Barrier systems for environmental contaminant containment and treatment

The USEPA’s Alternative Cover Assessment Program (ACAP) is also moni- toring the percolation rate from seven caps employing composite barrier layers consisting of a geomembrane underlain[r]

SYSTEMS for

ENVIRONMENTAL CONTAMINANT CONTAINMENT

and TREATMENT

A CRC title, part of the Taylor & Francis imprint, a member of the

Taylor & Francis Group, the academic division of T&F Informa plc.

Boca Raton London New York

BARRIER

SYSTEMS for

ENVIRONMENTAL CONTAMINANT CONTAINMENT

and TREATMENT

Edited by

Calvin C Chien • Hilary I Inyang

Lorne G Everett

Published in 2006 by

CRC Press

Taylor & Francis Group

6000 Broken Sound Parkway NW, Suite 300

Boca Raton, FL 33487-2742

© 2006 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group

No claim to original U.S Government works

Printed in the United States of America on acid-free paper

10 9 8 7 6 5 4 3 2 1

International Standard Book Number-10: 0-8493-4040-3 (Hardcover)

International Standard Book Number-13: 978-0-8493-4040-6 (Hardcover)

Library of Congress Card Number 2005047215

This book contains information obtained from authentic and highly regarded sources Reprinted material is

have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers

Danvers, MA 01923, 978-750-8400 CCC is a not-for-profit organization that provides licenses and registration for a variety of users For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Barrier systems for environmental contaminant containment and treatment / contributing editors, Calvin C Chien, Hilary I Inyang, Lorne G Everett ; prepared under the auspices of U.S Department of Energy, U.S Environmental Protection Agency, DuPont.

p cm.

Includes bibliographical references and index.

ISBN 0-8493-4040-3 (alk paper)

1 In situ remediation 2 Sealing (Technology) I Chien, Calvin C II Inyang, Hilary I III Everett, Lorne G IV United States Dept of Energy V United States Environmental Protection Agency VI E.I du Pont de Nemours & Company.

TD192.8.B375 2005

Visit the Taylor & Francis Web site at

and the CRC Press Web site at

Taylor & Francis Group

is the Academic Division of Informa plc.

4040_Discl.fm Page 1 Monday, September 26, 2005 11:08 AM

( http://www.copyright.com/ ) or contact the Copyright Clearance Center, Inc (CCC) 222 Rosewood Drive, quoted with permission, and sources are indicated A wide variety of references are listed Reasonable efforts

http://www.taylorandfrancis.com

http://www.crcpress.com

For permission to photocopy or use material electronically from this work, please access www.copyright.com

Contributing Editors

Calvin C Chien, Ph.D., P.E.

DuPont FellowDuPontWilmington, Delaware

Hilary I Inyang, Ph.D.

Duke Energy Distinguished Professor and Director,Global Institute for Energy and Environmental SystemsUniversity of North Carolina, Charlotte, North Carolina

Lorne G Everett, Ph.D., D.Sc.

President

L Everett and Associates, LLCSanta Barbara, CaliforniaPrepared under the auspices ofU.S Department of EnergyU.S Environmental Protection AgencyDuPont

With contributions by renowned experts on waste containment and waste ment science and technology

treat-20054040_C000.fm Page v Wednesday, September 21, 2005 4:38 PM

Technical Review Board

David E Daniel, Ph.D., Overall Book Reviewer

University of IllinoisUrbana-Champaign, Illinois

Craig H Benson, Ph.D., P.E.

University of WisconsinMadison, Wisconsin

Significant advances in subsurface containment technology occurred in the 1990s,both with the improvement of the technology and the broader acceptance andapplications as a measure for environmental remediation Since 1995, the U.S.Department of Energy (USDOE), U.S Environmental Protection Agency (USEPA),and DuPont have collaborated on a series of organized efforts to advance thistechnology In that year, these collaborators sponsored an international expertworkshop that led to the publication of the first major book on containmenttechnology Two international conferences were held by the same three partners

in 1997 and 2001, with individuals from all over the world attending

Although subsurface containment technologies are becoming increasinglyacceptable and popular in the environmental remediation field, questionsremained on the prediction and verification of long-term barrier performance andthis subject began to gain interest from the public, government agencies, and theU.S Congress With funding provided by USDOE, an executive committee, con-sisting of Skip Chamberlain (Chairperson, USDOE), Calvin C Chien (DuPont),and Annette M Gatchett (USEPA), was formed in October 2001 to plan andorganize an expert workshop Sixty invited international experts participated Themeeting was held between June 30 and July 2, 2002 in Baltimore, Maryland, andconsisted of five discussion panels — three on prediction and two on verification.Each panel was led by a panel leader and a co-leader to address particulartechnical topics in a designated area A designated graduate fellow, a graduatestudent whose research was related to these topics, recorded detailed notes forthe panel discussions The graduate fellow group was coordinated and supervised

by Jada M Kanak (DuPont) Each panel leader, assisted by the co-leader, wasresponsible for writing a chapter for this book, using the information generatedfrom the panel discussions and the detailed notes recorded by the graduatefellows The prediction chapters were reviewed and edited by Hilary I Inyang, andLorne G Everett reviewed and edited the verification chapters Calvin Chien hadthe responsibility for planning, coordinating, and editing the book, ensuring con-sistency and completeness, and resolving differences in opinions Skip Chamberlainprovided technical input and crucial support in working with experts from thenational laboratories on critical issues during the preparation of the book David E.Daniel (University of Illinois) conducted an initial review of the first draft andprovided high-level comments, which were useful in performing subsequentrevisions Dr Daniel also wrote the preface for the book, which provides anoutstanding introduction of containment technology history and book structure.Relevant new information that became available during the period of preparation4040_C000.fm Page xi Wednesday, September 21, 2005 4:38 PM

and editing was identified, evaluated, and added to the book to ensure that theinformation is as up-to-date as possible.

In addition to organizing and leading the graduate fellow group, Jada Kanakalso served as a special technical assistant for book preparation Her detailed andpatient efforts in reviewing and checking all of the references, figures, and tablescontributed greatly to the quality of this book Ms Kathy O Adams, a long-timeDuPont in-house contract technical writer, was responsible for ensuring the gram-matical accuracy of the book, and did an excellent job polishing the final draft.The team from Florida State University, consisting of Norbert Barszczewski,Sheryl A Grossman, Loreen Y Kollar, J Michael Kuperberg, and Laymon L.Gray, were responsible for the workshop planning and contributed greatly to thesuccess of the meeting

4040_C000.fm Page xii Wednesday, September 21, 2005 4:38 PM

The containment of buried waste, contaminated soil or groundwater, refers to

achieved with individual barriers or control technologies that, together, provide

a system of engineered control Containment is potentially applicable to anycircumstance in which contaminants exist in the subsurface (e.g., uncontrolledlandfills or dumps, chemical spills or leaks, pond or lagoon contaminant seepage)and can provide a safe and highly cost-effective mechanism for environmentalcontrol Containment is accomplished using physical, hydraulic, or chemicalbarriers that prevent or control the outward migration of contaminants

Containment has come full circle as an acceptable environmental controltechnology over the past 30 years Prior to the 1980s, containment was virtuallythe only technology available for managing subsurface contamination Althoughsome wastes were exhumed and treated, more often than not, if the pollutionproblem was recognized at all, the problem was managed via containment Duringthe 1980s, new environmental regulations emphasized treatment rather than con-tainment Research and development during this time dramatically expanded theportfolio of options available for treating or destroying contaminants at pollutedsites Technologies such as vapor extraction, oxidation, bioremediation, surfactantflushing, and heat-induced treatment became viable, though often expensive,treatment alternatives

In the 1990s, a dose of reality swung the pendulum back toward containment

It became apparent that it was not technically feasible to return contaminatedsites to pristine condition Further, as a nation, the United States came to realizethat it could not afford, nor did it need, the most sophisticated treatment technol-ogy available to manage pollution problems at every site effectively and safely

In addition, further research clearly showed that the subsurface has advantages

in addressing contamination problems — natural processes such as adsorptionand biodegradation can serve to contain or degrade contaminants For certainmaterials such as radioactive wastes, it became apparent that the exposure risksassociated with exhuming contaminants might be far greater than risks associatedwith managing the wastes in situ with containment Thus, for many reasons,interest in containment was revived in the 1990s Today, containment thrives as

a viable environmental management technology, and is often the preferred choicefor protecting human health and the environment

But a price was paid for putting containment “on hold” during the 1980s,when emphasis was placed on developing sophisticated treatment technologies:little research and development on containment technologies was achieved during4040_C000.fm Page xiii Wednesday, September 21, 2005 4:38 PM

this time As interest shifted back toward containment in the 1990s, the industryfound itself relying largely on pre-1980s technology Fortunately, in the past

10 years, important advances have occurred in several areas of containment, mostnotably in the area of permeable reactive barriers, which transform containmentbarriers into a passive treatment installation

In the early 1990s, the need to define the state of the art for containment wasunderstood by three visionary organizations: DuPont, the U.S EnvironmentalProtection Agency, and the U.S Department of Energy The DuPont CorporateRemediation Group (CRG) initiated the trio’s first collaborative effort in 1992.Experts from four nations experts were invited by DuPont to work with a team

at the State University of New York at Buffalo to conduct a comprehensive review

of the containment technology, the technology gaps, and future direction Theproduct of the work, a 1993 internal report, was published in 1995 by John Wiley &Sons, New York, titled Barrier Containment Technologies for Environmental

The principal chapters of the book focused on vertical barriers (walls), bottombarriers (floors), and surface barriers (caps) The three organizations joined againand organized an expert workshop on containment technology in 1995, inviting

115 international experts The book, Assessment of Barrier Containment

was edited by Ralph R Rumer and James K Mitchell and was published thenext year

With the rapidly increasing use of barrier technology in remediation, the needfor better understanding, prediction, and monitoring of the performance of bar-riers emerged The trio organized another expert workshop on the topic in 2002,which led to the development of this book The workshop planning committeeinvited many of the world’s most knowledgeable researchers and practitioners todiscuss the current state of the art and debate the appropriate applications anddirections for containment The participants then went home and collectivelycreated this book from their knowledge and exchanges This book is essentially

a diary of those discussions and assessments, recast into the form of an easilyreadable, comprehensive book that is rich with discussion and references toliterature, as well as further detail on specific topics of interest The first two

discussions in the first four chapters address caps, vertical walls, and permeablereactive barriers

how contaminants can get into the subsurface This is an important chapter,because one cannot understand how to contain something unless one knows howthe contaminants got into the subsurface in the first place, and how they mightspread and threaten the environment without containment This chapter not onlydescribes pathways, but also introduces the essential concept of risk No controltechnology is without risk Ultimately, a low risk of adverse environmental impact

4040_C000.fm Page xiv Wednesday, September 21, 2005 4:38 PM

chapters address prediction issues, Chapters 3 and 4 address monitoring niques, and Chapter 5 addresses the largely undeveloped field of verification The

tech-Chapter 1, “Damage and System Performance Prediction,” sets the stage for

draws from concepts in reliability of structures, and couples barrier structuralfailure to functional failure Relevant quantitative frameworks are presented foruse in assessing the long-term performance of containment systems.

basis for predicting the transport of water and contaminants through barriercomponents This chapter focuses on modeling the inflow of moisture to theburied waste (e.g., caps), or modeling the release of contaminants throughsubsurface barriers Fluid transport rate prediction is essential to the designprocess, because predictions can be integrated into the overall containment systemdetails on the current state of the art for performance prediction, but also clearlydelineates the limitations in modeling specific situations

in barriers, defining the properties of barrier materials and exploring how rials perform in the field The materials used for barriers include a myriad ofnatural and man-made materials, such as natural soil, stones and cobbles, imper-meable plastic lining materials, man-made filter fabrics, and chemical agentsdesigned to sorb or degrade contaminants that might come in contact with thematerial Factors such as clogging, deterioration, or alteration of physical, chem-ical, or hydraulic properties are explored, not only to define what is known aboutthese materials, but also to provide a learned and balance sense of what is notknown

mate-a thorough description of the mate-applicmate-ation of geophysicmate-al methods to subsurfmate-acebarriers Geophysical methods have been used widely to assist in identifyingpotential mineral resources deep within the subsurface, and in more recent years,

in the shallow environment, to help with identifying contaminant plumes andother anomalies When applied to subsurface barriers, geophysical methods arechallenged beyond their traditional role of identifying gross features that mightwarrant more detailed exploration (e.g., via a borehole), toward identifying moresubtle features, such as a leak in a subsurface barrier The techniques described

in this chapter include both near- and far-field devices, spanning equipmentdeployed in aircraft flying above a site to devices placed on the ground surfacethat probe the subsurface directly with electromagnetic or other sources of energy.The first half of this chapter describes the technologies that are available, and thesecond half addresses their applications to various types of barriers

the most challenging aspect of waste containment technology, i.e., validation offield performance Traditionally, monitoring has consisted of sampling of ground-water or soil gas from wells Although sampling soil, water, and air can provideinformation about the general performance of a system, it does not provide imme-diate, specific information about how a particular barrier component is meetingits design goals Further, there is little to motivate stakeholders to spend money

4040_C000.fm Page xv Wednesday, September 21, 2005 4:38 PM

should be maintained in a way that uses resources as wisely as possible Chapter 1

Chapter 2, “Modeling of Fluid Transport through Barriers,” addresses the

performance assessment scheme presented in Chapter 1 Chapter 2 provides

Chapter 3, “Material Stability and Applications,” addresses the materials used

Chapter 4, “Airborne and Surface Geophysical Method Verification,” provides

The subject of Chapter 5, “Subsurface Barrier Verification,” tackles perhaps

for performance verification, unless required for compliance with regulations.This chapter provides a comprehensive review of sensors and examples of howsensors can be used to document system performance, addressing the basicquestions: where, what, how, and what-if? Ultimately, the performance verifica-tion scheme should be linked to the performance prediction process It is perhapsthis linkage that is our most important end point, and one that requires morework, particularly in terms of assessing reliability and risk associated with theuse of waste containment as a technique for managing waste in the subsurface.The two well-known case studies in the United States that are presented in thischapter provide particular value to this need.

That which is buried in the subsurface, out of sight and out of mind, is thatwhich in some respects is the most challenging Nature has placed geologicmaterials in the subsurface in rather unpredictable and unknowable locations,with properties that are difficult to discern Individual barriers are constructed inmore controlled and documented ways, but still with considerable uncertainty inactual characteristics Systems comprised of multiple barriers enjoy considerableredundancy and tend not to rely on any single component for success Scientistsand engineers strive to understand, predict, design, and verify safe containmentschemes, both in terms of individual barriers and more complex containmentsystems This book provides a comprehensive report on the science and technol-ogy of waste containment, with a balanced presentation of what is and is notknown Subsurface containment will continue to be a widely used environmentalcontrol technology in the years ahead This book will provide a valuable reference,helping to chart the way to successfully managing many contaminated sites

David E Daniel

University of Illinois Urbana, Illinois

4040_C000.fm Page xvi Wednesday, September 21, 2005 4:38 PM

Calvin C Chien is a DuPont Fellow, one of only 13 individuals serving in thiscapacity in DuPont He has been working in the area of groundwater investigationand remediation since 1975 Since 1991, he has been responsible for evaluatingand developing transport modeling and containment technologies As such,

Dr Chien has played a leading role in improving the understanding of ment technology for use in environmental remediation He orchestrated the FirstInternational Expert Workshop (1995) and the publication (based on the work-shop) of the first comprehensive containment book: Assessment of Barrier Con- tainment Technologies: A Comprehensive Treatment for Environmental Remedi-

technology: the First International Containment Technology Conference Throughthese efforts, he has been recognized as a leading contributor to improving thescience of containment technology as well as its acceptance at the regulatorylevel He has authored and co-authored many technical papers for peer-reviewedjournals and books Currently, Dr Chien provides technical environmental sup-port and oversight for existing and new DuPont operations in the Asia-Pacificregion His contributions in the region led the Chinese Ministry of Science andTechnology to invite him to evaluate candidates for the 2005 State Natural ScienceAward of the People’s Republic of China This award is the most prestigiousaward for scientists and engineers in China

Hilary I Inyang is the Duke Energy Distinguished Professor of EnvironmentalEngineering and Science, Professor of Earth Science (GIEES), and Director ofthe Global Institute for Energy and Environmental Systems at the University ofNorth Carolina–Charlotte From 1997 to 2001, he was the Chair of the Environ-mental Engineering Committee of the U.S Environmental Protection AgencyScience Advisory Board, and also served on the Effluent Guidelines Committee

of the National Council for Environmental Policy and Technology He hasauthored and co-authored more than 170 research articles, book chapters, federaldesign manuals, and the textbook Geoenvironmental Engineering: Principles and

an associate editor and editorial board member of 17 refereed international nals, and contributing editor of three books, including the United Nations Ency-

served on more than 85 international, national, and state science/engineeringpanels and committees Since 1995, he has co-chaired several international con-ferences on waste management and related topics, and given more than 100 invited4040_C000.fm Page xvii Wednesday, September 21, 2005 4:38 PM

speeches and presentations on a variety of technical and policy issues at tions and agencies globally Professor Inyang holds a Ph.D with a double major

institu-in Geotechnical Enginstitu-ineerinstitu-ing and Materials, and a minstitu-inor institu-in Minstitu-ineral Resourcesfrom Iowa State University, Ames; a M.S and B.S in Civil Engineering fromNorth Dakota State University, Fargo; and a B.Sc (Honors) in Geology from theUniversity of Calabar, Nigeria His research has been sponsored by several agen-cies and corporations Dr Inyang’s research accomplishments and contributions

to geoenvironmental science and engineering have been rewarded with honors

by various national and international agencies among which are Fellow of theGeological Society of London; 2001 Swiss Forum Fellow selection by the Amer-ican Association for the Advancement of Science; 1991 Chancellor’s Medal forDistinguished Public Service awarded by the University of Massachusetts Lowell;and the 1992/93 Eisenhower Fellowship of the World Affairs Council to com-memorate the international achievements of the late U.S President Dwight Eisen-hower In 1999, Prof Inyang was appointed to Concurrent Professorship ofNanjing University, China and subsequently selected as an Honorary Professor

of the China University of Mining and Technology, Jiangsu, China He is thePresident of the International Society of Environmental Geotechnology (ISEG)and the Global Alliance for Disaster Reduction (GADR)

Lorne G Everett is the 6th Chancellor of Lakehead University in Canada,President of L Everett and Associates LLC, Santa Barbara, a Research Professor

in the Bren School of Environmental Science & Management at UCSB (LevelVII), and Past Director of the University of California Vadose Zone MonitoringLaboratory The University of California describes full professor Level VII as

“reserved for scholars of great distinction.” He has a Ph.D in Hydrology fromthe University of Arizona in Tucson, and is a member of the Russian Academy

of Natural Sciences In 1996, he received a Doctor of Science Degree (HonorisCausa) from Lakehead University in Canada for Distinguished Achievement inHydrology In 1997, he received the Ivan A Johnston Award for OutstandingContributions to hydrogeology In 1999, he received the Kapitsa Gold Medal —the highest award given by the Russian Academy for original contributions toscience In 2000, he received the Medal of Excellence from the U.S Navy, andthe Award of Merit, the highest award given by American Standards and TestingMaterials (ASTM) International In 2002, he received the C.V Theis Award, thehighest award given by the American Institute of Hydrology (AIH) for majorcontributions to groundwater hydrology In 2003, he received the CanadianGolden Jubilee Medal for “Significant Contributions to Canada.” He is an inter-nationally recognized expert who has conducted extensive research on subsurfacecharacterization and remediation Dr Everett has published over 150 technicalpapers, holds several patents, developed 11 national ASTM vadose zone moni-toring standards, and authored several books, including Vadose Zone Monitoring

book, entitled Handbook of Vadose Zone Characterization and Monitoring, is a4040_C000.fm Page xviii Wednesday, September 21, 2005 4:38 PM

best seller His book Groundwater Monitoring was endorsed by the U.S ronmental Protection Agency as establishing “the state-of-the-art used by industrytoday,” and is recommended by the World Health Organization for all developingcountries.

Envi-4040_C000.fm Page xix Wednesday, September 21, 2005 4:38 PM

Table of Contents

1.2.1 1.2.2 Types of Performance Prediction Approaches 11

1.2.2.1 Empirical Prediction Approaches 11

1.2.2.2 Semi-Empirical Prediction Approaches 12

1.2.2.3 Less Empirical (Theoretical) Modeling Approach 14

1.3 Relationship of Structural Failure to Functional Failure 15

1.3.1 Economic or Pseudo-Economic Criteria 18

1.3.2 Regulatory Criteria 19

1.3.3 Prescriptive Design Criteria 19

1.3.4 Risk Criteria 20

1.3.5 Demonstrating Compliance: The Safety Case Concept 22

1.3.6 Mixed Criteria 23

1.3.7 Qualitative and Indexing Analyses 23

1.4 Quantification of Long-Term Damage Scenarios, Events, and Mechanisms 24

1.4.1 Categories of Degradation Mechanisms 24

1.4.1.1 Slow Physico-Chemical and Biological Processes 24

1.4.1.2 Intrusive Events 29

1.4.1.3 Transient Events 30

1.4.1.4 Cyclical Stressing Mechanisms 32

1.4.2 Quantitative Linkage of Contaminant Release Source Terms to Risk Assessment and Compliance Limits 37

1.4.3 Frameworks for Assessment of Event Consequences and Connectivities Among Causes of Failure 42

1.4.3.1 Fault Trees 42

1.4.3.2 Event Trees 42

1.4.4 Estimation of Long-Term Failure Probabilities 42

1.4.4.1 System Failure Probability 43

1.4.4.2 Component Failure Probability 44

1.4.4.3 Random Resistance 47

1.4.4.4 Simplifications of Theory 48

1.4.4.5 The Multi-Dimensional Case 51

1.4.5 Component and System Failure in Containing Contaminants 53

1.4.6 Relating Probable Contaminant Concentrations to Risks 54

4040_C000.fm Page xxi Friday, September 23, 2005 4:37 PM Chapter 1 Damage and System Performance Prediction 1

Hilary I Inyang and Steven J Piet 1.1 Overview 1

Concepts and Analytical Framework 8

1.2 Long-Term Performance Analysis Framework 7

1.5 Use of Barrier Damage and Performance Models for Temporal

Scaling of Monitoring and Maintenance Needs 59

1.5.1 Updating 59

1.5.2 Effect of Updating on System Management 60

1.6 Life-Cycle Decision Approach and Management 61

References 62

Brent E Sleep, Charles D Shackelford, and Jack C Parker 2.1 Overview 71

2.2 Caps 72

2.2.1 Features, Events, and Processes Affecting Performance of Caps 72

2.2.1.1 Hydrologic Cycle 72

2.2.1.2 Layers and Features 74

2.2.2 Current State of Practice for Modeling Performance of Caps 75

2.2.2.1 Water Balance Method 75

2.2.2.2 HELP 81

2.2.2.3 UNSAT-H 82

2.2.2.4 SoilCover 82

2.2.2.5 HYDRUS-2D 83

2.2.2.6 VADOSE/W 84

2.2.2.7 TOUGH2 84

2.2.2.8 FEHM 85

2.2.2.9 RAECOM 85

2.2.3 Modeling Limitations and Research Needs for Caps 86

2.2.3.1 Role of Modeling 86

2.2.3.2 Data Needs 86

2.2.3.3 Code Quality Assurance and Quality Control 87

2.2.3.4 Verification, Validation, and Calibration 88

2.2.4 Unresolved Modeling Challenges 89

2.2.4.1 Time-Varying Material Properties and Processes 89

2.2.4.2 Infiltration at Arid Sites 90

2.2.4.3 Role of Heterogeneities 90

2.3 PRBs 90

2.3.1 Features, Events, and Processes Affecting Performance of PRBs 91

2.3.1.1 Groundwater Hydraulics 91

2.3.1.2 Geochemical Processes 92

2.3.1.3 Reaction Kinetics 98

2.3.2 Impacts on Downgradient Biodegradation Processes 98

2.3.2.1 Enhancement of Geochemical Conditions Conducive to Anaerobic Biodegradation 98

2.3.2.2 Overall Contaminant Concentration Reduction 99

4040_C000.fm Page xxii Wednesday, September 21, 2005 4:38 PM Chapter 2 Modeling of Fluid Transport through Barriers 71

2.3.2.3 Production of Hydrogen 99

2.3.2.4 Electron Donor Production 100

2.3.2.5 Direct Addition of Dissolved Organic Carbon 100

2.3.3 PRB System Dynamics 101

2.3.4 Geochemical Modeling 104

2.3.4.1 Speciation Modeling 105

2.3.4.2 Reaction Path Modeling 106

2.3.4.3 Reactive Transport Modeling 107

2.3.4.4 Inverse Modeling 108

2.3.5 Modeling Limitations and Research Needs of PRBs 109

2.4 Walls and Floors 110

2.4.1 Vertical Barriers 110

2.4.2 Horizontal Barriers 110

2.4.3 Current State of Practice for Modeling Performance of Walls and Floors 111

2.4.4 Contaminant Transport Processes 112

2.4.4.1 Aqueous-Phase Transport 112

2.4.4.2 Coupled Solute Transport 117

2.4.4.3 Modeling Water Flow through Barriers 119

2.4.4.4 Analytical Models 120

2.4.5 Modeling Limitations and Research Needs of Walls and Floors 123

2.4.5.1 Input Parameters and Measurement Accuracy 123

2.4.5.2 Time-Varying Properties and Processes 125

2.4.5.3 Influence of Coupled Solute Transport 125

2.4.5.4 Membrane Behavior in Clay Soils 126

2.5 Complicating Factors 128

2.5.1 Constant Seepage Velocity Assumption 128

2.5.2 Constant Volumetric Water Content Assumption 128

2.5.3 Anion Exclusion and Effective Porosity 129

2.5.4 Nonlinear Sorption 129

2.5.5 Rate-Dependent Sorption 130

2.5.6 Anion Exchange 130

2.5.7 Complexation 131

2.5.8 Organic Contaminant Biodegradation 131

2.5.9 Temperature Effects 132

References 132

Craig H Benson and Stephan F Dwyer 3.1 Overview 143

3.1.1 The Role of Barrier Material Mineralogy and Mix Composition on Performance 144

3.1.2 Approaches to Material Evaluation and Selection 147

3.1.3 Geosynthetics and their Durability in Barrier Systems 149

4040_C000.fm Page xxiii Wednesday, September 21, 2005 4:38 PM Chapter 3 Material Stability and Applications 143

3.2 Material Performance Factors in Caps 153

3.2.1 Material Performance Factors in Composite Barriers 155

3.2.2 Material Performance Factors in Water Balance Designs 160

3.2.3 Coupling of Vegetation and Material Performance Factors 163

3.3 Material Performance Factors in PRBs 167

3.3.1 Approach to Selection of PRB Materials 168

3.3.2 Evaluation of Field Performance Using Pilot Testing 170

3.3.3 Effects of Hydraulic Considerations on Reactive Material Performance 172

3.3.4 Structural Stability Factors in Performance 178

3.3.5 Material Durability Factors 183

3.3.5.1 Effect of Mineral Precipitation on Porosity and Hydraulic Conductivity 185

3.3.5.2 Effect of Mineral Precipitation on Reactivity 186

3.3.6 Applications of Geochemical Models in Reaction Tracking 187

3.4 Material Performance Factors in Cutoff Walls 191

3.4.1 In Situ Hydraulic Conductivity 193

3.4.2 Design Configuration 196

3.4.3 Geosynthetics in Vertical Cutoff Walls 198

3.4.4 Permeant Interaction Effects 199

References 201

Ernest L Majer 4.1 Geophysical Method Application and Use 209

4.1.1 Characterization and Geophysics 210

4.1.2 Performance Monitoring and Geophysics 212

4.1.3 Geophysical Methods for Site Characterization and Monitoring of Subsurface Processes 214

4.1.3.1 Seismic 214

4.1.3.2 Electrical and Electromagnetic 214

4.1.3.3 Natural Field and Magnetic 215

4.1.3.4 Remote Sensing 216

4.2 Specific Methods 216

4.2.1 Seismic Methods 216

4.2.1.1 Conventional and Advanced Ray and Waveform Tomography 220

4.2.1.2 Guided/Channel Waves 221

4.2.1.3 Scattered and Reflected Energy 221

4.2.1.4 Cross-Well/VSP/Single Well Imaging 222

4.2.1.5 Summary 224

4.2.2 Electrical and Electromagnetic Methods 224

4.2.3 Natural Field and Magnetic Methods 227

4.2.4 Airborne Geophysical Methods 228

4040_C000.fm Page xxiv Wednesday, September 21, 2005 4:38 PM Chapter 4 Airborne and Surface Geophysical Method Verification 209

4.2.5 State-of-the-Practice Remote Sensing Methods 231

4.2.5.1 Aerial Photography 232

4.2.5.2 Multi-Spectral Scanners 232

4.2.5.3 Thermal Scanners 233

4.2.6 State-of-the-Art Remote Sensing Technologies 233

4.2.6.1 Hyperspectral Imaging Sensors 234

4.2.6.2 LIDAR Systems 235

4.2.6.3 Laser-Induced Fluorescence (LIF) 236

4.2.6.4 Radar Systems 237

4.2.6.5 Fused Sensor Systems/Data Streams 238

4.3 PRBs 239

4.3.1 Requirements, Site Characterization, Design Verification, and Monitoring 239

4.3.1.1 Site Characterization 240

4.3.1.2 PRB Construction Verification 241

4.3.1.3 Short-Term Monitoring 242

4.3.1.4 Long-Term Monitoring 242

4.3.2 Case Histories 243

4.3.2.1 Electrical Imaging of PRB Construction and Installation (Kansas City, Missouri) 243

4.3.2.2 Cross-Hole GPR Investigations (Massachusetts Military Reservation, Massachusetts) 245

4.4 Vertical Barriers 246

4.4.1 Requirements, Site Characterization, Design Verification, and Monitoring 249

4.4.1.1 Design 249

4.4.1.2 Installation/Verification 249

4.4.1.3 Short-Term Monitoring 254

4.4.1.4 Long-Term Monitoring 254

4.4.2 Case Studies 254

4.4.2.1 Cross-Hole GPR 255

4.4.2.2 Seismic 259

4.4.2.3 ERT 260

4.5 Caps and Covers 261

4.5.1 Requirements, Site Characterization, Design Verification, and Monitoring 262

4.5.2 Case Histories 263

4.5.2.1 EMI and GPR 263

4.5.2.2 Apparent Conductivity Maps 267

4.5.2.3 Electromagnetic Radar for Monitoring Moisture Content 269

4.5.2.4 Aerial Photography 272

4.5.2.5 Multi-Spectral Scanners 273

4.5.2.6 Thermal Scanners 273

4.6.2.7 HIS Imagery 274 4040_C000.fm Page xxv Wednesday, September 21, 2005 4:38 PM

4.6 Summary 274

4.6.1 Primary Needs for Advancement 275

4.6.1.1 Integration 275

4.6.1.2 Processing and Interpretation 275

4.6.1.3 Code Development 276

4.6.1.4 Instrumentation 276

4.6.2 Future Developments 276

References 278

David J Borns, Carol Eddy-Dilek, John D Koutsandreas, and Lorne G Everett 5.1 Overview 287

5.2 Goals 288

5.3 Verification Monitoring 289

5.3.1 Methods 292

5.3.1.1 Moisture Change Monitoring Methods 292

5.3.1.2 Moisture Sampling Methods 294

5.3.1.3 Vadose Zone Monitoring Considerations 295

5.4 Verification System Design 296

5.5 Moving from State of the Practice to State of the Art 297

5.5.1 System Approach 298

5.5.1.1 Links to Modeling and Prediction 298

5.5.1.2 Optimization 299

5.5.1.3 Decision and Uncertainty Analysis 299

5.5.2 Smart Structures 300

5.5.2.1 Long-Term, Post-Closure Radiation Monitoring Systems (LPRMS) 302

5.5.2.2 Environmental Systems Management, Analysis, and Reporting (E-SMART™) Network 304

5.5.2.3 Direct Push Technologies 305

5.5.2.4 Nanotechnology Sensors 307

5.5.3 Advanced Environmental Monitoring System (AEMS) 307

5.5.4 A New DOE Barrier Design Code 308

5.6 Drivers for Implementation of New Approaches 309

5.6.1 Costs 309

5.6.2 Enabling Desired End States 309

5.7 Covers 310

5.7.1 Moving from State of the Practice to State of the Art 310

5.7.1.1 Methods 310

5.7.1.2 Verification Measurement Systems 311

5.7.1.3 Barrier Cap Monitoring 311

5.7.2 Case History: Mixed Waste Landfill 312

5.7.2.1 Cover Infiltration Monitoring 313

5.7.2.2 Neutron Moisture Monitoring 313

4040_C000.fm Page xxvi Wednesday, September 21, 2005 4:38 PM Chapter 5 Subsurface Barrier Verification 287

5.7.2.3 Fiber Optics Distributed Temperature Moisture

Monitoring 3145.7.2.4 Shallow Vadose Zone Moisture Monitoring 3145.7.3 Case History: Fernald On-Site Disposal Facility 3155.7.4 Verification Needs 3185.7.4.1 Optimization and Trend Analysis 3195.7.4.2 Sensors and Other Hardware 3205.8 PRBS 3215.8.1 Regulatory Framework 3245.8.2 Moving from State of the Practice to State of the Art 3255.8.2.1 Flow Characterization and Monitoring 3255.8.2.2 Verification of Geochemical Gradients and Zones 3275.8.3 Case History: Subsurface Monitoring 3295.8.4 Verification Needs 3295.8.4.1 Spatial and Temporal Flow Monitoring

Considerations 3305.8.4.2 Geochemical and Hydrological Process Monitoring

Considerations 3315.8.4.3 Acoustic Wave Devices 3315.9 Walls and Floors 3325.9.1 Moving from State of the Practice to State of the Art 3375.9.1.1 Neutron Well Logging 3375.9.1.2 Perfluorocarbon Tracer (PFT) Monitoring/

Verification 3385.9.2 Case History: Colloidal Silica Demonstration 3415.9.3 Case History: Barrier Monitoring at the Environmental

Restoration Disposal Facility (ERDF) 3435.9.3.1 Study Conclusions 3455.9.3.2 Study Recommendations 3455.9.4 Verification Needs 3465.9.4.1 Adequacy of the Containment Region 3475.9.4.2 Long-Term Performance of the Containment 3475.10 Conclusions 348References 349

Appendix AWorkshop Panels 353Panel 1 Prediction: Materials Stability and Application 353Panel 2 Prediction: Barrier Performance Prediction 353Panel 3 Prediction: Damage and System Performance Prediction 354Panel 4 Verification: Airborne and Surface/Geophysical Methods 355Panel 5 Verification: Subsurface-Based Methods 3554040_C000.fm Page xxvii Wednesday, September 21, 2005 4:38 PM

The diversity of waste types and desirable service lives for facilities undervarious regulatory programs are summarized in Tables 1.1 and 1.2, respectively.

* With contributions by James H Clarke, Vanderbilt University, Nashville, Tennessee; John B Gladden, Westinghouse Savannah River Company, Aiken, South Carolina; Horace K Moo-Young, Villanova University, Villanova, Pennsylvania; Priyantha W Jayawickrama, Texas Tech University, Lubbock, Texas; W Barnes Johnson, U.S Environmental Protection Agency, Washington, DC; Robert

E Melchers, University of Newcastle, Callaghan, NSW, Australia; Mark L Mercer, U.S mental Protection Agency, Washington, DC; V Rajaram, Black and Veatch Corporation, Overland Park, Kansas; and, Paul R Wachsmuth, University of North Carolina at Charlotte, Charlotte, North Carolina.

Environ-4040_book.fm Page 1 Wednesday, September 14, 2005 12:43 PM

2 Barrier Systems for Environmental Contaminant Containment & Treatment

Estimation of the long-term deterioration pattern of barriers is necessary toimprove the reliability of estimates of long-term contaminant release source termsfor input into human health and ecological risk assessments, as well as facilitymonitoring and maintenance planning Monitoring of barrier performance pro-vides useful but inadequate data for performance predictions, because of limitedfield experience with barriers of various configurations in many environments,and because epochal events such as floods and earthquakes produce transienteffects that cause deviations from performance patterns

The majority of quantitative methods that are currently used to estimate term barrier performance have time-invariant material characteristics andload/fluid application rates The use of these fate and transport models, most of

Hall, London With permission.)

Define Context

social, individual, organizational, political, technological

Define System

Hazard Scenario Analysis

• what can go wrong?

• how can it happen?

• what controls exist?

Estimate Probability

(of occurrence of consequences)

4040_book.fm Page 2 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 3

TABLE 1.1 Types of Hazardous Materials

Type

Typically Found

in Nature?

Importance of Chemical Form

to Toxicity

Does Hazard Decay Naturally?

Do We Know How to Destroy Hazard?

Radioactive isotopes

level of exposure to the hazard by altering the ingestion or inhalation uptake of isotopes

Natural decay is fixed for each isotope

Negligible prospects for

in situ

destruction or treatment

Ex situ treatment may be practical

to separate lived isotopes from short-lived isotopes Toxic

long-organic compounds b

and inhalation uptake

Decay generally slow (years, decades) and often dependent

on specific chemical environment, e.g., trichloroethylene

In situ decay may

be deliberately enhanced by microbes

Determines toxicity level

Ex situ

destruction generally possible, but the associated risks and costs of transportation and destruction are high Toxic metals Yes, although

sometimes not in the more hazardous chemical forms

Can affect ingestion or inhalation uptake

Metals won’t decay, but the chemical form may naturally change into less toxic forms

Destruction is not practical

Generally affects toxicity

In situ alteration

of chemical form can sometimes

be enhanced by microorganisms

4040_book.fm Page 3 Wednesday, September 14, 2005 12:43 PM

4 Barrier Systems for Environmental Contaminant Containment & Treatment

which are based on one-dimensional differential equations, for describing inant advection-dispersion has simplified barrier performance analyses but doesnot address long-term barrier system performance adequately As the performance

contam-TABLE 1.1 (continued)

Types of Hazardous Materials

Type

Typically Found

in Nature?

Importance of Chemical Form

to Toxicity

Does Hazard Decay Naturally?

Do We Know How to Destroy Hazard?

generally possible, but with associated risks and costs during transportation and destruction

a However, the specific radioactive isotopes are typically are not the specific isotopes found in nature.

b There are also some toxic compounds that are neither organic nor metals, e.g., asbestos.

Source: INEEL (2000) Environmental Laboratory Report INEEL/EXT-2000-01094; Piet et al (2001) INEEL technical report INEEL/EXT-2001-01485.

TABLE 1.2

Time frames for Waste Containment Performance under Various U.S Regulatory Programs

Time Frame Regulatory Program

10,000 years Nuclear Regulatory Commission and EPA regulations for high-level and transuranic

waste (10 CFR 60, 10 CFR 63, 40 CFR 191, 40 CFR 197)

1000 years EPA regulations for near-surface uranium and thorium mill tailings (40 CFR192)

and DOE policy for new land burial (DOE M 435.1)

500 years NRC regulations for near-surface burial of low-level waste (10CFR61)

30 years Baseline EPA RCRA time period for near-surface burial chemical hazards

(40 CFR264); EPA can increase or decrease this value for each case Indefinite Baseline EPA CERCLA time period for residual hazards (CERCLA requires a

5-year review to ensure the remedy is still protective of human health and the environment and is still performing as predicted)

4040_book.fm Page 4 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 5

analyses time frames extend from one or two decades to hundreds of years,changes in barrier material characteristics; cyclic changes, waning, or growth ofstressing events; and possible exhaustion of initially present parent contaminantsand/or generation of daughter contaminants combine to decrease the reliability

of contaminant release estimates

Long-term performance modeling of waste containment systems and ual barriers within such systems require identifying possible damage mechanismsand assessing the system resistance in all possible ways in which the systemmight fail Various techniques have been developed in practice (in differentindustries) and, hence, with different names, including the following:

individ-• Preliminary hazard analysis (PHA) (nuclear industry)

• Walk-down analysis consisting of on-site visual inspection, particularly

of pipe work (nuclear industry)

• Failure modes and effects analysis (FMEA), which uses generic terms

as prompts (various applications)

• Failure modes, effects, and criticality analysis (FMECA), which alsoassesses criticality of consequences

• Hazard and operability studies (HAZOP), which uses guide words asprompts (primarily chemical industry)

• Incident data banks, which contain data such as accident data and miss data

near-For the range of barrier applications available now and in the future, there is

a need for improved capacity to predict containment barrier damage and systemperformance Damage and system performance models must be:

• Responsive to the needs of a diverse set of decision makers(i.e., designers of new barriers, managers of barriers in service, regu-lators, funding agencies, and the public)

• Integrative of the most important mechanisms of failure (i.e., bothspatially uniform degradation and localized degradation; both contin-uously acting and discrete in time)

• Comprehensive with regard to the range of performance measuresrelevant to a given barrier design that solves a particular problem at aparticular location

• Stochastic to allow evaluation of the sensitivities of parameter tainties compared to performance measures

uncer-• Probabilistic in consideration of failure scenarios and mechanisms thatmay or may not occur during the service life

• Validated by data to the extent practical

• Adaptive to new information obtained during barrier service

4040_book.fm Page 5 Wednesday, September 14, 2005 12:43 PM

6 Barrier Systems for Environmental Contaminant Containment & Treatment

• Informative regarding barrier degradation to guide barrier surveillanceand maintenance, justification for reduction of surveillance and main-tenance, barrier lifetime extension while in service, and future barrierdesigns

• Graded in its implementation according to the severity and longevity

of the associated risks (barriers with lower severity or shorter durationhazards do not need all of the above)

Attempts to provide satisfactory system performance demands that one ormore criteria be available against which to measure system performance Thesetting or derivation of performance criteria is a problem with a fascinating andcomplex history, much of it based originally on issues associated with the nuclearindustry This history includes some deep philosophical questions, including

“Who is to bear what level of risk, who is to benefit from risk-taking, and who

is to pay? Where should the line be drawn between risks that are to be managed

by the state and those that are to be managed by individuals, groups, or rations? Who evaluates success or failure in risk management and how? Whodecides what should be the desired trade-off between different risks?” (Hood

corpo-et al., 1992) The decisions about these matters are influenced by judgments aboutthe following (Stewart and Melchers, 1997):

• Anticipation of system failure and resilience against unexpected trophe

catas-• Assumptions used to compute a numerical estimate of system risks

• Size of uncertainties in estimating system risks

• Organizational vulnerabilities to system failure

• Cost of risk reduction

• Size and composition of groups involved in decision-making processes

• Aggregation of individual preferences (i.e., distribution of benefits andrisks)

• Counter-risks (i.e., alternatives may have other societal risks)

Psychological aspects, such as risk perception and risk aversion, social andcultural preferences, as well as political processes and risk communication alsoplay a part The term “failure” can mean a variety of structural conditions or lack

of capacity to meet expected performance functions when it is applied to tainment systems Structural failure of a system component or the entire systemshould be differentiated from functional system failure as described by Inyang(1994) and Inyang et al (1995) Structural failure of a containment system maynot necessarily lead to immediate functional failure because the former is oftenindexed in terms of parameters that define the stability and hydraulic character-istics of the containment system, whereas functional failure is assessed in terms

con-of the risk con-of environmental and human exposure to contaminants that may bereleased from the system More broadly, for a given initial hazard inventory, the4040_book.fm Page 6 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 7

exposure generally depends on the five factors listed below and illustrated inFigure 1.2 Eventually, hazard either decays (with some half-life) or escapes

1 Hazardous half-life

2 Mobilization rate/year (e.g., leaching, diffusion in the absence of abarrier)

3 Time at which barrier begins to degrade

4 Barrier degradation rate/year

5 Transport time of escaped materials between barrier and recipientsFactors 2, 3, and 4 control when and how fast the hazard escapes Factor 5 controlshow much time (with additional hazard decay) will elapse before the escapedhazard impacts human health and the environment With reference to the range

of time horizons in various regulations, there is no systematic connection betweenthe hazard timescales and regulatory timescales that are summarized in Table 1.2.Different regulations were established at different times by different legislation

in response to different issues Thus, the appropriate framework for predictingbarrier system performance is not always clear: the time frames can differ greatlyand the appropriate assumptions on how long to monitor and manage the barriersystem can also differ

1.2 LONG-TERM PERFORMANCE ANALYSIS

FRAMEWORK

It is necessary to formulate a long-term performance analysis framework that enablesthe consideration of factors that are significant for a given class of containmentsystems The failure states of the constructed system in terms of both structuralfailure and functional failure need to be defined Also, the performance assessment

Hazard halflife

Hazard inventory

Escaping inventory

Time barrier starts to degrade Rate barrier degrades per year

Rates inventory mobilizes per year

Delay time in transport through environmental media Exposed individuals 4040_book.fm Page 7 Wednesday, September 14, 2005 12:43 PM

8 Barrier Systems for Environmental Contaminant Containment & Treatment

framework should incorporate nodes to which pre-failure performance modelscan be linked

1.2.1 C ONCEPTS AND A NALYTICAL F RAMEWORK

Several concepts and analytical frameworks have been proposed for use in ing the long-term performance of containment systems The concepts pertain tothe performance pattern of containment systems during service lives and post-closure time frames that can range from 30 years to thousands of years The focus

assess-of the analyses is the formulation and use assess-of performance prediction models thatare capable of determining contaminant release rates as a function of estimated,measured, or designed magnitudes of containment system design parameters,waste characteristics, stressing events and processes, and site/hydrological con-ditions The factors that need to be considered are numerous, as exemplified bythe case of a near-surface barrier illustrated in Figure 1.3

Several attempts have been made to establish the expected general pattern ofbarrier performance over long service lives Figure 1.4a shows the containmentsystem performance model that is implicit to current practice The facility isassumed to provide a constant level of service, or to be structurally sound untilexternal monitoring data indicate the release of contaminants at unacceptable

that influence the long-term performance of near-surface containment systems.

Plugging and surface tension

Output: Contaminant flow to the vadose zone

Waste zone

Hydrology (including micropores, capillaries)

Structure

Natural boundary conditions

(weather, climate, biota)

Engineered boundary conditions (design, maintenance, repair)

materials configuration Temperature

Precipitation Plant/animal intrusion

Soil type and thickness

Erosion

Subsidence Compaction

Wind/water erosion

Biochemical changes?

Ecological

Water 4040_book.fm Page 8 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 9

concentrations Figure 1.4b shows a more realistic performance pattern in whichthe performance degrades gradually during the immediate post-implementationperiod and then decays abruptly After abrupt decay, the performance decreasesmuch more gradually in a period that is characterized by large uncertainties Thereader should note that system damage vs time plots have configurations thatare the reverse of those of system performance (or effectiveness) vs time plots.Thus, Figure 1.5 shows an increase in the risk of containment system failure withtime It should be noted that although the system deterioration pattern may berepresented by a smooth curve, the performance pattern of a particular component

of the containment system could exhibit temporal fluctuations in response totransient stressing mechanisms, the passage of contaminant fronts, and mainte-nance activity In developing the conceptual framework for estimating the long-term performance pattern of containment systems, Inyang (1994) identified thevarious stages illustrated in Figure 1.6 Curve 1 shows the barrier degrading viacontinuous deterioration mechanisms The branching to Curve 2 shows a barriersuffering from a discrete (in time) negative perturbation, such as a flood or anearthquake The branching to Curve 3 reflects a barrier being upgraded orrepaired In the illustration, following Curve 1, the containment system effective-ness decays from an initial level of E to, to a minimum acceptable level of E tr attime, t r E tr corresponds to the functional performance level that is typically

(a) abrupt failure pattern implicit to current practice (b) gradual degradation pattern that

Time (b) 4040_book.fm Page 9 Wednesday, September 14, 2005 12:43 PM

10 Barrier Systems for Environmental Contaminant Containment & Treatment

specified by regulators or other authorities If the facility is repaired at a time,

t m, the effectiveness can abruptly increase to E tm so that an improved performance(described by Curve 3) results Essentially, repairs postpone the attainment of E tr

permission.)

Congress on Environmental Geotechnics, Calgary, Canada, pp 273–278 With permission.)

Damage and System Performance Prediction 11

by slowing down the deterioration of the repaired component(s) and, hence, thesystem The system can also degrade abruptly, as at t g, such that its effectivenessfalls to E tgand system performance follows Curve 2 to failure at t 2 (much soonerthan would result from the regular deterioration pattern)

1.2.2 T YPES OF P ERFORMANCE P REDICTION A PPROACHES

In order to serve practical purposes, performance patterns need to be quantified,requiring the development of rating systems and models Approaches to perfor-mance prediction can be categorized as empirical, semi-empirical, and less empirical(theoretical modeling)

1.2.2.1 Empirical Prediction Approaches

Empirical prediction approaches involve the extrapolation of current knowledge

of system behavior and/or similar system behavior to long-term system behavior.Such knowledge can also be acquired through accelerated testing in intensifiedenvironments Another example of an empirical approach is performance index-ing In most cases, indexing criteria do not explicitly include time functions withperformance factors Table 1.3 shows the ratings of single components and com-posite configurations of barriers (Piet et al., 2001) In general, the scores on

Estimated Cost (dollars/ft 2 )

Benefit/Cost Ratio

Ranking

in Group One-Barrier Layer

Source: Adapted with modification from Koerner, R.M and Daniel, D.E (1992) Civil neering, pp 55–57.

Engi-4040_book.fm Page 11 Wednesday, September 14, 2005 12:43 PM

12 Barrier Systems for Environmental Contaminant Containment & Treatment

overall benefit or utility of a particular design increase with the number of

components

Inyang and Tomassoni (1992) indexed the long-term performance pattern of

waste covers for use in regulatory impact analysis The scores are presented in

Table 1.4 The reader should note that these scores are general indices and are

not precise estimates of the performance of the components scored Other

researchers exemplified by Hagemeister et al (1996) developed detailed

perfor-mance indexing systems that incorporate ratings of barrier components,

contam-inant transport pathway factors, and human exposure potential

1.2.2.2 Semi-Empirical Prediction Approaches

These approaches involve the use of semi-empirical models to estimate the

damage time functions or deterioration pattern of containment systems or specific

containment system components Using adaptations from product reliability

anal-yses, a parameter that is generically referred to as the “failure rate” is used to

quantitatively describe the effectiveness or reliability of a barrier or containment

system with time The magnitude of the failure rate is the significant determinant

of the barrier degradation rate in the absence of transient events It is tempting

TABLE 1.4 Estimated Long-Term Effectiveness of Selected Waste Containment Measures

a Assumes addition of new clay cap at 100 years.

b Assumes addition of new synthetic cap at 100 years.

c Assumes addition of new composite clay and synthetic cap at 100 years.

d Assumes addition of new HDPE at 100 years.

e Assumes addition of new slurry wall at 30 years.

Source: Inyang, H.I and Tomassoni, G (1992) Indexing of long-term effectiveness of waste containment systems for a regulatory impact analysis.

A technical guidance document Office of Solid Waste, U.S Environmental Protection Agency, Washington, DC.

4040_book.fm Page 12 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 13

to erroneously assume that failure rates for containment systems are constant In

practice, the failure rates of most engineered systems are not constant with time

Generally,

(1.1)

where λ(t) is the time-variable failure rate of the containment system; λ0 is the

initial failure rate of the containment system; β is an exponent that describes the

variation (usually decay) of the failure rate with time, t Equation (1.1) represents

the general exponential form of the decay equation The linear and Weibull forms

of the equation are presented below as Equations (1.2) and (1.3), respectively

The parameters are as defined for Equation (1.1) The time parameter, t0, is the

time corresponding to the origin of the initial failure, λ0

(1.2)

(1.3)

For Equations (1.1) through (1.3), the value of the constant β determines the

shape of the failure rate function The failure rate is increasing with time if β > 0,

it is constant if β = 0, and it is decreasing if β < 0 For more information, the

reader is referred to Wolford et al (1992), who used this approach to estimate

the aging pattern of nuclear power plant equipment Such techniques have already

been successful in extending the license of 10 United States nuclear power plants

by 20 years Inyang (1994) observed that the Weibull format of failure analysis

provides the curve geometries that match the expected deterioration pattern of

most containment systems and proposed the use of Equation (1.4) with shape

parameters ranging from 2 to 5 The use of Equation (1.4) enables long-term

performance to be addressed within the context of system reliability

(1.4)

where R t is the reliability of the containment system at a future time of reference,

t is the future time of reference, and n is the scale or normalization parameter

that corresponds to the time duration at which the failure probability is 0.632

Generally, the larger the magnitude of β, the greater the deterioration rate

Considering that there is a complimentary relationship between the

probabil-ity of failure, P t , and reliability, R t, of a component or system as indicated by

Equation (1.5), initial values of reliability can be established

14 Barrier Systems for Environmental Contaminant Containment & Treatment

(1.5)The damage functions for each system component can be generated fromcurrent knowledge, testing, and extrapolations, and can be used to determine theprobability that barrier characteristics will meet specified standards at specifiedfuture times

1.2.2.3 Less Empirical (Theoretical) Modeling Approach

This approach involves modeling the stresses, deterioration processes, wastetransformations and release, barrier material durability, and flaw evolution for abarrier component or system In this approach, the failure probabilities of systemcomponents and the system itself are modeled Interactions among various param-eters that promote or negate effects are considered Considering that variousstressors and their impacts have different probabilities of occurrence within dif-ferent timescales, the challenge of deciphering the interactions among parameters

is quite great Therefore, an innovation within this modeling approach is the use

of modeling frameworks that enable the incorporation of various models and theestablishment of dynamic linkages among them This technique is nested in thesubdiscipline of system dynamics

System dynamics is the study of dynamic feedback systems using computermodeling and simulation (Forrester, 1961) Unlike other scientists, who study theworld by breaking it up into smaller and smaller pieces, system dynamicists look

at things as a whole The central concept of system dynamics is understandinghow all objects in a system interact with one another Visualization of the system

is one of the assets of this modeling technique However, beneath the visualexterior is a series of differential equations that define the behavior of the systemover time An example of software that can be used in this modeling exercise isStella Research (Stella, 2001) The calculations are performed using numericalintegration Although the interface makes the modeling look superficial andalmost trivial, a sophisticated mathematical engine performs the calculations.Using this modeling technique, it is possible to model complicated systems Athorough understanding of the structure of these complex systems can lead to anexplanation of their performance, both over time and in response to internal andexternal perturbations By understanding the underlying system structure, predic-tions can be made relative to how the system will react to change Systemdynamics models are descriptive in nature The elements in the models mustcorrespond to actual entities in the real world The decision rules in the modelsmust conform to actual practice and real-world phenomena A new project at theIdaho National Engineering and Environmental Laboratory (INEEL) is addressingbarrier degradation dynamics (Piet and Breckenridge, 2002) One component ofthe effort is the use of relatively simple but flexible system dynamics models toexplore possible interactions of processes These models provide a tool to explore

uncertainties in scenarios and mechanisms, whereas more sophisticated models

are tools for exploring sensitivities to parameter uncertainties

4040_book.fm Page 14 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 15

To illustrate the necessity of addressing interactions among various ters, the effects of the burrowing of covers by animals on evapo-transpiration areconsidered During the summer months, more water is lost from plots with animalburrows than from plots where no animal burrows are present During the wintermonths, both the plots with animal burrows and the control plots gain water Inaddition, water does not infiltrate below approximately 1 meter (m), even thoughburrow depths always exceed approximately 1.2 m The lack of significant waterinfiltration at depth and the overall water loss in the lysimeter plots are occurringdespite the following worst-case conditions:

parame-• No vegetative cover (no water loss through transpiration)

• No water run off (all precipitation is contained)

• Burrow densities in lysimeters greater than those in natural settings

• Extreme rainfall events applied frequently (i.e., three 100-year stormevents in three months)

• Animals burrowing deeper in the lysimeters than in natural settings

As part of the conclusion of the study described in the precedingparagraph, the investigators noted that “the overall water loss fromsoils with small-small burrows appears to be enhanced by a com-bination of soil turnover and subsequent drying, ventilation effectsfrom open burrows, and high ambient temperatures” (Gee and Ward,1997) Thus, in this case, animal intrusion had a net positive effect.Indeed, earlier work shows that soils were more dry beneath burrowsthan elsewhere (Cadwell et al., 1989; Link et al., 1995) Link et al.(1995) report that the increased moisture in burrows facilitatedvegetation response that increased plant transpiration as plants tookadvantage of the moisture and sent roots to use it, leading to dryzones under the burrows Indeed, Link et al (1995) note that “eco-logically, it is expected that a local abundance of a limiting resource,

in this case moisture, would be rapidly and therefore depleted.”

1.3 RELATIONSHIP OF STRUCTURAL FAILURE

TO FUNCTIONAL FAILURE

In real-world situations, defining satisfactory system performance can be difficult

It is a vector with many components, governed by different criteria, and driven

by different and perhaps interacting processes These processes may not be wellunderstood and, hence, can be represented analytically only with considerableuncertainty This situation is not too different from that in other spheres anddisciplines

It is usual in risk analysis to consider the consequences of failure, hence therecent focus of performance assessments has been on readily measurable barriercharacteristics (e.g., barrier permeability) with limited focus on various combi-nations of outflows and inflows Because the system properties and processes areuncertain, failure consequences can be described only with uncertainty Moreover,4040_book.fm Page 15 Wednesday, September 14, 2005 12:43 PM

16 Barrier Systems for Environmental Contaminant Containment & Treatment

the consequences usually are the critical outcome(s) of the system because thelarger community seldom has particular interest in the structural system itself.The foregoing discussion leads to the need to examine the performance factorsnecessary to evaluate containment systems These factors are divided into thefollowing two categories:

• Total system (parameters that define functional performance)

• Concentration of hazardous materials in surface/aquifer water

• Exposure to humans (e.g., water, air, intrusion pathways)

• Risk to humans

• Risk to ecologies

• Barrier and barrier subsystems (parameters that define structural formance)

per-• Resistance to human intrusion

• Water flux through barrier

• Gas flux through barrier

• Hazardous material flux through barrier

• Measures of individual degradation mechanisms (e.g., erosion,subsidence)

The satisfaction of both functional and structural design functions of thecomposite containment system requires that the various system components meetspecific design functions that contribute to overall system performance Thevariability in the combination of various containment system components impliesthat long-term performance under a given set of applied stresses will also bedifferent Inyang (1999) suggested the following nonexclusive criteria as indices

of containment system and component performance:

• Ability of the system to reduce the concentrations of aqueous phasecontaminants to acceptable levels through one or more contaminantattenuation processes (e.g., sorption, precipitation)

• Ability of the system to reduce the volume of contaminants that isreleased into protected media to acceptable levels

• Ability of the system to reduce the leaching of bound contaminantsfrom stabilized media to acceptable levels

• Ability of near-surface system components to attenuate radiation tonondamaging levels

Often, the locations at which measurements of contaminant volumes orrelease rates will be obtained are specified in documents that are used to establishthe compliance of a component or system at specified time intervals As anexample, in Table 1.5, Ho et al (2002a) summarized the design performanceobjectives for the Monticello Mill Tailings Repository in which performancestandards are specified in terms of specific quantities of contaminants that mustnot be exceeded

4040_book.fm Page 16 Wednesday, September 14, 2005 12:43 PM

Damage and System Performance Prediction 17

Period of Compliance Regulation

All pathways <100 mrem/year

Effective Dose Equivalent from all routine DOE activities

To a member of the public

5400.5 II 1.a

Atmosphere <10 mrem/year

Effective Dose Equivalent, excluding Rn

To a member of the public

1000 years if reasonably achievable and, in any case, for at least 200 years

40 CFR 192.02(a) and

40 CFR 192(b)(1) Atmosphere Annual average

concentration of Rn-222 in air

<0.5 pCi/L

At or above any location outside the landfill

1000 years if reasonably achievable and, in any case, for at least 200 years

40 CFR 192.02(a) and

40 CFR 192(b)(2) Groundwater Arsenic

Rn and U

<15 pCi/L a,b

Intersection of vertical plane with uppermost aquifer at downgradient limit of disposal area plus area taken by dike or other waste barrier

1000 years if reasonably achievable and, in any case, for at least 200 years

40 CFR 192.02(a) and

40 CFR 192.02(c)(4), and Table 1 to Subpart A of

40 CFR 192

Groundwater Beta particles, and

photons made from manmade radionuclides

<4 mrem/year

In community water supply systems

4040_book.fm Page 17 Wednesday, September 14, 2005 12:43 PM

18 Barrier Systems for Environmental Contaminant Containment & Treatment

1.3.1 E CONOMIC OR P SEUDO -E CONOMIC C RITERIA

Economic evaluation has the advantage (and disadvantage) of forcing all parties

to evaluate their objectives in monetary terms Pseudo-economic criteria, such asutility analysis, require a similar approach but in terms of a different unit ofmeasurement In principle, the maximum expected net present value criterion can

be stated as follows:

(1.6)

where k is the alternative or system configuration being considered, i is the state

of the system (e.g., normal operation, one or other mode of system failure), p i is

the probability of occurrence for each such state of nature, M is the number of such states, j is the attribute, N is the number of attributes, and X ji represents thevarious costs or benefits associated with each state There are some very signif-

icant problems associated with determining the X ji, and these are well known incost-benefit analysis literature (Layard, 1972; Dasgupta, 1993) Usually, the opti-mal decision is considered to be the maximization of the value of Equation (1.6),which then provides the possible decisions required Typically, this translates into

desired (maximum) values for the probabilities, p i These values are obtainable

through risk analysis, as are some of the values of X ji (where these are quences) In practice, the optimization of Equation (1.6) can be constrained byregulatory requirements (Stewart and Melchers, 1997)

Period of Compliance Regulation

a If background is below this level.

b An alternative concentration limit may be established under 40 CFR 192.02 (c)(ii)(A).

c Where secular equilibrium is obtained, this criterion will be satisfied by a concentration of 0.044 milligrams per liter For conditions of other than secular equilibrium, a corresponding value may be derived and applied, based on the measured site-specific ratio of the two isotopes of uranium.

d A unit gradient flow is assumed to equate percolation to hydraulic conductivity.

max EV k p i X

i

M

ji j