EPA/540/AR-93/501July 1993

perox-pureTM ChemicalOxidation Technology

Peroxidation Systems, Inc.

Applications Analysis Report

Risk Reduction Engineering LaboratoryOffice of Research and Development

U.S. Environmental Protection AgencyCincinnati, Ohio 45268

431 Printed on Recycled Paper

Notice

The information in this document has been prepared for the U.S. Environmental Protection Agency(EPA) Superfund Innovative Technology Evaluation (SITE) program under Contract No. 68-CO-0047.This document has been subjected to EPA peer and administrative reviews and has been approved forpublication as an EPA document. Mention of trade names or commercial products does not constitutean endorsement or recommendation for use.

ii

Foreword

The Superfund Innovative Technology Evaluation (SITE) program was authorized in the 1986Superfund Amendments and Reauthorization Act. The program is a joint effort between theU.S. Environmental Protection Agency’s (EPA) Office of Research and Development and Office ofSolid Waste and Emergency Response. The purpose of the program is to assist the development ofinnovative hazardous waste treatment technologies, especially those that offer permanent remedies forcontamination commonly found at Superfund and other hazardous waste sites. The SITE programevaluates new treatment methods through technology demonstrations designed to provide engineeringand cost data for selected technologies.

A field demonstration was conducted under the SITE program to evaluate the perox-pureTM chemicaloxidation technology’s ability to treat groundwater contaminated with volatile organic compounds. Thetechnology demonstration took place at the Lawrence Livermore National Laboratory site in Tracy,California. The demonstration effort was directed to obtain information on the performance and costof the technology and to assess its use at this and other uncontrolled hazardous waste sites.Documentation consists of two reports: (1) a Technology Evaluation Report, which describes fieldactivities and laboratory results, and (2) this Applications Analysis Report, which interpretsthe data anddiscusses the potential applicability of the technology.

A limited number of copies of this report will be available at no charge from EPA’s Center forEnvironmental Research Information, 26 West Martin Luther King Drive, Cincinnati, Ohio 45268.Requests should include the EPA document number found on the report’s cover. When the limitedsupply is exhausted, additional copies can be purchased from the National Technical Information Service,Ravensworth Building, Springfield, Virginia 22161, (703) 487-4600. Reference copies will be availableat EPA libraries in the Hazardous Waste Collection.

E. Timothy Oppelt, Director

Risk Reduction Engineering Laboratory

Abstract

This report evaluates the perox-pureTM chemical oxidation technology’s ability to remove volatileorganic compounds (VOC) and other organic contaminants present in liquid wastes. This report alsopresents economic data from the Superfund Innovative Technology Evaluation (SITE) demonstrationand three case studies.

The perox-pureTM chemical oxidation technology was developed by Peroxidation Systems, Inc. (PSI),to destroy dissolved organic contaminants in water. The technology uses ultraviolet (UV) radiation andhydrogen peroxide to oxidize organic compounds present in water at parts per million levels or less.This treatment technology produces no air emissions and generates no sludge or spent media thatrequire further processing, handling, or disposal. Ideally, the end products are water, carbon dioxide,halides (for example, chloride), and in some cases, organic acids. The technology uses medium-pressure,mercury-vapor lamps to generate UV radiation. The principal oxidants in the system, hydroxyl radicals,are produced by direct photolysis of hydrogen peroxide at UV wavelengths.

The perox-pureTM chemical oxidation technology was demonstrated under the SITE program atLawrence Livermore National Laboratory Site 300 in Tracy, California. Over a 3-week period inSeptember 1992, about 40,000 gallons of VOC-contaminated groundwater was treated in the perox-pureTM system. For the SITE demonstration, the perox-pureTM system achieved trichloroethene (TCE)and tetrachloroethene (PCE) average removal efficiencies of about 99.7 and 97.1 percent, respectively.In general, the perox-pureTM system produced an effluent that contained (1) TCE, PCE, andl,l-dichloroethane (DCA) below detection limits and (2) chloroform and l,l,l-trichloroethane (TCA)slightly above detection limits. The system also achieved chloroform, DCA, and TCA average removalefficiencies of 93.1, 98.3, and 81.8 percent, respectively. The treatment system effluent met Californiadrinking water action levels and federal drinking water maximum contaminant levels for TCE, PCE,chloroform, DCA, and TCA at the 95 percent confidence level.

The results from three case studies are also summarized in this report. All three case studiesrepresent full-scale, currently operating commercial installations of perox-pureTM chemical oxidationsystems. The contaminants of concern in these case studies include acetone, isopropyl alcohol (IPA),TCE, and pentachlorophenol (PCP). In the first case study, the perox-pureTM system treated industrialwastewater containing 20 milligrams per liter (mg/L) of acetone and IPA; the effluent met the dischargelimit of 0.5 mg/L for each compound. In the second case study, the perox-pureTM system treatedgroundwater that was used as a municipal drinking water source. The groundwater initially contained150 micrograms per liter (pg/L) of TCE, After treatment, the effluent TCE level was 0.5 rg/L, wellbelow the TCE drinking water standard of 5 pg/L. In the third case study, the perox-pureTM systemtreated groundwater at a chemical manufacturing facility. The groundwater contained 15 mg/L of PCP;after treatment the effluent achieved the target effluent PCP level of 0.1 mg/L.

Potential sites for applying this technology include Superfund and other hazardous waste sites wheregroundwater or other liquid wastes are contaminated with organic compounds. Economic data indicatethat groundwater remediation costs for a 50-gallon per minute perox-pureTM system could range fromabout $7 to $11 per 1,000 gallons, depending on contaminated groundwater characteristics. Of these,perox-pureTM system direct treatment costs could range from about $3 to $5 per 1,000 gallons.

iv

c

Contents

Notice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiForeword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i i iAbstract ........................................................... ivAcronyms, Abbreviations, and Symbols ............................................. ixConversion Factors .................................................... xiAcknowledgements .................................................... xii

1. Executive Summary ...................................................... 11.1 Introduction ................................................. 11.2 Overview of the SITE Demonstration ................................. 11.3 Results from the SITE Demonstration ................................ 21.4 Results from Case Studies ....................................... 31.5 Waste Applicability ........................................... 31.6 Economics .................................................. 3

2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.1 Purpose, History, and Goals of the SITE Program ..................... 52.2 Documentation of the SITE Demonstration Results ..................... 62.3 Purpose of the Applications Analysis Report .......................... 62.4 Technology Description ......................................... .6

2.4.1 Treatment Technology .................................... 72.4.2 System Components and Function ........................... 82.4.3 Innovative Features of the Technology ........................ 8

2.5 Key Contacts.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3. Technology Applications Analysis .......................................... 113.1 Effectiveness of the perox-pureTM Technology ......................... 11

3.1.1 SITE Demonstration Results ............................... 113.1.2 Results of Other Case Studies .............................. 12

3.2 Factors Influencing Performance .................................. 133.2.1 Influent Characteristics ................................... 133.2.2 Operating Parameters .................................... 133.2.3 Maintenance Requirements ................................ 14

3.3 Site Characteristics ............................................ 143.3.1 Support Systems ........................................ 143.3.2 Site Area and Preparation ................................. 153.3.3 Site Access ............................................. 153.3.4 Climate ............................................... 1.53.3.5 Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153.3.6 Services and Supplies .................................... 15

3.4 Material Handling Requirements .................................. 153.4.1 Pretreatment Materials .................................... 163.4.2 Treated Water ......................................... 16

3.5 Personnel Requirements ........................................ 163.6 Potential Community Exposures .................................... 16

V

Contents (continued)

3.7 Potential Regulatory Requirements ................................. 173.7.1 Comprehensive Environmental Response, Compensation,

and Liability Act ....................................... 173.7.2 Resource Conservation and Recovery Act .................... 173.7.3 Clean Water Act ................................, ....... 193.7.4 Safe Drinking Water Act ................................. 193.7.5 Toxic Substances Control Act ............................. 193.7.6 Mixed Waste Regulations ................................ 193.7.7 Federal Insecticide, Fungicide, and Rodenticide Act ............. 193.7.8 Occupational Safety and Health Act .......................... 20

4. Economic Analysis ...................................................... 214.1 Basis of Economic Analysis ...................................... 214.2 Cost Categories ............................................... 24

4.2.1 Site Preparation Costs .................................... 2.54.2.2 Permitting and Regulatory Requirements Costs ................. 254.2.3 Capital Equipment Costs .................................. 254.2.4 Startup Costs .......................................... 274.2.5 Labor Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2.6 Consumables and Supplies Costs ............................ 274.2.7 Utilities Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284.2.8 Effluent Treatment and Disposal Costs ....................... 294.2.9 Residuals and Waste Shipping and Handling Costs ............... 294.2.10 Analytical Services Costs .................................. 294.2.11 Maintenance and Modifications Costs ........................ 294.2.12 Demobilization Costs .................................... 30

5. References ....................................................................................................................... 31

Appendix A . Vendor Claims for the Technology .................................... 33A.1 Introduction ................................................. 33A.2 Description of the perox-pureTM System .............................. 33A.3 perox-pureTM Systems ........................................... 34

A.4 Design Improvements ..................... . ................... 34A.5 Pretreatment ................................................. 36A.6 perox-pureTM Applications ........................................ 36A.7 Advantages over Carbon Adsorption and Air Stripping Technologies ........ 36A.8 Other Advantages ............................................. 38A.9 Technology Combinations ....................................... 38

Appendix B . SITE Demonstration Results ........................................39B.lB.2B.3

B.4

B.5

B.6B.7

Site Description ............................................... 39Site Contamination Characteristics ................................. 39Review of SITE Demonstration ................................... 41B.3.1 Site Preparation ........................................ 41B.3.2 Technology Demonstration ................................. 42B.3.3 Site Demobilization ...................................... 43Experimental Design ............................................43B.4.1 Testing Approach ....................................... 44B.4.2 Sampling and Analytical Procedures .......................... 44Review of Treatment Results ..................................... 45B.5.1 Summary of the Results for Critical Parameters ................. 45B.5.2 Summary of Results for Noncritical Parameters- ................. 56Conclusions .................................................. 58References .................................................. 58

vi

Contents (continued)

Appendix C . Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59C.1 Wastewater Treatment System, Florida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

C.l.1 Site Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59C.1.2 System Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59C.1.3 costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

C.2 Municipal Drinking Water System, Arizona .......................... 60C.2.1 Site Conditions ........................................ 60C.2.2 System Performance ..................................... 60C.2.3 Costs ................................................ 60

C.3 Chemical Manufacturing Company, Washington ....................... 60C.3.1 Site Conditions ......................................... 60C.3.2 System Performance ..................................... 60C.3.3 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

vii

Acronyms, Abbreviations, and Symbols

ACL

AOX

BTEXBTXccl,CDEPCERCLACFRCWADCAl,l-DCE1,2-DCEDIMPDOEEPAFIFRAFS“FGCgpdgpmGSAH,O,hvIPAkWkWhLLNLMCLMeClMEKIrg/Lmg/LMSNPDESOH*O&MORDOSHAOSWERPAHPCB

Applications Analysis ReportAlternate concentration limitAtomic Energy ActAdsorbable organic halideApplicable or relevant and appropriate requirementBenzene, toluene, ethylbenzene, and xyleneBenzene, toluene, and xyleneCarbon tetrachlorideDepartment of Environmental Protection, State of ConnecticutComprehensive Environmental Response, Compensation, and Liability ActCode of Federal RegulationsClean Water Actl,l-dichloroethanel,l-dichloroethene1,2-dichloroetheneDiisopropyl methylphosphonateU.S. Department of EnergyU.S. Environmental Protection AgencyFederal Insecticide, Fungicide, and Rodenticide ActFeasibility studyDegree FahrenheitGas chromatographyGallons per dayGallons per minuteGeneral Services AreaHydrogen peroxideUltraviolet radiationIsopropyl alcoholKilowattKilowatt-hourLawrence Livermore National LaboratoryMaximum contaminant levelMethylene chlorideMethyl ethyl ketoneMicrograms per literMilligrams per literMass spectrometryNational Pollutant Discharge Elimination SystemHydroxyl radicalOperation and maintenanceOffice of Research and DevelopmentOccupational Safety and Health ActOffice of Solid Waste and Emergency ResponsePolynuclear aromatic hydrocarbonPolychlorinated biphenyl

ix

PCEPCPPOCPOTWPPEppmPSIQA/QCRIRCRASARASDWASITES V O CTCTCATCETERTICTOCTOXTSCAUCLU VV O C

Acronyms, Abbreviations, and Symbols (continued)

TetrachloroethenePentachlorophenolPurgeable organic carbonPublicly-owned treatment worksPersonal protective equipmentParts per millionPeroxidation Systems, Inc.Quality assurance/quality controlRemedial investigationResource Conservation and Recovery ActSuperfund Amendments and Reauthorization ActSafe Drinking Water ActSuperfund Innovative Technology EvaluationSemivolatile organic compoundTotal carbonl,l,l-trichloroethaneTrichloroetheneTechnology Evaluation ReportTentatively identified compoundTotal organic carbonTotal organic halideToxic Substances Control ActUpper confidence limitUltravioletVolatile organic compound

X

Conversion Factors

Length:

Area:

Volume:

Mass:

Energy:

Power:

Temperature:

To Convert From

inch

foot

mile

square foot

acre

gallon

cubic foot

pound

kilowatt-hour

kilowatt

(*Fahrenheit - 32)

To Multiply By

centimeter

meter

kilometer

2.54

0.305

1.61

square meter

square meter

0.0929

4,047

liter

cubic meter

3.78

0.0283

kilogram 0.454

megajoule 3.60

horsepower 1.34

*Celsius 0.556

xi

Acknowledgements

This report was prepared under the direction and coordination of Ms. Norma Lewis, U.S.Environmental Protection Agency (EPA) Superfund Innovative Technology Evaluation (SITE) ProjectManager and Emerging Technology Section Chief in the Risk Reduction Engineering Laboratory(RREL), Cincinnati, Ohio. Contributors and reviewers for this report were Messrs. Carl Chen, GordonEvans, John Ireland, and Ron Turner of EPA RREL, Cincinnati, Ohio; Mr. Chris Giggy of PeroxidationSystems, Inc., Tucson, Arizona; Ms. Lida Tan of EPA Region IX, San Francisco, California; MT. KaiSteffens of PROBIOTEC, Duren, Germany; Mr. Shyam Shukla of Lawrence Liver-more NationalLaboratory (LLNL), Liver-more, California; and Mr. Geoffrey Germann of Engineering-Science, Inc.,Fairfax, Virginia.

This report was prepared for EPA’s SITE program by Dr. Kirankumar Topudurti, Mr. MichaelKeefe, Mr. Patrick Wooliever, Mr. Jeffrey Swano, and Ms. Carla Buriks of PRC EnvironmentalManagement, Inc. (PRC). Special acknowledgement is given to Mr. John Greci of LLNL for his

invaluable support during the demonstration and to Ms. Carol Adams, Ms. Korreen Ball, Ms. ReginaBergner, and Mr. Tobin Yager of PRC for their editorial, graphic, and production assistance during thepreparation of this report.

xii

Section 1Executive Summary

1.1 Introduction

The perox-pureTM chemical oxidation technology,developed by Peroxidation Systems, Inc. (PSI), was evaluatedunder the U.S. Environmental Protection Agency SuperfundInnovative Technology Evaluation (SITE) program. Theperox-pure technology demonstration was conducted atLawrence Livermore National Laboratory (LLNL) Site 300in Tracy, California, over a 3-week period in September1992.

The perox-pure chemical oxidation technology isdesigned to destroy dissolved organic contaminants in water.The technology uses ultraviolet (UV) radiation and hydrogenperoxide to oxidize organic compounds present in water atparts per million levels or less. This treatment technologyproduces no air emissions and generates no sludge or spentmedia that require further processing, handling, or disposal.Ideally, end products are water, carbon dioxide, halides (forexample, chloride), and in some cases, organic acids. Thetechnology uses medium-pressure, mercury-vapor lamps togenerate UV radiation. The principal oxidants in the system,hydroxyl radicals, are produced by direct photolysis ofhydrogen peroxide at UV wavelengths.

The perox-pure chemical oxidation treatment system(Model SSB-30) used for the SITE technology demonstrationwas assembled from the following portable, skid-mountedcomponents: a chemical oxidation unit, a hydrogen peroxidefeed module, an acid feed module, a base feed module, aUV lamp drive, and a control panel. The oxidation unit hassix reactors in series with one 5-kilowatt UV lamp in eachreactor; the unit has a total volume of 15 gallons. The UVlamp is mounted inside a UV-transmissive quartz tube in thecenter of each reactor so that water flows through the spacebetween the reactor walls and the quartz tube, Circularwipers are mounted on the quartz tubes to periodicallyremove any solids that have accumulated on the tubes.

The perox-pure system requires little attention duringoperation and can be operated and monitored remotely, ifneeded. Remotely monitored systems can be connected todevices that automatically dial a telephone to notify

1

responsible parties at remote locations of alarm conditions.Remotely operated and monitored systems are hard-wiredinto centrally located control panels or computers throughprogrammable logic controllers.

The technology demonstration had the following primaryobjectives: (1) d te ermine the ability of the perox-puresystem to remove volatile organic compounds (VOC) fromgroundwater at the LLNL site under different operatingconditions, (2) determine whether treated groundwater metapplicable disposal requirements at the 95 percentconfidence level, and (3) gather information necessary toestimate treatment costs, including process chemical dosagesand utility requirements. The secondary objective for thetechnology demonstration was to obtain information on thepresence and types of by-products formed during treatment.

The purpose of this report is to present informationfrom the SITE demonstration and several case studies thatwill be useful for implementing the perox-pure”’ chemicaloxidation technology at Superfund and ResourceConservation and Recovery Act hazardous waste sites.Section 2 presents an overview of the SITE program,describes the perox-pure- technology, and lists key contacts.Section 3 discusses information relevant to the technology’sapplication, including pretreatment and posttreatmentrequirements, site characteristics, operating and maintenancerequirements, potential community exposures, and potentiallyapplicable environmental regulations. Section 4 summarizesthe costs associated with implementing the technology.Section 5 includes a list of references. Appendices Athrough C include the vendor claims for the technology, asummary of the SITE demonstration results, and summariesof three case studies, respectively.

1.2 Overview of the SITE Demonstration

Shallow groundwater at the LLNL site was selected asthe waste stream for evaluating the perox-pure- chemicaloxidation technology. About 40,000 gallons of groundwatercontaminated with VOCsswas treated during thedemonstration. The principal groundwater contaminantswere trichloroethene (TCE) and tetrachloroethene (PCE),

which were present at concentrations of about 1,000 and 100micrograms per liter @g/L), respectively. Groundwater waspumped from two wells into a 7,500-gallon bladder tank tominimize variability in influent characteristics. In addition,cartridge filters were used to remove suspended solidsgreater than 3 micrometers in size from the groundwaterbefore it entered the tank. Treated groundwater was storedin two 20,000-gallon steel tanks before being discharged.

The technology demonstration was conducted in threephases. Phase 1 consisted of eight runs using rawgroundwater, Phase 2 consisted of four runs using spikedgroundwater, and Phase 3 consisted of two runs using spikedgroundwater to evaluate the effectiveness of quartz tubecleaning. These phases are described below.

The principal operating parameters for the perox-pure”’system, hydrogen peroxide dose, influent pH, and flow rate(which determines the hydraulic retention time), were variedduring Phase 1 to observe treatment system performanceunder different operating conditions. Preferred operatingconditions, those under which the concentrations of spikedgroundwater effluent VOCs would be reduced to belowtarget levels, were then determined for the system.

Phase 2 involved spiked groundwater and reproducibilitytests. Groundwater was spiked with about 200 to 300 rg/Leach of chloroform; 1,1-dichloroethane (DCA); and l,l,l-trichloroethane (TCA). These compounds were chosenbecause they are difficult to oxidize and because they werenot present in the groundwater at high concentrations. Thisphase was also designed to evaluate the reproducibility oftreatment system performance at the preferred operatingconditions determined in Phase 1.

During Phase 3, the effectiveness of quartz tube wiperswas evaluated by performing two runs using spikedgroundwater and scaled and clean quartz tubes.

During the demonstration, samples were collected atseveral locations, including the treatment system influent;effluent from Reactors 1, 2, and 3; and the treatment systemeffluent. Samples were analyzed for VOCs, semivolatileorganic compounds, total organic carbon (TOC), totalcarbon, purgeable organic carbon (POC), total organichalides (TOX), adsorbable organic halides (AOX), metals,pH, alkalinity, turbidity, temperature, specific conductance,hydrogen peroxide residual, and hardness, as applicable. Inaddition, samples of influent to Reactor 1 and treatmentsystem effluent were collected and analyzed for acute toxicityto freshwater organisms. In the bioassay tests,Ceriodaphnia dubia (water fleas) and Pimephalespromelas (fathead minnows) were used as the testorganisms. Hydrogen peroxide, acid, and base solutionswere also sampled and analyzed to verify concentrations.

1.3 Results from the SITE Demonstration

For the spiked groundwater, PSI determined thefollowing preferred operating conditions: (1) influenthydrogen peroxide level of 40 milligrams per liter (mg/L);(2) hydrogen peroxide level of 25 mg/L in the influent toReactors 2 through 6; (3) an influent pH of 5.0; and (4) aflow rate of 10 gallons per minute (gpm). At theseconditions, the effluent TCE, PCE, and DCA levels weregenerally below detection limit (5 pg/L) and effluentchloroform and TCA levels ranged from 15 to 30 pg/L. Theaverage removal efficiencies, for TCE, PCE, chloroform,DCA, and TCA were about 99.7, 97.1, 93.1, 98.3, and81.8 percent, respectively.

For the unspiked groundwater, the effluent TCE andPCE levels were generally below detection limit (1 pg/L)with corresponding removal efficiencies of about 99.9 and99.7 percent. The effluent TCA levels ranged from 1.4 to6.7 pg/L with removal efficiencies ranging from 35 to84 percent.

The perox-pure system effluent met California drinkingwater action levels and federal drinking water maximumcontaminant levels for TCE, PCE, chloroform, DCA, andTCA at the 95 percent confidence level.

The quartz tube wipers were effective in keeping thetubes clean, and appeared to reduce the adverse effectscaling has on contaminant removal efficiencies.

Bioassay tests showed that the perox-pureTM systemeffluent was acutely toxic to freshwater test organisms,although the influent was not toxic. Comparison of effluenttoxicity data with that of hydrogen peroxide residual in theeffluent (10.5 mg/L) indicated that effluent toxicity may bedue to hydrogen peroxide residual rather than perox-pureTM

treatment by-products. Additional studies are needed todraw any conclusion on the effluent toxicity.

TOX removal efficiencies ranged from 93 to 99 percent,AOX removal efficiencies ranged from 95 to 99 percent.

For spiked groundwater, during reproducibility runs, thesystem achieved average removal efficiencies of 38 percentand about 93 percent for TOC and POC, respectively.

The temperature ,of groundwater increased at a rate of12 “F per minute of UV exposure in the perox-pureTM

system. Since the oxidation unit is exposed to thesurrounding environment, the temperature increase may varydepending upon the ambient temperature or otheratmospheric conditions.

2

1.4 Results from Case Studies 1.5 Waste Applicability

Information on the perox-pure’” t e chno logy ’ sperformance at three facilities was evaluated to provideadditional performance data. The three case studiesrepresent full-scale, currently operating commercialinstallations of perox-pure” systems. The contaminants ofconcern in these case studies include acetone, isopropylalcohol (IPA), TCE, and pentachlorophenol (PCP). Thecase studies are briefly summarized below.

The first case study involves wastewater treatment at theKennedy Space Center in Florida. At this facility, a liquid-phase carbon adsorption system had originally been installedto treat wastewater containing acetone and IPA at levels of20 mg/L. The treatment facility discharge requirement forboth acetone and IPA is 0.5 mg/L. Because the carbonadsorption system did not achieve the discharge requirement,the perox-pure” chemical oxidation technology was selectedto replace the carbon adsorption system. At this facility, theperox-pure”’ system initially treated wastewater in batches ofabout 5,000 gallons. Later, the system was converted to aflow-through mode. Currently, the system treats wastewaterat a flow rate of 5 gpm and produces an effluent that meetsthe discharge standard.

The second case study involves treating groundwaterused as a municipal drinking water source in Arizona. In1989, a municipal drinking water well in Arizona was foundto contain 50 to 400 pg/L of TCE. The well, capable ofproducing 2,000 gpm, was taken out of service while severaltreatment options were evaluated. Because the well islocated on a city lot in the middle of a large residential area,the city preferred using a low-visibility, quiet treatmentmethod that could consistently destroy TCE toconcentrations below the drinking water standard of 5 pg/L.Given these requirements, chemical oxidation using theperox-pure” system was selected. Currently, the systemtreats groundwater at a flow rate of 135 gpm and producestreated groundwater containing only 0.5 pg/L of TCE.

The third case study deals with treatment of PCP-contaminated groundwater at a chemical manufacturingcompany in Washington. Groundwater at the site was foundto contain about 15I.5 mg/L of PCP. In 1988, PSI installed aperox-pure” system at the site under a Full ServiceAgreement. Because the groundwater was also found tocontain high levels of iron (200 mg/L) and carbonates thatcould scale the quartz tubes and impair the treatmentefficiency, PSI recommended groundwater pretreatment.Pretreatment consists of iron oxidation and removal,followed by pH adjustment to about 5. Pretreatedgroundwater is then treated by a perox-pure” systemequipped with automatic quartz tube cleaners. Currently,the perox-pure” system treats groundwater at a flow rate of70 gpm and produces an effluent that meets the dischargestandard of 0.1 mg/L.

Potential sites for applying the perox-pure” technologyinclude Superfund and other hazardous waste sites wheregroundwater or other liquid wastes are contaminated withorganic compounds. The technology has been used to treatlandfill leachate, groundwater, and industrial wastewater, allcontaining a variety of organic contaminants, includingchlorinated solvents, pesticides, polynuclear aromatichydrocarbons, and petroleum hydrocarbons. In someapplications, where th.e contaminant concentration washigher than about 500 mg/L, the .perox-pure”’ system wascombined with other treatment technologies for costeffectiveness.

1.6 Economics

using information obtained from the SITEdemonstration, an economic analysis was performed toexamine 12 separate cost categories for perox-pure” systemstreating about 260 million gallons of contaminatedgroundwater at a Superfund site. This analysis examined twocases based on groundwater characteristics. In Case 1, thegroundwater was assumed to have five contaminants, two ofwhich are easy to oxidize (TCE and PCE) and the remainingthree are difficult to oxidize (chloroform, DCA, and TCA).In Case 2, the groundwater was assumed to have only twocontaminants that are easy to oxidize (TCE and PCE). Foreach case, costs for three different flow rates (10, 50, and100 gpm) were estimated. Detailed economic analysis forthe three flow rates of each case is included in Section 4.Costs for the 50-gpm flow rate scenario for each case aresummarized below.

For Case 1, capital costs are estimated to be about$906,000 of which the perox-pure” system direct capital costis $185,000. Annual operation and maintenance (O&M)costs are estimated to be about $188,000 of whichperox-pure” system direct O&M costs are $125,000.Groundwater remediation costs to treat 1,000 gallons ofcontaminated water are estimated to be about $11 of whichperox-pure” system direct treatment costs are $5.

For Case 2, capital costs are estimated to be about$776,000 of which the perox-pure” system direct capital costis $55,000. Annual O&M costs are estimated to be about$111,000 of which perox-pure”’ system direct O&M costs are$61,000. Groundwater remediation costs to treat1,000 gallons of contaminated water are estimated to beabout $7 of which perox-pure”’ system direct treatment costsare $3.

The case studies included in Appendix C have minimalcost data. According to PSI, in the case studies, the totalO&M costs ranged from $0.28 to S3.90 per 1,000 gallons ofwater treated.

3

Section 2Introduction

This section provides information about the SuperfundInnovative Technology Evaluation (SITE) program, discussesthe purpose of this report, and describes the perox-purechemical oxidation technology developed by PeroxidationSystems, Inc. (PSI), of Tucson, Arizona. The perox-puretechnology is designed to treat waters contaminated withorganic compounds. For additional information about theS I TE program, the perox-pure technology, or thedemonstration site, key contacts are listed at the end of thissection.

2.1 Purpose, History, and Goals of the SITEProgram

The Superfund Amendments and Reauthorization Act(SARA) of 1986 mandates that the U.S. EnvironmentalProtection Agency (EPA) select, to the maximum extentpracticable, remedial actions at Superfund sites that createpermanent solutions (as opposed to land-based disposal) forcontamination that affects human health and theenvironment. In doing SO, EPA is directed to use alternativeor resource recovery technologies. In response, EPA’sOffice of Research and Development (ORD) and Office ofSolid Waste and Emergency Response (OSWER) establishedthree programs: (1) a program to accelerate the use of newor innovative technologies to clean up Superfund sitesthrough field demonstrations; (2) a program to foster thefurther research and development of treatment technologiesthat are at the laboratory or pilot scale; and (3) a programto demonstrate and evaluate new or innovative measurementand monitoring technologies. Together, these threecomponents make up the SITE program.

The primary purpose of the SITE program is to enhancethe development and demonstration, and thereby establishthe commercial availability, of innovative technologiesapplicable to Superfund sites. The SITE program hasestablished the following goals:

l Ident i fy and remove impediments to thedevelopment and commercial use of alternativetechnologies

Demonstrate promising innovative technologies toestablish reliable performance and cost informationfor site characterization and remediation decisions

Develop procedures and policies that encourage theselection of alternative treatment remedies atSuperfund sites

Develop a program that promotes and supportsemerging technologies

EPA recognizes that a number of forces inhibit theexpanded use of new and alternative technologies atSuperfund sites. The SITE program’s goals are designed toidentify the most promising new technologies, developpertinent and useful data of known quality about them, andmake the data available to Superfund decision makers. Anadditional goal is to promote the development of emerginginnovative technologies from the laboratory- or bench-scaleto the full-scale stage.

Implementation of the SITE program is a significantongoing effort involving ORD, OSWER, various EPARegions, and private sector business concerns, includingtechnology developers and parties responsible for siteremediation. The technology selection process and thedemonstration program together provide objective andcarefully controlled testing of field-ready technologies.Through government publications, the SITE programdisseminates testing results to Superfund decision makers foruse in evaluating the applicability of technologies to site-specific remediation efforts.

The demonstration process collects the followinginformation for Superfund decision makers to consider whenmatching technologies with wastes, media, and sites requiringremediation:

l The technology’s effectiveness based on fielddemonstration sampling and analytical datacollected during the demonstration

The po ten t ia l need fo r p re t rea tment andposttreatment of wastes

The site-specific wastes and media to which thetechnology can be applied

Potential site-specific system operating problems aswell as possible solutions

The approx imate cap i t a l , ope ra t ing , andmaintenance costs

The projected long-term operation and maintenance(O&M) costs

Innovative technologies chosen for a SITE demonstrationmust be pilot- or full-scale applications and must offer someadvantage over existing technologies. Mobile technologiesare of particular interest. Each year the SITE programsponsors demonstrations of approximately 10 technologies.

2.2 Documentation of the SITE DemonstrationResults

The results of each SITE demonstration are reported intwo documents: the Applications Analysis Report (AAR)and the Technology Evaluation Report (TER). The AAR isintended for decision makers responsible for implementingspecific remedial actions and is primarily used to assist inscreening the demonstrated technology as an option for aparticular cleanup situation. The purpose of the AAR isdiscussed in the following section.

The TER is published separately from the AAR andprovides a comprehensive description of the demonstrationand its results. A likely audience for the TER includesengineers responsible for evaluating the technology forspecific site and waste situations. These technical evaluatorsseek to understand, in detail, the performance of thetechnology during the demonstration, as well as advantages,disadvantages, and costs of the technology for the givenapplication. This information is used to produce conceptualdesigns in sufficient detail to enable preliminary costestimates for the demonstrated technology. If the candidatetechnology appears to meet the needs of site engineers, amore thorough analysis will be conducted based on the TER,the AAR, and other site-specific information obtained fromremedial investigations.

2.3 Purpose of the Applications Analysis Report

Information presented in the AAR is intended to assistSuperfund decision makers in screening specific technologiesfor a particular cleanup situation. The report discusses theadvantages, disadvantages, and limitations of the technology.Costs of the technology for different applications areestimated based on available data for pilot- and full-scale

applications. The report discusses factors that have a majorimpact on cost and performance, such as site and wastecharacteristics.

To encourage the general use of demonstratedtechnologies, EPA will evaluate the applicability of eachtechnology for specific sites and wastes, other than thosealready tested, and will study the estimated costs of theapplications. The results are presented in the AAR. ThisAAR synthesizes available information on PSI’s perox-purechemical oxidation technology and draws reasonableconclusions regarding its range of applicability. This AARwill be useful to decision makers considering using theperox-pure technology. It represents a critical step in thedevelopment and commercialization of the treatmenttechnology.

Each SITE demonstration evaluates a technology’sperformance in treating an individual waste at a particularsite. To obtain data with broad applicability, priority is givento technologies that treat wastes frequently found atSuperfund sites. However, in many cases, wastes at othersites will differ in some way from the waste at thedemonstration site. Therefore, the successful demonstrationof a technology at one site does not ensure its success atother sites. Data obtained from the demonstration mayrequire extrapolation to estimate total operating ranges overwhich the technology performs satisfactorily. Anyextrapolation of demonstration data should also be based onother available information from case studies about thetechnology.

The amount of available data for the evaluation of aninnovative technology varies widely. Data may be limited tolaboratory tests on synthetic wastes or may include performance data on actual wastes treated by pilot- or full-scale treatment systems. In addition, only limitedconclusions regarding Superfund applications can be drawnfrom a single field demonstration. A successful fielddemonstration does not necessarily ensure that a technologywill be widely applicable or that it will be fully developed tocommercial scale.

2.4 Technology Description

In April 1991, EPA learned that PSI was contracted byLawrence Livermore National Laboratory (LLNL) toperform pilot-scale studies as part of remediation activitiesat the LLNL site. At that time, EPA and PSI discussed thepossibility of PSI participating in the SITE program todemonstrate how the perox-pureTM chemical oxidationtechnology could be used to treat contaminated groundwaterat Site 300 of LLNL in Tracy, California. EPA subsequentlyaccepted the perox-pure technology into the SITEdemonstration program. Through a cooperative effortbetween EPA ORD, EPA Region IX, LLNL, and PSI, the

6

perox-pureTM technology was demonstrated at the LLNL siteunder the SITE program.

2.4.1 Treatment Technology

The perox-pure chemical oxidation treatment systemwas developed by PSI to destroy dissolved organiccontaminants in water. The technology uses ultraviolet (UV)radiation and hydrogen peroxide to oxidize organiccompounds present in water at parts per million (ppm) levelsor less. In broad terms, oxidation is a chemical change inwhich electrons are lost by an atom or a group of atoms.Oxidation of an atom or group of atoms is alwaysaccompanied by the reduction of another atom or group ofatoms. Reduction is a chemical change in which electronsare gained by an atom or group of atoms. The atom orgroup of atoms that has lost electrons has been oxidized, andthe atom or group of atoms that has gained electrons hasbeen reduced. The reduced atom or group of atoms iscalled an oxidant. Oxidation and reduction always occursimultaneously, and the total number of electrons lost in theoxidation must equal the number of electrons gained in thereduction. In the perox-pureTM technology, organiccontaminants in water are oxidized by hydroxyl radicals, apowerful oxidant produced by UV radiation and hydrogenperoxide, Subsequently, the organic contaminants arebroken down into carbon dioxide, water, halides, and insome cases, organic acids.

A variety of organic contaminants can be effectivelyoxidized by the combined use of (1) UV radiation andhydrogen peroxide, (2) UV radiation and ozone, or (3)ozone and hydrogen peroxide. The principal oxidants in theperox-pure system, hydroxyl radicals, are produced bydirect UV photolysis of the hydrogen peroxide added tocontaminated water. The perox-pureTM system generates UVradiation by using medium-pressure, mercury-vapor lamps.

In principle, the most direct way to generate hydroxylradicals (OH*) is to cleave hydrogen peroxide (H,O,)through photolysis. The photolysis of hydrogen peroxideoccurs when UV radiation (hu) is applied, as shown in thefollowing reaction:

H,O, + hu --, 2 OH*

Thus, photolysis of hydrogen peroxide results in aquantum yield of two hydroxyl radicals (OH*) formed perquantum of radiation absorbed. This ratio of hydroxylradicals generated from the photolysis of hydrogen peroxideis high. Unfortunately, at 253.7 nanometers, the dominantemission wavelength of low-pressure UV lamps, theabsorptivity (or molar extinction coefficient) of hydrogenperoxide is only 19.6 liters per mole-centimeter. Thisabsorptivity is relatively low for a primary absorber in aphotochemical process. Because of the low absorptivityvalue for hydrogen peroxide, a high concentration of residual

hydrogen peroxide must be present in the treatment mediumto generate a sufficient concentration of hydroxyl radicals.According to PSI, the perox-pure system overcomes thislimitation by using medium-pressure UV lamps.

The hydroxyl radicals formed by photolysis react rapidlywith organic compounds, with rate constants on the order of10r to 10” liters per mole-second; they also have a relativelylow selectivity in their reactions (Glaze and others, 1987).However, naturally occurring water components, such ascarbonate ion, bicarbonate ion, and some oxidizable species,act as free radical scavengers that consume hydroxyl radicals.Free radical scavengers are compounds that consume anyspecies possessing at least one unpaired electron. Inaddition to naturally occurring scavengers, excess hydrogenperoxide can itself act as a free radical scavenger, decreasingthe hydroxyl radical concentration. Reactions with hydroxylradicals are not the only removal pathway possible in theperox-pure system; direct photolysis by UV radiation oforganic compounds also provides a removal pathway forcontaminants. With these factors affecting the reaction, theproportion of oxidants required for optimum removal isdifficult to predetermine. Instead, the proportion foroptimum removal must be determined experimentally foreach waste.

The principal operating parameters for the perox-pure”technology are hydrogen peroxide dose, influent pH, andflow rate (which determines hydraulic retention time).Typically, during treatability studies, initial values of theseparameters are selected based on (1) the technologydeveloper’s experience and (2) the anticipated effects of theoperating parameters on the treatment system’sperformance. These operating parameters are discussedbriefly below. Their effects on the system’s performance arediscussed in detail in Section 3.

Hydrogen peroxide dose is selected based on treatmentunit configuration, contaminated water chemistry, andcontaminant oxidation rates. Because hydrogen peroxide isa hydroxyl radical scavenger, excess hydrogen peroxide canresult in a net decrease in treatment efficiency. However, ifthe hydrogen peroxide dose is low, hydroxyl radicalformation will also be low, decreasing treatment efficiency.Therefore, a balance must be maintained between excess andlow levels of hydrogen peroxide.

Influent pH level controls the carbonate chemistry, whichcan affect treatment efficiency. Because carbonate andbicarbonate ions will scavenge hydroxyl radicals, groundwaterpH may need to be adjusted before treatment to shift thecarbonate equilibrium to carbonic acid, which is not ascavenger.

Flow rate through the treatment system will determinehydraulic retention time. Increasing or decreasing the flowrate will affect treatment efficiency by changing the time

7

available for hydroxyl radical formation and contaminantdestruction.

2.4.2 System Components and Function

The perox-pure’” chemical oxidation systems are typicallyassembled from the following portable, skid-mountedcomponents: a chemical oxidation unit, a hydrogen peroxidefeed module, a UV lamp drive, and a control panel unit. Inaddition to these main system components, other equipmentis used to address site-specific conditions or requirements,including contaminated water characteristics and effluentdischarge limits. For example, Figure 2-l presents aschematic diagram of the main and ancillary components ofthe perox-pure chemical oxidation system used for theSITE demonstration (Model SSB-30).

For the SITE demonstration, a skid-mounted acid feedmodule and a base feed module were used to adjust pH ofwater before and after treatment, respectively. PSI providedthe acid (sulfuric acid) and base (sodium hydroxide)solutions in drums. Two cartridge filters arranged inparallel, capable of screening suspended silt larger than 3micrometers, were used to remove particles from thegroundwater, which was primarily contaminated with volatileorganic compounds (VOC) including trichloroethene (TCE)and tetrachloroethene (PCE). A spiking solution feedmodule was used to spike effluent from cartridge filters withchloroform; 1,1-dichloroethane (DCA); and l,l,l-trichloroethane (TCA) for certain demonstration runs. A7,500-gallon bladder tank was used (1) as an equalizationtank and (2) as a holding tank to perform a fewdemonstration runs at flow rates greater than thegroundwater well yield. The bladder tank was useful inminimizing the volatilization of contaminants. To ensure arelatively homogeneous process water, static mixers wereused after chemicals were added at upstream locations in thetreatment system.

The SSB-30 model has six reaction chambers, orreactors, with one UV lamp in each reactor. Each UV lamphas a power rating of 5 kilowatts (kw), for a total systemrating of 30 kW. The UV lamps are mounted inside UV-transmissive quartz tubes at the center of the reactors, sothat water flows through the space between the reactor walland the quartz tube. Circular wipers are mounted on thequartz tubes housing the UV lamps. The wipers areperiodically used to remove any suspended particles thathave coated the quartz tubes. In a coating environment, thiscoating diminishes the effectiveness of the system byblocking some of the UV radiation.

Contaminated water is pumped to the treatment systemand enters the oxidation unit through a section of pipecontaining a temperature gauge, a flow meter, an influentsampling port, and hydrogen peroxide and sulfuric acidaddition points. Hydrogen peroxide is added to the

contaminated water before it enters the oxidation unit;:however, a splitter can be used to add hydrogen peroxide atthe inlet of each lamp section to allow for different dosesinto each reactor. Inside the oxidation unit, thecontaminated water follows a serpentine path that parallelseach of the six UV lamps. The water passes each lampindividually, allowing lamps to be turned on or off as needed.Sample ports are located after each reactor. Inside theoxidation unit, photolysis of hydrogen peroxide by UVradiation results in the formation of hydroxyl radicals; thesefree radicals react rapidly with oxidizable compounds, suchas organic contaminants.

Treated water exits the oxidation unit through aneffluent pipe equipped with a temperature gauge and sampleport. The hydrogen peroxide dose is usually set so that theconcentration of the residual hydrogen peroxide in thetreated water is less than 5 milligrams per liter (mg/L).Sodium hydroxide is then added to readjust the pH to meetdischarge requirements.

The control panel on the perox-pure” system monitorswater flow rate, total flow through the system, UV lampcurrent in each reactor, and alarm conditions for the perox-pure unit. Hydrogen peroxide and acid injection areactivated by switches on the control panel and are monitoredwith flow meters.

2.4.3 Innovative Features of the Technology

Common methods for treating groundwatercontaminated with solvents and other organic compoundsinclude air stripping, steam stripping, carbon adsorption,chemical oxidation, and biological treatment. As regulatoryrequirements for secondary wastes and treatment by-products became more stringent and expensive to complywith, oxidation technologies have been known to offer amajor advantage over other treatment techniques: chemicaloxidation technologies destroy contaminants rather thantransferring them to another medium, such as activatedcarbon or the ambient air. Also, chemical oxidationtechnologies offer faster reaction rates than othertechnologies, such as some biological treatment processes.However, the oxidation of organics by ozone, hydrogenperoxide, or UV radiation alone has kinetic limitations,restricting its applicability to a narrow range ofcontaminants. As a result of these limitations, conventionalchemical oxidation technologies have been slow to becomecost-competitive treatment options.

The combined use of UV radiation and hydrogenperoxide in the perox-pure system increases the destructionefficiency of the treatment system and allows for thetreatment of a wider range of contaminants. Hydroxylradicals formed by UV photolysis of hydrogen peroxiderapidly oxidize the contaminants and exhibit littlecontaminant selectivity.

8

Figure 2- 1 perox-pure” Chemicel Odd&ion Treatment System

The perox-pureTM treatment system produces no airemissions and generates no sludge or spent media thatrequire further processing, handling, or disposal. Ideally,end products include water, carbon dioxide, halides, and insome cases, organic acids. However, other oxidizable speciespresent in the water (including metals in reduced form,cyanide, and nitrite) can also be oxidized in the process andcan exert an additional oxidant demand.

Hydrogen peroxide is inexpensive, easy to handle, andreadily available. As a result, its use with UV radiation inthe perox-pure system offers considerable advantages overexpensive and difficult to handle chemicals.

Table 2-l compares several treatment options for watercontaminated with VOCs. Similar comparisons can be madefor semivolatile organic compounds (SVOC), polychlorinatedbiphenyls (PCB), and pesticides, although air stripping is notgenerally applicable to these types of contaminants.

2.5 Key Contacts

Additional information on the perox-pureTM chemicaloxidation technology, the SITE program, and Site 300 atLLNL can be obtained from the following sources:

1. The perox-pureTM Technology

Chris GiggyProcess Engineering ManagerPeroxidation Systems, Inc.5151 East Broadway, Suite 600Tucson, Arizona 85711(602) 790-8383

2. The SITE Program

Norma LewisSection Chief, EPA Project ManagerEPA SITE Program26 West Martin Luther King DriveCincinnati, Ohio 45268(513) 569-7665

9

Table 2-1 Comparison of Technologies for Treating VOCs in Water

Technology Advantages Disadvantages

Air stripping

Steam stripping Effective at all concentrations

Air stripping with carbon adsorption of

Vapors

Effective at high concentrations

Air stripping with carbon adsorption ofvapors and spent carbon regeneration

Effective at high concentmtions; nocarbon disposal costs; can reclaim theproduct

Carbon adsorption Low air emissions; effective at highconcentrations

Biological treatrnent

perox-pure” technology

Effective at high concentrations;mechanically simple; relativelyinexpensive

Low air emissions; relativelyinexpensive

No air emissions; no secondary waste;VOCs destroyed

Inefficient at low concentmtions; VOCsdischarged to air

VOCs discharged to air; high energyconsumption

Inefficient at low concentrations;requires disposal or regeneration ofspent carbon

Inefficient at low concentrations; highenergy consumption

Inefficient at low concentrations;requires disposal or regeneration ofspent carbon; relatively expensive

inefficient at high concentrations; slowrates of removal; sludge treatment anddisposal required

High energy consumption; not cost-effective at high concentrations

3. The Lawrence Livermore National Laboratory, Site 300

Albert LamarreSite 300 Section LeaderLawrence Livermore National Laboratory7000 East AvenueP.O. Box 808, L-619Livermore, California 94550(510) 422-0757

Shyam ShuklaProject Manager, Site 300Lawrence Livermore National Laboratory7000 East AvenueP.O. Box 808, L-528Livermore, California 94550(510) 422-3475

10

Section 3Technology Applications Analysis

This section addresses the applicability of theperox-pureTM chemical oxidation technology to treat watercontaminated with organic compounds. The vendor claimsregarding the applicability and performance of theperox-pureTM technology are included in Appendix A.Because results from the SITE demonstration provided anextensive data base, evaluation of the technology’seffectiveness and its potential applicability to contaminatedsites is mainly based on these results, which are presented inAppendix B. The SITE demonstration results aresupplemented by results from other applications of theperox-pure technology, which are presented in Appendix C.

This section summarizes the effectiveness of theperox-pure chemical oxidation technology and discusses thefollowing topics in relation to the applicability of theperox-pure technology: factors influencing performance,site characteristics, material handling requirements,personnel requirements, potential community exposures, andpotential regulatory requirements.

3.1 Effectiveness of the perox-pureTM Technology

This section discusses the effectiveness of theperox-pure technology based on results from the SITEdemonstration and three other case studies.

3.1.1 SITE Demonstration Results

The SITE demonstration was conducted at LLNL Site300 in Tracy, California, over a 3-week period in September1992. During the demonstration, a perox-pure unit (ModelSSB30) treated about 40,000 gallons of groundwatercontaminated with VOCs. Principal groundwatercontaminants included TCE and PCE, which were present atconcentrations of about 1,000 and 100 micrograms per liter(pg/L), respectively. Other VOCs (such as chloroform;DCA; 1,1-dichloroethene; 1,2-dichloroethene; and TCA)were present at average concentrations below 15 rg/L.Groundwater was pumped from two wells into a 7,500-gallonbladder tank to minimize variability in influentcharacteristics. In addition, cartridge filters were used toremove suspended solids greater than 3 micrometers in size

from the groundwater before it entered the bladder tank.Treated groundwater was stored in two 20,000 gallon steeltanks before being discharged.

The perox-pureTM chemical oxidation technologydemonstration performed under the SITE program had thefollowing primary objectives:

l Assess the technology’s ability to destroy VOCsfrom groundwater at the LLNL site under differentoperating conditions

l Determine whether the treated water meetsapplicable disposal requirements at the 95 percentconfidence level

l Obtain information required to estimate theoperating costs for the treatment system, such aselectrical power consumption and chemical doses

The secondary objective for the technologydemonstration was to obtain preliminary information on thepresence and types of by-products formed during thetreatment.

The technology demonstration was conducted in threephases. Phase 1 consisted of eight runs using rawgroundwater, Phase 2 consisted of four runs using spikedgroundwater, and Phase 3 consisted of two runs using spikedgroundwater to evaluate the effectiveness of quartz tubecleaning. The three phases are described below.

The principal operating parameters for the perox-pure”system are hydrogen peroxide dose, influent pH, and flowrate. During Phase 1 of the demonstration, each of theseoperating parameters was varied to observe treatment systemperformance under different conditions. Preferred operatingconditions, those under which the concentrations of spikedgroundwater effluent VOCs would be reduced to belowtarget levels, were then determined for the system. Thetarget levels for the VOCs are given in Appendix B.

11

During Phase 2, groundwater was spiked with knownconcentrations of contaminants, and reproducibility testswere conducted. Spiked groundwater contained 200 to 300pg/L each of DCA, TCA, and chloroform. Thesecompounds were chosen because they are difficult to oxidizeand because they were not present in the groundwater athigh concentrations. This phase was also designed toevaluate the reproducibility of treatment system performanceat the preferred operating conditions determined in Phase 1.

Phase 3 evaluated the effectiveness of the quartz tubewipers by performing two runs using spiked groundwater andscaled and clean quartz tubes.

During the demonstration, samples were collected at thefollowing locations: treatment system influent, effluent fromReactor 1, effluent from Reactor 2, effluent from Reactor 3,and treatment system effluent. Samples were analyzed forVOCs, SVOCs, total organic carbon (TOC), total carbon,purgeable organic carbon (POC), total organic halides(TOX), adsorbable organic halides (AOX), metals, pH,alkalinity, turbidity, temperature, specific conductance,hydrogen peroxide residual, and hardness, as applicable. Inaddition, samples of influent to Reactor 1 and treatmentsystem effluent were collected and analyzed for acute toxicityto freshwater organisms. Hydrogen peroxide, acid, and basesolutions were also sampled and analyzed to verifyconcentrations.

Appendix B summarizes information from the SITEdemonstration, including (1) site characteristics,(2) contaminated groundwater characteristics,(3) perox-pure system performance, and (4) technologyevaluation results. Key findings of the demonstration are asfollows:

l For the spiked groundwater, PSI determined thefollowing preferred operating conditions: (1)influent hydrogen peroxide level of 40 mg/L; (2)hydrogen peroxide level of 25 mg/L in the influentto Reactors 2 through 6; (3) an influent pH of 5.0;and (4) a flow rate of 10 gallons per minute (gpm).At these conditions, the effluent TCE, PCE, andDCA levels were generally below the detection limit(5 pg/L) and effluent chloroform and TCA levelsranged from 15 to 30 pg/L. The average overallremoval efficiencies for TCE, PCE, chloroform,DCA, and TCA were about 99.7, 97.1, 93.1, 98.3,and 81.8 percent, respectively.

l For the unspiked groundwater, the effluent TCEand PCE levels were generally below the detectionlimit (1 pg/L) with corresponding removalefficiencies of about 99.9 and 99.7 percent. Theeffluent TCA levels ranged from 1.4 to 6.7 pg/Lwith removal efficiencies ranging from 35 to84 percent.

The perox-pure system effluent met Californiadrinking water action levels and federal drinkingwater maximum contaminant levels (MCL) forTCE, PCE, chloroform, DCA, and TCA at the 95percent confidence level.

The quartz tube wipers were effective in keepingthe tubes clean and appeared to reduce the effectscaling has on contaminant removal efficiencies.

TOX removal efficiencies ranged from 93 to 99percent. AOX removal efficiencies ranged from 95to 99 percent.

For spiked groundwater, during reproducibility runs,the system achieved average removal efficiencies of38 percent and greater than 93 percent for TOCand POC, respectively.

The temperature of groundwater increased at a rateof 12 “F per minute of UV radiation exposure inthe perox-pur system. Since the oxidation unit isexposed to the surrounding environment, thetemperature increase may vary depending upon theambient temperature or other atmosphericconditions.

3.1.2 Results of Other Case Studies

The perox-pur technology has been used to treatcontaminated water at approximately 80 sites. Results fromthree of these applications are discussed as case studies inAppendix C. A brief summary of the effectiveness of theperox-pure” technology at the three sites chosen as casestudies is presented below.

All three case studies represent full-scale, currentlyoperating commercial installations of perox-pre” chemicaloxidation systems. The contaminants of concern in thesecase studies include acetone, isopropyl alcohol (IPA), TCE,and pentachlorophenol (PCP). In the first case study, theperox-pure system treated industrial wastewater containing20 mg/L of acetone and IPA; the effluent met the dischargelimit of 0.5 mg/L. In the second case study, the perox-puresystem treated groundwater that was used as a municipaldrinking water source. The groundwater initially containedan average concentration of 150I.50 pg/L TCE. Aftertreatment, the effluent TCE level was 0.5 pg/L, well belowthe TCE drinking water standard of 5 pg/L. In the thirdcase study, the perox-pure system treated groundwater ata chemical manufacturing facility. The groundwatercontained 15 mg/L of PCP; treatment achieved the targeteffluent PCP level of 0.1 mg/L.

12

3.2 Factors Influencing Performance

Several factors influence the effectiveness of theperox-pure chemical oxidation technology. These factorscan be grouped into three categories: (1) influeatcharacteristics, (2) operating parameters, and (3)maintenance requirements. Each of these is discussedbelow.

3.2.1 Influent Characteristics

The perox-pu eTM chemical oxidation technology iscapable of treating water containing a variety of organiccontaminants, i n c l u d i n g VOCs, SVOCs, pes t i c ides ,polynuclear aromatic hydrocarbons (PAH), PCBs, andpetroleum hydrocarbons. Under a given set of operatingconditions, contaminant removal efficiencies depend on thechemical structure of the contaminants. Removal efficienciesare high for organic contaminants with double bonds (suchas TCE, PCE, and vinyl chloride) and aromatic compounds(such as phenol, toluene, benzene, and xyleae), becausethese compounds are easy to oxidize. Organic contaminantswithout double bonds (such as TCA and chloroform) are noteasily oxidized and are more difficult to remove.

Contaminant concentration also affects treatment systemeffectiveness. The perox-pure”’ system is most effective intreating water with contaminant concentrations less thanabout 500 mg/L. If contaminant concentrations are greaterthan 500 mg/L, the perox-pure system may be used incombination with other treatment technologies, such as airstripping. For highly contaminated water, the perox-puresystem can also be operated in a “flow-through with recyclemode, in which part of the effluent is recycled back throughthe oxidation unit to improve overall removal efficiency.

The perox-pur TM system uses a chemical oxidationprocess to destroy organic contaminants; therefore, otherspecies in the influent that consume oxidants are consideredan additional load for the system. These species are calledscavengers. A scavenger may be described as any species inwater other than the target contaminants that consumesoxidants. Common scavengers include anions such asbicarbonate, carbonate, sulfide, nitrite, bromide, and cyanide.Metals present in reduced states, such as trivalent chromium,ferrous iron, manganous ion, and several others, are alsolikely to be oxidized. In addition to acting as scavengers,these reduced metals can cause additional concerns underalkaline pH conditions. For example, trivalent chromiumcan be oxidized to hexavalent chromium, which is moretoxic. Ferrous iron and mangaaous ion are converted to lesssoluble forms, which precipitate in the reactor, creatingsuspended solids that can build up on the quartz tubeshousing the UV lamps. Natural organic compounds, such ashumic acid (often measured as TOC), are also potentialscavengers in this treatment technology.

Other influeat characteristics of concern includesuspended solids, oil, and grease. These constituents canbuild up on the quartz tubes housing the UV lamps,resulting in reduced UV transmission and decreasedtreatment efficiency.

3.2.2 Operating Pammeters

Operating parameters are those parameters that can bevaried during the treatment process to achieve desiredremoval efficiencies. The principal operating parameters forthe perox-puee system are hydrogen peroxide dose, influeatpH, and flow rate.

Hydrogen peroxide dose is selected based on treatmentunit configuration, contaminated water chemistry, andcontaminant oxidation rates. Under ideal conditions,hydrogen peroxide is photolyzed to hydroxyl radicals, whichare the principal oxidants in the system. Direct photolysis ofeach molecule of hydrogen peroxide results in a yield of twohydroxyl radicals. The molar extinction coefficient ofhydrogen peroxide at 253.7 nanometers, the dominantemission wavelength of low-pressure UV lamps, is only 19.6liters per mole-centimeter, which is low for a primaryabsorber in a photochemical process (Glaze and others,1987). Therefore, although the yield of hydroxyl radicalsfrom hydrogen peroxide photolysis is relatively high, the lowmolar extinction coefficient requires that a relatively highconcentration of hydrogen peroxide exist in the water.However, because excess hydrogen peroxide is also ahydroxyl radical scavenger, hydrogen peroxide levels that aretoo high could result in a net decrease in treatmentefficiency. According to PSI, the perox-pure systemovercomes these limitations by using medium-pressure UVlamps.

The perox-pure system is equipped with a hydrogenperoxide splitter that allows the operator to inject hydrogenperoxide to the oxidation unit influent and directly to any ofthe individual oxidation reactors. The distribution of thetotal hydrogen peroxide dose is an important operatingparameter, because the hydroxyl radical has a short lifetime.If the total hydrogen peroxide dose is delivered to theinfluent, depending on other operating conditions, theresulting hydroxyl radical concentration in the last reactormay be zero. Consequently, removal eflicieacy in the lastreactor would decrease significantly. Distributing part of thehydrogen peroxide dose directly to the reactors guaranteesthat some hydroxyl radicals will be present throughout theoxidation unit.

Influent pH controls the equilibrium among carbonate,bicarbonate, and carbonic acid. This equilibrium isimportant to treatment efficiency because carbonate andbicarbonate ions are hydroxyl radical scavengers. If theinflueat carbonate and bicarbonate concentration is greaterthan about 400 mg/L as calcium carbonate, the pH should

13

be lowered to between 4 and 6 to improve the treatmentefficiency. At low pH, the carbonate equilibrium is shiftedto carbonic acid, which is not a scavenger.

Flow rate through the treatment system determines thehydraulic retention time. In general, increasing the hydraulicretention time improves treatment efficiency by increasingthe time available for contaminant destruction.Theoretically, at a certain point, the reaction proceedstoward equilibrium, and increasing the hydraulic retentiontime no longer significantly increases removal efficiency. PSIdid not observe this phenomenon in the range of hydraulicretention times provided by the perox-pure system.

3.2.3 Maintenance Requirements

The maintenancersystem summarized below are based on discussions with PSIduring and after the SITE demonstration. Regularmaintenance by trained personnel is essential for thesuccessful operation of temajor system component that requires regular maintenanceis the UV lamp assembly. A brief summary of themaintenance requirements for the UV lamp assembly andother miscellaneous components is presented below.

Regular UV lamp assembly maintenance includesperiodically cleaning the quartz tubes housing the UV lamps.Eventually, the lamps may need to be replaced. Thefrequency at which the quartz tubes should be cleaneddepends on the type and concentration of suspended solidspresent in the influent or formed during treatment. Cleaningfrequency may range from once every month to once every3 months. UV lamp assemblies can be removed from theoxidation unit to provide access to the quartz tubes, whichcan then be cleaned manually. The quartz tubes can also becleaned automatically during operation with wipers.Automatic tube cleaning is a standard feature on most PSItreatment units. The quartz tube wipers require replacementonce every 3 to 6 months depending upon the cleaning cyclefrequency.

Maintenance requirements for the medium-pressure,mercury-vapor, broad-band UV lamps used in the perox-pure system are similar to those for conventional, low-pressure UV lamps. The life of low-pressure UV lampsnormally cited by most manufacturers is 7,500 hours, basedon a use cycle of 8 hours. The use cycle represents thelength of time the UV lamp is operated between shutdowns.Decreasing the use cycle or increasing the frequency atwhich a UV lamp is turned on and off can lead to earlylamp failure.

A number of factors contribute to UV lamp aging.These factors include plating of mercury to the interior lampwalls, a process called blackening, and solarization of thelamp enclosure material, which reduces its transmissibility.

These factors cause steady deterioration in lamp output atthe effective wavelength and may reduce output at the endof a lamp’s life by 40 to 60 percent. This reduction in lampoutput requires more frequent replacement of the UVlamps. According to PSI, no significant decline in UV lampoutput occurs until after about 3,000 hours of operation.Therefore, PSI recommends replacing the UV lamps after3,000 hours. PSI guarantees the UV lamps in the perox-pure unit for 3,000 hours when they are turned on and offno more than two or three times a day.

The only other part of the UV lamp assembly requiringperiodic maintenance is the gasket between the UV lampand the reactor. This gasket, which is used to maintain awater-tight seal on each reactor, is generally replaced oncea year.

Other components of the perox-pure system, such asvalves, flow meters, piping, hydrogen peroxide feed module,acid feed module, and base feed module, should be checkedfor leaks once a month. In addition, the influent, hydrogenperoxide, acid, and base feed pumps should be checked oncea month for proper operation and maintenance. Feed pumpheads are usually replaced annually. PSI offers a full-serviceprogram to its customers that covers all regular maintenanceand replacement parts for the system.

3.3 Site Characteristics

In addition to influent characteristics and effluentdischarge requirements, site characteristics are importantwhen considering the perox ’ technology. Site-specificfactors can impact the application of the perox-puretechnology, and these effects should be considered beforeselecting the technology for remediation of a specific site.Site-specific factors include support systems, site area andpreparation, site access, climate, utilities, and services andsupplies. Tables 4-l and 4-2 in Section 4 identify examplesof categories that are specific to the perox-pure” system andto a hazardous waste remediation site.

3.3.1 Support Systems

To clean up contaminated groundwater, extraction wellsand a groundwater collection and distribution system mustbe installed to pump groundwater to a central facility wherethe perox-pure system is located. Because the perox-puresystem is normally operated as a continuous flow-throughsystem during site remediation, installation of severalextraction wells may be required to provide a continuoussupply of groundwater. An equalization tank may berequired if flow rates from the groundwater wells fluctuateor if contaminant concentrations vary. When installing agroundwater collection and distribution system, preventivemeasures should be considered to reduce volatilecontaminant losses.

14

Before choosing the perox-pure technology, thelocation, design, and installation of tanks, piping, and otherequipment or chemicals associated with any pretreatmentsystems should be considered. Pretreatment is often desiredto remove oil and grease, suspended solids, or metals. Anytanks that are part of pretreatment or other support systemsshould be equipped with vapor control devices (for example,floating lids) to prevent VOC losses.

If on-site facilities are not available for office andlaboratory work, a small building or shed may be requirednear the treatment system. The on-site building should beequipped with electrical power to run laboratory equipmentand should be heated or air-conditioned, depending on theclimate. The on-site laboratory should contain equipmentneeded to perform simple analyses of the physical andchemical water characteristics required to monitor treatmentsystem performance. Such characteristics may include pH,hydrogen peroxide dose, and temperature.

3.3.2 Site Area and Preparation

The perox-pure”’ units are available in several sizes,ranging in combined UV lamp power from 10 to 720 kW.The perox-pure units have operated at flow rates between5 and several thousand gpm, depending on the requiredhydraulic retention time. During the SITE demonstration,a 30-kW perox-pure unit with a total volume of 15 gallonswas used. A 10- by 20-foot area was adequate for theperox-pure unit and associated chemical feed units. Largersystems would require slightly larger areas. Areas requiredfor influent and effluent storage tanks, if needed, will dependon the number and size of tanks. Also, a 2O- by 1%foot areamay be required for an office or laboratory building.

The area containing the perox-pure unit and tanksshould be relatively level and should be paved or coveredwith compacted soil or gravel.

3.3.3 Site Access

Site access requirements for treatment equipment areminimal. The site must be accessible to tractor-trailer trucksof standard size and weight. The roadbed must be able tosupport such a vehicle delivering the perox-pure system andtanks.

3.3.4 Climate

According to PSI, below-freezing temperatures andheavy precipitation do not affect the operation of theperox-pure system. The system is designed to withstandrain and snow and does not require heating or insulation,because the chemical oxidation process generates heat,increasing the water temperature about 12 “F per minute ofcontact time. However, if below-freezing temperatures areexpected for a long period of time, chemical and influent

storage tanks and associated plumbing should be insulatedor kept in a heated shelter, such as a building or shed.Housing the system also facilitates regular system checks andmaintenance. The perox-pure unit requires a high-voltagepower supply, which should also be protected from heavyprecipitation.

3.3.5 Utilities

The perox-pure system requires potable water andelectricity. Potable water is required for a ‘safety shower, aneye wash station, personnel decontamination, and cleaningfield sampling equipment. The perox-pure unit requires480-volt, 3-phase electrical service. Electrical power for thechemical feed modules can be supplied through theperox-pure”’ unit. Additional 110-volt ,lO-volt , single-phase electricalservice is needed to operate the groundwater extraction wellpumps, light the office and laboratory building, and operateon-site laboratory and office equipment.

A telephone connection or cellular phone is required toorder supplies, contact emergency services, and providenormal communications.

3.3.6 Services and Supplies

A number of services and supplies are required for theperox-pure technology. Most of these services and suppliescan be readily obtained.

If UV lamps, pumps, flow meters, or pipingmalfunctions, an adequate on-site supply of spare parts isneeded. If an on-site parts inventory is not an option, siteproximity to an industrial supply center is an importantconsideration. In addition, an adequate supply of chemicals,such as hydrogen peroxide, sulfuric acid, and sodiumhydroxide, or proximity to a supply center carrying thesechemicals is essential.

Complex laboratory services, such as VOC and SVOCanalyses, that cannot usually be performed in an on-site fieldlaboratory require contracting a local analytical laboratoryfor an ongoing monitoring program.

3.4 Material Handling Requirements

The perox-pure system does not generate treatmentresiduals, such as sludge or spent media, that require furtherprocessing, handling, or disposal. The chemical oxidationunit and other components of the system, such as thechemical feed units, are air-tight and produce no airemissions. Material handling requirements for theperox-pure technology include those for (1) pretreatmentmaterials and (2) treated water. These are described below.

15

3.4.1 Pretreatment Materials 3.4.2 Treated Water

In general, pretreatment requirements for contaminatedwater entering the perox-pure system are minimal.Depending on the influent characteristics, pretreatmentprocessing may involve one or more of the following: oiland grease removal, suspended solids removal, metalsremoval, or pH adjustment to reduce carbonate andbicarbonate levels. These pretreatment requirements arediscussed below.

Water containing visible, free, or emulsified oil andgrease requires pretreatment to separate and remove the oiland grease. If not treated, oil and grease will scale UVlamps and reduce UV transmission, which makes theoxidation process less effective. Separated oil and greaseshould be containerized and analyzed to determine disposalrequirements.

Because suspended solids can also reduce UVtransmission, water containing more than 30 mg/L ofsuspended solids should be pretreated. Depending on theconcentration, cartridge filters, sand filters, or settling tanksmay be used to remove suspended solids. Solids removedfrom the influent by pretreatment precipitation, filtration, orsettling should be dewatered, containerized, and analyzed todetermine whether they should be disposed of as hazardousor nonhazardous waste.

Pretreatment also may be necessary for water containingdissolved metals, such as iron and manganese. These metalsare likely to be oxidized in the perox-pure unit and are lesssoluble at higher oxidation states under alkaline pHconditions. After oxidation, the metals will tend toprecipitate as suspended solids in the perox-pure unit,resulting in UV lamp scaling. Removing these metals isoften advised; however, removal may depend on theconcentration of oxidizable metals in the contaminatedwater. The economics of metals removal should becompared to (1) predicted decreases in contaminant removalefficiency without metals removal and (2) the economics ofmore frequent UV lamp cleaning or replacement. Metalsremoved from the influent by precipitation should becontainerized and analyzed to determine whether the metalsshould be disposed of as hazardous or nonhazardous waste.

If the contaminated water contains bicarbonate andcarbonate ions at levels greater than 400 mg/L as calciumcarbonate, pH adjustments are typically performed in-line.Carbonate and bicarbonate ions act as oxidant scavengersand present an additional load to the treatment system. Theonly material handling associated with pH adjustmentinvolves handling chemicals such as acids and bases; pHadjustment should not create any waste streams requiringdisposal.

Treated water can be disposed of either on or off site.Examples of on-site disposal options for treated waterinclude groundwater recharge or temporary on-site storagefor sanitary usage. Examples of off-site disposal optionsinclude discharge into surface water bodies, storm sewers,and sanitary sewers. Bioassay tests may be required inaddition to routine chemical and physical analyses beforetreated water is disposed of. Depending on permitrequirements and treatment unit operating conditions,treated water may require pH adjustment before discharge.

3.5 Personnel Requirements

Personnel requirements for the perox-pure system areminimal. Generally, one operator, trained by PSI, conductsa daily 30-minute system check. The unit operator should becapable of performing the following: (1) filling chemicalfeed tanks and adjusting chemical flow rates to achievedesired doses; (2) operating the control panel on thechemical oxidation unit; (3) collecting liquid samples andperforming simple physical and chemical analyses andmeasurements (for example, pH, hydrogen peroxideconcentration, temperature, a n d f l o w r a t e ) ;(4) troubleshooting minor operational problems; and(5) collecting samples for off-site analyses. Analytical workrequiring more technical skills, such as VOC analyses, can beperformed by a local laboratory. The frequency of samplecollection and analysis will depend on site-specific permitrequirements.

The unit operator should also have completed anOccupational Safety and Health Act (OSHA) initial 40-hourhealth and safety training course and an annual g-hourrefresher course, if applicable, before operating theperox-pure” system at hazardous waste sites. The operatoralso should participate in a medical monitoring program asspecified under OSHA requirements.

For remote sites, where daily system checks by anoperator are not feasible, the perox-pure system can bemonitored remotely, or it can be remotely operated andmonitored. Remotely monitored systems can be connectedto devices that automatically dial a telephone to notifyresponsible parties at remote locations of alarm conditions.Remotely operated and monitored systems are hard-wiredinto centrally-located control panels or computers throughprogrammable logic controllers.

3.6 Potential Community Exposures

The perox-pure system generates no chemical orparticulate air emissions. Therefore, the potential for on-sitepersonnel or community exposure to airborne contaminantsis low. If a system malfunction occurs, alarms will sound,and all components of the system will shut off automatically.

16

Contaminated-water pumps can be hard-wired to the perox-pure TM unit control panel so that alarm conditionsautomatically stop flow through the unit, reducing thepotential for a contaminated water release.

Hydrogen peroxide solution, which is a reactivesubstance, presents the greatest chemical hazard associatedwith the system. However, when handled appropriately, thepotential for exposure to hydrogen peroxide by on-sitepersonnel is low. Acids, bases, and hydrogen peroxiderequired for the perox-pure system are typically stored inpolyethylene totes housed in metal frames or cages. Therelatively small volumes of hydrogen peroxide used by thesystem and secure chemical storage practices result in a lowpotential threat of community exposure to hydrogenperoxide.

3.7 Potential Regulatory Requirements

This subsection discusses regulatory requirementspertinent to site remediation using the perox-pure”technology. Regulations applicable to a particularapplication of this technology will depend on site-specificremediation logistics and the type of contaminated waterbeing treated. Table 3-l provides a summary of thepotentially applicable regulations discussed below.

Depending on the characteristics of the water to betreated, pretreatment or posttreatment may be required forthe successful operation of the perox-pure system. Forexample, solids may need to be prefiltered using cartridgefilters, sand filters, or settling tanks. Metals, such as ironand manganese, may need to be removed by precipitation.Each pretreatment or posttreatment process may haveadditional regulatory requirements that need to bedetermined prior to use. This subsection focuses onregulations for the perox-pure system only.

3.7.1 Comprehensive Environmental Response,Compensation, and Liability Act

The Comprehensive Environmental Response,Compensation, and Liability Act (CERCLA), as amended bySARA of 1986, authorizes the federal government to respondto releases into the environment of hazardous substances,pollutants, or contaminants that may present an imminentand substantial danger to public health or welfare (FederalRegister, 199Oa). Remedial alternatives that significantlyreduce the volume, toxicity, or mobility of hazardousmaterials and provide long-term protection are preferred.Selected remedies must also be cost effective and protectiveof human health and the environment.

Contaminated water treatment using the perox-puresystem will generally take place on site, while treated waterdischarge may take place either on site or off site. On-siteactions must meet all substantive state and federal applicable

or relevant and appropriate requirements (ARAR).Substantive requirements pertain directly to actions orconditions in the environment (for example, effluentstandards). Off-site actions must comply with legallyapplicable substantive and administrative requirements.Administrative requirements, such as permitting, facilitatethe implementation of substantive requirements.

EPA allows an ARAR to be waived for on-site actions.Six ARAR waivers are provided by CERCLA: (1) interimmeasures waiver, (2) equivalent standard of performancewaiver, (3) greater risk to health and the environmentwaiver, (4) technical impracticability waiver, (5) inconsistentapplication of state standard waiver, and (6) fund-balancingwaiver. Justification for a waiver must be clearlydemonstrated (EPA, 1988). Off-site remediations are noteligible for ARAR waivers, and all substantive andadministrative applicable requirements must be met.

Additional regulations pertinent to use of the perox-pure TM system are discussed below. No air emissions orresiduals (such as sludge or spent filter media) are generatedby the perox-pure system. Therefore, only regulationsaddressing contaminated water treatment and discharge arepresented.

3.7.2 Resource Conservation and Recovery Act

The Resource Conservation and Recovery Act (RCRA),,as amended by the Hazardous and Solid Waste Amendmentsof 1984, regulates management and disposal of municipaland industrial solid wastes. The EPA and RCRA-authorizedstates [listed in 40 Code of Federal Regulations (CFR) Part272] implement and enforce RCRA and state regulations.

The perox-pure system has been used to treat watercontaminated with a variety of organic materials, includingsolvents, pesticides, PAHs, and petroleum hydrocarbons.Contaminated water treated by the perox-p re” technologywill most likely be hazardous or sufficiently similar tohazardous waste so that RCRA standards will berequirements. Criteria for identifying hazardous wastes areincluded in 40 CFR Part 261. Pertinent RCRArequirements are discussed below.

If the contaminated water to be treated is determined tobe a hazardous waste, RCRA requirements for storage andtreatment must be met. The perox-pure system mayinclude tank storage. Tank storage of contaminated andtreated water (if the waters are RCRA hazardous) mustmeet tank storage requirements of 40 CFR Parts 264 or 265,Subpart J. Although air emissions are not associated withthe perox-pure system, any fugitive emissions from storagetank vents would be subject to forthcoming RCRAregulations (see 40 CFR Part 269) on air emissions fromhazardous waste treatment, storage, and disposal facilities.When promulgated, these requirements will include

17

Table 3- 1 Reg&tionr Summary

Act Applicability Application to perox-pure” Chemical OxidationSystem

Citation

CERCLA Superfund sites

Superfund and RCRAsites

CWA Discharges to surfacewater bodies

SDWA Water discharges,water reinjection, andsole-source aqu i fe rand wellheadprotection

TSCA PCE contamination

AEA and RCRA Mixed wastes

FIFRA Pesticides

OSHA All remedial actions

The Superfund program authorizes and regulatesthe cleanup of environmental contamination. Itapplies to all CERCLA site cleanups.

RCRA defines and regulates the treatment, storage,and disposal of hazardous wastes. RCRA alsoregulates corrective action at generator, treatment,storage, and disposal facilities.

NPOES requirements of CWA apply to bothSuperfund and RCRA sites where treated water isdischarged to surface water bodies. Pretreatmentstandards apply to discharges to POTWs.

Maximum contaminant /eveIs and contaminant levelgoals should be considered when setting watercleanup levels at RCRA corrective action andSuperfund sites. (Water cleanup levels are alsodiscussed under RCRA and CERCLA.) Reinjectionof treated water would be subject to undergroundinjection control program requirements, and so/esource and protected wellhead water sourceswould be subject to their respective controlprograms.

I F PCB-contaminated wastes are treated, TSCArequirements should be considered whendetermining cleanup standards and disposalrequirements. RCRA also regulates solid wastecontaining PCBs.

AEA and RCRA requirements apply to the treatment,storage, and disposal of mixed waste containingboth hazardous and radioactive components.OSWER and DOE directives provide guidance foraddressing mixed waste.

FIFRA regulates pesticide manufacturing andlabeling. However, ifpesticidscontaminated wateris treated, RCRA regulations will generally apply.

OSHA regulates on-site construction activities andthe health and safety of workers at hazardous wastesites. Jnstallation and operation of the system atSuperfund or RCRA sites must meet OSHArequirements.

40 CFR Part 300

40 CFR Parts 260 to 270,Part 280

40 CFR Parts 122 to 125,Part 403

40 CFR Part 141

40 CFR Part 767

AEA and RCRA

40 CFR Part 165

29 CFR Parts 1900 to192629 CFR Part 1910.120(hazardous waste andemergency responseoperations)

Note: Acronyms used in this table are defined in text.

standards for emissions from equipment leaks and processvents. Treatment requirementSubpar t Q (Chemical, Physical, and Biological Treatment)would also apply. This subpart includes requirements forautomati c influe nt shut-off, waste analysis, and trial tests.

The perox-pureT system could also be used to treatcontaminate d water a t RCRA-regulate d facilities.

Requirements for corrective action at RCRA-regulatedfacilities will be included in 40 CFR Part, 264 Subpart F(Regulated Units) and Subpart S (Solid Waste ManagementUnits), as well a s 40 CFR Part 280 (Underground StorageTanks); these subparts generall y will apply to remediation atSuperfund sites. The regulations include requirements forinitiating and conducting RCRA corrective actions,remediating groundwater, and ensuring that corrective

18

actions comply with other environmental regulations(Federal Register, 199Ob).

Water quality standards included in RCRA (such asgroundwater monitoring and protection standards), the CleanWater Act (CWA), and the Safe Drinking Water Act(SDWA) would be appropriate cleanup standards and wouldapply to discharges of treated water or reinjection of treatedgtottndwater (EPA, 1989). The CWA and SDWA arediscussed below.

3.7.3 Clean Water Act

The CWA is designed to restore and maintain thechemical, physical, and biological quality of navigable surfacewaters by establishing federal, state, and local dischargestandards. If treated water is discharged to surface waterbodies or publicly-owned treatment works (POTW), CWAregulations will apply. On-site discharges to surface waterbodies must meet substantive National Pollutant DischargeElimination System (NPDES) requirements, but do notrequire a NPDES permit. Off-site discharges to a surfacewater body require a NPDES permit and must meet NPDESpermit limits. Discharge to a POTW is considered an off-site activity, even if an on-site sewer is used. Therefore,compliance with substantive and administrative requirementsof the national pretreatment program is required. Generalpretreatment regulations are included in 40 CFR Part 403.Any local or state requirements, such as stateantidegradation requirements, must also be identified andsatisfied.

3.7.4 Safe Drinking Water Act

The SDWA, as amended in 1986, required EPA toestablish regulations to protect human health fromcontaminants in drinking water. EPA has developed thefollowing programs to achieve this objective: (1) drinkingwater standards program, (2) underground injection controlprogram and (3) sole-source aquifer and wellheadprotection programs.

SDWA primary, or health-based, and secondary, oraesthetic, MCLs will generally apply as cleanup standards forwater that is, or may be, used for drinking water supply. Insome cases, such as when multiple contaminants are present,alternate concentration limits (ACL) may be used.CERCLA and RCRA standards and guidance should beused in establishing ACLs (EPA, 1987a).

Water discharge through injection wells is regulated bythe underground injection control program. Injection wellsare categorized in Class I through V, depending on theirconstruction and use. Reinjection of treated water involvesClass IV (reinjection) or Class V (recharge) wells andshould meet requirements for well construction, operation,and closure.

The sole-source aquifer protection and wellheadprotection programs are designed to protect specific drinkingwater supply sources. If such a source is to be remediatedusing the perox-pure system, appropriate program officialsshould be notified, and any potential regulatory requirementsshould be identified. State groundwater antidegradationrequirements and water quality standards may also apply.

3.7.5 Toxic Substances Control Act

Testing, premanufacture notification, and record-keepingrequirements for toxic substances are .regulated under theToxic Substances Control Act (TSCA). TSCA also includesstorage requirements for PCBs (see 40 CFR Part 761.65).The perox-pure system may be used to treat watercontaminated with PCBs, and PCB storage requirementswould apply to pretreatment storage of PCB-contaminatedwater. The SDWA MCL for PCBs is 0.05 pg/L; this MCLwould generally be the treatment standard for groundwaterremediation at Superfund sites and RCRA corrective actionsites. RCRA land disposal requirements for PCBs (see 40CFR Part 268) may also apply, depending on PCBconcentrations. For example, liquid hazardous wastecontaining PCB concentrations between 50 and 499 ppm thatwill be treated by incineration or an equivalent method willmeet the requirements of 40 CFR part 761.70.

3.7.6 Mixed Waste Regulations

Mixed waste contains both radioactive and hazardouscomponents, as defined by the Atomic Energy Act (AEA)and RCRA, and is subject to the requirements of both acts.When the application of both regulations results in asituation that is inconsistent with the AEA (for example, anincreased likelihood of radioactive exposure), AEArequirements supersede RCRA requirements. Use of theperox-pure” system at sites with radioactive contaminationmight involve the treatment or generation of mixed waste.

The EPA OSWER, in conjunction with the NuclearRegulatory Commission, has issued several directives toassist in the identification, treatment, and disposal of low-level radioactive, mixed waste. Various OSWER directivesinclude guidance on defining, identifying, and disposing ofcommercial, mixed, low-level radioactive and hazardouswaste (EPA, 1987b). If the perox-pure system is used totreat low-level mixed wastes, these directives should beconsidered. If high-level mixed waste or transuranic mixedwaste is treated, internal orders from the U.S. Departmentof Energy (DOE) should be considered when developing aprotective remedy (DOE, 1988).

3.7.7 Fedeml Insecticide, Fungicide, andRodenticide Act

The perox-pure technology can treat watercontaminated with pesticides. EPA regulates pesticide

19

product sale, distribution, and use through product licensingor registration under the authority of the Federal Insecticide,Fungicide, and Rodenticide Act (FIFRA) (see 40 CFR Part165). Use of a pesticide product in a manner inconsistentwith its labeling violates FIFRA Compliance with FIFRAby following labeling directions may not be required atSuperfund or RCRA corrective action sites, because thepesticide may be a RCRA hazardous waste at that point. Insuch cases, requirements for hazardous wastes containingpesticide constituents must be met.

3.7.8 Occupational Safety and Health Act

OSHA regulations, contained in 29 CFR Parts 1900through 1926, are designed to protect worker health and

safety. Both Superfund and RCRA corrective actions must’meet OSHA requirements, particularly Part 1910.120,Hazardous Waste Operations and Emergency Response.Part 1926, Safety and Health Regulations for Construction,applies to any on-site construction activities. For example,electric utility hookups for the perox-pure” system mustcomply with Part 1926, Subpart K, Electrical. Productchemicals, such as hydrogen peroxide, sulfuric acid, andsodium hydroxide, used with the perox-pure system must bemanaged in accordance with OSHA requirements (forexample, Part 1926, Subpart D, Occupational Health andEnvironmental Controls, and Subpart H, Materials Handling,Storage; and Disposal). Any more stringent state or localrequirements must also be met.

20

Section 4Economic Analysis

This section presents cost estimates for using the perox-pure system to treat groundwater. Two cases, based ongroundwater characteristics, are presented. In Case 1, thegroundwater contains both contaminants that are difficult tooxidize and contaminants that are easy to oxidize. In Case 2,the groundwater contains only contaminants that are easy tooxidize. In each case, treatment costs are compared at threedifferent flow rates: 10 gpm, 50 gpm, and 100 gpm.

Cost estimates presented in this section are basedprimarily on data compiled during the SITE demonstration.Costs have been placed in 12 categories applicable to typicalcleanup activities at Superfund and RCRA sites (Evans,1990). Costs are presented in April 1993 dollars and areconsidered to be order-of-magnitude estimates with anaccuracy of plus 50 percent and minus 30 percent.

Table 4-l presents a breakdown of costs for the 12categories for Case 1, and Table 4-2 presents those costs forCase 2. The tables also present total one-time costs andtotal annual O&M costs; the total costs for a hypothetical,long-term, groundwater remediation project; the net presentvalues of the long-term project; and the costs per 1,000gallons of water treated. Both tables highlight in boldfacetype the direct cost associated with using the perox-puresystem. Each table concludes with a presentation of totaldirect one-time costs, total direct annual O&M costs, andthe direct costs per 1,000 gallons of water treated.

4.1 Basis of Economic Analysis

A number of factors affect the estimated costs oftreating groundwater with the perox-pure chemicaloxidation system. Factors affecting costs generally includeflow rate, type and concentration of contaminants,groundwater chemistry, physical site conditions, geographicalsite location, contaminated groundwater plume size, andtreatment goals.

The perox-pure= technology can treat several types ofliquid wastes, including contaminated groundwater, landfillleachate, and industrial wastewater. Contaminatedgroundwater was selected for this economic analysis because

it is commonly found at Superfund and RCRA correctiveaction sites and because this waste treatment scenarioinvolves most of the cost categories. In addition, twogroundwater remediation cases based on groundwatercharacteristics are analyzed. In Case 1, the groundwatercontains five contaminants, three of which are difficult tooxidize and two of which are easy to oxidize. In Case 2, thegroundwater contains two contaminants that are easy tooxidize. The following presents the assumptions andconditions as they apply to each case. Next, the assumptionsand conditions used only for Case 1 are presented, and thenthose used only for Case 2.

For each case, this analysis assumes that theperox-pure- system will treat contaminated groundwater ona continuous flow cycle, 24 hours per day, 7 days per week.Based on this assumption, during a l-year period, the 10-gpm unit will treat about 5.2 million gallons, the 50-gpm unitwill treat about 26 million gallons, and the 100~gpm unit willtreat about 52 million gallons. Because most groundwaterremediation projects are long-term, this analysis assumesthat a treatment project will last 50 years using the 10-gpmflow rate, 10 years using the 50-gpm flow rate, and 5 yearsat the 100~gpm flow rate. Treating groundwater at theserates involves remediation of a contaminated groundwaterplume that contains approximately 260 million gallons ofwater. While it is difficult in practice to determine both thevolume of groundwater to treat and the actual duration of aproject, these figures are used in order to perform thiseconomic analysis.

The total costs for a groundwater remediation projectfor each case and each flow rate scenario are presented asfuture values. Using the time periods described above, thisanalysis assumes a 5 percent inflation rate to estimate thetotal costs. The future values are then presented as a netpresent value using a discount rate of 5 percent. A higherdiscount rate will make the initial costs weigh more heavilyin the calculation, and a lower discount rate will make thefuture operating costs weigh more heavily. Because thk costsof demobilization will occur at the end of the project, theappropriate future values of those costs presented in thetables were used in calculating the final values. The costs

21

Table 4-1 Costs Associated with the perox-pure” Technology - Case 1

Cost Categories10 gpm

She Preparation (a)AdministrativeTreatability StudySystem DesignMobilization

Permitting and Regulatory Requirements (a)

Capital Equipment (a)Extraction Wells, Pumps, and PipingShelter BuildingTreatment EquipmentAuxiliary Equipment

Startup (a)

Labor (b)Operations Staff (c)Health and Sakty Refresher Course

Consumables and Supplies (b)Hydrogen PeroxideSutfuric AddSodium HydroxideCartridge FittersCarbon ColumnsPPEDisposal Drums for PPEUV LampsSampling SuppliesPropane

Utilities (b)Treatment SystemAuxiliary Equipment

Effluent Treatment and Disposal (c)

Residuals and Waste Shipping andHandling (b)

Analytical Services (b)

Maintenance and Modifications (b)Treatment SystemAuxiliary Equipment

Demobilization (a)Treatment SystemAil Other

Total One-Time Costs

Total Annual O&M Costs

Groundwater Remediation:Total Costs (d,e)Net Present Value (f)Costs per 1,000 Gallons (g)

Total perox-pure” direct one-time costs

Total perox-pure” direct O&M costsCosts per 1,000 Gallons-Direct Costs (g)

S168,OOO $171,000

100 g p mS171.000

35,000 35,000 35,0005,000 5,000 5,000

120,000 123,000 123,0008,000 8,000 8,000

41,000 45,000 51,000

819,000 906,000 1,015,000746,000 158,000 158,000455,000 455,000 455,000

110,000 185.000 290,000708,ooo 108,000 112,000

5.00039,000

5,000

39,000

5,000

39,00036,000 36,000 36,000

3,000 3,000 3,000

13,000 30,000 54,000450 2,200 4,460600 3,000 6.890

2,000 10,000 20.000200 200 200

7,000 1,000 1,000600 600 600

50 50 504,500 9,000 18,000

1,000 1,000 1,0003,000 3,000 3,000

72,000 60,000 119,0009,200 45.900 91,7002,800 13,800 27,500

0 0 0

6,000 6,000 8,000

24,000 24,000 24,000

22,000 29,000 40,000lY.000 78,50070,800 10,800

40,000 40,000 40,00010.00030,000

10,00030,000

$1,073,000 $1,167,OOO $1,283,000

$176,000 $188,000 $285,000

$25,776,000 $3,558,000 $2,863,OOO$9.411.000 $2,747,000 $2,479,000

$36 $11 $10$744.000 $227,000 $342,000

564,000 $725,000 $205,000

$19 $5 5 5

Estimated Costs (1993 $)

50 GPM

29,00017,200

10,00030,000

Notes:hems in bold denote perox-pure” system direct costsa One-time costsb Annual O&M costsC See text for explanation of these costsd Future value using annual inflation rate of 5 percente To complete this project, it is assumed that the IO-gpm flow rate will take 50 years, the 5O-gpm flow rate will take 70 years, and the

lOO-gpm flow rate will take 5 years to treat 260 million gallons total.f Annual discount rate of 5 percentg Presented as a net present value using the same assumptions used elsewhere in this table.

22

Table 4-2 Costs Associated with the pemx-pure” Techology - Case 2

Cost Categories10 gpm

site Preparation (a) 5168.000AdministrativeTrea t&ii@ StudySystem DesignMobtYizhon

Permitting and Regulatory Requirements (a)

Capital Equipment (a)Extraction Wells, Pumps, and PipingShelter BuildingTreatment EquipmentAuxiliary Equipment

Startup (a)

Labor (b)Operations staff (clHealth and Safety Refresher Course

Consumables and Supplies (b)Hydrogen PeroxideSulfuric AcidSodium HydroxideCartridge FittersCarbon ColumnsPPEDisposal Drums for PPEUV LampsSampling SuppliesPropane

utilities (b)Treatment SystemAuxiliary Equipment

Effluent Treatment and Disposal (c)

Residuals and Waste Shipping andHandling (b)

Analytical Services (b)

Maintenance and Modifications (b)Treatment SystemAuxiliary Equipment

Demobilization (a)Treatment SystemAll Other

Total One-Time Costs

Total Annual O&M Costs

Groundwater Remediation:Total Costs (d,e)Net Present Value (flCosts per 1,000 Gallons (g)

Total parox-pure” direct one-time costs

Total perox-pure” direct O&M costs

Costs per 1,000 Gallons--Direct Costs (g)

38,000

35,0005,000

120,0008,000

764,000

39,000 39,000

776,000 780,000146,000 158,000 158,000455,000 455,000 455,00055,000 55,000 55,000108,000 108,000 112,000

5,000

39,0003 6 , 0 0 0 36,000 36,000

3,000 3,000 3,000

10,000 22,000 38,0003 0 0 1,600 3,1006 0 0 3,000 6,000

2 , 0 0 0 10,000 20,000200 200 200

1.000 1,000 1,000600 600 600

50 50 501 ,500 1,500 3,0001,000 1,000 1,0003,000 3,000 3,000

4,000 4,000 8,0003,100 3,700 6.700

900 900 1,800

0 0 0

6,000 6,000 8,000

24,000 24,000 24,000

16,000 16,000 17,0005,500 5,50010,800 10,800

40,00070,00030,000

$1,015,000 $1,032,000 $1,036,000

$99,000 $111,000 $134,000

$25159,000$8,091,000

$31

$84,000

$49,000

$15

Estimated Costs (1993 $)

50 gpm

$171,000

5,000

39,000

40,000

$2,453,000$1,894,000

$7

$84,000

$67,000

100 gpm$171,000

35,000 35,0005,000 5,000

123,000 123,0008,000 8,000

5,000

39,000

40,00010,00030,000

5,50011,200

10,00030,000

$3

$1,787,000$1547,000

$6$84,000

$80,000

$2Notes:Items in bold denote perox-pure,” system direct costsa Onetime costsb Annual O&M costsc See text for explanation of these costsd Future value using annual inflation rate of 5 percente To complete this project, it is assumed that the 1Ogpm flow rate will take 50 years, the 5Ogpm flow rate will take 10 years, and the

10Ogpm flow rate will take 5 years to treat 260 million gallons total.f Annual discount rate of 5 percent

g Presented as a net present value using the same assumptions used elsewhere in this table.

23

per 1,000 gallons treated are derived from the net presentvalues. Capital equipment is not depreciated in thiseconomic analysis.

Further assumptions about groundwater conditions andtreatment for each case include the following:

l Any suspended solids present in groundwater areremoved before entering the perox-pure system.

l Alkalinity is about 2.50 mg/L as calcium carbonate.

l The influent has a pH of 8 that is lowered to 5.5 viapretreatment.

l The treated effluent has a pH between 4 and 5 thatis adjusted to between 6.5 and 8.5 via posttreatmentto meet discharge standards.

This analysis assumes that treated water for each casewill be discharged to surface water, and MCLs specified inthe SDWA are the treatment target levels. Based on resultsof the SITE demonstration, the perox-pure system shouldachieve these levels.

The following assumptions were also made for each casein this analysis:

The site is a Superfund site located in a rural areaof the Midwest.

Contaminated water is located in a shallow aquifer,

The groundwater contains negligible amounts ofiron and manganese and will not requirepretreatment for metals.

Suitable site access roads exist.

Utility lines, such as electricity and telephone lines,exist overhead on site.

A 4,000-square-foot building will be needed tohouse the treatment system and auxiliary equipmentfor all three flow rate scenarios.

The treatment system operates automatically.

One technician will be required to operate theequipment, collect all required samples, andperform equipment maintenance and minor repairs.

One treated water sample and one untreated watersample will be collected daily to monitor systemperformance.

l Treated and untreated water samples will becollected monthly and analyzed off site for VOCs.

l Labor costs associated with major repairs are notincluded.

l PSI will dispose of spent UV lamps from the perox-pure system at no cost.

For Case 1, the chemical feed rates and the hydraulicretention time required to meet treatment goals listed inAppendix B were estimated based on the perox-pureTM

system performance during Runs 10, 11, and 12 of the SITEdemonstration. Based on the demonstration, VOC levels inthe groundwater, and perox-pureTM system operatingconditions assumed for Case 1 include the following:

0 Chloroform at 200 pg/L

0 DCA at 150 pg/L

l TCA at 130 pg/L

l TCE at 700 pg/L

l PCE at 100 pg/L

l Hydrogen peroxide dose of 85 mg/L

0 Hydraulic retention time of 0.75 minute

For Case 2, the chemical feed rates and hydraulicretention time required to meet treatment goals listed inAppendix B were estimated based on the perox-puresystem performance during Run 8. Based on thedemonstration, the VOC levels in the groundwater, andperox-pureTM system operations include the following:

l TCE at 1,070 pg/L

l PCE at 108 pg/L

0 Hydrogen peroxide dose of 60 mg/L

l Hydraulic retention time of 0.05 minute

4.2 Cost Categories

Cost data associated with the perox-pureTM technologyhave been assigned to the following 12 categories: (1) sitepreparation; (2) permitting and regulatory requirements; (3)capital equipment; (4) startup; (5) labor; (6) consumablesand supplies; (7) utilities; (8) effluent treatment and disposal;(9) residuals and waste shipping and handling; (10) analyticalservices; (11) maintenance and modifications; and (12)demobilization. Costs associated with each category arepresented in the sections that follow. Each section presents

24

the costs that are identical for each case. If applicable,differences between the costs of Case 1 and Case 2 are thendiscussed. Some categories end with a summary of thesignificant costs within the category. All direct costsassociated with operating the perox-pureTM system areidentified as perox-pureTM direct costs; all costs associatedwith the hypothetical remediation and auxiliary equipmentare identified as groundwater remediation costs.

4.2.1 Site Preparation Costs

Site preparation costs include administrative, treatabilitystudy, system design, and mobilization costs. For thisanalysis, administrative costs, such as legal searches, accessrights, and other site planning activities, are associated witha groundwater remediation project and are estimated to be$35,000.

A treatability study will need to be performed todetermine the appropriate specifications of the perox-pure”’system for the site as well as the amounts of chemicals andreagents needed for optimal performance. PSI estimates thecost for this study, which is directly associated with operatingthe perox-pureTM treatment system, to be about $5,000.

System design costs include designing the site layout andthe treatment system operations and are associated with agroundwater remediation project. Design costs are typically20 percent of the total construction costs. Construction costsinclude constructing a shelter building for the treatment andauxiliary equipment and installing extraction wells and piping(see Capital Equipment Costs). Construction costs areabout $600,000 for the l0-gpm flow rate scenario and about$613,000 for the 50- and lOO-gpm flow rate scenarios.Therefore, design costs would be about $120,000 for the 10-gpm flow rate scenario and about $123,000 for the 50- and100-gpm flow rate scenarios.

Mobilization involves transporting all equipment to thesite, assembling it, performing optimization and shakedownactivities, and operator training. Transportation costs aresite-specific and will depend on the location of the site inrelation to all equipment vendors. The skid-mountedperox-pureTM system is delivered to each site in onesemitrailer from Tucson, Arizona. PSI will position thesystem and perform optimization and shakedown activities aspart of the mobilization. Initial operator training is neededto ensure safe, economical, and efficient operation of thesystem. PSI estimates mobilization costs to be about $8,OOOand to take about 1 week to complete.

For each case, total site preparation costs are estimatedto be $168,000 for the l0-gpm flow rate scenario and$171,000 for the 50- and 100~gpm flow rate scenarios.System design constitutes about 71 percent of the total sitepreparation costs, and administrative costs make up about 20

percent. As such, about 7 percent of the site preparationcosts are associated with the perox-pureTM treatment system.

4.2.2 Permitting and Regulatory RequirementsCOStS

Permitting and regulatory costs depend on whethertreatment is performed at a Superfund or a RCRAcorrective action site and on how treated effluent and anysolid wastes generated are disposed of. Superfund sitesrequire remedial actions to be consistent with ARARs ofenvironmental laws, ordinances, regulations, and statutes,including federal, state, and local standards and criteria. Ingeneral, ARARs must be determined on a site-specific basis.RCRA corrective action sites require additional monitoringrecords and sampling protocols, which can increase thepermitting and regulatory costs by an additional 5 percent.

Permitting and regulatory costs are associated with agroundwater remediation project and are assumed to beabout 5 percent of the total capital equipment costs for atreatment operation that is part of a Superfund siteremediation project. This estimate does not include annualdischarge permit costs, which may vary significantlydepending on state and local requirements.

For Case 1, permitting and regulatory costs areestimated to be S41,OOO for the l0-gpm flow rate scenario;$45,000 for the 50-gpm flow rate scenario; and $51,000 forthe 100-gpm flow rate scenario. For Case 2, these costs areestimated to be $38,0OO for the lo-gpm, and $39,000 for the50- and 100-gpm flow rate scenarios. The costs in Case 2are similar because total capital equipment costs are nearlythe same for each flow rate. The difference in costs betweenthe two cases is due to the lower perox-pureTM system capitalequipment costs associated with Case 2.

4.2.3 Capital Equipment Costs

Capital equipment costs include installing extractionwells; constructing a building to shelter the treatment andauxiliary equipment; and purchasing and installing alltreatment equipment, including auxiliary equipment andmonitoring equipment.

Extraction well installation costs are associated with agroundwater remediation project and include installing thewell and pump and connecting the pumps, piping, and valvesfrom the wells to the perox-pure system. This analysisassumes that four, 150-foot extraction wells will be requiredto maintain the flow rate in each scenario. Extraction wellscan be installed at about $l50 per foot per well. Total wellconstruction costs for each case will be about $9O,OOO.

Pumps, piping, and valve connection costs are associatedwith a groundwater remediation project and will depend onthe following factors: the number of extraction wells needed,

the flow rate, the distance of the extraction wells from thetreatment system, and the climate of the area. This analysisassumes that four extraction wells are located about 200 feetfrom the perox-pure system. Four 2.5gpm pumps will berequired to maintain a l0-gpm flow rate. The total cost forthese four pumps is about $8,000. Four 12.5gpm pumps willbe required to maintain a 50-gpm flow rate, at a total costof about $20,000. Four 25-gpm pumps will be required tomaintain a 100-gpm flow rate, at a total cost of about$20,000. Piping and valve connection costs are about $60 perfoot, including underground installation. Therefore, totalpiping costs will be an additional $48,000.

A building will need to be constructed to shelter theperox-pure system and all auxiliary equipment, because thesite is assumed to be located in the Midwest, andremediation will likely be conducted in adverse weatherconditions. The building will require about 4,000 square feetand will cost about $100 per square foot, includingconstruction and materials, for a total cost of $400,000.Costs associated with designing the building are included inthe system design costs (see Site Preparation Costs).Heating costs included in this analysis assume that thebuilding site is remote and will require a propane or liquidpetroleum gas heating system with a 3,000-gallon tank andfuel delivery service. The propane tank, heating unit, andduct work will cost about $25,000 to install. Finally, utilityconnections to the building are estimated to cost $30,000.Total building costs for each case will be about $455,000 foreach flow rate scenario and are associated with agroundwater remediation project.

Treatment equipment typically consists of chemical feedmodules and the skid-mounted components of theperox-pureTM system. The cost of the perox-pure unit willvary depending on the size of the unit needed. PSI identitiesunit sizes by their kW usage. To meet the treatment goalsfor Case 1, PSI estimates that the lo-gpm unit will require15 kW, the 50-gpm unit will require 75 kW, and the 100~gpmunit will require 150 kW. However, the nearestcommercially available perox-pureTM units PSI offers are 30kW, 90 kW, and 180 kW respectively. These will cost$100,000, $175,000, and $280,000 respectively. PSIrecommends turning off any lamps in excess of the necessaryelectrical power.

To meet the treatment goals of Case 2, PSI estimatesthat the l0-gpm unit will require 1 kW, the 50-gpm unit willrequire 5 kW, and the 100-gpm unit w-ill require 10 kW.However, the nearest commercially available unit PSI offersis 10 kW. Therefore, perox-pureTM unit cost will be $45,000for each flow rate scenario of Case 2.

This analysis assumes that the site treatment scenariosfor each case will include acid feed and base feed modules.The chemical feed modules consist of one 2,000-gallonhydrogen peroxide tank and two feed pumps, one 500-gallon

acid storage tank and two feed pumps, and one 500-gallonbase storage tank and two feed pumps. The total additionalcost for all three chemical feed modules for each case will beabout $10,000.

Auxiliary equipment considered in this analysis isassociated with a groundwater remediation project andincludes one sedimentation tank, two equalization tanks, twocartridge filters, one filter press, and monitoring equipment.One 2,000-gallon sedimentation tank will be locateddownstream of the extraction wells and upstream of thecartridge filters to allow solids to settle out before treatment.This tank costs about $5,000. Two 5,000-gallon equalizationtanks are needed to minimize fluctuations in VOCconcentrations. While one tank is being filled, the other willbe emptied. These tanks will be located downstream of thecartridge filters. Both equalization tanks will cost a total ofabout $14,000. All tanks used during the remediation areassumed to have closed tops with vents. A venting systemthat includes duct work and carbon columns will be neededto eliminate fugitive emissions from the tanks. This ventingsystem will cost about $25,000.

Filtration will be required to remove any suspendedsolids from the sedimentation tank effluent. Two cartridgefilters will be installed on the perox-pureTM feed line.Cartridges cost about $2,000 each for the l0- and 50-gpmflow rate scenarios, for a total cost of $4,000. Cartridgescost about $4,000 each for the 100-gpm flow rate scenario,for a total cost of $8,OOO. The costs of replacement filtersare included in Consumables and Supplies Costs.

A filter press will be needed to dewater the sedimentcollected in the sedimentation tank and any other tanks thatmay accumulate sediment. The size of the filter press willbe determined after a bench-scale study is performed. Thisanalysis assumes that a 4-cubic-foot filter press will be used,at a cost of about $50,000.

Monitoring equipment includes a spectrophotometer tomeasure hydrogen peroxide concentration, a pH meter, athermometer, and other as needed. This equipment can bepurchased for about $10,000.

Total auxiliary equipment costs, including venting ductwork, for the l0- and 50-gpm flow rate scenarios will beabout $108,000. The total auxiliary equipment costs for the100-gpm flow rate scenario will be about $112,000.

For Case 1, total capital costs will be about $819,000 forthe lo-gpm flow rate scenario; $906,000 for the 50-gpm flowrate scenario; and $1,015,000 for the 100-gpm flow ratescenario. Of these costs, only the perox-pure and feedmodules are associated with operating the perox-puresystem. The costs of these components account for about 13percent of total capital costs for the lo-gpm flow ratescenario, 20 percent for the 50-gpm flow rate scenario, and

29 percent for the lOO-gpm flow rate scenario. In addition,the shelter building accounts for about 56 percent of thetotal capital costs for the l0-gpm flow rate scenario, about50 percent for the 50-gpm flow rate scenario, and about 45percent for the 100-gpm flow rate scenario.

For Case 2, total capital costs will be about $764,000 forthe l0-gpm flow rate scenario; $776,000 for the 50-gpm flowrate scenario; and $780,000 for the 100-gpm flow ratescenario. Of these costs, only the perox-pure system andfeed modules are associated with operating the perox-pure”’system. The costs of these components account for about 7percent of total capital costs for the each flow rate scenario.In addition, the shelter building accounts for about 60percent of the total capital costs for the l0-gpm flow ratescenario, about 59 percent for the 50-gpm flow rate scenario,and about 58 percent for the lOO-gpm flow rate scenario.The cost differences between the two cases are due to lowersystem costs associated with Case 2.

4.2.4 Startup Costs

Startup costs include the cost of developing a health andsafety program. For each case, developing a health andsafety program will also include providing a 40-hour healthand safety training course. This cost is associated with agroundwater remediation project and is estimated to costabout $5,000.

4.2.5 Labor Costs

Labor costs include the total staff needed for operatingand maintaining the perox-pure system, conducting anannual health and safety refresher course, and medicalmonitoring. The labor wage rates provided in this analysisdo not include overhead or fringe benefits. Once the systemis functioning, it is assumed to operate continuously at thedesigned flow rate, except during routine maintenance. Oneoperator will monitor the equipment, make any requiredhydrogen peroxide dose and pH adjustments, performroutine maintenance, and perform routine monitoring andsample analysis. PSI estimates that under normal operatingconditions, an operator will be required to work only 3.5hours per week. However, because finding a person willingto work for this short period of time may be difficult; thisanalysis assumes that the operator will work 8 hours duringthe weekdays, and 2 hours, at time and one-half, during eachday of the weekend. Annual staff costs for each case will beabout $36,000. The annual health and safety refreshercourse and medical monitoring are associated with agroundwater remediation project and will cost about $3,000per person.

For each case, total annual labor costs will be about$39,000. Of these costs, about 92 percent is associated withoperating the perox-pure treatment system.

4.2.6 Consumables and Supplies Costs

Consumables and supplies costs include hydrogenperoxide, sulfuric acid and sodium hydroxide solutions forpH adjustment, cartridge filters, activated carbon columns,disposable personal protective equipment (PPE) and PPEdisposal drums, UV lamps, sampling supplies, and propane.Costs of these items are discussed below.

Hydrogen peroxide is commercially available in solutionsof 30 to 50 percent by weight. It can be purchased in bulk,delivered to the site when needed, and stored in a 2,000-gallon tank (see Capital Equipment Costs). Hydrogenperoxide has a shelf life of over 1 year and a density ofabout 10 pounds per gallon. A 50 percent solution can bepurchased for about $0.12 per pound including delivery; thiscost is associated with operating the perox-pure treatmentsystem. The quantities of hydrogen peroxide consumeddepend on the system flow rate and the waste characteristics.Based on the SITE demonstration, the annual hydrogenperoxide costs for the l0-gpm flow rate scenario would beabout $450 for Case 1 and $300 for Case 2. Annualhydrogen peroxide costs for the 50-gpm flow rate scenariowould be about $2,200 for Case 1 and $1,600 for Case 2. Forthe 100-gpm flow rate scenario, annual hydrogen peroxidecosts would be about $4,450 for Case 1 and $3,100 forCase 2.

This analysis assumes sulfuric acid will be used to adjustpH before treatment, and sodium hydroxide will be used toadjust pH after treatment. In addition, this analysis assumesthat a 93 percent solution of sulfuric acid can be purchasedin bulk, delivered to the site when needed, and stored in a500-gallon tank (see Capital Equipment Costs). Sulfuric acidhas a shelf life of over 1 year and a density of 15 pounds pergallon. It can be purchased for about $0.09 per pound. Thequantities of sulfuric acid consumed depend on the flow rateand the initial pH of the contaminated groundwater. Basedon the SITE demonstration, annual sulfuric acid costs for theIO-gpm flow rate scenario would be about $600 for eachcase. Annual acid costs for the 50-gpm flow rate scenariowould be about $3,000 for each case. For the 100-gpm flowrate scenario, annual acid costs would be about S6,OOO foreach case.

Sodium hydroxide solution has a shelf life of over 1 yearand a density of 15 pounds per gallon. A 50 percent solutionof sodium hydroxide can be purchased for about $0.10 perpound. Based on the SITE demonstration, annual sodiumhydroxide costs for the l0-gpm flow rate scenario would beabout $2,000 for each case. Annual sodium hydroxide costsfor the 50-gpm flow rate scenario would be about $10,000 foreach case. For the 100-gpm flow rate scenario, the annualsodium hydroxide costs would be about $20,000 for eachcase.

27

This analysis assumes two cartridge filters capable ofscreening material larger than 3 micrometers will be installedupstream of the perox-pureTM unit and downstream of thesedimentation tank. These filters should remove anysuspended solids in the sedimentation tank effluent.Replacement frequency depends on the quality of thegroundwater to be treated and the flow rate. However, thisanalysis assumes one filter will need to be changed every 3months and is associated with a groundwater remediationproject. The dual filter system allows one filter to be usedwhile the other is replaced. For each case, replacementfilters will cost about $50 each or $200 per year.

Activated carbon columns on the venting system for thesedimentation and equalization tanks are assumed to bereplaced every 6 months and are associated with agroundwater remediation project. For each case,replacement columns cost about $500 each, for a total of$1,000 per year. The actual rate at which these carboncolumns will need replacement depends on theconcentrations of VOCs in the water being treated.

PPE is associated with a groundwater remediationproject and typically consists of nondisposable and disposableequipment. Nondisposable equipment consists of steel-toeboots and a full-face air respirator. For each case,disposable PPE includes latex inner gloves, nitrile outergloves, and safety glasses. Disposable PPE for each case isassumed to cost about $600 per year for the one operator,regardless of flow rate. Disposable PPE is assumed to behazardous and will need to be disposed of in 24-gallon fiberdrums. Any potentially hazardous wastes will also bedisposed of in these drums. One drum is assumed to befilled every 3 months. Drums cost about $12 each. For eachcase, total annual drum costs will be about $50.

UV lamp usage is associated with operating the perox-pureTM treatment system. PSI warranties its UV lamps for3,000 hours of use. At this rate, the lamps will need to bechanged three times per year. Five-kW lamps cost about$500 each and l5-kW lamps cost about $600. For Case 1,three 5-kW lamps are used during treatment in the l0-gpmunit for a total annual cost of $4,500. Five 15-kW lamps areused in the 50-gpm unit for a total annual cost of $9,000.Ten 15-kW lamps are used in the 100-gpm unit for a totalannual cost of $18,000. For Case 2, one 5-kW lamp is usedduring treatment in the l0- and 50-gpm units for a totalannual cost of $1,500. Two 5-kW lamps are used in the lOO-gpm unit for a total annual cost of $3,000. PSI will disposeof used UV lamps at no cost.

Sampling supplies are associated with a groundwaterremediation project and consist of sampling bottles andcontainers, ice, labels, shipping containers, and laboratoryforms for off-site analyses. For routine monitoring,laboratory glassware will also be needed. The number andtypes of sampling supplies will be based on the analyses to

be performed. Costs for laboratory analyses are presented in the Analytical Services Costs section. For each case, thesecosts are assumed to be $l,OtXl per year.

Propane use is associated with a groundwaterremediation project and will be needed to heat the shelterbuilding during the colder months of the year. Annualpropane usage will be based on the square footage of theshelter building, number of cold days, building materials, andother variables. For each case, propane is assumed to be$3,000 per year.

For Case 1, total consumables and supplies costs areestimated to be $13,000 for the lo-gpm flow rate scenario;$30,000 for the 50-gpm flow rate scenario; and $54,000 for100-gpm flow rate scenario. Of these costs, feed chemicalsand the UV lamps are direct costs of operating the perox-pure system. These costs account for about 58 percent ofthe total consumables and supplies costs for the lo-gpm flowrate scenario, about 81 percent for the 50-gpm flow ratescenario, and about 90 percent for the 100-gpm flow ratescenario. UV lamps account for between 30 and 35 percentof the total consumables and supplies costs. Feed chemicalsaccount for between 23 and 56 percent of the totalconsumables and supplies costs.

For Case 2, total consumables and supplies costs areestimated to be $10,000 for the lo-gpm flow rate scenario;$22,000 for the 50-gpm flow rate scenario; and $38,000 for100-gpm flow rate scenario. Of these costs, feed chemicalsand the UV lamps are direct costs of operating the perox-pureTM treatment system. These costs account for about 44percent of the total consumables and supplies costs for thelo-gpm flow rate scenario, and about 73 percent for the 50-gpm flow rate scenario and about 84 percent for the lOO-gpm flow rate scenario. Feed chemicals account for between29 and 77 percent of the total consumables and suppliescosts. UV lamps account for between 8 and 15 percent ofthe total consumables and supplies costs.

The difference in costs between the two cases isattributable to the higher power requirements and, hence,the higher number of UV lamps needed to treat thegroundwater characteristics of Case 1.

4.2.7 Utilities Costs

Total utility costs are based on the electrical power usedto operate the entire treatment system and all auxiliaryequipment including the shelter building. The mercury-vapor UV lamps, which are associated with operating theperox-pure unit, draw significant electricity. This analysisassumes that electricity costs about $0.07 per kilowatt-hour(kWh) inclusive of usage and demand charges. In addition,this analysis assumes that all auxiliary equipment, which isassociated with a groundwater remediation project, will draw

an additional 30 percent of the total electrical power of theperox-pure system.

For Case 1, the l0-gpm unit annually draws about131,000 kWh, the 50-gpm unit draws about 655,200 kWh, andthe 100-gpm unit draws about 1,310,400 kWh. Total annualutility costs for operating the lamps only will be about$9,200, $45,900, and $91,700 for the l0-, 50-, and 100~gpmflow rate scenarios, respectively. Auxiliary equipment usagewill cost an additional $2,800, $13,800, and $27,500 for therespective three flow rate scenarios. Operating the perox-pure system accounts for about 77 percent of total utilitycosts for all three flow rate scenarios.

For Case 2, the l0- and 50-gpm units annually drawabout 43,680 kWh, and the 100-gpm unit draws about 87,360kWh. Total annual utility costs for operating the lamps onlywill be about $3,100 for the l0- and 50-gpm flow ratescenarios, and $6,100 for the 100-gpm flow rate scenario.Auxiliary equipment usage will cost an additional $900 forthe l0- and 50-gpm flow rate scenarios, and $1,800 for the100-gpm flow rate scenario. Operating the treatment systemaccounts for about 77 percent of total utility costs for allthree flow rate scenarios.

The difference in costs behween the two cases isattributable to the higher power requirement to operate theUV lamps, needed for treating the groundwater in Case 1.

Electrical costs can vary by as much as 50 percentdepending on the geographical location and local utilityrates. A diesel gas-powered generator can also be used asa backup or alternate source of electric power, but it willcost considerably more than similar power supplied by localutilities.

4.2.8 Effluent Treatment and Disposal Costs

The perox-pureTM system does not generate sludge orspent carbon that requires further processing, handling, ordisposal. Ideally, the end products of the process are water,carbon dioxide, halides, and sometimes organic acids.Effluent monitoring will be conducted routinely by theoperator (see Labor Costs). The effluent can be dischargeddirectly to a nearby surface water body, provided theappropriate permits have been obtained (see Permitting andRegulatory Requirements Costs).

4.2.9 Residuals and Waste Shipping and Handlingcosts

Spent cartridge filters and PPE drums are associatedwith a groundwater remediation project. These wastes areconsidered hazardous and will require disposal at apermitted facility. For each case, this analysis assumes thatabout six drums will be disposed of annually for the l0- and

50-gpm flow rate scenarios, and eight drums will be disposedof for the 100-gpm flow rate scenario. The cost of shipping,handling, and transporting drums to a hazardous wastedisposal facility are assumed to be $1,000 per drum. Totaldrum disposal costs for each case will be about $6,000 forthe l0- and 50-gpm flow rate scenarios and $8,000 for thel00-gpm flow rate scenario.

In addition, filter cake from the falter press is consideredhazardous waste and will require disposal at a permittedfacility. Because the amount of filter cake generated willvary greatly from site to site, this analysis does not presentthe costs of filter cake disposal. PSI will dispose of used UVlamps at no cost.

4.2.10 Analytical Services Costs

Analytical costs are associated with a groundwaterremediation project and include laboratory analyses, datareduction and tabulation, quality assurance/quality control(QA/QC), and reporting. For each case, this analysisassumes that one sample of untreated water and one sampleof treated water will be analyzed for VOCs each month,along with trip blank, duplicate, and matrix spike/matrixspike duplicate samples. Monthly laboratory analyses willcost about $1,250; data reduction, tabulation, QA/QC, andreporting is estimated to cost about $750 per month. Totalannual analytical services costs for each case are estimatedto be about $24,000 per year.

4.2.11 Maintenance and Modifications Costs

Annual repair and maintenance costs apply to allequipment involved in every aspect of groundwaterremediation with the perox-pure”’ system. No modificationcosts are assumed to be incurred. Total annual maintenancecosts are estimated to be about 10 percent of capitalequipment costs. For Case 1, total annual maintenance costsfor the perox-pureTM system are estimated to be about$11,000 for a l0-gpm unit; $18,500 for a 50-gpm unit; and$29,000 for a 100-gpm unit. Total annual maintenance costsfor the auxiliary equipment are estimated to be about$10,800 for the l0- and 50-gpm units, and $11,200 for a 100-gpm unit.

For Case 2, total annual maintenance costs for theperox-pure treatment system are estimated to be about$5,500 for each flow rate scenario. Total annualmaintenance costs for the auxiliary equipment are estimatedto be about $10,800 for the l0- and 50-gpm units, and$11,200 for a 100-gpm unit.

The difference in costs between the two cases isattributable to the higher capital equipment costs in Case 1.These costs are higher because more power is required toachieve treatment goals.

29

4.2.12 Demobilization Costs

Site demobilization includes shutdown disassembly,transportation, and disposal of perox-pureTM equipment andauxiliary equipment at a licensed hazardous waste disposalfacility. This analysis assumes the perox-pure” system willhave no salvage value at the end of the project. Site cleanupand restoration are also included in demobilization costs.For each case, this analysis assumes the costs from shutdownto disposal for all activities associated with a groundwaterremediation project will be about $30,000, including sitecleanup, restoration, and building decontamination. Thisanalysis assumes the shelter building will remain standing atthe site. All disposal activities associated with operating theperox-pure” system are estimated to be about $10,000. Foreach case, demobilization is estimated to take about 1 weekto complete and will cost a total of about $4O,OOO.

The costs of demobilization, however, will occur at theend of the remediation project. To complete the project,this analysis assumes that the l0-gpm unit will be used for 50years, the 50-gpm unit will be used for 10 years, and the lOO-gpm unit will be used for 5 years. Therefore, based on theannual inflation rate of 5 percent, the net future values ofthis cost for l0-gpm, 50-gpm, and lOO-gpm units areestimated to be about $459,000; $65, 000; and $51,000,respectively. These figures were used to calculate the totalcosts for a groundwater remediation project presented inTables 4-l and 4-2.

30

Section 5References

Evans, G., 1990, Estimating Innovative Technology Costs forthe SITE Program. Journal of Air and WasteManagement Association. Volume 40, No. 7 (July).

Federal Register, 199Oa, U.S. Environmental ProtectionAgency (EPA), National Oil and Hazardous SubstancesPollution Contingency Plan; Final Rule. Volume 55, No.46 (March 8).

Federal Register, 199Ob, EPA Proposed Rules for CorrectiveAction for Solid Waste Management Units at HazardousWaste Management Facilities. Volume 55, No. 145 (July27).

Glaze, W., and others, 1987, The Chemistry of WaterTreatment Processes Involving Ozone, HydrogenPeroxide, and Ultraviolet Radiation. Ozone Science andEngineering.

U.S. Department of Energy (DOE), 1988, RadioactiveWaste Management Order. DOE Order 5820.2A(September 26).

U.S. Environmental Protection Agency (EPA), 1987a,Alternate Concentration Limit (ACL) Guidance. Part1: ACL Policy and Information Requirements,EPA/530/SW-87/017.

EPA, 1987b, Joint EPA-Nuclear Regulatory AgencyGuidance on Mixed Low-Level Radioactive andHazardous Waste. OSWER Directives 9480.00-14 (June29), 9432.00-2 (January 8), and 9487.00-8 (August 3).

EPA, 1988, CERCLA Compliance with OtherEnvironmental Laws: Interim Final, Office of SolidWaste and Emergency Response (OSWER).EPA/54O/G-89/006.

EPA, 1989, EPA Memorandum from OSWER to WasteManagement Division and Office of Regional Counsel,Applicability of Land Disposal Restrictions to RCRAand CERCLA Groundwater Treatment Reinjection.Superfund Management Review: Recommendation No.26 (December 27).

31

Appendix AVendor Claims for the Technology

A.1 Introduction

Because 50 percent of U.S. drinking water is obtainedfrom groundwater resources, the removal of contaminantsfrom polluted groundwater has become a majorenvironmental priority. Increasing groundwater pollution hasbeen caused by leaking underground fuel and chemicalstorage tanks, accumulations of pesticides and herbicides,landfill leachates, and improper handling and disposal ofindustrial chemicals and wastewater effluent. Contaminantsoften migrate through soil into municipal water wells,aquifers, and other water resources. The U.S.Environmental Protection Agency has identified 129 priorityorganic contaminants; with such diversity in molecularstructure and chemical properties, these contaminantspresent a challenge to traditional treatment technologies.

The perox-pureTM advanced chemical oxidation system,developed by Peroxidation Systems, Inc. (PSI), of Tucson,Arizona, provides an on-site treatment process that destroyscontaminants in groundwater and wastewater. The patentedsystem design has been used to meet the most stringentfederal and state groundwater cleanup regulations.

Since it entered the market in 1987, the perox-puresystem has been used on site to treat organic contaminationin groundwater at approximately 80 locations in the UnitedStates, Canada, Puerto Rico, and Europe. The perox-puresystem has been used to treat groundwater, wastewater,leachate, dredge water, bioeflluent, and municipal watercontaminated with volatile organic compounds, semivolatileorganic compounds, aromatic compounds, polynucleararomatic hydrocarbons, phenols, petroleum hydrocarbons,and a number of other organic compounds.

A.2 Description of the perox-pure System

The perox-pureTM system rapidly breaks the bondsbetween atoms in organic molecules, thereby allowing theatoms to form simpler, nontoxic compounds, such as carbondioxide and water. In the perox-pureTM system, waterpumped from a contaminated source is combined withhydrogen peroxide, a chemical oxidant. Hydrogen peroxide

33

is preferred to ozone as a chemical oxidant, because ozoneis not very soluble in water, which results in long oxidationtimes. The higher solubility of hydrogen peroxide in watergreatly simplifies the reactor design in terms of oxidantaddition, mixing of reactants, and reduction of toxic gasemissions. Hydrogen peroxide is commercially available insolutions of 30 to 50 percent by weight. Systems usinghydrogen peroxide also require significantly less storage andfeed equipment and are less expensive relative to those usingozone, because ozone must be generated on site. Finally,because ozone is a toxic gas, it requires a special gas phasedecomposition system and an air discharge permit.

The mixture of contaminated water and hydrogenperoxide passes through a series of reactor chambers thatexpose it to high intensity ultraviolet (UV) lamps mountedinside protective quartz tubes. Inside each of the UV lampchambers, a photochemical reaction occurs that forms highlyreactive hydroxyl radicals (OH*) from hydrogen peroxide(see Table A-l). Energy from the UV lamps absorbed bythe target compounds also increases the compounds’oxidation rate.

Hydrogen, peroxide injection points at the inlet of eachlamp section allow hydrogen peroxide to be added to eachreactor separately. This arrangement, called the splitter, aidsthe operator in maintaining the ideal balance of hydroxylradicals.

Hydroxyl radicals destroy the bonds in organiccontaminants between carbon and the attached groups ofatoms, forming simpler, nontoxic chemical compounds. Thetreated water exits the system at the top of the reactorchambers. The only additive to the system, hydrogenperoxide, is consumed in the process, and the systemgenerates no air emissions or by-products that requiredisposal.

The perox-pureTM system is effective in treating a varietyof organic compounds present in contaminated waters andis not limited by mass transfer, as are liquid-phase carbonadsorption and air stripping.

A.3 perox-pureTM Systems

The perox-pureTM systems are compact and designed tooperate unattended on a continuous basis. The skid-mounted unit consists of a control panel, lamp driveenclosures, and a reactor chamber. An accompanyinghydrogen peroxide module is also skid-mounted and providesstorage, feed pumps, and an eye wash station and safetyshower. For sites where pH adjustment of influentwastewater or effluent is necessary, a skid-mounted acid feedmodule or base feed module is used in addition to theequipment listed above.

The combined use of UV lamps and hydrogen peroxideresults in improved reaction rates. Moreover, a smaller,simpler design is possible, which reduces space requirements,the number of potential replacement parts, andcorresponding maintenance costs.

PSI can supply perox-pureTM units in various sizes. PSIhas designed, built, and installed systems capable of treatingflow rates varying from 5 gallons per minute to thousands ofgallons per minute. The unit size required for a particularapplication is determined through laboratory treatabilitystudies and on-site tests.

A diagram of the perox-pureTM unit is shown in Figure A-l. The perox-pureTM oxidation unit consists of multiplehorizontal sections connected in series. Inside each section,

one high intensity, medium-pressure UV lamp is mountedinside a quartz tube. The lamp and tube assembly arepositioned perpendicular to the side walls of the chamber.

A.4 Design Improvements

The advanced design of current perox-pureTM systemsdecreased the costs of earlier systems by 50 percent or more.Improvements include an innovative reactor design,anticorrosion lining in the reactors, better conversionefficiencies from electrical power to UV energy, andimproved methods for sequential hydrogen peroxideaddition. Engineering design improvements also make iteasier to adjust such parameters as flow rate, pH, UVenergy, and the addition of hydrogen peroxide to individualreactor chambers to maximize destruction rates. Anoptional programmable logic controller can be installed tointegrate the system into a fully automated remote treatmentsystem.

Advanced perox-pureTM systems include a patented,automated, self-cleaning mechanism for the UV lamps thatvirtually eliminates the scaling of the quartz tubes. Thecleaning mechanism consists of a wiper that fits in theannular space behveen the quartz tube and the oxidationchamber wall. Propelled by water being treated in thesystem, the wiper regularly removes deposits from the quartztubes. As a result, costs associated with unscheduledmaintenance to the system have been substantially reduced.

34

figure A- 1 isometric Diagram of perox-pure” Unit

35

These engineering improvements have transformed theperox-pureTM system into an efficient, low maintenance, andhighly reliable advanced oxidation system that is costcompetitive with carbon adsorption and air strippingtechnologies.

A.5 Pretreatment

A pretreatment system is sometimes needed to reducethe levels of certain inorganic species in the water. If highlevels of suspended solids or turbidity are present, filtrationand clarification may be required to provide optimumtreatment. Acid can also be added to increase treatmentefficiency by shifting the equilibrium from carbonate andbicarbonate species to carbonic acid.

A.6 perox-purem Applications

Although a broad spectrum of aqueous wastes can betreated with the perox-pureTM technology, the most commonapplication is for water containing less than 500 milligramsper liter of organic compounds. Higher concentrations canbe treated; however, the longer oxidation times needed totreat higher concentrations may require batch or partialrecycle treatment. In addition, treated water may need to becooled to keep it from overheating the system. A partial listof applications of the perox-pure technology is shown inTable A-2.

While the chemistry of the perox-pureTM technology hasa number of complex interactions which depend on thecharacteristics of the water, a large database is availablefrom which initial predictions can be made.

To determine if the perox-pure” system is applicable toa specific contamination problem, a team of applicationspecialists is available at the PSI corporate headquarters inTucson, Arizona (800-552-8064). After gathering theinformation available from the potential user, the applicationspecialists can assess the general applicability of theperox-pure technology and, in most cases, provide anestimate of the capital and operating costs for perox-pure”’technology. If requested, written estimates are preparedwhich document the figures and provide information on theavailable perox-pureTM systems and services.

PSI requires the following information to evaluate theapplication of the perox-pure technology:

l Contaminated water flow rate or production rate

l Identities and concentrations of organiccontaminants

l Treatment objectives

Other information that allows PSI to provide a morereliable estimate of treatment includes the following:

to

l Source of contaminated water

l Identity and concentrations of inorganics such asiron, chloride, and nitrate

l Levels of total organic carbon, chemical oxygendemand, pH, alkalinity, dissolved solids, turbidity,color, and suspended solids

l Current treatment and pretreatment systems in use

The estimates provided by PSI are intended to be useddetermine if the perox-pureTM technology is an

economically feasible method of treatment compared withthe costs of other potential treatment processes. Since thewater quality has such a significant impact on theperox-pureTM system performance, a process assessment atthe PSI Testing Laboratory, or in the field, with an actualsample of the water is necessary to provide definitivetreatment cost projections.

A.7Advantages over Carbon Adsorption and AirStripping Technologies J

The perox-pureTM advanced chemical oxidationtechnology has demonstrated major advantages over carbonadsorption and air stripping technologies. From a regulatoryperspective, the major advantage of the perox-pureTM

technology is that it creates no secondary pollutants to treator haul away. This benefit has become more important asfederal, state, and local regulations become more stringentregarding the allowable contaminant concentrations of airemissions generated using air stripping as well as thedisposal of secondary pollutants such as spent carbon. Theseregulations have increased the costs of monitoring anddisposal as well as the risk of legal liability associated withtreatment technologies. The perox-pure technology greatlyreduces the risk of future liability associated with thecreation of secondary pollutants.

The perox-pureTM technology is advantageous for treatingcontaminated groundwater in residential neighborhoods andhighly technical industrial complexes because the systemoffers a low-profile, on-site treatment that produces minimalnoise. In addition, the perox-pure systems are verycompact and can be positioned among existing equipment.As a result, public relations problems are minimized.

The perox-pureTM technology is ideal for thoseapplications where carbon adsorption and air stripping arenot viable. For example, some priority pollutants, such asmethyl-tert-butyl ether, vinyl chloride, and methylenechloride, are not readily adsorbed by carbon. Many commonorganic contaminants, such as isopropyl alcohol and acetone,

36

Table A-2 Partial List of perox-pore” Technology Applictions

waste .Stream Principal Contaminants’

Bioeffluent Chlorobenzene

Site Location

Connecticut

Dredge Water

Effluent

Groundwater

Polychlorinated Biphenyls

BTX

Phenol

MeCI, Phenol, PAH

Nitrated Esters

Phenol

Phenols, Nitrophenols

IPA. TOC, TCA, DCE, MEK

Acrylic Acid, Butyl Acrylate

BTEX

TCE

BTX

BTX

TCA, Freon, MeCI, BTX

TCA, TCE

TCE

TCE, PCE, BTX, TCA

TCE, PCE, DCE, TCA, MeCI, Chloroform

TCE, PCE, TCA, DCE

TCE, TCA, CC/, MeCl

TCE, TCA, PCE, DCE

Tetrahydrofuran

BTX

TCE, PAHs

TCE, PCE, TCA, DCE

BTX

MeCI, TCA

TCE, DCE, PCE, MeCl

TCE, DCE, PCE, TCA

Pentachlomphenol

Massachusetts

California

New Jersey

North Carolina

Pennsylvania

Pennsylvania

Arizona

Arizona

Arizona

California

California

California

California

California

California

California

California

California

California

California

Colorado

Louisiana

Maryland

Massachusetts

Massachusetts

New Jersey

New York

37

-

Table A-2 Partial list of perox-pure’” Technology Applications ~continuedl

Waste Stream Principal Contaminants’

Leachate Mixed Organic Acids, Ketones, and VOCs

Municipal Water Humic Acid/Color Control

Recycle Chemical Oxygen Demand

Bacteria, Phenol, Formaldehyde

Miscellaneous Hydrazine

Hydrazine, DIMP

Notes:

Acronyms used in this table are detined as follows:

BTEXBTX

CC4DCEDIMPIPAMeClMEKPAHPCETOCTCETCAVOC

Benzene, toluene, ethylbenzene, and xyleneBenzene, toluene, and xyleneCarbon tetrachlorideDichloroetheneDiisopropyl methylphosphonateIsopropyl alcoholMethylene chlorideMethyl ethyl ketonePolynuclear aromatic hydrocarbonTetrachloroetheneTotal organic carbonTtichlometheneTrichlomethaneVolatile organic compound

are not removed by air stripping. The perox-pureTM

technology provides reliable, consistent destruction of thesecontaminants.

A.8 Other Advantages

To minimize capital investment and avoid anunreasonably long-term commitment to the treatmentprocess, PSI offers a 5-year plan in which equipment,maintenance, technical services, and hydrogen peroxide areprovided in one monthly fee. This plan allows customers toaccurately budget costs over the life cycle of the project, and

Site Location

New Hampshire

California

Arizona

Ohio

Colorado

Colorado

it allows PSI to respond to changing regulations on behalf ofthe client and initiate process changes and improvements asneeded. In some cases, the monthly fee has been reducedas the process or water quality has improved.

A.9 Technology Combinations

In many water treatment systems, the perox-pureTM

technology has been paired with carbon adsorption, airstripping, or biological treatment. Depending on the water

quality and treatment objectives, the perox-pureTM technologycan be combined with other technologies to produce a morecost-effective solution than is achieved with any singleprocess.

38

Appendix BSITE Demonstration Results

In April 1991, the U.S. Environmental Protection Agency(EPA) learned that Peroxidation Systems, Inc. (PSI), wascontracted by Lawrence Livermore National Laboratory(LLNL) to perform pilot-scale chemical oxidation studies aspart of remediation activities at the LLNL site. At that time,EPA and PSI discussed the possibility of PSI participating inthe Superfund Innovative Technology Evaluation (SITE)program to demonstrate how PSI’s perox-pure”’ chemicaloxidation technology could be used to treat contaminatedgroundwater at LLNL Site 300 in Tracy, California. EPAsubsequently accepted the perox-pureTM technology into theSITE demonstration program. Through a cooperative effortbetween the EPA Office of Research and Development(ORD), EPA Region IX, LLNL, and PSI, the perox-pure”technology was demonstrated at the LLNL site under theSITE program. This appendix briefly describes the LLNLsite and summarizes the SITE demonstration activities anddemonstration results.

B.l Site Description

LLNL is a 640-acre research facility about 45 miles eastof San Francisco and 3 miles east of Livermore, California(see Figure B-l). Development of the site began in 1942,when it was used as a U.S. Navy aviation training base.Subsequent activities at LLNL varied considerably under themanagement of several government agencies, including theAtomic Energy Commission, the Energy Research andDevelopment Agency, and the U.S. Department of Energy,which is the present owner. Various hazardous materials,including volatile organic compounds (VOC), metals, andtritium were used and released at the site.

The SITE demonstration was conducted at Site 300,which is operated by LLNL but is located separately fromthe LLNL main campus (see Figure B-l). LLNL establishedSite 300 as a high-explosives test area in 1955. Site 300occupies 11 square miles in the Altamont Hills about15 miles southeast of Livermore and 8.5 miles southwest ofTracy, California. Site 300 operations consist of fouractivities: (1) hydrodynamic testing; (2) charged particle-beam research; (3) physical, environmental, and dynamictesting; and (4) high-explosive formulation and fabrication.

EPA chose a specific area of Site 300 for the technologydemonstration. This area is called the General ServicesArea (GSA). The GSA occupies about 80 acres in thesoutheastern corner of Site 300. Various administrative,medical, engineering, and maintenance operations areconducted in buildings located in the GSA. Before 1982,several GSA facilities used dry wells to dispose of wasterinse, process, and wash waters. Wastes from these facilitiesmight have included the following: photo laboratory rinsewater; water- and oil-based paint waste; automotive shopwaste containing degreasing solvents; and acid dip rinsewater. Between 1983 and 1984, the dry wells wereinvestigated and closed. After the dry well closure,wastewater from these activities was shipped off site fortreatment and disposal. Other wastes are currently storedon site in a permitted hazardous waste storage area. Thesuspected sources of groundwater contamination in the GSAare the dry wells, accidental releases and leaks during facilityoperations, and leaking underground fuel storage tanks.

LLNL’s Environmental Restoration Division submitteda remedial investigation (RI) report and a feasibility study(FS) report to EPA Region IX in May and December 1990,respectively. The FS report outlined a treatment system forcontaminated groundwater from the central GSA. Thesystem will be designed to treat both vapor and groundwaterobtained from extraction wells in the area. Groundwater willbe collected both on and off site for remediation. Currently,several treatment alternatives are being evaluated at the site.

B.2 Site Contamination Characteristics

Dry wells, accidental releases and leaks during facilityoperations, and leaking underground fuel storage tanks aresuspected sources of groundwater contamination in thecentral GSA. Data from the RI report (LLNL, 1990) wereused to select the shallow aquifer at the site as the candidatewaste stream for the technology demonstration.

In May 1992, LLNL performed an 8-hour drawdownpump test using existing groundwater extraction wells. Thegroundwater was sampled throughout the test and analyzedfor VOCs, semivolatile organic compounds (SVOC), metals,

39

tn Jose

\

Figure B- 1 LLNL Site Lochon

and a variety of other parameters, such as pH and alkalinity.Samples for VOC and SVOC analyses were collected afterapproximately 1, 3, 6, and 8 hours of pumping time. Theseanalyses showed that (1) only five VOCs were present abovedetection limits, (2) SVOCs were not present abovedetection limits, and (3) all five VOCs detected showedgradual decreases in concentration over the 8-hour testduration. At the end of 8 hours, trichloroethene (TCE);tetrachloroethene (PCE); 1,1-dichloroethene (l,l-DCE); 1,2-dichloroethene (1,2-DCE); and l,l,l-trichloroethane (TCA)were present at 1,200 micrograms per liter (pg/L), 95 pg/L,7.1 pg/L, 8.7 pg/L, and 7.5 pg/L, respectively.

The following parameters were also measured in thegroundwater samples collected after 6 hours of pumping:(1) pH was 7.8; (2) alkalinity was 300 milligrams per liter(mg/L) as calcium carbonate; (3) the concentration of totaldissolved solids was 930 mg/L; (4) the concentration of ironwas 10 pg/L, and (5) the concentration of manganese was 20tcglL*

B.3 Review of SITE Demonstration

The SITE demonstration was divided into three phases:(1) site preparation; (2) technology demonstration; and (3)site demobilization. These activities and a review oftechnology and perox-pureTM system performance during thephases are described below.

B.3.1. Site Preparation

Approximately 10,000 square feet of relatively flatground surface was used for the perox-pure= chemicaloxidation system and support equipment and facilities, suchas treated and untreated water storage tanks, nonhazardousand potentially hazardous waste storage containers, an officeand field laboratory trailer, and a parking area. Atemporary enclosure covering approximately one-fourth ofthe demonstration area was erected to provide shelter forthe perox-pureTM system during inclement weather. Sitepreparation included setting up major support equipment,on-site support services, and utilities. These activities arediscussed below.

Major Support Equipment

Support equipment for the perox-pureTM systemdemonstration included a cartridge filtration system toremove suspended solids from groundwater, storage tanksfor untreated and treated groundwater, an acid feed modulefor untreated groundwater, a base feed module for treatedgroundwater, a spiking solution feed system, a static mixer,two 55-gallon drums for collecting equipment wash waterand decontamination rinse water, a dumpster, a forklift withoperator, pumps, sampling equipment, health- and safety-related gear, and a van. Specific items included thefollowing:

l One cartridge filtration system containing two filtersupstream of the treatment unit; the filters werecapable of removing suspended solids greater than3 micrometers in size from groundwater.

l One 55-gallon closed-top, polyethylene drumcontaining spiking solution; the drum was equippedwith a floating lid and a mixing device. During thedemonstration, a spiking solution containingchloroform; l,l-dichloroethane (DCA); and TCAwas added inline to the groundwater to evaluate theperox-pure” system’s ability to treat compoundsthat are difficult to oxidize.

l One static mixer to mix the spiking solution andgroundwater before the mixture entered theuntreated groundwater storage tank

l One 7,500-gallon bladder tank to store untreatedgroundwater and minimize VOC losses duringstorage. The tank was used (1) as an equalizationtank and (2) as a storage tank for a fewdemonstration runs performed at flow rates greaterthan the groundwater yield.

l One pump for transferring contaminated water fromthe bladder tank to the perox-pure” system and onepump for adding spiking solution inline to thegroundwater

l One sulfuric acid feed module to adjust the pH ofthe influent to the perox-pureTM system; PSIprovided the module, which consisted of a 55-gallonacid feed drum, two pumps, and flow measuringdevices.

l One sodium hydroxide feed module to adjust thepH of the effluent from the perox-pureTM system;PSI provided the module, which consisted of a 55-gallon base feed drum, two pumps, and flowmeasuring devices.

l One solid waste dumpster to store nonhazardouswastes before disposal

l A number of 55-gallon drums to contain useddisposable field sampling and analytical equipment,used disposable health and safety gear, and fieldlaboratory wastes before disposal

l A forklift with operator for setting up equipmentand for moving drummed wastes

l Sampling equipment for aqueous media and processchemical solutions

41

l Analytical equipment for measuring fieldparameters at the demonstration site

l Two 20,000-gallon steel tanks to store treatedgroundwater before analysis and disposal

l Health and safety-related equipment, such as a first-aid kit and protective coveralls, latex or similarinner gloves, nitrile outer gloves, steel-toe boots anddisposable overboots, safety glasses, and a hard hat

l A van to transport oversight personnel and supplies

On-Site Support Services

On-site laboratory analyses were conducted in a fieldtrailer measuring 12 by, 44 feet. The field trailer also servedas an office for field personnel and provided shelter andstorage for small equipment and supplies. Two toilets wereavailable near the demonstration area.

Utilities

Utilities required for the demonstration included water,electricity, and telephone service. Water was required forequipment and personnel decontamination, for fieldlaboratory use, and for drinking purposes. During operationof the demonstration unit, personnel and equipmentdecontamination required about 10 gallons per day (gpd) ofpotable water. About 5 gpd of distilled, deionized water wasneeded for field laboratory use, and about 5 to 10 gpd wereneeded for drinking water.

Electricity was needed for the perox-pureTM system, theoffice trailer, and the laboratory equipment. The perox-pureTM system required 480-volt, 3-phase electrical service.Additional electrical power (IlO-volt, single-phase) wasneeded for operating the pumps, the mixing device in thespiking solution feed system, the office trailer lights, and theon-site laboratory and office equipment.

Telephone service was required mainly for orderingequipment, parts, reagents, and other chemical supplies;scheduling deliveries; and making emergencycommunications. LLNL provided most of the supportrequired to arrange utilities for the demonstration.

B.3.2 Technology Demonstration

This section discusses (1) operational/equipmentproblems and (2) health and safety considerations associatedwith the SITE demonstration.

Operational/Equipment Problems

The SITE team, consisting of EPA’S contractors,experienced a few operational/equipment problems duringthe demonstration. Some of these problems resulted inchanges in the demonstration schedule, while the othersrequired making decisions in the field to solve the problems.These problems and solutions include the following:

l Based on the S-hour drawdown test performed inMay 1992, LLNL estimated that during thedemonstration, contaminated groundwater could beextracted from Wells. W-7-O and W-875-08 atapproximately 9 gallons per minute (gpm) and 3gpm, respectively. The demonstration tests weredesigned assuming that the combined stream wouldbe the influent to the perox-pureTM system.However, based on observations made in earl)September 1992, LLNL informed the SITE teamthat the wells might not provide the estimated yieldthroughout the demonstration. The SITE teamresolved this issue by reducing the extraction ratesfrom both wells in the same proportion, so that theinfluent characteristics would be approximately thesame as those estimated before the demonstrationThe SITE team extracted groundwater from WellsW-7-O and W-875-08 at 6 gpm and 2 gpmrespectively. This approach did not affect thtdemonstration schedule or the technologyevaluation.

l PSI requested that one of its operating facilities shipthree scaled and three clean quartz tubes tcperform Phase 3 test runs. However, of the sixquartz tubes, one tube was broken in transit. PSdid not have enough time to replace the brokertube. Therefore, Phase 3 tests (Runs I.3 and 14were performed using only two ultraviolet (UVlamps, instead of three. As a result, perox-pure”system performance with scaled tubes and cleartubes was compared based on the removalachieved in two reactors, instead of those achievedin three reactors.

l Late arrival of the perox-pureTM system (particularlythe hydrogen peroxide feed tank) and otheauxiliary equipment (such as the bladder tankpumps, and other miscellaneous items) delayed thetechnology demonstration by 3 days. However, thSITE team completed the demonstration o!schedule by working late evenings and weekends.

l At the beginning of the demonstration, while settingthe operating parameters, water inside the oxidatio-unit overheated and burned the gaskets thamaintain a water-tight seal in two of the reactor:As a result, water leaked out of the treatment unit

42

!1

s‘.

1

tI 0

l

0

l

which PSI collected in a 55-gallon drum. PSIexplained that because of its oversight a fewpneumatically operated valves did not have an airsupply, resulting in a stagnant volume of water thatoverheated. PSI also stated that the temperaturesensor inside the unit, which is located in the topreactor, did not detect the high water temperaturebecause the unit was only partially filled. Later, PSIconnected an air compressor to the unit to avoidreoccurrence of this situation. Replacement gasketsarrived the following day, causing the demonstrationto be postponed 1 day.

During the initial stage of the demonstration, due toimproper operation of valves downstream of theperox-pureTM system, the pressure inside the perox-pureTM unit exceeded the design limit and thepressure relief gasket gave way. PSI immediatelycollected the leaking water in a drum and shut offthe influent. Because PSI had a replacement gasketon site, this operational problem did not cause asignificant delay.

Halfway through the demonstration, while one testrun was in progress, the sulfuric acid level in theacid feed tank decreased significantly. As a result,the influent pH could not be lowered to the desiredlevel, and the SITE team discontinued the run. Therun was repeated after PSI filled the acid feed tankwith sulfuric acid.

Plow rates through the perox-pureTM system forRuns 7 and 8 were planned to be 50 gpm. In orderto mainta in the preferred influent pH ofapproximately 5, the system flow rate was reducedto 40 gpm. PSI’s acid feed pumps were not capableof providing enough acid to the process flow toincrease the system flow rate. This deviation didnot alter the selection of preferred conditions fromPhase 1 of the technology evaluation despite theincreased hydraulic retention time (inverselyproportional to flow rate) resulting from the changein flow rate.

The SITE team initially encountered problemsmeasuring the effluent pH at the sampling locationdownstream of the sodium hydroxide addition point.Because no static mixer was used, sodium hydroxideadded to raise the effluent pH did not adequatelymix with the effluent. Lack of proper mixing causedproblems in measuring the true effluent pH aftersodium hydroxide addition. The SITE teamresolved this issue by installing another samplingport about 100 feet downstream, just before thetreated water entered the storage tanks. Thismodification significantly reduced fluctuations in pHand provided a good measure of effluent PH.

Health and Safety Considerations

In general, potential health hazards associated with thedemonstration resulted from the possibility of exposure tocontaminated groundwater and process chemicals, includinghydrogen peroxide, sulfuric acid, and sodium hydroxidesolutions. Although the treatment system was entirelyclosed, potential routes of exposure during thedemonstration included inhalation, ingestion, and skin andeye contact from possible splashes or spills during samplecollection.

All personnel working in the demonstration area had, ata minimum, 40 hours of health and safety training and wereunder routine medical surveillance. Personnel were requiredto wear protective equipment appropriate for the activitybeing performed. Steel-toe boots were required in theexclusion zone. Personnel working in direct contact withcontaminated groundwater and process chemicals woremodified Level D protective equipment, including safetyshoes, latex inner gloves, nitrile outer gloves, and safetyglasses.

B. 3.3 Site Demobilization

After the demonstration was completed and on-siteequipment was disassembled and decontaminated, equipmentand site demobilization activities began. Equipmentdemobilization included loading the skid-mounted units ona flat-bed trailer and transporting them off site, returningrented support equipment, and disconnecting utilities.

Decontamination was necessary for the perox-pureTM unit,the storage tanks, and field sampling and analyticalequipment. Demonstration equipment was either cleanedwith potable water or steam, as required. The treated watercollected during the demonstration was tested, and theresults were provided to LLNL. LLNL disposed of thiswater appropriately. LLNL also disposed of all hazardouswastes at a permitted landfill. All nonhazardous wastes wereroutinely disposed of along with similar wastes generated byLLNL.

B.4 Experimental Design

The technology demonstration had the following primaryobjectives: (1) determine VOC removal efficiencies in thetreatment system under different operating conditions, (2)determine whether treated water met applicable disposalrequirements at the 95 percent confidence level, and (3)gather information necessary to estimate treatment costs,including process chemical dosages and utility requirements.The secondary objective for the technology demonstrationwas to obtain preliminary information on the type of by-products formed during the treatment. To accomplish theseobjectives, the following test approach and sampling andanalytical procedures were used.

43

B.4.1 Testing Approach

The perox-pureTM chemical oxidation technology wasdemonstrated over a 3-week period in September 1992.During the demonstration the perox-pureTM unit treatedabout 40,000 gallons of groundwater contaminated withVOCs. Principal groundwater contaminants included TCEand PCE, which were present at concentrations of about1,000 and 100 pg/L, respectively. Groundwater was pumpedfrom two wells into a 7,500-gallon bladder tank to minimizevariability in influent characteristics. In addition, cartridgefilters were used to remove suspended solids greater than 3micrometers in size from the groundwater before it enteredthe bladder tank. Treated groundwater was stored in two20,000-gallon steel tanks before being discharged.

The technology demonstration was conducted in threephases (see Table B-l). Phase 1 consisted of eight runs ofraw groundwater, Phase 2 consisted of four runs of spikedgroundwater, and Phase 3 consisted of two runs of spikedgroundwater to test the effect of quartz tube cleaning.These phases are described below.

The principal operating parameters for the perox-pureTM

system include hydrogen peroxide dose, influent pH, andflow rate, which determines the hydraulic retention time.These parameters were varied during Phase 1 to observetreatment system performance under different operatingconditions. For Phase 1 runs, the initial operating conditionswere based on groundwater characterization performed byLLNL in May 1992 and PSI’s professional judgment andexperience. In Runs 1, 2, and 3 the influent pH was variedwhile the other parameters were held constant to determinepreferred operating conditions. The preferred operatingconditions were those under which the concentration ofeffluent VOCs would be reduced below target levels (seeTable B-2) for spiked groundwater. After the preferredvalue for pH was determined, that value was held constant,while the other parameters were varied. Preferred operatingconditions for each parameter were determined based onquick turnaround analytical data for three selected indicatorVOCs: TCE, PCE, and TCA. Even though TCE and PCEare easily oxidized, they were chosen because they werepresent in relatively high concentrations. TCA was chosenbecause it is relatively difficult to oxidize, although it waspresent at a low concentration. Based on quick turnaroundanalytical data, PSI selected Run 3 operating conditions asthe preferred operating conditions for spiked groundwater.

Phase 2 involved spiked groundwater and reproducibilitytests. Groundwater was spiked with sufficient chloroform,DCA, and TCA so that the spiked groundwater containedabout 200 to 300 pg/L of each of these VOCs. Thesecompounds were chosen because they are relatively difficultto oxidize and because they were not initially present in thegroundwater at high concentrations. Phase 2 increased theapplicability of the demonstration data to other sites that

may be contaminated with VOCs that are difficult to oxidize.Phase 2 was also designed to evaluate the reproducibility ofperox-pureTM system performance at the preferred operatingconditions determined in Phase 1. Specifically, Runs 10, 11,and 12 were performed at Run 3 conditions to evaluate thereproducibility of perox-pureTM system’s performance.

During Phase 3, the effectiveness of the quartz tubewipers was evaluated by performing two runs using spikedgroundwater at the preferred operating conditions. Thequartz tubes used in Phase 3 tests were obtained from ahazardous waste site where the water hardness and ironcontent caused scaling on the quartz tubes. PSI obtainedtwo sets of quartz tubes for Phase 3 tests. One set of quartztubes was relatively clean, because the wipers were routinelyused to minimize scaling. The other set of tubes hadsignificant scaling because wipers were not used. Becauseonly two tubes of each type (scaled and clean) wereavailable, only two reactors were used during Phase 3.Specifically, Run 13 was performed using scaled quartztubes, while Run 14 was performed using clean quartz tubes.In both runs only two UV lamps were used.

B.4.2 Sampling and Analytical Procedures

Liquid samples were collected from the perox-pureTM

treatment system at the locations shown in Figure B-2.Table B-3 lists analytical and measurement methods usedduring the demonstration. Total organic halides (TOX) andadsorbable organic halides (AOX) listed in Table B-3 wereadded to the analyte list as requested by German FederalMinistry of Research and Technology, under a U.S.-Germanbilateral technology transfer program. The followingparameters were considered critical for evaluating the perox-pureTM technology: (1) VOC, hydrogen peroxide, and acidconcentrations; and (2) flow rate and pH. VOCs weremeasured by both gas chromatography (GC) and GC/massspectrometry (MS) methods. Only GC measurement ofVOCs was considered critical, because GC data wereplanned for quantitative use; GC/MS data were planned forqualitative use.

Because the perox-pureTM technology was developed totreat organics, and because VOCs were the principalcontaminants in groundwater, four replicate samples werecollected for GC analysis of VOCs. For other analytes, thenumber of samples was based on (1) the intended use of thedata, (2) analytical costs, (3) sampling time, and (4) thediscretion of analytical laboratory. EPA-approved sampling,analytical, quality assurance, and quality control (QA/QC)procedures were followed to obtain reliable data. Details onQA/QC procedures are presented in the demonstration plan(PRC, 1992).

44

Table B- 1 Experimental Matrix for perox-pure” Technology Demonstration

Run Number Influent pHHydrogen Peroxide at lnfluent Hydrogen Peroxide at lntluent Flow Rate

to Reactor 1 (mg/L) to Reactors 2 to 6 (mg/L) (gpm)

8.0

2 6.5

3 5.0

4 5.0

5 5.0

6 5.0

7 5.0

8 5.0

9 5.0

10 5.0

11 5.0

72 5.0

13 5.0

14 5.0

Phase 1 (Raw Groundwater Runs)

40 25

40 25

40 25

70 50

30 15

240 Hydrogen Peroxidewas added at Influent

240 to Reactor 1 on/y

60

Phase 2 (Spiked Groundwater and Reproducibility Runa

70 50

40 25

40 25

40 25

Phase 3 (Quartz Tube Cleaner Runs)

40 25

40 25

10

10

10

10

10

10

40

40

10

10

10

10

10

10

B.5 Review of Treatment Results

This section summarizes the results of both critical andnoncritical parameters for the perox-pure” chemicaloxidation system demonstration, and it evaluates thetechnology’s effectiveness in treating groundwatercontaminated with VOCs. Data are presented in graphic ortabular form. For samples with analyte concentrations atnondetectable levels, one-half the detection limit was used asthe estimated concentration. However, if more than onereplicable sample had concentrations at nondetectable levels,using one-half the detection limit as the estimatedconcentration for all replicable samples with nondetectablelevels of contaminants will significantly reduce the standarddeviation of the mean and will affect the statistical inferencesmade. For this reason, 0.5, 0.4, 0.6, and 0.4 times thedetection limit were used as estimated concentrations for thefirst, second, third, and fourth replicate samples, respectively.Throughout this appendix, the terms “Reactor 6 effluent,”“perox-pureTM effluent,” and “effluent” are used synonymously.

B.5.1 Summary of the Results for CriticalPammeters

Results for the critical parameters are presented belowfor each phase of the demonstration.

Phase 1 Results

In Phase 1 (Runs 1 through 8), only three VOCs weredetected in the influent to and effluent from the perox-pureTM

system. In general, TCE and PCE were present abovedetection limits only in the influent. TCA could not bemeasured in the influent, because it was present atconcentrations two orders of magnitude lower than theaverage TCE concentration and was diluted out during theanalysis. However, because effluent samples did not requiredilution, the TCA concentration could be measured intreated groundwater. In general, TCA was present in theeffluent from the perox-pureTM system. Phase 1 VOCconcentration data are presented for TCE and PCE inFigures B-3 through B-5. TCA concentrations are notshown in figures because TCA levels in the influent couldnot be measured.

45

Table 52 Target L L for VOCs in Effluent Samples

VOCTargetLevel

&l/L)

Chloroform 100

l,l-Dichloromethane (DCA) 5

1, l-Dichloromethene (1, I-DCE) 6

1,2-Dichlomethene (l,2-DCE) 6

1, 1, l-Trichlomethane (KA) 200

Trichloromethene (TCE) 5

Tetrachloromethene IKE) 5

Figure B-3 presents TCE and PCE concentrations in theinfluent and Reactors 1,3, and 6 effluent for Runs 1,2, and3. Concentrations are expressed as a function of influentpH. In all three runs, the effluent TCE and PCEconcentrations were well below the target level of 5 pg/Land below the detection limit of 1 pg/L. Figure B-3 showsthat the perox-pureTM system performed best in Run 1, whenthe influent pH was 8 (the unadjusted pH of groundwater).In this run, the Reactor 1 effluent had lower levels of TCEand PCE than in Runs 2 and 3, and it had the same levelsof TCE and PCE as the Reactor 6 effluent in Runs 2 and 3.However, Reactor 6 effluent TCA concentration was lowestin Run 3 at 1.4 Fg/L (Reactor 6 effluent TCAconcentrations in Runs 1 and 2 were 6.7 and 3.1 pg/L,respectively). Because TCA is difficult to oxidize, PSIselected Run 3 as the preferred operating condition, with aninfluent pH of 5.0.

Figure B-4 presents a comparison of VOCconcentrations in Runs 3,4, and 5 as a function of hydrogenperoxide levels. Although the effluent TCE and PCEconcentrations were the same in all runs, the data show thatReactor 1 effluent TCE and PCE concentrations were thelowest in Run 4 (with the highest hydrogen peroxide level)and in Run 5 (with the lowest hydrogen peroxide level).Higher levels of TCE and PCE in the Reactor 1 effluent atintermediate hydrogen peroxide levels cannot be explained.The Reactor 6 effluent TCA concentrations in Runs 3, 4,and 5 were 1.4, 1.8, and 2.1 pg/L, respectively. No definitetrend can be identified based on TCE, PCE, and TCA datain Runs 3,4, and 5.

Figure B-5 presents TCE and PCE concentrations atdifferent flow rates and hydrogen peroxide levels. Runs 4and 6 were performed at a flow rate of 10 gpm. Runs 7 and8 were performed at a flow rate of 40 gpm. In Runs 4 and6, the same total amount of hydrogen peroxide was added to

the contaminated groundwater. However, in Run 4,hydrogen peroxide was added at multiple points in thesystem using the splitter, while in Run 6, all hydrogenperoxide was added at the influent to the system. Based ona comparison of TCE and PCE levels in Runs 4 and 6, theeffect of adding hydrogen peroxide at multiple points in theperox-pureTM system cannot be evaluated, because in bothruns, TCE and PCE levels in treated groundwater werebelow the detection limit of 1.0 ug/L. However, effluentTCA levels in Runs 4 and 6 were 1.8 and 3.0 pg/L,respectively. Based on this data, adding hydrogen peroxideat multiple points in the perox-pure” system appears toenhance the system’s performance.

A comparison of TCE and PCE levels in Runs 6 and 7shows that both TCE and PCE concentrations in Reactor 1effluent were higher in Run 7 than in Run 6. Similarly, theeffluent TCA level in Run 7 (3.9 pg/L) was higher than inRun 6 (3.0 pg/L). These observations are consistent withthe operating conditions, because contaminated groundwaterhad a much longer UV exposure time in Run 6 than in Run7. UV exposure times were 1.5 and 0.4 minutes in Runs 6and 7, respectively.

A comparison of TCE and PCE levels in Runs 7 and 8shows that both TCE and PCE concentrations in Reactor 1effluent were higher in Run 7 than in Run 8. Effluent TCAlevels were about the same in both runs (3.9 and 4.0 pg/Lin Runs 7 and 8, respectively). The higher Reactor Ieffluent TCE level in Run 7 may be attributed to higherinfluent TCE levels in that run. Reactor 1 effluent TCElevels correspond to 995 and 99.9 percent TCE removal inRuns 7 and 8, respectively. Similarly, the Reactor 1 effluentPCE levels correspond to 92.9 and 99.2 percent PCEremoval in Runs 7 and 8. These data seem to indicate thathigher doses of hydrogen peroxide may have scavengedhydroxyl radicals or excess hydrogen peroxide reduced UV

46

Figure B-2 perox-pure” Chemical Oxidation Treatment System Sampling Locations

transmittance through water, which resulted in lowerremoval efficiencies for Run 7 than those for Run 8.

Based on quick turnaround analyses performed duringRuns 1 through 6, PSI selected Run 3 operating conditionsas the preferred operating conditions for spikedgroundwater. As a result, Runs 10 through 14 wereperformed using Run 3 operating conditions.

Phase 2 Results

Phase 2 (Runs 9 through 12) results for VOC removalin the perox-pureTM system are presented in Figures B-6through B-8. A comparison of the perox-pureTM system’sperformance in treating spiked groundwater (Run 9) andunspiked groundwater (Run 4) is presented in Figure B-6.Figure B-6 shows that TCE and PCE levels in treatedgroundwater were higher in Run 9 (spiked groundwater)than in Run 4 (unspiked groundwater). These data suggestthat spiking compounds (chloroform, DCA, and TCA)affected the perox-pureTM system’s performance in removingTCE and PCE, perhaps because of the additional oxidantdemand. However, treated groundwater TCE and PCElevels plotted in Figure B-6 are estimated concentrations.Because the detection limit for TCE and PCE in Run 9 was

5 pg/L and in Run 4 was 1 pg/L, and because TCE andPCE were present at nondetectable levels in treatedgroundwater in both runs, the estimated concentrations inRun 9 are higher than in Run 4. Therefore, the data areinconclusive with regard to the effect of spiking compoundson the removal of TCE and PCE.

During the reproducibility runs (Runs 10, 11, and 12),the effluent TCE, PCE, and DCA levels were generallybelow detection limit (5 Fg/L) and effluent chloroform andTCA levels ranged from 15 to 30 Fg/L. VOC removalefficiencies in reproducibility runs are plotted in Figure B-7.Figure B-7 shows that for TCE and PCE, which are easy tooxidize, most of the removal occurred in Reactor 1, leavingonly trace quantities of TCE and PCE to be removed in therest of the perox-pure= system. However, for chloroform,DCA, and TCA, which are difficult to oxidize, considerableremoval occurred beyond Reactor 1. During the threereproducibility runs, average removal efficiencies for TCE,PCE, chloroform, DCA, and TCA after Reactor 1 were 99.5,95.9, 41.3, 67.0, and 17.4 percent, respectively. After Reactor6, overall removal efficiencies for TCE, PCE, chloroform,DCA, and TCA increased to 99.7, 97.1, 93.1, 98.3, and 81.8,respectively. The overall removal efficiencies of the perox-pureTM system were reproducible for all VOCs. However, for

47

Table B-3 Analytical and Measurement Methods

Parameter Method Type Method Source Name of Method

Acid

Alkalinity

Bioassay

Laboratory

ElectriCity

consumption

Flow rate

Field

Field

Hardness

Hydrogen peroxide

Laboratory

Field

Metals Laboratory(iron and manganese)

pH

Purgeable organiccarbon (POC)

SVOCS

Field

Laboratory

Laboratory

Specific conductance Field

Total carbon (TC) and LaboratoryTotal organic carbon

(TOC)

Temperature Field

TOX Laboratory

Turbidity Laboratory

VOCs (GC) Laboratory

VOCs (GC/MS) Laboratory

Acid-base titration

MCAWW 310.7’

DIN 36409 H14b

Acid-base titration

EPA/600/4-85/O 19

EPA/600/4-85/019

None

MCAWW 130.1

Boltz and Howell, 1979

SW-646 3010/6010d

MCAWW 150. la

SM 531Oti

sw-646 35 70/3640/a270d

SW-646 9050d

SM 531ocd

MCAWW 170.1’

SW-846 9020d

MCAWW 180.1’

SW-a46 5030/a070d

SW-a46 a240d

Tiition with sodium hydroxide

Alkalinity (titrimetric)

Adsorbable organic halide measurement

Titration with sulfuric acid

48-hour static acute toxicity test (detinitim)using Ceriodaphnia dubia

96-hour static acute toxicity test (definitive)using Pimeohales PtDmektS

Electricity consumption

Flow rate measurement using monitoringequipment on the treatment system

Hardness

Calorimetric method for hydrogen peroxidemeasurement

Metals by inductively coupled plasma-atomicemission spectroscopy

pH electrometric measurement

Carbon measurement by combustion-infrared method

GC/MS for SVOCs: capillary columntechnique

Specific conductance

Carbon measurement by persulfateultraviolet oxidation method

Temperature

Total organic halide measurement

Turbidity (nephelometriic)

Halogenated VOCs by GC: purge and trap

VOCs by GC/MS: capillary columntechnique

4

Notes:EPA, 1983.

b Steffins, 1992.EPA, 7985.

d EPA, 1986.d APHA, AWWA, and WPCF, 1989.

VOC Concentration, vg/L

TCE

PCE

/

Influent

Reactor I

Reactor3

m Reactor6

Figure B-3 Comparikon of VOC Concentrations at Different Influent pH Levels(Hydrogen Peroxide Level at Reactor 1 = 40 mg/L; Hydrogen Peroxide Level atReactors 2 through 6 = 25 mg/L; Flow Rate = 10 gpm)

4 9

VOC Concentration, pglL

I IO 100 1,000 I0,000

TCE

PCE(Hydrogen Peroxide Level

I ”Fsz$~~;~go”5 at Reactor I = 40 mglk” _“_l” ^_-.. _._._A at Reactors 2 through 6 = 25 mg/L)

T C E

PCE

at Reactor I = 70 mg/L

at Reactors 2 through 6 = 50 mg/L)

TCE

PCE Run 5(Hydrogen Peroxide Level

at Reactor I = 30 mg/L;at Reactors 2 through 6 = 15 mg/L)

/ 4

Influent

Reactor I

Reactor3

w Reactor6

figure B-4 Comparison of VOC Concentrations at Different Hydrogen Peroxide Levels(Influent pH = 5.0; Flow Rate = 10 gpm)

50

VOC Concentration, pg/L

I IO 100 11 ,000 10,000

TCE

PCE

TCE

PCE

TCE

PCE R u n 8(Hydrogen Peroxide Level

at Reactor I = 60 mglk

Flow Rate = 40 gpm)

/

Influent

Reactor I

@i@ Reactor 3k

Im Reactor 6

figure B-5 Comparison of VOC Concentrations at Different Flow Rates and Hydrogen Peroxide Levels(InfluentpH = 5.0)

TCE

PCE

Chloroform

DCA

TCA

VOC Concentration, VglL

I IO 100 1.000

Phase 2 - Run 9Spiked

TCE

PCE

/

Influent

Reactor I

Reactor3

E9 Reactor 6

Phase I - Run 4Unspiked

Figure B-6 Comparison of VOC Concentrations in Spiked and Groundwater(Influent pH = 5.0; Hydrogen Peroxide Level at Reactor 1 = 70 mg/L; HydrogenPeroxide Level at Reactors 2 through 6 = 50 mg/L; Flow Rate = 1O gpm)

52

100

0TCE PCE Chloroform

Run IO

DCA TCA

100

TCE PCE Chloroform DCA TCA

Run II

TCE PCE Chloroform

Run I2DCA TCA

Figure B-7 VOC Removal EfYicikw.~ in Reproducibility Runs(Influent pH = 5.0; Hydrogen Peroxide Level at Reactor 1 = 40 mg/L; HydmgenPeroxide Level at Reactors 2 through 6 = 25 mg/L; Flow Rate = 10 gpm)

53

TL=200- - - -

1 0 0 ;TL= 100

t

TCE PCE Chloroform DCA TCA

I Run 11

Run 72

gure B-8 Comparison of 95 Percent UCLs for Effluent VOC Concentrations with Target Levels in Reproducibility Runs(Influent pH = 5.0; Hydrogen Peroxide Level at Reactor 1 = 40 mg/L; HydrogenPeroxide Level at Reactors 2 through 6 = 25 mg/L; Flow Rate = 10 gpm)

ertain compounds, the removal efficiencies after Reactor 1 demonstration runs were used. In Run 13, scaled quartzwere quite variable (for example, chloroform removal tubes were used. These tubes had been exposed to anfliciencies ranged from 27.4 to 56.3 percent). This environment that encouraged scaling, but they had not beenariability may be associated with sampling and analytical maintained with cleaners or wipers. In Run 14, quartz tubesprecision. that had been maintained by cleaners or wipers were used.

Figure B-8 compares the 95 percent upper confidenceimits (UCL) of effluent VOC concentrations with targetevels in reproducibility runs. For this project, the targetllevel for a given VOC was set at the most stringent limit incases where the VOC has multiple regulatory limits. For allVOCs but chloroform, the most stringent limit is theZaliiornia drinking water action level. For chloroform, thenost stringent limit is the maximum contaminant level:MCL) specified in the Safe Drinking Water Act. Figure B-3 shows that perox-pureTM system effluent met the targetevels at the 95 percent confidence level in all threereproducibility runs, indicating that the system performance.vas reproducible.

Phase 3 Results

Figure B-9 presents VOC concentrations in Runs 12, l3,and 14, which were conducted to evaluate quartz tube

A comparison of removal efficiencies for TCE inReactors 1 and 2 shows that TCE removal efficiencies wereabout the same in all runs. PCE removal efficiencies wereabout 3 percent less in Run 13 than that in Runs 12 or 14.Removal efficiencies for chloroform, DCA, and TCA wereconsistently less in Run 13 than in Run 14, indicating thatperiodic cleaning of quartz tubes by wipers is required tomaintain the perox-pureTM system’s performance. Withoutsuch cleaning, the removal efficiencies will likely decrease inan aqueous environment that would cause scaling of quartztubes. For example, after Reactor 2, chloroform removalefficiency in Run 13 was 53.4 percent, compared to 61.3percent removal efficiency in Run 14. Because the quartztubes used in Run 12 had little coating, it was expected thatthe removal efficiencies in Run 12 would be higher thanthose in Run l3. However, the demonstration did notconfirm this for all VOCs. For example, Run 12 TCAremoval efficiencies were less than Run 13 TCA removal

.-p * . .I *- ‘.____!_*___.. -----* I.., ,.,l,:...-.A

TCE PCE Chloroform

Run I2DCA TCA

TCE PCE Chloroform

Run 13DCA TCA

TCE PCE Chloroform

Run 14DCA TCA

I

Reactor 1

Reactor 2

Figure B-9 VOC Removal Efficiencies in Quartz Tube Cleaner Runs(Influent pH = 5.0; Hydrogen Peroxide Lecrt41 at Reactor 1 = 40 mg/L; HydrogenPemxide Level at Reactors 2 through 6 = 25 mg/L; Flow Rate = 10 gpm)

B.5.2 Summary of Results for NoncriticalParameters

The technology demonstration also evaluated analyticalresults of several noncritical parameters. These results aresummarized below.

GC/MS analysis of influent and effluent samples forVOCs indicated that new target compounds or tentativelyidentified compounds (TIC) were not formed during thetreatment.

GC/MS analysis of influent and effluent samples forSVOCs showed that target SVOCs were not present atdetectable levels. However, several unknown TICs werepresent in both the influent and effluent samples.

Average influent TOX and AOX levels were 8 0 0 pg/Land 730 pg/L, respectively. The perox-pure= systemachieved TOX removal efficiencies that ranged from 93 to 99percent and AOX removal efficiencies that ranged from 95to 99 percent.

The TC, TOC, and POC concentrations in influent andeffluent samples in Runs 10, 11, and 12 are presented inFigure B-10. Average TC concentrations in the influent andeffluent were 75 mg/L and 55 mg/L, respectively. Thedecrease in TC concentration in the perox-pureTM system maybe due to the loss of dissolved carbon dioxide that occurredas a result of the turbulent movement of contaminatedgroundwater in the perox-pureTM system.

Figure B-10 shows a decrease in TOC of about 40percent during treatment. The decrease corresponds to theamount of organic carbon that was converted to inorganiccarbon during treatment, suggesting that about 40 percent ofthe organic carbon was completely oxidized to carbondioxide. However, the TOC data do not indicate whetherthe organic carbon that was completely oxidized hadoriginated from the VOCs present in groundwater or fromsome other compounds present in groundwater.

Effluent POC concentration was about 0.02 mg/L whichis below the reporting limit of 0.035 mg/L. POCconcentration data show that the average POC removalefficiency was about 93 percent. Assuming that the majorityof organic carbon associated with VOCs could be measuredas POC, this indicates that about 93 percent of volatileorganic carbon was converted to either carbon dioxide ornonpurgeable organic carbon.

During Runs 10, 11, and 12, bioassay tests wereperformed to evaluate the acute toxicity of influent to andeffluent from the perox-pureTM systems. Two freshwater testorganisms, a water flea (Ceriodaphnia ‘dubia) and afathead minnow (Pimephales promelas), were used in thebioassay tests. Toxicity was measured as the lethal

concentration at which 50 percent of the organisms died(LC,), and expressed as the percent of effluent (or influent)in the test water. One influent and one effluent sample weretested in each run. One control sample was also tested toevaluate the toxicity associated with hydrogen peroxideresidual present in the effluent. The control sample hadabout 10.5 mg/L of hydrogen peroxide (average effluentresidual in Runs 10, 11, and 12), and had characteristics(alkalinity, hardness, and pH) similar to that of effluent inRuns 10, 11, and 12.

In general, the influent was not found to be acutely toxicto either test organism. The effluent was found to be acutelytoxic to both test organisms. The influent LC, values forboth organisms indicated that in the undiluted influentsample more than 50 percent of the organisms survived.However, LC, values for the water flea were estimated to be35, 13, and 26 percent effluent in Runs 10, 11, and 12,respectively; and LC, values for the fathead minnow wereestimated to be 65 and 71 percent effluent in Runs 10 and11, respectively. In Run 12, more than 50 percent of thefathead minnows survived in the undiluted effluent. TheLC% value for the water flea was estimated to be17.7 percent in the control sample, indicating that the samplecontained hydrogen peroxide at a concentration that wasacutely toxic to water fleas. However, more than 50 percentof the fathead minnows survived in the undiluted controlsample indicating hydrogen peroxide was not acutely toxic tofathead minnows at a concentration of 10.5 mg/L. Thisobservation, however, is not entirely consistent withobservations made by the Department of EnvironmentalProtection, State of Connecticut (CDEP). The CDEP WaterToxics Section of Water Management Division reports LC,value of 18.2 mg/L of hydrogen peroxide with 95 percentconfidence limits of 10 mg/L and 25 mg/L (CDEP, 1993).

Comparison of the LC, value of the control sample withLC, values of effluent samples for water fleas indicates thetoxicity associated with the effluent samples is probably dueto hydrogen peroxide residual in the effluent. However, noconclusion can be drawn on the effluent toxicity to fatheadminnows because the control sample toxicity results from theSITE demonstration data are not entirely consistent with thedata collected by CDEP.

Iron and manganese were present at trace levels in theinfluent. In general, iron was present at levels less than 45Fg/L, and manganese was present at an average level of 15pg/L. Removal of iron or manganese did not occur in theperox-pureTM system, because these metals were present onlyat trace levels in the influent.

No changes in pH, alkalinity, hardness, or specificconductance were observed during treatment.

Average influent temperature was about 72 “F. Averageeffluent temperatures were about 90 “F and 76 “F, atinfluent flow rates of 10 gpm and 40 gpm, respectively.

56

Carbon Concentration, mg/L

0. I 1O 100

TC

T O C

POC . . . .._____ZZ.Z~“~~

iE&E~j3j-j2 j,.l..l_-...“I .I .._ ..Ic;;~~~:: f_--~------““_-“-~-“_--~---_““_.-.--.--.._--”_..--~~._

Run IO1..

0.0 I 0. I I 1O 100

Run I I

0.0 I 0. I IO 100

T C

T O C

POCL .I ..zT - _/

1- . “.“““_ _._. ~~-~_~_~~-~__~~ _^_ ._.__-._” ._._ ~...” _._. “1” __.._

Run I2

Figure & IO Carbon Concentrations in Reproducibily Runs(Influent pH = 5.0; Hydrogen Peroxide Level at Reactor 7 = 40 mg/L; HydrogenPeroxide Level at Reactors 2 through 6 = 25 mg/L; Flow Rate = 10 gpm)

Influent

m Reactor 6

57

Because 10 gpm corresponds to a hydraulic retention time of1.5 minutes and 40 gpm corresponds to a retention time of0.4 minutes, the average temperature increase due to 1minute of UV radiation exposure in the perox-pure’” systemis about 12 “F.

B.6 Conclusions

For the spiked groundwater, PSI determined thefollowing preferred operating conditions: (1) influenthydrogen peroxide level of 40 mg/L, (2) hydrogen peroxidelevel of 25 mg/L in the influent to Reactors 2 through 6, (3)influent pH of 5.0, and (4) flow rate of 10 gpm. At theseconditions, the effluent TCE, PCE, and DCA levels weregenerally below detection limit (5 pg/L) and TCA levelsranged from 15 to 30 pg/L. The average removalefficiencies for TCE, PCE, chloroform, DCA, and TCA wereabout 99.7, 97.1, 93.1, 98.3, and 81.8 percent, respectively.

For the unspiked groundwater, the effluent TCE andPCE levels were generally below the detection limit(1 pg/L), with corresponding removal efficiencies of about99.9 and 99.7 percent. The effluent TCA levels ranged from1.4 to 6.7 pg/L with removal efficiencies ranging from 35 to84 percent.

The perox-pure” system effluent met California drinkingwater action levels and federal drinking water MCLs forTCE, PCE, chloroform, DCA, and TCA at the 95 percentconfidence level.

The quartz tube wipers were effective in keeping thetubes clean and appeared to reduce the adverse effect scalinghas on contaminant removal efficiencies.

TOX removal efficiencies ranged from 93 to 99 percent.AOX removal efficiencies ranged from 95 to 99 percent.

For spiked groundwater, during reproducibility runs, thesystem achieved average removal efficiencies of 38 percentand greater than 93 percent for TOC and POC, respectively.

The temperature of groundwater increased at a rate of12 “F per minute of UV exposure in the perox pureTM

system. Since the oxidation unit is exposed to thesurrounding environment, the temperature increase may varydepending upon the ambient temperature or otheratmospheric conditions,

B.7 References

APHA, AWWA, and WPCF, 1989, Standard Methods forthe Examination of Water and Wastewater, 17th Edition.

Boltz, D-F., and JA. Howell, 1979. ColorimetricDetermination of Non-Metals. John Wiley & Sons, NewYork, New York.

State of Connecticut, Department of EnvironmentalProtection (CDEP), 1993, Personal communication fromA. Iacobucci to the U.S. Environmental ProtectionAgency (EPA) on the acute toxicity of hydrogenperoxide to freshwater organisms.

Lawrence Livermore National Laboratory (LLNL), 1990,Remedial Investigation of the General Services Area,Lawrence Livermore National Laboratory, Site 300(May).

PRC Environmental Management, Inc. (PRC), 1992,Demonstration Plan for the PSI perox-pureTM

Ultraviolet/Oxidation System. Final Report, submittedto EPA ORD, Cincinnati, Ohio.

Steffens, K, 1992, Personal communication with EPA on theanalytical procedure for AOX measurement.

EPA, 1983, Methods for Chemical Analysis of Water andWastes. EPA-600/4-79-020, Environmental Monitoringand Support Laboratory, Cincinnati, Ohio, andsubsequent EPA&O/4 Technical Additions.

EPA, 1985, Methods for Measuring Acute Toxicity ofEffluent to Fresh Water and Marine Organisms.EPA/600/4-85/013, Third Edition.

EPA, 1986, Test Methods for Evaluating Solid Waste.Volumes IA-IC: Laboratory Manual, Physical/ChemicalMethods, and Volume II: Field Manual,Physical/Chemical Methods, SW-846, Third Edition,Office of Solid Waste and Emergency Response,Washington, D.C.

58

Appendix CCase Studies

This appendix summarizes three case studies on the useof the perox-pureTM chemical oxidation system developed byPeroxidation Systems, Inc. (PSI). The perox-pureTM systemhas proven to be a technically and economically viablealternative to conventional technologies. It should beevaluated in cases requiring treatment of water contaminatedwith organic contaminants. All three case studies representfull-scale, currently operating commercial installations. Thecontaminants of concern in these case studies includeacetone, isopropyl alcohol (IPA), trichloroethene (TCE), andpentachlorophenol (PCP). The scope of the case studies islimited to basic information concerning the following topics:site conditions, system performance, and costs. Thefollowing case studies are discussed in this appendix:

l Wastewater Treatment System, Kennedy SpaceCenter, Florida

l Municipal Drinking Water System, Arizona

l Chemical Manufacturing Company, Washington

C.l Wastewater Treatment System, Florida

This case study describes the performance and treatmentcosts of the perox-pureTM system treating industrialwastewater containing acetone and IPA at the KennedySpace Center site in Florida.

C.l.1 Site Conditions

Effluent from the existing wastewater treatment systemat the Kennedy Space Center in Florida frequently exceededthe permitted levels for discharges of acetone and IPA.Discharges were reported at levels up to 20 milligrams perliter (mg/L) for both compounds. The treatment facilitydischarge requirement for both acetone and IPA is 0.5mg/L. A liquid-phase carbon adsorption system hadoriginally been installed as part of the treatment system, butwas found to be inadequate. Subsequently, perox-pureTM

chemical oxidation technology was selected as the newmethod of treatment.

After selecting the perox-pureTM chemical oxidationtechnology, the facility contractor installed a perox-pure”’Model SSB-30R in the fall of 1992. The system was used totreat 5,000- to 6,000-gallon batches of contaminated water.A demineraliition system operated by another contractorreceived effluent from the perox-pureTM unit. Facilityrequirements specified a maximum treatment time of 24hours for each batch.

C.1.2 System Performance

The perox-pureTM unit is currently treating maximumlevels of acetone and IPA (both 20 mg/L) to less than 0.5mg/L for each compound. The treatment objectives wereeasily met, even when the wastewater contained twice themaximum expected acetone and IPA concentrations.Treatment objectives were achieved for each batch in lessthan the specified 24-hour maximum treatment time. Inaddition, effluent from the perox-pureTM unit met thedemineralization discharge standards, making treatment bythe demineralization system unnecessary.

Due to the efficiency of treatment, the contractor wasable to treat wastewater in a flow-through mode rather thanin batches. The resulting flow rate of 5 gallons per minute(gpm) for a period of 20 hours per day requires 10 kilowatts(kW) to power the ultraviolet (UV) radiation lamps. Theadditional capacity of the perox-pureTM unit will be used incontaminant spill situations, which would produce muchhigher acetone and IPA concentrations. The influenthydrogen peroxide dosage for flow-through operation was100 mg/L.

c.1.3 costs

The operation and maintenance (O&M) costs for flow-through operation of the perox-pure= unit include electricity,chemicals (hydrogen peroxide), and general maintenance.The following items are used at the indicated rate for each1,000 gallons treated: electricity at $0.06 per kilowatt-hour(kWh) costs $2.00; 50 percent hydrogen peroxide at $0.35per pound costs $0.60; and estimated maintenance

59

t

requirements cost $1.00. The total O&M cost per 1,OOOgallons treated is $3.60.

C.2 Municipal Drinking Water System, Arizona

This case study describes the performance and treatmentcosts of the perox-pureTM system treating groundwatercontaining TCE. The groundwater was a source ofmunicipal drinking water in Arizona.

C.2.1 Site Conditions

In 1989, a municipal groundwater well in Arizona usedfor drinking water was found to contain 50 to 400micrograms per liter &g/L) of TCE. The well, capable ofproducing 2,000 gpm, was taken out of service whiletreatment options were considered. Because the well islocated on a city lot in the middle of a large residential area,the city preferred using a low-visibility, quiet treatmentmethod that could consistently destroy TCE toconcentrations below the drinking water standard of 5 pg/L.Moreover, treatment to below the drinking water standardwas desirable because of the high profile of the siteremediation. Given these requirements, the perox-pure”’chemical oxidation technology was selected.

An on-site performance evaluation was initiated inDecember 1989 using a perox-pureTM pilot system to treatcontaminated groundwater pumped from the well at a flowrate of 135 gpm. Testing showed that TCE could bedestroyed to below the analytical detection limit of 0.5 pg/L.

C.2.2 System Performance

A perox-pureTM Model SSB-30R is currently in operationat the site, treating organic contamination to belowdetectable levels at a flow rate of 135 gpm. Treatment isaccomplished with only 15 kW of power, one-half thecapacity of the unit. The average influent concentration ofTCE is 150 pg/L, and the eflluent TCE concentration is lessthan 0.5 pg/L.

C3 Chemical Manufacturing Company, Washington

This case study describes the performance and treatmentcosts of the perox-pureTM system treating groundwatercontaining PCP at a chemical manufacturing facility inWashington.

C.3.1 Site Conditions

PCP contamination was discovered in local groundwatersurrounding a chemical manufacturing company inWashington. The company had produced PCP for morethan 30 years. Because of the site geology, the groundwaterwas brackish and contained high concentrations of iron andcalcium carbonate. The chemical company initiated aremediation effort that included a pump-and-treat process.After bench-scale testing, the perox-pureTM chemicaloxidation system was selected to destroy PCP to below atarget level of 0.1 mg/L.

A full-scale perox-pureTM system was installed in 1988under a Full Service Agreement with PSI. When theremediation effort began, groundwater was contaminatedwith PCP at levels of up to 15 mg/L, three times higher thanexpected. Iron was detected at levels of up to 200 mg/L, 20times higher than expected. PSI made pretreatmentrecommendations and assisted with the selection of an ironoxidation and removal system, which included clarificationand multimedia filtration. Treatment with the perox-pure”system reduced iron concentrations in the groundwater toacceptable levels. The groundwater was stabilized and thescaling tendency was reduced by adding acid to lower thegroundwater pH. to approximately 5.

The original perox-pure”‘ unit did not have enoughcapacity to treat the unexpectedly high PCP levels.However, as part of the Full Service Agreement, a perox-pureTM Model CEBX-360R was installed to replace theoriginal unit. The new model incorporated the latestimprovements in the ,perox-pure”‘ unit, including anautomatic tube cleaning device.

C.2.3 CostsC.3.2 System Performance

O&M costs for continuous operation of the perox-pure=system at a flow rate of 135 gpm include electricity,chemicals (hydrogen peroxide), and general maintenance.The following items are consumed at the indicated rate foreach 1,OOO gallons treated: electricity at $0.06 per kWh costs$0.11; 50 percent hydrogen peroxide at $0.35 per pound costs$0.12; and estimated maintenance requirements cost $0.05.The total O&M cost per 1,OOO gallons treated is $0.28.

The groundwater is being successfully treated at a flowrate of about 70 gpm, and a power requirement of 180 kW.The pH of the influent groundwater is adjusted to 5 byadding acid, and hydrogen peroxide is added to the influentto achieve a concentration of 150 mg/L. The perox-pureTM

system is currently treating maximum influent PCPconcentrations of 15 mg/L to average effluent concentrationof 0.1 mg/L.

60

C.3.3 Costs

O&M costs for continuous operation of the perox-pure-system at a flow rate of 70 gpm include electricity, chemicals(hydrogen pero$de ad acid), and general maintenance.The following items are used at the indicated rate for each1,000 gallons treated: electricity at $0.06 per kWh costs$2.57; 50 percent hydrogen peroxide at $O.35 per pound costs$0.87; acid at $0.085 per pound costs $0.03; and estimatedmaintenance requirements cost $0.43. The total O&M costper 1,000 gallons treated is $3.90.

61~T~J.S.GOVERNMEN~PR~NT~NGO~FICE: 1993-753-292