an electron acceptor (oxidant), which has a higher affinity for electrons. These electron transfers result in the chemical transformation of both the oxidant and the reductant, in some cases producing chemical species with an odd number of valence electrons. These species, known as radicals, tend to be highly unstable and, therefore, highly reactive because one of their electrons is unpaired. Oxidation reactions that produce radicals tend to be followed by additional oxidation reactions between the radical oxidants and other reactants (both organic and inorganic) until thermodynamically stable oxidation products are formed. The ability of an oxidant to initiate chemical reactions is measured in terms of its oxidation potential. The most powerful oxidants are fluorine, hydroxyl radicals ( OH), ozone, and chlorine with oxidation potentials of 2.85, 2.70, 2.07 and 1.49 electron volts, respectively (Dorfman and Adams, 1973). The end products of complete oxidation (i.e., mineralization) of organic compounds such as MTBE or benzene are carbon dioxide (CO 2 ) and water (H 2 O). AOPs involve the two stages of oxidation discussed above: 1) the formation of strong oxidants (e.g., hydroxyl radicals) and 2) the reaction of these oxidants with organic contaminants in water. However, the term advanced oxidation processes refers specifically to processes in which oxidation of organic contaminants occurs primarily through reactions with hydroxyl radicals (Glaze et al., 1987). In water treatment applications, AOPs usually refer to a specific subset of processes that involve O 3, H 2 O 2, and/or UV light. However, in this analysis, AOPs will be used to refer to a more general group of processes that also involve TiO 2 catalysis, cavitation, E-beam irradiation, and Fenton s reaction. All of these processes can produce hydroxyl radicals, which can react with and destroy a wide range of organic contaminants, including MTBE. Although a number of the processes noted above may have other mechanisms for destroying organic contaminants, in general, the effectiveness of an AOP is proportional to its ability to generate hydroxyl radicals. The various chemical and physical mechanisms through which AOP technologies produce hydroxyl radicals are discussed in Section

136 114

137 3.3 General Process Principles and Implementability Issues As with the other treatment technologies discussed in this report, the design of an AOP is governed by the influent contaminant concentration, target effluent contaminant concentration, desired flow rate, and background water quality parameters such as ph, bromide concentration, and alkalinity. The key design parameters for AOPs include: chemical dosages and ratios with other chemicals, reactor contact time, and reactor configuration. The optimum dosages, ratios, and contact time are water-specific and treatment scenario-specific, and are often determined through pilot studies using the water matrix of interest. As can be expected, higher chemical dosages and contact times are typically expected to result in higher removal rates; however, increasing dosages results in higher O&M costs and possible by-product formation. However, in some cases, the formation of by-products can be limited by higher chemical ratios. This issue will be discussed in more detail in the discussions of specific AOP technologies. While AOPs have been found to be effective for a wide variety of organic contaminants, this analysis will focus on the practical implementation of AOPs in drinking water treatment specifically for removal of MTBE Water Quality Impacts As previously mentioned, there are many water quality parameters that may impact the effectiveness of any particular AOP. For example, nearly all dissolved organic compounds present in the source water will serve to reduce the removal efficiency of the target compound by consuming OH (Hoigne, 1998). Below is a brief discussion of each of these water quality parameters and the mitigation measures that can be taken to limit the detrimental impact of these parameters on AOP effectiveness. Alkalinity. The detrimental impact of alkalinity on the effectiveness of AOPs has been extensively studied (AWWARF, 1998). As mentioned previously, the hydroxyl radical is nonselective and, thus, can be exhausted by the presence of organic or inorganic compounds other than the contaminants of concern. Both carbonate and bicarbonate will scavenge hydroxyl radicals to create carbonate radicals which, in turn, react with other organic or inorganic compounds present, albeit at a much slower rate (Hoigne and Bader, 1976; AWWARF, 1998). The reaction for the scavenging of hydroxyl radicals by bicarbonate ions is shown below (Morel and Hering, 1993): OH + HCO 3 - CO3 + H 2 O The rate constants, k, for the reactions of the hydroxyl radical with carbonate and bicarbonate are 3.8 x 10 8 and 8.5 x 10 6 M -1 s -1, respectively (Buxton et al., 1988). These rate constants are much slower than the reaction rate constant of hydroxyl radicals with MTBE (10 9 M -1 s - 1 ). However, for these second order reactions, the actual reaction rate, r, is a function of both the rate constant and the concentration of the reactant, C: r = k [C]. Waters with medium to high alkalinities (>100 mg/l as CaCO 3 ) likely contain carbonate and bicarbonate ions at 115

138 concentrations several orders of magnitude greater than MTBE and, thus, the reaction of hydroxyl radicals with carbonate and bicarbonate can proceed as fast or faster than their reaction with MTBE. Consequently, MTBE-impacted waters high in bicarbonate ions may require a lowering of alkalinity (e.g., ph adjustment or carbon dioxide stripping) prior to treatment by AOP (AWWARF, 1998) or higher doses of oxidants coupled with increased reaction time. TOC and NOM. TOC measurement incorporates all organic compounds present in the water, both dissolved organic carbon (DOC) and particulate organic carbon (POC). Drinking water supplies typically contain TOC concentrations ranging from <1 mg/l to >7 mg/l and include naturally occurring compounds and anthropogenic compounds (e.g., pesticides, gasoline components, and chlorinated compounds). NOM, a subset of TOC, is commonly used to describe large macromolecular organic compounds present in water. These macromolecules can include humic substances, proteins, lipids, carbohydrates, fecal pellets, or biological debris (Stumm and Morgan, 1996) and, while not highly reactive, often contain reactive functional groups (Hoigne, 1998). Organic matter present in the water, whether anthropogenic or natural, will scavenge hydroxyl radicals and, thus, limit the effectiveness of AOPs. The rate constants reported in the literature for hydroxyl radical reactions with NOM range from 1.9 x 10 4 to 1.3 x 10 5 (mg/l) -1 s -1 (AWWARF, 1998). Given the high molecular weight of NOM (5,000 to 10,000 g/mole), these rate constants are comparable to the reaction rate constant for MTBE and, thus, high concentrations of NOM can result in significant reduction of MTBE destruction potential. The effects of high concentrations of NOM may be mitigated by the addition of higher dosages of oxidant and longer reaction times. Nitrates and Nitrites. Hydroxyl radicals can be formed by several mechanisms, including UV photooxidation of hydrogen peroxide. Any constituent present in the water that adsorbs UV light will decrease the formation of hydroxyl radicals and the subsequent destruction of MTBE. Nitrates and nitrites adsorb UV light in the range of 230 to 240 nm and 300 to 310 nm and, consequently, high nitrate (>1 mg/l) or high nitrite (>1 mg/l) concentrations have been shown to limit the effectiveness of UV technologies (see Section for a comparative discussion of UV lamps) (Calgon, 1996; Cater, 1999). Phosphates and Sulfates. While phosphates and sulfates are commonly present in low concentrations in source waters, these compounds have the potential to scavenge hydroxyl radicals. However, they are extremely slow in reacting with OH, and their scavenging effect can usually be neglected (Hoigne, 1998) for ozone/peroxide/uv systems. For TiO 2 systems, sulfates have been noted to significantly decrease the destruction rate of organic contaminants at concentrations above approximately 100 mg/l (Crittenden et al., 1996). Iron (II), Copper (I), or Manganese (II). While not well understood, the presence of these reduced metals in combination with NOM and hydroxyl radicals may lead to the formation of iron or copper organic complexes or the oxidation of Mn (II) to form permanganate (Hoigne, 1998; Calgon, 1996). The formation of permanganate has been observed to occur 116

139 with a rate constant of 3 x 10 3 to 2 x 10 4 M -1 s -1 (Reckhow et al., 1991). In addition, the presence of iron (absorptivity 200 to 400 nm) and other scaling agents may result in fouling of UV systems. Turbidity. Systems relying on UV irradiation for the dissociation of H 2 O 2 or O 3 exhibit a decrease in efficiency as turbidity increases. Turbidity lowers the transmittance of the source water and, thus, lowers the penetration of the UV radiation into the source water. In addition to possible interference with AOPs by the compounds described above, the recent emergence of UV irradiation as a disinfection technology has prompted the investigation of possible negative side effects of UV irradiation on drinking water containing NOM. In particular, there have been concerns regarding the potential effects of UV irradiation on by-product formation when UV is used in conjunction with chlorine addition. Chlorine is sometimes added at plant headworks as a pre-treatment step or, more likely, used as a primary or secondary disinfectant farther down the treatment train. The use of chlorine is associated with the formation of by-products, such as THMs and haloacetic acids (HAA 9 : HAA 5 plus tribromo-, bromochloro-, bromodichloro- and dibromochloroacetic acid), from NOM normally present in water. These by-products are suspected human carcinogens and are regulated by the EPA. A number of researchers have investigated the potential of UV irradiation to change the composition or distribution of these by-products by promoting the photolysis of NOM (e.g., humic acids) into smaller molecules that have higher potential for THM and HAA 9 formation (e.g., Stewart et al., 1993; Hengesbach et al., 1993). Zheng et al. (1999a) recently showed that UV irradiation at a dosage of 100 mj/cm 2 (the expected maximum UV dosage used for water disinfection applications) of water prechlorinated at 6 to 48 mg/l can result in a one to seven percent decrease in the THM concentration and a change in the HAA 9 concentration ranging from -3 to 4 percent. These results suggest that under typical treatment conditions, where pre-chlorination is usually less than 5 mg/l, there will be negligible changes in THM and HAA 9 formation. It was suggested that UV irradiation might have caused a fraction of the Cl 2 residual to decompose into chlorine radicals (Cl ), which then reacted with THM precursors and converted them into HAA 9 precursors. In a parallel study, Zheng et al. (1999b) also studied the effect of UV irradiation prior to chlorine addition on the subsequent formation of THMs and HAA 9 s after chlorine is added. The study found that UV irradiation (dosage range from 0 to 3,000 mj/cm 2 ) had insignificant effects on the formation of THMs after the addition of chlorine at a dosage of 5 to 10.5 mg/l. Similar results were observed for HAA 9 formation. Thus, it could be inferred that the use of UV, especially at doses less than 100 mj/cm 2, either prior to or after a chlorination process, has a minor impact on THM and HAA 9 formation. Zheng et al. (1999b) investigated the effects of pre-chlorination combined with H 2 O 2 /UV on THM and HAA 9 formation. Upon exposure of drinking water to a UV dose of 100 mj/cm 2 and H 2 O 2 dosages ranging from 3.6 to 51 mg/l, THM concentrations increased between two to six percent compared to values observed without H 2 O 2 /UV treatment. THM and HAA 9 117

140 formation generally increased linearly for UV doses up to 2,000 mj/cm 2. At a UV dosage of 2,000 mj/cm 2, the THM concentrations increased between 37 percent and 146 percent and HAA 9 concentration increased between 39 percent and 128 percent. UV doses for MTBE removal applications are expected to be greater than 2,000 mj/cm 2 and, thus, increases in THM and HAA 9 formation may be significant. Consequently, these preliminary results suggest that the combined effects of H 2 O 2 and UV irradiation on THM and HAA 9 formation potentials should be considered in drinking water situations General Advantages The specific advantages of each AOP will be discussed later in this chapter. The following list describes the advantages that are common to all AOPs. MTBE Destruction AOPs represent an alternative drinking water treatment option to air stripping, GAC adsorption, and resin sorption. Air stripping and sorption are phase-transfer processes in which organic contaminants like MTBE are physically transferred to a gas phase (air stripping) or solid phase (GAC and resins). Actual destruction of MTBE requires additional processes, such as thermal or catalytic oxidation of the off-gas from an air stripper or incineration of the GAC. In contrast, AOPs destroy primary organic contaminants directly in water through chemical reactions. Furthermore, the effectiveness of AOPs is facilitated by the relatively high solubility of MTBE while air stripping and sorption onto GAC and resins are limited by MTBE s high solubility. Disinfection Capability Several AOP technologies namely ozonation, ozonation combined with H 2 O 2, and certain types of UV irradiation are currently used for disinfection purposes in the water treatment industry. Currently, ozonation and MP-UV irradiation are the only state or federally approved disinfection technologies; (Reynolds and Richards, 1996; EPA/600R-98/160VS, 1999). Disinfection credit can be given for peroxide/ozone systems depending on the ozone residual remaining in the effluent water; this residual will decrease as the peroxide to ozone ratio increases. The EPA and NSF International verified the performance of a MP-UV system (Sentinel, Calgon Carbon, Markham, Ontario) under the Environmental Technology Verification (ETV) program and certified it for 3.9 log 10 removal of Cryptosporidium (EPA/600/R-98/160VS, 1999; Bukhari et al., 1999). Established Technologies for Drinking Water Treatment Although the use of AOPs for organic contaminant removal from drinking water has been limited in the past, many of the components of AOPs have been used by the water community and industry (e.g., UV and ozone for disinfection). Consequently, treatment plant operators 118

141 may already be familiar with operation and implementation of these established AOP components, suggesting that the implementation of some AOPs will be feasible with minimal training General Disadvantages Specific disadvantages of each AOP will be discussed later in this chapter. The following list describes some common disadvantages of AOP technologies. Oxidation By-products The reaction between ΟΗ and many organic contaminants occurs rapidly; however, this reaction by itself does not directly result in the mineralization of these contaminants but produces organic oxidation by-products, which can further react with OH. There are at least two proposed mechanisms in the literature for the complete oxidation of MTBE to carbon dioxide and water (Barreto et al., 1995; Kang and Hoffmann, 1998). If the reaction rate for a particular by-product is slower, it may be the rate limiting step in the complete mineralization of the target compound and that by-product will accumulate. Ideally, AOP systems are designed to completely mineralize the organic contaminants of concern to CO2 and H2O, but this may require more energy and greater chemical dosages and, ultimately, may prove to be cost prohibitive in certain applications. While highly dependent on the specific water quality, Table 3-1 shows experimentally determined hydroxyl radical rate constants for MTBE and its various oxidation by-products (water quality varies, as discussed in the literature). A possible mechanism for the oxidation of MTBE involves either direct conversion to TBA or oxidation of the methyl group to form tertiary-butyl formate (TBF). TBF can then be hydrolyzed to TBA and formaldehyde. TBA can lose a methyl group to form isopropyl alcohol, which can be further oxidized to acetone. In some AOPs, further oxidation of these by-products was found to be the rate-limiting step in the ultimate mineralization of MTBE. Finally, acetone may be converted to formaldehyde or formate, which can be mineralized to CO2 and H2O. As stated above, economic considerations can limit the complete mineralization of MTBE, leaving short-chained carboxylic acids, alcohols, aldehydes, and/or acetone in the effluent water (Karpel Vel Leitner et al., 1994; Liang et al., 1999a). These residual oxidation compounds are also produced by the partial oxidation of NOM (Hoigne, 1998). These compounds represent a source of concern in drinking water applications due to their high solubility and uncertain toxicity. In addition, the presence of these more easily degradable compounds can promote biological growth and fouling in the distribution system. Thus, these compounds are often further treated with a biological activated filter or some other polishing treatment step. 119

142 Table 3-1 Hydroxyl Radical Rate Constants, k, for MTBE and Its By-products Bromate Formation In addition to organic oxidation by-products, inorganic by-products can also be formed. Bromide concentrations in raw drinking water can range from non-detect to several hundred µg/l to several mg/l in brackish water (Hoigne, 1998). During the use of AOPs that employ ozone, bromate can be formed when bromide is present in the source water (Krasner, 1993; von Gunten and Hoigne, 1994; Siddiqui et al., 1999). The conversion of bromide to bromate can proceed as follows (Krasner, 1993; Siddiqui et al., 1994): Br - BrO - BrO - 2 BrO3 - Br - Br BrO BrO - 2 BrO2 BrO - 3 When bromide concentrations in the source water exceed 0.1 mg/l, bromate has been found to be produced at concentrations above the Stage 1 Disinfection/Disinfection By-product (D/DBP) MCL of 10 µg/l (Liang et al., 1999a). However, prior studies of ozone-based AOPs, specifically H 2 O 2 /O 3 systems, have shown that varying the chemical ratio of O 3 to H 2 O 2 is effective at minimizing bromate formation (Song et al., 1997). In addition, research has demonstrated that bromate formation is reduced by approximately 20 percent at a slightly acidic ph (~6.5), when compared to the ambient ph (Liang et al., 1999a). Interfering compounds As mentioned previously, alkalinity and/or the presence of TOC (including NOM), nitrites, and other inorganics can limit the effectiveness of AOPs due to their scavenging of hydroxyl radicals that would otherwise be used to destroy MTBE. Thus, as the concentrations of these 120

143 parameters increase, chemical dosages and contact times will also increase to maintain effluent concentrations at the treatment goal. Increasing chemical dosages will raise operational costs while increasing contact times will riase initial capital costs. The sensitivity of operational and capital costs to changing source water qualities is discussed in more detail in Section AOP Technologies AOPs can be divided into established and emerging technologies based on the existing literature and the water treatment industry s experience with the technology. Emerging technologies are defined here as technologies that have very limited, if any, full-scale applications in drinking water treatment. The following AOP technologies are discussed in this report: Established Technologies Hydrogen Peroxide/Ozone (H2O2/O3) Ozone/Ultraviolet Irradiation (O3/UV) Hydrogen Peroxide/ Ultraviolet Irradiation (H2O2/UV) Emerging Technologies High Energy Electron Beam Irradiation (E-beam) Cavitation (Sonication & Hydrodynamic) TiO2-catalyzed UV Oxidation Fenton s Reaction Each of the above AOP technologies is evaluated in Section 3.4 on the basis of its performance reported in the engineering literature, results of manufacturer or vendor studies, and the industry s experience with the technology. The following sections include detailed discussions of each technology s chemistry, advantages and disadvantages, key variables and design parameters, and available performance data from bench-, pilot-, and field-scale tests. Tables 3-2 to 3-5 present summaries of these discussions. A summary including brief descriptions, system components, and advantages and disadvantages of established and emerging AOP technologies is presented in Table 3-2. Table 3-3 lists the reactions, by-products, interfering compounds, and oxidant hierarchy associated with the various AOP technologies. Table 3-4 presents a summary of pilot and field studies and Table 3-5 lists vendor information for each AOP technology. 121

144 Table 3-2 Brief Descriptions, System Components, Advantages, and Disadvantages of Established and Emerging AOP Technologies 122

145 Table 3-2 (Continued) Brief Descriptions, System Components, Advantages, and Disadvantages of Established and Emerging AOP Technologies 123

146 Table 3-2 (Continued) Brief Descriptions, System Components, Advantages, and Disadvantages of Established and Emerging AOP Technologies / 124

147 Table 3-3 Reactions, By-products, Interfering Compounds, and Oxidant Hierarchy Note: hυ represents UV radiation. 125

148 Table 3-3 (Continued) Reactions, By-products, Interfering Compounds, and Oxidant Hierarchy Note: hυ represents UV radiation. 126

149 Table 3-4 Summary of AOP Pilot and Field Studies 127

150 Table 3-4 (Continued) Summary of AOP Pilot and Field Studies 128

151 Table 3-4 (Continued) Summary of AOP Pilot and Field Studies 129

152 Table 3-5 Summary of Vendor Information 130

153 Table 3-5 (Continued) Summary of Vendor Information 131

154 132

155 3.4 Established Technologies Hydrogen Peroxide/Ozone (H 2 O 2 /O 3 ) Process Description When O 3 is added to water, it participates in a complex chain of reactions that result in the formation of radicals such as the hydroxyl radical ( OH) and the superoxide radical (O 2 ) (Hoigne, 1998). Like O 3, these radical products ( OH and O 2 ) are oxidants capable of MTBE destruction. Of the radical intermediates formed in ozonated water, OH is the most powerful MTBE oxidant, even more powerful than O 3 itself. Direct oxidation of ethers by O 3 is known to occur very slowly; this reaction s second-order kinetic rate constant is less than 1 M -1 s -1 (Buxton et al., 1988). By contrast, oxidation of ethers by radical oxidants is extremely rapid. Hydroxyl radicals react with MTBE according to a rate constant of 1.6 x 10 9 M -1 s -1 (Buxton et al., 1988). H 2 O 2 can be combined with ozone to enhance the transformation of O 3 to OH in solution. H 2 O 2 is a weak acid, which partially dissociates into the hydroperoxide ion (HO 2 - ) in water. H 2 O 2 reacts slowly with O 3, whereas the HO 2 - ion can rapidly react with O3 to form OH (Hoigne, 1998): H 2 O 2 + H 2 O HO H3 O + O 3 + HO - 2 OH + O2 - + O2 These reactions and possible by-products are summarized in Table 3-3. Also listed in this table are the interfering compounds and oxidant hierarchy. System Description/Design Parameters In an H 2 O 2 /O 3 system, H 2 O 2 is used in conjunction with O 3 to enhance the formation of hydroxyl radicals. Since O 3 decomposes rapidly, it is typically produced on-site using a generator fed with dried compressed air or oxygen (Hoigne, 1998). The gas mixtures produced from air and oxygen by an ozone generator usually consist of 0.5 to 1.5 percent and 1 to 2 percent by volume O 3, respectively (Montgomery, 1985). The use of air to generate ozone requires dehumidification, which may be cost prohibitive relative to the use of pure liquid oxygen. In addition, larger quantities of ozone can be produced from a unit of liquid oxygen (14 percent O 2 by weight) compared to a unit of compressed air (2 percent O 2 by weight), which facilitates greater mass transfer of the ozone into the source water. Finally, ozone can be generated from liquid oxygen using less energy relative to compressed air. For AOPs, O 3 gas is fed through spargers, porous piping or plates, or Venturi-type injectors at dosages equivalent to 1 to 2 mg/l ozone per mg/l DOC; however, higher dosages are recommended for source waters with high alkalinity (>100 mg/l as CaCO 3 ) or NOM 133

156 (Hoigne, 1998). O 3 transfer efficiencies from the gas to the water of up to 90 to 95 percent can be achieved (Montgomery, 1985). H 2 O 2 is fed from an aqueous solution, at peroxide to ozone ratios ranging from 0.3:1 to 3:1 (Liang, 1999a, b; Applebury, 1999). The specific ratio will be a function of disinfection requirements, bromide concentration, contaminant concentration, and other water quality parameters. Since ozone residual can provide disinfection credit, a lower peroxide to ozone ratio is typically applied to source waters requiring disinfection (e.g., surface waters) in order to leave some ozone residual. However, researchers have shown that bromate formation is a strong function of the H 2 O 2 to O 3 ratio, and that H 2 O 2 to O 3 ratios can effectively reduce the concentration of bromate generated (Siddiqui et al., 1994; Liang et al., 1999a and 1999b; White, 1999). These counter-acting effects should be considered when trying to determine the optimal peroxide to ozone dosage ratio to apply for a specific water source with significant influent bromide concentrations (>0.1 mg/l). For source waters requiring minimal disinfection (e.g., some groundwaters), a higher peroxide to ozone ratio can be applied to minimize bromate formation. In the case of waters requiring disinfection, alternative bromate formation mitigation measures may be required if a high peroxide to ozone ratio is used for disinfection credit or an alternative disinfectant (e.g., Cl 2 ) may be used to fully meet disinfection standards. Two types of ozone contact configurations exist for application: 1) conventional 3 to 5 meter deep continuously stirred reactor basins, and 2) long (>100 feet) plug flow reactors. In a conventional ozone reactor, ozone is bubbled through the base of the reactor and allowed to diffuse through the reactor until it either escapes through the top or is completely reacted. This results in high ozone concentrations at the base of the reactor, independent of the contaminant concentrations, which promote the reaction of ozone with other chemical constituents to form regulated by-products (e.g., bromate). These reactors are typically covered so that excess O 3 can be collected and directed to an off-gas decomposer. Automatic monitoring and control systems are used to regulate chemical feed rates, ph, and other parameters. In addition, a variety of safety, monitoring, and control systems are included to facilitate operation. A schematic of a conventional H 2 O 2 /O 3 system equipped with UV lamps is shown in Figure 3-1a. 134

157 TREATED EFFLUENT STORAGE Figure 3-1a. A schematic of a conventional (continuously stirred tank reactor) H 2 O 2 /O 3 system equipped with UV lamps (drawing provided by Komex H2O Science, 1998). The second type of H 2 O 2 /O 3 contact system, referred to as HiPOx, has been commercialized by Applied Process Technology, Inc. (APT) (San Francisco, CA). HiPOx is a continuous, inline plug flow reactor where O 3 and H 2 O 2 are injected into the water stream in precisely controlled ratios at multiple ports along the flow reactor (see Figure 3-1b). The primary advantage of this system is that high dosages can be applied at the beginning of the flow reactor, where contaminant concentrations are high. As contaminant concentrations are reduced along the line, decreasing dosages can be applied, thereby controlling formation of regulated by-products (e.g., bromate). Using multiple injection ports, the concentration of molecular ozone in solution can be maintained at a lower concentration, typically below 0.5 mg/l, than in a large continuously stirred reactor. This keeps the H 2 O 2 to O 3 ratio high 135

158 which, in turn, increases the rate of molecular ozone being converted to the hydroxyl radical and increases the rate of hypobromite reduction to bromide (Staehelin and Hoigne, 1982; von Gunten and Oliveras, 1997). As a result, the HiPOx system has been shown to effectively reduce bromate formation, even under high concentrations of influent bromide (1,000 µg/l), by minimizing the molecular ozone available to oxidize the bromide to hypobromite and having excess hydrogen peroxide available to reduce any hypobromite produced (Applebury, 1999). In addition, this system can be operated without the loss of pressure experienced by bringing the source water in contact with the atmosphere, thereby reduce pumping costs. Figure 3-1b. A schematic of a plug flow H 2 O 2 /O 3 system manufactured by Applied Process Technology, Inc. (Applebury, 1999). The major components of both a continuously stirred tank reactor and a plug flow reactor include: A H 2 O 2 storage tank A H 2 O 2 injection system An ozone generator Liquid oxygen or compressed air tank 136

159 Ozone diffusers Ozone contactor An ozone off-gas decomposer Supply and discharge pumps and piping Monitoring and control systems Advantages and Disadvantages The advantages and disadvantages of the H 2 O 2 /O 3 system are briefly summarized in Table 3-2. The benefits of using an H 2 O 2 /O 3 system are: The combined H 2 O 2 /O 3 process has been demonstrated to be more effective at removing MTBE and other natural and synthetic organics than O 3 or H 2 O 2 alone. In addition, using a combination of O 3 and H 2 O 2 to produce hydroxyl radicals, rather than just O 3, allows a lower dosage of O 3 to be used. This is desirable for reducing costs and bromate formation potential. The theoretical yield of hydroxyl radicals via H 2 O 2 /O 3 technology is less than that of the H 2 O 2 /UV technology; however, the yield is less affected by water quality (i.e., turbidity, iron, and nitrates lower the yield for UV processes but not H 2 O 2 /O 3 processes). Once the hydroxyl radicals are formed, however, the chemical destruction and interferences are the same for both technologies. According to this literature review, H 2 O 2 /O 3 systems appear to be the most tested and applied AOP in remediation applications for groundwaters, relative to the other AOPs. Thus, the implementation of H 2 O 2 /O 3 systems has a field-proven history of operation and regulatory acceptance. The disadvantages and limitations of the H 2 O 2 /O 3 system are: The use of O 3 can result in the potential formation of bromate; however, bromate formation can be mitigated by lowering the ph to <6.5, increasing the H 2 O 2 to O 3 ratio, or adding another radical scavenger that will react with hydroxyl radicals prior to the bromide (e.g., ammonia). The H 2 O 2 /O 3 process typically requires an air permit for ozone emissions in addition to an off-gas treatment system for ozone destruction. The hydrogen peroxide reacts rapidly with most of the applied ozone and, thus, the air exiting the contactor has been observed to typically contain ozone concentrations less than 1 mg/l (Applebury, 1999). This concentration is significantly higher than the 1-hour Clean Air Act standard of 0.12 ppmv (CFR Title 40, Part 50). Current methods for removal of ozone in the off-gas include thermal destruction, catalytic reduction, or a combination of the two (Horst, 1982; White, 1999). 137

160 Thermal destruction takes advantage of the fact that ozone decomposes rapidly at high temperatures. Catalytic reduction involves passing the ozone off-gas across a surface (commonly iron or manganese oxide) that catalyzes the decomposition of ozone to elemental oxygen. These controls will add to the operational and capital cost of the system (AWWA/ASCE, 1997). Residual H 2 O 2 can serve as an oxygen source for microorganisms and can promote biological re-growth in the distribution system. Although there are currently no federal or state standards for residual H 2 O 2 in treated drinking water, drinking water purveyors are not likely to allow any detectable levels of H 2 O 2 in treated drinking water (detection limits range from 1 µg/l to 100 µg/l depending on the method and concentration) because of concerns over biological growth. Thus, depending on the effluent concentration, posttreatment of excess H 2 O 2 may be required to limit downstream biological fouling. However, if residual H 2 O 2 concentrations are limited to less than a few mg/l, treatment systems already in place for the removal of oxidation by-products from the H 2 O 2 /O 3 system effluent will also remove the residual H 2 O 2. In cases where residual H 2 O 2 generally exceeds a few mg/l, a treatment system specifically for H 2 O 2 removal (e.g., catalytic activated carbon) will need to be employed (Crawford, 1999). Bench-scale Studies Karpel Vel Leitner et al. (1994) studied the reaction of ozone combined with hydrogen peroxide on gasoline additives such as MTBE in a dilute aqueous solution. Experiments conducted in a semi-continuous reactor with MTBE showed that the use of H 2 O 2 /O 3 is a more effective treatment process than ozone alone. Applied dosages of 3.0 and 1.7 mg ozone per mg of MTBE were found to result in 80 percent reduction of MTBE under ozone alone (at ph 8) and under H 2 O 2 /O 3, respectively. TBA, TBF, and formaldehyde were identified as the by-products of the H 2 O 2 /O 3 -MTBE reactions. Pilot/Field Studies and Vendor Information Dyksen et al. (1992) performed pilot tests using in-line application of ozone and hydrogen peroxide to evaluate process issues for the removal of organic chemicals such as TCE, PCE, cis-1,2 dichloroethylene (cis-1,2 DCE), and MTBE. The results indicated that H 2 O 2 /O 3 is more effective than ozone alone for removal of TCE, PCE, cis-1,2 DCE, and MTBE. Nondetectable levels of MTBE were recorded using an ozone dosage of 8 mg/l, a contact time of 3 to 6 minutes, and a hydrogen peroxide to ozone ratio of 0.5. Liang et al. (1999a) conducted a pilot-scale study to investigate the effectiveness of ozone and H 2 O 2 /O 3 processes for MTBE removal in surface water. Using two treatment trains with a total flow capacity of 12 gpm, H 2 O 2 /O 3 was found to be more effective than ozone alone for MTBE removal under their tested conditions. The results indicated that 4 mg/l of ozone 138

161 and 1.3 mg/l of hydrogen peroxide can achieve average MTBE removals of approximately 78 percent for both source water supplies tested. Liang et al (1999b) also investigated the removal of MTBE from contaminated groundwater through the use of ozone and H 2 O 2 /O 3. Experiments conducted in a large-scale semi-batch reactor again demonstrated that H 2 O 2 /O 3 (at a H 2 O 2 to O 3 ratio of 1.0) was consistently more effective in oxidizing MTBE than ozone alone, even at ozone doses as high as 10 mg/l. Applied ozone doses greater than 10 mg/l were necessary to reduce MTBE concentrations from approximately 200 and 2,000 µg/l to concentrations below the California secondary drinking water standard of 5 µg/l. However, at this dosage, both ozone and H 2 O 2 /O 3 were also found to completely oxidize MTBE oxidation by-products, such as TBF and TBA. Most vendors who provide ozone technologies can also provide H 2 O 2 /O 3 systems by adding H 2 O 2 injection systems to their oxidation reactors. There are several vendors who currently exclusively provide H 2 O 2 /O 3 technology, but few have applied their process for MTBE removal (see Table 3-4). APT (San Francisco, CA) has performed several pilot/field-scale studies of their patented H 2 O 2 /O 3 system, HiPOx. As mentioned previously, unlike more conventional H 2 O 2 /O 3 systems, the HiPOx system has multiple oxidant injection ports and the reaction is carried out under pressure. Three of the APT studies involved MTBE removal applications (see Table 3-4). In one APT study involving highly brominated (bromide >1,000 µg/l) coastal water, a 10-gpm HiPOx system was able to reduce the MTBE concentration from 1,000 µg/l to 1 µg/l while maintaining the bromate concentration at less than 10 µg/l (Waters, 1999). TBA concentrations in the effluent were measured at approximately 60 µg/l. In a second field study, a 10-gpm HiPOx system was able to reduce MTBE (in a solution containing a mixture of BTEX compounds) from 33,000 µg/l to <5 µg/l MTBE. The third study (0.25 gpm) showed reduction of 660,000 µg/l MTBE to 2.3 µg/l. TBA and by-product concentrations were not measured for the latter two studies. Hydroxyl Systems (HSI) (Sidney, British Columbia, Canada) is currently conducting a remediation field study using H 2 O 2 /O 3 at the JFK Airport in New York. The objective of the study is to reduce MTBE from an initial concentration of 100 to 300 mg/l to a final effluent goal of 50 µg/l (local action level), at a flow rate of 20 to 60 gpm (Harp, 1999). Phase 1 of the field study is scheduled to commence in early The flow rates at this treatment facility will be gradually increased from 20 gpm to 60 gpm (Harp, 1999). U.S. Filter (Santa Clara, CA) also has several H 2 O 2 /O 3 installations across the nation but currently has no installations designed specifically for MTBE removal applications (Himebaugh, 1999; Woodling, 1999). Refer to Tables 3-4 and 3-5 for a summary of these case studies and vendor information, respectively. Summary H 2 O 2 /O 3 systems have been well studied at the bench-, pilot-, and field-scale levels for the removal of organic contaminants such as BTEX, TCE, and PCE. There are currently a 139

162 number of full-scale H 2 O 2 /O 3 systems in use for MTBE remediation (Table 3-4); however, use of this technology for drinking water applications has only been performed at the pilot scale. While concerns have been raised about the formation of bromate with these systems, this concern can be mitigated by increasing the peroxide to ozone ratio, decreasing the ph, or raising the concentration of other radical scavengers. The chemistry behind H 2 O 2 /O 3 systems is well understood; however, as with all AOPs, more pilot- and field-scale demonstration sites under a variety of water quality matrices are needed prior to general regulatory acceptance UV Systems UV light is in the high-energy end of the light spectrum with wavelengths less than that of visible light (400 nm) but greater than that of x-rays (100 nm). UV radiation (hυ) can destroy organic contaminants, including MTBE, through direct and indirect photolysis (Zepp, 1988). In direct photolysis, the absorption of UV light by MTBE places it in an electronically excited state, causing it to react with other compounds, and eventually degrade. In contrast, indirect photolysis of MTBE is mediated by hydroxyl radicals that are produced when ozone or peroxide is added to the source water either prior to or during UV irradiation. The most common sources of UV light are continuous wave low pressure mercury vapor lamps (LP-UV), continuous wave medium pressure mercury vapor lamps (MP-UV), and pulsed-uv (P-UV) xenon arc lamps. Both LP-UV and MP-UV mercury vapor lamps produce a series of line outputs, whereas the xenon arc lamp produces a continuous output spectra. The characteristics of typical LP-, MP-, and pulsed-uv lamps are presented in Table 3-6. Table 3-6 Characteristics of Typical Low Pressure (LP), Medium Pressure (MP), and Pulsed UV (P-UV) Lamps 140

163 For most traditional applications of UV irradiation with H 2 O 2 or O 3, LP-UV and MP-UV have been used; however, MP-UV is receiving increasing attention because of its greater potential for direct photolysis. In addition, MP-UV lamps radiate over a wider range of wavelengths (200 to 400 nm) than LP-UV lamps, which better facilitates the formation of hydroxyl radicals when hydrogen peroxide is present; hydrogen peroxide absorbs more in the higher wavelengths (250 to 300 nm). Furthermore, while an LP-UV lamp is more electrically efficient than an MP-UV lamp, the latter produces a greater UV output per lamp. Thus, MP-UV systems can be expected to use fewer lamps, take up less space, and require less maintenance. Finally, after extensive ETV testing, the EPA has recently credited Calgon Carbon s (Markham, Ontario, Canada) MP-UV lamp system (Sentinel ) with 3.9 log 10 inactivation for Cryptosporidium parvum (EPA/600/R-98/160VS, 1999; Bukhari et al., Calgon Carbon (Markham, Ontario) has decided not to use P-UV lamps due to their short lifetimes and minor observed benefits relative to MP-UV (Crawford, 1999). To describe the removal efficiency for organic contaminants using UV lamps, Calgon Carbon (Markham, Ontario, Canada) has defined the term Electrical Energy per Order of Removal (EE/O) as the kilowatt-hours (kwh) of electricity required to reduce the concentration of a compound (e.g., MTBE) in 1,000 gallons by one order of magnitude (or 90 percent) (Calgon AOT Handbook, 1996). The unit for EE/O is kwh/1,000 gal/order of removal and is defined at the optimum H 2 O 2 or O 3 concentration. According to Calgon Carbon, the EE/O provides a convenient way to compare the effectiveness of removal of various organic compounds, using UV irradiation for a single source water (i.e., the EE/O will change depending on the water quality). The higher the EE/O value of a contaminant, the more difficult and/or more costly it is to remove that contaminant relative to those with lower EE/O values. For example, MTBE is more difficult to treat than BTEX, with EE/O values of around 10 for MTBE and 2 to 5 for benzene. An EE/O of 10 for MTBE means that it would take ~10 kwh to reduce MTBE from 600 µg/l to 60 µg/l in 1,000 gal of water. It will take another 10 kwh to reduce the MTBE from 60 µg/l to 6 µg/l, and so on (Calgon AOT Handbook, 1996). Ozone/UV (O 3 /UV) Process Description Due to the relatively high molar extinction coefficient of ozone, LP-UV or MP-UV radiation can be applied to ozonated water to form highly reactive hydroxyl radicals (Wagler and Malley, 1994). The use of UV irradiation whether MP-UV, LP-UV, or P-UV to produce hydroxyl radicals with ozone occurs by the following reaction: hυ O 3 /UV process: O 3 + H 2 O O 2 + H 2 O 2 (λ <300 nm) 2 O 3 + H 2 O 2 2 OH + 3 O 2 As the above reactions illustrate, photolysis of ozone generates hydrogen peroxide and, thus, O 3 /UV involves all of the organic destruction mechanisms present in H 2 O 2 /O 3 and H 2 O 2 /UV 141

164 AOPs (Table 3-3). These mechanisms include direct reaction with ozone, direct photolysis by UV irradiation, or reaction with hydroxyl radicals (Calgon AOT Handbook, 1998). In most past applications of O 3 /UV, LP-UV lamps have been used (AWWA, 1990; Calgon AOT Handbook, 1996); however, MP-UV and P-UV are receiving increased attention due to their disinfection capabilities and direct photolysis benefits. System Description/Design Parameters Two basic UV reactor design configurations are used for the removal of organic contaminants from water. Calgon Carbon, Inc. (Markham, Ontario, Canada) currently uses both reactor designs for MTBE removal, depending on the flow rate (Crawford, 1999). For largescale drinking water applications (>500 gpm), a tower design is typically utilized. In the tower configuration, multiple UV lamps are arranged horizontally within a single large reactor vessel with the contaminated water flowing perpendicularly past the UV lamps. For example, a tower system may consist of kW UV lamps arranged horizontally throughout the tower. Heat transfer for MP-UV lamps is typically <1 C for every 4 kwh/1,000 gallons. Therefore, no cooling systems are needed for the large-scale tower configuration. For small-scale systems (<500 gpm), Calgon Carbon (Markham, Ontario, Canada) employs reactors where a single UV lamp per reactor vessel is arranged vertically. For example, a small-scale system may consist of three individual reactor vessels in series, each containing one 30-kW UV lamp in a vertical position. For very small systems (<50 gpm), these higher watt lamps operate at a higher temperature and, thus, require a cooling fan to effect heat transfer (Crawford, 1999). Safety interlocks are provided on Calgon UV reactors to protect personnel from both the UV radiation and high voltage supply (Calgon AOT Handbook, 1996). U.S. Filter (Santa Clara, CA) markets a LP-UV oxidation system, referred to as Ultrox, which can use either ozone, peroxide, or a combination of both as supplemental oxidants (Gruber, 1994). A typical Ultrox system can consist of a combination of the following four components: 1) a stainless steel reaction chamber with LP-UV lamps; 2) an air compressor/ozone generator; 3) a hydrogen peroxide feed system; and 4) a catalytic ozone decomposition unit. As a first step in the treatment process, the contaminated source water is mixed with peroxide and then fed into the reaction chamber where ozone is added, if necessary. The reaction chamber ranges in size from 325 to 3,900 gallons and is divided into a series of parallel sub-chambers, each housing a bank of LP-UV mercury vapor lamps (Gruber, 1994). As the water flows through each sub-chamber, it passes in front of each bank of UV lamps (the number of sub-chambers and the number of lamps depend on the size of the system and type of contaminant being destroyed). The Ultrox system employs lowintensity UV lamps; hence, the surface temperatures of the quartz sheath surrounding each lamp rarely exceeds 90 F (Gruber, 1994). 142

165 For O 3 /UV applications, ozone is introduced into the system at the bottom of each chamber by a stainless steel sparger. The ozone generator employed in the Ultrox system can electrically generate ozone from either air or liquid oxygen. Any ozone that is present in the off-gas is put through a fixed bed catalytic scavenger. This ozone decomposition unit operates at 150 F and uses a proprietary nickel-based catalyst to convert ozone to oxygen (Gruber, 1994). Ultrox systems can operate from flow ranges of 5 gpm to 1,200 gpm. Higher flowrates are attainable with multiple treatment trains (Himebaugh, 1999). To minimize problems associated with potential fouling of the UV lamp sleeves in cases where the influent water has high concentrations of fouling agents (e.g., iron, calcium, and magnesium), UV systems are equipped with automated cleaning devices. Quartz sleeves that separate the water from the UV lamps are periodically cleaned by pneumatically driven wipers. Quartz sleeve cleaning devices are common in UV oxidation technologies, and the costs are generally included in the total costs of the system (Crawford, 1999). The frequency of UV lamp cleaning is a function of the presence of iron and other scaling agents in the water. However, the wiping mechanisms used today are well designed and allow for troublefree operation for source waters containing concentrations of iron and other fouling agents. The two primary design variables that must be optimized in sizing a UV AOP system are the UV power radiation per unit volume of water treated more commonly referred to as UV dose and the concentration of hydrogen peroxide or ozone. UV dose, when applied to AOP, is a measure of the total lamp electrical energy applied to a fixed volume of water. The units are measured in kwh/1,000 gallons treated. This parameter combines flowrate, residence time and light intensity into a single term. The dose of UV light and peroxide/ozone required per unit volume of water treated may vary from one type of water to another. For a flowthrough system, the UV dose (kwh/1,000 gal) is given by: UV Dose = 1,000 x lamp power flow (gpm) x 60 Design tests are typically performed to measure the UV dosage required to achieve the desired effluent concentration. The dosage to be applied is determined in an iterative manner by examining the effect on treatment of selected process variables such as ph, oxidant concentration and retention time. The major components of an O 3 /UV system include: UV lamps, lamp sleeves, and lamp cleaning system Ozone generator and diffusers Ozone contactor Ozone off-gas decomposer 143

166 Liquid oxygen or compressed air tank Supply and discharge pumps and piping Monitoring and control systems Figure 3-1a previously showed a schematic of a conventional H 2 O 2 /O 3 system equipped with UV lamps. An O 3 /UV system is similar, with the exception of the H 2 O 2 feed system. Advantages and Disadvantages The advantages and disadvantages of the O 3 /UV system are briefly summarized in Table 3-2. The benefits of using the O 3 /UV system are: The removal efficiency of the combined O 3 /UV process is typically higher than the additive removal efficiencies of ozone and UV alone (Prado and Esplugas, 1999). The magnitude of this synergistic effect varies depending on the contaminant of interest (Prado and Esplugas, 1999). The combined O 3 /UV process is more efficient at generating hydroxyl radicals than the combined H 2 O 2 /UV process for equal oxidant concentrations using LP-UV. This is because the molar extinction coefficient of O 3 at 254 nm is two orders of magnitude greater than that of H 2 O 2, indicating that a lower UV intensity or a higher H 2 O 2 dose is required to generate the same number of hydroxyl radicals for these two processes (Glaze et al., 1987). However, for MP-UV lamps, H 2 O 2 /UV processes will generate more hydroxyl radicals than O 3 /UV processes, assuming the peroxide absorbs greater than 17 percent of irradiated light (200 nm to 300 nm) (Cater, 1999). The disadvantages of the O 3 /UV system are: As mentioned previously, the use of ozone for source waters with high bromide concentrations (>0.1 mg/l) can result in the formation of bromate. The O 3 /UV process typically requires an air permit for ozone emissions in addition to an off-gas treatment system for ozone destruction. These controls will add to the operational and capital cost of the system (AWWA/ASCE, 1997). Despite the fact that O 3 /UV is more stoichiometrically efficient at generating hydroxyl radicals than H 2 O 2 /UV or H 2 O 2 /O 3, the O 3 /UV process is less energetically efficient than H 2 O 2 /UV or H 2 O 2 /O 3 for generating large quantities of hydroxyl radicals due to the low solubility of O 3 in water compared to H 2 O 2. Thus, operational costs are expected to be much higher than these comparative processes. The hydroxyl radical yield can be decreased further by the presence of interfering parameters (e.g., nitrates, turbidity, or iron) in the source water. 144

167 Gaseous O 3 must be diffused into the source water, resulting in potential mass transfer limitations relative to H 2 O 2, which is fed as a liquid solution (Wagler and Malley, 1994). UV light penetration into the source water and, thus, process efficiency can be adversely affected by turbidity (Prado and Esplugas, 1999). As mentioned previously, there are many interference compounds that absorb UV light (e.g., nitrate and iron) and, thus, reduce process efficiency. UV lamp and sleeve failures can potentially contaminate treated water with mercury, although all lamp failures to date have resulted in aqueous Hg concentrations below drinking water standards (Crawford, 1999). Research suggests that the use of UV combined with pre- and/or post-chlorination can potentially result in the increased formation of THM and HAA 9 at UV dosages >100 mj/cm 2 (Zheng et al., 1999a,b). Bench-scale Studies To date, no bench-scale studies of MTBE destruction with the O 3 /UV process have been identified. Pilot/Field Studies and Vendor Information The O 3 /UV process has been employed for the following applications (Rice, 1997): Groundwater treatment to destroy TCE, PCE, and PCP. Remediation at Superfund sites to destroy VOCs and benzidines. Remediation at U.S. Army Ammunition plants to destroy explosive compounds such as 2,4,6-trinitrotoluene (TNT) and cyclonite (also referred to as RDX). Most O 3 /UV systems in place are for destruction of ordinance compounds. Currently, there are no field or pilot applications of O 3 /UV technology for MTBE remediation. This is likely because this technology is economically prohibitive relative to other AOPs for easily degradable compounds (Crawford, 1999). Some of the known vendors for O 3 /UV systems include: Hydroxyl Systems (Sidney, British Columbia, Canada), Calgon Carbon (Markham, Ontario, Canada), U.S. Filter (Santa Clara, CA) and HGC-UV Incorporated (Tucson, AZ). Table 3-5 summarizes vendor information for these technologies. 145

168 Summary The applications of ozone and UV are energy intensive processes and, hence, a combined O 3 /UV process may not be cost effective for treating waters with high TOC and MTBE concentrations. In addition, O 3 /UV process requires the expending of significantly more (electrical) energy than H 2 O 2 /UV or H 2 O 2 /O 3 processes. The use of ozone in potable water applications can result in the generation of bromate at concentrations above the Stage 1 D/DBP Rule of 10 µg/l. In conclusion, due to these economic and practical constraints, this technology will not be considered further for MTBE removal from drinking water in the remainder of this chapter. Hydrogen Peroxide/UV (H 2 O 2 /UV) Process Description As in the O 3 /UV process, the effectiveness of the H 2 O 2 /UV process relies on several synergistic oxidation mechanisms for the destruction of MTBE. The oxidation of organics can occur by either direct photolysis or reactions with hydroxyl radicals. Hydroxyl radicals are produced from the photolytic dissociation of H 2 O 2 in water by UV irradiation (Wagler and Malley, 1994; Calgon AOT Handbook, 1998). As in the O 3 /UV and H 2 O 2 /O 3 systems, the degradation of MTBE is primarily due to the oxidation reactions initiated by the highly reactive hydroxyl radicals: hυ H 2 O 2 /UV process: H 2 O 2 2 OH (λ <300 nm) OH + MTBE Oxidation by-products System Description/Design Parameters For the H 2 O 2 /UV system, higher radical generation results from the use of MP-UV lamps relative to the LP-UV lamps, due to the better absorptivity of H 2 O 2 at lower wavelengths (Cater, 1999). Peroxide dissociates to form hydroxyl radicals at wavelengths of 250 nm and below. Thus, while peroxide dissociation occurs with LP-UV, MP-UV emits a broader spectrum that promotes the dissociation of peroxide better than LP-UV. Calgon Carbon (Markham, Ontario, Canada) uses MP-UV lamps exclusively in its H 2 O 2 /UV processes due to the requirement for fewer lamps, the potential for direct photolysis, and smaller resulting system size (Cater, 1999). All of the reactor configurations discussed for the O 3 /UV process are applicable for the H 2 O 2 /UV process. H 2 O 2 /UV systems are equipped with hydrogen peroxide storage and injection systems in place of an ozone generator and diffuser system. Hydrogen peroxide is injected upstream of the reactor using metering pumps and mixed by in-line static mixers (Crawford, 1999). 146

169 The key design and operating parameters include the H 2 O 2 dose, the UV lamp type and intensity, the reactor contact time, and the control systems (ph and temperature). The low molar extinction coefficient for H 2 O 2 (Wagler and Malley, 1994) results in the use of MP-UV lamps for higher hydroxyl radical yields. UV doses typically range from 2.5 kwh/1,000 gallons to 15 kwh/1,000 gallons depending on water quality and contaminant concentrations (Crawford, 1999). The UV quartz sleeve cleaning frequency is a function of iron and other scalants that are present in the water. Hydrogen peroxide can be added either as a single slug dose or at multiple points in the system. The optimum dose of H 2 O 2 should be determined for each water source based on bench and pilot-scale testing, but is commonly estimated at twice the TOC and not less than 1 to 2 mg/l (e.g., TOC for drinking water ranges from less than 0.1 mg/l to greater than 7 mg/l, which would suggest a peroxide concentration of up to 14 mg/l). As previously noted, currently, there are no federal or state regulations for H 2 O 2 residual in treated drinking water; however, drinking water purveyors are not likely to allow any detectable levels of H 2 O 2 in treated drinking water because of concerns over biological growth. Thus, if H 2 O 2 is added at very high concentrations (>10 mg/l), effluent treatment will be required. Consequently, Calgon Carbon (Markham, Ontario, Canada) commonly keeps H 2 O 2 doses at less than 3 to 5 mg/l to minimize H 2 O 2 residuals. Once the optimum H 2 O 2 dose is determined, the EE/O for the target compound is applied to determine energy costs. Pulsar Environmental Remediation Technologies, Inc. (Auburn, CA) markets modular, P-UV reactors known as Riptide systems for remediation applications. These Riptide reactors are available in three different sizes: 1 to10 gpm (Riptide-8), 10 to 60 gpm (Riptide-20), and 60 to 400 gpm (Riptide-350) (Bender, 1998). The large Riptide reactor (Riptide-350) is a 6-foot vertical chamber with a 20-inch diameter (Bender, 1999). The Riptide system is comprised of a multi-pass reaction chamber containing a high-energy UV flashlamp. These Pulsar (Auburn, CA) UV lamps radiate UV light in a broad spectrum ranging from 185 nm to 400 nm, in a radiation profile known as blackbody or continuum radiation (Bender, 1998). The Pulsar (Auburn, CA) lamps also radiate visible and infrared light from 400 to 3,000 nm, in accordance with the blackbody profile (Bender, 1998). The major components of a H 2 O 2 /UV system include: UV lamps, lamp sleeves, and lamp cleaning system Hydrogen peroxide storage and injection system Reactor chamber In-line mixer Supply and discharge pumps and piping Monitoring and control systems 147

170 Figure 3-1a shows a schematic of a system capable of using O 3, H 2 O 2, and UV. A H 2 O 2 /UV system would look very similar, except for the absence of the O 3 feed system. Advantages and Disadvantages The advantages and disadvantages of the H 2 O 2 /UV system are briefly summarized in Table 3-2. The advantages of the H 2 O 2 /UV system are: No potential for bromate formation in the H 2 O 2 /UV process because the system does not rely on ozone for organic destruction (Siddiqui et al., 1999). Prior studies have demonstrated that the H 2 O 2 /UV process can oxidize >95 percent MTBE compared to <10 percent for UV or H 2 O 2 alone under similar test conditions (Wagler and Malley, 1994). Currently, the only full-scale drinking water treatment AOP in the United States is a H 2 O 2 /MP-UV system installed in Salt Lake City, Utah. According to this literature review, H 2 O 2 /UV systems appear to be the most tested and applied AOP in drinking water applications relative to the other AOPs, although not for MTBE. Thus, the implementation of H 2 O 2 /UV systems for drinking water applications has a history of operation and regulatory acceptance. MP-UV and P-UV irradiation can serve as an effective disinfectant for a variety of microorganisms (e.g., viruses); however, there is currently no regulatory authority for receiving disinfection credit as a result of using MP-UV or P-UV. H 2 O 2 is highly soluble and can be added to the source water at high concentrations, whereas O 3 is a much less soluble gas that must be bubbled into the source water. Consequently, H 2 O 2 /UV processes can generate larger amounts of hydroxyl radicals than O 3 /UV processes for equal amounts of energy used to add the oxidants to the source water. Furthermore, assuming the peroxide absorbs greater than 17 percent of the 200 to 300 nm light, H 2 O 2 /MP-UV processes will generate more hydroxyl radicals than O 3 /UV for equal concentrations of O 3 and H 2 O 2 in the source water (Cater, 1999). The disadvantages of the H 2 O 2 /UV system are: UV light penetration and, therefore, process efficiency can be adversely affected by high turbidity and elevated nitrate concentrations (Prado and Esplugas, 1999). UV lamp and sleeve failures can potentially contaminate treated water with mercury, although all lamp failures to date have resulted in aqueous Hg concentrations below drinking water standards (Crawford, 1999). 148

171 Research suggests that the use of H 2 O 2 /UV combined with pre- and/or post-chlorination can result in the increased formation of THM and HAA 9, especially at high UV dosages (>2,000 mj/cm 2 ). Note: 2,000 mj/cm 2 translates to approximately 0.6 kwh/1,000 gallons for a Calgon system (Cater, 1999) and thus, is well within the range of UV used for AOP applications. The theoretical yield of hydroxyl radicals via the H 2 O 2 /UV process is greater than that for the H 2 O 2 /O 3 process; however, due to interfering compounds in the water, this theoretical yield can be decreased to below that of the H 2 O 2 /O 3 process. Once the hydroxyl radicals are formed, however, the chemical destruction and interferences are the same for both technologies. The presence of residual hydrogen peroxide in the treated effluent will promote biological re-growth in the distribution system. Currently, there are no federal or state regulations for H 2 O 2 residual in drinking water; however, drinking water purveyors are not likely to allow any detectable levels of H 2 O 2 in treated drinking water (detection limits range from 1 to 100 µg/l depending on the method and concentration) because of concerns over biological growth. Thus, depending on effluent concentrations, post-treatment of excess H 2 O 2 may be required to limit downstream biological fouling. High concentrations of residual peroxide (exceeding a few mg/l) can be treated using catalytic activated carbon (Crawford, 1999). Bench-scale Studies Wagler and Malley (1994) conducted bench-scale studies to determine the effectiveness of UV light, H 2 O 2, and UV combined with H 2 O 2 in removing MTBE from contaminated groundwater in New Hampshire. In general, treatment of a simulated groundwater with ph between 6.5 and 8.0 by UV alone or by H 2 O 2 alone produced less than 10 percent removal of MTBE after 2 hours of exposure. In contrast, the combination of UV and H 2 O 2 within the ph range of 5.5 to 10 produced more than 95 percent removal of MTBE after only 40 minutes of exposure time. This study confirmed that the hydroxyl radical formed in the H 2 O 2 /UV process is the primary oxidant responsible for the oxidation of MTBE. During these oxidation experiments, methanol, formaldehyde, TBA, and 1,1-dimethylethyl-formate were identified as by-products of the H 2 O 2 /UV process. Furthermore, H 2 O 2 /UV oxidation of an actual groundwater containing MTBE and other VOCs resulted in 83 percent removal of MTBE after 2 hours of contact time. Chang and Young (1999) determined the kinetics of H 2 O 2 /UV degradation of MTBE by using a recirculating batch reactor with a LP-UV lamp. With a spiked MTBE concentration of 10 mg/l, H 2 O 2 /UV treatment resulted in 99.9 percent removal. The major by-product identified was TBF. The second order rate constant for the MTBE/ OH reaction under the H 2 O 2 /UV treatment process was found to be 4.82 x 10 9 M -1 s -1. The mean second order rate constant for the reaction of TBF with OH was found to be 1.19 x 10 9 M -1 s -1. The yield for 149

172 TBF formation from the MTBE/OH reaction was calculated to be 27 percent under the conditions of this experiment. In preliminary bench-scale studies, peroxide-assisted Pulsar (Auburn, CA) P-UV systems successfully reduced the MTBE from influent concentrations ranging from 40 to 2,000 µg/l to less than 5 µg/l (Bender, 1998). Presence of high turbidity, large particles and excess total dissolved solids (TDS) can affect the performance of P-UV systems (Bender, 1999), similar to other UV-dependent AOPs. Pulsar Environmental (Auburn, CA) is currently working with NSF International for certification of the Riptide system for use in drinking water applications (Bender, 1999). Pilot/Field Studies and Vendor Information In July 1998, a pilot treatment plant was constructed at the Charnock well field in Santa Monica, California to evaluate treatment technologies for removal of MTBE and TBA from drinking water. The treatment plant included several treatment processes, including an H 2 O 2 /UV oxidation system (for MTBE, TBA destruction), several carbon adsorption systems (for by-product destruction and polishing), and a packed tower air stripper. Additional systems, including a granular media filter and bag filter with an oxidant injection system, were installed to evaluate iron and manganese removal for pre-treatment to the H 2 O 2 /UV system. Calgon Carbon (Markham, Ontario, Canada) provided the H 2 O 2 /UV system, which consisted of a tower reactor with three MP-UV lamps. The reactor was approximately 42-inches in diameter and 6 feet in height. The pilot facility began operation on July 31, 1998 at a design flow of 140 gpm, but testing was conducted at flows as high as 350 gpm. The pilot testing was conducted over a period of approximately 12 months and included an optimization phase, a reliability phase, and a sensitivity phase for MTBE, TBA, and other by-product testing (Rodriguez, 1999). Raw groundwater from the Charnock well field was spiked at concentrations of approximately 1,000 µg/l MTBE and/or 200 µg/l TBA. MTBE and TBA were removed to less than 10 µg/l; however, the results of the pilot testing indicated that the costs were significantly higher than expected and the removal efficiencies were lower than predicted. The detection of by-products in the treated water (TBA, TBF, and acetone) mandated the use of an additional treatment unit, which would further increase the cost of treatment. Finally, residual H 2 O 2 in the treated water would require installation of a carbon system for its removal. Testing conducted indicated that the H 2 O 2 /UV technology worked for MTBE removal, but the high energy requirements, complications caused by several sleeve/lamp failures, formation of by-products, and requirement for additional treatment processes significantly reduced the advantages of this technology (Rodriguez, 1999). 150

173 Currently, Calgon Carbon (Markham, Ontario, Canada) has two installations where MTBE is being treated (Cater, 1999). One of these installations treats heavily contaminated wastewater with high concentrations of MTBE (100 mg/l), BTEX (40 mg/l), and chemical oxygen demand (COD) (2,000 mg/l). The other installation is for pipeline rinse water that contains BTEX (4 mg/l) along with MTBE (11 mg/l). Both these units were designed for BTEX treatment (Cater, 1999). While not for MTBE treatment, there are several full-scale drinking water applications of H 2 O 2 /MP-UV. Calgon Carbon (Markham, Ontario, Canada) currently maintains two fullscale AOP applications for drinking water treatment. These systems in Eastern Canada and Salt Lake City, Utah use H 2 O 2 /MP-UV to remove NDMA and PCE, respectively, at low concentrations. The Salt Lake City installation uses a 12-foot tower (4-foot diameter) with 12 layers of lamps aligned perpendicular to the flow to treat groundwater at 3,000 gpm. Treated water is discharged directly into the distribution system. Calgon Carbon (Markham, Ontario, Canada) is currently in the process of installing two additional H 2 O 2 /MP-UV systems in La Puenta, California and a second in Southern California. Some of the known vendors for H 2 O 2 /UV systems include: Hydroxyl Systems (Sidney, British Columbia, Canada), Calgon Carbon (Markham, Ontario, Canada), U.S. Filter (Santa Clara, CA), Pulsar Environmental Remediation Technologies (Auburn, CA), and Trojan Technologies (London, Ontario, Canada). Table 3-5 summarizes the vendors, their technologies and number of installations. U.S. Filter (Santa Clara, CA) has a patented H 2 O 2 and/or O 3 enhanced oxidation system known as Ultrox (Gruber, 1994). Ultrox systems are currently being used at more than 20 groundwater remediation applications and 10 wastewater treatment applications. None of these Ultrox systems treat MTBE (Himebaugh 1999; Woodling 1999). Trojan Technologies (London, Ontario, Canada) has more than 2,000 H 2 O 2 /UV systems for disinfection of wastewaters (Dewaal, 1999) and is currently planning to perform bench-scale studies in collaboration with HSI at the JFK Airport site discussed earlier (Dewaal, 1999). Summary H 2 O 2 /UV oxidation is one of the few AOPs used in drinking water treatment (e.g., Salt Lake City, Utah). In addition, a large amount of bench- and pilot-scale research has been conducted on the removal of MTBE and other gasoline contaminants using this AOP. A low potential for bromate formation coupled with compact reactor design makes the H 2 O 2 /UV system an attractive AOP option for treating potable waters. For highly turbid waters and for waters with high concentrations of scaling agents, appropriate pre-treatment may be required to enhance MTBE removal efficiency. As with all AOPs, more pilot- and field-scale demonstration sites under a variety of water quality matrices are needed prior to general regulatory acceptance. 151

174 152

175 3.5 Emerging Technologies E-beam Treatment Process Description E-beam treatment refers to the use of ionizing radiation from an electron beam source to initiate chemical changes in aqueous contaminants. In contrast to other forms of radiation, such as infrared and UV, ionizing radiation from an E-beam is absorbed almost completely by the target compounds electron orbitals, thus increasing the energy level of its orbital electrons. The energy level of radiation is sufficiently high to produce changes in the molecular structure of compounds, but is too low to induce radioactivity (Siddiqui et al., 1996; HVEA, 1999). Electron beam processes use the portion of the electromagnetic spectrum between 0.01 ev and 10 ev (Siddiqui et al., 1996). Within to seconds, E-beam irradiation (^^^) of water results in the formation of electronically excited species, including ions and free radicals, along the path of the electrons. The products of direct reactions of water molecules with the electron beam are formed in isolated volumes referred to as spurs. As these spurs expand through diffusion, a fraction of the initial products escape into the bulk solution and transfer their energy to other aqueous chemical species, causing more reactions to occur (Nickelsen et al., 1992). After approximately 10-7 s, oxidizing species, such as hydroxyl radicals, and reducing species, such as aqueous electrons and hydrogen atoms, are formed from the E-beam irradiation of water (Nickelsen et al., 1992; Allen, 1961). The net reaction is shown below: H 2 O + ^^^ 2.7 OH H e aq H H 2 O H 3 O + The combination of products that result from this reaction creates a unique environment where oxidizing and reducing reactions occur simultaneously (Allen, 1961). In particular, note that the oxidizing species, OH, and the reducing species, e aq -, are expected to be present in similar steady-state concentrations. These two species, along with another reducing species, the hydrogen atom ( H), are the most reactive products of this reaction and control the rate of the electron beam process for MTBE destruction. The reactions of these species with MTBE are as follows (Cooper and Tornatore, 1999): MTBE + e aq - (CH3 ) 3 C + - OCH 3 - OCH3 + H 2 O HOCH 3 + OH - MTBE + H Reduction by-products -H 2 OH + MTBE Oxidation by-products 153

176 The above-mentioned reactions are summarized in Table 3-3. The aqueous electron reacts with MTBE according to a rate constant of 1.75 x 10 7 M -1 s -1, while the rate constant for the reaction of MTBE with hydrogen atoms was found to be less than 8.0 x 10 4 M -1 s -1 (Cooper and Tornatore, 1999). The reaction rate of MTBE with hydroxyl radicals is approximately 90 and 20,000 times faster than with aqueous electrons and hydrogen atoms, respectively (Buxton et al., 1988; Cooper and Tornatore, 1999). System Description/Design Parameters In the E-beam process, a continuous stream of high-energy electrons irradiates contaminated water. The generation of high-energy electrons is accomplished through the use of an electron accelerator in which electrons emitted by a hot cathode (e.g., tungsten filament) are accelerated by means of a voltage differential (Nickelsen et al., 1998). The accelerated electrons are then deflected magnetically by a scanner to produce an E-beam, which scans the water surface with a particular radiation pattern and frequency. Once in the water, the electrons react with water molecules to form reactive intermediates, such as hydroxyl radicals, hydrated electrons, and hydrogen atoms, as discussed above (Cooper, 1999). The shape and frequency of this pattern is controlled to apply a uniform amount of electrons (dose) to the source water stream. A common unit of electron dose is the rad, defined as the energy absorption of 100 ergs per gram of material. The maximum depth of penetration of an E-beam is directly proportional to the energy of the incident electrons and inversely proportional to the density of the falling stream, the beam power, and the length of time the water is exposed to the electron beam. Most E-beam systems for drinking water treatment are designed such that the electron beam infiltrates less than a centimeter into the source water. For example, 1.5 MeV electrons have a depth of penetration of approximately 7 mm in water (Nickelsen et al., 1994). Currently, there is only one E-beam configuration used to treat drinking water. In this configuration, an E-beam scans in a raster pattern over a thin (approximately 4 mm) sheet of water. This configuration is designed to apply radiation doses up to several thousand krad and scanned at 200 Hz by 60 Hz to cover an area 1.2 m wide by 5 cm deep (Nickelsen et al., 1994). A key piece of equipment for the application of E-beam is the water distribution system. For E-beam to be effective, the source water must be spread over a plate at a sufficiently shallow thickness to allow electrons to penetrate most of the water. If the water thickness is too deep because of higher flow rates or limitations of the distribution system, multiple water passes through the electron beam may be required to meet effluent goals. Because E-beam systems can potentially emit x-rays, the electron accelerator, the beam scanner, and the contact chamber are usually completely surrounded by lead of varying thickness to attenuate any emitted x-rays to less than 0.2 mrem/h (Nickelsen et al., 1998). The major components of the E-beam system include: Electron accelerator with an insulating core transformer 154

177 Power source Beam scanner Contact chamber (concrete vault) Water distribution device Supply and discharge pumps and piping Resistance temperature devices (RTDs) Monitoring and control systems Lead shielding A schematic of the various components of an E-beam system is shown in Figure 3-2. CONTAMINATED WATER SURGE TANK INFLUENT SAMPLE WATER SPREADER CONCRETE VAULT ELECTRON BEAM GENERATOR POWER SOURCE THIN FILM OF WATER BEING RADIATED EFFLUENT SAMPLE CONTROL PANEL WATER TO DISTRIBUTION SYSTEM Figure 3-2. A schematic of a high energy electron beam system (drawing provided by Komex H2O Science, 1998). 155

178 Advantages and Disadvantages The advantages and disadvantages of the E-beam system are briefly summarized in Table 3-2. The advantages of the E-beam system are: Little potential for inorganic by-product formation (e.g., bromate) due to the large number of radicals produced. Some studies have indicated that E-beam irradiation can actually reduce the concentration of bromate in water (Siddiqui et al., 1996). Recent laboratory E-beam studies have demonstrated minimal organic by-product formation compared to other AOPs (Cooper, 1998; Cooper et al., 1999; Tornatore 1999); however, further studies are required to confirm this preliminary finding. The E-beam system can supplement the disinfection process, providing additional protection against pathogenic microorganisms (Kurucz et al., 1991). Studies have indicated that interference and turbidity have minimal effects on the performance of the E-beam treatment system (Cooper, 1998). The disadvantages of the E-beam system are: There are currently no full-scale drinking water applications of E-beam systems, although there have been many pilot-scale systems over the past several years. E-beam relies on irradiation of drinking water, a term synonymous in the public to radiation. Despite the relatively safe nature of this technology, radiation shielding is still required and, thus, there is significant public perception challenges that must be overcome prior to the implementation of E-beam. This is likely the largest disadvantage of E-beam, relative to the other AOPs evaluated in this chapter. E-beam systems are energy intensive and may prove to be cost prohibitive. The E-beam system requires specially trained skilled operators who are able to work near a radiation source. While not necessarily dangerous, this would likely require increased labor costs. Pilot/Field Studies and Vendor Information Over 700 E-beam systems have been implemented in materials applications and food and drug industry disinfection applications worldwide (Tornatore, 1999). There are presently no electron beam processes in continuous drinking water service, but over 200 pilot and demonstration studies have been performed on over 60 different organic contaminants. Some of the targeted contaminants include benzene (Nickelsen et al., 1992; Nickelsen et al., 1994), 156

179 phenol, TCE, and PCE (Lin et al., 1995; USEPA, 1998), THMs and THM precursors (Cooper et al., 1996), and MTBE (Cooper et al., 1998). Of the 200 pilot and demonstration studies, more than 10 have been designed to evaluate performance on MTBE, TBA, TBF, and formaldehyde (Tornatore, 1999). Many of these pilot scale studies have been completed using the mobile E-beam treatment system designed and built by High Voltage Environmental Applications (HVEA). This system can have process flow rates of up to 40 gpm at an applied power of 20 kw (Cooper et al., 1999; HVEA, 1999). HVEA also has a research station, formerly known as EBRF (Electron Beam Research Facility), at the Miami Dade Central District Wastewater Treatment Plant in Miami, Florida (HVEA,1999). At this (EBRF/HVEA) facility, there is an E-beam reactor designed for large scale research. The E-beam Reactor in Miami Dade is comprised of a horizontal 1.5 MeV insulated-core transformer (ICT) electron accelerator capable of delivering up to 60 ma of beam current (Nickelsen et al., 1994). Test results from MTBE pilot and demonstration studies have shown the ability of E-beam systems to reduce MTBE concentrations from 1,000 µg/l to less than 5 µg/l (Tornatore, 1999). Tornatore et al. (1999) also conducted large-scale E-beam experiments for MTBE destruction with a flow rate of 100 gallons per minute in a recycle mode. MTBE was reduced to below the detection limit (87 µg/l) after cumulative doses of 665 and 2,000 krads applied to initial MTBE concentrations of 2,300 and 31,000 µg/l, respectively. At the equivalent energy dose, primary reaction intermediates (TBA, TBF) were also treated to low residual concentrations. A dose response curve for TBA and TBF has been developed which suggests that the energy required for removal of these compounds to below detection levels is approximately equal to the energy requirement for MTBE removal. TBA and TBF reactions use reducing chemistry, and, thus, these reactions proceed concurrently with the process of MTBE destruction (Tornatore, 1999). A series of experiments were performed with the electron beam technology at Orange County Water District (Fountain Valley, CA) to treat MTBE and a variety of other contaminants in a number of different water sources (Cooper et. al., 1999). The results indicate that MTBE was readily treated in all experiments, and that treatment efficiency was dependent upon basic water chemistry, delivery system limitations, and the presence of competing organic and inorganic compounds. MTBE removal efficiencies of greater than 99.5 percent, with final concentrations less than 5 µg/l, were reported in several experiments (Cooper et al., 1999). Primary reaction intermediates (TBA, TBF) were reduced concurrently with MTBE, suggesting that the oxidizing and reducing chemistry involved is efficient in treating MTBE and its by-products. Background water quality was shown to have an impact on the treatment efficiency of MTBE and other compounds. Waters lower in TOC and ph demonstrated higher removal efficiencies. Researchers concluded that the application of the electron beam process is best suited for high flow rate, single or multiple constituent treatment scenarios where complete oxidation or mineralization of contaminants is the desired endpoint (Cooper 157

180 et al., 1999). Tables 3-4 and 3-5 present summaries of on-going field studies and vendor information, respectively. Summary E-beam systems are used widely in the food and drug industry for disinfection; however, over the past several years, a large number of pilot-scale studies have been completed at drinking water facilities. Due to the nature of the reducing and oxidizing species created in a E-beam reactor, MTBE concentrations can be reduced to well below action levels with minimal to no by-product formation. Despite this fact, the negative public perception resulting from the use of radiation combined with the requirement for skilled operators and the expected high capital and O&M costs for E-beam systems will likely result in their limited application. However, because this technology has been used in the past, there may be some treatment or remediation scenarios where E-beam will be selected because it may provide advantages relative to other treatment options Cavitation Process Description Cavitation is described as the formation of microbubbles in solution that implode violently after reaching a critical resonance size. These microbubbles can be produced by a number of mechanisms: 1) local increase in water velocity as in eddies or vortices, or over boundary contours; 2) rapid vibration of the boundary through sonication; 3) separation or parting of a liquid column owing to water hammer; or 4) an overall reduction in static pressure. The rapid implosion of cavitation microbubbles results in high temperatures at the bubble/water interface, which can trigger thermal decomposition of the MTBE in solution or thermal dissociation of water molecules to form extremely reactive radicals. The extreme conditions generated during cavitation decomposes water to create both oxidizing ( OH) and reducing ( H) radical species (Skov et al., 1997; Kang and Hoffman, 1998). As in other AOPs, the primary mechanism for MTBE removal by cavitation is through reaction with hydroxyl radicals. There are three known methods of producing hydroxyl radicals using cavitation namely, ultrasonic irradiation or sonication, pulse plasma cavitation, and hydrodynamic cavitation. Sonication causes the formation of microbubbles through successive ultrasonic frequency cycles until the bubbles reach a critical resonance frequency size that results in their violent collapse (Mason et al., 1988; Kang and Hoffman, 1998). Pulse plasma cavitation utilizes a high voltage discharge through water to create microbubbles. In hydrodynamic cavitation, microbubbles are generated using high velocity or pressure gradients (Pisani and Beale, 1997; Pisani, 1999a,b). 158

181 The production of OH through cavitation processes can be enhanced with the use of ozone (Table 3-3). Gas-phase ozone thermally decomposes in the microbubbles, yielding oxygen atoms and molecular oxygen. This results in a number of reactions that subsequently yield hydroxyl radicals (Kang and Hoffmann, 1998): O 3 + H 2 O O OH O 3 + OH HO O2 O 3 + HO 2 - OH + O2 - + O2 System Description/Design Parameters As discussed above, there are three known methods of producing hydroxyl radicals using cavitation namely, ultrasonic, hydrodynamic, and pulse plasma cavitation. A noble gas (e.g., krypton or argon) is sometimes used to achieve the optimal bubble production and size. MTBE removal occurs by both thermal decomposition at the bubble-water interface and by reaction with the radicals. The following sections will discuss the two most frequently studied and applied forms of cavitation: sonication and hydrodynamic cavitation. Ultrasonic Cavitation/Sonication When a liquid is irradiated with ultrasound, the ultrasound waves pass through the medium in a series of alternate compression and expansion cycles. When the acoustic amplitude is large enough to stretch the molecules during its negative pressure (rarefaction) cycle to a distance that is greater than the critical molecular distance to hold the liquid intact, microbubbles are created that then collapse in the subsequent compression cycle, giving rise to extremes of temperature and pressure. Estimates have suggested that temperatures greater than 5,000 C and pressures greater than 1000 atm can be produced locally during the collapse of these vapor bubbles (Pandit and Moholkar, 1996). The main factors that affect ultrasonic cavitation include: 1) the intensity of the ultrasound field (i.e., the frequency and amplitude of radiation); 2) the physical properties of the water (e.g., viscosity, surface tension, and vapor pressure); 3) the temperature; and 4) the presence of dissolved gas (Martin and Ward, 1993). There are several different kinds of sonication reactors that are currently available for commercial use, namely (Martin and Ward, 1993): Ultrasonic Cleaning Bath: Contaminated water is sonicated in a reactor with either external transducers (hooked to the container walls) or submersible transducer. This reactor is recommended for low-intensity irradiation applications. Probe System Reactors: In these reactors, the small magnitude oscillations of a piezoelectric crystal are amplified by placing it in a metal probe that is, in turn, immersed in the water. These reactors are available in both batch and flow through designs. 159

182 Tube Reactors: In these reactors, the water flows through pipes that are surrounded by transducers. These reactors are typically employed for large flowrate applications. The optimization of ultrasonic cavitation can be achieved by adjusting the ultrasonic frequency and saturating gas during sonication (Hua and Hoffman, 1997). Hua and Hoffman (1997) studied the production of hydroxyl radicals at ultrasonic frequencies of 20, 40, 80 and 500 khz, respectively, in the presence of four different saturating gases (Kr, Ar, He and O 2 ). The highest rate of OH production (0.391 um/min) was observed during the sonication of Kr-saturated solutions at 500 khz. Sonication of He-saturated solutions at 20 khz resulted in the lowest rate of OH production ( um/min) (Hua and Hoffmann, 1997). There are several mechanisms suggested to explain the higher hydroxyl radical production at higher frequencies. First, due to the shorter time allowed for bubble collapse at higher frequencies, there is less time for the hydroxyl radicals to recombine within the bubble and, thus, a higher hydroxyl radical production rate is observed (Petrier et al., 1992). Next, as the frequency increases, the bubbles may not completely collapse, but will rapidly oscillate and, in doing so, create a higher flux of hydroxyl radicals through the surface of the bubble (Hua and Hoffman, 1997). Finally, as an explanation for the apparent hydroxyl radical production rate dependence on inert gas, Hart and Henglein (1986) suggest that higher temperatures can be achieved within the bubbles for higher molecular weight gases. The nature of these arguments demonstrate that there are still many unknowns regarding the specific mechanism(s) for hydroxyl radical generation during sonication. Hydrodynamic Cavitation Hydrodynamic cavitation can be achieved when pressure at the orifice or any other mechanical constriction falls below the vapor pressure of the liquid, causing the formation of microbubbles. Once generated, microbubbles rapidly collapse downstream with a recovery of pressure giving rise to high temperature and pressure pulses. For water flowing through an orifice, a reduction in the cross-section of the flowing stream increases the velocity head at the expense of pressure head. During the re-expansion of flow, the fluid stream separates at the lower end of the orifice and generates eddies. At a particular velocity, the pressure during re-expansion falls below the vapor pressure of the water, causing the generation of microbubbles. If there is dissolved gas in the water, then cavitation is observed at pressures significantly above the vapor pressure because of the degassing that occurs at low pressures. The hydrodynamic cavitation reactor is simple and easy to operate. By changing the ratio of the orifice to the pipe diameter, the discharge pressure, and the pressure recovery rate, one can manipulate the outcome of hydrodynamic cavitation to suit the conditions of individual reactions or physical processes (Chivate and Pandit, 1993). Due to the longer life of the bubble and the higher velocity from which they are swept away from their point of generation, the actual volume of the bulk fluid exposed to cavitation effects is higher for hydrodynamic cavitation. Because bubbles under hydrodynamic cavitation show oscillatory behavior, a large number of smaller magnitude pulses are observed (Pandit and Moholkar, 1996). 160

183 The configuration of a hydrodynamic cavitation process is comprised of a centrifugal feed pump and a cavitation reactor that is connected to the effluent pipeline (Pisani and Beale, 1997). Oxidation Systems Incorporated has a proprietary hydrodynamic cavitation reactor called Hydrox process. This process facilitates multiple-pass cavitation, using either a recycle line downstream of the cavitation reactor or several cavitation reactors placed in series. When necessary, the cavitation reactor designs can be expanded to include UV treatment modules as well as the addition of hydrogen peroxide by placing these systems either upstream or downstream in line with the cavitation reactor. A schematic of a cavitation system is shown in Figure 3-3. The major components of a cavitation system include: Hydrodynamic/ultrasonic/pulse plasma cavitation generator Reactor chamber Chemical feed tanks and pumps Power source Temperature controller Supply and discharge pumps and piping Monitoring and control systems CONTAMINATED WATER RECYCLE LOOP SURGE TANK FEED PUMP REACTOR TO DISTRIBUTION SYSTEM PRIMARY CHEMICAL TANK SECONDARY CHEMICAL TANK SECONDARY CHEMICAL FEED PUMP POWER SOURCE PRIMARY CHEMICAL FEED PUMP Figure 3-3. A schematic of a cavitation system (HYDROX) with supplemental chemical oxidants (e.g., H 2 O 2 ) (drawing provided by Komex H2O Science, 1998). 161

184 Advantages and Disadvantages See Table 3-2 for a brief summary of the advantages and disadvantages of cavitation AOP processes. The advantages of the cavitation process include: The energy usage for cavitation systems is comparable to AOPs using UV lamps (Pisani and Beale, 1997). The only energy costs result from the use of pumps to create pressures of 50 to 100 psi. Cavitation systems use no moving parts, besides a feed pump and, thus, require minimal maintenance costs. The disadvantages of cavitation processes are: Supplemental oxidants such as O 3 and H 2 O 2 may be required to significantly increase (by a factor of 1.5 to 4) the rate of MTBE removal (Kang and Hoffmann, 1998). The use of these oxidants will raise O&M costs. Currently, no full-scale applications exist for this emerging technology. Hydrodynamic cavitation is currently a black box technology due to the reluctance of the primary vendor to share information regarding the specific operation of the cavitation device. Consequently, this technology is unlikely to be accepted for drinking water applications until all information concerning operation is generally publicized. Bench-scale Studies Ultrasonic cavitation assisted by ozone or peroxide addition has been studied for the destruction of a variety of compounds including NOM (Olson and Barbier, 1994), carbon tetrachloride (Hua and Hoffman, 1996), chlorophenols (Serpone et al., 1994), hydrogen sulfide (Kotronarou et al., 1992), and MTBE (Kang and Hoffmann, 1998). Kang and Hoffmann (1998) investigated the kinetics and mechanism of the degradation of MTBE in the presence of ozone at an ultrasonic frequency of 205 khz and power of 200 Watts/L. The observed first-order degradation rate constant for MTBE increased from 4.1 x 10-4 s -1 to 8.5 x 10-4 s -1 as the initial concentration of MTBE decreased from 89 mg/l to 0.8 mg/l. The presence of O 3 at 12 mg/l was found to accelerate the rate of MTBE destruction by a factor of 1.5 to 3.9 depending on the initial concentration of MTBE. Ozone had a larger effect for low initial MTBE concentrations, suggesting that at higher contaminant concentrations, oxidation is limited by mass transfer of the hydroxyl radicals to the contaminant (Kang and Hoffman, 1998). TBF, TBA, methyl acetate, and acetone were found to be the primary byproducts and intermediates of MTBE degradation, but were shown to disappear after 60 minutes of reaction. 162

185 Pilot/Field Studies and Vendor Information Ultrasonic cavitation has been well studied at the bench-scale level, but there are no full-scale installations. Pulse plasma cavitation is energy intensive and, hence, has not progressed commercially for organic oxidation applications. However, hydrodynamically induced cavitation has been in full-scale application at approximately 30 installations for removal of polyaromatic hydrocarbons, phenol, glycol, and polyhalogenated hydrocarbons. Oxidation Systems Incorporated (OSI), located in Arcadia, California, commercially supplies hydrodynamically induced (HYDROX) cavitation reactors (Pisani, 1999a,b). OSI has deployed several of their cavitation reactors for groundwater remediation applications. OSI s full-scale applications have flow rates ranging from 1 to 2,000 gpm (Pisani, 1999b), with the largest single unit capable of handling 2,000 gpm. Larger (>500 gpm) scale HYDROX units were found to perform more efficiently than the smaller (<100 gpm) units (Pisani, 1999b). OSI has just completed Phase-1 field studies for MTBE removal at March Air Force Base, Riverside, California (Pisani, 1999a). This study was conducted for the U.S. Army Corps of Engineers. The flow rates varied from 10 to 30 gpm, with an average influent MTBE concentration of 500 µg/l. At these trials, verified by third parties, greater than 80 percent reduction in MTBE was accomplished (Pisani, 1999a). Tables 3-4 and 3-5 summarize the case study and vendor information, respectively, for the hydrodynamic cavitation process. Summary Hydrodynamically induced cavitation appears to be a promising AOP option for organic contaminant removal, including MTBE. However, all applications identified to date for MTBE have required the use of an additional oxidant to achieve MTBE concentrations that meet drinking water standards. Thus, more pilot and field studies will facilitate a better understanding of this AOP s ability to meet drinking water standards for a reasonable cost. In addition, the largest vendor of hydrodynamic cavitation systems, OSI (Arcadia, CA), currently claims that the reactor chamber is proprietary information (i.e., a black box technology). Even if cavitation proves to be technically feasible for removal of MTBE to drinking water limits, it is unlikely that it will be adopted for widespread use in the drinking water industry until the OSI operational system is made generally available. In conclusion, while continued observation of this technology is warranted, further pilot- and field-scale remediation and drinking water applications are needed to prove its economic and technical feasibility TiO 2 -Catalyzed UV Oxidation (TiO 2 /UV) Process Description When TiO 2, a solid metal catalyst, is illuminated by UV light (380 nm), valence band electrons are excited to the conduction band and electron vacancies, or holes, are created (Kormann et al., 1991; Crittenden et al., 1996). This combination of excited-state electrons is capable of initiating a wide range of chemical reactions; however, hydroxyl radical 163

186 oxidation is the primary mechanism for organic contaminant destruction (Kormann et al., 1991; Crittenden et al., 1996). The production of hydroxyl radicals can occur via several pathways but, as with many of the other AOPs analyzed, is readily formed from hydrogen peroxide. The production of hydrogen peroxide primarily occurs through the following three reaction mechanisms (Kormann et al., 1988). In the first mechanism, peroxide is created by the reduction of oxygen with two conduction band (CB) electrons. As the concentration of electron acceptors (e.g., oxygen) is increased in solution, the yield of these CB electrons is increased, thereby increasing the yield of hydrogen peroxide (Kormann et al., 1988). The presence of electron acceptors decreases the combination of excited electrons with holes and, thus, increases the formation of hydrogen peroxide or other radicals (Crittenden et al., 1996). O 2 + 2H + + 2e - CB H 2O 2 Hydrogen peroxide is produced via the second mechanism through the oxidation of water by holes in the valence band (h VB ). This mechanism is thought to occur only in the absence of electron acceptors and the presence of electron donors (e.g., H 2 O, OH -, and HCO 3 - ) (Kormann et al., 1988; Hong et al., 1987; Turchi and Ollis, 1990). 2H 2 O + 2h VB + H 2 O 2 + 2H + Finally, hydrogen peroxide can be produced by secondary reactions between oxidized organic matter. These reactions are thought to be important at high TOC concentrations or after long illumination periods (Kormann et al., 1988). Once hydrogen peroxide is formed, it can dissociate in the presence of UV radiation to form hydroxyl radicals (see H 2 O 2 /UV discussion) or react with other radicals (e.g., hydroperoxyl or superoxide radical) to form hydroxyl radicals. The hydroperoxyl radical is formed when oxygen is reduced by a CB electron (Prairie et al., 1993; Sjogren, 1995): O 2 + H + + e - CB HO 2 Deprotonation of the hydroperoxyl radical at neutral ph results in the formation of a superoxide radical ( O 2 - ) which, in turn, reacts with hydrogen peroxide (Halliwell and Gutteridge, 1989): HO 2 H + + O 2 - H 2 O 2 + O 2 - OH - + O2 + OH Finally, hydroxyl radicals can be formed from the direct reduction of TiO 2 -absorbed H 2 O 2 by a CB electron (Al-Ekabi et al., 1989): 164

187 H 2 O 2 + e - CB OH - + OH In addition, hydroxyl radicals can be produced by the reaction of a hole with a hydroxide ion (Hong et al., 1987; Turchi and Ollis, 1990; Sjogren, 1995): OH - + h + VB OH The above reactions are summarized in Table 3-3. Also summarized in this table are the reaction by-products, interfering compounds, and hierarchy of oxidants. System Description/Design Parameters TiO 2 /UV systems experience interference due to the same radical scavengers that affect the other AOPs; however, TiO 2 /UV systems are also fouled by the presence of anions (e.g., chloride, phosphate, and bicarbonate), cations, and neutral molecules, which compete with the contaminant for reactive sites on the surface of the TiO 2 particles. The effect of cations and anions is strongly ph dependent. The ph of zero charge for TiO 2 is approximately ph 6 (Kormann et al., 1988). Kormann et al. (1991) note that at low ph (ph 3 to 4), reaction rates were significantly retarded due to anion adsorption onto the positively charged TiO 2 surface. At higher ph (ph >7), the TiO 2 particles are negatively charged and there was negligible anion adsorption; however, the presence of cations (e.g., cobalt [II], aluminum [III], and zinc [II]) was shown to decrease the reaction rate (Kormann et al., 1991). As a result of this decreased activity, TiO 2 systems may require ion-exchange pre-treatment to remove both anions and cations (Crittenden et al., 1996). In a TiO 2 /UV reaction system, catalysts can be either injected or dispersed (i.e., slurry design) into the system or attached to a support medium. For slurry design, rigorous benchand pilot-scale testing is required for each source water to determine the optimum TiO 2 dose. A low TiO 2 dose can result in a surface site limiting reaction and insufficient radical generation whereas a high TiO 2 dose can reduce the transmittance of the UV light. Kormann et al. (1988) found that a suspension of 500 mg/l TiO 2 allowed the absorption of greater than 95 percent of the UV light at 330 nm. TiO 2 particles can vary in size and shape; however, those particles used by Kormann et al. (1988) are spherical in shape with an average diameter of approximately 30 nm. As the above reactions suggest, bubbling air through the system results in higher dissolved oxygen (DO) concentrations, which yield faster reaction rates (Kormann et al., 1988; Venkatadri and Peters, 1993; Barreto et al., 1995). Significant change (from 6.8 to 4.2) in ph was observed under TiO 2 -catalyzed UV treatment (Barreto et al., 1994). When TiO 2 is attached to a support substrate (e.g., silica-based material, cobalt [II]-based material, or synthetic resins sorbents [see Chapter 5 for further discussion]), it eliminates the need for a post-treatment separation system, which is required for slurry designs (Hong et al., 1987; Crittenden et al., 1996). In one fixed TiO 2 design, TiO 2 was mixed into a silica gel, 165

188 which was subsequently hardened. The silica gel had 9-nm pore sizes with a total surface area of 480 m 2 /g (Crittenden et al., 1996). The UV light penetrates this porous silica gel to activate the catalyst, which in-turn oxidizes contaminants in the source water as it is run through the TiO 2 impregnated silica gel. The catalytic activity of imbedded TiO 2 is improved by the addition of metals such as silver or platinum to the TiO 2 surface (Venkatadri and Peters, 1993; Crittenden et al., 1996). Research has shown that destruction of BTEX compounds (2 mg/l) was slow when DO levels were below 3 mg/l and very rapid as DO levels increased to above 15 mg/l (Crittenden et al., 1996). The major components of a TiO 2 /UV system include: TiO 2 slurry injection and extraction system (Option 1) TiO 2 impregnated resin fluidized bed reactor (Option 2) UV lamps, lamp sleeves, and lamp cleaning system Static mixing device Supply and discharge pumps and piping Monitoring and control systems A schematic of a fluidized bed TiO 2 /UV system (Option 1) is shown in Figure 3-4. Advantages and Disadvantages The advantages and disadvantages of the UV/TiO 2 system are briefly summarized in Table 3-2. The advantages are: TiO 2 assisted photocatalysis can be performed at higher (300 to 380 nm) wavelengths than the other UV oxidation processes (Prairie et al., 1993; Sjogren and Sierka, 1994; Sjogren, 1995). The TiO 2 oxidation process has been studied for many organic compounds, including MTBE, under a variety of water qualities. The disadvantages of the TiO 2 system are: Currently, no full-scale applications exist for this emerging technology. In attached TiO 2 systems, pre-treatment is required to avoid fouling of the active TiO 2 sites and destructive inhibition of the TiO 2 catalyst (Dewaal, 1999). Significant fouling was observed due to deposition of NOM, inorganic particulates, photoreduced metal cations 166

189 TREATED EFFLUENT SAMPLE CONTAMINATED WATER FLUIDIZED BED/ UV REACTOR SETTLING TANK TO DISTRIBUTION SYSTEM SURGE TANK T 1 O 2 IMPREGNATED CATALYST UV LAMPS CATALYST PUMP RETURN CATALYST INFLUENT SAMPLE DELIVERY PUMP IN-LINE MIXER SPENT CATALYST H 2 O 2 SOLUTION (Optional) H SOLUTION PUMP POWER SOURCE Figure 3-4. A schematic of a fluidized bed TiO 2 /UV system (drawing provided by Komex H2O Science, 1998). and, to a lesser extent, prolonged exposure to UV radiation (Crittenden et al., 1996). Inhibition was observed to occur due to the presence of increased alkalinity and other anionic species (e.g., sulfates [>100 mg/l] and chlorides) (Crittenden et al., 1996). If TiO 2 is added to the system as a slurry, then a separation step is required to remove the solid TiO 2 from the treated water (Barreto, 1995; Sjogren, 1995). There is a potential for rapid loss of TiO 2 photocatalytic activity, resulting in the need for a large volume of replacement catalyst on-site or a catalytic regeneration process (Crittenden et al., 1996). If DO concentrations in the source water are low (e.g., as in some groundwaters) oxygen sparging may be required to increase the rate of contaminant destruction. 167

190 The reaction efficiency is highly dependent on the ph of the system, resulting in the need for close monitoring and control. Bench/Pilot/Field Studies and Vendor Information Several researchers have studied TiO 2 assisted UV oxidation processes for the following applications: To destroy a number of organic contaminants at a variety of concentrations (Mathews, 1988; Al-Ekabi et al., 1989; Kormann et al., 1991; Crittenden et al., 1996). To remove toxic inorganic species (e.g., cyanide) and heavy metal ions (Peral and Domenech, 1992; Sabate et al., 1992; Prairie et al., 1993). To inactivate bacteria (Ireland et al., 1993) and viruses (Sjogren and Sierka, 1994; Sjogren 1995). Barreto et al. (1995) performed bench-scale studies on MTBE removal in TiO 2 slurry systems. At an optimum TiO 2 dose of 125 mg/l and 2 hours of oxygen sparging, Barreto et al. (1995) found that 76 percent of the initial MTBE (88 mg/l or 1 mm) was removed in the first 20 minutes (first order rate constant of 1.2 x 10-3 s -1 ). After 20 minutes, the rate slowed, requiring nearly 4 hours to remove MTBE below detection levels (pseudo first order reaction rate of 1.3 x 10 4 s -1 ) and 8 hours to achieve 95 percent oxidation of by-products such as TBA and TBF (Barreto et al., 1995). In these experiments, there was no measured direct photolysis of MTBE, TBA, or TBF (Barreto et al., 1995). There are currently no full-scale applications of the UV/TiO 2 process for MTBE treatment. Hydroxyl Systems Incorporated (Sidney, British Columbia, Canada), has developed a fluidized bed TiO 2 system for commercial use (Harp, 1999). Trojan Technologies (London, Ontario, Canada) also has a UV/TiO 2 reactor for full-scale applications. In the Trojan Technologies (London, Ontario) process, the TiO 2 is attached to a substrate (Dewaal, 1999). Table 3-5 lists information for vendors of TiO 2 /UV systems. Summary TiO 2 catalyzed UV oxidation is a process recommended for use in ultra-pure water applications (e.g., semi-conductor industry) and for treating waters with low contaminant concentrations. Although this technology shows promise, it is still in the developmental stages, and additional studies are needed prior to its use in large-scale remediation or drinking water treatment applications. Consequently, this technology will no longer be evaluated in the remainder of this chapter. 168

191 3.5.4 Fenton s Reaction Process Description Hydrogen peroxide reacts with iron (II) to form Fenton s reagent (an unstable iron-oxide complex) that subsequently reacts to form hydroxyl radicals (Fenton, 1894). The net reaction is shown below: Fe 2+ + H 2 O 2 Fe 3+ + OH - + OH k = 76 M -1 s -1 (Walling, 1975) This reaction can occur either in homogeneous systems with dissolved ferrous iron or in heterogeneous systems in the presence of complexed iron such as goethite (FeOOH). The by-product, ferric iron, in turn reacts with peroxide or superoxide (O 2 - ) radical to reproduce ferrous iron as shown below: Fe 3+ + H 2 O 2 Fe 2+ + O H + O Fe 3+ Fe 2+ + O 2 The above three reactions cycle iron between the ferrous and ferric oxidation states until the H 2 O 2 is fully consumed, producing OH in the process. As in other AOPs, the destruction of organics (including MTBE) is primarily due to oxidation reactions initiated by the hydroxyl radical. Fenton s reactions are summarized in Table 3-3. Similar reactions can occur with copper (II) in place of iron (II). System Description/Design Parameters The use of Fenton s chemistry to destroy MTBE in drinking water requires the addition of iron and H 2 O 2 to the source water. The dosages of Fe(II) and H 2 O 2 are determined based on the organic contaminant removals required. The reactor must be configured to provide adequate mixing of Fe(II) and H 2 O 2 in order to optimize hydroxyl radical formation and destruction of MTBE. To keep iron in solution, a very low ph (~2.5) is required. For drinking water applications, an iron removal system is required prior to delivery to the distribution system. A diagram of a system utilizing Fenton s reaction is shown in Figure 3-5. The major components include: Fe(II) and hydrogen peroxide storage and injection systems Completely stirred tank reactor ph controllers Iron removal system 169

192 Supply and discharge pumps and piping Monitoring and control systems Advantages and Disadvantages A summary of advantages and disadvantages for Fenton s Reaction is shown in Table 3-2. The advantages of Fenton s Reaction are: This process requires very little energy compared to other oxidation technologies that utilize O 3 or UV. This process produces no vapor emissions and, therefore, requires no off-gas treatment or air permits. The disadvantages are: No full-scale applications exist for this emerging technology. An iron extraction system is needed to remove residual iron from the treated water, which may increase the costs for the system. CONTAMINATED WATER INFLUENT SAMPLE DELIVERY PUMP SURGE TANK NaOH SOLN. 2.5 ph CONTROLLER 7.5 H 2 SO 4 SOLN. ph CONTROLLER TREATED EFFLUENT SAMPLE ph ADJUSTMENT REACTOR RECYLE LOOP ph ADJUSTMENT IRON REMOVAL TO DISTRIBUTION SYSTEM Fe 2 SO 4 / H 2 O 2 SOLN. DELIVERY PUMP Figure 3-5. A schematic of a system utilizing Fenton s Reaction (drawing provided by Komex H2O Science, 1998). 170

193 A very low ph (<2.5) environment is necessary to keep the iron in solution (Potter and Roth, 1993; Mohanty and Wei, 1993; Huling, 1996). Therefore, ph adjustment before and after treatment will be required. The requisite acid and base injections will increase the O&M costs. Pilot/Field Studies and Vendor Information AOPs based on Fenton s Reaction and its associated reactions have been widely studied. Fenton s process has been employed to treat contaminants in drinking water and wastewater (Potter and Roth, 1993; Mohanty and Wei, 1993; Venkatadri and Peters, 1993) and to serve as a pretreatment for biologically recalcitrant contaminants (Koyama et al., 1994; Yeh and Novak, 1995; Huling, 1996). Yeh and Novak (1995) performed some bench-scale studies on MTBE degradation in soil systems. In these studies, the chemical oxidation of MTBE was found to be related to H 2 O 2 concentration, ph, and the presence of ferrous iron, but was found to be independent of the iron concentration (most likely because iron was not limiting). These findings were later confirmed by Chen et al., (1998). The application of Fenton s reaction for MTBE removal has been well studied in bench-scale systems, but has not yet been implemented in pilot or field studies. Calgon Carbon, Inc. (Markham, Ontario, Canada) has a patented AOP system that employs Fenton s chemistry. In this process, the contaminant is adsorbed to a proprietary carbon sorbent, which is regenerated by Fenton s reaction (Huling et al., 1999). In addition, other vendors currently market Fenton s chemistry for remediation of gasoline components. Refer to Table 3-5 for more information on Calgon Carbon, Inc. (Markham, Ontario, Canada). Summary Since Fenton s reaction is an emerging process, it is highly unlikely that it will be used in fullscale drinking water applications in the near future. For Fenton s reaction to be applicable for drinking water treatment, the catalyst (iron or copper) must be attached to a solid matrix. Otherwise, costly iron or copper removal must be performed. Catalyst attachment has not yet been done in a commercial application, other than for the use of Fenton s reaction as a carbon regeneration tool. In addition, ph adjustments and the potential for increased iron concentrations in the finished water suggest that this technology is currently not viable for drinking water treatment. In conclusion, while continued observation of this technology is warranted in a remediation context, it is not recommended for drinking water treatment. 171

194 172

195 3.6 Comparative Discussions of AOPS Permitting As with all drinking water treatment systems, the installation and operation of an AOP system will require multiple state and local construction permits; water, wastewater, and air discharge permits; and/or operational permits. A detailed discussion of all necessary permits is beyond the scope of this document; however, the key permitting issue that differentiates AOPs from other drinking water treatment technologies is the formation of oxidation by-products. Several of the oxidation by-products of MTBE are potential human carcinogens (e.g., formaldehyde and acetaldehyde). In addition, as mentioned previously, the combination of AOPs with preor post-chlorination may increase the formation of THMs or HAA 9 s, which are regulated under the Stage 1 D/DBP Rule. Consequently, whether the regulated compounds are THMs, HAA 9 s, or an oxidation breakdown product of MTBE, strict monitoring requirements will likely be enforced by the governing regulatory agency to ensure that treated water quality does not contain any of these organic secondary contaminants above drinking water standards. To mitigate these concerns, a GAC filter will likely be required to polish the effluent from AOPs. However, complicating this mitigation measure, it is likely that most polishing filters will sustain biological growth (due to the biodegradability of oxidation by-products). Biological processes for drinking water treatment are only now becoming accepted and, thus, the use of a biologically activated filter as a polishing step for an AOP will be under close regulatory scrutiny. In summary, control of AOP by-products will require further technical and regulatory study. Other relevant permitting considerations for AOPs include meeting the following standards: A 1-hour ozone effluent gas concentration of less than 0.12 ppmv according to the Clean Air Act (Code of Federal Regulations Title 40, Part 50) and less than 0.09 ppmv according to the California Code of Regulations Title 17, Section (H 2 O 2 /O 3, O 3 /UV). H 2 O 2 concentrations below 1 mg/l (1.4 mg/m 3 ) according to an OSHA permissible exposure limit (PEL) (NIOSH, 1997). Iron concentrations below 0.3 mg/l according to the SDWA Secondary MCL (Fenton s reaction). ph level between 6.6 and 8.5 according to SDWA Secondary MCL (all AOPs). THMs below 80 µg/l according to Stage 1 D/DBP Rule (all AOPs). HAA 5 below 60 µg/l according to Stage 1 D/DBP Rule (all AOPs). Bromate below 10 µg/l according to Stage 1 D/DBP Rule (H 2 O 2 /O 3, O 3 /UV). 173

196 3.6.2 Flow Rate Most of the AOP reactors that are discussed in the earlier sections are available from the manufacturers for treating waters at some pre-design flows (e.g., 100 gpm or 1,000 gpm). Typical ranges of AOP reactor capacities are shown in Table 3-7. Smaller or larger AOP reactors can be custom built. Most AOPs are modular processes; hence, more than one reactor can be employed in series (to obtain higher retention times) or parallel (to process larger volumes) mode to achieve the desired effluent goals for a given flow rate. Table 3-7 Range of AOP Reactor Capacities and MTBE Removal Efficiencies 174

197 3.6.3 Removal Efficiency Table 3-7 summarizes the reported MTBE removal efficiencies from field and pilot studies. Clearly, removal efficiencies will be a function of operating parameters (e.g., reactor residence time) and water quality parameters (e.g., alkalinity, NOM content). Table 3-7 presents some comparative removal efficiencies reported in the literature for the various processes. In general, higher MTBE removal can be obtained under longer retention times and greater chemical dosages. Refer to sections 3.4 and 3.5 for a more detailed discussion of removal efficiencies that have been observed Other Factors A comparative discussion of each of the AOP technologies relative to their applicability and effectiveness is presented below. Specifically, the AOP technologies are compared with respect to their reliability, flexibility, adaptability, potential for modifications and other related relevant factors. Reliability Reliability of AOP technologies can be addressed under two broad categories namely, process reliability and mechanical reliability. Technologies with fewer moving or replaceable parts are considered to be more mechanically reliable because they will likely require less frequent maintenance. Consequently, the H 2 O 2 /O 3 process receives the highest rating for mechanically reliability. In the H 2 O 2 /O 3 process, periodic checking and cleaning of the ozone generator and ozone gas diffusers is required (Cater, 1999). Fouling of spargers from precipitation of carbonates has been observed at potable water ozonation (disinfection) facilities (Cater, 1999). Sparger fouling can lead to inefficient ozone transfer. Hydrodynamic cavitation contains no moving parts, besides a pump. However, this technology is still a black box and, thus, requires vendor support if problems arise, resulting in a medium rating for mechanical reliability. The UV-based AOP technologies such as O 3 /UV and H 2 O 2 /UV receive a medium rating for mechanical reliability since they require periodic replacement and inspection of UV lamps and quartz sleeves to prevent leakage and scaling. Similarly, E-beam has a large number of specialty parts and equipment requiring experts for maintenance and possible replacement and, thus, receives a low score for mechanical reliability. TiO 2 /UV and Fenton s process also receive a low rating for mechanical reliability due to the required addition of TiO 2 or iron to the reactor. These technologies require significant operational and maintenance oversight in addition to continuous attention to mixing and ph controls. Process reliability for the various AOP technologies, defined as the ability of a given technology to consistently meet effluent goals, varies widely. Established technologies, including H 2 O 2 /O 3, O 3 /UV, and H 2 O 2 /UV, have been proven to consistently meet low effluent goals and are, thus, considered highly reliable. Currently, the lack of large-scale potable water treatment 175

198 applications for E-beam credits it with a medium rating; however, an optimized E-beam system should be able to consistently remove MTBE to below effluent goals. Other emerging technologies, such as cavitation, Fenton s reaction, and UV-catalyzed TiO 2, receive the lowest rating for process reliability due to their untested nature in drinking water applications and the secondary chemicals used for treatment that subsequently require removal (e.g., precipitated iron, TiO 2 slurry). For all AOP technologies, including those with high reliability ratings, monitoring and controls are recommended to optimize the treatment process. Flexibility Flexibility is defined as the ability of a technology to handle wide fluctuations in the influent water flow rate following design and installation. Occasionally, during the operation of a treatment process, the influent stream flows increase or decrease significantly compared to the design flow. A flexible technology should be able to handle these fluctuations with no major impact on the treatment process outcome. When designed with sufficient safety factors, established AOPs (H 2 O 2 /O 3, O 3 /UV, and H 2 O 2 /UV) can handle a large turndown ratio (i.e., ratio of maximum to minimum allowable flow rates). In addition, chemical additions or UV dose can be changed to respond to changing flow rates. These technologies receive a high rating. Hydrodynamic cavitation is known to perform better at higher (2000 gpm) flow rates compared to smaller (100 gpm) flow rates, and, thus, performance is expected to decrease if flow rates fall, suggesting a low rating. E-beam may require significant changes to its stream distribution or spreading system when the water flow rates increase, again suggesting a low rating. Finally, UV/TiO 2 and Fenton s reactions are likely performed in semi-batch reactors that can handle changes in flow rates, suggesting a medium rating; however, there is still significant uncertainty related to the design of these reactor systems. Flexible technologies, when necessary, can also be scaled up with little or no difficulty. The capacities of the modular AOP technologies, such as O 3 /H 2 O 2, O 3 /UV, H 2 O 2 /UV, UV/TiO 2, Fenton s reaction, and hydrodynamic cavitation, can be expanded by adding additional reactors either in series (to extend the reaction time) or in parallel (to increase flow rates). Adaptability In this report, adaptability of a technology is defined as its ability to handle fluctuations in water quality conditions, such as influent contaminant concentrations, hardness, alkalinity, and turbidity. All the AOP technologies discussed previously can achieve MTBE removal efficiencies that are independent of the influent MTBE concentration, but that vary widely with water quality conditions. If influent MTBE concentrations increase while effluent goals remain unchanged, it will be necessary to increase the contact time or oxidant doses in the reaction chamber to meet effluent goals. Since oxidation via hydroxyl radicals is the predominant mechanism for MTBE removal for each of the AOPs discussed above, the presence of radical scavengers will affect treatment 176

199 performance, independent of the selected AOP. However, those technologies that generate a larger number of hydroxyl radicals more rapidly will be less affected by the presence of radical scavengers. Thus, hydrodynamic cavitation, TiO 2 /UV, and E-beam, which rapidly generate a large number hydroxyl radicals due to the multiple oxidizing and reducing species introduced, receive a high rating. Alternatively, the removal efficiency of UV-based technologies, such as O 3 /UV, H 2 O 2 /UV, and TiO 2 /UV, is hindered by water quality parameters other than radical scavengers (e.g., excess turbidity [which masks the penetration of the UV light], the presence of nitrate [which absorbs effective UV radiation], and iron and other fouling agents [which scale the quartz sleeves]). Thus, O 3 /UV and H 2 O 2 /UV receive a low rating and TiO 2 /UV is reduced to medium rating. Finally, the effectiveness of H 2 O 2 /O 3 and O 3 /UV can be reduced by the presence of excess particulate matter or scaling parameters that foul the ozone gas diffusers these technologies, in addition to Fenton s reaction, receive a medium rating. The removal efficiency of each AOP technology is strongly dependent on the characteristics of the influent water quality. Hence, in the design of AOP systems, due consideration must be given to the concentrations (and expected fluctuations) of radical scavengers and other interfering compounds. Potential for Modifications The potential for modifications is defined as the ability to alter the installed system including the addition of any necessary pre- and post-treatments processes to accommodate changes in the design criteria and conditions (e.g., lowered target concentrations of MTBE, removal of high alkalinity or iron, by-products polishing). For example, most AOPs that treat source waters with medium to high (>100 mg/l) alkalinities may require pretreatment for alkalinity removal, which may include a ph adjustment step followed by CO 2 stripping. Most modular processes (e.g., H 2 O 2 /O 3, O 3 /UV, H 2 O 2 /UV, hydrodynamic cavitation, TiO 2 /UV, and Fenton s reaction) are more easily amenable to changes compared to non-modular processes (e.g., E-beam). Also, in modular processes, several modular units in parallel or series can supply additional contact time. When necessary, all of the AOP technologies evaluated can be supplemented with pre- and post-treatment systems. However, for some AOPs, these pre- or post-treatment systems are mandatory prior to drinking water distribution and, thus, these AOPs will receive a lower rating. For example, Fenton s reaction and TiO 2 /UV require post-treatment for removal of iron and TiO 2 from drinking water, resulting in a low rating. TiO 2 /UV systems also require pretreatment for removal of metal ions and addition of DO. For waters with high bromide concentrations (>100 µg/l), control of bromate formation will be necessary in ozone-based AOP systems (e.g., H 2 O 2 /O 3 and O 3 /UV), although this effect can be mitigated without pretreating, as mentioned previously. In addition, ozone-based processes require ozone off-gas treatment, resulting in a low rating. H 2 O 2 /UV processes may require a post-treatment temperature adjustment for small systems or pre-treatment to remove turbidity, nitrates, or scaling agents; however, these pre- and post-treatments are not always necessary, suggesting a medium rating. Similarly, hydrodynamic cavitation will likely require the use of additional 177

200 oxidants, such as ozone or hydrogen peroxide, to effect MTBE removal, suggesting a medium rating. E-beam requires no pre- or post-treatment processes, resulting in a high rating. Other Design and Implementation Factors In addition to reliability, flexibility, adaptability, and potential for modifications, there are other factors that could favor a specific AOP technology. Table 3-8 presents a comparison of AOP technologies with respect to other essential decision driving factors, including bromate formation potential, energy usage, costs, public acceptability, and ease of implementation. Bromate Regulatory Compliance. Bromate is classified by the International Agency for Research on Cancer (IARC) as a possible human carcinogen and is strictly regulated under the Stage 1 D/DBP Rule with a maximum contaminant level of 10 µg/l. This MCL may become more stringent in future rulemaking. Thus, AOPs that generate bromate (O 3 -based processes) receive a lower rating than alternatives; however, as discussed previously, bromate formation can be mitigated by varying chemical doses for the H 2 O 2 /O 3 process. Energy Efficiency. Energy usage is rated low for systems that use a combination of O 3 and UV light and is rated medium for AOPs that are based on either O 3 or UV alone or in combination with other oxidants. Fenton s reaction does not require electrical energy beyond the feed pumps, resulting in a high energy efficiency rating. The energy requirements for cavitation processes are stated to be comparable to those of the UV systems, suggesting a medium rating. E-beam requires significant energy for operation, resulting in a low rating. Public Acceptability. Systems that are widely used in remediation and drinking water treatment applications are rated as highly acceptable to the public whereas emerging AOPs with little or no field applications are classed as medium acceptability. E-beam is given a low rating for public acceptability due to its reliance on a radiation source for contaminant removal, which has received significant public criticism for use in the food industry. Fenton s reaction and TiO 2 /UV are also rated low due to the required addition of inorganic materials (i.e., iron and TiO 2 ) to the water. Finally, cavitation is given a low rating due to the industry s reluctance to install black box technologies for drinking water applications. Ease of Implementation. The number of field installations was used as a surrogate to determine the ease of implementation. Accordingly, the AOPs that use some combination of O 3, UV, and H 2 O 2 were rated high for ease of implementation whereas emerging AOPs such as Fenton s reaction and TiO 2 /UV with no field applications were given a low grade. E-beam and cavitation were given a medium rating due to the presence of a limited number of fieldand pilot-scale treatment systems. 178

201 Table 3-8 Comparative Analysis of Various AOPs 179

202 180

203 3.7 Cost Evaluation Cost estimates were developed to allow direct comparison among the various AOPs and for comparison with the costs developed for air stripping (Chapter 2), GAC (Chapter 4), and synthetic resin sorbents (Chapter 5). The costs for AOPs are highly dependent on the quality of the source water to be treated and effluent treatment goals. The cost comparison developed in this report should be used as a guideline. An understanding of actual costs will require pilot testing to determine site specific costs. As discussed in the previous section, cost is only one factor in the selection of an AOP, and other considerations may result in selection of an AOP that is not the most cost effective. In addition to those factors listed in Section 3.6, one should consider treatment plant location, duration of treatment required, environmental concerns, community impacts, and other considerations identified through the preparation of an initial study in compliance with the National Environmental Permitting Act (NEPA) or the California Environmental Quality Act (CEQA) Overall Costs of AOP Systems In order to compare the costs of the various AOPs, AOP equipment vendors were provided with a number of treatment scenarios and asked to provide costs for equipment, chemical dosages, electricity, and replacement parts. Four vendors provided detailed information to assist in this cost evaluation. 1. Calgon Carbon Corporation (Calgon) H 2 O 2 /MP-UV system 2. Applied Process Technology, Inc. (APT) H 2 O 2 /O 3 system 3. Oxidation Systems, Inc. (OSI) Hydrodynamic cavitation with H 2 O 2 4. Hydroxyl Systems, Inc. (HSI) TiO 2 -catalyzed/h 2 O 2 Other vendors, including Magnum Water Technologies (H 2 O 2 /MP-UV) and Calgon Carbon Corporation (Fenton s Reaction), also participated; however, these costs could not be verified with sufficiently detailed information or field data. Consequently, these costs are not included in this evaluation. In addition, Haley and Aldrich (E-beam system) provided cost estimates that suggest that E-beam may be cost-competitive with other AOPs. These costs were not included due their high degree of uncertainty resulting from the emerging nature of this technology for drinking water applications. The cost evaluation consisted of several treatment scenarios to evaluate a typical range of drinking water well production rates, MTBE influent concentrations, and effluent treatment goals. Influent flows of 60, 600, and 6,000 gpm. Influent MTBE concentrations of 20, 200, and 2,000 µg/l. Effluent MTBE discharge requirements of 20, 5, and 0.5 µg/l. 181

204 The vendors were provided with the following influent water characteristics: Hardness: 200 mg/l as CaCO 3 Alkalinity: 250 mg/l as CaCO 3 Bromide: ND Iron: <1 mg/l ph: 7.0 Temperature: 65 F TDS: 500 mg/l Nitrate: 25 mg/l as NO 3 or 5 mg/l as N One important issue in comparing AOPs is the formation and control of oxidation byproducts. Most of the vendors did not adequately identify or estimate the formation of byproducts, such as acetone, methyl acetate, formaldehyde, acetic acid, formic acid, pyruvic acid, oxalic acid, H 2 O 2, TBA, and TBF. Therefore, to facilitate the comparison of these AOPs with other drinking water treatment technologies, supplemental costs for biologically activated carbon polishing were developed for each AOP for removal of oxidation byproducts. Capital costs for this system are based on a Calgon GAC system using Filtrasorb 600 carbon. Operational costs were estimated to be similar to those identified in Chapter 4. Carbon replacement costs were estimated, but are difficult to predict due to the biological nature of this polishing process. Water quality may dictate some carbon changeouts, based on adsorption of contaminants onto the carbon. Relevant assumptions and costs for these polishing systems are included in Table 3-9. Table 3-10 provides a sample calculation of total capital costs, summary of annual costs, total annual costs, and unit treatment costs. As this table indicates, the capital costs provided by the vendors were used as the bases for estimating the complete installed system costs. Piping, valves, and electrical work was estimated at 30 percent of the system equipment costs. Site work was estimated at 10 percent of equipment costs, engineering was estimated at 15 percent of equipment costs, and contractor O&P was estimated at 15 percent of equipment costs. A contingency of 20 percent of the total costs was then added. Table 3-11 presents a summary of the capital costs for the four AOP technologies evaluated under the scenarios identified. The capital costs include the complete treatment system and installation. Table 3-12 presents a summary of the annual O&M costs for the four AOP technologies evaluated under the same scenarios described above. The O&M costs consist of replacement parts, labor costs, analytical costs, chemical costs, and electrical costs. The replacement part costs are based on vendor estimates and include replacement parts, such as UV lamps, and spare parts. Some of the vendors have estimated the replacement costs based on a percentage 182

205 of the capital cost of the equipment. The labor costs include labor for sampling of water, system operation and general maintenance for the specific type of AOP. The maintenance and sampling labor rate used was $80/hr. Analytical costs are based on weekly sampling of the influent and the effluent from each reactor and are estimated at $200 per sample. Chemical costs include H 2 O 2, O 3, and TiO 2 as they apply to the technology and were provided by the vendor (both dosage and costs). The electrical costs were based on power consumption and was estimated by the vendors based on $0.08/kWh. Details of the O&M costs are included in tables in Appendix 3A. Replacement part costs are presented in Table 3A-1, labor costs in Table 3A-2, analytical costs in Table 3A-3, chemical costs in Table 3A-4, and electrical costs in Table 3A-5. Labor costs are further broken down by technology and are presented in Tables 3A-6 to Tables 3A-9 in Appendix 3. Note that these costs do not include the polishing treatment required for removal of oxidation by-products. Amortized annual capital costs and annual O&M costs were combined to determine the total amortized operating costs for each system per 1,000 gallons of treated water as presented in Table The equipment was amortized at a discount rate of seven percent over a 30-year period. 183

206 Table 3-9 Cost of Hydrogen Peroxide Removal and Oxidation By-product Removal H 2O 2 Treatment Systems (Centaur Carbon) Flow Rate Capital Costs Annual O& M Costs Total Amortized Operating Costs Gallons Per Minute $ $ $/1,000 Gallons 60 $ 18,000 $ 39, $ 89,000 $ 44, ,000 $ 667,000 $ 194, Notes: 60 gpm system consists of one 500 pound GAC vessel. 600 gpm system consists of one 5,000 pound GAC vessel. 6,000 gpm system consists of ten 5,000 pound GAC vessels. O & M Costs Include: Capital Costs include: Carbon replacement Equipment (Costs provided by vendor) Analytical sampling Piping, valves, electrical (30%) Oversight during changeouts Site work (10%) General system O & M Contractor O & P (15%) Engineering (15%) Contingency (20%) Treatment for TBA, TBF, Acetone, & other By-products (Bio-GAC) Flow Rate Capital Costs Annual O& M Costs Total Amortized Operating Costs Gallons Per Minute $ $ $/1,000 Gallons 60 $ 16,700 $ 42, $ 67,800 $ 42, ,000 $ 667,000 $ 204, Notes: 60 gpm system consists of one 1,000 pound GAC vessel. 600 gpm system consists of one 10,000 pound GAC vessel. 6,000 gpm system consists of five 20,000 pound GAC vessels. O & M Costs Include: Capital Costs include: Carbon replacement Equipment (Costs provided by vendor) Analytical sampling Piping, valves, electrical (30%) Oversight during changeouts Site work (10%) General system O & M Contractor O & P (15%) Engineering (15%) Contingency (20%) 184

207 Table 3-10 Costs of H 2 O 2 /O 3 System for MTBE Removal (Applied Process Technology, Inc.) ITEM 1 Advanced Oxidation Unit COST $750,000 Piping, Valves, Electrical (30%) $225,000 Site Work (10%) $75,000 SUBTOTAL $1,050,000 Contractor O&P (15%) $157,500 SUBTOTAL $1,207,500 Engineering (15%) $181,125 SUBTOTAL $1,388,625 Contingency (20%) $277,725 TOTAL CAPITAL $1,666,400 AMORTIZED CAPITAL 1A $134,290 ANNUAL O&M $129,692 TOTAL ANNUAL COST $263,981 TOTAL COST PER 1,000 GALLONS TREATED $0.84 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Parts 2 Lump sum 1 $11,250 $11,250 Labor 3 Hour 724 $80 $57,920 Analytical costs 4 Sample 208 $200 $41,600 Chemical costs 5 $/1000 gal $0.03 $9,461 Power ($0.08/kWh) 6 kwh $0.08 $9,461 ANNUAL O&M $129,692 System Parameters: 600 gpm 200 µg/l influent MTBE concentration 20 µg/l effluent MTBE concentration 1 Cost of oxidation unit from vendor. 1A Amortization based on 30 year period at 7% discount rate. 2 Replacement is based on vendor s estimate of 1.5% capital cost. 3 Breakdowns of labor costs are given in Tables 3A-6 to 3A-9 in Appendix 3, based on a rate of $80/hr. 4 Sampling conducted weekly at 4 locations. 5 Chemical costs based on dosages and prices estimated by vendor. 6 Power is based on consumption estimates provided by vendor, priced at $0.08/kWh. 185

208 186 Costs exclude polishing treatment. Refer to Table 3-9 for additional treatment costs. Capital Costs include: Equipment Costs (provided by vendor) Piping, valves, electrical (30%) Site work (10%) Contractor O&P (15%) Engineering (15%) Contingency (20%) Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 177,700 $ 444,400 $ 134,400 $ 277, % $ 188,900 $ 444,400 $ 242,800 $ 426, % $ 177,700 $ 444,400 $ 156,600 $ 426, % $ 188,900 $ 444,400 $ 242,800 $ 439, % $ 188,900 $ 533,200 $ 260,000 $ 586, % $ 188,900 $ 533,200 $ 260,000 $ 586, % $ 188,900 $ 533,200 $ 260,000 $ 517, % $ 266,600 $ 622,100 $ 260,000 $ 691, % $ 266,600 $ 1,666,400 $ 356,600 $ 1,142, % $ 488,800 $ 1,777,400 $ 461,200 $ 1,344, % $ 337,700 $ 1,666,400 $ 356,600 $ 1,344, % $ 488,800 $ 1,777,400 $ 461,200 $ 1,730, % $ 695,400 $ 1,777,400 $ 792,100 $ 2,035, % $ 695,400 $ 1,777,400 $ 482,100 $ 2,035, % $ 811,000 $ 1,888,500 $ 482,100 $ 2,628, % $ 1,299,800 $ 1,888,500 $ 482,100 $ 3,092, % $ 999,800 $ 7,998,500 $ 1,446,400 $ 9,711, % $ 2,666,200 $ 8,887,200 $ 4,151,000 $ 9,711, % $ 1,666,400 $ 7,998,500 $ 3,209,400 $ 11,426, % $ 3,332,700 $ 8,887,200 $ 4,151,000 $ 14,703, % $ 4,221,400 $ 8,887,200 $ 4,339,200 $ 17,298, % $ 4,999,000 $ 8,887,200 $ 4,339,200 $ 17,298, % $ 6,665,400 $ 8,887,200 $ 4,339,200 $ 22,344, % $ 9,998,100 $ 9,775,900 $ 4,339,200 $ 26,288,300 Table 3-11 Captial Costs of AOPs

209 187 Costs exclude polishing treatment. Refer to Table 3-9 for additional treatment costs. O&M Costs include: Replacement Parts (Based on vendor s estimate). Labor costs at $80/hr. Analytical costs for sampling conducted weekly at $200 per sample. Chemical costs based on dose and price estimated by vendor. Power based on consumption estimates provided by vendor, priced at $0.08/kWh. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 54,400 $ 47,162 $ 60,200 $ 74, % $ 63,700 $ 48,800 $ 62,900 $ 78, % $ 58,900 $ 47,700 $ 60,400 $ 79, % $ 63,700 $ 48,800 $ 62,900 $ 80, % $ 70,600 $ 50,900 $ 66,300 $ 90, % $ 81,600 $ 60,800 $ 71,700 $ 100, % $ 94,200 $ 61,500 $ 74,600 $ 95, % $ 108,000 $ 63,900 $ 75,200 $ 107, % $ 157,800 $ 123,400 $ 167,700 $ 265, % $ 248,300 $ 139,900 $ 174,500 $ 277, % $ 198,200 $ 129,800 $ 167,700 $ 280, % $ 264,000 $ 139,900 $ 174,500 $ 330, % $ 343,500 $ 155,700 $ 208,100 $ 349, % $ 422,000 $ 193,300 $ 202,700 $ 373, % $ 487,700 $ 203,500 $ 233,100 $ 452, % $ 551,500 $ 222,500 $ 239,400 $ 483, % $ 930,600 $ 464,500 $ 1,101,900 $ 2,389, % $ 1,429,600 $ 628,100 $ 1,177,200 $ 2,515, % $ 1,190,400 $ 527,500 $ 1,109,800 $ 2,535, % $ 1,631,800 $ 628,100 $ 1,177,200 $ 3,103, % $ 1,989,000 $ 785,900 $ 1,512,900 $ 3,197, % $ 1,657,400 $ 1,061,700 $ 1,359,300 $ 3,431, % $ 3,386,100 $ 1,156,400 $ 1,662,700 $ 4,346, % $ 4,210,900 $ 1,351,600 $ 1,725,800 $ 4,504,000 Table 3-12 Annual O&M Costs of AOPs

210 188 Costs exclude polishing treatment. Refer to Table 3-9 for additional treatment costs. Note: Amortized at 7% over 30 years. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 2.18 $ 2.63 $ 2.25 $ % $ 2.50 $ 2.68 $ 2.61 $ % $ 2.32 $ 2.65 $ 2.32 $ % $ 2.50 $ 2.68 $ 2.61 $ % $ 2.72 $ 2.98 $ 2.77 $ % $ 3.07 $ 3.29 $ 2.94 $ % $ 3.47 $ 3.31 $ 3.03 $ % $ 4.11 $ 3.62 $ 3.05 $ % $ 0.57 $ 0.82 $ 0.62 $ % $ 0.91 $ 0.90 $ 0.67 $ % $ 0.71 $ 0.84 $ 0.62 $ % $ 0.96 $ 0.90 $ 0.67 $ % $ 1.27 $ 0.95 $ 0.86 $ % $ 1.52 $ 1.07 $ 0.77 $ % $ 1.75 $ 1.13 $ 0.86 $ % $ 2.08 $ 1.19 $ 0.88 $ % $ 0.32 $ 0.35 $ 0.39 $ % $ 0.52 $ 0.43 $ 0.48 $ % $ 0.42 $ 0.37 $ 0.43 $ % $ 0.60 $ 0.43 $ 0.48 $ % $ 0.74 $ 0.48 $ 0.59 $ % $ 0.65 $ 0.56 $ 0.54 $ % $ 1.24 $ 0.59 $ 0.64 $ % $ 1.59 $ 0.68 $ 0.66 $ 2.10 Table 3-13 Total Amortized Operating Costs (per 1,000 Gallons Treated) for AOPs

211 3.7.2 Evaluation of Cost Estimates for Specific AOP Technologies For nearly all of the cost estimates provided by vendors, the primary factors affecting system costs were flow rate and removal efficiency, independent of influent concentration. For example, the cost to reduce MTBE from 20 µg/l to 0.5 µg/l (97.5-percent reduction) at 6,000 gpm was nearly identical to the cost to reduce MTBE from 200 µg/l to 5 µg/l (97.5-percent reduction). However, as expected, the O&M costs increased substantially as the removal efficiency exceeded 99.9 percent. Capital costs for both Calgon and OSI were lower than APT and HSI for all of the flow rates. However, for high influent concentrations (2,000 µg/l) and high removal efficiencies (99.98 percent) capital costs increased substantially for Calgon. O&M costs were lowest for APT with Calgon and OSI O&M costs approximately 50 percent higher for the 60 and 600 gpm systems. HSI O&M costs were significantly higher under all but one flow rate and OSI O&M costs were significantly higher at 6,000 gpm. Combining capital and O&M, annual operating costs were the lowest for APT, Calgon, and OSI, ranging from approximately $2.18/1,000 gallons at 60 gpm to $0.32/1,000 gallons at 6,000 gpm. Again, it should be noted that these costs are intended for estimating purposes only and should not be used in place of sitespecific engineering cost estimates. Many assumptions were made to facilitate an equal comparison; however, these assumptions may not necessarily be accurate for each technology. For example, the application of standard multipliers for piping, valves, electrical, site work, engineering, and contractor O&P may not accurately reflect the actual costs of the system, but allowed for a more uniform comparison. The following is a detailed discussion of each of the cost estimates provided by vendors/ manufacturers: Applied Process Technology, Inc. (H 2 O 2 /O 3 ) Cost estimates provided by APT were for their H 2 O 2 /O 3 system. The costs per 1,000 gallons of water treated ranged between $0.35 (6,000 gpm, 20 µg/l) and $3.62 (60 gpm, 2,000 µg/l). These cost figures represent some of the lowest costs collected from any of the four vendors for this analysis. Although the capital costs (Table 3-11) for this system are significantly higher than those for Calgon and OSI, the lower operations and maintenance costs (Table 3-12), particularly with regard to chemical (Table 3A-4) and electrical (Table 3A-5) costs, make this system cost-competitive in terms of total amortized unit costs. Furthermore, under many circumstances, APT capital costs are expected to be lower since they provide a packaged treatment system that comes complete with piping, valves, electrical, and engineering. Thus, actual cost multipliers would be expected to be lower than the standard numbers applied in this report, making the APT system even more cost-effective than shown in the tables. 189

212 The cost estimates prepared by APT were based on effluent water with by-product formation, specifically TBA, TBF, and acetone, estimated at approximately 10 percent of the MTBE influent concentration for effluent MTBE treatment goals of 20 µg/l, and 5 µg/l (Applebury, 1999). When the effluent goal of MTBE is 0.5 µg/l, the applied ozone and peroxide doses were high enough to eliminate nearly all formation of TBA or TBF; however, acetone is still expected to be produced in the effluent water at about 10 percent of the MTBE influent concentrations (Applebury, 1999). APT has performed numerous pilot tests that confirm these results. APT estimates minimal peroxide residual due to the unique dosing mechanism (see Figure 3-1b) and, thus, the biologically activated polishing filter required for removal of oxidation by-products (see Table 3-9) is expected to be capable of reducing peroxide concentrations to non-detect levels. Calgon Carbon Corporation (H 2 O 2 /MP-UV) The cost estimates provided by Calgon Carbon Corporation were for their H 2 O 2 /MP-UV system. The cost per 1,000 gallons of treated water ranged from $0.32 (6,000 gpm, 20 µg/l) and $4.11 (60 gpm, 2,000 µg/l). Calgon had among the lowest capital costs, but O&M costs were higher than for APT or OSI for the 6,000 gpm system. The costs prepared by Calgon were based on meeting the specified effluent concentration of MTBE. However, by-products produced as a result of the oxidation process would require further treatment to meet drinking water standards. Calgon provided the most complete analyses on by-product formations and quantified by-product formation based on the data extrapolated from an actual study and provided estimates for the 600 gpm scenario (see Table 3-14). In addition, Calgon calculated the hydrogen peroxide residual remaining in the treated water. Because these concentrations are high (>10 mg/l), an additional treatment step will be required for H 2 O 2 removal. Calgon recommended using Centaur carbon for removal of the excess H 2 O 2 and a biologically activated carbon system for removal of the TBF, TBA, acetone, formaldehyde, and other acids prior to distribution of the treated drinking water. Costs for these two polishing systems were presented in Table

213 191 Estimates Provided by Calgon Carbon Corporation (Philadelphia, PA) Initial MTBE Concentration Final MTBE Concentration % Removal 99.00% 99.75% 99.98% 90.00% 97.50% 99.75% 75.00% 97.50% System Size, kw UV Dose, kwh/1,000 Gallons H Dose, mg/l By-product Formation/Residuals, µg/l, in the effluent TBF TBA Acetone Methyl Acetate Formaldehyde Acetic Acid Formic Acid Pyruvic Acid Oxalic Acid Total By-products, µg/l Residual H 2 O 2, mg/l Table 3-14 Estimated By-product Formations and Residual Oxidant Concentrations for H 2 O 2 /MP-UV (Calgon Carbon Corporation)

214 Oxidation Systems, Inc. (Hydrodynamic Cavitation with H 2 O 2 ) Oxidation Systems Inc. (OSI) provided cost estimates for hydrodynamic cavitation combined with H 2 O 2. Based on their cost figures, this AOP is comparable to the systems offered by Calgon or APT. Costs per 1,000 gallons of treated water range between $0.39 (6,000 gpm at 20 µg/l) and $3.05 (60 gpm at 2,000 µg/l). At high flows (600 gpm and 6,000 gpm) and high influent MTBE concentration (2,000 µg/l), this system had the lowest capital cost. This technology is expected to produce oxidation by-products as a result of incomplete oxidation. However, there is limited field information available to adequately estimate by-product formations and to confirm estimates by OSI. Phase II field-testing by OSI is expected to begin in 2000 and should address oxidation by-product formation and control. Regardless, this technology is expected to require a polishing system such as a biologically activated carbon system for removal of AOP by-products (see Table 3-9 for costs). Hydroxyl Systems, Inc. (TiO 2 -catalyzed UV) Hydroxyl Systems Inc. (HSI) provided cost estimates for TiO 2 -catalyzed UV. Based on these estimates, this process is less economical than the other AOPs evaluated. Costs per 1,000 gallons of treated water range between $1.01 (6,000 gpm at 20 µg/l) and $5.17 (60 gpm at 2,000 µg/l). Relative to the other AOPs, capital costs were the highest, particularly at the higher flow rates and influent MTBE concentrations. O&M costs were also high, although not significantly higher than for Calgon. There is limited information about by-product formation and, thus, vendor claims regarding by-product control currently cannot be verified. This technology is expected to require a polishing system such as a biologically activated carbon system for removal of AOP by-products (see Table 3-9 for costs) Sensitivity Analysis The presence of other chemical constituents in the source water will affect the performance and economics of AOPs. The constituents of concern are common gasoline aromatics, BTEX, and dissolved NOM expressed as TOC. A sensitivity analysis was performed to evaluate the impacts of BTEX and TOC on AOP drinking water treatment costs. In addition, the costs presented above are based on a treatment plant life of 30 years, as is standard for community drinking water treatment plants. However, some of the smaller treatment applications may be installed for a much shorter period and, thus, a sensitivity analysis was completed to evaluate the effect of facility lifetime on AOP drinking water treatment costs. All sensitivity analyses performed were based on information supplied by vendors and engineering judgement. Actual costs will vary depending on site-specific circumstances. The concentration of TOC in groundwater varies; therefore, TOC concentrations of 0.8 mg/l, 2 mg/l, and 8 mg/l were evaluated. Although BTEX is not expected to be detected in large community drinking water supplies due to the reliance of these supplies on deep aquifers, BTEX compounds are likely to be present in shallow aquifers contaminated with gasoline. 192

215 Thus, BTEX sensitivity analyses based on concentrations of 800 µg/l and 80 µg/l were completed to evaluate the impacts of BTEX on AOP treatment costs. Finally, cost estimates for 2-, 10-, and 30-year treatment facility lifetimes were also completed. The sensitivity analyses of AOPs were evaluated for the following base case: Flow rate of 600 gpm Influent MTBE of 200 µg/l Effluent MTBE of 5 µg/l Results of the sensitivity analyses are as follows: TOC Sensitivity The capital, operating, and total cost per 1,000 gallons of treated water are summarized in Table For three of the AOPs (H 2 O 2 /O 3 system [APT], H 2 O 2 /MP-UV system [Calgon], and hydrodynamic cavitation [OSI]), the vendors claimed that the levels of TOC included in the evaluation would not affect the capital cost (i.e., the size of the reactor would not be affected). In the fourth case, TiO 2 /UV, the vendor said that capital cost would significantly increase, as the TOC is expected to foul the catalyst and absorb some of the UV light; hence, requiring more lamps, a larger reactor, and more catalyst. Under high TOC concentrations, O&M costs are expected to increase for all systems with few exceptions. For the H 2 O 2 /O 3 (APT) and H 2 O 2 /MP-UV (Calgon) systems, TOC levels of 0.8 and 2 mg/l are not expected to increase O&M costs. The vendors claim that these levels of TOC do not significantly interfere with UV light or scavenge hydroxyl radicals in their design, and they have several field tests that support their claims. In the case of hydrodynamic cavitation, elevated levels of TOC will require greater operator maintenance and energy due to increased recycling. Elevated TOC levels are expected to have the greatest impact on the TiO 2 /UV system and make this technology cost prohibitive. Elevated TOC levels in these systems will necessitate more catalyst change-outs, more frequent reactor and lamp cleanings, and increased H 2 O 2 consumption. BTEX Sensitivity The capital, operating, and total cost per 1,000 gallons of treated water are summarized in Table 3-16 for this sensitivity analysis. In the case of H 2 O 2 /O 3 (APT), these levels of BTEX are expected to have no impact on cost at 80 µg/l, and only a slight impact (increase of $0.04/1,000 gallons treated) at 800 µg/l. Field data is available from APT to support their claim. In the case of H 2 O 2 /MP-UV (Calgon), BTEX at 80 µg/l is not expected to increase costs; however, BTEX at 800 µg/l is expected to increase costs by approximately 17 percent. In the case of hydrodynamic cavitation, elevated BTEX is not expected to increase capital cost but will increase O&M costs. For this AOP, 80 µg/l BTEX are expected to increase 193

216 O&M cost 10 percent while 800 µg/l BTEX are expected to increase O&M costs 25 percent. The vendor has demonstrated pilot units successfully for MTBE removal with complete BTEX removal at these concentrations. Finally, in the case of TiO 2 /UV, elevated BTEX is expected to increase capital and O&M costs. The increase in cost per 1,000 gallons treated is expected to be approximately 7 percent for 80 µg/l BTEX and 30 percent for 800 µg/l BTEX. However, there is currently no field data to support these assumptions. Design Life Sensitivity The results of the sensitivity analyses on the design life of the treatment system is presented in Table As can be expected, shortening the design life of these systems is expected to result in higher amortized capital costs. Reducing the design life from 30 years to 2 years, while maintaining a seven percent discount rate, results in an approximate doubling of the unit costs for the Calgon system ($0.96 to $1.69/1,000 gallons) and for the OSI system ($0.67 to $1.36/1,000 gallons). For APT and HSI, the costs increased even more significantly by reducing the life from 30 years to 2 years. The greater difference is attributed to the higher system capital costs. APT costs increased approximately fourfold ($0.90 to $3.56/1,000 gallons) while the HSI costs increased by almost threefold ($1.49 to $4.08/1,000 gallons). Table 3-15 Effect of TOC on AOP Treatment Costs TOC Concentration (mg/l) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc. Capital Cost 0.8 $488,800 $1,777,400 $461,200 $1,730,800 2 $488,800 $1,777,400 $461,200 $1,730,800 8 $488,800 $1,777,400 $461,200 $1,730,800 Annual Operation and Maintenance Costs 0.8 $264,000 $136,800 $185,500 $391,000 2 $264,000 $139,900 $204,400 $542,200 8 $316,900 $149,300 $229,600 $831,300 Total Amortized Operating Cost (Per 1,000 gallons Treated) 0.8 $0.96 $0.89 $0.71 $ $0.96 $0.90 $0.77 $ $1.13 $0.93 $0.85 $3.08 Base case is a 600 gpm system treating an influent of 200 µg/l MTBE with an effluent of 5 µg/l. 194

217 Table 3-16 Effects of Additional BTEX Contamination on AOP Treatment Costs BTEX Concentration (µg/l) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc. Capital Cost 80 $488,800 $1,777,400 $461,200 $1,903, $488,800 $1,777,400 $461,200 $2,423,100 Annual Operation and Maintenance Costs 80 $264,000 $139,900 $193,300 $347, $309,300 $139,900 $213,300 $413,000 Total Amortized Operating Cost (per 1,000 gallons Treated) 80 $0.96 $0.90 $0.73 $ $1.11 $0.90 $0.79 $1.93 Base case is a 600 gpm system treating an influent of 200 µg/l MTBE with an effluent of 5 µg/l. Table 3-17 Effects of Design Life on AOP Treatment Costs Design Life Years Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc. Capital costs 2 $488,800 $1,777,440 $461,200 $1,730, $488,800 $1,777,400 $461,200 $1,730, $488,800 $1,777,440 $461,200 $1,730,800 Annual Operation and Maintenance Cost 2 $264,000 $139,900 $174,400 $330, $264,000 $139,900 $174,400 $330, $264,000 $139,900 $174,400 $330,500 Total Amortized Operating Cost (Per 1,000 gallons Treated) 2 $1.69 $3.56 $1.36 $ $1.06 $1.25 $0.76 $ $0.96 $0.90 $0.67 $1.49 Base case is a 600 gpm system treating an influent of 200 µg/l MTBE with an effluent of 5 µg/l. 195

218 196

219 3.8 Conclusions and Recommendations for Future Research Recommended Technologies When compared to other drinking water treatment alternatives, such as air stripping and activated carbon, AOPs are an emerging technology. Currently, there are only a few cases where organic contaminants (e.g., PCE and NDMA) are being removed from drinking water using an AOP. Furthermore, there were no identified cases where MTBE is being removed from drinking water prior to distribution. Thus, thorough pilot- and field-scale testing of the selected AOP is required to demonstrate the capabilities and possible limitations of AOPs to produce drinking water from contaminated source water. Based on this evaluation, the two most promising AOP technologies appear to be H 2 O 2 /O 3 and H 2 O 2 /MP-UV. Both of these processes are well-understood and have been demonstrated at several bench- and field-scale sites to successfully remove MTBE from water to meet drinking water standards. Besides being the most technically feasible, these two technologies in addition to cavitation appear to be the most economically feasible. However, these costs are strongly dependent on source water quality and are difficult to verify due to the untested nature of these technologies in large-scale applications. Cavitation costs involve the most uncertainty because there are no pilot-, field-, or full-scale drinking water treatment applications for MTBE removal. Consequently, while there is significant uncertainty for all the cost estimates, H 2 O 2 /O 3 and H 2 O 2 /MP-UV technologies are essentially equivalent in cost and less expensive than the other AOPs evaluated. In addition to these two relatively established AOPs, E-beam and cavitation are two emerging AOPs that warrant future consideration due to their technical feasibility for removing MTBE from drinking water to meet standards. These technologies are still in their infant stages for removal of organic contaminants in drinking water applications; however, they have been widely demonstrated for disinfection and remediation applications Recommendation for Future Research As stated previously, there remains a significant amount of uncertainty regarding the technical and economic effectiveness of AOPs for removing MTBE from drinking water under a variety of water quality scenarios. More pilot- and field-scale studies need to be conducted to determine the removal efficiencies that can be achieved under different water quality conditions and operational parameters. In addition, the following specific topics warrant further research: 1) Water quality impacts on AOP effectiveness. The effectiveness of AOPs is directly related to water quality parameters such as ph, alkalinity, NOM, TOC, turbidity, and concentrations of other interfering compounds (e.g., nitrates and bromide). Future studies on AOP treatment of MTBE must independently evaluate the impact of each of the above-listed water quality parameters. The evaluation criteria must include MTBE removal efficiency, 197

220 oxidation by-product formation, DBP formation potential, and costs. For ozone-based AOPs, the effect of influent bromide concentration on bromate formation must also be evaluated. Similarly, a detailed analysis of the effect of influent water turbidity and nitrate concentrations on the effectiveness of AOPs relying on UV-light (LP, MP, pulsed) is warranted. 2) By-product formation and control. One of the most significant areas of future research is the issue of by-product formation and control. The oxidation of MTBE to carbon dioxide and water involves many steps and the formation of many oxidation by-products (e.g., TBA, TBF, acetone). If these by-products are not completely mineralized, they will be present in treated water, resulting in elevated concentrations of potentially toxic byproducts in the treated water. A better understanding of by-product formation mechanisms and subsequent mitigation strategies will be necessary prior to the acceptance of AOPs by the regulatory community for drinking water applications. This includes research to determine the most cost-effective treatment option, such as biologically activated carbon, for by-product removal in drinking water applications. 3) Cost evaluation as a function of water quality and contamination scenario. Finally, future research should evaluate engineering costs for MTBE oxidation by AOPs. Capital and O&M costs for each AOP process should be developed as a function of water quality, flow rate, influent MTBE concentration, and required removal efficiency. These cost evaluations must be performed under uniform design criteria (e.g., required removal efficiency) and operational assumptions (e.g., power rate). A unified costing approach will enable a direct comparison of the various AOPs for specific water qualities. 198

221 3.9 References Al-Ekabi H., Serpone N., Pelizzetti E., Minero C., Fox M. A. and Draper R. B., Kinetic Studies in Heterogeneous Photocatalysis 2. TiO 2 -Mediated Degradation of 4-CP alone and in a 3-component mixture of 4-CP, 2,4-DCP and 2,4,5-TCP in Air-Equilibrated Aqueous Media, Langmuir, v. 5, p , Allen, A. O., The Radiation Chemistry of Water and Aqueous Solutions, van Nostrand- Reinhold. Princeton, N.J., Applebury T., Personal Communication. Chief Technical Officer, Applied Process Technologies, Pleasant Hill, California, AWWA, Water Treatment and Quality: A Handbook of Community Water Supplies, Fourth Edition, McGraw-Hill, New York, AWWA/ASCE, Water Treatment Plant Design, 3rd Edition, McGraw Hill, New York AWWARF, Effect of Bicarbonate Alkalinity on Performance of Advanced Oxidation Processes, Prepared by: Peyton G. R., Bell O. J., Girin E., LaFaivre M. H. and Sanders J., Denver, Barreto R. D., Gray K. A. and Anders K., Photocatalytic Degradation of MTBE in TiO 2 Slurries: A Proposed Reaction Scheme, Water Research, v. 29, n. 5, p , Bender J. B., Photolytic Oxidation of Contaminated Water Using the Riptide Water Purification System, Toronto Environmental Trade Show and Conference, Seminar # 2, Toronto, Canada, May Bender J. B., Personal Communication. President, Pulsar Environmental Remediation Systems Incorporated, Bukhari Z., Hargy T. M., Bolton J. R., Dussert B. and Clancy J L., Medium-Pressure UV for Oocyst Inactivation, Journal of AWWA, v. 91, n. 3, p , Buxton G.V. et al. Critical Review of Rate Constants for Hydrated Electrons, Hydrogen Atoms and Hydroxyl Radicals (OH/O-) in Aqueous Solution. Journal Physical and Chemical Reference Data, v. 17, n. 2, p.13, Calgon AOT Handbook, Calgon Carbon Oxidation Technologies, Markham, Ontario, Canada, Cater S., Personal Communication. Research and Technical Manager, Calgon Carbon Technologies, Markham, Ontario, Canada,

222 Chang, P., & Young, T. Reactivity and By-products of Methyl Tertiary-Butyl Ether Resulting from Water Treatment Processes. University of California Toxic Substances Research & Teaching Program, Chen C. T., Tafuri A. N., Rahman M. and Forest M. B., Chemical Oxidation Treatment of Petroleum Contaminated Soil using Fenton s Reagent, Journal of Environmental Science Health, A33, p , Chivate M. M. and Pandit A. B., Effect of Sonic and Hydrodynamic Cavitation on Aqueous Polymeric Solutions, Industrial Chemical Engineering, v. 35, p. 52, Cooper W. J., Treatment of MTBE in Drinking Water Supplies using High Energy Electron Injection: Innovative, Proven, Economical, Orange County ACS Meeting, January, Cooper W. J., Leslie G., Tornatore P. M., Hardison W. C. and Hajali P. A., MTBE and Priority Contaminant Treatment with High Energy Electron Injection, Draft, Cooper W.J. and Tornatore P.M. Destruction of MTBE in Contaminated Groundwater using Electron Beam Injection. ACS Proceeding, Anaheim, California, March, Crawford M., Personal Communication. Calgon Carbon Incorporated, Suisun City, California, Crittenden J. C., Zhang Y., Hand D. W., Perram D. L. and Marchand E. G., Solar Detoxification of Fuel Contaminated Groundwater using Fixed-Bed Photocatalysts, Water Environment Research, v. 68, n. 3, p , Dewaal W., Personal Communication. Director of Marketing AOPs, Trojan Technologies, Ontario, Canada, October, Dorfman L. M. and Adams G. E., Reactivity of the Hydroxyl Radical, National Bureau of Standards, Report No. NSRDS-NBS-46, Dyksen J.E., Ford R., Raman K., Schaeffer J.K., Fung L., Schwartz B., and Barnes M. In-line Ozone and Hydrogen Peroxide Treatment for Removal of Organic Chemicals, AWWA Research Foundation, Denver, CO, EPA/600/R-98/160VS, Environmental Technology Verification Statement, Sentinel UV Reactor, USEPA Office of Research and Development, Washington D. C., May Fenton H. J. H., Oxidation of Tartaric Acid in Presence of Iron, Journal of Chemical Society, v. 165, p , Glaze W.H., Kang J.W., and Chapin D.H. The Chemistry of Water Treatment Processes Involving Ozone, Hydrogen Peroxide, and Ultraviolet Radiation. Ozone Science and Engineering. vol.9:

223 Gruber W., The Ultrox UV Oxidation System, EI Digest - Industrial and Hazardous Waste Management, p. 7-11, April Halliwell B. and Gutteridge J.M., Free Radicals in Biology and Medicine 2nd ed., Clarendon Press, Oxford, Harp, C., Personal Communication. Project Manager, Hydroxyl Systems, British Columbia Canada, October, Hart, E.J., and Henglein, A. Sonolysis of Ozone in Aqueous Solution. Journal of Physical Chemistry. Vol. 90 pp , Hengesbach B., Schoenen D., Hoyer O., Bernhardt H., Mark G., Schuchmann H.P., and Sonntag C. UV Disinfection of Drinking Water the Question of Bacterial Regrowth and the Photolytic Degradation of Biogenic High-Molecular Weight Substances. J. WSRT- AQUA. 42(1): Himebaugh W., Personal Communications. Industrial Wastewater Manager, Signal Hill, California, Hoigne J. and Bader H., The Role of Hydroxyl Radical Reactions in Ozonation Processes in Aqueous Solution, Water Research, v. 10, p , Hoigne, J. Chemistry of Aqueous Ozone and Transformation of Pollutants by Ozonation and Advanced Oxidation Processes in The Handbook of Environmental Chemistry, Vol. 5 Part C, Quality and Treatment of Drinking Water (edited by J. Hrubec) Springver-Verlag, Berlin, Hong A.P., Bahneman D.W. and Hoffmann M.R., Cobalt(II) tetrasulfopthalocyanine on titanium dioxide: a new efficient electron relay for the Photocatalytic Formation and Depletion of Hydrogen Peroxide in Aqueous Suspension, Journal Physical Chemistry, v. 91, p , Horst M., Removal of the Residual Ozone in the Air after the Application of Ozone, Ozone Manual for Water and Wastewater Treatments, John Wiley & Sons, New York, Hua I. and Hoffmann M. R., Kinetics and Mechanism of the Sonolytic Degradation of CCl 4 : Intermediates and Byproducts, Environmental Science and Technology, v. 30, p , Hua I. and Hoffmann M. R., Optimization of Ultrasonic Irradiation as an Advanced Oxidation Technology, Environmental Science and Technology, v. 31, p ,

224 Huling S. G., Microbial and Peat Effects on the Oxidation of 4-POBN by Hydroxyl Radicals in Soil, Ph.D. Dissertation, Department of Chemical and Environmental Engineering, University of Arizona, Tucson, Huling S. G., Arnold R. G., Sierka R. A., Jones P. K. and Fine D., Contaminant Adsorption and Oxidation via Fenton Reaction, Submitted to Journal of Environmental Engineering, HVEA (High Voltage Environmental Applications), Product Brochure, Rochester, New York, Ireland J., Klostermann P., Rice E. and Clark R., Inactivation of E-coli by TiO 2 Photocatalytic Oxidation, Applied Environmental Microbiology, v. 59, p , Kang J-W. and Hoffmann M. R., Kinetics and Mechanism of the Sonolytic Destruction of MTBE by Ultrasonic Irradiation in the Presence of Ozone, Environmental Science and Technology, v. 32, n. 20, p , Karpel Vel Leitner N., Papailhou A., Croue J., Peyrot J., and Dore M. Oxidation of Methyl Tert Butyl Ether (MTBE) and Ethyl Tert Butyl Ether (ETBE) by Ozone and Combined Ozone/Hydrogen Peroxide Ozone Science and Engineering, v. 16 (41), Komex H 2 O Science, Treatment Technologies for Removal of MTBE from Drinking Water (1st Edition), Chapter 2. Advanced Oxidation Processes. Principal Authors: Anthony Brown and Tom Browne, Prepared for MTBE Research Partnership, December, Kotranarou A., Mills G. and Hoffmann M. R., Oxidation of Hydrogen Sulfide in Aqueous Solution by Ultrasonic Irradiation, Environmental Science and Technology, v. 26, n. 12, Kormann C., Bahnemann D. W. and Hoffmann M. R., Photocatalytic Production of H 2 O 2 and Organic Peroxides in Aqueous Suspensions of TiO 2, ZnO, and Desert Sand, Environmental Science and Technology, v. 22, p , Kormann C., Bahnemann D. W. and Hoffmann M. R., Photolysis of Chloroform and Other Organic Molecules in Aqueous TiO 2 Suspensions, Environmental Science and Technology, v. 25, p , Koyama O., Kamagata Y. and Nakamura K., Degradation of Chlorinated Aromatics by Fenton Oxidation and Methanogenic Digester Sludge Water Research, v. 28, n. 4, p , Krasner, S. T.., Glaze W. H., Weinberg H. S., Daniel P. A. and Najm I. N.,. Formation and Control of Bromate During Ozonation of Waters Containing Bromide. Journal of AWWA, 85:1:73,Jan

225 Kurucz C. N., Waite T. D. and Cooper W. J., High Energy Electron Beam Irradiation of Water, Wastewater and Sludge, Advances in Nuclear Science and Technology, Editors: Lewins J. and Becker M., 22, p. 1-43, Plenum Press, NY, Liang S., Palencia S.L., Yates R.S., Davis M.K., Bruno J.M., and Wolfe R.L., Oxidation of MTBE by Ozone and Peroxone. Journal American Water Works Association, Vol. 91 (6), , 1999a. Liang, S., Yates. R. S., Davis D. V., Pastor S. J., Palencia L. S. and Bruno J-M., Treatability of Groundwater Containing Methyl tertiary-butyl Ether (MTBE) by Ozone and PEROXONE and Identification of By-products. Submitted to JAWWA for Publication. 1999b. Lin K., Cooper W. J., Nickelsen M. G., Durucz C. N. and Waite T. D., Decomposition of Aqueous Solutions of Phenol using HEEB Irradiation: a Large Scale Study, Applied Radiation Isotopes, v. 46, n. 12, p , Martin P. D. and Ward L. D., Reactor Design for Sonochemical Engineering Chemical Engineering Research and Design, v. 70, A3, p. 296, Mason, T., & Lorimer, J. P. Sonochemistry: Theory, Applications, and Uses of Ultrasound in Chemistry. Ellis Norwood, Ltd., Chichester, U.K., Mathews R. W., Kinetics of Photocatalytic Oxidation of Organic Solutes over Titanium dioxide, Journal of Catalysis, v. 111, p , Mohanty N. R. and Wei I. W., Oxidation of 2,4-DNT using Fenton s Reagent: Reaction Mechanisms and their Practical Applications, Hazardous Waste and Hazardous Materials, v. 10, n. 2, p , Montgomery J. M., Water Treatment Principles and Design, James M. Montgomery Consulting Engineers, A Wiley-Interscience Publication, New York, Morel F.M.M. and Hering J.G., Principles and Applications of Aquatic Chemistry, John Wiley & Sons, New York, Nickelsen M. G., Cooper W. J., Kurucz C.N. and Walte T. D., Removal of Benzene and Selected Alkyl-Substituted Benzenes from Aqueous Solutions Utilizing Continuous High Energy Electron Irradiation, Environmental Science and Technology, v. 26, n. 1, p , Nickelsen M. G., Cooper W. J., Lin K. and Kurucz C. N., High Energy Electron Beam Generation of Oxidants for the Treatment of Benzene and Toluene in the Presence of Radical Scavengers, Water Research, v. 28, n. 5, p ,

226 Nickelsen M.G., Kajdi D.C., Cooper W.J., Kurucz C.N., Waite T.D., Gensel F., Lorenzl H., and Sparka U. Field Application of a Mobile 20-kW Electron-Beam Treatment System on Contaminated Groundwater and Industrial Wastes. Environmental Applications of Ionizing Radiation. Edited by W. J. Cooper, R.D. Curry, and K.E. O Shea. John Wiley & Sons, Inc NIOSH, NIOSH Pocket Guide to Chemical Hazards, U.S. Department of Health and Human Services, Washington D.C., June Olson T. M. and Barbier P. F., Oxidation Kinetics of Natural Organic Matter by Sonolysis and Ozone, Water Research, v. 28, n. 6, p , Pandit A. B. and Moholkar V. S., Harness Cavitation to Improve Processing, Chemical Engineering Progress, v. 92, n. 7, p. 57, Peral J. and Domenech X., Photocatalytic Cyanide Oxidation from Aqueous Copper Cyanide Solutions over TiO 2 and ZnO, Journal of Chemical Technology and Biotechnology, v. 53, p , Petrier, C.; Jeunet, A.; Luche, J.L.; Reverdy, G.J.. Journal of American Chemical Society v pp Pisani J. A. and Beale S. E., Cavitation Induced Hydroxyl Radical Oxidation, AIChE Spring National Meeting, Session 95A, Houston, Texas, March Pisani J. A., Personal Communication, President, Oxidation Systems Incorporated, Arcadia, California, October 1999a. Pisani J. A., The HYDROX Process, Hydrodynamically Induced Cavitation, Oxidation Systems Incorporated, Arcadia, California, 1999b. Pontius F. W., New Horizons in Federal Regulations, Journal of AWWA, v. 90, n. 3, p , Pontius F. W. and Diamond W. R., Complying with Stage 1 D/DBP Rule, Journal AWWA, v. 91, i. 3, p , Potter F. J. and Roth J. A., Oxidation of Chlorinated Phenols using Fenton s Reagent, Haz. Waste and Haz. Materials, v. 10, n. 2, p , Prado J. and Esplugas S., Comparison of Different AOPs involving Ozone to Eliminate Atrazine, Ozone Science and Engineering, v. 21, n. 1, p , Prairie M. R., Evans L. R., Strange B. M. and Martinez S. L., An Investigation of TiO 2 Photocatalysis for the Treatment of Water Contaminated with Metals and Organic Chemicals, Environmental Science and Technology, v. 27, p ,

227 Reckhow D. A., Knocke W. R., Kearns M. J. and Parks C. A., Ozone Science and Engineering, v. 13, p. 623, Reynolds, T.D., and Richards, P.A. Unit Operations and Processes in Environmental Engineering, Second edition, PWS Publishing, Boston, MA, Rice R. G., Ozone - Advanced Oxidation Processes - Current Commercial Realities, Proceedings of 13th Ozone World Congress, Kyoto, Japan, Rodriquez, Rey. Project Manager for Santa Monica Drinking Water Treatment System. Personal Communication. November, Sabate J., Anderson M. A., Aguado M. A. and Gimenez J., Comparison of TiO 2 Powder Suspensions and TiO 2 Ceramic Membranes Supported on Glass as Photocatalytic Systems in the Reduction of Chromium(VI), Journal of Molecular Catalysis, v. 71, p , Serpone N., Terzian R., Hidaka H. and Pelizzetti E., Ultrasonic Induced Dehalogenation and Oxidation of 2-, 3- and 4-Chlorophenol in Air-Equilibrated Aqueous Media. Similarities with Irradiated Semiconductor Particulates, Journal of Physical Chemistry, v. 98, p , Siddiqui, M. S., and Amy, G. L. Factors Affecting DBP Formation During Ozone-Bromide Reactions. Journal of AWWA, 85:1:63 (Jan. 1993). Siddiqui M. S., Amy G., Ozekin K., Zhai W. and Westerhoff P., Alternative Strategies for Removing Bromate, Journal of AWWA, 81-95, October Siddiqui M. S, Amy G., Cooper W. J., Kurucz C. N., Waite T. D. and Nickelsen M. G., Bromate Ion Removal by HEEB Irradiation, Journal of AWWA, v. 88, n. 10, p , Siddiqui, M. S., Amy G., Ozekin, K., Zhai, W., and Westerhoff, P. Alternative Strategies for Removing Bromate. Journal of American Water Works Association, v. 86 (10), 81-95, Sjogren J. C. and Sierka R. A., Inactivation of Phage MS2 by Iron-Aided TiO 2 Photocatalysis, Applied Environmental Microbiology, v. 60, n. 1, p , Sjogren J. C., Inactivation of Phage MS-2 by Titanium Dioxide Photocatalysis, Ph.D. Dissertation, Department of Chemical and Environmental Engineering, University of Arizona, Tucson, Skov E. R., Pisani J. A. and Beale S. E., Industrial Wastewater Treatment using Hydroxyl Radical Oxidation by Hydraulically Induced Cavitation, AIChE Spring National Meeting Proceedings, Session 86a, Houston, Texas, March,

228 Song, R., Westerhoff P., Minear R. and Amy G., Bromate Minimization during Ozonation. Jour. AWWA, 89:6: 69, June Staehelin J. and Hoigne J., Decomposition of Ozone in Water: Rate of Initiation by Hydroxide Ions and Hydrogen Peroxide, Environmental Science and Technology, v. 16, p , Stewart, M.H., Hwang, C.J., Hacker, P.A., Yates, R.S., and Wolfe, R.L. Microbial and Chemical Implications of Using Ultraviolet Irradiation for Treatment of Biological Filter Effluent. Proceedings AWWA 1993 WQTC. Miami, FL Stumm W. and Morgan J. J., Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters, 3rd Edition, A Wiley Interscience Publication, John Wiley and Sons, New York, Tornatore P., Personal Communication. Vice-President, Haley & Aldrich, Rochester, NY, October Tornatore, P. M. and Cooper W. Remediation of MTBE Contaminated Drinking Water Supplies with High Energy Electron Injection. ACS Proceedings, Anaheim, California, March, Turchi C.S. and Ollis D.F., Photocatalytic Degradation of Organic Water Contaminants: Mechanisms Involving Hydroxyl Radical Attack. Jour. Catal., v. 122, p , USEPA, Oxygenates in Water: Critical Information and Research Needs. EPA/600/R- 98/048, December Venkatadri R. and Peters R. W., Chemical Oxidation Technologies: UV Light/ Hydrogen Peroxide, Fenton s Reagent and Titanium Dioxide-Assisted Photocatalysis, Haz. Waste and Haz. Materials, v.10, n. 2, p , Von Gunten, U. & Hoigne, J., Bromate Formation During Ozonation of Bromide- Containing Waters Proceedings 1993 Ozone World Congress, International Ozone Association, San Francisco. Von Gunten U. and Hoigne J., Bromate Formation during Ozonation of Bromide- Containing Waters: Interaction of Ozone and Hydroxyl Radical Reactions, Environmental Science and Technology, v. 28, p , Von Gunten U. and Oliveras Y., Kinetics of the Reaction between Hydrogen Peroxide and Hypobromous Acid: Implication on Water Treatment and Natural Systems, Water Research, v. 31, n. 4, p ,

229 Wagler J. L. and Malley J. P. Jr., The Removal of MTBE from a Model Groundwater using UV/Peroxide Oxidation, Journal of New England Water Works Association, p , September Walling C., Fenton s Reagent Revisited, Accounts of Chemical Research, v. 8, p , Waters C., Personal Communication. Applied Process Technologies, San Francisco, California, White G. C., Handbook of Chlorination and Alternative Disinfectants, A Wiley-Interscience Publication, John Wiley and Sons Inc., New York, Woodling R., Personal Communication. Senior Applications Engineer, U.S. Filter, Santa Clara, California, Yeh C. K. and Novak J. T., The Effect of Hydrogen Peroxide on the Degradation of Methyl and Ethyl Tert-Butyl Ether in Soils, Water Environment Research, v. 67, n. 5, p , Zepp, R. G., Factors Affecting the Photochemical Treatment of Hazardous Waste. Environ. Science Technology, 22 (3), 256 (1988). Zheng M., Andrews S. A. and Bolton J. R., Impacts of Medium-Pressure UV on THM and HAA Formation in Pre-UV Chlorinated Drinking Water, AWWA Water Quality Technology Conference, Tampa, Florida, 1999 (a). Zheng M., Andrews S. A. and Bolton J. R., Impacts of Medium-Pressure UV and UV/H 2 O 2 Treatments on Disinfection Byproduct Formation, AWWA Annual Conference, Chicago, 1999 (b). 207

230 208

231 4.0 Granular Activated Carbon Daniel Creek, P.E. James Davidson, P.G. 209

232 210

233 4.1 Background Nature of Problem Past use of GAC to treat MTBE-impacted water resulted in difficulties due to the physical and chemical characteristics of MTBE. Results of several full-scale applications suggest that some GACs perform poorly when MTBE is present in the water to be treated (e.g., McKinnon and Dyksen, 1984; Creek and Davidson, 1998). In consideration of this, an in-depth evaluation is needed to identify the most promising GACs for MTBE removal and to define conditions under which GAC is most likely to be a cost-effective treatment alternative for MTBE Objectives The primary objective of this chapter is to provide a detailed feasibility analysis regarding the use of GAC to remove MTBE from drinking water. This feasibility analysis includes determining the conditions (e.g., MTBE influent concentrations, background water quality) under which GAC is most likely to cost-effectively remove MTBE. Other specific objectives include the following: Clarify causes of the wide variability in existing data for MTBE removal using GAC. Determine state-of-knowledge regarding mass transfer parameters for different types of GAC. Evaluate the impact of NOM and SOCs (e.g., benzene) on the adsorption of MTBE. Present state-of-knowledge regarding desorption of MTBE from GAC. To address these objectives, this chapter presents general and detailed evaluations of GAC for treating MTBE-impacted water. The general evaluation includes a focused literature review and a compilation of vendor information to determine the benefits and limitations of GAC, key variables and design parameters, current usage of GAC for MTBE removal, etc. The detailed evaluation includes computer modeling results and cost estimates prepared to determine the impacts of influent MTBE concentrations, background organic matter, and the presence of other organic compounds upon the removal of MTBE using GAC. 211

234 212

235 4.2 General Evaluation of GAC Technology This section presents a general evaluation of using GAC for the treatment of MTBE. A brief description of carbon adsorption and its applications are presented, followed by discussions of advantages and disadvantages; key variables and design parameters; GAC vendors; developmental status of technology; ease of installation and operation; permitting; and current usage of technology. This section also presents brief reviews of related research for MTBE treatment with GAC Description of Technology Carbon adsorption technology is implemented by passing contaminated water through a vessel (or vessels) containing GAC. Intermolecular attraction between molecules of a dissolved chemical (adsorbate) and the GAC (adsorbent) surface results in adsorptive forces that physically attract the adsorbate to the GAC as the water passes through the vessel. As such, the adsorbate remains attached to the GAC matrix and the water leaves the system with a decreased contaminant concentration. Chapter 5 presents further discussion of the primary processes that govern adsorption. GAC for drinking water applications is created from carbonaceous source materials, such as bituminous coal, coconut shell, petroleum coke, wood, and peat. GAC is produced by grinding, roasting, and then activating the source materials with high temperature steam (water-gas reaction). This processing results in a porous material with very high internal surface area. Processing each of the different source materials results in GACs with significantly different adsorption properties. The adsorption potential of a given carbon is dependent on numerous factors, including the adsorption structure (i.e., pore sizes) and the characteristics of the adsorbate. Detailed discussions of these important GAC characteristics are presented in numerous references (e.g., Sontheimer et. al., 1988; Lehr, 1991; and, Nyer, 1992). To design an efficient carbon adsorption system for water treatment, a set of bench-scale and/or pilot-scale tests using the site-specific water is typically performed. These test results allow designers to optimize the type of GAC used and to estimate operating parameters for the full-scale system. Common types of bench-scale methods include isotherm tests (batch equilibrium tests) and dynamic column tests. These testing methods are discussed separately below. Isotherm Testing Isotherm testing is performed to determine the relationship between adsorption capacity of a GAC and concentration of an adsorbate under specific background water conditions. The results of isotherm tests can be used to select the most effective GAC for specific water conditions and to develop preliminary design parameters. 213

236 Isotherm testing, a static testing process, is performed by placing a measured amount of GAC into an aqueous solution containing a measured concentration of adsorbate. After equilibration, the adsorbate concentration in the water is measured, allowing the adsorption capacity of the GAC to be calculated. The procedure is performed several times at varying initial adsorbate concentrations in order to develop the relationship between carbon capacity and adsorbate concentration at a constant temperature. This relationship can be described using the well-established Freundlich Isotherm Equation: q = K F C n where q is the mass of adsorbate per dry unit weight of carbon [mg/g] and C is the concentration in the bulk solution of the adsorbate at equilibrium. The Freundlich isotherm parameters, K F and n, are experimentally determined constants at a given temperature. In general, it has been shown that the value of K F is related to the adsorbent capacity while the value of n is related to the strength of adsorption (Weber, 1972). Specific procedures for isotherm testing are presented in American Society for Testing and Materials (ASTM) Designation D (ASTM, 1996). Isotherm testing can be performed using distilled-deionized water or site-specific water that includes NOM and competing SOCs. The impacts of carbon fouling by NOM and competitive interaction between SOCs are discussed later in this chapter (Section 4.2.2). It is important to note that in full-scale applications of carbon adsorption, the GAC adsorption capacity for dissolved contaminants typically is lower than the maximum adsorption capacity determined from isotherm testing. Dynamic Column Testing Although analytical models that incorporate parameters determined from isotherm testing can be used to predict full-scale results, it usually is better to perform dynamic testing to quantify kinetic effects with site-specific water. Because of this, the design of carbon adsorption systems may be preceded by dynamic column testing with the selected GAC and the site water. These tests can also determine the need for pretreatment of the water prior to carbon adsorption. Dynamic testing is performed with a set of GAC columns connected in series, using either the rapid small-scale column test (RSSCT) method or the accelerated column test (ACT) method. Samples taken at the effluent end of the columns allow development of concentration breakthrough curves that are dependent on flow conditions in addition to GAC type and influent adsorbate concentration. These test results can be used for full-scale design through the use of scaling relationships and allow for evaluation of EBCT, length of the mass transfer zone (MTZ), and other critical design variables impacted by dynamic effects. Crittenden et al. (1989) present procedures and scaling relationships for the RSSCT method, which is currently being refined for presentation as an ASTM standard test method that is expected to 214

237 be finalized in the year 2000 (Graham, personal communication, 1999). Detailed discussions of column testing are presented in numerous references (Stenzel and Merz, 1988; Crittenden et al., 1991; and, Hand et al., 1989). Full-scale Operations Carbon adsorption systems can be used in various configurations (e.g., in-series, in-parallel, upflow, downflow). In general, adjusting the system configuration allows designers to optimize the adsorption characteristics and cost-effectiveness of the system based on the influent water conditions (e.g., adsorbate properties, flow rate), effluent treatment criteria, and regeneration/replacement methods. These design considerations for full-scale GAC systems are discussed in numerous references (e.g., Sontheimer et al., 1988; Hand et al., 1989; Nyer, 1992). Single vessel treatment systems are the simplest applications of GAC technology. These systems are often used in situations with low flow rates, low influent concentrations, or less stringent effluent criteria. In single vessel treatment systems (or in the lead vessel of an inseries system), the MTZ (see Figure 5-11 in Chapter 5) of the adsorbate moves down gradient as the GAC at the influent end of the vessel becomes saturated with adsorbate. Eventually, the MTZ of the adsorbate moves far enough into the vessel that measurable concentrations of adsorbate are observed in the effluent (i.e., breakthrough ). When the effluent from a single-vessel system reaches a specified adsorbate concentration, the vessel is taken off-line and the carbon is replaced. As shown in Figure 5-11, not all of the GAC will be saturated at breakthrough because of the shape of the MTZ. This limits the removal effectiveness of single-vessel systems and necessitates shutdown of the treatment system until new or regenerated carbon can be replaced into the single vessel. Figure 4-1 presents a schematic of in-series operation, which is often used for public drinking water systems. In-series operation allows the secondary, or lag, vessel(s) to maintain final effluent quality while GAC in the lead vessel is used to remove the majority of the adsorbate dissolved in the influent water. Once effluent concentrations from the lead vessel reach a specified level (e.g., 50 percent or more of the influent concentration), the lead vessel is taken off-line, replaced with fresh carbon, and then put back on-line as the lag vessel. Proper change-out of an in-series GAC system ensures that the MTZ of the adsorbate does not reach the effluent end of the lag vessel, thereby ensuring non-detectable concentrations in the system effluent at all times. In-series operation is typically required for weakly adsorbing compounds, such as MTBE, or in situations where effluent criteria are stringent (e.g., drinking water systems). Vessels can be used in series with more than two vessels, particularly with weakly adsorbing contaminants with long MTZs. Longer series can also optimize the cost-efficient use of the adsorptive capacity of a carbon batch, allowing almost complete saturation of the lead vessel without risking a breakthrough in the effluent, exceeding the effluent criteria. 215

238 Sampling Port Influent Sample Contaminated Water Delivery Pump GAC Vessel #1 GAC Vessel #2 GAC Vessel #3 Effluent Sample Midfluent Sample Note: Pretreatment May Be Required Note: In Some Cases Midfluent Sample To Water Distribution System Figure 4-1. Schematic for GAC using series operation. 216

239 Operation of vessels in parallel allows for larger flow rates to be treated and/or for increased EBCTs to be used. Because the adsorbate must be kept in contact with the GAC for a sufficient amount of time, it is sometimes necessary to split the influent flow into more than one series of carbon vessels. This effectively lowers the volumetric flow rate going through a given vessel (or series of vessels), increasing the residence time of the contaminated water in the GAC matrix. An important aspect of full-scale GAC operation is regeneration or replacement requirements. Carbon regeneration or replacement is necessary when the effluent from the lead vessel or one of the lag vessels in a series reaches a specified concentration. At that time, the lead vessel is taken off-line for either on-site regeneration or replacement with reactivated or virgin GAC. Because of stringent effluent standards, many drinking-water treatment applications will require replacement with virgin carbon, particularly for weakly adsorbing compounds such as MTBE. Another aspect of full-scale operation of GAC systems is pretreatment. In general, pretreatment is performed to modify the influent water stream in order to maximize the efficiency of the adsorption process for the primary adsorbate(s). Common pretreatment requirements, which are described briefly below, include filtration, precipitation, ph adjustment, and disinfection. Filtration is performed to remove suspended particles that would otherwise clog the pore structure of the GAC matrix in addition to taking up adsorption sites in the matrix. Influent water may require precipitation pretreatment in order to maximize the GAC adsorption efficiency by minimizing metal precipitation (and associated carbon fouling) within the GAC vessels. Research has shown that the ph of the influent water can impact carbon column performance (Semmens et al., 1986); thus, ph adjustment is another possible pretreatment requirement. With respect to ph, Semmens et al. (1986) concluded that conditions employed during pretreatment may have a profound influence on the subsequent performance and useful life of the GAC. However, Calgon Carbon Corporation (Pittsburgh, PA) has found that ph will not have an effect on [the adsorption of] compounds such as MTBE, BTEX, etc (Megonnell, personal communication, 1999). Disinfection is performed to reduce the growth of bacteria, which similar to suspended particles can clog the GAC matrix and take up adsorption sites. However, biological growth within the GAC matrix is sometimes beneficial, as bacteria can utilize organic contaminants as substrates for further growth. Consideration should be given to the potential beneficial effects of biological growth within the GAC vessels when designing pretreatment systems. This topic, which is currently being researched by University of California, Davis, is beyond the scope of this evaluation. 217

240 4.2.2 Benefits and Limitations Benefits Carbon adsorption has several advantages over other water treatment methods. The primary advantage is that it is a very simple technology that generally has stable operations. In addition, GAC is very well established for removing organic compounds (although, to date, its field application for treating MTBE in large-scale drinking water systems is limited). Carbon systems are very easy to implement due to wide commercial availability; numerous vendors can supply GAC and full-scale systems. Because of the simplicity of the equipment and materials, capital and installation costs are relatively low compared to more innovative technologies. Carbon systems require no off-gas treatment and the creation of by-products is limited to spent carbon that requires regeneration or disposal. In summary, the benefits of GAC for MTBE treatment are as follows: Simple technology/stable operations Well-established equipment and methods Easy to implement/commercially available Low capital and installation costs No off-gas treatment required The reader should note that there are other removal/destruction mechanisms that occur within the GAC matrix. Evaluation of these beneficial processes, which include biological activity, is beyond the scope of this report. Limitations The limitations of GAC effectiveness for the treatment of MTBE can be separated into the following primary categories, each of which is discussed separately below: MTBE characteristics Impact of NOM Impact of other SOCs Desorption MTBE Characteristics The physical and chemical characteristics of MTBE are generally considered poor for adsorption. In particular, MTBE s high solubility (48,000 mg/l at 25 C) causes the compound to preferentially remain in solution (API, 1994). This effect is also indicated by 218

241 MTBE s low octanol:water distribution coefficient (log K ow = 1.2), which demonstrates its high affinity for water (API, 1994). The poor adsorptive characteristics of MTBE can cause early breakthrough and/or frequent carbon changeout requirements. Research of small-scale point-of-entry (POE) treatment systems indicated that removal of BTEX compounds by GAC far exceeded that of MTBE, resulting in MTBE controlling GAC life in the systems (Malley et al., 1993). Isotherm studies of the widely utilized Calgon Filtrasorb 300 indicate that the adsorption capacity is approximately an order of magnitude lower for MTBE than for toluene, on an equivalent mass basis (Dobbs and Cohen, 1980). Because of the low adsorption capacity of carbon for MTBE, system design may require use of three or more vessels in series to contain the MTZ and to maximize carbon usage efficiency. As such, treatment of MTBE-impacted water using GAC may result in higher upfront capital costs and higher O&M costs due to frequent carbon changeout. Impact of NOM Numerous studies have shown that the presence of NOM in the source water to be treated can significantly reduce the adsorption capacity of GAC for SOCs (Summers et. al., 1989; Zimmer et al., 1988; and Speth, 1991). Based on these studies, it appears that GAC adsorption capacity decreases with increasing carbon exposure time to NOM. The magnitude of the capacity decrease is also dependent on variables such as type of NOM, type of carbon, and the SOC to be removed. The NOM fouls GAC in several ways: 1) NOM can take up adsorption sites within the MTZ of the SOC (i.e., competitive adsorption); 2) NOM can preload the GAC by moving more quickly than the SOC (i.e., ahead of the MTZ of the SOC); and, 3) NOM can physically clog the pore space of the carbon. Studies for several SOCs have shown that the Freundlich isotherm parameter, K F, generally decreases with increasing NOM exposure time (Zimmer et. al., 1988; Hand et. al., 1989). This indicates that the overall capacity of GAC for an SOC will decrease with time as more water is passed through the GAC matrix. Research has also shown that the decrease of K F varies significantly for different SOCs, different source waters (i.e., different types and concentrations of NOM), and different carbons (Sontheimer et. al., 1988). Because of this, it is difficult to predict the decrease of adsorption capacity without site-specific testing. Because of the kinetics of NOM adsorption, NOM fouling should have less effect on weakly adsorbing compounds that move relatively quickly through carbon. Based on laboratory data, Speth (1991) speculated that for dichloroethylene (DCE) and other weakly adsorbing compounds, the effects of NOM preloading in a carbon column are less significant on adsorption capacity than for strongly adsorbing compounds. Thus, for the weakly adsorbing MTBE, the impact of NOM may be less severe than for other SOCs. 219

242 At this time, there is no measurable water quality parameter that is clearly and consistently indicative of NOM fouling potential (Hand, personal communication, 1998). To date, NOM concentrations have been estimated by using DOC or TOC as surrogate parameters. However, it is unclear how the measured DOC or TOC concentrations in the water relate to the extent of GAC fouling. Although logic suggests that higher DOC or TOC loads will result in greater fouling, contradictory results have been found in research (Zimmer et. al., 1988; Summers et. al., 1989). Experience has shown that measured DOC or TOC concentrations are not always accurate indicators of GAC fouling potential (Hand, personal communication, 1998). This discrepancy is attributed to the fact that NOM fouling is controlled at least in part by the specific types of NOM present (e.g., humic acids, fulvic acids). These specific types of NOM are not quantified by measurements of DOC or TOC. At this time, there is no standard method for determining specific NOM types and concentrations present in a given water. As such, there currently is no readily measurable parameter that is clearly and consistently indicative of NOM fouling potential. No specific studies on the impact of NOM on MTBE removal by GAC were identified during this evaluation. However, based on the studies described above, NOM fouling is likely to decrease GAC capacity for MTBE to a lesser degree than for more strongly adsorbing chemicals such as the BTEX compounds. Testing under site-specific conditions (e.g., water characteristics, SOC concentrations, GAC characteristics, etc.) is needed to accurately assess the impact of NOM on MTBE removal by carbon adsorption. Impact of Other SOCs Similar to the effects of NOM, competitive adsorption due to the presence of other SOCs can reduce GAC capacity for MTBE. The capacity reduction depends on type and concentration of competing SOCs and the type of GAC being used. Because MTBE has relatively low affinity for adsorption onto carbon, other compounds with higher affinities will preferentially take up adsorption sites on the carbon. It should be noted that for strongly adsorbing SOCs, the MTZ will be established as a thin zone near the influent end of the lead vessel. As such, other SOCs may have limited impact on MTBE adsorption in the majority of the downgradient GAC matrix. Prediction of adsorption effectiveness for multi-component systems is complex, particularly in the presence of NOM. Similar to single component systems, competitive adsorption systems will be affected by NOM and other water quality characteristics. Although isotherm testing can be used to qualitatively assess competitive effects, dynamic column testing under site-specific conditions is generally required to accurately predict the impact of competitive adsorption on full-scale GAC capacity for MTBE. This topic has been researched extensively for numerous organic compounds (e.g., Kong and DiGiano, 1986), though no studies specifically evaluating competitive adsorption effects on MTBE were found for this evaluation. Further discussion of this topic is presented in Section

243 Desorption Since MTBE is only weakly adsorbed by GAC, other more preferentially adsorbed SOCs in the contaminated water will tend to displace previously sorbed MTBE from the GAC matrix. This effect, which is known as desorption, can also occur when influent MTBE concentrations decrease, allowing previously sorbed MTBE molecules to enter back into solution (re-equilibration). Evidence of MTBE desorption (i.e., effluent concentrations higher than influent concentrations) has been observed for several GAC systems designed to remove MTBE, including a service station in Massachusetts at which 33 MTBE desorption events occurred over a 6-year period (compared to seven benzene desorption events) (Creek and Davidson, 1998). Although desorption of MTBE (or other compounds) does not always constitute system failure (e.g., effluent concentrations may still be below treatment goals), it does signal that carbon changeout may be needed. Desorption is a critical variable in planning O&M efforts for GAC systems treating weakly sorbing compounds such as MTBE. The potential for desorption contributes to the need for design and operational variations such as multiple vessels in series and higher frequency sampling of midfluent and effluent. It should be noted that for a properly designed and maintenanced GAC system, desorption is a manageable problem Key Variables and Design Parameters This section reviews key variables and design parameters relative to the use of GAC for MTBE removal. GAC Type Different types of GAC have widely varying adsorptive properties for MTBE. Freundlich isotherm parameters, in addition to other relevant information for several GAC products, are presented in Table 4-1. Isotherm data for these products, including coal-based, coconut shell, and reactivated coal-based carbons, are plotted in Figure 4-2. It is important to note that not all of the isotherm data presented in Figure 4-2 and Table 4-1 were developed from laboratory testing. Computer models were used to generate the data for CARBTROL s (Westport, CT) CSL carbon and Calgon s (Pittsburgh, PA) Filtrasorb 300, PCB, and REACT carbons. However, the data for Barneby & Sutcliffe s (Columbus, OH) 207A and PC carbons, Calgon s (Pittsburgh, PA) Filtrasorb 400 and Filtrasorb 600 carbons, and U.S. Filter/Westates (Los Angeles, CA) CC-602 carbon were developed using laboratory isotherm testing methods. 221

244 222 A Units of K F are (mg/g)(l/mg) ṇ Freundlich Modeling/ Range of MTBE Vendor/Source Product Carbon Parameters Testing Concentrations Name Source K A F n Conditions (µg/l) Barneby & Sutcliffe 207A coal laboratory tests 6, ,000 Corporation T = 25 o C PC coconut distilled water Calgon Carbon FS 300 coal computer model 10-1,000,000 Corporation T = 25 o C PCB coconut "pure" water REACT reactivated coal FS 600 coal laboratory tests T = 25 o C distilled water CARBTROL CSL coconut computer model 1-10,000 distilled water U.S.Filter/Westates CC-602 coconut laboratory test 1-1,000,000 T ~ 20 o C organic-free water Table 4-1 Summary of MTBE Isotherm Data for Activated Carbons Speth and Miltner, FS 400 coal laboratory test (Calgon) T = 24 o C dist./deion. water A it ( / )(L/ )

245 1000 MTBE Isotherms for Granular Activated Carbon Figure 4-2. MTBE isotherms for GAC. Adsorbed Concentration (mg/g) Adsorbed Concentration (mg/g) Aqueous Concentration (mg/l) 207A (coal; Barneby & Sutcliffe, 1998) PC (coconut; Barneby & Sutcliffe, 1998) FS 300 (modeled coal; Calgon, 1998) PCB (modeled coconut; Calgon, 1998) REACT (modeled react. coal; Calgon, 1998) CC-602 (coconut; U.S. Filter/Westates, 1998) CSL (modeled coconut; CARBTROL, 1998) FS 400 (Calgon coal; Speth and Miltner, 1990) FS 600 (coal; Calgon, 1999)

246 Based on the isotherm data available for MTBE, coconut shell carbons appear to have a higher adsorptive capacity for MTBE than coal-based carbons. Figure 4-2 shows that for MTBE concentrations ranging from 1 µg/l to 1,000,000 µg/l, coconut shell carbon (U.S. Filter/Westates CC-602 or Calgon PCB) has consistently higher adsorption potential than the coal-based carbons. As shown on Table 4-1, the Freundlich adsorption capacity parameter, K F, is 10 for Calgon s PCB coconut shell carbon. In contrast, the K F values are 5.6 and 7.8 for Calgon s Filtrasorb 300 and Filtrasorb 600, both coal-based carbons. Calgon predicted carbon costs for removal of MTBE from pure water and found that using coconut shell carbon (PCB) was significantly more cost-effective than using coal-based carbon (Filtrasorb 300) for MTBE concentrations below 100 mg/l (McNamara, personal communication, 1998). U.S. Filter/Westates (Los Angeles, CA) also recommended coconut shell carbon based on extensive isotherm testing and in-house computer modeling of performance (Graham, personal communication, 1998). However, it is important to note that properties of GAC will vary from lot to lot, depending on the source material. It is likely that a given coconut shell carbon will vary in quality more than a coal-based carbon due to the less uniform nature of the source materials. Another possible advantage to coconut shell GAC is that research has shown that it may be less vulnerable to immediate fouling effects due to NOM. Zimmer et al. (1988) tested three different GACs for adsorption of TCE in the presence of NOM. The results of the study showed that the microporous coconut carbon maintained a greater capacity for TCE up to NOM preloading times of 25 weeks. However, after about 25 to 50 weeks, all three of the carbons were measured to have approximately the same capacity for TCE. This research suggests that the adsorption of MTBE may be less likely to be impacted by NOM fouling if coconut shell carbon is used. Calgon has recently introduced a high quality coal-based carbon (Filtrasorb 600) that reportedly performs very well for removal of MTBE. As shown on Table 4-1, the modelderived K F -value for Filtrasorb 600 is 7.8. This GAC is produced by carefully screening particle sizes to maximize the quantity of micropore sizes that are most effective for adsorption of MTBE. This GAC is still in early stages of development and, as such, it does not yet have extensive use in full-scale operations. For drinking water applications, the use of virgin carbon may be necessary in order to meet the low treatment goals for weakly adsorbing compounds like MTBE. As shown on Table 4-1, reactivated carbon (Calgon REACT) has a relatively low K F value (i.e., 3.5) in comparison to those for virgin coal and coconut shell carbons. Although reactivated carbon is generally less expensive per pound, higher usage rates and associated carbon changeout requirements and an increased risk of contaminant breakthrough are expected to outweigh the unit cost advantages of using reactivated GAC for MTBE removal from drinking water. 224

247 Background Water Quality Background water quality can significantly impact the removal efficiency for SOCs. As discussed in Section 4.2.2, NOM and the presence of competing SOCs can significantly reduce carbon capacity for MTBE (note: the impact of these variables are evaluated in detail in Section 4.3). In addition, other compounds, such as iron, manganese, and calcium carbonate, can precipitate in the carbon bed, changing the flow structure and limiting diffusion of organics into the micropores (Hall and Mumford, 1987). Because of undesirable background water quality, it may be necessary at some sites to pretreat influent water going into carbon systems in order to optimize treatment efficiency (see Section 4.2.1). Process Flow Configuration and Operating Parameters Proper design of a carbon adsorption system is critical for maximizing removal effectiveness. The primary design variables include empty bed contact time (EBCT) and process flow configuration. Other operating parameters include hydraulic loading and carbon bed depth. The EBCT, calculated as bed volume (BV) divided by flow rate, is proportional to the residence time of water in a GAC vessel (residence time is determined by multiplying EBCT by the porosity of the GAC). The optimum EBCT for a system depends on the length of the MTZ, which is affected by numerous factors including carbon type, concentrations of SOCs present in the water, and background water quality. For many contaminants, municipal water treatment systems have EBCTs ranging from a few minutes to more than 10 minutes, with 15 minutes being a relatively safe upper limit for most groundwater contaminants (Lehr, 1991). However, for weakly adsorbing compounds such as MTBE, longer EBCTs (up to approximately 20 minutes) may be required. As mentioned previously, the optimum EBCT for specific water conditions can be evaluated by performing dynamic column tests with the water in question. The configuration of a GAC system can be used to increase the volume of water treated per mass of carbon used (i.e., the specific volume ). As discussed in Section 4.2.1, the use of multiple carbon beds in-series allows for more complete saturation of the lead bed(s) prior to changeout, with the secondary (lag) bed(s) in the series used as a polish or guarantee against total system breakthrough. The potential increase in specific volume due to in-series operation depends on the shape of the MTZ, which varies for different compounds (and different influent conditions). For TCE and PCE, in series operation was found to increase specific volume by 35 to 50 percent in comparison to single-bed operations (Hand et. al., 1989; Zimmer et. al., 1988). Of course the added cost for multiple beds is only justified if the savings in carbon consumption compensates for the higher capital and labor cost associated with the in-series operation. Other operating parameters that will impact the effectiveness of carbon adsorption systems include hydraulic loading (i.e., linear flow rate) and bed depth. Typical hydraulic loadings 225

248 (calculated as volumetric flow rate divided by the cross-sectional area of the GAC vessel) for GAC systems range from 2 to 6 gpm/ft 2. Bed depth for a given vessel determines the overall volume of GAC available in the treatment zone. Deeper bed depths provide larger treatment zones and also enhance the flow distribution and water contact within the vessel. Carbon Changeout Requirements For weakly adsorbing compounds such as MTBE, frequent carbon changeouts may be required. This is a critical variable to account for in cost analyses and system design. As discussed previously, most drinking water treatment applications will require virgin carbon in order to achieve stringent effluent standards, particularly for weakly adsorbing compounds such as MTBE. Because of this, operations will include periodic replacement with virgin carbon and off-site regeneration or disposal of the spent carbon Carbon Vendors Table 4-2 presents product information for the vendors that were contacted for this study, including: Barneby & Sutcliffe (Columbus, OH) Calgon Carbon Corporation (Pittsburgh, PA) Carbon Link Corporation (Columbus, OH) CARBTROL (Westport, CT) U.S. Filter/Westates (Los Angeles, CA) Each of these vendors was solicited for data pertaining to promising carbons for MTBE treatment. The carbon products listed in Table 4-2 were those recommended by the responding vendors for treatment of MTBE. As shown, most of the vendors recommended coconut shell carbon. Also included are other carbon products (i.e., coal-based and reactivated coal-based) that have established isotherms for MTBE. It should be noted that several other carbon vendors were contacted for this study, but either did not respond or the response was too limited to include in this report. The unit costs presented in Table 4-2 are based on price quotes given by the respective vendors in It is important to note that unit costs for GAC will vary as a function of quantity purchased and location Level of Development of Technology Carbon adsorption is a very well-established technology with decades of practical experience and research. As such, this technology has well-defined methods and widely available 226

249 equipment. There currently are numerous systems operating for drinking water treatment for a variety of SOCs. Knowledge and experience with carbon systems for treatment of MTBE are limited. As presented in Section 4.2.3, MTBE adsorption isotherms for several GACs are available. However, very little detailed information and data regarding full-scale performance, impact of NOM, and competitive adsorption effects specific to MTBE removal were encountered in the literature reviewed for this chapter. Table 4-2 Vendor Information Vendor Name Address/Phone Product Name Carbon Source Unit Cost Barneby & Sutcliffe P.O Box A coal - Corporation Columbus OH PC coconut shell $1.10/lb Calgon Carbon P.O Box 717 FS 300 coal - Corporation Pittsburgh, PA PCB coconut shell $1.25/lb REACT reactivated coal GAC - FS 600 coal - Carbon Link CARBTROL U.S.Filter/Westates 4174 Fisher Road Columbus, OH Microcarb SXO coconut shell Riverside Avenue Westport, CT CSL coconut shell $0.65/lb South Boyle Ave. Los Angeles, CA CC-602 coconut shell $1.35/lb A Note: A Includes changeout labor and transport to local regeneration facility. 227

250 4.2.6 Technical Implementability The implementation of carbon adsorption technology is expected to be straightforward because the methods and equipment are very well established. The following is a list of important issues related to implementation of GAC for MTBE removal: Many vendors supply GAC and treatment equipment. These vendors have extensive knowledge of installation and operation of carbon systems. Activated carbon technology can be implemented quickly due to its commercial availability. Site-specific bench- and/or pilot-scale testing is recommended for most treatment scenarios to allow for efficient design. Carbon adsorption systems are mechanically simple (i.e., very few moving parts). Pretreatment may be required (e.g., filtration, disinfection) to maximize removal efficiency and limit biofouling, though the presence of MTBE does not cause the need for any specialized pretreatment. Power costs are relatively minor; water flow can be gravity- or pump-driven. The primary maintenance requirements are associated with carbon changeouts (i.e., regeneration or disposal of spent carbon), which may be frequent due to MTBE characteristics. Periodic backflushing of carbon vessels may be required to remove accumulated biological growth. Careful monitoring of influent, midfluent, and effluent streams is required to assess system effectiveness and limit breakthrough events. In addition to monitoring for dissolved-phase MTBE, sampling plans need to include analyses for DOC, metals such as iron and manganese, and other site-specific variables that can cause carbon fouling and limit adsorption efficiency. Monitoring frequency should be based on site-specific conditions such as expected breakthrough time and regulatory requirements Permitting The permitting of carbon adsorption systems is generally straightforward. Standard permitting required for any type of on-site treatment system may be needed as required by local regulatory authorities. No off-gas is produced from carbon adsorption systems therefore no air discharge permits are needed. This is a significant advantage in some areas. Permitting for drinking water applications is discussed in Chapter 1. The carbon adsorption method does produce a significant amount of byproducts/residuals in the form of spent carbon. As such, transport and disposal permits may be needed. If spent carbon is regenerated onsite, permits for the regeneration facility will be required and may include air discharge permits. In general, permitting for application of carbon adsorption systems should be relatively simple in comparison to other technologies. 228

251 4.2.8 Current Usage of Technology As described previously, GAC is a well-established and widely used technology for removing organic compounds from water. Unfortunately, documentation of full-scale applications for MTBE removal is limited. Creek and Davidson (1998) compiled carbon treatment data from six confidential MTBE point-source remediation sites (i.e., not drinking water treatment). Relevant results from these sites are presented as part of Table 4-3. This table also presents results of a recent application of carbon adsorption at the Arcadia well field in Santa Monica, California (for extracted water from an aquifer pump test; per Komex H2O Science, 1998) and results of a well-documented drinking-water treatment case study in Rockaway Township, New Jersey (McKinnon and Dyksen, 1984). As shown in Table 4-3, the performance of GAC for MTBE removal varied widely. For the eight sites reviewed, estimated carbon usage rates ranged from less than 0.4 lbs/1,000 gallons to over 20 lbs/1,000 gallons treated. At the two sites where carbon performed most effectively (i.e., lowest carbon usage rates), influent MTBE was less than approximately 50 µg/l, BTEX was not present, and other contaminants were present at low concentrations. Conversely, the two sites where GAC performed least effectively (i.e., highest carbon usage rates) had relatively high MTBE and BTEX concentrations. In addition to SOC concentrations, other variables that can affect usage rates include variations in GAC type and quality, background water quality, and system operating parameters (e.g., EBCT). At the well-documented Rockaway site, coal-based carbon (Calgon Filtrasorb 300) was used for removal of MTBE, di-isopropyl ether (DIPE), and TCE (McKinnon and Dyksen, 1984). Based on carbon usage rates, the system at this site appears to have performed better for MTBE removal in comparison to most of the remediation sites reviewed. However, five complete carbon changeouts were required during the first 15 months of operations, causing the system operators to switch to air stripping as the primary treatment method. For the five sites at which BTEX compounds were present, estimated usage rates ranged from 2 to 23 lbs/1,000 gallons. Influent MTBE concentrations at these five sites varied widely but were always greater than 270 µg/l. It is unknown what type of GAC was used for these systems. For treatment of the extracted water from the pump test at the Arcadia well field, reactivated coal-based carbon (Calgon REACT) was used. As indicated on Table 4-3, the system performed very poorly, with an estimated usage rate of 4.9 lbs/1,000 gallons for an average influent concentration of 37 µg/l MTBE. This system used EBCTs ranging from 12 to 20 minutes, which is an adequate range for removal of most organic compounds. Based on the data available from this site, it appears that the primary cause of the poor performance is the type of carbon used (i.e., reactivated coal-based carbon). 229

252 Table 4-3 Estimated Carbon Usage Rate for Full-scale Systems Other Estimated A Site/Project System Influent MTBE Contaminants Usage Rate Description Design (µg/l) Present (lb/1000 gal) Aquifer Pump Test B, 2-15,000 lb vessels in series none 4.9 Arcadia Well Field, avg flow rate = 260 gpm average 37 Municipal Drinking Water Supply, EBCT = minutes Santa Monica, California reactivated coal GAC Calgon REACT Municipal Drinking Water 2-20,000 lb vessels in parallel DIPE: µg/l 0.4 Treatment System C, avg flow rate = 800 gpm TCE Rockaway Township, EBCT = 10 minutes New Jersey bituminous coal GAC Calgon F-300 Service Station Remediation D, 2-1,000 lb vessels in series 15,000-23,000 BTEX present 23 California avg flow rate = 1 gpm benzene: 23,000 µg/l unknown carbon Leaking UST Remediation D, lb vessels in series 400-2,720 BTEX present 13 Vermont avg flow rate = 3 gpm benzene: 1,270 to unknown carbon 3,970 µg/l Service Station Remediation D, 2 vessels in-series (2000 lb, 500 lb) 1,100-5,000 BTEX present 3 Massachusetts avg flow rate = 5 gpm benzene: 1,200 to unknown carbon 2,100 µg/l Service Station Remediation D, lb vessels in series ,400 BTEX present 2.3 Massachusetts flow rate = 1-2 gpm benzene: 636 to unknown carbon 11,000 µg/l Service Station Remediation D, GAC used for air stripper polish < TBA: up to 630 µg/l <0.4 New Jersey single 500 lb vessel estimated avg flow rate = 0.5 gpm unknown carbon Service Station Remediation D, 3-1,000 lb vessels in series up to 14,700 BTEX present 2.2 Massachusetts avg flow rate = 21 gpm unknown carbon A Carbon usage rates estimated based on approx. breakthrough time/changeout frequency and average flow rate. It is important to note that usage rates are affected by numerous variables including GAC type and quality, concentrations of SOCs, background water quality, and system operating parameters such as EBCT. B See Komex H20 Science, C See McKinnon and Dyksen, D See Creek and Davidson,

253 An example of desorption of MTBE occurred at a service station site in Massachusetts, where a 2 gpm GAC system (two 300-lb vessels in-series) was used to remove MTBE and BTEX from a contaminated groundwater source (Creek and Davidson, 1998). At this site, desorption caused effluent MTBE concentrations from the first carbon vessel to be higher than influent concentrations in 22 of 73 monthly sampling events. In addition, effluent MTBE concentrations from the second vessel exceeded influent concentrations 11 times over the 6 years of operation. Other examples of MTBE desorption are presented elsewhere (e.g., three field sites in API, 1990) Summary of Ongoing Research The University of California, Los Angeles (UCLA) and Calgon Carbon Corporation (Pittsburgh, PA) are currently performing parallel research projects with GAC and MTBE. The scopes of these projects, which are being performed under contract with the MTBE Research Partnership, are presented below. In the year 2000, the California MTBE Research Partnership plans to publish a report that summarizes the results of these studies in addition to revised computer modeling and cost analyses. UCLA Research On behalf of the California MTBE Research Partnership, UCLA is currently performing a series of RSSCTs using MTBE influent concentrations ranging from 20 µg/l to 2,000 µg/l. Two different coconut shell carbons are being tested, including Calgon s PCB carbon and U.S. Filter s CC-602 carbon. Three different background water sources are being tested to evaluate the effects of NOM on carbon adsorption. The waters being tested include groundwater from the Arcadia well field in the City of Santa Monica (California), groundwater from South Lake Tahoe Public Utility District (California), and water from Lake Perris, a drinking water source for the Metropolitan Water District (Los Angeles, California). For several of the tests, influent water is being spiked with BTEX compounds or TBA to quantify the effects of competitive adsorption. The results of this study are expected to be ready for publication in the year Calgon Research On behalf of the California MTBE Research Partnership, Calgon Carbon Corporation (Pittsburgh, PA) is performing a study that parallels the work being performed by UCLA. Calgon is running a series of column tests using PCB coconut shell carbon and the ACT method. Calgon is evaluating the same range of MTBE influent concentrations for the same source waters as the UCLA research. Several of the tests will incorporate spiking of BTEX compounds or TBA. Similar to the UCLA study, results of this project are expected in the year

254 232

255 4.3 Detailed Evaluation of GAC This section presents a detailed evaluation of carbon adsorption technology for the treatment of MTBE-impacted water. The results of computer modeling and cost estimates for various MTBE treatment scenarios are presented in order to determine the most cost effective applications for GAC. The treatment scenarios considered for this evaluation were based on the following range of conditions: Flow Rates: 60 gpm, 600 gpm, and 6,000 gpm. Influent MTBE: 20 µg/l, 200 µg/l, and 2,000 µg/l. Effluent MTBE Targets: 20 µg/l, 5.0 µg/l, and 0.5 µg/l. Section presents a discussion of the computer modeling including a description of the AdDesignS model (Mertz et al. 1994) and the primary assumptions used. Detailed evaluations of flow rate (Section 4.3.2), removal efficiency (Section 4.3.3), and other characteristics (reliability, flexibility, and adaptability) that may influence the selection of GAC technology for MTBE removal (Section 4.3.4) are then presented. Section presents results of the cost estimates based on carbon usage rates estimated from the computer modeling. Section discusses sensitivity analyses performed for varying background water-quality conditions (NOM, BTEX loading) and other variables that impact the cost estimates Computer Modeling In order to evaluate GAC under the range of conditions listed above, computer modeling was performed using the Michigan Technological University AdDesignS model (Mertz et. al., 1994). This model consists of equilibrium and mass transfer models that can be used to simulate multi-component adsorption in fixed-bed adsorbers. The primary mass transfer model used for this study (the Pore Surface Diffusion Model, or PSDM) accounts for both surface diffusion and pore diffusion as intraparticle mass transfer mechanisms. These mechanisms are modeled using a series of well-established analytical equations, as presented in the program manual (Mertz et al., 1994). The PSDM can account for carbon fouling due to NOM and competitive effects due to the presence of multiple SOCs. The model accounts for the effects of competitive adsorption using the Ideal Adsorbed Solution Theory (Fritz and Schundler, 1981). The AdDesignS modeling package also includes several databases that allow the user to access compound physical properties, adsorbent properties, manufacturer fixed bed adsorber specifications, adsorption equilibrium properties, and kinetic properties. Further information regarding the computer model, which has recently become commercially available, is presented in the program manual (Mertz et al., 1994). 233

256 Assumptions Used in Computer Modeling Numerous assumptions were required in order to utilize the AdDesignS computer model for predicting carbon performance. In addition to the primary assumptions regarding flow rates and concentrations of MTBE in the influent water, detailed assumptions of carbon characteristics, background water quality, etc., were used for the modeling. These assumptions are discussed below. Carbon Specifications The carbon specifications for the modeling were selected in order to be representative of a high-grade coconut shell carbon. The specifications listed below match those of U.S. Filter/Westates CC-602 carbon, as presented in product literature. Mesh size: 12x30 Apparent density: g/cm 3 Porosity: Particle radius: cm Freundlich Parameters Although complete isotherm data for MTBE and BTEX with U.S. Filter/Westates CC-602 were available, adjusted parameters for MTBE were used to better reflect typical values found for coconut-based carbons. The values used in the modeling are as follows: K F = 11.0 (mg/g)(l/mg) n n = 0.50 These selected parameters are mid-range for the values reported by the vendors for coconutbased carbons (see Table 4-1 for the range of Freundlich parameters). Based on the values reported by Barneby & Sutcliffe (Columbus, OH), U.S. Filter/Westates (Los Angeles, CA), and Calgon Carbon Corporation (Pittsburgh, PA), these values are considered representative of high-grade, virgin coconut shell carbon. Background Water Quality The AdDesignS model can simulate time-dependent fouling that occurs due to the presence of NOM in the water being treated. The model has several waters available, representing a range of fouling effects. As recommended in the AdDesignS manual, Karlsruhe (West Germany) groundwater was used to represent the background water quality for the modeling scenarios. The moderate-level fouling parameters for this water were determined by preloading isotherm studies using Calgon F-100 carbon (equivalent to Filtrasorb 400). The 234

257 carbon fouling for MTBE was represented by the chemical correlation for aromatic compounds. Although it is recognized that MTBE is not an aromatic, none of the other chemical correlation types available describe MTBE and, based on an abbreviated sensitivity analysis performed for this study (not presented), the correlation for aromatic compounds gives conservative results. Temperature The assumed temperature for the modeling was 20 C. Variations in GAC capacity can be expected for different temperatures. However, these variations are considered to be within the accuracy of the evaluations reported here. Vessel Specifications The AdDesignS model has standard specifications available for several Carbonair fixed-bed adsorbers, including the following types: PC-20 (2,500 lb capacity), PC-28 (5,000 lb capacity), and PC-78 (20,000 lb capacity). For each of the flow rate/influent concentration scenarios, one of the vessel types listed above was selected for modeling based on flow rate capabilities and/or preliminary modeling. The vessel was then modeled at the appropriate flow rate/influent conditions in order to determine the carbon usage rate and breakthrough time for each of the effluent criteria (i.e., 20 µg/l, 5.0 µg/l, and 0.5 µg/l). The vessel specifications for each of the treatment scenarios are listed in Table 4-4. EBCTs As shown on Table 4-4, the EBCTs for the modeling ranged from 10.7 minutes to 22.4 minutes. The effect of optimizing EBCTs was briefly evaluated by varying bed depths for vessels under specific flow rate/influent conditions. Based on the modeling, it was determined that decreases in carbon usage rate due to EBCT optimization were minor and within the accuracy of the evaluations. Therefore, standard EBCTs were selected for each flow rate/influent concentration scenario (constant regardless of the effluent treatment goal). System Configuration For the higher flow rate scenarios (600 gpm and 6,000 gpm), it was necessary to assume inparallel operations because of limited flow capacities of the carbon vessels available. For example, the maximum listed flow capacity of the Carbonair PC-78 is 550 gpm. Detailed assumptions for these scenarios are as follows: 600 gpm systems - two parallel lines of PC-78 vessels operating at 300 gpm per line. 6,000 gpm systems - 12 parallel lines of PC-78 vessels operating at 500 gpm per line. 235

258 Table 4-4 Results of AdDesignS Modeling Modeled Modeled MTZ Single-Vessel Single-Vessel Flow Rate Influent Effluent Adsorber Type/ EBCT Length VTM Breakthrough Usage Rate (gpm) (µg/l) (µg/l) Specifications (min) (cm) (m 3 /kg) (days) (lb/1000gal) Carbonair PC mass = 5300 lbs bed depth = 6.4 ft int. area = 28 ft Carbonair PC mass = 5300 lbs bed depth = 6.4 ft int. area = 28 ft Carbonair PC mass = 2500 lbs bed depth = 4.6 ft int. area = 20 ft A Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft A Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft A Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft B Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft B Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft B Carbonair PC mass = 20,000 lbs bed depth = 9.1 ft int. area = 78.5 ft 2 A Maximum flow capacity of PC-78 is 550 gpm; modeling assumed two vessels in-parallel at 300 gpm each. B Modeling assumed twelve vessels in-parallel at 500 gpm each. EBCT = empty bed contact time. MTZ = mass transfer zone. VTM = water volume treated per carbon mass. 236

259 Results of Computer Modeling Results of the computer modeling are summarized in Table 4-4. As shown on this table, the estimated carbon usage rate for single-vessel systems varies from approximately 0.2 lbs/1,000 gallons of water treated (lb/1,000 gallons) to approximately 2.1 lb/1,000 gallons, depending primarily on the influent concentrations and treatment goals. Detailed discussion of the results, as they pertain to flow rate (Section 4.3.2), removal efficiency (Section 4.3.3), and other characteristics (Section 4.3.4), is presented below. Comparison of the modeling results to data from the full-scale applications presented in Table 4-3 indicates that the modeling predicts significantly better GAC performance than realized under field conditions. However, it is important to note that the modeling was performed assuming favorable conditions with respect to the presence of other SOCs. In contrast, most of the full-scale applications reviewed had BTEX or other organic compounds in the water. In addition, the modeling assumed use of high-grade coconut shell carbon, which is expected to perform better that the coal-based carbons used in many of the full-scale applications reviewed. It should also be noted that the results of the computer modeling for MTBE in natural water compare well to modeling results prepared by Calgon Carbon Corporation (Pittsburgh, PA) and CARBTROL (Westport, CT). In summary, the carbon usage rates predicted by AdDesignS are considered reasonable and accurate within the limitations of computer modeling Flow Rate Table 4-4 presents predicted carbon usage rates for carbon adsorption systems ranging from 60 gpm to 6,000 gpm. As mentioned previously, the predicted usage rates for all the scenarios vary from approximately 0.2 lb/1,000 gallons to approximately 2.1 lbs/1,000 gallons. However, as shown in Table 4-4, the variation in usage rates is not impacted by system flow rate. For example, at an influent concentration of 2,000 µg/l and treatment goal of 5.0 µg/l, the carbon usage rate varies from 1.9 lb/1,000 gallons (60 gpm system) to 2.0 lb/1,000 gallons (6,000 gpm system). The slight difference in usage rate noted here is attributable to variations in treatment system parameters such as EBCT and vessel configuration. Carbon adsorption for MTBE treatment is not technically limited to any range of flow rates. However, in order to treat high flow rates within an acceptable range of EBCTs, it may be necessary to operate carbon beds in parallel, thereby reducing the flow rate that goes through a given bed. For weakly adsorbing compounds that require relatively long EBCTs, it may not be feasible to install high flow rate systems (i.e., capital costs for numerous parallel lines of several beds in-series may be prohibitive). Detailed discussion of cost-effectiveness as a function of system size is presented in Section

260 It should be noted that the capacity of the largest, standard carbon vessel is 20,000 lbs. For modeling of the 600 gpm and 6,000 gpm systems, multiple parallel lines of 20,000-lb vessels were used Removal Efficiency Carbon adsorption for MTBE treatment is not technically limited to a specific range of influent concentrations or treatment goals. For any given treatment scenario, removal of MTBE to a non-detectable concentration can be achieved if enough carbon and adequate contact time are used. However, as discussed below, carbon usage rates (hence cost effectiveness) are highly dependent on influent MTBE concentrations and, to a lesser degree, on effluent treatment goals. Figure 4-3 and results of computer modeling presented in Table 4-4 show that carbon usage rates are closely related to influent MTBE concentrations. For influent concentrations of 20 µg/l MTBE (with no other SOCs and NOM fouling for typical groundwater), the estimated carbon usage rate for single-vessel systems varies from 0.2 to 0.3 lb/1,000 gallons. In contrast, for influent concentrations of 2,000 µg/l, the estimated carbon usage rate under the same conditions varies from approximately 1.9 to 2.1 lbs/1,000 gallons. Table 4-4 shows that carbon usage rates of single-vessel systems are only mildly impacted by effluent treatment goals ranging from 20 µg/l to 0.5 µg/l. For influent MTBE of 20 µg/l, predicted usage rates vary from 0.22 lbs/1,000 gallons (effluent goal = 5.0 µg/l) to 0.26 lbs/1,000 gallons (effluent goal = 0.5 µg/l). For the higher influent concentrations (200 µg/l and 2,000 µg/l), the predicted usage rates vary by even less. It is important to note that most carbon adsorption systems are operated such that effluent remains at non-detectable concentrations. This is accomplished by utilizing multiple beds inseries. As discussed in Section 4.2.1, the lead vessel is taken off-line and replaced with new carbon prior to breakthrough of the lag vessel(s). The lead vessel is then reinstalled as the lag vessel, ensuring that contaminant breakthrough of the overall system is avoided. The carbon usage rate for treatment of MTBE is strongly dependent on the influent MTBE concentration (Figure 4-3). Although it is possible to operate multiple large carbon beds inseries to remove high MTBE concentrations, the carbon usage rate increases significantly with increasing influent MTBE concentrations. This suggests that GAC is more likely to be cost-effective in situations where influent MTBE concentrations are relatively low (e.g., as a polishing step). 238

261 Single Vessel; Removal Efficiency = % ,000 10,000 Influent MTBE (µg/l) 6,000 gpm 600 gpm 60 gpm Predicted Usage Rate (lb/1000 gal) Figure 4-3. Predicted carbon usage rate vs. influent MTBE concentration. Single vessel; removal efficiency = 97.5 to 99 percent. 239

262 4.3.4 Other Characteristics Reliability The mechanical reliability of GAC technology is very high due to the simple nature of the systems. The primary concerns include pump O&M and tank/line leakage. Continuity of operations is typically very high because of the mechanical simplicity of the systems. Although carbon changeout requirements can cause delays in system operation, these stoppages should be predictable as long as influent water conditions are relatively consistent. Also, with the expected multiple vessel designs for MTBE applications, one vessel can usually be taken out of use for changeout while the system operates with the remaining vessels. The duration of a carbon changeout event typically is less than 1 day. The adsorption process is expected to be reliable for removal of MTBE as long as influent conditions are relatively consistent. Flexibility As discussed above, carbon adsorption technology can be implemented for any flow rate, though prohibitively high costs may be encountered at higher flow rates due to equipment size and O&M efforts associated with carbon changeout. Based on results of the adsorption modeling, the primary site factors that determine the effectiveness of GAC for MTBE removal include influent MTBE concentrations, background NOM, and the presence of other SOCs in the influent stream. Although systems can be designed for most influent conditions, the cost-effectiveness of carbon systems may be severely impacted by any one of these factors. This is further discussed in Section Adaptability The adsorption process for MTBE is sensitive to the presence of other more readily adsorbed SOCs (e.g., xylene). If concentrations of other SOCs begin to increase in the influent water, competitive adsorption effects may reduce the removal efficiency for MTBE. Results of the sensitivity analyses (Section 4.3.6) for varying BTEX loads show that carbon usage rates increase significantly if competing SOCs are present. In comparison to natural water with only MTBE (at 20 µg/l), water with a modest additional load of 80 µg/l Total BTEX increases the predicted carbon usage rate by about 17 percent (see Table 4-5). Water with a higher additional load of 800 µg/l Total BTEX increases the predicted usage rate by greater than 50 percent. This conclusion is consistent with data from full-scale applications (see Table 4-3) that show high carbon usage rates at sites with both MTBE and BTEX present. Arrival of other SOCs in the influent water can also cause an additional problem the desorption of MTBE already adsorbed onto the carbon. Desorption of MTBE can occur due to displacement of MTBE molecules from GAC by preferentially adsorbed SOCs, resulting in higher effluent MTBE concentrations. Desorption can also occur from MTBE-loaded 240

263 carbon as influent MTBE concentrations of the extracted water naturally decrease (which changes the equilibrium of the solid-liquid partitioning). If the influent conditions change in either of these manners, desorption of MTBE can happen quickly. Because of the potential for desorption of the weakly adsorbed MTBE, frequent monitoring of influent, midfluent, and effluent waters is critical to assess the process effectiveness. The arrival of other more strongly adsorbed compounds in the influent stream should be taken as a warning that removal efficiency for influent MTBE will decrease and that desorption of previously adsorbed MTBE may begin very soon. Because of desorption and the effects of competitive adsorption, frequent sampling is critical to monitor system effectiveness and limit the potential for MTBE breakthrough. The feasibility of carbon adsorption technology for MTBE is not expected to be impacted by regulatory requirements for drinking water. Based on the computer modeling, it appears that variation in effluent treatment goals from 0.5 µg/l to 20 µg/l has only a mild effect on the carbon usage rate (hence the cost-effectiveness). In addition, GAC treatment systems generally are operated such that no detectable contaminants are present in the effluent, particularly in drinking water applications. Because of this, changing regulatory requirements for MTBE effluent standards are not expected to significantly impact the feasibility of GAC for MTBE treatment. 241

264 Table 4-5 Results of AdDesignS Modeling for Sensitivity Analyses Modeled Modeled Single-Vessel Single-Vessel Predicted B Predicted B Predicted B Water Type A VTM Breakthrough Usage Rate Usage Rate Vessel Life Changeouts (m 3 /kg) (days) (lb/1000gal) (lb/1000gal) (days) Per Year Rhine River water (high fouling) Karlsruhe ground water (moderate fouling) Wausau ground water (low fouling) Organic-Free Water Moderate BTEX C each at 200 µg/l Low BTEX C each at 20 µg/l No BTEX C each at 0 µg/l Organic-Free Water EBCT - empty bed contact time VTM - volume treated per carbon mass A Water types are those available in the AdDesignS model (Mertz et al., 1994); calculated values are based on time-dependent fouling correlations determined for coal-based GAC and other organic compounds (not MTBE). B Values estimated using assumed capacity increase of 50% due to in-series operation. C BTEX and MTBE in Karlsruhe groundwater. Breakthrough, VTM, and calculated usage rate based on equilibrium column modeling using the AdDesignS model. Assumptions: Influent MTBE = 20 mg/l; effluent contains no detectable MBTE (<0.5 mg/l). Total flow rate = 600 gpm; two parallel lines with 2 vessels in series (300 gpm per line) System type: Carbonair PC-78, bed mass = 20,000 lbs, bed depth = 9.1 ft EBCT = 17.8 minutes 242

265 4.3.5 Cost Estimates The purpose of this section is to present the estimated costs for carbon adsorption technology under a range of MTBE treatment scenarios. The cost estimates were developed for direct comparison with the estimated costs developed for air stripping (Chapter 2), advanced oxidation (Chapter 3), and resins (Chapter 5). Comparing the relative costs of the technologies allows for determination of the most promising applications for GAC. In addition to estimates for the range of MTBE treatment conditions discussed at the beginning of Section 4.3, cost estimates for several other scenarios are presented to investigate the cost impact sensitivity of varying degrees of NOM fouling and BTEX loading, and of different system design lifetimes. Cost Estimating Methods and Assumptions Feasibility-level cost estimates were developed using vendor-supplied costs and standardized assumptions for design/engineering, contractor overhead and profit (O&P), and contingency. Preliminary design configurations and results of computer modeling were used to predict carbon usage rates and vessel changeout frequencies. The primary assumptions used in development of the cost estimates include: Flow rates: 60 gpm, 600 gpm, and 6,000 gpm. Influent MTBE concentrations: 20 µg/l, 200 µg/l, and 2,000 µg/l. No detectable MTBE in effluent (<0.5 µg/l). Preliminary system designs established based on computer modeling and engineering judgement; in-series operation (three vessels in-series) assumed for all scenarios; number of parallel lines varies. Carbon usage rates and vessel breakthrough times determined by computer modeling using the AdDesignS model by Michigan Technical University (Tables 4-4, 4-6). It was assumed that adsorption efficiency is increased by 50 percent due to in-series operation (see Note). Modeling used NOM fouling correlations for groundwater from Karlsruhe, West Germany. This water type is considered to cause moderate carbon fouling and is recommended by the authors of AdDesignS as representative of typical groundwater conditions. Unit costs for carbon and system capital/installation costs based on 1998 vendor quotes from Calgon Carbon Corporation (Pittsburgh, PA), U.S. Filter/Westates (Los Angeles, CA), CARBTROL (Westport, CT), and Carbonair (New Hope, MN). Standard percentage rate assumptions for design/engineering, contractor overhead and profit, and contingency. Standard system design life of 30 years with seven percent interest rate for capital amortization. 243

266 Note: The adsorption capacities and breakthrough times calculated by the AdDesignS model for single-vessel systems were increased to account for the benefits of in-series operation. As mentioned previously, in-series operation was found to increase specific volume for compounds with long MTZs (i.e., TCE and PCE) by up to 50 percent in comparison to single-bed operations (Hand et. al., 1989; Zimmer et. al., 1988). Preliminary results of bench-scale column testing by UCLA (Suffet, personal communication, 1999) show that MTBE adsorption requires a relatively long MTZ. These results indicate that in-series operation is capable of significantly increasing specific GAC volume for MTBE. As presented in Table 4-6, the single-vessel breakthrough times predicted were increased by 50 percent for in-series operation with effluent having no detectable MTBE (<0.5 µg/l). Similarly, the volume treated per carbon mass was increased by 50 percent. These assumptions are supported by field experience and industry expertise (Graham, personal communication, 1998). Table 4-7 presents summaries of the cost evaluations for the MTBE treatment scenarios. Further details of methods and assumptions used for the cost estimates are presented in Appendix 4A. Individual costs for each of the scenarios are presented in a series of spreadsheets (Table 4A-3 in Appendix 4A). Discussions regarding critical parameters that impact the costs for GAC are presented below. Effect of Flow Rate Figure 4-4 shows the relationship of unit treatment cost (i.e., $/1,000 gallons) vs. system flow capacity for varying influent MTBE concentrations. Based on this figure and the costs presented in Table 4-7, it is clear that there is economy of scale, particularly for the lower range of flow rates (600 gpm and lower). In this lower range, unit costs decrease significantly as system flow capacity increases. However, the differences in estimated unit costs for the 600 gpm and the 6,000 gpm systems are small to negligible. This lack of economy of scale is primarily because the largest standard carbon vessels are designed for maximum flow rates of approximately 500 to 600 gpm. Flow rates above this level are handled by operating multiple vessels in parallel; hence, there are no significant capital savings for higher flow rates. In addition, as discussed in Section 4.3.2, carbon usage rates (lbs/1,000 gallons) are not impacted by flow rate; as such, O&M costs (and unit treatment costs) increase proportionally with flow rate. 244

267 Table 4-6 Predicted Carbon Usage Rates and Breakthrough Times Using In-series Operation System Flow Rate = 60 gpm; single vessel EBCT = 22.4 minutes Modeled Modeled Influent Single-Vessel Single-Vessel Predicted B Predicted B Predicted B MTBE Breakthrough A Usage Rate A Usage Rate Vessel Life Changeouts (µg/l) (days) (lb/1000gal) (lb/1000gal) (days) Per Year C System Flow Rate = 600 gpm; single vessel EBCT = 17.8 minutes Modeled Modeled Influent Single-Vessel Single-Vessel Predicted B Predicted B Predicted B MTBE Breakthrough A Usage Rate A Usage Rate Vessel Life Changeouts (µg/l) (days) (lb/1000gal) (lb/1000gal) (days) Per Year System Flow Rate = 6000 gpm; single vessel EBCT = 10.7 minutes Modeled Modeled Influent Single-Vessel Single-Vessel Predicted B Predicted B Predicted B MTBE Breakthrough A Usage Rate A Usage Rate Vessel Life Changeouts (µg/l) (days) (lb/1000gal) (lb/1000gal) (days) Per Year A Results of AdDesignS computer model for effluent containing 5 mg/l MTBE. B Values estimated using assumed capacity increase of 50% due to in-series operation. C For this 60 gpm scenario (20 mg/l influent), smaller GAC vessels used with EBCT = 11.1 minutes. 245

268 Table 4-7 Summary of Cost Estimates Flow Rate System Influent MTBE Capital Cost Annual O&M Total Annual Unit Cost (gpm) Configuration (µg/l) ($) ($) Cost ($) ($/1,000 gal) lb beds, 20 $150,000 $61,000 $73,000 $ in series lb beds, 200 $234,000 $79,000 $98,000 $ in series lb beds, 2,000 $234,000 $127,000 $146,000 $ in series 20,000 lb beds, parallel lines, 20 $1,019,000 $161,000 $243,000 $ in series 20,000 lb beds, parallel lines, 200 $1,019,000 $282,000 $364,000 $ in series 20,000 lb beds, parallel lines, 2,000 $1,019,000 $665,000 $747,000 $ in series 20,000 lb beds, 6, parallel lines, 20 $5,979,000 $1,091,000 $1,573,000 $ in series 20,000 lb beds, 6, parallel lines, 200 $5,979,000 $2,575,000 $3,056,000 $ in series 20,000 lb beds, 6, parallel lines, 2,000 $5,979,000 $6,526,000 $7,008,000 $ in series Note: 1. Cost estimates developed from results of adsorption modeling using the AdDesignS model (Mertz et al., 1994). 2. Detailed cost estimates and assumptions presented in Appendix 4A. 3. Cost estimates developed for in-series operation; effluent contains no detectable MTBE (<0.5 mg/l). 246

269 Estim ted Unit Tre tment Cost vs. System Flow C p city In-Series Operation ,000 10,000 System Flow Capacity (gpm) Infl. MTBE 2000 ug/l Infl. MTBE = 200 ug/l Infl. MTBE 20 ug/l Infl. MTBE = 2,000 µg/l Infl. MTBE = 200 µg/l Infl. MTBE = 20 µg/l Unit Treatment Cost ($/1000 gal) Unit Treatment Cost ($/1,000 gal) Figure 4-4. Estimated unit treatment cost vs. system flow capacity for in-series operation. 247

270 In-Series Operation ,000 10,000 Influent MTBE (µg/l) 60 gpm 600 gpm 6,000 gpm 60 gpm 600 gpm 6,000 gpm Unit Treatment Cost ($/1,000 ($/1000 gal) Figure 4-5. Estimated unit treatment cost vs. influent MTBE concentration for in-series operation. 248

271 Effect of Influent MTBE Concentration Figure 4-5 presents unit treatment cost vs. influent MTBE concentration for in-series operation. This figure illustrates that unit costs increase significantly with increasing influent concentrations. As shown on Table 4-7, for the 60-gpm systems, the estimated unit costs vary from $2.30/1,000 gallons to $4.43/1,000 gallons. For the 600-gpm and 6,000-gpm systems, estimated unit costs vary from $0.77 to $2.37/1,000 gallons, and from $0.50 to $2.22/1,000 gallons, respectively. Effect of Removal Efficiency As discussed previously, computer modeling indicates that effluent treatment goals between 20 µg/l and 0.5 µg/l result in minimally different carbon usage rates. Therefore, it appears that removal efficiency will have only a minor impact on treatment costs for a single-vessel system. Because in-series operation with no detectable MTBE in the effluent were assumed for the cost estimates, no further conclusions can be drawn regarding cost effectiveness for varying removal efficiencies Sensitivity Analyses Sensitivity analyses were performed for several critical parameters using the AdDesignS computer model and the cost estimating methods described in Section and Appendix 4A. The parameters evaluated using the adsorption model include NOM and BTEX loading (Table 4-8). These parameters were varied to determine their relative impact on carbon usage rates and cost effectiveness. In addition, system design life was varied to evaluate impacts to unit treatment costs. This section summarizes assumptions used in the sensitivity analyses and presents discussions of the modeling results (Table 4-5) and associated cost estimates (Table 4-8). Assumptions Used in Sensitivity Analyses All sensitivity analyses were performed for a single treatment scenario (influent MTBE of 20 µg/l, flow rate of 600 gpm) using the same general assumptions discussed in Section Brief discussions of additional assumptions used in these sensitivity analyses are presented on the following pages. 249

272 Table 4-8 Cost Estimates for Sensitivity Analyses Sensitivity Changeouts Capital Cost Annual O&M Total Annual Unit Cost Parameter per Year A ($) ($) Cost ($) ($/1,000 gal) NOM Fouling Wausau ground water 1.1 $1,019,000 $156,000 $238,000 $0.75 (low fouling) Karlsruhe ground water 1.2 $1,019,000 $161,000 $243,000 $0.77 (moderate fouling) Rhine River water 1.6 $1,019,000 $183,000 $265,000 $0.84 (high fouling) BTEX Load No BTEX present 1.2 $1,019,000 $161,000 $243,000 $0.77 each at 0 µg/l Low BTEX 1.4 $1,019,000 $172,000 $254,000 $0.81 each at 20 µg/l Moderate BTEX 1.9 $1,019,000 $199,000 $281,000 $0.89 each at 200 µg/l Design Life 2 years 1.2 $1,019,000 $161,000 $701,000 $ years 1.2 $1,019,000 $161,000 $287,000 $ years 1.2 $1,019,000 $161,000 $243,000 $0.77 A Predicted based on AdDesignS modeling; see Table 4-5. Primary Assumptions for cost estimates given above: 600 gpm systems 2 parallel lines of 3 beds in-series Influent MTBE = 20 mg/l Effluent contains no detectable MTBE (<0.5 mg/l) NOM fouling for Karlsruhe groundwater unless otherwise noted. 250

273 NOM The impact of varying background NOM in water was evaluated by changing the carbon fouling parameters during the adsorption modeling. The AdDesignS model has fouling correlation data for several different water types with varying degrees of fouling potential. The different water types used in the modeling, and their relative fouling effects, are as follows: Surface water from Rhine River, Germany Groundwater from Karlsruhe, Germany Groundwater from Wausau, WI High fouling Moderate fouling Low fouling As discussed in Section 4.3.1, groundwater from Karlsruhe, West Germany was used to model the majority of the treatment scenarios because it is considered representative of typical groundwater. Crittenden et al. (1989) report further details on the characteristics of these waters. Presence of Other SOCs Two different scenarios were modeled to evaluate the effects of competitive adsorption from other SOCs. The compounds evaluated, and their concentrations, are as follows: BTEX at 20 µg/l each (Total BTEX = 80 µg/l). BTEX at 200 µg/l each (Total BTEX = 800 µg/l). The assumed Freundlich isotherm parameters used for the BTEX compounds were those reported by U.S. Filter (Los Angeles, CA) for CC-602 GAC (Graham, personal communication, 1998). These values are as follows: Benzene: K F = 50 (mg/g)(l/mg) n n = 0.53 Toluene: K F = 97 (mg/g)(l/mg) n n = 0.43 Ethylbenzene: K F = 163 (mg/g)(l/mg) n n = 0.41 Xylene: K F = 184 (mg/g)(l/mg) n n = 0.47 System Design Life Cost estimates for three system design life durations (2 years, 10 years, and 30 years) were completed to evaluate the effects on unit treatment costs. Cost estimates were prepared by amortizing capital costs over each of these different system design periods. 251

274 Impact of NOM Results of computer modeling and cost estimates using the predicted carbon usage rates for the three different water types are summarized in Tables 4-5 and 4-8. These results show that the carbon usage rate for MTBE removal varies by as much as approximately 50 percent, depending on the degree of NOM fouling (Table 4-5). This difference in carbon usage rate results in differences in unit treatment costs of up to 12 percent for in-series operation under the assumed treatment conditions. For example, Table 4-8 shows that estimated unit treatment costs vary from $0.75/1,000 gallons to $0.84/1,000 gallons for water with low and high fouling, respectively. Impact of BTEX Loading The impact of BTEX loading on the adsorption of MTBE was investigated by varying influent BTEX concentrations from Total BTEX loads of 80 µg/l (each of the four components at 20 µg/l) to 800 µg/l (each of the four components at 200 µg/l). Results of adsorption modeling and the cost estimates using the predicted carbon usage rates for the BTEX loading scenarios are summarized in Tables 4-5 and 4-8. The adsorption modeling predicts that the carbon usage rate is increased by more than 50 percent if Total BTEX loads of 800 µg/l are added to the water being treated. This difference in carbon usage rate results in increased unit treatment costs of up to 16 percent. Table 4-8 shows that predicted unit costs range from $0.77/1,000 gallons to $0.89/1,000 gallons as Total BTEX loads increase from non-detect up to 800 µg/l (for the assumed treatment scenario used in the sensitivity analysis). Impact of System Design Life In order to investigate how the assumed design life affects unit treatment costs, the assumed design life was varied between 2 years and 30 years. Using the selected treatment scenario discussed above (i.e., 600 gpm, 20 µg/l MTBE influent, 0.5 µg/l effluent), unit costs were estimated for 2-year and 10-year design periods, in addition to the 30-year period used for the majority of the estimates. The results of these estimates, which are presented in Table 4-8 and Appendix 4A, indicate that the unit treatment cost increases by almost 200 percent when design period is changed from 30 years to 2 years. Specifically, as shown on Table 4-8, predicted unit costs increase from $0.77/1,000 gallons for a 30-year design life to $2.22/1,000 gallons for a 2-year design life. Although this may seem to be a dramatic change, it is less than the unit cost increases expected for more capital-intensive technologies such as resins and AOP. Because the total annual costs for GAC systems are highly dependent on O&M costs, the assumed system design life does not impact the unit treatment cost as much as for higher capital technologies. As such, expected project duration is an important variable to consider when comparing costs of treatment technologies. 252

275 4.4 Conclusions and Research Recommendations Conclusions Based on the literature review, the computer modeling, and the cost analyses, conclusions regarding the feasibility of GAC for aqueous-phase treatment of MTBE are as follows: 1. As indicated by isotherm data (Table 4-1 and Figure 4-2), activated carbons from different source materials have a wide range of adsorption capacities for MTBE. Coconut shell carbon, which often has a slightly higher cost per pound, generally has better adsorption characteristics for MTBE than coal-based carbon. However, carbon quality is more difficult to control for coconut shell carbon than coal-based carbon due to the heterogeneity of the source material (coconut husks). 2. As shown on Table 4-4 and Figure 4-5, carbon usage rates and unit treatment costs are highly dependent on influent MTBE concentrations. As influent MTBE concentrations are increased, carbon usage rate and unit treatment costs increase significantly. This suggests that GAC is most likely to be cost-effective for removal of relatively low MTBE concentrations (e.g., less than 2,000 µg/l). 3. Carbon usage rates and unit treatment costs are highly dependent on the characteristics of the background water. As shown on Table 4-5, computer modeling predicts that carbon fouling from NOM can cause up to 50 percent increases in carbon usage rates for removal of low MTBE concentrations (i.e., 20 µg/l). Due to this sensitivity, GAC is more likely to be cost-effective for waters that are relatively clean with respect to NOM (e.g., some groundwaters). Research suggests that there is no single water quality parameter (e.g., DOC) that is clearly indicative of fouling potential. As such, isotherm and column testing on site-specific waters should be performed to predict fouling effects on carbon usage rates. 4. Carbon usage rates and unit treatment costs are also highly dependent on the presence of other SOCs. Adsorption modeling shows that moderate loads of Total BTEX (800 µg/l) can cause greater than 50 percent increases in carbon usage rates for GAC systems treating influent with 20 µg/l MTBE Recommendations for Future Research Based on the literature review, the computer modeling, and the cost analyses, there are several topics that require more research before GAC usage for MTBE removal from drinking water is fully understood. These topics are: 1) reproducible isotherms for different GAC types; 2) dynamic GAC adsorption capacities for different background water qualities; and, 3) full-scale performance of GAC under field conditions. 253

276 Reproducible Isotherms Although there are MTBE isotherms available for various GAC products, these isotherms were produced using a variety of laboratory testing conditions (e.g., temperatures, background waters). As such, comparisons between these isotherms are of limited value. In addition, based on documented use of GAC for removing MTBE, performance of different GAC types varies widely (Table 4-3). For these reasons, it is recommended that standardized testing be performed to obtain comparable and reproducible isotherms for a range of GAC types, including high-grade coconut shell carbon and coal-based carbon. Prior to testing, consideration should be given to the variability in specific GAC products in order to ensure that isotherm data are developed for representative samples of GAC. Dynamic GAC Adsorption Capacities Dynamic column tests should be performed to determine GAC usage rates, optimum EBCTs, and other operating parameters for a variety of background-water conditions and GAC types (i.e., coconut shell and coal-based). Several waters with differing NOM characteristics (e.g., surface water and groundwater) should be tested under a range of MTBE influent concentrations to allow for better prediction of full-scale performance of GAC for removing MTBE. In addition, more information is needed on MTBE desorption from GAC systems and on the competitive effects of other SOCs (e.g., BTEX, TBA) in the impacted water. Dynamic column testing is currently being performed by UCLA and Calgon Carbon Corporation to investigate several of the issues listed above. Full-scale Performance To date, there are limited data regarding the successful use of full-scale GAC systems for removing MTBE from drinking water. As such, it is recommended that GAC performance for MTBE removal be evaluated under full-scale, field conditions. Collection of cost and operational data, including long-term NOM fouling effects and pretreatment requirements, will allow for meaningful comparison with results of dynamic column testing and cost analyses. 254

277 4.5 References American Petroleum Institute, A Compilation of Field-Collected Cost And Treatment Effectiveness Data For The Removal of Dissolved Gasoline Components From Groundwater, API Document 4525, November. American Petroleum Institute, Transport and Fate of Dissolved Methanol, Methyl- Tertiary-Butyl-Ether, and Monoaromatic Hydrocarbons in a Shallow, Sand Aquifer. API Publication American Society for Testing and Materials, Standard Practice For Determination of Adsorptive Capacity of Activated Carbon By A Micro-Isotherm Technique For Adsorbates At ppb Concentrations, ASTM Designation D Creek, D.N., and Davidson, J.M., The Performance and Cost of MTBE Remediation Technologies, Proceedings of the Petroleum Hydrocarbons and Organic Chemicals In Ground Water; Prevention, Detection, and Remediation Conference, November, 1998, pp Crittenden, J.C., Reddy, P.S., Aroro, H, Trynoski, J., Hand, D.W., Perram, D.L., and Summers, R.S., Predicting GAC Performance With Rapid Small-Scale Column Tests, Journal AWWA, January, pp Crittenden, J.C., Reddy, P.S., Hand, D.W., and Arora, H., Prediction of GAC Performance Using Rapid Small-Scale Column Tests, AWWA Research Foundation, September. Dobbs, R.A., and Cohen, J.H., Carbon Adsorption Isotherms For Toxic Organics, EPA Document EPA-600/ , April. Fritz, W. and Schundler, E.U., Competitive Adsorption of Two Dissolved Organics Onto Activated Carbon, Chemical Engineering Science, Vol. 36, No. 3. Graham, J., Technical director for activated carbon, U.S. Filter/Westates, Los Angeles, CA. Personal communication with D. Creek, Alpine Environmental, Inc., Fort Collins, CO. May Graham, J., Technical director for activated carbon, U.S. Filter/Westates, Los Angeles, CA. Personal communication with D. Creek, Alpine Environmental, Inc., Fort Collins, CO. October Hall, D.W. and Mumford, R.L., Interim Private Water Well Remediation Using Carbon Adsorption, Ground Water Monitoring Review, p

278 Hand, D., Senior Research Engineer, Environmental Engineering Center for Water and Waste Management, Michigan Technological University, Houghton, MI. Personal communication with D. Creek, Alpine Environmental, Inc., Fort Collins, CO. May Hand, D.W., Crittenden, J.C., Arora, H., Miller, J.M., and Lykins, B.W. Jr., Designing Fixed-Bed Adsorbers to Remove Mixtures of Organics, Journal AWWA, p Komex H2O Science, Literature Review of Technologies for Treatment of Methyl Tertiary Butyl Ether (MTBE) in Drinking Water. Prepared for City of Santa Monica, CA. April Komex H2O Science, Tom Browne, personal Communication with D.Creek, Alpine Environmental, Inc. Fort Collins, CO. May Kong, E.J. and DiGiano, F.A., Competitive Adsorption Among VOCs on Activated Carbon and Carbonaceous Resin, Journal AWWA, April, p Lehr, J.H., Granular-Activated Carbon (GAC): Everyone Knows of It, Few Understand It, Ground Water Monitoring Review, Vol. XI, No. 4, p Malley, J.P., Eliason, P.A., Wagler, J.L., Point-of-Entry Treatment of Petroleum Contaminated Water Supplies, Water Environment Research, Vol. 65, No. 2, p McKinnon, R.J. and Dyksen, J.E., Removing Organics From Groundwater Through Aeration Plus GAC, Journal AWWA, May, p McNamara, D., Carbon application specialist, Calgon Carbon Corporation, Pittsburgh, PA. Presentation to California MTBE Research Partnership. February, Mertz, K.A., Gobin, F., Hand, D.W., Hokanson, D.R., and Crittenden, J.C., Adsorption Design Software for Windows (AdDesignS), Michigan Technical University. Megonnell, N., Carbon specialist, Calgon Carbon Corporation, Pittsburgh, PA. Submittal to California MTBE Research Partnership, October, Munz, C., Walther, J-L, Baldauf, G., Boller, M., and Bland, R., Evaluating Layered Upflow Carbon Adsorption For The Removal of Trace Organic Contaminants, Journal AWWA, March, p Nyer, E., Groundwater Treatment Technology, Van Nostrand Reinhold Press, New York. Randtke, S.J. and Snoeyink, V.L., Evaluating GAC Adsorptive Capacity, Journal AWWA, August, p Semmens, M.J., Norgaard, G.E., Hohenstein, G., and Staples, A.B., Influence of ph on the Removal of Organics by Granular Activated Carbon, Journal AWWA, May, p

279 Snoeyink, V.L., Adsorption of Organic Compounds, Water Quality and Treatment. American Water Works Association. Editor: F.W. Pontius. McGraw-Hill, New York, Sontheimer, H., Crittenden, J.C., and Summers, S., Activated Carbon For Water Treatment, DVGW-Forschungsstelle & AWWA Research Foundation. Speth, T.F. and Miltner, R.J., Technical Note: Adsorption Capacity of GAC for Synthetic Organics, Journal AWWA, February, p Speth, T.F., Evaluating Capacities of GAC Preloaded with Natural Water, Journal of Environmental Engineering, January/February, Vol. 117, No. 1., p Stenzel, M.H. and Merz, W.J., Use of Carbon Adsorption Processes In Groundwater Treatment, Proceedings of American Institute of Chemical Engineers, 1988 Summer National Meeting, Denver, Colorado, August, Paper No. 6c. Suffet, I. H., Personal communication with California MTBE Research Partnership, June, Summers, R.S., Haist, B., Koehler, J., Ritz, J., Zimmer, G., and Sontheimer, H., The Influence of Background Organic Matter on GAC Adsorption, Journal AWWA, May, p Weber, W.J. Jr., Physicochemical Processes for Water Quality Control. Wiley Interscience, John Wiley and Sons Inc., New York, NY Zimmer, G., Crittenden, J.C., Sontheimer, H., and Hand, D., Design Considerations For Fixed-Bed Adsorbers That Remove Synthetic Organic Chemicals In The Presence of Natural Organic Matter, Proceedings of AWWA Conference, Orlando, Florida, June, p

280 258

281 5.0 Synthetic Resin Sorbents Amparo Flores Andrew Stocking, P.E. Michael Kavanaugh, Ph.D., P.E. 259

282 260

283 5.1 Introduction Background In the 1970s, synthetic resin sorbents became available as an alternative to GAC (see Chapter 4 for a detailed discussion of GAC). One of the major advantages of these resins over GAC is their on-site regenerability through steam stripping, solvent extraction, or microwave irradiation. The ability to regenerate resins on-site potentially offers economic advantages over GAC, which typically requires off-site high-temperature incineration for regeneration. In addition, because the manufacturing process of synthetic resins is a controlled process, resins can have specially designed functional groups and pore size ranges that can be manipulated to optimize their performance for specific applications. Currently, synthetic resin sorbents are used in a variety of industrial applications such as purification processes in the food and drug industry, odor control, and industrial wastewater treatment. They have also been used in a number of groundwater remediation applications (e.g., removal of halogenated organic compounds) and landfill leachate purification. However, the use of resin sorbents in municipal drinking water applications has been limited. In particular, there are currently no full-scale installations of resin systems for MTBE removal from drinking water. The higher unit costs of resins compared to the more traditional sorbent GAC has been mainly responsible for the limited applications of resin sorbents in drinking water treatment scenarios. The effectiveness of resin sorbent systems for MTBE removal is evaluated in this document because recent developments have suggested that synthetic resin sorbents may be economically competitive with other more established treatment technologies (air stripping, AOPs, and GAC) for this application. First, it is now recognized that groundwater contaminated with MTBE may also be contaminated with TBA a contaminant in some fuel grade MTBE and a by-product of microbially mediated MTBE degradation. Because TBA is more hydrophilic and has a lower Henry s constant than MTBE, TBA is even more challenging to remove by air stripping or GAC, and can be practically removed only through advanced oxidation processes and, potentially, biological treatment. Some preliminary findings suggested that synthetic resins may have sufficiently better sorption capacities for TBA relative to GAC to present a practical alternative treatment technology for sites that contain TBA. Secondly, improvements in resin regeneration processes such as steam regeneration, solvent regeneration, and microwave regeneration may make the life cycle cost of a resin system competitive or, perhaps, more economical than other options. Thirdly, resins (unlike AOPs) do not produce oxidation by-products. Finally, a greater demand for resin sorbents may lead to lower unit prices that would make them more cost-effective Objectives of the Evaluation In light of the developments noted above, this chapter will evaluate the applicability of resins for MTBE removal from water under typical drinking water treatment applications encountered 261

284 in the field. In addition, costs are also estimated for a flow rate (6 gpm) likely to be associated with site remediation applications. Using the available published literature, manufacturer/ vendor information, communication with consultants, researchers, resin manufacturers, and other knowledgeable parties, the following major areas of interest and concern regarding the use of resins are addressed: Synthetic resin sorption capacities for MTBE and TBA in water relative to GAC. Effects of background water quality parameters such as temperature, ph, NOM, and the presence of other synthetic organic compounds on the sorption capacities of resins. Various process flow configurations for a synthetic resin system. Regeneration alternatives for synthetic resin systems. Capital, operation, and maintenance costs of synthetic resin systems under various configurations and combinations of flow rates, influent concentrations, and effluent goals: a) Flow rates: 6 gpm, 60 gpm, 600 gpm, and 6,000 gpm. b) Influent concentrations: 20 µg/l, 200 µg/l, and 2,000 µg/l MTBE. c) Effluent goals: 0.5 µg/l (representing non-detect), 5 µg/l, and 20 µg/l MTBE. The literature review involved published literature and information provided by manufacturers and vendors, and by major researchers and consultants on the use of synthetic resins in site remediation and water treatment applications. The economic analysis was performed using the available literature and AdDesignS, a computer software designed by Mertz et al. (1994) to aid in the design of sorbent treatment systems. The information generated by AdDesignS was incorporated with the available information on regeneration alternatives to develop economical resin system design(s). Life-cycle cost analyses were performed based on 30 years of use and a seven percent discount rate. Field sites where resins have been used to remove MTBE from water were also identified. Using the limited data available from field studies, factors that may render the use of synthetic resins impractical or economically unattractive (e.g., potential resin foulants or regeneration limitations) were identified and taken into consideration in the design of a practical resin system. Upon completion of the literature review, economic analysis, and field site identification, research needs and data gaps were identified. 262

285 5.2 Process Principles Synthetic resins, like the widely used GAC, rely on sorption processes to remove organic compounds from water. Sorption processes fall into two broad categories: adsorption and absorption. Adsorption is the physical and/or chemical process in which a substance is accumulated specifically at an interface between two phases (typically a gas or solution phase in contact with a solid), while absorption involves the intermixing of a substance with the matrix of the second phase (typically an amorphous or gel-like phase rather than a true solid). The substance being removed from one phase is referred to as the sorbate while the sorbent is the phase onto or into which the accumulation occurs. In the context of water treatment, the process of interest generally involves the sorption of contaminants, such as certain ions or organic compounds from water by porous solid or semi-solid sorbent particles. The sorption of compounds by such sorbents occurs because of two primary driving forces: the hydrophobic (water-disliking) character of the sorbate and/or the high affinity of the sorbate for the sorbent. For the majority of systems encountered in water and wastewater treatment systems, sorption results from the net effect of the combined interactions of these two driving forces (Weber, 1972). The hydrophobicity (water-disliking character) of a compound is inversely proportional to its solubility in water. An extremely hydrophobic compound has a low aqueous solubility and, thus, may prefer to sorb onto a solid surface or into an amorphous matrix rather than remain surrounded by water molecules. On the other hand, hydrophilic (water-liking) compounds like MTBE tend to be stable in aqueous solutions and will leave the solution only if the sorbent provides an attractive force sufficient to overcome the strong bonds between the compound and water molecules. This attractive force, or affinity of the compound for the sorbent, can result from physical or chemical mechanisms. Physical mechanisms include dipole-dipole interactions and van der Waals interactions. Dipole moments in molecules are caused by a net separation of positive and negative charges resulting from the configuration of their atoms and electrons. When dipoles from two molecules are near each other, they tend to orient their charges to lower their combined free energy; the negative poles of one molecule tend to approach the positive pole of another, and vice versa. This intermolecular interaction results in a net attraction between the two molecules (Montgomery, 1985). When two neutral molecules that lack permanent dipoles approach each other, a weak polarization is induced in each because of quantum mechanical interactions between their charge distributions. The net effect is a weak attraction between the two molecules known as van der Waals force. As a general rule, van der Waals interactions increase with increasing size or surface area of the molecules involved (Schwarzenbach et al., 1993). Van der Waals interactions are generally weaker than dipole-dipole interactions. Chemical sorption, or chemisorption, is based on functional chemical group interactions. The principal difference between physical sorption and chemisorption is that the former is less specific with respect to which compounds sorb to which surface sites, has weaker forces 263

286 and lower energies of bonding, and operates over longer distances between sorbate molecules and sorbent sites (Montgomery, 1985). In chemisorption, the attraction between sorbent and sorbate approaches that of a covalent or electrostatic chemical bond between atoms, with shorter bond length and higher bond energy. Sorbates bound by chemisorption to a surface generally cannot accumulate to much more than one molecular layer (monolayer) because of the specificity of the bond between the sorbate and the surface. The location of the sorption sites tends to be very specific since only certain functional groups on a sorbate molecule are able to form these chemical bonds. In general, synthetic resins designed as sorbents for organic compounds have lower densities and varieties of chemical functional groups than activated sorbents such as GAC, and sorbate interactions with their surfaces are, thus, primarily through physical, rather than chemical, mechanisms. When compounds such as MTBE are removed from solution through direct sorption on surfaces within the very small pores of porous sorbents, their accumulation may be markedly enhanced through condensation within the pores. In the very narrow (less than 10-9 m) pore spaces of some synthetic resin sorbents and some GACs, intramolecular interactions (e.g., between MTBE molecules) can occur. These attractive intramolecular forces can result in the build-up, or condensation, of small pockets of pure solutions of these compounds in pore spaces. The following section describes two equilibrium models that are commonly used to characterize sorption onto various sorbents and that can be used to facilitate comparisons of the effectiveness of different resins among themselves and with GAC Equilibrium Sorption Models The sorption of chemical compounds from solution onto a surface or into a matrix may be viewed as an energetic process driven by thermodynamics. Various models have been developed to describe the thermodynamic equilibrium that exists in a system containing a sorbent, a sorbate, and a solvent. These models take the form of sorption isotherm equations that relate the concentration of the sorbate on or in the sorbent phase to the bulk concentration of the sorbate in solution (solvent) at a given temperature. For a description of isotherm testing procedure, see Chapter 4. The Freundlich and Dubinin-Astakov models are two that are commonly used to describe sorption of organic compounds from gases and aqueous phases by synthetic resins. While the latter model generally provides a more precise representation of the sorption phenomena involved in such cases, the Freundlich model is more widely used to compare the performance of resins with GAC because of its mathematical simplicity and its broader range of applicability. GAC sorption is represented well by the Freundlich model; therefore, much of the available information for GAC isotherms is provided in the form of Freundlich parameters. 264

287 The Freundlich Model The Freundlich model is particularly suitable for describing processes involving heterogeneous sorbents having broad ranges of sorption site energies. This model has the form: q = K F C n where q is the concentration of the sorbate on the sorbent [mg/g] and C is the sorbate concentration in the bulk solution [mg/l]. The K F value is related to the sorption capacity while the exponent n is related to the energy of sorption and to site heterogeneity. For homogeneous sorbents of uniform site energy, the value of n approaches unity. The logtransformed version of this equation (log q = log K F + n log C) is useful for regression analysis of experimental data. K F and n can be calculated from the y-intercept and slope, respectively, of a plot of log q vs. log C. Studies designed to compare the effectiveness of resins and GAC typically use log K F and n values as bases for comparison. The Dubinin-Astakov Model Davis and Powers (1999) reference the work of several researchers who have shown that sorption isotherms for carbonaceous resins generally are not continuously linear over residual aqueous concentration ranges spanning several decades when plotted on a log-log scale (Weber and van Vliet, 1981; Kong and DiGiano, 1986; Hand et al., 1994; Parker, 1995; Gallup et al., 1996). Sorbent concentrations (q) approach a maximum value at high aqueous concentrations (C). As suggested by Weber and van Vliet (1981), the curvilinear nature of these isotherms indicates a micropore (pore diameters less than 2x10-9 m) filling process described by Dubinin, a process more similar to the capillary condensation phenomenon discussed earlier than to a build-up of sorbate film layers on the sorbent surfaces (Gregg and Sing, 1992). The model developed by Dubinin and Astakov is described by the following isotherm equation: q = q m exp[-(a/e) b ] where A is the sorption potential (J/mol) defined as (Parker, 1995): A = RT ln (C s /C) and q = sorption capacity or concentration of the sorbate on the sorbent (mg/g) q m = capacity at monolayer coverage (mg/g) E = characteristic energy (J/mol) R = ideal gas constant (8.314 J/mole K) T = temperature ( K) C s = aqueous solubility of the adsorbate (mg/l) C = concentration of the adsorbate in the bulk solution (mg/l) 265

288 The parameters q m, E, and b are typically used as fitting parameters even though they represent physical characteristics of the sorbate and the sorbent (Davis and Powers, 1999). A comparison of Freundlich model and Dubinin-Astakov model fits to experimental data for a resin (Ambersorb 572) is presented in Figure 5-1 (Davis and Powers, 1999). Figure 5-1. Comparison of Freundlich model and Dubinin-Astakov (DA) model fits to experimental data for a carbonaceous resin (Ambersorb 572). The r2 values were and 0.994, respectively. (Davis and Powers, 1999) Sorption Rates Equilibrium isotherms are useful for estimating ideal or theoretical sorption performance, but under dynamic conditions, the efficiency of the process will be controlled by the rates at which equilibrium conditions are approached. In GAC and resin systems, the efficiency of contaminant removal from water is usually determined by the method of contact between the adsorbate and sorbent. The process of sorption can be broken down into a series of steps (Table 5-1) which are described by individual rate relationships (Montgomery, 1985). In water treatment applications, the steps of bulk transport and chemical or physical bonding (steps 1 and 5) are generally rapid, and the overall rate of sorption is controlled by film and/or pore diffusion (steps 2 and 3). Characterization of mass transfer coefficients and surface diffusivities useful for adsorber system design can be obtained by performing bench-scale or pilot-scale kinetic studies (Liu and Weber, 1981; Crittenden et al., 1991). For a description of dynamic column tests, see Chapter

289 Table 5-1 Steps in the Process of Sorption (Montgomery, 1985) STEP CONTROLLING FACTORS 1 Transport of sorbate from the bulk solution to the rates of advection and turbulent mixing boundary layer or surface film surrounding the sorbent particle. 2 Transport of sorbate across the film boundary rate of molecular diffusion layer to the exterior surface of the sorbent particle. 3 Diffusion of sorbate within the pores, from the rate of molecular diffusion exterior to the interior surfaces of the sorbent particle. 4 Transport of the sorbate along the surfaces of pore rate of surface diffusion walls. 5 Physical or chemical interactions of the sorbate rate of chemical interactions at surfaces and at internal surfaces and in micropores of the sorbent. in micropores 267

290 268

291 5.3 Application of Synthetic Resins to Water Treatment Resin Production The largest manufacturers of synthetic resin sorbents used for the removal of organic compounds and the products market applications are listed in Table 5-2. Not surprisingly, these companies are also some of the biggest manufacturers of ion-exchange resins. Sorbent resins and ion-exchange resins are manufactured through very similar processes and are often created from the same base material or polymer backbone. The main difference between the two products is that ion-exchange resins contain charged functional groups, which can form chemical bonds with ions in solution while sorbent resins rely on physical or non-ionic interactions to remove contaminants from water. Table 5-2 Largest Manufacturers of Polymeric and Carbonaceous Resins Used for the Removal of Organic Compounds MANUFACTURER PRODUCT LINE APPLICATIONS Rohm and Haas Company Amberlite, Ambersorb, Duolite, XAD groundwater remediation, spill cleanup, wastewater treatment, vapor phase treatment of toxic air emissions, odor control, decolorization of food products, purification of food and pharmaceutical products, removal of chlorinated solvents, phenols, and aromatic compounds from waste streams Dow Chemical Company DOWEX Optipore removal of phenol and other organic compounds from water; processing of corn syrups Bayer Lewatit water and wastewater treatment Purolite Hypersol-Macronet color, taste, and odor removal from corn and cane syrups, pesticide and herbicide removal from potable water sources, removal of aromatic compounds (phenol and chlorinated phenols), dyes, mineral oil, and detergents from wastewaters, caffeine isolation, desugarization of molasses, decolorization and debittering of juices Based on Manufacturer Literature as of In addition to their chemical composition, resins are differentiated on the basis of their pore size distributions. As a matter of convention, micropores are defined as pores less than 20 angstroms (Å) (2x10-9 m) in diameter, mesopores are between 20 to 500 Å (2x10-9 to 5x10-8 m) in diameter, and macropores have diameters greater than 500 Å. Synthetic resins generally have a more controlled and even distribution of pore sizes than GAC. In order to be useful for sorptive applications in water treatment, resins have an extensive network of micropores, similar to GAC, which creates high surface areas and abundant sorption sites. To produce superior kinetics over GAC, resins are also designed with a significant percentage of pores in the mesopore and macropore size range, which can provide access to the inner surfaces of resins. A gram of a synthetic resin can have more than 1,000 m 2 of surface area. Synthetic sorbents can be classified into two categories: polymeric resins and 269

292 carbonaceous resins. In general, polymeric resins are non-ionic versions of ion-exchange resins, and carbonaceous resins are based on a particular type of ion-exchange resin that has undergone partial pyrolysis, or a slow heating of the base material in the absence of air. Although polymeric resins are widely applied for the removal of organic compounds from aqueous solutions (e.g., removal of aromatic compounds from industrial wastewater), current evidence suggests that carbonaceous resins are more effective than polymeric resins in the removal of MTBE. The results of several bench-scale studies comparing the MTBE sorption capacities of polymeric vs. carbonaceous resins are discussed in Section Polymeric Resins Polymeric resins are typically based on cross-linked polymers having polystyrene, phenolformaldehyde, or acrylate matrices (Figure 5-2) (Faust and Aly, 1998). Most commercial macroporous polymeric sorbents are based on polystyrene-divinylbenzene copolymers (Neely, 1982) in which the divinylbenzene serves as a cross-linking agent that makes the styrene insoluble and confers physical strength to the resin (DeSilva, 1995). Although they can be based on the same matrices, polymeric resins differ from traditional ion-exchange resins in their lack of ionic functional groups. Figure 5-2. Various matrices used for polymeric resins (Faust and Aly, 1998). To produce macroporous polymeric resins, the polymerization process is carried out in the presence of an inert material. Small amounts of an inert material yield a non-macroporous, three-dimensional network while a high inert material content leads to the formation of microstructures, or nuclei. As the polymerization progresses, the nuclei agglomerate and crosslink to form microspheres. Aggregates of microspheres form irregularly shaped particles, which constitute the resin beads (Malley et al., 1993). The pore size distribution of polymeric resins can be controlled during their manufacture by varying the amount of extender used in the polymerization reaction; this governs the degree of cross-linking and the ultimate pore structure created (Weber and van Vliet, 1981). 270

293 Carbonaceous Resins Synthetic carbonaceous resins are formed by the partial pyrolysis of macroporous polymer beads (Weber and van Vliet, 1981). Ambersorb, a line of carbonaceous resins developed and patented by Rohm and Haas (Neely, 1997 and Maroldo et al., 1989), is created from the partial pyrolysis of macroporous sulfonated styrene-divinylbenzene ion-exchange resin (Parker, 1992). This process produces resins that have lost their ionic functional groups and have relatively non-polar surfaces (Davis and Powers, 1999). Different levels of pyrolysis will yield slightly different pore size distributions and surface areas for the same synthetic polymer base. While the macropores and mesopores are maintained during the process of pyrolysis, more micropores are generated. The presence of micropores is critical to the ability of carbonaceous resins to function well in water treatment applications. The micropores are only accessible to smaller molecules and, thus, reduce fouling of and competition for sorption sites associated with larger molecules typical of NOM Physical and Chemical Properties of Resins A summary of the physical and chemical properties associated with some commercially available synthetic resins and GAC is presented in Tables 5-3a and 5-3b. The resins included in the table are those that have been evaluated or recommended for MTBE removal by various researchers and/or manufacturers and for which data were available. All of the resins come in the form of spherical beads with diameters of approximately 1 mm or less. For comparison, data is also presented for two GAC products that have been evaluated for MTBE removal by various researchers and/or manufacturers and for which data was provided. Filtrasorb 400 (Calgon Carbon; Pittsburgh, PA) is a coal-based GAC commonly used for drinking water applications. CC-602 a a coconut shell-based GAC manufactured by U.S. Filter/Westates was found to have the highest sorption capacity among the GAC products evaluated in Chapter 4. In general, coconut-based GAC has been found to perform better than coal-based GAC for MTBE applications (see Chapter 4). The flexible polymeric structure of synthetic sorbents, compared to the rigid structure of GAC, allows resins to be regenerable through steam stripping or microwave irradiation. When resin beads are heated using steam, for example, the polymeric matrix relaxes or loosens and the pores widen. As the pores widen, they become accessible to steam, which can then re-solubilize formerly adsorbed organic compounds. In addition, heating the resin and widening the resin pores can volatilize pockets of condensed organic compounds in micropores (see Section 5.2). 271

294 Table 5-3a. Physical and Chemical Properties of Polymeric Sorbents PROPERTIES Rohm & Haas Amberlite XAD-4 Rohm & Haas Amberlite XAD-7 Dow L-285 (XUS ) Dow L-323 (XUS ) Dow Optipore L- 493 Dow XUS Bayer Lewatit VP OC 1066 PHYSICAL porosity ~55% % (relative) total pore volume (cm 3 /g) minimum macropores (> 500 A) mesopores ( A) micropores (< 20 A) average pore diameter nm 7 bulk density (g/ml) particle density (g/ml) appearance white translucent RESINS Polymeric Sorbents beads 1 beads 1 spherical beads 9 spherical beads 9 spherical beads 9 spherical beads 9 beads 7 white translucent light tan to amber colorless to off-white orange to brown orange to brown white, opaque porous particle size (US sieve series) 20 to to to to to particle diameter range (mm) mm BET* surface area (m 2 /g) ~ CHEMICAL structure/matrix polystyrene crosslinked with divinylbenzene 5 cross-linked polyacrylate matrix 5 polystyrene crosslinked with divinylbenzene 9 divinylbenzene, ethylvinylbenzene copolymer 9 polystyrene crosslinked with divinylbenzene 9 polystyrene crosslinked with divinylbenzene 9 polystyrene crosslinked with divinylbenzene 7 H2O adsorption at 80% relative humidity (mg/g) to 40 (84% --- R.H.) 9 functional groups none none dimethylamine and chloromethyl groups 9 none none 1 Rohm and Haas ( ), 2 Faust and Aly (1998), 3 Malley et al. (1993), 4 Davis and Powers (1999), 5 Browne and Cohen (1990), 6 Isacoff et al. (1992), 7 Bayer (1999), 8 Calgon Carbon (1999), 9 Dow Company (1999), and 10 US Filter/Westates (1999) 272

295 Table 5-3b. Physical and Chemical Properties of GAC and Carbonaceous Resins PROPERTIES GRANULAR ACTIVATED CARBON Calgon Filtrasorb 400 (coal-based) US Filter/Westates CC-602 (coconut-based) Rohm & Haas Ambersorb 563 RESINS Carbonaceous Resins Rohm & Haas Ambersorb 572 PHYSICAL porosity % (void fraction) total pore volume (cm 3 /g) macropores (> 500 A) mesopores ( A) micropores (< 20 A) average pore diameter bulk density (g/ml) particle density (g/ml) appearance black, granular black, granular hard black spherical black spherical beads 1 beads 1 particle size (US sieve series) 12 to 50 12x to to 50 1 particle diameter range (mm) BET* surface area (m 2 /g) CHEMICAL structure/matrix stacked layers of fused hexagonal rings of carbon atoms 2 stacked layers of fused hexagonal rings of carbon atoms 2 sulfonated polystyrene crosslinked with divinylbenzene 1 sulfonated polystyrene crosslinked with divinylbenzene 1 H2O adsorption at 80% relative humidity (mg/g) functional groups acidic surface oxides (e.g., carboxylic groups, phenolic hydroxyl groups, quinone-type carbonyl groups) and basic surface oxides (e.g., chromene) 2 acidic surface oxides (e.g., carboxylic groups, phenolic hydroxyl groups, quinone-type carbonyl groups) and basic surface oxides (e.g., chromene) 2 none none 1 Rohm and Haas ( ), 2 Faust and Aly (1998), 3 Malley et al. (1993), 4 Davis and Powers (1999), 5 Browne and Cohen (1990), 6 Isacoff et al. (1992), 7 Bayer (1999), 8 Calgon Carbon (1999), 9 Dow Company (1999), and 10 US Filter/Westates (1999) 273

296 5.3.3 Results of Bench-scale Studies Sorption Capacity/Static Isotherm Studies Malley et al. (1993) Study Amidst concerns over gasoline contamination of drinking water supply wells by leaking USTs, the New Hampshire Department of Environmental Services (NHDES) performed an evaluation of the various POE treatment technologies available. The study concluded that a combination of air stripping and GAC was cost-effective and reliable for the removal of all of the volatile components of gasoline except for MTBE. Thus, it was necessary to find an alternative treatment technology for the removal of MTBE from contaminated groundwater. Malley et al. (1993) was asked to evaluate several alternative options, one of which was the use of synthetic resins. In their study, Malley et al. evaluated seven synthetic polymeric and carbonaceous resins: the polymeric resins Amberlite XAD-4 and XAD-7 (Rohm and Haas; Philadelphia, PA) and XUS and XUS (Dow Chemical Company; Midland, MI), and the carbonaceous resins Ambersorb 563, 572, and 575 (Rohm and Haas; Philadelphia, PA). Filtrasorb 400, a coal-based carbon, was being used by the NHDES in POE units at the time of testing; therefore, it was used as a benchmark for comparison. Preliminary kinetic studies were conducted to determine how quickly the same mass of each sorbent reached equilibrium with a solution containing an initial MTBE concentration of 1,026 µg/l. Equilibrium was said to have been reached when the MTBE concentration in the solution was no longer changing. Testing revealed that the polymeric resins generally had lower ultimate sorption capacities compared to the GAC at equilibrium. Conversely, the carbonaceous resins were found to remove more MTBE than the GAC and at a faster rate. All three carbonaceous resins reached equilibrium after 5 days as opposed to 10 days for the GAC. Based on the results of the preliminary kinetic studies, Malley et al. subsequently performed batch sorption isotherm studies in simulated groundwater for the three carbonaceous resins and Filtrasorb 400. Overall, Ambersorb 563 and 572 were found to have the highest MTBE capacities followed by Ambersorb 575 and Filtrasorb 400, which the researchers found to have the lowest capacity for MTBE. At an aqueous MTBE concentration of 500 µg/l, the sorption capacities of Ambersorb 563 and 572 were approximately 2.5 times greater than that of Filtrasorb 400 (10.8 and 11.4 mg/g vs. 4.2 mg/g). Column studies performed on the Ambersorb 563 indicated that it can remove up to 1.7 times as much MTBE as Filtrasorb 400 per gram of wet sorbent at an MTBE concentration of approximately 1,200 µg/l. Based on the results of their study, Malley et al. concluded that Ambersorb sorbents warrant further investigation for use in MTBE removal. In particular, they recommended the investigation of regeneration methods that could make resins more cost-effective than GAC. 274

297 Davis and Powers (1999) Study Davis and Powers performed a preliminary screening of various sorbents (GAC, carbonaceous resins, C18 bonded silica, acrylic resin, and porous graphitic carbon) based on their MTBE sorption capabilities. Their isotherm studies indicated that the porous graphitic carbon (Hypercarb - Life Sciences International; London, England) and the carbonaceous resins (Ambersorb 563 and 572) had greater capacities for MTBE than the GAC (Filtrasorb 400). However, the required addition of a pre-wetting agent (e.g. methanol) to suspend and activate the Hypercarb made this sorbent unsuitable for treatment applications. The Ambersorb 563 and 572 resins were found to have sorption capacities four to five times greater than the Filtrasorb 400 at an aqueous MTBE concentration of 1,000 µg/l (16.2 and 13.8 mg/g, respectively, vs. 3.1 mg/g). The observed superior performance of Ambersorb 563 and 572 over Filtrasorb 400 supports the findings of Malley et al. (1993). Suffet et al. (1999) Study A total of 11 sorbents were evaluated by Suffet et al. based on their MTBE sorption capabilities, including two polymeric sorbents (Amberlite XAD-4 and XAD-8), one carbonaceous resin (Ambersorb 572), and several GACs. The results showed that within the experiment range of approximately 5 to 600 µg/l MTBE, the coconut-based GAC (GRC-22) (Calgon Carbon; Pittsburgh, PA) and the Ambersorb 572 had the highest sorption capacity ranges at 1.5 to 40 mg/g and 0.5 to 15 mg/g, respectively. The data for these two sorbents also fit the Freundlich model for equilibrium isotherms. On the contrary, the Amberlite XAD-4 and XAD-8 were found to have inferior sorption capacities compared to the other GAC products examined and were poorly represented by the Freundlich model. Industry/Manufacturer/Vendor Studies American Purification, Inc. An isotherm study performed by AmeriPure, Inc., a division of American Purification, Inc. (Newport Beach, CA) that designs regenerative liquid and vapor phase sorption systems, showed that Ambersorb 563 had greater capacity than Dowex Optipore L-493 at aqueous MTBE concentrations less than approximately 10 mg/l. Beyond 10 mg/l, Dowex Optipore L-493 surpassed the capacity of Ambersorb 563. The capacity range for Ambersorb 563 was found to be 0.7 to 21 mg/g over an aqueous MTBE concentration range of to 72 mg/l. Dow L-493 was found to have a capacity range of 0.25 to 45 mg/g over an aqueous MTBE concentration range of 0.04 to 94 mg/l. Equilon Enterprises, L.L.C. MTBE isotherms were generated by a research group at Equilon Enterprises, L.L.C. for Ambersorb 563 and Dow Optipore L-493 resins (Sun, 1999). The sorption capacity range of 275

298 Ambersorb 563 (0.5 to 18 mg/g) was found to be approximately an order of magnitude greater than that of Dowex Optipore L-493 (0.03 to 1 mg/g) over an aqueous concentration range of to 1 mg/l. The slopes of the isotherms are similar for the two resins and suggest a consistently superior performance for Ambersorb 563 over Dow L-493, contradicting the results observed by AmeriPure, Inc. at the higher concentrations discussed above. Ion-exchange resin manufacturers Several ion-exchange and sorbent resin manufacturers were contacted to determine if they currently have, or are in the process of developing, any products that would be suitable for MTBE sorption. Purolite (Philadelphia, PA) is currently testing their Hypersol-Macronet sorbents for MTBE application (Boodoo, 1999). Hypersol-Macronet is a diverse line of products that includes both ionic and non-ionic resins. To date, Purolite has provided no isotherms. Bayer Company (Pittsburgh, PA) is also researching potential applications of their products for MTBE removal (Fatula, 1999). No isotherms are currently available but Lewatit VP OC 1066, a nonionic resin, was suggested for consideration by Bayer Company. Absorbent Products (MicroClean Services Co. [Castro Valley, CA] and Guardian Environmental Technologies [GET] [Kent, CT]) MicroClean Services Co. (Castro Valley, CA) and GET (Kent, CT) are two relatively new companies that are marketing disposable or nonregenerable absorbent products. These two products are composites of GAC and a polymeric resin. Unlike polymeric and carbonaceous resin sorbents, which rely on surface interactions, these non-porous products absorb contaminants into their internal polymeric matrix. Thus, the absorption of organic compounds causes these products to swell. This mechanism allows for potentially high contaminant loadings, which manufacturers claim can exceed 100 percent on a mass basis. According to MicroClean Services Co., their product PetroLOK, a polymer-enhanced activated carbon material, has been demonstrated in the company s laboratory and limited field studies to have MTBE equilibrium absorption capacities between 500 to 1,000 mg per gram of the media at concentrations approximately 10 mg/l and above (MicroClean Services Co., 1999). No isotherms were provided, but MicroClean indicated that they are willing to offer a guaranteed price for their product based on this range of absorption capacity. The MTBE absorption capacity of GET s PolyGuard was tested by an independent laboratory in a closed-loop, continuous flow system and was found to be 1,530 mg per gram of the absorbent (Baron Consulting Co., 1995). However, this capacity has not been observed under field conditions and most likely indicates an ultimate capacity under highly optimal conditions (i.e., a single-component [MTBE only] solution and a long contact time of

299 hours). A pilot study of PolyGuard (Section 5.5.4) observed an MTBE absorption capacity of 150 mg/g under an average influent concentration of 263 mg/l. For comparison, Ambersorb 563, which appears to be the industry s best resin candidate for MTBE removal, has equilibrium sorption capacities ranging from approximately 1 to 200 mg MTBE/g sorbent over an aqueous concentration range of 0.01 to 1,000 mg/l. Independent quantitative testing of these absorbent products in the laboratory and in the field needs to be performed to more accurately characterize their absorption capacities relative to polymeric and carbonaceous resins and GAC. Summary of Existing MTBE Isotherm Data A summary of the existing MTBE isotherm data for synthetic resins and two GAC products is shown in Figure 5-3. Linear regression data based on the log-transformed Freundlich equation were compiled from various sources and re-formatted, when necessary, so that the data can be compared. It is important to note that the data comes from various sources and does not necessarily reflect the same experimental conditions. The experimental conditions, and the resulting Freundlich model parameters, associated with the various sources of MTBE isotherms are presented in Table 5-4. In looking at the isotherms presented in Figure 5-3, the experimental conditions, specifically background water quality parameters, need to be taken into consideration since they can impact resin and GAC sorption capacities. MTBE isotherms were included for three synthetic resins: Ambersorb 563, Ambersorb 572, and Dowex Optipore L-493. Amberlite XAD-4 and XAD-8 isotherms were not included for reasons previously cited. Isotherm data for two GAC products were also included for comparison to synthetic resins. As noted previously, CC-602 is a coconut shell-based GAC while Filtrasorb 400 is a coal-based GAC commonly used for drinking water applications. Figures 5-4a and 5-4b present the observed sorption capacities of these various resins and GACs at aqueous concentrations of 100 µg/l and 1,000 µg/l MTBE, respectively. The results of two independent comparative studies (Malley et al., 1993 and Davis and Powers, 1999) suggest that Ambersorb 563 is a promising alternative to GAC. However, these studies compared Ambersorb 563 with Filtrasorb 400, a coal-based GAC. Since industry data suggest that coconut-based products generally offer superior sorption capacities for MTBE over coal-based products, tests comparing resins with the best coconut-based GAC available should be performed to fully quantify the advantage of resins over GAC. In addition, Calgon Carbon (Pittsburgh, PA) recently introduced Filtrasorb 600, a new coalbased GAC that was specifically designed for the removal of trace levels of MTBE and other organic compounds from water. Although data from manufacturer-independent sources are currently not available, isotherms generated by Calgon Carbon for a deionized water solution and a field sample are promising (Appendix 5A). It is recommended that Filtrasorb 600 be considered in future studies as well. 277

300 1000 MTBE Isotherms Ambersorb 563 (Malley et al., 1993) Ambersorb 563 (Sun, 1999) Ambersorb 563 (Davis and Powers, 1999) Ambersorb 563 (AmeriPure, 1999) Ambersorb 572 (Malley et al., 1993) Ambersorb 572 (Davis and Powers, 1999) Ambersorb 572 (Suffett et al., 1999) Dowex Optipure L-493 (Sun, 1999) Dowex Optipure L-493 (AmeriPure, Inc., 1999) CC602 (US Filter/Westates, 1999) Filtrasorb 400 (Davis & Powers, 1999) Filtrasorb 400 (Malley et al., al, 1993) Aqueous Phase Concentration (mg/l) Sorbed Phase Concentration (mg sorbate/g sorbent) Sorbed Phase Concentration (mg sobate / g sorbent) Figure 5-3. MTBE isotherms. The endpoints do not mark actual data points and are used for identification purposes only. The heavy lines represent Ambersorb resins (continuous for 563 and dashed for 572). Dowex L-493 is represented by long dashed lines and GAC products are represented by short dashed lines. 278

301 Table 5-4. MTBE Isotherm Studies and Their Associated Experimental Conditions and Freundlich Parameters EXPERIMENTAL CONDITIONS RESIN/GAC T ( o C) ph Water Quality Aqueous MTBE Conc. s (mg/l) n log KF # of data points R 2 SOURCE POLYMERIC RESINS Dowex 25 ~7 organic-free water; buffered 0.04 to AmeriPure, Inc. (1999) Optipore L unknown synthetic groundwater to Sun (1999) CARBONACEOUS RESINS Ambersorb simulated groundwater: Milli-Q water with 5x10-4 M NaHCO3; and 0.1 N HCl or 0.1 N NaOH; and 1 N NaCl 0.07 to Malley et al. (1993) simulated groundwater: distilled, deionized water buffered with 8x10-5 M NaHCO 3 and 7.8x10-3 M NaCl 0.6 to Davis & Powers (1999) 25 unknown synthetic groundwater to Sun (1999) 25 ~7 organic-free water; buffered to AmeriPure, Inc. (1999) Ambersorb simulated groundwater: Milli-Q water with 5x10-4 M NaHCO3; and 0.1 N HCl or 0.1 N NaOH; and 1 N NaCl 0.07 to Malley et al. (1993) simulated groundwater: distilled, deionized water buffered with 8x10-5 M NaHCO 3 and 7.8x10-3 M NaCl 1 to Davis & Powers (1999) organic-free water: glass distillation of Milli-Q water then passed through Filtrasorb 200 GAC to Suffet et al. (1999) GAC Filtrasorb simulated groundwater: Milli-Q water with 5x10-4 M NaHCO3; and 0.1 N HCl or 0.1 N NaOH; and 1 N NaCl 0.06 to Malley et al. (1993) simulated groundwater: distilled, deionized water buffered with 8x10-5 M NaHCO 3 and 7.8x10-3 M NaCl 0.4 to Davis & Powers (1999) CC ~7 organic-free water; not buffered to US Filter/Westates (1999) 279

302 6 Sorbent Sorbent Capacities at at Aqueous Concentration = 100 = 100 µg/l µg/l US (a) 20 Sorbent Capacities at Aqueous Concentration = 1,000 = 1,000 µg/l µg/l Ambersorb 563 Ambersorb 563 (a) Ambersorb 563 (e) U.S. US Filter/Westates CC602 (f) Ambersorb 572 (e) Ambersorb 563 (d) Dowex Optipore L-493 (d) U.S. Filter/Westates CC602 (f) (f) Ambersorb 563 (a) Ambersorb 572 (b) Ambersorb 572 (c) Ambersorb Ambersorb (b) Ambersorb 563 Ambersorb 563 (d) (s) Amberlite XAD- Amberlite XAD- (c) 4 (c) Dowex Dowex Optipore Optipore L-493 (d) L-493 (d) mg mg MTBE/g MTBE/g sorbent mg mg MTBE/g sorbent (b) Figures 5-4a and 5-4b. Sorption capacities of various synthetic resins and GAC at aqueous concentrations of 100 and 1,000 µg/l MTBE. Note: a = Sun (1999); b = Malley et al. (1993); c = Suffet et al. (1999); d = AmeriPure, Inc. (1999); e = Davis and Powers (1999); f = U.S. Filter/ Westates (1999). 280

303 An upcoming pilot test in Santa Monica, California may provide more solid data on the sorption performance of resins relative to GAC under field conditions. For information regarding this pilot test, see Section Effects of ph Malley et al. (1993) performed studies on Ambersorb 563, 572, and 575 to evaluate the effects of ph on the removal of MTBE. The ph 6.5, 7, and 8.5 isotherms for all three sorbents did not appear to be different from each other, indicating that the ph of the simulated groundwater did not affect the capacity of the sorbents. This finding is consistent with the fact that resins do not have functional groups and generally do not rely on chemical interactions that could be affected by ph. Effects of Temperature Malley et al. (1993) performed studies on Ambersorb 575 and Filtrasorb 400 to evaluate temperature effects on their MTBE sorption capacities. The results for the GAC confirmed previous observations that showed an inverse correlation between capacity and temperature, implying an exothermic sorptive process (Weber and Moris, 1964). On the contrary, the sorption capacities of Ambersorb 575 appeared to be the same at 10 C and at 25 C (Malley et al., 1993), suggesting that the sorption capacity of Ambersorb 575 is unaffected by temperature over this range. There is currently very limited data available on the effects of temperature on resin MTBE sorption capacities. Controlled tests need to be performed on individual resins to fully quantify any temperature effects. Effects of Oxidants The effects of various chemical oxidants such as sodium hypochlorite (NaOCl), O 3, and H 2 O 2 on carbonaceous resins have been investigated (Isacoff, 1999; Suri et al., 1999; and Crittenden et al., 1997). Isacoff of Rohm and Haas (1999) performed a study where Ambersorb 563 was exposed to 1,000 mg/l of a hypochlorite solution for 42 days. After exposure, there were no changes observed in the resin s sorptive capacity for chloroform. In addition, the resin s degree of hydrophobicity was unaffected, as indicated by water sorption isotherm studies performed before and after exposure to hypochlorite solution. Similar results were observed by Isacoff (1999) for the exposure of Ambersorb 563 to a 32 percent H 2 O 2 solution. 281

304 This finding is consistent with unpublished results by Suri (1999), which indicate that Ambersorb 563 resins are unaffected by exposure to H 2 O 2 and O 3. There have also been studies (Crittenden et al., 1997 and Suri et al., 1999) in which Ambersorb 563 resin beads were impregnated with the metal oxide catalysts TiO 2 and Pt-TiO 2 and subjected to multiple cycles of sorption and steam regeneration. These tests seem to indicate that Ambersorb 563 is unaffected by any hydroxyl radicals ( OH) generated through the reaction of these catalysts with water under exposure to sunlight or UV radiation. Interference of NOM To evaluate the effects of the presence of humic substances on the MTBE sorption capacities of synthetic resins vs. GAC, Suffet et al. (1999) performed isotherm studies on Ambersorb 572 and GRC-22, a coconut-based carbon, using groundwater from Santa Monica, California wells and compared their results with isotherms derived from organic-free water. On average, the groundwater contained 0.5 mg/l of total organic carbon (TOC) and had a ph of 6.5. There was a marked decrease in the sorption capacity of the GRC-22 under exposure to humic substances while the Ambersorb 572 appeared to be unaffected (Figures 5-5a and 5-5b). These findings support results from a previous study by Hand et al. (1994), which showed that Ambersorb 563 is resistant to fouling by NOM. Hand et al. performed TCE isotherm studies on Ambersorb 563 and Filtrasorb 400 that had been pre-exposed to groundwater containing NOM (1-3 mg/l) for 2.5, 10, and 24-week intervals. After 10 weeks of NOM exposure, there was no observed decrease in the capacity of the Ambersorb 563 for TCE compared with an average decrease of 35 percent for the GAC. Even after 24 weeks of NOM exposure, the capacity of the Ambersorb 563 resin only decreased by an average of 12 percent. Biofouling Studies performed by Rohm and Haas indicate that Ambersorb products are not prone to biofouling (Isacoff, 1999). According to Isacoff, the narrow pores in carbonaceous resins make them inaccessible to microorganisms that could cause internal fouling of the resins (1999). In addition, as noted earlier, the internal pores in carbonaceous resins are inaccessible to NOM that are necessary for sustained microbial growth. However, as with GAC, resins may be subject to biofouling on their external surfaces under conditions of high NOM content and long EBCTs (see Section 5.4.5). In these cases, disinfection of the influent water prior to contact with the resins may be necessary. 282

305 Log Aqueous Phase MTBE Concentration (µg/l) (a) Coconut-based GAC (GRC-22) Log Sorbed Phase Concentration (µg MTBE/g sorbent) Log Sorbed Phase Concentration (µg MTBE/g sorbent) Log Aqueous Phase MTBE Concentration (µg/l) (b) Ambersorb 572 Figures 5-5a and 5-5b. Effects of background humic substances in Santa Monica water on the sorption capacities of coconut-based GAC (GRC-22) and Ambersorb 572 (Suffet et al., 1999). 283

306 Competitive Sorption BTEX Davis and Powers (1999) performed bi-solute isotherm studies on Ambersorb 563 and Ambersorb 572 to determine their effectiveness in the presence of other gasoline hydrocarbons. m-xylene was used as a representative BTEX compound and was added at a concentration of 43.2 mg/l in different samples containing a constant mass of the sorbents and varying MTBE concentrations (5 to 2,500 mg/l). m-xylene was removed to below detection limits (0.1 mg/l) in all the competitive sorption experiments. MTBE, however, was detected for the entire concentration range studied. These results imply that the resins preferentially sorb m-xylene over MTBE at an initial m- xylene concentration of 43.2 mg/l. Davis and Powers point out that given the hydrophobic nature of these resins, the preferential sorption of m-xylene, which is more hydrophobic than MTBE, is to be expected. A comparison of the chemical properties of MTBE and m-xylene is presented in Table 5-5. Table 5-5 Chemical Properties of MTBE, m-xylene, and TBA (NSTC, 1997 and Malley et al., 1993) Figure 5-6 illustrates the observed decrease in MTBE sorption capacities of Filtrasorb 400, Ambersorb 563, and Ambersorb 572 in the presence of 43.2 mg/l m-xylene and an aqueous MTBE concentration of 1 mg/l. Further studies should be conducted to investigate the potential for competitive sorption under lower BTEX concentrations. 284

307 MTBE Sorbed Phase Concentration (mg/g) at at C C = = 1 1 mg/l mg/l no m-xylene m-xylene Co = 43.2 mg/l Filtrasorb 400 Ambersorb 572 Ambersorb 563 Figure 5-6. Effects of m-xylene (Co = 43.2 mg/l) on the MTBE sorption capacities of Filtrasorb 400, Ambersorb 572, and Ambersorb 563 (Davis and Powers, 1999). TBA For comparison to MTBE, a list of the chemical properties of TBA is included in Table 5-5. TBA isotherms were provided for Ambersorb 563, Dowex Optipore L-493, and Filtrasorb 400 by Sun (1999) (Figure 5-7). Isotherm tests for Ambersorb 563, Dowex Optipore L-493, and Filtrasorb 400 were performed for synthetic groundwater (buffered, organic-free water with a concentration of salts representative of groundwater) spiked with TBA. An isotherm test was also performed for Ambersorb 563 and a water sample from one of the wells in the Charnock well field in Santa Monica, California (discussed in Section 5.5). The results suggest that Ambersorb 563 has a significantly greater sorption capacity for TBA than both Filtrasorb 400 and Dow L-493. These experiments strongly suggest that Ambersorb 563 should be considered for use in sites where MTBE and TBA co-exist. Further testing should be performed to confirm the results from these tests and to determine any decrease in MTBE and TBA sorption capacities resulting from competitive sorption. 285

308 Sorbed Phase Concentration (mg sorbate/g sorbent) Sorbed Phase Concentration (mg sorbate/g sorbent) TBA Isotherms Ambersorb Charnock sample (Sun, 1999) Ambersorb 563 (Sun, 1999) Dowex Optipore L-493 (Sun, 1999) Filtrasorb 400 (Sun, 1999) Aqueous Phase Concentration (mg/l) (mg/l) Sorbent log K F n Resins Ambersorb Charnock sample (Sun, 1999) Ambersorb 563 (Sun, 1999) Dowex Optipore L-493 (Sun, 1999) GAC Filtrasorb 400 (Sun, 1999) Figure 5-7. TBA Isotherms. One isotherm was conducted for a field sample (Charnock well field in Santa Monica, CA). All other isotherms were conducted for synthetic groundwater at 25 C. The endpoints do not mark actual data points and are used for identification purposes only. To evaluate the effects of the presence of TBA on the MTBE sorption capacities of synthetic resins vs. GAC, Suffet et al. (1999) performed isotherm studies on Ambersorb 572 and GRC- 22 using organic-free water spiked with 100 µg/l TBA. Samples contained varying amounts of the sorbent and the same initial concentrations of MTBE at 1,000 µg/l. The sorption capacity of the Ambersorb 572 for MTBE appeared to be unaffected by competition from TBA while there was a significant decrease in the MTBE sorption capacity of the GRC-22 (Figures 5-8a and 5-8b). In a separate study, Sun (1999) found that the MTBE sorption capacities of the synthetic resins Ambersorb 563 and Dow Optipore L-493 were approximately an order of magnitude greater than their TBA counterparts (Figure 5-9). 286

309 Log Sorbent Phase Concentration (µg MTBE/g sorbent) Log Aqueous MTBE Concentration (µg/l) (a) Ambersorb 572 Log Sorbent Phase Concentration (µg MTBE/g sorbent) Log Aqueous MTBE Concentration (µg/l) (b) Coconut-based GAC (GRC-22) Figures 5-8a and 5-8b. Effects of TBA (100 µg/l) on the equilibrium sorption capacities of coconut-based GRC-22 and Ambersorb 572 resin (Suffet et al., 1999). 287

310 Sorbed Phase Concentration (mg sorbate/g sorbent) Sorbed Phase Concentration (mg sorbate/g sorbent) MTBE vs vs. TBA TBA Isotherms MTBE: Ambersorb 563 (Sun, 1999) TBA: Ambersorb 563 (Sun, 1999) MTBE: Dowex Optipore L-493 (Sun, 1999) TBA: Dowex Optipore L-493 (Sun, 1999) Aqueous Phase Concentration (mg/l) (mg/l) Figure 5-9. MTBE vs. TBA isotherms for Ambersorb 563 and Dowex Optipore L-493 resins. The lines shown are based on a series of data from laboratory tests performed over the given ranges. The endpoints do not mark actual data points and are used for identification purposes only. Continuous and dashed lines correspond to Ambersorb 563 and Dow L-493, respectively. Desorption As in the case of GAC, desorption from resins is most likely to occur when the influent concentration of MTBE decreases significantly, reversing the concentration gradient between the bulk liquid and the resin pore space, or when the influent concentration of other organic compounds increase, resulting in competitive sorption. If desorption occurs to a significant extent, one will observe spikes in effluent concentrations that may exceed influent concentrations. Thus, the occurrence of desorption may create the need for multiple resin vessels in series, a greater number of sampling locations, a higher frequency of sampling, and a more frequent regeneration of resins. Due to the limited number of field studies for MTBE treatment, there are currently no published data available on the topic of MTBE desorption from resins; however, the upcoming pilot test at the Charnock well field in Santa Monica, California (see Section 5.5) is expected to provide some valuable information regarding desorption. Regeneration As stated previously, the ability to regenerate resins on-site may result in an economic advantage over the use of GAC. The following sections describe the regeneration methods currently available for resins and present the results of studies that have evaluated the effectiveness of these methods. These methods include steam regeneration, solvent regeneration, and micro- 288

311 wave regeneration. A detailed analysis of the economics of resin systems using steam regeneration is presented in Section 5.6. Steam Regeneration In steam regeneration, saturated steam is passed through a loaded/saturated resin bed, condensed, and collected. The steam is used to desorb organic compounds from the resin and to transport them away from the column. The condensed steam is typically discarded or treated through superloading. In superloading, the condensate is passed through a sorbent column just prior to the column s regeneration cycle, taking advantage of the additional capacity of the sorbent at higher concentrations. The use of steam has been demonstrated to be an effective means of regenerating Ambersorb 563 resins loaded with a variety of organic compounds such as THMs (Vandiver and Isacoff, 1994) and TCE (Parker and Bortko, 1991). In these cases, steam regeneration was able to fully restore the resin s sorption capacity for these compounds. Suri et al. (1999) confirmed these findings in a study which showed that 28 to 40 BVs of steam (160 C) per regeneration cycle effectively regenerated resins saturated with p-polychlorinated biphenyls (p-pcb), PCE, and CCl 4 over six regeneration cycles. In the case of o-pcb, there was a 20 percent loss in capacity observed after the first regeneration cycle; however, subsequent cycles did not result in further loss of sorption capacity (Suri et al., 1999). Limited studies have been published on the specific application of steam regeneration to MTBE-loaded columns. A preliminary study by Sun found a five percent decrease in the MTBE sorption capacity of an Ambersorb 563 column after the first steam regeneration cycle (1999). However, consistent with the results observed for o-pcb described above, no further loss in capacity was observed in succeeding regeneration cycles. A pilot plant scheduled to be operated in Santa Monica, California (Section 5.5.5) may provide more longterm quantitative information on the effectiveness of steam regeneration specifically for MTBE removal applications. There is research currently being performed on a process that combines steam regeneration with a metal oxide-catalyzed photooxidation process. The objective of this research is to develop a combined process that would not only remove organic contaminants from water, but also destroy them directly during the regeneration process. Suri et al. (1999) published results from a study that evaluated the steam regeneration of Ambersorb 563 beads impregnated with the metal oxide catalysts, TiO 2 and Pt-TiO 2. It had previously been reported that Ambersorb 563 impregnated with metal oxide catalysts can provide significant destruction of certain organics in the gas phase at 250 C (Brendley et al., 1993 and Vandersall et al., 1993). However, while Suri et al. (1999) found that 28 to 40 BVs of steam at 160 C effectively regenerated resins saturated with p-dichlorobenzene (p-dcb), o-pcb, PCE, and CCl 4, they observed minimal or insignificant destruction of these adsorbed compounds during regeneration. Suri et al. noted that higher destruction of adsorbed organics 289

312 on Ambersorb 563 may potentially be achieved by using more active catalysts, greater steam contact times (>0.9 seconds), or higher steam temperatures. The development of steamregenerable resins for combined removal and destruction of organic compounds such as MTBE is an important issue that should be explored in future studies. Solvent Regeneration In solvent regeneration, a solvent in which the adsorbate is highly soluble is passed through the saturated bed. Studies by Rohm and Haas have demonstrated that the use of solvents such as methanol or acetone at a flow rate of 2 BVs an hour can successfully regenerate resin columns (Rohm and Haas, 1992). In one study, methanol was found to extract more than 99 percent of 280 mg of TCE adsorbed per gram of Ambersorb 563 (Parker and Bortko, 1991). A second study showed that 4 to 5 BVs of methanol at a flowrate of 1 BV per hour can remove more than 95 percent of 1,2-dichloroethane (DCA) loaded on an Ambersorb 563 column (Isacoff et al., 1992). A rinse with water or steam is typically performed after regeneration to remove any residual regenerant prior to the next sorption cycle. Limited studies have been published on the effectiveness of solvent regeneration specifically for MTBE-saturated columns. However, in one experiment by Malley et al. (1993), methanol was demonstrated to be ineffective in regenerating an MTBE-saturated Ambersorb 563 column. In this experiment, methanol was passed through an Ambersorb 563 column supersaturated with MTBE at a rate of 4 BVs an hour for 4 hours. After being returned to service, the regenerated column had lost a significant portion of its sorption capacity and was able to remove only between 0 and 15 percent of MTBE in the simulated groundwater. Although there is currently insufficient data to draw conclusions on the effectiveness of solvent regeneration for MTBE applications, it is unlikely that solvent regeneration will be approved for drinking water applications due to concerns over potential contamination of the effluent with the solvent, inefficiencies of the process, and uncertainties in the final disposal. Microwave Regeneration Most studies on the use of microwave regeneration of sorbents examined vapor-phase applications in which the sorbents are used to remove VOCs from gaseous emission streams. However, the use of microwave regeneration for a liquid phase sorbent system is likely to be a similar process. Unlike steam regeneration, which uses steam to heat up the sorbent system, microwave irradiation generates heat directly in the sorbent bed by exciting sorbent and adsorbate molecules. Contaminants adsorbed onto a resin column are volatilized and subsequently extracted through an induced vacuum (AmeriPure, Inc., 1999). Microwave heating has been shown to effectively eliminate the heat and mass transfer resistances that limit the rate of regeneration in conventional steam systems (Price and Schmidt, 1997). 290

313 A bench-scale study by Price and Schmidt (1997) demonstrated the ability of microwave irradiation to desorb VOCs such as methyl ethyl ketone (MEK), toluene, and n-propyl acetate from the polymeric resin Dowex Optipore L-502. A pilot scale of a moving bed column desorber (Salinas et al., 1999) found effective regeneration and recovery of the polar solvents isopropyl alcohol and MEK from Dowex Optipore L-502 and the carbonaceous resin Ambersorb 600. Non-polar toluene was found to be recoverable through microwave regeneration of Ambersorb 600. Dowex Optipore L-502 and Ambersorb 600 are vapor-phase application versions of Dowex Optipore L-493 and Ambersorb 563 and 572, respectively. As in the cases of steam and solvent regeneration, further work is needed to more completely quantify the applicability of microwave regeneration to MTBE applications. 291

314 292

315 5.4 Key Variables in the Design of Synthetic Resin Sorbent Systems Type of Synthetic Resin The various synthetic resins for which MTBE isotherms are available were presented in Figure 5-3. Based on the data currently available, Ambersorb 563 and Ambersorb 572 appear to have the most competitive sorption capacities of the resins that have been evaluated. It is recommended that these two resins be considered in bench-scale or pilot-scale designs. Although isotherm studies are useful for screening resins, it is important to note that these isotherm data are based on batch equilibrium sorption studies that may not be directly representative of dynamic performance configurations. Like the case of GAC, the actual mass loading on a resin may be significantly lower than that predicted from an isotherm study depending on numerous variables such as competitive sorption effects, background water quality, contact time, etc. Tests performed by Calgon Carbon (Pittsburgh, PA) have shown that operating carbon usage rates, based on capacity at the time of breakthrough, can be estimated at 45 to 55 percent of the equilibrium capacity for VOCs (Stenzel and Merz, 1988). In order to determine a similar relationship for resins, it would be necessary to conduct dynamic column tests. For use in drinking water applications, the synthetic resin has to be certified by the National Sanitation Foundation (NSF) under Standard 61, which regulates the use of products in contact with potable water. Of the products listed in Table 5-6, only Ambersorb 563, Ambersorb 572, Filtrasorb 400, and CC 602 have been confirmed to meet this criterion Background Water Quality Limited data are currently available on the effect of background water quality on the MTBE removal efficiency of resins. As discussed in Section 5.3.3, the data currently available suggest that the performance of Ambersorb resins is unaffected by ph (6.5 to 8.5), temperature (10 C vs. 25 C), oxidants (HOCl, H 2 O 2, and O 3 ), the presence of NOM, and the presence of TBA. Ambersorb resins have also been found to be unsusceptible to biofouling. However, m-xylene, which could be considered representative of BTEX compounds, has been found to compete with MTBE sorption. Site-specific studies should be conducted to investigate the extent to which other synthetic compounds could compete with MTBE sorption. In addition, as discussed in Section 5.3.3, the issue of MTBE desorption would need to be addressed through site-specific studies to determine the potential need for multiple resin vessels in series, a greater number of sampling locations and a higher frequency of sampling, and a more frequent regeneration of resins. Depending on the background water quality, pretreatment of influent water may be necessary to optimize the efficiency of resins for MTBE removal. Influent water may need to be filtered 293

316 to remove particulate matter, which may clog the top of the resin column and create poor distribution. To reduce the organic loading on the resin, it may be prudent to use GAC columns at the front end of the process flow to remove BTEX compounds. GAC is cheaper on a per unit basis and is generally more effective for the removal of highly hydrophobic compounds like BTEX. In cases of high NOM fouling, disinfection of the influent water may be necessary Process Flow Configuration The process flow configuration of synthetic resin systems is very similar to that of GAC systems. The main difference between a resin system and a GAC system is the provision for a regeneration process. A typical process flow configuration for a resin system is shown in Figure Sorption columns can be used in either a downflow or upflow service mode. In general, the system configuration will be dependent on a number of factors, including the effluent standard, regeneration technique, and vessel design constraints. In situations where low effluent standards must be met (such as primary or secondary drinking water standards for MTBE) and, thus, low leakage levels are allowed, a resin system works best under countercurrent operation and regeneration, with operation in the upflow mode and regeneration in the downflow mode (Rohm and Haas, 1992). Using the downflow mode for regeneration has been found to result in better removal efficiencies and lower leakage levels once the column is returned to service after regeneration (Rohm and Haas, 1992). Sorbent columns can be operated in series, in parallel (also referred to as carousel), or as a combination of the two configurations depending on a number of factors, including the need for continuous operation, space constraints, effluent criteria, service cycle time constraints, operation logistics, and requirements for multi-barrier treatment. The operation of columns in-series (partially or completely) confers several advantages (EPA, 1995): The system can continue to operate at full flow while the lead column is being regenerated. A lag column provides extra insurance that the effluent water quality will meet stringent effluent criteria. A lag column allows complete exhaustion of sorption capacity for the lead column. In a single column mode, the vessel would have to be regenerated once the effluent quality reached the effluent criteria. With two columns operating in series, the lead column would not be subjected to final effluent standards and, thus, can take a higher loading of the adsorbate. 294

317 Figure Typical process flow configuration for a resin system (based on EPA, 1995). 295

318 However, from a cost perspective, the specific magnitude of the in-series advantage is a function of the length of the breakthrough curve as determined by the length of the MTZ (Figure 5-11). If the MTZ is very short (i.e., < 0.5 meter for a 6,000 gpm system), the increased utilization of the leading column will be a small percentage of the breakthrough time and, thus, the cost-savings resulting from in-series operation may not offset the increased capital costs. Alternatively, if the MTZ is long, the second vessel will significantly increase the time until the first vessel requires regeneration. Many field systems do not exhibit ideal, narrow breakthrough curves, but show elongated curves with tailing on the front and back end (Suffet, 1999 and Sun, 1999). This would suggest that in-series operation offers an economic advantage over carousel operation. This hypothesis is evaluated in the economic analysis (Section 5.6) Regeneration The ability to regenerate a resin sorbent bed on-site is a key factor in determining the economics of a resin system. Several alternatives available for regeneration were discussed in Section These options include steam regeneration, solvent regeneration, and microwave regeneration. As stated in Section 5.3.3, solvent regeneration is unlikely to be approved for treatment applications. Therefore, this method will not be considered further. Salinas et al. (1999) reference a study by Schweiger et al. (1993) that found the regeneration of sorbent beds by conventional heating methods (i.e., steam) to be an inefficient process. The study found that most of the energy is used to heat the sorbent and the vessel, and only about one-fourth to one-third of the energy is actually used for desorption of the sorbent. This factor may be an important consideration in choosing between the use of steam vs. microwave irradiation. Influent 50% removal efficiency Used sorbent 100% removal efficiency Unused sorbent Concentration of MTBE distance in sorbent vessel MTZ Effluent Figure Hypothetical breakthrough curve and its MTZ at one point in time. 296

319 Steam Regeneration For their Ambersorb products, Rohm and Haas recommends a steam pressure of 20 to 30 psig and a temperature between 125 to 135 C. A total steam volume of 0.5 to 2.0 BVs measured as condensate is typically used. Regeneration efficiency can be optimized while minimizing the cycle time by using a lower flow rate for initial steam introduction followed by higher flow rates during later stages (Rohm and Haas, 1992). The total number of BVs of steam condensate can be optimized based on the effluent targets and required minimum service cycle times. Chlorinated organics typically require up to 20 BVs of steam as condensate to consistently meet stringent effluent criteria such as MCL drinking water standards. In the case of chloroform (CHCl 3 ), 85 to 90 percent is generally removed from the resin after the first 5 to 10 BVs of steam (measured as condensate) are passed through the bed. There were no MTBE-specific data available. Ambersorb beds should be cooled down for at least 2 hours after steam regeneration. Rohm and Haas (1992) recommends that treated water be introduced upflow at a low flow rate (0.25 gpm/ft 2 ) to rehydrate the bed and remove air pockets without disturbing the bed. MTBE- and site-specific optimal operating conditions should be determined through benchor pilot-scale studies. Microwave Regeneration The following variables should be considered in the design of a microwave regeneration system: Purging Efficient desorption requires a purging of the desorbed compounds from the sorbent columns to prevent build-up of the gas phase concentration. For microwave regeneration, purging can be done by inducing a vacuum in the sorbent column or by flowing a purge gas through it. An evaluation by Price and Schmidt (1998) indicate that vacuum purging offers substantial performance and economic benefits, including a smaller condenser and a lower microwave power consumption rate. Regeneration Pressure The operating pressure of a microwave system is a key parameter that affects desorption kinetics, as well as the size and power consumption of the recovery system (Price and Schmidt, 1998). While a lower pressure (i.e., higher vacuum) improves the extraction of volatilized compounds, it requires a larger vacuum pump, which consumes more power. However, microwave power consumption and generator capacity decrease since reducing the pressure lowers the final temperature to which the bed must be heated to achieve a given 297

320 degree of desorption. A preliminary cost evaluation by Price and Schmidt (1998) revealed that, overall, the benefits of decreased pressure outweigh the disadvantages. Mechanical vacuum pump energy is generally cheaper and has a stronger effect on desorption equilibria than microwave heating. The optimal pressure for MEK on Dowex Optipore L-502 was found to be in the 1-5 torr range (Price and Schmidt, 1998). Due to the limited amount of data and experience available on the use of microwave regeneration for MTBE sorption systems, the optimal operating range for an MTBE and sorbent resin system will need to be determined from a bench- or pilot-scale study. Post-regeneration Sorbent Concentration The degree to which the sorbent is regenerated is primarily dependent upon the heating temperature. Unlike the case of steam regeneration, where complete desorption of contaminants is seldom economically practical, experimental studies (Price and Schmidt, 1998) indicate that microwave regeneration allows for an economically attractive complete regeneration of resin beds. This is true for several reasons: 1) microwaves heat the entire bed while the vacuum pump maintains a uniform gas pressure (consequently, there are no moving heat and MTZs and the bed can be efficiently desorbed to completion); 2) high bed temperatures are readily achieved with microwave heating since there is no limiting heat source temperature; and, 3) regenerating the bed to near completion will minimize the possibility of early breakthrough in the subsequent sorption cycle. AmeriPure, Inc. (Newport Beach, CA) is presently marketing a patented microwave regeneration system for MTBE applications. Pilot- and full-scale units have been installed for vaporphase sorbent systems involving the removal of toluene, benzene, acetone, and other organic compounds from an industrial waste air stream but, to date, no units have been installed for MTBE applications (AmeriPure, Inc., 1999). AmeriPure, Inc. s microwave-regenerated system for MTBE applications is designed with Ambersorb 563. When a resin column is ready for regeneration, the resin beads are mechanically removed from the vessel, dried to 12-percent water content (by mass), and transported to a stand-alone microwave regeneration system. The regeneration system consists of a cylindrical stainless steel vessel with a microwave guide running along its centerline. A ceramic screen separates the guide from the resin beads while a stainless steel mesh holds the resin beads away from the tank walls. The resin beads are irradiated with microwaves and heated to 350 to 400 F to volatilize any adsorbed organic compounds. An induced vacuum is used to extract the volatilized compounds, which can then be treated through thermal oxidation. After regeneration, the beads are cooled using a cold surface as a heat exchanger or by misting the beads with deionized water. The cooled resin beads are then transported back to their original vessels. The regeneration process is approximately a 6-hour operation, from the time the resin column is ready for regeneration to the time the column is placed back 298

321 on-line. AmeriPure, Inc. estimates that a 50 gpm sorbent system will require a 2-kW microwave power generator Operating Parameters EBCT or Flow Rate Loading The EBCT which is calculated as the BV divided by the flow rate, or its reciprocal, the flow rate loading is used to estimate the volume of sorbent required and the number of reaction vessels necessary. The recommended flow rate loading depends on the effluent criteria, service cycle time constraints, pressure drop, or other site constraints (Rohm and Haas, 1992). Resin sorption system flow rates vary from 0.25 gpm/ft 3 up to 8 gpm/ft 3 sorbent (Rohm and Haas, 1992). Preliminary tests conducted for the Charnock well field in Santa Monica, California (see Section 5.5.5) indicated that resins, specifically Ambersorb 563, require a shorter EBCT of ~5 minutes (1.5 gpm/ft 3 ) compared to GAC for MTBE applications. In general, GAC systems in municipal water treatment plants have EBCTs ranging from several minutes to more than 10 minutes (see Chapter 4). However, for MTBE removal, a higher EBCT is required for GAC systems (10 to 20 minutes) due to the low affinity of MTBE for carbon, relative to other organic contaminants in water treatment plants (e.g., THMs) (Sun, 1999). The lower EBCT required for a resin system results from the kinetics of sorption on resins compared to GAC. As discussed in Section 5.3.1, resins are designed with a greater percentage of mesopores and macropores, which facilitate the transport of adsorbate molecules to the sorption sites. A lower EBCT for a sorption system can provide the following design advantages: 1) a smaller volume of sorbent is required for a given flow rate; 2) a vessel with the same size can take higher flow rates; or 3) fewer units running in parallel will be required for a given flow rate. Hydraulic Loading Hydraulic loading rates (or linear flow rates) for GAC units vary from 0.4 to 12 gpm/ft 2, with typical values ranging from 3 to 4 gpm/ft 2 (Faust and Aly, 1998). It is recommended that similar hydraulic loading ranges be used in the preliminary design of resin adsorber systems. Hydraulic loading rates of 4 and 10 gpm/ft 2 result in pressure drops of 0.7 and 2 psi/ft of Ambersorb resin bed depth, respectively. Vessel Height The height to diameter ratio of the adsorber vessels is a function of flow distribution requirements, pressure drop, or space constraints (EPA, 1995). A minimum bed height of 2 299

322 to 3 feet is typically recommended for each adsorber vessel. A deeper bed provides a margin of safety by providing a larger treatment zone for the less strongly adsorbed compounds. The deeper bed also enhances flow distribution and water contact within the sorption vessel. However, a deeper bed results in a greater head loss across the column, which may increase energy costs Manufacturers and Resin Unit Costs The costs of various polymeric and carbonaceous resins commercially available are listed in Table 5-6. Costs for Filtrasorb 400 and CC-602 are also included for comparison. Polymeric resins are generally less expensive than carbonaceous resins. Synthetic resins can be 5 to 30 times more expensive than GAC on a per unit basis. 300

323 Table 5-6. Manufacturer Information and Unit Costs Manufacturer Address/Phone Products Polymeric Resins Rohm Rohm and and Haas Haas Company Company Dow Dow Chemical Chemical Company Company Bayer Bayer Purolite Purolite Carbonaceous Resins Rohm Rohm and and Haas Haas Company Company Home Home Office Office Independence Independence Mall Mall West West Philadelphia, Philadelphia, PA PA (215) (215) Customer Customer Info Info Group Group Building Building Midland, Midland, Michigan Michigan (800) (800) Corporate Corporate Communications Communications Bayer Bayer Rd. Rd. Pittsburgh, Pittsburgh, PA PA (412) (412) Corporate Corporate Office Office Monument Monument Road, Road, Bala Bala Cynwyd Cynwyd Philadelphia, Philadelphia, PA PA (610) (610) Home Home Office Office Independence Independence Mall Mall West West Philadelphia, Philadelphia, PA PA (215) (215) Amberlite XAD-4 $12.60-$12.98 Amberlite XAD-7 $13.41-$13.78 Unit Cost $/lb $/ft 3 $530-$545 $530-$545 $550-$565 $550-$565 DOWEX L-285 $8.33-$13.10 $350-$550 $350-$550 DOWEX L-323 $8.85-$13.08 $345-$510 $345-$510 DOWEX L-493 $11.11-$15.14 $466-$636 $466-$636 Lewatit VP OC 1066 $14.70 $665 $665 Hypersol-Macronet --- $400 $400 Ambersorb 563 $35-$40 Ambersorb 572 $50-$55 $1,155-$1,320 $1,155-$1,320 $1,550-$1,705 $1,550-$1,705 Absorbents MicroClean Services Co Vannoy Ave. PetroLOK $33 $726 Castro Valley, CA (510) Guardian Environmental P.O. Box 517 PolyGuard $9 $165 Technologies 25 North Main Street Kent, CT (860) GAC Calgon Carbon Corporate Headquarters Filtrasorb 400 $1.22-$1.88 $38-$ Calgon Carbon Drive Pittsburgh, PA US Filter/Westates 6611 San Leandro St. CC 602 $ $26-$29 Oakland, CA (510)

324 302

325 5.5 Pilot-scale/Field Studies Major Oil Refinery in Bakersfield, California (AmeriPure, Inc., 1999) Pilot testing of a resin unit using Dowex Optipore L-493 followed by Ambersorb 563 was conducted at a major refinery in Bakersfield, California to treat influent concentrations of 140 to 160 µg/l MTBE. 1,250 gallons of water contaminated with MTBE, BTEX, and other gasoline components were pumped through the unit at an average rate of 0.5 gallons per minute over a 2-week period in December 1997 (2,500 minutes of operation time). MTBE in the effluent was non-detectable by U.S. EPA method 8240 throughout the duration of testing. The resins were regenerated using 5 gallons of steam measured as condensate. Steam concentrated the MTBE by a factor of approximately 250, producing condensate with 38.7 mg/l MTBE concentration. Recovery of MTBE from the resins after regeneration was greater than 90 percent World Oil Service Station (AmeriPure, Inc., 1999) Influent concentrations of 200 µg/l MTBE were treated using a unit similar to the one described above (Dowex Optipore L-493 followed by Ambersorb 563). 500 gallons of water containing MTBE and BTEX compounds were treated at a flow rate of 1.0 gallon per minute. The concentration of MTBE in the L-493 effluent ranged from 1 to 3 µg/l while it was nondetectable in the Ambersorb 563 effluent throughout the test (EPA method 8240). The L-493 column was regenerated using 2.5 gallons of steam (measured as condensate), which recovered approximately 93 percent of the MTBE BP Oil Company (Alisto Engineering Group, 1996) Field pilot-scale testing was conducted to assess the sorption capacity and removal effectiveness of Ambersorb 563 for MTBE and BTEX. The testing unit consisted of two 9-inch diameter by 44-inch high stainless steel canisters connected in series. Each canister was filled with 33 pounds of resin. 6,000 gallons of contaminated groundwater with concentrations ranging from 49,000 to 110,000 µg/l of MTBE were pumped continuously through the unit at a rate of 0.5 gallons/minute. Groundwater samples were collected daily before and after each canister until the resin in the second canister was exhausted. Ambersorb 563 was found to be effective in adsorbing MTBE with a preference and higher capacity for BTEX compounds. No BTEX breakthrough was observed throughout the duration of testing while MTBE was detected in the effluent of the first canister after a throughput of 1,590 gallons. MTBE was not detected above the reported detection limit in the effluent samples of the second canister until 4,070 gallons had been treated. After 6,018 gallons, the resins in both canisters were exhausted and no more MTBE sorption was taking place. 303

326 Overall, 1 pound of resin was found to remove 0.05 pound of MTBE (50 mg MTBE/g Ambersorb 563) in this site s groundwater. At a flow rate of 0.5 gallon per minute, the life expectancy of fresh resin for a 2,000-lb system was expected to be about half a year. At a rate of $40/lb, resin costs would be $160,000/year, not including additional labor, sampling, and energy expenses. Due to the high projected costs, the resin system was not considered further and no attempt to regenerate the resin was made Gas Station in Bellingham, MA (Winkler, 1999) A pilot demonstration of groundwater remediation using PolyGuard was conducted by the Massachusetts Strategic Envirotechnology Partnership (STEP) program in cooperation with GET and Environmental Compliance Services (Winkler, 1999). The site involved groundwater contaminated with BTEX and MTBE from a gasoline spill. Concentrations of BTEX and MTBE were reported to be 75 and 125 mg/l, respectively, prior to the demonstration. However, these concentrations decreased by a factor of three or more during the testing. The demonstration, which began in August 1997 and ended in December 1998, involved several phases and various design configurations but, in general, the PolyGuard vessels were constructed in series and were followed by a GAC vessel which served as a polisher. A minimum contact time of 15 minutes was provided with operating rates of up to 3 gpm. Under the varying conditions of the test, the highest absorption capacity observed was 150 mg MTBE/g PolyGuard (15 percent capacity, by weight) under an average flow rate of 0.98 gpm and an average influent concentration of 263 mg/l. The PolyGuard system never reached its full capacity during the evaluation because all tests had to be ended prematurely to meet discharge requirements. BTEX was removed at 100 percent during all of the tests. The results of this work and other unpublished studies suggest that PolyGuard is best applied in bulk removal situations where influent concentrations of MTBE are greater than 25 mg/l (Litwin, 1999). Further pilot testing of PolyGuard was scheduled to commence in August 1999 at a major oil refinery site where concentrations of MTBE and total petroleum hydrocarbons (TPH) are 10 mg/l and 0.3 mg/l, respectively (Litwin, 1999). The operating flow rate is designed to be 5 gpm Charnock Well Field in Santa Monica, CA In Progress (Rodriguez, 1999) All five drinking water wells located at the Charnock well field in Santa Monica, California were shut down due to MTBE contamination with concentrations of up to 610 µg/l. Resins are currently being evaluated as a potential treatment technology for this site due to the combined presence of MTBE and TBA. Pilot testing may be conducted in 2000 to: 1) determine the effectiveness of resin technology in removing MTBE and TBA to levels below primary MCLs or other limits imposed by CDHS; 2) identify potential problems associated with background water quality and regeneration; and, 3) collect data for calibration of AdDesignS, a sorption system design software. The calibrated model will then 304

327 be used to simulate different inlet concentrations which were not evaluated through actual testing. The incidental removal of other VOCs such as 1,1-DCE and TCE during testing will also be evaluated as part of this study. The tests will be conducted at one of the wells which was shut down due to contamination. The pilot plant will consist of four operational resin beds, each containing 1.67 ft 3 of Ambersorb 563. The pilot plant will be operated at flow rates of 5 and 4 gpm with corresponding EBCTs of 2.5 and 3.1 minutes, respectively, per vessel. Regeneration of the resin beds will be performed using steam with a pressure range of 35 to 75 psi and a temperature range of 280 to 328 F. A 5 BHp electric boiler will be used to generate the steam. Spent steam and dissolved organics will be delivered to a condenser skid where it will be cooled by a forced draft heat exchanger and then passed through a subcooler inside the condenser. The steam condensate will be collected in a drum for disposal. The regenerated resin will be cooled down for 2 hours then rehydrated for 1 hour at a flow rate of 0.25 gpm/ft

328 306

329 5.6 Economic Analysis Objective and Background The primary purpose for evaluating the use of resins for treating drinking water contaminated by MTBE is to determine the potential cost savings relative to air stripping, GAC, and AOPs. Based on the review of the available literature, these potential cost savings originate from: a) Minimal interference from NOM. Elevated concentrations of NOM will 1) reduce the sorption capacity of activated carbon for MTBE; 2) foul air stripping systems; and, 3) competitively interfere with the destruction of MTBE in advanced oxidation systems. All of these phenomena result in higher treatment costs for MTBE. As previously mentioned in Section 5.3.3, the resin sorptive capacity for MTBE appears to be unaffected by NOM. b) Regenerative properties of resins. Despite the high initial capital investment in resins (~$35+/lb resin) compared to GAC (~$1.25/lb GAC), resins can be regenerated and reused on-site whereas carbon must be taken off-site, reactivated, and replaced. Consequently, if resin regeneration costs are sufficiently low and resin usage rates are equal to or less than carbon usage rates, the lifecycle costs for a resin system (30 years) will be less than those for a carbon system. c) Shorter required EBCT. While still under debate by carbon and resin manufacturers, a preliminary test indicated that resins, specifically Ambersorb 563, require a much shorter EBCT (~5 minutes) compared to GAC (10-20 minutes) (Section 5.4.5). Consequently, resin vessels can sustain a much higher flow-through than equivalently sized carbon vessels, resulting in either a reduction in the total number of required resin vessels or a reduction in the size of the resin vessels. The smaller volume of resins required may lead to lower costs. d) Affinity for TBA. Removal of TBA from groundwater has recently become a concern due to its frequent co-occurrence in MTBE-contaminated groundwater. The physio-chemical properties of TBA are such that it cannot be easily air stripped, nor does it readily adsorb to activated carbon. Conversely, it has been suggested that resins have a relatively high sorption affinity for TBA and may, therefore, reduce treatment costs if TBA is present in high concentrations. This suggestion is further evaluated below. e) Minimal interference with other organics. Initial results from bench scale testing by Davis and Powers (1999) suggest that at high BTEX concentrations (>43 mg/l), the resin sorptive capacity decreased by approximately 22 percent for MTBE (aqueous concentration of 1 mg/l). The bench scale testing did not analyze for decreased sorptive capacity at lower BTEX concentrations. It is unlikely that BTEX concentrations will reach 43 mg/l for an extended period of time in most drinking water or remediation scenarios and, thus, BTEX interference with MTBE sorption is expected to be low. f) No by-products formation. Unlike other treatment technologies currently being evaluated for MTBE removal from drinking water (e.g. advanced oxidation processes and biological 307

330 treatment), the use of resins produces no MTBE by-products that may further compromise drinking water quality. Each of these differences suggests a decreased resin cost relative to other well-established treatment technologies for MTBE removal from groundwater. This analysis is intended to identify the least expensive treatment process for resin system use and regeneration. All of the assumptions and background information can be found in Appendix 5B Cost Scenarios For this economic analysis, all combinations of the following flow rates, influent concentrations, and effluent goals were evaluated. These parameters, with the exception of the 6 gpm system, are consistent with the evaluations of air stripping, activated carbon, and advanced oxidation systems in the other chapters. a) Flow rates: 6 gpm, 60 gpm, 600 gpm, and 6,000 gpm. b) Influent concentrations: 20 µg/l, 200 µg/l, and 2,000 µg/l. c) Effluent goals: 0.5 µg/l (representing non-detect), 5 µg/l, and 20 µg/l. In addition, the economic implications of using two resin vessels operated in series vs. two resin vessels operated in parallel or carousel were evaluated. Finally, a number of steam regeneration scenarios were evaluated. Costs of microwave regeneration were not evaluated in the economic analysis due to the uncertainty in its application and effectiveness. AmeriPure, Inc. (Newport Beach, CA) owns a patent for liquid phase applications of a microwave system (Section 5.4.4) and provided their estimate of the annual operating costs of this system for 100 gpm and 2,000 µg/l influent to 0.5 µg/l effluent concentrations ($1.20 to $1.50/1,000 gallons) (Hodge, 1999). Sufficient information and backup assumptions were not available to rigorously review these estimates; however, they are within the range presented in this analysis and warrant further investigation of microwave regeneration. Solvent regeneration was not considered in this evaluation because it is unlikely to be approved for drinking water treatment applications. Thus, the following five scenarios were evaluated for regeneration: a) Steam regeneration followed by commercial hazardous waste disposal of regenerant by a disposal company (e.g., Safety Kleen of Oakland, CA). b) Steam regeneration followed by air stripping the regenerant and treating the off gas with a catalytic oxidizer. c) Steam regeneration followed by a superloaded resin column. Upon breakthrough, this superloaded column was steam regenerated and the regenerant was disposed by hazardous waste disposal company. 308

331 d) Steam regeneration followed by a GAC column. Upon breakthrough, this GAC column was emptied and refreshed with virgin carbon. e) Steam regeneration followed by microbial degradation of regenerant Assumptions All cost estimates were based on engineering judgement, past experience, and computer modeling. All of the capital costs for the resin system, steam regeneration system, and regeneration units were obtained from vendors with correction factors applied to account for a) electricity, valves, and piping; b) site work; c) contractor overhead and profit; d) engineering; and e) contingency. The application of these corrections is illustrated in Appendix 5B. All modeling work to determine breakthrough times and BVs treated was completed on AdDesignS (Mertz et al., 1998). However, AdDesignS predicts a steep breakthrough curve, which conflicts with laboratory and field data that suggest a wider breakthrough curve. Consequently, we have increased the predicted sorption capacity of each vessel by approximately five percent when estimating the time to complete column exhaustion. We have also assumed that two vessels in series can be operated to allow complete exhaustion of each vessel (i.e., the breakthrough curve can be contained within one vessel). Additional cost information, analytical sampling frequency assumptions, labor assumptions, and relevant calculations are presented in Table 5-7 through Table 5-11 and Appendix 5B. While we believe that all necessary information is provided, any additional detailed costs and spreadsheets can be obtained from NWRI upon request Results Annualized capital and O&M costs (1999 $) for the 28 scenarios evaluated range from $0.30/1,000 gallons (Option 2; 6,000 gpm; 20 µg/l to 5 µg/l) to $26.42/1,000 gallons (Option 2; 6 gpm; 2,000 ppb to 0.5 ppb) and can be found in Tables 5-7 and 5-8. These costs are based on 30 years of operation and a seven-percent discount rate. These costs include steam regeneration and one of the previously mentioned regeneration treatment trains, as depicted in Figure Series vs. Carousel Operation As mentioned previously, one objective of this cost estimate was to determine the cost effectiveness of two vessels in series vs. two vessels in carousel configuration. As noted in Section 5.4.3, one major advantage of series operation (Option 1) is that each vessel can be completely exhausted prior to resin regeneration (see Figure 5-13) whereas carousel operation (Option 2) requires regeneration of each vessel some time before complete exhaustion. As a result, vessels operated in series require less frequent regeneration and less monitoring to determine when effluent goals have been reached. Despite a decrease in O&M costs to account for these benefits (i.e., less frequent regeneration), there will be a slight increase in pumping costs for the series option due to a larger head loss across two vessels 309

332 instead of just one. In addition, vessels operated in series will result in an increase in capital costs due to increased piping and controls (i.e., two vessels in series are designed to operate in both directions or alone) and increased number of required vessels (see Figure 5-13). For example, the largest vessels produced can handle maximum flow rates of 670 gpm and, thus, 6,000 gpm requires at least nine resin vessels in parallel to treat the desired flow rate. In a parallel-only or carousel operation, 14 vessels are required to treat the flow (i.e., continuous flow through nine rotating vessels while regenerating the other five vessels) whereas the in-series operation requires nine parallel combinations of two vessels in series, resulting in 18 total vessels (see Figure 5-12). As Table 5-7 and Table 5-8 show, in the 6,000 gpm scenarios, the increased O&M costs of carousel operation (Option 2) are insufficient to offset the increased capital costs of series operation (Option 1) and, thus, carousel operation is the least expensive for each of the influent concentration/treated water goal combinations. For the highest influent concentrations (2,000 ppb) of the 600 gpm, 60 gpm, and 6 gpm scenarios, the comparison is slightly different; series operation is less expensive due primarily to the decreased sampling and regeneration frequency. However, for both of these scenarios and for all other influent concentrations, the cost difference between series and carousel operation is small (<10 percent difference) and well within the estimated uncertainty of this analysis. In conclusion, unless significantly fewer vessels can be used in carousel operation (e.g., the 6,000 gpm scenario), series operation is expected to be easier to operate due to the allowed complete exhaustion of each column prior to regeneration, resulting in continuous non-detect concentrations in the effluent. Regeneration Technique Detailed capital and O&M costs are shown in Tables 5-9 to Air stripping and off-gas treatment result in a high initial capital expenditure compared to GAC or superloaded resins. Consequently, air stripping regeneration is only cost effective in two situations: when the O&M costs for GAC and resins are high enough to offset the amortized capital costs of air stripping, or when influent concentrations are low and air permitting requirements are minimal such that off-gas treatment is not required. This concept is depicted by comparing the 6,000 gpm/2,000 µg/l influent scenario to the 600 gpm/2,000 µg/l influent scenario. For both scenarios, air stripping regeneration costs remain about the same, due to the fixed capital cost of the air stripping and off-gas treatment units. However, the GAC carbon usage rate and power consumption costs fall by an order of magnitude between the 6,000 gpm system and 600 gpm system because there are a ninth as many vessels requiring regeneration. Consequently, GAC regeneration is more expensive for the 6,000 gpm system due to high O&M costs and air stripping regeneration is more expensive for the 600 gpm system due to high capital costs. 310

333 Table 5-7. Option 1 (Series Operation): Capital, O&M, Regeneration, and Total Costs Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Resin + Steam Capital Resin + Steam O&M Resin + Steam $/1000 gallons Least Expensive Regeneration Technique Annual Cost of Regeneration Total (Resin, Steam, and Regeneration) $/1000 gallons % $1,007,690 $814,800 $0.58 Low Profile Air Stripper with OGT $41,455 $ % $1,007,690 $819,231 $0.58 Low Profile Air Stripper with OGT $41,520 $ % $1,007,690 $819,231 $0.58 Low Profile Air Stripper with OGT $41,520 $ % $906,069 $453,107 $0.43 GAC + GAC Disposal $32,047 $ % $1,007,690 $461,320 $0.47 GAC + GAC Disposal $34,482 $ % $1,007,690 $463,842 $0.47 GAC + GAC Disposal $35,184 $ % $804,448 $308,685 $0.35 GAC + GAC Disposal $8,295 $ % $1,007,690 $321,182 $0.42 GAC + GAC Disposal $9,324 $ % $212,454 $180,124 $1.24 GAC + GAC Disposal $22,606 $ % $212,454 $180,435 $1.25 GAC + GAC Disposal $22,800 $ % $212,454 $180,435 $1.25 GAC + GAC Disposal $22,800 $ % $212,454 $151,811 $1.16 GAC + GAC Disposal $5,183 $ % $212,454 $152,598 $1.16 GAC + GAC Disposal $5,411 $ % $212,454 $152,879 $1.16 GAC + GAC Disposal $5,495 $ % $212,454 $105,359 $1.01 GAC + GAC Disposal $2,807 $ % $212,454 $106,618 $1.01 GAC + GAC Disposal $2,911 $ % $26,710 $113,160 $4.44 GAC + GAC Disposal $2,772 $ % $26,710 $113,212 $4.44 GAC + GAC Disposal $2,804 $ % $26,710 $113,212 $4.44 GAC + GAC Disposal $2,804 $ % $26,710 $103,580 $4.13 GAC + GAC Disposal $1,063 $ % $26,710 $103,662 $4.13 GAC + GAC Disposal $1,086 $ % $26,710 $103,685 $4.13 GAC + GAC Disposal $1,092 $ % $16,307 $61,928 $2.48 GAC + GAC Disposal $825 $ % $26,710 $61,962 $2.81 GAC + GAC Disposal $835 $ % $1,885 $76,146 $24.74 GAC + GAC Disposal $374 $ % $1,885 $76,146 $24.74 GAC + GAC Disposal $374 $ % $1,885 $76,146 $24.74 GAC + GAC Disposal $374 $ % $1,885 $64,495 $21.05 GAC + GAC Disposal $206 $ % $1,885 $64,509 $21.05 GAC + GAC Disposal $211 $ % $1,885 $64,510 $21.05 GAC + GAC Disposal $211 $ % $1,885 $61,444 $20.08 GAC + GAC Disposal $185 $ % $1,885 $61,456 $20.09 GAC + GAC Disposal $185 $

334 Table 5-8. Option 2 (Carousel Operation): Capital, O&M, Regeneration, and Total Costs Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Resin + Steam Capital Resin + Steam O&M Resin + Steam $/1000 gallons Least Expensive Regeneration Technique Annual Cost of Regeneration Total (Resin, Steam, and Regeneration) $/1000 gallons % $791,602 $844,862 $0.52 Low Profile Air Stripper with OGT $41,766 $ % $791,602 $874,649 $0.53 Low Profile Air Stripper with OGT $42,250 $ % $791,602 $998,501 $0.57 Low Profile Air Stripper with OGT $44,277 $ % $741,709 $444,335 $0.38 GAC + GAC Disposal $32,885 $ % $791,602 $451,566 $0.39 GAC + GAC Disposal $35,339 $ % $791,602 $460,102 $0.40 GAC + GAC Disposal $37,509 $ % $641,924 $297,560 $0.30 GAC + GAC Disposal $8,384 $ % $791,602 $322,052 $0.35 GAC + GAC Disposal $9,446 $ % $210,619 $182,759 $1.25 GAC + GAC Disposal $23,293 $ % $210,619 $191,280 $1.27 GAC + GAC Disposal $26,149 $ % $210,619 $197,965 $1.30 GAC + GAC Disposal $28,107 $ % $210,619 $150,923 $1.15 GAC + GAC Disposal $5,253 $ % $210,619 $151,698 $1.15 GAC + GAC Disposal $5,522 $ % $210,619 $153,742 $1.16 GAC + GAC Disposal $5,964 $ % $210,619 $104,255 $1.00 GAC + GAC Disposal $2,816 $ % $210,619 $105,189 $1.00 GAC + GAC Disposal $2,922 $ % $25,965 $115,069 $4.47 GAC + GAC Disposal $2,862 $ % $25,965 $115,196 $4.48 GAC + GAC Disposal $2,928 $ % $25,965 $115,244 $4.48 GAC + GAC Disposal $2,945 $ % $25,965 $104,080 $4.12 GAC + GAC Disposal $1,071 $ % $25,965 $104,165 $4.13 GAC + GAC Disposal $1,099 $ % $25,965 $104,206 $4.13 GAC + GAC Disposal $1,108 $ % $16,157 $61,780 $2.47 GAC + GAC Disposal $826 $ % $25,965 $61,818 $2.78 GAC + GAC Disposal $837 $ % $1,769 $80,976 $26.24 GAC + GAC Disposal $400 $ % $1,769 $81,020 $26.25 GAC + GAC Disposal $412 $ % $1,769 $81,113 $26.28 GAC + GAC Disposal $439 $ % $1,769 $64,486 $21.01 GAC + GAC Disposal $207 $ % $1,769 $64,503 $21.01 GAC + GAC Disposal $213 $ % $1,769 $64,514 $21.02 GAC + GAC Disposal $215 $ % $1,769 $61,433 $20.04 GAC + GAC Disposal $185 $ % $1,769 $61,443 $20.04 GAC + GAC Disposal $186 $

335 Table 5-9a. Regeneration Capital Costs for Option 1 (Series Operation) Operation Parameters OPTION 1 Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Regeneration Flow Rate (gpm) MTBE Regenerant Conc. [mg/l] Total Regen. Time [hrs/yr] Amortized Hazardous Disposal Capital Amortized GAC Annual Capital Amortized Resin Annual Capital Amortized Air Strip. + OGT Annual Capital % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $35, % $0 $4,946 $25,093 $35, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $35, % $0 $2,473 $14,561 $35, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17,

336 Table 5-9b. Regeneration Capital Costs for Option 2 (Carousel Operation). Operation Parameters OPTION 2 Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Regeneration Flow Rate (gpm) MTBE Regenerant Conc. [mg/l] Total Regen. Time [hrs/yr] Amortized Hazardous Disposal Capital Amortized GAC Annual Capital Amortized Resin Annual Capital Amortized Air Strip. + OGT Annual Capital % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $36, % $0 $4,946 $25,093 $35, % $0 $4,946 $25,093 $35, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $36, % $0 $2,473 $14,561 $35, % $0 $2,473 $14,561 $35, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $791 $2,876 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17, % $0 $181 $1,088 $17,

337 Table 5-10a. Regeneration O&M Costs for Option 1 (Series Operation) Operation Parameters OPTION 1 Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Regeneration Flow Rate (gpm) MTBE Regenerant Conc. [mg/l] Total Regen. Time [hrs/yr] Hazardous Disposal GAC O&M (Power and Carbon) Resin O&M (Power and Resin) Air Stripper O&M (Power and Fuel) % $22,114,123 $201,305 $704,857 $4, % $22,431,172 $204,191 $714,962 $4, % $22,431,172 $204,191 $714,962 $4, % $4,486,234 $27,101 $84,371 $ % $4,889,294 $29,536 $91,952 $1, % $5,006,638 $30,238 $94,159 $1, % $839,235 $3,349 $8,879 $ % $1,090,296 $4,378 $11,584 $ % $2,211,725 $20,133 $70,489 $ % $2,233,033 $20,327 $71,175 $ % $2,233,033 $20,327 $71,175 $ % $448,623 $2,710 $8,437 $ % $486,270 $2,938 $9,145 $ % $500,264 $3,022 $9,408 $ % $83,178 $334 $884 $ % $108,992 $438 $1,158 $ % $217,629 $1,981 $11,722 $ % $221,147 $2,013 $11,911 $ % $221,147 $2,013 $11,911 $ % $44,820 $272 $1,288 $ % $48,590 $295 $1,396 $ % $49,516 $301 $1,423 $ % $8,378 $34 $223 $ % $10,900 $44 $176 $ % $31,607 $193 $7,456 $ % $31,585 $193 $7,451 $ % $31,585 $193 $7,451 $ % $5,950 $25 $865 $ % $7,073 $30 $1,035 $ % $7,077 $30 $1,036 $ % $1,334 $4 $114 $ % $1,584 $4 $136 $6 Regeneration O&M includes: power ($0.08/KWhr) and/or natural gas ($5/MMBtu). 315

338 Table 5-10b. Regeneration O&M Costs for Option 2 (Carousel Operation). Operation Parameters OPTION 2 Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Regeneration Flow Rate (gpm) MTBE Regenerant Conc. [mg/l] Total Regen. Time [hrs/yr] Hazardous Disposal GAC O&M (Power and Carbon) Resin O&M (Power and Resin) Air Stripper O&M (Power and Fuel) % $23,616,215 $207,668 $725,941 $4, % $25,946,505 $219,214 $755,931 $5, % $35,720,873 $257,968 $839,713 $7, % $4,754,642 $27,939 $86,226 $ % $5,172,146 $30,393 $93,133 $1, % $5,794,719 $32,563 $98,742 $1, % $880,211 $3,438 $9,049 $ % $1,158,944 $4,500 $11,819 $ % $2,351,594 $20,820 $72,367 $ % $3,022,018 $23,676 $79,308 $ % $3,549,598 $25,634 $83,443 $ % $473,038 $2,780 $8,579 $ % $518,930 $3,049 $9,344 $ % $675,113 $3,491 $9,908 $ % $87,910 $343 $904 $ % $115,702 $449 $1,180 $ % $236,972 $2,071 $10,934 $ % $246,172 $2,137 $11,196 $ % $249,720 $2,154 $11,261 $ % $47,699 $280 $1,312 $ % $52,586 $308 $1,436 $ % $55,140 $317 $1,475 $ % $8,947 $35 $229 $ % $11,763 $45 $180 $ % $36,284 $219 $8,429 $ % $40,795 $231 $8,492 $ % $50,453 $258 $9,181 $ % $6,512 $26 $889 $ % $8,023 $32 $1,080 $ % $9,135 $34 $1,125 $ % $1,439 $4 $117 $ % $1,796 $5 $143 $6 Regeneration O&M includes: power ($0.08/KWhr) and/or natural gas ($5/MMBtu). 316

339 Table Regeneration Totals ($/1000 Gallons) Option 1 (Series Operation) Option 2 (Carousel Operation) Flow Rate [gpm] Influent [µg/l] Goal [µg/l] Removal Efficiency Hazardous Disposal $/1000 gallons GAC $/1000 gallons Resin $/1000 gallons Air Stripper $/1000 gallons Hazardous Disposal $/1000 gallons GAC $/1000 gallons Resin $/1000 gallons Air Stripper $/1000 gallons % $7.01 $0.07 $0.23 $0.01 $7.49 $0.07 $0.24 $ % $7.11 $0.07 $0.23 $0.01 $8.23 $0.07 $0.25 $ % $7.11 $0.07 $0.23 $0.01 $11.33 $0.08 $0.27 $ % $1.42 $0.01 $0.03 $0.01 $1.51 $0.01 $0.04 $ % $1.55 $0.01 $0.04 $0.01 $1.64 $0.01 $0.04 $ % $1.59 $0.01 $0.04 $0.01 $1.84 $0.01 $0.04 $ % $0.27 $0.00 $0.01 $0.01 $0.28 $0.00 $0.01 $ % $0.35 $0.00 $0.01 $0.01 $0.37 $0.00 $0.01 $ % $7.01 $0.07 $0.27 $0.12 $7.46 $0.07 $0.28 $ % $7.08 $0.07 $0.27 $0.12 $9.58 $0.08 $0.30 $ % $7.08 $0.07 $0.27 $0.12 $11.26 $0.09 $0.31 $ % $1.42 $0.02 $0.07 $0.12 $1.50 $0.02 $0.07 $ % $1.54 $0.02 $0.08 $0.12 $1.65 $0.02 $0.08 $ % $1.59 $0.02 $0.08 $0.12 $2.14 $0.02 $0.08 $ % $0.26 $0.01 $0.05 $0.11 $0.28 $0.01 $0.05 $ % $0.35 $0.01 $0.05 $0.11 $0.37 $0.01 $0.05 $ % $6.90 $0.09 $0.46 $0.57 $7.51 $0.09 $0.44 $ % $7.01 $0.09 $0.47 $0.57 $7.81 $0.09 $0.45 $ % $7.01 $0.09 $0.47 $0.57 $7.92 $0.09 $0.45 $ % $1.42 $0.03 $0.13 $0.56 $1.51 $0.03 $0.13 $ % $1.54 $0.03 $0.14 $0.56 $1.67 $0.03 $0.14 $ % $1.57 $0.03 $0.14 $0.56 $1.75 $0.04 $0.14 $ % $0.27 $0.03 $0.10 $0.56 $0.28 $0.03 $0.10 $ % $0.35 $0.03 $0.10 $0.56 $0.37 $0.03 $0.10 $ % $10.02 $0.12 $2.71 $5.68 $11.51 $0.13 $3.02 $ % $10.02 $0.12 $2.71 $5.68 $12.94 $0.13 $3.04 $ % $10.02 $0.12 $2.71 $5.68 $16.00 $0.14 $3.26 $ % $1.89 $0.07 $0.62 $5.65 $2.06 $0.07 $0.63 $ % $2.24 $0.07 $0.67 $5.65 $2.54 $0.07 $0.69 $ % $2.24 $0.07 $0.67 $5.65 $2.90 $0.07 $0.70 $ % $0.42 $0.06 $0.38 $5.64 $0.46 $0.06 $0.38 $ % $0.50 $0.06 $0.39 $5.64 $0.57 $0.06 $0.39 $

340 6000 gpm Option 1: In-Series Operation Option 2: Carousel Operation Multiple Barrier (Redundant GAC) Multiple Barrier (Redundant GAC) 600 gpm, 60 gpm, and 6 gpm Multiple Barrier (Redundant GAC) Multiple Barrier (Redundant GAC) Figure Process flow configurations for in-series operation vs. carousel operation of resin systems. Note that for 6,000 gpm, the 200 µg/l to 20 µg/l and 20 µg/l to 5 µg/l scenarios require fewer vessels. 318

341 Steam Generator Steam Generator 1 2 Regenerant Regenerant Steam Generator Steam Generator 3 4 Regenerant Regenerant Contaminated water flowing through the vessel Vessel is regenerating Figure Four operational modes for two vessels in series. 319

342 In all cases the superloaded resin option is more expensive than the GAC option; however, this is primarily due to the assumption that once the superloaded resin vessels are exhausted, the regenerant will be disposed by a hazardous waste disposal company. If an alternative option, such as microbial degradation, could be found to degrade the high concentrations of MTBE in this secondary regenerant (MTBE concentrations exceeding 10,000 mg/l), the superloaded resin option may prove to be economically beneficial. Biological degradation of the regenerant was not economically evaluated due to the high degree of uncertainty and untested nature of this option at the concentrations and flow rates in question. However, this option has been qualitatively discussed in the next section. Biological Degradation Biodegradation of MTBE has been the subject of a significant amount of research over the past several years. Recently, several researchers at University of California, Davis and Equilon have identified MTBE-degrading cultures that can grow on MTBE as the only carbon source. In addition, Envirogen (Lawrenceville, NJ) and U.S. Filter/Envirex (Waukesha, WI) sell biofilters that have been used to treat high concentrations of MTBE in water. Consequently, under the right conditions, microbial cultures could be cost-effectively used to degrade the high concentrations of MTBE present in the regenerant stream. However, to date, limited studies have been conducted on the biodegradation of MTBE at concentrations greater than 200 mg/l and flowrates more than several gallons per minute. In addition, Envirogen (Lawrenceville, NJ) cites a field study where 12,000 mg/l MTBE were degraded to less than 5 mg/l for an initial capital expenditure of $200,000. Nonetheless, under appropriate conditions (e.g., a relaxed discharge criteria), potential cost savings could be realized using biological treatment of MTBE for treating MTBE-contaminated regenerant water. A detailed cost estimate is warranted once the appropriate microbes have been identified; however, the following is a qualitative discussion of necessary components of this cost estimate. Components A biological system should include the following components: biodegradation tanks, aeration system, pumps, solids separation, and a chlorination or disinfection system for discharge. The size and number of tanks will be a function of a) the influent MTBE levels; b) the retention time required for the microorganisms to degrade MTBE to desired levels; and, c) the volume of contaminated water requiring treatment. Salanitro indicated a solid retention time of 100 days and a hydraulic retention time of several days for his BC-4 culture, which is currently degrading 200 mg/l of MTBE to <5 mg/l at Port Hueneme, California (1999). At 200 and 2,000 µg/l influent concentrations, the regenerant contains approximately 400 and 1,000 mg/l MTBE. If the BC-4 culture were not able to degrade these concentrations of MTBE, either dilution of the regenerant stream or an alternative microbial strain would be required. If dilution is selected, additional tanks would be required and, thus, costs could dramatically increase. Finally, when choosing a treatment tank, one should consider tank material and its effect on the control of ph, corrosivity, and temperature of the microbial reactor. 320

343 Microbial Requirements All currently identified microbes capable of degrading high concentrations of MTBE are aerobic organisms and, therefore, require oxygen for biodegradation to occur. Salanitro (1999) suggests that oxygen levels should be maintained between 2 and 4 mg/l for MTBE biodegradation with his BC-4 culture. Submersed diffusers or surface aerators could be used to aerate the system; however, maintaining a homogenous, well-aerated system will be difficult. In addition to oxygen, microbes will also require nutrient supplements consisting of phosphorus and nitrogen. Tests should be conducted to determine the required ratios and whether any additional micronutrients are necessary for biodegradation. Finally, microbes will require a steady stream of MTBE contaminated-water to ensure high cell populations. Only for the 2,000 µg/l influent scenarios are resin vessels constantly regenerated; for all other scenarios, there is a lag time between regeneration. If this lag period lasts too long, the microbial population will decrease in size and require a start-up period prior to the next batch of regenerant water. Other considerations Effluent from this biodegradation process will have to conform with all necessary rules, permits and regulations (e.g., the National Pollutant Discharge Elimination System [NPDES] permit). These permits will likely specify the concentration of MTBE and biomass allowed in the effluent. Due to the reliance on large populations of microbes, one should expect biomass and solids production. As a result and if specified in the discharge permit, solid separation facilities may be required to separate the solids from the discharge water and possibly to concentrate solids that need to be wasted from a reactor to maintain its stability. Pilot studies should be conducted to determine the volume of solids that will be produced and the need for effluent polishing, sludge dewatering, and disposal. In addition, the effluent may require disinfection (e.g., chlorination or UV/H 2 O 2 disinfection) to meet permit requirements. Affinity for TBA The results of one study (Sun, 1999) suggest that the carbonaceous resin (Ambersorb 563) sorptive capacity for TBA is higher than that of GAC (Filtrasorb 400); however, there are currently limited published field data to support this claim. According to Sun s data for a field sample (1999), TBA has Freundlich parameters of 1.8 for K F and 0.85 for n for sorption on Ambersorb 563. These parameters are higher than those for most field-tested carbons; however, the results are still not encouraging due to the relatively high resin usage rate. Table 5-12 illustrates the time before complete column exhaustion for TBA relative to MTBE based on modeling with these parameters. This modeling assumes that an EBCT similar to that of MTBE is sufficient for sorption (~8 min). 321

344 Table Time to Column Exhaustion for MTBE vs. TBA. Flow MTBE: MTBE: Time to TBA: TBA: Time to Rate Influent Column Influent Column [gpm] Conc. [µg/l] Exhaustion [days] Conc. [µg/l] Exhaustion [days] 6 2, , , Sensitivity Analysis Table 5-13 presents the results from a sensitivity analysis. As noted earlier, the sorptive capacity of resins for MTBE is unaffected by NOM and is expected to be minimally affected at low concentrations (<43 ppm) of BTEX. The third analysis presents the results from shortening the project lifetime; as the project lifetime shortens the effective annual capital costs increase and the $/1,000 gallons increases. Nonregenerable Absorbent Polymers Manufacturer studies have shown that, under laboratory conditions, sorbent polymers can have extremely high capacities for BTEX and MTBE. However, a significant disadvantage of these sorbents is that they are not regenerable and need to be disposed of after use. Further field investigation is necessary to perform a cost comparison between super absorbent polymers and GAC. In particular, the laboratory and field capacities of these absorbents need to be evaluated through independent testing. Based on the data currently available, it appears that these absorbents are more suitable as bulk removers of MTBE and BTEX in groundwater remediation applications rather than drinking water applications. For most situations, the use of these absorbent polymers will require the use of a polishing unit such as a GAC system. 322

345 323 1) All sensitivity analyses for Option 1 (Series Operation); 600 gpm; 200 mg/l to 5 mg/l. 2) Includes Resin and Steam regeneration costs; does not include GAC. 3) Includes regeneration capital and O&M costs. Sensitivity Parameter 1 NOM Fouling Low Fouling Moderate Fouling High Fouling BTEX Load Resin Capital Cost ($) 2 Resin Annual O&M ($) 2 Total Resin Annual Cost ($) Unit Cost ($/1000 gal) Total Unit Cost ($/1000 gal) 3 No BTEX present $212,454 $152,598 $365,052 $1.28 $1.30 BTEX at 20 µg/l each BTEX at at 200 µg/l each Design Life $212,454 $152,598 $365,052 $1.16 $1.30 Literature review indicates that NOM does not affect adsorption of MTBE or TBA. Literature review indicates that NOM does not affect adsorption of MTBE or TBA. Not sufficient laboratory or field data to determine BTEX interference at low concentrations (<43 mg/l) Not sufficient laboratory or field data to determine BTEX interference at low concentrations (<43 mg/l) 2 years $1,458,143 $152,598 $1,610,741 $5.11 $ years $375,357 $152,598 $527,955 $1.67 $ years $212,454 $152,598 $365,052 $1.28 $1.30 Table Sensitivity Analysis of Cost Estimates

346 5.6.5 Conclusions and Limitations For this analysis, a thorough feasibility-level economic investigation of the capital and operational costs for resin treatment systems was completed (+50 percent/-30 percent). For the 6,000 gpm scenario, carousel operation (Option 2) followed by either low profile air stripping or GAC treatment of the regenerant is the most cost effective treatment option, with costs (1999 $) ranging from $0.30/1,000 gallons (75 percent removal efficiency) to $0.58/1,000 gallons (99.98 percent removal efficiency), as demonstrated in Table 5-8. For all other scenarios, series operation (Option 1) followed by GAC treatment of the regenerant will be the least expensive option (based on 1999 $) ranging from $1.02/1,000 gallons (600 gpm; 20 ppb to 0.5 ppb) to $24.86/1,000 gallons (6 gpm; 2,000 ppb to 0.5 ppb), as demonstrated in Table 5-7. For all cases the cost differential between series and carousel operation was small (<16 percent) and, thus, is well within the estimated uncertainty of this analysis. In some cases alternative assumptions have been made to improve the specific accuracy for resins (see Appendix 5B). These costs are highly contingent on the results from the AdDesignS model, which was used to determine breakthrough time and BVs treated. This model assumes a narrow breakthrough curve that does not likely reflect actual field experience, as previously discussed. If this breakthrough curve is wider in the field than predicted by the model, the time between complete exhaustion and 20, 5, and 0.5 µg/l breakthrough will increase, thereby, increasing the cost-effectiveness of series operation. To account for this, we have added five percent to the column capacity when estimating complete column exhaustion; however, this should be verified in the field. In conclusion, the results from this economic analysis suggest that resins are sufficiently cost-effective to warrant further testing under field conditions to verify the advantages previously noted. Specifically, field evaluation of alternative regeneration scenarios should be undertaken (e.g., microbial degradation and microwave regeneration). Furthermore, if field studies determine that the resin sorptive capacity for TBA is much better than previously thought and better than activated carbon, resins could prove to be much more cost effective than alternative treatment options (e.g., air stripping and activated carbon). Finally, it should be noted that costs are expected to be very site-specific and may vary from the costs presented in this analysis. 324

347 5.7 Recommendations for Future Work Based on the preceding literature review, there are four primary areas where additional information is critical to cost-effectively design a treatment system for MTBE removal from water: 1) dynamic resin sorption capacities for MTBE, TBA, BTEX, and NOM; 2) mechanisms of regeneration and optimization of regenerative processes; 3) the effectiveness of biological degradation as a regenerant treatment; and, 4) the synergistic advantage of using resins in combination with an alternative treatment process. Dynamic Resin Sorption Capacities As noted earlier, batch equilibrium sorption studies are generally not directly representative of dynamic sorbent systems. Dynamic column tests should be performed in order to determine the optimum EBCT and other operating parameters for treating MTBE-impacted groundwater. The effect of fluctuating temperature, ph, and other water quality parameters should be evaluated, possibly for multiple resins. In addition, more information is needed on the sorptive capacity of resins for TBA, BTEX, and NOM and their interference with the sorption of MTBE. This includes a more detailed analysis of competitive sorption using a dynamic model. An analysis of MTBE leakage rates (i.e., MTBE breakthrough caused by competitive sorption) and MTBE/TBA desorption as a result of fluctuating influent concentrations and the presence of other organic compounds is also needed. Finally, initial research suggests that super-absorbents might be effective for the removal of high concentrations of MTBE from water; laboratory research should be completed to fully determine the applicability of absorbent in a drinking water context (e.g., determination of absorption capacity, swell rates, and variability with changing water quality). Mechanisms of Regeneration and Optimization of Regenerative Processes One of the primary economic advantages of synthetic resins over GAC is their regenerability; however, only a limited number of bench- and pilot-scale studies have been completed to investigate the effectiveness of steam and microwave regeneration for MTBE applications. Further information on the feasibility and economics of these regeneration processes is necessary to perform a detailed cost evaluation and comparison. Specifically, microwave regeneration techniques should be evaluated. Effectiveness of Biological Degradation as a Regenerant Treatment If steam regeneration is used, more research is needed to determine the most cost-effective treatment solution for treating the MTBE-concentrated regenerant solution. This review has evaluated air stripping, GAC, and superloaded resin columns; however, the economic analysis has noted that biological degradation of concentrated regenerant solutions may be a viable cost-effective treatment option. Prior to a detailed economic analysis of this option, however, additional research is needed to determine: a) whether microbes currently identified are 325

348 capable of degrading >800 mg/l MTBE to less than 5 mg/l; b) the retention time required at various flow rates for this degradation; c) oxygen and nutrient requirements to support this degradation; and, d) the reliability of a biological system to consistently degrade MTBE independent of interferences (e.g., other organics; lag time before degradation; slow growth rate and rapid decay rate). In addition, this analysis should explore the potential difficulties of obtaining and complying with a discharge permit for a biological system. Based on preliminary economic estimates from Envirogen (Lawrenceville, NJ) (i.e., $200,000 capital expenses for a 2 gpm, 1,200 mg/l MTBE system), biological systems are not cost-effective relative to GAC and air stripping; however, detailed answers to the above questions are needed in order to verify these cost estimates. Synergistic Advantage of Resins in Combination with Alternative Treatment Process While resins have been shown to be comparable in cost to air stripping with off-gas treatment, activated carbon, and advanced oxidation, the combination of resins with these treatment processes may prove to be more cost-effective than resins alone. For example, using resins for polishing the effluent from an air stripping unit could prove advantageous because the regenerant solution could be fed back into the influent of the air stripper, thereby eliminating the need for an additional regenerant treatment option. A similar treatment process train could be evaluated using advanced oxidation processes. Alternatively, as suggested in Section 5.3.3, the combination of resins with an oxidation process, such as TiO 2 catalysis or treatment with peroxide or ozone, to destroy adsorbed organic chemicals without the need for a regeneration cycle warrants further research. 326

349 5.8 Acknowledgments This report was prepared by Malcolm Pirnie, Inc. under the direction of Dr. Michael Kavanaugh, P.E., Vice-President of Malcolm Pirnie, Inc., for the California MTBE Research Partnership and the National Water Research Institute. Andrew Stocking, P.E. was the project manager and performed the economic analysis. Amparo Flores performed the literature review and analysis. Questions regarding the content of this report should be addressed to Andrew Stocking, P.E. and Amparo Flores at Malcolm Pirnie, Inc., 180 Grand Avenue, Suite 1000, Oakland, California ; The authors would like to thank the following individuals and organizations for their contributions to this work: Academic and Industrial Research Community Prof. Susan Powers at Clarkson University (315) Prof. Mel Suffet at the University of California in Los Angeles (UCLA) (310) Prof. David Hand at Michigan Technological University (906) Prof. Eric Winkler at the University of Massachusetts Amherst (413) Dr. Joseph Salanitro and Dr. Paul Sun Equilon 3333 Highway 6 S. Houston, TX (281) David Pierce Chevron Research and Technology Co. 100 Chevron Way Richmond, CA (510) Resin Vendors and Remediation Design Firms Rohm and Haas Company 100 Independence Mall West Philadelphia, PA (215) Haley and Aldrich 189 North Water St. Rochester, NY (716) American Purification, Inc. 23 Corporate Plaza, Suite 145 Newport Beach, CA (949) Bayer 100 Bayer Rd. Pittsburgh, PA (412) Dow Chemical Company 690 Building Midland, Michigan (800) Purolite 150 Monument Road, Bala Cynwyd Philadelphia, PA (610)

350 MicroClean Services Co Vannoy Ave. Castro Valley, CA (510) Guardian Environmental Technologies P.O. Box North Main Street Kent, CT (860) Case Studies and Additional Information BP Oil Company 295 SW 41st St. Bldg. 13 Suite N. Renton, WA (425) H 2 O R 2 Consultants 653 E. Michelle St. West Covina, CA (626) Alpine Environmental, Inc. 203 W. Myrtle St., Suite C Ft. Collins, CO (970) Cost Estimate Information Carbonair, Inc Fairview Avenue E. Seattle, WA (800) Northeast Environmental Products, Inc. 17 Technology Drive West Lebanon, NH (603) Calgon Carbon La Palma Avenue, Suite H-142 Yorba Linda, CA (714) Safety Kleen 400 Market St. Oakland, CA (510) Advanced Environmental Systems 2440 Oldfield Point Rd. Elkton, MD (410) Envirogen 4100 Quaker Bridge Rd. Lawrenceville, NJ (609) Johnston-Boiler Co. 300 Pine St. Ferrysburg, MI (616)

351 5.9 References Alisto Engineering Group. Treatability Study of Methyl Tertiary Butyl Ether and Petroleum Hydrocarbons in Groundwater Treat Boulevard, Suite 201, Walnut Creek, CA AmeriPure, Inc. Personal Communication and Manufacturer Literature. 23 Corporate Plaza, Suite 145, Newport Beach, California June Baron Consulting Co. PolyGuard Hydrocarbon Absorption Media Total Absorption Capacity Testing Results for Aqueous Solution. 273 Pepe s Farm Road, Milford, CT October 2, Bayer. Manufacturer Literature. 100 Bayer Rd. Pittsburgh, PA Boodoo F. Personal Communication. The Purolite Company. 150 Monument Road, Bala Cynwyd, PA June Brendley, W., Drago, R., Petrosius, S., and Grunewald, G. Deep Oxidation of Chlorinated Hydrocarbons. Proc. Symp. On Envir. Catalysis; 205th American Chemical Society National Meeting and Exposition Program Browne, T.E. and Cohen, Y. Aqueous-Adsorption of Trichloroethene and Chloroform onto Polymeric Resins and Activated Carbon. Industrial Engineering and Chemistry Research. 29, Buscheck T.F., Gallagher D.J., Kuehne D.J., and Zuspan C.R. Occurrence and Behavior of MTBE in Groundwater. Chevron Research and Technology Company Calgon Carbon. Manufacturer Literature. 400 Calgon Carbon Drive Pittsburgh, PA Crittenden J.C., Reddy P.S., Arora H., Trynoski J., Hand D.W., Perram D.L., and Summers R.S. Predicting GAC Performance with Rapid Small-Scale Column Tests. Journal AWWA. 83(1): Crittenden J., Suri R., Perram D., and Hand D. Decontamination of Water Using Adsorption and Photocatalysis. Water Research. 31(3): Davis S.W. and Powers S.E. Alternative Sorbents for Removing MTBE from Gasoline- Contaminated Groundwater. Journal of Environmental Engineering, in press DeSilva F. J. Essentials of Ion Exchange. Water Quality Association Annual Convention and Exhibition. Nashville, TN. March

352 Dow Company. Manufacturer Literature and Personal Communication (Technical Representatives). 690 Building, Midland, MI Environmental Protection Agency (EPA). Emerging Technology Summary: Demonstration of Ambersorb 563 Adsorbent Technology. Prepared by Roy F. Weston, Inc. EPA/540/SR-95/516. August Fatula, Phil. Personal Communication. Bayer Co. 100 Bayer Rd., Pittsburgh, PA June Faust S.D. and Aly O.M. Chemistry of Water Treatment, Second Edition. Ann Arbor Press, MI Gallup D.L., Isacoff E.G., and Smith D.N. III. Use of Ambersorb Carbonaceous Adsorbent for Removal of BTEX Compounds from Oil-Field Produced Water. Environmental Progress. 15(3), Garrett, D.E. Chemical Engineering Economics. Van Nostrand Reinhold, NY Guardian Environmental Technologies (GET). Manufacturer Literature. P.O. Box 517, 25 North Main Street, Kent, CT Gregg S.J. and Sing K.S.W. Adsorption, Surface Area, and Porosity, Second Edition. Academic Press, NY Hand D.W., Herlevich J.A., Perram D.L., Crittenden J.C. Synthetic Adsorbent versus GAC for TCE Removal. Journal AWWA. 86(8), Happel A.M., Beckenbach E.H., and Halden R.U. An Evaluation of MTBE Impacts to California Groundwater Resources. Lawrence Livermore National Laboratory and UC Berkeley, prepared for the California State Water Resources Control Board UST Program, DOE Office of Fossil Fuels, and Western States Petroleum Association Hodge, Philip. Personal Communication. American Purification, Inc. 23 Corporate Plaza, Newport Beach, California August 16, Hull, David. Personal Communication. MicroClean Services Co Vannoy Ave. Castro Valley, California June/July Hunter, David - Editor. News - Top of the Week. Chemical Week. April 7, Isacoff E.G., Bortko S.M., and Parker G.R. The Removal of Regulated Compounds from Groundwater and Wastewater Using Ambersorb 563 Carbonaceous Adsorbent. Presented at the American Institute of Chemical Engineers Annual Conference, Miami Beach, FL. November

353 Isacoff, Eric. Personal communication of unpublished results. Rohm and Haas. 100 Independence Mall West, Philadelphia, PA Kong E.J. and DiGiano F.A. Competitive Adsorption Among VOCs on Activated Carbon and Carbonaceous Resin. Journal AWWA, 78(4): Litwin, Bill. Personal Communication. Guardian Environmental Technologies (GET), P.O. Box 517, 25 North Main Street, Kent, CT June/July Liu K.T. and Weber W.J. Characterization of Mass Transfer Parameters for Adsorber Modeling and Design. Journal of the Water Pollution Control Federation (WPCF). 53(10): Mace R.E. and Choi W.J. The Size and Behavior of MTBE Plumes in Texas. Proceedings of the Petroleum Hydrocarbons and Organic Chemicals in Ground Water Conference Malley J.P., Locandro R.R., Wagler J.L. Innovative Point-of-Entry (POE) Treatment for Petroleum Contaminated Water Supply Wells. Final Report. U.S.G.S. New Hampshire Water Resources Research Center. Durham, NH. September Maroldo S.G., Betz W.R., Borenstein N. U.S. Patent 4,839,331 to Rohm and Haas Company Megonnell, Neal. Personal Communication. Calgon Carbon., 400 Calgon Carbon Drive, Pittsburgh, PA June Mertz K.A., Gobin F., Hand D.W., Hokanson D.R., and Crittenden J.C. Adsorption Design Software for Windows (AdDesignS). Michigan Technological University MicroClean Services Co. Manufacturer Literature Vannoy Ave. Castro Valley, California (Montgomery) James M. Montgomery Consulting Engineers, Inc. Water Treatment Principles and Design. John Wiley and Sons, Inc National Science and Technology Council (NSTC). Interagency Assessment of Oxygenated Fuels. NSTC. Washington D.C Neely J.W. U.S. Patent 4,040,990 to Rohm and Haas Company Neely J.W. and Isacoff E.G. Carbonaceous Adsorbents for the Treatment of Ground and Surface Waters. Series: Pollution Engineering and Technology, 21. M. Dekker, NY Parker G.R. and Bortko S. Groundwater Remediation Using Ambersorb Adsorbents. Presented at the Florida Environmental Chemistry Conference. October 30 - November 1,

354 Parker G.R. Comparison of Ambersorb 563 Carbonaceous Adsorbent and Granular Activated Carbon for the removal of TCE from Water at Short Empty Bed Contact Times. Presented at the American Institute of Chemical Engineers Annual Conference. Miami Beach, FL. November Parker, G.R. Optimum Isotherm Equation and Thermodynamic Interpretation for Aqueous 1,1,2-Trichloroethene Adsorption Isotherms on three Adsorbents. Adsorption Price D.W. and Schmidt P.S. Microwave Regeneration of Adsorbents at Low Pressure: Experimental Kinetics Studies. Journal of Microwave Power and Electromagnetic Energy. 32(3), Price D.W. and Schmidt P.S. VOC Recovery through Microwave Regeneration of Adsorbents: Process Design Studies. Journal of the Air and Waste Management Association. 48: Rodriguez, Rey. Personal Communication. H2O R East Michelle Street, West Covina, California August Rohm and Haas, Inc. Manufacturer Literature. 100 Independence Mall West, Philadelphia, PA Salanitro, J.P., Spinnler, G.E., Neaville, C.C., Maner, P.M., Stearns, S.M., Johnson, P.C. and Bruce, C. Demonstration of the Enhanced MTBE Bioremediation (EMB) In Situ Process. The Fifth International In Situ On-Site Bioremediation Symposium Salinas M.J., Price D.W., and Schmidt P.S. VOC Recovery Using Microwave Regeneration of Adsorbents: Pilot-Column Studies. Air and Waste Management Association, 92nd Annual Meeting, St. Louis. June Schwarzenbach R.P., Gschwend P.M., and Imboden D.M. Environmental Organic Chemistry. John Wiley and Sons, Inc Schweiger T., Le Van M.D. Steam Regeneration of Solvent Adsorbers. Industrial Engineering and Chemistry Research. 32(10): Snoeyink V.L. Adsorption of Organic Compounds, Water Quality and Treatment. American Water Works Association. Editor: F.W. Pontius. McGraw-Hill, NY Stenzel M.H. and Merz W.J. Use of Carbon Adsorption Processes in Groundwater Treatment. Proceedings of the American Institute of Chemical Engineers Summer National Meeting, Denver, CO. Paper No. 6c. August,

355 Suffet, I.M., Shih T., Khan E., Rong W., Wangpaichitr M., and Kong J. Sorption for Removing Methyl Tertiary Butyl Ether from Drinking Water. Submitted to the University of California Toxic Substances Research and Teaching Program (UC TSR&TP) Suffet, I.M. Presentation to the MTBE Research Partnership Research Advisory Committee - unpublished results. June 13, Sun, Paul. Personal Communication of unpublished results. Equilon Enterprises, L.L.C., 3333 Highway 6 S. Houston, TX June Suri, Rominder. Personal Communication of unpublished results. Department of Civil and Environmental Engineering, Villanova University. 800 Lancaster Avenue, Villanova, PA October Suri, R., Crittenden, J, and Hand D. Removal and Destruction of Organic Compounds in Water Using Adsorption, Steam Regeneration, and Photocatalytic Oxidation Processes. Journal of Environmental Engineering. 125 (10): U.S. Filter/Westates. Personal Communication and Manufacturer Literature South Boyle Avenue, Los Angeles, California July Vandersall, M., Maroldo S., Brendley, W., Jurczyk, K., and Drago, R. Low Temperature Deep Oxidation of Aliphatic and Aromatic Hydrocarbons. Proc. Symp. On Envir. Catalysis; 205th American Chemical Society National Meeting and Exposition Program Vandiver M. and Isacoff E.G. THM Reductions with Ambersorb 563 Adsorbent. Presented at the Society of Soft Drink Technologists 41st Annual Conference, Albuquerque, NM. April, Weber, W.J. Physico-Chemical Processes for Water Quality Control. Wiley Interscience, NY Weber, W.J. and van Vliet, B.M Synthetic adsorbents and activated carbons for water treatment: overview and experimental comparisons. Journal of the American Waterworks Association (AWWA). 73(8), Winkler, E. Technology Demonstration Report: PolyGuard (DRAFT). Prepared for the Massachusetts Strategic Envirotechnology Partnership (STEP). July

356 334

357 6.0 Conclusions and Recommendations Amparo Flores Andrew Stocking, P.E. Daniel Creek, P.E. James Davidson, P.G. Rey Rodriguez Michael Kavanaugh, Ph.D., P.E. 335

358 336

359 6.1 Summary of Key Findings and Conclusions This chapter summarizes the key findings and conclusions of the technology evaluations discussed in the previous chapters. Air stripping, AOP, GAC, and synthetic resin sorbents are compared on the bases of permitting, cost effectiveness, reliability, flexibility, adaptability, and potential for modifications. This chapter concludes with recommendations for future research Air Stripping The tendency for a compound to be removed from water by air stripping is characterized by its Henry s constant. Because MTBE s Henry s constant is several times lower than those of other organic compounds commonly treated through air stripping (e.g., TCE and benzene), air stripping of MTBE is more difficult and more costly than for these other compounds. However, air stripping is a proven technology that has been used successfully to remove MTBE from drinking water. To optimize the performance of air strippers, the contact between air and water is maximized while energy costs associated with the equipment design are minimized. This optimization provides the highest rate of mass transfer between the water and air at the lowest operating cost. The most common mass transfer design for air stripping systems is randomly packed towers. Packed tower air strippers are being used successfully for drinking water treatment in La Crosse, Kansas and Rockaway Township, New Jersey. In addition to packed towers, established and emerging air-stripping technologies potentially applicable for MTBE treatment in drinking water include low profile air strippers, bubble diffusion strippers, spray towers, and aspiration air strippers. Packed tower aeration was found to be superior to the other air stripping technologies from a cost perspective, regardless of hydraulic capacity, removal efficiency requirements, or initial MTBE concentrations. At higher flow rates (>600 gpm) and higher removal efficiencies (>95 percent), packed towers are not only less expensive but, often, the only technology capable of achieving the treatment goal. However, for lower flow rates (<100 gpm), low profile air strippers become cost competitive with packed towers ($1.80 to $1.86 vs. $1.75 per 1,000 gallons treated, respectively, for 97.5-percent MTBE removal). In addition, low profile air strippers are generally easier to install, maintain, and modify for changing flow and water quality conditions than packed towers. Thus, for hydraulic capacities less than 100 gpm, which may be a remediation or a small drinking water application, a low profile air stripper is recommended. For drinking water applications requiring hydraulic capacities greater than 100 gpm, the packed tower aeration technology is recommended. A summary of the total amortized costs ($/1,000 gallons; in 1999 $) associated with packed tower and low profile air strippers is presented in Table

360 Table 6-1 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Air Stripping Technologies System Flow (gpm) Packed Tower ($/1,000 gal) Low Profile ($/1,000 gal) 90% removal 1 99% removal 2 90% removal 1 99% removal 1 60 $1.66 $1.79 $1.70 $ $0.32 $0.36 $0.85 $ $0.15 $0.17 $0.41 NE NE = not evaluated due to lack of data. System may require custom design. Costs are in 1999 dollars. 1 90% removal is for 200 µg/l influent concentration. 2 99% removal assumes 2,000 µg/l influent concentration. The evaluation in Chapter 2 assumed that off-gas treatment is required when MTBE gas phase concentrations exceed 1 lb/day. If off-gas treatment is required and MTBE influent concentrations are low (<200 µg/l), vapor phase GAC is generally the most cost-effective off-gas technology because carbon usage rates are low (as a result of the very dilute MTBE stream) and, thus, O&M costs remain low. If MTBE influent concentrations are higher (e.g., the 2,000 µg/l scenario), oxidation is the recommended technology for an air stream from a packed tower system. The cost analysis indicates a small difference in costs between catalytic and thermal oxidation. Thermal oxidation is the recommended technology to evaluate in conjunction with the selected aeration technology because, like GAC, it is a commonly used and proven technology. Both GAC and thermal oxidation demonstrate equally high levels of reliability, flexibility, removal efficiencies, and cost effectiveness. Table 6-2 lists the costs associated with the most cost-effective off-gas treatment technologies, vapor phase GAC, and thermal oxidation. See Chapter 2 for the assumptions used in the cost analysis of air stripping and off-gas treatment systems as well as more detailed cost breakdowns. Table 6-2 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Off-gas Treatment Technologies System Flow Water (gpm) Air 1 (cfm) 1,200 12, , ppmv Influent MTBE $0.54 $0.24 $0.23 Vapor Phase GAC ($/1,000 gal) 5 ppmv Influent MTBE $1.86 $1.56 $1.55 Thermal Oxidation 2 ($/1,000 gal) 0.5 ppmv Influent MTBE $1.18 $0.54 $ ppmv Influent MTBE $1.18 $0.54 $0.44 Costs are in 1999 dollars. 1 Based on an AWR of Recuperative thermal oxidation at 60 and 600 gpm and recuperative flameless thermal oxidation at 6,000 gpm. 338

361 6.1.2 AOPs AOPs destroy MTBE and other organic contaminants directly in the water through chemical oxidation, as opposed to simply transferring them from the liquid phase into a gas phase (as in the case of air stripping) or solid phase (as in the case of GAC and resins). Removal of organic compounds from water by AOPs is primarily accomplished through the reaction of organic contaminants with highly reactive hydroxyl radicals ( OH) that can be produced through a variety of mechanisms. Compared to more established drinking water treatment alternatives such as air stripping and GAC, AOPs are generally considered an emerging technology. Currently, there are several full-scale applications where organic contaminants (e.g., PCE and NDMA) are being removed from drinking water using an AOP. There are, however, no full-scale installations of AOPs for removal of MTBE from drinking water. Thus, thorough pilot- and field-scale testing of an AOP system is required before it can be implemented for MTBE removal in drinking water applications. Some of the challenges with respect to the implementation of AOPs in drinking water treatment are associated with the formation and fate of oxidation byproducts (e.g., TBA and TBF), non-selective radical oxidation, radical scavenging, and bromate formation (for ozonebased AOPs). Although it is possible to overcome these challenges, costs will increase as a result of the required greater energy usage, greater chemical dosage, and/or secondary treatment polishing steps. Chapter 3 evaluated the following established and emerging AOPs potentially applicable for the removal of MTBE from drinking water: Established Technologies Hydrogen Peroxide/Ozone (H 2 O 2/ O 3 ) Ozone/Ultraviolet Irradiation (O 3 /UV) Hydrogen Peroxide/Continuous Wave Medium-Pressure Mercury Vapor Lamps (H 2 O 2 /MP-UV) Emerging Technologies High Energy Electron Beam Irradiation (E-beam) TiO 2 -catalyzed UV Sonication/Hydrodynamic Cavitation Fenton s Reaction The two most promising AOP technologies appear to be H 2 O 2 /O 3 and H 2 O 2 /MP-UV. Both of these processes are well understood and have been demonstrated at several bench- and field-scale sites to successfully remove MTBE from water to meet drinking water standards. In addition to these two relatively well-established AOPs, E-beam and cavitation are two emerging AOPs that warrant further consideration due to their technical feasibility for removing MTBE from drinking water to meet standards. These technologies are still in their commercial developmental stages for removal of organic contaminants in drinking water applications; however, they have been widely demonstrated for disinfection and remediation applications. 339

362 Besides being the most technically feasible, H 2 O 2/ O 3 and H 2 O 2 /MP-UV in addition to cavitation appear to be the most economically feasible. However, these costs are strongly dependent on source water quality and are difficult to verify due to the untested nature of these technologies in large-scale applications. Cavitation costs involve the most uncertainty because there are no pilot, field, or full-scale drinking water treatment applications for MTBE removal. While there is significant uncertainty for all the cost estimates, H 2 O 2 /O 3 and H 2 O 2 /MP-UV technologies appear to be equivalent in cost and less expensive than the other AOPs evaluated. A summary of the total amortized costs ($/1,000 gallons; in 1999) for H 2 O 2 /O 3 and H 2 O 2 /MP-UV are summarized in Table 6-3. Table 6-3 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended AOP Technologies System Flow (gpm) H 2 O 2 /O 3 ($/1,000 gal) H 2 O 2 /MP-UV ($/1,000 gal) 90% removal 1 99% removal 2 90% removal 1 99% removal $2.65 $0.84 $0.37 $3.29 $1.07 $0.56 $2.32 $0.71 $0.42 $3.07 $1.52 $0.65 Costs are in 1999 dollars. Note: Costs do not include polishing treatment for oxidation by-products. 1 90% removal assumes 200 µg/l influent concentration. 2 99% removal assumes 2,000 µg/l influent concentration GAC Carbon adsorption technology is implemented by passing contaminated water through a vessel containing GAC. Intermolecular attraction between the dissolved organic contaminant (adsorbate) and the GAC surface (adsorbent) results in adsorptive forces that physically attract the adsorbate to the GAC as the water passes through the vessel. As such, the adsorbate remains attached to the GAC matrix while the water leaves the system with a decreased concentration. The adsorption potential of a type of GAC for a given contaminant depends on numerous factors, including the GAC structure (i.e., pore size distribution) and the physical and chemical characteristics of the adsorbate. Based on the MTBE isotherms currently available, coconut shell GAC appears to have better adsorption characteristics than coal-based GAC. Simplicity and stable operations are the primary advantages of GAC relative to other water treatment technologies. In addition, GAC is well established for removing organic compounds from water. However, field application specifically for MTBE removal in large-scale drinking water systems is currently limited. GAC systems are easy to implement due to wide commercial availability for other applications. Numerous vendors can supply GAC and the necessary equipment (e.g. vessels, piping, and pumps). Because of the simplicity of the 340

363 equipment and materials, capital and installation costs are relatively low compared to more innovative technologies. Finally, there is no off-gas treatment required for GAC systems and the creation of by-products is limited to spent carbon that can either be thermally regenerated or discarded. The effectiveness of GAC for the treatment of MTBE has been limited by its poor physical and chemical adsorption characteristics. In particular, MTBE s high solubility causes the compound to preferentially remain in solution rather than be adsorbed onto a solid surface. In addition, NOM and other SOCs compete with MTBE for the adsorption sites of GAC. Since MTBE is only weakly adsorbed by GAC, other more preferentially adsorbed SOCs in the contaminated water can result in the desorption or the displacement of previously-sorbed MTBE from the GAC matrix. As presented in Chapter 4, carbon usage rates and unit treatment costs are highly dependent on influent MTBE concentrations, background water quality, and the concentration of other SOCs. The cost analysis suggests that GAC is most cost-effective for the removal of lower MTBE concentrations, which result in lower carbon usage rates and, consequently, lower O&M costs. GAC is also more likely to be cost-effective for waters that are relatively clean with respect to NOM (e.g., some groundwaters). For example, computer modeling predicts that carbon fouling from NOM can cause approximately 50 percent increases in carbon usage rates for the removal of 20 µg/l MTBE. Finally, GAC is more cost-effective for waters contaminated solely with MTBE since other SOCs will preferentially occupy adsorption sites, thereby increasing carbon usage rates. Adsorption modeling and cost estimates show that moderate loads of total BTEX (800 µg/l) can cause greater than 50 percent increases in carbon usage rates for MTBE removal for systems treating influent with 20 µg/l MTBE. The total amortized costs ($/1,000 gallons; in 1999 $) associated with GAC systems are summarized in Table 6-4. The cost estimates developed for GAC systems are all based on non-detect (<0.5 µg/l) effluent concentrations, resulting in removal efficiencies ranging from 97.5 to >99 percent, depending on influent concentrations. These costs were developed assuming a 30-year design life. Due to GAC s high O&M to capital ratio, this technology is more likely to be cost-effective for shorter duration projects. 341

364 Table 6-4 Summary of Total Amortized Costs ($/1,000 Gallons; in 1999 $) for Recommended GAC Systems GAC Systems ($/1,000 gal)* System Flow (gpm) Influent Concentrations and Removal Efficiencies 20 µg/l (97.5%) 200 µg/l (99.75%) 2,000 µg/l (99.975%) $2.30 $0.77 $0.50 $2.92 $1.15 $0.97 $4.43 $2.37 $2.22 Costs are in 1999 dollars. *Effluent concentrations assumed to be non-detect (<0.5 µg/l) for influent concentrations of 20, 200, and 2,000 µg/l Synthetic Resin Sorbents Synthetic resin sorbents, like GAC, rely on the process of sorption to remove organic compounds from water. The primary advantages of resin sorbents over GAC is their on-site regenerability and their resistance to competitive NOM sorption. Resins used for drinking water applications could be regenerated on-site through steam stripping or microwave irradiation. Resin isotherm data for MTBE suggests that Ambersorb 563, a carbonaceous resin manufactured by Rohm and Haas (Philadelphia, PA), is currently the resin industry s most promising candidate for MTBE removal from water. Two independent studies found Ambersorb 563 to have superior sorption capacity for MTBE compared to Filtrasorb 400, a coal-based GAC widely used by the water industry. Limited data are available on the effects of background water quality toward the MTBE removal efficiency of resins. Available data suggest that the performance of Ambersorb resins is unaffected by ph (6.5 to 8.5), temperature (10 C vs. 25 C), oxidants (HOCl, H 2 O 2, and O 3 ), the presence of NOM, and the presence of TBA. Ambersorb resins have also been found to be unsusceptible to biofouling. However, m-xylene, which can be considered representative of BTEX compounds, has been found to decrease resin MTBE sorption capacity when it is present at relatively high concentrations (43.2 mg/l) (Davis and Powers, 1999). An economic evaluation of resin systems was performed for various combinations of flow rates (6, 60, 600, and 6,000 gpm), influent concentrations (20, 200, and 2,000 µg/l MTBE), and effluent goals (0.5, 5 µg/l, and 20 µg/l). In addition, a comparison between the use of two resin vessels operated in series vs. two resin vessels operated in parallel was performed. The use of steam regeneration, currently the more established regeneration method for resin systems, was also evaluated in a variety of process configurations. 342

365 For the 6,000 gpm flow scenario, parallel resin sorption followed by either low profile air stripping or GAC treatment of the regenerant was found to be the most cost effective treatment option, with costs ranging from (1999 $) $0.30/1,000 gallons (75-percent removal efficiency) to $0.58/1,000 gallons (99.98-percent removal efficiency). For all other flow rates, series operation followed by GAC treatment of the regenerant was found to be the least expensive option ranging from (1999 $) $1.02/1,000 gallons (600 gpm; 20 µg/l to 0.5 µg/l) to $24.86/1,000 gallons (6 gpm; 2,000 µg/l to 0.5 µg/l). A summary of the total amortized costs ($/1,000 gallons; in 1999 $) associated with resin systems is presented in Table 6-5. As this range indicates, for low flow rates, resin systems are impractical due to their excessive unit costs. For all cases evaluated, the cost differential between series and parallel operation was small (<16 percent) suggesting that both configurations should be carefully considered depending on site-specific conditions. These costs are highly contingent on the breakthrough times predicted by the AdDesignS model, which should be verified in the field. In addition, the costs are dependent on limited field data regarding regeneration effectiveness and resin lifetime. Costs could vary depending on site-specific conditions, especially with respect to the presence of other SOCs. For information regarding the assumptions used in the cost analysis of resin sorbents, as well as more detailed cost breakdowns, see Chapter 5. Table 6-5 Summary of Total Amortized Costs ($/1,000 Gallons) for Recommended Synthetic Resin Systems System Flow (gpm) Synthetic Resin with Regeneration System ($/1,000 gal) % Removal 1 $4.16 $1.16 $ % Removal 2 $4.56 $1.32 $0.53 Costs are in 1999 dollars. 1 90% removal assumes 200 µg/l influent concentration. 2 99% removal assumes 2,000 µg/l influent concentration. 343

366 344

367 6.2 Comparative Discussion of the Different Technologies Within each technology class, specific technology(ies) have been selected from each chapter as the preferred choice. Air Stripping Packed towers (high flow rates) Low profile (low flow rates) GAC Coconut shell-based GAC Resins Carbonaceous resin sorption, steam regeneration, GAC regenerant treatment AOP H 2 O 2 /O 3 H 2 O 2 /MP-UV The final analysis prior to selection and implementation of a specific technology is a detailed comparison between the most promising alternatives. The following sections compare the technologies with respect to their ease of permitting, reliability, flexibility, adaptability, potential for modifications, and cost effectiveness Permitting While a detailed evaluation of permitting requirements for the four technologies was beyond the scope this report, it is possible to perform a cursory analysis of permitting for the various technologies. For each technology, permitting issues can be divided into two categories: 1) start-up requirements and 2) steady-state operational requirements. As discussed in the DHS extremely impaired source water policy, permitting for start-up usually requires several months of pilot-scale demonstration with consistent contaminant removal prior to regulatory acceptance of treated effluent water as drinking water. Steady-state operational requirements encompass all the permits related to by-product formation, off-gas discharges, and regeneration issues. Air stripping has been proposed by DHS as the BAT for MTBE removal from drinking water and, thus, is expected to require the least amount of pilot-scale demonstration prior to being permitted. However, an air stripper will require an off-gas treatment or discharge permit, which may dramatically increase the permitting difficulty. In addition, the height of packed towers may increase the permitting difficulty, especially in residential neighborhoods or near schools. For example, due to the regulatory difficulty of obtaining an off-gas treatment permit and the political challenges of installing a packed tower within 1,000 feet of a school, the use air stripping for MTBE removal from drinking water was not pursued in Santa 345

368 Monica, California. Alternatively, GAC was not originally considered by DHS as a BAT, but is now being evaluated in the selection of a BAT for early GAC is expected to be easier to permit once operational because it creates no by-products or secondary contamination (e.g., air) and has been used extensively for organic contaminant removal in drinking water applications. Resin sorption and AOPs are expected to be relatively difficult to permit due to the limited use of these technologies in drinking water treatment. There are no full-scale drinking water applications using resins for MTBE removal from drinking water, nor are there systems where full-scale regeneration of resins has been performed in a drinking water application. AOPs have been used in limited cases for drinking water treatment, although no applications are for MTBE removal. Consequently, extended start-up pilot-scale testing will be required to demonstrate consistent MTBE removal before either of these technologies will be approved for delivery of potable water to consumers. Once permitted for operation, AOP steady-state operational permitting is expected to be difficult due to the formation of oxidation by-products and the presence of residual oxidant concentrations. Similarly, the use and operation of on-site steam regeneration for resins and the treatment of the regenerant are expected to incur strict regulatory scrutiny for large-scale resin applications Reliability The reliability of a technology can be divided into two components: mechanical reliability and process reliability. Technologies with fewer moving parts are considered to be more mechanically reliable because they have fewer parts subject to wear and tear and, thus, are less susceptible to mechanical failures and require less frequent maintenance work (e.g., lubrication, replacements of seals, etc.). Minimizing the number of other parts required for operation (e.g., monitoring and control systems) also improves system mechanical reliability. GAC involves no moving parts other than a pump and, therefore, is expected to be the most mechanically reliable of the technologies evaluated. Air stripping (both low profile and packed tower) is also highly mechanically reliable because it relies only on a pump and blower for MTBE removal. Air stripping may also necessitate off-gas treatment; however, even with this ancillary treatment step, it still involves fewer moving parts than resins or AOPs. Resin sorption and AOPs are the least mechanically reliable due to the large number of additional parts (e.g., steam generators, ozone diffusers), chemical additions (e.g., ozone, peroxide), and ancillary technologies (e.g., super-adsorber columns, regenerant disposal equipment, ozone off-gas destruction equipment) required for MTBE removal. However, it should be noted that within each of these general classes of technologies, such as AOPs, specific technologies can vary in their degree of mechanical reliability. For example, APT s H 2 O 2 /O 3 process is expected to be as mechanically reliable as a low profile air stripper. 346

369 Process reliability is evaluated based on the ability of a technology to consistently produce effluent that meets the necessary drinking water standards. Sorption technologies, such as GAC and resins, are expected to be highly reliable for achieving consistently high removal efficiencies of MTBE since these technologies are typically designed to remove contaminants to non-detectable levels. A single packed tower air stripper has been demonstrated to consistently achieve greater than 95 percent MTBE removal at higher flows (i.e., 600 to 6,000 gpm), while a single low profile air stripper is capable of high removal efficiencies for MTBE at lower flows (<100 gpm). Bench-scale testing suggests that AOPs will also be reliable; however, further studies at higher flow rates need to be conducted for verification of consistent MTBE removal Flexibility Occasionally during the operation of a treatment process, the influent flow rates increase or decrease significantly compared to the design flow rate. A flexible technology is able to handle these fluctuations with no major impact on the treatment process outcome. Each of the selected treatment technologies can handle a fluctuating flow rate, both higher flow rates or lower flow rates than expected, while maintaining process reliability. However, some modifications or adjustments may be necessary. Thus, to compare technologies, one can look at the relative decreased costs when the flow rate falls and increased costs as the flow rate raises. For example, as the flow rate decreases, adsorption technologies can be operated longer prior to regeneration or carbon change-out, thereby reducing the O&M cost of these technologies. Similarly, decreasing the amount of chemical additions during lower flow rate periods can reduce the cost for AOPs. For an air stripper, costs can be lowered by maintaining a constant AWR (i.e., as the water flow rate falls the air flow rate can also be decreased to maintain a constant AWR). However, not all air stripper blowers are capable of turn down, (e.g., low profile units have minimal turndown) and, thus, air stripping represents the technology for the lowest potential cost savings with decreasing flow rates. As the flow rate increases, operation of each technology will require modification or adjustment to maintain constant removal efficiencies. Sorption technologies will experience faster breakthrough and, thus, require more rapid carbon or resin regeneration. AOPs will require more chemical additions. Air stripping will require an increase in the air flow rate by either turning up the blower or installing a larger blower. If the flow rate increase beyond the design flow rate, each technology will require additional units operated in parallel Adaptability Adaptability of a technology is defined as its ability to handle fluctuations in water quality conditions, such as influent contaminant concentrations, hardness, alkalinity, and turbidity. Based on the literature review, the performance of air stripping and resins are least affected by background water quality. While air stripping can be adversely affected by the presence 347

370 of scaling agents, both of these technologies are unaffected by the presence of other contaminants or changing background water quality to achieve consistent MTBE removal from drinking water. Alternatively, the presence of other contaminants or NOM can significantly reduce MTBE removal efficiencies for GAC due to the competitive desorption of target contaminants. Similarly, high concentrations of alkalinity, bromide, and TOC in source water, among other factors, can decrease the effectiveness of AOPs. If influent contaminant concentrations change and the effluent goal remains unchanged, treatment costs will rise and fall proportionally to the influent concentration. For increasing influent concentrations, the carbon or resin usage rate will increase, the AOP chemical dose will increase, and the AWR for an air stripper will increase. In general, sorption and AOPs can handle fluctuating contaminant concentrations better than air stripping because of the difficulty in turning up the blower power. Similarly, if contaminant concentrations decrease, air stripping is probably the most difficult technology to turn-down, for the same reasons discussed in Section for a decrease in flow rates Potential for Modifications The potential for modifications is defined as the operator s ability to alter the installed system, including addition of any necessary pre- and post-treatment systems, to accommodate changes in the design criteria and conditions. To comply with regulations, each of the technologies discussed in the report requires a redundant treatment system prior to drinking water distribution. This redundant treatment system is effectively a post-treatment polishing system. All of the technologies can handle this system or any pre-treatment systems with similar ease. However, some of the systems may require pre- or post-treatment systems prior to regulatory acceptance and, thus, these technologies should be rated lower than those technologies which can stand alone. GAC is the only stand-alone technology, requiring no pre- or post-treatment systems for drinking water applications, other than the redundant carbon vessel. Air stripping may require an off-gas treatment unit, depending on local air quality restrictions. AOPs will require ozone off-gas treatment units, hydrogen peroxide catalytic oxidizers, and/or a polishing filter to remove oxidation by-products. Finally, resin systems require regeneration units, super-adsorber columns, and other equipment for on-site regeneration and re-use of the resins Cost Effectiveness Detailed cost estimates have been completed for each of the four technologies evaluated in this report; however, these costs are intended for comparative purposes only and should not be used in place of a detailed site-specific engineering cost analysis. Furthermore, it may be difficult to compare costs between technologies due to differences in vendor assumptions and technology effectiveness. For example, all technologies can remove MTBE 348

371 concentrations to below drinking water standards; however, AOPs create oxidation byproducts, which may cause effluent water quality to exceed regulatory thresholds. Keeping these complicating factors in mind, a summary of the total amortized costs ($/1,000 gallons; in 1999 $) for the various treatment technologies is presented in Table 6-6. A detailed description of assumptions used is provided in each respective chapter. A direct comparison of these costs indicate that, under the assumptions stated for the cost analysis of each technology, packed tower aeration with off-gas treatment (when required) is generally the most cost-effective treatment option. However, packed tower aeration is expected to be ineffective for cases where very high removal rates are required. In these cases, H 2 O 2 /O 3 systems become the most cost-effective option for small and medium-sized systems (60 and 600 gpm, 2,000 µg/l to 0.5 µg/l, percent removal). For large systems (6,000 gpm), resin sorbent systems appear to be the most cost-effective option for both low (20 µg/l) and high (2,000 µg/l) concentrations. For medium concentrations (200 µg/l), H 2 O 2 /O 3 systems are most cost-effective for lower removal efficiencies (90 percent) while packed towers are most cost-effective for higher removal efficiencies (97.50 and percent). However, it is important to note that the differences in unit costs amongst 6,000 gpm systems for these technologies (packed tower aeration with off-gas treatment, H 2 O 2 /O 3 systems, and resin sorbent systems) are generally in the order of a few cents per thousand gallons. Considering the uncertainty in these cost estimates, all three technologies can be considered equivalent in costs for a large-scale system. 349

372 Table 6-6 Summary of Total Amortized Costs ($/1,000 Gallons) for Various Recommended Treatment Systems Flow (gpm) Influent (µg/l) Effluent (µg/l) Packed Tower AIR STRIPPING AOPs Removal GAC Low Profile Packed Tower w/ OGT H2O2/MP- UV O3/H2O2 RESIN SORPTION Lowest Unit Cost Amongst the Technologies Evaluated % $1.66 NE NR $2.18 $2.63 NE $2.50 Packed Tower % $1.75 $1.86 NR $2.50 $2.68 $2.30 $2.81 Packed Tower % $1.66 $1.70 NR $2.32 $2.65 NE $4.16 Packed Tower % $1.75 $1.80 NR $2.50 $2.68 NE $4.16 Packed Tower % $1.82 $1.89 NR $2.72 $2.98 $3.10 $4.16 Packed Tower % $1.79 $1.90 $3.08 $3.07 $3.29 NE $4.56 Packed Tower with OGT % $1.82 $2.02 $3.20 $3.47 $3.31 NE $4.57 Packed Tower with OGT % NE NE NE $4.11 $3.62 $4.61 $4.57 O3/H2O % $0.30 $0.78 NR $0.57 $0.82 NE $1.01 Packed Tower % $0.34 $0.92 NR $0.91 $0.90 $0.77 $1.01 Packed Tower % $0.32 $0.85 $0.57 $0.71 $0.84 NE $1.16 Packed Tower with OGT % $0.34 $0.96 $0.59 $0.96 $0.90 NE $1.17 Packed Tower with OGT % $0.37 $1.09 $0.62 $1.27 $0.95 $1.15 $1.17 Packed Tower with OGT % $0.36 $0.96 $0.90 $1.52 $1.07 NE $1.32 Packed Tower with OGT % $0.37 $1.09 $0.91 $1.75 $1.13 NE $1.36 Packed Tower with OGT % NE NE NE $2.08 $1.19 $2.37 $1.38 O3/H2O % $0.13 $0.34 $0.36 $0.32 $0.35 NE $0.30 Resin Sorption % $0.16 $0.48 $0.39 $0.52 $0.43 $0.50 $0.36 Resin Sorption % $0.15 $0.41 $0.38 $0.42 $0.37 NE $0.39 O3/H2O % $0.16 $0.48 $0.39 $0.60 $0.43 NE $0.41 Packed Tower with OGT % $0.17 $0.64 $0.40 $0.74 $0.48 $0.97 $0.41 Packed Tower with OGT % $0.17 NE NE $0.65 $0.56 NE $0.53 Resin Sorption % $0.18 NE NE $1.24 $0.59 NE $0.54 Resin Sorption % NE NE NE $1.59 $0.68 $2.22 $0.58 Resin Sorption 6, gpm 600 gpm 60 gpm 60 gpm Costs are in 1999 dollars. 1 NE = not evaluated due to lack of data. 2 NR = off-gas treatment not required. 3 Boxed numbers indicate the lowest unit cost amongst the technologies evaluated. 4 Air stripping costs are italicized and underlined when off-gas treatment is expected to be required based on 1 lb/day emission standards. 5 OGT = off-gas treatment. 6 AOP Treatment Costs for by-product and residual oxidant removal not included. 350

373 6.3 Recommendations for Future Research In general, there are only a few full-scale installations of MTBE treatment systems for drinking water treatment. Therefore, for each technology, there are a number of areas that are still lacking information critical for effective implementation in MTBE applications. Recommendations for further research to address these areas of uncertainty are discussed below Air Stripping Air stripping is a well-understood technology with many full-scale installations across the country. However, other than the packed tower air strippers at LaCrosse, Kansas and Rockaway Township, New Jersey, there appears to be a lack of published data for air stripper applications for MTBE removal from drinking water. Consequently, there is a need to collect cost and other operational data from a variety of air stripping sites to better evaluate the applicability and cost effectiveness in MTBE treatment scenarios. Cost data should include both real capital and O&M costs. Operational data should include influent concentrations, removal efficiencies, air and water flow rates, off-gas concentrations, and costs AOPs AOPs are not as well understood as air stripping and adsorptive processes due to the large number and variety of chemical and physical processes involved in advanced oxidation reactions. Thus, there remains a significant amount of uncertainty regarding the technical and economic effectiveness of AOPs for removing MTBE from drinking water under a variety of water quality scenarios. More pilot- and field-scale studies need to be conducted to determine the removal efficiencies that can be achieved under different water quality conditions and operational parameters. In addition, the following specific topics warrant further research: 1) Water quality impacts on AOP effectiveness. The effectiveness of AOPs is directly related to water quality parameters such as ph, alkalinity, NOM, TOC, turbidity, and concentrations of other interfering compounds (e.g., nitrates and bromide). Future studies on AOP treatment of MTBE must independently evaluate the impact of each of these water quality parameters. The evaluation criteria must include MTBE removal efficiency, oxidation by-product formation, disinfection by-product (DBP) formation potential, and costs. For ozone-based AOPs, the effect of influent bromide concentration on bromate formation must also be evaluated. Similarly, a detailed analysis of the effect of influent water turbidity and nitrate concentrations on the effectiveness of AOPs relying on UVlight (LP, MP, pulsed) is warranted. 2) By-product formation and control. In addition to more testing and demonstration, one of the most significant areas of future research is the issue of by-product formation and control. The oxidation of MTBE to carbon dioxide and water involves many steps and the 351

374 formation of many oxidation by-products (e.g., TBA, TBF, acetone). If these by-products are not completely mineralized, they will be present in treated water, resulting in elevated concentrations of potentially toxic by-products in the treated water. A better understanding of by-product formation mechanisms and subsequent mitigation strategies will be necessary prior to the acceptance of AOPs by the regulatory community for drinking water applications. This includes research to determine the most cost-effective treatment option, such as biologically activated carbon, for by-product removal in drinking water applications. 3) Cost evaluation as a function of water quality and contamination scenario. Finally, future research should evaluate engineering costs for MTBE oxidation by AOPs. Capital and O&M costs for each AOP process should be developed as a function of water quality, flow rate, influent MTBE concentration, and required removal efficiency. These cost evaluations must be performed under uniform design criteria (e.g., required removal efficiency) and operational assumptions (e.g., power rate). A unified costing approach will enable a direct comparison of the various AOPs for specific water qualities GAC Based on the literature review, the results of the computer modeling, and the cost analyses, there are several topics that require more research before GAC usage for MTBE removal from drinking water is fully understood. These topics are: 1) Reproducible Isotherms. Standardized testing should be performed to obtain comparable and reproducible isotherms for a range of GAC types, including high-grade coconut shell-based carbon and coal-based carbon. 2) Dynamic GAC Adsorption Capacities. Dynamic column tests should be performed to determine GAC usage rates, optimum EBCTs, and other operating parameters for a variety of MTBE influent concentrations, background water quality conditions, and GAC types. These tests will allow for better prediction of full-scale performance of GAC systems for MTBE removal. In addition, more information is needed on MTBE desorption from GAC and on the competitive effects of other SOCs (e.g., BTEX, TBA) in the source water. 3) Full-scale Performance. To date, there are limited data regarding the successful use of fullscale GAC systems for removing MTBE from drinking water. As such, it is recommended that GAC performance for MTBE removal be evaluated under full-scale field conditions. Collection of cost and operational data, including long-term NOM fouling effects and pretreatment requirements, will allow for meaningful comparison with results of dynamic column testing and cost estimates. Currently, a 400-gpm GAC system is planned for installation in Santa Monica, California. This system will provide the California MTBE Research Partnership with financial and performance data from pilot-scale testing. 352

375 6.3.4 Resins There are four primary areas where additional information is critical to design a costeffective resin treatment system for MTBE removal from water: 1) Dynamic Resin Adsorption Capacities. Dynamic column tests should be performed in order to determine the optimum EBCT and other operating parameters for treating MTBEimpacted groundwater under varying background water quality conditions, such as temperature, ph, and the presence of other SOCs and NOM. In addition, more information is needed on the adsorptive capacities of various resins for TBA, BTEX, and NOM and their interference with the adsorption of MTBE. 2) Mechanisms of Regeneration and Optimization of Regenerative Processes. To date, only a limited number of bench- and pilot-scale studies have been completed to investigate the effectiveness of steam and microwave regeneration for MTBE applications. Further information on the feasibility and economics of these regeneration processes is needed to allow for a detailed cost evaluation and comparison. 3) Effectiveness of Biological Degradation as a Regenerant Treatment. Further research is necessary to determine the most cost-effective treatment for the concentrated MTBE solution created by regeneration processes. In particular, biological degradation of concentrated regenerant solutions should be evaluated as a potentially cost-effective treatment option. 4) Synergistic Advantage of Resins in Combination with Alternative Treatment Process. The potential economic advantages of combining resin systems with the other technologies evaluated in the report should be investigated. 353

376 354

377 Appendices

378

379 Appendix 2A-1 Cost Assumptions for Packed Tower Aeration Vendor: Verified by: Capital costs for packed tower air stripping systems were originally provided by the Layne Christensen Company (Bridgewater, NJ) in To verify the costs, quotes were also obtained from Carbonair Environmental Services (New Hope, MN), Branch Environmental (Somerville, NJ), and Delta Cooling Towers (Fairfield, NJ). These cost estimates were verified again by Layne Christensen Company (Bridgewater, NJ) in In addition, Layne Christensen Company provided actual capital costs for sold equipment that supported the accuracy of these cost estimates. Assumptions given to vendors: Flow rate: 60, 600, 6,000 gpm Influent concentration: 20, 200, 2,000 µg/l Influent water temperature: 65 F ph: 8.0 Alkalinity: 250 mg/l as CaCO 3 Hardness: 250 mg/l as CaCO 3 Iron: <1 mg/l TDS: 750 mg/l Assumptions made by vendors: Liquid loading rate: 10.0 to 12.0 gpm/ft 2 AWR ratio: 200:1 Henry s constant (dimensionless) = Aluminum tower construction Packing Medium: 2-inch Jaeger Tri-Pack 357

380 Calculations made by vendors: Table 2-A1 Packed Tower Aeration Assumptions Flow (gpm) Removal Rate (%) Packing Height (ft) Power Requirement (HP)

381 Appendix 2A-2 Cost Assumptions for Low Profile Aeration Vendor: Verified by: Capital costs for low profile air stripping systems were originally provided by North Eastern Environmental Products (West Lebanon, NH) in To verify the costs, quotes were also obtained from Carbonair Environmental Services (New Hope, MN). These cost estimates were verified again by North Eastern Environmental Products (West Lebanon, NH) in In addition, North Eastern Environmental Products provided actual capital costs for sold equipment that supported the accuracy of these cost estimates. Assumptions given to vendors: Flow rate: 60, 600, 6,000 gpm Influent concentration: 20, 200, 2,000 µg/l Influent water temperature: 65 F ph: 8.0 Alkalinity: 250 mg/l as CaCO 3 Hardness: 250 mg/l as CaCO 3 Iron: <1 mg/l TDS: 750 mg/l Assumptions made by vendors: AWR: see the following table. Influent air temperature: 50 F Henry s constant = cost estimates were based on empirical data, not theoretical calculations (i.e., NEEP generated a performance curve at 1 gpm, 5 gpm, 15 gpm, and 45 gpm for various removal efficiencies and influent concentrations [maximum influent of 5 mg/l] and extrapolated for other flow rates and removal efficiencies). Safety Factor = 1 Calculations made by vendors: The footprint of a single unit, including the blower, ranges from 5.5 feet by 6.2 feet for low flow models of the 3600 series used at 60 gpm to 7.5 feet by 12.5 feet for the series models used for 6000 gpm flows. 359

382 Table 2A-2 Low Profile Aeration Assumptions 360 Flow (gpm) Influent (µg/l) Removal Rate (%) Air/Water Ratio Number of Trays Number of Units HP per Unit

383 Appendix 2B 9/1/98 La Crosse, Kansas Trip Report Introduction The water treatment plant for La Crosse, Kansas is one of only two known large-scale facilities in the country treating public drinking water for MTBE removal and delivering the water to their customers. On September 1, 1998, four members of the MTBE Research Partnership met with representatives from the State of Kansas and their consultant to learn about the La Crosse situation and share information about MTBE occurrence and treatment. La Crosse is located along Highway 183, about 20 miles south of Hays, Kansas. In attendance: Todd Anderson, Santa Clara Valley Water District Jim Davidson, Alpine Environmental, Inc. Dave Smith, ARCO Andy Stocking, Malcolm Pirnie, Inc. Craig Hofmeister, Handex Phil Koontz, State of Kansas Department of Health and Environment Bill Reetz, State of Kansas Department of Health and Environment Background La Crosse operates a drinking water treatment plant for water softening and disinfection. Groundwater is the only water source for the plant; one source well is located 8/10 of a mile east of the town of Rush Center; the other is 1/10 of a mile further east. Rush Center is located approximately 5 miles south of the La Crosse treatment plant. Hydrated lime softening reduces the water s hardness from about 700 to 110 mg/l (as calcium carbonate). Chlorine is applied at wellheads and at the plant to maintain a free chlorine residual of between 0.5 and 1.0 mg/l in the finished water; this residual is typically 0.7 mg/l. The plant is only run during the day, and treats between approximately 300 gpm (winter) and 450 gpm (summer). In May 1996, a nearby resident noticed chemical-like odors from his irrigation well water. A sample was sent to a laboratory for analysis; MTBE was found and was acknowledged to be the source of the odors. One of La Crosse s nearby water supply wells was sampled; this water contained about 200 µg/l of MTBE. Shutting down any wells was not an option for La Crosse because of water supply needs. A shallow tray air stripper was quickly installed at the plant as a temporary measure; a 40 percent reduction in MTBE concentration was achieved at 250 gpm. A permanent packed tower air stripping system was subsequently installed at the plant, and has been generally successful in treating the influent MTBE to 361

384 below 5 µg/l (non-detectable concentrations using a detection limit of 0.2 µg/l have been reported, but this is probably an unrealistically low detection limit). Other than the irrigation well owner, no residents have reported MTBE-related taste, odor, or other problems. The drinking water treatment plant has incorporated the air stripping towers into its treatment sequence and operations and maintenance program, and continues to deliver drinking water to its customers. MTBE Source Investigation Three UST sites have been identified as the sources of MTBE. These are located on three corners of the main intersection in Rush Center, Kansas, about 5 miles south of the La Crosse plant. The closest of the two affected water supply wells (Well #5) is located about 8/10 mile due east of the service station cluster. The next well is about 1/10 mile further east. Contamination from leaking USTs at each of the three sites has resulted in a comingled total petroleum hydrocarbon (TPHg) and BTEX plume, which has been undergoing investigation for a number of years. Contamination immediately underneath the sites is fairly high and free product has been present. Until sampling was conducted for MTBE, the extent of groundwater contamination was thought to be fairly well defined with numerous monitoring wells, and the contamination appeared to reach about 800 feet downgradient (east). All three stations have had the leaking tanks removed or replaced. MTBE was not thought to be a contaminant of concern in the investigation of the three UST sites until after it was noticed in the nearby irrigation well and the La Crosse supply well. MTBE has been added to local gasoline since 1979 as an octane booster, and since the early 1990s as an oxygenate. MTBE is reportedly the oxygenate of choice in about 80 percent of Kansas gasoline. MTBE was found in site monitoring well samples in concentrations as high as 77,000 µg/l. La Crosse s water wells are screened at about 50 to 70 feet bgs. A similarly screened monitoring well was installed about halfway between the UST sites and Well #5, and produced a 1,600 µg/l sample. This well was installed following MTBE being noticed in the irrigation and municipal wells and confirmed the concern over the MTBE presence in the plume. A shallow monitoring well installed in the same area produced water with about 5 µg/l of MTBE. Geologic Setting and UST Site Remediation The UST sites are located in a locally low topographic area, near an existing creek, and likely over a buried stream channel. Silts and clays are generally present to 25 feet bgs, and sands are generally present from 25 feet bgs to bedrock around 70 feet bgs. The high permeability and preferred pathway of the buried stream channel, as well as the easterly groundwater gradient, promoted the MTBE migration to the water supply wells. 362

385 Since heavy rains in 1993, groundwater levels in the area have remained about 10 feet higher than in previous years. Groundwater has risen over the zone with previously detected free product, and has submerged the screens of the soil vapor extraction wells on site. This has delayed the remediation plan for the three sites, which originally consisted of soil vapor extraction and groundwater pump-and-treat. Shallow tray air stripping, carbon adsorption, and soil vapor extraction equipment are currently sitting unused at the site. The stripping unit was temporarily relocated to the drinking water treatment plant and operated for a number of months there when MTBE was first noticed in the water. It was replaced by the two air stripping towers and was returned to the service station remediation area. The site remediation system was originally designed for just gasoline contamination, but will hopefully remove MTBE from the subsurface as well. MTBE that escapes the on-site remediation system will continue to be addressed by the point-of-use treatment at the plant s air strippers. The UST site occupying the southeast corner of the Rush Center intersection has been excavated to a depth of 20 feet to remove contaminated soil; air sparging in conjunction with soil vapor extraction is planned for this site. Two Oxygen Releasing Compound (ORC) lines were installed in May of 1997 in an attempt to decrease the concentration of BTEX and MTBE at the site and the concentration of MTBE in downgradient wells, including the La Crosse municipal wells. The first ORC line was installed, bisecting the BTEX and MTBE plume near the downgradient third of the plume. ORC was injected from 20 to 70 feet below grade, including the entire saturated zone. The second ORC line was installed approximately 900 feet upgradient of the municipal wells. ORC was tremied in the geoprobe borings for this installation. Quarterly monitoring following the ORC applications indicated a significant increase in dissolved oxygen concentrations downgradient of the ORC lines. BTEX and MTBE concentrations also declined in wells located downgradient of the ORC lines. However, recent monitoring results indicate that BTEX and MTBE concentrations in these wells are again beginning to increase. MTBE Treatment After MTBE was first identified in the drinking water, a low profile shallow tray air stripping system was used to treat the water prior to distribution. Water was pumped to a clear well, pumped through the five-tray stripping system, and then discharged back into the clear well. The residence time in the clear well was such that the water could run through the stripping system several times prior to distribution. As stated previously, this system was only capable of achieving a 40 percent reduction in influent MTBE concentrations (ranging from 200 µg/l to 500 µg/l) at the La Crosse plant. The currently operating permanent MTBE treatment system at La Crosse s water treatment plant consists of two 35-foot tall (6-feet diameter) packed tower air strippers. Each contains about 30 feet of 2-inch Jaeger Tripacks; the towers are operated in series. Water is pumped 363

386 to the strippers following lime addition, flocculation, and settling. Effluent from the strippers is returned to the conventional treatment system and piped to the sand/anthracite filters. The consultant for the State of Kansas claims that the operating air/water ratio in each of the two stripping towers is 175. No off-gas treatment is required or performed; MTBE odors are not evident at any distance from the top of the towers, although they are strong at the immediate top of the towers. Plant operators report that the stripping system does not require significant operations and maintenance time; there have not been problems with fouling or scaling to date. The towers are operated from 8 am to 4 pm, 6 days/week. There are three 10-horsepower pumps to pump water to the top of the two towers and back to the filtration unit, and two 15-horsepower blowers used for the packed tower system. The air stripping towers were designed by Process Equipment and Engineering (PEE) of Lakeland, Florida. PEE was selected because they provided the lowest bid of the four contractors chosen for quotations. Originally, PEE had stated that only one air stripper was needed for complete removal of MTBE to below 10 µg/l (the treatment goal); however, it has been necessary to use both towers in series to reach the goal. During summer operation, the first tower achieves approximately 90 percent removal and the second tower generally reduces the first tower s effluent concentration by an additional 90 percent. These two numbers are each about 80 percent during the winter. Water temperatures are lower during the winter due to the approximately 1 hour residence time of the influent water in the settling basin prior to treatment in the air strippers. Therefore, this may affect the removal percentages. However, the difference in removal percentages is mostly due to the large difference in air temperature between the seasons; ambient air is used in the stripping towers. Each tower was designed to handle a maximum water flow rate of 500 gpm and the blowers were designed to circulate 11,500 cubic feet/minute of air through each tower. The tower operator stated that the blowers were not turned down. This implies that the actual air/water ratio fluctuates between approximately 183 in the summer (at a flow rate of 470 gpm) and 287 in the winter (at a flow rate of 300 gpm). Well Operation and Sampling Well #5 (~500 gpm) is only operated 1 day per week (Saturday) to ensure that it remains operational. Well #4 (~500 gpm) is used during the rest of the week because its MTBE concentrations are generally lower. Sampling for MTBE at the plant is also performed once a week (Tuesdays). Influent concentrations generally range from 100 to 200 µg/l for well #4 and 400 to 600 µg/l for well #5; effluent concentrations are nearly always below 10 µg/l. Public and Regulatory Relations Working with the public and regulatory agencies has been a smooth process. A public meeting was advertised and held in La Crosse soon after the MTBE problem was discovered. Residents were informed of the situation and the system that was planned to treat the water. There has been little public or press interest in the treatment system or MTBE contamination 364

387 since the beginning, as evidenced by the lack of public comments and the low turnout at the public meeting (one press person and two residents attended). There have also been no obstacles in dealing with regulatory agencies. Several homes are located between the supply wells and the La Crosse treatment plant, and sometimes receive direct well water. Since chlorine is applied at the wellhead, water quality (other than hardness) has historically not been a concern for these residents. Since the MTBE discovery, the State of Kansas has installed and is maintaining carbon adsorption treatment systems at each of these homes for MTBE removal. These systems consist of two carbon canisters in series; spent carbon is changed out about once every 6 to 9 months. Samples to confirm the treatment systems effectiveness were initially collected every month and are now collected once each quarter. Effluent samples from these systems have generally not contained detectable concentrations of MTBE. Because these residents are located upstream of the treatment plant, they cannot get the benefit of the large air stripping units at the water treatment plant during daytime use (when a well is pumping and they are receiving water directly from the well). When the plant is not operating, the flow is reversed and the pipeline feeding their houses is back-pressured with treated water from the plant s clear well. The residents can then fill small individual storage tanks with the treated water. This water can be used as a daytime source of water. Taste and Odor Although there have been no reports from drinking water customers about adverse tastes or odors from La Crosse water, the presence of MTBE can be noticed. Prior to stripper installation, MTBE odors were occasionally evident to water treatment plant operators during filter backwashes. The taste of MTBE was also noticed in nearby restaurant drinking water by State of Kansas representatives during this time. When the shallow tray air strippers were operated at the Rush Center intersection as part of the UST plume remediation, there were odor complaints. No odor complaints have resulted from the air stripping operations at the La Crosse drinking water treatment plant. Most individuals can detect the taste and/or odor of MTBE in untreated plant influent water. On the day of our visit, the influent MTBE concentration was 108 µg/l and the taste of MTBE was evident in influent water samples when drawing air through the mouth while tasting. Strong MTBE odors are apparent in the off-gas at the tops of the two air stripping columns. On the day of our visit, effluent water samples from the first air stripping column and the second air stripping column reportedly contained 15.2 and 3.03 µg/l of MTBE, respectively. Follow-up The Partnership members and the Kansas representatives agreed to continue to share information regarding their experiences with MTBE. Partnership members were particularly 365

388 interested in ongoing air stripping performance data and MTBE soil vapor sampling results (if and when the sparging/vapor extraction system is operated at the UST cluster). Kansas representatives were presented with current copies of the Partnership s source protection and treatability documents. The Kansas consultant was encouraged to publish a peer-reviewed article on the experiences at La Crosse, and the Partnership members expressed a willingness to discuss the possibility of helping to fund the preparation of such a report. 366

389 Appendix 2C Air Stripping Equipment Vendors 1. Packed Tower Air Strippers QED Environmental Systems, Inc. P. O. Box 3726-T Ann Arbor, MI Toll Free: (800) Telephone: (313) Fax: (313) Layne Christenson Company (Formerly HydroGroup, Inc.). 97 Chimney Rock Road Bridgewater, NJ Toll Free: (800) Fax: (732) Delta Cooling Towers, Inc. 134-T Clinton Road Fairfield, NJ Toll Free: (800) Buy-Delta Telephone: (201) Fax: (201) Carbonair Environmental Systems, Inc Nevada Avenue North New Hope, MN Telephone: (612) Fax: (612) Tonka Equipment Company P.O. Box Watertower Circle Plymouth, MN Telephone: (612) Fax: (612) EPG Companies Inc. P.O. Box 224 Maple Grove, MN Telephone: (612) Fax: (612) Duall Industries 1550-TR Industrial Drive Owosso, MI Telephone: (517) Fax: (517) Spray Tower Strippers Branch Environmental Corporation P.O. Bax Route 22 East Somerville, NJ Telephone: (908) Fax: (908) GDT Corporation N 19 Av Phoenix AZ Telephone: (602) Fax: (602) Bubble Aeration Air Strippers The Stripper Lowry Engineering, Inc. P.O. Box 189 Unity, ME Telephone: (207) Fax: (207)

390 Remtech Bubble Lance Low-Profile Stripper Remtech Engineers Whitewater Business Center 200 North Cobb Parkway Building 100, Suite 124 Marietta, GA Toll Free: (800) Telephone: (770) Fax: (770) BREEZE Compact Air Stripper AEROMIX Systems, Inc N. Second Street Minneapolis, MN Toll Free: (800) Telephone: (612) Fax: (612) Carbonair Environmental Systems, Inc Nevada Avenue North New Hope, MN Telephone: (612) Fax: (612) Low Profile Strippers Shallowtray Low Profile Air Stripper North East Environmental Products, Inc. 17 Technology Drive West Lebanon, NH Telephone: (603) Fax: (603) Carbonair Environmental Systems, Inc Nevada Avenue North New Hope, MN Telephone: (612) Fax: (612) Carbtrol Corporation 51 Riverside Avenue Westport, CT Toll Free: (800) Telephone: (203) Fax: (203) EZ Tray Strippers QED Environmental Systems, Inc. P.O. Box 3726 Ann Arbor, MI Telephone: (313) Fax: (313) Lowry Engineering, Inc. P. O. Box 1239 Blue Hill, ME Toll Free: (800) Telephone: (207) Fax: (207) Northeast Environmental Services, Inc. Maguerite Drive West Canastota, NY Telephone: (315) Fax: (315) Geotech Environmental Equipment, Inc East 40th Avenue Denver, Colorado Telephone: (303) Fax: (303) Aspiration Strippers Maxi-Strip System Hazleton Environmental 125 Butler Drive Hazelton, PA Telephone: (717) Fax: (717)

391 Appendix 2D Off-gas Treatment Equipment Vendors 1. Thermal and Catalytic Oxidizers Flameless Thermal Oxidizer Thermatrix Inc. 101 Metro Drive, Suite 248 San Jose, California Telephone. (408) Fax. (408) Regenerative Thermal/ Recuperative Cataltic Oxidizers Advanced Environmental Systems 2440 Oldfield Point Road Elkton, MD Toll Free: (800) Telephone: (410) Fax: (410) Durr Environmental, Inc Sheldon Rd., Suite 300 Plymouth, MI Telephone: (734) Fax: (734) Megtec Systems 830 Prosper Road P.O. Box 5030 De Pere, WI Telephone: (920) Fax: (920) HiTemp Technology Corporation PO Box 903, Flemington NJ Toll Free: (800) Telephone: (908) Fax: (908) ABB Air Preheater Inc PO Box 372 Wellsville, NY Telephone: (716) Fax: (716) Catalytic Products International 980 Ensell Road Lake Zurich, IL Telephone: (847) Fax: (847) North American Manufacturing Company 4455 East 71st Street Cleveland, OH Telephone: (216) Fax: (216) CVM Corporation 402 Vandever Avenue Wilmington, DE Telephone: (302) Fax: (302) Carbon Adsorbers and GAC Nixtox Adsorbers Tigg Corporation 800 Old Pond Road, Suite 706 Bridgeville, PA Toll Free: (800) Telephone: (412) Fax: (412) Carbonair Environmental Systems, Inc Nevada Avenue North New Hope, MN Telephone: (612) Fax: (612)

392 Calgon Carbon Corporation PO Box 717 Pittsburgh, PA Toll Free: (800) - 4 CARBON Telephone: (412) Fax: (412) Norit Americas Inc Crown Pointe Pkwy, Suite 1500 Atlanta, Georgia Toll Free: (800) Envirotrol, Inc. P.O. Box 61-T Sewickley, PA Telephone: (412) Fax: (412) Carbochem, Inc. 326 W. Lancaster Avenue Ardmore, PA Telephone: (610) Fax: (610) Biofilters BioCube AMETEK Rotron Biofiltration Products 75 North Street Saugerties, NY Telephone: (914) Fax: (914) Bohn Biofilter Corporation P.O. Box Tucson, AZ Telephone: (520) Monsanto Enviro-Chem S. Outer 40 Road St. Louis, MO Telephone: (314) US Filter/Westates 5375 S. Boyle Avenue Los Angeles, CA Toll Free: (800) Telephone: (610) Fax: (610) Indusco Environmental Services, Inc. P.O. Box Atlanta, GA Telephone: (770) Fax: (770) Nucon International, Inc. P.O. Box Huntley Road Columbus, OH Telephone: (614) Fax: (614)

393 Appendix 3A Assumptions for AOP Economic Analysis Introduction This appendix presents the assumptions and backup data for the range of MTBE treatment scenarios presented in Chapter 3. The summary tables presented in Chapter 3, which compile the data from the tables in this appendix, will not be repeated here. Cost estimates were requested from five AOP vendors for 24 different treatment scenarios along with additional scenarios to evaluate effects of BTEX and TOC on treatment costs. Those vendors who participated in the detailed cost evaluation are listed below. The costs were critically reviewed by performing data validation where field pilot testing data was available for comparison. Vendors: Applied Process Technology, Inc. (San Francisco, CA) Calgon Carbon Corporation (Markham, Ontario, Canada) Hydroxyl Systems, Inc. (Victoria, British Columbia, Canada) Oxidation Systems, Inc. (Arcadia, CA) Assumptions Given to Vendors: Influent flow rates of 60, 600, and 6,000 gpm. Influent MTBE concentrations of 20, 200, and 2,000 µg/l. Effluent MTBE discharge requirements of 20, 5, and 0.5 µg/l. Hardness: 200 mg/l as CaCO 3 Alkalinity: 250 mg/l as CaCO 3 Bromide: ND Iron: <1 mg/l ph : 7.0 Temperature: 65 F TDS: 500 mg/l Nitrate: 25 mg/l as NO 3 or 5mg/L as N Assumptions for Capital Costs: 30-year system design life. Seven percent rate for capital amortization. Piping, valves, and electrical: 30 percent AOP unit costs. Site work (e.g., clearing, excavations, foundation): 10 percent of AOP unit costs. Contractor O&P: 15 percent of capital cost for AOP unit, site work, piping, valves, and electrical. Engineering: 15 percent of capital costs for AOP unit, site work, piping, valves, electrical, and contractor O&P. Contingency: 20 percent of all capital costs (engineering costs, contractor O&P AOP equipment, site work, piping, valves, and electrical). 371

394 The tables summarizing the capital costs were presented in Chapter 3. Table 3-9 presents the treatment costs for the hydrogen peroxide removal and oxidation by-product removal systems. Table 3-10 presents an example showing the breakdown of the calculations for the capital costs along with the O&M costs for one AOP system. Table 3-11, 3-12, and 3-13 present a summary of the capital, O&M, and total amortized costs for the 24 treatment scenarios. Tables showing the effects of TOC, BTEX, and design life are presented in Tables 3-15, 3-16, and 3-17, respectively. Assumptions for O&M Costs: Replacement parts: Costs are based on vendor s estimates (Table 3A-1). Replacement costs may include bulb replacements for MP-UV sytem, ozone generator for ozone system, catalyst replacement for TiO2 system, and other components. Labor costs: The labor costs include water sampling, general and specific system O&M (Tables 3A-2, 3A-6, 3A-7, 3-8, and 3A-9). Analytical costs: Costs are based on the estimated sampling frequency required as presented in Tables 3A-6 through 3A-9. Chemical costs: Cost were provided by vendors and are summarized in Table 3A-4. This includes the consumable chemicals for the applicable technology. Electrical costs: Electrical costs were based on power consumption estimates provided by the vendors and are presented in Table 3A-5. Power cost was based on $0.08 per kilowatthour. 372

395 373 Replacement costs are based on vendor's estimates. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 4,800 $ 3,000 $ 1,200 $ 22, % $ 5,200 $ 3,000 $ 3,300 $ 25, % $ 4,800 $ 3,000 $ 1,400 $ 25, % $ 5,200 $ 3,000 $ 3,300 $ 22, % $ 5,200 $ 3,600 $ 3,500 $ 28, % $ 5,200 $ 3,600 $ 3,500 $ 28, % $ 5,200 $ 3,600 $ 3,500 $ 24, % $ 10,500 $ 4,200 $ 3,500 $ 30, % $ 10,500 $ 11,300 $ 1,600 $ 63, % $ 25,300 $ 12,000 $ 2,100 $ 69, % $ 15,700 $ 11,300 $ 1,600 $ 69, % $ 27,900 $ 12,000 $ 2,100 $ 82, % $ 38,000 $ 12,000 $ 4,300 $ 91, % $ 45,600 $ 12,000 $ 2,200 $ 91, % $ 60,800 $ 12,800 $ 4,300 $ 113, % $ 71,000 $ 12,800 $ 4,300 $ 126, % $ 62,900 $ 54,000 $ 6,500 $ 504, % $ 152,000 $ 60,000 $ 18,700 $ 567, % $ 99,600 $ 54,000 $ 14,400 $ 567, % $ 162,500 $ 60,000 $ 18,700 $ 725, % $ 235,900 $ 60,000 $ 39,100 $ 756, % $ 277,900 $ 60,000 $ 19,500 $ 756, % $ 429,900 $ 60,000 $ 39,100 $ 1,009, % $ 592,400 $ 66,000 $ 39,100 $ 1,072,200 Table 3A-1 Replacement Part Costs for AOPs

396 374 Breakdown of labor costs are given in Tables 3A-6 to 3A-9, based on a rate of $80/hr. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 21,000 $ 22,100 $ 25,900 $ 18, % $ 21,000 $ 22,100 $ 25,900 $ 18, % $ 21,000 $ 22,100 $ 25,900 $ 19, % $ 21,000 $ 22,100 $ 25,900 $ 19, % $ 21,000 $ 22,100 $ 25,900 $ 19, % $ 29,400 $ 30,400 $ 34,200 $ 29, % $ 29,400 $ 30,400 $ 34,200 $ 29, % $ 29,400 $ 30,400 $ 34,200 $ 29, % $ 46,700 $ 57,900 $ 56,000 $ 91, % $ 46,700 $ 57,900 $ 56,000 $ 91, % $ 47,200 $ 57,900 $ 56,000 $ 93, % $ 47,200 $ 57,900 $ 56,000 $ 93, % $ 47,200 $ 57,900 $ 56,000 $ 93, % $ 74,000 $ 82,900 $ 81,000 $ 117, % $ 74,000 $ 82,900 $ 81,000 $ 117, % $ 74,000 $ 82,900 $ 81,000 $ 117, % $ 259,400 $ 211,500 $ 234,200 $ 868, % $ 259,400 $ 211,500 $ 234,200 $ 868, % $ 261,800 $ 211,500 $ 234,200 $ 888, % $ 261,800 $ 211,500 $ 234,200 $ 888, % $ 261,800 $ 211,500 $ 234,200 $ 888, % $ 424,300 $ 361,300 $ 384,000 $ 1,122, % $ 424,300 $ 361,300 $ 384,000 $ 1,122, % $ 424,300 $ 361,300 $ 384,000 $ 1,122,600 Table 3A-2 Labor Costs for AOPs

397 375 Breakdown of labor costs are given in Tables 3A-6 to 3A-9, based on a rate of $200/sample. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 20,800 $ 20,800 $ 20,800 $ 20, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 31,200 $ 41,600 $ 31,200 $ 41, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322, % $ 135,200 $ 72,800 $ 72,800 $ 322,400 Table 3A-3 Analytical Costs for AOPs

398 376 Chemical costs include that for H 2 O 2, O 3, and TiO 2. Amounts used and their cost were supplied by vendor. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Carbon Corporation Applied Process Technology, Inc. Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 2,800 $ 631 $ 5,000 $ % $ 2,800 $ 1,600 $ 5,000 $ % $ 5,400 $ 900 $ 5,000 $ % $ 2,800 $ 1,600 $ 5,000 $ % $ 2,800 $ 2,200 $ 7,900 $ % $ 5,400 $ 3,200 $ 5,000 $ % $ 18,000 $ 3,500 $ 7,900 $ % $ 5,400 $ 4,400 $ 8,500 $ % $ 28,400 $ 6,300 $ 50,500 $ 6, % $ 41,000 $ 15,800 $ 50,500 $ 6, % $ 41,000 $ 9,500 $ 50,500 $ 6, % $ 41,000 $ 15,800 $ 50,500 $ 6, % $ 69,400 $ 22,100 $ 78,800 $ 6, % $ 82,000 $ 28,400 $ 50,500 $ 6, % $ 69,400 $ 34,700 $ 78,800 $ 6, % $ 82,000 $ 44,200 $ 85,100 $ 6, % $ 220,800 $ 63,100 $ 504,600 $ 63, % $ 283,800 $ 157,700 $ 504,600 $ 63, % $ 283,800 $ 94,600 $ 504,600 $ 63, % $ 410,000 $ 157,700 $ 504,600 $ 63, % $ 410,000 $ 220,800 $ 788,400 $ 63, % $ 410,000 $ 283,800 $ 504,600 $ 63, % $ 693,800 $ 346,900 $ 788,400 $ 63, % $ 693,800 $ 441,500 $ 851,500 $ 63,100 Table 3A-4 Chemical Costs for AOPs

399 377 Electrical costs were based on power consumption estimate provided by the vendors, at $0.08/kWh. Flow (gpm) Influent (µg/l) Effluent (µg/l) Removal Efficiency (%) Calgon Oxidation Technologies Applied Process Technologies Oxidation Systems, Inc. Hydroxyl Systems, Inc % $ 5,000 $ 631 $ 7,300 $ 12, % $ 13,900 $ 1,300 $ 7,900 $ 13, % $ 6,900 $ 900 $ 7,300 $ 13, % $ 13,900 $ 1,300 $ 7,900 $ 16, % $ 20,800 $ 2,200 $ 8,200 $ 21, % $ 20,800 $ 2,800 $ 8,200 $ 21, % $ 20,800 $ 3,200 $ 8,200 $ 21, % $ 41,900 $ 4,100 $ 8,200 $ 26, % $ 41,000 $ 6,300 $ 28,400 $ 63, % $ 104,100 $ 12,600 $ 34,700 $ 69, % $ 63,100 $ 9,500 $ 28,400 $ 69, % $ 116,700 $ 12,600 $ 34,700 $ 107, % $ 157,700 $ 22,100 $ 37,800 $ 116, % $ 189,200 $ 28,400 $ 37,800 $ 116, % $ 252,300 $ 31,500 $ 37,800 $ 173, % $ 293,300 $ 41,000 $ 37,800 $ 192, % $ 252,300 $ 63,100 $ 283,800 $ 630, % $ 599,200 $ 126,100 $ 346,900 $ 693, % $ 410,000 $ 94,600 $ 283,800 $ 693, % $ 662,300 $ 126,100 $ 346,900 $ 1,103, % $ 946,100 $ 220,800 $ 378,400 $ 1,166, % $ 410,000 $ 283,800 $ 378,400 $ 1,166, % $ 1,702,900 $ 315,400 $ 378,400 $ 1,829, % $ 2,365,200 $ 410,000 $ 378,400 $ 1,923,700 Table 3A-5 Electrical Costs for AOPs

400 60 gpm Systems 1 reactor Influent Sampling Analytical Sampling Estimated UV Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Lamp Changeouts B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $29, $21, $21, gpm Systems 2 reactors Influent Sampling Analytical Sampling Estimated UV Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Lamp Changeouts B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $74, $47, $46,700 6,000 gpm Systems 12 reactors Influent Sampling Analytical Sampling Estimated UV Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Lamp Changeouts B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $424, $261, $259,400 A - 1 hour/sample B - Lamp change-outs every 3,000 hours; 1 hour labor per lamp. Number of lamps for each system assumed as follows: for 60 gpm system: 1 at 2,000 µg/l, 200 µg/l, and 20 µg/l for 600 gpm system: 14 at 2,000 µg/l; 6 at 200 µg/l; 4 at 20 µg/l for 6,000 gpm system: 96 at 2,000 µg/l; 48 at 200 µg/l; 24 at 20 µg/l C - Inspection/replacement of pump components, calibration and flow verification, change of H 2O 2 storage vessels(1hr/wk per reactor) D - general system oversight and maintenance (e.g., pressure checks, backflushing) for 60 gpm system: 4 hr/wk at 2,000 µg/l; 2 hr/wk at 200 µg/l and 20 µg/l for 600 gpm system: 12 hr/wk at 2,000 µg/l; 6 hr/wk at 200 µg/l and 20 µg/l for 6,000 gpm system: 72 hr/wk at 2,000 µg/l; 36 hr/wk at 200 µg/l and 20 µg/l E - labor $80/hr Table 3A-6 Estimated Labor Costs for H 2 O 2 /MP-UV Systems

401 60 gpm Systems 1 reactor Influent Sampling Analytical Sampling Estimated O 3 Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A System O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $30, $22, $22, gpm Systems 3 reactors Influent Sampling Analytical Sampling Estimated O 3 Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A System O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $82, $57, $57,900 6,000 gpm Systems 6 reactors Influent Sampling Analytical Sampling Estimated O 3 Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A System O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $361, $211, $211,500 Table 3A-7 Estimated Labor Costs for H 2 O 2 /O 3 System A - 1 hour/sample B - O 3 generator replacement (one year service life, 2 days per replacement) C - Inspection/replacement of pump components, calibration and flow verification, change of H 2O 2 storage vessels (1hr/wk per reactor) D - general system oversight and maintenance (e.g., pressure gauges, control panel, leak checks) for 60 gpm system: 4 hr/wk at 2,000 µg/l; 2 hr/wk at 200 µg/l and 20 µg/l for 600 gpm system: 12 hr/wk at 2,000 µg/l; 6 hr/wk at 200 µg/l and 20 µg/l for 6,000 gpm system: 72 hr/wk at 2,000 µg/l; 36 hr/wk at 200 µg/l and 20 µg/l E - labor $80/hr

402 gpm Systems 1 reactor Influent Sampling Analytical Sampling Estimated Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $34, $25, $25, gpm Systems 2 reactors Influent Sampling Analytical Sampling Estimated Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $81, $56, $56,000 6,000 gpm Systems 6 reactors Influent Sampling Analytical Sampling Estimated Estimated H 2O 2 General O & M Total Annual Total Annual Concentration Frequency Annual Labor A O & M B System O & M C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $384, $234, $234,200 A - 1 hour/sample B - Inspection, replacement, and repair based on 1,000 hour service life C - Inspection/replacement of pump components, calibration and flow verification, change of H 2O 2 storage vessels (1hr/wk per reactor) D - general system oversight and maintenance (e.g., pressure gauges, control panel, leak checks) for 60 gpm system: 4 hr/wk at 2,000 µg/l; 2 hr/wk at 200 µg/l and 20 µg/l for 600 gpm system: 12 hr/wk at 2,000 µg/l; 6 hr/wk at 200 µg/l and 20 µg/l for 6,000 gpm system: 72 hr/wk at 2,000 µg/l; 36 hr/wk at 200 µg/l and 20 µg/l E - labor $80/hr Table 3A-8 Estimated Labor Costs for Ultrasound/H 2 O 2 Systems

403 gpm Systems 1 reactor Influent Sampling Analytical Sampling Estimated Catalyst Estimated UV General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Maintenance B Lamp Changeouts C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $29, $19, $18, gpm Systems 3 reactors Influent Sampling Analytical Sampling Estimated Catalyst Estimated UV General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Maintenance B Lamp Changeouts B Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $117, $93, $91,200 6,000 gpm Systems 30 reactors Influent Sampling Analytical Sampling Estimated Catalyst Estimated UV General O & M Total Annual Total Annual Concentration Frequency Annual Labor A Maintenance B Lamp Changeouts B Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) (hours/year) (hours/year) (hours) (hours) ($) 2, $1,122, $888, $868,600 Table 3A-9 Estimated Labor Costs for TiO 2 /MP-UV System A - 1 hour/sample B - Maintenance includes spent catalyst removal, recharge, and delivery C - 1 hour/ lamp changeout for UV/H 2O 2 (Calgon) and UV/titanium dioxide (Hydroxyl) Systems D - general system oversight and maintenance (e.g., pressure checks, backflushing) for 60 gpm system: 4 hr/wk at 2,000 µg/l; 2 hr/wk at 200 µg/l and 20 µg/l for 600 gpm system: 20 hr/wk at 2,000 µg/l; 16 hr/wk at 200 µg/l and 20 µg/l for 6,000 gpm system: 200 hr/wk at 2,000 µg/l; 160 hr/wk at 200 µg/l and 20 µg/l E - labor $80/hr

404 382

405 Appendix 4A Cost Estimates for GAC Introduction This appendix presents cost estimates for a range of MTBE treatment scenarios using GAC technology. The cost estimates were developed based on results of carbon adsorption modeling using the AdDesignS computer model (Mertz et al., 1994), 1998 price quotes from carbon vendors, and standard cost estimating assumptions for feasibility-level evaluations. The cost estimates presented here are considered to be accurate within ±30 percent. Details regarding assumptions used in the development of the predicted carbon usage rates, which impact the estimated O&M costs, are presented in Chapter 4. Actual carbon usage rates for site-specific conditions should be obtained via testing with site water if higher accuracy cost estimates are required. Assumptions used for development of O&M labor costs for the primary treatment scenarios and the sensitivity analyses are presented in Tables 4A-1 and 4A-2, respectively. Summaries of the cost estimates are presented in Tables 4A-3 and 4A-4. The assumptions used for development of the cost estimates are listed below. General Assumptions Influent MTBE concentrations: 20 µg/l, 200 µg/l, 2,000 µg/l. Effluent water contains no detectable MTBE (<0.5 µg/l). System flow rates: 60 gpm, 600 gpm, 6,000 gpm. No pretreatment of influent water is required. Assumptions for Capital Costs Carbon adsorption unit: Standard Carbonair vessels rated for appropriate range of flow rates; costs for adsorber systems based on 1998 price quotes from Carbonair for purchase and installation of specific models. Capital cost includes initial fill with virgin, coconut shell GAC at $1.25/lb (unit cost based on 1998 vendor price quote). All systems designed as single line or parallel lines of three GAC vessels in-series. Piping, Valves, and Electrical: 30 percent of capital cost for carbon adsorption unit. Site Work (e.g., clearing, grubbing, foundation placement): 10 percent of capital cost for carbon adsorption unit. Contractor O&P: 15 percent of capital cost for carbon adsorption unit, site work, piping and valves, and electrical. 383

406 Engineering: 15 percent of capital cost for carbon adsorption unit, site work, piping, valves, electrical, and contractor O&P. Contingency: 20 percent of all other capital costs. Total capital amortized over 30-year system design life using seven percent discount rate. Assumptions for O&M Costs Replacement GAC: Carbon changouts using virgin, coconut shell GAC at $1.25/lb (unit cost based on 1998 vendor price quotes). Estimated changeout frequency based on results of AdDesignS modeling (Tables 4-4 and 4-5), assuming increased bed life of 50 percent for in-series operation (see Chapter 4 for further details). Changeout Labor/Transport: Estimated costs based on 1998 vendor price quotes ($0.10/lb for transport to off-site regeneration facility). Assumed $1,000 minimum cost per changeout event. O&M Labor: Estimated costs for analytical sampling, GAC changeout oversight, and general system O&M (e.g., pressure checks, backflushing). Detailed assumptions presented in Tables 4A-1 and 4A-2. Analytical Testing: 60 gpm systems - three samples per week per line (one influent, one midfluent, one effluent); 600 gpm systems - five samples per week per line (one influent, two midfluent, two effluent); 6,000 gpm systems - 25 samples per week per line (1 influent, 12 midfluent, 12 effluent); Assumed testing cost - $200 per sample, Method Power: Assumed unit cost at $0.08 per kilowatt-hour; kilowatt-hours estimated based on flow rate and bed depth. 384

407 gpm Systems 1 line of 3 vessels in-series Influent Sampling Analytical Sampling Predicted GAC Changeout General O&M Total Annual Total Annual Concentration Frequency Annual Labor A Changeouts Annual Oversight C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) Per Year B (hours) (hours) (hours) ($) 2, $32, $22, $21, gpm Systems 2 lines of 3 vessels in-series Influent Sampling Analytical Sampling Predicted GAC Changeout General O&M Total Annual Total Annual Concentration Frequency Annual Labor A Changeouts Annual Oversight C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) Per Year B (hours) (hours) (hours) ($) 2, $61, $40, $38,000 6,000 gpm Systems 12 lines of 3 vessels in-series Influent Sampling Analytical Sampling Predicted GAC Changeout General O&M Total Annual Total Annual Concentration Frequency Annual Labor A Changeouts Annual Oversight C Annual Labor D Labor Labor Cost E (µg/l) (samples/wk) (hours) Per Year B (hours) (hours) (hours) ($) 2, $307, $211, $153,000 A 1 hour/sample. B See Table 4-6. C 4 hours/changeout for each line of GAC vessels. D General system oversight and maintenance (e.g., pressure checks, backflushing): For 60 gpm system: 4 hr/wk at 2,000 µg/l, 2 hr/wk at 200 µg/l, and 20 µg/l. For 600 gpm system: 8 hr/wk at 2,000 µg/l, 4 hr/wk at 200 µg/l, and 20 µg/l. For 6,000 gpm system: 32 hr/wk at 2,000 µg/l, 20 hr/wk at 200 µg/l, and 10 hr/wk at 20 µg/l. E Labor $80/hr. Table 4A-1 Estimated Labor Costs for GAC Systems

408 386 Water Sampling Analytical Sampling Predicted GAC Changeout General O&M Total Annual Total Annual Type Frequency Annual Labor A Changeouts Annual Oversight C Annual Labor D Labor Labor Cost E (samples/wk) (hours) Per Year B (hours) (hours) (hours) ($) Rhine River $38,000 (high fouling) Karlsruhe gw $38,000 (moderate fouling) Wausau gw $38,000 (low fouling) Moderate BTEX F $38,000 each at 200 µg/l Low BTEX F $38,000 each at 20 µg/l No BTEX F $38,000 each at 0 µg/l A 1 hour/sample. B See Table 4-5. C 4 hours/changeout for each line of GAC vessels. D 4 hrs/wk; includes general system oversight and maintenance (e.g., pressure checks, backflushing). E Labor $80/hr. F BTEX and MTBE in Karlsruhe groundwater. 600 gpm Systems. 2 lines of 3 vessels in-series. Influent MTBE = 20 µg/l; effluent contains no detectable MTBE (0.5 µg/l). Table 4A-2 Estimated Labor Costs for Sensitivity Analysis

409 Table 4A-3 Estimated Costs for Carbon Adsorption System Parameters: 60 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no detectable MTBE (0.5 µg/l) Effluent contains no detectable MTBE (<0.5 mg/l) High fouling (Rhine Riverwater) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $182,640 TOTAL ANNUAL COST $264,731 TOTAL COST PER 1,000 GALLONS TREATED $0.84 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.6 $50,000 $80,000 Changeout Labor/Transport event 1.6 $4,000 $6,400 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $182,640 1 Carbon system size: three 2,500 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 2,500 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 3 samples per weekly event, 52 weeks per year. 387

410 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: 60 gpm 60 gpm 200 µg/l influent MTBE 200 concentration µg/l influent MTBE concentration Effluent contains no detectable Effluent contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $105,110 SUBTOTAL $105,110 Piping, Valves, Electrical (30%) $31,533 Site Work (10%) $10,511 SUBTOTAL $147,154 Contractor O&P (15%) $22,073 SUBTOTAL $169,227 Engineering (15%) $25,384 SUBTOTAL $194,611 Contingency (20%) $38,922 TOTAL CAPITAL $233,533 AMORTIZED CAPITAL 2 $18,820 ANNUAL O&M $73,319 TOTAL ANNUAL COST $92,139 TOTAL COST PER 1,000 GALLONS TREATED $2.92 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 2.7 $6,250 $16,875 Changeout Labor/Transport event 2.7 $1,000 $2,700 O&M Labor 4 year 1 $22,000 $22,000 Analytical Testing 5 samples 156 $200 $31,200 Power ($0.08/kWh) kwhr 6,800 $0.08 $544 ANNUAL O&M $73,319 1 Carbon system size: three 5,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 5,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 3 samples per weekly event, 52 weeks per year. 388

411 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 60 gpm 60 gpm 2,000 µg/l influent MTBE 2,000 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $105,110 SUBTOTAL $105,110 Piping, Valves, Electrical (30%) $31,533 Site Work (10%) $10,511 SUBTOTAL $147,154 Contractor O&P (15%) $22,073 SUBTOTAL $169,227 Engineering (15%) $25,384 SUBTOTAL $194,611 Contingency (20%) $38,922 TOTAL CAPITAL $233,533 AMORTIZED CAPITAL 2 $18,820 ANNUAL O&M $121,019 TOTAL ANNUAL COST $139,839 TOTAL COST PER 1,000 GALLONS TREATED $4.43 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 7.9 $6,250 $49,375 Changeout Labor/Transport event 7.9 $1,000 $7,900 O&M Labor 4 year 1 $32,000 $32,000 Analytical Testing 5 samples 156 $200 $31,200 Power ($0.08/kWh) kwhr 6,800 $0.08 $544 ANNUAL O&M $121,019 1 Carbon system size: three 5,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 5,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 3 samples per weekly event, 52 weeks per year. 389

412 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $161,040 TOTAL ANNUAL COST $243,131 TOTAL COST PER 1,000 GALLONS TREATED $0.77 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.2 $50,000 $60,000 Changeout Labor/Transport event 1.2 $4,000 $4,800 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $161,040 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 390

413 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 600 gpm 600 gpm 200 µg/l influent MTBE 200 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $281,840 TOTAL ANNUAL COST $363,931 TOTAL COST PER 1,000 GALLONS TREATED $1.15 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 3.4 $50,000 $170,000 Changeout Labor/Transport event 3.4 $4,000 $13,600 O&M Labor 4 year 1 $40,000 $40,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $281,840 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 391

414 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 600 gpm 600 gpm 2,000 µg/l influent MTBE 2,000 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 mg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $664,640 TOTAL ANNUAL COST $746,731 TOTAL COST PER 1,000 GALLONS TREATED $2.37 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 10.1 $50,000 $505,000 Changeout Labor/Transport event 10.1 $4,000 $40,400 O&M Labor 4 year 1 $61,000 $61,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $664,640 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 4 Based on 5 samples per weekly event, 52 weeks per year. 392

415 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 6,000 gpm 6,000 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $2,690,940 SUBTOTAL $2,690,940 Piping, Valves, Electrical (30%) $807,282 Site Work (10%) $269,094 SUBTOTAL $3,767,316 Contractor O&P (15%) $565,097 SUBTOTAL $4,332,413 Engineering (15%) $649,862 SUBTOTAL $4,982,275 Contingency (20%) $996,455 TOTAL CAPITAL $5,978,730 AMORTIZED CAPITAL 2 $481,806 ANNUAL O&M $1,091,000 TOTAL ANNUAL COST $1,572,806 TOTAL COST PER 1,000 GALLONS TREATED $0.50 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.9 $300,000 $570,000 Changeout Labor/Transport event 1.9 $24,000 $45,600 O&M Labor 4 year 1 $153,000 $153,000 Analytical Testing 5 samples 1,300 $200 $260,000 Power ($0.08/kWh) kwhr 780,000 $0.08 $62,400 ANNUAL O&M $1,091,000 1 Carbon system size: 12 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 240,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 25 samples per weekly event, 52 weeks per year. 393

416 Table 4A-3 (Continued) Estimated Costs for Carbon Adsorption System Parameters: Parameters: 6,000 gpm 6,000 gpm 200 µg/l influent MTBE 200 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $2,690,940 SUBTOTAL $2,690,940 Piping, Valves, Electrical (30%) $807,282 Site Work (10%) $269,094 SUBTOTAL $3,767,316 Contractor O&P (15%) $565,097 SUBTOTAL $4,332,413 Engineering (15%) $649,862 SUBTOTAL $4,982,275 Contingency (20%) $996,455 TOTAL CAPITAL $5,978,730 AMORTIZED CAPITAL 2 $481,806 ANNUAL O&M $2,574,600 TOTAL ANNUAL COST $3,056,406 TOTAL COST PER 1,000 GALLONS TREATED $0.97 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 6.3 $300,000 $1,890,000 Changeout Labor/Transport event 6.3 $24,000 $151,200 O&M Labor 4 year 1 $211,000 $211,000 Analytical Testing 5 samples 1,300 $200 $260,000 Power ($0.08/kWh) kwhr 780,000 $0.08 $62,400 ANNUAL O&M $2,574,600 1 Carbon system size: 12 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 240,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 25 samples per weekly event, 52 weeks per year. 394

417 Table 4A-3 (Concluded) Estimated Costs for Carbon Adsorption System System Parameters: Parameters: 6,000 gpm 6,000 gpm 2,000 µg/l influent 2,000 MTBE µg/l concentration influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Item Cost Carbon Adsorption Unit 1 $2,690,940 SUBTOTAL $2,690,940 Piping, Valves, Electrical (30%) $807,282 Site Work (10%) $269,094 SUBTOTAL $3,767,316 Contractor O&P (15%) $565,097 SUBTOTAL $4,332,413 Engineering (15%) $649,862 SUBTOTAL $4,982,275 Contingency (20%) $996,455 TOTAL CAPITAL $5,978,730 AMORTIZED CAPITAL 2 $481,806 ANNUAL O&M $6,526,200 TOTAL ANNUAL COST $7,008,006 TOTAL COST PER 1,000 GALLONS TREATED $2.22 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 18.2 $300,000 $5,460,000 Changeout Labor/Transport event 18.2 $24,000 $436,800 O&M Labor 4 year 1 $307,000 $307,000 Analytical Testing 5 samples 1,300 $200 $260,000 Power ($0.08/kWh) kwhr 780,000 $0.08 $62,400 ANNUAL O&M $6,526,200 1 Carbon system size: 12 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 240,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 25 samples per weekly event, 52 weeks per year. 395

418 Table 4A-4 Estimated Costs for Sensitivity Analysis System Parameters: Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) High fouling (Rhine High Riverwater) fouling (Rhine River water) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $182,640 TOTAL ANNUAL COST $264,731 TOTAL COST PER 1,000 GALLONS TREATED $0.84 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.6 $50,000 $80,000 Changeout Labor/Transport event 1.6 $4,000 $6,400 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $182,640 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 396

419 Table 4A-4 (Continued) Estimated Costs for Sensitivity Analysis System Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Moderate fouling (Karlsruhe Moderate fouling groundwater) (Karlsruhe ground water) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $161,040 TOTAL ANNUAL COST $243,131 TOTAL COST PER 1,000 GALLONS TREATED $0.77 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.2 $50,000 $60,000 Changeout Labor/Transport event 1.2 $4,000 $4,800 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $161,040 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 397

420 Table 4A-4 (Continued) Estimated Costs for Sensitivity Analysis System Parameters: Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Low fouling (Wausau Low groundwater) fouling (Wausau ground water) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $155,640 TOTAL ANNUAL COST $237,731 TOTAL COST PER 1,000 GALLONS TREATED $0.75 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.1 $50,000 $55,000 Changeout Labor/Transport event 1.1 $4,000 $4,400 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $155,640 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 398

421 Table 4A-4 (Continued) Estimated Costs for Sensitivity Analysis System System Parameters: Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Moderate BTEX load Moderate (200 µg/l BTEX each) load (200 g/l each) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $198,840 TOTAL ANNUAL COST $280,931 TOTAL COST PER 1,000 GALLONS TREATED $0.89 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.9 $50,000 $95,000 Changeout Labor/Transport event 1.9 $4,000 $7,600 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $198,840 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 399

422 Table 4A-4 (Continued) Estimated Costs for Sensitivity Analysis System Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) Low BTEX load (20 Low µg/lbtex each) load (20 ppb each) Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $82,091 ANNUAL O&M $171,840 TOTAL ANNUAL COST $253,931 TOTAL COST PER 1,000 GALLONS TREATED $0.81 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.4 $50,000 $70,000 Changeout Labor/Transport event 1.4 $4,000 $5,600 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $171,840 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 400

423 Table 4A-4 (Continued) Estimated Costs for Sensitivity Analysis System Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) 10 year design life for 10 year capital design amortization life for capital amortization Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $125,592 ANNUAL O&M $161,040 TOTAL ANNUAL COST $286,632 TOTAL COST PER 1,000 GALLONS TREATED $0.91 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.2 $50,000 $60,000 Changeout Labor/Transport event 1.2 $4,000 $4,800 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $161,040 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 10 year period. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 401

424 Table 4A-4 (Concluded) Estimated Costs for Sensitivity Analysis System Parameters: 600 gpm 600 gpm 20 µg/l influent MTBE 20 concentration µg/l influent MTBE concentration Effluent contains no Effluent detectable contains MTBE no detectable (<0.5 µg/l) MTBE (0.5 µg/l) 2 year design life for 2 capital year design amortization life for capital amortization Item Cost Carbon Adsorption Unit 1 $458,490 SUBTOTAL $458,490 Piping, Valves, Electrical (30%) $137,547 Site Work (10%) $45,849 SUBTOTAL $641,886 Contractor O&P (15%) $96,283 SUBTOTAL $738,169 Engineering (15%) $110,725 SUBTOTAL $848,894 Contingency (20%) $169,779 TOTAL CAPITAL $1,018,673 AMORTIZED CAPITAL 2 $540,124 ANNUAL O&M $161,040 TOTAL ANNUAL COST $701,164 TOTAL COST PER 1,000 GALLONS TREATED $2.22 Summary of Annual O&M Costs Item Unit Quantity Unit Cost Cost Replacement Carbon 3 event 1.2 $50,000 $60,000 Changeout Labor/Transport event 1.2 $4,000 $4,800 O&M Labor 4 year 1 $38,000 $38,000 Analytical Testing 5 samples 260 $200 $52,000 Power ($0.08/kWh) kwhr 78,000 $0.08 $6,240 ANNUAL O&M $161,040 1 Carbon system size: 2 lines of three 20,000 lb vessels in series. 2 Amortization based on 30 year period at 7% discount rate. 3 Based on $1.25/lb and 40,000 lb/event; changeout frequency per Table Includes analytical sampling, oversight during changeouts, and general system O&M. 5 Based on 5 samples per weekly event, 52 weeks per year. 402

425 Isotherms provided by Calgon Carbon (Pittsburgh, PA): Appendix 5A Filtrasorb 600 Isotherm Adsorption Isotherms for the Removal of MTBE from DI Water mg MTBE/cc GAC MTBE (mg/l) Figure 5A-1 MTBE Isotherms for Filtrasorb 600 and PCB, a coconut-based GAC. 403

426 404

427 Appendix 5B Assumptions for Synthetic Resin Sorbents Cost Estimates A. Assumptions made by Malcolm Pirnie, Inc.: Seven percent discount rate 30-year project life time All costs are 1999 $ Cost: $25/lb for large quantity orders (6,000 gpm system) $30/lb for 600 gpm system $45/lb for 60 and 6 gpm systems Labor: See Table 5B-4 Analytical: See Table 5B-4 Pump Requirements: 600 gpm/vessel (40 ft head loss in series; 25 ft for parallel) 74 gpm/vessel (35 ft head loss in series; 20 ft for parallel) 60 gpm/vessel (35 ft head loss in series; 20 ft for parallel) 6 gpm/vessel (20 ft head loss in series; 15 ft for parallel) Freundlich Isotherm Parameters for 15 C: MTBE: K F = 17; n = 0.35 TBA: K F = 1.797; n = Modeling assumptions: Breakthrough time in days and BVs treated listed in Table 5B-1; Five percent increase in model-predicted capacity for full column exhaustion (based on laboratory and field data showing a wider breakthrough curve) Split flow to achieve removal efficiency. Treated flow rates and corresponding carbon vessels listed in Table 5B-2. B. Regeneration Assumptions made by Malcolm Pirnie, Inc.: Steam Regeneration: Components: Boiler, Condensing Tank 10 BVs for regeneration (1 BV/hour) Five percent one-time loss in sorptive capacity $5/Mbtu natural gas Costs estimated from Chemical Engineering Economics (Garrett, 1989) Flow rate: 12 hours to treat regenerant for 6,000, 600, and 60 gpm 1 hour to treat regenerant for 6 gpm For 6000 gpm: staggered regeneration Effluent Goal: 3 mg/l for GAC, Resins, or Air Stripping to be fed back into resin influent Freundlich Isotherm Parameters for 15 C: MTBE: K F = 11.0; n = (used by Alpine Environmental, Inc. for GAC report) 405

428 C. Regeneration Assumptions Made by Vendors (or using Vendor Models): ShallowTray Air Stripping with Off-gas Treatment: Vendor: Northeast Environmental Products 6,000 gpm and 600 gpm: AWR = gpm/2,000 µg/l influent: AWR = gpm/200 µg/l and 20 µg/l influent: AWR = gpm/2,000 µg/l influent: AWR = gpm/200 µg/l and 20 µg/l influent: AWR = 137 Vendor: Advanced Environmental Systems Catalytic Off-gas Treatment with Air Stream Preheating Temperature = 90 F 6,000 gpm, 600 gpm: 1,500 cfm catalytic oxidizer 60 gpm, 6 gpm: 250 cfm catalytic oxidizer Safety Kleen costs for Hazardous Waste Disposal (drum = 55 gallons) <20 barrels: $327/drum barrels: $298/drum barrels: $278/drum barrels: $261/drum barrels: $196/drum Activated Carbon Regeneration: Vendor: Calgon Carbon $1.25/lb for use and disposal 406

429 Table 5B-1. MTBE Sorption Modeling Results Applicable for Option 1 Applicable for Option 2 Flow Rate [gpm] Influent [µg/l] Effluent Goal [µg/l] Removal Efficiency Time until Column Exhaustion [days] Bed Volumes Treated at Column Exhaustion Point Time to Breakthrough of Effluent Goal [days] Bed Volumes Treated at Breakthrough of Effluent Goal % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

430 Table 5B-2. MTBE Assumptions Used in AdDesignS Flow Rate [gpm] Influent [µg/l] Effluent Goal [µg/l] Removal Efficiency Water Requiring Treatment to Meet Effluent Goal [gpm] Single Adsorber Flow [gpm] Option 1: Carbonair Adsorber Name Option 1: Number of Double (2 in series) Vessels in Parallel Option 2: Carbonair Adsorber Name Option 2: Number of Vessels in Pure-Parallel % PC78 9 PC % PC78 9 PC % PC78 9 PC % PC78 8 PC % PC78 9 PC % PC78 9 PC % PC78 7 PC % PC78 9 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC78 1 PC % PC13 1 PC % PC13 1 PC % PC13 1 PC % PC13 1 PC % PC13 1 PC % PC13 1 PC % PC7 1 PC % PC13 1 PC % 6 6 PC1 1 PC % 6 6 PC1 1 PC % 6 6 PC1 1 PC % 5 5 PC1 1 PC % 6 6 PC1 1 PC % 6 6 PC1 1 PC % 5 5 PC1 1 PC % 6 6 PC1 1 PC

431 Table 5B-3. Calculation of Capital, Annual, and Unit Treatment Costs Line Item Calculation Calculation of of Capital, Capital, Annual, and Unit Treatment Costs Costs 600 gpm system, 2000 µg/l to 20 µg/l, Option 1: Series Operations 600 gpm system, 2000 µg/l to 20 µg/l, Option 1: Series Operation Cost Treatment & Regeneration Units $603,750 Piping, Valves, Electrical (30% of resin vessel & steam regeneration equipment costs) $165,333 Site Work (10%) $60,375 SUBTOTAL $829,458 Contractor O&P (15%) $124,419 SUBTOTAL $953,877 Engineering (15%) $143,082 SUBTOTAL $1,096,958 Contingency (20%) $219,392 Resin Costs with a 10% mark-up $1,320,000 TOTAL CAPITAL $2,636,350 Amortized Annual Capital $212,454 Annual O&M $180,124 TOTAL ANNUAL COST $392,578 Annual Flow Treated (kgal) 315,360 UNIT TREATMENT COST ($/kgal)* $1.24 *To convert unit treatment cost to $/acre-ft, multiply by 326. Amortization based on a 30-year period at a 7% discount rate. Capital expenses include: equipment, piping, valves, and electrical components (30%), site work (10%), contractor O&P (15%), engineering (15%), and contingency (20%). O&M Costs include: 1. Power costs at $0.80/kWhr 2. Labor costs at $80.00/hr 3. Analytical costs at $200 per sample. See table B-4 for more details on the labor and analytical cost assumptions. 409

432 Table 5B-4. Labor and Analytical Assumptions. Series Operation Carousel Operation Flow Rate Influent Concentration (µg/l) # Weekly Samples # Annual (Series Operation) 1 Regen. Samples 2 Analytical Costs (Series Operation) 3 # Weekly Samples (Carousel Operation) 1 # Annual Regen. Samples 2 Analytical Costs (Carousel Operation) 3 % Time Labor Required for Operation 4 Costs $173, $200, Time $320, $119, $120,800 $240, $107, $107,000 $160, $28, $30, Time $120, $22, $22,600 $120, $21, $21,200 $80, $29, $31, Time $80, $22, $23,200 $80, $21, $21,400 $40, $35, $40, Time $40, $24, $24,400 $40, $21, $21,400 $40,000 1) Includes 1 sample at influent and 1 sample at effluent. 2) Includes approximately 3 samples each time the resin system is regenerated. 3) Samples cost $200/sample for ) Labor includes sampling, boiler maintenance, regeneration system upkeep; see Table 6-3a and 6-3b for annual regeneration times 5) Labor rates assume $80/hour or $160,000/year 410