Process Engineering for Pollution Control and Waste Minimization (Environmental Science and Pollution Control Series)

  • 49 710 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Process Engineering for Pollution Control and Waste Minimization (Environmental Science and Pollution Control Series)

EDITED BY DONALD L. WISE Northeastern University Boston, Massachusetts DEBRA J. TRANTOLO Cambridge Scientific Inc. B

2,587 39 98MB

Pages 743 Page size 336 x 500.16 pts Year 2008

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

EDITED BY

DONALD L. WISE

Northeastern University Boston, Massachusetts

DEBRA J. TRANTOLO

Cambridge Scientific Inc. Belmont, Massachusetts

Marcel Dekker, Inc.

New YorkaBaselaHongKong

Library of Congress Cataloging-in-Publication Data

Process engineering for pollution control and waste minimization / edited by Donald L. Wise, Debra J. Trantolo. p. cm. -- (Environmental science and pollution control: 7) Includes bibliographical references and index. ISBN 0-8247-9161-4 (alk. paper) 1. pollution.2.Wasteminimization. I. Wise,DonaldL.(DonaldLee). U. Trantolo, Debra J. III. Series. TD191.5.W6 1994 628.5-dc20 93-4601 CIP

The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional Marketing at the addressbelow. This book is printed on acid-free paper. Copyright 0 1994 by Marcel Dekker, Inc. All Rights Reserved. be reproduced or transmitted in anyform or by any means, electronic Neither this book nor any part may or mechanical, including photocopying, microfilming, and recording, or by an information storage and retrieval system, without permission in writing from the publisher.

Marcel Dekker. Inc. 270 Madison Avenue, New York, New

York 10016

Current printing (last digit): l0987654321 PRINTED IN THE UNITED STATES OF AMERICA

Preface

A clean environment is a goal to which we all strive. However, we have been the victims of activities. The severe environmental damage as a result of industrial growth and defense-related damage to our environment is substantially affecting our overall health and welfare. It is a credit to our human spirit that we remain optimistic and share an enthusiasm about environmental issues. The numbers of registered wastesites are alarming, and continue to grow daily. No longer can we casually consider wastean acceptable by-product of our everydayactivities. While the consumer hasbegun to embrace the concept of waste reductionas, for example,in the practice of recycling, the large-scaleindustrialconcern has also turned to wastecontrolmethods. Whether driven by governmentmandate,socialresponsibility,economics, or other forces, waste control and waste minimization practices are increasingly welcomed. Process Engineeringfor Pollution Control and Waste Minimization provides an up-to-date source of technical information relating to current and potential pollution control and waste minimization practices. Overfifty recognized experts provide an in-depth treatmentof this rapidly growing field that draws its resources frommany disciplines. We have deliberately solicited input from governmental,industrial, and academic specialiststo ensure a multidimensional presentation of the pollution control and waste minimization schemes that are shaping our environmental outlook. The text is divided into five parts. It begins with the presentation of general engineering considerations and the regulatory, ethical, and technical framework within which these processes are managed, then enters into specific wastelwastewater pollution control technologies that are used throughout industry. Models for potential control and minimization techniques are offered, and industry-specific case studies complete the text. Throughout, we have attempted to provide a sense that the scopeof waste control and minimizationmay be immense, but it is not overwhelming. We trust that this book will provide a contribution to this important field and emphasize the need for continued progress. One way to better our environment is to eliminate or reduce iii

iv

Preface

pollution at the source. Potentially great benefits await us if we can develop economical, effective, and efficient solutions to our waste generation problems. All readers of this text will contribute something to the environment of tomorrow. Donald L. Wise Debra J. Trantolo

Contents

Preface Contributors

Part I: Engineering Issues In Pollution Control and Waste Minimization Process Engineering for Pollution Control and Waste Minimization John Hanna and Osawaru A. Orumwense

iii ix

3

Selection of Least Hazardous Material Alternatives Alvin F. Meyer

17

Multiple Approaches to Environmental Decisions Douglas M . Brown

25

Introduction to Engineering Evaluation for Contaminated Sites David S. Wilson, Alan C . Funk, Ronald G . Fender, and Marilyn Hewitt

47

Innovative Approaches to Cleanup Level Development Ronald J. Kotun, Richard F. Hoff, Robert J. Jupin, Diane McCauslund, and Patrick B. Moroney

87

Designing to Prevent Pollution James Lounsbury

145

Biochemical, Genetic, and Ecological Approaches to Solving Problems During 171 in Bioremediation situ and Off-site 0. A. Ogunseitan V

vi 8

9

allenges Competitive

10

Contents

Commandments of Waste Management Donald K. Walter A Proactive Approach to Environmental Management: Meeting and Environmental William E. Schramm and Stella S. Schramm HealthHazardsAssociatedwithPollutionControlandWasteMinimization Patrick D . Owens

193

213

227

Part 11: Methodologies of Waste Control 11 Techniques for Controlling Solid and Liquid Wastes Hsai-Yang Fang and Jejhrey C. Evans

247

l2

Solidification and Stabilization Techniquesfor Waste Control A. Samer Ezeldin and George P. Korj?atis

271

Soil Remediation with Environmentally Processed Asphalt @PATM) M. Testa and D. L. Patton

297

14

Lead Decontamination of Superfund Sites Ann M . Wethington, Agnes Y. Lee, and Vernon R. Miller

311

15

A Secure Geologic Repository for Hazardous Waste Residuals Thomas R . Klos

331

Photocatalytic Degradation of Hazardous Wastes

363

l3

S.

16

M. S. Chandrasekharaiah, S. S. Shukla, J. L. Margrave, and S. C. Niranjan 17

Photocatalytic Oxidation of Organic Contaminants Allen P. Davis

377

18

Biodegradation of Organic Pollutants in Soil Paul D . Kuhlmeier

405

19

Siallon: The Microencapsulationof Hydrocarbons Withina Silica Cell Tom McDowell

425

20

Remediation of Heavy Metal Contaminated Solids Using Polysilicates George J. Trezek

441

21

Fluidized Bed Combustion for Waste Minimization: Emissions andAsh Related Issues E. J. Anthony and F. Preto

Part 111: Wastewater Treatment 22 An Overview of Physical,Biological, andChemicalProcesses for Wastewater Kanti L. Shah

467

489

vii

Contents 23

FreezeConcentration: Its Application in HazardousWastewaterTreatment Ray Ruemekorf

24

OrganoclaySorbents for Selective Removalof OrganicsfromWater and Wastewater Steven K. Dentel, Ahmad I. Jamrah, and Michael G. Stapleton

513

525

25

Removal of Chromate,Cyanide,and Heavy MetalsfromWastewater Klaus Schwitzgebel and David M . Manis

535

26

NeutralizationTacticsforAcidicIndustrialWastewater Christopher A. Hazen and James I. Myers

557

Part Iv:Modeling for Pollution Control 27 IntroducingUncertainty ofAquifer Parametersinto an OptimizationModel Robert L. Ward 28

29

30

Application of Total QualityManagement(TQM)Principles Prevention Programs Prasad S. Kodukula

569

to Pollution 591

PC Software for OptimizingGroundwaterContaminantPlumeCapture and Containment Richard C . Peralta, Herminio H. Suguino, and Alaa H. Aly

597

HorizontalWellsforSubsurfacePollutionControl George Losonsky and Milovan S. Berjin

619

Part V Industry-Specific Pollution Control Pollution Control and Waste Minimization in Military Facilities 31 Merrit R Drucker

637

32

Waste Reduction Strategies for Small Businesses Dan A. Philips

643

33

Contaminated Soils in Highway Construction Namunu J. Meegoda

663

34

Management of Waste Compressed Gases Dan Nickens

685

35

Pollution Control in the Dairy Industry T. Viraraghavan

705

36

Landfill Gas Collection and Destruction Systems: Evaluating Toxic Emissions and Potential Health Risk Karnig Ohannessian, Anna Peteranecz,and Thomas Kear

Index

715 727

This Page Intentionally Left Blank

Contributors

AlaaH.Aly Utah State University,Logan,Utah E.J. Anthony CANMET, Ottawa, Canada Milovan S. Beljin University of Cincinnati,Cincinnati,Ohio Douglas M. Brown TheLogistics Management Institute,Bethesda, Maryland M. S. Chandrasekharaiah Houston Advanced Research Center,TheWoodlands, T~xas Allen F? Davis University of Maryland,CollegePark, Maryland Steven K. Dentel University of Delaware,Newark, Delaware Merrit F? Drucker Army Management Staff College, Fort Belvoir,Virginia JeffreyC.Evans Bucknell University,Lewisburg, Pennsylvania A. Samer Ezeldin Stevens Institute of Technology,Hoboken, New Jersey Hsai-YangFang LehighUniversity,Bethlehem, Pennsylvania Ronald G. Fender Environmental Resources Management Group,Exton, Pennsylvania Alan C. Funk Environmental Resources Management Group,Exton, Pennsylvania John Hanna TheUniversity of Alabama,Tuscaloosa, Alabama Christopher A. Hazen MilesInc., New Martinsville, West Virginia Marilyn Hewitt Environmental Resources Management Group,Exton, Pennsylvania Richard F. Hoff ChesterEnvironmental,Monroeville, Pennsylvania Ahmad 1. Jamrah University of Delaware, Newark, Delaware Robert J. Jupin ChesterEnvironmental,Monroeville, Pennsylvania ThomasKear OP&L,Inc., San Diego, California Thomas R. Klos Envirovest Management,Houston, Texas Prasad S. Kodukula Woodward-Clyde Consultants, Overland Park, Kansas George F? Korfiatis Stevens Institute of Technology,Hoboken, New Jersey RonaldJ.Kotun ChesterEnvironmental,Monroeville, Pennsylvania

X

Contributors

Paul D. Kuhlmeier Consulting Environmental Engineer,Boise,Idaho Agnes Y. Lee US.Bureau of Mines,Rolla,Missouri George Losonsky Eastman Christensen Environmental Systems,Houston, Texas James Lounsbury National Roundtable of State Pollution Prevention Programs, Silver Spring, Maryland David M. Manis EET,Austin, Texas J. L. Margrave Houston Advanced Research Center,TheWoodlands, Texas DianeMcCausland Chester Environmental,Monroeville, Pennsylvania TomMcDowell SiallonCorporation, Laguna Niguel, California Namunu J. Meegoda New Jersey Institute of Technology, Newark, New Jersey Alvin F. Meyer A. F. Meyer and Associates,Inc.,MeLean,Virginia Patrick B.Moroney Chester Environmental,Monroeville, Pennsylvania Vernon R. Miller U.S.Bureau of Mines,Rolla, Missouri James 1. Myers MilesInc., New Martinsville, West Virginia DanNickens Earth Resources Corporation,Ocoee, Florida S. C. Niranjan RiceUniversity,Houston, Texas 0.A. Ogunseitan University of California,Irvine, California KarnigOhannessian OPdiL,Inc., San Diego,California Osawaru A. Orumwense The University of Alabama,Tuscaloosa, Alabama Patrick D. Owens Tosco Refining Company,Martinez,California D. L. Patton Applied Environmental Services, Inc., San JuanCapistrano, California Richard C. Peralta Utah State University,Logan,Utah AnnaPeteranecz OPdiL,Inc., San Diego, California Dan A. Philips Pensacola JuniorCollege,Pensacola, Florida F. Preto CANMET, Ottawa, Canada RayRuemekorf NIRO,Inc.,Columbia, Maryland Stella S. Schramm University of Tennessee,Knoxville, Tennessee William E. Schramm Oak Ridge National Laboratory, Oak Ridge, Tennessee KlausSchwitzgebel EET,Austin, Texas KantiL.Shah OhioNorthernUniversity, Ada, Ohio S. S. Shukla* Houston Advanced Research Center,TheWoodlands, Texas Michael G. Stapleton University of Delaware,Newark, Delaware Herminio H. Suguino Utah State University,Logan,Utah S. M. Testa Applied Environmental Services, Inc., San JuanCapistrano,California George J. Trezek Greenfield Environmental, Carlsbad, Californiaand University of California at Berkeley, Berkeley,California T. Viraraghavan University of Regina,Regina, Canada Donald K.Walter US.Department of Energy,Washington, D.C. Robert L. Ward Ohio Northern University, Ada, Ohio Ann M. Wethington US.Bureau of Mines,Rolla, Missouri David S. Wilson Environmental Resources Management Group,Exton, Pennsylvania _________

*Current affiliation: Lamar University, Beaumont, Texas

'

This Page Intentionally Left Blank

Part I

ENGINEERING ISSUESIN POLLUTION CONTROL AND WASTE MINIMIZATION

This Page Intentionally Left Blank

Process Engineering for Pollution Control and Waste Minimization

John Hanna and Osawaru A. Orumwense The University of Alabama Tuscaloosa, Alabama

1.

INTRODUCTION

As a part of the material cycle, ores and fossil fuels are extracted from the earth, processed, and converted into metals, chemicals, and other processed (high value added) materials. Hence, any expansion in the world economy increases the demand for minerals and metals with subsequent increases in the amount of waste generated. Wastes are generated by the mining, mineral processing, metallurgical, and chemical industries at an estimated annual rate of over 2.3 billion tons. The accumulated solid wastes at both active and inactive mining sites approach a whopping 30 billion tons [l]. These wastes include gases, dusts, sludges or solutions, ashes, and a variety of massive solid materials such as overburden, waste rocks, tailings, and slags that must be disposed of at low cost with a minimum of environmental degradation. A large volume of the wastes is normally disposed of at locations close to either the mining sites or processing plants. Evidence of these can be found in Minnesota, Utah, Alabama, California, Tennessee, Idaho, Montana, and other states having high mining and industrial activities. A significant amount of the tailings is disposed of in impoundments, which range in size from a few acres to large ponds covering thousands of acres. Wastes from processing plants pose the most difficult disposal and environmental problems in view of the physical and chemical properties of the wastes as well as the enormous volumes involved, and consequently a large expanse of land must be used for the disposal [2]. A typical example is the Florida phosphate slimes. The overburden and waste rocks with characteristic high contents of pyrite, heavy metals, and radioactive materials also present potential environmental and health problems. Unfortunately, most of the mineral processing wastes have been excluded from the Resource Conservation Recovery Act of 1976 (RCRA). However, this situation is changing with the stringent environmental regulations introduced in recent years, which have necessitated that precautions be taken to both minimize and control waste disposal. Hence, the mineral or metal

3

Orumwense 4

and

Hanna

constituents in wastes, whether of little or no economic value, must be amendedfor their environmental impact or as a source of resource supply.

II. SOURCE OF WASTES A. Mining and Processing Solid Wastes Most operations in the extraction and processing cycle generate wastes (refer to Figure l), but the extent to which a material can be classified as a waste depends on a number of factors. These include 1. Sources andvolumesofwasteproduced 2. Potential dangers to health and the environment 3. Long-term reactivity with air and/or water and mobilization to the environment 4. Present disposal practices and alternative disposal methods 5 . Cost of disposal and potential use of the waste Overburden Sub-grade mlnerals -Slurries - cI Flnes

RAW MATERINS EXTRACTION mlnlng, quanylng. dredglng. exploratlon

""

I I

-

I

I RAW MATERINS BENEFICIATION

I

I I

I I

Concentrates. refined 011.gas. minerals

U

Dusts

stags Smoke Fumes Muds -Dresses solutlons Resldues Ash

t

Ingots. plgs. chemicals. energy

1

"S1urrlea

Chemlcals

I

7

-

I +I

"

SolUUOns

I

$

Talllngs Sands

--

mllllng. washlng. concentratJon. upgmdlng

MANUFACTURING AND SERVICES assembly, packing. transportation.energydlstrlb.

I

vI

I I

-

Pulp Dust Smoke Fumes solutlons

-- -

Coods and servias

I I I I I

I I

Ores. crude 011. coal. etc.

I

I I

-

Spoils

-

Drosses Grlndlng CIlpplngS

I

4

"

'3

I

I

Fumes Dusts

s01uu0ns Metals. glass. -paper. p~asuc.etc.

L-

Figure 1 Mineralwastematerialssupply,utilization,anddisposalsystem.

Smokes

I

--J

Process Engineering Control for Pollution

5

The following are typical examples of the waste generatedby mining and related processing industries. It has been reported that in the production of about 1.6 million tons of copper in 1976, 1 billion tonsof materials were processed. This breaks down to684 million tonsof overburden, 264 million tonsof tailings, 5 million tons of slag, 3.3 million tons of sulfur dioxide,and about 100 billion gallons of process water [3]. The iron industry is one source of enormous amounts of waste, since most of the iron concentrates used in the manufacture of iron and steel are derived from relatively low grade ores. spically, raw ores assaying25-33% Fe are mined and beneficiated to producehigh quality pellets assaying 60-65% Fe and 5% Si for the manufacture of iron and steel. About 330 million long tons of iron ore was mined in the United States in 1976, and the amount of wastes generated was about 200 million tons [4], excluding the slag and dust wastes from the steelmaking step. Conventional magnetic and gravity separation processing of magnetic and nonmagnetic taconites of the Lake Superior Region resulted in substantial iron losses of about 20-30% in the tailing products [5]. On the other hand, the more advanced beneficiation of the tailings from techniques such as flocculation and flotation processes reduced the iron lost in the rejectsto as low as 10%. The loss is partly due to the mineralogical compositionof the ores and the grain size of both iron and gangue. High iron losses are observed for Birmingham red hematite ore, for instance, because it produces more slimes than taconite ores. This is oneof the factors responsible for the relatively poor recovery of iron from run-of-mine material and the generation of large tonnages of wastes, particularly in large-scale beneficiation operations. The Florida phosphate industry is another sourceof a tremendous volumeof wastes. In the production of phosphate, the soft minerals in the matrix, particularly clays and the very fine phosphate aggregates,are dispersed readily in water, forming slimes duringthe hydraulic mining, transportation, and separation steps. These slimes are difficult to recover, and in addition they impair the beneficiationoperation. About one-third of the phosphate contentof the matrix is lost in the slimes, which are generally discarded as wastes. The Florida phosphate slimes are characterized byveryslow settling and trap a highvolume ofwater. Currently, impounded slimes are stored behind earth dams and pose a serious threat to the environment. The reclamation of the land and enormous volumes of water are important for resource conservation and in order to comply with stringent environmental regulations. The recovery of the phosphate values discarded in the slime and tailingfractions containing about 30%-40% of the phosphate present in the mined matrix would enhance the economy of the phosphate industryand expand the available resources. The impact of this on reducing potential environmental hazards is enormous. The phosphate losses at thecurrent rate of rock production of about 40 million tons per year includes over 11 million tons of high grade phosphate that is lost in the slimes annually. Vasan [6] has estimated that about1.5 billion tons of phosphate slimes is accumulated over the years in dams together with about 4.5 billion tons of water. The coalmining industry is another sourceof a large volume of solid wastes. The methods used in the past for cleaning coal were highly inefficient and resulted in high coal losses in waste streams duringthe mining and washing operations [2]. The washer waste fines are normally storedin above-ground impoundments. Quitea number of processing plants still indulge in the practice of discarding coal fines.As a result, about 25% of the coal mined is disposed of as waste.Based on the current rate of coal production of about 1 billion tons, about250 million tons of wastes is produced annually, and out of this, about 200 million tons is coarse particles and 50 million tons is fines. The amountof coal in coarse waste particles is more than 30 million tons of carbon per year, while the corresponding amount in the fine fraction is about 30 million tons on an annual basis. The disposal of coarse waste particles is not a serious problem

6

Hanna and Orumwense

in most coal preparation plants,as they are usedas landfill. The fine-size wastes, on the other hand, are a problem because of the difficulties experienced in dewatering and the characteristic relatively low structural strengthof fine particles, which prevent fines from being used as landfill [7].

B. Mining and ProcessingLiquidWastes Effluents from coal preparation plants and drainage from waste disposal sites have a characteristic dark color and have high concentrations of suspended particles that cause not only siltation due to the settling of coarse particles, but also water pollution, both of which have negative effects on aquatic lives[2,7]. Effluents from coal cleaning plantsand mines are also reputed to have a great impact on the environment through the phenomenon as known acid mine drainage (AMD). This is oneof the causes of the destruction of forestsand vegetation today. Acid drainage is reportedto be causedby the reaction,between oxygen, water, and iron sulfides such as pyrite and marcasite. Microorganisms are known to enhance the rate of this reaction. The most common techniques for mitigating acid drainage are neutralization using either lime, limestone, soda ash, or caustic soda; reverse osmosis; and treatment involving silicates [7]. These techniques are discussed in detail later. The highly acidic solutions produced dissolve several heavy metals in the waste pilesor impounded material and become loaded with a host of environmentally undesirable heavy metal species, sulfates, and other anions. On the other hand, the water discharged from some mines contains valuable metals such as copper and uranium that couldbe recovered economically. Copper is usually extracted from such discharges by either cementation or liquid ion exchange. Mine drainage containing9-12 ppm U,O, is stripped by ion exchange as exemplified by the operation in the Ambrosia Lake district [8]. The acid mine drainage containing300-600 ppm A1,0, and 10-20 ppm U,O,, on the other hand, is stripped by a combination of ion and liquid ion exchanges [9].

C.Coal

UtilizationWastes

As a result of burning coal in boilersand electric power plants, a large quantityof ash is produced. The amount of ash generated by power plants in 1977 is estimated to have been about 67.8 million tons, of which 48.5 million tons wasfly ash, 14.1 million tons bottom ash, and the remainder boiler slag. During that year, about6.3 million tons of the fly ash, 4.6 million tons of the bottomash, and 3.1 million tonsof the boiler slag,or approximately 21%of the total ash generated, was recycled in such products as concrete blocks, asphalt, and roofing materials [lo-121.

D. MetallurgicalWastes The production of alumina by the Bayer process each year is accompanied by simultaneous formation of about 7 million tons of red mud that consists of a substantial amount of valuable minerals and dissolved salts. These wastes are estimated to contain a large amount of caustic soda, 1.2 million tons of alumina, 1.7 million tons of iron, and about 450,000 tons of titania [13]. These pose severe environmentaland health hazards. In steelmaking, over 2 million tons of dust and gases is generated by electric and basic oxygen furnaces annually. The dust contains a substantial quantity of lead (0.4-2.6%), zinc (6.3-24.8%), manganese (0.5-5.3%), and copper (0.03-0.27%) in addition to iron [14]. Similarly, in manufacturing stainless steel, a large amount of metals is as lost wastes. Powell et al. [l51 estimated that approximately5 million poundsof molybdenum is lost in stainless steel furnace dust each year.

Process Engineering Control for Pollution

7

Table 1 Characterization of FoundryDust Analysis (wt. %) locationSample Alabama Ohio Michigan New Hampshire Pennsylvania Massachusetts West Virginia

cuPb

Zn

13.52 0.50 7.50 0.24 0.79 0.75 4.90

65.34 63.70 56.70 44.70 54.80 65.04 78.25

Fe 3.01 1.10 6;30 0.58 5.80 0.30 2.95 0.15 5.80 460 0.06 6.86 0.15 2.40 39 0.61

Toxicity

c1

(mg Pb/L)

0.06 0.54

530 440 764 188

0.66 0.40 1.30 0.5 1 0.05

6

A large amount of dust is producedby brass and bronze foundries and secondary smelters annually in the United States. The baghouse dusts vary in composition, but the main constituents are zinc(40-78%), copper (10-15%), and small amountsof lead and tin. Most of the zinc is present at ZnO, while the remainder is in the form of brass or bronze alloys. A typical characterization of dust from some foundries is presented in Table1. These wastes are considered to be hazardous because of the high lead contents. During the production of elemental phosphorus using an electric furnace, a large amount of toxic wastes such as sludge, slag, gases, and phossy water are also generated. It is known that between 5 and 10% of the elemental phosphorus that is produced is left behind in the sludge. The compositionof the other solid constituentsof the sludge is4040% SO,, 5-15% CaO, 2-4% Fe,O,, and 2-5% P,O, [16]. In general, the ratio of phossy water and sludge that are formed to the amount of elemental phosphorus produced is about5: 1. Phosphorus wastes pose both environmental and fire hazards,and these wastes are producedat a rateof 1.5 million tons annually. 111.

METHODS OF CONTROL AND TREATMENT OF BULK SOLID WASTES

A number of measures are taken to minimize or render bulk solid wastes safe for disposal. These include the extraction of heavy metals or toxic constituents from the waste materials using either physical, chemical,or bioremediation techniques.On the other hand, some wastes are either recycledor used directly, but more often a combination of these techniques is applied to achieve maximum process efficiency. The following methods are classified according to the source of the solid wastes.

A. Copper Mine Wastes Copper mine wastes are increasingly important because of thelow very grade of most available copper ores. Rule and Siemens [l71 have shown that the bulk flotation method is effective in extracting such metal values as copper, cobalt, and nickel from copper mine wastes with recoveries in the range of 54-95%. The primary problem in using the flotation method for this purpose is the intimate association of the valuable minerals or metals (minor) with the predominant gangue materials. Consequently, a high degree of fineness is necessary in order to ensure liberationand subsequent separationof the metal values. However, reagent consumption is also expected to be high. In most instances, the residues still contain fairly high levels of valuable minerals or heavy metals and as a result mustbe subjected to further treatment. Pressure leaching or bacterial leaching (bioleaching) is often used for this purpose.

Hanna and Orumwense

8

B. IronOre Wastes In the past, many iron tailing ponds were subjected to gravity concentration[l81 to recover the iron contents. Jones and Laughlin Steel Corporationin Calmet, Minnesota, is an example of a company that at one time combined flotation and gravity concentrationfor treating iron wastes to recover the metal values. The presence of a large amount of slimes and the high impurity contents of either the initial ores or the wastes impaired the recovery of iron from the wastes. In contrast, selective flocculation and high-intensity and high-gradient magnetic separations [l91 are some of the other techniques that canbe used effectively totreat such materials. Waste materials can also be subjected to reductive roastingand magnetic separation to reduce the energy required for processing.

C.Phosphate

Rock Wastes

Phosphate slimes are known to be not only difficult to recover but also economically unsound. However, the associated adverse environmental impact necessitates treatment. Laboratory tests on waste pond materials, low-grade washer products, and some raw Tennessee phosphate ores have shownthat some of the phosphate can be recovered. Market grade phosphate concentrates assaying 60-82% P205can be obtained in substantial amounts using the anionic flotation method [20]. Direct digestion of the phosphate matrix with sulfuric acid is an alternative approach for the minimization of slime disposal problems.This process producesa simple waste consisting of gelatinous slime, sand, and gypsum. The composite is a compact sandy cake that could be 95%of the P205is recovered as useful used as a filling materialin mined-out areas while about material [2l ,221.

D. Fine CoalWastes TWO major

techniqueshave been proposedfor treating coal wastes. Theseare gravity separation and flotation [23,24]. The use of Humphreys spirals to treat coal wastes has been established. Although such treatmentsare capable of yielding high-grade coalconcentrates, the recovery is relatively low. Also these techniquesare only applicableto feeds withparticle size coarser than 200 mesh. Besides,a substantial amountof the coal is lostin the tailings-about 10-71% [24]. Therefore, techniques thatare suitable for fine particles processing are required to supplement the spirals in order to improve coal recovery.This has led to the development of a process that is based on a combination of gravity separation and froth flotation. In this process, Humphreysspirals are used to recover the coarse coalparticles while column flotation is employed for the minus 200 mesh size fractions [24]. Mechanical flotation can also be used in place of spirals to separate the coarse particles. In this manner, both the quality and the recovery of coal are improved significantly. Similarly, thepyrite present in the wastes can be removed, and by doing so, acid drainage problems can alsobe mitigated. It is also possible to employ a bioleaching techniqueto eliminate the pyrite constituents from coal wastes. This can be achieved by allowing bacteria to oxidize the pyrite in coal wastes as feed.

E. Phosphorus Wastes The methods of treating phosphorus waste include physical, ch’emical, and bioleaching techniques. The physical methods include sizing, sedimentation, centrifugation, cycloning, and flotation [25-27, 311, while air oxidation, chlorine oxidation, electrolytic oxidation, catalytic

gineeringProcess

for Pollution Control

9

oxidation, distillation, CS2 extraction[28-331, and ion exchangeconstitute the chemical methods. Most of these processes either partially separate or oxidize phosphorus from the impurities.Therefore, a combination of twotreatmenttechniques is necessary for complete remediation of phosphorus wastes. Another factor necessitating this methodology is the associated low operating costs for such schemes. A combination of clarification and chlorination techniques has been developed for extracting elemental phosphorus from phossy water [26]. However, the associated residual chlorine has an adverse environmental impact, and this renders the technique impractical, The ERCO process is based on the use of nascent oxygen to oxidize elemental phosphorus prior to subsequent separation [33]. Another method uses distillation as the basis for the remediation of phosphorus from sludges [29]. The high operating costs associated with these methods have limited their application. In many cases a major part of the phosphorus wastes are present in the coarse particles. Anazia et al. [31] have shown that between 26 and 29% by weight of the particles in the tested sludge samples (obtained from FMC Corporation of Pocatello, Idaho, andthe TVA at Muscle Shoals, Alabama) are coarse phosphorus particles containing 82-91% P4. It was also demonstrated in the same study that about61-88% of the coarse phosphorusparticles can be recovered by screening. The fine fractions represent 71-74% byweightof the sludge and assay 5-21% P4. The as-received unsized sludge can also be subjected to flotation to separate phos61 and 78%. P4 with a recovery in the range of 71-79% phorus concentrates assaying between depending on the characteristics of the wastes. The tailings assayed between9 and 18% P4and constitute about 59-68% of the sludge [31]. It is obvious, therefore, that the fine fractions or tailings must likewise be subjected to further treatment using other methods. The phosphorus remainingin the physical separation rejects can be extracted after air oxidation treatment at ambient temperatures. These form the basis for the proposed two-step method comprising either flotation or screening and conventional air oxidation for the treatment of phosphorus sludges [31]. However, the P4 concentrations in the refuse from the oxidation step can be as much as 4%, which is still high in terms of toxicity. A long air oxidation periodof several days or weeks may be necessary to achieve 90-95% conversion of P4 to H3P04 at an ambient temperatureof 30°C. Under these conditions the oxidation rate of P4 in water is slow and is influenced by many factors such as pH, oxygen content, temperature, presence of metal ions, and degree of dispersion of colloidal material [34]. Therefore, an incomplete conversion of P4 to oxyphosphorus compounds occurs during the conventional oxidation process because the reaction kinetics appear to be influenced by other factors such as agitation, particle size, and surface coating [34]. This process has been further developed at the University of Alabama such that the oxidation and conversion of P4 to soluble oxyphosphorus compoundsare enhanced significantly [32, 351. In the new process, the insoluble P4 is converted to highly soluble and nontoxic compounds thatare easy to extract from the rest of the sludge. This improvement has been achieved by employing a novel reactor design known as HSAD to expedite the remediation operation. Thus, depending on the P4 contentof a sludge, an almost complete oxidationof phosphorus is achieved in about 1-3 hr, and the resultant acid solution can be employed in the manufacture of either phosphoric acid or fertilizer by-product by neutralization. The chemically inactive solid waste can be dried and safely disposed of as nonhazardous landfill product. Someof the results obtained employing this processare given in Table 2. The advantagesof the HSAD technique includeshort processing duration, high efficiency, simple configuration, low cost, and applicability to various phosphorus wastes. The process requires no catalysts, chemical oxidants, or high temperatures [35].

Hanna and Orumwense

10

Table 2 Test No.

mid Results of HSAD West Oxidation of Phosphorus Sludge [32] Product Weight

(g)

Weight (%)

P4 Analysis (%) ~

1

2

Solution Residue 35.40

50.20 27.51

Feed

77.71

63.44 Solution 36.56 Residue

49.76 26.68

Feed

78.44

64.60

100.00

100.00

P4 Removal (%)

~~

53.39" 0.02

99.97 0.03

34.50

100.00

54.35" 0.05

99.94 0.06

34.50

100.00

"Equivalent P4analysis of oxyphosphorus compounds.

F. 'Brass and Bronze Foundries Dust Baghouse zinc dust is processed by using sulfuric acid leaching and electrolysis or crystalli[36,37]. The zinc extractionattainable with this method zation to recover zinc and other metals is in the range of 89-99%. Basically, the method involves the use of strong sulfuric acid and intense aeration to dissolve the zinc oxide and metallic copper from the dust. The lead, tin, and zinc metal alloys present in the dust are not dissolved by sulfuric acid and remain with the solid residues. The leach residues, which account for 20-50% by weight of the dust and are rich in a number of metals, can be further treated to extract the metallic components [36]. The pregnant leach liquor is subjected to electrowinning to produce metallic zinc. However, the zinc electrowinning operation is adversely affectedby the presenceof chloride ions or some metals. Hence, additional measuresare required to eliminate chloride ions andother impurities in order to produce high-grade metallic zinc. This can render the whole process expensive. Alternatively, the crystallization technique is employed to recover the zinc from the leach liquor as zinc sulfate salt.

IV. ADVANCEDREMEDIATIONTECHNOLOGY A. Leaching The most common method of recovering the metal values from low-grade sources such as waste dumps or heaps is leaching. Leaching is a process in which a solid material is contacted with a solvent in order to selectively dissolve some of the components. The objectives of leaching metals from sludge include the dissolution of the metal valuesfor recycling or subsequent separation by other methods, to render wastes nonhazardous,or to render wastes amenable to further treatment. Leaching is known to account for about 10% of the yearly copper production. The commonly used leaching agents are sulfuric acid, hydrochloric acid, ferric chloride, nitric acid, ferric sulfate,ammonia or ammonium carbonate, hydroxide, and microorganisms such as bacteria, yeast, and fungi. Unfortunately, many factors concerning leaching such as thesize and heightof dumps and factors affecting solution percolation andthe kinetics and recoveryof the valuable metals from the leach of pregnant liquors in general still require detailed studies and information dissemination. The fact that many of the minerals in wastes canbe recovered inexpensivelyby leaching implies that some of the problems associated with the disposal of fine wastes canbe alleviated. Biological remediation of wastes is accomplished by using naturally occurring microorganisms such as bacteria, yeast, and fungi to treat contaminants. Its use is rapidly increasing. However, the microorganisms requirea wide rangeof macro and micro nutrientsfor their met-

Process Engineering Control for Pollution

I1

abolic activities and growth. The environment is generally poor in the nutrients such as nitrogen, phosphorus,andcarbonrequired by the microorganismsforsustenance,andsome contaminants exhibit a certain degree of resistance to different microorganisms. These are the primary causeof the slow rate of breakdown of contaminants. Therefore,a successful bioleachingoperationrequiresthegrowth of appropriatemicroorganismsthatcan be inducedby manipulating conditions suchas the availability of nutrients, temperature, electron acceptors, and aeration.

B. Precipitation Precipitation is one of the common means of remediating wastewater. In this method, chemicals are used to alter the physical state of dissolved or suspended metals and to enhance subsequent separation using sedimentation techniques. Chemicals such as caustic soda, lime, soda ash, sodium borohydroxide, sodium phosphate, ferrous sulfide, and sodium sulfides are used to induce precipitation. It is sometimes necessary to subject wastewater to some form of pretreatment such as filtration, destruction or organic matter and cyanides, metal reduction, neutralization, and/or oil separation prior to precipitation. Some metals as typifiedby hexavalent chromium are difficult to precipitate in the form in which they occur and must be reduced if the operation is to be successful. Reducing agents commonly used include sulfur dioxide, sodium bisulfite, sodium metabisulfite, and ferrous sulfate. Similarly, to effect sedimentation and subsequent separation of precipitates, flocculants are sometimes required. Lime,alum, and polyelectrolytes are used for this purpose. The major characteristics of wastes that have an impact on precipitation operations are the type and concentration of metals, amount of total dissolved solids, concentration of residual complexing agents, and amount of oil and grease present in the wastes. Metal-laden wastewater resulting from electroplating, pigment manufacture, the photographic industry, battery manufacture,and nonferrous metal industriesare usually subjected to precipitation treatment.

C.IonExchange In the ion-exchangeprocess,metalions in a dilute solution are substituted for identically charged ions electrostatically bonded to the surface of an immobile solid medium. The solid medium can be either a naturally occurring inorganic zeolite or a synthetic organic resin. Ion exchange is a reversible chemical reaction. Therefore,the loaded resin or exchange medium is placed in a pure solution of appropriate pH and the trapped metal ions are released. This method is applicable only to liquid wastes or pregnant solutions. The performanceof the process depends on(1) the concentration and valenceof the metal constituents, (2) the presence of competing ion species, and (3) the presence of dissolved or suspended solids and organic compounds. Therefore, the feed toan ion-exchange system must be subjected to pretreatment. Thismethod results in about 95% metal recovery and high purity products. This method is fully developed and is used commercially to remove chromium, copper, nickel, cadmium, silver, and zinc from wastewater.

D. ElectrolyticRemediation Electrolytic cell is the primary device used in electrolytic remediation. It consists of an anode and a cathode immersed in an electrolyte. When an electric current is applied to an electrolyte solution, the dissolved metals are reduced and subsequently deposited at the cathode. The electrolytic remediation technique isalso known as electrowinning because the metals recovered are of high purity. This is one of the most effective methods for remediating chelated

12

Hanna and Orurnweme

metals, which are difficult to retrieve by other techniques. This method has the advantage of producing metal-laden free sludge, but it is limited to solutions containing a fewtypesof elements. Electrolysis can be used to remediate cadmium, chromium, copper, lead, tin, and zinc. However, such treatments involve a high energy expenditure. Wastes containing copper and certain other elements must be leached with hydroxides before being subjected to electrolytic treatment. A variation of the electrolytic technique known as electrodialysis is obtained when a membrane is placed between the anode and cathode such that the mobility of some ions throughthe membrane is obstructed. Electrodialysis can be used for remediating wastes from such sources as gold-, chromium-, silver-, zinc-, nickel-, and tin-plating operations where the ion concentration is low and would not be economical for electrowinning. Most feeds for electrodialysis is a compulsory treatment must be filtered to remove suspended solids. Besides, pH control pretreatment measure because of the effect on metalseparation. When electrodes having a high surface area are employed, metals removalof about 98% can be achieved.

E. MembraneSeparation The membrane separation method encompasses such techniquesas filtration (microfiltration, ultrafiltration, etc.), reverse osmosis, and electrodialysis. Thefiltration technique is used after the sludge hasbeen pretreated for the removal of metals. The techniqueis also usedto pretreat feeds destined for subsequent treatmentby both reverse osmosis and electrodialysis. Reverse osmosis and electrodialysis are used to retrieve metalsor plating compounds from wastewater. The electrodialysis method is described in the preceding subsection. Reverse osmosis (RO) systemsare characterized by having a number of modular units connectedeither in parallel or series or a combination of the two. The applicationof this method to the remediation of metal-laden sludgeis limited by the pH range in which the membrane can be used. Cellulose acetate membranes are not suitable for use at pH above 7, while amide and polysulfone membranes can be used in the pH range between 1 and 12. The performance of R 0 systems is impaired by the presence of colloidal matter, dissolved organics, and insoluble constituents. It is recommended that the feeds to R 0 systems be subjected to such pretreatments as pH adjustment, carbon adsorption, and filtration. The method is used commercially to remove brass, chromium, copper, nickel, and zinc from metal-finishing wastes. These techniques canbe used to produce effluents with very low metal constituents, provided, of course, that adequate pretreatmentshave been carried out. Metal removal onthe order of 99% can be achieved by making use of a combination of precipitation and filtration.

F. Evaporation Evaporation is a simple method for remediationof mixed materials based onthe difference in volatility. Hence, the concentration of metals is brought about by the reduction in the volume of the waste. The primary instrumentation used for this purpose includes rising film, flash, and submerged tube evaporators. Cadmium, chromium, nickel, zinc, copper, and silver from platingbaths are retrieved in theelectroplatingindustry by using this method.However, this method of remediating wastes is cost-effective only when a very small volume of waste is involved.

G. Encapsulation Soluble silicates and their derivatives are very effective for the stabilization and fixation of hazardous wastes.Silicates are used in waste treatment becauseof their inherent characteristics

ngineering Process

Control for Pollution

13

such as alkaline nature (pH 20-14), ability to form gels, and reactivity with multivalent cations, and because their disposal poses no potential dangerto the environment. Soluble silicates are polymeric and condense on aging to form anions having a siliconoxygen-silicon linkage that are complex and exist in various chain lengths and cyclic structures. Silicates react with metal ionsto form insoluble amorphous metal silicates. These metal pH range compared with simple metal hydroxides. silicate complexesare insoluble over a large This is responsible for the increased resistance to leaching of metals in solidified wastes and is perhaps the main feature of silicates in waste treatment. Soluble silicates are made by fusing sodiumcarbonate or potassiumcarbonateandsandinafurnace at 1450°F. Theresultant nSiO,Na,O compound has silica (SiO,) to alkalinity (Na,O) ratio in the rangeof 1.6:3.9. The SiO,Na,O ratio has great significancein subsequent useof silica because only compounds having high ratios are employed in the manufacture of products such as gels, precipitated silica, and zeolites and in the treatmentof wastes. Setting agents commonly used in waste treatment include Portland cements, pozzolanic fly ashes, and cement or lime kiln dust. The active components in setting agents are such derivatives as the mono-, di-, and tricalcium silicates formed when the agent is mixed with water. The physical properties and behavior of setting agents are strongly influencedby the calcium silicate content, as this is directly related to the number and strength of the resultant bonds formed. Silicates also reduce the permeability by reducing calcium hydroxide inclusion formation or the presence of voids in the structure of the material. The treatment of hazardous waste with setting agents can be subdivided into two categories, stabilization and fixation. Stabilization is a chemical processof transforming a liquid waste into a solid. The setting agents are mixed with the waste, and when they “set up” or harden, the waste material is entrapped in the structure. The procedure used in the stabilization operation involves premixing the waste and setting agents before introducing soluble silicate. The role of silicates in the stabilization process includes the reduction of setting time, decreasing of the permeability, increasingofthecompressivestrength,andreductionofboththeamountofsettingagents employed and the volume of the treated waste. Fixation, on the other hand, is similar to stabilization in many respects, but rather than merely entrapping the wastes as inclusions, the wastesare modified and bonded into a cementlike matrix. Hence the solubility or leachability of hazardous components is reduced dramatically. In this manner, the toxicity and mobility ofheavymetalwastesarechangedbythe treatment. The treatment steps involve mixing the waste with cement or kiln dust as a setting agent and water. Thereafter, a soluble silicate is introduced and mixed thoroughly. The procedure is recommendedif a good result mustbe obtained, and cementmust be used when a waste canused with silicates is tobe fixed. Portland cement is the most effective setting agent that be for this purpose. The reason is that during hydration cement produces gels that help to encapso sulate waste. Lime-based materialsdo not produce a large amount of gels during hydration, the amount of bonded wastes is reduced. Therefore, lime-based setting agents should not be used for waste fixation.

REFERENCES Hill, R. D., and Auerbach, J. L., Solid waste disposal in the mining industry, in Fine Particles Processing, Vol. 2 (€?Somasundran, e d . ) , SME-AIME,NewYork, 1980, pp. 1731-1753. 2. Hanna, H. S., and Rampacek, C., Resources potential of mineral and metallurgical wastes, in Fine Particles Processing,Vol. 2 (F?Somasundaran. ed.), SME-AIME, New York, 1980, pp. 1709-1730. 3. Mineral Trends and Forecasts, US. BureauofMines, 1979. 4. Klinger, F. L., Mineral facts and problems-iron, Bull. 667, U.S. Bureau of Mines, 1976, pp. 5251.

545.

Hanna and Orumwense

14

5 . Rampacek, C., The impact of R&D on the utilization of low-grade resources,

Chem. Eng. Prog.,

February, 57-68, 1977. 6. Vasan S., Utilization of Florida phosphate slimes, Proc. 3rdMineral Waste UtilizationSymp., Chicago, 1972, pp. 171-177. 7. Moudgil, B. M., Handling and disposal of coal preparation plant refuse, in Fine Particles Processing, Vol. 2 (F!Somasundran, ed.), SME-AIME, New York, 1980, pp. 1754-1779. 8. Spendlove, M. J., Bureau of Mines research on resource recovery, IC 8750, U.S. Bureau of Mines, 1977. 9.

IO. 11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Ross, J. R., and George D. R., Recovery of uranium from mine waters by countercurrent ion ex-

change, RI 7471, U.S. Bureau of Mines, 1971. Faber,J.H., Coal Technology Conference, Houston, Texas, 1978. Brackett, G . E., Production and utilization of ash in the United States, Proc. 3rd. Int. Ash Utilization Symp., Pittsburgh, Penna., IC 8640, U.S. Bureau of Mines, 1974, pp. 12-18. Jackson, J., Total utilization of fly ash, Proc. 3rd Miner. Waste Utilization Symp., Chicago, Ill., 1972, pp. 85-93. Dean, K. C.,Utilizationofminemillandsmelterwastes, Proc. 2nd MineralWaste Utilization Symp., Chicago, Ill., 1978, pp. 138-141. Dressel, W. M.. Barnard, F! G., and Fine, M. M.. Removal of lead and zinc and the production of prereduced pellets from iron and steel making wastes, RI 7027. U.S. Bureau of Mines, 1974. Powell, H. E., Dressel, W. M., and Crosby, R. L.,Converting stainless steel furnace flue dust and wastes to a recyclable alloy, RI8039, U.S. Bureau of Mines, 1975. Slack, A. V. (ed., Phosphoric Acid, Parts 1 and 2, MarcelDekker,NewYork, 1965. Rule, A. R., and Siemens, R. E., Recovery of copper, cobalt and nickel from waste mill tailings, Proc. 5th Mineral Waste Utilization Symp., 1976, pp. 62-67. Fine, M. M., and Heising, L. F., Iron ore waste occurrence, beneficiation and utilization,Proc. 1st Mineral Waste Utilization Symp., Chicago, Ill., March 1968. pp. 67-72. Colombo, A. F., Jacobs, H. D., and Hopstock, D. M.,Beneficiation of Western Mesabi Range oxidized taconite, RI 8325, U.S. Bureau of Mines, 1978. Lamont, W. E.,etal.,LaboratoryflotationstudiesofTennesseephosphatesinthepresenceof slimes, RI 7601, U.S. Bureau of Mines, 1972. White, J. C., Fergus, A. J., and Goff, T. N., Phosphoric acid by didect sulfuric digestion of Florida land pebble matrix, RI 8086, U.S. Bureau of Mines, 1975. White, J. C., Goff, T. N., and Good, I? C., Continuous circuit preparation of phosphoric acid from Florida phosphate matrix, U.S. Bureau of Mines, 1978. Browning, J. S . , Recovery of fine-size waste coal, Final Rep. U.S.Dept. of Energy, Contract ET76-G-01-9005, Univ. Alabama, May 1978. Hanna, J., and Kalathur, R., Recovery of fine size coal from impounded wastes, Miner. Metall. Processing, November, 174-179. (1992). Fleming, J. D.,Removalofphosphorus,aliteraturesurvey,TennesseeValleyAuthority,Muscle Shoals, Alabama, 1970. Barber, J. C., Recovery of phosphorus from dilute waste streams, U.S. Patent 4,595,492 (July 17. 1969).

27. 28. 29. 30. 3 1. 32.

Crea, D.A., et al., Recovery of phosphorus from electric furnace sludge, U.S. Patent 3,615,218 (October 1986). Post, L. B., et al., Recovery of phosphorus from electric furnace sludge, U.S. Patent 3,615,218 (October 1971). Holmes, W. S., Lowe, E. J., and Brazier, E. R., Phosphorus distillation, U.S. Patent 4,081,333 (Mar. 28, 1978). Hinkebein, J. A. Recovering phosphorus from sludge, U.S.Patent 3,436,184 (April 1969). Anazia, I., Jung, J., and Hanna J., Recovery and removal of elemental phosphorus from electrical furnace sludge, Min. Metall. Processing, May, 64-68 (1992). Hanna, J., and Jung, J., Phosphorus removal by dispersed air oxidation, Miner. Metall. Processing November, 200-205 (1992).

Process Engineering Control for Pollution

15

33. Deshpande, A. K., Oxidation of phosphorus in aqueous medium, U.S. Patent 3,971,707 (July 27, 1976). 34. Sullivan, J. H., Jr., et al., A summary and evaluation of aquatic environmental data in relation to establishing water quality criteriafor munitions-unique compounds. Part 3. White phosphorus,Final Report, Water and Air Research, Inc., Gainesville, Ha., April 1979. streams, Proc. Haz35. Hanna, J., and Jung,J., Remediation of phosphorus from electric furnace waste ardous Materials Control, HMC-South '92. New Orleans, La., Feb. 26-28, 1992, pp. 34-39. 2nd Annual Environmental 36. Hanna,J.,andRampacek.C.,Recoveringzincfrombaghousedust, Society, Milwaukee,Wisc.,Aug. 23-24,1989, Affairs Conf. of theAmericanFoundrymen's pp. 119-126. 36. CommodityDataSummaries,Phosphates, U.S. Bureau of Mines, 1978. 37. Powell, H. E.,et al., Recoveryof zinc, copper and lead-tin mixtures from brass smelter flue dusts, RI 7637, U.S.Bureau of Mines, 1972. 38. Proceedings 3rd International Symposium, IC 8640, U.S.Bureau of Mines, 1974.

This Page Intentionally Left Blank

2 Selection of Least Hazardous Material Alternatives

Alvin F. Meyet A. E Meyer and Associates, Inc. Mckan, Virginia

1.

INTRODUCTION

A. Summary This chapter addresses the underlying purposesof accomplishing measures to select the least hazardous of alternative materials usedor planned for use by industrial and governmental entities. It then discusses the relationship of substitution processes to other considerations in the decisionprocess. An overview of someapproaches for selectionmethodsispresented.A “methodology” originally developed for the U.S. Navy is described, along with examples.

B. Substitution Methods as an Element of Pollution Prevention 1. Substitutionvs.Controls A longstanding principle of environmental control and industrial hygiene is that the first and basic consideration in control of hazards is the elimination of the hazardous component, or if that is not feasible, then the substitution of a lesser hazard[l]. The concept of eliminating the source or reducing the amount of toxic materials includes in addition to substitution of materials such other measuresas process or operation2 changes, properdesign of operations, and housekeeping. The importance of eliminating the need for costly environmental control measures by substituting less dangerous or less offensive materials also is of longstanding recognition[2]. Likewise, the economic advantages of industrial waste recovery is not a new concept. Nonetheless the primary approachto environmental control, until the late198% was that of using &‘end of the pipe” control measures. This was primarily in response to the focus of environmental regulations specifying limits to be metby treatment or other control measures. In the early 1990s there emerged a steady increase in recognition both in regulatory agenties [e.g., u.S. Environmental Protection Agency (EPA),the Department of Defense1 and in the private sector that a comprehensive, cost-effective approach to environmental quality in17

18

Meyer

cludes source reduction, recovery, and reutilization, with treatment and ultimate disposal as the last resort. The Pollution Prevention Act of 1990 established these concepts in a national policy that “pollution should be prevented or reduced at the source whenever feasible.”

II. HAZARDOUS MATERIAL SUBSTITUTION ACTION A. Statutory and Regulatory Requirements There are many statutory and regulatory requirements that directly or indirectly create the need for hazardous material assessment. These include federal statutes and their implementing regulations or standards, executive orders requiring federal agency compliance, and state and local codes, standards, and regulations. The Department of Defense (DOD) and the military departments and agencies have requirements that implement the federal mandates and some requirements that predate them. 1 . Federal Codes, Standards, and Regulations The primary federal statutes and their implementing regulations regarding environment, safety, and health are the Clean Air Act, Resource Conservation and Recovery Act (RCRA), Clean Water Act, Safe Drinking Water Act, Toxic Substances Control Act (TSCA), Emergency Planning and Community Right to Know Act (EPCRA), Pollution Prevention Act of 1990, Occupational Safety and Health Act (OSHA), Hazardous Materials Transportation Act, and the National Environmental Policy Act (NEPA). These acts taken together impose a need to examine the feasibility of using materials that are less hazardous, are less costly, or impose fewer administrative or other regulatory compliance resource requirements.

2. Possible Application of Department of Defense Methodologies A Risk Assessment Code (RAC) procedure was developed by the Department of Defense in the early 1960s [3]. Initially, it was designed to provide a means of ranking hazards associated with new weapon systems. Subsequently, the procedure was adapted in 1981 to rate occupational safety and health deficiencies. In its simplest form, the procedure provides a rating scheme based on a matrix to estimate the severity of effects of the hazard and the probability of occurrence, with the results stated as a risk assessment code. The range is from RAC 1 (catastrophic impact) to RAC 5 (negligible) (see Table 1). The later (1981) procedure, which is still in use, includes a cost effectiveness index and an abatement priority ranking. The revised procedure takes into account, with a numerical algorithm, such circumstances of exposure and resultant effects as the OSHA Permissible Exposure Limits (PEL), number of employees involved, effects of exposure (ranging from death to minimal lost time, disease, or injury), and duration of exposure. It is firmly established in the military services procedures for evaluating and prioritizing occupational safety and health hazard abatement requirements. The RAC schemes deal primarily with chemical and safety hazards ratings, with no consideration for environmental ramifications. Recently there has been an increased focus on the environmental aspect of hazardous materials use on the decision-making process. Unlike the long history of chemical and safety hazards rating schemes, there are no universally accepted systems for environmental hazards and risk acceptance. One possible method is a European model, described below. 3. A European Method for Priority Selections and Risk Assessment A study requested by the European Community Commission to provide a practical approach for priority setting among existing chemicals was prepared by Sampaolo and Binetti [4]. Using this

Least Hazardous Material Alternatives

19

Table 1 Risk Assessment Code Rating Schemea Hazard Severity (HHSC)

I I1 111 IV

Mishap probability (MPC) A 1 1

2 3

B

C

D

1

2 3

4

2 3 4

4 5

3 5 5

aInterpretation of HM selection Risk Assessment Code: RAC 1 = high risk (imminent danger of life or property; possible civil or criminal action) RAC 2 = serious risk (may result in severe injury or illness on or off site, potential for major damage to environment, and resulting notice of violation) RAC 3 = moderate risk (may cause few illnesses or injuries or significant property damage or environmental impact on or off site) R A C 4 = low risk (can result in only minor impact on or off site or only violation of a standard without damage) RAC 5 = negligible (insignificant impact)

model, an individual property or a number of properties of a given chemical can be evaluated and then ranked with those of other substances. This flexibility allows for evaluating different relationships. For example, one might want to compare only the intrinsic properties with respect to direct personal exposure in a particular circumstance or with respect to environmental exposure. Certain chemicals might have different relative rankings for these two categories. This model offers a number of advantages: The system is simple and flexible enough to be adapted to different and specific needs (i.e., personal exposure to general exposure, risk from domestic exposure vs. professional exposure, etc.). It is a self-improving system because new information can be input and the result can be refined further. This model uses three sets of parameters to evaluate risk and the priority of a given chemical: assessment of intrinsic properties, risk assessment or potentiality, and priority assessment.

Intrinsic Properties. Intrinsic properties of a substance are based on the set of physicochemical , toxicological, and ecotoxicological properties that are considered fundamental (or intrinsic) to the first evaluation of the substance. Each element of the intrinsic property (e.g., molecular weight under the physicochemical category) is assigned a numerical value that corresponds to its level of danger. From this information a score is developed for each intrinsic property, which also addresses the availability or nonavailability of the data. These intrinsic properties are considered additive and determine the intrinsic danger of the substance independently of external agents or factors that may influence it. External Factors. Risk assessment or potentiality includes not only the intrinsic danger of a substance but also the external factors that can influence the danger. These external factors include the quantity of the substance on the market, the plurality of possible exposures, and the size of the risk population. As an example, a substance may be highly dangerous, but if it is not on the market it will not pose any effective risk, and thus its intrinsic risk will be minimal. Priority Meusurement. Priority assessment involves both the known or presumed danger of the chemical and the degree of the lack of knowledge of the substance’s properties. A priority measurement can be made by calculating the ratio of the weighted figures for properties without data to those figures with available data. Both the risk assessment and priority assessment parameters can affect the intrinsic properties of a substance by multiplying or canceling them.

20

Meyer

LIFECYCLE AND MANAGEMENT CONSIDERATIONS

111.

A. LifeCycleConsiderations Regardless of the methodology usedin rating hazardous properties of a material, the selection process for the “least hazardous” involves more than environmental, safety, and health considerations. In addition to assessmentsto determine the least harmful material based ona hazardassessment(suchasthealgorithmdescribed later in this chapter), there are major considerations that must be carefully assessed. 1. BasicFeasibilityandEngineeringConsiderations A fundamental question that should be addressed is: Does the substitute perform adequately for its intended use? This requires determination of the following: 1.

2. 3.

4. 5. 6.

“Favorable” vs. “adverse” effects on required performance of the material(s) in production, operations, and maintenance situations. Creation of new or different hazards (suchas substitution of a less toxic material witha fire hazard potential for one that is highly toxic but has a low hazard or no hazard). Durability and life cycle times to failure (as with a low volatile organic compound(VOC) paint that may or may not last as long in a very hot or very cold climate). Maintainability of equipment involved in using a substitute. Possible process or equipment changes that may be needed. Environmental and/or OSHA controls required even if it is the lesser hazard.

2. LifeCycleCostConsiderations There are costs and benefits associated with the engineering and feasibility considerations that need to be assessed. In addition, there are many other costs associated with the life cycle of hazardous material. That life cycle extends from the time of concept through procurement, storage, use, and disposal. It is beyond the scopeof this chapterto do more than highlight such costs. It is also beyond its scope to describe the economic analyses required to evaluate the relative costs and benefits of two or more candidate materials for selection. Among the many costs that should be taken into account in the selection process are those shown in Table 2. A very useful guide for comparing alternatives is the EPA Pollution Benefits Manual [5]. The EPA Pollution Benefits Manualprovides for a financial analysis approachto compare alternatives for pollution prevention. It involvesa four-tier cost analysis from which economy feasibility of alternatives can be evaluated. The four tiers are as follows: Tier 0,Usual costs. The alternatives are identified, and all normal costs associated with each are determined. These include investment (depreciable capital, expenditures), operating costs, and operating revenues. Table 2 Life Cycle HazardousMaterialsCostsandCost Avoidance Considerations Acquisition Supply and storage Use

Waste treatment Other disposal Emission control Inventory control Engineering/process controlkhange Training

Safety Hazard/risk assessment ENEIS Permits Personal protection Medical monitoring Spill prevention and control Regulatory overhead Liability

Least

21

Tier l , Hidden Costs. These include such investments as monitoring equipment, protective equipment, control technology, and operating costs such as reports, monitoring training, and medical costs. Tier 2, Liability Costs. Included are penalties, fines, and future liabilities. Tier 3 , Less TangibleCosts. These include costs such as those associated with labor relations, and public relations. The results of the tier analyses are thenincorporatedintocostsummariesandfinancial worksheets, which result in an assessment and/or comparison of any cost savings for each alternative. This procedureallowsforcomparison of relative costs and benefits of selected alternatives.

B. Management Decisions and Actionsfor Selection of Least Hazardous Material 1. DrivingForces The basic driving forces for managementto consider in the selection of less hazardous materials, in addition to regulatory requirements, include the life cycle cost considerations discussed above and such needs as planning for new products or processes (or changes to existing ones), avoidance of new and long-term liabilities, and the possible benefitsof participation in such voluntary programs as the U.S. EPA Industrial Toxics Project (ITP). Other benefits include improved employee and community relations.

2. “Closing the Loop” Once decisions are made for substitution, a large number of follow-on actions are needed. These include planning for phase-out of the existing in-usematerials, development of new specifications and technicaldata documents, trainingof personnel, provision of any necessary controls, and compliance with any new permit or similar requirements.

W. DESCRIPTION OF METHODOLOGY A. Overview of the Navy Substitution Algorithm The hazardous material substitution algorithm developed for the U.S.Navy also had wide POtential for selection or substitution of least hazardous materials in civilian applications [6]. 1. The NavyModel The hazardous material substitution algorithm is sufficiently flexible that it can either serve as a preliminary screeningto identify the most likely candidatefor further study or become a part of a much more detailed and sophisticated decision model. In the first instance, the model would be useful to an industrial or commercial concern or to a military installation comparing materials proposedby vendors as substitutesfor existing materials not conforming to regulatory requirements. The second application would be for screening as part of an in-depth decision process for changes to production operations, or comparison of newly synthesized materials for possible large-scale application throughout a major industry. For maximum utility, the Navy model is adaptable to either simple manual computations or computer applications.

2. Description of the Algorithm The algorithmis used to assign numerical “points” for various hazard descriptor elements such as toxicity, duration of personnel exposures, number of persons exposed, related medical effects, fire and explosion potential, requirements for personal protective equipment, anda limited assessmentof environmental impact and control requirements. These latter include volatile

22

Meyer

organic chemicals, EPA Reportable Quantities, and EPA’s List of Lists materials (40 CFR 302,4) among others. The “points” assigned are totals that providea numerical score anda risk assessment code (RAC). This provides for determination of a hazardous material selection factor (HMSF), which allows for materials to be compared with one anotherin numerical terms. The resultcan then be used as an entry point into the foregoing overall decision process. The input data are readily available. Principal among these are the Material Safety Data Sheets (MSDS) required by 29 CFR 1910.1000, OSHAPermissible Exposure Limits (Table 2 29 CFR 1910.100), and EPA Publication 56014-90-011“Title 111 List of Lists.” The RAC procedure is based on a commonly used system safety analysis method (MIL-STD-882), and the basic approach to the “point” algorithm is the previously described DoD system for rating occupational, safety, and health hazards.

B. Understanding the Basis for the “Points” The following brief information providesan understanding of the basis for selecting the range of numerical values for the algorithm’s points. 1.

Toxic Effects The evaluation should include the frequency and duration of possible worker exposure. This includes whether the material presents toxic hazardson brief, short-term exposures associated with high concentrations and accidental releases or primarily causes harm from extended exposure to relatively low concentrations. Materials that are irritants skin or sensitizers or that are suspect or known carcinogens, teratogens, or mutagens require special attention even if the projected quantities are small. In many instances, the MSDS will only summarize the toxicity data of the individual components of the mixture andwill not provide information concerning specific toxicological studies on the material itself. In such cases, judgments will have to be based on consultation with also must be given such approved sourcesas the Navy Environmental Health Center. Attention to any information indicating that the material is a known skin sensitizer or possesses allergenic is the National Institute of properties. A suggested source of reference regarding toxic hazards Occupational Safety and Health (NIOSH) Pocket Guide to Chemical Hazards available from the US. Government Printing Office. 2. Characteristics Physical Characteristics. Materials with a high vapor pressure are more likely to be easily dispersed into the environment than those with lower vapor pressures. Those with low flash point and low boiling point (flash point lower than 73°F and boiling point below 100°F) are extremely hazardous froma fire and explosion viewpoint compared with those with flash points greater than 100°F. Liquids with specific gravities less than 1.O present fire-spreading hazards because such materials float on water. A “toxic material” with a high vapor pressure is more of a hazard in a confined work area than one with the same toxicproperties but a much lower vapor pressure. This is because the higher vapor pressure will afford a greater risk of room atmospheric contamination. ChemicalCharacteristics. Wheremixtures are involved, it is importanttounderstandthat those that include aromatic organic chemicals are generally more toxic (and often pose greater fire and explosion hazards) than those classed as aliphatic chemicals. Among the chemical characteristics that must be considered are stability, reactivity with other chemicals (forexample, is the material an oxidizer or corrosive?), and solubility, not only in water but in other media.

Least

23

Circumstancesof Exposure. In addition to the specifics of probable work areas, questions on the distribution of material throughout theweapon system life cycle or on-shore activity need to be considered. Localized use (in a single work area) of a material determined to be highly hazardous presentsa different setof concerns with respectto approval decisions than those that apply to a material with moderate hazard potential that is widely used. Among the considerations that should be examined are size of the work force or number of persons at a work site, present and/or needed engineering or other controls, and work area environmental conditions that affectthe hazard (temperature, humidity, the presence of other chemicalsthat may be synergistic or additive, etc.). During a general review and evaluation of a proposed material, questions need to be examined with respect to the interactionof the proposed material with others already approved and its use in the system or work areas and with nearby operations. For example, it would be a mistake to approvea new cleaning solvent with a high vapor pressure and low flash point for use in shops in which arc welding is conducted. EnvironmentalImplications. The potential for hazardous waste (HW) generation and compliance with various federal, state, and local codes, standards, and regulations must be evaluated. Insomegeographical areas, regulationson useand/or release of volatileorganic compound air pollutants are very severe and may require special controls if a material is approved. Similar concerns must be examined with regard to air quality and water permits. Because of the, widevariety ofsuch requirements,the “points” used in thismethod are simplified. More detailed ratings may have to be developed by the user for some analyses.

V. CONCLUSIONS AS TO UTILITY OF THE METHODOLOGY The hazardous material substitution algorithm developed for the U.S. Navy has been tested extensively and found to be a useful first screening tool.It also fills the need for a wide variety of applications in the civil sector. As indicated earlier in this chapter, itis only one element of the decision process. It is also essential to note that although one goalmay be the elimination of hazardous materials that affect people or the environment, in many instances complete elimination is not feasible. The selection method, and other considerationsin the decision process, provide fora rational and cost-effective determinationof the most suitable material. As stated by EPA (Pollution Prevention 1991, EPA 21P-3003) and the Pollution Prevention Act of 1990, when pollution cannot be prevented, reduced at the source, or recycled, it “should be treated in an environmentally safe manner . . . and disposal or other release to the environment should be employed only as a last resort and should be conducted in an environmentally safe manner.”

ACKNOWLEDGMENTS This chapter is based in part on AFMA-TR-91001, Development of Guidance for Selection/ Substitution of Less Hazardous Materials, for the U.S. Naval Supply Systems Command, underUSAFcontractF3361589-D-4003,Order16,A. F. MeyerandAssociates,Inc.with Engineering-Science, Inc. Publication rights to this chapter are retained by the U.S. Government. Copies of the basic technical report canbe obtained from Defense Technical Information Center.

REFERENCES 1. Patty, F. T.,IndustrialHygieneToxicology, Vol. 1 , Inter-SciencePublishing Co., Chicago, 1948. 2. Meyer, A. F., Jr., Engineeringbiotechnologyinoccupationalhealth, Trans. Am. Soc. Civil Eng., 121: Paper No. 2798 (1956).

24

U.S. Department of Defense, Deparfment of Defense Occupational Safety and Health (OSH) Programs, DODI 6055.1A. 9 Sep 87. 4. Sampaolo, A., and Binetti. R., Regulatory Toxicology and Pharmacology, Vol. 6, 1986. 5. U.S. EnvironmentalProtectionAgency, Pollution Benefits Manual, October1989. 6. U.S. Navy, Naval Supply Systems Command, Development of guidance for selection substitution of less hazardous materials, Tech. Rep. AFMA-TR-91001. 1992. 3.

~

Meyer

3

Multiple Approaches to Environmental Decisions

Douglas M. Brown The Logistics Management Institute Bethesda, Maryland

1.

THE IMPORTANCE OF DECISION MAKING

It would be difficult to overstate the importance of the environment as a policy issue. Aside from the ecological implications of decisions in many “nonenvironmental” policy fields, environmental policies have impacts on other policyfields. The recent controversy over whether jobs of timber protecting the spotted owl should weigh more or less heavily than protecting the industry workers is not going to be solved here. The important thing is to realize that environmental policies, often considered to be based on scientific analysis, must include consideration of nonscientific issues such as fiscal realities, economic growth policies, and cultural values. Even race has surfaced as an issue in this field [l]. Because of the weakness of the current state of the art in fundamental measurability of environmental policies, an appreciation of the impact of such policies can only be hintedat by using other proxy measures. While environmentalactivists prefer to see environmental protection as a universally superior good not subject to such comparisons,the fact is that protective activities incur costs. Whether theyare continuing expenses or just investments that will result in lower costs later on is a matter of interest, but it does not relieve societyof the obligationto pay the bills until the investments mature. In the end, all policy costs are experienced by society’s consumers and taxpayers. Individual firms, of course, can be punished with criminal sanctions or forced into bankruptcy over environmental breaches; but as a rule governments and entire sectors of industry simply pass the costs along in the form of coerced tax hikes or industry-wide price hikes. Thus, neither government or industry (in a wide sense) “pays” for environmental protection except to the extent that when consumers or taxpayers find themselves with no more money to spend, the popular taste for government or the industrial product may evaporate. Those who believed thatthe 1970s’ Great Society programsor the 1980s’ defense buildups nearly achieved this national policy bankruptcy point should look closely at the environment. The cost to the taxpayer of dealing with environmental problems is expected to exceed 6%of 25

26

Brown

the Gross National Productby mid-decade. That is the size of the entire defense budget during the peak of the Reagan buildup. Simply cleaningup known hazardous waste sitesat federally owned facilities is expected to cost over$125 billion [2], with the figurebeing adjusted upwards every year. Newly recognized threats are being discussed that will only add to the potential size of this burden of this burden[3,4]. In addition, consumers will bear anadditional burden. Some is hidden in the priceof consumer goods, as those private-sector firms that continue in business rather than declaring bankruptcy either pay for required cleanups, self-insure against the need to clean up in the future, or develop new processes to avoid becoming a party to a future cleanup. In some cases, businesses will choose to go out of business rather than risk personal or corporate liabilities of staggering proportions. At the least, this will reduce the number of choices available to consumers, and at the worst, employees will be thrown out of work and further impact on other taxpayers. All in all, there are compelling fiscal and social reasons for ensuring that our very real environmental problemsare identified and dealt with in prudent and responsible ways. Money spent on the environmenteither directly by governments on behalf of consumersor indirectly by consumers through higher prices chargedby producers as a result of regulations, cannot be spent on other worthy causes such as consumer and national savings and debt reduction, urban issues, transportation networks, and national security: whatever your policy preferences are, environmental spending competes with it. And, if not properly thought out, environmental policy can compete with itself. For instance, the EPA has spent years convincing the public that toxic wastesites are a tremendous sourceof health risks and must be dealt with promptly whatever the cost. The publication of the Unfinished Business report [4] requires the EPA to reeducate the public that in its new view many other threats are more risky; EPA competes for funding for those higher risks against an established, costly effort that the EPA itself established and plansto continue. A micro version of the same argument canbe made at the level of the individual producer facility. Statutory responsibility or not, an organization cannot devoteso many resources to environmental protection that it can no longer afford to remain in business. Environmental actions, even where deemed socially worthy, must compete for funding with other programs, and where the available funding does not cover all perceived needs, then environmental spending itself must be prioritized. In short, for regulator, policy analyst, and facility manager,a sound basis for making environmental decisionsis essential to the development and effective execution of a holistic, complex, and credible program for the protection of health and resources. While some may argue for a policy based strictly on scientific evidence, others argue for environmental policies based onemotion, and yet others argue thatthe costs of delay on the one hand and regulation on the other are socially destructive, environmental managers are faced with a situation where something hasto be done that will satisfy all sides without bankruptcy. Thus, while decision theoryis not the cornerstone of environmental science, it may well be the keystone of environmental management.

II. ENVIRONMENTALROLES Environmental threatsare produced and dealt with by organizations whose missionsare broader than simply protectionof the environment. Environmental agencies, however, havethe mission of ensuring that producers do not forget their environmental responsibilities. Those responsibilities are to the third player in this process: the public. Through the political process, the public caused an environmental policy to be put in place to protect healthand the environment, and at the grass-roots level the public maintains oversight onthe specific actions of both reg-

Management 27

Environmental

Table 1 PolicyRoles Characteristic PolicyviewpointSinglefocus on narrow portion of environmental issues. Mission pollution Prevent all and punish all polluters. Objectives Focus resources worst on Resources Cleanup

Approach Policy effect

threat. costs do not detract from primary mission. Extract payment for pollution. Deters, does not repair, problem.

Complete operation with environment as part of the whole. Produce products and pay for environmental protection. Minimize resource drain. Cleanup costs are taken from funds otherwise available for mission. Focus on avoidance, then cleanup. Avoids, evades, or repairs problem.

Does not generally under-

stand or get involved.

Seeks products and protection. No perceived threat is acceptable. Pay in either case, through taxes or prices. Passivity until aroused; then paranoia. Suffers consequences.

ulators and producers. The differencesin the roles of regulators, producers, and the publicare summarized in Table 1. The most familiar enforcement agency is a police department. It has a single focus on a statutory area (in thiscase, public order). The primary responsibilitiesare (preferably) to deter criminals, which itself deters further crime or, when that fails, toseekoutandapprehend crimes. In addition to fines levied through the punishment process, such departments may collect fees to recoup their costs of doing business, thereby reducing the burden on their budget (and, in theory, on the taxpayer). Frequently, however, there is no payor, either because the guilty party hasnot been apprehended or because the enormityof the crime makes financial restitution, even with damages, unacceptable or so high as to be unpayable. Such cases, which are the norm more than the exception, make it necessary for the department to absorb the cost of enforcement; but such expenses are budgeted for and appropriated over and abovethe cost of normal operations, not at the expense of those operations. We expect them to deter and apprehend, not to repair, the problem of crime. In those few jurisdictions where victim compensation is considered, it does not come out of the police operating budget. It is not generally expected that the enforcer will have to police itself (although such occasions do arise, and have arisen in most of our major cities over the past 15 years, and are generally poorly handled). In some jurisdictions, we see a cooperative enforcement approach [5]: “community policing” or the “cop on the beat,” tomatch the environmental metaphor witha police example. Nonetheless, thosecooperative approaches are part of an overall enforcementstrategy; the regulator coaches the producer toward compliance rather than taking over the operating responsibility itself. Finally,we expect the police to focus on the worst problem-catching murderers rather than staking out shoplifters. The producers have a completely different set of responsibilities, the foremost of which is the factthat they must continue on with their production task in order to survive; environmental issues are a secondary concern. When a cleanup does become necessary, the producer must pay, and its payment comes out of its normal operating expenses. To some degree, consumers will absorb someof this cost, but in general, passing on too much of the cost will simply drive

28

Brown

the producer out of business (unless the producer happens to be a government agency). With limited discretionary funding available for environmentalrestoration, then, producers need to accomplish as much as possible with the resources they have. Normally, government regulatory agencies act as enforcers. Once enforcement has occurred, however, one is faced with the need to restore the situation. Then, and especially in the case of the Superfund, the government becomesa “producer” and needs to act and think like one. Finally, there is the public. The public tends to be easily excited over health and safety issues, although a much smaller (but more active) group maintains vigilance overnon-human health and natural resources, anda very small groupis both active in and knowledgeable about global ecology issues. Inaddition, the public is concerned aboutjobs, general economic issues, property values, and the quality of life in communities. Thus, the public concerns tend to be more diffuse than the single focus enjoyed by enforcers and producers. Because of that diffuseness, the public seldom speaks with a coherent voice, which makes it easier for activists and extremists on all sides of an issue to misrepresent or override the public will. While the federal government’s National Environmental PolicyAct provides processes for public involvement, as do a number of state statutes, there is at present no real requirementto go along with public preferences as long as the pro forma requirementsare met. Thus, on any given decision, the public can be and often is ignored. The more this happens, of course, the more the public comes to see the regulator as well as the producers as its enemy, especially as these parties will be on different sides at different times. Given this disparity in roles, it may come as no surprise that there are different perspectives on what the general objectives of the environmental effort shouldbe. Ihave identified seven primary goals; others may exist. While most of them appear desirable, at least in isolation, they are not all consistent. They are presented here in no particular order; indeed, the ordering process itself is one of the most important facets of the environmental decisionmaking process. Risk. Eliminate risks to human health and the environment. Cost. Engage in environmental projects that achieve organizational objectives without posing an unacceptable risk to the organization’s economic competitivenessor viability. Time. Accomplish objectives rapidly, if in fact there are risks to the environment and especially to our health. Acceptability. Satisfy publicandorganizationalexpectations. All environmentaldecisions, whether takenby public or private organizations, occur underthe observation of a political structure and still remain consistent with the organization’s own value system.’ Deterrence. Ensure that environmental offenders, and particularly the worst offenders, are caught and prosecuted. Administration. Minimize the debate over the intent and application of environmental laws and regulations. Practicality. Develop policy alternatives that are executable, compromising the ideal to maximize what can be accomplished. If the budget is unlimited, decision making or prioritization is unnecessary: one Simply does everything that is wanted as soon as it is feasible to do so. However, increasing sophistication in regulations and increasing effectiveness of environmental compliance efforts are ‘This objective was titled“Politics” in earlier work [l]. Some managers objectto the idea that “politics” intrudes on the pristine pursuit of public and ecological health but feel quite comfortable with the need to tolerate public value systems. “Acceptability” seems less threatening.

ndition

29

Environmental Management

Table 2 DecisionApproaches Knowledge Multiple-rule solution; (one-rule) Universal Certainty solutions (e.g., cost-benefit) Statistical uncertainty Static Dynamic Outcome uncertainty

Expected value Simulation Aspiration level Subjective decisions Minimax

uncertainty Total allocation Political

combining to produce requirements estimates for environmental work that are increasing exponentially. For federal agencies alone, workingoff existing project requirementsand dealing with a number of pending interagency agreements with the EPA or stateagencies could require three- to fivefold increasesin current environmental spending, even assuming that the historical inflation of environmental costs can be controlled. But as soon as we say that some projects expectations, the question of priorities will have tobe delayed to meet more reasonable funding emerges: which projects should be delayed, and why? In short, under reasonable resource conditions, not all of these objectives can be accomA total focus modated in full simultaneously. A blind emphasis on speed usually wastes money. on scientifically assessed risk may be impractical if the necessary science is incomplete. Administrative simplicity may be translated into rigidity, resulting in the carrying out of regulations blindly, resulting in large costs with no appreciable improvement in the environment. And so on. It is in just such cases-multiple, conflicting objectives-that the use of decision methodologies is needed.

111.

DECISION-MAKINGMETHODOLOGIES

This chapter is not intended as a single referencefor explaining the details of decision theory. Such textsare available-indeed required-in every college’s management course work.’This chapter proposes rather to explain why decision analysis is needed in environmental decisions as much as any other. In general, decision-making methodologiesare selected based on the degree of uncertainty surrounding a condition or situation; Table 2 shows that alignment.

A. Certainty Under the condition of certainty, we know eachof the outcomes; it is simply a matter of choosing the programthat benefits us the most. In that case, there is a universally superior solution or decision rule. Even where such a rule is deemed to exist, it must be tested to determine its universality and the existence of underlying rule structures, which may require reversion to a more complex decision approach. Forinstance, the accepted rule may be that projects that reduce the most human health risks will take priority over all others.The real world presents us %ere are too many such texts to list, and each university has its own preferred texts. Each manager has, or should have, several. On my bookshelf, perhaps the most frequently used is Fleischer’s Engineering Economy [6]. Others include Lapin’s Quanrirarive Methods for Business Decisions [71, Quade’s Analysis for Public Decisions [g], and Douglas’s Managerial Economics [9].

30

Brown

with resource constraints, not necessarily limited to constraints on ready operating funds: the total capacity of administrative, technical, industrial,capital, and labor resources formsan effective obstacle to unlimited activity levels. Given a clear goal and normal capacity constraints, the basic rule can be expected to meet a quick challenge: if two projects are equally effective in risk reduction but there are not enough resources to accomplish both, anothercriterion must be applied as a tie-breaker. And, as a result of limited resources, at some point the residual funding may be adequate only for a low-cost but also low-priority project while a higher priority but more expensive activity must be forgone. Given a high degree of certainty, we know all potential cases and all potential outcomes (and the relationships between the two). Then rule-based decisions are the appropriate solution to ensure that the best possible outcome is achieved from each possible situation. The decision makers can incorporate multiple factors simply by adding more complex rules, but in a rule-based system the rules cannot be waived.Norshould there be anyreasonto do so, because all possible cases and outcomes can be predicted and the best course of action identified. Generally, we do not have such perfect information. Nonethelessmany environmental decisions are based entirely on methods that address certainty. This occurs for two reasons: the real information is unknown or the real information is too complex to be usedin its full detail. When the real information is unknown, decision makers should apply appropriate decision approaches under uncertainty, suchas those displayed in Table2. However, environmental decisions are frequently made by bureaucracies (governmental organizationsor large industries) that do not subscribe to subjectivity or political acceptability as an explanation for how decisions were reached [lo]. Thus, where the real information is too complex to be analyzed effectively (as is often the case), managerssimplifyittoits essentials [ll]. If a modewill describe the behavior of a natural phenomenon adequately(e.g., saying thatthe prevailing wind is at 5 mph from the east, when in fact over the courseof a year it blows from mostdirections at various speedsfor some periodof time), and if the resulting policiesdo not appear to be too badly flawed, the approach is validated in terms of the value of the time and effort requiredof the manager. Indeed, contrast this behavior with the warning to remain consciousof the value of perfect information issued later in this chapter. Again, however, true certainty seldom prevails.And in acting “as if,” we are simply making assumptions: takinga mode to be the only possible outcomerather than the most frequent one. The real situation may take oneof four alternative forms, representing increasing degrees of uncertainty. Statistical uncertainty means that wehave a good idea of what might happen and that we have some probabilistic statements of how, when, or how often. Technically speaking, this is called decision making under conditions of risk, but because of the extensive use of the word “risk” in the environmental sense of a threat, I have called it statistical uncertainty instead.

B. StatisticalUncertainty When statistical uncertainty exists, we must use our statistical knowledge to make decisions “as if” the outcome were to be as predicted by the probabilities, with appropriate consideration for the fact that it might not. This is basically done using an expected value approach, which some of you may know as a weighted average approach. Rule-based approaches become meaningless ifwehave uncertainty, because we do not know what situation exists or which rule should be applied. Despite the masses of data collected to support science’s continuing assault on the mysteries of the ecosystem, we remain extremely uncertain in our understanding both of the system as a whole and of most of our

Environmental

Management 31

individual “facts.” There is actually nothing wrong with an uncertain situation; in fact, our economic system and our national security systems function tolerably well under admitted uncertainty. The Apollo missions reached themoon and returned safely under conditions of uncertainty. Decisions under uncertainty are quite possible: the recognition of uncertainty simply requires some adjustments to our decision pattern. Indeed, the bulk of decision theories address this uncertainty directly, either in factoring in the possibility that multiple outcomes may occur (through weighting) or through safety factors. The more serious error is to conduct business under uncertain conditions as if there were no uncertainty. Unfortunately, in many cases, environmental policies have been devised and implemented in exactly that manner.For instance, one of the primary“rules” that dominated implementation of the Superfund program was the “worst-first”rule, under which priority of effort was givento the most hazardous known contaminated sites. Thisrule was implemented through a detailed process resulting in a numeric score that was deemed at the time to be somehow a measure of the site’s risk. As detailed site investigations were pursued, it began to appear that in many cases the scores did not reflect risk at all [12]. Part of the reason for this is that so that the worst are probably not first[13]. the scoring system used is mathematically skewed Subsequently, the EPA issued a revision of its scoring systems, but EPA is refusing to rescore the sites now found on the National Priorities List. Although the reasonsfor that decision are more related to politics and face saving, it forms a good example of the fact that under uncertainty the desired or statistically most likely outcome may not in fact occur. In addition to using expected value methodologies to overcome statistical uncertainty problems, the technique for calculating the valueof perfect information is one with which all environmental decision makers should be familiar (at least in concept). The specific equation is not particularly relevant, because there is often no “variance” (one of the terms in the perfect information equation) in environmental data, which tend to be highly site-specific.But a great portion of the funds in environmental activities are expended on “just one more round of testing,” a round that generally proves as inconclusive as the original round. There are a numberof activities that are worthwhileof themselves but very much subject to the question of perfect information. Facility managers are asked to comb large tracts of land looking for endangered species that might be there but often are not. Products are banned or consigned to expensive disposal programs because they might be dangerous. Although protective measures are laudatory, all environmental professionals (whether they are regulators, fato be conscious of the fact thatat some cility managers,or taking the public’s perspective) need point enough is enough. Part of any action discussion should be an explicit understanding of what “enough” is, what (specifically) is expectedto be gained by achieving “enough” as opposed to some lower level of effort, and what the costs and other implications of getting to “enough” will be.

C. DynamicUncertainty Many managerial texts address uncertainty as if it remained constant, albeit unknown. In the environmental world, things do not remain the same. Technology advances or is discovered to be ineffective, regulatory requirements change, the ecosystem changes, and specific activities evolve. Additionally, many decision theory models assume a single decision, even if a protracted and complex one. Environmental compliance activities often do not fit this mold. They occur within a continually evolving process where the outcomeat any point is dictated As a result, more complex modonly inpart by the objective facts of the original circumstance. els than the basic decision tree are needed to deal with the statistical uncertainty facedby environmental managers.

32

Brawn

As an example, compliance requirementsfor wastewater treatment facilities have changed only superficially over the past two decades. Therefore, techniques of static statistical analysis are useful when looking at a large population(the view enjoyed by the EPA of a large number of wastewater plants, for instance) because the baseline for sucha view has been well established and changes slowly. And EPA-level decisions, because of the mission of regulation and enforcement, addresshow to regulate entire populations andhow to scan those populations to identify recalcitrants. For a single facility, the domainof most environmental managers, this aggregate information is of little value. Decisions are not made on how to comply; that has already been spelled out. Generally, they are not made on whether to comply. The question is how to maintain the operation in compliance with no deficiencies, or at least as few as are practical, in the face of any number of unpredictable events. The ability to comply with regulatory requirementsis only partially a result of the original adequacy of a plant’s technical design (the only static aspect of the operation). Much more important are a myriad of specificcircumstances arising, or not arising, onanygiven day. What is needed is a model of the entire system in operation, from water flow production to effluent disposal, so that each point wherea problem might occur can be identified and dealt with. Even wherea violation may occur, itdoes not follow that an enforcement action will result. That too isa dynamic process, dependentin part on theplant’s record from the past, on teambuilding efforts by the regulated facilitystaff, on the sheer coincidence of inspection scheduling on days when the facility does or does not suffer a reversal of fortune, and on whether or not the facility has implemented anyof the improvements recommendedby the regulators on their previous visits. This situation, in which the probabilistic variables themselves vary over time, can only be of a scenario over multiple addressed by simulation tools. Such tools can represent the running iterations to represent the effects of the passage of time or to try out the effect of assigning different probabilities to variables.Thus, in addition to the decision trees and contingency tables often found in managerial textbooks, simulation techniques must be employed. Generally, managerial texts restrict their discussion (if any) of simulations to the Monte Carlo technique. This does providea very powerful tool. However, even in its most basic form it requires recomputationof known equations in a number of iterations, which effectivelydemands automated tools. As simulationsare being applied to increasingly challenging problems in industry and commerce, the computational power of simulations is being enhanced with graphics to provide comprehensibility to problems and solutions thatwould otherwise be nothing but piles of computer printout. For environmental purposes,which tend to address problemsan order of magnitude more complex than industrial process modeling, dynamicsimulations that incorporate visual effects are needed both to complete a reasonably accurate representation and to enable the functional manager to see what the computer is trying to communicate. Itis also important to understand that elaborate graphical presentations, even though they may be very data-intensive, are not the useful to a rational decision processunless they canbe used to develop relationships among data. High-end geographic information systems generally display informationin multiple layers, making it accessible to intuitive analysis. They may provide a database management system that permits the display of selected information, but usually they have no capacity for mathematical analysis. Simulation tools are very rarely found in environmental use, despitetheir obvious utility. One reason is the complexity of environmental issues,which often forcesa long tool develop-

Environmental

Management 33

ment time. Also, the policy system under which simulations are developed (asmany withother areas of environmental activity) allows for endless researchand refinement rather than product fielding. Another reason is that a circular logic is in place. There is understandable reluctance at the EPA to certify any software products that have not been thoroughly validated by the EPA. However, the experience of many producers with the EPA’s unwillingness to accept anyway but the EPA way has made them wary of attempting to present EPA with any new products or approaches, and warier still of investing in the creation of new suchways until the EPA approves the product for general use. As a result, managerial innovation is restricted. The final reason is that formany environmental managers on the job today, the computer remains a fearsome tool. But the very complexity of environmental issues and the increasing cost of environmental solutions will soon make an effective decision-making process dependent on the ability to exploit the power of automation. More familiarity with the use, applicaof environtions, and abuse of computers in general and simulations in particular is required mental professionals.

D. OutcomeUncertainty Outcome uncertaintyis technical talk that means basically we have no idea of whether, how, or when things might happen, although we may be able to make some guesses as to what the things that might happen are. Under outcome uncertainty, we are forced into more subjective decision-making processes. The “aspiration level” approach [6] is seldom acknowledged but frequently seen. In such an approach, not having confidencein predictions of what may happen, and finding the risk unacceptable, we decide t6 guard against the worst possible event. U.S. defensestrategy:anuclearattack by theformerSovietUnion Thisislikethe wasalwaysconsideredextremelyunlikely,butthepotentialdamagethatsuchanattack could cause was so great that we spent enormous sums protecting ourselvesas best we could from such an event, even though we knew that by doing so many other needs would simply have to go unaddressed. The present worst-first policy is also an aspiration-level policy: it assumes that we know little about costs and remedies of cleanups, but we believe that we have identified the worst situations and we are committed to removing those situations, beginning with the worst. Even under such conditions, however, cost-’benefit considerations are at work, although less obviously; they are squeezed in through the back door using the potential for public outrage as the vehicle. We accepted the cost of the defense programas necessary in conceptand generally affordable. And our national policy is to save endangered species. However, where a human community must give up its current livelihood in order to save a species (as is threatened in the effort to preserve the spotted owl habitat), the regulation enters the political arena where the EPA may win or it may lose. The contest willbe presented to the public on the one hand as preservation of the quality of human life in preference to unproven allegations of harm to what is onlyone of millions of species, and on the other hand as preservation of defenseless creatures against greedand callousness. The essence of this argument is the cost-effectiveness of the effort: is the public willing topay the price for a particular environmental project? Another methodof dealing with uncertainty is subjective decision making using group processes. There are any number of approaches available, from public meetings and roundtables to expert opinions eitheras individual contributionsor controlled through a Delphic process.An example of the weighting of subjective preferences is seen in the U.S. Department of Energy (DOE) approachto its overall capital facilities investment strategy[14]. Successful results (as

34

Brown

proven by the subsequent accuracy of the predictions) have been experienced with using an even more structured approach known as the analytical hierarchy process (for instance, our work on community fiscal impact analysis [15]). TheEPA has begun looking into that process in its negotiations withthe U.S. Department of Energy over howDOEwill conduct its cleanup operations [16]. The process in a nutshell is that stakeholders and/or experts come to agreement over the relative significance of factors taken two at a time;at the end of making many such comparisons, the relative significance of each of the factors can be mathematically arrayed into a numerical weight table. Another perfectly valid approach under uncertainty is problem avoidance, an extreme form of the minimax principle in which we minimize the maximum cost without much concern for probable benefits. If the issues and the costs are imperfectly defined, or if the impact of the events (including enforcement) is seen as unlikely or highly arbitrary, the whole system may not be worth worrying about. It is probably cheaper to be dragged intocourt from timeto time precisely the strategy than to go around solving problems that may not exist. In our view,is this in place among many producers today. We have named it “problem avoidance,” although one could also characterize it as foot dragging or passive resistance.

E.TotalUncertainty The final possible approach, under total uncertainty (if we have little confidence in the risk data or the cost data), is to simply allocate the available funds on some arbitrary or politically acceptable basis and hope for the best. That division may occur on a social basis or on a geographical basis,and there are some suggestions that this is exactly what is occurring, either deliberately or as a consequence of the problem avoidance strategy. However, using similar arguments to achieve the opposite result (as in the present enthusiasm over racial equity in environmental issues) is analytically no more pure than the original failure. The approach is not covered as fully as others in this text because such issues are not amenable to environmental management and because there is some evidence that, at least in the Superfund probe gram, fundsdo flow to higher scoringsites [17]. However, the environmental manager must aware that such a schoolof thought exists and that when skillfully used it can awaken powerful political pressures.

W. HOW CERTAIN IS ENVIRONMENTAL SCIENCE? The range of approaches notedin Section I11 varies according to the amount of certainty that is attached to the situation. Clearly, it varies or there would be no need for so many approaches. In this section, we briefly consider some of the issues that lend uncertainty tomany cases. to considerable controversy. Each of the major objectives to be accomplished is still subject Research needs to be applied in each of those areas if we are to move forward effectively in managing the environment andto move from the imaginary certaintyof acting “as if” (when really acting under outcome uncertainty if not total uncertainty) to at least acting under statistical Uncertainty. The following brief summary is provided hereto allow the environmental manager a glimpse of all that we do not know.

A. Risk The definition of chemicals as hazardous is done through a series of separate environmental regulations, and there is little comparability of threat. In many cases, two chemicals have overlapping effects levels: the question becomes whether it is worse to die immediatelyor to get

Environmental

Management35

lung cancer. In such cases, how do we establish “worse”? Even when we stick with a single chemical class, there is a great deal of doubt as to what risk, if any, a regulated material poses. While the chemical data are confusing and contradictory (see, e.g., reference 18, from which the original Hazard Ranking System (HRS) factors were derived,or reference 19), we cannot always rely on the accepted approximation systems: current measurement systems (the original and revised HRS) cannot be shown to approximate risk. In our earlier paper [13], we showed that the original HRS was mathematically inconsistent, leading notto the potential but to the actuality of score inversion. While the’revised system corrected several glaring deficiencies of the original system, there are enough departures from the generally accepted threat-pathwaydose-receptor paradigm to raise doubts as to whether risk is related to cost under the revised scoring scheme.

B. Cost Given a reasonably accurate description of the problem, a reasonably accurate cost estimate A common criticismof current modshould be possible for the engineering cost of the remedy. els is that they are usually off by 50% or more and usually understate the final cost. This is partly because of the lackof data points, partly a resultof inaccurate input, and partly dueto the continuing ballooning of legal intervention costs. However, it should be noted that a 50% error may be less than the current errors in risk estimation.

C. Time Time has interesting impacts on environmental decision making thatto need be explored. Time, for instance, is tied to risk issues. Givenlow risks now and high long-term risks, or moderate transient risks now, how do we choose? What about the tendencyof pollution problems to expand over time? Time has extensive impacts on the cost side of our decision model. What does the slope of the environmental learning curve look like? What is the real rate of environmental inflation, and can it be separated into its components: engineering, legal, procedural, and so on? Even more significantly, what would be the most appropriate way to integrate the concerns into a decision model, even if the answers were known?

D. Acceptability A great deal of work remains to be done in the integration of legitimate political and public concerns into the decision process. If political distribution of benefits is a real show-stopping issue, how can that be integrated into a decision process in an open manner so that it can be assigned a proportionate role?Is the distribution of pollution problems consistent with industrialization and hence with population density, and does this ensure an acceptably proportionate share in the program?How do we distinguish between self-interested “NIMBY-ism” and valid public safety issues, or does it not really matter?

E. Deterrence Generally, deterrence is achieved through a combinationof other objectives: effective definition of the risk factors, approaches that maximize the benefits of compliance, and a smooth A great deal of research is going administrative process that makes apprehension more likely. on today on the question of the right amount of fees to charge polluters. Less research addresses and what the costs of checking the questionof how to make producers stop polluting altogether on it are.

36

Brown

F. Administration Research into the other bureaucratic goal-administrative smoothness-would be well served. The primary attraction of the “as-if‘’ approaches are that they are very easy to administer. HRS, for instance,is self-operated by polluters or agency contractors;sites simply go on a list. No effort has to be expended by the agency on the messy questions of cost, time, risk assessment, and so on; all of that is up to the polluter. From the producers* perspective, problem avoidance has advantages administratively because one hasto fund only whatever the agency requires, so there is little need for decision making and a good probability that the major exof time. But the primary pense of restoration canbe spread out overan almost indefinite period reason for such behavior is that the process required to do anything else is so convoluted that it is not worththe cost. The simple actof declaring that a site should not be listed as dangerous becomes a very expensive process. Research is needed to address the valueof each of the existing steps of the process and todetermine whether it is possible to design a less cumbersome administrative process with less reliance litigation on that still protectsthe interestsof the agencies and the producers.

G. Practicalities Practical problems include the capacityof the environmental industry to handlethe workload, to handle projects above a the availability of technology, the capability of project managers certain size and speed, and the limited experience at all levels with actually executing many types of environmental activities. of today’s feasible actions. Are There are plenty of other practical issues within the scope existing contract vehicles adequate? To what standards should contractors be held, and how effectively can they be managed? What are the lessons learned from past programs, andhow are they being disseminated? Answersto any of these questions would improve the process of planning and executing environmental cleanup.

V. DECISION APPROACHES

AND ENVIRONMENTAL OBJECTIVES

We noted earlier that there are several possible objectives for environmental programs and the people and organizations charged with carrying out those programs.How do the approaches we have discussed meetthose objectives? Table 3 compares the major approaches in terms of the objectives of the environmental program. The table shows a plus sign where the approach enhances that objective, a minus sign where it detracts from an objective, and a zero where it has no effect. The single-rule approach (exemplifiedby the Superfund worst-firstrule) addresses the risk and acceptability issues by focusing onthe worst cases. In ignoring certain other objectives, the single-rule case in particular engenders a number of practical problems. Most notably, ignoring cost issues results in a number of cases of self-protective evasion by responsible parties. In addition, an oversimplified approach ignoresmany other practical limitations, such as the capacity of the environmental industry to undertake many projects or the availability of reliable technology with which to do the work. And it ignores the value of time in terms of inflation, opportunity costs, and continued public exposure to hazards. More recent policies (if not many efforts) have focused on maximizing risk reduction[20], a multiple-rule policy based on both risk and cost. Cost as a consideration at least forces inestimates will need direct considerationof some unspecified issues. Effectively performed cost to consider practicality andthe value of time, as a minimum. Such policies will necessarily be

Management 37

Environmental Table 3 Decisions and EnvironmentalObjectives Static

Multiple (expected Aspiration Single Subjective mle rule Risk reduction

cost Time Acceptability Deterrence Administration

Practicalities

+ -

+

+ +

-

value)

+ + + 0 + + +

Simulation weighting level

+ + +

+ + + + + 0 +

0 0

-

+

Maximin Politics

+-

+ + +-

-

-

0

+-

0

+ 0

-

+

0

0

+ +

+

0

-

-

Key: +, enhances: - deqacts; 0, no impact.

more complexto administer and may, because of their complexity, be less acceptable. Indeed, in the chart shown in Table 3, the multiple-rule-based approach appears almost ideal, with many pluses and no minuses. But this assumes that the rules can be developed in such a way as to enhance the objective, in short, that we are operating with certainty, which as we have noted above is usually not the case at all. Operating under statistical uncertainty,we find that a more detailed understanding of the variables should result in a more targeted approach to the problemto be solved; thus risk, cost, time, and practicality are well addressed. The added complexity of the approach makes it less acceptable until it proves its worth. Likewise, deterrence may be stronger after the approach has been proved but will initially be poorer than a straightforward policy until it proves effective. And of course such a policy will always be complicated to administer. Introducing the dynamic aspect of uncertaintywould be a policy disaster without the appropriate tools. While an academic case might be made for the power of formal automated modeling and simulation in enhancing almost every objective except administration, the lack of appeal of a massive computer printout (often the only form of output available) would make it untenable to either politicians or bureau executives. However, given the right tool (a powerful simulation engine with an attractive graphic interface), this method becomes very useful. The scoring in Table3 reflects this latter case. The effectiveness of a good simulation can helpto account for practical constraints and can maximize risk reductions while minimizing time and money spent. This also maximizes the policy's deterrent effect. Effective graphics can make the conclusions more acceptable to both politiciansand the public, and a well-programmed tool will ease the administrative taskby making the decision rationale cleareror perhaps providing a recommendation. So, why is everybody not using such tools? As noted earlier, the EPA often refuses to endorse them, and wisely so because the state of the science is such thatone can barely be said to be operating under statistical uncertainty. A case of outcome uncertainty is more often the case. Earlier, three typical approaches were noted under outcome uncertainty. The worstfirst policy is an aspiration-level choice if we confess that not much is known about risks or that we care little for costs. This has the principal advantage of addressing the mission directly: every effort will be expended to meet it. This may also be well accepted until it starts to cost too much, which in the case of the environment it will. In fact, in every way, its strengths and weaknesses (as shown in Table3) parallel those of the single-rule-based decision; indeed it differs only in that the single rule assumes that the outcome is known, whereas the aspiration level accepts the risk that the outcome may not transpire but feels it to be worth guarding against. '

38

Brown

Environmental management today is taking a close look at subjective decision making. The principal vehicle for this in recent efforts has been the analytic hierarchy process (pairwise comparisons). In addition to being used with some success (in terms of acceptability to activists and regulators alike)at the state level, it is under development now between DOE and EPA. It maximizes participationearly in the processby the groups that most delay action through legal and regulatory maneuvering. Because all parties have an opportunity to participate in the decision-making process, and a properly designed process will leave few completely disgruntled, all are co-opted into the solution. Those few who continue their challenges will have limited credibility given the ecumenical composition of those who have agreedto the solution. In addition, the process can create a rule-based framework with which to make decisions, giving the illusion of a quantitative universal solution under certainty. This capability also presents the biggest trap presentedby the method. As a rule, the process is conducted in conjunction with computer programs to present the questions, record the opinions, and perform the weighting calculations. The programs sell themselves with a pleasing and simple interface that encourages their use by novices, but the comparison structure must be set up very carefully. This has often been overlooked, with the result that the comparisons are inherently flawed. Even where the analysis is properly performed, the illusion of quantified objectivity remains. Decisionsby subjective processes necessarily mean that objective information is lacking or inadequate. The form placed on the structure simply provides a facade of quantifiability. In fact, however, the process is structuring a political decision, made in this case by consensus among expertsor activists rather than duly elected or appointed authority. While there is nothing wrong with that, it is important to recognize it for the political process that it is. In short, subjective decisions maximize political acceptanceand basically admit a lack of hard knowledge in other areas, especially the core objective. Finally, within outcome uncertainty, there is the minimax approach, principally represented in environmental policyby problem avoidance. This approach is characterized by conthe spread of definite tinually seekingnew facts while making temporary investments to control risks. Since nothing is actually accomplished by this, the final cost of a solution is unaffected, and becauseof the tendency for environmental problems to diffuse themselves that cost may be much higher althoughthe payment date is set further into the future. The primary positive outputs of delay are that it can be given an acceptable facade, through continuing studies (especially ifnobody isconcernedaboutcosts), and itminimizestheopportunitycosttothe organization, which is free to go about its business with minimal budgetary impact. In terms of our objective function, problem avoidance deals with neither risk nor effectiveness, but in a perverse sort of way it does keep an eye on costs, at least in current years, by simply refusing to do the needed work. Finally, there is the straightforward political approach. As far as we know fromthe limited studies available, thisis seldom seen (environmental equity authors might differ with that statement). An arbitrary decision by an administrator has the same effect. In suchcases, the basis for the decisionrules is very different from thosewe have noted so far and is unlikely tomaximize any value except political acceptability.

VI. APPLICATIONS OF THE DECISION MODELS In the remainder of this chapter we review five applications of decision theoryto a single enare the use of a singlerule and vironmental problem: investment decisions. Those applications the use of multiple rules under certainty, the useof an expected value approach in a dynamic statistical uncertainty mode, and the use of public opinion under uncertainty. However, the

Management 39

Environmental

state of management science within the environmental field is such that only decisions under certainty can be reviewed in any detail: the other approaches are still being considered but do not as yet represent any official policy.

A. Certainty The first cases to be analyzed use rules to solve the capital allocation question. Theobjective of this exercise is to conduct remediationwork on a group of toxic waste sites that exceeds our ability to pay. One approach is to use a single rule: worst-first. Under that rule, all available resources will be focused on the site posing the worst risk to public health. Remaining funds will be applied to the next-worst site, and so on. An aIternative approachappliesmultiple rules, or at least a morecomplexrule.In this case, the rules are that all available funds will be devotedin sequential priority to the site where the risk reduction per dollar is the maximum. Once work ona site is started under this formula it must be finished. Thisrule is called the “results-first” rule [l]. The situation to be addressed is displayed in Table 4, where a third approach is also displayed for convenience. That approach, problem avoidance, is in a way a rule-based approach (do the minimum necessary for as long as possible) but, as was described earlier,is a manifestation of an approach under uncertainty.

1. Applying the Rules to a Hypothetical Problem Table 4 shows a notional five-site universe given a budget of 100 money units (call them millions of dollars, if you will). It provides the agreed risk (in whatever units) posedby each site and the estimated cost to restore each. Additionally, the table shows the cost of “doing nothing” at each site:these are the costs of the minimum essential security measures,studies, monitoring, and so on (not to mention legal fees) that must be carried out even when no action is possible or desired. Under the worst-firstapproach, only the site risk needbeknown.Wewould attempt to fund site A first; with luck, the sumof the studies over the next 12 years would result in a partial resolution of the problem, but it is more likely that at the end of the period we would have to spend another 800 units of money to execute the remedy (requiring several else could have yearstoaccomplish, at a rate of 100 unitsperyear).Meantime,nothing been accomplished because the balance of our funds each year (20 units) is insufficient to remediate the next site, B. Note also that remediation at either site C or E, which could have been completed with the residual funds, willnotbeaddressed because these sites are not “worst first.”

Table 4 Example of Approaches” Project

Risk

Cleanup cost

Risk/$

Avoidance cost

A

100 70

800 35

0.12 2 3 0.5 5

80 8 3

B

C

Db

E

60

40 20

20 80 4

Total budget available = IOO/yr. bPmject D is already under way. Source: Brown et al. [l].

40 1

40

Brown

The table shows the risk reduction per dollar ratio, which isthemeasureneededfor a results-first decision. This is quite simply the risk dividedby the cost. There is one twist to this: project D was already started, andwe assume that 40 money units has already been expended on it. That is, its total cost was originally greater than 80 units and so its ratio was evenlessthanone-half.Unfortunately, we haveauniqueenoughsituation at Dthatsimply keeping it alive through studies is up to one-half as expensive as finishing it up (a situation that occurs more frequently than one might think). Using results first, we see that the best risk reduction per dollar gives the highest ratio, making our order of preference E, C, B, D, and E, C, and then A. Withthis preference, wesee that within our budgetwe can complete projects B. We also have 41 units left over to fund some of the work at A and D. By the end of the second year we would have completed all the sites except A, and we would be left with the same problem as in the worst-first case-the inability to restore A-except that while only being 2 years further behind at A (which had not really gone anywhere anyway), we have completed all our other work and in the process we may have learned something abouthow to deal with site A. us to fund studiesat all sitesand to finishD, since work Problem avoidance would require there was already started and it is too much trouble to stop. In either case, we do not have enough money to do the full studies requiredat A, so we skimp along on whatwe do have after funding everything else. This does happen in practice, since the durationand scope of tests and studies are fairly arbitrary and highly negotiable. The result of this strategy is that site D, which offers one of the lowest returns in risk reduction per dollaras well as being a relatively low ranking site, is the only one cleaned up, because every year the entire budget is consumed in studies. After4 years, we have spentas much in studiesat site E as we will eventually spend in cleaning it up-another event that is not infrequent in reality. Witness the fact that most of the sites removed from the National Priorities List have not been the “worst” sites that expensive studiesto have posed should have been first, but rather sites that proved after ofyears little risk.

2. ApplyingRealData Setting up Table 4 for illustrative purposeswas easy enough butmay be considered to be misleading. Would the superiority of a results-first approach be demonstrated with real data?An analysis was performed using the scoring data from1600sites containedin the EPA’s National Priorities List Technical Data Files database, and estimated remediation costs were extracted from the 523 then-existing records of decision (RODs).The result was a joint database of 412 unique sites with cost data (some RODs are for operable units within sites or are updates of earlier RODs). These data are acknowledged to be less than perfect. Aside from the fact that the HRS score may not really be a measure of risk, in some cases the factor valuesmay have been improperly assigned during the site assessment and scoring processes. The costs are quite the for operable units are fluid. More important, itis not always clear in the ROD whether data cumulative or separate (one can sometimes get enough of an impression to know that either is used on occasion), so that the decision to add to or replace the previously recorded cost is subject to error. The model uses an arbitrary cleanup budget ranging from 1 to 3 billion dollars per year over a 30-year period, reflecting the reasonable expected range of Superfund appropriations $2 billion). In addition, no more than an arbitrary limit of $5 (presently being scaled back from million per year maybe spent onany one site; this limitation was designed to preventmodel the from miraculously fixing a complex site suchas Commencement Bay or Rocky Mountain Arof cash.In fact, theenvironmentalprogram senalinoneyearwithamassiveinfusion

Environmental Management

41

Table 5 ResultsUsing NPL Data Worst-first Results-first Results ratio Sites cleaned HRS points reduced Total risk reduction including time effect

5 365 5,608

69 2,877 69,989

1 41 8:I 12:1

Source: Brown et al. [l].

has shown someof the same tendency to throwaway money underoverstimulation aswas seen in the Defense Departmentin the 1980s: sometimes thereis a limit to how much money one can spend wisely in a given period. Finally,the model recognizes the continuationof risk incurred by failure to remediate. Because of the funding limitby time period, and because of the life-cycle risk approach, the model requires an iterative solution based on the initial decision rules. It is important for environmental managersto note, however, that this relatively complex calculation canbe (and was, for this exercise) accommodated with plain old dBase 111 Plus rather than using a more complex simulationtool. The working of the model is described in more detail in reference 1. The simulation determined which projects were to be funded in each year using the onerule and multiple-rule methodologies. The results are shown in Table 5 . Clearly, a more sophisticated rule system that acknowledges the effect of realistic constraints (i.e., resource limitations) outperforms onethat adheres more closely tothe original goal. One might argue that for the specific sites we have selected, the HRS scores do not represent risk and do not represent toxicity or time effects. However, the purpose of this analysis is not to show how many HRS points can be cleaned up or what strategic approach is superior. Rather, it is to point out that effective management depends on effective identification of objectives, requirements, resources, and constraints. The use of tools because they are simpler (such as a one-rule system)will often be less productivethan the use of tools that more closely reflect real conditions. The manager’sjob is to identify the relevant aspectsof any decision and to ensure that all aspects are properly considered.

B. StatisticalUncertainty The rule-based system just described was fairly complex, even when operating with the assumption of certainty. Incorporating probabilities or ranges of values into such a model as the multiple-rule model would require automated systems support. There are many models in existence that assist in the solution of particular problems. There are few, however, that provide information to a manager before an event occurs.The EPA’s CAMEO model may be the best known; designed for emergency response use, it can be used for “what-if” analysis under selected circumstances. (Contact theUSEPA Office of Solid Waste and Emergency Response in Washington, DC., for additional informationon CAMEO.) Even with CAMEO, however, the situation needs to be specified quite closely. Other “screening” models have been produced under various development initiatives by the EPA and by private vendors; those models generally addressa specific pollution mode (air, soil, surface water, or groundwater) and usually address fate and transport issues rather than providing impact analyses. In addition, many site-specific models have been developedfor individual environmental projects,and in a number of cases those models have been placed inthe

42

Brown

public domain in case they may be found useful. In other words, there are literally hundreds of dynamic models available; some address known a situation over time,and some aremuch more s~phisticated.~ What is needed is a simulation capability that can portray sources and receptors and model (GIS) systems pathway behavior. Unfortunately,at present the Geographic Information System that contain the needed data cannot be interfaced with the programs that replicate the pathways. Some systems development is under way that may help in this area [21]. The requirement is for a system that will represent the behavior not of a single site, but of a typical environmental situation with multiple potential sources and receptors. Such a model must include the possibility that currently contained sources may fail, must allow for changes to be made to reflect proposed upgradesor corrective actions, and must be able to correlate these activities to costs. Finally, the systemmust be visually interpretable and preferably simple to use. These requirements will not be simple to meet, which is why there are no such systems today.

C. Outcome Uncertainty It was demonstrated earlier that the problem avoidance (minimax) strategyis self-defeating in environmental operations because in many cases it costs almost as much to do the minimum (once the effectof time is considered) as it does to go ahead and take careof the problem. We have also seen, thanks to the “worst-first” model under certainty, that aspiration-level approaches (which may be expressed as “just do it”) may well fall apart when faced with the reality of resource or capacity constraints. In fact, the very insistence on mission focus at the expense of realism results in less of the actual mission being accomplished than in the more realistic multiple-rule or expected value approach. This leaves the subjective decision-making approach. This method of analysis is receiving increasing attention. Two effortsaddressingcapitalfacilitiesdemonstratedifferentapEPA’s effort to develop policies for DOE facilities. EPA DOE’S and use proaches. The first is the of such an approach was encouraged by the successof earlier efforts in developing underground storage tank replacement strategies at the state level, with New Mexico being the leader in this regard. 1. AnalyticHierarchy

The analytic hierarchy process(AHP), often known as “pairwise comparison,” works by considering all of the factors involved in reaching a decision and comparing their relative importance in that decision. For instance, when trying to select an item of equipment, one may consider purchase cost, reliability, and efficiency. Then these elements are compared. One could just assign weights directly; AHP assumes that effective quantitative judgments on these qualitative issues wouldbe illusory. Instead, the participants are asked to assess the relative importanceof one factor over another. Based on the amount by which one element is said to be more important than another, a weight is established (unseen to the participants) for each factor. Factors can be tiered: thus, each factormay have subordinate indicators (in the example to above, reliability may be assessed in terms of both manufacturer’s advertised mean time failure and the corporate experience with similar machines, with one rated more heavily than the other).

’A helpful referenceis the USEPA Informution Resources Directory, publishedby the Information and Resources Management Division of the USEPA. My copy is document number OPA 003-89. March 1989, but newer versions may subsequently be published under different document numbers.

Environmental

Management 43

The apparent simplicity of the AHP method is deceptive.It leads people who are not very familiar withit to attempt its use, and because the result is produced by the computer to three decimal places it has the aura of authenticity. But if the factors are improperly constructed, the result willbe meaningless. While the computer can assess the technical consistency of the rankings with each other based on the pairwise comparisons, it cannot answer whether the comparisons were meaningful. The best test of structure is to ask the same question the software asks; using the example above,“In assessing whatequipment to buy, which is more important: reliability or flciency?” If the sentence with the blanks filled in seems nonsensical, the structure may be improper. Another indication of a problem with the structure of the model, or a lack of general understanding of the issues, is the finding of generally very high or very low scores (near 0% and 100% weighting) for numerous factors. In theEPA’s implementation of the Risk Information System (RIS), the major factors were health impacts and environmental impacts. Tiered beneath these were various intervals within which pollution would be expected to arrive. These in turn were constructed in terms of the individual and population risks, which in turn had three layers of other indicators. In early efforts, the panels returned numerous zero-weight responses; where decision trees have only two branches, eliminationof one branch renders the decisiontree useless. To the EPA’s credit, they recognized the problem this created and went back to review the model structure; at the time of writing, the revision has not been published. This servesto reinforce the point made earlier in offering multiple decision models: not all will work in every situation. When the model doesn’t work, one has to make assumptions to develop the best possible outcome. Criticalto that process, of course, is to recognize the fact that assumptions have beenmade.

2. WeightingWithoutAHP An alternative formulationof the same problem came from DOE‘SResource Allocation Support System (RASS) project. In this case, DOE did not use the AHP approach, electing instead to rate the need for facilities based on a series of scales, each scale having a weight of its own and the raw data being used to construct the scales being weighted within each scale. Raw data supplied by local managers to describe their proposed project were converted to scales presented as “letter grades.” The siting strategy considered four primary factors: cost, health issues, the volume of waste that an alternative could handle, and the degree to which the alternative complied with applicable regulations. An extensive amountof preparation would be required to obtain the raw data needed to make the system work. However, it was not intended as a detailed engineering analysisas much as a first-tier estimate to allow general project decisions and funding allocations tobe made. This process in still in the pilot project phase, and a report on the results is not yet available. The item that environmental managers should consider in using such a system is whether it is in fact appropriate to the problem to be solved. The difference between the RASS application and the AHP-based RIS application is that RIS attempted to capture nonquantifiable values and norms (the relative significanceof carcinogenic and noncarcinogenic pollution, for instance). RASS deals exclusively with highly quantifiable values (waste tons, costs, etc.). Considering that the compliance question is really fairly irrelevant,as no facility will be permitted that does not comply with all applicable regulations, one could ask whether the objective function could have been simplified to maximize the waste or risk reduction (as was demonstrated in the “results-first” model) without need the for a weighting system that distorts the true dimensional valuesof the data. Nonetheless, the RASS approach is a relatively complete attack on a multifaceted problem and is worthy of consideration as an approachto solving environmental problems.

44

Brown

VII. CONCLUSION This chapter does not answer any questions about the environment. has,Ithopefully, stimulated a number of questions about how decisions are madeand what basis is offered for those decisions. Environmental managers needto understand the full range of tools available for decision support. More important, they need to understand the degree to which the way in which the the solution objective is stated caninfluence the choice of decisionapproachandhence adopted. And they need to be aware of the strengthsand limitations both of the decision methods employed and of the data used to support those decisions. Few decisions will ever satisfy all parties involved in such a complex and controversial policy area as the environment. Elaborate mathematical proofs, eye-catching simulations, or cutely worded weighting scales will neither improve the decision nor overcomethe reluctance of an opponent to concede that a position has some merit. But a better understanding of the premises used to arrive at a decision can ensurethat discussions of the decision are based on the same understandingof the problem,an essential first step in arriving at a sustainable public policy decision.

ACKNOWLEDGMENTS This chapter represents a significant advancementfrom, but nonetheless capitalizes onmany of the frameworks and concepts found inan earlier article [l]. The permissionof Hazardous Material Control to reproduce major portions of the text of that article are gratefully acknowledged.

REFERENCES 1. Brown, D. M., Dienemann, P. F.,and Kline, R. C., Results first: an enhancement of the worst-first approach, Hazardous Mater. Control, 5(4): 20-33 (1992). 2. Hembra, R. L. (U.S. EPA), The need to establish environmental priorities, presentedat the Southeastern Conference on Public Administration, Am. Soc. Public Administration, Charlotte, N.C. Oct.16,1991. 3. U.S. Environmental Protection Agency,Comparing Risks and Setting Environmental Priorities, Office of Policy, Planning and Education (PM-220). August 1989. 4. U.S. Environmental Protection Agency, Unfinished Business: A Comparative Assessment of Environmental Problems, 1987. 5 . Hawkins, K. and Thomas, J. (eds.), Enforcing Regulation, Kluwer-Nijhoff, Boston, Mass. 1984. 6. Fleischer, G. A. Engineering Economy: Capital Allocation Theory, Wadsworth, Belmont, Calif., 1984. 7. Lapin, L. L., Quantitative Methodsfor Business Decisions, 2nd e d . , Harcourt Brace Jovanovich, NewYork,1981. 8. Quade, E. S., Analysis for Public Decisions, 2nd ed., North-Holland, New York, 1982. 9. Douglas, E. J., Managerial Economics: Theory, Practice and Problems, 2nd ed., Prentice-Hall, Englewood Cliffs, N.J., 1983. 10. Weber, Max,c. 1922, Economy and Society: An Outline of Interpretive Society, edited and translated by G . Roth and K. Wittich, 2nd e d . , University of California Press, Berkeley, Calif., 1955. 11. Simon, H. A., Administrative Behavior, 3rd ed., Free Press, New York, 1976. 12. Doty, C. B.andTravis,C.C., Is EPA's national priorities list correct? Environ. Sci. Technol., 24( 12): 1778-1780 (1990). 13. Brown, D. M.,and Kline, R. C., Are the worst first? A review of national priority list sites, Proc. Region VI Conf. Am. Soc. Public Administration.Dayton, Ohio, 3 Oct. 1991, pp. 87-104, Political Science Department, Univ. Dayton, October 1991.

Environmental 14.

15.

16.

17. 18. 19. 20.

Management 45

U.S. Department of Energy, Objectives, scales and scoring instructions for a pilot study of waste management's resource allocation support system (RASS), draft, U.S. Dept. of Energy, Germantown, Md., June 1992. Contact Mr. Kevin Donovan of DoE for more information on the status of the RASS study. Moore, W. B., Brown, D. M., and Hutchinson. R. A., Updatedfiscal impact analysisfor the Naval Submarine Base, Kings' Bay, Georgia, LMIRept.No.FF'605R1,LogisticsManagementInst., Bethesda, Md., December 1986. Walsh, W., Susel, I., and Ronayne, A., Setting risk-based priorities: a method for ranking sites for response, Proc. HMCRI R&D Conf. Anaheim, Calif., February 1991, HMCRI, Silver Spring, Md., 1991, pp. 112-123. Hird, J. A., Superfund expenditures and cleanup priorities: distributive policiesor the public interest? J . Policy Anal. Manage., 4455-483 (1990). Sax, I. N., Dangerous Properties of Industrial Materials, 5th ed., Van Nostrand Reinhold. New York, 1979. Weiss, G . , Hazardous Chemicals Data Book, 2nd ed., Noyes Data Corp., Park Ridge, N.J. 1986. Fiorino, D. J., Can problems shape priorities? The case of risk-based environmental planning, Public Admin. Rev. 50( l): 82-90 (1989).

21.

Brown, D. M., Integrated environmental management: a GIS system you can afford, Proc. HMC Superfund 92 Conf. (Hazardous Materials Control and Research Institute), Washington, D.C., Dec. 1-3.1992.

This Page Intentionally Left Blank

4 Introduction to Engineering Evaluation for Contaminated Sites

David S. Wilson, Alan C. Funk, Ronald G. Fender, and Marilyn Hewitt Environmental Resources Management Group Exton, Pennsylvania

1. CHARACTERISTICS OF ENVIRONMENTAL CONTAMINANTS There are literally thousands of substances that can contaminate the environment. In some cases, this contamination is related to natural conditions and is thus a “background” condition. Most often, however, environmental contaminants are related to human activity. Current regulatory programs governing contaminated sites generally focus on a class of contaminants called hazardous substances. These regulated substances have been identified by the United States Environmental Protection Agency (USEPA) on the basis of their relative toxicity in the environment. Many other substances may be released to the environment; some are unregulated, and others may be regulated by agencies other than the USEPA. Any given contaminated site is usually characterized by one primary contaminantor suite of contaminants related to activities conducted at the site. This contaminant profile is often referred to as the site “fingerprint.” Once this fingerprint hasbeen established by general environmental sampling and comprehensive chemical analysis, detailed studies of the site are generally focused on the fingerprint compounds. The contaminants present at any given site are highly dependent on the natureof the acof ortivities at the site. For example, at chemical industry manufacturing facilities, a variety ganic compoundsmay be present, whereasat a facility that does substantial electroplating, the contaminants may be principally heavy metals. The electronics industry and metal products industry will also commonly be characterized by the presence of organic solvents as environmental contaminants. A wood-preserving operation, by contrast, would be characterized by phenols and creosote-related compounds. A commercial hazardous waste disposal facility could be characterized by any combination of organic and inorganic contaminants. A municipal waste disposal landfill is generally characterized by some amount of organic compounds and metals, but mostly by liquid leachate containing organic acids, iron, and ammonia from the decomposition of trash and garbage. Environmental contaminants are generally divided into three classes: 47

Wilson et al.

48

Organic Compounds. These are human-produced compounds that do not occur naturally in the environment to any significant degree. Inorganic Compounds. Heavy metal ions are the principal concerns. Although these are naturally occurring elements, human activity tends to concentrate them to levels that have higher potential toxicity than natural levels. Other contaminants such as the anions chloride and sulfate are also inorganic contaminants. Inorganic contamination can also result in environmentally unacceptablepH levels in the environment at some sites. Biological. The principal concerns for biological contaminants are bacteria and pathogens associated with sewage. These are not generally concerns at contaminated sites related to industrial or disposal activities. A brief summary of the nature of common organic and inorganic contaminants follows.

A. Organic Contaminants The USEPA divides organic contaminants intoclasses related to chemical characteristics that drive required analytical methods. These classes are discussed below. Volatile Organics By far the most common environmental contaminants are volatile organic compounds (VOCs). This is due to their presence in petroleum hydrocarbons that are widely used as fuel in our society (gasoline, fuel oil, etc.) and to the almost universal use by American industryof volatile organic solvents for degreasing during manufacturing operations. Volatile organics are also used as raw materials in some industrial processes, such as those of the pharmaceutical industry. Table l presents USEPA's Target Compound List for VOCs. Volatile organic compounds are characterized principally by their high vapor pressure, that is, their tendency to volatilize to the gaseousstate at standard temperature and pressure. These compounds are also characterizedby relatively low levels of water solubility, generally froma few hundred parts per million 1% in water. However, due to toxicological characteristics, these solubilities, although limited, are significant in terms of potential environmental and human health impact. A third majorcharacteristic of volatile organic compounds is their specific gravity. Those compounds without chlorine atoms tend to be lighter than water and therefore will float. Compounds that have been chlorinatedare generally heavier than water and will sink in water under the influence of gravity. This characteristic is very important in the dynamics of migration of VOCs in the environment. Chlorinated aliphatic compounds are among the most common environmental contaminants. These are the Compounds that have been most commonly usedby industry as degreasing solvents. The principal compounds of concern are tetrachloroethene, which is used in industrial applications and as commercialdry cleaning fluid; trichloroethene, widely used for metal degreasing; l,l,l-trichloroethane, also widely used for metal degreasing; and methylene chloride, whichhas a variety of industrialuses,includingitsuseastheprincipalsolvent for decaffeination of coffee and for wood-stripping operations. Of these compounds, trichloroethene, also knownas TCE, is perhaps the most ubiquitous and notorious. Depending upon environmentalconditions, this compound undergoes reductive dehalogenation to form dichloroethenes,and finally vinyl chloride, the most toxic of the chlorinated aliphatic compounds. These chlorinated compounds are heavier than water and thus tend to sink under the influence of gravity in water systems. Also very common in the environment are aromatic compounds, which are used as solvents and are present in petroleum products. Perhaps the most common class of these aromatics is referred to as the BTEX series-benzene, toluene, ethylbenzene, and xylene. These com-

Evaluation Engineering

forSites Contaminated

49

Table 1 TargetCompoundlAnalyteList Volatiles ~~

~~

Acetone Benzene Bromochloromethane Bromodichloromethane Bromoform Bomomethane/methyl bromide ZButanone/MEK Carbon disulfide Carbon tetrachloride Chlorobenzene Chloroethane Chloroform Chloromethane/methyl chloride Dibromochloromethane

1,2-Dibromo-3-chloropropane 1 ,ZDibromoethane 1 ,2-Dichlorobenzene 1,3-Dichlorobenzene 1 ,4-Dichloroethane 1.1-Dichloroethane l ,2-Dichloroethane 1, I-Dichloroethene cis-l ,2-Dichloroethene trans- 1,2-Dichloroethene 1 ,ZDichloropropane cis-l ,3-Dichloropropene trans- 1,3-Dichloropropene Ethylbenzene

2-Hexanone Methylene chloride 4-Methyl-2-pentanonelMIBK

Styrene 1,1,2,2-Tetrachloroethane Tetrachloroethene Toluene 1,1,1 -Trichloroethane 1,1,2-Trichloroethane lkichloroethane Vinyl chloride Xylenes (total)

Semivolatiles Acenaphthene Acenaphthylene Anthracene Benzo[a]anthracene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[ghi]perylene Benzo[a]pyrene bis(2-Ch1oroethox)methane bis(2-Chloroethy1)ether bis(2-Ethylhexy1)phthalate CBromophenyl phenyl ether Butyl benzyl phthalate p-Chloroaniline p-Chloro-m-CreSOl 2-Chloronaphthalene 2-Chlorophenol CChlorophenyl phenyl ether Chrysene o-Cresol/2-methylphenol

p-CresollCmethylphenol Di-n-butyl phthalate Dibenz[a,h]anthracene Dibenzofuran 3,3’-Dichlorobenzidine 2,CDichlorophenol Diethyl phthalate 2,CDimethylphenol Dimethyl phthalate 4,6-Dinitro-o-cresol 2,CDinitrophenol 2,4-Dinitrotoluene 2,6--Dinitrotoluene Di-n-octyl phthalate Fluoranthene Fluorene Hexachlorobenzene Hexachlomyclopentadiene Hexachloroethane Hexachlorobutadiene

Indeno[ 1,2,3-cd]pyrene Isophorone 2-Methylnaphthalene Naphthalene o-Nitroaniline m-Nitroaniline p-Nitroaniline Nitrobenzene o-Nitrophenol p-Nitrophenol n-Nitrosodiphenylamine n-Nitrosodi-n-propylamine 2,2‘-0xbil( I-chloropropane) Pentachlorophenol Phenanthrene Phenol F‘yrene 1,2,4-Trichlorobenzene 2,4,5-TrichlorophenoI 2,4,6-Trichloropheno1

PesticidesPCBs Aldrin a-BHC p-BHC y-BHC (Lindane) &BHC a-Chlordane y-Chlordane 4,4’-DDT 4.4”DDE 4,4‘-DDED

Dieldrin Endosulfan I Endosulfan I1 Endosulfan sulfate Endrin Endrin aldehyde Endrin ketone Heptachlor Heptachlor epoxide Methoxychlor

Aroclor 1016 Aroclor 1221 Aroclor 1232

Aroclor 1242 Aroclor 1248 Aroclor 1254 Aroclor 1260 Toxaphene

de,

Wilson et a!.

50

Table 1 (Continued) Inorganics ~

Nickel

Aluminum Antimony Arsenic Barium Beryllium Cadmium Calcium Chromium

~~

~

~

~

~~~

Cobalt

pounds are present in refined petroleum products, principally in gasoline, and are thus found widely distributed through the environment due to leaking subsurface gas tanks and surface spills and discharges of petroleum products. In pure form they are also used by certain industries as raw materials and/or solvents. These compounds, being unchlorinated,are lighter than water and therefore float. Semivolatile Organics Semivolatile organicsare by-products of many types of industrial and nonindustrial processes. Table 1 shows the USEPA's Target Compound List of semivolatile compounds. Examples of common compounds found in the environment are phthalates (used as plasticizers), polynuclear hydrocarbons (PAHs) (produced by the combustion offossil fuels), and phenol compounds (which havemany industrial applications inthe chemical industry andare used as wood preservatives). The characteristics of semivolatile organic compounds includelower vapor pressure than the volatile organics. These compounds are therefore of low volatility and tend to exist in the solid or liquid state at standard temperature and pressure. They may be present in the environment either in liquid form or in some cases solidparticulate form (e.g., PAHs in soot produced by incompletecombustion in residential wood stoves). Many semivolatileorganic compounds are heavier than water in their liquid state and thus will tend to sink in water. In general, semivolatile organics alsohave relatively low solubility in water. Phenol compounds are the general exception,with solubilities in the tens ofparts per million range; PAHs, by contrast, have solubilities limited to the low parts per billion range. Pesticides and PCBs Pesticides and polychlorinated biphenyls (PCBs) are also semivolatile compounds. They are lumped togetherby the USEPA on the basis of similar analytical protocol requirements. These compounds have been found to be fairly ubiquitousin the environment at very low levels (i.e., low parts per million to parts per billion). Table 1 shows USEPA's Target Compound List of pesticides and PCB compounds of concern. Pesticides enter the environment through both agricultural use and heavy historical residential use for pest control and weed control. PCBs, on the other hand, have had principally industrial applications. Because of excellent heat resistance properties, PCBs were used for many years in electrical transformers and as hydraulic fluids in heavy equipment. PCBs were generally dissolved in oil for their application. The manufactureof PCBs was of use in the halted by the USEPA in the 1970s. They have been almost completely phased out United States today. Pesticides may enter the environment either in particulate form as powdersor dissolved in solvents. In their liquidform, pesticides are usually lighter than water, as they tend to be dis-

Evaluation Engineering

forSites Contaminated

51

solved for commercial application in BTEX solvents. PCBs are also lighter than water when present at percent levels or less in oil; however, pure PCB liquids are heavier than water. Perhaps the most controversial semivolatile organic compounds are dioxin and its related compounds, the furans. Dioxin is principally a by-product of the manufacture of certain other substances, the most well known processes being the production of defoliants such as Agent Orange, and the bleaching of paper in the commercial paper industry. Dioxinand furans can also be produced as combustion by-products of some organic materials. Small amountsof dibe present in uncontained emissions from municipal waste incinerators. oxin have been found to As these compoundsare by-products of other processes, they are not found in concentratedor by pure form; rather, they may be found in a wastewater effluent stream, in media affected incinerator emission fallout, or as contaminants in certain waste streams produced from the manufacture of other substances.

B. InorganicContaminants As previously noted, inorganic contaminants are naturally occurring substances that can become concentrated in the environment due tohuman activities. The principal contaminants of concern in the environment are theheavy metals. Metals on the USEPA's Target Analyte List are listed in Table 1. Heavy metals are of relatively low vapor pressure and are thus effectively nonvolatile at standard temperature and pressure. However, some of them, for example, lead and mercury, become very volatile at elevated temperatures. This is very significant to environmental engineers considering thermal treatment techniques for wastes or contaminated media containing such metals. In addition to the more toxic heavy metals, less toxic metals such as iron and manganese are common environmental contaminants. These metals in particular can create difficulties for environmental engineers in water collection and treatment systems at contaminated sites. In addition, salt ionsin solution (e.g., calcium, sodium, magnesium, and potassium chlorides and sulfates) are often present at contaminated sites.At some sites, theremay be elevated levelsof nutrient compounds (i.e., nitrogen and phosphorus compounds). Both the salts and nutrients present significant difficulties and costs for environmental engineers when they must be removed from water.

C. Summary In summary, of the thousands of substances that can contaminate environmental media, the USEPA has developed a list of hazardous substances that are of primary concern. Volatile organic compounds, semivolatile organic compounds, and inorganics, principally heavy metals, are the classesofhazardous substances commonly dealtwith by environmentalengineers at contaminated sites. The following discussion addresses the occurrence of these classes of compounds in various environmental media and the implications for remediationof contaminated sites.

II. CHARACTERISTICS OF CONTAMINATED MEDIA The media that may require remediationat any given contaminated site may include waste materials that were disposed or of spilled in the environment, soils contaminated either directly by spillage and discharges of liquids or indirectly by infiltrating contaminated water, groundwater contaminated either directly by nonaqueous-phase liquids (NAPLs) or by contaminants in solution, surface water containing nonaqueous-phase liquids and/or contaminants in solutionor

52

Wilson et al.

suspension, and air,which may transport contaminated dustor Contaminants in gaseous form. Each of these media is described briefly in this section.

A. Waste Materials There are a great variety of types of waste that can cause environmental contamination. Some of the materials commonly encounteredare listed below. LiquidChemicals. Raw materials,waste,and/oroff-specliquidchemicals mayhave been spilled at a site, disposed of in bulk on the ground, or placed in containers such as drums or tanks. Such spillage or disposal can contaminate soil, groundwater, and surface water. Thus, these types of waste may have to be remediated either as bulk materials still in containers or as constituents absorbed in soil or water. Solid Materials. Solid waste materials are usually residuals generated by a manufacturing or treatment process. Examples include sludgesfrom treatment of electroplating wastewaters and refinery sludges containing oilsand BTEX compounds. Sludges to be remediated either may be very low in solids content or may have been dried to essentially solid form. or are present as precipitates in treatment These typesof materials m often found in drums or holding lagoons that must be remediated. Other types of industrial solids may include set organic resins, powders, and materials such as slag or mine spoils. MunicipalSolid Waste. This class represents materials disposedofin municipallandfills. Older landfills requiring remediation commonly contain principally household refuse, usually mixed with at least some wastes generatedat industrial facilities. Low levels of hazardous substances may be present in leachate from these landfills, originating both from industrial disposal and from the disposalof household products containing hazardous substances. Interestingly, codisposal of industrial and municipal wastes at many such landfills has resulted in biodegradation of most hazardous organic compounds. Also, the essentially neutral pH of municipal landfill leachate does not favor migration of heavy metals from many landfills. Thus, at many such facilities, the principal problemis leachate containing iron and ammonia and having a high biological oxygen demand.

B. Soil Contaminated soil must be remediated principally for two reasons: 1.

Potential adverse health affects on humans and/or animals coming into direct contact with the contaminants 2. The fact the contaminants in soil often provide a continuing source of contamination to underlying groundwater and/or surface water The degree to which soil at a site has been contaminated is dependentupon the nature of the disposal activities, the quantity of materials disposed of, and the chemical nature of those materials. When selecting a remedial action forsoil, several key characteristics of that material must be taken into consideration: 1. 2.

Soil grain size and cohesiveness. It is generally much easier to extract contaminants from a permeable sandy soil than froma low permeability clay soil. Organic carbon and clay content. Certain classes of contaminants, such as heavy metals, PAHs, PCBs, and pesticides, readily adhere to soils that have either high organic carbon content or significant cation-exchange capacity relatedto clay mineralogy. Other contaminants, such as the volatile organics, have very low absorption coefficients on soil materials regardless of organic carbon or clay content.

Evaluation Engineering

forSites Contaminated

53

3. Soil moisture content. This can be an important factor both for handling soils during ex-

cavation and for treating soils, either in situ or after excavation. For example, effective the removal of VOCs from soilby volatilization, either in situ or above ground, depends on rate at which the soil can be dried and disaggregated. 4. BTU content. If soils are to be incinerated, the BTU content is important to determining the feed rate to the incinerator. Since most soilsare very low BTU materials, the degreeto which the contamination itself contributes BTUs is often thekey factor. Another very important factor, unrelated to the soils themselves, is the required cleanup levels for the contaminants of concern. Where very low cleanup levels (generallylow ppm to ppb levels) are required, it is fairly unrealistic to expect that in situ technologies will reach those levels. Although significant contaminant massmay be removed (for example, by soil vapor extraction of volatile organics or in situ biodegradation of biodegradable organics), these methods will not generally achieve verylow cleanup levels. This is due to the heterogeneityof the subsurface environment,which prevents in situ processes from effectively reachingareas all of contamination. Excavation and carefully planned above-ground treatmentof soils, although they can become very expensive, are generally required to meet very low cleanup levels.

C. Groundwater Remediation of groundwater requirestwo key components: 1. Interception and/or collection of the contaminated water in the subsurface 2. Treatment of the contaminated water that has been intercepted and/or collected

Interception and/or collection of the contaminant “plume” is designed by a hydrogeologist or hydrologist with assistance from geotechnical engineers. The hydrogeologist must determine aquifer characteristics such as mechanisms of flow, directions of flow, extent of the contaminant plume, aquifer permeability, and subsurface stratigraphy in order to design a collection system. For example, selection of collectiodinterception technologies such as pumping wells, interception trenches,barrier walls, and potentialfor in situ treatment are all highly dependent upon the aquifer characteristics. Full remediation of contaminated groundwater to cleanup levels below regulatory standards cannot generallybe achieved with existing technology. The degreeto which contamination can be removed from an aquifer system depends upon two main factors: 1. The nature of the contaminants released-their degree of solubility and the aquifer material’s absorption capacity for them. 2. The nature of the release. If water materials or contaminated soilsin the unsaturated zone are not completely removed,they will serve as continuing sourcesof contamination to the groundwater below.If liquids were released in sufficientquantity to physicallyreach the groundwater table in free form, then either a light nonaqueous-phase liquid (LNAPL) or a dense nonaqueous-phase liquid (DNAPL) willbe present in the groundwater system. LNAPLs can be well defined in many hydrogeologic systems and removed to a significant degree. However,DNAPLs migrate downwardbeneath the water table, in response to gravity, along natural pathsof permeability contrast withinan aquifer. It is commonly difficult if not impossible to track the migration of all DNAPLs. In addition, if found, they cannot usually be completely removed.

Groundwater that is collected for treatment above ground is usually readily treatable using well-developed, proven treatment technologies. However, common pitfalls for recovery/treatare ment systems are the effects of iron and manganese on those systems. Iron and manganese

54

Wilson et al.

often presentat contaminated sites at elevated levels,either naturally or related to the sitecontamination. Well screens, trenches, and treatment systemsin media may be fouled by corrosion andlor bacteria related to iron and manganese. This can greatly affect operation and maintenance requirements andcosts for groundwater remediation facilities.For this reason, it is imperativethatthese characteristics of thegroundwater be determinedbeforerecoveryand treatment systems are designed. To assess the potential forin situ treatment, thecharacteristics of an aquifer must be well defined. To date, in situ treatment of groundwater has been limited becauseof the limited biodegradability of many compounds and heterogeneitiesin the subsurface aquifer materials.The one exception to this generality is the use of biological degradation in homogeneous unconsolidated permeable aquifers where nonchlorinated petroleum product contaminants are present.

D. SurfaceWater Direct discharges of contaminants flowing to surface water are dispersed rapidly; therefore, remediation of surface water problems usually requires that the source of the contaminationbe mitigated. Direct treatment of surface water at contaminated sites is not commonly required except in the case of a large-scale emergency spillage. At a contaminated site, surface water is generally affected via three pathways:

1. Contaminants in solution or suspension inrunoff 2. Discharge of contaminants via the groundwater system 3. Contaminant accumulation in bottom sediments, providing a continuous source of release of soluble contaminants intothe surface water Remediation of surface water requires knowledge of the mass balance of contaminant input from these various sources.Only then can the proper combination of runoff and erosion conbe selected for meeting trols, groundwater interception, and/or sediment treatment and removal surface water cleanup requirements. Due to the low solubilities and high adsorption coefficients, contaminants such as metals, PAHs, pesticides, and PCBs can accumulatein bottom sediments. Volatile organics, and sometimes metals, can enter surface water in solution via groundwater. Oils and other NAPLs can also be discharged directly into surface water via groundwater.

E. Air Air quality issues at contaminated sites are generally associated with two potential concerns: (1) airborne dust emission containing semivolatilesor metals, and (2) direct gaseous emissions of volatile organics by volatilization from othersite media. These pathways are sometimes of concern under ambient site conditions, that is, at the existing unremediated site. This is particularly true if heavy surface soil contamination is present and there is little or no vegetation to prevent fugitive dust emissions fromoccurring. It also may be true where NAPLs discharge from subsurface via seepsor to surface water bodies. Even more commonly, however,these two pathways are of concern during remedial action itself. Excavation of soils, movement of heavy equipment, and exposure of volatile contaminants deep in the subsurface by excavation can create air emission problems during remediation. Also, emissionsmay be created by treatment processes. For example, any air stripping of volatile organics will result in the generation of gaseous state compounds that may require treatment. Soil vacuum extraction systemsfor volatile organics commonly require treatmentof the air stream, usually by carbon absorption or fume incineration. The use of heat for treating environmental mediamay also result in the emission ofsuch substances as volatile metals, PAHs, dioxins, and furans. The potential emissions depend upon the natureof the contaminants at the site and the nature of the treatment method selected.

Evaluation Engineering

for Sites Contaminated

55

F. Summary In summary, media requiring remediationat contaminated sites most commonly include waste materials, soil, groundwater, surface water, and air. The characteristicsof the media themselves as well as the characteristics of the contaminants present must be well known before effective remedial measures can be taken. The degree to which remediation of the site is practical is often a function of the degree of cleanup desired or required. Removalof trace levels of environmental contaminants from environmental media is often beyond the means of available technology.

111. REGULATORY STANDARDS FOR SITE REMEDIATION To mitigate risks to human health and the environment andto define levels for cleanupof existing contaminated sites, the USEPA and state regulatory agencieshave established standards for environmentalmedia discussed respectively in Sections I and 11. Because the list of chemical parameters subject to regulatory standards is extensive, our presentation of standards is limited to some of the most common contaminants in environmental media. These contaminants and the applicable standards are listed in Tables 2-5. The standards discussed herein include those from federal regulations suchas the Federal Drinking Water Criteria (FDWC) and the Resource Recovery and Conservation Act (RCRA) action levels. As an example of state regulatory standards, New Jersey cleanup standards are included because they are some of the most stringent and most recently updated cleanup criteria. At the time of writing, the New Jersey standards were proposed but not yet promulgated on all ongoing as law. However, NewJersey hasbeen using the proposed standards for guidance and proposed cleanups.

A. SolidWastes The standards for cleanup of solid wastes will,in most cases, be the same as the standards for cleanup of contaminated soils(see Table 2). Handling of solid wastes is dependenton whether they are classified as hazardousor nonhazardous. They may be classified as hazardous on the basis of being a listed waste (per 40 CFR 261 Subpart D) or on the basis of being hazardous by characteristic (per 40 CFR Subpart C). Characteristics of hazardous wastes include ignitability, corrosiveness, reactivity, and toxicity, as defined by the results of standard tests. In particular, the Toxicity Characteristic Leaching Procedure (TCLP) is currently used to characterize the toxicity of waste.If the waste is classified as hazardous, it will usually trigger land ban restrictions (LDRs), which call for treatment of the wastes to RCRA cleanup levels prior to disposal at an RCRA landfill and within 90 days of removal using the RCRA best demonstrated available technology (BDAT) or other technologies capableof reaching the same goals.

B. SoilsandSediments Table 2 presents RCRA and New Jersey soil standards. RCRA action levels dictate when a corrective measures study (CMS) is to be performed to evaluate solutions for a site cleanup. RCRA cleanup levelsare determined ona case-by-case basisby USEPA based on potential risk to human health and the environment. When determining levels of constituents in soils, it is important to examine the natural levels (i.e., background concentrations)of constituents forthe area being investigated. Possible sources of interference that can produce elevated background conditions mayinclude runoff from roadways (which can contribute lead and other metals, semivolatile compounds, and

mg/kg)

Wilson et al.

56

Table 2 SoilandSedimentStandards New Jersey cleanup standardsb

Subsurface Nonresidential Residential RCRA soil soil surface soil surface levels' action Parameter

(mag) ~

Arsenic Benzene Benzo[a]pyrene' Cadmium Chlordane Chromium 4,4-DDT Endrin Lead Lindane Nickel PCBs Selenium Tetrachloroethylene Toluene Trichloroethylene 1,1,1 -Trichloroethane Vinyl chloride Xylenes Zinc

8.00E + 01

4.00E + 01

5.00E - 01 4.00E + 02(Cr6) 2.00 + 00 2.00E + 01 5.00E - 01 2.00E + 03 9.00E - 02

-

-

1.00E + 01 2.00E + 04 6.00E + 01 7.00E + 03

2.00E + 05 -

2.00E + 01 3.00E + 00 6.60E - 01 1.00E + 00

2.00 + 01 1.30E + 01 2.50E - 01 1.00E + 02

2.00E + 00 1.70E + 01 1.00E + 02 5.20E - 01 2.50E + 02 4.50E - 01 1.00E + 00 9.00E + 00 1.00E + 03 2.30E + 01 2.10E + 02 2.00E + 00 3.60E + 02 1.50E + 03

9.00E + 00 3.10E + 02 6.00E + 02 2.20E + 00 2.40E + 03 2.00E + 00 1.00E + 03 3.70E + 01 1.00E + 03 1.00E + 02 3.80E + 03 7.00E + 00 6.30E + 03 1.50E + 03

-

-

1.00E + 00 1.00E + 02

1.00E + 02 5.00E + 01 -

1.00E + 00

-

1.00E + 02

-

1.00E + 00 5.00E + 02 1.00E + 00 5.00E + 01 1.00E + 00 1.00E + 01

'USEPA, ECRA Corrective Action Proposed Rules; FR30798/27 July 1990. bNJDEPE Cleanup Standardsfor Contaminated Sites-Proposed Rule; N.J.A.C. 7:26D NJ Register 3 February 1992. CBenze[a]pyrene (BaP) is presented to be representative for semivolatiles.

chlorides), roof drains (semivolatiles), and small amounts of debris on the ground surface (a small piece of plastic can result in a positive result for phthalates).

C. Groundwater and Surface Water

Table 3 presents FDWC (40CFR 141 et al), RCRA, and New Jersey water standards.As with soils and sediments, runoff from roadways can cause interferences with surface water results and some groundwater results. Other possible interferences include septic tanks, sewer lines, and water lines, which can sometimes contribute chloroformand lesser amounts of other organics that canbe formed during the chlorination of surface water for municipal water systems. One common pitfall is use to municipal waterfor drilling and installing a monitoring well without testing the municipal water along with other background samples. Many false positives for chloroform have been overlooked thisway.

D. Air Since theearly years of air quality management, air qualityrules and regulations inthe United States have been based on a set of air quality standards known as the National Ambient Air Quality Standards,or NAAQS. The NAAQS represent a maximum concentration or "threshold

Evaluation Engineering

57

forSites Contaminated

Table 3 GroundwaterandSurfaceWaterStandards Federal drinking water criteria Parameter (m@.)) Amnic Benzene Benzo[a]pyrene' Cadmium Chlordane Chromium 4.4-DDT Endrin Lead Lindane Nickel PCBs Selenium Tetrachloroethylene Toluene Trichloroethylene 1,l ,l-Trichloroethane Vinyl chloride Xylenes Zinc

MCUj (mg/kg) m (& )$

-

OBOE + OOb OBOE + 00" 5.00E - 03' O.WE + 00" 1.00E - 01'

-

2.00E - 03d OBOE + 00' 2.00E - 0 4 '

-

0.00E + 00" 5.00E - 02' OBOE + 00" 1.00E + 00" OBOE + 0Ob 2.00E - O l b O.WE + 0Ob 1.00E + 01'

-

New Jersey standards cleanup RCRA

levelsg action MCL (mg/L)

for Class IIA GWh

5.00E - 02" 5.00E - 03b 2.00E - 04d 5.00E - 03' 2.00E - 03' 1.00E - 01'

5.00E - 02 5.00E - 03

-

-

-

5.00E - 03 3.00E - 05 5.00E - 02 (CR6) 1.00E - 04 2.00E - 03d 2.00E - 03 1.50E - 02 1.50E - 02' 2.00E - 04 4 ' 2.00E - 0 7.00E - 01 5.00E - 06 5.00E - 0 4 ' 5.00E - 02' 5.00E - 02 5.00E - 03' 7.00E - 04 1.00E + 01 1.00E + 00" 5.00E - 03 5.00E - 03b 2.00E - Olb 3.00E - 00 2.00E - 03 2.00E - 03b 7.00E + 01 1.00E + 01'

8.00E - 03 1.00E - 03 2.00E - 02 4.00E - 03 5.00E - 04 1.00E - 01 1.00E - 04 2.00E - 03 1.00E - 02 2.00E - 04 1.00E - 01 5.00E - 04 5.00E - 02 1.00E - 03 1.00E + 00 1.00E - 03 3.00E - 02 2.00E - 03 4.00E - 02 5.00E + 00

'USEPA, National Primary and Secondary Drinking Water Regulation; FR 5956924 December 1975. bUSEPA, National Primary and Secondary Drinking Water Regulations; FR 256 90/8 July 1987. 'Benzo[a]pyrene (BaP) is presented to be representative for semivolatiles. dUSEPA, National Primary and Secondary Drinking Water Regulation; F R 31776/17 July 1992. 'USEPA, National Primary and Secondary Drinking Water Regulation; FR 3526/30 January 1991. 'USEPA, National Primary and Secondary Drinking Water Regulation FR/7 May 1991. WSEPA, RCRA Corrective Action Proposed Rules; FR 30798/27 July 1990. hNJDEPE, Cleanup Standards for Contaminated Sites-Proposed Rule; N.J.A.C.7:26D NJ Register 3 February 1992.

level" of a pollutant in the air above which humans or the environmentmay experience some adverse effects. The actual threshold levelsare based on years of epidemiological,health, and environmental effects research conductedby the USEPA. The USEPA has developed twotypes of NAAQS: primary standards, which are set at levels that are designed to protect the public health, and secondarystandards, which are designed to protect the public welfare (such as vegetation, livestock, building materials, and other elements of the environment). TheNAAQS differentiate between the effects from short-term exposureandthosefromlongertermexposure to air pollutants. Thus, there are short-term NAAQS based on l-hr or 8-hr average concentrations and long-termNAAQS based on annual concentrations. Because the NAAQS representnumerical criteria, the reports inwhich the USEPA presentsinformation on thedevelopment of an NAAQS are called criteria documents. The pollutantsfor which the USEPA hasdeveloped standards are thus known as criteria pollutants.

58

Wilson et al.

Table 4 National AmbientAir Quality Standards Primary

Secondary

(pg/m3)

NAAQS @@m3)

NAAQS Averaging period pollutant Criteria PM10 (particulate matter) SO2 (sulfur dioxide)

NO, (nitrogen dioxide) Ozone CO (Carbon monoxide) Lead

Annual 24-hr Annual 24-hr 3-hr Annual 1-hr 8-hr l-hr

50 150 80 365

-

100 235 10,Ooo 40O , oo 1.5

50 150

1,300 100 235 10,Ooo 40,Ooo 1.S

The USEPA has establishedNAAQS for six compounds sincethe concept of NAAQSwas established in the CAA amendments of 1970. Table 4 lists these six criteria pollutants and their standards. In Table 4, the NAAQS concentrations are expressed in terms of micrograms per cubic meter (pg/m3);however, you will sometimes see them expressed in terms of parts per million (ppm). This table also presents some additional information regarding thecriteria pollutants.

IV. SUMMARY OF REMEDIAL ENGINEERING TECHNOLOGIES This section presents the identification and evaluation of many available remedial engineering technologies for various environmental media. This presentation is not intended to be comprehensive, but it can serve as a general summary of common remedial technologies. The technologies are grouped into the following broad categories:

No Action. Monitoring and inspection technologies that do not contribute to actual remediation of site conditions. Institutional Actions. Indirect methods of reducing exposure to site hazards. Containment. Physical isolation of solid waste, groundwater, and/or other affected media. Removal. Physical removal of solid waste, groundwater, and/or other affected media. Treatment. Alteration of solid waste, groundwater, and/or other affected media to reduce the toxicity, mobility, or volume of site constituents. Disposal. Placement of solid waste, treatment residuals, and/or affected media into a secure disposal facility, or discharge of treated water to the environment. If site remediation is required, two or more technologies may be used in combination to provide a comprehensive approach to site cleanup. An example of combining technologies would be the use of a treatment technology to reducethe toxicity and volume of affected material combined with a containment technologyto reduce the mobility of residual constituents in the treated product. Remedial technologiesare identified for each of the categories listed abovefor the various environmental media found at contaminated sites. The identified technologiesare then evaluated to present some of the major advantages and disadvantages associated with each one. Where appropriate, specific data to be collected to facilitate further evaluation andselection of a remedial technology are also presented. c

Engineering Evaluation for Sites Contaminated

59

Table 5 Notes on the Criteria Pollutants

PM 10 is composed of solid particles, less than 10pm in diameter, that are small enough to be inhaled by humans. The particles can be composed of any number of compounds, from road dust to heavy metals, depending on the source. Sulfur dioxide(SO,) is a compound formed both naturally and by the combustion of fossil fuels. Exposure to higher concentrations ofSO, can cause respiratory problems for some people. When combined with water, SO, forms sulfur compounds that are one of the main components of acid rain. Nitrogen dioxide (NO,) is another compound formed by the combustion of fossil fuel that contributes to the formation of both acid rain and smog. NO, is one of several nitrogen oxide (NO,) compounds present in the atmosphere. The NAAQS is established for NO,, but NO, is more commonly measured. Ozone itself is not emitted directly into the air but rather is formed through a series of complex physical and chemical reactions in the atmosphere. Therefore, discussions about the criteria pollutant ozone often focus on a group of gaseous pollutants known as volatile organiccompounds or VOCs, which are carbon-based, organic compounds that tend to evaporate into the air easily. Solvents, cleaners, and paints are among the hundreds of compounds in use that contain VOCs. In the presence of sunlight, VOCs and other chemical compounds, including NO,, react to form ozone. Carbon monoxide (CO) is a colorless, odorless, and tasteless gas that occurs naturally in the atmosphere and is also formed in the combustion of fossil fuel. teud (Pb) is a common metal that is released into the atmosphere from a number of sources, including burning of leaded fuels. Lead has the lowest NAAQS, since exposure to low levels can present health problems.

A list of the technologies presented in this section and the applicable environmental media is presented in Table 6. A brief description and evaluation of the potentially applicable technologies identified in this section are presented in Table 7 . A summary of the major data requirements identified for each technology is presented in table 8.

A. No ActionTechnologies “No action’’ implies that no remedial actions are to be conducted on the media of concern. Actions such as groundwater monitoring and site inspections are included as no action technologies because theyare intended to detect changes in site conditions rather than to actually remediate existing contamination. Monitoring Description. Long-term, periodic groundwater and/or surface water monitoring to detect changes in site conditions, such as the migration of constituents in groundwater, due to natural processes. Evaluation. Although groundwater monitoring does not reduce site constituents, it is a proven method of detecting changes in site conditions and is commonly required as a component of remediation. Site Inspection Description. Long-term, periodic site inspections to detect visible changes in site conditions due to natural processes. Evaluation. Although site inspections do not reduce site constituents, they are effective in site conditions andare commonly requiredas a component of for identifying visible changes remediation.

60

Wilson et al.

Table 6 Potential Corrective Measure Technologies and Applicable Environmental Media Applicable Media ~ _ _ _ ~

Technology No Action Monitoring Site inspections Institutional Actions Physical barriers Deed restrictions Containment Storm water controls Capping Vertical barriers Filter barriers Subsurface drains Removal Excavation Dredging Recovery wells Interceptor trench Vacuum extraction Treatment Air stripping (soil) Biological (soil) Asphalt batching Soil flushing Stabilization Incineration Air stripping (water) Biological (water) Chemical precipitation GAC adsorption Ion exchange Oxidation-reduction Steam stripping Filtration Neutralization Off-site water treatment Disposal On-site landfill Off-site landfill Surface water discharge Reinjection

Solid waste

Soil

Sediment

Groundwater

Surface water

X

X X

X X

X X

X

X X X

X

X

X X X X X

X

X

X X

X X

X

X X X X

X

X

X X X X

X

X

X X

X X X X X X X X

X X

X X

X X

X X X

X

X

X X X X

X X X X

X X

X

X X

Evaluation Engineering

61

forSites Contaminated

Table 7 Summary of Potential Corrective Measure Technologies Technology

No Action Monitoring

Site inspections

Institutional actions Physical barriers

Deed restrictions

Containment Storm water controls

Capping

Subsurface vertical barriers

Filter barriers

Long-term periodic groundwater and/or surface water monitoring. Long-term periodic inspection of the site.

Does not reduce conDetect changes in site stituents. Long-term conditions. Monitor expense. effectiveness of corrective measures. Can help detect changes D o e s not reduce constituents. Long-term in site conditions. Enexposure. sure continued effectiveness of corrective measures.

Physical barriers such as a chain-li& fence or vegetation around waste areas.

Periodic maintenance Reduce risk of exposure to site constituents and inspection required. Does not reby restricting site acduce constituents. cess. Reduce risk of site disturbanceby unauthorized intruders. Restricts future usability Reduce hypothetical of the site. risks from future land use and/or groundwater use. Reliability is dependent upon continued enforcement.

Legal limitations placed on future property and/or groundwater use.

May require some Reduces the potential waste disturbance. for erosion of coversoils. Can reduce infiltration of water and migration of constituents. Does not reduce conMinimizes surface waConstruction of a soil stituents. Could ter infiltration and conor multilayer cap to restrict future stituent migration. contain waste areas. development at Reduces risk of contact the site. with waste. Reduces watedwind erosion. Could disturb adjacent Low-permeability verti- Reduces mobility of wetland areas. Effecconstituents and subsecal subsurface barrier tiveness could be quent risk of exposure. to groundwater flow low. Relatively high Could increase the efsuch as a slurry wall. cost. fectiveness of groundwater recovery. Effectiveness is not Could reduce cation A subsurface wall or proven. Could concentrations in surface blanket of require periodic groundwater. Could glauconitic sands replacement. reduce migration of with a high capacity constituents without for adsorbing metal affecting wetlands. cations.

Improve storm water drainage by regrading, vegetation, swales, or pipes.

cription

62

Wilson et al.

Table 7 (Continued) Technology Subsurface drains

Removal Excavation

Dredging

Groundwater recovery wells

Interceptor trench

Vacuum extraction

Treatment (soil/waste) Air stripping

Subsurface drain or trench upgradient of disposal facilities to reroute clean groundwater.

Can be used to lower groundwater levelsto below waste levels to reduce groundwater infiltration and constituent migration.

Limited effectiveness. Could require significant disturbance of the site for construction.

Removal of solid waste and/or affected soils.

Required for subsequent Treatment and/or distreatment and/or disposal Would be reposal of solid waste. quired. Risks Removes potential associated with waste sources of constituexcavation and transents from the site. portation. Removal of affected Reduces potential risks Dredged material would sediments from water associated with afrequire disposal and fected sediments. courses. may require treatReduces potential ment. Dredging acfor migration of contivities would disturb stituents and risk of aquatic life and could exposure. increase constituent migration. Reduces potential miCollection of affected Requires long-term gration of constitugroundwater with operation and mainents via groundwater. recovery wells. tenance. Site condiFacilitates groundtions must be water treatment. suitable for this technology to be effective. Requires long-term A subsurface trench to Reduces potential mioperation and maingration of constitufacilitate groundwater tenance. Site condicollection. ents via groundwater. tions must be Facilitates groundsuitable for this techwater treatment. nology to be effective. Economically limited to shallow depth. Effective for the recov- Not effective for recovExtraction of VOCs ery of heavy metals. from pore spaces in ery of VOCs from Requires high porossoils. Does not resoils. quire excavation of ity and low moisture waste or affected content. Requires air material. treatment. Mechanical screening of soils to increase effective surface area.

Effective for theremoval of VOCs from soils. Implementation is fairly easy. Proven technology.

Not effective for heavy metals or semivolatiles. Air emission controls would be required.

63

Engineering Evaluationfor Contaminated Sites Table 7 (Continued)

Advantages

Technology organic Biological Breakdown of

Not effective for removal of heavy metals. Effectiveness can be lowered by a number of site conditions. Limited to certain types of petroleum of constituents. Aphydrocarbonplicable only for containing soils in coarse-grain soils. asphalt pavements. Regulatory agencies may restrict applications. Unrecovered solution Proven to be effective In situ injection of can contribute to for the removalof flushing solution to groundwater degradaheavy metals. Does facilitate collection tion. Only practical not require waste of constituents. for a limited range of excavation. soil conditions. Increases waste volStabilization of constit- Proven to be effective ume. Not effective for reducing the uents by mixing with for volatile organics. leachability of heavy stabilizing agents. Does not destroy metals. Can be conconstituents. ducted in situ. Readily available. Combustion of organic Proven effective for the Does not destroy heavy metals. Residual ash destruction of organic constituents with may require further constituents. Would high-temperature oxitreatment before disremove constituents dation, on-site or posal. Very expenfrom the site. off-site. sive. constituents by microorganisms.

Asphalt batching Reuse

Soil flushing

Stabilization

Incineration

Treatment (water) Air stripping

Disadvantages

Effective for the removal of many organic solvents and petroleum hydrocarbons. Beneficial reuse of affected media. Relatively low cost.

Removal of VOCs from liquids with an airstripping column or other mechanical facilities

Effective for the removal of VOCs from water. Easily implemented and well proven.

Biological

Breakdown of organic constituents by microorganisms.

Effective for the removal of many organic solvents.

Chemical precipitation

Commonly used for the Alteration of pH to reremoval of metals duce the solubility of from water. Proven constituents and facileffectiveness. Could itate precipitation. be used in combination with other treatment technologies.

Not effective for the removal of heavy metals. Air discharge permit would be required. Air treatment may be required. Not effective for removal of heavy metals. Effectiveness can be lowered by a number of site conditions. Recovered sludge may require treatment. Not effective for removal of organic constituents.

cription

64

Wilson et al.

Table 7 (Continued) Technology GAC adsorption

Passing a waste stream through activated carbon to remove constituents by adsorption.

Ion exchange

Removal of toxic metal ions from waste streams.

Oxidation-reduction

Alteration of waste Could be effective for stream to reduce toxthe removal of mercury from waste icity or solubility or to create a waste that streams or affected is easier to handle. waters. May be applicable to the organic constituents at the site. Use of superheated More effective than air steam to stripVOCs stripping for less volatile organics. More from water. cost-effective than air-stripping for highVOC concentrations. Proven effective for the Removal of suspended removal of suspended solids by passing solids. Could be used through a porous in combination with medium. other treatment technologies. Easily implemented and Adjustment of pH to well proven. Relareduce corrosiveness tively inexpensive. and acidity. Could be used in combination with other treatment technologies. Off-site facilities exist Treatment of affected for the treatment of a water at an off-site wide variety of contreatment facility. stituents. A new treatment facility would not have to be built. Best for small quantities of liquids.

Steam stripping

Filtration

Neutralization

Off-site water treatment

Well suited for the removal of VOCs, and has some effectiveness for mercury removal. Readily available and well proven technology. Proven effective for removal of heavy metals.

May not be effective for all site constituents. GAC unit requires regeneration and/or disposal. May not be costeffective. Spent reagent solutions may require treatment. Appropriate reactions must be determined. May not be effective for some combinations of constituents.

Not effective for the removal of heavy metals. Stripped effluent may require treatment before discharge. Not effective for dissolved constituents. Filter backwash would require treatment. Limited effectiveness.

Off-site treatment may not be practical for large volumes of waste. Some risks involved with waste transportation.

Evaluation Engineering

65

forSites Contaminated

Table 7 (Conrinued) Technology Disposal On-site landfill Placement treated of

or

untreated wastes in a secure disposal facility on site. Off-site landfill

Transportation and disposal of untreated waste or treatment residues at an approved off-site landfill.

Surface water discharge

Discharge of treated water to an existing surface water body.

Reinjection

Injection of treated water into the ground through injection wells or infiltration galleries.

Could provide for secure containment of wastes and/or treatment residues. Reduces mobility of constituents and subsequent risk of exposure.Removes constituents from the site. Increases potential for future use of the site. Proven means of disposing treated water. Surface water body is readily available. Has been proven to be effective.

Construction would involve high costs and site disturbance. Long-term maintenance would be required. Limited off-site landfill capacity. Potentially high costs. Potential long-term liability. Untreated materials may be restricted from land disposal. NPDES permit would be required. Periodic sampling and maintenance would be required. May not be practical in low-permeability aquifers. Groundwater modeling may be required. Could increase constituent migration.

B. InstitutionalActions Institutional action technologies reduce potential exposures to site constituents by indirect methods rather than by containment or treatment of the affected media. Institutional actions include physical barriers and deed restrictions. Physical Barriers Description. Physical barriers provide an easily implemented, low-cost method for restricting pedestrian and animal traffic across areas of concern, thus decreasing the potential for exposure to site media or damage to on-site storageor containment structures. Physicalbarriers could range from chain-link security fencing to closely grouped rows of obstructive vegetation. be required to maintain the integrity of the barrier. Periodic inspection and maintenance would Evaluation. Physical barriers do not reduce site constituent levels, but they canbe effective for protecting human health and the environment by preventing exposureto affected media. Deed Restrictions Description. Deed restrictions place legal limitations on future property use. These restrictions can prohibit future property and/or groundwater uses that could result in increased exposure to site constituents. Deed restrictions can be easily implemented, but their effectiveness is dependent upon continued enforcement.

Wilson et al.

66 Table 8 Potential Corrective Measure Technologies and Applicable Data Requirements Required data items"

No Action Monitoring Site inspections Institutional Actions barriers Physical restrictions Deed Containment Storm water controls Capping barriers Vertical Filter barriers Subsurface drains Removal Excavation Dredging Recovery wells Interceptor trench

I

Air stripping (water) Chemical precipitation GAC adsorption Ion exchange Oxidation-reduction Steam stripping Filtration Neutralization Off-site water treatment Disposal landfill On-site Off-site landfill Surface water discharge Reinjection

.. . 0

* I

l

1 . 1 .

.

.

Evaluation Engineering

forSites Contaminated

67

Table 9 Description of Data Items Groundwater chemistry(Conr.) Topographylsetting Conductivity Ridges Dissolved oxygen Valleys Nutrients Hills Chemical oxygen demand Surface water bodies Total organic carbon Forests Constituent concentrations Wetlands Heavy metals Drainage patterns Source area characterization Buildings Location Vegetation “ypeldesign Climatological data Operating practices and Evaporation history Evapotranspiration Physical condition Temperature Wind speed and direction Age Method of closure Precipitation Type of wastelclassification Atmospheric pressure Volume/extent Relative humidity Constituent chemical properties Regionallsite geology Geologic units PH Molecular weight Strata Hydrolysis Strike and dip Chemical class Folding Viscosity Faulting Solubility Depositional history Oxidation/reduction Soillrock types potential Aquifer characteristics Vapor pressure Hydraulic conductivity Sorption Porosity Biodegradability Grain size distribution Photodegradability Saturated and unsaturated Chemical transformations zones Migration potentiall Attenuation characteristics leachability Extent Constituent physical properties Depth Physical form Thickness Potential migration pathways 5pe Temperature Rechargeldischarge areas Density and amounts Boiling point Aquifer leakagelinteractions Soil physical properties Groundwater Flow Soil classification Water level contours Grain size distribution Vertical and horizontal flow Soil profilelstratigraphy TidaVseasonal influences Permeability Man-made influences Density Groundwater chemistry Porosity PH Moisture content Total dissolved solids Infiltration Total suspended solids Storage capacity Biological oxygen demand Mineral content Alkalinity

Soil physical properties(Cont.) Settlement potential Soil index properties Erosion potential Soil chemical properties PH Organic content Sorptive capacity Ion-exchange capacity Constituent types Constituent concentrations Surface water characteristics Location Elevation Area Depth Velocity Width Inflowloutflow Temperature Seasonal fluctuations Flood plain Stream cross sections Surface water chemistry PH Total dissolved solids Total suspended solids Biological oxygen demand Alkalinity Conductivity Dissolved oxygen Nutrients Chemical oxygen demand Total organic carbon Constituent concentrations Heavy metals Sediment characteristics Depositional area Thickness Grain size distribution Density Organic carbon Ion exchange capacity PH Constituent extentlvolume Horizontal and vertical extent in waste Horizontal and vertical extent in soil Horizontal and vertical extent in sediments

68

Wilson et al.

Table 9 (Continued) Constituent extent/volume (Cont.) Horizontal and vertical extent in groundwater Horizontal and vertical extent in surface water Total volume Concentration profiles Constituent migration Horizontal and vertical migration direction

Contituent migration (Cont.) Horizontal and vertical migration rate Factors influencing migration Potential future movement Potential receptors Age Location Use Population

Potential receptors (Cont.) Ecology Endangered species 'Iteatability Study Bench-scale tests Residual characteristics TCLP Waste samples SoiYsediment samples Water samples

Evaluation. If enforced, deed restrictions can be effective in reducing the potential for disturbance of affected site media.

C. Containment Containment technologies reducethe potential for direct exposure to site constituents and the potential for their migration by physically isolating the affected media or wastes. Storm Water Controls Description.Stormwatercontrolssuch as surfaceregrading,increasedvegetation, drainage swales, and drainpipes can be used to improve the drainage of surface water away from waste disposal facilities such as landfills and impoundments. Improved drainage can reduce the potentialfor erosion of cover materials, reduce infiltration of surface water, and minimize the potential for ponding of surface water. Evaluation. This technology can be effective for reducing the potential volume of leachate produced by infiltration of surface water. Surface water controlsalso minimize erosion and subsequent migrationof constituents. Data Requirements. Data needs include: topography (extent of wetlands, existing drainage patterns, drainage area), soilphysical properties (erosion potential), andclirnatological data. Caps and Liners Description. Capping is a common containment technology thatis used to prevent direct contact with wastes, reduce the infiltration of surface water and subsequent leaching of constituents, prevent erosionof waste materials, and control surface runoff. Low-permeability caps such as clay caps and multilayer capsare the most effective caps for waste containment. Multilayer caps typically consistof an upper vegetative layer underlainby a drainage layer, an impermeable synthetic membrane liner, and a low-permeability clay layer and are usually more effective than clay caps at restricting surface water infiltration. Evaluation. This technology is one of the most common technologies for the containment of hazardous waste, and it has been proven to be effective. Multilayer caps are appropriate for improving waste containment and reducing the potential for leaching of constituents. Data Requirements. Data requirements include topography (extent of wetlands, surface slopes, vegetation conditions), climatological data, source areacharacterization (extent, depth, and volume of waste), constituent chemicaland physical properties, and soil physical properties (infiltration, permeability of existing covers and waste, bearing capacity of waste, potential settlement).

Evaluation Engineering

forSites Contaminated

69

Vertical Barriers Description. Subsurface vertical barriers (e.g., slurry walls, membrane walls, grout curtains, sheet piling) can contain land-disposed wastesby restricting the lateral flow of groundwater through the waste, thereby reducing the potential for migrationof constituents into the groundwater. Vertical barriers can also allow pumping within the contained area without significant water level drawdown of nearby water bodies (rivers, lakes, etc.). Evaluation. Vertical barriers have been used successfully in a number of waste containment applications. Vertical barriersare most effectiveat shallow depths and when they canbe keyed (or tied) into a confining substratum that can restrict the flow of groundwater beneath the barrier. Because vertical barriers restrict the flow of groundwater, they can impact nearby ecosystems such as wetlands, rivers, and streams. Data Requirements. Data requirements include topography (extent of wetlands), aquifer rate characteristics (physical propertiesof confining layer, Columbia Aquifer profile, discharge to surface water), groundwater flow (depth to water table, groundwater flow rates and direction, hydraulic gradient), and constituent extent and migration. Filter Barriers or ablanketfilterbarriercould be created Description. A verticalwallfilterbarrier by the placement of natural sediments having high cation-exchange capacity (e.g., glauconitic greensand) in a vertical trench or as a blanket, respectively. Glauconitic greensands are sediments with a high capacity for adsorbing cations (e.g., lead, mercury) from liquid solutions. Evaluation.Bench-scaleexperimentshavedemonstratedthatgreensanddepositsfrom the Delaware Coastal Plain can remove heavy metals from spiked water samples and landfill leachates. Greensands havebeen shown to retain more mercury from basic (high pH) solutions than from acidic (low pH) solutions.To date, the effectivenessof natural greensand filter barriers has not been proven in field applications, and a pilot field study wouldbe required prior to full-scale implementation. Data Requirements. Data needs include sources of greensand or other media with high cation exchange capacity; extent of groundwater degradation, if any; extent of wetlands; depth to water table; groundwater flow rates and direction; hydraulic gradient; aquifer profile; disbe required to charge to surface water; and surface water flow rates. Bench-scale testing would evaluate this technology on a site-specific basis. Subsurface Drains Description. A subsurface drain or drains could be constructed hydraulically upgradient of any disposal facility impacting groundwater to lower the groundwater table to a level below the bottomof the impoundment(s). This technology is considered a containment action because it physically isolates the waste from the underlying groundwater and reduces the potential for leachateproductionand/orleachatemigration.Groundwatercollectedintheupgradient drain(s) could be discharged to a nearby stream or storm water drainage system. Subsurface drains can also be used to collect groundwater affected by site constituents. This technology is further discussed in Section D under groundwater removal. Evaluation. If the groundwater table is periodically above the bottom of a landfill or disof posal impoundment, this technology couldbe effective for reducing the potential migration constituents into groundwater. Data Requirements. Data requirements include depth to groundwater table, groundwater flow rates, depth of impoundment bottoms, soil and waste hydraulic conductivities or transmissivities, groundwater gradient, and aquifer profile.

70

et

Wilson

al.

D. Removal Removal technologies focus on the physical removal of affected media from the site, usually to facilitate treatment and/or disposal. Solid Waste, Soil,andlor Sediment Removal Excavation Description. Excavation is applicable for the removal of solid wastes and affected soils and sediments. If materials must be removed, they can typically be excavated using standard methods andpractices, although the volume of material to be removed could necessitate staged excavation or other special handling requirements. Evaluation. Although excavation alone is not a remedial technology, it may be required in conjunction with other treatment and/or disposal actions. Fugitivedust and volatilization of constituents into theair during excavation could pose short-term risks to human health and the environment. Data Requirements. Data requirements include extent, depth, and volume of waste and/or affected soils and sediments; soil physical properties; and constituent concentrations. Dredging Description. Dredging is the removal of sediments from under water, which may be accomplished using hydraulic systems, clamshells, or other means of excavation. Evaluation.Dredgingcaneffectively remove affectedsedimentsfromsurfacewater courses. However, dredging operations can disturb aquatic life and potentially result in increased mobilization of constituents downstream. DataRequirements.Datarequirementsincludestreamflow rates, stream profile,and sediment characteristics (extent, depth, and volume of affected sediments). GroundwaterRemoval Recovery Wells Description. Affected groundwater can be collected with pumping wells located in the affected aquifer, hydraulically downgradient of the constituent source area(s). By intercepting groundwater, the migration of constituents from the source areas can be restricted. Collected groundwater would probably require some type of treatment prior to discharge. Evaluation. Recovery wells can be effected for long-term groundwatercollection in certain aquifers, although this technology may not be appropriate under all site conditions. Recovery well systems are most effective when measures are taken to remediate the source(s) of constituents migration. DataRequirements. Data requirementsincludedepth to water table, groundwater flow rates and direction, hydraulic gradient, aquifer characteristics, and extent of affected groundwater. interceptor Trench Description. An interceptor trench is typically constructedby excavating a narrow trench into a stratum of concern, placing a perforated drainpipe alongthe bottom of the trench, and backfilling the trench with some type of drainage aggregate. The drainpipe conveys waterby gravity to a collection sump, and the water collectedin the sump is pumped out for treatment and/or discharge to surface water. Evaluation. Interceptor trenches havebeenfound to be effective for the collection of shallow groundwater in unconsolidated aquifers, but they are not appropriate for all site conditions. At shallow depths(10-20 ft), interceptor trenchesare typically more cost-effective and perform better than recovery wells in unconsolidated aquifers.

Evaluation Engineering

for Sites Contaminated

71

Data Requirements. Data requirements include depth to water table, groundwater flow rates and direction, hydraulic gradient, and aquifer characteristics. Vapor Extraction Description. A vacuum extraction system consists of a network of vapor withdrawal (or vacuum) wells installed in the vadose (unsaturated) zone to remove volatile organic compounds (VOCs). A vacuum is applied to promote movement of air and VOCs to the extraction well points. An in-line water removal system can be provided to remove condensate and limited quantities of recovered groundwater. Recovered vapors are treated in an in-line vapor-phase carbon adsorption system (forVOC capture) or an exhaust fume incineration system (forVOC destruction). Steam can be injected through injection wells to enhance the vacuum extraction process. Vacuum extraction is considered for the removal of VOCs from wastes and affected soils. Evaluation. Vacuum extraction has been provento be effective for the removal ofVOCs from the vadose zone. This technology can be used to remediate soils in place, thereby minimizing disturbanceto waste facilities that result from excavation and removal operations. This technology is only effective for removal of volatile constraints. Data Requirements. Data requirements include depth to groundwater, physical and chemical properties of soils and wastes, soil and groundwater quality, and extent of affected soil and waste.

E. Treatment Treatment technologies reducethe toxicity, mobility, or volume of affected media or wastes, thus reducing the potential for constituent exposure to human health and the environment. Removal and disposal technologies may be required in conjunction with treatment alternatives, although some treatment technologies can be implementedin situ. Solid Waste, Soil, andlor Sediment Treatment Air Stripping Description. Air stripping of volatile organic compounds (VOCs) in soils can be accomplished by a process of mechanical screening that involves the passage of excavated soils through oneor more vibrating screens commonly used in construction projects. In this process, the rate of volatilization is maximizedby soil disaggregation and the resulting increase of effective soil particle surface area. Evaluation. The removal of VOCs from wastes and/or affected soils can reduce the potential for migrationof VOCs into groundwater. The equipment, materials, and labor required to perform VOC stripping of soils are readily available. Excavation of materials is required prior to implementation of this technology, which would disturb the site and potentially result in short-term exposure. Air emission and erosion and sedimentation controls and permits would likely be required. DataRequirements.Datarequirementsincludewastevolume,chemicalandphysical characteristics of constituents, and soil physical properties. Biological Treatment Description.Biologicaltreatment,sometimesreferred to as bioremediation,generally refers to the breakdown of organic constituents by microorganisms. The most common processes are based on aerobicor anaerobic bacteria, suchas those processes used in the treatment of muncipal wastewaters. In situ, pump-and-treat, solid-phase, slurry phase, and soil heaping biological treatment techniques have been used to remediate contaminated sludges.

72

Wilson et al.

Effectiveness. The effectiveness of biological treatment can be influenced by a number of parameters, including pH, temperature, availability of nutrients, and the presence ofheavy metals. Some biological treatment methods have been successful for the treatment of certain organic solvents,but this technology is not effective for the removal of heavy metals from solid waste or soils. Data Requirements. Data requirements include physical chemical characteristics of soils and physical and chemical properties of constituents. Asphalt Batching Description. Asphalt batching can be used as an alternative to landfillingor on-site treatment of soil containing petroleum hydrocarbons to recycle it into useful products such as asphalt paving material. After testing for petroleum hydrocarbon content (and the presence of hazardous substances), the soil is delivered to the batch plant, where it is crushed and sieved through screens to remove wood, metal, or other undesirable debris. Next, the soil is passed through a gas- or oil-fired rotary kiln, where it is heated to approximately 350°F, which evaporates all the water and burns the petroleum hydrocarbon components. Thereafter, the soil is blended with other aggregates and asphalt and delivered to the paving location. Evaluation. Recycling by asphalt batching removes the contaminants from the original site, destroys undesirablematerials, and providesa valuable constructionmaterial. Reasonable safety cautions during the initial haulingand storage and adequate quality testingare the only major technical difficulties related to asphalt batching. Local regulatory agencies may have administrative control and shouldbe contacted prior to using this process. This process is best for sandy soils rather than clay soils, because a high percentage of clay is not desirable in asphalt paving materials. Data Requirements. Data requirements include chemical types and concentrations, physical properties of the soil, particularly grain size analysis, and the extent of constituents. Soil Flushing Description. Soil flushing involves the in situ injection or percolation of large volumes of flushing solution to an area of waste and/or soil requiring remediation and the subsequent collection or recovery of the flushing solution. The solution is intended to dissolve or precipitate constituents as it passes through the affected media. Water is a common flushing solution, although aqueous surfactant solutions and organic solvents have also been used. Well points, subsurface drains, or anothertype of collectionsystemmustbe installed to collect the constituent-laden solution. The recovered solution requires treatment. Evaluation. Soil flushing has been proven to be effectivefor the removalof heavy metals from solid wastes and soils. Soil flushingmay not be effective for all organic constituents,but the recovered groundwater and flushing solution can typically be treated with an alternative method for the removal of organics. Cloggingof soil pores can limit the flushing of the soils, thereby reducing the effectiveness of the treatment and preventing recoveryof the flushing solution. Unrecovered flushing solution can contribute to increased groundwater degradation. Data Requirements. Data requirements include chemical and physical properties of soil, aquifer characteristics, groundwater flowrate and direction, physical and chemicalproperties of constituents, and the extent of constituents. Stabilization Description. Stabilization technologies have been used to immobilize organic and inorganic compounds in wet or dry media, using reagents to produce a stable mass. Stabilization methods include cement-based methods, silicate-based (pozzolanic) methods, thermoplastic methods, organic polymer methods, and others. Waste materials and/or affected soils can be mixed in place with shallow or deep soil mixing systems, or the affected materials can be ex-

Evaluation Engineering

forSites Contaminated

73

cavated and consolidated into oneor more locations before mixing. 'I)lpically, this technology does not destroy constituents but incorporates them into a dense, homogeneous, low-porosity structure that reduces their mobility. Stabilization has been proven to be effective for immobilizing inorganic constituents (e.g., metals) but hashad limited success for volatile organics. Because a reagent must be added to the soil, the volumeof treated soilmay be greater than the original soil volume by as much as 20-100%. Evaluation. This technology can restrict the migration of constituents from the waste impoundments and reduce the potential for subsequent exposure. It can be implemented in place. However, its applicability to heterogeneous wastesmay be limited, as the proper selectionof a stabilizing reagent for amixed waste may be difficult. DataRequirements.Datarequirementsincludebench-scale or pilot-scaletestresults, leachability of unstabilizedandstabilizedmasses,physicalandchemicalcharacteristics of waste, and the extent and volume of waste. Incineration Description. Incineration is a thermal treatment method that uses high-temperature oxidation under controlled conditions to degrade waste materials into by-products that include carbon dioxide, water vapor, ash, nitrous oxide, sulfur dioxide, and hydrochloric acid gases. Air pollution controlsare typically needed to ensure that air quality standards are not violated. Residual incinerator ashmay require stabilization priorto disposal due to high concentrations of metals. Typesof incinerators that are commonly used for the remediation of solid waste and transsoils include rotary kiln, fluidized bed, and infrared incinerators. Excavated soils can be ported off-site for incineration,or they can be treated with a mobile incinerator assembled on site. Incineration is not effective for the destruction of metalsin solid media. Evaluation. Incineration is a feasible and well-developed technology for the destruction of organic constituents, although residual ash would require stabilization prior to disposal if high metals concentrations are present. Due to cost considerations, off-site incineration is most appropriate for smallto moderate volumes of materials; on-site incineration would only be appropriate for the incineration of large volumes of material. Waste handling and stack emissions could present risks to human health and the environment unless appropriate healthand safety equipment and emission controls are employed. Limited availability and capacity of off-site incinerators can result in lengthy remediation schedules if large volumes of waste are present. Public oppositionand permitting requirements could hinder the sitinganofon-site incinerator. Data Requirements. Data requirements include a treatability study (test burn); ash content, heat value, and halogen content; and physical and chemical characteristics of waste. Liquid Treatment The liquid treatment technologies presented here may be applicable to the treatment of affected groundwater, surface water, or water generated or collected during solid waste treatment. Off-Site Water Treatment Description. Affected groundwater recovered from the siteor water resulting from some type of on-site treatment process is treated at an off-site treatment facility. Water is pumped or hauled via tank trucksto an off-site treatment facilityor discharged to a publicly owned treatment works (FQTW) upon authorization by the POTW. Evaluation. Off-site facilities have been used successfully for the treatment of water containing various constituents.If a large volume of water requires treatment, hauling the to water an off-site facility can be extremely expensive and impractical. If the water is hazardous, a permit is required for off-site transportation. Data Requirements. Data requirements include identification of a facility, pilot-scaletest, ground and/or surface water chemistry, and physical and chemical properties of constituents.

74

Wilson et al.

Air Stripping Description. Air stripping to remove organics from groundwateris most often performed by passing the water through a countercurrent air-stripping column tofacilitate the transfer of volatile organics fromthe liquid phase to the gas(air) phase. Off gas treatmentmay be required for vapor-phase VOC removal. Other methods of air stripping are employed that do not entail countercurrent air andliquidflow(diffused aeration andmechanical aeration); however, packed tower stripping is most widely used. Evaluation. Air stripping has been proven to be effective for the removal ofVOCs from groundwater, although it is not effective for the removal of metals. Stripping efficiency is related to the air-to-water ratio, system temperature, tower packed depth, and volatility of the constituents. An air discharge permit is required, and offgas treatment may be required. Air stripping can be prone to fouling, depending on mineral and bacterial levels in the water. Data Requirements. Data requirements include groundwater and/or surface water chemistry, physical and chemical properties of constituents, and effluent target levels. Biological Description. In situ biotreatment is the aerobic or anaerobic microbial degradationof organic constituents in groundwater. The complete biodegradation process converts organics to carbon dioxide and water (aerobic) or carbon dioxide and methane (anaerobic). In biotreatment, a microbial population (biomass)is maintained, by injection of nutrients and oxygenor methane, to feed on the organic substances present in the water. The microbial populations available for forming the biomass are numerous, and many alternative designs are available for both aerobic and anaerobic systems. Effectiveness. Biotreatment is well documented in attaining high organic removal efficienciesmeasured as COD or BOD reduction. However,some organicsubstances do not biodegrade but are stripped to the atmosphere as the water is treated. Biotreatment methods are most effective for water that has high VOC concentrations. High levels of inorganic cations (specifically calcium, iron, and magnesium) can exert toxic effects on anaerobic treatment processes. Data Requirements. Data requirements include groundwater and/or surface water chemistry, physical and chemical properties of constituents, and effluent target levels. Chemical Precipitation Description.Chemicalprecipitationprocesses are commonlyused to removemetals from water. Chemical precipitation is pH dependent in that acid or base is added to a solution to adjust the pH to a point where the constituents to be removed have their lowest solubility. Frequently a bulking agent such as lime or a soluble salt that forms insoluble metalprecipitates is employed as a precipitant. The precipitant (lime, caustic, or sulfide) is added to the metalcontaining water, and thenthe solution is mixed with a flocculating agent (polymer)to promote solidsagglomeration andsettling.Someorganic or inorganicsubstances (e.g., chelants) present in the water may inhibit precipitation. Evaluation. This technology could be effective for the removal of metals from collected groundwater and/or process streams from other treatment technologies. Sludge produced during the process would have to be dewatered or treated before disposal. Data Requirements. Data requirements include bench-scale (iar) test results (precipitant and/or surface water chemistry; physdoses, setting rates, and sludge production); groundwater ical and chemical properties of constituents; and effluent target levels. Granular Activated Carbon Adsorption Description. Carbon adsorption involves contact of a waste stream with activated carbon, usually by downward flow througha series of packed bed reactors. Molecular adsorption

Evaluation Engineering

forSites Contaminated

75

onto granular activated carbon (GAC) occurs through physical and/or chemical forces in which molecules are held on the surface of the carbon particle. A compound’s affinity for carbon depends on its molecular size and water solubility. Constituents with high molecular weights and low water solubilities exhibit a high affinity for carbon. Evaluation. Granular activated carbon adsorption is well suited forthe removal of VOCs, with the exception of vinyl chloride and chloromethane compounds, and some inorganic constituents such as mercury have also shown a limited potential for adsorption onto GAC. AlthoughGACadsorptiondoesnotproduceanyoff-gases, the spentcarbon would require regeneration or treatment prior to disposal. Adsorption of metals onto GAC may limit the regenerability of the carbon. GAC is also subject to bacterial fouling. Data Requirements. Data requirements include groundwater and/or surface water chemistry and physical and chemicalproperties of constituents. Also, a pilot test would be required to determine the adsorption capacity of the carbon for the organics to be removed and to estimate the carbon bed life. Ion Exchange Description. GAC adsorption is commonlyused for the removal of toxic metal ions from solutions when recoveryof the concentrated metal isfeasible and/or a relatively low proportion of competing non-heavy metal ions are present. The design of an ion-exchange system must consider the specific ionsto be exchanged, the controlof influent bacteria and suspended solids, and the treatment of spent regenerant solutions. Evaluation. Ion exchange could be effective for the removal of metals from collected groundwater and/or process waste streams. Highly concentrated waste streams can usually be separated with more cost-effective technologies. Data Requirements. Data requirements include pilot-scale test results, groundwater and/ or surface water chemistry, and physicaland chemical properties of constituents. Oxidation-Reduction Description. Chemical oxidation-reduction reactions are used to reduce toxicity or solubility or to transform a substance to one that is more easily handled. Chemical reduction is primarily used for the treatment of wastes containing hexavalent chromium. Common reducing agents include sulfurdioxide, sulfite salts, and ferrous sulfate. Chemical oxidationcan be used for the treatment of oxidizable organics. Chlorine (hydrochloride), hydrogen peroxide, and ozone are common oxidizing agents. Ultraviolet light is used as a catalyst to facilitate chemical oxidation reactions. Evaluation. Chemical reduction could be effective for the removal of reducible organics from collected groundwater and/or process waste streams. Chemical oxidation may be applicable to some organic constituents. DataRequirements.Datarequirementsincludepilot-scaletest results, groundwater and/or surface water chemistry, physical and chemical properties of constituents, and benchscale testing. Steam Stripping Description. Steam stripping consists of passing superheated steam countercurrent to a preheated groundwater stream in a packed or tray tower to strip VOCs and other organics into the vapor phase. The off-gases from the columnare routed to a condenser, which can be cooled using groundwaterin a non-contact mode. From the condenser, the condensate flows to a decanter, where the organic layer is drawn off and the bottoms are routed to the stripping tower influent. Effluent from the stripping tower can be routed to a polishing treatment or discharged. The efficiency of stream stripping can be increased by operating the column at a negative pressure.

76

Wilson et al.

Evaluation. Steam stripping will treat less volatile and more soluble wastes than air strip be effective ping andis more cost-effectivefor high VOC concentrations. Steam stripping could for the treatment of collected groundwater and/or process water containing highoflevels VOCs. Stripped effluent may require treatment prior to discharge, and air pollution controls would likely be required to reduce toxic emissions. DataRequirements.Datarequirementsincludepilot-scale test results, groundwater and/or surface water chemistry, physical and chemical properties of constituents, and benchscale testing. Filtration Description. Filtration is a process of separating and removing suspended solids from a liquid by passing the liquid througha porous medium. Filtrationis a proven technology for the removal oflow levels of suspended solids from liquid waste streams. In particular, granular media filters have been demonstratedto be effective as partof mobile treatment systems as well as on-site systems for hazardous wastewater treatment. filter The backwash would require treatment prior to disposal due to high solids levels. Evaluation. This technology could be effective for the treatment of collected groundwater and/or process waters that have constituents, such as metals, in the form of suspended as ion exchange solids. Filtration canbe an effective pretreatment process for technologies such and carbon adsorption. Data Requirements. Data requirements include groundwater and/or surface water chemistry and physical and chemical properties of constituents. Neutralization Description.Neutralization is used to treat wasteacids and wastealkalies(bases) to eliminate or reduce their corrosiveness. More frequently, neutralization is used as a pretreatment step prior to other liquid waste treatment processes. Care should be taken to avoid the formation of hazardous compounds. Evaluation. Neutralization may be required prior to the treatment or disposal of collected groundwater and/or process waste streams. Data Requirements. Data requirements include pilot-scale test results, groundwater and/ or surface water chemistry, and physical and chemical properties of constituents.

F. Disposal Disposal technologies provide secure, permanent containment of affected media or wastes, thus reducingthe potential for exposure to or migration of constituents. A removal action would be required prior to the implementation of any disposal action.

Disposal of Solids On-Site Landfills Description. On-site landfilling is the placement of treated or untreated wastes in a disposal unit (such as a landfill, surface impoundment, or vault) constructed on site to meet the relevant standardsof RCRA and anyother applicable federal and state requirements. An on-site landfill would provide containment and secure storageof affected solid media. Therisk of exposure to constituents would be reduced by minimizing the mobility of the waste. Evaluation. The construction of a new, approved, on-site landfillmay be impractical due to the required disturbance of the waste, limited site area, and shallow groundwatertable. The construction of an on-site landfill would involve high capital costs and long-term maintenance. Data Requirements. Data requirements include site characteristics, permit requirements, and physical and chemical properties of the waste.

Evaluation Engineering

for Sites Contaminated

77

Off-Site Landfill Description. Off-site landfilling is the transportation and disposal of untreated wastes or treatment residues at an approved off-site landfill. An off-site landfill could provide for the secure containmentof affected mediaor wastes. Wastesor treatment residues (e.g., incinerator ash) would have to meet certain analytical parameters (e.g., TCLP) before off-site disposal would be permitted. The risk of exposure to site constituents would be eliminated by removing the affected mediaor wastes from the site. Excavationwould be required prior to the disposal of the waste. Evaluation. This technology could be a component of remediation for alternative corrective measures that generate wastes or treatment residues that cannot be contained on site. However, off-site disposalof waste would be relatively expensive, and existing landfill capacity is limited. Permits would be required for transportation of wastesto a permitted facility. Data Requirements. Data requirements include identification of a facility, permit requirements, and physical and chemical properties of the waste. Disposal of Liquids Surface Water Discharge Description. Surface water discharge is the discharge of treated water to an existing body of surface water. Priorto discharge, the water would have to meet all applicable environmental standards including Surface Water Quality Criteria (SWQC) and National Pollution Discharge Elimination System (NPDES) standards. Evaluation. Discharge to surface water is a proven means of disposing of treated water. Because discharged water would have to meet all applicable environmental standards, there would be no adverse impacts on human health or the environment. The equipment, materials, and labor requiredto construct a surface water discharge line are readily available. An NPDES permit and other permitsmay be required prior to discharge, or a current NPDES permit may require modification to include the new discharge. Periodic water quality sampling and maintenance of the discharge line would be required. Data Requirements. Data requirements include surface water characteristics and effluent limitations. Reinjection Description. Reinjection refers to the pumping or gravity flow of treated groundwater back into the ground through injection wells or infiltration galleries. Reinjection could be used for the disposalof treated waterif surface water discharge is not possible. Treated water would have to meet all applicable environmental standards before reinjection. Reinjection can also be used to provide more control during groundwater collection. Evaluation. Reinjection has been used successfully at a number of sites for the disposal of treated groundwater. Clogging of reinjection wells from silting, mineralization, and/or biological growth can reduce the effectiveness of reinjection. Reinjection of water into aquifers with low permeabilities may not be practical. Groundwater modeling couldbe required to determine the implementability of reinjection and to determine which reinjection method (injection wells or infiltration galleries) would be the most appropriate. Data Requirements. Data requirements include aquifer and groundwater flow characteristics and effluent limitations.

V. DEVELOPMENT AND EVALUATION OF REMEDIAL ALTERNATIVES This section is intended to serve as a primer on the process of remedy selection. It focuses on the identification, development, evaluation, and selection of appropriate remedial measures,

78

Wilson et al.

technologies, techniques, and procedures. The term “remedy selection” is used in the broad sense of referring to the series of evaluations and assessments that take place inthe practice of determining the most appropriate remedial action fora contaminated site. The reader can and should read the various guidance documents issued by the USEPA and other regulatory agencies that outline in detail the proper procedures for evaluating remedial technologies and alternatives.Thesedocumentseffectivelypresent the agencies’requirements.This chapter attempts to present, froma practitioners’ point of view, how the process really worksor sometimes doesn’t work. In this section we review the basics of the remedy evaluation and selection process from an engineeringperspective. The properengineeringperspectiveis essential tomaintain a solution-oriented focus, as it is easy for the practitioner to become overwhelmed and sidetracked in the regulatory procedural requirements of remedy selection and losethe focus of the real goal-identification and selection of a balanced, protective, and cost-effective approach to remediation of a given site. Obviously, this must be achieved in a manner that satisfies the regulatory requirements, but it remains essential to keep the proper technical perspective. It is important for the engineering practitioner to understand that remedy selection, like many technical endeavors, is an art as well as a science. We firmly believe that in most cases there is not just one “technically correct” remedy for a site or that a singular correct remedy will be identified by any qualified engineer if one follows the specified procedures as defined in various guidance documents prepared by state and federal regulatoryauthorities. Rather, the appropriate remedy for a site is a balance of several factors that include technical, business, and financial considerations as well as environmental protectiveness, compliance with standards, and effectivenessand other criteria prescribed by the USEPA in the National Contingency Plan (which serves as the main regulatory document outliningthe requirements for proper remedy selection under the federal Superfund program andwill be addressed later).

A. Setting Remedial Objectives The fmt step in selecting an appropriate remedyfor a site is to set realistic objectives. Remediation objectives should be focused on ensuring a suitable level of protectiveness of human health and the environment. Political, regulatory, and financial considerations should come into play when assessing how to achieve the objectives, not in setting them. Some of the key factors that must be integrated into setting the objectives are as follows: 1. Real environmental and human health risk (based on realistic exposure scenarios) Current andplanned site use Useof surroundingproperty Impact of site on surrounding areas Assessment of active versus historical contaminant migration 6. Public opinion/support/opposition 7. Technicalfeasibility of achieving objectives

2. 3. 4. 5.

With these factors in mind, there is one overriding questionthat frames the core of any set of objectives regarding remedy selection: What remedy should be implemented? This may seem like an obvious question, but very often the remedy selection process focuses on Whatremedy can be implemented? Whenthe focus is on what is possible rather than and cost-effective,an unnecessarily expensive remedy what is necessary, appropriate, feasible, will almost certainly be selected. The possible remedy, which often attempts to remediate to background or preexisting conditions, often imparts little or no added risk reduction but costs substantially more andimparts an increase in short-term risk duringthe implementation period

Evaluation Engineering

for Sites Contaminated

79

due to the inevitable site disruption, which is typically proportional to the aggressiveness of remediation. Therefore, like most endeavors, there is a point of diminishing returns beyond which additional beneficial effectsare not obtained and in fact negative effects are probable. The assessment of should versus can goes to the most basic, yet controversial issues dominating site remediation projects after nearly 12 years of implementing Superfund and other remediation programs: How clean is clean? The answer to this fundamental question is different if you assess it from a technical or scientific perspective than from a political, cost-benefit, regulatory, or financial perspective. Therefore, by definition, the final answer is site-specific and must properly integrate and bafance the many factors that drive the remedy selection process at each site. A negotiation process must therefore take place between the parties that have genuine interests at each site. Federal and state environmental regulatory agencies usually allow and even facilitate this negotiationprocess by offeringtothe parties having a financialliability and responsibility to remediate the site the privilegeof taking the lead technicalrole in defining and evaluating alternative remedies. The financially liable parties may include the current owner/operator, all past owners/operators, or a party that sent any waste to the site (e.g., a company that formerly used a permitted landfill); these are known as potentially responsible parties (PRPs).

B. RemedySelectionCriteria The National Contingency Plan (NCP),as revised in March 1990, sets forth the selection criteria to be used in the evaluation of remediation alternatives. Remedies are evaluated in three steps in accordance with the NCP and USEPA guidance documents: 1. Identification and evaluation of remedial technologies-usually targeted at one or more specific site problems 2. Combining of suitable technologiesintoremedialalternatives-usuallyintendedtoaddress the entire site 3. Detailed analysis of alternatives-intended to support subsequent remedy selection The first two steps are initially evaluated according to three criteria: effectiveness, implementability, and cost. Upon review of the definitions of these three criteria, as stated in the NCP, the reader will see that they are essentially a repackaging of the expanded set of nine criteria prescribed for the third step. Therefore, in this text, the focus is on the nine criteria for three reasons: ( l ) The step-by-step procedureis already described in the NCP; (2) the focus of this section is on the overall concept and thought process of remedy selection, not on strict procedures; and (3) the practitioner is well served to keep all nine criteria in mind from the beginning. The nine criteria are listed below in the prioritized order the NCP has assigned to them. Each listing includes an explanationof what the criterion means, or should mean, in practical application. Threshold Criteria These two criteria must be met. Protectiveness of Human Health andtheEnvironment. Attaining an acceptableresidual risk presented by the site. The target residual risk must be within the range of one additionalcancerriskper 10,OOO population (1 X to oneadditionalriskper1 O , OOO , OO (1 X Thesecalculatedrisklevelsareverysensitiveto the set of exposure asare sumptions made and the toxicological parameters and factors applied. These factors

Wilson et al. continuously being updated; however, itoften takes years for new information to be accepted in practice, especially if it shows less risk than previously assumed. Compliance with Applicable or Relevant and Appropriate Requirements (ARARs). Compliance with promulgated laws and standards. Often, proposed standards, guidelines, and agency policiesare incorrectly “applied” as if they were promulgated standards. This criterion is often the focus of controversy between the agency and the owner or PR€! In addition, there are waiver provisions in the NCP that are sometimes granted when ARARs are particularly burdensome or expensive or when compliance actually increases site risk (usually short-term risk). Balancing Criteria These five criteria should optimize the remedy. The first three are relatively more important: The ability of the remedy to achieve and mainLong-Term Effectiveness and Permanence. tain the target protectiveness on a long-term and permanent basis.It is important to note thatoftentheconcept of permanence is incorrectlyinterpretedtomean “destruction.” Treatment and containment remedies can also be permanent with proper control and maintenance. Reduction in Toxicity, Mobility, or Volume. The degree to which the contaminantsare altered or recycled to render them less toxic or less likely to migrate or reduce their volume. Zmplementability. An assessment of whether the remedy can be constructed and executed in a technicallyfeasiblemanner to achievetheremediationobjectives.Availabilityand proven performance are addressed under this criterion. The other two balancing criteria are considered to be relatively less important: Short-Term Effectiveness. The risk and impacts imparted by the implementation of the remedy,such as risk to remediation workers and local residents, including transportationrelated risks for off-site remedies. Cost. The capital and operations and maintenance costsof implementing the remedy and the net present value ofthe remedy. It is interesting to note that for government-led Superfund (lead-fund) cleanups, the cost criterion often becomes relatively more important than if an owner or PRP is the funding entity. Modifying Criteria State Acceptance. The special concerns requiredor requested by the state agency in the case of a federally led Superfund project. Community Acceptance. The special concerns required or requested by the local community or local agency in the case of a federally led or state-led Superfund project. These criteria are used to evaluate remedial alternatives prior to actual remedy selection. This evaluation effort may be completed by either the regulatory agencyor the owner/operator or a PR€? Only the agency can select the remedy. However, when an owner has accepted the responsibility to conduct the evaluation, it should also anticipate remedy selection and in fact go thmugh the process of “selecting” the remedy to facilitate effective interchange with the regulatory agency. We propose that the following additionalconsiderations be integrated withthe nine selection criteria stated above: 1. Efficacy of the remedy. Does it achieve the objectives? It is not a success to efficiently solve the wrong problem.

Evaluation Engineering 2.

\

3.

4.

5.

6. 7. 8.

9. 10.

11.

for Sites Contaminated

81

Availability of target remedy. Is the technology commercially available or just in the laboratory development phase? This consideration is often sidestepped, even though it is included in the implementability criterion, in the name of promoting innovative technologies. We support the development and useof innovative technology. However,this must be balanced with the needs of the specific site. Necessary pre-or post-treatment or other treatmenthandling requirements. These requirements add cost and often add time to the remediation schedule.Too often the focus is only on the core technology, when in reality the sum of supplemental unit processesmay cost more or impart higher short-term risk than the core process. Total cost of the remedy overthe life of the remedy (capital, operation and maintenance, and monitoring). Use of innovative technology. Should be selected only when conventional technology cannot achieveobjectives, or innovative technology can achieve equal or superior results at a lower cost, or innovative technologycan achieve equalor superior resultsat the same cost but in less time or with less site disruption. Cash flow demand-the rate at which the remedy must be funded. This is critical in determining affordability. Technical and regulatory precedent. Has the remedy been implemented previously? Degree of uncertainty in problem definition, especially in terms of the volume or area requiring remediation. Unfortunately, many projects advance to the remedy selection stage before the “problem” has been adequately defined. Agency objectives, predispositions, and national or logical policies. Legal situation. The negotiate/remediate/litigatedilemma. Corporate commitment of owner or PRF!

These additional considerations are consistent with the concept, as proposed in this section, of targeting a remedy that is necessary rather than one that is merely possible. It is important to note that the “cost-benefit” concept is not often adopted by the regulatory agencies in the remedy selection process in any formal manner (except in the case of a fund-led cleanup). Cost is delegated to a low-priority criterion, which often is used only to select from among remedies that are essentially equal in all other criteria. This can lead to a circumstance whereany marginal reduction in residual riskis considered appropriate regardless of the incremental cost increase. This text proposes that properly balanced remediation can be achieved only through a concept of risk-based remedy selection.

C. ”Selecting”theRemedy A formal three-phase procedure for identification and evaluation of remedies is presented in “Guidance for Conducting Remedial Investigationsand Feasibility Studies Under CERCLA” issued by the USEPA in October 1988. This procedure can be overly prescriptive and rigid if misinterpreted; however, it does facilitate the development of appropriate remedial alternatives for a given site. Unfortunately, in the absence of the proper context, criteria, and objectives, it is easy to follow this procedure throughthe development and selection of extremely expensive and unnecessarily complex and burdensome remedies. The first step in the remedy selection process is a basic understandingof the problem-the nature and extentof contamination. It is not necessary to have completed the problem definition phase, known as the Remedial Investigation (RI), to begin to develop and evaluate solutions, that is, remedies. In fact, it is beneficial to beginto focus onthe remedy as soon as there is a basic understanding of the site contamination and remedial objectives. This “early focus on the remedy” allows the problem to be defined in the context of the possibldprobable

82

Wilson et al.

solutions. This serves to streamline the RI and ensure that those conducting it obtainthe site and contaminant information necessaryto properly assess and select an appropriate remedy. The initial information should definethe media affectedby contamination-groundwater, soil, sludges; the basic geologyand hydrogeology of the site, and the type of contaminantsvolatiles, semivolatiles, pesticides/PCBs,and inorganics. A preliminary assessmentof whether . or not there are active source areas of ongoing contaminant migrationis very important at this early stage, as it will be one of the main factors to consider in the remedy evaluation process. In addition, one will need a preliminary assessment of the real risk imposed by the site contaminants. This will form the beginning steps of what is known as a “baseline” risk assessment, which will prove essential in proper and informed remedy selection. Evaluating Remedial Technologies Once you know these factors, you can begin to target the types of remedial technologies that will abate specific unacceptable conditions and satisfy remedial objectives. Each typeof problem can be addressed by a technology or linked technologies. Itis vitally important to buildan array of remedial technologies beginning with one that ensures satisfaction of the protectiveness criterion for the targeted problem. This is often the least controversial criterion, as all parties in a remediation project are interestedin ensuring protection of human health and the environment. Assumptions madein calculating risk to assess protectivenessare sometimes the subject of disagreement between the agencies and the owner, but the concept of protectiveness is rarely in dispute. Many of the other evaluation criteria are the subject of greater disagreement. Therefore, it is a good practice to identify a limited set of technologies that arguably achieve the protectiveness goal. The potential remedies may fall into one or more of the following categories of remedial technologies presented in Section I V no action, institutional actions, containment, removal, treatment, and disposal. Initially, one or more of the remedial technologies from these categories should be considered. At the early stages of the remedial investigation, one can afford to be somewhat broadbased in assessing which remedies show real promise in achieving the remedial objectives, including cost.However, it is a good idea to streamline the target remedies as soon as possible. Concurrent progress onthe RI effort will help inthe streamlining process. Once a focused set of technologies and remedies have been tentatively identified that satisfy the protectiveness criterion, they must be compared to the additional criteria to see if the remedies are still in compliance. If the remedies do not comply with successive criteria, they must be modified or additional remedies developedas necessary to ensuresatisfaction of the criteria. After a targeted subset of remediation categories are identified, the next step is to define the basic method,or subcategory, within that category. That is to say, if treatment is one of the targeted categories, one must then identify whether that is best achieved through physicall chemical, chemical, biological, or thermal treatment technologies. The next step is to define the particular technology or process within the subcategoryjust defined. For example, within the physicakhemical subcategory for treating groundwater, there are several choices: air stripping, steam stripping, vacuum-enhanced stripping, carbon adsorption, resin adsorption, or other. For soil remediation within the physical containment subcategory, choices include clay cap, RCRA (layered) cap, soil cover with synthetic membrane, simple soil cover, and others. Some practitioners will take the position that refinement of the remedy to the level displayed above should be deferred untilthe remedial design (RD) stage. However, the RD stage comes after the remedy is selected. At that point the flexibilityto refine or modify the remedy is minimal and is burdened by regulatory procedural limitations. Therefore, it is advantageous

Evaluation Engineering

for Sites Contaminated

83

to evaluate remedies down to the specific technology stage prior to remedy selection. This will sometimes require that treatability studies be conducted. This effort should also precede remedy selection. It is more difficult and requires more skill and knowledge to evaluate specific technologies in the feasibility study (FS) stage and still maintain a focused and limited set of alternative technologies; however, that is the best way to lead to the selection of the optimal remedy; one that meetsthe remedial objectives including cost-effectiveness. It is often true that a wide array of technologies can be “made to work” at many sites. This reflects the concept of possible versus necessary. The goal of the technology evaluation and selection process is not just to identify any workable technology but to define the most balanced remedy. Evaluating Remedial Alternatives A remedial alternative is the combination of remedial technologies that have been targeted to address the sum of the individual problems at a site. Many times, it is appropriate to modify one or more component remedial technologies or partial remedies becauseof beneficial or negative “side effects” resulting from the combination of component remedies. For example,s u p pose there are two types of sludges on a site and the optimal technology for each is different when viewed separately. However, the volume of sludge A is five times that of sludge B. If sludge B is amenable to treatment via the targeted technology for sludge A, it may be more cost-effective overall to combine sludges A and B and treat them bythe sludgeA method, even though there is a more efficient technology available for sludgeB. Overall, the elimination of the sludge B technology is likely to more than pay for the increased volume to be handled by the sludge A technology. To put remedy selection in a logical context, it is useful to array the remedial alternatives as a function of residual site risk versus total cost. Thisis shown graphically in Figure 1. This “cost-risk” curve, when calibrated to an actual site, is a powerful tool that clearly shows the choices available and their implications. If implementation of alternative A yields an acceptable residual site risk, then it should be chosen unless there are compelling and rational reasons to implementone of the other alternatives. Thegraph clearly demonstratesthat to the residual risk curve approaches an asymptote at some point while total cost continues increase dramatically with respect to risk. This is the point of diminishing returns concept previously mentioned. After protectiveness is ensured, there are two of the remaining eight criteria that tend to move the “acceptable” remedy from left to right on Figure 1: compliance with ARARs and long-term effectiveness andpermanence. An oftenquoted rule of thumb is that if it costs X dollars to achieve protectiveness, it will cost 3 X 5X dollars to achieve compliance with ARARs and 10-15X dollars to achieve “permanence” (when is viewed as “destruction” of contaminants). This is why it is essential to begin by developing remedies that achieve the initial threshold criterion, protectiveness. Then expand or modify the remedy, only as necessary, to progressively comply with the additional criteria. This “bottom-up” approach is the best way to ensure compliance in the most cost-effective manner achievable. Unfortunately, many remedies are selected onthe basis of a “top-down” approach, which tendstoresult in unnecessarilyexpensiveremedies.If you define a 15-20X remedy first, nearly all other remedies could be viewed as relatively cost-effective. One of the fundamental areas of controversy in remediation projects between owners or PRPs and regulatory agencies is this cost escalationafter protectiveness hasbeen ensured. The disagreements essentiallycenter around the argument that what is protective today may not be so in the future, especially if proper controls are not maintained.

84

Wilson et al.

Predominant Remedial Action

Predominant Remedial Action

Containment

Excavatioflreatment

/

Degree of Cost Uncertainiy

Al. 1

0

5

10

15

20

25

30

Estimated RemediationCosts, Millions of Dollars

Figure 1 Representativecost-residualriskrelationship.

D. Typical Problems with Remedy Evaluation and Selection The most pervasive problem regardingremedy selection in the years since the Superfund program began is the dominance of a “define the problem” mentality instead of a “define the solution” focus. By farthe greatest amountof effort, dollars, and timeat remediation sites has, to date, been spent in the investigation of contamination, with RIs often taking years and the feasibility study effort being limited to a few months. Unfortunately, this has led to situations where the selected remedy was unable to achieve the objectives, was extremely and unnecessarily expensive, or was commercially unavailable,or to combinations of these andother factors. Obviously, this has been partly responsible for the poor reputation and extensive criticism that plagues the Superfund program as well as other remediation programs. This situation is gradually changing. Public demand and frustrated agencies and corporations are bringing pressures, sometimes opposing pressures, on the process. In the last few years, the remedy selection process has gone from being very sequentialto showing some increased degreeof integration regarding the definitionof the problem and the definition of the solution. Additional necessary changesare expected. Although overstudyingthe problem (the “Study it to death” mentality) is a real shortfall, it is also possible and equally disadvantageous to “under-define” the contaminant situation and prematurely proceed tothe remedial design and remedial action(RDRA) stage. Often public or political pressures forcethe RUFS effort to be unrealistically acceleratedto the point that the engineers who facethe task of identifying and evaluating remedies find that the problem is

Evaluation Engineering

forSites Contaminated

85

not well defined. This inevitablyleads to a situation where allor most of the site-specific thinking is deferred to the subsequent remedial design phase. This may lead to the implementation of a remedy that is substantially less than optimal. Premature remedy selection can lead not only to a waste of funds but also to a situation where thewrong problem is “solved,” leaving the real problem unremediated. Sometimes the problem that remains unsolved is due to the contaminants that pose the “principal threats” at the site. In such a case (and unfortunately, there are several), the contaminants that causethe lesser concernmay have been “efficiently” abated, giving the illusion of remediation and protectiveness while the true site risk is minimally affected. One of the main deficiencies arising from a poor understanding of the site problems is a high degree of uncertainty regarding the volume or area of contaminated media requiring remediation. This is a major burden in the effort to define a protective, balanced, and costeffective remedy because remedy selection is not a linear process. The optimal remedy for 40,000 cubic yardsof soil contaminated with volatile and semivolatile organics may not be the optimal remedy for 4OOO cubic yards of soil with the same level of the same contaminants. Many factors prevent the simple adoption of a remedy for a similar application but a site of substantially different magnitude.

VI. SUMMARY This chapter has presented some of the issues that face the site remediation engineer in professional practice andhas suggested someof the techniques applicablein the endeavorto clean up the nation’s hazardous waste sites with a knowledge-based, cost-effective, and systematic but integrated approach. The key concept presented in this chapter is one of balance. There are no perfect remedies;they all have disadvantages and limitations. Each site presents a unique set of trade-offs. The challenge is to balance the pros and cons in a manner that satisfies the essential objectives, is implementable, and is affordable to the funding entity. Without this balance, a strategy of either stall or litigate will likely result instead of a remediation strategy. The business ofremedy evaluation and selection is extremely complex and requires a working knowledgeof environmental regulation, environmental law, process engineering, cost estimating, and construction engineering.No single chapter,or even a full volume, can attempt to convey more than the very basics of this profession. To compound the situation, it is very dynamic and changes continuously,so what was optimal in the past may not be acceptable in the future. This makes the practice very challenging, but also very exciting.

This Page Intentionally Left Blank

Innovative Approaches to Cleanup Level Development Ronald J. Kotun, Richard F. Hoff, Robert J.Jupin, Diane McCausland, and Patrick B. Moroney Chester Environmental Monroeville, Pennsylvania

1.

INTRODUCTION

Recently proposed regulations on the federal and state levels establish uniform cleanup standards to provide objectives for remediating contaminated sites. The purpose of these standards is to restore contaminated sites to levels that ensure protectionof human health and the environment. These regulations may result in consistent cleanup decisions and may expedite the remediation of contaminated sites by eliminatinglengthynegotiationsovertheextent of cleanup. However, this “consistency” may result in overconservative remediations and excessive expense or, more important, inadequate remediation. As part of a remedial investigation, a baseline risk assessment is conducted to determine if there is any potential for adverse health effects due to hazardous substances released from a site in the absence of controls. The risk assessment identifies constituents of interest (COIs) exposure scenarios, and likely receptors (residents, construction workers, etc.) and quantifies the amount of chemical intakeby a receptor at the site. The amount of intake can be translated into a risk value that indicates the extent to which public health may be affected. The results of the baseline risk assessment document the magnitude of risk and identify its primary cause. The results also provide a basis for a decision as to whether remedial action is necessary. The U.S.Environmental Protection Agency (EPA) has established guidelinesthat provide a consistent process for evaluating and documenting threats to the public health and the environment. Consequently, risk assessment provides a means to determine site-specific cleanup goals that are adequately protective of human health. Other scenarios consider the environmental fate and transport of chemicals to determine whether chemicalsin one medium might have an impact on another medium. These approaches consider factors suchas the spatial distribution of chemicals, interim measures currentlyin place to control chemicaltransport, plausible exposure scenarios, likely receptors to chemical exposure,and the physical and chemical properties of the chemicals. Therefore, the results generated from these approachesmay span several orders of magnitude yet still resultin protection of human health and the environment. More important, these approaches derive levels that are technologically and economically feasible to attain. 87

88

Kotun et al.

In this chapter we present four case studies that consider the factors previously discussed to establish cleanup levels. These cases illustrate various means to derive a cleanup level that is technically defensible. However, these levels may not necessarily be the optimum for site remediation. Usually, more than one approach should be explored to establish the optimum cleanup level. Case I explores an innovative approachto the derivation of risk-based cleanup levels for carcinogenic polycyclic aromatic hydrocarbons (cPAHs). Using relative potency factors and specific exposure scenarios and considering apparent remedial strategies, a cleanup level for cPAHs in pond sediments was determined. The target cleanup level was developed by establishing a correlation between the cPAH concentration in a pond sediment sample and the corresponding risk. Plotting thesedata to generatea regression equation andselecting a target risk level allows one to identify a cleanup level. Case I1 employs a similar statistical and graphical approachto deriving cleanup levels for cPAHs, but the approach is more complex because cleanup levels are also being derived for pentachlorophenol (PCP) and arsenic. Risk-based cleanup levels can be derived for exposures to single constituents by back-calculating a soil constituent concentration corresponding to a target risk level using risk assessment methodology. However, when the risks result from exposure to more than one constituent, the target risk level should be apportioned among the individual constituents. The availablesite data were evaluated to determine the relative distribution of the chemicals across the site. Target risk levels were then apportioned among the individual chemicals according tothese relative distributions. Case I11 uses EPA's recently developed Multimedia Exposure AssessmentModel (Multimed) to derive soil cleanup levels for a source area of chemical release that would be protective of groundwater at a designated downgradient receptor location. In this study, Multimed was used to model the unsaturated and saturated zone fate and transport of PAHs. Groundwater investigations revealed that the PAHs may leach from source soils and migrate horizontally through an underlying aquifer toa downgradient well location and an adjacentestuarine river. The cleanup levelswere developed to be protective of humans consuming waterat the receptor well location, organisms inhabiting the river, and humans consuming organisms from the river. Soil cleanup levels were calculated byusing the dilution-attenuation factors derived by the model in combination with partition coefficients to meet applicable and appropriate performance standards in the water. Finally,Case IV employs other environmental fate and transportmodels,specifically EPA's Organic Leachate Model (OLM) and Vertical and Horizontal Spreading (VHS)model, PAHs that is protectiveof underlying groundto determine a soil cleanup goal for carcinogenic water quality. These two models account for the leaching of cPAHs into groundwater and their dilution in the shallow saturated zone affordedby dispersion. Use of these analytical solutions to derive a soil cleanup level providesa timely, cost-effectivealternative to complex numerical modeling. Furthermore, the conservatism of the results may account for some of the temporal and spatial variations inherent in even the simplest groundwater flow domainsnot accounted for in complex models.

II. CASE STUDY I: REGRESSION ANALYSIS AND APPARENT REMEDIAL STRATEGIES TO DERIVE CLEANUP LEVELSFOR CARCINOGENIC PAHs A. Introduction The Comprehensive Environmental Response, Compensation, and Liability Actof 1980, as amended (Superfund),is a national program that establishesa means to respond to releases of

Cleanup

89

hazardous substances.The National Oil and Hazardous Substance Pollution Contingency Plan (NCP) is theregulationthatimplementsSuperfund and establishesthegeneralapproach for remediation of hazardous waste sites. The objective of the program is to protect human health and the environment fromthe potential harm posed by the constituents presentat these sites. The U.S. Environmental Protection Agency (EPA), administrator of Superfund, has established a framework for assessing risks to human health that ultimately determines whether remedial action is necessary at a site. The risk assessment process, as set forth in EPA's Risk Assessment Guidancefor Superfund, Volume I, Human Health Evaluation Manual(Part A) [l], provides an analysis of baseline risks, determines theneed for remedial action, and establishes a basis for calculating cleanup levels. This process provides an alternative to meeting regulatory cleanup criteria and allows for the use of site-specific information to establish practical remedial alternatives. Cleanup levelsare being established for a pond that is adjacent to a former wood-treating facility that operatedfor 18 years. The facility used creosote asthe preserving chemical.Of the 27 shallow sediment samples collected from the pond, cPAHs were detected in all samples with total concentrations ranging from 0.807 to 144.82 mgkg (Table 1). In many cases, the high degree of contamination is limited to a small area and is notcharacteristic of the entire site. A frequency distribution of the cPAHs in the shallow pond sediments indicates that only fourof the 27 samples had total cPAHs exceeding 30 mg/kg and seven had total cPAHs exceeding 10 mgkg (Table 2). A risk assessment was prepared in accordance with Superfund for this pond to establish baseline risks for all potential receptors and exposure scenarios. Based on the data, children with the cPAH-impacted sedimentsare those wading in the pond and coming into direct contact at the greatest potential risk. Therefore, development of cleanup levels for pond sediments is based on this most sensitive receptor. The primary objective of establishing cleanup levelsat this site is to be protective of human health, specifically the health of children wading in the pond. These children may come in contact with the cPAHs in the sediments through dermal contact and incidental ingestion. Baseline risks associated with exposure to the cPAH-contaminated sediments are based on the application of Chu andChen's relative potency factor approach [2]. Using the site-specific data and the relative potency factors,a cleanup level canbe calculated. Sediments withcPAH concentrations above a certain concentration would be removed, thus resulting in a reduced average concentration of cPAHs and a concomitant reduction in risk.

B. Chu and Chen's Relative Potency Factor Approach

.

Originally, risks attributed to exposure to cPAHs were derivedby assuming that all cPAHsare equipotent to benzo[a]pyrene. Realizing that there is much scientific evidence that demonare strates that benzojalpyrene is regardedas one of the most potent cPAHs and that all cPAHs not equipotent [2], alternative methods have been developed that assess risks more fairly. A study conducted by Margaret Chu and Chao Chen of EPA's Carcinogenic Assessment Group provides evidence to support these contentionsand also provides relative potency factorsthat allow for a more appropriate evaluation of risks associated with these less potent cPAHs. Relative potency factors (RPFs) for six cPAHs are presented in Table3. These cancer slope factors are derived by multiplying benzo[a]pyrene'soral cancer slope factorof 11S 3 [mg/(kg/ day)]" [3] by the RPF. Ultimately, potential risks are calculated by summing the products of the cPAH concentrations and the corresponding cancer slope factors[l]. The total risk can be defined by the equation

2

l l l l $ l l l l 0

Kotun et al.

Cleanup Level Development

91

Table 2 Frequency Distribution of Total cPAHs in Pond Sediments" Concentration range

(mag)

Number of samples

BCLDL DL5

0 9 8 4 2 1 0 0 0 0 2 0 0 0 0 0 1 0

5- 10 10-20 20-30 30-40 40-50

50-60 60-70 70-80 80-90

90- 100 100-1 10 110-120 120-130 130-140 140-150 >150 "BDL = below detection limits; DL = detection limit.

x n

RL =

-

(CSi CSFi EXP)

i= 1

where RL is the risk level from exposure to cPAHs, CS, is the concentration of cPAH,, CSF, is the cancer slope factor for PAH,, and EXP is the product of intake parameters to establish long-term daily intakeof cPAH,. EXP is defined as EXP =

CR EF ED

BW-AT

where CR is the contact rate (ingestion rate or surface contact), EF the exposure frequency, ED the exposure duration, BW the average body weight, and AT is the averaging time. For children, an exposed skin surface area of 3768 cm2 was assumed to account for 50% of the total body surfacearea. The exposed surface areais limited to 50% because itis unlikely that the total surfacearea would be covered with sediment [4]. Absorption factorswere used to reflect the desorption ofchemicalfrom the sedimentand the absorption of thechemical through the skin to the bloodstream. A value of 5% was used for cPAHs[4]. Adherence to skin was estimated at 1.45 mg/cm2 [4]. An exposure frequency of24 days per year(2 days per week during the summer) for an exposure duration of 13 years was used. The average body weight was 41.5 kg [5]. The averaging time was 70 years [l]. There are no recommended values for the amount of sediment a child may incidentally ingest; therefore,the recommended incidental ingestion rate of 100 mg/day was used [6], with only 50% of the sediment coming from the site. The exposure frequency, exposure duration, body weight, and averaging timeare the same parameters thatwere used to assess risks due to dermal exposure.

92

Kotun et al.

Table 3 Chu and Chen Relative Potency Factors for cPAH factorpotency Relative Compound 1 .o 0.69

Benzo[a]pyrene anthracene Dibenzo[a,h] fluoranthene Benzo[b] Chrysene Indeno[ 1,2,3-cd]pyrene anthracene Benzo[a] pyrene Benzo[k] Source: Chu and

0.08

0.00122 0.0171

0.0134 0.00444

Chen [2].

With no remedial action taking place, the average cPAH concentration in the pond sediment is 20 mgkg. The risk to children exposedto these sediments through dermal contact and incidental ingestion using the previously stated assumptions is 7.3 X

C. Site-Specific Development

of Risk-Based Cleanup Levels

Each sediment sample consists of a different compositionof cPAHs that results in a correspondingly different probability of risk. For example, two sediment samples may have equivalent total cPAH concentrations. However, the one sample that has a greater percentage of benzo[a]pyrene, one of the most potent cPAHs, will pose the greater probability of risk. To illustrate this relationship among the sediment samples, a log-log plot of the probability of risk 1). Cleanup levels corresponding to versus the total cPAH concentration was developed (Figure

1E44

1E47 0.1

1

10

100

Total Carcinogenic PAHs (rngwg)

Figure 1 Risks to children associated with total cPAHs resulting from exposure

to pond sediments.

Cleanup Level Development

93

Table 4 Total cPAHConcentrations

and Their

Corresponding Target Risk Levels Total cPAH concentration (mgncg)

Risk

280 28

10~-4 IOE-~

10E-6 target risk levels can be determined by simply locating them on the regression line. For example, a target risk level of lo-’ corresponds to an average total cPAH concentration of 28 mg/kg (Table 4). This pond has only fourof its 27 samples with totalcPAH concentrations exceeding a concentration of 30 mgkg (approximately the target risk of lo-’)). Remediating areas with elevated levels of total cPAH to this concentration or “pickup level” would reduce the average total cPAH concentration, or cleanup level,in the pond and consequently reduce the associated risk. Table 5 presents various pickup levels and the risks that correspondto the cleanup levels for children who wade in the pond. A pickup level of 30 mgkg results in a reduction of the cleanup level for total cPAH concentration from 20 mgkg to 12 mgkg with a corresponding reductioninriskfrom 7.3 X loF6to 4.3 X This approachresultsinlimitingtheremediation to the heavily impacted areas and minimizing unnecessary cleanup.

D. Conclusion Risk assessment provides a means to determine whether chemicals present on a site exist in concentrations that pose a potential threat to public health or the environment. The risk assessment performed for this pond indicated that exposure to sediments from thepond presented a significantrisk, with children wading in the pond being the most sensitive receptor. There are be impractical isolated areas in the pond that obviously require remediation; it would certainly to remediate the entire pond to pristine conditions. Using risk assessment principles, acPAH level can be estimated to which one can be exposed and still have a negligible probability of developing cancer. Our assessment of the site indicated that remediation to an average concentration of 30 mgikg would result in a risk level of l X lo-’. Without any remedial action, exposure to the Remediating averagecPAHconcentrationthroughoutthesiteresultsinarisk of 7.3 X areas in excess of 30 mgkg, the so-called pickup level, will lower the site average concentra-

Table 5 Pickup Levels, ResultingCleanupLevels,andTheir Corresponding Risks to Children Wading in the Pond and Exposed to Impacted Sediments Pickup level (mg/kg)

Site cleanup level ( m a g )

No action 7.3E-06

20 18 16 14 12 10 7

90

70 6.OE-06 50 5.2E-06 30 4.3E-06 20 3.7E-06 10 2.6E-06

Risk level

6.68-06

Kotun et al.

94

tion to a cleanup level of 12 mgkg and concomitantly reduce the risk by 41%. In essence, determination of a site pickup level for cPAHs through regression analysis to meet a target risk level can result in a practical remediation that is protective of public health, minimizes the volume of sediments to be removed, and is most likely cost-effective.

111.

CASE STUDY II: STATISTICAL METHODS TO DERIVE CLEANUP GOALS FOR A SITE IMPACTED BY MORE THAN ONE CHEMICAL

A. Introduction Cleanup levelsare being established to address contaminated surface and subsurfacesoils at a 65 years. The facility used creosote and penwood-treating site that operated for approximately tachlorophenol as wood-preserving chemicals.In general, polynuclear aromatic hydrocarbons (PAHs) and pentachlorophenol were detected in nine areas of the site, ranging in concentration from 1 to 10,800 mgkg and from 0.1 to 240 mgkg, respectively, The areas showingthe highinadjacent to the treating area est concentrationsof PAHs or pentachlorophenol were located or and in a former disposal area. Although wood was never treated with chromated copper arsenate (CCA) at this site, arsenic and copper were detected above background levels in the wood storage yard, at concentrations ranging from 24 to 369 mgkg and from 0.005 to 1.13 mg/kg, respectively. A risk assessment was prepared .in accordance with Superfund for this site to establish baseline risks for all potential receptors and exposure scenarios. Based on the data, on-site workers and construction workers coming into direct contact with the impacted surface and subsurface soilswould be at the greatest potentialrisk. Therefore, development ofcleanup levels for surface and subsurface soilsis based on these most sensitive receptors. The primary objective of establishing cleanup levels at thissite is to be protective of human health, specifically the health of on-site workers and construction workers. These individuals may come into contact with the constituents in the surfaceand subsurface soils through dermal contact, incidental ingestion, and inhalation. Using site-specific the data and targetrisk levels, a cleanup level can be calculated. Constituents detected at the site exhibit both carcinogenic and noncarcinogenic adverse health effects. Potential adverse health effects due to carcinogenic substances are derived differently from those due to noncarcinogenic substances. Therefore, cleanup levels have to be developed to address both carcinogenic and noncarcinogenic adverse health effects. Theseare dealt with in turn in the following sections.

B. Cleanup Levels for Potentially Carcinogenic Constituents Results of the risk assessment indicated that potential risks for exposureof on-site workers and construction workers to surface and subsurface soils were above the U.S. EPA's target risk range of 10-4-10"j [7];therefore, risk-based levels were derived for these potential human receptor groups. Constituents evaluatedin the risk assessment included PAHs, pentachlorophenol, volatile organic compounds (VOCs), and metals. Cleanup levels can be developed for all of the evaluated constituents of interest (COIs), but this is not necessary if the potential risks from the individual COIs are insignificant. Following EPA Region 111 guidelines, cleanup levels were developedfor only those constituents that contribute to 99% of the total potentialrisk. Tables 6 and 7 present a summary of the potential risks and the percent contribution of the individual constituentsto the total risk for the evaluated human receptor groups. As can be seen from these tables, PAHs, pentachlorophenol, and arsenic were the COIs responsiblefor 99% of

95

Cleanup Level Development

Table 6 Summary of Potential Risks of Surface Soils to On-Site Workers E Area

D

B Area Constituent A Area Potential risk PAHs Pentachlorophenol Benzene Styrene Arsenic Chromium Total potential risk

4.7E-04 8.6E-06 O.OE+OO O.OE+OO 2.8E-05 O.OE+OO 5.1E-04

4.4E-03 1.4E-06 1.5E-10 7.3E-11 1.OE-04 O.OE+OO 4.5E-03

1.7E-03 1.4E-04 2.3E-09 5.5E-10 O.OE+OO O.OE+OO 1.9E-03

7.8E-05 8.8E-06 4.6E-09 8.9E-10 5.8E-05 3.1E-07 8.5E-04

15e-03

1.5E-06 O.OE+OO O.OE+OO 1.1E-05 3.5E-07 1.6E-03

Percent contribution to total potential risk ~~~

99.14% 92.04% 92.66% 97.73% 92.75% PAHs %7.34%0.03% Pentachlorophenol 1.69% .00% Benzene 00% Styrene 0.00% Chromium 56% Arsenic

. 0.04%

0.10% 0.02%

Table 7 Summary of Potential Risks of Surface and Subsurface Soils to Future Construction Workers B Area Constituent A Area

D

E Area

Potential risk PAHs Pentachlorophenol Benzene Styrene Arsenic Chromium Total potential risk

2.8E-05 l .4E-07 O.OE+OO 4.6E- 13 7.OE-07 4.2E-08 2.9E-05

4.6E-04 1.8E-08 7.9E- 12 1.6E- 12 2.8E-06 4.2E-07 4.6E-04

2.3E-05 1.7E-06 3.8E- 11 1.4E-09 O.OE+OO O.OE+OO 2.5E-05

1.8E-05 1.6E-07 4.7E- 11 9.3E- 12 1.6E-06 5.OE-08 1.9E-05

6.2E-05 1.OE-07 2.2E- 10 1.9E- 11 3.8E-06 8.OE-08 6.6E-05

Percent contribution to total potential risk PAHs Pentachlorophenol Benzene Styrene Arsenic Chromium

96.99% 0.47% 0.00% 0.00% 2.40% 0.14% 100.00% 100.00%

99.30% 0.00% 0.00% 0.00% 0.61% 0.09%

93.25% 6.74% 0.00% 0.01% 0.00% 0.00% 100.00%

90.48% 0.85%

0.00% 0.00% 8.42% 0.26% 100.00%

93.98% 0.16% 0.00% 0.00% 5.75% 0.12% 100.00%

the potential risk to on-site workers and construction workers; thus cleanup levels were derived only for these COIs. Potential risks were estimated in the risk assessment by evaluating exposuresto site surface and subsurface soils via dermal contact, incidental ingestion, and inhalation exposure pathways. Applying the 99% contribution criteria mentioned aboveto each exposure pathway indicated that cleanup levels shouldbe derived using eachof the evaluated exposure pathways.

Kotun er aZ.

96

Risk-based cleanup levels were derived using the procedures and exposures assumptions presented in the risk assessment forestimating potential risk. Instead of using the sampled CO1 concentration in soil and solving the risk equations for potential risk, a target risk level was assumed and the risk equations were solved for the concentration of the CO1 in soil. The following equation was used in the risk assessment to estimate potential risk (PR):

-

PR = C, CSF EXP where C, is concentrationof COI, CSFis cancer slope factor, andEXP is the product of intake parameters used to establish long-term daily intakeof COI. EXP is defined as

EXP = (CR EF ED)/(BW AT) where CR is contact rate, EF is exposure frequency, ED is exposure duration, BW is average body weight, and AT is averaging time. Solving Equation (3) to yield a soil concentration results in

C, = PR/(CSF EXP)

(4)

Soil cleanup levels were derived for target risk levels and of as required by the U.S. EPA. Using the above method and the same intake assumptionsas were used in the risk assessment, cleanup levels corresponding to were and calculated pentachlofor rophenol and arsenic and are presented in Table 8. A different methodwas used for deriving cleanup levels for PAHs. PAHs comprise several constituents, and itis more desirable to express a cleanup level for PAHs as total PAHs than to derive cleanup levels for each individual constituent. First, the potential risk resulting from exposure to only PAHs for on-site workers and construction workers was calculated at each soil-sampling location. Second, the concentrationsof total PAHs were plotted against the corresponding potential risks as shown in Figures 2 and 3. Having statistically significant correlov5,and risklevels were located lations (R2), the cleanup goalscorresponding to in Table8 are appropriate on the plot and are presented in Table8. The cleanup levels presented for those areas of the site that contain only PAHs, pentachlorophenol, or arsenic and not a combination of the three COIs.

Table 8 CleanupLRvels (mgkg)with No Apportioning of Target Risk Levels Target risk level Constituent

10-6 10-~

10-5

On-site workers

Pentachlorophenol Arsenic Total PAHs Pentachlorophenol Arsenic Total PAHs

700 314

70 31

229 16 Construction workers

7 3 1

30,500

3050

305

6,620 20,000

662

1300

66 85

97

Cleanup

Total PAHs (mg/Kg)

Figure 2 Potentialrisk vs. total PAHs-on-site workers.

1. o E a

Y

.v, 1.oE-05 a

l.OE-07

1. o E a

Figure 3 Potentialrisk vs. total PAHs-construction workers.

Kotun et al.

98

Table 9 Distribution of Risk to On-site Workers from Exposure to Surface Soils Arithmetic Standard

Minimum mumdeviation mean Constituent Potential risk PAHs Pentachlorophenol Arsenic Total

4.7E-04 3.8E-06 3.7E-05 5.1E-04

1.2E-03 8.9E-06 3.5E-05 1.2E-03

2.OE-06 O.OE+00 1.2E-06 8.OE-06 Percent contribution

PAHs Pentachlorophenol Arsenic

64.88% 1.75% 33.37%

30.31% 2.93% 30.04%

4.51% 0.00% 0.37%

6.3E-03 4.1E-05 1.5E-04 6.5E-03

7.8E-05 8.2E-07 2.6E-05 1.1E-04

99.09%

71.16%

12.40% 95.28%

25.51%

0.46%

Since the total potential riskfor most areas of the site is the result of exposure to several COIs, the target risk levels must be apportioned among the COIs. The target risk level was apportioned basedon the contributions of the risk levelsof the individualCOIs. For example, and the site average potential risk from PAHs if the site average potential risk level is 1 X is 7.5 X itwould be assumedthat (7.5 X 10-5)/(1 X or 75%, of thetargetrisk level would be apportioned amongthe PAHs. This is the preferred method for apportioning the target risk level among the COIs of the site, since potential risk is based on both the toxicity and the concentration of the COI. Polycyclic aromatic hydrocarbons, pentachlorophenol, and arsenic were identified above as the major contributors tothe potential risk at the site. Eighty-seven surface samples and65 subsurface samples were collected at the site and analyzed for PAHs and pentachlorophenol, but only 32 of the surface soil samples and 27 of the subsurface samples were analyzed for arsenic. Only those soil samples analyzed for all three COIs were used in determining the apportionment of the risk levels. The potential risk of exposure to PAHs, pentachlorophenol, and arsenic for on-site workers and construction workers was calculated for each of the selected soil samples. The average contribution of each CO1 to the total risk for on-site workers is presented in Table 9. Table 9 illustrates that 65% of the total risk for on-site workers is due PAHs, to 33%is due to arsenic, and 2% is due to pentachlorophenol. Table 10 shows that 75% of the potential risk for construction workers is due to PAHs, 23% is dueto arsenic, and 1.4%is due to pentachlorophenol. The target risk levels were apportioned according to these distributions. As an example, asfor on-siteworkers,cleanuplevelswouldbederived suming a targetrisklevel of 1 X or a 6.5 X lo-’ targetrisklevel. A for PAHs corresponding to 65% of the 1 X 3.3 X lo-’ target risk level would be used for arsenic,and a 2 X targetrisklevelfor pentachlorophenol. Cleanup levels for on-site workers and construction workers based on the apportioning method described above are presented in Table 11.As can be seen from thetable, cleanup goals for on-site workers are lower than the corresponding goals for construction workers.

C. Cleanup Levels For Noncarcinogenic Constituents Results of the risk assessment indicated that total hazard indices for on-site workers and construction workers were above the “acceptable” level of 1.O for some areas of the site. Tables 12 and 13 present a summary of the hazard indices and the percent contribution of the indi-

Cleanup Level Development

99

Table 10 Distribution of Risk to Construction Workers from Exposure to Surface and Subsurface Soils Arithmetic Standard

tion mean Constituent

Minimum

Maximum

Median

8.2E-01 7.5E-03 1.9E-03 8.2E-01

1.1E-03 1.4E-04 1.2E-05 2.2E-03

99.%%

94.36% 1.78% 0.65%

Potential risk ~~

PAHs Arsenic Pentachlorophenol Total

2.6E-02 9.6E-04 1 .OE-04 2.7E-02

3.6E-05 2.OE-08 1.9E-07 3.7E-05

l.lE-01 1.6E-03 3.OE-04 l.lE-01

Percent contribution PAHs Arsenic Pentachlorophenol

31.77% 31.78% 2.16%

75.64% 22.94% 1.42%

2.49% 0.00% 0.02%

97.43% 9.25%

Table 11 CleanupLevels ( m a g ) with Apportion Apportionment of Target Risk Levels Target risk level Constituent

10-4

10-5

10-6

On-site workers Pentachlorophenol Arsenic Total PAHs

14 104 140

Pentachlorophenol 15 Arsenic 152 PAHs Total

427 1515 5350

1.4 10 10

0.14 1 .o 0.7

Construction workers 43

4

445

35

vidual constituents to the total hazard index for the evaluated human receptor groups. As can be seen from these tables, PAHs, pentachlorophenol, arsenic, copper, chromium, and zinc are the COIs thatare responsible for 99%of the hazard index for on-site workers and construction workers. When the total hazard index exceeds 1.0, EPA guidelines [l] recommend that COIs be segregated by target organ effects and mechanism of action to derive separate hazard indices for each group. The COIs were segregated by effect and mechanismof action, and the resulting hazard index still exceeded unity, indicating that the cleanup levels would also have to be derived for noncarcinogenic COIs. Past experience has shown that when cleanup levels are developed for constituents thatr exhibit both carcinogenic and noncarcinogenic effects, those developed to address the carcinogenic effects are always lower than those developed to address the noncarcinogenic effects. If the cleanup levels developed for the carcinogenic effects are also protective of the noncarcinogenic effects, then cleanup levels to address the noncarcinogenic effectsneed not be separately developed. Plots of the total hazard indices (for exposures to allCOIs, not just PAHs) versus total PAHconcentrations are shown in Figures4 and 5. The targetcleanup level fortotal PAHs corresponding to the target risk range presented in Table 8 is 140 mgikg for on-site workers and 5350 mgikg for construction workers. The loF4cleanup level of 140 mgkg for

Kotun et al.

100

Table 12 Summary of Hazard Indices-Surface Soils and On-site Workers BArea C Area

Area A Constituent Area

D

Area E

Hazard Index PAHs Pentachlorophenol Phenol Toluene Ethylbenzene Styrene Xylenes (total) Arsenic Chromium Copper Zinc Hazard index0.29

2.3E-02 6.7E-03 O.OE+OO O.OE+OO O.OE+OO O.OE+OO O.OE+OO 4.5E-02 3.5E-03 4.1E-03 2.3E-03 1.22 0.09

2.6E-01 1. IE-03 3.9E-05 2.2E-07 2.9E-07 5.1E-07 l.lE-07 1.6E-01 3.3E-02 8.lE-02 3.1E-02

1.1E+00 l.lE-01 2.6E-06 4.5E-06 1.4E-05 3.8E-06 1 .OE-04 O.OE+OO O.OE+OO O.OE+OO 1.4E-03

1.7E-01 6.88-03 1.OE-05 7.1E-06 1.2E-05 6.2E-06 4.8E-06 9.OE-02 6.6E-03 7.OE-03 1.9E-03

0.57

3.7E-01 1.2E-03 1.9E-06 O.OE+OO O.OE+OO O.OE+OO O.OE+OO 1.8E-02 7.6E-03 2.4E-03 l.lE-03 0.40

Percent contribution to total hazard index PAHs 92.54% 60.70% 91.08% 45.87% 26.90% Pentachlorophenol 2.39% 8.79% 0.20% 7.89% 0.01% Phenol 0.00% Toluene 0.00% 0.00%0.00% 0.00% 0.00% 0.00% Ethylbenzene 0.00% Styrene 0.00% Xylenes0.01% (total) 0.00% 0.00% 4.40%Arsenic 31.46% 0.00% 28.37%53.48% Chromium 1.89% 2.32% 0.00% 5.78% 4.14% 14.35% 4.86% Copper Zinc 0.12% 5.43% 2.72% 0.28% 0.67%

0.29%

0.00%

0.00% 0.00%0.00%

0.00%

0.00%

0.00%

0.00% 0.00%

0.00% 0.00%

0.00%0.61%

0.00%

2.45%

on-site workers corresponds to a total hazard index of 0.01 on Figure 4. The cleanup level to a total hazard index of 1.O on Figure5 . of 5350 mgkg for construction workers corresponds Remediating the site to the cleanup levels based on potential risk will lower the total hazard indices for all the Cols to 1.O or lower; therefore, separate cleanup levelswere not derived for noncarcinogenic compounds.

D. Conclusion Risk assessment provides a means to determine whether constituents presentat a site exist in concentrations that pose a potential threat to public health or the environment. The risk assessment performed as part of the remedial investigation study for the site indicated that exposuretosurfaceandsubsurfacesoilspresented a significant risk to on-site workersand construction workers. Risk assessment methodology along withstatistical and graphical methods were used to derive cleanup levels for the site. risk level were 140 For areas that contain all COIs, the risk-based cleanup goals at the mg/kg for total PAHs, 104 mgkg for arsenic, and 14 mgkg for pentachlorophenol. For areas of the site that contained only a single constituent, the risk-based cleanup goals at the risk levelwere 230 mg/kg for totalPAHs, 314 mgkg for arsenic,and 700 mg/kg for pentachlorophenol.

101

Cleanup

Table 13 Summary of Hazard Indices4urface and Subsurface Soils, Future Construction Workers

ea A Constituent Area

E Area D Area B C Area Hazard Index

PAHs 9.5E-01 5.3E-03 Pentachlorophenol 2.OE-06 Phenol 7.6E-08 Tolulene Ethylbenzene 2.5E-07 1.6E-07 Styrene l.lE-07 Xylenes (total) 4.68-02 Arsenic 2.9E-02 Chromium 4.2E-03 Copper 5.9E-03 Zinc 1.3E-04 4-Methylphenol 8.3E-05 2-Methylphenol 6.9E-04 2,4-Dimethylphenol 1.88 0.28 1.040.76 Hazard Index

1.3E+00 7.OE-04 1 SE-05 6.8E-07 4.6E-07 5.7E-07 1.4E-07 1.8E-01 2.8E-01 9.8E-02 3.5E-02 1.4E-05 O.E+00 2.4E-05 1.93

7.OE-01 6.5E-02 61.E-06 4.7E-06 1.1E-05 2.1E-06 5.3E-05 O.OE+OO O.OE+OO O.OE+OO 1.9E-03 9.9E-05 9.9E-05 2.5E-04

1.2E-01 6.4E-03 3.7E-06 3.8E-06 6.1E-06 3.2E-06 2.5E-06 l.lE-01 3.4E-02 1.4E-02 3.OE-03 O.OE+OO O.E+00 O.OE+00

1.6E+00 4.1E-03 1.9E-05 7.7E-06 1.8E-05 6.6E-06 3.8E-06 2.5E-01 5.4E-02 1.2E-02 3.3E-03 3.7E-04 3.7E-04 1.OE-03

Percent contribution to total hazard index PAHs Pentachlorophenol Phenol Phenol Tolulene Ethylbenzene Styrene Xylenes (total) Arsenic Chromium Copper Zinc CMethylphenol 2-Methylphenol 2,CDimethylphenol

91.26% 0.51%

0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 4.41% 2.77% 0.40% 0.57% 0.01% 0.01% 0.07%

68.85% 0.04%

0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 9.50% 14.71% 5.08% 1.82% 0.00% 0.00% 0.00%

91.21% 8.47% 0.00% 0.00% 0.00% 0.00% 0.00% 0.01%

0.00% 0.00% 0.00% 0.25% 0.01% 0.01%

0.03%

41.61% 2.27% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 37.97% 12.06% 5.03% 1.05% 0.00% 0.00% 0.00%

82.79% 0.22% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 13.16% 2.89% 0.66% 0.17% 0.02% 0.02% 0.06%

IV. CASE STUDY 111: DERIVATION OF SOIL PAH CLEANUP GOALS USING THE MULTIMED MODEL A. Introduction A remedialinvestigatiodfeasibilitystudy (RUFS) was conductedat a former wood-treating and storage facility in Virginia that is currently included onEPA's National Priority List (NPL) of Superfund sites. A finalRI report, which characterized the natureand extent of contamination of surface and subsurface soils at the site as well as groundwater in the shallow, unconfined Columbia Aquiferand the deep Yorktown Aquifer beneath the site, was submittedto the EPA and VDWM and was subsequently approved. Analytical groundwater data acquired for the RI revealed that detectable concentrationsof polycyclic shallow, unconfined aquifer during the aromatic hydrocarbons (PAHs) were present in the groundwater. Analytical data acquired during the RI also revealed the presenceof the samePAHs in the unsaturated zone soils above the

102

Kotun et al. 10

1

v

2

0.1

U

0.001

o.Ooo1 10

100

1

Total PAHs (mg/Kg)

Figure 4 Totalhazardindex vs. total PAHs-on-site workers.

100

10

1

0.1

0.01

0.001

1

10

100

1O , oo

Total PAHs (mg/Kg)

Figure 5 Totalhazardindex vs. totalPAHs-constructionworkers.

Cleanup Level Development

103

Table 14 Summaryof Polycyclic Aromatic Hydrocarbons (PAHs) Detected in Soil and Groundwater Samples During Remedial Investigation Potentially carcinogenic PAHs

Noncarcinogenic PAHs

Benzo[u]anthracene Benzo[u]pyrene Benzo[b]fluoranthene Benzo[k]fluoranthene Chrysene Dibenzo[a,h]anthracene Indeno[ 1,2,3-cdlpyrene

Acenaphthene

Acenaphthylene Anthracene Benzo[g,h,qperylene Fluoranthene Fluorene Naphthalene Phenanthrene Pyrene

shallow aquifer. Both potentially carcinogenic and noncarcinogenic PAHs (cPAHs and nPAHs, respectively) that are commonly associated with wood-treating and storage activities were identified as contaminants in a public health and environmental assessment (PHEA) that was performed as part of the final RI for the site (Table 14). l h o areas, areas A and B, were delineated as potential source areas on the western and eastern portions of the site, respectively (Figure 6). The presence of PAHs in the soil and groundwater samples collected from these areas implies the possibility of downward vertical movement of organic leachate from the unsaturated zoneto the groundwater of the Columbia Aquifer. Once in the groundwater, the potential exists for the transport of PAHs from beneath source areas A and Bto a receptor domestic welland an adjacent estuarine river, respectively. be used as Although the aquifer is not currently being used as a potable water supply, it could such in the future. Hypothetical off-site domestic wells in areas west of the site are therefore considered to be potential receptors in a future scenario. The adjacent river is considered be to a potential environmental receptor for those PAHs that have leached from the soil into the shallow aquifer at the site. For the purpose of conducting a focused feasibility study, it became necessary to develop soil cleanup goals for PAHs that were protective of (1) humans consuming groundwater at a receptor well location impacted by area A, (2) organisms inhabiting the river, and(3) humans consuming organisms from the river impacted by area B. It was suggested by EPA that this task be performed by applying EPA's Multimedia Exposure AssessmentModel (Multimed) to designated source area(s) at the site [8]. The monitoring well designated MW-l02 (Figure6 ) , installed in the Columbia Aquifer in the southwestern portion of the site, was selected as the receptor welllocationfor PAHs migrating fromareaA, sinceitissituateddowngradient of source area Aand PAHs weredetected in groundwater samples collected from this location. In addition, property west of the site is more likely to undergo residential development than property at any other on-site or offsite location. The adjacent river, as stated previously, was determined to be the potential environmental receptor for constituents migrating from source area B. The purpose of Multimed in the development of soil cleanup goals was the derivationof dilution-attenuation factors (DAFs) that are used as multipliers for selected applicableand appropriate performance standards at the receptor location of interest. This will be explained in greater detail in a later discussion.

104

- +-

I

a CI

Kotun et al.

d W

B

Cleanup Level Development

105

B. Description of Multimed Multimed is a recently developed user-friendly computer model that is capable of simulating the release of chemicals in leachate form from a source (or designatedarea) at the site to soils directly beneath the source. In addition, Multimed can be used to further simulate chemical fate and transport in the unsaturated and saturated zones, followedby possible interceptionof the subsurface plume by a specified receptor (e.g., a well or surface stream). The fate and transport of a chemical released from a source is simulated by incorporating the known responses of the chemical toa number of complexphysical, chemical,and biological processes the chemical encounters as moves it in the multimedia environment. These responses are incorporated as chemical-specificvariable input data by the model user. Other variable input data, such as source-specific and aquifer-specific data, must also be incorporated by the user. For some of the variable input data, the model provides the user with an optionto either manually specify values for the variable input data (constant input) or have the model mathematically derive the variable input data from other constant inputs (derived input). After all relevant input data have been defined and the type of output desired has been specified, the multimedia transport of each contaminant is mathematically simulated by the model. An output file is then generated showing final concentrations of specific constituents and any other pertinent information(i.e., times of concentration Occurrences or statistical distributions resulting from multiple iterations). The final downgradient concentration(s) calculated by the model can be used to represent potential toxic exposureconcentration(s)to which human and/or environmental receptors may be subjected. For this site, deterministic and Monte Carlo simulations of steady-state unsaturated and saturated zone flowand transport were performed using Multimed.A Gaussian boundary condition was applied to the saturated zone transport of the contaminants away from the source, with the maximum concentration occurring at the source. Steady-state conditions in the model were used for the approximation of a system mass balance in which water entering the flow systemis balanced by the water leaving the system. There is no significant temporal variationin the system. Thus, the assumption of a steady-state system basically simplifies the mathematical equations used in describing the flow and transport processes and reduces the amount of input data, since no information on temporal variability is necessary. The primary assumption of steady-state flow and transport is that the source is of a sufficiently large chemical mass to ensure that the final downgradient contaminant concentration in the groundwater is maintained at the receptor location. The source is assumed to be continuous and constant, without decay or any other temporal variation. In the deterministic model of steady-state conditions, each inputvariable is of fixed value and is assumed to havea fixed mathematical relationship with the other variables. Each run of a deterministic model can result in either the output of one maximum concentration or timestepped concentrations occurring over a specified time interval. For this site, the output selected was the maximum concentration that wouldOccur over an arbitrarily selected 500-year period. The deterministic mode of the Multimed model should only be applied to one or more particular modeling situations in which all values for the input variables are known or can be assumed with a high level of confidence. If there is uncertainty as to the values of input variables, then the simulation(s)may be performed within the MonteCarlo framework, wherethe randomness and uncertaintiesof values inherent in the modeled system can be evaluated. Input values in the deterministic model were either constant or derived values. Tables 15-17 present

its

Kotun et al.

106

Table 15 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Unsaturated Zone for Modeling to MW-l02 and the River-Deterministic Model of Steady-State Conditions variable

Input Unsaturated zone material variables m Depth of unsaturated zone Number of layers 4.42" cmlhr Saturated hydraulic conductivity Unsaturated zone porosity Unsaturated zone function variables cm" Alfa coefficient Residual water content Van Genuchten exponent Unsaturated zone transport variables Bulk density of soil for layer @cm3 0.042' m Longitudinal dispersivity of layer Percent organic matter m Thickness of layer

type

1 1 0.38b

0.075' 0.065* 1 .89'

1.4P 0.59

1

Constant Constant Constant Constant Constant Constant Constant Constant Derived Constant Constant

"Literature value obtained from Mulrimed User's Manual, Table 6-2. for sandy loam. bValue obtained from Mulrimed User's Manual, Table 6-3. Represents average porosity for sand (fine and coarse), gravel (fine and coarse), silt, and clay. 'Literature value obtained from Mulrimed User's Manual, Table 6-5, for sandy loam. dLiterature value obtained from Mulrimed User's Manual, Table 6 - 4 , for sandy loam. =Literature obtainedfrom Mulfimed User's Manual, Table 6-8, for sandy loam. 'Value obtained from the calculation av = 0.02+ O.O22L, where UY is longitudinal dispersivity (unsaturated flow in the vertical direction) and L is the depth of the unsaturated zone = Im. Xiterature value obtained fromMultimed User's Manual. Table 6-7,for group B soils.

all values used in defining the unsaturated and saturated zone input variable parameters assumed for the site in the deterministic model. The Monte Carlo method provided a means of applyingthe known uncertainty associated with an input variable to that variable. This uncertainty is expressed as a cumulative probability distribution. For each uncertain inputvariable, a probability distribution must be specified that best describes the frequenciesof Occurrence of measured valuesfor that variable.As the Monte Carlo simulation is run overa large number of iterations (the number of iterations is specified by the user), random values generated froma specified probability distribution are assigned to the variable. For this site, 500 Monte Carlo simulations were performed by Multimed. The probability distribution may be specified as uniform, log,, uniform, normal, log,, normal, exponential, empirical, or the Johnson system of distributions, Relating the input variableto any one cumulative probabilitydistribution may be difficult. The difficulty arises from the fact that the specification of a distribution for an input variable requires a large amount of site-specific data that may not be available. The types of values assigned to the Monte Carlo, steady-state input variables were either constant, derived, or ranges of uniform distribution. For this site, a uniform distribution of values was specified for each select input variable due to the lack of site-specific data for those variables. All values used in defining the unsaturated and saturated zone inputvariable parameters assumed forthe site in the Monte Carlo steady-statemodel are presented in Tables 18-2 1.

107

Cleanup

Table 16 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Saturated Zoneof the Columbia Aquifer to Receptor Well Location MW-102Deterministic Model of Steady-State Conditions Input Chemical-specific variables Acid-catalyzed hydrolysis Base-catalyzed hydrolysis rate Biodegradation coefficient (sat. zone) Dissolved phase decay coefficient Distribution coefficient, Kd Neutral hydrolysis rate constant Normalized distribution coefficient, Koc Overall chemical decay coefficient Overall first-order decay coefficient Reference temperature Solid-phase decay coefficient Source-specific variables Area of western source Duration of pulse Infiltration rate Initial concentration of leachate Length scale of source Near-field dilution Recharge rate Source decay constant Spread of contaminant source Width scale of source Aquifer-specific variables Angle off centerline of plume Aquifer porosity Aquifer thickness Bulk density Distance to receptor,X, Groundwater seepage velocity Hydraulic conductivity Hydraulic gradient Longitudinal dispersivity Mixing zone depth f, Organic carbon content (fraction), Particle diameter Retardation Coefficient Temperature of aquifer Transverse dispersivity Vertical dispersivity pH

0" 0"

0" 0"

-

0"

-a

v 0" 25

0" 8260'

-

0.3178 l00h

90.88'

>lg 0.317g

v

15.1' 90.88' 0

0.38' 5.34 1 .49k 183'

-m

442 0.006

18.3"

-m

0.005"

0.03p

-

18 6.1" 1.02" 6.5

Constant Constant Constant Constant Derivedb Constant Constant Constant Derived Constanf Constant

'

Constant Constant' Constant Constant Derived Derived Constant Constant Derived Derived Constant' Constant Constant Constant Constant Derived Constant Constant Constant Derived Constant Constant Derived Constant' Constant Constant Constant'

.Literature value obtained from Ref. 11. bDistributioncoefficient (not presented by model output) was derived from the calculation K,, = K, f,, where K, is the normalized distribution coefficient and f, is the organic carbon content (fraction). 'Conservative input value assumed. dValue will be zero since it is derived from solid and dissolved phase coefficients, which themselves were assigned a value of zero.

Kotun et al.

108

Table 16 Continued Temporal factors are ignored under steady-state conditions. 'Area was designated as shown in Figure 7. Model assumed area is square, approximated area at 8260 mz with length = width = 90.88 m. %filtration and recharge rates selected for model represents minimum values derived from Hydrological Evaluation of Landfill Performance (HELP) model. hAssumedvalue. 'Spread of Gaussian contaminant source = (width of source)/6. jValue represents mean of the mean porosity values for materials ranging from fine sand to clay-Table 6-3, Multimed User's Manuul. 'Literature value obtained from Multimed User's Manual, Table 6.8. for sandy loam. 'Determined from Figure 7. "'Modelderived value unknown to user. "Longitudinal dispersivity,a, = 0. I X,;transverse dispersivity = aU3; vertical dispersivity = 0.056 d. "No site-specificf, data available. Input value of0.005 is a model default value that falls within range off, values derived fromfom values obtained fromMulrimed User's Manual, Table 6-7,for group B soils using the equationf, = foJ172.4. Wean particle diameter assumed from range givenfor medium sand in Table6-10 of Multimed User's Manual. SActual modelderived R,, value unknown to user.

Table 17 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Saturated Zone of the Columbia Aquifer to the Adjacent River-Deterministic Model of Steady-State Conditions

Variable

Input

type

Chemical-specific variables Acid-catalyzed hydrolysis rate Basecatalyzed hydrolysis rate Biodegradation coefficient (sat. zone) Dissolved phase decay coefficient Distribution coefficient, Kd Neutral hydrolysis rate constant Normalized distribution coefficient. K, Overall chemical decay coefficient Overall first-order decay coefficient Reference temperature Solid-phase decay coefficient Source-specific variables rea of eastern sourcef Duration of pulse Infiltration rate Initial concentration of leachate Length scale of sourcef Near-field dilution Recharge rate Source decay constant Spread of contaminant source Width scale of sourcef Aquifer-specific variables Angle off centerline of plume Aquifer porosity

Input Units

Value 0" 0'

v

oe -

0"

-a 0'

Od 25 0'

100

Constant Constant' Constant Constant Derived Derived Constant Constant Derived' Derived

0 0.38'

Constant' Constant

10,Ooo

-

0.317g 100h 100 >1g d y r

0.317g

m m

16.7

yr-'

Constant Constant Constant Constant Derivedb Constant Constant Constant Derived Constant' Constant

0'

109

Cleanup

Table 17 Continued Input

Aquifer Bulk X, to Distance velocityseepage Groundwater ivity Hydraulic Hydraulic rsivity Longitudinal Mixing fraction), content carbon Organic Particle Retardation coefficient Temperature of aquifer vity Transverse Vertical PH

m

m d y r

-

5.34 1.49k 244'

-

820 0.006

24.4

-

f,

-

'

"C m

-

0.005

0.03

-

18 8.13 1.37 6.5

Constant Constant Constant

Derived"'

Constant Constant Constant"' Derived'" Constanto ConstantP Derivedq Constante Constant" Constant" Constant'

'Literature value obtained from Ref. 1 I . bDistribution coefficient (not presented by model output) was derived from the calculation Kd = K,f,, where K, is the normalized distribution coefficient andf, is the organic carbon content(fraction). cConservative input value assumed. dValue will zero since it is derived from solid and dissolved phase coefficients, which themselves W- assigned a value of zero. Temporal factors are ignored under steady-state conditions. 'Area was designated as shown in Figure 7. Model assumed area is square, approximated area at 10,ooO m* with length = width = IOOm. %filtration and recharge rates selected for model represents minimum value derived from Hydrological Evaluation of Landfill Performance (HELP) model. hAssumed value. 'Spread of Gaussian contaminant source = (width of sourceY6. jValue represents mean of the mean porosity values for materials ranging from fine sand to clay-Table 6-3, Mulrimed User's Manual. 'Literature value obtained from Uulrimed User's Manual, Table 6-8, for sandy loam. 'Determined from Figure 7 . "Modelderived value unknown to user. "Longitudinal dispersivity, aL = O . l X r ; transverse dispersivity = d 3 ; vertical dispersivity = 0.056oL. "No site-specificf, data available. Input value of 0.005 is a model default value that falls within range off, values derived fromf, values obtained fromUulrimed User's Manual, Table 6-7,for group B soils using the equationf, = fJ172.4. Wean particle diameter assumed from range given for medium sandin Table 6-10 of Multimed User's Manual. qActual modelderived Rd value unknown to user.

C. Use of Output Data For Derivation of Site-Specific Soil Cleanup Goals

As stated previously,the purpose of Multimed inthe development of site-specificcleanup goals is the derivation of dilution-attenuation factors.These derived factors are then used as multipliers for selected performance standardsat the receptor locations of interest. Rather than dePAH compound individually, one soil cleanup goal veloping a soil cleanup level for each for total PAHs was calculated that will be protective of the groundwater at monitoring well MW-102,marine organisms in the river, and humans consuming organisms from the river.

Kotun et al.

110

Table 18 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the unsaturated Zone for Modeling to MW- 102-Monte Carlo Model of Steady-State Conditions variable

Input ~

~~~~~

~

Unsaturated zone material variables Depth of unsaturated zone Number of layers Saturated hydraulic conductivity Unsaturated zone porosity Unsaturated zone function variables Alfa coefficient Residual water content Van Genuchten exponent Unsaturated zone transport variables Bulk density of soil for layer Longitudinal dispersivity of layer Percent organic matter Thickness of layer '

@cm3 m

0.613-1.33 1 l .O-150 0.250-0.500

Uniform Constant Uniform Uniform

0.005-0.145' 0.034-0. 100b 1.09-2.68'

Uniform Uniform Uniform

1.25-1.76'

Uniform Derivedd Uniform Uniform

-

0.180-1.30'

m

0.500-2.00'

'Literature values obtained from Multimed User's Manual, Table 6-5, for sandy loam. kiteramre values obtained from Multimed User's Manual, Table 6 - 4 , for sandy loam. 'Literature values obtained from Multimed User's Manual, Table 6-8, for sandy loam. dDerived values obtainedfrom the calculation av = 0.02+0.022L, where av is the longitudinal dispersivity (unsaturated flow in the vertical direction) and L is the depth of the unsaturated zone = lm. 'Literature values obtained from Multimed User's Manual, Table 6-7, for group B soils.

Table 19 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Saturated Zone of the Columbia Aquifer to Receptor Well Location MW-102-Monte Carlo Model of Steady-State Conditions Value(s)

Input

Chemical-specific variables Acid-catalyzed hydrolysis rate yr)" Base-catalyzed (Mhydrolysis rate Biodegradation coefficient (sat. zone) coefficient Dissolved decay phase coefficient, Distribution of K,, constant hydrolysis Neutral rate Normalized distribution coefficient, Overall chemical decay coefficient yrOverall first-order decay coefficient temperature Reference coefficient decay Solid-phase Source-specific variables source'western of Area rate Infiltration concentration leachate mg/L Initial of ourcef of scale Length dilution Near-field te Recharge constantdecaySource

(MY r r l

Yr- I Yr" Yr" K,

&g I

Yr- ' Yr" dYr

0" 0" 0' 0"

-

08 14.2-5,500,0008 0' Od 25' 0"

8260 0.317-0.587'

100"

dYr

Yr- I

90.88 l' 0.308-0.744 OC

Constant Constant Constant Constant Derivedb Constant Uniform Constant Derived Constant Constant Constant Uniform Constant Derived Derived Uniform Constant

Cleanup

111

Table 19 Continued Input

Spread Width Aquifer-specific variables e off Angle Aquifer Aquifer Bulk X, o Distance WYrvelocity Groundwater seepage ctivity Hydraulic Hydraulic gradient ersivity Longitudinal ne Mixing Organic carbon content (fraction),f, Particle Retardation coefficient Temperature16-25' of aquifer Transverse dispersivity Vertical dispersivity PH

type

of ale

v Uniform

183'

m

-

-

Uniform 0.0011-0.0100 18.3"'

m Constant

-

-

Uniform 0.0010-0.0076"

-

-

"C m m

-

Constant

0.26-0.57j

Constant

6.1"' 1.02"' 6.00-9.00'

Constant Derived

Derived Uniform Derived Uniform Constant Constant

"Literature values obtained from Ref. 11. bDistribution coefficient was derived from the equation Kd = K,f,, where K, is the normalized distribution coefficient andf, is the organic carbon content (fraction). 'Conservative input value assumed. dValue will be zero since it is derived from solid and dissc!ved phase coefficients, which themselves were assigned a value of zero. 'Assumed value(s). ' h a was designated as in Figure 7. Model assumed area is square, approximated area at 8260 m*, with length = width = 90.88 m. %filtration rates selected for model represent the minimum and maximum values derived from Hydrological Evaluation of Landfill Performance (HELP) model. hAssumed value. 'Spread of Gaussian contaminant source = (width of source)/6. @orosity values represent range fromfine sand to Clay-Mulrimed User's Manual, Table 6-3. 'Literature values obtained from Mulrimed User's Manual, Table 6-8. for sandy loam. 'Determined from Figure7. "'Longitudinal dispersivity, OL. = O.lX,; transverse dispersivity = oU3; vertical dispersivity = 0.056aL. "No site-specificf, data available. Input value range obtained fromMulrimed User's Manual, Table 6-7, for group B soils using the equationf, = f0,,,/172.4. OParticle diameter range assumed for particle types ranging from fine silt to coarse gravel, given in Table6-10 of the Mulrimed User's Manual.

The groundwater performance standard assumed at well location MW-l02 for the derivation of a soil cleanup goal for total PAHs at area A was the proposed maximum contaminant level (MCL) for benzo[a]anthracene,O.OOO1 mg/L [9]. Although there are various MCLs and DWELs established for the individual PAH compounds, this MCL was selected for the total PAHs since it represents the most health-conservative drinking water standard. The perforof soil cleanup goals for total PAHs that mance standards assumed at the river for the derivation were protectiveof marine organisms and humans consuming organisms from the river impacted by area B were 0.3 and O.oooO311 mg/L, respectively [lo].

Korun er al.

112

Table 20 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Unsaturated Zone for Modeling to the Adjacent River-Monte Carlo Model of Steady-State Conditions

variable

Input

type ~

Unsaturated zone material variables Depth of unsaturated zone Number of layers Saturated hydraulic conductivity Unsaturated zone porosity Unsaturated zone function variables Alfa coefficient Residual water content Van Genuchten exponent Unsaturated zone transport variables Bulk density of soil for layer Longitudinal dispersivity of layer Percent organic matter . Thickness layer of

0.500-2.00 1 1.0-150 0.250-0.500

Uniform Constant Uniform Uniform

0.005-0.145"

Uniform Uniform Uniform

0.034-0.l00b 1.09-2.68'

@cm3 m m

1.25-1.76' 0.180-1 .30' 0.500-2.00"

Uniform Derivedd Uniform Uniform

'Literature values obtained from Mulrimed User's Manual, Table 6-5, for sandy loam. bLiterature values obtained from Mulrimed User's Manual, Table 6-4, for sandy loam. 'Literature values obtained from Mulrimed User's Munual. Table 6-8, for sandy loam. dDerived values obtained from calculation m = 0.02+0.022L, where av is longitudinal dispersivity (unsaturated flow in the vertical direction) and L is depth of the UnSaNrated zone = 1 m. 'Literature values obtained from Mulrimed User's Manual, Table 6-7,for group B soils.

Table 21 Summary of Multimed Input Parameters Used in Modeling PAH Leachate Flow and Transport Through the Saturated Zone of the Columbia Aquifer to the Adjacent River-Monte Carlo Model of Steady-State Conditions ~

Input Variable

type

Input Value(s) Units

Chemical-specific variables Acidcatalyzed hydrolysis rate Basecatalyzed hydrolysis rate Biodegradation coefficient (sat. zone) Dissolved phase decay coefficient Distribution coefficient, K,, Neutral hydrolysis rate constant Normalized distribution coefficient.K, Overall chemical decay coefficient Overall firstader decay coefficient Reference temperature Solid-phase decay coefficient Source-specific variables k e a of eastern sourcef Infiltration rate Initial concentration of leachate Length scale of sourcef Near-field dilution Recharge rate

Od 25" 0"

Constant Constant Constant Constant Derivedb Constant Uniform Constant Derived Constant Constant

10,Ooo 0.317-0.587' 100h 100 'l 0.308-0.744

Constant Uniform Constant Derived Derived Uniform

0"

0" 0" 0"

-

0" 14.2-5,500,0008

W

Cleanup

113

Table 21 Continued Input

Source decay constant Spread of contaminant source Width scale of sourcef Aquifer-specific variables off Angle Aquifer Aquifer Bulk o Distance X, velocity Groundwater seepage uctivity Hydraulic ent Hydraulic ersivity Longitudinal e Mixing (fraction), content carbon Organic r Particle cient Retardation Temperature of aquifer Transverse dispersivity Vertical dispersivity PH

type 0'

p-'

m m

16.7 100

v

.76k

m

mm

0.26-0.57' 4.57-6.10 1.25-1 244'

-

dYr

142-3.780 0.0059-0.0068 24.4"

-

0.0010-0.0076" 0.0004-0.2000"

-

-

f,

"C m m

-

-

16-25" 8. 13" 1.37" 6.00-9.00'

Constant Derived' Derived Constant Derived Uniform Uniform Constant Derived Uniform Uniform Constant Derived Uniform Uniform Derived Uniform Constant Constant Constant

'Literature values obtained from Ref. 11. bDistribution coefficient was derived from the equationKd = K, f,, where K, is the normalized distribution COefficient andf, is the organic carbon content (fraction). CConservative input value assumed. ''Value will be zero since it is derived from solid- and dissolved-phasecoefficients, which themselves were assigned a value of zero. 'Assumed value(s). fh was designated as in Figure 7. Model assumed m is square. approximated area at 1O.ooO m*, with length = width = 100 m. Infiltration rates selected for model represents the minimum and maximum values derived from the Hydrological Evaluation of Landfill Performance (HELP) model. hAssumed value. 'Spread of Gaussian contaminant source = (width of sour~e)/6. jPorosity values represent range from fine sand to clay-Table 6.3. Mulrimed User's Manual. 'Literature value obtained from Multimed User's Manual, Table 6-8, for sandy loam. 'Determined from Figure 7. mLongitudinal dispersivity, al. = 0. I X,;transverse dispersivity = aLJ3; vertical dispersivity = 0.056aL. "No site-specific f, data available. Input value range obtained fromMultimed User's Manual. Table 6-7,for group B soils using the equation f, = fJ172.4. OParticle diameter range assumed for particletypes ranging from fine silt to coarse gravel, given in Table 6-10 of the Multimed User's Manual.

Prior to discussing the actual calculation process, Table 22 presents and defines the parameters that were used in the development of soil cleanup goals. 1. GroundwaterApproach of groundwater at the designated The calculations for developing soil cleanup goals protective receptor well location MW-l02 are as follows.

Kotun et al.

114

Table 22 Parameters Used for Derivation of Soil Cleanup Goals

Riverconcentrationofacontaminant,equaltotheappropriate AWQC DAFDilution-attenuationfactorderivedfromMultimedastheratiobetweentheinitialleachate concentration at the source and the modeled downgradient groundwater concentration at the receptor RM Rivermixing(dilution)factorforcontaminantsintheriver Dd Soilwaterequilibriumpartitioningcoefficientusedforderivinginterimsoilcleanup'goals from steady-state modeling ci Initial contaminant leachate concentration at source Cf Finaldowngradientgroundwaterconcentrationatreceptorlocation CS, Contaminantconcentrationingroundwateratpointofdischargeintoriver,back-calculated from river concentration,C,,, C,, Contaminantconcentrationingroundwateratmonitoringwell MW-102,equaltotheappropriate groundwater performance standard Cl Leachateconcentrationatsource,back-calculatedfromdowngradienttargetgroundwater concentration (performance standard) at receptor location CS Soilconcentrationcorrespondingtointerimsoilcleanupgoal C,,,,

1. Initial PAH concentrations (Ci)are modeled from the designated source area through the unsaturated and saturated zones to the receptor location at monitoring well MW-102. A groundwater concentration at this location (C') is output by the model. 2. The following relationship is then used to calculate a dilution-attenuation factor: DAF -*Ci/Cf 3. Assuming that the target PAH groundwater concentration at location MW-l02 is mul-

tiplying by the model-derivedDAF(derived in step 2) givesthe PAH leachateconcentration (C,) at the source, or C/ = C,, 4.

*

DAF

Finally, multiplication of C, by the distribution coefficient (Kd) results in a soil concentration at the source corresponding to the soil cleanup goal. This partitioning is expressed as

c,

= C&

2. River Approach The calculations for developing soil cleanup goals protective of the river are as follows.

Initialleachate PAH concentrations (Ci) are modeledfrom the designatedsourcearea through the unsaturated and saturated zones to the point of groundwater discharge at the river. A groundwater concentration at this location (C/>is output by the model. 2. The following relationship is then used to calculate a dilution-attenuation factor: 1.

DAF = CiCf 3. Assuming that the target river concentration is Caw,,

the targeted PAH concentration in the groundwaterat the point of discharge intothe river (C,) can be calculated by multiplying with RM, or C,

= Cawv RM

I15

Cleanup Level Development

4.

Multiplying Cgwby the derived DAF from the model gives the leachate concentration at the source (Cl): Cl =

c,

*

DAF

5 . Finally, multiplication of Cl by the distribution coefficient (Kd) results in a soil concentration at the source correspondingto the soil cleanup goal. Thispartitioning is expressed as CS=

c&

3. Soil Cleanup Goals for TotalPAHs The calculation for developing soil cleanup goals for total PAHs that are protective of the groundwater at monitoring well location MW-l02 and the river is as follows. 1. The leachate concentration of total PAHs (C,,TpAH) at each sourceis calculated in the same manner as follows. For protection of groundwater at MW-102, C1,TPAH

=

c s td

DAF

For protection of the river, C~,~= AH C,,,,

RM DAF

Since the MCLs and DWELs are different for each PAH, the lowest groundwater perforis being used for mance standard (O.OOO1 mg/L - proposed MCL for benzo[a]anthracene) totals PAHs as a conservative approach. ( C s , p A H ) must take into account 2. Since the calculationof a soil cleanup goal for total PAHs the Kdof each PAH (Kd,pAH) as well as the percent distribution of each individual PAH (%DpAH) across the area of interest, the following assumptions can be made regarding the derivation of the soil cleanup goal. For each PAH, C1,PAH

=

Cs.PAH/Kd.PAH

(5)

where C~,PAH = C~.TPAH %DPAH

(6)

Substitution of Equation (6) into Equation ( 5 ) yields the relationship c/,PAH

=

(Cs.TPAH

%DPAH)/Kd,PAH

(7)

For total PAHs, Equation (7) becomes

By moving Cs H ,. outside the summation and rearranging, the following expression for the calculation of a soil cleanup goal for total PAHs is obtained

Table 23 presents a summary of the mean percent distribution of each PAH in soil at the site (%D),the distribution coefficient (Kd)of each PAH, and the adjusted distribution coefficient obtained for each PAH by dividing each Kd value by the corresponding %D.

Kotun et al.

116

Table 23 Summary of Diskibution Coefficients for Polycyclic Aromatic Hydrocarbons ~

Distribution Mean percent Adjusted coefficient distribution," Polycyclic aromatic hydrocarbon (cm3/g) %D Benzo[a]anthracene Benzo[a]pyrene Benzo[b]fluoranthene Benzo[k]fluoranthene Chrysene Dibenzo[a,h]anthracene Indeno[l,2,3-cdjpyrene Acenaphthene Acenaphthylene Anthracene Benzo[g,h,i]perylene Fluoranthene Fluorene Naphthalene Phenanthrene Pyrene

Kd

6.33 5.33 10.0

9.72 7.66 1.S5 3.01 2.96 1.45 5.75 2.97 15.6 3.23 2.74 9.83 11.8

value,b K; (cm3/g)

109,005 515,947 27,500 28.292 13,055 1,064,516 8OOO 265,78 1 23 777 12.5 862 70 1,217 8o00 269,360 190 1,218 36.5 1.130 4.7 172 70 712 190 1,610 Sum K: = 2,301,155

6900 27500 2750 2750 loo0 16500

'%D for each PAH = (mean PAH conc./total mean PAH conc.) 100%. bKd =

K,J(%D)/100].

D. Results 1. Dilution-Attenuation Factors Table 24 presents a summary of the dilution-attenuation factors (DAFs) that were derived from the final downgradient groundwater concentrations estimated at two receptor locations (MW-102 and the river) by the Multimed simulationsof steady-state flow and transport from

Table 24 Summary of Dilution-Attenuation Factors from Flow andTransport Models Through Unsaturated and Saturated Zones-U.S. EPA Exposure Assessment Multimedia Model, Steady-State Conditions Modeled groundwater finalModeled concentration dilutionAssumed for total PAHs attenuation initial leachate receptor facto? at for concentration location (mg/Ll total PAHs for total PAHs conditions source" at (mg/L) MW-IO2 River MW-102 River Model type for steady-state PAHs of Deterministic models 9.8Carlo 45 10.2 2.23 100 of PAHs' models Monte

11.71 0 011.9 8.52 8.38 ~~~

'Initial leachate concentration at source location is in the unsaturated zone. 100

m& was assumed for modeling

presentation.

bDilution-attenuation factors (dimensionless)are used for calculations of interim soil cleanup levels. 'Values presented for Monte-Carlo simulations representthe 95th percentile.

117

Cleanup

Table 25 Summary of Interim Soil Cleanup Goals for PAHs Protective of Groundwater and the Adjacent River Protective of humans consuming groundwater from Well MW-102 (mgkg) Deterministic model 10,355

2738

Monte Carlo model"

Protective of humans consuming aquatic organisms from river (mgkg) Deterministic model

Monte Carlo model"

Protective of marine organisms in river (mgkg) Deterministic model

403,852,703 35,067 41,866

Monte Carlo model" 338,269,785

'Interim soil cleanup goals presented for the Monte Carlo model represent the 95th percentile.

the source areas of the site, through the saturatedand unsaturated zones. The table shows that two DAFs were determined for each receptor location. One DAF was derived based on the results of a deterministic model, the other based on the 95th percentile results of a Monte Carlo model of 500 iterative simulations. The groundwater concentration of all PAHs estimated at receptor location MW-102by the deterministic and Monte Carlo simulations of steady-state conditions were 8.38 and 2.23 mg/L, respectively. Sincethe original leachate concentration was arbitrarily assumed to be 100 mg/L (for ease of presentation), the corresponding DAFs derived from the results of the deterministic and Monte Carlo modeling efforts to receptor location MW-102 are l l.9 and 45, respectively. Thegroundwaterconcentrations of all PAHs estimated at thepoint of groundwater discharge to the river by the deterministic and Monte Carlo simulations of steady-state conditions were 8.52 and 10.2 mg/L, respectively. Since the original leachate concentration was 100 mg/L, the corresponding DAFs derived from the results of the deterministic and Monte Carlo modeling efforts to the river are 11.7 and 9.8, respectively. 2.SoilCleanupGoals the The DAFs derived and summarizedin Table 24 were usedto calculate soil cleanup goals for site that are protective of (1) the groundwater at monitoring well MW-102, (2) humans consuming aquatic organisms from the river, and (3) marine organisms in the river. These soil cleanup goals are presented in Table 25. Table 25 shows that the soil cleanup goals for totalPAHs in area A, derived from the results of the deterministic and Monte Carlo simulations, for protection of humans consuming groundwater at the receptor well location were 2738 and 10,355 mgkg, respectively. The soil cleanup goals for total PAHs in area B,derived fromthe results of the deterministic and Monte Carlo simulations, for protection of the humans consuming river organisms were 41,866 and PAHs in area B, derived from the 35,067 mgkg, respectively. The soil cleanup goals for total results of the deterministic and MonteCarlo simulations, for protection of river organismswere , OOO , OO parts per million. Based on these soil cleanup goals and the exboth greater than 1 O isting PAH concentrationsdetected at the site, no soil remediationof PAHs would be necessary in either area A or B.

E. Conclusion Environmental fate and transport modeling of contaminants in the multimedia environment provides an alternative means of developing and establishing cleanup goals for potential source

118

Kotun er al.

areas at hazardous waste sites. As was shown in this case, cleanup goals canbe derived from modeling outputs that protect potential human and/or environmental receptors from contaminants as they become mobilized following release into the environment. The soilcleanup goals derived for thissite were to protect both human and environmental receptors fromPAHs originating from two sourcelocations, with the most conservativesoil cleanup goals being derived from both deterministic and Monte Carlo models for the protection of humans consuming groundwater as drinking water. These soilcleanup goals were 2738 and 10,355 mgkg, respectively. The uncertainties associated with applying Multimed to the derivation of soil cleanup goals for this site are discussed in the following paragraphs. In the deterministic models of steady-state conditions, literature valueswereused for chemical and physicalproperties of individual PAHs. These values may not be appropriate for the actual existing conditionsat the site. Many of the literature values were obtained from laboratory conditions or field conditions different from those atthe site. Many site-specific conditions may cause the chemical and physical properties and behaviors of the PAHs to deviate from values reported in the literature. In addition, the model evaluates chemicals separately. The behavior in the environment of chemicals that are constituents of mixtures, such as PAHs in creosote, may be different from what their behaviorwould be if they were interacting individually with the environment. The MonteCarlo model of steady-stateconditions assumesa constant, nondecaying source of large area and sufficient chemical mass to forcethe modeled system into steady-state conditions and equilibrium, such thata constant downgradient groundwaterPAH concentration is maintained at all times. In reality, however, the source strength may decay over time as PAHs migrate away (downgradient) from the sourceor degrade naturally. An uncertainty associated withthe Monte Carlo mode exists in the random generation of values from a specified distribution. It is uncertain whether the model considers interdependencies thatmay exist betweenor among many of the input variables. For example, the organic carbon partition coefficient (K,) may, in reality, change with the changing pHof a system. This is probably ignored by the model, especially when K, values are entered as constant input. Another consideration for uncertainty also existsin Monte Carlo simulations. Since there was a very limited base of site-specific data for each input variable, the uniform probability distribution was best suited for the input variables because of the degree of uncertainty associated with them. Hydraulic conductivity,for example, is estimated tofollow a log-normal distribution, and application of a uniform distributionmay not be appropriate, but due to the lack of data for this parameter, it was the only option available. Other overall uncertainties were associated with the use of the Multimed model for this site. These include (1) the uncertainty resulting froma lack of sufficient aquifer-specificdata for calibration of the model to actual conditions beneath the site; (2) the uncertainties that exist in parameter estimation from literature values, especially for values presented fora particular variable for differenttypes or classifications of unsaturated andsaturated zone materials (i.e., soils), noneof whichmay adequately match the materials in the unsaturated and saturated zones at the site; (3) uncertainty associated with theselection of a representative location and area geometry, since the size of each source area;and (4) the uncertainty associated with source model assumed that the geometry of each source area at this site was square, which may not represent the actual geometryof the area. Selection of the area geometry will affect how the plume is modeled.

Cleanup Level Development

119

V. CASE STUDY W: DETERMINATION OF SOIL CLEANUP GOALS PROTECTIVE OF GROUNDWATER QUALITY USING ANALYTICAL SOLUTIONS A. Introduction During the 1980s, a large number of hazardous wastesites were investigated to determine the potential impactsof these sites on human health and the environment. The detail with whichthe investigations were conducted varies greatly fromsite to site. Databases for sites can be composed of several to several thousand sample analyses and observations. Despite the reliability of a given database, emphasis is currently on the remediation of sites having U.S. Environmental Protection Agency (U.S.EPA) promulgated Records of Decision (RODS). Within a ROD, remedial goals are established for those media presenting unacceptable risk upon exposure or for any medium representing sources or potential sources of continuing contaminant release. Remedial goals designed to abate human health risks can be derived once the rate and duration of exposure to a given medium have been determined. Rate andduration of exposure are based onsite-specificinformation such as area demography and land use patterns. Exposure is defined in the Public Health and Environmental Assessment (PHEA)section of the remedial investigation (RI) report for any site. However, establishing cleanup goals protectiveof actual or potential contaminant releaseis more difficult and may require somefate and transportmodeling, particularly in the case of soil contamination and groundwaterprotection. Soil cleanup goals protectiveof underlying groundwater canbe derived by developing simple, logical models of site-specific contaminant migration from affected soils to groundwater and, if necessary, to an alternate point of compliance (APC). An APC is a hypothetical boundary beyond which contamination cannot extend at concentrations exceeding some minimum performance standard. APCs are established byEPA and can extend to the nearest potential household or be limited to the downgradient source area boundary. Once the APC is established, the site-specific conceptof contaminant migration can then be approximated by the use of site data and/or mathematical models. The use of mathematical models to describe the movement of groundwater is not new. Mathematical models representativeof groundwater flow regimes have been used by engineers and scientists since the late 1800s [12]. Application of groundwater models to pollution transport and management problems has become increasingly popular over the last twenty years due to the enactment of legislation such as the Federal Water Pollution Control Act of 1972, the Safe Drinking WaterAct (SDWA) of 1976, the Clean Water Act of 1977 (CWA), andthe Comprehensive Environmental Response Compensation and LiabilityAct of 1980 (CERCLA). The complexity of the available groundwater models varies greatly. Selection of a model for use at a site is therefore based on factors such as the complexity of the groundwater flow regime,the importance of modeling resolution, and the data available to the engineer or scientist doing the modeling. Under relatively simple hydrogeological conditions or when simplifying assumptions about the flow regime are appropriate, contaminant migration can be approximated by the useof mathematical models such as analytical solutions or analytical models. The distinction between the two is more than semantic. Analytical solutions solve a very simple process equation by hand calculations. They require few site-specific analytical data and provide conservative estimates of contaminant transport. Analytical models solve more complex process equations

120

Kotun et al.

with the aid of a computer program. The data needs and the need to develop mathematical solutions that describe boundary conditions for analytical models are more extensive and are open to conjecture. This section discusses the useof analytical solutions in the determination of soil cleanup goals protectiveof underlying groundwater quality accepted by EPA, RegionV (EPA V), at an active CERCLA site in Texas. The site is a former wood-treating facility that operated for over 50 years. Data necessary for the use of more complex analytical modelsare not available. All process-related structures have been removed, andportions of the site are currently occupiedby other industries. The ROD issued for this particular site by EPA V specifies that all soils containing in excess of 700 ppm of carcinogenic polynuclear aromatic hydrocarbons(cPAHs) must be remediatedtoprotectindividualspotentiallyexposed by directcontact(dermalcontactand accidental ingestion). The ROD also specifies that the remaining soils must be remediated to achieve the “no leaching potential” in order to protect underlying groundwater quality. Since the underlying shallow aquifer is not currently usedas a drinking water supply and Texas legislation currently prohibits the further development of groundwater for potable use in the area, the attainment standard for cPAHs in groundwater is less than 10 ppb (not detected) at the APC, designatedas the downgradient property boundary. Possible impacts on the deep aquifer of an improperly abandonedwell thought to exist inthe southern portion of the site must also be addressed. The deep aquifer is a potential drinking water source; therefore, an attainment standard of 0.2 pg/L corresponding to the proposed maximum contaminant level (MCL) for benzo[a]pyrene is warranted. The site conceptual model of contaminant migration (site conceptual model) at the site uses EPA’s Organic Leaching Model (OLM) [l31 to account for cPAH soil leaching andEPA’s Vertical and Horizontal Spreading (VHS) model [l41 to account for dilution in the shallow saturated zone afforded by dispersion. A two-dimensional horizontal flow model representing the presence of a continuous solute line source representsthe potential migration (and subsequent dilution) of cPAHs in the deep aquifer emanating from the improperly abandoned well. Contaminant attenuation is afforded primarilyby the leaching portion of the site concep tual model. Dispersion in the saturated zone is minimal becauseof the relatively shallow saturated zone under consideration and proximity of the soil sourceareas to the APC. Attenuation by leaching is also a controlling factorfor the potential impacts on groundwater quality inthe deeper aquifer. Derivation of soil cleanup goalsby these methods provided a timely, cost-effective alternative to complex numerical modeling. Furthermore, the conservatism of the results may account for someof the temporal and spatial variations inherent in even the simplest groundwater flow domains not accounted for in complex models. Upon completion of the modeling effort and a qualitative sensitivity analysis, site-specific data were obtained to verify the findings of the leachateportion of the model that provided the majority of contaminant attenuation. These data will be summarized and analyzed to ensure that the theoretical site concept of cPAH migration does not underestimate the site-specific potential for continuing release of cPAHs from soils. If site data do show that the modeling resultsare underestimates or overestimates of actual conditions at the site, the model will be modified to account for the difference.

B. Purpose The ROD issued in September 1988 for the site stipulates that during the initial stages of the remedial design, contaminated soilareas will be remediated if they exceedeither the risk-based soil cleanup goal of 700 ppm of cPAHs or the leaching potential-based cleanup goal (“no

Cleanup

121

leaching potential”). The leaching potential-based goal is defined as the residual cPAH soil concentration that will not leach and impact groundwater beneath the site. Although a detailed hydrogeological assessment was conducted at the site during the remedial investigation, chemical analytical data and data requirements for groundwater flow modeling are limited. A visual soil investigation was attempted to determine the extent of contaminated soils, and only select samples of the visually impacted soils were analyzed for the presence of cPAHs. The remedial investigation was concluded in 1986, and nofurther studies were conducted. The development of a remedial design work plan was initiated by one of the potentially responsible parties (PRPs) in 1989 in responseto the promulgation of the ROD by EPA V. In early 1991, acceptance of the Remedial Design Work Plan byEPA V was contingent upon addressing the “no leaching potential.”) Meetings with the PRP and EPA V resulted in defining the no leaching potentialas a total cPAH concentration in leachate emanating from contaminated soils not exceeding 10 pg/L at the downgradient property boundary. A detection is not categorized as a potential drinklimit of 10 pg/L was chosen because the shallow aquifer ing water unit because of its low yield and general water quality.If the aquifer were categorized been selected as a potential potablesource, the proposed MCL for benzo[a]pyrene would have as the minimum attainment standard for cPAHs. MCLsare enforceable standards for drinking water suppliesor potential potable sources.EPA V was concerned not only with protecting the shallow water-bearing unit fromfurther deterioration but also with the possible impact on the deep aquifer, in which an improperly abandoned wellis thought to exist. An improperly abandoned deep well could act as a conduit fromthe shallow aquifer tothe deep aquifer. The deep well was installed around 1912 but is no longer in service and has not been used since the late 1950s. However, records of proper abandonment do not exist. EPA V therefore stipulated that soil cleanup goals protective of shallow groundwater qualitymust also be protective of the deep aquifer, which is used as a supply of potable water in the immediate area of the site. A minimum attainment standardof 0.2 pg/L corresponding to the proposed MCLfor benzo[a]pyrene was established as the minimum attainment standard for this water-bearing zone. EPA V also concluded thatthe use of models if applicable would be acceptable but subject to EPA’s review and final approval. The purposeof this section is to present the rationale for the development of a soil cleanup goal protective of groundwater quality in the shallow and deep aquifers using available site information and analytical solutions that will withstand an EPA V review. Acceptance of the soil cleanup goal by EPA V would provide the ROD-specifiedno leaching potential and result in the completion of the remedial design work plan for the site.

C. Application 1. Development of the Site Conceptual Model The conceptual model for the site is based upon information about the source areas and hydrogeology at the site presented in the final remedial investigation report. The regional geologic setting for the site is the Quaternary Gulf Coastal Plain of Texas. This region comprises a series of sedimentary depositional plains, the youngest of which is of recent, postglacial deposition (Holocene deposits). Sediments of the Holoceneare deposited alongthe coast and in the alluvial flood plains of existing river systems. The site is located specifically in the surface sediments of the Beaumont Formation. The average depthof the Beaumont Formation is 0-20ft below grade. The Lissie Formation lies below the Beaumont and extends to approximately200 ft below the surface. Themajority of the historical site soil borings were drilled to a depth of 65 ft, although deeper borings

122

I. I I l I I

I I

I I I I I

I I I l I I I l

Kotun et al.

123

Cleanup

were drilled to depths of200 ft below the surface. Sediments making up these units consist of a top stratum of cohesive soils (sandy claysand silty clays) and a substratum of cohesionless soils (silty sands, clayey sands, and poorly graded sands). Primary water-bearing zones and their corresponding thickness encountered at the site are organized as follows: shallow zone (10-21 ft), intermediate zone (115-127 ft), and deep zone (174-200 ft).A generalized cross section of these zones is presented in Figure 7. The 1986 RI report states that the shallow water-bearing units consist predominantly of silty and clayey sands with occasional gradations to sand and clayey silts.The base of this unit is irregular and slopes gently to the east. Shallow zone groundwater trends at the site generally slope to the westat a gradient averaging about 20 ft/mi. Site surveys indicated that the shallow zone extends continuously offsite to the west, toward a bayou. This drainage feature may as act a groundwater discharge area, thereby influencing westward-trending gradients. A surface impoundmentlocatedoffsitealongtheeastpropertylinemaycauselocalizedgroundwater mounding, which also would produce westward gradients in the shallow zone. The hydraulic conductivityof the shallow zone was evaluated by conducting falling-head field permeability tests at selected well locations. Table 26 presents the field hydraulic conductivity test results for wells screenedin the shallow water-bearing zone. Soilsof the waterbearingunit are alsotypedforfutureconsideration.Measuredfieldhorizontalhydraulic 2.4 X to 2.2 X c d s e c andaverconductivitiesfortheshallowzonerangedfrom aged 8.3 X low4cdsec. The next significant water-bearing unit is the intermediate water-bearing zone, which is overlain by a clay layer and is not considered to be affected by historical site activities. Hydraulic conductivities for this zone are not currently available. The continuous nature of the intermediate aquitard excludes this zone from the modeling effort. The deep zone is also considered to be a confined water-bearing unit. However, historical data indicate that a deep well was screened in this unit, and records of the well’s abandonment are not available. Physicaldata are not available for this particular zone. It is believed that the deep zone gradient slopes to the south-southwestat approximately 6 ft/mi. Vertical seepage rates between shallow and intermediate water-bearing units are extremely low because of the nature of the confining strata. However, a potential conduit between the shallow zone and the deep may exist in the form of a leaking improperly abandoned well in the southern portion of the site. For this reason, the deep water-bearing unit is a concern EPA of V and is considered in the modeling effort. Given the natureof the site and the characteristics of the underlying strata, contamination of the shallow zone by infiltration of precipitation is the most obvious groundwater contami-

Table 26 Field Hydraulic Conductivity Test Results4hallow Groundwater Zone Field horizontal hydraulic Monitoring well MW0 1 MW02 OW07 OW08

Soilconductivity Screen type interval (ft)

Sand Silty sand Sand

Sand

9.0-21 .O 9.0-21 .O 11.0-16.0 14.0-19.0

(cdsec) 2.4 4.5 4.5 2.2

X 10-~ X IO-^ X 10-~ X 10-~

Avg. 8.3 X

Kotun et al.

124

nation mechanism. Contamination reaching the shallow zone would in all likelihood move in the westward direction of shallow zone groundwater migration. Vertical and horizontal dispersion would then occur, withvertical movement bounded below by the intermediate aquitard. If the improperly abandoned wellexists, contaminants might alsoenter the deep water-bearing zone in which the well was screened. Groundwaterin the deep water-bearing zonein which the well was screened. Groundwater in the deep water-bearing zone trends toward the southern property boundary. The site conceptual model consists of three elements: (1) leaching of constituents of interest (COIs) from affected soils, (2) migration and subsequent dispersion of the COIs with respect to shallow groundwater flowdirection, and (3) migration from the shallow zone tothe deep water-bearing unit throughthe open conduit and subsequent dispersion. The site conceptual model is presented in Figure 8. An important aspect of this conceptual model isthat it is chemically conservative, assuming that the cPAHs are not biodegraded or adsorbed to soils before, during, or after migration.

2. ConstituentMobility Polynuclear Aromatic Hydrocarbons (PAHs), in general, are immobile constituents in environmental media. Recent research shows that PAHs bind to soil surfacesas a result of their van der Waals forces [15]. Van der Waals forces act solely between molecules within close proximity of each other. As a rule, the larger the molecular size, the greater the van der Waals forces. cPAHs, being generally larger than noncarcinogenic nPAHs, are even less mobile in environmental media. Similarly,relatively low water solubilitiesand vapor pressures add to the inherent environmental immobility of cPAHs. A semiquantitative assessment of theoretical mobility developed by Laskowski et al. [l61 can be used to describe the immobility of cPAHs in the environment based onknown physical and chemical constants. The basis for the assessment is an algorithm that utilizes water solubility (S), vapor pressure (VP), and the organic carbon partition coeffkient (K,) of the constituents to determine a relative mobility index (MI). The MI is defined as MI = log [(S VP)&]

(10)

A relative scale is then used to evaluate the MI derived for eachcPAH [17]. The scale is a descriptive one, comparing a numerical MI to the categories extremely mobile, very mobile, slightly mobile, immobile, and very immobile.

Description indexMobility >S

0 to 5 -5 to 0 -10 to -5

220 7.62 ND

mag

1

"F

5

-

ND

mg/L

.01 1 .5 .5

ND

mag "F

1 5

>220 8.43

ND ND

mag

-

mag mglL mgIL

0.01

1 .5 .5

ND = not detected above the respective dection limit.

For the best chemical performance, the asphalt should have high contents of pyridinic, phenolic, and ketone groups, which can be achieved by carefully choosing the source material. If the situation requires special stability or redundancy, small amounts of shale oil and lime can be used as additives. Situations and conditions that favor the presence of inorganic sulfur, monovalent salts, and high ionic strength solutions in the asphalt should be avoided, because these conditions decrease the chemical stabilityof the asphalt cementby disruption of the functional groupaggregate bonds and by increasing the overall permeability. However, these conditions are not expected in the anticipated uses of asphalt cement to stabilize contaminantsin metals-affected soil using environmentally processed remedial technology.

XI. DISCUSSION OF USE Environmentally processed asphalt can at best be described as user-friendly. There currently exist a multitude of uses for cold-mixEPA incorporating affectedsoil. One of the more viable and creative is keeping contaminated soils outof landfills as a waste and placing itin landfills as an end product [7]. The imminent closure of many of the nation's Class 111 and municipal landfills creates the potential use for hundreds of thousands of tons of contaminated soils incorporated into asphalt for use as a landfill liner or cap. The cost effectivenessof this method of capping landfills is very attractive to financially strained municipalities.Prior to the advent of EPA for useas a liner or a cap, clay wasthe specified material. In addition to environmental concerns associated with mining vast quantities of clay for these uses,the cost of landfill closure had no cost recovery options. By using EPA, the municipalities and landfill owners can charge attractive fees for the acceptance of affected soil.In mostcases, this acceptance fee pays for the cost of on-site processing of the affected soil into the asphaltend product. The effectiveness of the cost recovery is obvious as the capping materials production process becomes a profit center. By the use of on-site material, not only is the cost of obtaining the clay canceled, but transportation costs also are eliminated. In essence, the capping processof landfill closure is more affordable, makesuse of a product far superior to the traditional clay method, and reduces a broad spectrum of environmental concernsby keeping affected soil outof landfills as a waste. Instead it places affected soilas environmentally sound end products such as caps or liners.

Soil Remediation

305

Studies of asphalt, clay, and other membrane liners subjected to a variety of aging tests in exposure columns at various temperatures, pH conditions, oxygen concentrations, and hydroand static pressures havebeen discussed [ 8 ] . The conclusions were that the asphalt liners membranes were extremely stable chemically and physically. An aging period equivalentto 7 years produced penetration of reaction products to only 0.5 mm (0.5% of the 10-cm liner thickness). 6%, these linerswould The results showed that if the asphalt content of the liner exceeded about perform adequately under impoundment conditions for over 1000 years, conditions that are similar to those expected for cold-mix EPA [17]. Catalytically blown asphalt was considered the best liner material and was selected for long-term field testing. Field testsof catalytically blown asphalt over a 2-year period showed superior performance of the asphalt liners compared be especially true for the petroleum constituents in cold-mix to thatof the clay liners. This will EPA’s liner; overall, asphalt is a much better liner material for this application than clay. Its use as a cap or liner is only one exampleof this product’s cost effectiveness and versatility, but what of its more traditional useas a pavement? To best describe “user-friendly,” one should visualize a typical multilane high-traffic-volume freeway and the load-bearing capability and durability that must be designed into the asphalt product used in its construction. Now visualize the typical bicycle path windingway itsthrough our urban areas. The point being that both the freeway and the bicycle path are asphalt pavements, but their end uses are drastically different. Cold-mixEPA pavement is certainly nothingnew. There are very few, if any, state and county road departments that do not use variations of cold-mix EPA. The end use of asphalt dictates its specifications, or better said, if the asphalt mix will perform its required function, from freewayto bicycle path,it is within specifications. In fact, the ASTM procedure for cold-mix asphalt design includes a section that states that the mix must fulfill the requirements of its intended application. Recalling the term “user-friendly” it becomes apparent that the function of the end product will determine the asphalt mix design. Pavement for a heavy equipment yard has been constructed from EPA made with affected soil recovered from leaking underground tanks. By producing parking lot pavement for on-site of their contaminated soil in a use, the generator eliminated the inherent liability of disposing dump site. Approximately $80.00/ton of disposal taxes were saved as the materials were recycled and not disposed of. The pavement produced not only kept the project’s pricing below any other option but created a paved parking lot of extremely low permeabilityto prevent further adverse subsurface impact. The mix design was not the same as that required to construct a freeway, but then a freeway was not the intended use. The intended use was for low traffic volume but required extremely high load-bearing strength. Another project used affected soil from an oil tank spill for paving loading and unloading facilities at an oil refinery. Again, the affected soil was not disposed as of hazardous waste but was recovered and used in a cold-mix asphalt pavementto remediate the affected soil and prevent further contamination, and the mix design was consistent with the end use. Stability or strength (as measured by the Marshall test) achievedby various mix designs of cold-mix EPA is presented in Table 4. The minimum Marshall stability required for paving mixtures, for example, is 2224 [18]. Mix designs used for actual applications range 95% from a contaminated soil (native silt, sand, and gravel contaminated with diesel fuel to 32,000 ppm total petroleum hydrocarbons) with a 5% emulsion, to a 5% contaminated soil (heavy black clay contaminated with machine cutting oils to 55,000 ppm total petroleum hydrocarbons) plus 90% Class I1Y4-in. or less base rock and 5% emulsion. To date, EPA has been successfully used on projects ranging from road base and road pavement to containment dikes and drain channels. The procedure wasto determine the requirements, then design the EPA mix to fit the use. AS the equipment used to produce EPA is portable and certainly not complex, field test field batches of20 tons or more are used rather than bench-scale tests. In this manner the actual

306

Patton

Testa and

Table 4 Summary of Marshall Test Results for EPA Stability number:5 Sample

4

3

1

2

6

62-64 5 5.2

62-64 5 5.2

62-64 5 5.2

62-64 6 5.2

62-64 6 5.2

62-64 6 5.2

2.08 1116.1 584.4 1120.1 2 5/8" 3100 2880 30 NTW

2.07 1125.1 586.5 1129.0 2 11/16" 2350 2090 27.5 2520

2.11 1130.0 599.0 1134.1 2 518" 2800 2600 31

2.05 1120.1 577.2 1124.0 2 11/16" 2450 2180 31

2.05 1120.7 577.5 1124.9 2 11/16" 2200 1960 24

NT

NT

2.07 1122.6 584.4 1126.5 2 11/16" 2250 2000 30 2050

X

X

75/25 Blend; 314-in. Class I1

base and contaminated soil Asphalt in emulsion" Residual asphalt in mixtureasb Total mix watef Compacted specimen data (emulsion in percent) Bulk density Weight in air Weight in water Weight SSD Thickness Stability Adjusted stability Flow Average stability number:

Sample 9

8

7

85/15 Blend; 314-in. Class I1 base and contaminated soil Residual asphalt in mixtureasb 6.5 6.5 6.5 Bulk density 2.002.022.00 3548 Stability 24-Hour soak NT 14number:

total

Sample 13

12

85/15 Blend; 314-in. Class I1 base and contaminated soil mixtureamb in Residual asphalt 2.03 Bulk density2.01 2.03 Stability 24-Hour soak absorbedMoisture Maximum

' In percent Emulsion e

Not tested

NT

10

6.5 2.02

NT

3260 NT

1410

NT 1056

11

6 . 2.01 3158 NT X

6

6

6

NT

2640

NT

1577

NT

1201

X

X X

'

Soil Remediation with Asphalt

307

3000

I

0

z -

2500

-

2000

-

W

$

Moximumdrydensity = 122pcf Optimummoisturecontent = 10.47,

0 CK

1500

X Standard

W

Q

v,

100.4

1000

n

% Standard 95.5

Z

1 500

2

0 0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

PENETRATION INCHES Figure 4 California bearing ratio for Class I1 base using cold-mix EPA.

mix is tested rather than a small hand-mixed batch. The bearing ratio for processed EPA for Class I1 base is presented in Figure 4.

XII. CONCLUSIONS In consideration of certain factors including durability, chemical resistance andaging, biological resistance, permeability, and leachability, cold-mix EPA is anticipated to perform more than adequately under normal conditions fora long period of time, probably over lo00 years. The use of EPA as a liner, cap, or any number of other site-specific applications has vast potential. Hazardous waste cleanup projects become cold-mixasphalt production projects. Contaminated soil becomes a recoverable resource and is within the letter, spirit, and intent of current regulations. Under Californiaregulations, for example, non-RCRA regulated recyclablematerials used in a manner such that they are not considered to be “used in a manner constituting disposal” are not subject to the provisions of Health and Safety Codes of the State of California, subsection 25143.2(b). Thus, if the recyclable materials satisfy the conditions of subsection 25143.2, then (1) they are not considered hazardous waste and (2) they are conditionally exempt from the California Departmentof Health Services hazardous waste regulations. Providing the conditions summarized above are met, permanent fixation of petroleum- and metalscontaminated soils via asphalt incorporation is a viable, cost-effective soil remediation option that can be accomplished withina relatively short period of time with minimal long-term liability [18]. Furthermore, although highway-type paving material as a resultant product is limited, multiple secondarymarketsexist,includinglinersandlandfill caps, roadbase, dust abatement, bank stabilization, and paved storage areas, among other uses.

308

Testa and Patton

Table 5 EPA Costs vs. Disposal Costs for Non-RCRA Hazardous Wastea ~

Task description

~

~~

costs Disposal costsEPA

Difference

Site investigation Characterization and work Plan Profiling and agency approval Excavation Transportation

Same Same

Same Same

0 0

Same Same 0 for on-site processing $5.00 per ton for central plant processing

Same Same

0 0 0

NA

$25.00 per ton $65.00 per ton

$ 10.00

Disposal costs Processing costs Taxes and fees' Superfund (HS) HAMY County tax (10% of disposal costs) Site restorationd Product cost recovery

$35.00 per ton (avg)

NA

$ 30.00

0 0 0

$7.88 $26.25 $6.50

$ 7.88 $ 26.25 $ 6.50

Same

Same

$20.00 per ton (avg)

0

NA

0 $ 20.00

Total per ton

$100.63

All costs in 1992 dollars. NA = notapplicable. State taxes and fees shown are from a recap of hazardous waste fees for fiscal year 1991-92 compiled by the State Board of Equalization. Disposal costs shown above do not includegenerator fees or hazardouswastereporting surcharge. Site restoration costs where cold-mix EPA was used on site would be the same; however, the product cost recovery would be deducted from the gross site restoration costs, thereby providing a much lower actual net cost.

REFERENCES 1. Asphalt Institute, Principles of Construction of Hot-Mix Asphalt Pavements, The Asphalt Institute, College Park, Md., Manual Ser. No. 22, 1982. 2. Preston, R. L., and Testa, S. M., Permanent fixation of petroleumcontaminated soils. Proc. Nut. Res. Develop. Conf. on the Control of Hazardous Materials, Anaheim, Calif., 1991, pp. 4-10. 3. Testa, S. M., and Patton, D. L., Paving market shows promise, Soils, Nov.-Dec. 1991, pp. 9-11. 4. Testa, S. M., Patton. D. L., and Conca, J. L., The use of environmentally processed asphalt as a contaminated soils remediation method,Proc. Hazardous Materials Control Res. Inst. Conf. (Hazardous Materials Control South), New Orleans. La., 1992. 5. Testa, S. M., Patton, D. L., and Conca, J. L., The use of environmentally processed asphalt as a contaminated soil remediation method,Petroleum in Conmminated Soils, Vol. 4, E. J. Clarence and F? T. Kostecki, eds., Lewis, Chelsea, Mich., 1992. 6. Testa, S. M., Patton, D. L., and Conca, J. L., Fixation of petroleum contaminated soils via cold-mix asphalt for use as a liner, Proc. Hazardous Materials Control Res. Inst. Conf., HMC South, New Orleans, La., 1992, pp. 30-33. on 7. MRM Partnership, Bituminous and Asphaltic Membranes for Radioactive Waste Repositories Land, Report to Dept. of the Environment, DOuRW187.009. Bristol, England, 1988. 8. Buelt, J. L., Liner Evaluationfor Uranium Mill Tailings, Final Report,PHL-4842, Pacific Northwest Laboratory, Richland, Wash., 1983. 9. Eschrich, H., PropertiesandLong-TermBehavior of BitumenandRadioactiveWaste-Bitumen Mixtures, SKBF KBS Tech. Rep. 80-14, Swedish Nuclear Fuel and Waste Management Company,

Stockholm.

Soil

Asphalt

309

10. Benedetto, A. T., Lottman, R. F?,Cratin, F? D., and Ensley, E. K. Asphalt Adhesion and Interfacial Phenomena, Highway Research Record No.340, Nat. Research Council Highway Research Board,

Washington, D.C., 1980. 11. Hickle, R. D., Impermeable asphalt concrete pond liner, Civil Eng., 1976, 56-59. 12. Atlas, R. M. Microbial degradation of petroleum hydrocarbons,an environmental perspective,Microbiol. Rev. 1981, 180-209. 13. Harris, J. O., Preliminary studies on the effect of micro-organisms on the physical properties of asphalt, Trans. Kansas Acad. Sci., 61, 110-113 (1958). 14. Jones, T. K., Effects of bacteria and fungi on asphalt,Material Protection, 4 , 39 (1965). 15. Testa, S . M., and Conca, J. L., When contaminated soil meets the road, Soils, December 1992, pp.32-38. 16. Conca, J. L.,and Testa, S. M., Chemical aspects of environmentally processed asphalt, Int. Symp. Asphaltene Particles in Fossil Fuel Exploration, Recovery, Refining and Production Process, Las Vegas,Nev.,1992. 17. Haxo, H. E., Jr., Assessing synthetic and admixed materials for liner landfills, including gas and leachate from landfills, in Formation, Collection and Treatment (E. J. Genetelli and J. Circello. eds.), EPA Rep. 600/9-76-004,U.S. Environmental Protection Agency,MTIS Rep. PB251161, Cincinnati, Ohio, 1976, pp.130-158. Soils, May-June 18. Testa, S. M., and Patton, D. L. (1992). Add zinc and lead to pavement recipe, 1992, pp, 22-35.

Additional Reading Asphalt Institute (1990). Asphalt Cold Mix Manual, 3rd d.,Asphalt Inst. Manual Ser. No. 14 (MS-14), The Asphalt Institute, College Park, Md. of Hazardous Waste, Van Nostrand Reinhold, Conner. J. R. (1991). Chemical Fixation and Solidification New York. U.S. Environmental Protection Agency (1991). Treatment of lead-Contaminated Soils, Superfund Engineering Issue.

This Page Intentionally Left Blank

14

Lead Decontamination of Superfund Sites

Ann M. Wethington, Agnes Y. Lee, and Vernon R. Miller US. Bureau of Mines Rolla. Missouri

1.

INTRODUCTION

The U.S. Bureau of Mines and the EPA entered into a Memorandum of Understanding in June 1987, that provides for the Bureau to supply technical assistance to the EPA in the area of treatment of inorganic wastes on Superfund sites. Bureau researchers had already developed an award-winning processto reclaim lead fromscrap batteries electrolytically [l-31. On the basis of this experience, interagency agreements were signed to conduct the treatability studies of waste at five battery-recycling Superfund sites. In recent years, the numberof lead-acid battery recycling plants in the United States has been reduced from over 1 0 0 to about 20. When some of the plants closed, they left behind hundreds of thousands of tons of lead-contaminated wastes and soil. To date, 23 such sites, ranging in size from 26 acres to about 4 acres, have been designated as Superfund sites by the EPA [4]. Environmental engineers have investigated new techniques to excavate, stabilize, vitrify, incinerate, biodegrade, and encapsulate materials considered to be hazardous. Most of these treatment methods will leave heavy metals in the waste body, which will require subsequent monitoring, and do not result in a permanent solutionto the problem. None of these methods can remove lead or lead compounds from the lead battery casing wastes, the adjunct sulfateoxide sludge, or the contaminated soils that surround the casing wastes. This chapter discusses the studyon a site in Ohio, but material from other lead-acid battery sites has been treated by the Bureau-developed process. Although the battery casing wastes

The Comprehensive Environmental Response Compensation and Liability Act or Superfund, as it is now commonly to help pay for cleanup of hazardous known, was passed in1980. This trust fund, administered by the EPA, was created waste sites that threatened the public health or environment.

311

312

Wethingtun et al.

from various sites are very similar,the soil composition and level of contamination vary widely from site to site. The 26-acre Ohio site is located in a rural area with farmland to the north and south. The north boundary of the site is bordered by a gravel road. The south boundary is a tributary of the Miami river witha small industrial plantto the south of it. A railroad is adjacent to the east boundary with woods and undeveloped land beyond. To the west, the site is bounded by four residential or business properties and a county road. There is one residence across the road, directly west of the site. The site drains to the southeast into the tributary, through a swampy area overgrown withtrees and understory materials. Theoffice, loading dock, and a couple of sheds remain on the site from the original operation. The estimated amount of material to be cleaned is55,000 yd3 of waste batter casings and the waste pile. approximately 85,000-100,000 yd3 of lead-contaminated soil under and around These battery wastes consist of ebonite battery cases, lead sulfate-oxidesludge, metallic lead, and the associated soils and debris.Lead contamination of the cases ranged from 900 to 3000 ppm Pb, the lead content of the adherent sludge rangedfrom 20 to 36%, and the lead content of the soils ranged from 0.05 to 2%. Initially, EPA's criteria for success were that the battery waste material and soils had to pass the EP toxicitytest and contain less thanlo00 ppm Pb after cleanup. Later, the criteria for the battery casings (since they will be shipped off-site) were changedso that the casings had to pass the TCLP test instead of the EP toxicity test. The EP toxicity was test retained forthe soils, and the standard for lead content in the soil and casing residue was reduced to 500 ppm. The criteria were accepted by the State of Ohio. Generally, state laws cannot be less strict than federal laws or conflict with the intent and purpose of the federal law. After initial treatability studies, a process was developedto decontaminate the ebonite battery casings and lead-contaminated soils. The process consists of physical separation of the different components, carbonation to convert PbSO, into PbC03, and acid leaching to solubilize the PbC03. After a solid-liquid separation and rinsing, the lead levels in the casings and soils meet the EPA requirements. For the carbonation, two reagents [(NH,),CO, and Na2C03] were investigated. For the acid leach, two effectiveacids (HNO, and HzSiF6) were investigated and compared.An electrowinning method was used to strip lead fromthe H,SiF, leachate prior to acid recycling. The cleaning battery chips, with a heating value of 13.78 X lo6 Jikg, may be suitable as a fuel, and the decontaminated soils canbe replaced on the site. The metallic lead and sludge removed from the battery casings pile may be recycled by a secondary smelter.

II.CHARACTERIZATIONSTUDIES A. ScreenAnalysis Casing and soil materials shipped from the Ohio site were split and sampled. Throughout the treatment process,the casings and soilwere handled separately. The lead contentof the sludge clinging to the battery casings, the different handling characteristics of the soil, and the subsequent changein the criteria for the casings were all factorsin that decision. Screen analyses were conducted onthe as-received materials to determine the relative weight of each constituent and the lead concentration in each fraction. The plus 18 mesh fractions from battery casing wastes were mixtures of casing chips, rocks or gravel, metallic lead grids, and twigs. A screen analysis on battery casing material is listed in Table 1. (In a screen analyses, standard nomenclature is to refer to material that remains on a given screening surface as the plus size and material that passes through the screening surface as the minus size.)

Lead Decontamination of Superfund Sites

313

Table 1 ScreenAnalysis of As-ReceivedBatteryCasings Size

wt %

Plus 318 in. Minus 318 in.,17.1 mesh plus 8 Minus 8, plus 18 mesh Minus 18 mesh

78.0 0.3 4.6

Table 2 BatteryCasingsSample-LeadContentandDistribution Casings Rocks Sludge

(%)

metal Pb

~

Sample weight distribution Pb weight distribution Pb analyses

60.55 0.53 0.13

4.95 0.08 0.25

30.7 74.1 35.8

3.75 25.3

96"

'kur to six percent antimony is associated with the lead metal.

Table 3 Screen Analysis of As-Received Soil Sample Size

Wt %

Plus 318 in. 318 Minus mesh in.,8 plus Minus 8, plus 18 mesh Minus 18 mesh

9.4 16.9 2.0 71.7

Table 4 SoilSampleDistribution (%) Size

Other

Plus 318 in. Minus 318 in., plus 18 mesh Minus18mesh

WtRocks %

Casings

9.4 94.3 18.9 71.7

2.7 1

-

3"

99

-

-

-

b

'Wood, tramp iron, and metallic lead. No lead metal was discerned in any other mesh

size. Samples from other sites contained significant amounts of metallic lead in the other mesh sizes, depending upon the individual battery recycling operation. bo.71-0.85% Pb compounds; 15-20% moisture.

A typical lead analysis and the distribution of the different fractions in the casing wastes are listed in Table 2. The total lead distribution in the sludge and metal fractions was 99.4%. The screen analyses of a composite of as-received soil waste sample is listed in Table 3, and the distribution of this composite sample is shown in Table 4.

B. MicroscopicExamination One set of casing samples was studied using an electron probe X-ray microanalyzer[7]. These studies were performed to (1) determine what mineral phases, particularly the lead phases, were present in the soils and casings; (2) quantify the various lead phases; and (3) determine complete chemical composition. The casing sample was washed free of sludge, ground, cast in epoxy, and polished before being examined with the electron probe. These studies showed the presence of PbSO,, PbO,, Pbo, and metallicPb in all fractions. The characterization studies of the sludge removed from the battery casings indicated almost

314

Wethington et al.

Table 5 Analysis of theWashedChips (13/8-in. Chips) Element

PPm

W As Ba

3000 11.1 44 1.1 22.2

Cd Cr

55

Sb Se

0.001

60% of the lead compounds were PbSO, and the remainder was a mixture of oxides and metallic Pb. The examinationof the casing fragments underthe electron probe revealed numerous small fractures throughout that were filled with PbSO, and, occasionally, metallic Pb. This was a significant finding sinceany successful lead treatment would be dependent on removal of the PbSO, from these cracks.

C.ChemicalAnalyses A partial chemical analysis of the washed 3/8-in. battery cases is listed in Table 5 . A partial chemical analysis of the sludge washed from the casing material is shown in Table 6 . The partial chemical analysis of as-received soil samples is given in Section V.B.

111.

PRELIMINARYTREATABILITYSTUDIES

Initially, studies were done on the battery casings and soils to determine if physical methods could be used to concentrate the contaminants and reducethe volume of material that had to be treated. Since lead compounds have a relatively high density, gravity separation was a logical choice. The finenessof the soils ( '/4 in. Size > % in. Size > % in. Treated size Waste m t e d (g)

'Zteatment reagents

Extraction information

Silicate 1 (g) Silicate 2 (g) Cement (g) Cement type Lime (g) Lime type Fly ash (g) Fly ash type Kiln dust (g) Kiln dust type Polysulfide (g) Polysulfide type Phosphate (g)

Ext. date (totals) Ext. date (TCLP) Ext. date (STLC)

Pre- and post-treatment concns.

Ag SIZC Ag TCLP Ag TTLC A1 STUJ

Al TCLP

E Ni STLC Ni TCLP Ni TTLC PbsTLc Pb TCLP P b m

AlTTLC AssLC As TCLP A s m

Sb SIZC Sb TCLP Sb TTLC

Ba STLC Ba TCLP Ba TTLC Be STLC

Se STLC Se TCLP Se TTLC T1 STLC

Y

5%

Phosphate type Acid (g) Acid type Base (g) Base type Oxidant (g) Oxidant type Reductant (g) Reductant type Other 1 (g) Other 1 type Other 2 (g) Other 2 type

Be TCLP Be TIu3 Cd sIu3 Cd TCLP Cd TTLC c o STLC Co TCLP c o TTLC

cam

= =

Cr(II1) TCLP Cr(1II) TTLC cr(w Cr(VI) TCLP c r ( w TTLC

cu STu=

Cu TCLP c u TIu3 FTTZC

T1 TCLP TI TTLC Vslzc V TCLP V TTLC zn STLC Zn TCLP ZnTTLC ca STLC Ca TCLP

B

5 3

%

2 Q

4 % r8

E.

c1 STLC

2

C1 TCLP

CI TTLC F e r n Fe TCLP FeTTLc

F TCLP

Hg Hg !m-.c Hg TCLP Mo sTu= Mo TCLP Mo TTLC

Na STLC Na TCLP Na TTLC Final dry wt. pH solid DHextract

=

r8

cam

Mg STLC Mg TCLP Mg

FSlZC

F

3

3

Table 2 Total Metal and he-and Posttreatment Concentrations for Metal Gmum 1-5 ~

Extraction procedure

Character

Units

Arsenic

~~~

Copper

Lead

Nickel

Cadmium

Zinc

1.24 1S 2

0.01 0.01

70

0.88

0.01

Metal system 1

Totals 3050 STLC

STLC Totals 3050

TCLP #1 TCLP #2

Initial initial Treated

mg/kg m a

lnitial Initial Treated

mgkg mg/L mg/L

Totals 3050 STLC

Initial Initial

STLC

Treated

Totals 3050

snc

Initial Initial TEated

Totals 3050 TCLP #2 TCLP #2

Initial Initial Treated

STLC

16,300

2

m a

0.005 589 11.23 0.001 Metal system 2 3217 140 0.09

I359 1.11 0.12 Metal system 3

0.001

mg/L mglL mag mg/L mg/L

mglkg m gn

mpn

3917

o.OoO1

383 O.OOO1 0.88 Metal system 4 N/A NIA NIA

350 19.01 7.28 Metal system 5

2,759 158.3 0.068

1,627

34.7 0.075

2 0.27 0.02

21 0.06 0.05

770

72 0.075

537 31.99 0.62

1,550

0.33 0.28

29 0.1 0.1 30

2 0.01 0.01

40

2.89 0.01

0.28 0.005

92 2.56 1.64

4.95 0.46 0.005

470

41.25 20.91

48 1.54 0.005

0.01 0.01

398 2.2 0.003 5400

915 21 -0.5 1,012 87.04 0.84

325 15.7 0.534

Remediation of Heavy Metal Contaminated Solids

?

W

8

.p!

B

5

B

451

1 I

Fa

3.2

Trezek

452

Figure 2 Equipment used in the delivery of the STS technology.

introduced at the midpoint of the mixer. The feed rate canbe adjusted by controlling the variable-speed drive on the silo rotary vane feeder.The residence timein the mixer is controlled by the blade angles. For soil, a 22" blade angle is used in the first half of the mixer to enhance retention (i.e., increase the contact timebetween the silicates and the material). In the section after the cementitious material is added, the blade angles are set at approximately 45" to enhance mixing and removal of the treated material from the chamber. As the treated material exits the pug mill unit, a radial stacking conveyor pilesthe material. The process is complete after the material has cured in the stockpiles. The treated stockpiles typicallyare turned with a front-end loader on a daily basis for several days. A Bomag unit can also be used to cure treated material. Here, the material is arranged in 3-44? lifts. After the material has partially set, the Bomag unit traverses the lift, creating suitably sized material. The final aspect of the treatment includes the sampling and subsequent analytical evaluation of the cured material. This involves obtaining representative samples from treated stockpiles and subjectingthe material to an extraction test. If the EPA TCLP test is used, the treated material (crystalline cementitious matrix) is extracted witha sodium acetate-acetic acid buffer solution. In this process, metallic elements can be leached from the matrix to form metal acetates. As shown below, the final extraction product is an equilibrium mixture of metal polysilicates, metal acetates, acetic acid, metal hydroxides, metal oxides, etc. a CH3COO-

+ b H+ + c Na+ + d H20 + e M(OH), + fCa(OH)2

+ g metalpolysilicates + - -

4

+

h (CH3COO),M + i CH3COONa + j CH3COOH k H20 n NaOH o ca3(C~H&)~ p metal polysilicates

+

+

+

+ l M(OH), + m Ca(OH)2

Remediation Contaminated Solids Metal of Heavy

453

test, a stable matrix Thus, in order for a treatment to be successful with regard to the TCLP must be created that resists the attackof the aggressive leachingfluid, i.e., one in which only negligible quantities of soluble metallic elementsare present. Upon completion of the analytical data, the treatedfriable material can then be backfilled on the site with conventionalearthmoving equipment.

V. CASESTUDIES Two examples of applying the technology to actual site remediations are given here. These particular case studies were selected becauseof their diversityin terms of both site conditions and types of metals. The first case study involves a heavy metal-contaminated site in the Port of Los Angeles where there had been an extensive metal salvaging operation dealing with a variety of operations that included ship breaking. Asa result of these activities, the soil was contaminated with lead, zinc, cadmium, nickel,and copper. The treatmentoperations were carried out on a clay pad constructed on site. The second case study is an example of “treatment in tank” that requiredthe construction of an RCRA tank before treatmentoperations were begun. In this case, arsenic was the principal metal of concern, and its remediation required a modification of the processing system to include an additionalsilo for the delivery of a second cementitious material.

A. Case Study I This case study deals with the remediation of a 23.5-acre site located in the Terminal Island District of the Portof Los Angeles [4]. The initialsite characterizationanalytical data indicated that approximately 18-24 in. of top soil material, or about 60,OOO tons of soil, would require treatment. In actuality, 106,700 tons of soil was treated in the overall project. A clean areawas prepared onsite for the mobile equipment treatment operations. The contaminated layer was removed in a 300 ft by 300 ft area, stockpiled in an adjacent location on the site, and replaced or backfilled with clean decomposed granite soil. This area provided a working pad for the equipment and the curing of treated material. The material requiring treatment was not typical soil. Because of the prior metal salvaging activities, the material containeda variety of ferrous and nonferrousmetals, rocks and stones, pieces of wood and asphalt, and other miscellaneous items. The size distribution of these materials spanned several orders of magnitude ranging from less than l in. to several feet. Occasionally, various parts of ships (i.e., riveted and welded beams, parts of anchors, sections of mechanical equipment, etc.) were uncovered in the excavation of the site and found their way into the stockpile for treatment. Consequently, the heterogeneous natureof the material dictated the type of preprocessing unit operations prior to mitigating the heavy metals. Although the project was permitted to operate from 6:00 A.M. to 6:00 P.M. five days a week, the South Coast Air Quality District imposed the added restriction that all operations including operating rolling stock cease by 5 0 0 P.M. Consequently, the effective daily treatment window was approximately 10 hr or less, depending upon downtime. A period of at least 1 hr was required for cleanup, maintenance, moving piles, etc. at the end of each shift. Thus, in order to meet the project schedule,a nominal 1OOO tons of material per day had to be processed within these time constraints. Material requiring treatmentwas arranged in 1000-ton stockpiles, 30 by 150 by 8 ft high, on the site adjacent to the clean soil equipment zone. Samplesof this material for laboratory analysis of heavy metals were taken as the piles were generated. Thesedata were used to s u p plement the original site characterization data and provide guidance in establishing the daily

454

Trezek

treatment protocols.The sampling protocol also involved the collection of samples of untreated and treated material at 15-min intervals during operation. These samples formeda daily composite, which was split for independent certified laboratory analysis. After receiptof the laboratory report and acceptance by the Port inspectors, the material was backfilled on the site. Additional samples of the in-place material were also taken. 1. TreatmentLevels Testing of the contaminated soil for all 17 metals revealed that only five had elevated levels requiring treatment. The range of these metals in terms of both the soluble (STLC) and total (TTLC) concentrations is summarized as follows: (1) lead, STLC 11-121 mg/L; TTLC 271500 mg/kg; (2) zinc, STLC 14-320 mg/L; TTLC 242-3130 mgkg; (3) cadmium, STLC 0.11.9 mg/L; TTLC 2-12 mgkg; (4) nickel, STLC 0.2-7 mg/L; TTLC 30-600 m a g ; and (5) copper, STLC2-96 mgL; TTLC 70-2610 mgkg. The soluble concentrationswere determined by the CAM wet extraction method, which involves milling to pass a No. 10 standard sieve followed by 48 hr of extraction in a sodium citrate solution. The relationship between the total and soluble concentrations summarized inFigure 3 shows the respective ranges for each metal. In effect, the treatment process must deal with metals whose concentrations cover a range of four orders of magnitude. It should be noted that the values of lead shown in Figure 3 are plotted as Pb/100 to aid in pattern recognition. Thus, the concentrationsof lead are in the same general band as copper and the lower range of zinc. 2. TreatmentResults The actual treatmentactivities began on October9, 1989 and terminated on April 12, 1990. The treatment of the initial 60,OOO tons of material was completed by the contracted scheduledate of January 15, 1990. The project period was then extended to treat the additional 47,000 tons of soil. Approximately 16,000 tons of nonhazardous oversize material was removed in the screening operation. With the exception of the ferrous metals, this fractionwas disposed of in a Class 111 landfill. The quantities of polysilicates and cementitious material were adjusted to coincide with the concentrations of metals in the in-feed material. Because of a combination of logistical, economic, andtreatability considerations, cement wasused as thecementitiousmaterial. Throughout the course of the treatment, the additionof cement ranged from 10.14 to 11.08%. Even with the wide range of STLCconcentrations of the various metals,the use of polysilicates varied over a relatively narrow band, which ranged from 0.513 to 0.59 gaton of soil.

B. Case Study II This study involves the remediation of 40,OOO tons of heavy metals-contaminated soil at the Thompson-Isaacson site in 'hkwila, Washington. In this soil, arsenic was the principal metal of concern. Theother metals thatwere evaluated were barium, chromium, copper, nickel,lead, and zinc. The need for the remediation arose asa result of the planned developmentof the site, which entailed the construction of an industrial structure necessitating substantial excavations for foundations and pedestrian tunnels[5]. Elevated levelsof arsenic are believed to have occurredas a result of the dredging,filling, and straighteningof the Duwamish channel beginning aroundthe turn of the century. Presently, one of the site boundaries is the Duwamish Waterway, which once flowed through the approximate center of the site before the channel was straightened and redirected. Various site investigations [6,7] indicated thatthe arsenic contaminationmay have resulted froma variety of fill materials generated by smelting and other ore processing operations.

Remediation of Heavy Metal Contaminated Solids

0

0

-

0

0 0

9 0

0 0

0

455

456

Trezek

A number of factors entered intothe decision to use on-site treatment forthe remediation of the heavy metals. Basically, thisalternative allowed the site owner to maintain control in that the treated material would remain on-site, thereby reducing the reliance on off-site hazardous waste landfill disposal and its associated long-term liabilities. Further, the useof on-site treatment at high processing rates (100 tons/hr) eliminates the excessive time associated with trucking large quantities of material considerable distances to Class I disposal sites. Finally, in terms of construction, since the characteristics of the backfilled STS-treated materialare suitable for the placement of buildings, the need and costs associated with imported fill materials can be significantly reduced or eliminated. The conditions for theuse of on-site treatment were consistent with the treatment in tank by generator regulations. Consequently,an RCRA tank was designed and constructed withthe appropriatemembraneliner,asphaltconcretepavement,catchbasins, etc. Basically,this RCRA tank was designed to not only contain the soil during treatment, but also to collect all contact and noncontact liquids,which could be subsequently transferred toappropriate storage tanks. A secondary containment system was also installed for the purpose of monitoring any leakage ftom this system. Normally, full-scale or commercial processing would follow the treatability study. However, in thisproject, the Washington State Department of Ecology, as part of the approval of the Site Remediation Action Plan[g], requested a 4000-5000-ton pilot field test prior to full-scale remediation for the purposeof verifying the treatability data and the effectiveness of the treatment under actualfield conditions. It was necessary to first construct the previously mentioned RCRA tank before the test could begin. The mobile treatment system was then erected in the completed tank. This system consistedof the feed hopper, mixing unit, chemical delivery system, and associated feed and discharge conveyors. In this project, the curing and subsequent stockpiling of the treated material occurred in the tank. Approximately 4500.tons of soil was treated at commercial processing rates of 80-100 tonskr during the 5-day pilottest; Material requiring treatmentwas excavated to the water table (about 12 ft below the surface) using a standard backhoe machine. Prior to entering the treatment unit, the excavated soil was screened to a 314 in. particle size through a two-step process using grizzly bars and a trommel screen. Stockpilesof the screened material were sampled and evaluatedfor heavy metals concentrations before treatment. Composite samples of the material were also evaluated before and after treatment. Control of the treatment protocol was accomplished through a mass balance, which entails continuous measurement of the rate at which material and reagentsenter the treatment unit. The necessary instrumentation included a certified belt scale on the material in-feed conveyor, calibrated and computer-controlled rotary feeders for dry reagents, and in-line flow meters and calibrated delivery pumps for liquids. 36,000 tons of material After the successful completion of the pilot test program, the remaining was processed and subsequently backfilled into the excavation. 1. TreatmentLevels In this particular soil, as in Case I, a comparison of the total (TTLC) and soluble concentrations (TCLP) for arsenic, copper, and zinc (Figure 4) illustrates that the concentrationsof heavy metals can vary over several orders of magnitude throughout the site. Arsenic is the principal metal of concern; copper and zinc exhibitedthe next highest concentrations but did not exceed TCLP limits. The data shown in Figure 4 were obtained during the pilot test program. During the production phase, the evaluation of total concentration levelswas not included in the routine data collection. An analysis of the data in Figure 4 indicates that the degreeof solubility is a function of total concentration. In the case of arsenic, the soluble concentrations were about five times

Remediation Contaminated SolidsMetal of Heavy

457

Figure 4 Relationship between soluble and total concentration levels in untreated soil (Case Study 11).

greater at the highest TTLC levels. In other words, the TCLP levels range between 200 and lo00 times less than the TTLC levels as they decrease from lO,o00 to 100. Copper exhibits similar behavioreven though the ranges in TTLC and TCPL levels only span an order of magnitude. On the other hand, the levels of zinc are clustered such thatthe TCLP levelis about 100 times less than the TTLC concentration. Itis interesting to comparethis soil with that of Case I, which also exhibited concentration levelsof the aggregate or mixture of metals that spanned several orders of magnitude (Figure 3). However, unlike the arsenic in this soil, the metals in the Case I soil were grouped or clustered in relatively narrow ranges withinthe overall distribution. Also, another notable exceptionwas zinc, which was not clustered. In these particular arrangements (Figure 3), the average soluble concentrations (determined by the STLC procedure) were about 20 times less than the total concentration. Extraction procedures will affect the behavior of soluble concentrations in a mixture of metals. In terms of the above comparison, this would involve differences inthe leaching characteristics between the previously discussed TCLP procedure and the STLC extraction, which uses a sodium citrate buffer with a tenfold dilution factor. 2.TreatmentResults The treatment protocol used two cementitious materials, cement and lime, in the ratios of 20:3 or 15:5, with polysilicate additionsof approximately 0.5 gaton. An example of the ability of the process to reduce the soluble concentration is given for arsenic treatment (Figure 5) during the production phase.It is important to notethat in this case the treatment standard for arsenic was set at 1 ppmby the site owner instead of the regulated level of 5 ppm. The analytical evaluation protocol was established at 0.2ppm, so values belowthis level were not determined.

458

Trezek

Number of Samples

figure 5 Comparison of untreated and treated TCLP levels of arsenic (Case Study 11).

In this project, a daily production goalof 750 tons of treated material was established as being reasonable for the site conditions, which were constrained by the ability to deliver material to the process (i.e., excavation, screening, sampling,etc.), the cycle time for curing and stockpiling in the tank, and the subsequent backfilling. The ability of the system to meet the processing goals is illustrated in Figure 6 in terms of daily tonnage and daily average tonnage over the production phase of the project. At times it was not possible to meet the daily production schedules, the reasons being (1) a mechanical failure in the lime delivery system on days 3 and 4, (2) the unavailability of feed material on day 3 1, and (3) the reduced quantityof feed material at the end of the major portion of the project during days 37 and 38. An additional feature of this project concerned the contaminated water collected on-site in the process. The RCRA tank allowed liquids to be collected from the processing areas that included both the screening and treatment activities. Approximately 12,000 gal of water per day was generated froma combination of spraying for dust controland rain. An on-site systemwas implemented that allowed this water to be incorporated into the treatment process. The water from the collection system was introduced into the first of a series of four Baker tanks. This provided suitable capacity and time for the settling of suspended solids. The liquid from the fourth tank was then returned to the polysilicate-water blending tank in the chemical mixing portion of the treatment system. The concentration of arsenic in the liquid entering this tank, as determined by EPA methods SW-846,3010, and 6010 was on theorder of 4 mg/L. Blending this liquid with the process water at a ratio of 1:20 did not impact the treatment protocol. In addition, the project benefited from a significant savings in off-site disposal costs.

VI. ANCILLARY TREATMENT PERFORMANCE PARAMETERS In addition to the task of reducing metal solubility,other important featuresof this technology are considered. These include (1) the nature of hydration and curing; (2) the compactability of the treated material as related to on-site backfilling; (3) the effect of pre- and posttreatment particle sizes, which influence metal concentrations in dispersed material; and (4) the long-termviabilityof the treatment as measuredthroughmultipleextractionprocedures.

459

Remediation of Heavy Metal Contaminated Solids

10

15

20

Days Running

25

30

35

Figure 6 h e s s i n g system throughput performance (Case Study II).

A. HydrationandCuring As previously mentioned, curing the treated material is the final step in the process. In terms of the field application of the technology, this basically amounts to liberating the moisture in the treated material, i.e., allowing it to dry. However, this process has other fundamental and serious impacts on the efficacy of the technology in terms of(1) the proper hydration of cementitious materials, which is relatedto the transformation of ionic bonds to covalent bonds, and (2) the use of sampling procedures that ensure that the curing process has been completed before the material is subjected to TCLP or other extraction protocols. The following discussion considers the interrelationship between hydration, curing, and sampling, which were extensively evaluated for the soil treated in Case Study 11. The treatment protocol developed for the remediation of the arsenic-contaminated soilof Case Study I1 used cementnime ratios of either 20:3 or 15:5 on the basis of weight. The effect of these mixtures on the hydration characteristics of the treated material is illustrated in Figure 7 for a bench-scale test using an initial 500-g sample of soil having a moisture content 12%. of After the addition of reagents and water in the amounts of 85 and 100 mL, respectively, the mixture weight was about 700 g for both cement-lime combinations. Here water essentially behaves as a reaction catalyst. Based on prevailing accepted stoichiometric data, the amountof water bound to cement and lime is on the order of 13% and 30% of their weight, respectively. This would indicate that approximately 17-18 g of water would be retained in the matrix. In

460

Trezek

-

Time hours

Figure 7 Hydration characteristics of cement-lime mixtures.

other words, the dilution factor due to water would be approximately 3.5%. As expected, the mixture with the higher percentage of lime begins to initially lose its free water at a faster rate (Figure 7) due to the increased heat of hydration. After thisinitial period, a similar linear rate of water loss occurs in both mixtures over a 72-h period. During this time, the samples at 8-hrintervals were subjected to a 70°F, 65% relative humidity environment and were broken to expose new surface. Between 72 and 75 hr into the test period, heat lamps were used to accelerate the water loss. The samples were allowed to equilibrate to room conditions for 1 hr before final weighing. The data obtained fromthe above exerciseillustrate that (1) the majority of the free water used as a catalyst will be liberated and(2) the retentionof water in the matrix appearsto be less than that derived from conventional stoichiometry. However, more important in terms of operations, using artificial means such as heat lamps to accelerate the loss of free water does not alter the fundamentals of the treatment. In certain cases, a procedure of this nature, referred to as a lab cure, is used to expedite the turnaround time of obtaining analytical laboratory TCLP data. Operationally, this can allow naturally cured stockpiles to be backfilled at the time the analytical data are available. A comparison of the lab and field cure TCLP arsenic data (Figure 8 ) illustrates that the two methods yield the same results with proper field curing. It shouldbe noted that in Figure 8 , arsenic data are given for both the pilot test and the production phase where the detection limits were 0.01 and 0.2 mg/L, respectively.The acceptable TCLP levelof arsenic in the treated material was set at 1.0 mg/L. Basically, the elevated levels in the field data shown in Figure8 indicate that the material was prematurely sampled.Sufficient free water was still present; that is, the process shown in Figure 7 was not completed. The tendency of the treated material to form monoliths is another ramification of improper field curing. Stockpiled material must be turned, usually on a daily basis, to (1) enhance the rate of curing through accelerated heat and mass transfer to the environment and(2) mechanically overcome the tendency of cementitious mixtures to fuse or solidify into a continuous substance, which would negate the desirable feature of a friable matrix structure.

Remediation of Heavy Metal Contaminated Solids

461

Number of Samples

Figure 8 Comparison of laboratory and field curing on TCLP arsenic concentrations.

B. Compactability An important feature of the process is the ability to backfill or recompact the treated material into the excavation. Standard tests of treated material using ASTM Method D-l557 yield a to achieve this compaction of about 97%. The following procedure is typically used in the field compaction. Stockpiled cured material is loaded into a dump truck and moved to the excavation. Here the dumped pile is spread with a bulldozer into an 8-12-in. lift. The moisture content is adjusted according to the ASTM test data to be consistent with optimum compaction. A drum vibrating compactor is then usedto complete the placementof the lift. The volume expansion due to treatment is estimated to be on the order of 20%. Operationally, after the removal of the oversize fraction, the backfilled treated material occupies the volumeof the excavation. Several field observations indicate that the features of the backfilled treated material are suitable for construction. For example, backhoe digging tests on material in place for 28 days showed that it remained friable with sufficient structural stability for trenching; that is, sharp vertical cuts couldbe made. The equipment operators indicated that the backfilled material was similar to digging hardpan.

C. Effect of Particle Size An additional featureof the STS polysilicate technology is its ability to indirectly mitigate total metal concentrations through increased posttreatment particle The size.basic mechanism stems from the fact that the emissionrate for total respiration particulate matter(E,,) has significant E,, is illustrated through the work of Cowherd et al. [9] particle size sensitivity. This effect on and summarized in the California Site Mitigation Decision Tree F’rocess [lo]. The basic calculational methodology follows. 1.ParticleSizeDistribution The f m t step in the procedure involves the determination of the characteristic particle size. This is typically accomplishedby subjecting representative samples of pre- and posttreated ma-

462

c

Trezek

Size mm Figure 9 Size distribution of pre- andposttreated soil.

terial to a sieve analysis andarranging the results in terms of percent passing vs. size (Figure 9). It is interesting to note that the characteristics of the treated material (straight line on a semilog plot) are representative of a Rosin Rammler distribution, which will have a characteristic particle size (X,) corresponding to 63.2% passing. Consequently, for illustrative purposes, the value of X, for the particular treated material shown in Figure 9 would be given by a size of about 6.4 mm. The comparable size for the untreated material can thenbe taken as 0.8 mm. Thus, as illustrated in Figure 9, the agglomerative featuresof creating the metasilicate and cementitious matrix during the treatment process hasthe effect of shifting the size distribution of the treated materialto larger sizes. The increase inthe characteristic size is usually about an order of magnitude. Here this increase is a factor of 8. 2.ThresholdFriction Velocity As illustrated in the work of Cowherd [9], the threshold friction velocity (Uf) is logarithmically related to the so-called aggregate size distribution mode. Basically, this parameter is intended to be representative of the wind speed at ground level corresponding to the occurrence of soil erosion. In termsof the characteristic particle size, the valueof Uffor the 0.8-mm and 6.4-mm particles are 60 c d s e c and 140 cdsec, respectively. Thus, an eightfold increase in particle size slightly more than doubles the threshold friction velocity. 3. Roughness Height Roughness height is used to convert the value of U, at ground level to the wind speed at a typical 7-m-high weatherstation. Values of roughness height(Z,) range from0.1 cm for natural snow to IO00 cm for high-rise buildings. Similarly,plowed fields and grasslands are given by Z, values of 1.O cm and 2.0-4.0 cm, respectively. A Z, value of 1 cm is typically used for soil being remediated.

Remediation Contaminated Solids Metal of Heavy

463

4.Threshold WindVelocity The threshold wind velocity (Ur),i.e., the wind speed necessary to initiate soil erosion as determined by 7-m-high weather station data, is given in terms of the value of Ufaccording to a relation developed by Cowherd,

Ut = U, (13.1 - 2.5 In G) The corresponding values of U,for the 0.8- and 6.4-mm particles are 7.86 and 18.34 m/ sec, respectively.

5. Respirable Particulate EmissionRate The following relation developed by Cowherd was used to evaluatethe emission rate of respirable particle matter from erodible surfaces: El0

= 0.036 (1 - V ) ( I Y / U , )F(x) ~

where E,, is the emission rate for total respirable particular matter [g/(m 2.hr)]; V , the fraction of exposed contaminated area thatis vegetated (V = 0 for bare soil); and U,the mean annual wind speed (m/sec). The parameter F(x) is given by the relation

F(x) = 0.18(82

+ 12x)/8*

where x = 0.886

(UJU)

Assuming a nominal 10 mph wind speed (4.47 m/sec), the value of E,, for the 0.8-mm particle size is 5.15 X low3g/(m 2*hr).The corresponding value for the 6.4-mm particle is 7.3 X lo-* g/(m 2+r). Thus, an eightfold increasein particle size reduces the value of E,, by in threshold friction velocity due to increased 7 X lo4. Alternatively, relatively small increases particle size result in large decreases in respirable emission rates. The E,, behavior with wind speedfor 0.8-mm and 6.4-mm particles is illustrated in Figure 10. Each particle exhibits a characteristic steep rise in concentration at elevated wind speed. Basically, the larger particles can sustain higher wind speeds for a given concentration. For g/(m2-hr)at 10 mph. A similar example, the 0.8-mm particle has an E , , value of 5.15 X value of E,, for the 6.4-mm particle would be reached at a 23 mph wind speed. 6. DownwindConcentrations The effect of particle size on concentrations of specific constituents reachinga downwind receptor can be obtained from therelation

X = Q/n ayazU where X is the concentration of the constituent in ambient air (pg/m); Q is the emission rate (pg/sec); ayand a, are the standard deviationof horizontal and vertical dispersion, respectively (m);and U is wind speed (m/sec). The following assumptions are made to illustrate the effect of particle size on downwind concentrations for a working area of IO00 m' and a length of 100 m to the receptor. Using the data of lbrner in the Workbook of Atmospheric Dispersion Estimates[1l] and ClassC stability, the values of ayand a, are 13 and 7 m, respectively, fora 100-m distance. Thus, for a 10 mph ng/m3for the 0.8 and 6.4wind speed, the valuesof X become 1.12 X IO3 and 1S 8 X mm particles, respectively. Thus, an estimate of the downwind concentration of any specific

46.4

Trezek

A. 3

0

U C

1.OOE-(

_

c

6.4 mm PARTICLE

X I- o.amm PARTICLE _

i 38

1.00E-07

E10 (gms/sqm.hr)

, l .OOE-

1H 1 .OOE-0'

1.00E+00

Figure 10 Influence of wind speed on concentration. constituent can be calculated by multiplying the downwind concentration X by the mass fraction of that constituent. The results of this calculation are summarized in Table3 for the total 2 for metal system5 . Basically, these results follow the E,, metal concentrations given in Table effect such that the downwind concentrations become negligible for the larger particle size.

VII. MULTIPLEEXTRACTIONPROCEDURE The standard extraction tests (TCLPor California STLC) provide a one-time equilibrium solubility valueof the heavy metal compounds in the extraction fluid do butnot give an indication of the time-dependent stability of the residual matrix.The EPA multiple extraction procedure, designed for evaluating the leachabilityof materials exposedto so-called acid rain conditions, was used to test the stability of the polysilicate cementitious treatment protocol.

Table 3 Effect of Particle Size on Constituent Downwind Total Metal Concentrations for Metal System 5" Compound Arsenic Copper Lead Nickel Cadmium 0.364 Zinc ~

Concn, Mass fraction X

0.174 0.076

2.759 2.57 1.627 12.45 S5 0.47 0.048 0.325

~~

'Mean wind speed = IO mph.

Concn, 0.8 mm particle (ng/m3)

6.4 mm particle(ng/m3 X 10"~)

3.09 1 .S2 1.74 0.526 0.054

4.36

465

Remediation Contaminated Solids Metal of Heavy

STLC

Ar4 Arl Ar3 Ar2

Ar5 Ar7 Ar6

STLC - Acid Rain Extractions

Ar8

Ar9

Arlo

Figure 11 Effect of multiple extractions on soluble metal concentrations.

The evaluation consistedof an initial STLC extraction (as described in CCR title 26, Ch22a solution of 66261 Appendix 2) followed by 10 consecutive acid rain extractions (using 60/40 sulfurichitric acid with apH of 3 as described inSW846 Volume C, Ch. 6,Method 1320).The predominant metals in the sample being treated were copper, lead, and zinc with initial STLC concentrations of 22, 110, and 106 ppm, respectively. Thus, in this sample the summed value be 238 ppm. Following treatment, the summed for the STLC soluble metal concentration would 11.7 ppm. The results of the subsequent acid rain extractions for STLC concentration value was both the treated and untreated materials are shown in Figure 11. Basically, the stability of the treated material is illustratedby the fact that the equilibrium leachable value was reached following the second acid rain extraction and remained unchanged. On the other hand, the metalliccompoundsintheuntreatedmaterialcontinuedtoionize andleachfollowingeach additional extraction step.

VIII. CONCLUDING REMARKS The STS polysilicate technology provides an effective and relatively low cost method of remediating soils contaminated with heavy metals. In this system, the metals become partof a covalent bonded matrix created through the action of the silicates and cementitious materials. of severalorders of magnitude are Reductioninheavymetalsolubleconcentrationlevels achievable. In terms of site remediation, the processing equipment can be configured in the form of a mobile system capable of operatingat high throughput rates. When properly cured, 97% compaction. An orderthetreatedmaterial is friableandcan be backfilledwitha of-magnitude increase in the mean particle size of mated the material significantly reduces the

466

Trezek

downwind total metal concentrationsof potentially airborne material. Finally, multiple extraction procedures validate the long-term survival of treated material in an acid rain environment.

ACKNOWLEDGMENT I wish to express my appreciation to Ms. Nancy Wright for her patience and competent assis-

tance in preparing the manuscript.

1.

2. 3. 4. 5.

6. 7. 8. 9. 10.

11.

PhysicallChernical Processes,In'Ifezek, G . J., Polysilicate heavy metals mitigation technology, in novative Hazardous Waste Treatment Technol. Ser., Vol. 2, Technomic, Lancaster, Penna., 1987, pp.155-163. Trezek, G . J., Polysilicate treatment of heavy metals in soil, Fifth Annual Hazardous Materials Management Conf./West, November 1989. Trezek, G . J., Raphael, G . , Wilber, J. S., and VanPelt, R. E., Remediation of arsenic contaminated soil using polysilicates, HMCRI R&D '92 Conf., San Francisco, February 1992. Trezek, G . J., Heavy metal contaminated soil remediation at high throughput, Proc. Superfund '90, Washington, D.C., November 1990, pp 673-676. Landau Associates, Inc., Edmonds, Wash., Thompson-Isaacson Site Soil Stabilization Pilot Test Summary Report, Prepared for The Boeing Company, July 1991. Wicks, F! H., Bothell. Wash., Proposed Remedial Action at Isaacson Corporation Property, Seattle, Washington, Prepared for Isaacson Corporation, 1984. PropDames & Moore, Seattle, Wash., Report of Evaluation of Site Contamination, Isaacson Steel erty, Prepared for the Boeing Company, 1983. LandauAssociates,Inc.,Edmonds,Wash.,andParametrix, Inc.,Bellevue,Wash.,ThompsonIsaacson Site Soil Remedial Action Plan, Prepared for The Boeing Company, 1990. Cowherd, C. M., Muleski, G . E., Englehart, F! J., and Gillette. D. A. Rapid Assessment of Exposure to Particulate Emissions from Surface Contamination Sites, Midwest Research Institute, Kansas City, MO., 1984. California Department of Health Services, Toxic Substances Control Division,The California Site Mitigation Decision Tree Manual, Sacramento, Calif., 1986. 'lbmer. D. B., Workbook ofArtnospheric Dispersion Estimates, HEW, Washington, D.C., 1970.

21

Fluidized Bed Combustion for Waste Minimization: Emissions and Ash Related Issues

E. J. Anthony and F. Preto CANMET Ottawa, Canada

1. INTRODUCTION Fluidized bed combustion (FBC) is a versatile technology that can be used in burning a wide variety of fuels. Fuels may range from conventional premium fuels (coal, oil, and natural gas) to municipal solid wastes (MSW), refusederived fuels (RDF), coal rejects (up to 70% ash or more than 50% moisture), biomass and biomass wastes, sludges, pitches, and tars [l-51. Another advantage of FBC is that, due to lower combustion temperatures, NO, levels are considerably lower than for conventional combustion systems. Furthermore, SO2emissions can be reduced by using limestone as the bed material to capture sulfur (as CaSO,)[6-81. Paradoxically,despitethelowercombustiontemperatures(typically800-950°Ccompared to 1200-1400°C for conventional combustion), FBC also produces very low levels of polyaromatic hydrocarbons (PAH) and other volatile organic compounds(VOC), due to a number of factors. First, the residence times of gases and solids in the combustion zone are relatively long compared to conventional combustion; second, FBC systems provide excellent, bed turbulentmixing;finally, it appearsthat at FBCcombustiontemperatures,limestone particles act as a catalyst to destroy organics [9-131. In a study on both FBC and pulverized fuel boilers burning coal, it was concluded that “in general, the levels for the limited polychlorinated dibenzofurans and dioxins that have been detected in fly ashes from these power stations are several orders of magnitude lower than levels reported in municipal incinerator ashes” [Ill. The use of sorbents as bed material results in the generation of ashes with unique properties. The management, i.e., disposal or utilization, of these ashes must be taken into account as part of the overall waste minimization scheme. For high sulfur fuels as such coal, petroleum coke, and some pitches and tars, the ash produced has high calcium oxide content and be can associated with exothermic reactions, dimensional instabilityin waste disposal sites, and high pH leachate (typically pH 11-12). However, the alkaline nature of these ashes does confer the benefit that leaching of heavy metals from such residues in minimal [14,15]. 467

468

Anthony and Preto

CANMET (the research division of Energy, Mines and Resources Canada, the federal government department responsible for energy-related activities in Canada) has been investigating fluidized bed combustion for coal, industrial wastes, and biomass since 1975. Research using bench-scale and pilot-scale reactors throughout this period has shown that fluidized bed combustion appears to be an ideal technology for efficiently burning a wide variety of fuels and wastes while also achieving verylow emissions of pollutants. The only significant difficulties have been related to feeding nonconventional fuels using pilot-scale equipment. These difficulties are resolvable in full-scale units.

II. FLUIDIZEDBEDCOMBUSTIONPRIMER

A fluidized bed consist ofbed a of particles suspended above a grid (e.g., perforated plate) by a fluid (air) moving upward through holes in the grid. The velocity of the fluid must be sufis too ficient to suspend the particlesso that they move freely in the bed region. If the velocity low, the particles are immobile in a “fixed bed”; at velocities greater than the particle’s terminal velocity, the particles are blown completely out of the bed, i.e., elutriated. In the fluidized bed regime, particles remain in the bed (hence good residence time) yet move about freely (hence good mixing). These characteristics, i.e., good mixing and long residence time, are the primary reasons for the numerous benefitsof using fluidized beds. Fluidized bed combustion (FBC) is the technique whereby a fluidized bed of particles is used as a medium for the combustion process. The bed particles themselves are not usually consumed in the combustion process but may play important roles in emissions control, for example, limestone particles are used to capture sulfur dioxide (forming calcium sulfate, i.e., anhydrite). The fuel particles typically comprise a very small fraction of the bed, e.g., less 1% for burning coal. Good particle mixing results in nearly uniform bed temperatures. This together with high thermalinertia due to the considerable mass of particles in thebed allows for virtually isothermal reaction at the optimum operating temperature. This confers upon FBC a number of advantages over conventional combustion systems. Some of FBC’s advantages ca be summarized as follows.

Fuel particles are rapidly heated to stable combustion temperatures. Almost any fuel can be handled, including nonhomogeneous fuels withlow volatility or high moisture content. are minimal; washing, drying, and pulverization are genThe requirements for fuel preparation erally unnecessary. Judicious selection of bed material can result in the in-bed capture of gaseous pollutant sp Low combustion temperatures (typically800-950°C) ensure that the formation of nitrogen oxides from air nitrogen is negligible.

1. A blower suppliesair to the A schematic diagramof the FBC process is shown in Figure windbox, which then discharges it through the distributor plate into the bed. The particle bed is traversed by heat exchanger tubes to provide for heat extraction from the bed. Above the particle bed a “freeboard” region is providedto allow particles ejected fromthe bed to settle back and to allow gas-phase reactionsto proceed to greater completion.The cyclone separator removes all but the finest particles carried by the exhaust gases. The cyclone product can be either recycled to the bed for further reaction or removed from the system. The gases leaving the cyclone are passed through a fabric filter “baghouse” to remove fly ash. The type of FBC unit shown in Figure1 is generally referred toas a bubbling bed because bed through whichair in excessof that required for fluidization passesas there exists a distinct “bubbles.” Superfkial fluidizing velocities are typically 1-4 &sec. If the fluidizing velocity

. 469

Fluidized Bed Combustion

L

I

I

1

CYCLONE

+

STACK

I

BAGHOUSE

f

COMBUSTOR

OVERBED FUEL & S *O R B E N T

+

* *

"

COOLIN1 -. I3 WATER -.

UNDERBED FUEL & *BENT

AS RECYCL -

-

WINDBOX

?

V

CiYCLONE P'RODUCT

AIR

UNDERBED OVERBED BAGHOUSE PRODUCT PRODUCT PRODUCT

Figure 1 Schematic diagram of FBC process.

is increased substantially, i.e., to 4-8 d s e c , the bed particles become entrained and are carried out of the combustor. These particles can, however, be captured by a "hot" cyclone and returned to the bed. A hot particle circulation loop is thus established that allows particle reactions to continue throughout the complete cycle. This type of system offers some advantages for capture of pollutant species and combustionof low-reactivity fuelsdue to the considerably increased residence time of fine particles in the bed. The actionof this systemis referred to as circulating fluidized bed combustion (CFBC). This section is included only as a basic introduction toFBC.The literature on FBC is quite voluminous, a recommended source beingthe American Society of Mechanical Engineers' series of conferences, themost recent of which is the 12th International Conference onFBC held in San Diego in May 1993.

Anthony and Preto

470

111.

FBC FOR WASTE MINIMIZATION

Due to low organic emissions levels, FBC is a promising technology for disposal of organic material-laden wastes. Donlee Technologies (USA) has developed a CFBC boilerto incinerate infectious hospital wastes by cofiring with coal [16]. Based on pilot-scale testing, they report that PAH, dioxin, and furan emissions were typically well below those achieved with other typesofhospitalwasteincinerators. Furthermore, suppression ofHClofup to 50% was achieved. Organic emissions metboth the Pennsylvania Department of Environmental Resources and proposed California restrictions. FBC-based technology meeting all emission limits for PAH, dioxins, and furans has also been offered by Ogden Environmental Services (USA) for the successful destruction of a number of hazardous wastes, including PCB-contaminated soil [17,18]. FBC is being examined in Italy for the combustion of organic material-laden industrial, agricultural, and domestic wastes. Published results[l91 indicate emission levelsof dioxins and furans much lower than proposed European legislated limits.The potential of FBC technology as an inexpensive method for disposal of hazardous wastes is being examined in India by Sandoz (India) and Thermax [20,21]. Thermax also reports that there are now 200 FBC installations operating in India burninga wide rangeof fuels fromlow grade coals to agrowastes such as rice hulls [21]. In India, FBC is overtaking rotary kilnsas the preferred technology for the incineration of solid and liquid wastes [21]. The use of FBC (primarily bubbling bed technology) for incineration is strongly established in Japan [22,23]. FBC is routinely used to burn MSW, plastics, and tires among other materials (one supplier alone, EBARA Corp., has 40 FBC incinerators operating [23]). It has been reported that there are currently 113 FBC incinerators operating in Japan [24]. In Europe there are over 50 FBC incineration plants [25]. Studies have also shown that FBC can burn RDF and industrial waste plastics either directly or by cofiring with coal [26,27]. A particularly interesting design concept, the fast internally circulating bed [25], used FBC technology to achieve “100%” elimination of all pollutants. The principal disadvantageof the concept was that the payback periodin the Austrian contextwas inherently long, i.e., 15 years for a 10 MW(th) unit. In North America, the use of bubbling fluidized bed combustion is increasingly finding more acceptanceas a means for the production of energy from the combustion of used tires and municipal wastes [28]. In the UnitedStates, circulating fluidized bed combustion is being used for disposal of hazardous wastes [17,18]. In Canada, a 10 MW(e) revolving fluidized bed unit has recentlybeen built to cleanly burn residues fromtar ponds near Sydney,Nova Scotia [29]. Although the main thrust of this project is to clean up 700,000 tons of coke oven residues,the energy production is not unimportant. This unit is due to become fully operational in 1993. Cofiring involvesthe burning of a waste “fuel” in conjunction with a premium fuel(coal, gas, or oil). Cofiring is used to supplement a waste fuel that may not be able to sustain combustion on its own or as a topping fuel if insufficient waste fuel is available. Cofiring, also known as “smart-burn,” does not prevent recycling programs,as any reduction inthe quantity of waste (particularly MSW) can be made up with the premium fuel. Cofiring thus appears to offer a palatable solution to the considerable municipal waste disposal problems now being experienced across North America. Traditional resistance to incinerators is likely to be reduced since as little or asmuch MSW canbe burned or recycled as desired, based on local economic or legislative requirements. Recently, the idea of cofiring coal and MSW or RDF has begun to find favorin the United States [30]. This is especially true in California, where the use of such wastes allows positive returns (extension of the lifetime of disposal sites; reduction in waste site emissions of green-

471

Fluidized Table 1 BubblingFBCCombustion of Biomass paper husks waste basis Corn pellets Rice grass Alfa Dry

cobs

59.8 83.1 15.2 0.2 1.6 HHV (MJkg) 18. l Particle size 40-50 mm diam, 150mm-long cylinders Feed rate 40-80 (kg/hr) Fluidization 0.9-1.2.4 velocity (dsec) >98% Combustion efficiency 100-400 ppm CO % Volatiles % Fixed C % Sulfur % Ash

80.9 14.6

-

13.4

-

4.4 22.2 18.8 14.1 3-cm fragments 5 10-mmhusks

-

Wood pellets

-

-

-

-

l-cm pieces

2-3-cm pieces

70

80-90

-

-

61

20-70

2 1.4

0.4-2.2

1.3

96% >96%

>97%

>90%

1.9%

>200(up ppm

0.1-0.3%

0.6-1.0%

to 0.5 %)

60-150 NO, @Pm) 100-180

so, (PPm)

100-200

>450

155

-

50-150

-

60-80

-

house gases and other toxic products; and generation of energy in the form of heat, process stream, and/or electricity) while still meeting the strict state environmental guidelines[31]. Elliot [31] quotes a figure of 3.6-4.1 cents/kWh for cofiring MSW compared with approximately 5.5 cents/kWh for a new coal-burning plant.Furthermore, these figures do not include a credit for the extension of the lifetime of the waste disposal facility or reduced liability costs for storage of potentially hazardous materials such as used tires. The economics of cofiring are therefore very favorable, findings that have been reinforced by a recent detailed literature survey on application of FBC to MSW [32]. In Canada, a number of companies are evaluating this technology and CANMET is also actively supporting FBC technology for biomass combustion [33,34].

A. Firing of Waste Biomass CANMET has tested a number of waste biomass fuels in pilot-scale bubbling fluidized bed combustors (40cm X 4 cm units with a total height of 5 m) [35]. The fuels that have been burned are paper waste pellets, alfa grass, wood chips, corncobs, and rice husks. Proximate analyses of the fuels, where available, operating conditions, combustion performance, and emission levels are given in Table 1. The only serious problem associated with burning these fuels was feeding them into the bed. This was largely due to the pilot scale of the equipment; the alfa grass and wood chips had a tendency to “bridge” in the fuel hopper; the cyclone was overwhelmed by ash during the alfa grass trials. Both the alga grass and corncob exhibited significant particle carryover. Freeboard combustionwas found to be significantfor all biomassfuels, particularly if the bed temperature was low. For instance, for the paper pellet tests, when the bed temperatures

472

Anthony and Preto

were between 600 and 800”C, the freeboard temperatures were typically in the range of 10001100°C. However, if the bed temperature was increased to 700-95OoC, the freeboard temperatures dropped to 650-950°C. In the case of this particular feedstock there were also some problems associated with defluidization of the bed. This was theorized as being due to agglomeration, possibly caused by the chlorine contentof the fuel. Scanning electron microscopic examination of the bed material showed high levels of chlorine in the limestone particles. Rice husks proved to be an ideal fuel for pilot-scale trails as their small uniformsize and low compressibility allowed them to be easily screw-fed into the bed. Their combustion characteristics were excellent (Table 1). The principal finding of this work was that the primary challenge for FBC utilization of waste biomass fuels is that of feeding the fuel. Once this can be done successfully, complete combustion and low emissions of both SO, and NO, can be achieved. Reduction in CO emissions and particle carryover require only a longer residence time, i.e., a substantial freeboard. Combustion of wood waste (sawdust and hog fuel) has beencarried out at the University of British Columbia using a 150-mm internal diameter circulating fluidized bed reactor [33]. NO, emissions were found to be low (50%) from wastewater treatment operations. These materials typically contain high moisture and have low heating values. Approximately half of this material is incinerated, and the remainder is used as landfill. Increasing concerns about heavy metal contamination may make the landfill option less attractive in the future. A study has been directed by CANMET, using a 150-mm-diameter CFBC unit, to inves[34]. Eight trails were pertigate the cofiring of these sludges with high sulfur bituminous coal formed with coaYsludge mass ratios varying between 3:1 and 1 :3. Sludge moisture ranged form 50 to 60%. One test, burning only sludge, quickly demonstrated that the sludge alone could not sustain combustion. In general, combustion efficiencies of 93-98% were achieved (based on carbon conversion). Emissions were low; NO, varied from 100 to 200 ppm, and CO levels ranged from 30 to 700 ppm. The only significant problem was, once again, feeding due to the fibrous nature of the sludges. To echo an earlier conclusion: If feed problems can be resolved, there are no barriers to using FBC technologyfor minimization of pulp mill sludges. The economics of sludge combustion would probably be significantly improved by better dewatering techniques. In a joint project, CANMET and Air Products and Chemicals, Inc. (USA) have used a 40 cm X 40 cm FBC pilot plant [35] to study the combustion performance of deinking sludges cofired with propane. Again,once feed problemswere overcome, all of the combustion results were successful. Due to low sulfur content anda high degree of sulfur capture inherentin the sludge, SO, concentrations remained below 5 ppm. NO, and CO concentrations were easily maintained below 150 ppm and 100 ppm, respectively. HCl levels were typically less than 15 ppm. At low temperatures (

To

L. Bound 0.D0000

11 12 13 14 14 15 16 16

0.00000 0.00000 0.00000

0.ooooo 0 .ooooo 0.00000 0.00000

Optimal 0.00055 0.00003 0.00019 0.00000 0.00157 0.00000 0.00087 0.00000

U. Bound

0.01D00 0.01000 0.01000 0.01000 0.01000 0.01000 0.01000 0.01000

Marginal 0.000 0.000 0.000 1.17E+7 0.000 3.26E+7 0.000 1.17E+7

SECONDTIMEPERIOD

L. Bound

From To 1 -> 1 1 3 -> 12

5 -> 6 -> 7 -> 8 -> 9 -> 10 ->

0.00000 0.00000 0.00000

13 14 14 15 16 16

0.00000 0 .ooooo

0.00000 0.00000 0.00000

Optimal 0.00082 0.00241 0.00027 0.00014 0.00115 0.00000 0.00081 0.00000

U. Bound 0.01000 0.01000 0.01000 0.01000 0.01000 0.01000 0.01000 0.01000

Marginal 0.000 0.000

0 IO00 0.000 0.000

2.88E+8 0.000 0.000

Figure 2 Continued.

B. US/REMAXBfor Heterogeneous Multilayer Systems 1 . Model Background For optimizing management of complex heterogeneous systems, one would rather useUS/ REMAXB [l61 than US/WELLSD.This is the basic version of the Utah State response matrix model. To develop influence coefficients, it uses code modified fromMODFLOW, a modular finite-difference groundwater flow simulation model [4], and STR, a related stream routing the same module [17]. The physical system data neededby US/REMAXDcanbeinputin format as is used by MODFLOW and STR. Internally, US/REMAXB also uses a portion of PLUMAN, a decision support system for optimal groundwater contaminant plume management [18], and other code. The optimization model formulation capabilities are similar to those of US/WELLSD (Table 2). For steady state, the generic objective is to minimize J

K

where Wjis the weight assigned to pumpingin cellj, dimensionless or [$.T/L3].US/REMAXB can employ constraints 1-3 of US/WELLSDfor multiple layers. Similar to the US/WELLSD constraint 4, US/REMAXB can force total extraction to exceed, equal, or be less than total injection. Again, via the sign on the weighting coefficients, one can perform maximization. One can also achieve multiobjective optimization by the weighting method. Whereas in US/ WELLSD the same weight must be applied to all extraction wells in a time step (and a different weight can be used for injection wells, but the same must be applied to all such wells in a particular time step), in US/REMAXB eachwell can employ a different weight.

PC Software Capture for Plume Optimizing

607

2.Application andResults Introduction. For illustration, we discuss addressing a contaminant plume in a representative study area. First, the study area is described and the results of continuing current management are predicted, using MODFLOW+STR for flow simulation and MOC [l91 for transport simS/O model is ulation. Then an approach to developing an optimal strategy is discussed, the applied, and an optimal strategy is computed. Next, the system response to implementing the optimal strategy is verified using MODFLOW+STR and M W . Finally, slight variations in themanagementgoal or situation areassumed andnew optimalstrategiesaredeveloped. Computed optimal strategies are compared. Suguino [20] first addressed this study area using PLUMAN. Some of the discussion below follows his development. Study Area Description and Situation. The area (Figure 3) measures about 4.3 km by 4.3 km. It is bounded on the north by a large saltwater body; on the south, east, and northwest by impermeable material; and on the west by a lake. A river transects the area from south to north. Aquifer parameters of this example study area were obtained from ranges reported by Todd [21]. For the unconfined upper layer (layer l), parameters are as follows. Hydraulic conductivity: 1stzone: 2nd zone: 3rd zone:

45 d d a y (coarsesand)fromlaketocontaminantspillarea(columns 57-58) 30 d d a y (medium sand) in irrigated area (columns 51-56) 450 &day (fine gravel) in contaminant spill area (columns 37-50).

Specific yield: 1st zone: 0.27 (coarse sand) 2nd zone: 0.28 (medium sand) 3rd zone: 0.25 (fine gravel)

Figure 3 Finite-difference grid for the area addressable with US/REMAXB.

1-36and

608

Peralta et al. Recharge by deep percolation and/or irrigation:

1.167 X 1.928 X

d s e c in nonirrigated area d s e c in irrigated area

In the confined lower layer (layer 2): 0.1564 m*/sec Transmissivity: Saturated thickness: 30.0 m Storage coefficient: O.OOO1 Finite-difference modelsare to be used in this study. This requires systemdiscretization. The resulting block-centered cell grid (Figure 3) has 58 columns and 39 rows. Cell side lengths range from 3 to 400 m. Because MOC will be used for transport simulation near the plume, cells of uniform size are specified for that region. The resulting 17 row by 20 column region (subsystem) near the plume has squarecells of 15.2 m (50 ft) side length. A conservative (nonreactive) contaminantis assumed to be spilled in the top aquifer layer (layer 1) of cell (22, 18) or (ll,, 3,). (The subscript “S” after a cell row or column index indicates that the cell is in the subsystem.) This cell is treated as a continuous source during the management period. Initially, pumping for water supply occurs in two cells between the plume and the river. One well is in layer 1 of (23, 15) or (12s, 15,). The other well is in layer 2 of (18, 18) or (7,, 18,). There is immediate concern about the potential for contamination reaching the supply well in layer l . Nonoptimal System Response Determination (Step 1). Before one attempts to develop an optimal strategy, one usually demonstrates the need for such a strategy. This requires predicting system response if no optimal strategy is implemented. Frequently, simulation models are used for this action. Here, MODFLOW+STR computes the potentiometric surface that will result from assumed steady-state conditions (Figure 4). Because of the gradient, the contaminant will tend to migrate toward the supply wells. M W is used to quantify the migration resultingin the subsystem from the steady flow. Figure 5 shows the 210 ppb contour expected to result 60 days after contamination begins. Furthermore, concentrationin the cell containing the drinking well (12,, 15,) reaches 3 17 ppb 8 months after the spill. We assume that this concentration level exceeds the health advisory for human consumption and that developing a plume capture strategy is desirable. Management Goals Specification and SI0 Model Formulation for Scenario l (Step 2). The assumed goal is to minimize the steady pumping (extraction and injection) needed to capture the plume. Plume capture will presumably be achieved when hydraulicgradients, just outside the plume boundary, all point toward the plume interior. We also want the head at extraction wells not to drop too far (to avoid reducing saturated thicknessby more than about 10%)or the head at injection wells not to rise to the ground surface. These criteria identify the example problem termed Scenario 1 . The SI0 model formulation for this scenario is shown below. The model computes the pumping strategy that minimizesthe value of the objective function, subject to the stated constraints and bounds. Locations of potential injection and extraction wells to be considered by the model are shown in Figure. 5. Figure 6 identifies head difference (gradient) control cell pairs and shows the direction that will be imposed on the hydraulic gradient by any computed optimal strategy. Theseare placed to enclose the plume projected to exist by day 60.A modeler can select potential well locations on the basisof practical experience.For example, the closer the injection wells are to the head gradient control locations, the less pumping is needed to

PC Sofnvare for Optimizing Plume Capture

609

I H

Figure 4 Nonoptimal (unmanaged) steady-state potentiometric surface contour map for the study area of Scenario 1 (meters above MSL). Js 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 ~ 1 1 5 1 5 ~ 1 8 1 9 2 0 legend 12

l3

14

U 16

l7 18 19 20 21

8 9 ‘S

10 11 12

22

23

l3

24

14

25

K

26 27 28

16 17

16ll79

21

23

25

27

29

31

33



unmanaged wen [ I 1 layer 1 [ 2 ) layer 2 potential managed injection well potential managed extraction well contamlnantsource

B 210 ppb concentration contour

0

meter 30 45

K

5

35

J

Figure 5 Subsystem discretization, potential well locations for Scenario 1, and 210 ppm contour, 60 days after contamination begins.

610

Peralta et al.

J*

1 2 3 4 5 ' 6 7 8 9101112l314lSl617181920

1

12 13 14

2 3 4 5

unmanaged wen ( 1 1 layer 1 ( 2 1 layer2

Is l6

6

'S

Legend:

potential managed

lnJectionwell (upper bound on head)

l7

7

#)

8 9

19

14 15 16

2o I 21 22 23 24 25 26 27

l7

28

ta 11

l2 l3

16 17

l9 2321

25 2927 d

31

33

Q potential managed extraction well (lower bound on head)

contaminant source +

gradient canstraint 0

IS 30 45 m

tS¶

35

Figure 6 Head-difference constraint locations applied within the S/O model in Scenario 1.

satisfy the head-difference constraint. Thus, the modeler might want the model to consider pumping sites near the location where heads need most to be affected. The model objective is to minimize the value of Equation (3), using weights of 1, subject to

. . . , 22

Gb 2 0.01,

for B = 1,

hp

for 2 = 1,

(5)

for 2 =

(6)

2

15.0,

h* I25.0,

. . . ,6 1, . . . , 25

(4)

where GBis the difference in head between a pair of cells, the first located farther from the plume. A positive value denotes a higher head farther from the plume, [L]; ha is the hydraulic head just outside the casingof pumping well P located in the center of a pumping cell, [L]; 6 is the index denoting pair of cells head-difference (gradient) control pair; and 6 is the index denoting pumping well at the center of cell j or k. Here j = 6 and k = 25. Note that identifyingthe location of potential extraction and injection wells for the model (Figure 5 ) does not mean that the model will choose to pump at those locations. Via the optimization process, the model might choose to pump at only a few of the potential sites. The computed strategy will require less total pumping than any other strategy possible forthe specified potential well locations and imposed bounds and constraints.Furthermore, since this isa steady-state problem, steady-state system responseto implementing the strategy computedby the model will satisfy all those bounds and constraints. This is verified in the next step. Optimal Strategy Computation and Verificationfor Scenario I (Step 3). The optimal strategy computed for Scenario 1 is shown in Table 3. Because the model is minimizing pumping only for plume containmentin layer 1, no extraction is shown for layer 2. The original unmanaged pumping does continue from original supply wellsin both layers (Figure3) but is not included in Table 3 because the model is not optimizing that pumping.

ware PC

611

Capturefor Plume Optimizing

Table 3 Pumping Results for the Sample Scenarios g(eW

b(inj.)

(g

+ b) total

(@m) (m’lsec) Layer 2nd1st Layer

Constraints Scenario

~~~

1

2 3

onconstraint Gradient same the located on heads layer, head constraint on injection and extraction well. constraint: pumping Added = total extraction of sum total sum of injection. heads Gradient constraint on same and located theon on different layers, head constraint on injection and extraction wells.

0.01338 (212.05)

-

0.03358 0.02020 (320.13) (532.18)

0.01702 (269.74)

-

0.03404 0.01702 (269.74) (539.48)

-

-

~

0.00300 0.00329 0.03786 0.04415 (47.54) (52.14) (600.03) (699.69)

Figure 7 shows the locations of wells that will pump, according to the optimal strategy. It also shows the head-difference constraints [Equation (4)] that will be tight. Tight constraints are those that are satisfied exactly. The other gradient constraints are also satisfied, but the so. These latter head-difference constraints are “loose” (there model had no difficulty in doing at the two cells coupled by an arrowin Figure is more than0.01 m difference between the heads 6 but not shown at all in Figure7). No heads are against their bounds. Therefore neither Equation (5) nor (6) is tight. It is appropriate to verify that the computed strategy accomplished its goal of plume capture. MODFLOW+STR can be used to demonstrate how quickly the optimal steady pumping

%

l 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1819 20

12 13 14 15

16

n

18

Is

19 20 I 21

9 10

22 23 24 25 26

11 12 l3 14 15

16 ti

27

28 16 l7

19

21 23

25272931 J

33 35

I.egend:

unmanaged well ( 1 1 layer 1 c21 layer 2 activemanaged injection well actiuemanaged extraction well 1contaminant MUI ’ce

+ lightgradient constraint meter 0 63045

Ea

612

Peralta et al.

strategy will cause the desired gradients to occur. Transient simulation demonstrated that the gradient constraintswould be satisfied 30 days after implementing the optimal pumpingstrategy (Figure 8). Figures 9 and 10 show the ultimate steady-state surface resulting from strategy implementation. Clearly, a groundwater divide has been formed between the plume and the supply well. MOC is used to predict the pollutant transportthat would result from strategy implementation. No contaminant moved past the injection wells. Theoretical verificationof the optimality of the computed strategyis beyond the scope of this document. However, many texts onoperations research and linear programming assure the optimality of solutions to models havinga linear objective function and constraints'. Alternative Scenarios. Scenario 2. This scenario differs from the previous in the addition of a constraint forcing total injection to equal total extraction around plume. the Again, pumping fromthe two supply wells is not included in the total. aquifer optimal a l l gradients contam. punping constraints beg ins ach i eved I 0

-I

I

60

90

check of cont.conc. I 240

days

Figure 8 Time scale of Scenario 1.

J -

column

3

0

c l

H

Figure! 9 Subsystem potentiometric surface resulting from implementing the optimal pumping strategy for Scenario 1 (meters above MSL).

613

PC Software for Optimizing Plume Capture Ips 1.36

Well

I 2

3

4 S

0.62 "0.60 9.69 0.85

Well 6 7 8

gprn

2l.56

9.89 9.49

- -EX55

9 10

13.46

Ips gpm -449.02 -3.09 79-58 5.02 'M.67 4.46 7.88 l2496

-100.00 "158482

Figure 10 Subsystem potentiometric surface resulting after6 months of optimal pumping for Scenario 1 (meters above MSL).

Results in Table 3 show an increase in extraction and a decrease in injection. Total pump the phenomeingneeded for plume containment increased slightly (1.4%). This illustrates of the non-increasing the number or restrictiveness of constraints does not improve the value objective function. Although total pumping increased, one less extraction well is used in this strategy than in the previous (Table 4). The same number of gradient constraintsare tight, but the locations of the tight gradient constraints differ slightly. Scenario 3. This scenario demonstrates what might happen if involved managers have conflicting goals. It differs slightly fromScenario 1. In addition to controlling the plume, the agency wishes to extract more from layer 2 for water supply. Three new potential extraction wells are located in cells (19,25), (20, 25), and (21, 25), as if along a nearby road. Pumping is not permitted to change at the two initial supply wells.

Table 4 Numbers of Managed Wells that Will Pump Under the Optimal Strategies for the Tested Scenarios g(extr.1 Scenario l 2 3

Layer Layer 2nd 1st

Hinj.)

3 2

-

4

3

6 6 14

(g

+ b) total 9 8 21

614

Peralta et al.

As a result, the objective function is altered to maximize new extraction from layer 2 while still minimizing the pumping in layer 1 needed to capture the plume. This is achieved by assigning a negative sign to extraction from the supply wells, and minimizing: K

I

Since minimizing a negative number is the same as maximizing a positive number, minimizing negative extraction in layer2 means maximizing that extraction. Also added are new constraints imposed on vertical flow in cells (21, 21) and (22, 21). There, the head in the lower layer is forced to exceed that in the upper layer by 0.01 m, preventing the downward migrationof contaminant. Figure 11 shows the resulting optimal injection and extraction well locations and tight gradient constraints. The optimal pumping strategy includes seven extraction wells and 14 injection wells. Although extractionof polluted water decreases, injection increases with respect to Scenario 1 (Table 3). Extraction of water for public supply increases by 31% above the unmanaged rate. Although the gradient constraints are all satisfied by the optimal strategy, subsequent simulation demonstrated that the vertical gradient is reversed in some plume-containing cells in be careful in placing head which the gradient was unconstrained. This illustrates that one must or gradient control in appropriate locations. In practice, another optimization would be performed, using additional vertical head-difference or gradient constraints. Processing Considerations. It is useful to consider the resources required to address optimization problems. First, the total computer time needed to solve an optimization problem is of Js

1 2 3 4 5 ' 6 7 8 9101112I3MlSl6~181920 l2 1 l3 2 14 3 15 4 16 5 n 6 18 7 19 8 9

~~~d

B unmanagedWen ( 1 1 layer 1 ( 2 I layer 2

actiuemanaged inJection well actiuemanaged wtraction well

20 I

21 22 23

Is 10 l1 l2 l3 14 15

24

25 26 27 28

16 17 16 17 9.

21 23

25

27 29

31

contaminant source tight grad. wnstr. (heads on samelaywl ++ tight grad.constr. (heads ondiff. layer1 meter

c

0 15 3045

Ea

33 35

J

Figure 11 h a t i o n of optimal pumping wells andtight head-difference (gradient) constraints for Sce-

nario 3.

min)

615

PC Software Capturefor Plume Optimizing

interest. Table5 illustrates the time neededto address Scenario 1. Included stages useeither the discussed simulation models or the PLUMAN code on a 386 PC running at 33 MHz and having 4 MB RAM. Time required for US/REMAXB is comparable to thatof PLUMAN, since it uses many of the same solution procedures. Clearly, the stage of computing influencecoefficients, arranging the optimization model, and calculatingan optimal strategyis the most computationally intensive.For this scenario and stage, two steps can be distinguished. Thefirst involves computing influence coefficients. The second is model organization and optimization problem solution. Here, the step of generating influence coefficients requiresby far the most time. This results because this act essentially involves one simulation of a modified MODFLOW+STR per potential pumping location. Since thereare 31potential pumpinglocations, 31 simulations are performed to develop the influence coefficientsneeded for the response matrix. Themore decision variables (potential pumping rates), the more computer time involved in this step. The step involving model formulation and calculation of the optimal strategy is fairly short. The time needed to perform the optimization is a function of the number of decision variables (potential pumping rates) and state variables (heads or gradients that must be constrained within the optimization model). The larger these numbers, the more time required. Second, the size of the optimization problem being solved is of interest. For example, the special versions of US/WELLSD and US/REMAXB that are released in shortcourses are limited in the number of nonzero values they can have in the optimization formulation. (Even optimization algorithms that are not part of water management models are commonly limited either in the number of nonzeros or in the number of rows and columns in their constraint equations.) By way of explanation, there is one row in the response matrix per heador gradient constraint equation per time step of constraint. There is one column in the matrix per decision variable. For a steady-state problem, total matrix size is the product of the number of control locations andthe number of decision variables.The matrix contains one nonzero coefficientfor each potential pumping location-head control location pair (per time step of active constraint). For the steady-state Scenario 1, there are 31 X (22 31), or 1643, nonzeros due to influence coefficients. Thereare also 31nonzeros due to the weighting coefficients (even if they are 1in value) assigned to decision variables in the objective function. Thus, the optimization model formulation for Scenario 1employs almost 1700 nonzeros. (That of Scenario A using by considering US/WELLSD includes919 nonzeros.) This number can be reduced significantly injection in only every other cell on the plume periphery rather than in each cell. For example, 12 injectionwellswere considered, thenumber of nonzeroswouldbeabout ifonly 18 X (22 18) 18, or 738. In addition to reducing problem size, this would significantly reduce computational time.

+

+

+

Table 5 Computer Time Required to PerformEach Activity for Scenario 1 Step

Time

Software used

1

MODFLOW+STR head) nonoptimal (compute

2 3

MOC nonoptimal (predict atransport solute potent surface) in PLUMAN (compute influence coefficients, formulate management model and determine optimal pumping strategy) MODFLOW+STR (compute transient head response to optimal pumping) MOC (compute head and solute transport response to optimal pumping)

4

5

5.0 35.0 150.0 1.3 8.0

Peralta et al.

616

Reducing the number of nonzeros below IO00 is important becausethat is theupper limit on problem size in the inexpensive “special” versions of US/REMAXB and US/WELLSD.If problem size increases beyond that, software price increases dramatically. The full professional versions of the software can address problemsof virtually unlimited size.

V. SUMMARY Use of simulationloptimization modelscan significantly aid managementof groundwater contamination. It can speed the design process and reduce manpower costs. It can improve the produced remediation designs and reduce remediation costs. It can easily address problems previously considered very difficult. S/O modeling methods for groundwater flow management have been well established in research literature. Now, generally applicable S I 0 models are available for use on Pcs. The discussed models, US/WELLSD and US/REMAXB, use linear systems theory, influence coefficients, and superposition. These models can addressa wide rangeof problems. Easyto use, they include all simulation and optimization algorithms neededto compute optimal strategies. US/WELLSD and US/REMAXBare perfectly applicable to linear (confined) aquifer systems and can be applied to nonlinear systems. The former is most appropriate for fairly homogeneous aquifer and stream-aquifer systems.The latter can address complex heterogeneous multilayer stream-aquifer systems. Increasing use of these PC-based S I 0 models is anticipated, especially as user-friendly options increase. Even the special versionsof these models (releasedat shortcourses),can solve important real-world problems.

REFERENCES 1. Peralta, R. C., and Willardson, L. S . , Optimizing ground water planning and management, U.S. Committee on Irrigation and DrainageNewsletter, AprilIJune 1992, Denver, Colo., pp. 61-65. 2. Clarke, D., Microcomputer Programsfor Groundwater Studies, Elsevier, New York, 1987. 3. Glover, R. E., and Balmer. G. G., River depletion resulting from pumping a well near a river, Trans. AGV, 35(3), (1954). 4. McDonald, M. G., and Harbaugh, A. W., A modular threedimensional finite-difference groundwater flow model, inTechniques of Water-Resources Investigations,U.S. Geological Survey, 1988, Chapter Al, Book 6. 5. Gorelick, S. M., A review of distributed parameter groundwater management modeling methods, Water Resources Res., 19(2), 305-319 (1983). 6. Morel-Seytoux,H. J., A simple case of conjunctive surface-ground-water management, Ground Water, 13(6) (1975). 7. Verdin, K. L., Morel-Seytoux, H. J., and Illangasekare, T. H., Users Manual for AQUISIM: FORTRAN N Programs for Discrete Kernel Generationand for Simulation of Isolated Aquifer Behavior in 2 Dimensions, HYDROWAR Program, Colorado State Univ., Fort Collins, Colo., 1981. 8. Heidari, M., Application of linear systems theory and linear programming to groundwater management in Kansas, Water Resources Bull., 18(6), 1003-1013 (1982). 9. Illangasekare,T. H.,Morel-Seytoux, H. J., and Verdin, K. L., A technique of reinitialization for efficient simulation of large aquifers using the discrete kernel approach, Water Resources Res., 20(11), 1733-1742 (1984). 10. Danskin,W. R.,and Gorelick,S. M., A policy evaluation tool: management ofa multisystem using controlled stream recharge, Water Resources Res., 21( 1l), 1731-1747 (1985). 11. Lefkoff, L. J., and Gorelick, S . M.,AQMAN. Linear and Quadratic Programming Matrh Generator Using lbo-Dimensional Ground-waterFlow Simulation for Aquifer Management Modeling,

U.S.Geological Survey Water-Resources Investigations Rep. 87-4061, 1987.

oftware PC

Capture forPlume Optimizing

617

12. Reichard, E. G., Hydrologic influences on the potential benefits of basinwide groundwater management, Water Resources Res., 23(1), 77-91 (1987). 13. Ward,R. L., andPeralta,R.C.,EXEIS-ExpertScreeningandOptimalExtractionllnjection Pumping Systems for Short-Term Plume Containment, Rep. ESLTR-87-57, Air Force Engineering

and Services Center, Tyndall AFB, Fla., 1990. 14. Willis, R., and Finney, B., Optimal control of nonlinear groundwater hydraulics: theoretical development and numerical experiments, Water Resources Res., 21(10), 1476-1482 (1985). 15. Aly, A. H., and Peralta, R.C., USIWELLSD, Extractionllnjection Well System for Optimal Groundwater Management: User’sManual, Biological and Irrigation Eng. Dept., Utah State Univ., Logan, Utah, 1992. 16. Peralta. R. C., Aly, A. H., Suguino, H. H.,Belaineh, G . , and Miyojim, M., USIREMAX’, Utah State Model for Optimizing Management of Stream-Aquifer Systems Using the Response Matrix LoMethod: User’s Manual, Version f.05,Biological and Irrigation Eng. Dept., Utah State Univ., gan, Utah, 1992. 17. Prudic, D. E., Documentation of a Computer Program to Simulate Stream-Aquifer Relations Using

a Modular, Finite-Difference, Ground-water Flow Model, U.S. Geological Survey, Open-File rep. 88-729, 1989. 18. Suguino, H., and Peralta, R. C., PLUMAN A Decision Support System for Optimal Groundwater Contaminant Plume Management: User’s Manual, Version 1.0, Software Eng. Div., Dept. Biological and Irrigation Eng., Utah State Univ., Logan, Utah, 1992. 19. Konikow. L. F., and Bredehoeft, J. D., Computer model of two-dimensional solute transport and dispersion in ground water, in Techniques of Water Resources Investigations,USGS, Washington, D.C., 1984, Book 7, Chapter C2. 20. Suguino, H., A decision support system for optimal groundwater contaminant plume management, Ph.D. dissertation, Dept. Agricultural and Irrigation Eng., Utah State Univ., Logan, Utah, 1992. 21. Todd, D. K., Groundwater Hydrology, Wiley.NewYork, 1980.

ADDITIONAL READING Colarullo, S. J., Heidari, M., and Maddock, T.,111 (1984). Identification of an optimal groundwater management strategy in a contaminated aquifer, Water Resources Bull., 20(5), 747-760. Gharbi, A, (1991). Optimal groundwater quantity and quality management with application to the Salt Lake Valley, Ph.D. Dissertation, Dept. Agricultural and Irrigation Eng., Utah State Univ., Logan, Utah. Peralta, R. C., Killian, F? J., Yazdanian, A., and Kumar,V. (1989). SSTAR UsersManual; Rep. IIC-8913, International Irrigation Center Utah State Univ., Logan, Utah.

This Page Intentionally Left Blank

30

Horizontal Wells for Subsurface Pollution Control

George Losonsky Eastman Christensen Environmental Systems Houston, Texas

Milovan S . Beljin University of Cincinnati Cincinnati, Ohio

1.

INTRODUCTION

Horizontal wells are emergingas a new technology for solving problems in the environmental industry. Horizontal wellscreen orientation complements typical aquifer geometry, groundwater flow patterns, and common site logistics. Soils are naturally stratified, and individual aquifers or water-bearing zones are much wider than they are thick. Contaminant concentrato remedial options are often highest directly beneath buildings, landfills, and other obstacles erations, so treatment facilities are constructed tens or hundreds of feet away from the target zone of remediation. Despite the dominance of the horizontal direction in aquifer shapes and groundwater flow, the predominant tool for extracting contamination from subsurface sources is a vertical well. However, in many environmental remediation scenarios, a horizontal well offers a better matchof form and function than a vertical well. The tabular geometry ofmany aquifer zones renders horizontal wells more productive than vertical wells. The specific capacity ratio of horizontal-to-vertical wells increases with decreasing aquifer thickness. Extraction of contaminated groundwater is often more efficient with horizontal Flow wells. A horizontal well placed characteristics ofmany aquifers create elongated contaminant plumes. through the coreof a plume can recover higher concentrations of contaminants at a given flow rate than a vertical well. Hydraulic barriersare most efficiently created using horizontal wells oriented perpendicular to groundwater flow direction. Logistical advantagesof horizontal wells are obvious. Horizontal wells avoid theneed for installing wellheads inside buildingsor in the midst of complex manufacturing or process fapenetrated to extract leachateor cilities. Landfills, spoil mounds, and landfill liners needbenot other underlying contaminants. Fractures in an aquiferare commonly vertical. Because fluidor vapor recovery from fractured zones requires penetration of numerous fractures, a horizontal well oriented normally to vertical fractures is the optimal tool for pump-and-beat or soil vapor extraction systems in vertically fractured zones. By analogy, vertical wells are efficient in highly stratified soils with 619

620

Losonsky and Beljin

little vertical communication between strata, where fluid or vapor recovery from many thin layers t h u g h a single wellbore is required. Injection of groundwater is part of some remediation systems, either to create a water table mound or to reinject treated effluent from a manufacturing plant into the subsurface. Water table mounds can help to control flow of contaminants toward recovery wells or trenches, or they can serve as hydraulic barriers. Manufacturing plants can avoid high sewer discharge costs if their treated plant .effluent can be reinjected into a non-drinking water aquifer. Reinjection can cause mounding, but the mounding can be minimized by using horizontal wells.

II. DRILLINGTECHNOLOGIES Various technologies can be used for installing horizontal wells for subsurface pollution control. Such wells are typically installed in unconsolidated soils 10-200 feet deep. Selection of drilling technique depends on surface access, well placement and completion requirements, and subsurface hydrogeology. Jetting and moling are the most common techniques used to create horizontal brings in shallow, unconsolidated soils. Both are mechanically simple and require minimal labor, using small tubulars that can be handled by one person. Small trucks can carry the hardware for both techniques. Moling employs a rotating bit, compressing soil into the hole wall. This requires a soft, compressible soil. Directional control is limited andrelies chiefly on trial-and-error targeting methods. Holes are only a few inches in diameter, limiting completion options. No drilling fluid is used, and no cuttings are generated. Clayey soils may become “damaged” (i.e., their permeability may be reduced) because the cuttings are not removed from moling boreholes. Jetting employs an off-axis, high-pressure water jet to fluidize the formation. Downhole transmitters anda walk-over receiver are used for surveying thewell location and determining jet orientation. The operator can steer the system accuratelyin unconsolidated, homogeneous soils. Jetting requires that there be no interference sources between the surface receiver and downhole transmitter.The transmitter must not be located deeper than25 ft below the surface. Drilling fluid lubricates the drill string. Cuttings mixed with drilling fluid enter the soil formation along the wellbore, which may cause formation damage as with moling. Rotary drillingemploys a stabilized assembly allowing control of well path inclination but no horizontal directional capability. Thedrilling rig must be large enough to providerotating, push, and pull forces. The technique is used for oil and gas recovery and for installing utility lines under rivers. Low drilling fluid flow rates are used in utility line installation, where the fluid serves only to carry cuttings throughthe hole wall to keep it open and lubricate to the soil. High fluid flow rates are used in hydrocarbon recovery to remove cuttings from the wellbore and thereby preserve formation permeability. A large radius of curvature is required in rotary drilling to prevent failure of tubulars. The larger the tubular diameters, the larger the curve radius required to prevent failure of tubulars. Surveying toolsare similar to those used for magnetic and gravitational orientation surveys. These tools work at any depth, but only in the absence of magnetic interference. Positive displacement steerable motor technology offers vertical and horizontal directional control. The radiusof curvature is small compared to that of rotary drilling, because the drillstring rotation is minimal. Penetration rates are high because ofrapid bit rotation. Drilling fluid forms an impermeable layer on the borehole wall, preventingfluid and cuttings from entering the formation, as with rotary drilling. The impermeable layer also prevents hole collapse by maintaining positive fluid pressure within the wellbore. The impermeable layer must be removed following installation of the screened casing, to expose undamaged formation. Drilling

Horizontal Wellsfor Pollution Control

621

fluid must remove cuttings from the wellbore. Viscosity and turbulence of the drilling fluid allow it to clear the wellbore. By comparison, vertical wellbores rely solely on viscosity. Turbulence prevents cuttings from accumulating along the bottom of the wellbore. Since turbulence increases with decreasing viscosity, ideal drilling fluid balances both factors. Thixotropic gels are well suited as drilling fluids. lbrbulence is also maintained with the help of centralizers, which provide a uniform annulus around the tubulars [l]. Percussion and vibration drilling techniques use air as a drilling fluid. These techniques are particularly suitable for drilling through heterogeneous soils containing bouldersor coarse gravel or through highly fractured rock formations. The wellbore remains dry, so fluid loss is avoided. However, percussion and vibration drilling techniques currently lack well-developed steerability and directional control.

111.

HORIZONTALWELLHYDRAULICS

Recently, due to increased interest in horizontal wells for oil production, a large number of papers have been published regarding the reservoir engineering aspects of horizontal drilling and reservoir simulation[e.g., 2-41. In the groundwater industry, thefirst theoretical analysis of groundwater flowto horizontal drains (collector wells) can be traced back the to early 1960s [ 5 ] . In the last few years, there has been renewed interest in horizontal wells for subsurface remediation [6-81 and for groundwater monitoring [9]. Langseth [lo] performed a numerical analysis of horizontal well performance and comparedinstallation and operation costs of horizontal and vertical wells.

A.Steady-StateFormulas Whereas a vertical well drains a cylindrical volume, a horizontal well of length L drains an ellipsoid. The zoneof influence is elliptical, with endpoints of the well constituting the foci of , is the ellipse. The area of the drainage ellipse, A

A,

=

n Reva

(1)

where R,, is the effective drainage radius of a vertical well in the same aquifer, anda is half the major axis of the ellipse [ll]: a = In order to compare the drainage area of a horizontal well with that of a vertical well, the drainage radius of a horizontal well, R,, measured in the horizontal plane that contains the well, is defined suchthat the corresponding circular areaA, equals the elliptical drainage area A, of the well: 2 A, = A, = n Rch

(3)

Combining Equations (1H3) and solving for a , a = (1512)[ O S

+ J0.25 + (We&)4]

0.5

A formula forestimating steady-state flow to a horizontal well in a homogeneous and isotropic aquifer is given as [12,13] 2nKBAs

(5)

622

Losonsky and Beljin

where Qh is the flow rate, [L3iT]; As is the drawdown, [L]; L is the length of horizontal well, [L]; r, is the well radius, [L]; K is the hydraulic conductivity, [ m ] ; and B is the aquifer thickness, [L]. The specific capacity of a horizontal well, Jh = Q,,/&, is usually greater than that of a comparable vertical well, J , = QJAs, except in relatively thick and highly permeable aquifers. In Figure 1, the ratio of the specific capacityof a 500-ft-long horizontal well,Jh, to that of a fully penetrating vertical well, J,, is plotted as a function of the hydraulic conductivityK (gpd/ft2) for three different aquifer thicknesses(B = 5 , 20, and 100 ft). The figure reveals that the ratio of the horizontal to vertical well specific capacity (productivityratio) is the greatest for relatively thin aquifers with low hydraulic conductivity. Thus, horizontal wells are most effective in thin, low-permeability aquifers where vertical wells commonly fail to produce significant volumes of groundwater.

B. Anisotropy Effect When the vertical conductivity,K,, is different from the horizontal conductivity,Kh, the flow rate to a horizontal well can be computed as follows:

where

.

=

l0O0? Aquifer Thickness:

.001

.01

.l

1

K [gpdlft2] Figure 1 Productivityratiovs.hydraulicconductivity.

10

mmmmmm~

b=20ft

*M*UUVU

b=lOOft

100

1000

Wells

Horizontal

623

for Pollution Control

Figure 2 plots the specificratio of a 500-ft horizontal well as a functionof the conductivity contrast and the aquifer thickness. In the case of anisotropic aquifers, the specific capacity ratios of horizontal to vertical wells increase if the hydraulic conductivity ratios, K,,/K,, decrease. This increase is inversely proportional to aquifer thickness.

C. Formation Damage and Effective Well Radius During drillingof a well, drilling mud can invade the aquifer and change its permeability in the vicinity of the well and cause formation damage or skin effect. The thickness of the “skin zone” will depend not only on drilling technology but also on the permeability of the aquifer. High-permeability aquifers exhibit a larger skin zone than low-permeability aquifers. However, the reduction in permeability is smaller in a high-permeability zone than in a low-permeability zone. The additional drawdown due to the change in permeability and the turbulent flow around well losses. Because of lower flow rate per unit screen length, horizontal wells the well is called show smaller well losses due to drilling mud invasion than vertical wells. Van Everdingen [l41 defined a dimensionless “skin factor” S as s = [ ( W

- 11log(rirw)

(7)

where r, is the radius of the skin zone with permeability k, in an aquiferof permeability k, [L2]. The effective radius of a well,rk, is defined as the theoretical radius of the well required to match the observed pumpingrate and drawdown. The effective radiusof a hypothetical vertical well that pumpsat the samerate as a horizontal well canbe computed using the equation

1000 3

100

10

1

Figure 2 Productivityratiovs. anisotropy.

Aquifer thickness:

Losonsky and Beljin

624

The following formula relates the skin factor, S, and the effective radius of a well, rk:

rk = r, exp(-S)

(9)

Renard and Dupuy [l51 gave the following solution for computing flow rate, Qh, from a horizontal well with the skin factor S: Qh =

cosh”(2a/L.)

2nKBAs (B/L)log(B/23crb)

+

+

S

If the length of a horizontal well greatly exceeds aquifer thickness, L % B , or if well length is small compared to drainage radius, R,, Equation (5) reduces to the well-known Thiem’s equation,

with the effective well radius being equalto one-fourth of the horizontal welllength, rk = L/4.

D. Off-Center Horizontal Well All solutions presentedso far have assumed that the horizontal well is located at midheight in the aquifer cross section. The distance, 6, from aquifer mid-height to the horizontal well is called the eccentricity of the well. The flow rate of an eccentric horizontal well can be calculated using the formula

2nK#As Qh =

[

a+-

1%

L12

]

-k

log

[

+

]

(pB/2)2 p2a2 pBrJ2

As long as thehorizontal well eccentricityis relatively small, 6 < 2 B/4, the performance of a horizontal well is not significantly affected [16].

IV. TRANSIENT SOLUTION Gringarten et al. [l71 used source and Green’s functions to solve a problem of transient flow into a well with a single vertical fracture. For the uniform-flux fracture, i.e., a fracture with uniform fluid entry along its length, the pressure distribution is given by

where S, is dimensionless drawdown;rD is dimensionless time;,x = x/xf; yo = y/xf;xf is half fracture length, [L]; x is thedistance measured fromthe well center along the fracture, [L]; and y is the distance perpendicular to the fracture, [L]. The concept of uniform flux and infinite-conductivity fracture can be applied to a horizontal well by assuming that the horizontal well is a vertical fracture with a width equal tothe radius of the well [16]:

Horizontal Wellsfor Pollution Control

625

where LD = [ W 2 B ] m ; rwD = 2rJL; XD = YD = 2yIG zWD= z JB; and z, is the vertical distance from the aquifer bottom, [L].

Z,

= zIB;

For a long horizontal well,the summation termin Equation (14) approaches zero, and the horizontal well solution reduces to the fracture solution, Equation (13).

V. EXAMPLEAPPLICATIONS A. Case 1: Texas Gulf Coast Industrial Landfill Site Before the advent of horizontal wellbore technology, vertical wells were used to recover contaminated groundwater under landfills. In aquifers with high hydraulic conductivity, vertical wells placed along the periphery of a landfill can influence the aquifer below the landfill. If the size of the plume is small comparedto the size of the landfill, migration of contaminants from to vertical extraction wells along the periphery spreads the coreof the plume below the landfill the contamination. Low hydraulic conductivity requires placement of vertical wells within the landfill, penetrating the landfill liner. Long-term sealing of the liner around the vertical well casing cannot be guaranteed. Horizontal wells avoid these problems by placing a production screen directly in the plume without penetrating the landfill. In 1991, a horizontal well was installed beneath an RCRA facility on the Gulf Coast of 900 ft Texas to recover contaminated groundwater. The landfill occupies an area approximately by 900 ft. Its depth is 30 ft, and the slope along its sides is 27" from horizontal (Figure 3). Numerical modeling prior to installation predicted that a single horizontal well wouldbe hy-

RED-BROWN CLAY GRAY-TAN SILTY CLAY RED-BROWN CLAY FINE SAND

""""""""""""I""""""""""""""""""""""""""". GRAY CLAY

Figure 3 Texas Gulf Coast landfill-cross section with horizontal well path.

626

Losonsky and Beljin

draulically more efficient than a vertical well system [18]. The single horizontal well replaces five vertical recovery wells. Fewer pumps and a lower volume of recovered water save operation and maintenance costs of the remediation system. 1. Hydrogeology Interbedded, 20-30 ft-thick reddish brown clay and gray and tan silty or sandy clay characterize the subsurface from0 to 98 ft depth, corresponding to 26 ft above mean sea level(MSL) to 72 ft below MSL (Figure 3). The uppermost silty clay layer and the top of the underlying reddish-brown clay layer have been excavated to accommodate the landfill. Tan fine sand occurs at 100-128 ft depth from the surface,or 70-95 ft below the base of the landfill. Blue-gray and black-gray clay underlies the sand. The silty or sandy clay layers and the sand layer are saturated. The reddish-brown clays are aquicludes or aquitards. The tan sand layer is the target of the remediation effort. The hydraulic conductivity, K, of the sand is 30 ftlday, and storativity, S, is 1 X lo-', indicative of a confined aquifer. There is no significant potentiometric gradient in the absence of pumping. 2. Well Specifications Sixty feet of 65/-in. slotted stainless steel screen was installed beneath the landfill at a total vertical depthof 114 ft. This depth is 4.7 ft below mid-height in the aquifer, givingthe well an eccentricity of 4.7 ft. The wellhead protrudes at a 41 S " angle, 20 ft from the edgeof the landfill. The horizontal displacement from the wellhead to the beginning of the screen is 248 ft, reflecting a radius of curvature of 275 ft. The curved sectionof the well was cased with 10 -in. of high-density polyethylene liner to isolate soil zones during drilling and to provide structural support for the inner stainless steel casing. 3.DrillingProcedure Drilling commenced with augering of a 16-in. wellbore to a measured depth of 37 ft, which corresponds to the kickoff point forthe curved section of the wellbore. A steel conductorpipe, 14 in. in nominaldiameter, was cemented in the 16-in. wellbore. The cement was allowed to set for 12 hr. The curve drilling assembly then drilled and simultaneously caseda 13-in. borehole with an effective drilling radius of 275 ft to a measured depthof 272 ft. The curved sectionwas drilled using a 13-in.wing bit. High-densitypolyethylenecasing 10 in.thickwasthencehad set, the horizontaldrilling mented intothe curved section of the wellbore. After the cement assembly drilled an 8%-in. wellbore to a measured depth of 352 ft, with a final inclination of 90.5" from vertical. The horizontal section was drilled using an 8%-in. rock bit. Schedule 40, entire 3 16 SS ERW stainless screen casing, with6Ys-in. outer diameter, was installed along the length of the wellbore. Only the60-ft section of stainless steel casing installed withinthe target zone was slotted. Slot size was 0.02 in. 'henty-foot sections of stainless steel casing were welded together, increasing installation time. Six 1 0 4 sections of prepacked screen were installed inside the stainless steel screen in the target zone. The prepacked screen was attached directly to an electric submersible pump oriented horizontally within the unscreened stainless steel casing. Well development required 18 hr,after which clear water was recovered. Residual drilling fluids were flushed out of the well with a submersible waterjetting tool. Drilling, installation, and well development required approximately 10 days. Duringdrilling, continuous updatesof inclination and tool faceorientation were provided to ensure proper wellbore placement. TruTracker surveys were taken as needed, at least once every 20 ft, to provide actual bit locations with respect to the proposed well path direction. The horizontal wellbore remained within a 4-ft vertical target range, thus exceeding the required accuracy of k 5 ft (Figure 3). The well remained within a 20-ft-wide horizontal target zone (Figure 4). Approximately 18,000 gal of cuttings, drilling fluids, and cement returns were generated dur-

627

Horizontal Wellsfor Pollution Control

Target Zone

Figure 4 Texas Gulf Coast landfill-plan view

of horizontal well path.

ing the installation of the horizontal well. Nearly half of that volume was generated during installation of the curved section of the well. 4.

WellPerformance 1-2 ft at distances of upto 300 The well produces upto 7 gpm of water, creating drawdown of ft radially from the midpoint of the well. The drawdown created by the horizontal well is consistent with results of analytical modeling shown in Figure 5 (distances are given in feet, and the contour interval is 0.1 ft). The model assumes steady-state flow in a homogeneous, isotropic aquifer. Less than 100 ft from the well, the potentiometric surface defines an elliptical

0

200

400

600

Figure 5 Texas Gulf Coast landfill-horizontal

800

1000

well drawdown contour map.

628

Losonsky and Beljin

trough of depression. Beyond 100 ft, potentiometric surface contours are circular. The model indicates that drawdown along the well is 2.7 ft at a pumping rate of 7 gpm. The specific capacityof the horizontal well is only 1.16 times that of a comparable vertical well with fully penetrating screen. Therefore,the primary advantageof this horizontal well is not hydraulic, but logistical. The risks involved in penetrating an existing landfill liner, this scenario, outweigh the cost whichwould be requiredfor vertical wellinstallationin difference.

B. Case 2: Industrial Chemical Plant in Southern Louisiana Contaminated clayey soils beneath industrial plantsare commonly excavated to prevent migration of contaminant5 into underlying aquifers. Low hydraulic conductivities of such soils require close spacing of vertical wells, but dense wellfields are difficult or impossible to install because of logistical obstructions such as underground utility lines and overhead steel structures. Furthermore, drilling vertical wells in the midst of active operating facilities raises concerns about worker health and safety as well as plant productivity losses. Recently, several horizontal wells have been installedas an alternative to excavation in sandy clay soils beneath petrochemical complexes alongthe Mississippi River in the industrial corridor between Baton Rouge and New Orleans, Louisiana. Two such wells were installedat a petrochemical complex located southof Baton Rouge. The wells were installed to recover dissolved ethylene dichloride and monochlorobenzene from shallow silty clay soils. 1. Hydrogeology Clay and silt dominate the subsurface to a depth of 60 ft at thesite. Eight soilstrata have been identified. The top 8-10 ft consists of dark brown to gray clay and orange, green, and brown to gray silty clay. Below that is a 10-ft-thick orange and gray to brown clayey silt, which is underlain by alternating clay and silt.The 10 ft of silt isthe target of the remediation effort. Its hydraulic conductivity is 1.7 Wday in the vicinity of the contaminant plume. Vertical permefffday. The clay ability in the clayey soils aboveandbelow the silt is approximately 3 X sediments behave likeaquitirds. The potentiometric gradient in the 10 ft of silt is 0.002 eastnortheast.

2.Well Placement The shallower of the two wells (Well A)was installed at a total vertical depth of 12 ft beneath an existing superstructure. The overheadstructures and the concrete foundation below ground level containedactive unit process equipment.The horizontal section of the well is 356 ft long, and it was installed alongthe midline of a 20-ft-wide corridor bounded by 60-ftdeep concrete and wood pilings. A vertical well was located only 5 ft away from the wellpath, demanding high placement accuracy. The deeper well (Well B) was installedat a total vertical depth of 14 ft beneath an existing concrete road, along a pipe rack and. railroad loading dock. The horizontal screenof the well is 400 ft long, and it was installed between pilingssupporting the pipe rack on one side and a freshwater drainage ditch onthe other side. 3. Well Specifications

The curved section of Well A (Figure 6) is constructed of 10%-in. -high-density polyethylene casing cementedin a curved 12!4-in. wellbore. The curve ends at a measured depth (measured along the wellbore) of 80 ft. The horizontal section is completed with 65h-in. highdensity polyethylene slotted liner (slot size is 0.02 in.) installed within an 8Ya-in. wellbore. The horizontal section ends at a measured depth of 436 ft. Blank (unslotted)highdensity polyethylene

ells

Horizontal

629

for Pollution Control

CROSS SECTION

200

100

300

400

DISTANCEIN FEET

Figure 6 Louisianachemicalplant-horizontal

well path, well A.

casing is installed inside the loyein. curved casing, rising to the wellhead. An electric submersible pump, rated at 10 gpm, was placed inside the @&in. casing at the bottom of the curved section of the well. Stainless steel wire-wrapped prepacked screen was installed inside the horizontal high-density polyethylene casing and attached the to pump to filter silt-sized formation material. Horizontal displacement between the wellhead and the beginningof the screen is approximately 70 ft. The horizontalsection remains withina 3-ft tolerance envelope, ranging in total vertical depth from 11.8 to 14.9 ft. Horizontal accuracy is within 2 ft of the planned termination point. Well B is constructed similarly to Well A (Figure 7), with total vertical depth ranging from 12.4 to 17 ft. Four undisturbed horizontal core samples were extracted during drilling-three from the shallower well and one from the deeper well. Eachcore is 5 ft long and 2 in. in diameter. Coring is accomplished by setting a hydraulic coring tool (Figure8) into the soil at the end of the

PLAN VIEW

20

'

l 100

200 DISTANCE IN FEET

Figure 7 Louisianachemicalplant-horizontal

300

well path, well B.

400

630

Losonsky and Beljin

Figure 8 Hydrauliccoringtool.(Source:EastmanChristensenEnvironmentalSystems.)

wellbore with a moderate amount of pushdown force to isolate the inner barrel from the drilling fluid and contaminated soiland groundwater. Pressure is hydraulically appliedto create a constant load on the outer tubeof the coring device.An accumulator located on the drillingrig is plumbed in parallel with the drill string circulation path to maintain a constant punch force. Pressure is raised to the calculated punch release force, which breaks shear pinsand accelerates the inner tube into the formation. Pressure is maintained in the system to hold the outer tube against the formation to prevent the drilling medium from coming into contact with the sample as it is pulled back into the outer barrel. The core barrel is then retrieved to the surface for recovery of the undisturbed sample. The sample is contained within a disposable plastic liner. The core barrel requires a radiusof curvature of at least 100 ft to pass through a 5Vi-in. borehole. The total groundwater recovery of the two wells is 23 gpm, which is comparable to over 50 vertical wells in the same water-bearing zone. Vertical wellsas far as 70 ft away from the horizontal wells exhibit drawdown causedby pumping from the horizontal wells. 4.

AnalyticalModeling Results of analytical modeling of capture zones that develop in response to groundwater recovery from the two horizontal wells are shown in Figure 9. Recovery of the contaminant tank car unloading areas, process sump, and glyoxal unit, shown in plume in the vicinity of the two 400-ft-long horizontal wells placed within the waterFigure 9, canbeachievedwith bearing silt zone.The southern well would be placed 30 ft north of the glyoxal unit,as shown

-

horizontal screen

capture zones:

400

FT

0 4years 8 years

Figure 9 Louisianachemicalplant-groundwatercapturezones.

Horizontal Control Wellsfor Pollution

631

in Figure 9. Shaded areas indicate 4-year and 8-year capture zones of the two wells. The capture zones are calculated using a groundwater code called PAT [19], which assumes a homogeneous, isotropic aquifer. The assumed aquifer parameters, input data, and the results are listed in Table 1. Results of analytical modeling of aquifer response to groundwater recovery fromthe two horizontal wells are presented in Figure 10 (distances are in feet and the contour interval is 1 ft). The shaded region represents the chlorobenzene (MCB) plume. Groundwater recovery rates are calculated using W E L L , an analytical model [20]. Aquifer response tothe two proposed horizontal recovery wellsis estimated by simulating each horizontal well by a series of closely spaced vertical wells with total pumping rate equal to that calculatedby W E L L . This method provides a good estimate of drawdown curves generated by more complex, numerical methods, and hence it is a good tool for predicting actual drawdown contours.The model used for this simulation is a modified version of THWELLS [21], which solves the Theis equation for multiple wells and accounts for unconfined aquifer conditions. The model assumes non-steady-state conditionsin a homogeneous, isotropic aquifer with infinite areal extent and with a hydraulic gradient of 0.002. Time elapsed since the onset of pumping is 100 days. The analytical models suggest thatthe two horizontal wells installed at the plant could capturethe plume during an 8-10-year period if the total pumpingrate were only 5 gpm. Actual recovery rates are over four times that amount, so closure levels of contaminant concentrations may be achieved sooner.

VI. CONCLUSIONS Horizontal wells offer significant advantages over vertical wells in environmental remediation and protection in many hydrogeological scenarios. These include thin and/or low-permeability aquifers, where many closely spaced vertical wells are replaced by a single horizontal well. Current high installation cost of a horizontal well compared to a vertical well is offset by operation and maintenance cost savings. New developments in horizontaldrilling technology will further reduce the cost of installing horizontal wells, and subsurface pollution control with horizontal wells should become as common as vertical wells. Table 1 Input data Hydraulic conductivity Storage coefficient Flow rate of horizontal well Time of pumping (capture zones) Time of pumping (drawdown contours) Aquifer thickness Horizontal well length Wellbore radius Radius of influence of horizontal well Horizontal well eccentricity Regional flow gradient Results Half the major axis of drainage ellipse Specific capacity, horizontal well Drawdown in absence of flow gradient

1.67 ft/day 0.05 2.5 gpm

4and8yr 100 days 8 ft

mft 0.7 ft 330 ft 2 ft

0.0021

361.9 ft 68.34 gpdft2 7.04 ft

632

Losonsky and Beljin

1000

500

Figure 10 Louisianachemicalplant-aquiferresponsetohorizontalwells.Potentiometricsurface contours interval is 1 ft. Distance scale isin ft. Initial potentiometric surface elevation is 10 ft at origin (SW comer).

REFERENCES 1. Chin, W. C., Borehole Flow Modeling in Horizontal, Deviatedand Vertical Wells,Gulf Publ. Co.,

Houston, Tex., 1992. 2. Babu, D. K., and Odeh, A. S., Productivity of a horizontal well, Paper SPE 18298, SPE Reservoir Evaluation, November 1989, pp. 417-421. SPE 3. Giger, F. M., Horizontalwellsproductiontechniquesinheterogeneousreservoirs,Paper 13710, presented at the 1985 SPE Middle East Oil Technical Conference and Exhibition, Bahrain, Mar.11-14,1985. 4. Goode, F! A., and Kuchuck. F. J., Inflow performance of horizontal wells, SPE Reservoir Eng.. August 1991, pp. 319-323. 5. Hantush, M. S., and Papadopulos, I.S., Flow of ground waterto collector wells,J. Hydraul. Div., Proc ACSE, 1962, pp. 221-244. 6. Dickinson, W., Dickinson. R. W., Mote, F!, and Nelson, J., Horizontal radials for hazardous waste remediation, NSWMA Waste Tech '87 Conference, San Francisco, Calif., Oct. 26-27, 1987. 7. Kaback, D., Looney, B., Corey, J., Wright, L., and Stele, J., Horizontal wells for in-situ remediation of groundwater and soils, NWWA Outdoor Action Conf., Orlando,FL., May 22-25, 1989. 8. Looney, B., Kaback, D. S., and Corey, J. C., Field demonstration of environmental restoration using horizontal wells, Third Forum on Innovative Hazardous Waste Treatment Technologies, June 11-13, 1991, Dallas, Tex. 9. Karlsson, H., and Bitto,R., New horizontal wellbore system for monitor and remedial wells, Proc. Superfund '90, pp. 357-362. 10. Langseth, D. E., Hydraulic performance of horizontal wells, Superfund '90, Washington, D.C., NOV. 26-28,1990.

Horizontal Wellsfor Pollution Control

633

11. Fritz, R. D., Horn, M. K., and Joshi, S. D., Geological Aspects of Horizontal Drilling, ASFG Course Notes 33, 1991. 12. Borisov, J. P., Oil Production Using Horizontal and Multiple Deviation Wells, Nedra,Moscow, 1964. 13. Joshi, S. D., Augmentation of well productivity with slant and horizontal wells, JPT, June 1988, pp.729-739. Trans. 14. van Everdingen, A. F., The skin effect and its influence on the productive capacity of a well, AIME, 198, (1953). 15. Renard, G . I., and Dupuy, J. M., Influence of formation damage on the flow efficiency of horizontal wells, Presented at the Formation Damage Control Symposium, Lafayette, La., Feb. 22-23, 1990. 16. Joshi, S. D., Horizontal Well Technology, Pen Well Books, 'hlsa, Okla., 1991. 17. Gringarten, A. C., Ramey, H. J., Jr., and Raghavan, R., Unsteady-state pressure distributions created by a well with a single infinite-conductivity vertical fracture,SPE J . , August 1974. pp. 347360. 18. Speake, R. C., 'Itpjan, M.. and Wang, Z. Z., Modeling the performanceof a horizontal groundwater recovery well, Proc. Outdoor Action Conf., Las Vegas. Nev., Natl. Ground Water Assoc., 1991, pp. 399-415. Model, Kassel, Stutt19. Kinzelbach, W., and Rausch, R.,PAT: Pathlines and Traveltimes Groundwater gart, Germany, 1990. 20. Beljin, M. S . , and Losonsky, G., HWELL a horizontal well model, Solving Ground WaterProblems with Models, NGWNIGWMC Conference, Dallas, Texas, 1992, pp. 45-54. 21. van der Heijde, P. K. M., THWELLS, International Ground Water Modeling Center, Holcomb Research Institute, Indianapolis, Ind., 1991.

This Page Intentionally Left Blank

Part V

INDUSTRY-SPECIFIC POLLUTION CONTROL

This Page Intentionally Left Blank

31

Pollution Control and Waste Minimization in Military Facilities

Merrit l? Drucker Army Management StaffCollege

Fort Belvoir, Virginia

1.

INTRODUCTION

Military facilities include a wide range of bases, installations, maintenance facilities, manufacturing plants, factories, as well as installations in foreign countries and temporary facilities established in support of combat operations. Some military facilities are similar to cities and towns; others are highly specialized facilities producing unique munitions;others resemble office complexes. Some facilities are rented. Military facilities present a variety of problems for the environmental engineer. First, military bases are generally organized to support a wartime or contingency mission. This means that environmental considerations may not be as prominent as in municipalities and private industry, although this attitude is changing rapidly. Second, because of their organization and function, the waste stream from military facilities is likely to be more complex and, in some cases, more toxic or hazardous than the waste stream from a comparable civilian facility. Finally, military facilities are operated undera different budgeting and financial system than municipalities and private industry. Environmental engineers employedby military facilities ought of the culture, or sociology, to be aware of these differences and should make themselves aware of the military facility where theyare working, the military chainof command, andthe peacetime and wartime missions of the facility. This is not merely “nice-to-know” information. It is the essential first step to designing a pollution control and waste minimization strategy for an installation.

II. MATERIALS

AND PROCESS ANALYSIS

The first step in controlling pollution and minimizing waste from a military facility is to find out how much material and what type of material is being brought into the facility. This information is not easily obtained, and few if any installations can tell for certain what all their material imputs are.However, even imprecise estimates of the amountof material brought into

637

638

Drucker

the facility can helpdetermine start points for control strategies. Often, after merely lookingat what the facility is ordering, substantial reductions or effective substitutions can be made. A centralized computer system that could identifythe total amount of material coming into the installation would be a useful management tool. Unusually large orders or inputs of toxic or hazardous materials are good initial places to look for reductions. vpically, pollution control and waste minimizationhave been thought of as logistical or engineering functions.A more effective approachwould require managers at all organizational levels to know what all of their inputs are (materials, energy, utilities) and to examine their first reducing the amount of inputs caneliminate much input stream for reductions. Focusing on waste. Often smaller quantities can be ordered, less toxic material substituted, or some items eliminated entirely. Working with suppliers can be especially helpful, as suppliers of hazardous materials are under intense pressure to reduce the hazard level of theirproducts, to reduce packaging, and to accept empty containers for reuseor recycling. The second step is to examine the various processes at the facility to find economies and efficiencies. Most industrial and materials processes at military facilities were not designed with pollution control or minimization in mind until relatively recently. All processes mustbe continuously evaluatedso they can be made to consume fewer inputs of materials, energy, and hazardous materials. Process improvement may entail improved housekeeping, better maintenance of equipment, training of employees, replacementof older equipment with newer, more efficient models, or completely new ways of performing required industrial tasks. Engineers ought to be especially active in asking employees who actually runthe processes how savings can be achieved.Often, the employee closest to the process can providethe best suggestions on ways to make the process less polluting, less energy-intensive, or less expensive. Nor should relatively inexpensive, simple solutions be overlooked. Actions such as providing plastic covers for metal drums can prevent contamination from rainwater, thus protecting products stored outside and avoiding large disposal fees. Since many military installations share similar processes andfacilities, it is often possible to obtain information fromother facilities on how to improve operations. Informationsharing and exchange are vital elements in any plan to reduce waste or minimize pollution. Processes that have been implementedat one installation can often be implemented at another with little modification. The United States Department of Defence is undertaking a department-wide effort to reduce the amount of hazardous waste generated. Specialized agencies throughoutthe be called department serve as consultantsor centers of expertise for waste reduction and should on for assistance. Environmental engineers assigned to military facilities should aggressively pursue efforts to replace toxic material with less toxic substances. It is especially importantfor military facilities to plan and budgetfor pollution control and waste minimization. These facilities do not have the flexibility of private industry, and procurement times tend to be longer. Installation and organizational plans, both long-term and short-term, as well as budgets, must include planning and funding for new equipment and facilities. The military facility’s higher headquarters must be kept fully informed as to the status of minimization efforts and the need for capital investment. The third, and final, step is to control and properly dispose of the wastes that are generated. This requires disposal systems and options, and education of the installation population. The primary type of disposal of wastes from militaryfacilities should be recycling. Almost all military facilities have recycling systems, and thereis an accounting system in place to return profits from the recycling program to the facility. However, recycling could be greatly expanded, resulting in significant cost savings as disposal fees for waste increase. All wastes should be evaluatedfor recycling before disposal, and consideration should be given to giving away waste material such as scrap lumber or metal rather than disposing of it as pure waste.

cilities Pollution Military Control at

639

Recycling programs on military facilities shouldbe fully integrated with the local community. The military recycling program should be made available to the local community, and those who live and work on the installation should be encouraged to participate in community recycling programs. It is essential that recycling programs be fully coordinated and integrated into programs sponsored by other military facilities and the local community, inboth the United States and in foreign countries. Current and pending legislation, primarilyat the state level, will require that certain percentages of the waste stream be recycled. It is impossible to determine if the military facility is meeting this goal unless the waste managers have accurate information on both the amount of material coming into the installation and the amount of waste being generated. This reinforces the need for automated materials tracking systems. There is a considerable amount of technical information available on pollution control and waste minimization, much of which is readily applicable to military facilities. Employees and managers at all organizational levels should be required to be on the lookout for alternative processes and methods and ways to improve existing processes. It is critically important that waste minimization and pollution controlbe seen as a primary responsibility of those who actually work with the systems and materials and not the responsibility of the environmental enbe seen as a line, and not a staff, gineer.Pollutioncontrolandwasteminimizationmust function; these requirements shouldbe part of efficiency reports and performance appraisals. While supply personnelare in a critical position, leaders and managers must specify that they want nontoxic materials whenever possible, and must specify that they prefer products made from recycled, postconsumer waste. One extremely valuable, yet sometimes overlooked, resource available to military facilities is audit capability. Most military facility commanders have professional auditors available to support them. These auditors can examine all phases of the pollution control, waste minimization, and recycling program and can assist in achieving considerable savings and efficiencies. Environmental engineers working on military facilities are encouraged to seek auditing support. The payoff can be enormous.

111.

LEGALREQUIREMENTS

In the United States, federal facilities must comply with local, state, and federal environmental, public health, and safety laws. Although there are some minor exceptions, the general thrust of environmental legislation is toward more and stricter regulation; military facilities are no exception. For the environmental professional, a detailed knowledge of the law is essential. Law and technology combine to form an interactive process.As technology improves and allows us to detect and control more pollutants, laws are amended or written to make the technological ability a requirement.New laws, on the other hand, often contain requirements that serve as a spur or stimulus to technological innovation. Legal requirements are important and must be complied with, but they are only a small part of the pollution and waste control effort at military facilities. Further, focusing only on legal requirements may mask or distract attention from the possibility for considerable waste minimization and economic savings resulting from a more systemic, global view of pollution control and waste minimization. Environmental engineers must be looking for additional ways to prevent or minimize pollution to levels well below those required by law. All U.S. military commands have legal staff support available to them. Environmental engiearly very in the neers are well advisedto bring the legal staff into all environmental engineering planning stage. Continuous coordination and consultation with the legal staff during all phases of base operations will help eliminate environmental problems before they become legal problems.

640

Drucker

Military commanders and federal employees can be held both civilly and criminally liable for violating environmental laws. This makes compliance with the law quite important for mil itary facilities and increases the importanceof close consultation with supporting legal staffs. Legal problems with state or federal regulatory agencies have the potential to delay projects or military missions. Military commanders are extremely concerned about the possibility of delay on military missions andwant to be kept fully informed on possible problems or delays. Engineers working on military facilities should ensure that they have open communications channels with the installation commander.

IV. MOBILIZATIONANDDEPLOYMENT A unique military environmental problem is the amount of waste generated by military mobi-

lizations and deployments. During times of national emergency or war, the population of an of 2 or 3 days. Industrial and production activities, installation can doubleor triple in the space such as maintenance and ammunition manufacturing, cango from very low rates to very high rates in a matter of days. Surging troop populations preparing for overseas deployment dema all manner of goods and services, resulting in the generation of large amounts of waste of all types in a short time period. It is difficult to minimize waste during mobilizations, since they are short-duration, resource-intensive activities. Good planning, however, can prevent pollution and recover such waste for recycling. It is important that military facilities be planned and operated in such a way that all environmental laws and practices followed in peacetime can be followed during rapid wartime expansion. In most cases, this canbe done if the installation or facility infrastructure has been properly constructed and if adequate environmental planning has been accomplished. Typical, specific problems have emerged from past mobilizations and deployments; theseare predictable problems that can be prevented by good planning. Perhaps the single biggest environmental problem during mobilizations is wastewater resulting from increased troop populations. Large flows from troop housing areas, mess halls, vehicle washing, and industrial operations can quickly overwhelm wastewater treatment sysbe designed to accommodate tems. To prevent this, military wastewater treatment systems must the maximum or peak troop population during mobilization. This may seem wasteful or too expensive; however, the costs of environmental cleanup, fines, and health problems may well exceed the initial cost of larger capacity wastewater systems. Vehicle painting operations present another problem. Due to the toxicity of military cambe painted in special booths intended to protect worker safety and ouflage paints, vehicles must prevent the release of toxic fumes to the atmosphere. While a typical facility may haveorthree four booths, a wartime emergencymay require the rapid painting of hundreds of military vehicles. Unless sufficient booths to accommodate wartime surges have been constructed, the vehicles may be painted in the open or workers not protected. Environmental planners must ensure that adequate facilitiesare constructed or that contract arrangementsare made to have the vehicles painted in appropriate facilities. A more effective long-term solution wouldbe to design a paint system that does not have toxic components. Painting and industrial coating p cesses for military equipment (ammunition, weapons, naval vessels, aircraft electronic equipment) tend to present the same types of problemsas vehicle painting, particularly when there is a surge in production. These “pollution surges” must be planned for in all phases of the engineering process. Increased populations will generate substantially more quantities of solid waste, and increased industrial operations will result in the generationof increased quantities of hazardous waste. These additional quantities can usually be disposed of by expanding existing disposal

ntrol Pollution

at Military Facilities

641

contracts for waste disposal and recycling. Should planning estimates reveal that likely mobilization surges will overwhelm existing capacity, arrangements shouldbe made to have “contingency” contractsprepared for immediate implementation when needed. The to keys effective waste disposal are good recycling programs, good contracting, and thoughtful planning. One potentially significant pollution problem resulting from increased military training is the pollutionof streams and waterways with soil or other material eroded from military training areas. Large military tracked vehicles, such as tanks and armored personnel carriers, can severely damage the land, making erosion possible. Careful training area management and appropriate revegetation and repair of damaged land must be planned for during mobilizations.

V. WARTIMEOPERATIONS A time of crisis or war does notmean that environmental standards are discarded. Rather, peacetime pollution control and waste minimization should continue in domestic facilities as well as in the theaterof operations. Thiswill present somedifficulty; however, intelligent prior planning and appropriate discipline in the theater of operations can prevent large amounts of environmental destruction. First, logistic support plans should clearly specify what will be done with any hazardous waste as well as solid waste generated in the war zone.Much of this material will be generated at fixed, semipermanent, or permanent facilities at rear areas of combat zones, so intense enemy interferencewith many support activities will be limited.Waste maybedisposedof through the system of the host country or retrograded back to the United States for disposal. or Wastes may be pretreated to reduce bulkor hazard. It is essential that waste not be dumped abandoned, even if that is standard practicein the country our forcesare in. For legal, political, ecological, and moral reasons, wastes shouldbe disposed of, whenever possible, in accordance with existing U.S. standards. Second, nonhazardous solid waste should,to the greatest extent possible, be made available to local recyclers. Open burning and burial should be avoided unless otherno options exist be constructed and the waste becomes a health problem.If possible, engineered landfills should and the location and closure plans made available to the appropriate national environmental agency. Excess supplies or equipment should be returned for reuse or abandoned, butnot burned or buried. Third, battlefield cleanup must be plannedfor and conducted as soon as possible after the cessation of combat. Normally, troop units can remove and dispose ofthe vast majority of battlefield debris. Specially trained and equipped units willbe required to recover explosives, unexploded projectiles, mines, and chemical weapons. Someitems, such as vehicles destroyedby depleted uranium rounds, may be returned to the United States for secure disposal.

VI. TRAINING An effective training program is essential if military facilitiesare to minimize waste and control pollution. Military facilities usually have a diverse work force of military and civilian personnel; if the facilityis in another country there will most likely be employees from the host nation. All members of the facility will need to be informed on the procedures for recycling, waste disposal, and pollution control. The best, most expensive, most technologically sophisticated pollution control or recycling system will be of little use if it is not understood by those who must use it. There are a variety of training systems available. The environmental engineer or recycling coordinatorcanprovideinstruction or information to unitenvironmentalrepresentatives

642

Drucker

through a variety of installation councils and meetings. The unit environmental representatives can then train the members of their organization and serve as a means of communicating new information to their organizations. Many facilities use installation newspapers or periodicals to communicate information be a very useful adjunct to a formal training quickly to all members of the installation. This can PWPam. Some trainingis quite technical and is requiredby federal law. This training mustbe conducted by qualified instructors,and careful records maintainedto ensure that reporting requirementsaremet.Militaryfacilitiesinthesamegeneralgeographicareashouldconsider consolidating this type of training to cut training expenses. All training should be very high quality, and standards kept high. Training must be repeated periodically, especially for new members of the installation team. Trainingat facilities in foreign countries must be conducted in the language of the employees. This is especially important, as many logistics functions for U.S. forces overseas are performed by foreign nationals.

32 Waste Reduction Strategies for Small Businesses

Dan A. Philips Pensacola Junior College Pensacola, Florida

1.

INTRODUCTION

Some historians believe that the 1990s will be rememberedas theenvironmental decade.Even small businesses cannot remain in operation today without considering the repercussions of each of their actions on the natural environment, the regulatory environment, and the public perception of the environment. The late 1960sand early 1970sbrought aboutmany changes. Hair became longer and skirts shorter. For any number of reasons, the public consciousness was raised along with hemlines. One of the major controversiesto arise concerned the role of business in maintaining the ecology, includingthe detrimental effects that many industrial processeshave on the integrity of the environment. By exerting a great deal of political pressure, the public forced government into the position of “protector of the environment.” The government lived up to its newly established role by developing a myriad of bureaucratic agencies, committees, subcommittees, and review boards to address environmental issues. The late 1970s brought about concrete, tangible legislation on waste treatment and pollution prevention. The basic legislation was merely a small snowball balancing precariously on the edge of a ski slope. Over the years to follow, regulations grew until small businesses were facing a veritable avalanche of rules, acts, regulations, and amendments. EPA, DOT, CERCLA, SARA, RCRA, and OSHA are not charitable foundations requesting tax-deductible donations.They are regulatory agencies and regulatory acts. The cumulative results of their efforts comprise an extensive list of emission limits, exposure limits, disposal limits, etc. Failure to comply with their standards can result in tremendous fines. Onefine from any of these agencies could translate into bankruptcy for a small business. Today, small business operators may be confronted by over 10,OOO pages of regulations, and the number continues to grow. It is estimated that each new page of federal regulations costs businesses about $10 million. Unfortunately, over 90% of the money, efforts, and regulations focus on end-of-the-pipe wastes. Little emphasis is placed on the nature of the process 443

644

Philips

being regulated, i.e., the beginning of the pipe. Obviously, what goes in directly affects what comes out. Most federal andstate environmental funds are spent on pollution control and remediation. $70 billion, with two-thirds National spending on pollutionand waste control is approximately of that amount being suppliedby industry. That represents a large sumof money that could be used for investments, expansions, and capital outlay if the waste stream could be reduced or eliminated [13. To make matters worse, the land available for disposal is quickly disappearing, and government regulations on land burial of waste are becoming more stringent. Once a waste is created, the creator is responsible forever. Long-term waste liability may be 200 times as costly as the initial disposal. Obviously, this can drastically affect the cost of doing business. Approximately 300 million tons of hazardous waste is buried annually in the United States. Over 7 billion tons of solid waste is generated each year by 72,000 businesses. Such estimates provide a driving forceto regulate what is happeningto these wastes. The reduction of waste has been a regulatory requirement for several years, providing a strong motivationto comply; however, this has traditionally not been sufficient to ensure full compliance with existing regulations. Businesses generally consider wastes as end products requiring disposal and not as expenses that need to be reduced. Traditionally, most businesses have preferred to use waste management techniques such as treatment, incineration, or land burial. Such techniques are applied strictly to the end of the process[2]. These old techniques pose environmental risks, require meeting costly regulatory compliance standards, and increase future liabilities. Disposal costs will continueto increase as regulationsconcerningwastebecomemorerestrictive.HalfofallU.S.landfillsarealready nearing or exceeding their designed capacity. Some government agenciesare already stressing a zero-discharge philosophy. In meeting these new regulations, companies will need to place a strong emphasis on source reduction and on-site recycling while shifting away from off-site recycling or disposal [3]. In 1976, the Resource Conservation and RecoveryAct (RCRA) established a “cradle-tograve” responsibility for the generator of a hazardous waste. Theneed for better waste man1984, agement was increased by the Hazardous and Solid Waste Amendments (HSWA) of which applied even more restrictions. The intent of these laws is to prevent environmental degradation and to protect natural resources. This philosophy is clearly stated in the Pollution Prevention Act of 1990, which reads: The Congress hereby declares it to be the national policy of the United States that, polbe prelution be prevented or reduced at the source wherever feasible; pollution that cannot vented should be recycled in an environmentally safe manner whenever feasible; pollution that cannotbe prevented or recycled should be treated in an environmentally safe manner; and disposal or release into the environment shouldbe employed only as a last result. The Comprehensive Environmental Response, Compensation and Liability Act (CERCLA,

or Superfund) was passed in1980, raising the stakes in the waste disposal game considerably. CERCLA is responsible for repairing environmental damage done because of past and present improper management of wastes. In 1986, the Superfund Amendments Reauthorization Act (SARA) was passed. SARA assigned liabilityto the responsibleparties and provided a means by which to hold the parties financially responsible.The principle of jointand several liability, upheld in the U.S. Supreme Court decisionU.S. v Stringfellow, makes all parties whose waste is contained at a Superfund site equally responsible for cleanup costs[4]. The reduction of waste has been a regulatory requirement for several years. However, there was no common description of what constituted waste reduction, there were few data

Strategiesfor Small Businesses

645

on the extent of industrial waste reduction, and data that were collected were often measured incorrectly [5]. Under the Clinton-Gore administration, waste reduction will probably receive increased governmental emphasis. In Vice President Gore’s book Earth In The Balance, he points out that it will be increasingly important for all companies to incorporate standards of environmental responsibility into their entire operation [6]. By using materials more efficiently, industry can reduce waste. This in turn will allow greater protectionof human health and the environment while ensuring regulatory compliance. Other factors are beginning to have a considerable impacton how waste is handled. Of these, probably the most overwhelming incentive is liability. The generator of a hazardous waste isultimately and solely responsible for what happens or has happened to that waste in the past, present, and future. The imminent hazard provisions of SARA and RCRA affect waste managed on-site as well as waste shipped off-site and can close down a business as well as imposing heavy fines and penalties. EPA can assess triple damages (three times the amount of the cleanup) and assess a lien on a business property to collect [l]. There are many compelling reasonsto minimize waste; Table1 summarizes someof them. RCRA regulations require that generators of hazardous waste “havea program in place to reduce thevolume and toxicity of waste generated tothe extent that is economically practical.” A waste reduction program isan organized, comprehensive, and continual effort to systematically reduce the generation of waste. It should become part of a company’s everydayoperating policy. While the main goal is to reduce or eliminate waste, it will also bring about an improvement in production efficiency. Under the more stringent legislation being considered, waste reduction programs would include all environmental pollutants regardless of whether their disposal is permitted. This includes solid waste, wastewater, hazardous waste, andair pollutants. Also being consideredare bills restricting waste reduction options by considering only source reduction or on-site recyor any cling techniques. The strictest legislative packages would not consider off-site recycling treatment separate from the production process within the scope of waste reduction. Although these proposals have developed support fromcertain congressional quarters and are supported Table 1 WasteMinimizationIncentives ~

~~

~~~~~~

~

Economics Landfill disposal cost increases Costly treatment technologies Savings in raw material and manufacturing costs Tax incentives for pollution prevention Regulations Certification of a waste minimization program on the hazardous waste manifest Biennial waste minimization program reporting Land disposal restrictions and bans Increasing permitting requirements for waste handling, and treatment Liability Potential reduction in generator liability for environmental problems at both on-site and off-site treatment, storage, and disposal facilities Potential reduction in liability for worker safety Public image and environmental concern Improved image in the community Concern for improving the environment

446

Philips

by the Office of Technology Assessment, such restrictive legislation is not favorable to either manufacturers or the developing recycling industry. Strong efforts by generators to reduce waste generation are encouragedby the EPA in order to avoid more restrictive regulation[ 5 ] .

II. BARRIERS TO WASTE REDUCTION Although economic factors often work in favor of pollution prevention approaches, there is always some resistanceto change of any kind. Waste reduction projects can reduce operating costs and improve environmental compliance, but they frequently bring out conflicts between different groups within acompany. Economic barriers to pollution prevention include the following: Inaccurate market signals. Sometimes the immediate cost of releasing toxic substances is less than the cost of implementing a pollution prevention project. This occurs when the longrange cost of the release is not included in the calculations. Incomplete accounting. Indirect benefits, such as lower future liabilities and improved public image, are not commonly considered in a financial analysis. Fear of lowering quality. This is very common in situations where unused feed materials are recovered from a waste and recycled back into the process. If not done properly, such processes may affect quality. Workerfear ofjob loss. If employees or labor groups look upon pollution prevention as a threat to their jobs, these concerns may pose a barrier to new processes. Fear of losing market share. Surveys suggest that a significant barrier to pollution prevention is reluctance to tamper with proven processes for fear of adverse effects on product quality [7]. Attitude-related barriers must be overcome before any pollution prevention program is tried, or it is destined to fail. A prevailing attitude is to maintain the status quo and avoid the unknown. There is also a fear that a new program may not work as advertised. Without the commitment of everyone involved, a pollution prevention program is doomed to failure. The Case of the Proverbial Potato' In 1986, a coatings manufacturer in Hamburg, Germany hired a bright young process chemist for one of its factories.The young chemist was involved in preparing a high quality varnish to be used as a sealer on musical instruments. While reviewing the formula, he was surprised to see that it called for a potatoto be added after each new ingredient was mixed into the batch. Curiosity got the best of him. He asked head the chemist what physicalor chemical properties the potato addedto the batch. The head chemist abruptly told the young man that potatoeshad been added to the varnish batches for over50 years and reprimanded him for his insolence. Not easily discouraged, the young chemist began asking other chemists in the lab the about addition of potatoes. Unable to provide an answer, they too became curious about the use of potatoes in the varnish batches. One of the older employees on the staff suggested that the young man contact a retired chemist who lived in a nearby village. The old chemist had worked at the factory for many years before his retirement. He explained to the inquisitive young man that during the war years, thermometers had beem very hard to acquire. Potatoes had been used to tell when the batchhad reached the proper temperature for adding the next ingredient. When this information was conveyed to management, the use of potatoes was halted at the factory and thermometers were substituted. 'Adapted from Ref. 8.

Strategiesfor Small Businesses

647

This story exemplifies our reluctance to change. It also demonstrates that one concerned individual, through persistence, can make a difference. Of course, a potato is not considered a hazardous or toxic substance, but it is symbolic. Often, less toxic and equally effective substitutes are available for a process, but they are not used because “this way is we’ve the always done it.” When a person does not fully understand the nature of a proposed option and its impact, a common attitude is that “it just won’t work.” Attitudinal changes are more likely to occur when people are presented with stimuli that encourage self-persuasion. Techniques of persuasion used by others may force the confused individual into a defensive posture. When this happens, creative options may be dropped before they can be evaluated. Oneway to avoid this is to use idea-generating sessions (brainstorming), encouraging participants to propose a large number of options without regardto cost, technology, or barriers. The more impractical ideas will be dropped anyway, but previously undiscovered problem-solving solutions may emerge. Many business managers are surprised to discover that waste reduction does not always necessitate large-scale investments. Significant savings have resulted from such simple, commonsense improvementsas better housekeeping, preventive maintenance, balancing inputs and outputs (inventory control), and minor process changes. Waste minimization shouldbe viewed as an investment in the company’s future. It begins with a conceptas simple as “Waste-If you don’t produceit, you won’t haveto dispose of it.” At first, this statement seems oversimplified and naive; however, upon further review, many managers have become impressed with how often this principle can be applied. Waste minimization is a relatively low-cost but effective procedure for reducing liability and increasing profits. Often, waste minimization processes do not require high technologyor expensive equipment. In many cases, very simple shop practices and procedures, combined with low-cost equipment, can greatly reduce the amount of hazardous waste generated. In addition, 42 states presently offer tax credits or exemptions for companies purchasing pollution prevention equipment [9].

111.

LEVELS OF WASTE MINIMIZATION

There are three major levels in the waste minimization hierarchy. These levels canbe viewed as steps to improving production processes and saving money. The levels, in orderof priority, are source reduction, recycling, and treatment.

A. Source Reduction Source reduction involves the minimization or elimination of wastes at their source. It often includes changesin procedures rather than in technology or machinery, makingit simple, easy, and cost-effective. Because littleor no wasteis generated, source reduction alleviates the problems associated with handling and disposing of wastes. It is therefore the most desirable option in the waste reduction hierarchy.

1. ImprovedOperatingPractices Operatingpracticesincludeprocedural,administrative,andinstitutionalmeasuresthata company can useto reduce waste. Many of these measures have been used for decades by successful businesses as efficiency improvements and sound management practices. They can ofGood operating ten be implemented with little cost and have a high return on investment. practices include Management and personnel involvement Material handling and inventory practices

Philips Loss prevention Waste segregation Complete accounting practices Production scheduling Management and personnel involvement includes employee training, incentives and bonuses, and other programsthat encourage employees to conscientiouslystrive to reduce waste. Material handling and inventory practices include programsto reduce loss of input materials due to improper handling, expired shelf life, or improper storage conditions. Loss prevention minimizes wastes by such means as avoiding leaks and spills from equipment and preventing evaporation by keeping solvent containers closed tightly. Waste segregation practices reduce the volume of hazardous wastesby preventing the mixing of hazardous and nonhazardous wastes. Complete accounting practices involve allocating waste treatment and disposal costs directly to the departments or groups that generate wasterather than chargingthese costs to general company overhead accounts. In doing so, the departments or groups that generate the waste become more awareof their treatment and disposalpractices, thus providing a financial incentive to minimize waste. By judicious production scheduling of batch runs, the frequency of equipment cleaning and its resulting waste can be reduced. This technique is particularly useful in printing and painting operations. Example: Good OperatingPractices-Management. A printing companyreducedpaper plates, and presseswere waste by 30% simply by implementing process control measures. Inks, kept at peak efficiency, reducing the need for reprinting. Example: Good Operating Practices-Material Handling and Inventory. A consumer product company in California adopted a corporate policy to minimize the generation of hazardous waste. In order to implement the policy, the company mobilized quality circles made upof employees representingareas within the plant that generated hazardous wastes. The company experienced a 75% reduction in the amount of wastes generated by instituting proper maintenance procedures suggestedby the qualitycircle teams. Since the team members were also line supervisors and operators, they made certain that procedures were followed. Process changes provide the best opportunities for reducing waste. One way to alter the process is to use different raw materials. It may be possible to choosea material that allowsfor a greater percentage of product in the end. A material that is nontoxic, evenif it does become a waste, will not pose as great a danger to the environment. Employees’ assignments canbe planned carefully in advance so they are done efficiently and thoroughly. Putting excessive pressure on workers to complete jobs quickly may mean that the job will not bedone properly thefirst time because itwas rushed. Consequently, moretime and materials will be required to correct the mistakes. Another way to reduce waste is through better housekeeping. Keeping an operation clean and orderly improves safety, delays deterioration of equipment, reduces breakdowns, and increases efficiency. It will also assist in discovering leaks and spills when they first occur. Improving inventory control has been found to be a big step in reducing waste. Records should keep track of which materials are being used and howquickly they are being used. Buying large quantities of materials just because the unit cost is lower will not save money if a majority of the material sits on the shelf anddries out or becomes unusable because of chemical decomposition.

Businesses Strategiesfor Small

649

To avoid exceeding an expiration date, it is also useful to rotate stock periodically. The freshest material cango to the back of the shelf, with the oldest moved to the front. Ordering environmentally safe products greatly reduces disposal costs should a material be stored past its useful shelf life. Preventive maintenance will also contribute greatly to waste reduction.If a tool or piece of equipment is maintained properly, it will last longer and provide a more productive of the use process for which it was designed. are more procedural than technological. Many helpful changes that foster waste reduction Improved operating proceduresdo not have to be elaborateor expensive; they just need to get the job done. 2.

TechnologyChanges Technology changes are oriented toward industrial process and equipment modifications that will reduce waste, primarily in a production setting. Technology changes may range from minor improvements that can be implemented in a matter of days at low cost to new processes with startup costs that require large capital outlays. These changes include Alterations in the production process Changes in equipment, layout, or piping Use of automation Changes in process operating conditions, suchas flow rates, temperatures, pressures,and residence times Example: Technology Changes-Equipment, Piping, or Layout. In many dry cleaning facilities, existing equipment is retrofitted with ventilation and vapor recovery machinery. It is costeffective to invest in modern equipment that can practically eliminate emissions and human contact with solvents. Example: Technology Changes-Process. Many companies are installing equipment to recycle wastewater by separating the hazardous components from the water, which can then be reused. Example:TechnologyChanges-Process. A manufacturer of fabricated metal products foran alkaline chemical bath priorto using the wire in merly cleaned nickel and titanium wire in the product. In 1986, the company began to experiment with a mechanical abrasive system. The wire was passed through the system, which uses silk and carbide pads and pressure to brighten the metal.The system worked but required passing the wire through the unit twice for complete cleaning. In 1987, the company bought a second abrasive unit and installed it in series with the first unit. This allowed the company to completely eliminate the need for the chemical cleaning bath. 3. InputMaterialChanges

Input material changes accomplish waste reduction by reducing or eliminating hazardous mabe made to avoid terials that enter a production process. Also, changes in input materials can generation of hazardous wastes within a production process. Input material changes include material purification and material substitution. Example:InputMaterialChanges. An electronics manufacturing facility originally cleaned printed circuit boards with organic solvents. The company found that by switching from an organic-based cleaning system to a water-based system, the same operating conditions and workloads could be maintained and the aqueous system cleaned six times as effectively. Thisresulted in a lower product reject rate and eliminated theneed for disposal of a hazardous waste.

650

Philips

ProductImprovements Product improvements are performed by the manufacturer of a product with the intent of reducing waste resulting from a product’s use. Product improvements include product substitution, product conservation, and changes in product composition.

4.

Example: Product Improvements-Product Substitution. In the paint manufacturing industry, water-based coatings are finding increasing application where solvent-based paintswere previously used. These products do not have the toxic or flammable characteristics that make solvent-based paints hazardous.Also, it is not necessary to clean the applicators with solvent. The use of water-based paints instead of solvent-based paintsalso greatly reduces volatile organic compound emissions to the atmosphere. In many cases, water-based productsare cheaper than those based on organic solvents.

B. Recycling 1. Extended Use andReuse

Recycling via use and/or reuse involves either returning a waste materialto the originating process as a substitute for an input material or using it in another process as an input material. Return to original process Raw material substitution for another process Example: Reuse-Return to Original Process. A printer of newspaper advertising purchased an ink recycling unit to produce black newspaper ink from its various waste inks. The unit blends different colors of waste ink together with fresh black ink and black tonerto create the new black ink. This ink is then filtered to remove flakes of dried ink. The recycled inkis used in place of fresh black ink and eliminates the need for the company to ship waste ink off-site the savings for disposal.The price of the recycling unit was paid off18inmonths based only on in fresh black ink purchases. The payback period improved to 9 months when the costs for disposing of ink as a hazardous waste were included. Reclamation Reclamation is the recovery of valuable material from a hazardous waste. Reclamation techniques differ from use and reuse techniques in that the recovered material is not used in the facility but is sold to another company. Waste exchanges provide valuable information to link companies that wish to transfer wastes and those that wish to use them [8]. Materials to be reclaimed may be processed for resource recovery or processed as a by-product.

2.

Example: Reclamation. A photoprocessing company uses an electrolytic deposition cell to recover silver from the rinsewater from film processing equipment. The silver is then sold to a small recycler. When the silver is removed from this wastewater, the wastewater can be discharged to the sewer without additional pretreatmentby the company. This unit paid for itself in less than 2 years due to the value of the recovered silver. The company also collects used film and sells it to the same recycler. The recycler bums the film andcollects the silver from the residual ash. Removing the silver fromthe ash renders the ash nonhazardous. Many environmental groups focus on recycling as the solution to solid and hazardous waste problems;however, in most recycling projects, somematerials are lost, contaminated, or leached into the environment. Evenin Japan, where industries and municipalities have the world’s most successful recycling program, all of their recycling practices combined reduce waste by only 65%. Recycling is no substitute for source reduction [l].

Strategiesfor Small Businesses

651

C. Treatment Treatment shouldbe considered onlyas a last resort when implementing a waste reduction program. Waste treatment may involve conversion of hazardous waste to a less toxic form by chemical, biological,or physical meansor it couldbe as simple and dangerous as placement in landfills, deep-well injection, or ocean dumping [6]. Treatmentmayinvolvemoreexoticprocessesforcleaningup or usingcontaminated wastes. Many shop wastes canbe cleaned up using activated carbon filtration, biological treatment, and extraction. Some wastes can be put to good use as a fuel source for on-site heating units. Creative processes and ideas can accomplish wonders in reducing wastes. Even though new ways of thinking and reformulating processes may be required, it is less difficult to improve on waste minimization techniques thanto improve on waste treatment techniques. Some of the real incentives that waste minimization has going forit are that most of the procedures are that can be carried out are not expensive, save money over treatment costs in the long run, not time-consuming, and will help protect business interestsas well as the environment.

IV. ORGANIZING A WASTE REDUCTION PROGRAM Because a waste reduction program affects many groups within a company, a program task force should be assembled. This group must include members of any department that has a significant interest in the outcome of a waste reduction program. The formalityor informality of a waste reduction program will depend on the nature of the company. A program in a large, highly structured company will probablybe quite formal, one in a small company, less formal. Table 2 lists the typical responsibilities of a waste reduction program task force. The task force draws on the expertise of everyone in a small company. Even in a small business, several people willbe required to implement a successful waste reduction program. People with responsibility for waste treatment and disposal, production, facility maintenance, and quality control should be included on the team. At a small facility, a single person, for example, an owneror manager, may have all of these responsibilities; however, even with a small facility,at least two people shouldbe involved to get a variety of viewpoints and perspectives. Some larger companies have developed a system in which assessment teams visit different facilities within thecompany. Benefits result through sharingof ideas and experiences. Similar results can be achieved for smaller businesses by visiting other facilities with similar operations. Table 2 Responsibilities of the Waste Minimization F %m TaskForce Get commitment and a statement of policy from management.

Establish overall waste minimization program goals. Establish a waste-tracking system. Prioritize the waste streams or facility areas for assessment. Select assessment teams. Conduct (or supervise) assessments. Conduct (or monitor) technical/economic feasibility analysesof favorable options. Select and justify feasible options for implementation. Obtain funding, and establish the schedule for implementation. Monitor (and/or direct) implementation progress. Monitor performance of the option once it is operating.

652

Philips

Table 3 Attributes of EffectiveGoals Acceptuble to those who will work to achieve them Flexible and adaptable to changing requirements Measurable over time Motivarioml for all employees Suitable to the overall corporate goals and mission Understandable by all employees Achievable with a practical level of effort

Planning a waste reduction strategy is similar to a big league manager getting his team ready for baseball season. It is best to Establish a game plan. Set specific long-term and short-term goals. Keep your eye on the ball. Concentrate on the worst streams first. Know your spike zone. Set realistic objectives that center on production activities. Keep score. Examine possible cost savings for various process changes. Get the best equipmentfor your players. Do not be afraid to make improvements that will pay for themselves in the long run. Remember that expansion teams do not per&orm well early in the season. Invest sufficient time and effort before evaluating results.

A. SettingGoals The first priority of a waste reduction task forceis to establish goals that are consistent with management policy. Waste reduction goals can be qualitative, for example, “a significant reduction of toxic substance emissions into the environment”; however, it is better to establish measurable, quantifiable goals, since qualitative goals canbe interpreted ambiguously. Quantifiable goals establish a clear guideas to the degree of success expected of a program. Other attributes of effective goals are listed in Table 3. Waste reduction must be reviewed periodically. As the focus of a waste reduction program are becomes more defined, the goals should reflect any changes. Waste reduction assessments not one-time projects. Periodic reevaluation of goals is necessary because changes occur in available technology, raw material supplies, environmental regulations, and economic climate.

B. Planning a Strategy The steps involved in planning and organizing a waste reduction program are summarized in Table 4. Most small businessesdo not realize how much waste they produce. Ask most managers if they use all of their raw material,or input, and. they will probably say that they do. What they actually mean is that they process allof the raw material needed, not that they use the entire amount purchased. If they feel they are getting their money’s worth, most managers will not give a second thought to what is lost or wasted. This philosophy is common in smaller businesses because they often do not have the personnel, the expertise, or the budget to spend on developing adequate inventory control. The larger the business, the more resources are available to scrutinize inventory control. Waste reduction in any size industry can only work if it is embraced by top management. Small operations often have an advantage here because the owner is usually the immediate supervisor and remains on-site for most of the workday. Through leadership and example, employees will be motivated to participate in the overall program.

Strategies for SmallBusinesses

653

Table 4 PlanningandOrganizationActivitiesSummary Setting Up the Program

Get management commitment to Establish waste minimization as a company goal. Establish a waste minimization program to meet this goal. Give authority to the program task force to implement a program. Starting the Program Task Force

Find a “cause champion” with the following attributes: Familiar with the facility, its production processes, and its waste management operations. Familiar with and respected by the employees. Familiar with quality control requirements. Good rapport with management. Familiar with new production and waste management technology. Familiar with waste minimization principles and techniques and environmental regulations. Aggressive managerial style. Getting Company-wide Commitment

Find people who know the facility, processes, and procedures. Find people from the affected departments or groups. Incorporate the company’s waste minimization goals into departmental goals. Solicit employee cooperation and participation. Develop incentives and/or awards for managers and employees. Several large corporations, e.g., General Dynamics, Borden Chemical, and the 3M Corporation, have pioneered effective waste reduction programs with procedures thatfit equally well in smaller operations. These large industries were driven by upper management pressures, profit margins,and corporate images. Inherent in each program is a strong waste reduction plan and direct involvement by employees from all departments [ 8 ] . Small businesses may not have the capital to spend on expensive equipment or to change existing processes overnight, but theydo possess a valuable resource in their employees. Creative ideas and active participationby all employees can translate into unexpected cost savings and profits for any entrepreneur. Acting on an employee’s recommendation, General Dynamics was able to eliminate disposal costs of a caustic solution by finding a market for it. Borden Chemical greatly reduced effluent wasteby encouraging good housekeeping techniques. In the area of waste reduction, employers should never overlook the obvious and never reject another’s ideas as being too simplistic.

V. THE ASSESSMENTPHASE

A. Selecting a Team A program task force should be concerned with reducing waste for an entire facility; however, an assessment team should focus on one particular waste streamat a time if the stream can be linked to a single process.Team membership should notbe limited to environmental managers

654

Philips

and engineers. Teams should include all people with direct responsibility and knowledgeof a particular waste stream or area of a plant. In addition to internal staff members, assistance from outsiders should be considered, especially during assessment and implementation phases. These outside people may be trade association representatives, consultants, or experts from a different facilityof the same company. One or more outsiders can bring in new ideas and providean objective viewpoint. An outsider is more likely to counteract bias brought about by the “sacred cow” syndrome, such as that encountered when an old process area rich in tradition undergoes an assessment. Free or inexpensive assistance may be available froma variety of souices, including universities and community colleges, management associations, state agencies, insurance companies, and retired employees [ 8 ] . Although their services are expensive, professional consultants bring a variety of experiare especially usefulto smaller ence and expertise to a waste reduction assessment. Consultants companies, which may not have in-houseexpertise in relevant waste reduction techniques and technologies. Trade associations, community colleges, and public agencies will sometimes have very knowledgeable p p l e available to provide advice forlittle or no compensation. Production operators and line employees should not be overlooked as a source of waste reduction suggestions. They possessfirsthand experience with processes, and their assistance may be useful in assessing operational changesor equipment modifications thataffect the way they perform their work. “Quality circles” have been instituted by many companies, particularly in manufacturing industries, to improve product quality and production efficiency. These quality circles (sometimes referred to as TQM, total quality management) consist of workers and supervisors sharing ideas on proposed improvements. Qualitycircles are often successful because they involve production people who are closely associated with the operations and truly appreciate being part of the decision-making process. Several companies with quality circles have used them effectively to solicit suggestions for waste reduction that have saved millions of dollars.

B. Collecting and Compiling Data In order to develop options, a detailed understanding of a plant’s wastes and operations is a must. An assessment should beginby examining information about processes,operations, and waste management practices. There are several questions that need to be answered [lo]: 1. What waste streams does the business generate? How much? 2. With which processes or operations do these waste streams originate? are not?Whatmakesthem 3. Whichwastes are classified as hazardous,andwhich 4.

5. 6.

7. 8. 9. 10.

hazardous? What are the inputmaterials used that generate the waste streamsof a particular process or plant area? How much of a particular input material enters each waste stream? How much of a raw material can be measured throughfugitive losses? How efficient is the process? Are unnecessary wastes generatedby mixing otherwise recyclable hazardous wastes with other process wastes? What types of housekeeping practices are used to limit the quantity of wastes generated? What types of process controls are used to improve process efficiency? Additional types of information for assessing wastes are listed in Table 5.

Strategiesfor SmallBusinesses

655

Table 5 Facility Information for Waste Minimization Assessments Design information

Process flow diagrams Material and heat balances (both design balances and actual balances) for (1) production processes and(2) pollution control processes Operating manuals and process descriptions Equipment lists Equipment specifications and data sheets Piping and instrument diagrams Plot and elevation plans Equipment layouts and work flow diagrams Environmental information ;

Hazardous waste manifests Emission inventories Biennial hazardous waste reports Waste analysis Environmental audit reports Permits andlor permit applications Raw materiallproduction information Product composition and batch sheets Material application diagrams Material safety data sheets Product, utility, and raw material costs Operating and maintenance costs Departmental cost accountingreports Economic information

Waste treatment and disposal costs Product, utility, and raw material costs Operating and maintenance costs Departmental cost accountingreports Other information

Company environmental policy statements Standard procedures Organization charts

C. Site Inspection Once an assessment team isin place and a specific areaor waste stream has beenselected, the assessment may continue with a visit to the site. Although collected informationis critical to gaining an understanding of the processes involved, it is helpful to witness an actual operation. In many instances, a process unit is operated differently from themethod described in an operating manual. Modifications may have been made that were not recorded on flow diagrams or equipment lists. An assessment team should preparea list of needed information andan inspection agenda prior to conducting the site inspection. This can bea checklist detailing objectives, questions and issues to be resolved, andlor further information needed. An agenda and information list should be given to appropriate plant personnel beforethe visit to allow them time to assemble the requested information.With a carefully thought-out agenda andchecklist, important points are less likely to be overlooked during an inspection. the In performing a site inspection, an assessment team should observe each process from point where rawmaterials enter to the point where products and wastes leave an area. The team

656

Philips

should identify suspected sources of waste. Thismay include the production process, maintenance operations, storage areas for raw materials, finished product, and work-in-progress. Even a plant’s waste treatment area may itself offer opportunities to minimize waste. An inspection often results in preliminary conclusions being drawn aboutthe causes of waste generation. Full confirmationof conclusions may require additionaldata collection, analysis, and/ or site visits. The following are some guidelines for asite inspection [lo]. 1. Prepare an agenda in advance that covers all points that still require clarification. Provide staff contacts in each area being assessed with an agenda several days before the inspection. 2. Schedule an inspection to coincide with a particular operation that is of interest (make-up chemical addition, bath sampling, bath dumping, start-up, shutdown, etc.). 3. Monitor an operation at different times during shifts, especially when waste generation is highly dependent on human involvement (e.g., painting or parts cleaning operations). 4. Interview operators, shift supervisors, and foremen in an assessed area. Do not hesitate to question more than one person. Assess operators’ and supervisors’ awareness of waste generation aspects of their operation.Note their familiarity (or lack thereof) with the impacts their operation may have on other operations. 5 . Photograph an area of interest. Photographs are valuable in the absence of plant layout be forgotten drawings. Many details canbe captured in photographs that otherwise would or inaccurately recalled at a later date. 6. Observe “housekeeping” aspects ofan operation. Check for signsof spills or leaks. Visit the maintenance shop and ask about problems in keeping equipment leak-free. Assess the overall cleanliness of the site. Pay attention to odors and fumes. 7 . Assess the organizational structure and the levelof coordination of environmental activities between various departments. 8. Assess administrative controls, such as cost accounting procedures, material purchasing procedures, and waste collection procedures.

D. Flow Diagrams and Material Balances Different materials and proceduresare used to execute each process; however, every operation maoffers an opportunityto track material flow. Figure1 provides an extremely oversimplified terial flow chart. Before a waste reduction plan can be formulated, it is crucialto determine the amount of waste that the business is already producing. How? Perform a waste audit. What is a waste or audit? It is the direct application of the law of conservation of matter taught in Physics Chemistry 101: Matter is neither created nor destroyed. In other words, nothing ever disappears. Everything thatwas here on the first day this planet was created is still with us today, although it may be in a different form. or goes “away,” it must be somewhere. So to find out where Since nothing ever disappears the “matter” is going, a waste reduction team must perform a material balance. In this process, a quantitative relationship is established between raw materials, product, and waste. Material balances allow for quantifying many losses or emissions that were not previously recorded by any bookkeeping or inventory system. In its simplest form, a material balance is representedby the equation Mass in = mass out

+ mass accumulated

657

Strategies for Small Businesses AIR EMISSIONS

SPILLS

Figure 1 Simplified flow chart indicating process components to be included in a material balance study.

An independent material balance should be made for all componentsthat enter and leave a process. When chemicalreactions take place in a system, an “elemental balance” should be performed for specific chemical elements. Material balances assist in determiningconcentrationsof waste components where data are limited. Theyare particularly useful when there are points in a process where itis difficult (due to inaccessibility) or uneconomical to collect analytical data. A material balance candetermine whether or not fugitive losses are occurring. For example, the evaporation of solvent from a the tank and solvent cleaning tank canbe estimated as thedifference between solvent put into removed from the tank. effort; however, Characterizing wastestreamsby material balance can require considerable the result is a complete picture of a waste stream. This helps to establish a focus for waste reduction activities and provides a baseline for measuring performance. 1.

Sources of Material Balance Information

Material balance includes materials entering and leaving a process. The following list exemplifies potential sources of material balance information: Samples, analyses, and flow measurements of feedstocks, products, and waste streams Raw material purchase records Material inventories Emission inventories Equipment cleaning and validation procedures Batch make-up records Product specifications Design material balances Production records Operating logs Standard operating procedures and operating manuals Waste manifests

658

Philips

Material balances are easier, more meaningful, and moreaccurate when done for individual units, operations, or processes. It is important to define a material balance envelope properly. An envelope should be drawn around a specific area of concern rather than a larger group of areas or an entire facility. An overall material balance fora facility can then be constructed from individual unit material balances. This highlights relationships between units and helps to point out areas for waste reduction. Several factors mustbe considered when preparing material balances in order to avoid errors that could significantly overstate or understate the waste streams. The precision of analytical data flow measurements may not allow an accurate measure of a stream. In processes with very large inlet and outletstreams, the absolute error in measurement of these quantities may be greater in magnitude than the waste stream itself. Whenever a reliable estimate of a waste stream cannot be obtained by simply subtracting the quantity of material in a product from that in a feed, the team must be creative in developing a more advanced methodof quantitative analysis. Time spanis important whenconstructing a material balance. Material balances performed over the duration of a complete production run are typically the easiest to construct and are reasonably accurate. Time duration also affects the use of raw material purchasing records and on-site inventories for calculating input material quantities. The quantities of materials purchased during a specific time period may not necessarily equal the quantity of materials used in production duringthe same time period. Purchasedmaterials can accumulatein warehouses or stockyards. Developing material balances around complex processes can bea complicated undertaking, especialy if recycle streams are present. Such tasks are generally performed by chemical engineers, often with the assistance of computerized process simulators. Material balances will often be needed to comply with SARA Title I11 reporting requirements to establish emission inventories for specific toxic chemicals. EPA's Office of Toxic Substances has prepared a guidance manual entitled Estimating Releases and Waste Treatment Eflcienciesfor the Toxic Chemicals InventoryForm (EPA 560/4-88-02). This manual contains additional information on developing material balances for listed toxic chemicals [ll]. 2. TrackingWastes By tracking wastes ona regular basis, seasonalvariations in waste flows or single-batch waste streams are distinguished fromcontinual, constant flows. Changesin waste generation cannot be meaningfully measured unless the information is collected both before and after a waste reduction option is implemented. Fortunately, it is easier to do material balances the second time, and it continues to becomeeasier as more are completed. It is relatively simple and inexpensive, with today's computer software,to use computerizeddatabase systems to track and monitor waste streams.

E. Prioritizing Waste Streams and/or Operationsto Assess Ideally, all waste streams and plant operations should be assessed. Prioritizing waste streams and/or operations to assess is necessary when available funds, personnel, or time is limited. Waste reduction assessments should concentrate on the most important waste problems first and then move to lower priority problems as time, personnel, and budget permit. Setting priorities for waste streams or facility areas requires a great deal of care and attention, since this step focuses the direction of an assessment activity. Important criteria to consider when setting these priorities include the following. Compliance with current and future regulations

Strategiesfor Small Businesses

659

Costs of waste management (treatment and disposal) Potential environmental and safety liability Quantity of waste Hazardous properties of a waste (including toxicity, flammability, corrosivity, and reactivity) Other safety hazards to employees Potential for removing bottlenecks in production or waste treatment Available budget for waste minimization assessment program and projects Small businesses with only a few waste-generating operations should assess their entire facility. It is also beneficial to lookat an entire facility when thereare a large number of similar operations. The implementationof good operating measures such as soliciting employee suggestions, awareness-building programs, better inventory and maintenance procedures, and internal cost accounting changes can be implemented on a facility-wide basis. Many of these options do not require large capital expenditures and can be put into operation in a relatively short time. Performing the mass balance may be the most difficult part of the waste audit, but it will make the remainder of the waste reduction process easier and more effective.

VI. WASTEREDUCTIONOPTIONS Once the origins and causes of waste generationare understood, the assessment process enters the creative phase. Theobjective of this step is to generate a list of waste reduction options for further consideration. Following the collection of data and site inspections, members of the team will have begun to identify possible ways to minimize waste in an assessed area. The identification of potential options relies on both the expertise and creativity of all team members. Much knowledge comes from education and on-the-job experience; however, the use of technical literature, contacts, and other sources is always helpful. When identifying options, team members should follow the waste reduction hierarchy in which source reduction possibilitiesare explored first, followed by recycling options. With an ever-increasing emphasis being placed on zero-discharge and on-site recycling, waste treatment technology will eventually be phased out ofmany operations. Source reduction is preferred because it reduces or eliminates waste from the beginning of a process. Recycling should be considered only when all attempts at performing source reduction have been exhausted, and then on-site recycling should bethe primary method considered.

A. Methods of Generating Options The process whereby waste minimization options are generated should occurin an environment that encourages opennessand independent thinkingby members of the assessment team. While individual team members will suggest many potential options on their own, the process can be enhanced,by using some common group decision techniques. These techniques allow an assessment team to identify options that individual members might not have come up with on their own. Brainstorming sessions with team members are an effective means of developing options. It is likely that all or most of the options being presented have some merit and some drawbacks. During a brainstorming session, it is best not to dwell on details but to simply list as many options as possible.

B. Screening and Selecting Options for Further Study Several waste reduction options will be identified in a successful brainstorming session. It is then necessary to identify the options that offer the best potential for minimizing waste and

660

Philips

reducing costs. Because a detailed evaluation of technical and economic feasibility is often costly and time-consuming, the proposedoptions should be screenedto identify the ones that deserve further evaluation. An initial screening will eliminate options that appear unreasonable, impractical, or inferior without a detailed and more costly feasibility study. Screening procedures can range from a simple vote by team members to sophisticated quantitative decision-making formulas. Generally, an informal evaluation utilizes an unstructured conferencesetting by which an assessment teamor program task forceselects options that appear tobe feasible. This methodis used most often by small businessesor in situations where only a few options are practical.

VII. FEASIBILITYANALYSIS Once the assessment team has agreed to evaluate selected options, a feasibility analysis is conducted to determine which options are technically and economically practical. The level of analysis required is directly proportional to the complexity of the option being considered. Simple options, such as housekeeping improvements, would not require as thorough an evaluation as equipment changes[7]. The final product of an assessment phaseis a list of feasible options for an assessed area. The assessment will have screened out impracticalor unattractive options. The next step is to determine if remaining options are technically and economically feasible. An evaluation should be conductedto determine whether a proposed option will workfor a specific application. The evaluation of an option should include facility constraints, product requirements, capital costs, and operating costs. A fiscal evaluation should consider standard measures of profitability such as payback period, return on investment, and net present value. Each organization has its own economic criteria for selecting projects; however, as in any project, the cost elementsof a waste reduction project can be broken down into long-term and short-term capital costs and operating costs. All affected groups in a facility should contribute to and review the results of a technical evaluation. Prior consultation and review with affected groups is needed to ensure the viability and acceptance of an option. If the option calls for a change in production methods or input materials, the project’s effects on final product quality must be determined. If, after technical evaluation, the project appears impossible or impractical, it should be dropped.

VIII. IMPLEMENTING WASTE REDUCTION OPTIONS The assessment team’s evaluation report should providethe basis for obtaining company funding of waste reduction projects. Projects should notbe marketed ontheir technical or environmental merits alone. An analytical, fiscally sound description of both short- and long-term benefits can help edgea proposed project past competing projects for funding. Itis also helpful to have someone from the manufacturing, purchasing, or accounting section of the facility present the proposal. This will enlarge the imageof the projectbeyond environmental improvements, to depict the proposed action as a sound business investment. The presenters of the proposal should be flexible enough to develop alternatives or modifications. They shouldalso be sufficiently committedto the projectto obtain background data and support studies and to anticipate potential problems that may be encountered during implementation. Above all, team members should keep in mindthat an idea will not sell if team members are not sold on it themselves. This is why management and production staff people should support the concept of waste reduction from the beginning of the process.

Strategies for Small Businesses

661

A. ObtainingFunding Waste reduction projects generally call for improvements in process efficiency and/or reductions in operating costs of waste management. These are the points that shouldbe emphasized in the decision-making process. An assessment team made up of financial and technical personnel helps to ensure that a sponsor’s enthusiasm is balanced with objectivity. Even if a project promises a high rate of return, a small company may have difficulty acquiring funds for capital investment. In this case, the company should look to outside financing. It generally has two major sources to consider: private sector financing and government-assisted funding. If the venture is prudent and cost-effective, many lenders will gladly provide loans for pollution prevention or waste reduction projects, particularlyif they have already backed the company with other loans. Government grants and loans arealso available for pollution prevention projects. If the project involves a new or unusual approach to waste reduction, the entire proposalmay be funded by a research grant.

B. Installation Waste reduction proposals that involve operational, procedural or material changes canbe implemented as soon as funding has been approved. For projects involving equipment modificabe involvedwithplanning,design,comparisons, tions or new equipment,theteamwill procurement, and construction of thenew equipment.

C. MeasuringWasteReduction Once a waste reduction project has been implemented, it mustbe monitored to determine how effective the option actually turns outto be. Projects that do not measure up to their original performance expectations may require reworking or modification. One measureof project effectiveness is its payback time. A project should pay for itself in a reasonable time through reduced waste disposal costs and/or reduced raw materials costs. The primary goal of the project, however, is to reduce waste. This should be the major consideration when implementing a program. Accurate measurements must be takento express the extent of waste reduction. The process by which the wastes are reduced must be fully understood and compared to previous processes to determine the total impact of the new method.

D. AssessingNewProcesses It is important to avoid trading one waste problem for another. Waste reduction principles should be applied tonew projects experimentally on a small scale. Iteasier is to observe waste generation during research and development than to go back and modify a process after it has already been installed. With new emphasis on attaining zero discharge and eliminating off-site recycling, companies must strive to implement fully contained systems.

E. KeepingtheFaith A waste reduction program is an ongoing rather than a one-time effort. Once high priority streams havebeen assessed and reducedor eliminated, an assessment team should look at low priority streams. The ultimate goal of the program should be to reduce the generation of waste to the maximum extent achievable.

Phillips

662

To be truly effective, a philosophy of waste reduction must be developed in the organization. This meansthat waste reductionmust be an integral part of the company’s operations. The most successful waste reduction programs to date have all developed this philosophy within their companies.

REFERENCES 1. 2. 3. 4. 5.

6. 7. 8. 9. 10.

11.

Long, R. E., The problem of waste disposal, Rd. Shelf, 60, 41-48 (1989). University of Tennessee Teleconference,Promote Landfill AlternativesNow. PLAN Booklet, 1992. Council for Solid Waste Solutions, The Solid Waste Management Problem, 1989, pp. 7-10. Marburg Associates, Site Auditing: Environmental Assessments of Property, STP Specialty Technology Publishing, 1992, Appendix U.S. Office of Technology Assessment,Serious Reductionof Hazardous Waste,Booklet U.S. Congress, Washington, D.C., 1991. Gore, A., Earth In the Balance, Houghton Mifflin, Boston, 1992, pp. 342-343. Theodore, L. and Mffiuinn, Y., Pollution Prevention, Van Nostrand Reinhold, New York, 1992, p. 170. Philips, D. A., Waste Reduction in Industrial Processes, Florida Dept. of Environmental Regulation,1991,pp.1:16-24. Environmental Information, Ltd., Tax incentives for pollution abatement, Environ. Dig. State Programs, Minneapolis, Minn., May1992,pp.18-21. Freeman, H. A.,Hazardous Waste Minimization, Mffiraw-Hill, New York, 1990, pp. 77-107. HowToComply Handbook SARA Title III FloridaStateEmergencyResponseCommission, 1990, p. 7.

33 Contaminated Soils in Highway Construction

Namunu J. Meegoda New Jersey Instituteof Technology Newark, New Jersey

1.

INTRODUCTION

In 1988, there were approximately4 million milesof roads in the UnitedStates of America, of which 2.3 million miles were surfaced with asphalt or concrete. Of the surfaced roads, approximately 96%, or 2.2 million miles, had asphalt pavements[l]. More than 95% of the2 trillion vehicle miles traveled each year occur on asphalt-paved roads. In 1988, expenditures for highways amounted to over $68 billion at all levels of government [l]. During 1988, 500 million tons of hot mix asphalt (HMA) was produced and placed, and about 250 million tons of HMA was used for construction, rehabilitation, and maintenance of the highways [l]. The HMA industry directly employs 300,000 people and indirectly accounts for an additional 600,000 jobs [l]. The hot mix asphalt concrete consists of a combination of aggregates blended and uniformly mixed, coated with asphalt cement, and compacted into a dense material. The materials in HMA consist of (1) coarse aggregates with sizes ranging from 1.5 in.to U.S. sieve No. 4, (2) fine aggregate or sand with sizes passing U.S. sieveNo. 4 and retained on U.S.sieve No. 200, (3) mineral filler such as crushed stone dust or lime passing U.S. sieve No. 200, and (4) asphalt cement. A typicalHMA composition consistsof 50% coarse aggregates, 40% fine aggregate, 5% mineral filler, and 5% asphalt cement. Asphalt cement is obtained by distillation of petroleum crude. The asphalt cements obtained from refineries are classified as AC-2.5, AC-5, AC-10, AC-20, AC-30, or AC-40 based on viscosity. To obtain sufficient fluidityof asphalt cement for proper mixing and compaction, both the aggregate and the asphalt cement are heated prior to mixing; hence the product is called hot mix asphalt (HMA) concrete. qpically 5-10% waste products such as recycled asphalt pavements, tire rubber, glass, municipal solid waste (MSW)ash, roofing shingles, polythene waste,fly ash, bottom ash, ore slug, and petroleum-contaminated soils are addedto HMA without sacrificing its strength and performance [2-51. U.S. Department of Transportation data [l] suggest that an estimated 25 million tons of industrial waste can be recycled and consumed annually by the U.S. asphalt

663

664

Meegoda

industry, by simply adding 5% waste productsto all HMA mixes. Therefore, all levels of government in the UnitedStates emphasize the inclusion of waste products in HMA. Usually separate mix designsare not performedto include such waste material. Either the amount of materi in the original mix is proportionately reduced or waste products replace the mineral filler. Petroleum-contaminated soil (PCS), one of the industrial solid waste products used in HMA, is generated from leaking underground storage tanks (USTs), including their piping systems. During the 1950s and 1960s, the construction of many gasoline stations and chemical manufacturing and processing facilities led to the installation of millions of USTs. Several million USTs in the United States contain petroleum products. Tens of thousands of these USTs, including their piping systems, are currently leaking [6].The U.S. Environmental Protection Agency estimates that there are more than400,OOO leaking USTs with petroleum hydrocarbons. Many more are expected to leak in the future. Most states vigorously encourage the removal of all tanks after 25 years of service. It is estimated that the removal of a leaking tank generates approximately 50-80 yd3 of contaminated soil. In addition, soils that surround petroleum refineries and crude oil wells are contaminated with petroleum products. Since groundwater is a source of drinking water, federal legislation seeks to safeguard our nation’s groundwater resources. Congress responded to the problem of leaking USTs by adding Subtitle I to the Resource Conservation and RecoveryAct (RCRA) in 1984. The current federaland state statutes require that leaking USTs be removed to prevent further contamination. Petroleum-contaminated soils consist of mixtures of natural sands, silts, and clays with petroleum products. A small portion of light petroleum product mixed with asphalt cement merelyproduces an asphaltcementofslightlydifferentspecification or characterization. Therefore, it is believedby many involved with the asphalt industry that a small increase in the quantity of light petroleum substanceswould not damage the HMA. This is, in fact, the basis for the theory that contaminated soils can be used in asphalt concrete paving. However, the inclusion of natural soils inHMA and the environmental impact dueto such inclusion require an in-depth study. When PCSs are added to HMA, three beneficial actionsoccur: incineration, dilution, and solidification. Part of the petroleum is used as a fuel and is burned during the production of asphalt concrete; thus a majority of the contaminantsare eliminated beneficially, which reduces fuel costs. Since only5-3096 petroleum-contaminated soilis added to virgin aggregates during the production of asphalt concrete, there is spreading and dilution. The asphalt cement actsas a binder in asphalt concrete; therefore, the remaining diluted contaminants are solidified and stabilized in the final asphalt concrete matrix. Strength or stability, durability, and workabilityare the primary factorsto be considered in hot asphalt mix designs [7]. The secondary factors are flexibility, permeability,fatigue resis[7].Since the HMA produced withPCSs is used for tance, skid resistance, and stripping action paving, it must satisfy all the engineering and environmental requirements specified by the appropriate federal, state, and local agencies. In 1990, a major laboratory and field research project fundedby the New Jersey Department of Environmental Protection and Energy (NJDEPE) was initiated to investigate the feato sibility of using PCSs in HMA. An extensive laboratory and field study was conducted evaluate the use of PCSs as an aggregate replacement in the production of hot mix asphalt (HMA) concrete. In the laboratory and field investigations, petroleum-contaminated soils were , added to HMA, and the resulting asphalt concrete mixes were tested for strength, durability, permeability, and leachability. During the field production of HMA and PCSs, emissions of volatile organic compounds(VOCs) were also monitored. In this chapter, the engineering performance of HMA with K S s is summarized and the environmental impact of the process is described in detail.

Soils Contaminated

665

in Highway Construction

Table 1 DataonSixContaminatedSoilsfromNewJersey Soil 1 sand

Soil 2

Soil 3

Soil 4

Soil 5

classification Soil Well-graded Clayey Silty sand Poorly siltSilty clay graded sand

Soil 6 Poorly with silt

moisture In situ 7.3 24.1 14.3 content (a) qpe and amount 1100 ppm 1200 ppm of contamination heating heating oilheating oilgasoline oil gasoline gasoline

6600 ppm

14.4

19.6

10.1

25 ppm

1500 ppm

330 ppm

Source: Meegoda et al. [ 8 ] .

II. EXPERIMENTAL PROCEDURE

AND RESULTS

A. Engineering Performance of HMA with PCSs 1.SoilClassification Six contaminated soils providedby NJDEPE from sites identifiedas containing soils with less than 1% total petroleum hydrocarbons (TPHs) were selected for testing and for characterization. Three soils were contaminated with heating oil, and the other three were contaminated with gasoline. The degree of contamination for oil-contaminated samples was determinedby the Soxhlet oil and grease extraction method (USPHS standard method for the analysis of water and wastewater). Table 1 gives the soil classification data and lists the type and amount of contaminants in each soil. 2. Stability or Strength of HMA with PCSs All the asphalt concrete samples were designed for New Jersey surface coarse mix (NJ 1-3). A control mix was designed and tested for comparison. The mix designs to include soils were prepared for each of the six soils based on the sieve analysis data. (Please refer to Meegoda data for soilsand aggregates and the NJ 1-3 specifications.) The et al. [8] for grain size analysis optimum percentages that may be usedin the NJ 1-3 mix based on aggregate blending for the above six soils wereas follows: Soil I , 35%; soil 2, 10%; soil 3, 20%; soil 4, 15%; soil 5, 10%;andsoil 6 , 15% [8]. These percentages are muchhigherthanthecurrentpracticeof 5% PCSs. Once the maximum amountof PCS that may be added to HMA was determined, the suitability of such an addition was evaluated.The Marshall stability test was performed using ASTMD1559.BasedontheNewJerseyDepartmentofTransportation(NJDOT)requirements for Marshall strength, bulk density, air voids, voids in mineral aggregates (VMA), and flow, the optimum asphalt content for each mix was selected. The optimum asphalt contents based on Marshall test results for the control mix and for the six contaminated soils are shown in Table 2 [8]. Table 2 also shows the dry density, Marshall stability, air voids, VMA, and flow values at the optimum asphalt contents for the control as well as for HMA made with 1-3 mix are also shownin each soil type. The New Jersey specifications for high traffic volume Table 2. If an asphalt concrete meets all state specifications and if it is a workable mix, then it is accepted as a paving material. Exceptfor the control mix and HMA mixed withPCS No. 4, as shown in Table2, all the HMA mixes with PCS were acceptableas paving materials. The control mix and the HMA mix with soil No. 4 had low flow values. A higher flow value can be selected for all three mixes from Marshall test results with higher asphalt contents, but it 8006 N). If a lower Marshall strength value will result in lower stability values (lower than

Meegoda

666

Table 2 Optimum Properties of Asphalt Concrete with PCSs for NJ 1-3 Mix Asphalt concrete Allowable for property NJ 1-3 Control Mix Strength (N) Flow (0.25 mm) Air voids (%) VMA (%) Density (W/m3) Optimum asphalt content (%)

>8006 >6.0 2.0-8.0 >13.0

8006 4.0 7.0 18.0 24.3 5.0

NIA 4-8

Soil 1

Soil 2

Soil 3

Soil 4

Soil 5

Soil 6

8228 11.0 7.5 17.8 24.8 4.5

8450 8.0 3.0 14.0 24.5 4.5

10229 7.5 5.7 16.8 24.1 5.0

8450 3.5 8.0 18.0 23.4 4.5

8317 6.5 4.0 14.7 24.6 4.5

10452 7.7 3.4 14.2 24.5 4.5

N/A= not available. Source: Meegoda et al. [8].

(say 6671N) is acceptable, then the control mix and the HMA made with soil No. 4 are acceptable as paving materials. Based on the test results shown in Table 2, it can be stated that the HMA with PCSs produced better asphalt concrete thanthe control. This may be due to the better blend obtained by adding natural soils. 3. Durability ofHMA with

PCS

ASTM D4867 describes the test procedure for determining the effect of moisture on asphalt concrete mixtures,a factor that is very important for the durabilityof hot mix asphalt concrete. It has a section on freeze-thaw conditioning of a mixture. However, the freeze-thaw and wetdry tests require onlyone cycle eachof freeze-thaw or wet-dry. There was no rationale for the selection of one cycleof freeze-thaw or wet-dry to evaluate the moisture damage when, in the real world, asphalt concrete pavements are subjected to several freeze-thaw and wet-dry cycles under service conditions before they are removed for resurfacing. Therefore, an experimental program was designed to evaluate the moisture damage of the control mix and one mix prepared with PCS No. 3. In this experiment HMA specimens were subjected to several cycles of freeze-thaw and wet-dry. For this experiment, 18 specimens of control mix and 18 specimens of HMA mixed with PCS No. 3 were used. The experimental test resultsof an extended freeze-thaw test are plotted in Figures 1 and 2. Figures 3 and 4 show similar data for wetdry tests. In this extended durabilitytest, the control mix and HMA with PCS No. 3 were tested for one, three, seven, and 14 cycles, with each cycle taking approximately48 hr. Upon completing 14 freeze-thaw and wet-dry cycles, data were collected and graphically displayed. It can be concluded from thesetest results that as the temperature drops to freezing temperatures for the freeze-thaw specimen, the asphalt concrete contracts and becomes brittle, creating tiny cracks on the surface of the sample, which provide entry points for water. Water inside the specimen causes moisture damage due to strippingand volume expansions during subsequentfreeze cycles. With the increasein the number of freeze-thaw cycles, the cracks get larger,letting more water into the specimen and eventually leading to the failure of the specimen. The data from cyclic freeze-thaw tests indicate that the percentage swell increased rapidly during the first cycle and then gradually reacheda maximum before the specimen failed. It is believed thatthe specimen reached its maximum percentage swell when it was totally saturated, and stripping occurred before complete failure. The tensile strength ratio also decreased rapidly during the first cycle and also began to level off to zero strengthafter 14 cycles. The behavior seemsto be similar for the control mix and the mix containing PCS No. 3, suggesting comparable durability

Contaminated Soils in Highway Construction

.-

667

c

70- ......................................................................................................

.............

5

60-........................................................................................................ ...............................................

.............

c 2

50 -

iz

40-

d

m

-.U)

c

F

PCS#3

30- ..........

............

"

............. .......

..

"

.....

20-

"

10...........................................

0

0

1

2

I

4

I

I

6 8 10 NUMBER OF CYCLES

I

12

1

14

'

5

Figure 1 Extended freeze-thaw test for control mix and HMA with PCS No. 3-TSR Meegoda et al. [9].

values. (From

100 ap

h v)

t

0 .+-

90

......................

80

70

a .F U)

C

0

T

0

2

4

12 8 10 NUMBER OF CYCLES 6

16

14

Figure 2 Extended freeze-thaw test for control mix and HMA with PCS No. 3-swell values. (From Meegoda et

al. [9].)

values for these two mixes. It also seems that there is a correlation between the percentage swell and the tensile strength ratio. The data show that most of the strength is lost during the f i i t cycle, and hence there was no need to test beyond one freeze-thaw cycle as suggested by ASTM D4867. The cyclic wet-dry test also indicated that the first cycle was the p i n t where attention must be focused. For the wet-dry test, the tensile strength ratio declined during the f i t '

668

Meegoda 6-

5 5-

5 - ........................................................................................................... 4 5-

p ..c# ..

.................

........

4 - .................

g 3 5 - ............................................................................................................................................................................

h

ti -I W

3-........ I

................................................

-

.................. .......... ...-..............." "

2

"

4

6

"

........... .....................................................................................

"

"

.................. .............................. ..................... ............................................ + .............. ""

0

.

...........

...............................

... ................

^

"

"

"

"."..._..............". ........ "

8

10

....".

"

12

"

14

16

NUMBER OF CYCLES

Figure 3 Extended wet-dry test for control mix and HMA with PCS No. 3-TSR goda et al. [9].)

values. (From Mee-

.-

h

0

2

4

6 8 10 NUMBER OF CYCLES

16 12

14

Figure 4 Extended wet-dry test for control mix and HMA with PCS No. 3-swell values. (From Meegoda et al. [9].)

cycle and increased thereafter. This is believed to be due to oxidation of asphalt during the drying cycle. Therefore, it was concluded that the first wet-dry cycle should yield the critical conditions. It can be concluded from these tests that freeze-thaw and wet-dry tests with one cycle indicate whethera specimen is durable, and a strength lossof more than20% indicates that the specimen will not withstand harsh weathering conditions.

Contaminated Soils in Highway Construction

669

Table 3 Durability of HMA with PCSs HMA mix

Freeze-thaw Wet-dry test test

82.3Control 89.0 with soil 1 HMA 100.0 with soil 2 HMA HMA with soil 3 87.0 HMA with soil 4 98.4 HMA with soil 5 HMA with soil 6 Source: Meegoda et

91.7 98.0 89.3 87.2 83.8 93.4 100

93.9

100

al. [9].

Table 3 shows the tensile strength ratio (TSR) values of wet-dry and freeze-thaw tests performed based on ASTM D4867 for the control mix and for the HMA made with six soils. TSR values for HMA with PCSs were not significantly different from that of the control, indicating that HMA with PCSs produced durable asphalt concrete. Some of the wet-dry and freeze-thaw test results show that the HMA withPCS are better than theHMA produced with virgin aggregates. 4.

Permeability of HMA Prepared with PCSs Hydraulic conductivity tests were performed on laboratory-compacted HMA specimens. The of a 100 mm diameter flexible-wall permeameter. A cell pressure specimens were placed on top of 50 psi (344kPa), a back-pressure of 30 psi (206 P a ) , and a desired pressure difference (mainly 1 psi or 7 kPa) were applied to the specimen. Twenty-four hours after the in-flowbecame equalto the outflow and when the hydraulic conductivity didshow not further reduction, the permeability test was stopped. The permeability test was conducted concurrently on three PCS. A bladder accumulator was connected between specimens of HMA made with the same the permeameter and the pressure panel to separate the permeant from the distilled water used in the pressure panel. This procedure eliminated the contaminationof the pressure panel. 5 shows At the end of the permeability test, the specimen height was measured. Figure typical permeability test results where the variation of hydraulic conductivity with time is shown. The average saturated hydraulic conductivities of the control mix and HMA mixes with PCSs are shown in Table 4. Table 4 shows the saturated hydraulic conductivity data for HMA with and withoutPCSs. Only one mix (with PC No. 2) showed a lower saturated hydraulic conductivity value than the control mix. However, the saturated hydraulic conductivity values of all the HMA mixes with PCSs werelessthan 2.0 X acharacteristicvaluefor low permeableclay-typesoilsand hence canbe considered acceptable. Table4 also shows the other parameters of asphalt matrix that contribute to the saturated hydraulic conductivity. It appears that the combination of air voids, asphalt content, and dlo size (aggregate size correspondingto the 10% finer fraction in the gradation curve) control the saturated hydraulic conductivity. With a higher percentage of air voids, one would expect a higher hydraulic conductivity as a larger fractionof asphalt concrete matrix will be porous, allowing fluid to flow through those voids. In geotechnical engiof with soils neering, it is an accepted fact d,, thatsize controls the hydraulic conductivitysoils, that have higher d l , sizes having higher hydraulic conductivity values. However, since all the HMA mixes tested in this research were NJ 1-3 mixes, there should not be drastic differences in d l , size as shown in Table 4; hence its influence here is minimal. The asphalt content will also play a major role when it is higher than optimum, as the excess asphalt cement, after coating all the particles, will block the interconnecting voids,

670

Meegoda

; inlet 0 :outlet

......

I

‘0

Time (hrs)

Figure 5 A typical hydraulic conductivity test result. (From Meegoda

et al. [9].)

Table 4 Hydraulic Conductivities and Other Related Parameters of HMA with PCSs Saturated hydraulic HMA conductivity mix (cm/sec) (mm) mix voids Air 4.75 Control HMA with soil HMA with soil HMA with soil 4.25with soil HMA HMA with soil HMA with soil

1

2 3 4 5 6

4.7 2.3E-07 3.3E-07 1.6E-07 1.6E-06 6.9 1 .OE-06 8.3847 4.6E-07

d,, size inthe (%)

0.18 4.50 0.21 0.13 0.12 0.27 4.50 0.14 4.50 0.15

Asphalt content

(%)

5.6 6.9

7.5

4.50 5.00

7.3 6.3

Source: Meegoda et al. [9].

causing a reduction in hydraulic conductivity. However, since all the mixes were tested around the optimum asphalt content values for the corresponding mixes, its contribution to the test results are minimaltoo.Therefore,thehydraulicconductivityvaluesshouldincreasewith the percentage of air voids, and there appears to be a direct correlation between percentage of air voids with the hydraulic conductivity except for mix having PCS No. 3. We suspected measurement errors for this mix [9]. Therefore, it canbe concluded that the additionof PCSs may not change the saturated hydraulic conductivityof the asphalt concrete and the change is due only to the difference in air voids, dlo size, and asphalt content associated with the mix design [9].

B. Environmental Impact of Adding PCSs to HMA 1. Leachability Test for HMA with PCS The increase in the number of organic contaminants being detected in groundwater as well as in surface water is causing concern because of potential health risks claimed to be associ-

Contaminated Soils Construction in Highway

671

ated with human exposure to these substances. Soil contaminated with petroleum product is a potential threat to both surface water and groundwater; thuskey a issue is whether hydrocarbonwillleach out from the asphalt mixture when it is mixedwith PCSs andused to pave roads. The short-term test conducted by Eklund [IO] on the environmental effects of paving showed that after allowing a pavement to leach for7 days with distilled water, the amount of petroleum-based compounds leached out and detected was less than 2 ppm. In another series in cold of experiments by Eklund [lo] with 6% waste oil with an appreciable lead content mix asphalts (CMA), when the CMA was leached for a week in an acid rain simulation, the leachate contained less than 3 ppm of lead. Although short-term tests showed no sign of any harmful leachate generation, tests should be conducted to determine the environmental impact of long-term exposure of asphalt produced with contaminated soils on the environment. A uniform leaching test was used to estimate the quality of leachate that would be produced by the asphalt mixture. A new leachability test was designedto simulate the rate of release of contaminants when HMAs mixed with PCSs are exposed to the actual performance environments when used as pavements. The test methodology and test standards were developed in cooperation with the New Jersey Departmentof Environmental Protection and Energy (NJDEPE). The experimental protocol was evolved based on the information and experience of EPA toxicity characteristics leaching procedure (TCLP) and waste stabilizatiodsolidification program, the waste solidification program of the U.S.Army Corps of Engineers, and the nuclear waste research program at Brookhaven and Oak Ridge National Laboratories. The uniform leaching procedure gives an indicationof the amount of each organic compound that is leachable under specific experimental conditions. The structural integrity of sami.e., inthose plewaskeptinthis test, whereparticlesizereductionwasinappropriate, instances where solidification of the waste is needed to meet the best demonstrated available technology provision of environmentallaw. Grinding may not adequately represent the actual process. Particle reduction alters the physical charactermany of solidified wastesby destroying the cementitious property of these wastes in such a way as to show an unrealistically high leaching rate. This water-leachability test was based on the EPA draft on Solid Waste Leaching Procedure (SWLP). This document recommends using laboratory reagent wateras the leaching mebe chosen to mimic the environment where the dium. Chemical and physical conditions should HMA with PCS is used. The leaching medium in this test was chosenas reagent water with a pH value of 6.8 at a temperature of25°C (77°F). Since the asphalt mixturesare used as paving materials, the reagent water should be more representative of the actual environment than an acidic leaching medium. The reagent water was prepared by boiling deionized water for 15 min. Subsequently, while maintaining the water at about 90°C (194"F), nitrogen gas was bubbled through the water for 1 hr. The HMA specimen was treated as a monolithic waste (i.e., the to testing). The asphalt concrete with PCSs were compacted specimen was not pulverized prior into a specially designed stainless steel mold 2 in. X 2 in. X 2 in. before being tested. Since 2-in. cubes were needed for this test, HMA could not be compacted to the same densityas in Marshall tests. However, a loose matrix should produce a higher and conservative leaching rate. HMA specimens were prepared as for the Marshall test, where the blended aggregates were heated to 130°C(268°F). Then heated asphalt cement was added to the dry mix and wet-mixed for 1 min. Then a sufficient quantity of HMA was placed inside the steel molds and compacted with 50 tamps and the HMA specimen was allowed to cool. The compacted HMA specimens with PCSs were placed in 480-mL glass containers and sealed with Teflon septum caps. The volume of the container was at least three times that of the sample,so sufficient space was available in the container and the sample was surrounded on all sides by

672

Meegoda

leaching medium. There was zero headspace to minimize the effects of volatilization. Containers were placed on an Orbit shaker table at 75 rpm for a period of 96 hr. After the end of the shaking period, the containers were opened and a 25-mL water sample was immediately drawn using a gastight syringe. The extracted solutionwas analyzed using an EPA purge-andtrap method. The Purge and Trap Method and GC Settings. The analytical system usedin this test was an ALS 2016 desorber and Tekmar 2000 purge-and-trap concentrator interfacedto a Varian 3400 gas chromatograph (GC) with a high resolution capillary column and a flame ionization detector (FID). The parameters of A L S 2016 were as follows:

Preheat pwe Dry Purge Cooldown Desorb preheat Desorb

Inject Bake

2 min 15 min. helium flow at 40 d m i n 4 min - 150°C 150°C 5 min at 180°C 3 min at 190°C 10 min at 240°C

The column used for GCFID analysis was a cross-linked 5% phenyl and 95% methyl silicone gum, 50 m long, 0.2 mm in diameter, and 0.5 pm film thickness fused silica-bonded high resolution capillary column. Flow rates for the GC were

Hydrogen Air Carrier gas Make-up gas

30 d m i n

30mUmin 1 mUmin 30 d m i n

The GC parameters were

Initial column temperature Initial column hold time Program 1 final column temperature Program 1 column rate Program 1 column hold time Program 2 final column temperature Program 2 column rate Program 2 column hold time Inject temperature Detect temperature

40°C 5 min 65°C 2"Umin 0

190°C 8"Umin 5 min 210°C 250°C

I

Contaminated Highwuy Soils in

Construction

673

Table 5 Method Specifications and Regulatory Limits for Fraction Leached (proposed NJ) in ppb Compound

MDL

Dichloromethane 1 ,ZDichloroethane l,l,l-TCE Hexane Benzene Toluene F’erchloroethene Ethylbenzene p,m-Xylene o-Xylene

0.29 0.47 0.39 0.22 0.26 0.25 0.13 0.28 0.28 0.24

f S,

CV and concentration

Regulatory level

82 f 32% 76 f 21% 73 f 28% 106 f 25% 88 f 16% 94 f 18% 108 f 24% 79 f 17% 95 f 20% 91 f 21%

12.6%112 24.8%126 2 1.4%18 18.5%18 14.3%114 16.8%114 19.7816 19.7%/6 9.6%/5 12.6%15

270 820 110

P

NIA

410 350 100 360 340 340

N/A = not available. Source: Meegoda et al. [9].

An external standard was used for the calibration. The standard consistedof a mixture of targetcompounds in methanol. Theywere dichloromethane.1 ,Zdichloroethane, 1,1,1trichloroethane, hexane, benzene, toluene, perchloroethylene, ethylbenzene, p,m-xylene, and o-xylene. Calibration and Quantification of Data. To determine the precision of the measurements, three injections were made for each concentration level, and the results were analyzedto obtain the mean and the standard deviation. Seven standard samplesof the same concentration were analyzed to obtain the coefficient of variation (CV)to determine the reproducibilityof his analytical method for each compound. The detection limit of the method (MDL) is defined as the minimum concentration of a substance that can be identified, measured, and reported with 99% confidence when the analytic concentration is greater than zero and is determined from analysisof a sample in a given matrix containing analytic compounds. In this test, reagent water was used to estimate the MDL concentration. Standards (analytic in reagent water) at concentrations equal to 1 to 5 times the estimated MDL were used to calculate the MDL. The MDL was reported in concentration units as the standard deviation (SD) of the replicates multiplied by the appropriate Student’s t-value (for a one-tailed test at 99% confidence) for the number of replicates. In this test, the number of replicates was chosen as seven, so MDL was defined as MDL = 3.143

X

SD

Each day, three or four sampleswere spiked withat least 10% of the samples and analyzed to monitor and evaluatethe experimental data quality. The percentage recovery was calculated

using the equation P = 100 (A - B)% / T where A is concentration after spiking, B is the background concentration, and Tis the known true value of the spike. After the analysis of 10 spiked samples,the average percentage recovery( P ) and the standard deviationof the percentage recovery(Sp) values were calculated and expressed as P f Sp. Table 5 shows those values. A major sourceof interference in this test was cross-contamination of samples. To prevent such cross-contamination, reagent water blankswere run twice before running each sampleto

674

Meegoda

Table 6 Leachate Concentration (ppb) from Monolithic HMA Specimens with Pc& Compound Dichloromethane 1 ,2-Dichloroethane l,l,l-TCE Hexane Benzene Toluene Perchloroethene Ethylbenzene p,m-Xylene o-Xylene

Pcs 1

Pcs 2

Pcs 3

Pcs 4

Pcs 5

PcS 6

ND ND ND 0.23 ND ND ND ND ND ND

ND ND ND ND ND ND ND ND 0.32 ND

2.% 0.88 8.01 6.46 3.11 ND 1.19 ND 0.36 0.37

ND 0.16 ND ND ND ND ND ND ND ND

ND 0.18 5.43 2.43 2.19

ND 1S 1 5.14 0.85

ND

ND ND

0.97 ND 0.33 ND

ND ND 1.69 ND

ND=none detected or less than the detection limit. Source: Meegoda et al. [9].

demonstrate that interferences from the analytical system were under control. The following procedure was used to clean the vials and glass containers: 1.

2.

3. 4. 5.

Wash all the vials and containers thoroughly in hot water using a detergentto remove the particulate matter and contaminants. Rinsethoroughlyusingtapwater. Rinsethreetimesusingdeionizedwater. Place in a vacuum oven at 105°C for 12 hr to bake all volatile compounds. Cool for 30 min, screw the lids tightly, and store in an area not subject to contamination by air or other sources.

Test Results. This method can estimate the concentrations of target chemicals up to a minimum concentration of around one-tenth of 1 part per billion (ppb). Table6 shows the testresults. Ethylbenzene did not leach out from any of the six mixes. Benzene, toluene, and xylene 8 ppb. For the HMA mix containing PCS No. concentrations in all the leachates were less than 3 with an initial petroleum concentration of 6600 ppm in the soil, the total leachate concenPCS No. 3 was tration of 10 keychemicals in leachate from solidified and stabilized HMA with less than 25 ppb. The experimental results clearly demonstrate that the HMA with PCSs solidifies and stabilizes most of the petroleum contaminants within the asphalt matrix and the very small concentrations of organic chemicals that may be leaching are from the soil particles that are not completely coated with asphalt cement. To further evaluate the long-term leaching of contaminants, three specimens of HMA and shaken for durations of made with PCS No. 3 were prepared, placed on top of the shaker, 1 day, 1 week, and 1 month. Those test results are shown in Table 7 and are graphically displayed in Figure6 . Results do not show any significant increase in contaminant concentrations with time for any of the compounds tested.

2. Analysis of Air Quality During the Production of HMA with PCS Czarnecki [4] obtained a permit from the Massachusetts Department of Environmental Quality Engineering (DEQE) to process petroleum-contaminated soils subjected to the following conditions: It should process 95% virgin aggregate and 5% PCS, and 95% of the hydrocarbons should be incinerated during the process.To demonstrate the95% destruction, a mass balance of pure chemicals was performed for the incineration system. The test used sand contaminated with a 3% concentration of a 50150 blend of xylene and toluene. Three points in the system

Contaminated Soils in Highway Construction

675

Table 7 Leachate Concentration (ppb) with TimefromMonolithic HMA Specimen with PCS No. 3

eekOne day Compound One 2.33 Dichloromethane 0.81 1 ,2-Dichloroethane 1,l.l-TCE 6.29 Hexane 6.32 Benzene 2.32 Toluene 1.11 Perchloroethene Ethylbenzene 1.65 p,m-Xylene o-Xylene 0.28

1.82 0.82 5.35 6.01 2.19 ND 1.01

ND

3.12 1.03 9.43 6.87 2.29 ND 1.12

ND

ND

ND

1.80 0.30

0.33 0.26

ND=none detected or below detection limit for each compound. Source: Meegoda et al. [9].

10

g

P v

0

"i

." .............-

"'B

"

"

..... .... "

"

"

A

x

X

i

Time (days)

30

Figure 6 Long-term leaching test results. (From Meegoda et

al. [9].)

were selected for analysis: incoming sand from the conveyer belt before the aggregate dryer, sand stripped of contaminants from the storage silo, and air samples from the stack. For three tests, Czarnecki [4] showed 99% destruction of hydrocarbons. Following are the test results from this study; the ai+ sample collection method was not documented:

Incoming sand concentration 30,000.0 ppm Sand coming out from aggregate dryer 0.9 ppm Air samples from the stack 0.2 ppm The measurement of volatile organic compounds (VOCs) has become an important aspect in understanding photochemical reactions and providing an index of hydrocarbons present in the atmosphere. Knowledge of the levels of such materials in the ambient atmosphere isalso

676

Meegoda

required in order to determine human health impacts. However, the determination of toxic organic compounds in ambient air is a complex task, primarily because of the wide variety of compounds of interest and the lack of standards and procedures for analysis. The U.S. Environmental Protection Agency (EPA) has developed several standard analytical methods forthe measurement of volatile organic compounds present in ambient air. Sample collection has been reported as the weakest link in the analytical chainfor the determination of airborne organics, which is critical to the accuracyof the results. The sampling methodsused for measurement of VOCs can be categorizedas adsorption by a solid, cryogenic trapping,or whole air collection. Adsorption by a solid. The solid adsorbents include(1) organic polymers (Tenax, XAD2), (2) inorganic materials (silica gel, florisil), and (3) carbon (activated carbon, carbon molecular sieves). In this technique VOCs are collected on a solid sorbent material while the bulk constituents (e.g., nitrogen, oxygen) are allowed to pass through the sorbent. The VOCs adsorbed are then desorbed, and the sample is injected in a gas chromatograph (GC). Stripping of the adsorbed analytes from the adsorbent is typically accomplishedby either thermal or solvent desorption. Solvent desorption, although useful in many applications, generally requires sample preconcentration before analysis.This adds to the complexity of the final method and increases the sample handling time and the possibilityof contamination. The thermal desorption of analytes has the advantages of reduced sample handling and increasedsensitivity because of the transfer of all analytes onto the chromatographic column. The main advantages of the use of a solid adsorbent for organic polymers are that (1) little water is collected in the sampling process and (2) a large volume of air can be sampled relative to other techniques such as cryogenic sampling. The analytes adsorbed onto Tenax, desorbed, and analyzed using GC/MS can determine volatile nonpolar organics (e.g., aromatic hydrocarbons, chlorinated hydrocarbons) having boiling pointsin the range of 80-200°C [1l]. The PUF/XAD-2 adsorption with GC and HPLC detection can be applied todetermine polynuclear aromatic hydrocarbons. A major disadvantage of these materials is the breakthrough of volatile compounds. Degradation products of the trapping materials are frequently found in adsorbent tubes such as Tenax. Incomplete desorption is also a problem with these methods. Inorganic adsorbents includesilica gel, alumina, florisil, and molecular sieves. Thesematerials are considerably more polar than the organic polymeric adsorbents, leading to the efficient collection of polar materials. Unfortunately, water isalso efficiently captured, leading to rapid deterioration of the adsorbents. Carbon adsorbentsare relatively nonpolar compared to the inorganic adsorbents, and hence water absorption is not a significant problem. Carbon tends to exhibit much strongeradsorption properties than organic polymeric adsorbents, allowing the efficient collection of volatile materials such as vinyl chloride. However, the strong adsorption of carbon adsorbents can be a disadvantage. The desorption of target compounds from the carbon tubes is a common problem. For example, carbon molecular sieves used in EPA Method TO-'2 bind aromatic compounds tightly, and a high temperature (=4OO"C) is required to desorb them. Finally, moisture affects the trapping and desorption efficiencyof the charcoal tubes. The adsorption on carbon molecular sieve followed by desorption and GUMS analysis can be used to determine highly volatile nonpolar organics (e.g., vinyl chloride, vinylidene chloride, benzene, toluene) having boiling points in the range of - 15°C to + 120°C. Cryogenic Trapping. The collection of atmospheric organics by condensation in a cryogenic trap is an attractive alternative to adsorption or impinger collection. The primary adthe collection of a wide range of organic materials, (2) vantages of this technique include(I) avoidance of the contamination problem associated with adsorbents other and collection media, (3) the availability of the sample for immediate analysis without further work, and (4) consis-

Soils Contaminated Construction in Highway

677

tent recoveries. But the disadvantage of this method is that it is suitable only for volatile and nonpolar organics having boiling points the in range of - 10°C to +200"C. This technique provides quantitative concentrations of identified species of lower molecular weights suchas CT C,, compounds typically observed in ambient air.However, an important limitation to this technique is the condensation of large quantities of moisture and carbon dioxide and lesser amounts of certain reactive gases. Whole Air Collection. Collection of whole air samples using stainless steel canisters, evacuated glass bulbs, or similar devices is probably the simplest sampling approach. This approach is most useful for relatively stable volatile compounds such as hydrocarbons and chlorinated hydrocarbons with boiling points below 150°C. The canister samplers have the following advantages compared to solid sorbent tubes: (1) Breakthrough does not occur with canister sampling because the actual air sample is collected; (2) no thermal desorption is required; (3) canister pressure canbe used as an indicator of correct sampler operation; (4) analysisof the canister sample can be repeated by using the remainder of the sample in the canister; and (5) the evacuated canisters can be used for sampling without power in the sampling location. SUMMA passivated canister sampling with GC was developed by USEPA to determine semivolatile and volatile organic compounds. The canister is a sampling deviceused to collect and store whole air samples. It can be pressurized, thereby increasing the volume of the collected air sample. It has been demonstrated that the SUMMA passivation process, in which a pure chrome-nickel oxide layer is coated on the inner metal surface, increases the stability and the storage life of many organic compounds. However, certain compounds pose stability problems associated with storageby the formation of an oxide coating. Selection of an alternative container material can circumvent these problems in many cases. The most difficult problem associated with this method is the quantity of moisture collected in the canister. Too much moisture can clog the cryogenic trap and the capillary interface.

GC Due to the complexity of ambient air samples, only high resolution (capillary column) techniques are acceptable for most of the above methods. The GC/MS system should be capable of programing subambient temperatures. Unit mass resolution shouldbe better than 800 amu, and GC/MS shouldbe capable of scanning the 30-440 arnu region every 0.5-1 sec. The measuring device shouldbe equipped with a data system for instrument controlas well as data acquisition, processing, and storage. Impinger bubblershave traditionally beenused for the collection of volatile organics from air. They are charged with a trapping solvent and require much more sampling time for collecting sufficient amounts of the analyte. The growing demand forgreater sensitivity and time resolution suggestedthe development of a system based on solid adsorbent sampling and thermal desorption of VOCs followed by gas chromatographic analysis with cryogenic refocusing techniques. An air sampling and analysis systemfor monitoring VOCs being emitted from the stacks of asphalt plants was designed. It was based on solid adsorbent sampling tubes with thermal desorption GC analysis using capillary column separations with cryogenic refocusing techniques. Air samples were collected using the solid adsorbent Tenax in stainless gas collection . tubes. Stripping of the adsorbed analytes from the sampling tube was accomplished by thermal desorption and followed by gas chromatographic analysis. Preparation of Tenux Cartridges. The following procedure was usedto prepare a 5/8 in. tube containing Tenax cartridges. Prior to use, the Tenax resin was subjected to a series of solvent extraction and thermal treatment steps. The operation was conducted in an area where levels of volatile organic com-

Meegoda

678

pounds (other than the extraction solvents used) are minimal. All glassware usedinTenax purification and all cartridge materials were thoroughly cleanedby rinsing with water followed by rinsing with acetone and drying in an oven at 250°C. The bulk Tenax was placed in a glass extraction thimbleand held in place with a plug of clean glass wool. The resin was then placed in the Soxhlet extraction apparatus and sequentially extracted with methanol and then withpentane for 16-24 hr for each solvent for approximately 6 cycleshr. Glass wool for cartridge preparation was cleaned in the same manner as that for Tenax. The extracted Tenax was placed immediately in an open glass dish and heated under an infrared lamp for 2 hr inside a hood. Care was exercised to avoid overheating the Tenax with the infrared lamp. The Tenax was then placed in a vacuum oven (evacuated using a water aspirator) without heating for 1 hr. Then it was purged with an inert gas (helium or nitrogen) at a rate of2-3 m u m i n to aid the removal of solvent vapors. Then the oven temperature was increased to 110°C while maintaining the flow of inert gas for 1 hr. Then oven temperature control was shut off, and the oven was allowedto cool to room temperature. Before opening the oven, it was slightly pressurized with nitrogen to prevent contamination with ambient air. The Tenax was removed from the oven and sieved through a 4 / 6 0 mesh sieve (acetone rinsed and oven-dried) into a clean glass vessel. in a clean glassjar with a Teflon-lined screw If the Tenax was to be used later, it was stored cap and placed insidea desiccator. The cartridge used for the monitoringof air was packed by placing a 0.5-1 cm long glass wool plug at the bottom of the cartridge and then filling the cartridge with Tenax up to approximately 1 cm from the top. Then a 0.5-1 cm long glasswool plug was placed on top ofthe Tenax. Thecartridges were thermally conditionedby heating for 4 hr at 270°C while purging with an inert gas (helium at 100-200 mumin). The Desorption and GC Settings. The air sampling and analysis system was based on solid adsorbent sampling tubes and thermal desorption gas chromatographic analysis. After collection of air samples, sampling tubes were put into a desorber. Then the VOCs were twice condensed in cryogenic trap to improve capillary column resolution. Before cryogenic refocusing, a purge step served to remove any oxygen that remained inthe tube. This eliminated the problem of the solid adsorbent reactingwith the oxygen when heated, and it also removed traces of water from the tube. Collected air samples in Tenax traps were first desorbed using a Tekmar Model 5010 automatic desorber connected to a Varian 3400 high resolution capillary GC with a flame ionization detector. The two systems were interfaced to automate the entire analysis. The desorption conditions are as follows.

Prepurge Desorb

Cryotrap 1 Cryotrap 2 Transfer Inject

5 min at 10 mUmin 8 min, 210°C, 10 mUmin 150°C 150°C 10 min. 210°C 0.5 rnin, 210°C

The column used for GC/FID analysis was a cross-linked 5% phenyl and 95% methyl silicone gum, 50 m long, 0.2 mm diameter, and 0.5 pm film thickness fused silica bonded high resolution capillary column. Flow rates for the GC were

Contaminated Soils in Highway Construction

679

30 mUmin 30 mUmin, 1 mUmin

Hydrogen Air

Carrier gas Make-up gas

30 mUmin

The temperature program was (1) an initial temperature of 15°C for 8 min and (2) temperatures programed up to 210°C at 4"CImin. The detector range was selected as 10. A schematic diagram of the analytical system is shown in Figure 7. Calibration and QuantificationofData. External standards were used to calculate a response factor for each compound of interest. The process involvedthe analysis of four calibration levels for each compound during a given day for the determination of the response factor (area/ picomole). The linear least squares fit of a plot of picomoles versus area was used for the determination of the response factor. If substantial nonlinearity is present in the calibration curve, a nonlinear least squares fit should be employed. This process involvesfitting the data to the equation

Y

=

A

+ BX + DX2

where Y is the peak area, X is the quantity of the component in picomoles, and the constants

A, B , and D are coefficients to be determined fromthe regression analysis. If the instrumental response is linear over the concentration rangeof components, a linear equation (D = 0 in the equation above) can be employed. The systemdetection limit for each component can be calculated from the calibration standards, where the detection limit is defined as

DL

=A

+ 3.3SD

where DL is the calculated detection limit in picomoles, A is the intercept of the regression analysis equation, and SD is the standard deviation of the samples with replicate determinations of the lowest concentration level. The standard gas used for calibration consisted of a mixture of chloromethane, hexane, chloroform, 1,1, l-trichloroethane, carbon tetrachloride, trichloroethene, toluene, trichloroethylene, benzene, mpxylene, and o-xylene. These compoundswere injected into an evacuated and clean 13-L stainless steel cylinder, and the cylinder was pressurized with zero grade helium. The preparation and analysis of the standard was done by Alphagaz, Edison, New Jersey. A 2-mL volume sampleloop was used for the GC analysis. When the sample loop wasflushed, the standardgas mixture passed throughthe sample loop. A three-port valve witha switch was used to allow helium gas to pass throughthe loop, flushing the standard gas onto a Tenax cartridge. The compoundsin the standard gas were adsorbed by Tenax and quantifiedby gas chromatography. The following ideal gas equation was used for the calculation of the component concentrations in the standard gas. n = PVC/(R/T) where P is pressure (inatm), V is the volume of the sample loop, C is the concentration of each compound expressed as a fraction, R is the universal gas constant (0.082), Tis temperature (in kelvins), and n is the amount of each standard gas injected (in moles). At 25"C, sample concentration in ppb can be expressed as Sample concentration =

area of sample X C X 24.5 X lo9 area of the xv

Meegoda

680

Automatic Desorber Tenax trap with contaminants

i

IOpsig, I mVmin and then desorborganics

Printer

t Computer

t GC Server 4 A

i

,

,..., .~,...

.,

Air, 6Opsig 30 d m i n

Hydrogen, 4Opsig 30 mVmin ' , -

.

. . . .

Nitrogen, 8Opsig 30 mumin ;make up for capillary column

...........

:

*

.....................................: *

High Resolution CapillaryG a s Chromatograph Figure 7 A schematic diagramof the analytical system usedfor the air quality analysis. (From Mesoda et al. [g].)

where C is concentration of the compound in standardgas and V is the volume of the sample in liters. If V is at a different temperature and/or pressure, then the volume of gas should be converted to that at 25" and a pressure of 1 atm. The air samples were collected by drawing air through theTenax cartridge, a 5/8 in. stainless steel tube packed with 1.5 g of 60/80 mesh Tenax using a vacuum pump. Figure 8 shows air samples from stacks. Samples were drawn at approximately 500 the schematic for extracting &min. In this experiment, 10 Tenax blank traps were spiked with known quantities of standard and then desorbed into the analytical system. The reproducibility can be expressed by the coefficients of variation(CV) of the target compounds. The CV andDL values for all the tested organic compounds are shownin Table 8.

Contaminated Soils in Highway Construction

681

Tenax Cartridge

Meter

L

x

l

+

Vacuum Pump

Valve

Reducing Union

I

.

/

l Tenax

EndCap

\

Glass Wool

Swagelok Fitting

Vent

Metal Cartridge

Figure 8 Schematic diagrams of the setup used for air quality analysis. Prom Meegoda et al. [8].) Test Datu. An air sampling and analysis system was designed to monitor volatile organic compounds emitting from stacks of asphalt plants. It was based on solid adsorbent sampling tubes with thermal desorption gas chromatographic analysis and used capillary column separations with cryogenic refocusing techniques. Air samples were collected using the solid adsorbent Tenax in stainless gas collection tubes. Stripping of the adsorbed analytes from the sampling tube was accomplishedby thermal desorption and followedby gas chromatographic analysis. During the field study, air samples were drawn for3 min from the stackin the asphalt plant PCS was prepared.The air samples were drawn through while the HMA concrete blended with a filter to remove the dust going into the Tenax cartridge. The Tenax cartridge was connected 3 min for test, and the Tenax samples to a vacuum pump. Three air samples were obtained for soils were analyzed for concentrations of target organic compounds. Test results (firsttwo day, are shownin withfourdifferentblends;secondday,onesoilwiththreedifferentblends) Table 9.

Meegoda

682

Table 8 Method Specifications and Regulatory Limits for Fraction Released (Proposed N.J.) (ppb)

(PPW

DL Compound 270 Chloromethane NIA Dichloromethane NIA Hexane 6 Chloroform 494 1,1,1 -Trichloroethane 13.8 Trichloroethylene 16.7 Benzene 988 Toluene 9.9Tetrachloroethylene 148p,m-Xylene 148o-Xylene

Regulatory level as average flux (mg/m2* day)

CV

0.54 0.13 0.03 0.44 0.22 0.05

0.07 0.06 0.03 0.05

0.06

20.5% 9.2% 11.8% 9.8% 10.5% 21.4% NIA 6.7% 8.7% 7.5% 9.5%

N/A =not available.

Areas of all peaks were added to computethe total nonmethane organic carbon (NMOC) emissions. A standard propane/heliumgas mixture was usedto obtain the GC calibration curve for concentration. Various known amounts of the mixture were injected to the GC and the peak areas were obtain to establish the calibration curve. Then the concentration corresponding to sum of all peak areas was calculated from thecalibration curve and reported as total NMOC in Table 9. Figure 9 shows a comparison of totalNMOC emission from a regular asphalt plant and that uses PCSs (day 2 test 3 in Table 9). The concentrations of the target chemicals were less than 1 ppm, and total concentrations of VOCs were lower than the NJDEPE specification for regular asphalt plants (c250 ppm). Table 9 AirQualityTestResults

Concentration in ppb

otal Cs Hexane 0-Xylene Xylene P&M Toluene No. Test 8

1

ant

2 3 4 2 2 2 plant around

Day581 Test 1 52 Day 1 Test Day 1 Test Day 1 Test Day Day Day Conc. levels 2,650 149 after stopped Background conc. Stack conc. from an ordin. HMA plant without PSCs

52 90

69

53 14,880

99155,230 11 4

2

25

2

24

16

Source: Meegoda et al., 1992 [9].

6

11 283

0

1,210

0

26,700

Soils Contaminated Highway

in

Construction

683

NMOC

100 I

80

60

40

20

n "

Normal HMA

=

PPmC

IIMA with PCSs I b Carbonlhr

Figure 9 Comparison of non-methaneorganiccarbon (NMOC)emission by aregularasphaltplant and one that uses PC%. Solid bars, ppm carbon; hatched bars, pounds of carbon per hour. (From Meegoda et al. [ g ] . )

Furtherdetailscan be fohndinthe 1991 report to NJDEPE ontheuseofpetroleumcontaminated soils in construction material production [ 121.

111.

SUMMARY AND CONCLUSIONS

Leaking underground storage tanks (USTs) are one of the primary sources of groundwater contamination in the United States. The soils contaminated by leaking USTs are treated as solid waste. The quantities of such soils are projected to increase substantially over the next few years. In this report the feasibility of using petroleum-contaminated soil (PCS) in the production of asphalt concrete is discussed. When F'CSs are used in asphalt production, three beneficial actions occur: incineration, dilution, and solidification. An in-depth laboratory study was performed to determine the feasibility of producing HMA with PCSs. For the heating oil-contaminated soil used in this study, it was possible to include upto 35% PCS based on the total weight of the aggregates in the HMA mix. This value is much higher than any reported in the literature. The impact with respect to strength and durability of asphalt concrete dueto the additionof PCS to HMA was also evaluatedby performing the Marshall stability test. The test results showed that HMA with PCS produced a much better paving material than the control. The extended durability test showed that one cycle of freeze-thaw and wet-dry was sufficient to evaluate the durability of HMA with PCS. The durability of HMA produced with PCS was found to be the same as that of the control mix,

684

Meegoda

suggesting that there are no harmful effects from the addition ofPCS. The hydraulic conductivity of HMA with allPCSs produced asphalt concrete with hydraulic conductivities less than 2.0 x A new leachability test was designed for the monolithic sample. The leachability of hydrocarbons based on the EPA purge-and-trap method showed that maximum release was less than 25 ppb after the specimens were immersedin analytical water and subjected to4 days of vigorous shaking. To further evaluate the leaching of contaminants, a long-term leaching test was performed. The test results show no significant increase in contaminant concentrations with time for all the compounds tested. An extensive field studywas conducted to demonstrate the applicability of this process. An air sampling and analysis system was designed for monitoring volatile organic compounds being emitted from asphalt plant stacks. It was based on solid absorbent sampling tubes with thermal desorptiongas chromatographic analysis using capillary column separations with cryogenic refocusing techniques. The concentrations of the target chemicals were less than1 ppm, and the concentrations ofVOCs were lower than the NJDEPE specification for regular asphalt plants (