Agency
£ EPA
INNOVATIVE SITI
REMEDIATION
TECHNOLOGY
Chemical
Treatment
-------
,:!lll!f I"1 H PlfiPlllllljih ''IlllllllillPllilll'llPllllPI i 111! illll||||nillUllllliJ:i|i|lllllilll!lllllllllnlllllllllllll'll|ll '
iitiijniinnnnii/iiiiiiiiiiiiiiiiiiiiiiiiiiii'fiiijijijiiiiiiiiiii^
.:' ", ,i!
,'.).•• vi' ., v, ,; I is i'' ;' ' •.. •
• P " ' , '!:" .. •
i 1 1 i: I : • I! : , | i : ^ I : , • ! ,:,: • ; '; , i
I'S; lijpp, PUPBinnp, pniPPiiPPPP litiiiWpJiiui :!JB^^
, iH'i't liiilli
^ pi ppu;,' HHiipppni1!iipiniiniijj p, PI, ppp,:1 mtfit 1
JIlilI'lllllllLIIIIIIblKll-IIIIIIITIIllli!]!^!^
Jll d: ', i : i , ', ; "l
_
II'VmlnE 'T!!!!!!!!!!!!!!!!!!!!*!'!''!!!!!!!!!!!!!!!!!!!',!!!!!!!!!!!!!!!!'!!!!!!!!!!!!!!
'nil ll» 1|"1 iv . '". i iilklliililwIFIIIIIIIIWIIi'ttll".. "li;!;:!'l" "IL, 'h millnlill
''':!'-'!'''!" :'l )"'" «•-• FIT1 Iliiill1:. Hi W«'Jiia
iiiiiiiii'K'ii!' '"iifliii! iifiiiiin^p'iiiiigiiriiiiinnniinniiinii'iitii ! iiiiiiiiP'iiHiiiiiiin iivsniipi!:1! «iir;; liiiiiiiiiiiiiiiiifjiiiiiiiiiiiiiiiiiiFiiiiiiiih r'niiiPiiiiiinniii'iniiff I'liiiiiiiiiiiiii^iiiiiiiiiii1:!!
niniiiii'iiip ih'!:ii iiiEiuainldiiiiiiiiiii
iilljlliljlj''!!^
' "' ":
!i«^^^^^^^^^^^ Kii--*Ki f ram •i«i'l'1.
.......... ........... "•• Hi":.. ' ' ..... iin^^^^^^^^^ ....... •)' >" .'i: ..... i!i:"!:i i WH
JIH
•ii|W^ Jiiyihflijilii^^^^^^^^^^^^^^ iiif ;
-------
INNOVATIVE SITE
REMEDIATION TECHNOLOGY:
DESIGN AND APPLICATION
CHEMICAL TREATMENT
One of a Seven-Volume Series
Prepared by WASTECH*, a multiorganization cooperative project managed
by the American Academy of Environmental Engineers* with grant assistance
from the U.S. Environmental Protection Agency, the U.S. Department of
Defense, and the U.S. Department of Energy.
The following organizations participated in the preparation and review of
this volume: *.
*' Air & Waste Management
• Association
P.O. Box 2861
Pittsburgh, PA 15230
American SocVety of
Civil Engineers
345 East 47th Street F
New York, NY 10017
American Academy of
Environmental Engineers*
130 Holiday Court, Suite 100
Annapolis, MD 21401
Hazardous Waste Action
Coalition
1015 15th Street, N.W., Suite 802
Washington, DC 20005
American Institute of
Chemical Engineers
345 East 47th Street
New York, NY 10017
Soil Science Society
of America
677 South Segoe Road
Madison, WI 53711
Water Environment
Federation
601 Wythe Street
Alexandria, VA 22314
Monograph Principal Authors:
Leo Weitzman, Ph.D., Chair
Irvin A. Jefcoat, Ph.D. Byung R. Kim, Ph.D.
Series Editor
William C. Anderson, P.E., DEE
-------
Library of Congress Cataloging in Publication Data
Innovative site remediation technology: design and application.
p. cm.
"Principle authors: Leo Weitzman, Irvin A. Jefcoat, Byung R. KimM-V,2, p. Hi.
"Prepared by WASTECH."
Includes bibliographic references.
Contents: ~[2] Chemical treatment
1. Soil remediation-Technological innovations. 2. Hazardous waste site remediation-
Technological innovations. I. Weitzman, Leo. II. Jefcoat, Irvin A. (Irvin Ally) III. Kim. B.R.
IV. WASTECH (Project)
TD878.I55 1997
628°5;5--dc21 97-14812
CIP
'" ISBN 1-883767-17-2 (v. 1) ISBN 1-883767-21-0 (v. 5) j
ISBN 1 -883767-18-0 (v. 2) ISBN 1-883767-22-9 (v. 6) y
ISBN 1-883767-19-9 (v. 3) ISBN 1-883767-23-7 (v. 7) . ,
ISBN 1-883767-20-2 (v. 4) ?
Copyright 1997 by American Academy of Environmental Engineers. All Rights Reserved.
Printed in the United States of America. Except as permitted under the United States
Copyright Act of 1976, no part of this publication may be reproduced or distributed in any
form or means, or stored in a database or retrieval system, without the prior written
permission of the American Academy of Environmental Engineers.
The material presented in this publication has been prepared in accordance with
generally recognized engineering principles and practices and is for general informa-
tion only. This information should not be used without first securing competent advice
with respect to its suitability for any general or specific application.
The contents of this publication are not intended to be and should not be construed as a
standard of the American Academy of Environmental Engineers or of any of the associated
organizations mentioned in this publication and are not intended for use as a reference in
purchase specifications, contracts, regulations, statutes, or any other legal document.
No reference made in this publication to any specific method, product, process, or
service constitutes or implies an endorsement, recommendation, or warranty thereof by the
American Academy of Environmental Engineers or any such associated organization.
Neither the American Academy of Environmental Engineers nor any of such associated
organizations or authors makes any representation or warranty of any kind, whether
express or implied, concerning the accuracy, suitability, or utility of any information
published herein and neither the American Academy of Environmental Engineers nor any
such associated organization or author shall be responsible for any errors, omissions, or
damages arising out of use of this information.
Printed in the United States of America.
WASTECH and the American Academy of Environmental Engineers are trademarks of the American.
Academy of Environmental Engineers registered with the U.S. Patent and Trademark Office.
-------
CONTRIBUTORS
PRINCIPAL AUTHORS
Leo Weitzman, Ph.D., Task Group Chair
President
LVW Associates, Inc.
Irvin A. Jefcoat, Ph.D. Byung R. Kim, Ph.D.
University of Alabama Ford Motor Research Laboratory
Department of Civil Engineering Energy Systems Division
REVIEWERS { '
>,
The panel that reviewed the monograph under the auspices of the'^reject
Steering Committee was composed of: .• '
?
Richard A. Conway, P.E., DEE, Chair Steven McCutcheon, Ph.D., P.E.
Union Carbide Corporation EPA — Athens, Georgia
Herbert E. Allen, Ph.D. Michael G. Nickelsen, Ph.D.
University of Delaware High Voltage Environmental
Applications, Inc.
William J. Cooper, Ph.D. Steven Shoemaker, P.E.
High Voltage Environmental DuPont Environmental
Applications, Inc.
Ernest Gloyna, Ph.D., P.E., DEE Michael H. Spritzer
University of Texas General Atomics
Eric Lindgren, Ph.D. Walter J. Weber, Jr., Ph.D., P.E., DEE
Sandia National Laboratories University of Michigan
-------
STEERING COMMITTEE
This monograph was prepared under the supervision of the WASTECH® Steering
Committee. The manuscript for the monograph was written by a task group of experts
in chemical treatment and was, in turn, subjected to two peer reviews. One review w.as
conducted under the auspices of the Steering Committee and the second by professional
and technical organizations having substantial interest in the subject.
Frederick G. Pohland, Ph.D., P.E., DEE Chair
Weidlein Professor of Environmental
Engineering
University of Pittsburgh
Richard A. Conway, P.E., DEE, Vice Chair
Senior Corporate Fellow
Union Carbide Corporation
William C. Anderson, P.E., DEE
Project Manager
Executive Director
American Academy of Environmental
Engineers
Colonel Frederick Boecher
U.S. Army Environmental Center
Representing American Society of Civil
Engineers
Clyde J. Dial, P.E., DEE
Manager, Cincinnati Office
SAIC
Representing American Academy of
Environmental Engineers
Timothy B. Holbrook, P.E.
Engineering Manager
Camp Dresser & McKee, Incorporated
Representing Air & Waste Management'
Association
Joseph F. Lagnese, Jr., P.E., DEE
Private Consultant
Representing Water Environment Federation
Peter B. Lederman, Ph.D., P.E., DEE, P.P.
Center for Env. Engineering & Science
New Jersey Institute of Technology
Representing American Institute of Chemical
Engineers
George O'Connor, Ph.D.
University of Florida
Representing Soil Science Society of America
George Pierce, Ph.D. ,
Manager, Bioremediation Technology Dev.
American Cyanamid Company
Representing th
Microbiology
. >y V .
Representing the Society of Industrial
Peter W. Tunnicliffe, P.E., DEE
Senior Vice President
Camp Dresser & McKee, Incorporated
Representing Hazardous Waste Action
Coalition
Charles 0. Velzy, P.E., DEE
Private Consultant
Representing, American Society of
Mechanical Engineers
Calvin H. Ward, Ph.D.
Foyt Family Chair of Engineering
Rice University
At-large representative
Walter J. Weber, Jr., Ph.D., P.E., DEE
Gordon Fair and Earnest Boyce Distinguished
Professor
University of Michigan
Representing Hazardous Waste Research Centers
FEDERAL REPRESENTATION
Walter W. Kovalick, Jr., Ph.D.
Director, Technology Innovation Office
U.S. Environmental Protection Agency
George Kamp
Cape Martin Energy Systems
U.S. Department of Energy
Jeffrey Marqusee
Office of the Under Secretary of Defense
U.S. Department of Defense
Timothy Oppelt
Director, Risk Reduction Engineering
Laboratory
U.S. Environmental Protection Agency
iv
-------
REVIEWING ORGANIZATIONS
The following organizations contributed to the monograph's review and acceptance
by the professional community. The review process employed by each organiza-
tion is described in its acceptance statement Individual reviewers are, or are not,
listed according to the instructions of each organization.
Air & Waste Management
Association
The Air & Waste Management
Association is a nonprofit technical and
educational organization with more than
14,000 members in more than fifty
countries. Founded in 1907, the
Association provides a neutral forum
where all viewpoints of an environmen-
tal management issue (technical,
scientific, economic, social, political,
and public health) receive equal
consideration.
Qualified reviewers were recruited
from the Waste Group of the Technical
Council. It was determined that the
monograph is technically sound and
publication is endorsed.
The reviewers were:
Lee Dodge
RUST Remediation Services
Pleasanton, CA
James S trunk
Union Carbide Corporation
Bound Brook, NJ
American Institute of
Chemical Engineers
The Environmental Division of the
American Institute of Chemical Engineers
has enlisted its members to review the
monograph. Based on that review the
Environmental Division endorses the
publication of the monograph.
American Society of Civil
Engineers
The American Society of Civil
Engineers, established in 1852, is the
premier civil engineering association in
the world with over 124*000 members.
Qualified reviewers were recruited from
its Environmental Engineering division.
These individuals reviewed the
monograph and have determined that it
is acceptable for publication.
The reviewers were: .
G. Fred Lee, Ph.D., P.E., DEE
G. Fred Lee & Associates
El Macero, CA
Richard Reis, P.E.
EMCON
Bothell, WA
Hazardous Waste Action
Coalition
The Hazardous Waste Action
Coalition (HWAC) is the premier
business trade group serving and
representing the leading engineering
and science firms in the environmental
management and remediation industry.
HWAC's mission is to serve and
promote the interests of engineering and
science firms practicing in multi-media
environment management and
remediation. Qualified reviewers were
recruited from HWAC's Technical
Practices Committee. HWAC is
-------
pleased to endorse the monograph as
technically sound.
The lead reviewer was:
James D. Knauss, Ph.D.
President, Shield Environmental
Lexington, KY
Soil Science Society of
America
The Soil Science Society of America,
headquartered in Madison, Wisconsin,
is home to more than 5,300 profession-
als dedicated to the advancement of soil
science. Established in 1936, SSSA has
members in more than 100 countries.
The Society is composed of eleven
divisions, covering subjects from the
basic sciences of physics and chemistry
through soils hi relation to crop
production, environmental quality,
ecosystem sustainability, waste
management and recycling,
bioremediation, and wise land use.
Members of SSSA have reviewed the
monograph and have determined that it
is acceptable for publication.
The lead reviewer was:
John Zachara, Ph.D.
Pacific Northwest Laboratory
Richland, WA
Water Environment
Federation
The Water Environment Federa-
tion is a nonprofit, educational
organization composed of member
and affiliated associations throughout
the world. Since 1928, the Federation
has represented water quality
specialists including engineers,
scientists, government officials,
industrial and municipal treatment
plant operators, chemists, students,
academic and equipment manufac-
turers, and distributors.
Qualified reviewers were
recruited from the Federation's
Hazardous Wastes Committee and
from the general membership. It has
been determined that the document is
technically sound and publication is
endorsed.
The reviewers were:
James A. Kent, Ph.D.
t
Morgantown, WV
William Butler *
DuPont Environmental Remediation
Services -
Wilmington, DE
Vl
-------
ACKNOWLEDGMENTS
The WASTECH* project was conducted under a cooperative agreement
between the American Academy of Environmental Engineers* and the Office
of Solid Waste and Emergency Response, U.S. Environmental Protection
Agency. The substantial assistance of the staff of the Technology Innovation
Office was invaluable.
Financial support was provided by the U.S. Environmental Protection
Agency, Department of Defense, Department of Energy, and the American • V
Academy of Environmental Engineers*.
This multiorganization effort involving a large number of diverse profes-
sionals and substantial effort in coordinating meetings, facilitating communica-
tions, and editing and preparing multiple drafts was made possibly by a
dedicated staff provided by the American Academy of Environmental Engi-
neers* consisting of: •• '
William C. Anderson, P.E., DEE
Project Manager & Editor
John M. Buterbaugh
Assistant Project Manager & Managing Editor
Robert Ryan
Editor
Catherine L. Schultz
Yolanda Y. Moulden
Project Staff Production
J. Sammi Olmo
I. Patricia Violette
Project Staff Assistants
vii
-------
•/
t
-------
TABLE OF CONTENTS
Contributors jjj
Acknowledgements vii
List of Tables xv
List of Figures xvii
1.0 INTRODUCTION 1.1
1.1 Background 1.1
, •' >'
1.2 Chemical Treatment ,,1.4
1.3 Development of the Monograph *> 1.7
1,3.1 Background f 1.7
-------
2.5.4 Pretreatment Processes 2.18
2.5.5 Posttreatment Processes 2.19
2.5.6 Process Instrumentation and Control 2.20
2.5.7 Safety Requirements 2.20
2.5.8 Specification Development 2.20
2.5.9 Cost Data 2.21
2.5.10 Design Validation 2.23
2.5.11 Permitting Requirements 2.24
2.5.12 Performance Measures 2.24
2.5.13 Design Checklist 2.24
2.6 Implementation and Operation 2.'25
*
2.6.1 Implementation Strategies 2/25
2.6.2 Start-up Procedures <; 2.26
2.6.3 Operations Practices ^ ' 2.26
2.6.4 Operations Monitoring 2.27
2.6.5 Quality Assurance/Quality Control 2.27
2.7 Case Histories 2.27
2.7.1 Laboratory-Scale Tests 2.27
2.7.2 Pilot-Scale Tests 2.31
3.0 IN SITU PERMEABLE, ELECTROCHEMICALLY ACTIVE
METAL BARRIERS 3.1
3.1 Scientific Principles 3.1
3.2 Potential Applications 3.4
3.3 Treatment Trains 3.6
3.4 Remediation Goals 3.6
3.5 Design 3.6
3.5.1 Design Basis 3.6
3.5.2 Design and Equipment Selection 3.8
3.5.3 Process Modification 3.9
3.5.4 Pretreatment Processes 3.11
-------
idDie or cjonrenrs
3.5.5 Posttreatment Processes 3.11
3.5.6 Process Instrumentation and Control 3.11
3.5.7 Safety Requirements '3.12
3.5.8 Specification Development 3.12
3.5.9 Cost Data 3.13
3.5.10 Design Validation 3.15
3.5.11 Permitting Requirements 3.15
3.5.12 Performance Measures 3.15
3.5.13 Design Checklist 3,16
3.6 Implementation and Operation 3j&
3.6.1 Implementation Strategies 5:16
3.6.2 Start-up Procedures * 3.16
3.6.3 Operations Practices ,; 3.17
3.6.4 Operations Monitoring .« ' 3.17
3.6.5 Quality Assurance/Quality Control 3.17
3.7 Case Histories 3.17
4.0 SUPERCRITICAL WATER OXIDATION 4.1
4.1 Scientific Principles 4.1
4.2 Potential Applications 4.6
4.3 Treatment Trains 4.7
4.4 Remediation Goals 4.11
4.5 Design 4 13
4.5.1 Design Basis 4.13
4.5.2 Design and Equipment Selection 4.17
4.5.2.1 Materials of Construction/Corrosion Management 4.18
4.5.2.2 Heat Transfer 4.22
4.5.3 Process Modification 4.30
' 4.5.4 Pretreatment Processes 4.31
4.5.5 Posttreatment Processes 4.31
4.5.6 Process Instrumentation and Controls 4.33
x!
-------
icoe or v-ornenrs
4.5.7 Safety Requirements 4.34
4.5.8 Specification Development 4.34
4.5.9 Cost Data 4.36
4.5.10 Design Validation 4.38
4.5.11 Permitting Requirements 4.39
4.5.12 Performance Measures 4.39
4.5.13 Design Checklist 4.39
4.6 Implementation and Operation 4.40
4.6.1 Implementation Strategies 4.40
4.6.2 Start-up Procedures 4.40 ,
4.6.3 Operations Practices 4.40
4.6.4 Operations Monitoring . 4.41
4.6.5 Quality Assurance/Quality Control t; 4.41
4.7 Case Histories / ' 4.42
4.7.1 Commerical Activities 4.42
4.7.1.1 Eco Waste Technologies Me. 4.42
4.7.1.2 General Atomics 4.42
4.7.1.3 Sandia National Laboratories 4.45
4.7.2 Laboratory-Scale Study, Chemical Agent Treatment 4.46
4.7.2.1 GB Agent Treatment 4.46
4.7.2.2 VX Agent Treatment 4.47
4.7.2.3 Mustard Agent Treatment 4.49
4.7.3 Pilot-Scale Studies 4.53
. 4.7.3.1 Hydrolyzed Rocket Propellant Treatment 4.53
4.7.3.2 Paper Mill Wastes Treatment 4.54
4.8 Conclusion 4.59
5.0 EX-SITU HIGH VOLTAGE ELECTRON BEAM TREATMENT 5.1
5.1 Introduction 5.1
5.2 Process Description 5.2
5.3 Scientific Principles 5.4
xii
-------
Table of Contents
5.4 Potential Applications 5,6
5.5 Treatment Trains 5.7
5.6 Design 5.9
5.6.1 Design Basis 5.9
5.6.2 Design and Equipment Selection 5.10
5.6.3 Process Modification 5.10
5.6.4 Pretreatment Processes 5.11
5.6.5 Posttreatment Processes 5.11
5.6.6 Process Instrumentation and Control 5.11
5.6.7 Safety Requirements 5Mr
5.6.8 Specification Development 5.12
5.6.9 Cost Data * 5.12
5.6.9.1 Assumptions «/ 5.18
5.6.9.2 Site Preparation Costs g ' ' 5.20
5.6.9.3 Permitting and Regulatory, Costs 5.21
. 5.6.9.4 Mobilization and Start-up Costs 5.21
5.6.9.5 Equipment Costs 5.22
5.6.9.6 Labor Costs , 5.22
5.6.9.7 Supply Costs 5.23
5.6.9.8 Utility Costs 5.23
5.6.9.9 Effluent Treatment and Disposal Costs 5.24
5.6.9.10 Residual Waste Shipping and Handling Costs 5.24
5.6.9.11 Analytical Services Costs 5.25
5.6.9.12 Equipment Maintenance Costs 5.25
5.6.9.13 Site Demobilization Costs 5.25
5.6.9.14 Economic Analysis Conclusions 5.25
5.6.10 Design Validation 5.26
5.6.11 Permitting Requirements 5.28
5.6.12 Performance Measures 5.28
5.6.13' Design Checklist 5.28
xiii
-------
Table of Contents
5.7 Implementation and Operation 5.28
5.7.1 Implementation Strategies 5.28
5.7.2 Operation 5.29
5.8 Case History 5.30
5.8.1 Demonstration Procedures 5.31
5.8.2 Sampling and Analytical Procedures 5.32
5.8.3 Removal Efficiency 5.33
5.8.4 Effect of Treatment on Toxicity 5.38
5.8.5 Reproducibility of Treatment System Performance 5.41
Appendices
A. Ex-S!tu Electrochemical Treatment Processes A.I
A.I Electrochemical Coagulation A.2
A.1.1 ACE Technology t; A.4
A. 1.1.1 Process Description £ r A.4
A. 1.1.2 Technology Testing A.5
A. 1.1.3 Costs A.9
A. 1.1.4 Conclusions A. 10
A.1.2 Andco's Electrocoagulation Pilot Study A.12
A.l.2.1 Process Description A.12
A. 1.2.2 Treatment Levels A. 14
A.l.2.3 Test Results A.20
A. 1.2.4 Chemical Consumption A.21
A.2 Electrochemical Oxidation — Silver (H) Process A.25
B. List of References B.I
C. Suggested Reading List C.I
xlv
-------
LIST OF TABLES
Table Xiile. Page
1.1 Technologies Reviewed 1.9
2.1 Cation-Exchange Capacities of Common Clay Minerals 2.5
2.2 Contaminant Concentrations for In Situ Electrochemical
Pilot-Scale Tests 2.31
3.1 Halogenated Hydrocarbons Evaluated for Electrochemical
Reduction Using Iron as a Sacrificial Metal ^ 3.5
4.1 Kinetic Parameters for Key Rate-Controlling Intermediates • 4.4
4.2 SCWO Destruction Efficiency for Selected Organic !f
Compounds £ 4.10
4.3 SCWO Destruction Efficiency for Selected Organic Wastes 4.12
4.4 Global Kinetic Models for Supercritical Water Oxidation
of Organic Substances 4.14
4.5 Corrosion Results Summary , 4.20
4.6 Power Cost Analysis for SCWO 4.37
4.7 GB Agent Bench-Scale Test Matrix 4.46
4.8 Analytical Results for GB Agent Bench-Scale Tests 4.48
4.9 VX Agent Bench-Scale Test Matrix 4.48
4.10 Analytical Results for VX Agent Bench-Scale Tests 4.49
4.11 Mustard Agent Bench-Scale Test Matrix 4.50
4.12 Analytical Results for Mustard Agent Bench-Scale Tests 4.51
4.13 Composition of Feed Sludge and Product Ash 4.55
5.1 Comparison of Technologies for Treating VOCs in Water 5.5
5.2 Summary of Percent Removal of Various Organic
Compounds by Treatment Application Area 5.8
xv
-------
List of Tables
Table Title Page
5.3 Costs Associated with the E-Beam Technology — Case 1 5.14
5.4 Costs Associated with the E-Beam Technology — Case 2 5.16
5.5 E-Beam Treatment System Direct Costs 5.27
5.6 VOC Concentrations in Unspiked and Spiked Groundwater
Influent 5,34
5.7 VOC Removal Efficiencies (REs) * 5.36
5.8 Compliance with Applicable Effluent Target Levels , * 5.37
5.9 Acute Toxicity Data , ? '' 5.39
p
A.I Optimum Operating Conditions for Parallel Electrode
Unit Based on Bench-Scale Tests A.7
A.2 Electrochemical Iron Treatment Levels A. 14
A.3 Electrochemical Precipitation — Days 2 and 3 Pilot-Scale
Treatability Data • A. 15
A.4 Electrochemical Precipitation — Days 4, 5, and 6 Pilot-Scale
Treatability Data A. 18
A.5 Chemical and Electrical Power Consumption Per-Million
Gallons Treated A.22
xvl
-------
LIST OF FIGURES
Figure Title Page
2.1 . Schematic of Electromigration 2.3
2.2 Schematic of Electroosmosis 2.6
2.3 Indium-Coated Titanium Anode 2.13
2.4 Zinc-Coated Wire Cathode 2.14
.; *•
2.5 Galvanized Steel Electrode 2/15
2.6 Precipitation of Uranium Hydroxide at Cathode "** 2.18
2.7 Bench-Scale Electrochemical Cell ,; 2.28
1
2.8 One-Ton Soil, Pilot-Scale Electrochemical Cell ,; ' 2.29
2.9 Full-Scale Electroosmosis/Electromigration Treatment
System . 2.30
3.1 Schematic of Permeable Barrier 3.2
3.2 Conversion of TCE in Columns Packed with Mixtures of
Iron and Pyrite 3.10
4.1 Simplified Process-Flow Diagram of SCWO 4.8
4.2 Generalized Idealized Regimes for SCWO Reactor
Operations 4.16
4.3 Transpiring-Wall Platelet Reactor 4.21
4.4 . Overall Heat Transfer Coefficient as a Function of Core
Temperature 4.23
4.5 Viscosity of Water and Water Vapor in the Critical Region 4.25
4.6 Gross Separation Efficiency (as Penetration) for Two
Hydrocyclones 4.27
4.7 Grade Efficiency for Hydrocyclone Separation of Silica 4.29
4.8 Air Force SCWO Pilot Plant 4.43
xvii
-------
LIST or ngures
Figure Title Page
4.9 SCWO Reactor Temperature Profile 4.57
5.1 Elevation of the Electron Beam Research Facility,
Key Biscayne, Florida 5.3
5.2 VOC REs in Reproducibility Runs 5.40
A.I Schematic of an ACE Separator™ Used in Alternating-Current ,
Electrocoagulation ^ A.3
A.2 Simplified Schematic of the Silver (n) Process f A.28
xviil
-------
Chapter 1
INTRODUCTION
This monograph, covering the design and applications of Chemical Treat-
ment, is one of a series of seven on innovative site and waste remediation
technologies. This series was preceded by eight volumes published in 1994
and 1995 covering the description, evaluation, and limitations of the pro-
cesses. The entire project is the result of a multiorganization effort involving
more than 100 experts. It provides the experienced, practicing professional
guidance on the innovative processes considered ready for full-s^ale applica-
tion. Other monographs in this design and application series and the com-
panion series address bioremediation, liquid extraction: soil washing, soil
flushing, and solvent/chemical, stabilization/solidification, thermal desorp-
tion, thermal destruction, and vapor extraction and air sparging.
7.7 Background
An earlier book on chemical treatment (Weitzman et al. 1994) categorizes
the technology into three processes:
Substitution Processes that substitute a different functional group
for one or more functional groups on a target molecule. For
example, a mixture of potassium hydroxide and polyethylene
glycol (PEG) is used to replace one or more chlorine atoms on a
polychlorinated biphenyl (PCB) molecule with a PEG moiety.
The resulting molecule is not legally a PCB nor is it regulated by
the Toxic Substances Control Act. The hazard of the PEG moiety
is unknown.
Oxidation Processes that use an oxidizing agent, such as air,
oxygen, ozone, or hydrogen peroxide, to destroy organic
1.1
-------
irmoaucTion
molecules. Numerous techniques, such as Iron n catalysis, ultra-
violet light, or ionizing radiation, have been used to improve
oxidation by Or
Precipitation Processes that use techniques, such as pH adjust-
ment, addition of carbonates or sulfides, and reducing agents, to
transform a soluble compound of a metal into a less soluble form.
Precipitation is used for the treatment of aqueous materials con-
taminated with toxic inorganic elements and compounds. Its use
in the treatment of soils would normally be considered stabiliza-
tion, a technology which is covered in another monograph. The
procedure has been routinely used to treat wastewaters. Its appli-
cation to remediation situations is less common, buylata from
wastewater applications are applicable.
Recently, little new activity has taken place involving precipitation pro-
cesses. No specific uses of the technology were found that fit the; definition
of "innovative technology" used in this monograph. Therefore; these pro-
cesses are not discussed herein and the reader is referred to the first mono-
graph of this series, Chemical Treatment (Weitzman et al. 1994), for further
information on the subject.
No information could be obtained regarding the use of substitution pro-
cesses in pilot- or full-scale systems beyond the projects described in the
earlier monograph (Weitzman et al. 1994). Hearsay reports of the use of the
Base Catalyzed Decomposition (BCD) process for treating condensate from
thermal desorption systems (Beeman 1995; Lyons 1995) were encountered.
However, repeated efforts to obtain written reports or data from these field
programs were unsuccessful. A development program for the BCD process
is currently underway at the Naval Facilities Engineering Service Center in
Port Hueneme, California. The process under development consists of a
rotary reactor operating at 343°C (650°F) and a chemical treatment unit.
According to the developer, the soil, mixed with 5 to 10% sodium bicarbon-
ate is fed to the rotary reactor. PCBs are driven out of .the soil and collected
by condensation into a stirred tank reactor where they are chemically de-
stroyed. According to the information submitted by IT Corporation, the
contractor performing the work, efforts to date have focused primarily on the
rotary reactor portion of the process. For further information, the reader is
referred to the: Naval Facilities Engineering Service Center, 1100 23rd Av-
enue, 414ST, Port Hueneme, CA 93043.
1.2
-------
Because the applicability of substitution processes is limited to special or
isolated cases combined with the lack of field data, substitution processes are
not addressed in this monograph.
While this monograph focuses on innovative treatment methods, many
traditional oxidation and other wastewater treatment technologies are also
applicable to contaminated site remediation. Commonly used chemical
oxidants are air (oxygen), chlorine compounds (hypochlorous acid, chloram-
ines, chlorine dioxide, bromine chloride), permanganate, ozone, hydrogen
peroxide, and Fenton's reagent. Principles and applications of chemical
oxidation can be readily found in the literature (Weber 1972; Glaze 1990;
Cornwell 1990; James M. Montgomery Consulting Engineers 1985). Chlo^
rine compounds have been frequently used for disinfection in water treat-
ment. Other contaminants and undesirable properties that are'amenable to
chemical oxidation are iron, manganese, cyanide, phenols, taste arid odor,
color, disinfection byproducts, and other synthetic organics. W#ite (1972)
described the use of chlorine for treating potable water, waste^vater, and
cooling water. Design considerations for iron and manganese removal and
taste and odor control are described in Water Treatment Plant Design
(American Society of Civil Engineers and American Water Works Associa-
tion 1990).
Coagulation/flocculation has been used to coalesce small colloidal par-
ticles (clay and silt particles in natural water, chemical precipitates, etc.) to
form larger aggregates that can be removed by sedimentation and filtration.
The stability of colloidal particles is controlled by electrostatic interactions
and has been described using the theory of electrical double layer which is
directly related to the phenomenon associated with electroosmosis covered
in this monograph. Principles and applications of coagulation/flocculation
are described in terms of coagulant types (inorganic and organic), destabili-
zation mechanisms, and design considerations (Weber 1972; Amirtharajah
and O'Melia 1990; James M. Montgomery Consulting Engineers 1985;
American Society of Civil Engineers 1992; American Society of Civil Engi-
neers and American Water Works Association 1990).
Chemical precipitation precedes coagulation in removing dissolved metal
contaminants such as iron, manganese, hardness (calcium and magnesium),
phosphorus, and various heavy metals. Principles and applications of this
process can be readily found in the literature (Benefield and Morgan 1990;
James M. Montgomery Consulting Engineers 1985; Snoeyink and Jenkins
1.3
-------
1980; American Society of Civil Engineers and American Water Works As-
sociation 1990). The removal of phosphorus from municipal wastewater is
described in Design of Municipal Wastewater Treatment Plants (American
Society of Civil Engineers 1992).
Ion exchange has been used to remove inorganic ions (ammonium, heavy
metal, etc.). Ion-exchange materials are either inorganic (e.g., zeolite) or
organic (organic-polymer-based synthetic resins). Principles and design
factors of ion exchange can be readily found in the literature (Weber 1972;
Clifford 1990; Helfferich 1962; James M. Montgomery Consulting Engi-
neers 1985).
Adsorption has been used to remove dissolved organics from water. The,
most commonly used adsorbent is activated carbon, while some^ynthetic
resins are also used for selective organic compounds. The use of granular /
and powdered activated carbon and synthetic resins for removing organics
was described by Weber (1972), Snoeyink (1990), and in James f#. Mont-
gomery Consulting Engineers (1985). Design considerations
-------
b. convert the hazardous constituents into a less mobile form, for
example, by precipitation;
c. convert the hazardous constituent into a more mobile form, im-
proving the performance of a second treatment process that re-
moves the modified hazardous constituent from the nonhazard-
ous matrix (While such mobilization is conceptually possible, no
commercial or developmental chemical processes for doing this
were identified, and the concept is not covered herein.);
d. convert the hazardous constituent into a form that is more ame-
nable to subsequent treatment by another process. An example is,
the partial oxidation of contaminants in groundwater to convert
refractory (difficult-to-degrade) organics into compounds that are
amenable to biodegradation. ;
For organic contaminants, the ideal goal is complete mineralization — for
example, conversion of PCBs to sodium chloride, carbon dioxide, and water.
Realistically, however, the goal of most chemical treatment processes is
more modest; it is the conversion of selected target contaminants into un-
regulated or less toxic chemical forms. For example, replacing chlorine on a
PCB or chlorodibenzodioxin (dioxin) molecule with an aryl or alkyl group
using, for example, a sodium naphthalide reagent or with another functional
group such as a polyethylene glycol. Placement of the chlorine legally con-
verts the hazardous compound to a nonregulated substance. In many cases,
the long-term stability or environmental effects of such treatment is not
well-understood. For example, Hong et al. (1995) studied the genotoxicity
profiles of treated extracts from the dehalogenation of wood preserving
waste using the KPEG process (see Weitzman et al. 1994 for a description of
the KPEG process). Results showed that the KPEG process effectively
dehalogenated the pentachlorophenol in the wood preserving waste and that
the genotoxicity of the waste was reduced throughout the dechlorination
reaction. However, the genotoxicity was not completely eliminated and
further treatment was recommended to completely detoxify the waste.
For inorganic contaminants, the ideal goal is conversion to a nonhazard-
ous form. This can be achieved with elements such as chromium that have a
highly hazardous (i.e., hexavalent chromium) and a relatively nonhazardous
(i.e., trivalent chromium) oxidation state or with organometallic compounds
such as nickel carbonyl. When complete conversion is not possible, chemi-
cal treatment operates to convert metals to a less soluble, and hence, a less
1.5
-------
innoauciiori
teachable form. This latter goal impinges on stabilization and solidification,
treatment technologies that are covered in another monograph in this Series
by that name.
Chemical treatment is rarely used as the sole process. It may be used as a
pretreatment technique to enhance the efficiency of subsequent processes or
as a posttreatment step to polish an effluent. For example:
• various advanced oxidation techniques have been successfully
employed to soften organic compounds to improve their biode-
gradability;
• chemical dechlorination can be used to treat the contaminated
eluate from solvent extraction of chlorinated organics from
soil; and * /'
• chemical destruction can be used to treat the offgas from a
vapor-phase extraction process. (
Chemical treatment must be performed with a knowledgepof the
chemical reactions involved. Also, the nature of the treated material is
an important consideration when using chemical treatment. When the
reagents are mixed with the contaminated material to destroy or modify
the target contaminants, the "decontaminated material" still contains the
chemical reaction products and any residual reagent. These remains,
which are usually mobile, may be toxic, have a significant environmen-
tal impact on the surrounding ecosystem, or pose legal or safety con-
cerns. These impacts of chemical treatment remains have been a major
impediment to its use for direct treatment of soils.
Chemical treatment is technique — rather than process — oriented. To
determine the proper treatment method it is first necessary to identify the
target contaminant and then determine its availability and reactivity. Chemi-
cal knowledge is used to ascertain the type of chemical reactions to which.
the target compound(s) is amenable, to evaluate the available equipment, and
to select or design the appropriate treatment system — this monograph is
organized in a similar manner, grouping technologies by types of chemical
reactions used.
The technologies discussed in this monograph are shown in Table 1.1.
They have been grouped based on whether their application is in situ or
ex-situ. All of the techniques are based on some form of electron transfer,
where the target compound is (1) made available for treatment, (2) converted
1.6
-------
to a nonregulated form (this process should not be encouraged unless those
responsible will do the work necessary to be certain that the nonregulated
form is in fact nonhazardous to public health and the environment), or (3)
oxidized completely. Electrical processes are used to migrate materials to a
collection point, remove chlorine atoms from organic compounds, and co-
agulate and oxidize the target containments. The other processes for which
detailed design and application data are provided are supercritical water
oxidation and high voltage electron beam treatment, both ex-situ processes.
Appendix A discusses ex-situ chemical treatment processes which are
classified as emerging technologies. These technologies have been exten-
sively studied, but insufficient commercial application data exists to fully ,' /'
discuss them in the detail required. They are electrochemical coagulation'
(alternating-current electrocoagulation processes), and electrochemical oxi-
dation/reduction (the Silver (II) process). i
fr
If
f '
7.3 Development of the Monograph
1.3.1 Background
Acting upon its commitment to develop innovative treatment technologies
for the remediation of hazardous waste sites and contaminated soils and
groundwater, the U.S. Environmental Protection Agency (US EPA) estab-
lished the Technology Innovation Office (TIO) in the Office of Solid Waste
and Emergency Response in March, 1990. The mission assigned TIO was to
foster greater use of innovative technologies.
In October of that same year, TIO, in conjunction with the National
Advisory Council on Environmental Policy and Technology (NACEPT),
convened a workshop for representatives of consulting engineering
firms, professional societies, research organizations, and state agencies
involved in remediation. The workshop focused on defining the barriers
that were impeding the application of innovative technologies in site
remediation projects. One of the major impediments identified was the
lack of reliable data on the performance, design parameters, and costs of
innovative processes.
1.7
-------
The need for reliable information led TIO to approach the American
Academy of Environmental Engineers®. The Academy is a
long-standing, multidiscipliriary environmental engineering professional
society with wide-ranging affiliations with the remediation and waste
treatment professional communities. By June 1991, an agreement in
principle (later formalized as a Cooperative Agreement) was reached
providing for the Academy to manage a project to develop monographs
describing the state of available innovative remediation technologies.
Financial support was provided by the US EPA, U.S. Department of
Defense (DoD), U.S. Department of Energy (DOE), and the Academy.
The goal of both TIO and the Academy was to develop monographs
providing reliable data that would be broadly recognized and accepted'
by the professional community, thereby eliminating or at least minimiz-
ing this impediment to the use of innovative technologies. t
The Academy's strategy for achieving the goal was founded^on a
multiorganization effort, WASTECH* (pronounced Waste Te^h), which
joined in partnership the Air and Waste Management Association, the
American Institute of Chemical Engineers, the American Society of
Civil Engineers, the American Society of Mechanical Engineers, the
Hazardous Waste Action Coalition, the Society for Industrial Microbiol-
ogy, and the Water Environment Federation, together with the Academy,
US EPA, DoD, and DOE. A Steering Committee composed of highly-
respected representatives of these organizations having expertise in
remediation technology formulated the specific project objectives and
process for developing the monographs (see page iv for a listing of
Steering Committee members).
By the end of 1991, the Steering Committee had organized the
Project. Preparation of the initial monographs began in earnest in Janu-
ary, 1992, and the original eight monographs were published during the
period of November, 1993 through April, 1995. In Fall of 1994, based
upon the receptivity of the industry and others of the original mono-
graphs, it was determined that a companion set, emphasizing the design
and applications of the technologies, should be prepared as well. At this
time the Soils Science Society of America joined the WASTECH® con-
servation. Task Groups were identified during 1995 and work com-
menced on this second series.
1.8
-------
Table 1.1
Technologies Reviewed
Process
Contaminant Types
Process Description
Means of Treatment
1. In Situ Electromigration,
Electroosmosis
All dissolved organics and inorganics
2. In Situ Electrochemical
Reduction
3. Ex-Situ Oxidation by
Supercritical Water
Oxidation
Chlorinated organic compounds and oxidized metals
All organic compounds and inorganic salts
4. Ex-Situ Destruction by
Electron Beam Irradiation
All organic compounds and some inorganic ions
Electrodes are embedded in a contaminated
site and a DC voltage is applied between them.
In the simplest applications, only two
electrodes, an anode and a cathode, are used;
larger sites require anodes and multiple
cathodes.
A permeable barrier consisting of iron metal
powder is installed downgradient of the
contaminated site. Impermeable barriers may
be added to funnel the groundwater through
the permeable barrier.
An aqueous stream containing (usually) high
concentrations of organic materials is mixed
with an oxidant (usually oxygen gas or
hydrogen peroxide) in water at temperatures in
the range of 350'C (662°F) to 600'C (1,112'F)
and pressures of 17 MPa (2,500 psi, 170 atm)
or greater.
An aqueous stream containing relatively low
concentrations of organic materials is passed
over a weir at ambient temperature and
pressure. The cascading"film of water it
irradiated with a scanning high;energy electron
beam (analogous to a cathode rayJTV tube).
The electron beam forms hydrogen and
hydroxyl radicals which react with the organic
compounds. Metals in solution may have their
oxidation state's altered. ""
Metals and soluble organic
chemicals migrate to the
electrodes where they
concentrate in the groundwater
which is pumped out to treatment.
The iron in the barrier reacts with
the chlorinated organic
compounds in the water to
remove the chlorine.
The organics are oxidized;
inorganic salts precipitate out at
these conditions.
The organic compounds are
mineralized to a high level. Metal
oxides or hydroxides may
precipitate out.
I
-------
Introduction
1.3.2 Process
For each of the series, the Steering Committee decided upon the technolo-
gies, or technological areas, to be covered by each monograph, the mono-
graphs' general scope, and the process for their development and appointed a
task group of experts to write a manuscript for each monograph. The task
groups were appointed with a view to balancing the interests of the groups
principally concerned with the application of innovative site and waste
remediation technologies — industry, consulting engineers, research, aca-
deme, and government. . .
The Steering Committee called upon the task groups to examine and
analyze all pertinent information available, within the Project's financial
and time constraints. This included, but was not limited to, th6 compre-
hensive data on remediation technologies compiled by US EPA, the
store of information possessed by the task groups' members^ that of
other experts willing to voluntarily contribute their knowledge, and in-
formation supplied by process vendors.
To develop broad, consensus-based monographs, the Steering Com-
mittee prescribed a twofold peer review of the first drafts. One review
was conducted by the Steering Commi'ttee itself, employing panels con-
sisting of two members of the Committee supplemented by at least four
other experts (see Reviewers, page iii, for the panel that reviewed this
monograph). Simultaneous with the Steering Committee's review, each
of the professional and technical organizations represented in the Project
reviewed those monographs addressing technologies in which it has
substantial interest and competence.
Comments resulting from both reviews were considered by the Task
Group, appropriate adjustments were made, and a second draft published.
The second draft was accepted by the Steering Committee and participating
organizations. The statements of the organizations that formally reviewed
this monograph are presented under Reviewing Organizations on page v.
1.10
-------
Chapter 1
1.4 Purpose
The purpose of this monograph is to further the use of innovative chemi-
cal treatment site remediation and waste processing technologies, that is,
technologies not commonly applied, where their use can provide better,
more cost-effective performance than conventional methods. To this end, the
monograph documents the current state of chemical treatment technology.
7.5 Objectives * ( >.
The monograph's principal objective is to furnish guidance fjSr experi-
enced, practicing professionals, and users' project managers.f'The mono-
graph, and its companion monograph, are intended, therefore, not to be
prescriptive, but supportive. It is intended to aid experienced professionals
in applying their judgment in deciding whether and how to apply the tech-
nologies addressed under the particular circumstances confronted.
In addition, the monograph is intended to inform regulatory agency per-
sonnel and the public about the conditions under which the processes it ad-
dresses are potentially applicable.
1.6 Scope
The monograph addresses innovative chemical treatment technologies
that have been sufficiently developed so that they can be used in full-scale
applications. It addresses all aspects of the technologies for which sufficient
data were available to the Chemical Treatment Task Group to review the
technologies and discuss their design and applications. Actual case studies
were reviewed and included, as appropriate.
The monograph's primary focus is site remediation and waste treatment.
To the extent the information provided can also be applied elsewhere, it will
provide the profession and users this additional benefit.
1.11
-------
Introduction
Application of site remediation and waste treatment technology is site
specific and involves consideration of a number of matters besides alterna-
tive technologies. Among them are the following that are addressed only to
the extent that they are essential to understand the applications and limita-
tions of the technologies described:
• site investigations and assessments;
• planning, management, specifications, and procurement;
• regulatory requirements; and
• community acceptance of the technology. ^
/
I
i
y
1.7 Limitations *
The information presented in this monograph has been prepared in accor-
dance with generally recognized engineering principles and practices and is
for general information only. This information should not be used without
first securing competent advice with respect to its suitability for any general
or specific application.
Readers are cautioned that the information presented is that which
was generally available during the period when the monograph was pre-
pared. Development of innovative site remediation and waste treatment
technologies is ongoing. Accordingly, postpublication information may
amplify, alter, or render obsolete the information herein about the pro-
cesses addressed.
This monograph is not intended to be and should not be construed as a
standard of any of the organizations associated with the WASTECH® Project;
nor does reference in this publication to any specific method, product, pro-
cess, or service constitute or imply an endorsement, recommendation, or
warranty thereof.
1.12
-------
Chapter J
1.8 Organization
This monograph is organized under a uniform outline and addresses the
design and application of five innovative chemical treatment technologies
available for site remediation.
For each, the following are discussed:
• scientific principles on which the technology is founded;
• potential applications of the technology;
• treatment trains, including definition of the point of application
in a complete remediation and essential pre- and post^treatment
processes; ^
i
• design related guidance covering: .• '
• remediation goals,
• design basis,
• design and equipment selection,
• process modifications,
• pretreatment processes,
• posttreatment processes,
• process instrumentation and controls,
• safety requirements,
• specification development,
• cost data,
• design validation,
• permitting requirements, and
• performance measures.
• implementation and operation issues including:
• implementation strategies,
• start-up procedures,
• operations practices,
1.13
-------
Introduction
• operations monitoring, and
• Quality Assurance/Quality Control; and
case histories of laboratory- and pilot-scale applications of the
technology.
1.14
-------
Chapter 2
IN SITU
ELECTROCHEMICALLY INDUCED
PROCESSES
r
,, j
2.1 Scientific Principles
In situ electrochemical remediation uses electric current and potential to
enhance the transport of contaminants in groundwater or to convert metal
compounds to less mobile precipitates. To apply the technology, electrodes
are embedded in the contaminated region and a DC Voltage is applied be-
tween them. In the simplest applications, only two electrodes, an anode and
a cathode, are used; larger sites require multiple anodes and cathodes. The
electric current between the anodes and cathodes:
• electrolyzes a fraction of the groundwater, forming acidic and
caustic zones near the anodes and cathodes, respectively;
• increases the relative mobility of some soluble organic and inor-
ganic compounds, causing the compounds to migrate to the elec-
trodes; and
• reduces the solubility of some metals near the cathodes by form-
ing less-soluble metal hydroxides or carbonates.
Once soluble contaminants concentrate at the electrodes, the contami-
nated groundwater is pumped out and treated. In theory, insoluble organic
compounds could migrate as oil droplets. However, the low permeability of
most soils makes this impractical.
2.1
-------
In Situ Electrochemically Induced Processes
A variation of electrochemical remediation replaces the electrodes with a
sacrificial grid or powdered metal that is embedded in an area where con-
taminated groundwater passes through and makes contact with the powdered
metal (O'Hannesin and Gillham 1992; Gillham and O'Hannesin 1994). The
metal is oxidized, serving as an electrochemical galvanic half-cell, while
contaminants are reduced to complete the electrochemical cell. This varia-
tion is called permeable barrier treatment.
The concept of using in situ electrodes to chemically oxidize or reduce , -
contaminants remains a possibility. Laboratory tests using ex-situ electro- •
chemical cells have shown that it is possible to destroy organic compounds
such as PCBs (Zhang 1995) and other organics (see the discussion of the :
Silver (II) process, Appendix A) by a variety of chemical means. £Jo data
beyond the laboratory phase were found. At this stage, the use of jnduced
electric current in situ appears to be restricted to improving the mobility of
metals and organic compounds and the following discussion is restricted to
this application.
The fundamental electrochemical processes whereby contaminants are
removed or destroyed are electromigration, electrophoresis, electroosmosis,
electrocoagulation, and electrochemical reduction. Electromigration occurs
when a charged ion in solution is transported under an electric field, whereas
electrophoresis occurs when, instead of an ion, a charged particle is in-
volved. However, since the movement of colloidal particles in compacted,
low-permeability soils is not practical, electrophoresis is not usually consid-
ered in remediation. Electroosmosis occurs when a thin liquid layer around
a charged particle, which contains charged ions, moves relative to a station-
ary and oppositely-charged surface under an electric field. Electrocoagula-
tion occurs when metal ions, which come from the electrochemical oxidation
of an anodic metal such as iron or aluminum, are used as coagulants for the
coagulation of contaminant metal ions. This process has not been used for
in situ remediation. Electrochemical reduction occurs when a sacrificial
metal (e.g., iron) is oxidized to induce the chemical reduction of organic
compounds (e.g., chlorinated organic solvents). Electrochemical reduction
has been tested in the form of permeable barriers for treatment of
solvent-contaminated groundwater. Permeable-barrier treatment that does
not require an external input of electricity is covered in Chapter 3.
Electromigration and electroosmosis that require an external input of elec-
tricity are covered in this chapter.
2.2
-------
Chapter2
2.1.1 Electromigration
The process of electromigration has been described by many researchers
and developers (Acar 1992; Acar 1993; Acar, Alshawabkeh, and Gale 1993;
Acar et al. 1995; Lindgren, Mattson, and Kozak 1994; Marks, Acar; and
Gale 1994; Mattson and Lindgren 1995; Probstein and Hicks 1993). In this
process, an array of electrodes (cathodes and anodes) is inserted into soil
with a potential difference on the order of a few hundred V/m (Acar and
Alshawabkeh 1993). In this electric field, cations (e.g./metal ions) move to,
the cathodes whereas anions (e.g., cyanide complexes, metal-hydroxide ..,'.'
complexes, anionic dyes, and chromate) move to the anodes (see Figure 2^1).
Figure 2.1
Schematic of Electromigration
S//SSSSS/SSS
Process Control System
Extraction/
Exchange
Extraction/
Exchange
- CATHODIC
'-'PROCESS
VFLUID
ACID FRONT
and/or ANODIC
PROCESS FLUID
s'D?8 a?r?s- f1""16 «°. £car et al •Etectrokinetic Remediation: Basis and Technology
Thei Netherlands permission of Elsevier Science - NL, Sara Burgerhartstraat 25,1055 KV Amsterdam,
2.3
-------
In Situ Electrochemically Induced Processes
At the electrodes, electrolysis of water takes place as follows:
At the anode,
2H20->4H++02(g) + 4e- (2.1)
At the cathode,
2H20 + 2e -»20H" + H2 (g) (2-2)
This electrolysis results in an acidic front at the anode and an alkaline -
front at the cathode, which move to the cathode and the anode, respectively.
The propagation of the acid and base fronts promotes the dissolutio^ of
metal ions near the anode and the precipitation of the metal ions near the
cathode. These conditions significantly affect (1) the pH and ioni<£ strength
of pore water, (2) the mobility and solubility of metal contaminants,
(3) charge conditions of soil particles, and (4) the hydraulic conductivity of
the porous media. Depending on the type of contaminants of concern, these
conditions could be significant or minimal. For example, the pH could drop
to around 2 at the anode and increase to around 12 at the cathode (Acar and
Alshawabkeh 1993; Kahn and Alam 1993).
The pH affects the precipitation equilibria of metal hydroxides and car-
bonates (high pH effects more precipitation). Khan and Alam (1993) dem-
onstrated in laboratory studies the dissolution of metal precipitates (lead,
manganese, and zinc) in soil due to the acid front from the anode and the
subsequent migration of the dissolved metal ions to the cathode. Runnells
and Wahli (1993) observed the transport of copper and sulfate ions toward
the cathode and anode, respectively, and the precipitation of copper hydrox-
ide near the cathode in a column of fine quartz. Probstein and Hicks (1993)
determined that zinc removal from clay occurred mainly by electromigration
and diffusion (electroosmotic flow was negligible) and that zinc precipitation
occurred at the isoelectric point where the acid and base fronts converge.
Marks et al. (1994) discussed the role of H+ in the cation-exchange equi-
libria of a soil-metal ion system. Hydrogen ions in the acid front displace
metal ions that are adsorbed on soil particles by simple ion exchange and
surface complexation mechanisms, resulting in more mobile metal ions in
pore water that could be transported by electromigration. Cation-exchange
capacities of common clay minerals are shown in Table 2.1. Because reduc-
tion reactions occur at the cathode, some metal ions could be reduced to
2.4
-------
Chapter 2
elemental metals and deposited on the surface of the electrode. Therefore, in
electromigration, several processes can take place simultaneously: dissolu-
tion, ion exchange, migration, precipitation, and reductive deposition.
The transport of charged metal ions under an electrical field was de-
scribed as similar to diffusive transport due to a concentration gradient
(Shapiro and Probstein 1993; Acar and Alshawabkeh 1993). Acar et al.
(1995) reported an ionic migration of 1 to 80 cm/day (0.39 to 31.5 in./day)
under an electrical field of 100 V/m.
Table 2.1 V
Cation-Exchange Capacities of Common Clay Minerals
Mineral
Kaolinite
Halloysite(2H2O)
Hal!oysite(4H2O)
niite
Allophane
Montmorillonite
Vermiculite
Source: Marks, Acar,
Structural Control
Unsatisfied valences on edges of structures
Unsatisfied valences on edges of structures
Unsatisfied valences on edges of structures
Octahedral/tetrahedral substitutions, edges and K*
deficiency between layers
Amorphous structure, unsatisfied valences
Octahedral/tetrahedral substitutions and edges
Replacement interlayer cations, substitutions
and Gale 1994; Carrels and Christ 1965
Exchange Capacity
(meq/100gatpH7)
3-15
5-15
40-50
10-40
70
70-100
100-150
2.1.2 Elecfroosmosis
The movement of a thin, charged layer is responsible for electroosmosis
in a system similar to the one shown in Figure 2.1 for electromigration. The
concept of an electrical double layer has frequently been used to describe
electrostatic interactions between negatively charged particles (e.g., clay and
silt) and positively charged ions in water (added as a coagulant) when
2.5
-------
In Situ Electrochemically Induced Processes
removing the particles from water via coagulation (Amirtharajah and O'Melia
1990). The particles are negatively charged in natural water because of their
negative zeta potential, -20 to -40 mV according to Amirtharajah and O'Melia
(1990) and -10 to -100 mV according to Probstein and Hicks (1993).
In this double-layer concept, there are two electrical layers around a
charged particle: (1) a compact layer of negative ions on the surface of the
particle and (2) a diffuse layer of positive ions that are electrostatically at-
tracted to the compact layer and somewhat dispersed into the water because
of their thermal motion. This positively-charged diffuse layer (1 to 10 nm ,,
according to Probstein and Hicks (1993)) becomes mobile in electroosmosis
as the layer is pulled to the cathode under an electrical field (see Figure 2.2).
Electroosmosis is only effective for low-permeability, fine-grained, soils that
have a hydraulic conductivity of less than 1 • 10'5 cm/sec (3.9 • l$6in./
sec)(Segall and Bruell 1992). For a hydraulic conductivity greater than 1 •
10'5 cm/sec, the electroosmotic effect is nullified by backflow from the cath-
ode. The effectiveness of electroosmosis is also reduced by the production
of H+ at the anode and other sources of cations, since the diffuse layer be-
comes compressed as the concentration of positive ions in water increases.
Figure 2.2
Schematic of Electroosmosis
w
r
ELECTROOSMOTIC HEAD 4
"'•t,1*'
£- * u, 11LI| ,,, -n - - r- fr
v
DC CURRENT/VOLTAGE
Cathode
Reprinted from Journal of Hazardous Materials, Volume 40, Acar et al.,-EtectroWnetic Remediation: Blasfe and Technology
Status," PP 117-137.1995 with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25,1055 KV Amsterdam.
The Netherlands.
2.6
-------
Electroosmosis causes a convective movement of pore water, which is
different from electromigration or diffusion. Therefore, any contaminants
(either ionic or neutral) in the fluid can be transported to the cathode along
with the fluid. The movement of ionic species would be enhanced if they
were cationic or hampered if they were anionic because of the effect of
electromigration. Comparing phenol and acetic acid, Shapiro, Renauld, and
Probstein (1989) attributed the lower electroosmotic flow rate for acetic acid
to its lower pH, which compresses the diffuse layer. They also observed a
lower removal of acetic acid when the pH was high at the cathode because of
the back-migration of ionized acetic acid molecules to the anode.
Jk
The aqueous solubility and adsorbability of organic compounds can Effect
their removal by electroosmosis. Bruell et al. (1992) observed that organic
solvents with relatively high aqueous solubility and low adsorbability (e.g.,
benzene, toluene, trichloroethylene, and m-xylene) were easpiy removed
from water-saturated kaolin clay. They also observed that solvents with
relatively low aqueous solubility and high adsorbability (hexane and isooc-
tane) were not transported easily. The possibility of solubilizing relatively
insoluble compounds with surfactant was suggested (Probstein and Hicks
1993). Marks, Acar, and Gale (1994) proposed an in situ bioremediation
method in conjunction with electroosmosis to deliver a nutrient-containing
solution by electroosmosis to microorganisms in low-permeability soil.
Electroosmotic flow is described by an equation similar to Darcy's law
(Marks, Acar, and Gale 1994; Segall and Bruell 1992):
ue = 2s. = keie (2.3)
where: ue = electroosmotic velocity (m/sec);
Qe = electroosmotic flow rate (m3/sec);
A = cross-sectional area (m2);
ke = electroosmotic conductivity or electroosmotic coefficient
of permeability (m2/V-sec); and
ie = electrical gradient (V/m).
The variable ke can be related to zeta potential and the viscosity of water
according to the Helmholts-Smoluchowski theory (Hunter 1982; Probstein
1989; Acar and Alshawabkeh 1993). According to Segall and Bruell
(1992), ke varies from 10"9to 10'10 mW-sec for a wide range of soils, result-
ing in an electroosmotic velocity of 10'5 to 10'6 cm/sec under an electric
gradient of 100 V/m. The maximum electroosmotic flux was reported to be
2.7
-------
In Situ Electrochemically Induced Processes
approximately 10"4 cm/sec at 100 V/m (Acar et al. 1995), indicating that the
electroosmotic velocity is expected to be approximately 0.1 to 10 cm/day (0.04
to 4 in./day) at 100 V/m. Overall, transport of a compound is determined by the
combination of electroosmosis, electromigration (if ionized), and diffusion.
The transport of metal ions in an electric field occurs because of both
electroosmosis and electromigration. Although the transport of metal ions is
dominated by electromigration in most cases, it is difficult to determine
precisely which is the primary one because it depends on soil characteristics,.
applied energy level, and production and transport of hydrogen and hydrox-
ide ions at the electrodes. Therefore, the transport of metal ions is discussed
in the potential applications, Section 2.2, relative to both electroosmosis and
electromigration. ,;
2.2 Potential Applications
Electromigration involves the movement of ionic contaminants in pore
water toward oppositely-charged electrodes and does not require the move-
ment of the water being treated. Therefore, electromigration is not depen-
dent on pore size. Although it can be applied to both high- and
low-permeability soils, electromigration may not be suitable in
high-permeability soils because a simple pump-and-treat method may be
more convenient, flexible, and cost-effective. Thus, electromigration is typi-
cally used in conjunction with electroosmosis for low-permeability soils.
Electromigration can be applied to only ionic contaminants (e.g., metal ions
and dissociated organic acids and bases) and is not suitable for the removal
of neutral contaminants (e.g., undissociated organic acids and bases and
organic solvents).
Electroosmosis, on the other hand, does involve the movement of pore
water — any contaminants that are dissolved in the water are transported to
the cathode. Therefore, this process can be used for both ionic and nonionic
contaminants. At the same time, electromigration will take place because of
the electric field and will affect the transport of ionic contaminants.
When compared to other in situ technologies that only target one group of
contaminants, organic or inorganic, electroosmosis is advantageous because it
can be applied to a broad range pf contaminants, forms, and concentrations. In
2.8
-------
Chapter 2
addition, most in situ remediation technologies are ineffective for re-
moval of contaminants from low-permeability soils (e.g., fine-grained
soils). However, with electroosmosis, the electric field provides a high
degree of hydraulic control. The direction of flow can be controlled by
appropriately placing anodes and cathodes, and the electroosmotic flow
can be initiated or stopped by applying or discontinuing electrical cur-
rent. Since electroosmosis is not a pressure-driven process, channeling
is also minimized.
Electroosmosis depends on porosity and zeta potential and is not affected''
by pore-size distribution. Acar and Alshawabkeh (1993) indicated that the'
maximum electroosmotic flow often occurs in low-activity clays ("activity"
is defined as the plasticity index divided by the percent of clay particles less
than 2 jjm in size) with high water content and low ionic strength'. Elec-
troosmosis was first used in the 1930s in Germany to dewater-and stabilize
soils (Probstein and Hicks 1993). Stabilization occurs becaus^ consolidation
through dewatering alters the physical and chemical properties of the soils
(Cabrera-Guzman et al. 1990).
In summary, electroosmosis can be applied in conjunction with
electromigration to remove metals and organics in low-permeability soils.
Required energy input depends on soil types and conditions and types of
contaminants. An acid or a base may be introduced at the electrodes as a
pretreatment or as part of the overall treatment to enhance the transport and
removal of contaminants.
2.3 Treatment Trains
In situ electrochemical treatment is rarely the only form of treatment at a
contaminated site. This method concentrates contaminants and is usually
part of a remediation program that includes one or more of the following
components:
• diversion systems, such as reduced-permeability walls, to reduce
groundwater infiltration to the contaminated area;
• covers or caps that reduce rainwater infiltration to the area;
• monitoring wells to allow sampling of the groundwater;
2.9
-------
In Situ Electrochemically Induced Processes
wells enabling the injection of solutions to modify contaminant
migration or the reinjection of treated water;
wells through which groundwater is pumped to the surface to
depress the groundwater level and/or remove contaminated
groundwater;
treatment systems to remove contaminants from the extracted
groundwater; and
air pollution control equipment to capture volatile organic com-
pounds (VOCs) that may be released from the wells, water treat-
ment system, or pumps. ^
2.4 Remediation Goals *
The goal of electrochemically-induced processes is to concentrate the
contaminants preferentially in the vicinity of the electrodes so that the con-
centrations of the contaminants in the contaminated zone would be below
target contaminant levels.
2.5 Design
2.5.1 Design Basis
Design of an electrochemical treatment system is highly site-specific.
The system's geometry is governed by the site characteristics and the proper-
ties of the soil and the contaminants. Materials of construction for the sys-
tem, especially those of the electrodes, depend on the nature of the contami-
nants, the natural materials found at the site, and the products of electrolysis
of these substances.
Once the area of contamination is determined, the contaminated soil must
be defined with respect to permeability, moisture content, cation-exchange
capacity, organic content, and pore water characteristics (pH, alkalinity, ionic
2.10
-------
Chapter 2
strength, etc.). This process is most suitable for low-permeability soils. For
high-permeability soils, conventional pump-and-treat methods may be more
suitable and should be investigated first. An adjacent high-permeability
region could adversely affect the decontamination of a low-permeability
region by electroosmosis by providing a return flow.(Segall and Bruell
1992). Therefore, spatial variation in permeability and contamination needs
to be carefully assessed before electroosmosis is applied.
Moisture content is important because both electroosmosis and
electromigration require moisture. Electromigration was reported to be • ''
effective in soil with a moisture content as low as 7% (Lindgren, Kozak, and
Mattson 1991). Electroosmosis was originally used to dewaterand stabilize
soils, mine tailings, and mineral sediments (Probstein and Hicks 1993). •
Cation-exchange capacity is important for transport of ions by
electromigration and movement of the acid from the anode, typical
cation-exchange capacities of clay minerals are provided in Table 2.1. Soil
with a high cation-exchange capacity is expected to slow the transport of
contaminant cations to the cathode by exhibiting a high affinity for the ions
and keeping the pH of the pore water from decreasing (i.e., keeping metal
hydroxides and carbonates from dissolving).
The organic content of soil has been reported to be responsible for
adsorbing hydrophobic organic compounds (Mills et al. 1985). The parti-
tioning between the pore water and the soil has been frequently described by
using octanol/water partition coefficients. Therefore, the transport of organ-
ics via electroosmosis will depend on the affinity of the soil for the com-
pound of interest, especially when a purge solution is introduced to push the
contaminant through the soil pore. As mentioned earlier, Bruell, Segall, and
Walsh (1992) established that organic solvents with relatively high aqueous
solubility and low adsorbability (relatively hydrophilic) were easier to re-
move than solvents with low solubility and high adsorbability (relatively
hydrophobic). The use of surfactants in the purge solution may enhance the
mobility of hydrophobic compounds.
The pH of the pore water affects the solubility of metal hydroxides and
carbonates and thus the transport of metal ions by electromigration. The
alkalinity of pore water represents its buffering capacity — high alkalinity
pore water resists pH changes that could be caused by the arrival of the acid
and base front from the electrodes. High alkalinity may result in:
2.11
-------
In Situ Electrochemically Induced Processes
• relatively small amounts of metal-ion transport due to low disso-
lution of metal hydroxides and carbonates near the anode; and
• reduced metal precipitation near the cathode.
The ionic strength of the pore water may affect the thickness of the dif-
fuse double layer by changing the zeta potential and, therefore, the rate of
electroosmotic flow. High ionic strength could be caused by either back-
ground ions or contaminant ions. Probstein and Hicks (1993) suggested that
a two-step process may occur for the case of high ionic strength due to con-, .
taminant ions consisting of:
• the removal of contaminant ions by electromigratioif, which
eventually increases zeta potential and thus the thicknes? of the
double layer; and
-------
Chapter 2
Figure 2.3
Iridium-Coated Titanium Anode Used by Electrokinetics, Inc.
Reproduced courtesy of Electrokinetics, Inc. (1995)
The configuration of electrode systems with respect to the number of
electrodes, spacing, and orientation needs to be considered for maximum
hydraulic control. Since most contaminants to be removed are metal ions
and nonionic organics, the contaminants are transported toward the cathode.
In this case, an electrode system, consisting of a cathode and multiple an-
odes around the cathode is frequently used to collect the contaminants at the
cathode (e.g., a polygon shape with a cathode at the center and anodes at the
corners). The electrodes could be placed horizontally or vertically.
Geokinetics, a European remediation contractor, used an .electrode system
that consisted of one long horizontal cathode (0.5 m (1.65 ft) below the
ground surface) and a row of vertical anodes (up to 1 m (3.3 ft) deep and 1 m
(3.3 ft) apart) (Lageman 1993). Bruell, Segall and Walsh (1992) suggested
3 m (10.9 ft) electrode spacing. Probstein and Hicks (1993) suggested an
2.13
-------
In Situ Electrochemically Induced Processes
electrode spacing of 2 to 10 m (6.6 to 33 ft) and an electrode depth of 2 to
20 m (6.6 to 66 ft). For a relatively large area, several electrode systems can
be installed.
The net flux of contaminants toward either the cathode or the anode due
to electromigration, electroosmosis, and diffusion can be estimated using the
equations in Section 2.1. For example, Equation 2.3 represents the contribu-
tion of electroosmosis by considering the magnitude of applied electric po-
tential, zeta potential, viscosity of water, and electroosmotic permeability as
described by Acar and Alshawabkeh (1993) and Shapiro and Probstein
(1993). An ionic migration of 1 to 80 cm/day (0.4 to 31 in./day) and an
electroosmotic velocity of 0.1 to 10 cm/day (0.04 to 4 in./day) afan electric
gradient of 100 V/m (2.5 V/in.) were reported. I '
i/
Figure 2.4
Zinc-Coated Wire Cathod Used by Electrokinetics,lnc.
Reproduced courtesy of Electrokinetics, Inc. (1995)
2.14
-------
Chapter 2
Power requirements can be estimated based on a desirable contaminant
flux. Probstein and Hicks (1993) suggested the following ranges of power
requirements:
• 40 to 200 V for applied electrical potential;
• 20 to 200 V/m (0.5 to 5 V/in.) for applied electric field
strength; and
• 50 to 500 mA/cm2 (320 to 3,200 mA/in.2) for current density.
The range of current density is consistent with the values used by Electfd
kinetics (US EPA 1995c) and Geokinetics (Lageman 1993). Acar et al.
(1995) noted that increasing current density does not necessarily increasl
removal efficiency.
, i
Figure 2.5
Galvanized Steel Electrode Used by Electrokinetics, Inc.
Reproduced courtesy of Electrokinetics, Inc. (1995)
2.15
-------
In Situ Electrochemically Induced Processes
Photovoltaics may be the ideal DC source because they produce a DC cur-
rent, the voltages from individual photovoltaic panels are in the appropriate
range, and storage requirements are likely unnecessary (Probstein and Hicks
1993). When AC is used as a power source, the AC has to be converted to DC.
When anionic contaminants (e.g., cyanide complexes, metal-hydroxide
complexes, anionic dyes, and chromate) are to be removed by
electromigration, the flux toward the anode due to electromigration should
be greater than the flux toward the cathode due to the combined effect of
electroosmosis and diffusion.
In electroosmosis, the flow rate will eventually diminish if a*purging solu-
tion is not introduced at the anode or if precipitation at the cathode decreases
the efficiency of the process. In most cases, removal of one pore,,volume is
not sufficient to remove contaminants to regulatory levels. Alspjthe compo-
sition of the purging solution must not adversely affect the zetl potential of
the soil (and thus the thickness of electrical diffuse double layer) so that the
electroosmotic flow rate can be maintained.
2.5.3 Process Modification
The main way to adapt electrochemical processes'to accommodate vary-
ing site conditions is through the use of purge solutions. Purge solutions can
accelerate the removal of contaminants by increasing their solubility in wa-
ter. When the alkalinity of the pore water is relatively high, an acid solution
can be introduced at the anode to increase the solubility of metal hydroxides
and carbonates by lowering the pH of the water to an optimum value. The
added acid also affects the ion-exchange equilibrium and facilitates the des-
orption of metal ions from soil particles. However, the strength of the acid
solution must be carefully selected to preclude the resulting pH of the pore
water from adversely affecting the zeta potential of the soil, and thus the
electroosmotic flow rate. Because the zeta potential of a typical soil is nega-
tive, electroosmotic flow is toward the cathode. If the pore water pH is too
low, the zeta potential could be reversed, causing the electroosmotic flow to
flow toward the anode.
A key condition in controlling the electroosmotic flow direction is under-
standing the zero point of charge (ZPC). ZPC represents a pH where the
zeta potential of a particle is zero. For example, kaolinites have a ZPC of
3.3 to 4.6 depending on the clay source (Parks 1967; Shapiro and Probstein
2.16
-------
Chapter 2
1993). Clay has a positive zeta potential below the ZPC and a negative po-
tential above the ZPC. If the strength of the acid solution is selected such
that the resulting pH of the pore water drops below the ZPC, the electroos-
motic flow direction will be reversed. If the resulting pH is close to (but not
below) the ZPC, the flow direction may not be reversed; however, the flow
rate may decrease .significantly because of a compressed double layer. Of
course, other cations can also compress the double layer and thus affect the
zeta potential. Therefore, the ionic strength of the purge solution and the
type of cation used (e.g., the charge of the cation) also need to be consideredr
to determine the composition of the purge solution. ';/
The acidic purge solution may also complement the hydrogen ions pro-
duced at the anode to neutralize the hydroxide ions produced at the* cathode
and prevent metal hydroxides and carbonates near the cathode. )por example,
an acetic acid solution (0.05 M) was used in a treatment proems'to enhance
the removal of uranyl ion and to prevent the precipitation of the ion near the
cathode (Acar and Alshawabkeh 1993) (see Figure 2.6). When the electroos-
motic flow rate is low, an acid solution or water may be introduced around
the cathode to flush or neutralize the hydroxide ions produced, maintain a
neutral pH, and prevent the metal ions from precipitating.
Flushing the hydrogen ions produced at the anode may be necessary when
the resulting pore water pH is too low to cause an adverse impact on electroos-
motic flow or on the integrity of the clay mineral structure by dissolving silica
and alumina. An alkaline solution or water can be used for this purpose.
Instead of using an acid or alkaline solution to control the production of
hydrogen and hydroxide ions at the anode and cathode, respectively, the use
of membrane or ion-exchange materials around the electrodes has been sug-
gested (Probstein and Hicks 1993; Marks, Acar, and Gale 1994). The use of
a chelating agent (e.g., ethylenediamine tetraacetic acid [EDTA]) was re-
ported to enhance the mobility of metals by forming metal-EDTA complexes
(Allen and Chen 1993). The use of a surfactant solution as a purge solution
was also suggested to enhance the solubility of relatively hydrophobic or-
ganic contaminants (Marks, Acar, and Gale 1994).
The use and selection of a purge solution needs to be carefully evaluated
in terms of regulatory requirements and potential environmental impacts. A
selected purge solution has to be acceptable to regulatory agencies, and all
precautions have to be taken to minimize potential dispersal of contaminants
beyond the zone of contamination.
2.17
-------
In Situ Electrochemically Induced Processes
Figure 2.6
Precipitation of Uranium Hydroxide at Cathode
("Yellow Cake")(Electrokinetics, Inc.)
Reproduced courtesy of Electrokinetics, Inc. (1995)
2.5.4 Pretreatment Processes
Lageman (1993) suggested removal of conducting objects of larger than
10 cm (4 in.) (e.g., tins, barrels, reinforcing rods) as a pretreatment whenever
possible because these objects may function as preferential flow paths for
the electrical current and delay the movement of contaminants. He also
indicated that nonconducting objects (e.g., wooden beams, plastic sheets,
concrete blocks) may interfere with electroosmosis/electromigration. In
addition, any subsurface pipes and cables should be located prior to treat-
ment so that they can be protected.
2.18
-------
Chapter 2
Pretreatment methods which can improve the mobility of groundwater
through the soil or to improve the contaminants' solubility and hence their
mobility should be evaluated. In addition, solutions of acids or caustics can
be injected into the contaminated formations to open up the soil structure.
Such pH modification could also increase the solubility, and hence the mo-
bility of metal contaminants.
2.5.5 Posttreatment Processes
In electrochemical processes, the groundwater collected from the t].
region around the cathode (or the anode if the contaminants of concern
are anionic) will be concentrated with contaminants. The ground water
must be pumped to the surface, collected, treated, and disposed. The
treatment methods used will depend on the types of contaminants at the
site. If the contaminants are organic, the liquid can be treated using
many traditional separation or destruction processes such'as adsorption,
stripping, biological treatment, chemical or photochemical oxidation, or
any of their combinations. If the contaminants are inorganic, they could
be separated by chemical precipitation, ion exchange, or reverse osmo-
sis. The description of these processes can be readily found in a number
of books, papers, and reports listed in Appendix B.
The recovery of some purging agents (e.g., complexing agents and surfac-
tants) may be necessary. Allen and Chen (1993) described an electrolytic
process to recover EDTA from an EDTA-lead solution which was used to
chemically extract lead from a contaminated soil. In the recovery process, as
EDTA-lead complexes are electrolytically destroyed, the lead is deposited on
a copper cathode while the EDTA is released. To prevent the EDTA from
being electrolytically oxidized at the anode, the anode was separated from
the solution by a cation-exchange membrane. This recovery process may be
directly applied to in situ electromigration/electroosmosis by recovering
heavy metals at the cathode as a deposit and returning the EDTA solution to
the anode, which is separated by a cation-exchange membrane.
Used surfactants may be recovered by ultrafiltration, a membrane separa-
tion process, or by sieving out organic contaminants which are generally
larger than surfactants. However, the cost of recovering spent purging agents
should be compared with that of continuously adding the agents and treating
them as contaminants.
2.19
-------
In Situ Electrochemicaily Induced Processes
2.5.6 Process Instrumentation and Control
The power applied and the composition of purging agents need to be
monitored and controlled based on design parameters. Additionally, the
effectiveness of the process needs to be assessed by monitoring the quality
of treated groundwater (see discussion regarding design validation [Section
2.5.10]).
2.5.7 Safety Requirements
The potential health hazards associated with in situ electrochemical site'
remediation are generally chemical in nature and involve the contaminants of
•concern (both organics and inorganics) and the chemicals used for j
remediation. To protect a worker from the chemical hazards, the ^ireshold
limit values, the short-term exposure limits, and the immediate,danger to life
and health levels for the chemicals involved need to be identifild. After
determining these levels for the chemicals involved in all forms (vapors,
dust, and liquids), appropriate control actions need to be taken to provide a
safe environment for workers. Depending,on the types and nature of the
chemicals involved, protective devices (e.g., respirators, gloves, clothing,
boots, safety glasses, etc.) must be specified to protect the workers against
potential hazards to the respiratory system, skin, and eyes.
The use of electricity also presents a hazard either as a direct contact with
electrified positive anodes or contact with soil surfaces that are in contact
with electricity. The treatment areas where the maximum voltage is ex-
pected to be over 2 to 10 V should be fenced to exclude entry while power is
applied. In addition, warning signs about the hazards associated with elec-
tricity need to be posted. Before work is performed using any equipment at
the site, the electricity should be turned off to avoid any electric shock. Use
of protective devices to avoid electric shock is also necessary.
2.5.8 Specification Development
The key requirements that must be incorporated in specifications for an
electrochemical treatment application depend on whether the bids are for
equipment which is to be installed by others, or for a turnkey electrochemi-
cal system with performance guarantees.
2.20
-------
Chapter 2
If a vendor is to provide a turnkey system with performance guarantees,
such guarantees must be based on the overall site characterization and the
overall treatment scheme (beyond just the electrochemical system) that is to
be used at the site. The site description should incorporate all of the issues
discussed in the design checklist in Section 2.5.13. Namely, the vendors
should have a clear understanding of the nature of the contaminants and how
they are distributed. It is important to provide information on the actual
chemical form of the contaminants. For example, simply specifying that a
site contains mercury is insufficient. Mercury can be found as a metal, as • •' >'
methyl mercury, or in another chemical form; each form will behave in ail'
entirely different way under the influence of the electric field^To success-
fully integrate an electrochemical process into the overall remediation, tKe
vendor must be apprised of the overall scheme. An in situ electrochemical
process is only part of a system that can include infiltration confrols, ground-
water diversion systems, injection wells, and pumping system's.
Specification of individual equipment to be installed (for example, elec-
trodes and power supply) is fairly straightforward. Electrical equipment
should be specified to allow for a far wider variation in operation than that
envisioned in the system design. For example, the DC power supply must be
capable of withstanding a sudden rise in the level of the groundwater that
would otherwise cause it to short circuit. The types of electrical safeguards
required to prevent destruction of the power supply must be clearly defined
to the vendors.
Electrode specifications must clearly state the types of metals and alloys
used for different portions. Corrosion resistance is a crucial concern. It is
relatively easy to reduce the cost of the electrodes by, for example, shorten-
ing the length of the parts made of a unique metal; however, this could be
offset by a reduced electrode life.
It is preferable to give equipment vendors performance, rather than de-
sign, specifications. Vendors may have proprietary systems that can meet
the performance requirements at a lower cost. Vendor involvement in the
specification process is essential.
2.5.9 Cost Data
Cost data for electrochemical remediation are still being developed; how-
ever, the processes appear to be competitive with other in situ processes.
2.21
-------
In Situ Electrochemically Induced Processes
Installation and equipment costs for electrochemical systems are relatively
modest. The major operating cost elements (above those for a normal
groundwater flow modification and pumping system) are electric power and
electrode replacement.
Probstein and Hicks (1993) estimated energy costs for electroosmosis to
be approximately $1.10/tonne ($I/ton) of soil treated using:
• a value of 20 kWh/m3 (15kWh/yd3), which was taken from the
range of 10 to 20 kWh/m3 (8 to 15 kWh/yd3) reported by Segall , > /
and Bruell (1992);
• $0.10/kWh,
• a soil porosity of 50%; 4;
• a dry specific gravity of 3; and ,. »
• two pore volumes of purging.
They also estimated the energy costs for electromigration to be ap-
proximately $2.20/tonne ($2/ton)— an estimate that is more problem-
atic because electromigration is concentration-dependent and
voltage-specific. This estimate was based on power requirements for
electromigration of approximately 40 kWh/m3 (30 kWh/yd3) as reported
by Hamed, Acar, and Gale (1991).
Considering a safety factor of 10, Probstein and Hicks (1993) predicted
the energy costs for electroosmosis in conjunction with electromigration to
be approximately $22 to $33/tonne ($20 to $30/ton). This is comparable
with an energy cost estimate of $16 to $33/tonne ($15 to $30/ton) with a
power usage of 60 to 200 kWh/m3 (46 to 150 kWh/yd3) for removing lead
from kaolinite specimens (Acar et al. 1995). However, it is not clear why
these cost estimates were so close for quite different values of power usage.
The cost is expected to vary substantially because of a wide range of re-
ported power usage: 18 to 39 kWh/m3(14 to 30 kWh/yd3) for removing phe-
nol (Acar, Li, and Gale 1992); 200 kWh/m3 (150 kWh/yd3) for removing
hydrocarbons (Bruell, Segall, and Walsh 1992); 300 to 700 kWh/m3(230 to
540 kWh/yd3)(US EPA 1995c), and 65 to 300 kWh/m3 (50 to 230 kWh/
yd3)(Lageman 1993). Some of these values are an order of magnitude higher
than the value used by Probstein and Hicks.
2.22
-------
Chapter 2
The cost of electrode replacement depends to a large extent on the charac-
teristics of the water and contaminants at the site and can vary widely. No
meaningful general estimate of this cost can be made at present.
2.5.10 Design Validation
In almost all applications, electrochemical remediation is part of a
long-term remediation program. As such, it is crucial that the design be
validated early in the program. The system's purpose is to concentrate con- /•
taminants at the electrodes. Hence, the most direct validation program is,to
measure the contaminant concentration in the groundwater iiuand around the
contaminated zone before and after an electric potential is applied to the?
electrodes. The data analysis would validate whether or not the,/treatment is
effective in removing the contaminants and also that the contanjfinants stay in
the zone. Such a sampling and analysis program must use statistical tech-
niques that account for normal variability in contaminant concentration;
Several techniques can be used to validate the design.
The simplest would be to monitor the discharge wells around the elec-
trodes for a period of one year prior to energizing them. Sampling times
would be chosen to be representative of a variety of weather and climatic
conditions. A one-year interval would allow for sample collection during the
various seasons. At the start of the second year, the electrodes would be
energized and samples would be collected at times that appear to best dupli-
cate the site conditions during the first year of sampling. Statistical analysis
of results from the two sets of samples will identify if the treatment is effec-
tive or if further adjustment is warranted.
The above validation program has several limitations. One is that there is
no guarantee that the weather conditions between the years would be the
same. Such variability could mask the system's performance. Another limi-
tation is that it may take a long time to determine whether remediation sys-
tem enhancement is necessary.
An alternative validation scheme is to implement a cyclic technique. Ini-
tially, samples would be collected for a few days after installing but before
energizing the electrodes. Then, the system would be energized for the same
number of days, the sampling would be repeated, and the system would be
shut off. This process could be repeated as necessary for validation. The
length of time for the cycles should be greater than the time required for
2.23
-------
In Situ Electrochemically Induced Processes
contaminants to migrate to the electrode regions. This "pulsed" validation
technique should reduce the impact of seasonal variability on sampling and
analysis results and can be used to fine-tune the system's operation, e.g., the
system could be adjusted based on the contaminants' migration velocities.
2.5.11 Permitting Requirements
The types of information that must be supplied to the regulatory agencies
to justify use of electrochemical processes will generally be the same as that
required for a traditional pump-and-treat system. In addition, it will be nec-
essary to provide complete information on the design and operating condi-
tions for the system, chemicals (chelating agents for metals, buffering agents
for electrodes, etc.) to be used for treatment, and any potential byproducts
such as evolution of gases at cathodes. Finally, it will be necessary to dem-
onstrate that the system can achieve the remediation goals without adverse
local environmental impact.
2.5.12 Performance Measures
The intent of the electrochemical treatment systems is to increase the flow
of the contaminants to wells for removal from the site. Therefore, the main
performance measure for such systems is simply the amount of contaminants
removed per unit of groundwater pumped for treatment.
2.5.13 Design Checklist
A. Local conditions, soils, and geologic formations
1. Porosity and permeability of the soil(s) in the system area of
influence
2. Location of natural and artificial barriers (buried metallic or
non-metallic objects) to flow and electric current and their impact
on the migration of groundwater and contaminants toward the
electrodes
3. Changes in water levels that might short electrodes
4. Naturally-occurring materials (i.e., salt) that might impact the
process
5. Special site-specific requirements to protect the system from
weather
2.24
-------
Chapter 2
B. Contaminant types
1. Presence of inorganic materials that might clog pores over time
because of electrochemical precipitation
2. Behavior of all contaminants (not just those targeted) under the
influence of the electric field
3. Interaction between contaminants and naturally-occurring mate-
rials at the site
.»
C. Electrode composition ,., .
1. Corrosion characteristics of the electrodes *.
/
2. Anticipated lifetime in situ i
-------
In Situ Electrochemlcally Induced Processes
Rather, the implementation of such a system must begin with a good under-
standing of the site and the contaminants. This information should be
coupled with treatability studies using simple electrical cells and soil
samples collected from the site. Conducting treatability studies on actual
•site samples is especially crucial for this technology since small differences
in chemistry can have a major impact on process performance.
The vendor used for these treatability studies can be the supplier of the
equipment or an independent party; however, a strong background in electro7
chemical processes and diffusion under electric fields is essential. The ven--
dor should be capable of interpreting the results of the treatability studies
and creating a set of performance specifications for the equipment. The :
equipment itself can be acquired through normal procurement channels.
While competitive bidding is desirable, this application has not bejfen applied
extensively in the field, and ultimately, vendor selection shouldlbe based not
only on a cost comparison, but also on an assessment of the vendor's experi-
ence with similar applications.
2.6.2 Start-up Procedures
Once installed, startup of an electrochemical system is relatively straight-
forward. After baseline contaminant concentrations have been established,
the power is slowly ramped up to the pre-established design levels. A slow
increase in power at startup insures that there are no unknown contingencies
such as buried metal or pockets of salt that could short circuit the system.
Injection of solutions into the site to improve the mobility of contaminants
(i.e., chelating agents) should not be started until the system's performance
without these agents has been established. Injection should also be gradu-
ally and electrosomotically made in increments to reach the target levels.
2.6.3 Operations Practices
No specific operation practices appear necessary for electrochemical
processes other than maintaining liquid injection and withdrawal rates and
power levels. However, groundwater levels and power usage should be
monitored, and the process should be adjusted to reflect changes in these
parameters over time.
2.26
-------
Chapfer 2
2.6.4 Operations Monitoring
Monitoring programs for electrochemical systems should track two com-
ponents: (1) the degree of concentration of the contaminants and pH and
temperature changes at and around the electrodes, and (2) process conditions
such as electricity usage, current, and potential.
For the second component, process conditions must be monitored to iden-
tify changes in the site over time. For example, under constant temperature,
a change in current at constant potential (voltage) or potential at constant' ''
current could indicate a chemical change in the system. Power usage in'd
DC system is the product of current and potential. * f
(
2.6.5 Quality Assurance/Quality Control (QA/QC)f
,, ?
As with any remediation effort, QA/QC is an integral partfof electro-
chemical process implementation. The QA/QC program should be devel-
oped on two general levels. The first is to meet the operating needs of the
system. This should include the tests discussed in Section 2.5.10 which
ascertain overall system performance.
The second level is intended to satisfy regulatory requirements; the QA/
QC program must clearly demonstrate that the system is meeting all environ-
mental and legal goals and requirements.
2.7 Case Histories
Both laboratory- and pilot-scale tests have proven electrochemical pro-
cesses to be successful techniques for site remediation. The results of some
of these test efforts are shown in Figures 2.7 through 2.9 and described in the
following sections.
2.7.1 Laboratory-Scale Tests
At the laboratory-scale, electroosmosis has been applied to a wide range
of soils and contaminants. Laboratory tests have shown effective cleanup is
possible with electroosmosis, but it depends on many variables including
types of contaminants, pH, initial concentrations, and adsorption and
cation-exchange capacity of soil.
2.27
-------
In Situ Electrochemlcally Induced Processes
Figure 2.7
Bench-Scale Electrochemical Cell
Reproduced oourtwy of EtotrokJmrtlo*, Inc. (1995)
Organic compounds are amenable to electroosmosis. Bruell, Segall, and
Walsh (1992) achieved a removal of 15-25% for benzene, toluene, trichloro-
ethylene, and m-xylene from kaolin clay in only 2-5 days of treatment.
Shapiro and Probstein (1993) and Probstein and Hicks (1993) removed over
90% of phenol with an initial concentration of 450 mg/L from compacted
kaolin clay samples by extracting less than 1.5 pore volumes — an indica-
tion that electroosmosis could be a very effective remediation method. A
lower removal was observed at a lower initial concentration (45 mg/L) and
was attributed to the adsorption of phenol on the clay. The effect of initial
concentration on removal efficiency was more pronounced for acetic acid
(Shapiro and Probstein 1993), where acetate ions electromigrated to the
anode (the degree of dissociation of acetic acid is higher at a higher pH,
which is caused by a low concentration of acetic acid).
2.28
-------
Chapter 2
Figure 2.8
One-Ton Soil Pilot-Scale Electrochemical Cell
Reproduced courtesy of Electrokinetics, Inc. (1995)
For removal of metal ions, Pamukca and Wittie (1992) demonstrated,
using laboratory-prepared synthetic samples, that 85 to 95% of cadmium,
cobalt, nickel, and strontium were removed from commercially-obtained
kaolinite and bentonite, prepared clayey-sand, and prepared/washed New
Jersey beach sand. They attributed the removal to the movement of the acid
front toward the cathode which caused dissolution of metal precipitates and
the desorption of adsorbed metal ions. Of the soils tested, metals were the
2.29
-------
In Situ Electrochemically Induced Processes
most difficult to remove from the bentonite. This is consistent with the
higher ion-exchange capacity of montmorillonite, a major ingredient of ben-
tonite, as indicated in Table 2.1. In addition, the treatment of clayey sand
and bentonite was most influenced by the chemistry of the individual metal
ions. The removal of metals is also affected by the buffering capacity (alka-
linity) of soils, which resists a pH drop caused by an advancing acid front.
The buffering capacity is due to the cation-exchange capacity and precipi-
tates such as calcium carbonate (Acar and Alshawabkeh 1993). To dissolve
metal precipitates in highly-buffered soils, additional hydrogen ions may be • "
needed and can be introduced in the form of an acid solution at the anode.'
Figure 2.9 £
Full-Scale Electroosmosis/Electromigration
Treatment System by Electrokinetics, Inc.
Light-weight HOPE liner material on surface reduces evaporation and escape of volatiles.
Reproduced courtesy of Electrokinetics. Inc. (1995)
2.30
-------
Chapter 2
Other ions were also found to be amenable to removal such as lead, man-
ganese, and zinc (Khan and Alam 1993); copper and sulfate (Runnells and
Wahli 1993); nitrate (Segall and Bruell 1992); uranium (Acar and
Alshawabkeh 1993; Ugaz et al. 1994); and zinc (Probstein and Hicks 1993).
The removal of radium and thorium was found to be poor due to the forma-
tion of insoluble precipitates in soil (US EPA 1995c; Ugaz et al. 1994).
2.7.2 Pilot-Scale Tests
, •' r
Two companies conducted pilot-scale tests and cleanup projects: (I)'/ '
Electrokinetics, Inc. (Baton Rouge, Louisiana) used 1-ton specimens of
kaolinite and a mixture of fine sand and kaolinite for its pilot-scale test6
(US EPA 1995c) and (2) Geokinetics (The Netherlands) reported on two
field pilot-tests and three cleanup projects at various sites (paiijlt factory,
galvanizing plant, timber-impregnation plant, landfill, and r&ilitary depot)
covering a surface area of 50 to 2,800 m2 (60 to 3,350 yd2) and a depth of 1
to 2.6 m (3 to 8 ft)(Lageman 1993). The contaminants (all metals) and
concentrations in the tests are listed in Table 2.2.
Table 2.2
Contaminant Concentrations for In Situ
Electrochemical Pilot-Scale Tests
Contaminant Concentration in mg/L (test)
Arsenic 400-500 (Geokinetics)
Cadmium 2-3,400 (Geokinetics)
Chromium less than 300 (Geokinetics)
Copper 500-1,000 (Geokinetics)
Lead 100 to less than 5,000 (Geokinetics)
850-5,322 (Electrokinetics)
Nickel 860 (Geokinetics)
Zinc 7,010 (Geokinetics)
2.31
-------
In Situ Electrochemically Induced Processes
Electrodes were separated by 0.7 m (2.3 ft)(Electrokinetics) and 1 m (3.3 '
ft) (Geokinetics). Electrokinetics applied a one-dimensional electric field for
the tests, whereas Geokinetics used a system of horizontal cathodes buried
0.5 m (1.6 ft) below the ground surface and vertical anodes 1 m (3.3 ft) deep
and 1 m (3.3 ft) apart). Reported ranges of power consumption were similar:
300 to 700 kWh/m3 (230 to 540 kWh/yd3)(Electrokinetics) and 65 to ap-
proximately 300 kWh/m3 (50 to 230 kWh/yd3)(Geokinetics). However, a
value of as high as 800 kWh/m3 (600 kWh/yd3)was suggested to achieve a
treatment goal in one Geokinetics test. Values of charge density reported •
were 133 |Wcm2 (858 |jA/in.2)(Electrokinetics) and 400-800 uA7cm2 (2,600
to 5,200 ^A/in.2)(Geokinetics).
Most tests resulted in successful remediation (70 to >90% remoy^l) after
one to several months of operation. The following suggestions anf observa-
* f
tions were noted: . J*
• Electrokinetics suggested the depolarization of the cathode using
acetic acid to prevent the precipitation of lead near the cathode;
• the buffering capacity (alkalinity) of soil influences the transport
of metal ions. A low-pH soil (pH 4) facilitated the mobilization
of lead at a low-energy dosage, whereas a highly-buffered soil
retarded the mobilization of zinc requiring an additional energy
input to lower the pH to between 3 and 4;
• metallic objects (>10 cm in size) in soil interfered with the re-
moval by providing preferential paths for electrical current and
should be removed in a pretreatment step along with insulating
objects (plastic, wood, and concrete), whenever possible; and
• concretions of cadmium sulfide (a few mm to several cm in size)
were found to prolong the removal. Two pretreatment methods
were suggested: (1) the use of acid to dissolve cadmium from the
concretions and (2) the removal of concretions by sieving.
In summary, the researchers found that effective cleanup of contaminated
low-permeability soil is possible with electroosmosis/electromigration.
However, many design and operational variables need to be considered for
this technology to be cost-effective.
2.32
-------
Chapters
IN SITU PERMEABLE,
ELECTROCHEMICALLY ACTIVE
METAL BARRIERS
•'/•
3.1 Scientific Principles
In situ permeable electrochemically-active metal barriers can treat
groundwater contaminated with dissolved halogenated organic compounds
and certain types of oxidized metals. The process is based on the fact that
many common contaminants, both organic and inorganic, react with iron and
other metals in their elemental (zero-valence) state. Conceptually, the treat-
ment process is very simple, see Figure 3.1. A permeable barrier consisting
of a trench or a wall structure filled with granular iron or other metal is
placed in the flow path of the contaminated groundwater passing through the
site. Contaminants in the water react with the metal and are reduced to less
environmentally objectionable and more controllable forms as the water
flows through the permeable barrier. This technology has been described in
the literature by a number of different names such as porous-reactive walls,
permeable walls, reactive iron walls, and permeable reaction walls.
The organic chemical reduction process is similar to the electrochemical
processes previously discussed with the exception that the electrodes in
those processes are replaced with a sacrificial metal such as aluminum,
brass, copper, iron, or zinc. Permeable barriers can treat halogenated or-
ganic nitrates, organic nitrites and certain metals. Halogenated organic con-
taminants are dehalogenated. Metals, such as hexavalent chromium, are
reduced to less hazardous or less soluble forms.
3.1
-------
In Situ Permeable, Electrochemically Active Metal Barriers
Figure 3.1
Schematic of Permeable Barrier
m
I
1
Because of its comparatively low price and ready commercial availability,
iron has been the most widely-tested and used metal. Field trials have
shown granular iron barrier walls to be successful in dechlorinating chlori-
nated organic compounds in groundwater. Laboratory data indicate that an
iron barrier can also reduce hexavalent chromium to trivalent chromium.
However, this reduction process may form precipitates which could cause
significant plugging problems. When contaminated groundwater passes
through this "permeable wall of iron metal" (O'Hannesin and Gillham 1992;
Gillham and O'Hannesin 1994), the metal is oxidized, serving as an electro-
chemical galvanic half-cell, and contaminants and any other electron accep-
tors (e.g., O2) are reduced to complete the electrochemical cell.
In field application, permeable barriers can be installed to either cover the
entire plume of contaminated groundwater or intercept the plume using a
"funnel-and-gate" approach with impermeable barriers depending on site
conditions (Focht et al, 1996; Shoemaker et al. 1995). The permeable bar-
rier consists of ground iron or another metal appropriate for the application.
The finer the particle size of the metal, the greater the surface area to which
the contaminated groundwater is exposed, and hence, the faster the rate of
chemical destruction. However, with finer paniculate, the permeability of
the barrier decreases. Therefore, design requires testing to ensure a balance
between these two characteristics. In either case, permeable barriers are
3.2
-------
Chapter 3
subject to plugging as oxidation products, carbonates, and other precipitates
accumulate in the pores.
When iron is selected as a sacrificial reactive metal, the following corro-
sion or electrochemical reactions take place within the barrier (Matheson
andTratnyek 1994; Wilson 1995; Sivavec and Homey 1995):
Fe-»Fe+++2e' (3.1)
At the same time, contaminants are reduced. An example of groundwatef
contaminated with a halogenated hydrocarbon and hexavalent chromium' -' '
follows: ^
Cr*+3e--»Cr+3
X- ,; (3.2)
I* ' (33)
where: R = an alkyl group; and
X = a halogen.
The following are the overall chemical reactions:
X- (3.4)
2Cr+6 + 3Fe -> 2Cr+3 + 3Fe++ (3.5)
The trivalent chromium that is formed in the preceding chemical reaction
is the relatively insoluble precipitate, chromium hydroxide. The aliphatic
chlorides are converted to the relatively environmentally-benign and readily
biodegradable aliphatic compounds. The chlorine in the-aliphatic chlorides
is converted to chloride.
The chemical reduction of many chlorinated orgam'cs using iron metal
was found to follow first-order kinetics (O'Hannesin and Gillham 1992;
Gillham and O'Hannesin 1994; Tratnyek 1966; Cipollone et al. 1997). Re-
search and field results have shown that degradable hydrocarbons can be
degraded to below detection limits given sufficient contact time.
One of the major factors that influence the electrochemical reduction is
pH. As the reaction proceeds, the pH of water increases as shown in Equa-
tions 3.2 and 3.4. When the pH is greater than 8, the reaction slows substan-
tially (Senzaki and Kumagai 1988, 1989; Senzaki 1991; Matheson and
3.3
-------
In Situ Permeable, Electrochemically Active Metal Barriers
Tratnyek 1994). The high pH also causes metals to precipitate, leading to
the loss of porosity of the flow path. In the case of the reduction of
hexavalent chromium, the production of chromium hydroxide is beneficial
for immobilizing the chromium within the barrier; however, this same pre-
cipitation clogs the pores of the permeable barrier, and a concomitant de-
crease in the rate of water flow through the barrier.
For additional information on the use of semi-permeable metal barriers in
the field and a general description of the chemical processes and some of the
design considerations, the reader is referred to Matheson and Tratnyek
(1994). For a more detailed discussion of the chemistry involved and of
recent research and filed results, see Johnson, Sheerer, and Traffiyek (1996).
{
J
I*
3.2 Potential Applications
If permeable barrier treatment is used as an alternate containment
strategy, high permeability in contaminated soil is not required. How-
ever, if permeable barrier treatment is used as a primary strategy to clean
up contaminated soil and groundwater, it requires 'relatively high perme-
ability (a hydraulic conductivity of much higher than 1 • 10'5 cm/sec) in
the contaminated area so that contaminants can move out of the soil and
through the permeable barrier. In addition, the treatment has to be de-
signed such that contaminants do not escape the area without being
treated. For example, the barrier needs to be more permeable than the
surrounding formation to avoid the buildup of hydraulic pressure behind
the barrier. The addition of impermeable barriers at strategic locations
may be necessary depending on site conditions.
According to Vogan et al. (1995) and Gillham and O'Hannesin (1994),
various halogenated hydrocarbons can be reduced by this form of treatment.
Table 3.1 lists halogenated hydrocarbons that were tested for electrochemical
reduction. Many of these hydrocarbons rapidly degrade with half lives rang-
ing from a few minutes to a day where the iron surface area is 1 m2/rnL.
Gillham and O'Hannesin (1994) compared the half lives from electrochemi-
cal reduction with those from natural abiotic degradation. As shown in Table
3.1, the natural abiotic degradation rates were many orders of magnitude
slower than the electrochemical reduction rates.
3.4
-------
Chapter 3
Table 3.1
Haiogenated Hydrocarbons Evaluated for Electrochemical
Reduction Using Iron as a Sacrificial Metal
Initial Concentration,
Compounds (Mg/L)
Dibromomethane, CH2Br2
Dichloromethane
(Methylene Chloride), CH2Q2
Trichloromethane
(Chlorofonn), CHCl,
Tribromomethane
(Brpmoform), CHBrj
Tetrachloromethane
(Carbon Tetrachloride), CC14
Chloroethane, CHjCHjQ
1 ,2-Dichloroethane, C1CH 2(3^0
1,1,1-Trichloroethane, CHjCCl3
1 , 1 ,2,2-Tetrachlorocthane,
QjCHCHClj
1,1,1 ,2-Tetrachloroethane,
QjCCHCl
Hexachloro ethane
(Perchloroethane), CCljCdj
Chloroethene
(Vinyl Chloride), CH2CHC1
1 ,1 -Dichloroethene
(Vinylidene Chloride), CHjCCl 2
Trans- 1 ,2-Dichloroethene,
C1CHCHC1
Cis-l,2-Dichloroethene, C1CHCHC1
Trichloroethene, CC12CHC1
Tetrachloroethene
(Perchloroethene), C12CCC12
Methanes
-
-
2013
2120
1631
Ethanes
-
-
683
2513
2334
3621
Ethenes
3663
2333
1774
1949
1555
2246
HalfLife1-5 HalfLife*
Electrochemical- Natural Abiotic
Reduction Degradation
(hr) (hr)
. •* '
degraded2 - , _, '.
no degradation 2 •*• -
(
1.49 i.ei. io7
V
0.041 t 13.6.10s
|(
0.020 6.0 . IO6
no degradation2 -
no degradation 2 -
0.065 9.6 • IO3
0.053 3.5 • 1(P
0.049 4.1 • 10s
0.013 1.6. IO3
1155
5.47 1.1. IO12
6.41 3.9« 10U
19.7 3.9 « 10U
0.67 1.1. 10'°
0.28 8.7 • 10'°
Trifluorotrichloroethane, C 2F3 Q 3
(Huorocarbon-113)(FC-113)
Others
degraded2
'Gillham and CfHannesin (1994)
^ogan et aJ. (1995)
3Jetfereeta). (1989)
'Vogel, Criddlo, and McCarty (1987)
5Assuming that the barrier has an iron surface area of 1 m2 /mL.
Source: Gillham and OHannesin 1994; Vogan et al. 1995
3.5
-------
In Situ Permeable, Electrochemlcally Active Metal Barriers
3.3 Treatment Trains
An in situ permeable, electrochemically-active metal barrier treatment
system needs to be designed such that contaminants and contaminated
groundwater do not escape the area without passage through the permeable
barriers. The installation of impermeable barriers may be necessary to run-
nel the groundwater to the face of the permeable barriers. The impermeable
barriers can be sheet piling, grout curtains, or other standard designs appro-
priate for the site. Once the treatment system is installed, operation of the •
system is self-regulating; however, monitoring is required.
f
3.4 Remediation Goals f
The general application for a permeable wall treatment system is similar
to that for a pump-and-treat system including aboveground water treatment
— to reduce the concentration of the target contaminant hi the groundwater
flowing through the walls to acceptable levels. Table 3.1 provides the initial
concentrations and half lives of a variety of contaminants found in ground-
water when passed through an iron bed.
3.5 Design
3.5.1 Design Basis
The design of a permeable treatment wall is mainly based on the ground-
water velocity and seasonal direction, aquifer hydraulic conductivity, distri-
bution of conductivity, and the concentration and distribution of contami-
nants of concern. The distribution of contaminant and the-hydraulic conduc-
tivity determines whether a continuous permeable wall or funnel-and-gate
arrangements are most cost-effective. The wall, optimally, is placed perpen-
dicular to the plume center line, transverse to groundwater flow. Seasonal
variation of the flow direction with respect to the wall is taken in account
into designing the thickness of the wall. The wall either should extend
3.6
-------
Chapter
vertically to an aquitard or flow modeling performed to design the depth to
which the wall must be extended in the aquifer to capture the entire plume.
Currently, walls are limited to a trenching depth of about 50 ft, perhaps 75 ft
if scalable sheet piles can be driven to that depth (requires absence of boul-
ders and limited friction with the soil involved). Groundwater models and
wall porosity selections are used to design the width of a wall beyond plume
limits to ensure complete capture. Wall permeability is a critical design
parameter necessary to direct the complete contaminant plume into the reac-
tive media. Generally, walls are constructed to be more permeable than tl)e /•
aquifer into which the wall is placed. Precipitation and clogging must be/ '
considered in designing the wall to be adequately permeable jpr the treat-
ment period intended. Otherwise, the estimate of the present cost of thef
system should reflect the need to replace or regenerate the wall once clog-
ging interferes with wall performance. Currently, a 5- to IQ-yefc design life
is assumed. It is best if a wall can be extended to an aquitard to prevent
leakage of a contaminant under and around a wall. However, cost-effective,
hanging walls can be designed if careful attention is given to porosity in
designing and installing the wall. Wall thickness is based on the flow veloc-
ity through the media (including increased velocity if impermeable walls are
used to direct flow), kinetics of degradation, the type of iron (i.e., granulated
reagent grade, blast furnace dust, steel furnace dust) selected, and the surface
area of the iron, as well as porosity. Highly oxidized waters may require a
pretreatment through a vertical layer consisting of a mixture of granular cast
iron and large size sand or pea gravel. The pretreatment layer distributes the
flows and precipitates dissolved solids in a layer with a high void volume.
High dissolved solids are generally not known to be a problem, but ground-
water chemistry in the presence of zero-valent iron must be taken into ac-
count when designing wall porosity and thickness. High concentrations of
nitrate; sulfate, carbonate, and other oxidized species are important because
these species may interfere with the reduction of chlorinated compounds
(i.e., TCE, PCE, vinyl chloride). Multiple species of chlorinated solvents
can be degraded simultaneously. However, the wall thickness must be cho-
sen considering the reaction kinetics, influent concentration, and desired
effluent concentration of each species. In addition, it should be noted that as
the parent species (e.g., TCE) degrades within the wall, it is likely that
dehalogenation daughter products (e.g., DCE) will appear in some propor-
tion to the parent concentration. The daughter product may react more
slowly than the parent and become the limiting factor in overall treatment,
3.7
-------
In Situ Permeable, Electrochemically Active Metal Barriers
particularly if the daughter product already is present in the influent Since
these reaction kinetics are to a degree interactive and quite complicated, many
designers select wall thickness on the basis of laboratory treatability testing and/
or field pilot results. Alternative mixtures of granular iron and pyrite or other
compounds may be used to control pH and thus control precipitation of carbon-
ates and reaction byproducts and the life of the reactive wall.
3.5.2 Design and Equipment Selection
The most important consideration in the design and equipment selection for '
permeable barrier installations is the materials of construction for she barrier
walls and grout curtains. Permeable treatment walls are usually made pf grarui-
lar iron. The barrier walls in existing installations have been made from inter-
locking sheet piling. These walls, formed from commercially-availaple, corru-
gated metal are installed using standard construction techniques. Mother types
of barrier walls, such as slurry wall, jet grouted barrier, or any other types of
impermeable barrier, can be used as appropriate.
An important consideration for permeable barrier treatment is the nature
of the bottom barrier to groundwater flow. The existing field installations
have taken advantage of impermeable clay lenses and other layers underly-
ing the contaminated areas. If such impermeable layers do not exist, or if
they do not form a continuous impermeable barrier to groundwater flow
under the site, then use of grouting techniques, possibly combined with
horizontal drilling methods, may be necessary to seal the bottom of the treat-
ment region. The effectiveness of horizontal drilling methods for this appli-
cation has not yet been demonstrated.
Another potential concern is the nature of the iron used for the reactive
barrier wall. To date, all work has used high-grade iron containing known
levels of impurities. Other sources of granular iron are available, such as
blast furnace dust (a high-grade iron). The Air Force Armstrong Laboratory
is in the process of developing a protocol for selection of reactive media that
is useful in defining specifications for suitable granular iron requirements
(McCutcheon 1996). Iron from these other sources can be significantly
cheaper than the high-quality material currently used to date. However, such
iron can be contaminated with high levels of toxic impurities such as lead,
mercury, zinc, cadmium, selenium, and arsenic. These metals cannot be
introduced into aquifers without extensive investigation of their mobility and
toxicity. As operating experience with permeable treatment walls increases,
3.8
-------
Chapter 3
there may be an incentive to reduce costs by using these less-expensive
sources of iron, but care must be taken to ensure that iron from such sources
does not inadvertently contaminate the site.
3.5.3 Process Modification
The physical layout and construction of permeable electrochemically-active
metal barriers is largely determined by site conditions. The in situ treatment
system must be constructed to runnel all contaminated water flow into the per-.., r
meable treatment barrier. One area of flexibility in the design is the composi- /
tion of the barrier. Most field applications of this technology to dajte have used
granular cast iron. It is possible, however, to mix the iron metal with either an
inert solid or with magnetite (Fe2O3). No specific data on the potential effect of
this alteration were found. It is possible to mix the iron metal with ^nd, but
this is mainly used for porosity control. Typically, the greater theftiilution of the
iron surface area, the lower the mass of iron and the larger and deeper (in the
direction of groundwater flow) the bed needs to be. Given an influent concen-
tration and iron with a certain reactivity, the important parameter in design of a
granular iron treatment barrier is the mass of iron placed in the path of the influ-
ent groundwater. Sufficient iron must be placed so as to achieve treatment to
the desired effluent concentration. How this iron is configured is somewhat
irrelevant in terms of the reactions. It could be placed as pure iron in a single
trench, mixed with sand and placed in a larger trench, or placed into the aquifer
through other means such as jetting or deep soil mixing. Whichever method is
most cost effective will depend on how much iron is needed (in terms of equiva-
lent wall thickness) and how deep it must be placed. However, the statement
that mixing the iron with sand may aid in porosity control (i.e., versus a pure
iron-filled trench) is correct as is the statement that ferrous iron may assist in the
degradation process if used as the mixing agent with zero-valent iron.
Use of an equilibrium model, MINTEQU-A (Felmy, Girvin, and Jenne
1984), indicates that dissolution of pure iron metal in some ground waters
could cause a sharp increase in pH which would then result in the precipita-
tion of naturally-occurring dissolved solids. The precipitates might ad-
versely affect the performance of a barrier treatment system by:
• the formation and accumulation of precipitates within the reac-
tive zone which could reduce the porosity of the system and
change its hydraulic properties;
3.9
-------
In Situ Permeable, Electrochemlcally Active Metal Barriers
• the precipitate could form impermeable blocks in' the groundwa-
ter flow pathway, thereby diverting the groundwater away from
the permeable wall; and
• the precipitates could deposit on the active surface and
de-activate it.
Ongoing laboratory studies (Holser, McCutcheon, and Wolfe undated-a)
indicate a mixture of 25% (by weight) granular pyrite (iron disulfide) and
granular iron may control pH. The results of the tests using three different •
iron-to-pyrite ratios imply that a 9:1 mixture of iron to pyrite may improve'
the destruction of chlorinated organic compounds primarily through control
of pH (see Figure 3.2). {
¥
Figure 3.2
Conversion of TCE in Columns Packed with Mixtures of Iron and Pyrite
1
g0.8
c
_o
1 0.6
u
H 0.4
02
dr
X
X * 1
X .
T
* *
X
- 1 I
1 1 1 1 1
0 1 2 3 4 56
0(min)
+ 50% Fa/50% Pyrits
x 75%F«y25%Pyrito
° 100%F«
Mixtures of iron and pyrile were prepared and the pH of the effluent was monitored over a period of 500 residence times.
The total amount of packing for each of these columns was 4.0 grams. TCE conversions were similar for the different columns
despite the variation in the amounts of iron.
Source: Holser et aL undated-a
3.10
-------
Chapter 3
Other process modifications that may be required relate to the possibility
of plugging or deactivation of the treatment medium. It is noted that no
significant precipitates were observed in the in situ reactive wall at the Uni-
versity of Waterloo Borden test site two years after installation (Tratnyek
1996; Matheson and Tratnyek 1994). The wall has performed consistently
for about 3.5 years with no noted problems. Data from in situ systems in-
stalled in California in December 1994 and September 1995 and from other
in situ applications will generate further data to provide additional under-
standing of plugging and deactivation. . - >•
i ! '
3.5.4 Pretreatment Processes
/
No pretreatment is required for application of this process. .. !
. >
3.5.5 Posttreatment Processes fr
If porosity is carefully designed to account for precipitation and clogging
or pH is controlled to avoid precipitation, no posttreatment is expected to be
required. However, there may be a need for periodic treatment of the reac-
tive material in the permeable barrier wall to remove precipitates. No unac-
ceptable quantities of precipitates were found at the Waterloo-Borden test
site during the (approximately) first five years of operation. Subsequently,
some severely-clogged column treatability tests indicate that clogging is a
possible problem with some groundwaters. According to the manufacturer
of that system, laboratory measurements of alkalinity losses indicate the
possibility of a 2 to 15% loss in porosity per year. However, the vendor has
suggested that the amount of precipitation that will occur in situ will be sig-
nificantly less than predicted from laboratory studies due to the groundwater
used for laboratory studies. He indicates that sampling and transport satu-
rates the samples with oxygen and shifts the carbonate equilibrium in the
groundwater used for laboratory studies. These changes, which are caused
by the sampling and transport result in plugging of the iron column used for
the laboratory tests.
3.5.6 Process Instrumentation and Control
The process requires no unique instrumentation or controls. Other than
monitoring wells, the only addition to a site would be provisions for sam-
pling at different levels up and downstream of the treatment system and
3.11
-------
In Situ Permeable, Electrochemically Active Metal Barriers
continuous groundwater level monitoring. Water level monitoring is neces-
sary to indicate a reduction in the barrier's permeability or occurrence of
blocking that may be due to precipitation.
Monitoring is expected to be more expensive as compared to other
remediation methods for two reasons. First, sampling at many locations and
at multilevels is needed to monitor the performance of the treatment system
because of the in situ nature of the treatment. Second, environmental regula-
tors may require additional information to ensure the treatment works as
expected because it is an emerging technology. With experience gained at,
various ongoing demonstrations, monitoring requirements should soon de-
crease, however. f
«
3.5.7 Safety Requirements t [
The iron metal used in permeable barriers, especially if it is a finely-di-
vided form, is a potential fire hazard. The metal will oxidize exothermally
upon exposure to air. If the metal is placed in contact with combustible ma-
terial, the mixture might ignite. However, once in-place in a wet environ-
ment, the risk of fire is minimal. The material safety data sheet (MSDS) for
iron supplied by the vendor should be consulted for proper handling proce-
dures. Other sources of iron mentioned previously would avoid this hazard.
In addition, the formation and accumulation of hydrogen gas from hydrogen
ions is possible. Therefore, all monitoring wells will have to be designed to
vent all gases to prevent hydrogen gas buildup.
3.5.8 Specification Development
The specifications for installation of a permeable barrier treatment system
focus on on-site geology. It is imperative that the location of permeable and
impermeable zones in the area be well known so that appropriate imperme-
able barriers can be installed to funnel the contaminated groundwater flow
into the treatment barrier. Since the cost of construction and placement of
the barrier is the greatest expense in the implementation of this technology,
relatively small variations in the site's physical characteristics can have a
major impact on the system cost.
3.12
-------
Chapter 3
3.5.9 Cost Data
The overall cost of a permeable electrochemically-active metal barrier
system include:
• site survey and detailed geochemical assessment costs;
• installation costs for impermeable barriers and for the permeable
barrier treatment system; and
• monitoring costs.
Since the system is passive, once installed, there are no unique systeni '
operating costs. Monitoring is common to any in situ remediation system,
but as indicated earlier, because of the developing nature of this technology,
additional monitoring may be required.
-------
In Situ Permeable, Electrochemically Active Metal Barriers
precipitates from the reactive media. No information on the costs of such
regeneration were found. Should flushing prove unsuccessful in clearing the
blockage, reactive media removal and replacement will be necessary. The
cost of such replacement depends on the quantity and cost of the iron.
Because of the high front-end and low operating costs of the permeable
reaction wall treatment process compared with those for competing processes,
the cost analysis for any application must be performed on a present-worth (or
annual cost) basis. Costs supplied by EnviroMetal Technologies, Inc. indicate •J
that the costs can be lower for this process than for competing processes when
treating groundwaters contaminated with volatile organic compounds (although
no underlying assumptions were given). Also, the exact cost of the permeable
barrier treatment portion of the overall treatment systems quoted abdve appears
to be relatively small. The costs quoted are: , I
.*
Industrial facility, California (3 yr net present value)
• EnviroMetal process $2.9M
• Pump-and-treat $7.8M
• Dewater, soil vapor extraction $4.1M
Somersworth Sanitary Landfill, New Hampshire
• EnviroMetal Process
Installation $12.74M
Operation $2.22M
Total S15.0M
Of this total, EnviroMetal process capital costs amount to about $1.5M
• Pump-and-treat alternative with impermeable cap, etc.
Installation $16.5 to 18.4M
Operation $2.8 to 3.2M
Total $19.7to21.2M
Landfill, New York
The present-worth values reported below supposedly include other as-
pects of the site remediation, such as monitoring, soil treatment, etc., how-
ever, details were not provided.
3.14
-------
Chapter 3
Total Present Worth Q&M Present Worth
• EnviroMetal process $2.7M $0.6M*
• Air sparging $2.5M $0.7M*
• Air sparging/in situ bio $2.9M $0.9M
• Extraction (pump) $2.9M $0.8M
*Of these installation costs, the EnviroMetal process involved about
$500,000 in capital costs.
# .' *
Industrial Facility, midwest U.S.
• EnviroMetal process $0.7M , •
• Soil treatment and pump-and-treat $0.8M *J
• Soil excavation and pump-and-treat $1.5M j»
Note: The EnviroMetal process is a proprietary permeable reactive bar-
rier process marketed by EnviroMetal Technologies Inc.
3.5.10 Design Validation
The principal design validation required is regular monitoring of the
groundwater before and after the treatment barrier and continuous monitor-
ing of the groundwater level at points upstream of the groundwater barrier.
3.5.11 Permitting Requirements
There appears to be no unique permitting requirements for a reactive
barrier treatment system beyond approval of the remediation plan re-
quired for all types of site cleanups. The technology is passive, it has no
NPDES discharges, produces no hazardous waste that must be disposed
or further treated, and it has no point sources of air emissions that might
require permitting.
3.5.12 Performance Measures
The process has no unique measure of performance beyond the results of
continued monitoring through sampling and analysis.
3.15
-------
In Situ Permeable. Electrochemlcally Active Metal Barriers
3.5.13 Design Checklist
The process is too new to enable providing a meaningful design checklist.
3.6 Implementation and Operation
,> >'
3.6.1 Implementation Strategies
Jk
The implementation of such a system must begin with a good understand-
ing of the site and the contaminants. This information should be coupled
with treatability studies using groundwater samples collected frofrn the site.
Conducting treatability studies on actual site samples is especially crucial for
this technology since small variations in chemistry can have a major impact
on process performance.
The vendor used for these treatability studies can be the supplier of the
equipment or an independent party; however, a strong background in electro-
chemical processes is essential. The vendor should be capable of interpret-
ing the results of the treatability studies and creating a set of performance
specifications for the equipment. The equipment itself can be acquired
through normal procurement channels. While competitive bidding is desir-
able, this application has not been applied extensively in the field, and ulti-
mately, vendor selection should be based not only on a cost comparison, but
also on an assessment of the vendor's experience with similar applications.
Although the basic research on permeable barriers appears to have been
conducted at U.S. and Canadian facilities, certain components of the tech-
nology may be proprietary to EnviroMetal Technologies, Inc. Legal guid-
ance should be sought to avoid possible patent or other infringements.
3.6.2 Start-up Procedures
The barrier treatment process is a passive system. Once installed, it re-
quires no "start-up", and operates without the need for adjustment. Compli-
ance monitoring should be delayed until one or two pore volumes (volume in
the soil available for liquid or gas) has passed through the barrier.
3.16
-------
Chapter 3
3.6.3 Operations Practices
After installation, the operation of a permeable barrier treatment system
involves monitoring and correction of blockages if they occur.
3.6.4 Operations Monitoring
Regular monitoring of the groundwater before and after the treatment
barrier and continuous monitoring of the groundwater level at points up- ,,.
stream of the groundwater barrier are required. , / •'
**
3.6.5 Quality Assurance/Quality Control (QA/QC) <
Standard QA/QC procedures related to permeable barrier*^reatment
need to be followed for all analyses. Successful system installation
requires that the permeable wall be homogeneous so that the groundwa-
ter flow takes advantage of all of the system's reactive volume. In addi-
tion, care must be taken to ensure that impermeable barriers to flow are
not inadvertently breached.
3.7 Cose Histories
Permeable barrier treatment has been the subject of several studies at
various scales. This section describes two programs — one system at the
pilot-scale and a second, which is ongoing, at full-scale.
Gillham (1995) described a study at the Canadian Forces Base Borden field
site where a 1.4 m (4.5 ft) thick permeable wall was used that contained 22%
(by weight) of granular iron and 78% (by weight) of sand. The residence time
of groundwater in the wall was 16 days at a flow rate of 9 cm/day (3.5 in./day).
The contaminants were trichloroethene and perchloroethene at concentration of
270 mg/L and 43 mg/L, respectively. As the groundwater passed through the
wall, the concentrations decreased by 90 and 88%, respectively. During con-
taminant degradation, the production of trans-1,2-dichloroethene and
1,1-dichloroethene was observed. These byproducts represented less than 10%
of the trichloroethene and perchloroethene removed and were found to be de-
graded. All the degradation of the chlorinated hydrocarbons was attributed to
3.17
-------
In Situ Permeable, Electrochemically Active Metal Barriers
electrochemical reactions; no appreciable biological degradation of these
compounds was found.
Yamane et al. (1995) and Wilson (1995) described the installation of a
permeable barrier at a former semiconductor manufacturing facility
(Intersil) in Sunnyvale, California, to replace an above-the-ground treatment
facility. The site was contaminated with 1,2-dichloroethene,
trichloroethene, CFC 113, and vinyl chloride. The barrier contained 100%
iron and was 4 m (13 ft) deep and 1.2 m (4 ft) wide with two slurry walls at -
the ends of the barrier to guide the contaminated groundwater. The resi- ',
dence time of the groundwater in the barrier was greater than 2*days. The
cleanup standards to be met were equivalent to the maximum contaminant
levels in groundwater set by the California Department of Health; Services
or the US EPA. The residence time of 2 days was selected basecf on the
degradation of vinyl chloride, which degrades slowly and has k long half-
life as shown in Table 3.1.
3.18
-------
SUPERCRITICAL WATER OXIDATION
4.1 Scientific Principles , '• '
• ' ^
Supercritical water oxidation (SCWO), also referred to as, h^drothermal
oxidation (HTO), is a technology for the destruction of hazardous and non-
hazardous wastes. Relative to ponventional incineration technologies, it
offers the following advantages:
• equivalent levels of organic destruction;
• potential for complete containment of all effluents until accept-
able treatment has been verified;
• lower temperature oxidation;
• negligible likelihood of the formation of dioxins or furans;
• ability to physically remove normally-soluble salts from an aque-
ous solution;
• enhanced process stability;
• negligible NOX and SOX production;
• negligible airborne particulates;
• compact equipment; and
• minimal pollution abatement equipment required.
SCWO destroys organic materials using an oxidant (usually air, oxygen,
or hydrogen peroxide) in water above its critical point of 374°C (705°F) and
pressures of 22.1 MPa (3,205 psi, 218 atm). Oxidation occurs under
near-homogenous, single-phase conditions, which provide excellent mixing,
4.1
-------
Supercritical Water Oxidation
high mass, and heat transfer rates. The organic destruction occurs in a rela-
tively small volume reactor. Typical products from a SCWO process include
carbon dioxide, water, nitrogen, metal oxides, and inorganic salts.
A common criticism of supercritical water oxidation is that the high pres-
sures make the process dangerous. As explained in detail hi Section 4.5.2,
this is not true. Because of the rapid destruction times (on the order of 5 to
60 seconds) and high density of the material being treated, the actual volume
of the reactor and ancillary piping and equipment under high pressure is very ,;,
small. For example, a 76 L/min (20 gal/min) system typically requires a ,,
reactor volume on the order of 6 to 76 L (15 to 20 gal). This small volume
can be safely maintained at the required pressures if materials of construe-/
tion are appropriately selected. , *
Conceptually, SCWO is similar to wet oxidation [also called w$t air oxi-
dation (WAO)], a technique that has been used to treat sewage sludge for
nearly one hundred years (Gloyna and Li 1993). WAO is discussed in an-
other monograph of this series, Thermal Destruction. The difference be-
tween the two is that a SCWO system operates in the supercritical region of
water, where water, organic materials, and oxygen are miscible forming a
single phase. In water's subcritical region (where WAO systems operate) the
water exists in both a liquid and gas phase.
The critical point of an element or compound is defined as the tempera-
ture and pressure at which the liquid and gaseous phases merge.
Phase-change properties, such as heat of vaporization cease to have a mean-
ing in the supercritical region. Therefore, for the purposes of this discussion,
the material in the supercritical region is referred to as a "fluid" to differenti-
ate it from liquids and gases.
In a SCWO system, the single-phase fluid greatly increases mass transfer
rates between the oxygen and the organic compounds enabling high organic
compound conversion levels with short residence times. A SCWO system
can accomplish the desired treatment in times ranging from seconds to one
or two minutes as compared to the many minutes or even hours required by a
subcritical WAO system. A second important consideration when operating
in the supercritical region is that normally-soluble inorganic salts are highly
insoluble in the supercritical water. Removal of this salt is a design consid-
eration. A variety of techniques offer potential solutions for removal of
these salts from supercritical water.
4.2
-------
Chapter 4
As previously mentioned, the key parameter that differentiates SCWO
from WAO is that a SCWO system operates above the critical temperature
and pressure of water, defined above. Operation in the supercritical region
tends to create a relatively homogeneous, nearly single-phase reaction me-
dium between the oxygen and the organic materials to be oxidized. There-
fore, the general chemical reaction for the destruction of the organic con-
stituents can be reasonably well-defined as follows:
CcHhNnPpCl,Ss + 02 -> C02 + H20 + HC1 + H3PO4 + N2 + H2SO4 (4.1,)
Intermediate products such as organic acids may form, but they are
also oxidized at SCWO reaction conditions. As indicated 1>y Equation
4.1, the SCWO environment can be highly corrosive. The cqrribination
of oxygen and inorganic acids requires the various corrosion/resistant
materials of construction, such as nickel alloys, titanium, ^nti platinum
liners. Some designs minimize the need for such corrosion resistant
materials by injecting a non-corrosive liquid stream that forms a protec-
tive layer over the system's surfaces.
Several generalized kinetic models, based on simplified reaction schemes
involving the formation and destruction of rate-controlling intermediates, are
available (Li et al. 1991). It is possible for the formed intermediate products
to be destroyed within seconds, depending on the reaction rates. These rates
are controlled by the operating temperature. If intermediate products are
detected in the effluent, the temperature can be raised or the flow rate de-
creased to increase the residence time. Table 4.1 (Gloyna and Li 1993) lists
the rate-controlling intermediates and the oxidation endproducts from sev-
eral laboratory studies of various compounds at WAO and SCWO condi-
tions. The kinetic parameters for the assumed first-order reaction are also
given. As the rate parameters for the hydrocarbons and oxygenated com-
pounds indicate, the reaction rates are substantially higher for the SCWO
than for the WAQ systems.
Reaction mechanisms and byproduct analyses indicate that short-chain
carboxylic acids, ketones, aldehydes, and alcohols are the major oxidation
intermediates under WAO conditions (Bailed, Faith, and Masi 1982). Ki-
netic studies of refractory compounds under SCWO conditions, such as
acetic acid (Lee 1990), methanol (Rofer and Streit 1989), ammonia (Webley,
Tester, and Holgate 1991), and carbon monoxide (Helling and Tester 1987)
are well documented. For nitrogen-containing organic compounds, nitrogen
4.3
-------
Table 4.1
Kinetic Parameters for Key Rate-Controlling Intermediates
Organic
Compound
Category
Hydrocarbons
and Oxygenated
Hydrocarbons
Nitrogen-
Containing
Organics
Chlorinated
Organics
Kinetic Parameters *
Key Intermediate
(Alternative)
CHjCOOH
NH3
(N20)
CHjQ
(CH3OH)
Oxidation End
Product
COj.HjO
N2,H20
Hd,H20
Condition
(Water)
Subcritical
Supercritical
Supercritical "
Supercritical
Supercritical
k
(I/sec)
4.40 • 10 >2
2.55. 10"
2.63« 10 10
3.16 «10«
231-10"
(IcJ/mol)
182
172.7-
167.1
157
395
Reference
Foiissard, Debcllefontain, and Besombes-Vailhe 1989
Wightman 1981
Lee 1990
Webley, Tester, and Holgate 1991
Refer and Streit 1989
'Pseudo first-order reaction model using oxygen.
"Using hydrogen peroxide.
'"Obtained from fitting the reported seven data points.
Reprinted from Proceedings of the Second International Symposium on Environmental Applications of Advanced Oxidation Technologies , Gloyna and Li, "Supercritical Water Oxidation:
An Engineering Update," 1993 with permission of EPRI.
-------
Chapter 4
gas is generally the predominant SCWO end product regardless of the oxida-
tion state of nitrogen in the contaminated material treated (Killilea, Swallow,
and Hong 1992). Ammonia and nitrous oxide are formed under a variety of
operating conditions (Shanableh 1990; Killilea, Swallow, and Hong 1992).
The heat to bring a waste stream to the temperature required for the
SCWO process is derived from three sources: (1) heat released by the
oxidation of the organic contaminants in the stream being treated, (2)
heat released from materials added to the stream being treated, and (3).,,.
external heat supplied to the reactor or the feed streams. If the organic
concentration in the feed stream is high enough (usually on the order of
10%) then the reaction is autogenous and external heat may be unneces-
sary. If not, then external heat will be necessary. It is necessary, there-
fore, to calculate the heating value of the organic constituents of the
waste stream in order to specify the SCWO system. jl '
The heat released from the oxidation of the organic compounds during the
SCWO process is approximately equal to the Higher Heating Value of each
of the constituents in the waste stream. The Higher Heating Valuefor many
substances can be found in numerous combustion handbooks, such as Gill
and Quiel (1993). As an alternative, heat release can be calculated based on
the heat of formation of the constituents. This latter method allows for more
precise calculation since it considers the heat consumed in the formation of
the products of the oxidation reaction such as hydrochloric, sulfuric, and
phosphoric acids.
The heat of reaction is calculated by (1) summing the heats of formation
of the products of the chemical reaction, and (2) subtracting the sum of the
heats of formation for the reacting species. The heat of formation is defined
as the enthalpy change occurring during a chemical reaction where 1 mol of
a product is formed from its elements. The heat of reaction, AHr, is calcu-
lated from the heat of formation using the following formula:
' (4.2)
where: Hf = the heat of formation of the products and reactants
(subscripts p and r, respectively) and the subscript 298 °K
(25"C, 77°F) refers to the reference temperature for the
reactants.
4.5
-------
Supercritical Water Oxidation
Gill and Quiel (1993) include an extensive list of heats of formation for
most organic and inorganic compounds commonly encountered in environ-
mental situations. The heat of formation of most organic and inorganic com-
pounds can also be found in chemistry and chemical engineering handbooks
as well as in standard thermodynamic references. The effect of pressure on
the heat of reaction hi the liquid phase is relatively small and can be ignored
in the preliminary design of a SCWO system.
The following example indicates how the heat of reaction of
1,3-dichloropropane (liquid) is calculated from the heat of formation. The, '
heat of formation for 1,3-dichloropropane is 615 Btu/lb, and its molecular
weight (MW) is 113. The oxidation equation is as follows: /
C3H6C12(L) + 402-»3C02+-2H20(L) + 2HC1 '/ (4.3)
The heat of formation for each of these compounds is as follows:
C3H6C12(L) 02(G) C02(G) H20(L) HC1
Btu/lb -615 0 -3,848 -6,832 -1,088
MW 113 32 44 18 36.5
Btu/lb-mol -69,480 0 -169,294 -122,971 -39,713
Therefore, the heat of combustion of 1,3-dichloropropane liquid is:
3(-169,294) + 2(-122,971) + 2(-39,713) - (-69,500)
= -763,769 Btu/lb - mol (4.4)
= -6,759 Btu/lb
In other words, 6,759 Btu are released when 1 Ib of liquid
1,3-dichloropropane is oxidized, and the water resulting from the chemical
reaction is condensed after forming.
4.2 Potential Applications
SCWO is applicable to the treatment of aqueous wastes and streams con-
taining organic compounds including highly energetic and toxic materials
4.6
-------
Chapter 4
such as propellants, explosives, or chemical warfare agents and their prod-
ucts of neutralization. It can also remove soluble inorganic compounds from
wastewaters, and at the same time, destroy the organic constituents. While
the SCWO system can be applied to dilute aqueous streams (i.e., those with
a relatively low heating value), the cost of heating the stream to the required
supercritical water temperature is significant.
SCWO is most appropriate for treating concentrated aqueous solutions of
organic compounds where the heat of oxidation of the organic compounds can
provide a significant fraction of the heat needed to operate the system. Assum-
ing no energy recovery, the system becomes autogenous (i.e., all of the required
heat can be provided by the chemical reaction, and the system bfecomes
self-sustaining) at approximately 10-15% organics, depending on thp heating
value of the total organic content (Killilea 1996). Inclusion of a heat recovery
system can lower this value significantly. The process has been applied to slud-
ges such as those from wastewater treatment, but engineering problems can
occur when pumping abrasive materials such as soil/water mixtures at the high
pressures required by the system. SCWO is at present being considered as a
serious alternative to incineration for the treatment of hydrolyzed propellant
waste, hydrolyzed chemical agent (i.e., nerve gas, mustard gas), paper mill
wastes, and other hazardous wastes. Section 4.7 describes a number of case
histories of sites where SCWO has been identified as a potential solution to
difficult contaminant destruction problems.
4.3 Treatment Trains
Figure 4.1 is a simplified schematic of the treatment train incorporating a
basic SCWO system. Water and fuel can be added to the waste to be treated
to decrease or increase its heating value, respectively. Water is added only if
the organic content of the waste is higher than needed for an autogenous
process. If the waste has too little organic material to maintain the desired
operating temperature, fuel oil (kerosene or #2 heating oil) or other organic
constituents (e.g., methanol) can be added to the waste stream to increase the
overall amount of heat released and reduce the need for external heating.
Alternatively, additional external heating is often preferred. If the waste
contains halides (e.g., organic chloride) sulfur, or phosphorus, caustic
4.7
-------
Supercritical Water Oxidation
additives can be added to neutralize the acids (HC1, HF, H2SO4, H3PO4) that
form in the chemical reaction. Alternatively, acidic products might not need
to be neutralized if the system's materials of construction can withstand the
corrosive effects of the acids.
Figure 4.1
Simplified Process-Flow Diagram of SCWO
Gas Effluent
High Pressure Pump
PreheaterJ— Reactor j— Cooling
Liquid Effluent-*
Gas/Liquid
Separator
Liquid
Neutralization
and Treatment
Solids Effluent
Reproduced courtesy of Genera) Atomics
Once premixed, the waste stream is pressurized, preheated, and mixed
with an oxidant such as compressed air, oxygen, or hydrogen peroxide. The
mixture is then fed into the reactor.
The fuel added to the reactor reduces the amount of heat needed from
outside sources; however, it increases the amount of oxygen that must be
injected into the system. The decision of whether to inject additional fuel
and oxygen into the reactor or to introduce the heat externally depends on
the nature of the waste and on the relative cost of the two options.
4.8
-------
Chapter 4
The treated supercritical effluent from the reactor must be cooled and
reduced in pressure prior to discharge or further treatment. Several methods
of cooling may be used. The simplest cooling method is to "quench" cool,
whereby the effluent is passed through a pressure reduction valve and mixed
with cold water. This method of cooling is rapid and it minimizes or even
eliminates the deposition of precipitates on equipment. A second means of
cooling is by the use of heat exchangers. The working fluid which absorbs
the heat from the supercritical effluent might be an external fluid, such as
non-contact cooling water, or it might be the incoming fluid to the SCWQ -y
reactor. In the former case, a heat exchanger could be used to make proce'ss
steam. In the latter case, the waste heat from the effluent is used to preheat
the influent with a commensurate reduction in the system's energy require-
ment. The design tradeoffs between these approaches need to be examined
from the perspective of economics and of the potential of a specific waste to
foul the heat transfer tubes. The need for treatability studies.to make this
determination is readily apparent.
Pressure let-down can similarly be performed by various means. The
most simple is by passing the effluent through an expansion valve. This has
the inherent advantage of low cost, but the behavior of solids under the high
turbulence of the expansion valve must be evaluated. Erosion of the valve is
virtually unavoidable, although proper design and choice of materials can
keep the level of erosion within acceptable limits. It is possible to use tur-
bines for pressure let-down and, thereby, to recover some of the mechanical
energy used for pumping; however, discussions with various practitioners
have led to the conclusion that this is presently not a common practice,
largely because the cost of pumping is a relatively minor component of the
overall cost of treatment (this is discussed hi greater detail later in this chap-
ter) and because the corrosive environment and the potential for plugging
and erosion reduce the turbine's life to the point where the cost of energy
saved is less than the cost of the equipment and its maintenance.
After pressure let-down, the effluent usually enters a knockout drum
where the gases are separated from the liquid. The liquids may be subject to
further treatment prior to discharge. The gases, which consist mostly of
excess oxygen, nitrogen, carbon dioxide, and water vapor, are passed
through condensers and demisters before discharge to the atmosphere. Fur-
ther gas treatment is usually unnecessary.
4.9
-------
Supercritical Water Oxidation
Tabio 4.2
SCWO Destruction Efficiency for Selected
Organic Compounds
Concentration
Compound
2-BuUnone
Methyl Ethyl Ketone
(MEK)
p-Chlorophenol
0-Cresol
2,4-Dichloropbenol
DicthyleneGlycol
Diethyl Ether
2,4-Dinitrotoluene
Ethylene Glycol
Pentachlorophenol
Temp*
CC)
400
400
450
500
400
450
450*
400
400
450
450
500
400
450
500
450*
450*
410
528
450*
450*
400
450
500
Time*
(min)
5
10
5
5
10
5
3
12
10
12
10
12
10
5
5
4
2
3
3
1
2
2
2
2
h
(mg/L)
6210
6210
5,140
6210
5,140 -
5,140
1,000
10,040
10,040
10,040
10,040
10,040
300
300
500
1,000
1,000
84.0
180.0
200.0
1.000
500
500
500
out
(mg/L)
251
197
136
J
71
273
136 £
00.1
4453
71.9
1302
. 20.9
511
12
09
16
<0.01
<1D
14.0
<0.04
<0.04
<0.04
Destruction
Efficiency*
(%)
95.96
. •* ''
96.83
97.35
•t
98.86 /
y 94.68
i
97.35
>99.99
55.7
99.3
87.0
99.8
94.9
• 99.6
99.7
99.7
>99.999
>99.9
83.0
>99.0
>99.5
>99.9
>99.99
>99.99
>99.99
4,10
-------
Chapter 4
Table 4.2 cont.
SCWO Destruction Efficiency for Selected Organic Compounds
Concentration
Compound
Pyridine
Trichloroethylene
2,4,6-Trichlorophenol
Temp'1'
CC)
400
450
500
500
450
- 450
500*
Time*
(min)
5
10
5
20
1
5
2
n
(mg/L)
500
500
1,000
500
1,827
1,827
200
out
(mg/L)
3516
4.1
24.3
1.8 •*
32
,3 |
<0.01 £
Destruction
Efficiency*
29.5
99.2 ' •'.''
97.6 '
99.6 f
1 98.2 '
99.3
>99.995
All SCWO data were obtained from the batch tests with excess oxidant loading.
Pressure for all SCWO tests was about 27.6 MPa (4000 psi).
•Hydrogen peroxide was used; and oxygen was used for all other tests,
1These results are for systems operating at the low end of SCWO temperatures. Reactors operating in the 600-650'C
range and less than one minute residence time routinely achieve 99.999% destruction efficiencies for all of these, and
other compounds. Please see Section 4.7 for results of treatment at these elevated temperatures.
Reproduced with permission of the American Institute of Chemical Engineers from Environmental Progress, Volume 14,
Number 3, Gloyna and Li, "Supercritical Water Oxidation Research and Development Updates " p 185 Copyright
©1995AICHE. All rights reserved.
4.4 Remediation Goals
The remediation goal of supercritical water oxidation is to destroy the
organic materials in an aqueous stream or sludge and, if necessary, concen-
trate the inorganic materials in a small-volume side stream. Very high levels
of destruction of organic contaminants can be achieved; total organic carbon
reductions of >99.999% have been reported for SCWO systems. Table 4.2
presents destruction data on selected compounds at various residence times
and temperatures. Table 4.3 reports SCWO system performance on a variety
of industrial organic wastes.
4.11
-------
Supercritical Water Oxidation
Table 4.3
SCWO Destruction Efficiency for Selected Organic Wastesi-2
Concentration
Compound
Industrial Wastewater3
Industrial Sludge4
Mixture of Industrial
Wastewater and Sludge4
Municipal Sludge4
Contaminated Soils4
Temp1
CO
400
450
500
400
450
450
400
450
500
400
450
400
450
500
Time*
(min)
1
I
1
30
10
S
4
4
4
8
4
8
4
2
in
(mg/L)
1,840
1,840
1,840
30,300
30,300
30,300
39,000
39,000
39,000
14,020
14,202
170
170
170
out
(mg/L)
27
15
4
120
50
400
4,520
831
429
687
84
13
9
9
Destruction
Efficiency5
98.5
99.2
^ 99.7
99.6 '.
-------
Chapter 4
4.5 D0$ign
4.5.1 Design Basis
The key to successful SCWO process design is the integration of various
unit operations. Important design considerations include (Gloyna and Li
1993):
• reactor residence times and associated temperatures;
• reactor and ancillary equipment configuration;
Jk
• system pressures and related temperatures; {
• materials of construction for each unit operation; y
• control and removal of solids either from the supercritical fluid or
the treated effluent; and
• operation and maintenance of the facility, including safety, ana-
lytical support, regulatory monitoring, and disposal requirements.
Generally, SCWO research has covered such areas as chemical reaction
mechanisms and kinetics, salt formation and solubility, mass and heat trans-
fer, transformation product identification, corrosion, catalysts, and additives.
Process development has focused on materials of construction, reactor de-
sign, heat exchange and recuperative heat recovery, solid-liquid separation,
gas-liquid separation, control systems, effluent handling, ash disposal, safety
requirements, and process system integration.
A SCWO reactor of tubular design behaves like an ideal plug flow
reactor as defined by Levenspiel (1962). Therefore, the kinetics of the
reaction are a necessary consideration in its design. Table 4.4 lists the
various global kinetic models for common SCWO reactions including
kinetic parameters and waste types. The experimental conditions for
each reaction are also included.
4.13
-------
Table 4.4
Compounds
Acetamide
Acetamide**
Acetic Acid
Acetic Acid
Acetic Acid
Acetic Acid
Activated Sludge
(COD)
Ammonia
2-Butanone
Carbon Monoxide
Carbon Monoxide"
o-Cresol
Digested Sludge
(COD)
Global Kinetic
Reactor
Oxidant Type
HA flow
HA fl°w
HA flow
HA ft°w
Oj flow
Oj flow
Oj batch
Oj flow
Oj batch
Oj, flow
O2 flow
Oj batch
O2 batch
Models
for Supercritical
Kinetic Parameters '
k'
2.75.10s
5.01 • 10*
2.63. 10 '»
9.23. 107
9.82.10"
2.55.10"
-1.5*102
3.16« 10«
1.20.10
3.16. 10«
3.16.10'
3.16.10°
4.36. 10 3
E,
88.3
94.7
167.1
131
231
172.7
-54
157
36.2
112
134
28.5
20.4
m
1.15
1
1
1
1
1
1
1
1
1
0.96
1
1.86
Water
Oxidation
Temperature
n
0.05
0.17
0
0
1
0
0
0
0
0
0.34
0
0
CK)
673-803
673-803
673-803
673-773
611-718
611-718
573-723
913-973
673-773
673-814
693-844
673-773
573-723
of Organic Substances
Pressure
(atm)
240-350
240-350
240-350
240-350
394-438
394-438
240-350
246
240^00
"Vet
246
240-400
240-350
(g^)°
1.5-4.0
1.5-4.0
1.3-3.3
1.0-5.0
0525
0525
46.5
0.03-0.11
-6
0.02-fc.ll
-fr.01-0.098
-10
» -v, %
46.5 '" \
References
Lee 1990
Lee 1990
Lee 1990
Wilmanns 1990
Wightman 1981
Wightman 1981
Sharmbleh 1990
WebleyetaL1991
Griffith and Gloyna 1992
Helling and Tester 1987
Holgateetal. 1992
Griffith and Gloyna 1992
Tongdhamachart 1991
-------
Ol
2,4-Dichlorophcnol
Ethanol
Formic Acid
Glucose (TOG)
Methane
Methane
Methane
Methanol
Methanol
Phenol
Phenol
Pyridine
°2
Oz
Oz
Oz
Oz
Oz
Oz
Oz
Oz
Oz
o,
Oz
flow
flow
flow
batch
flow
flow
flow
flow
flow
flow
flow
flow
1.94. 10<
6.46.10"
not reported
not reported
1.26- 107
2.51-10"
2.04. 107
2.51. 10 M
3. 16* 10 M
2.61 • 105
-
3.44* 10 "
71.9
340
-96
130
156.8
178.9
141.7
395.0
408.4
63.8
-
227
1
1
1
05
1
0.99
1
1
1.1
1
05
1
0.38
0
1
1
0
0.66
0
0
-0.02
1
0
02
683-788
755-814
683-691
653-683
913-973
833-903
833-903
723-823
723-823
557-702
653
698-800
276
241
408-432
-400
245
245
245
243
243
292-340
188-278
276
0.4-0.8
0.03-0.036
1.0
-10
-
-
-
-
0.038-0.17
0.1-0.4
0.25-1.0
1-3
Grain and Gloyna 1992
Helling 1986
Wightman 1981
WhMock 1978
Refer and Streit 1989
Webley, Tester, and Holgate 1991
Webley, Tester, and Holgate 1991
Refer and Streit 1989
Webley etal. 1990
Wightman 1981
Thornton and Savage 1990
Grain and Gloyna 1992
'Kinetic parameters are defined by -d[CJ/dt - klC^O]" and k - k'exp(-Ee /RT), where [C] and [O] are concentrations of organic reactants and oxidant, respectively; E. te inkj/mol; T is in K;
R - 8.314 J/mol-K; and k* - 1/sec (first-order), etc. - Not available. [C]0 - feed concentration. The concentration of compounds labeled with COD is quantified by chemical oxygen demand
method; the concentration of other compounds is quantified by chromatographic techniques. The excess oxidants are used in all tests. Kinetic parameters are reported for the overall
reaction in water unless otherwise indicated.
Parameters have been obtained for oxidation only (e.g., excluding reactions with water).
Reprinted from Proceedings of the Second International Symposium on Environmental Applications of Advanced Oxidation Technologies, Gloyna and U, "Supercritical Water Oxidation:
An Engineering Update,* 1993 with permission of EPRI.
-------
Supercritical Water Oxidation
Three general chemical reaction kinetics principles apply in a SCWO
system. First, the oxidation rate is independent or only weakly dependent on
the oxidant concentration. Therefore, only a small excess (over the stoichio-
metrically required) amount of oxygen needs to be fed to the SCWO reactor
to result in very high oxygen utilization. Second, the oxidation reactions
generally follow pseudo first-order kinetics with respect to the concentration
of starting compounds. Third, the activation energy for organic compounds
treated in a SCWO system ranges from about 20 kJ/mol (4.8 kCal/mol) to
408 kJ/mol (98 kCal/mol). As a result, the residence times required for a t ' '"
desired level of treatment can be determined by applying batch reactor treat-
ability study results to a first-order reaction model for a plug flow reactor,,
f
Figure 4.2
Generalized Idealized Regimes for SCWO Reactor Operations
9.000
3,000
350
400
450 500
Temperature CQ
550
600
650
Reprinted from,Proceedings of,the Second International Symposium on Environmental Applications of Advanced Oxidation
7ee/>/wtog«s.Gloynaand U, "Supercritical Water Oxidation: An Engineering Update.' 1993 with permission of EPRI.
4.16
-------
Chapter 4
4.5.2 Design and Equipment Selection
The first step in the design of a SC WO system is a series of treatability
tests to establish the reactor's operating regime. Figure 4.2 illustrates the
various conceptual regimes which have been proposed for SCWO reactor
operations (Gloyna and Li 1995).
The next step in the design of a SCWO system is specification of the size
and shape of the reactor. The reactor's size is based on the required fluid
residence time determined from treatability results and the desired waste • "'
throughput Since the chemical reactions are approximately first order and the
rates are independent of the oxygen concentration, testing of relatively few
concentrations of organics can establish the required reactor volume for the'
"worst-case" conditions. Using the equations describing first-ordef reaction
kinetics (Levenspiel 1962) allows for sizing of the reactor over a wide range of
concentrations from tests at only two conditions; however, it is recommended
that the treatability tests include several additional contaminant concentrations
to verify that first-order kinetics apply to the specific case.
Once the reactor's volume has been determined, it is necessary to select
its diameter and length. The diameter is set to maintain the fluid velocity
above a minimum value for tubular reactors (determined through treatability
studies on the specific waste streams) that minimizes deposition. During
treatment, inorganic compounds will precipitate and can clog reactor compo-
nents. It is not possible to predict the particle size of the precipitate from
first principles, so precipitation and deposition studies for the waste to be
treated should be conducted as part of the treatability study. These tests
should be conducted on a pilot-scale system to allow for realistic liquid ve-
locities and fluid dynamics.
The next design determination is the amount of heat that must be supplied
to the process. The use of external heating reduces the amount of oxygen
that must be used and compressed into the system. External heat can be
supplied by gas or oil burners or by electric heaters. External gas or oil heat-
ing results in a lower oxygen requirement, but increases the amount of fuel
needed because of combustion and heat transfer inefficiencies. Direct injec-
tion of the fuel results in essentially complete utilization of its heating value,
but requires that oxygen (or another oxidizer) be supplied and energy be
expended pumping the oxygen or oxidizer into the high-pressure system.
The tradeoff must be evaluated on a case-by-case basis. The quantity of heat
that is needed will also depend on the efficiency of the heat recovery system.
4.17
-------
Supercritical Water Oxidation
Again, a site-specific evaluation of the tradeoff between the cost of the heat
exchange equipment and the cost of fuel is required.
4.5.2.1 Materials of Construction/Corrosion Management
For most SCWO applications, the heat-exchange equipment and the reac-
tor will have to be made of specialty alloys and some parts may need to be
lined with materials such as platinum or titanium for corrosion protection.
In some applications, special corrosion resistant materials of construction ,,
may not be required for transpiring wall reactors, discussed below. As a .,
result, the increased cost of the corrosion-resistant heat recovery system
must be carefully weighed against the cost savings from reduced fuel usage.
This analysis is further complicated by the fact that increased fuel usage can
result in either a greater volume for the reactor (if the fuel is injefcted) or a
greater heat transfer area if external heating is used. Such increased size
requirements increase the need for high-priced materials of construction and,
therefore, increase system capital cost.
The point at which the liquid effluent, which contain the inorganic acids,
cool below the critical point typically is also the point where aggressive
corrosion attack is most likely. Various means have been employed to mini-
mize this attack. The most obvious is the use of materials or liners which are
resistant to this type of corrosion. Data on the corrosion resistance of a
range of materials is presented elsewhere in this chapter. An alternative
means of protection is the use of sacrifical liners. These are attacked prefer-
entially and replaced on a regular schedule. The electrochemical potentials
thus set up protect the lines and fittings. An alternative method for protect-
ing the lines is employed in the Aerojet transpiring wall reactor whereby a
non-corrosive clean fluid is injected into critical parts of the system in such a
way as it forms a non-corrosive boundary layer at these corrosion-prone
points. Clearly, the type of protection used will be determined by the nature
of the material to be treated and the relative costs of the alternatives.
Specification of the SCWO reactor and ancillary high-pressure/
high-temperature equipment must address two materials problems: (1) cor-
rosion and (2) high pressures at high temperatures. Under these conditions,
metal alloys tend to embrittle and experience creep. This material degrada-
tion, coupled with the possibility of corrosion-induced pitting, cracking, or
crazing, creates a potential design problem (Blaney et al. 1995). In some
cases, passive corrosion control will require the use of some form of lined or
4.18
-------
Chapter 4
composite materials of construction. Reactors which use a dynamic means
of corrosion control, such as the transpiring-wall platelet reactor, described
below, may require less expensive materials of construction. For example, a
passive design developed by Kimberly-Clark (patent pending) consists of a
reactor surrounded by a carbon steel pressure vessel which is maintained at a
substantially lower temperature than the reactor. The reactor is designed to
withstand the high temperature and corrosive environment, but is incapable
of withstanding the full pressure. The surrounding vessel is strong enough
to withstand the pressure. The space between the reactor and the outer ves*' >'
sel is filled with an insulating fluid. Control systems maintain the pressure'
of the insulating fluid at approximately the same pressure as tkat of the reac-
tor, resulting in lower pressure stresses on the temperature- and . •
corrosion-resistant reactor. .;
1
General Atomics conducted corrosion testing in developing a SCWO system
to treat propellants and chemical warfare agents. The results ofthis testing are
summarized in Table 4.5. Platinum was identified as the most chemically resis-
tant material of construction for SCWO processing of GB and VX agents.
Therefore, a platinum-lined Hastelloy C276 reactor was used for GB and VX
treatment. For processing the hydrolysates of mustard agent or solid propellant,
titanium was identified as the best material of construction, again as a thin liner
within a Hastelloy C276 pressure-bearing wall (Hazlebeck, Downey, and Rob-
erts 1994; Turner 1993). Both of these applications required relatively high
temperatures and pressures to achieve the desired endpoints. Because treatment
conditions required for many remediation projects are not as rigorous, less ex-
otic materials of construction may be acceptable.
Corrosion testing conducted in a MODAR vessel reactor (INEL 1995) has
shown that proper choice of materials of construction can allow operation of
a SCWO system with corrosion maintained at an acceptable level.
A new type of reactor design that was recently developed attempts to
overcome corrosion and deposition problems in a dynamic manner. The
reactor, developed by GenCorp Aerojet, is termed a platelet liner technology.
It has been successfully tested by Sandia National Laboratories and is being
incorporated by Foster Wheeler Development Corporation into SCWO dem-
onstration units for destruction of certain smokes and dyes for the U.S. Army
and of shipboard hazardous materials for the Navy.
4.19
-------
Supercritical Water Oxidation
Table 4.5
Corrosion Results Summary
HFandH3PO4 HjS
(GB)
Material 350-C
Pi D
Pt/Ir °
Pt/Rh o
Iff 0
Ti A
Tiroet21S D
Zr704 0
Mo 0
Nb A
Nb/Ti A
Ta °
A^O, 0
A1N o
Sapphire O
S,N« 0
SiC o
ZiO2 A
C22 D
HastC276 A
Hayn. 188 A
HR-160 A
Inc. 825 °
Inc. 625 o
450'C
0
o
a
0
O
O
O
o
A
O
o
0
o
0
o
0
o
o
o
0
o
o
o
550-C
a
a
a
O
A
O
O
O
O
O
O
O
O
o
o
o
o
o
o
o
o
o
o
350-C
o
a
a
A
o
o
A
o
A
a
o
A
a
A
o
o
A
A
A
A
O
A
A
\O4 and H,PO4
(VX)
450'C
a
G
a
A
O
O
A
O
O
O
a
O
O
A
O
O
O
O
0
o
o
o
o
550*C
a
a
0
A
A
A
A
O
O
A
O
O
A
A
O
O
A
O
O
O
o
o
o
HClandHjSC^
350-C
O
o
0
»
A
A
O
O
O
A
a
O
O
0
o
o
o
o
0
o
0
o
o
(Mustard)
450*C
A
A
*A
0 i
*
¥
,' 'A
O
N/A
o
A
O
O
0
o
o
o
0
o
o
A
O
0
A
550*C
a
4 '* ''
a
D
o-
A
A
N/A
O
N/A
A
N/A
A
O
A
O
O
A
O
O
O
O
o
o
o Good (<10 mi'yr corrosion rale)
A Moderate (10-200 mi'yr conosion rate)
o Poor (>200 mi'yr corrosion rate)
N/A Not available
Reproduced courtesy of General Atomics
4.20
-------
Chapter 4
Figure 4.3
Transpiring-Wall Platelet Reactor
Manifold
Transpired
Platelet
Liner
Reproduced from Rousar, \foung, and Sieger, Development of Components for Waste Management Systems Using Aerospace
Technology. 1995 courtesy of Aerojet Corp.
The transpiring wall platelet reactor consists of an outer cylindrical pres-
sure housing, which encloses a concentric cylindrical platelet liner with a
small, but finite gap between the two (Figure 4.3). Clean high-pressure wa-
ter enters the pressure housing through inlet nozzles and feeds into inlet
manifolds located strategically on the platelet liner outer surface. Following
an intricate circuitry through several platelets, the high-pressure water
stream is metered, split repeatedly, and delivered to the liner inner surface
through numerous injection points. The injected water forms a nonreactive
barrier between the platelet liner and the reactants and reaction products
(Rousar, Young, and Sieger 1995).
4.21
-------
Supercritical Water Oxidation
The transpiring-wall reactor offers additional advantages:
• The temperature of the platelet liner and the pressure housing is
controlled by the transpiration water. This isolates the reactor
from any high-temperature swings in the reaction. Furthermore,
the reactor allows the option of operation at higher reaction tem-
peratures which dramatically improves destruction efficiency and
throughput.
• The reactor configuration allows design optimization by injection''
of transpiration water at different temperatures and flow rates '
along the length by manifolding the outer housing. *
Long-term performance data on this type of reactor are now bejrig col-
lected using a pilot-scale system where throughput is 30 cc/sec $8.5 gal/hr).
The costs and potential failure modes for such a dynamic system'versus
traditional, static methods of corrosion control must be considered in the
overall determination of which design is to be used for a specific application.
4.5.2.2 Heat Transfer
Heat transfer is another important consideration. Water exhibits marked
changes in heat transfer properties near its critical point (Michna 1990).
Limited heat transfer data for supercritical water under turbulent-flow condi-
tions (mass velocities ranging from 75 to 4,000 kg/m2-sec) have been re-
ported (McAdams, Kennel, and Addoms 1950; Dickinson and Welch 1958;
Yamagata et al. 1972). A University of Texas at Austin study focused on
heat transfer to supercritical water under laminar- to transient-flow condi-
tions (mass velocities ranging from 2.6 to 49 kg/m2-sec)(Michna 1990).
Heat transfer to water was enhanced for bulk temperatures just below the
critical point. Increased natural convection effects due to the extremely low
kinematic viscosity of water near the critical point are believed to be partly
responsible for such enhancement.
The critical point of a solution is similar in concept to the critical point of a
pure compound. Each pure compound has a unique critical point defined by the
temperature, pressure, and density (T-V-p), although from the "phase rule," if
any two of these variables are specified, the third is fixed. For a solution, the
critical T-V-p is not usually a unique point Rather, it is a range of values for
each variable within which the solution exhibits properties similar to those of a
4.22
-------
Chapter 4
pure compound at its critical point The T-V-p above which the solution exhib-
its the critical properties needed for a given application is commonly termed the
critical point of (he solution, but the fact that it is actually a range should be kept
in mind. The critical temperature is the apparent critical temperature of the
mixed materials, that is, the temperature and pressure at which the liquid and
gas properties of the mixture approach one another; the critical temperature
refers to the critical temperature of a pure substance.
. •»/*
Figure 4.4 *
Overall Heat Transfer Coefficient as a Function of Core Temperature
4000 -
3500
?
| 3000-
1 "~
| 2500
a
.§ 2000
H
g 1500
3
| 1000-
500
0
350
_L
J_
_L
JL
_L
360
370 380 390
Core Temperature ('C)
400
410
9.53 mm OD Tube
0.39 mm Wall
5.44J
-------
Supercritical Water Oxidation
As shown in Figure 4.4, the heat transfer coefficient rapidly increases
with increasing bulk temperature as the pseudo-critical temperature is ap-
proached. Deterioration in heat transfer occurs for bulk temperatures just
above the critical temperature. Such deterioration appears to be largely the
result of variations in the physical properties between the fluid at the surface of
the tube and the bulk fluid. Therefore, deterioration is greatest for pressures
close to the critical pressure where the physical properties change most rapidly.
Design correlations developed for high-temperature/high-pressure steam'
boilers and turbines can be used in the design of SCWO systems. Figure1
4.5, which shows the viscosity of water/steam in the critical tnd supercritical
region, illustrates the rapid changes in physical properties associated with
this region. As can be seen, as the critical temperature is apprtiached
(374.1°C [705.4'F], and 2.210 MPa [3,206 psi]) the fluid's y-iscosity drops
rapidly, approaching that of a gas rather than of a liquid. :
The operating conditions for the SCWO process can be modified to take
advantage of the changes in the solubility of inorganic materials at different
temperatures and pressures so as to remove inorganic contaminants as a concen-
trated stream. For example, at high temperature (800°C [1,472°F]) and rela-
tively low pressure (240 bar [3,500 psi]) most sticky salts will precipitate and
then melt, thereby, tending to adhere to the walls of the system. Metal oxides,
on the other hand, tend to have higher melting points and therefore, do not ad-
here to system surfaces. Most salts will precipitate at pressures of about 242
bar (3,500 psi) and temperatures of about 400°C (752°F). However, at a higher
pressure (1 kbar or 14,570 psi) and a lower temperature (500°C [932°F]) and
most of the salts will remain in solution. The higher pressure (1 kbar) also
causes a relatively large increase in water density (0.3 to 0.8 g/mL) without an
accompanying increase in viscosity (approximately 0.05 cP).
A variety of methods have been developed (such as filters or
hydrocyclones) to remove a large fraction of the inorganic solids (salts and
inorganic oxides) from the effluent. Salts are generally sticky under SCWO
conditions and, hence, they are more difficult to remove and tend to clog the
system. Metal oxides are generally not sticky and, hence, are easier to re-
move and cause fewer clogging problems than salts.
One means of reducing the impact of sticky salts on a system is to reduce
the number of surfaces on which they can deposit and lead to plugging. The
transpiring-wall reactor described above is one solution being proposed to
minimize the deposition of sticky solids on the reactor walls; however, it
4.24
-------
Chapter 4
Figure 4.5
Viscosity of Water and Water Vapor in the Critical Region
1.60 -
f
•i 1.20 -
0.80 -
0.40 -
700
800
Temperature (T)
H^S^JSfl «•» permi88ion of The McGraw-Hill Companies from Perry, Chilton, and Kirkpatrick. Chemical Engineers
1993.
4.25
-------
Supercritical Water Oxidation
does not eliminate these problems entirely. Deposition and plugging can
still occur at the inlet and outlets to the transpiring-wall reactor as well as on
the surfaces of ancillary heat transfer equipment. Finally, some salts can
redissolve when the effluent from the SCWO reactor is cooled and depres-
surized to below the critical point of water. Impact of these salts on the ef-
fluent quality must be considered. Development is continuing on the use of
filtration, hydrocyclones, and flushing methods and vendors should provide
data validating their proposed method of dealing with solids deposition for
each particular application.
Even though they are not sticky, oxide particulates can also cjog narrow
passages, inlets, and outlets of the reactor; however, these are more easily
removed from the system than sticky salts by means such as settling, filtra-
tion, or hydrocyclones. The performance of small hydrocyclone^ for
non-sticky solids is illustrated with data derived from a UT pil&t plant
(Dell'Orco 1991; Dell'Orco 1993, Dell'Orco et al. 1991a). Mln-U-Sil 5
(quartz silica, U.S. Silica Corp.), exhibiting density and particle-size charac-
teristics similar to the oxides formed in reactors during SCWO operations, is
typical of various particles evaluated. To evaluate hydrocyclone perfor-
mance, it is necessary to determine solids separation efficiency and
particle-size distributions. Equation 4.5 can be used to estimate solids sepa-
ration efficiency as a function of Stokes' Number (W):
EG=1-AXPB (4.5)
where: ¥ = {(pp- pw) 'd^v.} / 18|iD
p , pw = density of particle and water respectively, kg/m3
D = diameter of cyclone at feed port, m
v. = fluid inlet velocity, m/sec
d = characteristic particle diameter, m
A,B = empirical constants (0.018 and 0.64, respectively, for
10 mm (0.4 in.) cyclone and silica, titania, and zirconia
powder, to be determined by treatability testing for
specific application; and
EG = gross separation efficiency (dimensionless).
.Figure 4.6 shows experimental gross separation efficiency data for two
hydrocyclones (10 mm and 25.4 mm diameters [0.4 and 1 in.]) and the
model calculation for the 10 mm (0.4 in.) hydrocyclone. Gross separation
efficiencies near 80% are achievable for silica at temperatures above
4.26
-------
Chapter 4
Figure 4.6
Gross Separation Efficiency (as penetration) for Two Hydrocyclones
(Hydrocyclone A: 10 mm diameter; Hydrocyclone B: 25.1 mm diameter)
IV
1
?
3.
1 «
f
1
'0.01
0.001
0.00
"
_ IBB %s-JI • B
E D ri^^BB •
I 0 » 9 ® N- 0 |
000 Sx» B<* *
: D C^%
: n
I n
"** *
= o e
: o
1 — i — i i n nl 1 i i i 1 1 ni i i i i 1 1 ni i i i i 1 1 n
01 0.001 0.01 0.1 l
Stokes' Number (dimensionless)
° Hydrocyclone A, Titania
o Hydrocydona A, Zrconia
• Hydfocydone A, Silica
• Hydrocyclone B, Silica
... Data Fit, Hydrocyclone A
>odings oftho Second International Symposium on Environmental Applications of Advanced Oxidation
and U, Supercritical Water Oxidation: An Engineering Update." 1993 with permission of EPRI.
4.27
-------
Supercritical Water Oxidation
300"C (572* F), while at the same temperature range, gross separation effi-
ciencies for the more dense zirconia particles are greater than 99%. The
individual silica particle-size (grade) separation efficiencies are shown in
Figure 4.7. The separation efficiency of these solids is directly tied to par-
ticle size, temperature, and pressure, although particle size would be less of a
problem if filtration were employed instead of a cyclonic separator. The
objective is to accurately predict separation efficiencies for individual par-
ticle diameters. Equation 4.6 can be used to predict separation efficiencies
of bulk solid streams and individual particle sizes (Leith and Licht 1972): '
h = 1 - exp{-2C¥(n + l)"/'2-2" ~ (4.6)
i
where: h = separation efficiency; <>
C = geometrical parameter (dimensionless); . »
n = vortex exponent (dimensionless); and •*
*¥ = as defined for Equation 4.5.
Chromium speciation during SCWO is of considerable importance (Rollans,
Li, and Gloyna 1992). While the primary concern is the chromium in the waste
streams being treated, a secondary concern is the presence of chromium in
many "stainless" steels that might be used for the reactor or other components
and could leach due to the corrosive environment. Chromium behavior was
studied in a bench-scale, vertical, concentric-tube reactor system treating mu-
nicipal wastewater sludges to determine the hexavalent and soluble trivalent
chromium concentrations in the reactor bottom and the treated effluent. The
reactor material was Stainless Steel 316.
Under SCWO conditions, hexavalent and trivalent chromium corrosion
products were generated and removed by precipitation. At 400°C (752°F)
and a Reynolds number of approximately 8,000, the hexavalent and trivalent
chromium concentrations in the effluent were <0.004 mg/L and 0.163 rng/L,
respectively. Similarly, with an influent feed of 8.76 kg (19 Ib) of sludge,
the concentrations of hexavalent and trivalent chromium in the bottom of the
reactor were 0.288 mg/L and 3.712 mg/L, respectively.
Optimum separation efficiency, species distribution, and corrosion effects
depend on the type of sludge and treatment conditions. Two mechanisms
appear to be responsible for separation of hexavalent and trivalent chromium
from effluents: (1) the precipitation of chromate complexes and trivalent
chromium salts, and (2) the settling of solid residues, including sorbed
4.28
-------
trivalent chromium. In the case of the municipal sludge, hexavalent chro-
mium, in the form of insoluble chromate complexes, settled from the bulk
supercritical fluid. The concentration of the settled hexavalent chromium in
the reactor bottom was at least 72 times greater than the chromium concen-
tration in the effluent. At the subcritical and laminar-flow conditions, the
concentrations of chromate complexes in the effluent and reactor bottoms
were 0.046 mg/L and 0.035 mg/L, respectively.
Figure 4.7
Grade Efficiency for Hydrocyclone Separation of Siiiap '
1
-o.o
0.1
0.2 0.3
Log dp (dp in microns)
0.4
0.5
0.6
o 70'C
• WC
• 213'C
o 371'C
inor of the Second International Symposium on Environmental Applications of Advanced Oxidation
K3 U, Supercritical Water Oxidation: An Engineering Update," 1993 with permission of EPRI.
4.29
-------
Supercritical Water Oxidation
For both temperature regimes used during the treatment of the industrial
sludge, the concentration of hexavalent chromium was below the detection
limit in the effluent and reactor bottom. A large portion of divalent chro-
mium was sorbed onto the solid residues which settled in the reactor bottom.
Concentrations of soluble trivalent chromium in the effluent and reactor
bottom were comparable.
In general, moderate amounts of chromium can be tolerated in the
reactor's materials of construction. The possibility of trivalent chromium .,
being converted to hexavalent chromium appears small. Hexavalent chro-,
mium present in the influent appears to precipitate out of solutions at typical
SCWO reactor conditions and, if chromium is an effluent problem, it ap-
pears possible to incorporate particulate removal into the system anil collect
the chromium as a separate concentrated stream. J
£
4.5.3 Process Modification
The SCWO process is highly flexible and can be modified to accommo-
date a wide variety of conditions and waste streams. Modifications that have
been proposed include:
• use of deep wells to reduce pumping costs and the pressure drop
across the reactor lining; and
• use of catalysts to increase the rates of chemical reaction.
Deep well reactors are an adaptation that have been used for WAO sys-
tems, but apparently have never been applied to SCWO systems where the
much higher temperatures and pressures make it difficult, if not impossible,
to implement Conceptually, the approach is simple. A well is drilled into
the earth (a stable and impervious geologic formation is a requisite) down to
a sufficiently deep level so that the hydrostatic pressure equals the reactor's
operating pressure. To illustrate, the well would have to be approximately
2,252 m (7,400 ft) deep to achieve the critical pressure of water at its bottom.
Then, the well is lined, and a smaller annular pipe is dropped to within a foot
or two of the bottom. The annular pipe is braced against the outer casing in
such a manner that liquid will freely flow past the bracing. Heaters and
pipes feeding pressurized oxygen or another oxidizer would also need to be
lowered into the lower part of the well.
The contaminated liquid is pumped down into the well where it is heated as
the liquid's hydrostatic pressure increases. The combination of hydrostatic
4.30
-------
Chapter 4
pressure and heat generated by the oxidation of the organic compounds in the
waste turns the volume at the bottom of the well into a SCWO reactor. The cost
of pumping is reduced and, because the reactor is in the well, the pressure drop
across its lining is lessened. The pressure stresses on the reactor are similar to
those of the pressure stresses across the inner reactor in the Kimberly-Clark
reactor described in Section 4.5.2. While the approach is intriguing, the diffi-
culties associated with maintaining a tight seal and maintenance of the system
2.2 kilometers (almost one and one half mile) underground diminishes its vi-
ability at the current state of development.
' / '
Catalysts can enhance the total conversion of complex organo-nitrogen
compounds (including lower molecular weight transformation products),
shorten the reaction time, and lower the required reaction tempejaWe.
However, the presence of inorganic materials can result in depositions on the
catalysts and reduce catalyst life. The use of catalysts has bejen tested at a
laboratory-scale on WAO systems to good effect and a large body of research
data on the subject is available (Gloyna and Li 1993).
4.5.4 Pretreatment Processes
The only physical pretreatment that may be required for SCWO is some
form of screening followed by masceration to remove or destroy solids too
large to pass through the pump. Generally, abrasive solids need to be re-
moved or size reduced sufficiently to protect the high-pressure pumps,
valves, and seals.
A common form of pretreatment used for many highly-reactive organic
compounds such as chemical agents or rocket fuels is hydrolysis. During
hydrolysis, the waste is mixed with water, often containing a caustic. Hy-
drolysis reduces the wastes reactivity and, in the case of many chemical
munitions agents, toxicity to the point where they are safer to handle. Suffi-
cient water is added to bring the wastes' heating value to within that required
for the SCWO reactor. SCWO is currently under active evaluation as a
means of treating such hydrolyzed highly-reactive wastes.
4.5.5 Posttreatment Processes
The SCWO system produces three by-product streams: (1) aqueous prod-
uct, (2) solid precipitate or filtrate, and (3) gas. The aqueous product con-
sists of the water and dissolved solids and residual organic compounds, if
4.31
-------
Supercritical Water Oxidation
any, that are products of the SCWO chemical reactions. The solid precipi-
tate consists of oxides and other insoluble inorganic materials, and the gas
stream consists of nitrogen (if air is used as an oxidizer), excess oxygen,
carbon dioxide, and water vapor with possible traces of carbon monoxide,
sulfur oxides, nitrogen oxides, and organic constituents.
The aqueous product coming from a properly designed and operated
SCWO system contains extremely low levels of organic material, often be-
low the levels of detectability. Low concentrations of acetic acid or ammo-
nia are found. These materials can be removed to below detection limits by,
increasing the temperature or the residence time of the SCWO reactor. Alterna-
tively, they can be removed by subjecting the effluent to normal biological treat-
ment in a wastewater treatment plant. The amounts of hazardous products of
reaction (the equivalent of products of incomplete combustion, PICsV in an
incinerator) that occur in the SCWO effluent, and indeed in all th^ waste
streams from a properly specified and operated SCWO system, are extremely
small. The case study and results presented in this chapter indicate the levels
that have been encountered during testing under various conditions.
Inorganic materials, such as metals, will pass through the SCWO system
or be deposited in it, the latter being undesirable. SCWO conditions oxidize
most metals and these will precipitate. In some applications, where the
waste being treated has a high metals content, some form of precipitation or
ion exchange may be needed.
The solid wastes from a SCWO system are the precipitates, filter cakes,
and internal system deposits which are removed during maintenance. Their
composition is completely site-specific and no general posttreatment meth-
ods can be recommended.
The small-volume gas stream emanating from the SCWO system consists
largely of the following constituents:
• excess oxygen pumped into the process;
• water vapor;
• nitrogen from any air that may have been pumped into the reactor as
oxidizer and from nitrogen-bearing constituents in the waste; and
• carbon dioxide which is the product of oxidation of the organic
compounds in the waste.
4.32
-------
Chapter 4
In addition to these relatively major constituents, trace gases may be
found in very low concentrations in the emission stream from a SCWO sys-
tem. The concentrations found are usually much lower than those found in
emissions from any type of combustion equipment. More important, when
oxygen (rather than air) is the oxydent, gas emissions are much smaller than
from combustion systems. The following materials might also be present in
the emissions:
• CO2, traces of CO sometimes occur as normal equilibrium prod?
ucts of the oxidation reaction;
• ammonia, if nitrogen is present (depends on reactor conditions);
• oxides of sulfur, if sulfur is present in the waste (depend on reac-
tor conditions); V
• acetic acid. $
While traces of PICs such as chlorodibenzodioxins and chlorodibenzofurans
(dioxins, furans) have been found in the effluent from a SCWO reactor operated
at lower temperatures (see Section 4.7.3), these compounds have a very low
volatility and are not carried into the gas stream in a SCWO reactor. It is noted
that higher operating temperatures and pressures can. destroy these contami-
nants directly in the SCWO reactor.
The particulate concentration in the gas stream is negligible and because
of the very low flow rates the stream can be readily treated by standard air
pollution control techniques, such as adsorption, to control the other trace
contaminants. In most cases, such treatment will not be required.
4.5.6 Process Instrumentation and Controls
The process requires careful control to maintain the required temperature
and pressure of the reactor and ancillary units. Because of the large changes
in physical properties and inorganic compound solubilities that occur in the
super-critical region, this control has relatively small tolerance. The process
is controlled through the flow rate at the waste pump, the water pump, the
oxidant system, and the temperature in the reaction zone. Feedback is re-
ceived through pH and TOC analysis of the liquid effluent. Other than these
special requirements, the instrumentation and control needs of the system are
equivalent to those of a very small hydrocracker which is common in the
petroleum industry.
4.33
-------
Supercritical Water Oxidation
4.5.7 Safety Requirements
A SCWO system operates at relatively high pressures — typically on the
order of 276 bar (4,000 psi). This fact is sometimes pointed out as a safety
concern. While this is an understandable assessment, a more careful analy-
sis of a SCWO system reveals that the risk is less than or equivalent to those
of common industrial processes, primarily because of the minimal amount of
energy in the high-pressure portion of the system.
The SCWO system requires routine safety systems and temperature and ,
pressure controls as well as routine safety procedures associated with the . /
industrial use of pressurized liquid and oxygen. The safety requirements are
analogous to those of very small capacity industrial processes used in the
petroleum refining industry. ,;
_,, r
4.5.8 Specification Development £
The specifications for a SCWO system include the following major items:
• Quantity of material to be treated and rate of treatment.
This information is required to size the system and to establish
the system's anticipated life. The anticipated life is needed to
both establish a cost and to specify materials and a maintenance
and parts replacement schedule so that the system will withstand
the harsh environment for its anticipated life.
*
• Total organic content of the waste stream.
This information is needed to determine the amount of dilution,
heat recovery, external heat, added fuel that is needed to achieve
the desired temperatures, and residence times.
• Maximum quantities of organic halogens, phosphorus,
and sulfur.
This information is used to establish the amount of neutralizing
chemicals that will be required, to select corrosion-resistant ma-
terials, solids separation processes needed, and operational pro-
grams required to achieve the desired treatment
4.34
-------
Chapter 4
• Specific contaminants that must be destroyed and acceptable
minimum levels of destruction.
This information is needed to establish the reactor residence time
and volume.
• Quantity and size of suspended solids in the feed material.
This information is needed to determine size reduction and
pumping requirements.
• The types and quantities of soluble inorganic constituents - T
that are present. ' /
This information is needed to establish the type and amount x>f
insoluble material that is present at the reactor conditions and to
ascertain the "stickiness" of the precipitates, where 4 the system
they may come out, and whether these solids pose apposition or
clogging problem during operation. For example, if non-sticky
solids such as metal oxides are found to precipitate from the
supercritical fluid in the reactor, a tubular reactor must be de-
signed so that the fluid flow rates are high enough to sweep them
out of the system. For a transpiring-wall reactor, these inorganic
concentrations will be used to establish the required water flow
rates through the liner.
Sticky solids such as mineral salts also precipitate in the
supercritical fluid and these may adhere to the reactor wall. Vari-
ous proprietary methods have been developed by different
SCWO vendors to minimize or avoid buildup of these salts on
system surfaces. The transpiring-wall reactor is one example of
the different approaches to this design challenge.
Discharge requirements for organic and inorganic constituents.
This information will establish the temperature and residence
time required to achieve the desired level of organic compound
destruction, and whether posttreatment systems such as filtration
or ion exchange systems are needed to remove suspended or
dissolved inorganic materials.
Available utilities at the site.
4.35
-------
Supercritical Water Oxidation
4.5.9 Cost Data
Cost analyses of a SCWO system for treating paper mill sludge were
performed independently by Kimberly-Clark Corporation and by Charles
Eckert of the Georgia Institute of Technology (Blaney et al. 1995). The
details of the system and of the treatment parameters are given in the case
study in Section 4.7. The results were comparable and indicated a cost of
$33 to $44/wet tonne ($30 to $40/wet ton) of sludge (dewatered to 50%
solids). These costs included a credit for recovered calcium carbonate ash „,.
from the SCWO unit.
There is no technical reason why SCWO cannot be applied te wastewa-
ters containing a wide range of organic material concentrations. However,
below concentrations which are autogenic (i.e., the organies in the ground or
surface waters being treated provide sufficient energy for heating and pump-
ing it (and the oxygen source) required affects the process cost-effectiveness.
Table 4.6 provides the steps of a simple cost analysis that can be used to
determine the amount and approximate cost of external energy required for
an SCWO unit This analysis is a "worst-case" because to assumes no heat
recovery and no energy released from the oxidation of the waste.
The calculations of Table 4.6 are for a 37.85 L/min (10 gal/min) system
operating at 242 bar (3,500 psi) and 538'C (l,OOOeF).«The calculation as-
sumes that electricity is used to drive the compressors and that either elec-
tricity, natural gas, or fuel oil is used for external heat. External gas heating
was assumed to have a 75% thermal efficiency, oil heating a 60% efficiency,
and the overall pump/motor efficiency of 50%. Electricity is assumed to
cost 6.00 per kWh, natural gas $7.00 per MBTU, and fuel oil at 600 per gal.
Other assumptions are shown in Table 4.6.
As this calculation shows, the cost of pumping the material up to the op-
erating pressure of the reactor is only a minor component of the overall cost
of the system, about 6.30 per 1,000 gal. The cost of heating can be signifi-
cant ranging in cost per 1,000 gal from $204.72 for electric heating to
$85.70 for oil heating. It is estimated that these costs can be cut in approxi-
mately half with appropriate use of heat recovery. The economics clearly
favor SCWO for the treatment of wastes containing high concentrations of
organic compounds whose energy content can be used towards heating the
waste streams. Optimum economics for any applications requires a far more
detailed analysis than this one which is only intended to illustrate the magni-
tude of the various costs.
4.36
-------
Chapter 4
Table 4.6
Power Cost Analysis for SCWO
Cost of Pumping to SCWO Operating Pressure
Critical Pressure of Water
Critical Temperature of Water .
1. Pressure of Water
2. Assumed Flowrate
3. Conversion Factor
4. Assumed Flowrate
5. Conversion Factor
6. Pressure of Water
7. Pumping and Compressor Power
8. Pumping Power
9. Conversion Factor
10. Pumping and Compressor Power
11. Conversion Factor
12. No Loss Power
13. Assumed Pump/Motor Efficiency
14. Power Usage for Pumping and
Compressor
15. Conversion Factor
16. Power Usage for Pumping and
Compressor
17. Assumed Cost/kWh
18. Pumping Power Cost
19. Pumping Power Cost
3,206.00 Ib-fl/in.2
705.4T
3^00.00 lb-ft/in.2 ,
lOgal/min
7.4805 ftVgal
1.3368 [2/3] ft3Anin
144in.2/ft2
504,000 [1«5] Ib-fW2
673,751.75[8»6]ft-lb/min
11,229.20 ft-lb/sec
0.00181818 horsepower/(ft-lb/sec)
20.417 [9 • 8] horsepower
0.001356 kW/(ft-lb/sec)
15.23 [10-11] kW
40.00%
38.0670112/13] kW
3,600 sec/hr
0.00105742 [14/(2 • 15)] kWh/gal
$0.060 $/kWh
$0.00006344 [17 • 16] per gal
$0.06344496 per 1,000 gal
From Perry's
From Perry's
P 3-192
P3-192
Cost of Heating to SCWO Operating Temperature
Feed Water Temperature
Water Enthalpy
Water Enthalpy
AH
Conversion Factor
Heat Requirement
60T
28.06 Btu/lb® 1,000'F
1,424.5 Btu/lb
1396.44 Btu/lb
8.341b/gal@60T
11,646.31 Btu/gal
From Perry's P 3-191
From Perry's P 3-192
4.37
-------
Supercritical Water Oxidation
Table 4.6 cont.
Power Cost Analysis for SCWO
Gas Heating
Cost of Gas ($/MBtu) $7.00
Heat Cost per gal @ 100% Efficiency $0.0815
Assumed Heating Efficiency 75.00%
Heating Cost per gal $0.1087
Electric Heating
Cost of Power $0.06
Conversion Factor kWh/Btu 0.000292875 ^
Power per gal @ 100% Efficiency 3.4109 £
Heating Cost per gal $0.2047
Oil Heating
CostofOil.($/gal)
Heating Value of O3, Btu/gal
Heat Cost per gal @ 100% Efficiency
Assumed Heating Efficiency
Heating Cost per gal of Water
$0.60
136,000
$0.0514
60.00%
$0.0856
Total Power Cost Electric Gas O»
per gal of Water
per 1,000 gal of Water
$0.2047
$204.72
$0.1087
$108.76
$0.0856
$85.69
4.5.10 Design Validation
The major design problems for SCWO systems have focused on finding
suitable materials to withstand the high temperature, pressure, and corrosive
environment and to manage inorganic materials within the reactor for con-
tinuous operation. These problems, while not completely solved, appear to
be sufficiently overcome to allow design, construction, and operation of
4.38
-------
Chapter 4
full-scale systems. For example, corrosion problems are being addressed
through the use of special metals as in the General Atomics, EcoWaste Tech-
nologies, Inc., Foster-Wheeler, and ceramic cladding as in the Kimberly-
Clark Reactor. Plugging problems are being addressed in tubular reactors by
maintaining the fluid flow high enough to scour the surfaces as, for example
in General Atomics reactors or by the use of unique designs which prevent
the fluid being treated from coming in contact with the walls, as in the
Aerojet transpiring-wall reactor.
SCWO appears appropriate to those situations where (1) the waste '"
stream contains a sufficient organics content to provide the majority of'the
heating value needed for the high temperatures, or (2) the cdhtaminants are
so highly toxic that safe treatment demands a fully-enclosed process. '
SCWO is becoming increasingly competitive with incineration for hazard-
ous waste disposal. , 7
4.5.11 Permiff ing Requirements
Because it is a fully-enclosed technology, a SCWO system requires far
fewer permits than, for example, an incinerator. If the system is treating a
hazardous waste (as defined by the Resource Conservation and Recovery Act
[RCRA]), then the requisite RCRA permits would be required. In addition
the system will require discharge permits for the liquid effluent and air per-
mits for the small gaseous emissions. None of these permits should pose
significant difficulties since a SCWO system would not be considered a
major source of contaminants.
4.5.12 Performance Measures
There are no unique or unusual performance measures for SCWO sys-
tems. Performance measures are the percentage of organic content de-
stroyed, length of time of continuous operation, mean time between failures
(operational reliability), reactor life cycle, and the ability for the system to
react to a change in the waste stream characteristics.
4.5.13 Design Checklist
Refer to the specification development guidance in Section 4.5.8.
4.39
-------
Supercritical Water Oxidation
4.6 Implementation and Operation
4.6.1 Implementation Strategies
Because of its present stage of development, implementation of SCWO
requires a vendor experienced in the design, construction, and operation of
such systems. The vendor must also have available a range of laboratory and
pilot-scale equipment for conducting treatability studies. This equipment
must be capable of achieving the range of operating conditions that encom7/
pass those required for the installation. ^
/
4.6.2 Start-up Procedures
-------
Chapter 4
4.6.4 Operations Monitoring
SCWO systems require a high degree of process control for proper opera-
tion. The systems should be controlled and monitored by an advanced pro-
cess control system which incorporates an automatic waste-feed shutoff and
orderly shutdown in case of process failure or excursion of key control pa-
rameters. In addition to the automatic controls, proper operator training and
experience is essential. Items that must be monitored during operation are
typical of a high-pressure industrial process and include the following:
• liquid flow rates; • '''
'/ •
• inlet and outlet absolute pressure for the reactor and all pieces of
high-pressure equipment. Interlocks must be installed to stop the
influent flow in case of exceedences; . {
y
• pressure drop across the reactor and all other pieces of equip-
ment to identify possible plugging. The pressure drop must
be monitored independent of the inlet and outlet pressures
since the very high overall pressure would tend to mask the
gradual increase in pressure drop that might be caused by
blockage. Interlocks must be installed to stop the flow in
case pressure drops exceed design levels;
• safety shutoffs;
• tank or other storage systems for holding the influent during a
shut-down; and
• reactor inlet, outlet, and intermediate temperatures.
4.6.5 Quality Assurance/Quality Control (QA/QC)
No unusual QA/QC procedures are required for SCWO; however, peri-
odic nondestructive testing of the high-pressure components is necessary to
ensure safety and operability. All ANSI and ASME codes for high pressure/
temperature equipment design, testing, operation and maintenance must be
scrupulously adhered to and an appropriate record of compliance and testing
must be maintained.
4.41
-------
Supercritical Water Oxidation
4.7 Case Histories
A number of groups were identified who are currently actively conducting
research or marketing commercial SCWO systems. It is impractical to cite
all activities in this field herein; however, the following three groups were
identified with ongoing commercial activities and technology.
4.7.1 Commerical Activities
.r
4.7.1.1 Eco Waste Technologies Inc.
Eco Waste Technologies (EWT) currently operates a 151.41 L/hr (40 gal/
hr) pilot plant at the University of Texas at Austin that has been used for
treatment of a wide variety of wastes in numerous tests. A commercial
SCWO plant went on-line in the fall, 1995 at Huntsman's Corporation, Aus-
tin, Texas plant. The facility is based on the technology developed by EWT
and has a capacity to process 27,,254.48 L/day (7,200 gal/day) of wastewater.
The combined waste feeds to the system consist of process wash water and
other chemical wastes containing methanol, polyols, amines, ammonia, and
oxygenated organic compounds. The combined stream is about 10% (by
weight) organic compounds with a total organic carbon (TOC) content of
about 50,000 mg/L (Weismantel 1996). The plant's effluent is claimed to be
of consistently high quality. Tests of its performance are planned in Sweden
in the near future.
4.7.1.2 General Atomics
General Atomics (GA)[also MODAR, Inc. which was recently acquired by
GA, and Organo, the licensee in Japan] currently operates three pilot-scale test
systems. The work is being supported by the Environics Directorate of
Armstrong Laboratory at Tyndall Air Force Base in Florida. The first pilot plant
has a capacity of 5.68 L/min (1.5 gal/min) and is rated for operation at a maxi-
mum temperature and pressure of 650°C (1,202°F) and 306 atm (4,500 psig),
respectively. It is located at the GA facility in San Diego and was developed for
the Defense Advanced Research Project Agency (DARPA) for the treatment of
' chemical agents, propellants, and other hazardous wastes. This pilot plant has
been used most recently for development tests leading to a shipboard SCWO
system for destruction of Navy excess hazardous materials. The second pilot
plant, located at a site near Brigham City, Utah shown in Figure 4.8 was
4.42
-------
Chapter A
developed to treat up to 20.9% (by weight) of hydrolyzed solid rocket pro-
pellant in aqueous solution. Operating ranges that have been tested are from
425 to 610°C (797 to 1,130°F) and flow rates of 1.14 to 1.89 L/min (0.3 to
0.5 gal/min). The third pilot plant, also located at the GA facility, was ac-
quired during the recent acquisition of MODAR. This pilot plant had been
used extensively by MODAR to treat a wide variety of chemical plant wastes
in demonstrations conducted for industrial plants including the following.
Figure 4.8
Air Force SCWO Pilot Plant
The reactor is in the "slatted box" in the background, the liquid oxygen tank in the center of the photograph, and the heat
exchange systems is to the left outside the area of the photograph.
Reproduced courtesy of General Atomics
4.43
-------
Supercritical Water Oxidation
Permitted and developed a process design for a 18,926.72 L/day
(5,000 gal/day) demonstration unit to handle a variety of wastes
at an operating RCRA-permitted TSD facility in Texas.
Fabricated a skid-mounted pilot unit for treatment of PCBs, oils,
solvents, and sludges at Niagara Falls, NY (G.T. Hong "Hydrottier-
mal Oxidation: Pilot Scale Operating Experiences" Paper No.
IWC-95-51, presented at the 56th Annual International Water Con-
ference, October 30-November 1,1995, Pittsburgh, PA). This unit,,
was also used for treating solvents, biological wastes, and ammonia
at a Smith Kline and French manufacturing facility in Pennsylvania
(Johnson, J.B., Hannah, R.E., Cunningham, V.L., Daggy, B P., f
Sturm, F.J., and Kelly, R.M., "Destruction of Pharmaceutical and
Biopharmaceutical Wastes by the MODAR SCWO Process." Bio-
technology, Volume 6, pp 1423-1427, December, 198$).
Tested more than 400 pure and complex mixtures in bench and
pilot-scale SCWO systems.
Evaluated more than 100 metals, ceramics, and coatings during
more than 5,000 hours of operation under supercritical conditions
over the past 14 years.
Developed and piloted a patented concept for the control and
handling of sticky solids generated during the processing of halo-
genated feedstocks.
Adapted, implemented, and debugged commercial off-the-shelf
hardware and software for computer control of an operating pilot plant
Permitted SCWO processes for operation in New York and in
Massachusetts.
Under subcontract to Stone & Webster, GA/MODAR completed
a program for the U.S. Department of Energy (DOE) to demon-
strate the MODAR-designed pilot plant's ability to effectively
process mixed (radioactive and hazardous) wastes representative
of those generated in the DOE Weapons Complex. This pro-
gram, which started in September 1994, included modifications
to the existing 1,892.67 L/day (500 gal/day) pilot plant to im-
prove the process performance and facilitate the processing of
Trimsol oil contaminated with a variety of heavy metals
(Bettinger, Ferland, and Killilea 1994).
4.44
-------
Chapter 4
4.7.1.3 Sandia National Laboratories
Sandia National Laboratories (Livermore, CA) in cooperation with Foster
Wheeler Development Corporation and Aerojet GenCorp currently operates
four supercritical water oxidation reactors, two flow, one batch reactor, and
one reaction cell for studying hydrothermal flames at supercritical condi-
tions. Further descriptions are now provided for the flow reactors: the Engi-
neering Evaluation Reactor (EER) and the Supercritical Fluids Reactor
(SFR). The EER is a second generation reactor developed to evaluate the
engineering aspects of SCWO technology. Its modular design facilitates,,' /'
different test configurations. The EER has a maximum operating tempera-
ture of 650'C (1,202°F) at an operating pressure of 345 bar (5$00 psi). The
total flow capacity is 30 cc/sec (28.5 gal/hr). The system can provide either
air or hydrogen peroxide as the oxidizer.
-------
Supercritical Water Oxidation
The following case studies are representative of pilot-scale SCWO activi-
ties. They were selected for discussion herein because the tests were con-
ducted under the auspices of an independent third party or because the de-
**'"•-•"* " tailed results have been subjected to peer review. There are numerous other
examples of successful application of the technology. While the examples
given herein are of pilot-scale programs, the results could be readily scaled
up to operation in the 37.85 L/min (10 gal/min) range, a size which is con-
sidered to be commercial scale for a system which is designed to treat
wastes with very high organics content. •'/'
.- 4.7.2 Laboratory-Scale Study, Chemical Agent Treatment ,
This section presents a set of tests conducted by General Atomics'at ap-
proximately 50 niL/min on chemical warfare agents (GB, VX, and mustard)
and a second set of tests conducted at approximately 1.51 L/mi$ (6.4 gal/
min) treating effluent from base hydrolysis of rocket propellant.
4.7.2.1 GB Agent Treatment
The tests were conducted at the Illinois Institute of Technology Research
Institute (UTR!), a facility especially designed and certified to safely handle
chemical agents, beginning on May 10,1993. All tests were performed at GB
concentrations of approximately 1% (by weight) with 100% excess oxygen.
Table 4.7 presents the GB-test matrix.
Table 4.7
GB Agent Bench-Scale Test Matrix
Test No.
1
2
3
4
5
Pressure
(psig)
4000
4000
4000
4000
4000
Temperature
CO
550
450
550
450
500
Total Flow Rate
(mlVmin)
505
39.4 •
503
43.4
46.4
Residence Tune
(sec)
16
29
16
26
20
Test Duration
(min)
42
15
54
71
58
4.46
-------
Chapter 4
Gas and liquid samples were collected and analyzed throughout the test
series. No agent was detected in any liquid samples, signifying a destruction
in excess of 99.99999%. Higher destructions have been achieved, but
99.99999% is the maximum that can be measured given an influent agent
concentration of 10,000 ppm (1% by weight) and the 1 ppb detection limit
for GB in liquid samples. Additionally, no agent above the allowable expo-
sure limit (AEL) was found in the gaseous effluent samples analyzed on-line
by a Minicams* analyzer.
After the gaseous and liquid effluent samples were confirmed to be agent*
free, they were shipped to the Institute of Gas Technology (IGT) and the' / '
University of Texas JJ. Picklo Research Campus (UTPRC), respectively, for
further analysis. Gas samples were found to contain oxygen, nitrogen, al-
gon, carbon dioxide, and trace amounts of methane. Liquid samples showed
essentially quantitative conversion of the GB agent to complete (oxidation
products, i.e., hydrofluoric and phosphoric acids. Small amounts (<400
ppm) of methyl phosphonic acid (MPA) and acetone were detected in the
samples from the 450eC (842'F) reactor temperature with significantly less
detected in the samples from the 500 and 550eC (932°F and 1,022°F) reactor
conditions. Table 4.8 shows the analytical results for the GB test series. '
4.7.2.2 VX Agent Treatment
VX agent bench-scale testing commenced on June 29, 1993. Six separate
tests were performed, investigating temperatures of 450 to 550°C (842eF to
l,022°F)(see Table 4.9). As with the GB test series, all tests were performed
at agent concentrations of approximately 1% (by weight) with 100% excess
oxygen. Typically, four liquid and two gas samples were taken for each test.
Additionally, 4-5 Minicams* analyses of the gaseous effluent were per-
formed during the course of testing. No agent was detected in any liquid
samples, signifying destruction in excess of 99.99999%. Higher destruc-
tions may have been achieved, but 99.99999% is the maximum that can be
measured given an influent agent concentration of 10,000 ppm (1% by
weight) and the 1 ppb detection limit for VX in liquid samples. No agent
above the allowable exposure limit for VX was detected during on-line
Minicams* analysis of the SCWO gaseous effluent.
Gas and liquid samples were sent to IGT and UTPRC, respectively, for fur-
ther analysis following verification by HTRI personnel of the absence of detect-
able agent. The VX agent was essentially quantitatively converted during test-
ing to sulfuric and phosphoric acid. Small quantities of transformation products
4.47
-------
Supercritical Water Oxidation
such as acetic acid and acetone were observed. Gas analyses showed the pres-
ence of N2O and very low concentrations of NOx and SO^. The results of gas
and liquid analyses are summarized in Table 4.10.
Table 4.8
Analytical Results for GB Agent Bench-Scale Tests
Liquid
Analysis
Gaseous
Analysis
Component
Fluoride
Phosphate
MPA
Acetone
CO
HZ
Methane
Unit of
Measure
ppm
ppm
PP01
ppm
vppm
vppm
vppm
ajo
' /
Reaction Temperature
450'C
1431
6524
122
385
BDL
BDL
BDL
500'C
1719
7785 £
83
106
BDL
BDL
300
550'C •
^'1216
5683
48
<2
BDL
BDL
100
BDL below detection limit
vppm parts per million by volume
•ORE of >99.99999% achieved for all test samples.
Affluent data are averaged for multiple runs performed at the same temperature.
Test No.
1
2
3
4
5
6
vx
Pressure
(psig)
4000
4000
4000
4000
4000
4000
Table
4.9
Agent Bench-Scale Test
Temperature Total Flow Rate
CC) (mL/min)
500
450
500
550
450
550
505
39.4
505
495
39.4
505
Matrix
Residence Tune
(sec)
18
29
18
16
29
16
Test Duration
(min)
68
45
93
44
56
48
4.48
-------
Chapter 4
Table 4.10
Analytical Results for VX Agent Bench-Scale Testsa99.99999% achieved for all test samples.
bEffluent data are averaged for multiple runs performed at the same temperature.
4.7.2.3 Mustard Agent Treatment
Unlike GB and VX, mustard agent has a low solubility in water. GB and
VX agents were mixed with the water/oxidizer solution in line just upstream
of the reactor. They readily went into solution to yield a uniform, miscible
solution. Because of its low solubility, mustard agent would yield a
two-phase mixture upon feed to the SCWO reactor if fed in the same manner
as the VX and GB agents. The two-phase mixture could present operational
and performance difficulties in the compact test rig. Therefore, steps were
taken to solubilize the mustard agent by hydrolyzing in hot water, thereby
producing a uniform feed for SCWO testing.
A small test sample (2 g) was mixed with water in a 10% (by weight)
solution and heated. The mixture initially showed distinct water and mus-
tard agent phases, but after heating to 80-90'C (176-194T) for 7 minutes, it
was converted it to a single, clear, uniform phase. A hydrolysis apparatus
4.49
-------
Supercritical Water Oxidation
was assembled to allow controlled hydrolysis of larger samples. Hydrolysis
testing showed that mustard agent hydrolysis in water can be completed
within approximately 5 minutes at 80 to 100'C (176 to 212eF), if suitably
agitated. Hydrolysis at 60eC (140°F) requires approximately four times
longer. Following hydrolysis, the solution was cooled and stored for later
SCWO use.
Mustard agent testing began on April 5,1994. Five separate tests were
performed, investigating temperatures of 450 to 550*C (842 to l,022eF)(see
Table 4.11). As before, all tests were performed at agent concentrations of,
approximately 1% (by weight) with 100% excess oxygen. Typically, four
N liquid and two gas samples were taken for each test. Additionally, multiple/
Minicams® analyses of the gaseous effluent were performed during, tlie
course of testing.
Table 4.11
Mustard Agent Bench-Scale Test Matrix
Test No.
1
2
3
4
5
6
Pressure
(psig)
4000
4000
4000
4000
4000
4000
Temperature
CQ
450
450
550
500-525
500
500
Total Flow Rate
(mL/min)
33.3
31.3
aborted
30
32.5
31.5
Residence Time
(sec)
34
37
aborted
34
28
29
Test Duration
(rnin)
55
55
aborted
92
47
47
No agent was detected in any liquid samples, signifying destruction in
excess of 99.9999%. Higher destructions may have been achieved, but
99.9999% is the maximum that can be measured given an influent agent
concentration of 10,000 ppm (1% by weight) and the 10 ppb detection limit
for mustard agent in liquid samples. Also, except for Runs 3 & 4, where a
system upset resulted in agent contamination of the effluent collection lines,
4.50
-------
Chapter 4
no readings above the mustard agent AEL of 0.003 mg/m3 were detected.
During Run 3, an equipment malfunction necessitated test termination. Re-
sidual feed material was flushed at reduced pressure and temperature
through the reactor into the effluent collection lines, thus contaminating
them. This was not discovered until after the start of Run 4.
Gas and liquid samples were sent to specialized laboratories for further
analysis. The results are listed in Table 4.12. The data for runs performed at
the same temperature have, been combined and averaged. Meaningful chlqr/.
ride measurements could not be made because sodium chloride has preyf,'
ously been added to the wastes in order to stabilize them for jjtorage and
shipping. Low levels (<600 ppm) of the intermediate transformation prod-
ucts acetic and formic acids were observed at450°C (842eF), with less ob-
served at 500°C (932°F), and none observed at 525°C (977°F).melatively
high concentrations of CO were observed at lower temperature's, decreasing
to less than 2,000 ppm at 525°C (977°F). Higher operating temperatures
will even further reduce observed CO levels.
Table 4.12
Analytical Results for Mustard Agent Bench-Scale Testsab99.99999% achieved for all test samples.
"Effluent data are averaged for multiple runs performed at (he same temperature
cSeveral unknown, nonagent peaks of -50-100 ppm each were detected.
4.51
-------
Supercritical Water Oxidation
SOx levels of 1000 to 2600 ppm were observed during testing. A major
factor contributing to these relatively high levels is thought to be poor mix-
ing/mass transport limitations caused by the low flow rates and short lengths
of the test system. Even so, the sulfur present in the observed SOX repre-
sented only about 2% of the available sulfur, the remaining having been fully
converted to sulfate. Improved mixing in a redesigned pilot-plant reactor
should result in lower SOx levels.
The DARPA HTO (High Temperature Oxidation) pilot plant is designed .
to provide a transportable pilot-scale demonstration unit for extremely haz-> /
ardous wastes such as chemical warfare agents, as well as solid pjopellants
. and other U.S. Department of Defense (DoD) hazardous wastes. The system
is designed to provide a very flexible test bed. The maximum operating
pressure and temperature are 310 bar (4,500 psi) and 650°C (l,202f F). It has
a nominal flow rate of 3.79 L/min (1 gal/min) with a typical feefl concentra-
tion of 5% (by weight) organics and up to 12% (by weight) inert solids.
All of the high-pressure equipment, other than .the oxygen supply system
and the high-pressure water pump, are contained on the reactor skid. There
are numerous flanges hi the lined system to allow installation of instruments
and special test equipment. The system can be operated with or without heat
recovery and preheaters. The reactor skid is enclosed with polycarbonate
shielding to provide personnel protection from all high pressure components
and to contain the effluent in the event of a system rupture prior to discharge
to the facility ventilation system.
In the fall of 1996, this pilot plant was used to test the ability of SCWO to
treat hydrolized VX for the U.S. Army. The tests were considered successful
with destruction of organic compounds achieved to no measurable VX-thiol,
the most prevalent VX compound present in the hydrolysate. The detection
limit for these measurements is equivalent to 99.9999% destruction. Based
on the results of these tests in early 1997, the Army selected hydrolysis fol-
lowed by SCWO treatment of the hydrolysate as the method to be used to
treat VX, GB and mustard stockpiles around the U.S. The full-scale system
is currently being designed.
4.52
-------
Chapter 4
4.7.3 Pilot-Scale Studies
4.7.3.1 Hydrolyzed Rocket Propellent Treatment
For the Air Force, the prototype hydrolysis and HTO systems built by Gen-
eral Atomics were installed at a Thiokol site near Brigham City, Utah where
demonstration tests were recently completed. The Air Force prototype HTO
system installed at Thiokol has a maximum operating pressure of 310 bar
(4,500 psi), a maximum temperature of 650'C (1,202°F) and a nominal flow
rate of 1.59 L/min (0.42 gal/min). All of the high pressure equipment, othej/'•!''
than those associated with the oxygen supply system, are contained on the reac-
tor skid. Like the ARPA system, there are numerous flanges in the lined piping
to allow installation of instruments and special test equipment. Th,e reactor skid
is enclosed with polycarbonate shielding to provide personnel protection from
all high pressure components. A RCRA RD&D permit was oblaihed for the Air
Force system at Thiokol in a period of about six months.
Two test runs were conducted on effluent from the base hydrolysis of rocket
propellant over a wide range of operating conditions. The initial test processed
-25 pounds of double base hydrolyzed propellant as a 1% (by weight) hydro-
lyzed solution. The result of a continuous run exceeding 24 hours was total
organic carbon (TOC) destruction to below the detection limit
A second 695 Ib batch of hydrolyzed double base propellant at a concen-
tration up to 21% was processed during a continuous 34-hour run and dem-
onstrated reliable system performance over the range of temperatures from
450 to 580°C (842 to 1,076°F) and flow rates of 1.14 to 1.67 L/min (0.3 to
0.44 gal/min).
Effluent sample analysis confirmed pretest predictions of hydrolyzed
effluent treatment with TOC and NOx levels below the detection limit (1 and
5 mg/L, respectively) for operating temperatures in the 570 to 580°C (1,058
to 1,076'F) range at a pressure of 276 bar (4,000 psi).
Propellant throughput rates of up to 800 Ib/day (24-hr/day equivalent)
were demonstrated over several hour run times at several operating points.
Instantaneous propellant throughput rates, extrapolated to 24-hr/day opera-
tion, of 1,100 Ib/day were achieved. According to General Atomics, the
operator, with fuel additives, these rates could potentially be significantly
higher. No evidence of corrosion was found in the titanium-lined sections of
the preheater, reactor, and cool-down heat exchangers. The high solids
4.53
-------
Supercritical Water Oxidation
content feed was processed with stable pressure control and with no evi-
dence of plugging or erosion. The system operated within the requirements
specified by the RCRA RD&D permit obtained for the SRMD Prototype
Facility for processing of hazardous wastes.
4.7.3.2 Paper Mill Wastes Treatment
The University of Texas at Austin currently owns and operates an ap-
proximately 227.12 L/hr (60 gal/hr) capacity, continuous-flow SCWO sys-
tem (Blaney et al. 1995). Kimberly-Clark Corporation has performed nu- , /
merous experiments at this facility and has shown that paper mill sludges
can be converted (oxidized) to clean water, clean calcium carbonate ash, and
clean gas (carbon dioxide and residual oxygen). Experiments were rjer-
formed on de-inking sludge from a paper recycling operation. Treatment
results for virgin sludges are reported elsewhere (Hossain 1991$.
The paper mill wastes contained small traces of polychlorinated
dibenzodioxins and dibenzofurans (PCDD/PCDF). The SCWO system's
ability to destroy these compounds were of particular concern during these
tests. Two tests were conducted at the following SCWO reactor conditions:
• pressure 245 atm (3,600 psi)
• flowrate 75.71 L/hr (20 gal/hr)
• reaction temperature 450 and 500°C (842 and 9328F)
The overall organic destruction was quantified in terms of total organic car-
bon (TOC). The thoroughness of destruction of trace chlorinated organics was
quantified by measuring the trace quantities of PCDDs, PCDFs, and PCBs
before and after SCWO, using gas chromatography and high resolution mass
spectrometry following standard methods specified by US EPA (S W-846).
The major components in the pilot plant consisted of a high pressure dia-
phragm pump, a number of double pipe heat exchangers, an electric heater, a
coiled tube reactor, an air-driven oxygen booster, and a hydrocyclone. The
heat exchanger between the feed and reactor effluent allowed the recovery of
some thermal energy, reducing the electrical power input required to main-
tain the process. The oxygen flowrate, electrical power input to the heater,
and process pressure and temperature were controlled and monitored by a
computer. The feed flowrate was controlled by manually adjusting the
stroke length dial at the pump head and was monitored by the computer.
The reactor volume was 13.08 L (49.51 gal).
4.54
-------
Chapter 4
The feed was aqueous de-inking sludge collected at a paper recycling mill
before biological treatment. It contained 5 to 7% solids, the solids portion
being 30% inorganic (clays, fillers) and 70% organic (cellulosic fiber fines,
residual pulping and de-inking chemicals, and trace chlorinated organics).
The sludge pH was about 5.7, and total chlorides about 1.1 mg/L. The
sludge was diluted to 2.5% solids, and the diluted mixture, termed the "feed
sludge," was subjected to SCWO treatment. In order to increase the feed
sludge's heating value for the higher-temperature test, Experiment B, a small
amount of methanol was added to the feed holding tank and blended evenly'/'
throughout the feed sludge. Table 4.13 presents the feed sludge values fo/ '
total organic carbon (TOC), PCDDs, PCDFs, and PCBs on a dry basis, not
including the added methanol. {
i»
Table 4.13
Composition of Feed Sludge and Product Ash1
TOC
2,3,7,8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
OCDD
Total PCDD
2,3,7,8-TCDF
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
OCDF
Total PCDF
Feed Sludge2
440,000 ppm
3.99 ppt
6.41 ppt
<2.46ppt*
15.36 ppt
33.81 ppt
470.70 ppt
530 ppt
35.09 ppt
72.42 ppt
9.37 ppt
<2.97ppt*
12.06 ppt
15.19 ppt .
144 ppt
Expt A Product Asji3
3400 ppm
<0.325ppt'(DE>99%)
18 ppt
<2.5ppt'
<1.23ppt*
5.25 ppt
<7ppt*
23.3 ppt (DE>96%)
2.85 ppt (DE>99%)
63.5 ppt
<1.05ppt*
<0.8 ppt'
<1.19ppt*
<1.5ppt*
66.4 ppt (DE>96%)
Expt. B Product Ash3
<100 ppm
<0.23ppf(DE>99%)
1.5 ppt
<0.81ppt*
<0.445 ppt*
<0.64ppt*
1.6 ppt
3.1 ppt (DE>96%)
<0.505ppt'(DE>99%)
7 ppt
<0.275 ppt'
<0.34ppt*
<0.22ppt'
<0.53ppt*
7 ppt (DE>96%)
4.55
-------
Supercritical Water Oxidation
Table
Composition of Feed
Feed Sludge2
rnono-chlor-biphcnyl <0.32 ppb*
Di-Chl-Bp 8.65 ppb
Tri-CM-Bp <1.51 ppb*
Tetrm-Chl-Bp <0.65 ppb*
PenU-Chl-Bp <0.69 ppb*
Hex»-Chl-Bp <0.65 ppb*
Hepta-Chl-Bp <1 .29 ppb*
Octo-Chl-Bp <1.29ppb*
Nona-Chl-Bp <3.23ppb*
Deca-Chl-Bp <3.23ppb*
Total PCBs 8.65 ppb
TCDD tetrachlorodfcenzo-para-dioxin
PeCDO pentachlorodibenzo-para-dioxin
HxCDD hexachkwodtoanzo-para-dioxin
HpCDD heptachlorodibenzo-para-dioxin
PCDD polychlorinated dibenzo-para-dioxin
PCDF polychlorinated dibenzofuran. etc.
ppb parts per bilEon (ng/L)
ppt parts per trilion
'All data reported on dry basis
2Average of 4 analyses
^Average of 2 analyses
* Below detection limit
Reprinted with permission from Hutchenson and Foster,
Copyright 1995 American Chemical Society.
4.13 cont.
Sludge and Product
Expt A Product Ash J
46.5 ppb
20.5 ppb
<0.945ppb*
<0.4ppb*
<0.4ppb*
<0.4ppb*
<0.8 ppb*
<0.8ppb*
<2ppb*
<2ppb*
67 ppb (PCBs formed)
Innovations 'm Supercritical Fluids
Ashi
Expt. B Product Ash3
90%)
.Series 608, p 448.
The feed sludge was pressurized to approximately 245 atm (3,600 psig)
using a high-pressure diaphragm pump, and then heated with electrical heat-
ers to reach the desired reaction temperature. While the heating value of the
sludge was sufficient to maintain the reactor's temperature, poor insulation
of the pilot-scale test reactor required additional heat which was provided by
electrical heaters and a number of double-pipe heat exchangers recovering
energy from the processed reactor effluent.
4.56
-------
Chapter 4
Figure 4.9
SCWO Reactor Temperature Profile
I
550
500
450
400
J_
J_
I
I
4 8 12 16
Distance Along Reactor (m)
20
° Experjment A
+ Experiment B
Reprinted with permission from Hutchenson and Foster, Innovations in Supercritical Fluids, Series 608, p453.
Copyright 1995 American Chemical Society.
The influent flowrate was 75.71 L/hr (20 gal/hr). Reactor temperature
profiles for the experiments are shown in Figure 4.9. The residence time at
reactor conditions was about 50 seconds. Two sets of tests were conducted
at the following conditions:
Experiment A. Average reactor temperature was maintained at
approximately 450"C (842°F), all other variables as described
above.
Experiment B. Average reactor temperature was maintained at
approximately 500'C (932°F).
For Experiment B, 2% (by weight) methanol was mixed with the feed in
the influent holding tank; The methanol acted as an auxiliary fuel and its
added heat of combustion brought the reactor temperature to 500°C (932°F).
Since methanol is very easily oxidized to carbon dioxide and water at
SCWO conditions, it is assumed that adding this small amount of methanol
would only minimally affect the product mix.
4.57
-------
Supercritical Water Oxidation
Feed, product water, and product ash from both Experiment A and Ex-
periment B were analyzed within two weeks of the trials. Analyses were for
all congeners of PCDDs, PCDFs, and PCBs, Total Organic Carbon (TOC),
percent solids, and metals. The analyses were conducted at an independent
laboratory which utilized gas chromatography and high-resolution mass
spectrometry for trace chlorinated organies analyses following standard US
EPA procedures. In addition, numerous analyses of the aqueous phase for
suspended solids, chemical oxygen demand, acetic acid, ammonia, chlorides,
and pH were performed. Tests on the solid (ash) phase included volatile
solids and the Tpxicity Characterization Leaching Procedure (TCLP).
Data for the product streams are shown in Table 4.13 adjacent to the feed /
data. Values of PCDDs, PCDFs, and PCBs in the product water werejess man
one percent of that in the product solids (ash) so these data were not included.
For both experiments the majority of the PCDDs and PCDFs were^destroyed.
In Experiment B, at 500'C (932°F), the oxidation appeared to approach
completion, as the TOC in the solids was under the detection limit of 100
mg/L, and virtually all of the PCDDs and PCDFs were destroyed. Destruc-
tion Efficiencies (DEs) of 2,3,7,8- TCDD and 2,3,7,8-TCDF (widely ac-
cepted as the most toxic congeners) were over 99%, and DEs of total PCDDs
and PCDFs were over 96%. Over 90% of the PCBs were destroyed.
In Experiment A, the destruction efficiencies for PCDDs and PCDFs were
roughly the same as for Experiment B (>99%). However, at the lower tem-
perature of 450°C (842°F) in Experiment A, the data indicate that PCBs form
in the parts per billion level. In addition, TOC destruction was unacceptably
low. SCWO treatment reduced the TOC in the product ash to 3,400 mg/L
from 440,000 mg/L in the feed solids. Possible PCB formation was not
observed at the higher SCWO temperatures and TOC destruction was also
greater. The formation of PCB was clearly significant. No other literature
was found showing such results and the results were not replicated at the
higher temperatures. Site-specific testing is needed to confirm and expand
upon these findings. Until these results are confirmed or refuted it appears
prudent in those applications where chlorinated organic compounds are
treated to operate at temperatures above 500'C (932'F) and to test the efflu-
ent for PCB.
4.58
-------
Chapter 4
At first glance it may appear that some TCDDs were formed in experi-
ment A; this is not the case. A complete mass balance which takes into ac-
count the fact that the 18 ppt total TCDD in the product ash is based on inor-
ganic only (dry basis), whereas the 6.41 ppt total TCDD in the feed solids is
based on a mixture of inorganic plus organic, also on a dry basis.
The solid residues (ash) derived from the SCWO tests were characterized by
the Toxicity Characteristic Leaching Procedure. Concentrations of the metals in
the leachate from the ash were lower than the regulatory levels set by the US
EPA. Most heavy metals, including As, Cd, Cr, Hg, Ni, Pb, Se, Tl, and V, were'
nonleachable (below the detection limit of 0.0005 mg/g solid).
Supercritical water oxidation at temperatures of 500°C (9i&0F), pressures
of 245 atm (3,600 psi), and a residence time of about 50 sec, was fehowri to
be effective in destroying over 99% of the most toxic dioxin-ty^e congeners,
2,3,7,8-TCDD and 2,3,7,8-TCDF, over 96% of the total PCDD/PCDFs, and
over 90% of the PCBs. However, at a lower temperature of 450'C
(842°F)(other conditions remaining constant), destruction of chlorinated
organics was not as thorough, and PCBs may have actually formed and sur-
vived for a short time.
4.8 Conclusion
SCWO appears to be a technology that is reaching commercial scale.
Destruction efficiencies for organic materials equivalent to those achieved by
incinerators have been demonstrated and, sufficient knowledge appears to
have been accumulated to scale (at least some SCWO system designs) to the
10 gallon per minute range which the authors generally consider to be com-
mercial scale. The cost of treatment is still projected to be somewhat higher
than for incineration so that the initial applications will, most likely, be for
the treatment of specialized wastes, such as rocket propellant or chemical
warfare agents which (because of technical or social objections) cannot be
incinerated. It appears likely that additional experience will reduce the cost
and this technology should be considered for those applications where the
waste includes aqueous streams contaminated with high concentrations of
organic compounds.
4.59
-------
r f
-------
Chapter 5
EX-SITU HIGH VOLTAGE
ELECTRON BEAM TREATMENT
5.7 Introduction
r
The electron beam (E-beam) technology is a means of treating wastewaters
and groundwaters contaminated-with organic compounds. The technology was
developed by High Voltage Environmental Applications, Inc. (HVEA) of Mi-
ami, Florida, in conjunction with Florida International University and the Uni-
versity of Miami. A full-scale system utilizing the E-beam technology was
installed at the Miami-Dade Water and Sewer Authority Wastewater Treatment
Plant in Key Biscayne, Florida in approximately 1983 as a means of sterilizing
the sewage sludge from the wastewater treatment plant (Kurucz, Waite, and
Cooper 1995). No longer used for this purpose, it serves as an experimental
unit for research and treatability studies. This fixed system has a capacity of
460 L/min (120 gal/min) and uses a 1.5 MeV, 50 mA (75 kW) electron accel-
erator as the radiation source for treatment.
HVEA operates a second, mobile pilot-treatment system (model
M25W-48S) with a capacity of 19 to 190 L/min (5 to 50 gal/min) which has
been used for a number of demonstrations in the U.S. and Europe. This
system uses an electron beam with a maximum power output of 25 kW (US
EPA 1995d). The majority of the information presented is based on
pilot-scale demonstrations using this mobile system.
A bench-scale batch treatment system with a capacity of approximately
two gallons of wastewater is also available for treatability studies. The
bench-scale system uses gamma as a source of electrons for the beam pro-
duced by ^Co jacketed in stainless steel. The ^Co produces, on decay, one
beta particle and two gamma rays. The stainless steel is constructed to stop
5.1
-------
Ex-Sliu High Voltage Electron Beam Treatment
the beta particles, allowing only the highly-penetrating gamma rays to es-
cape into the surrounding medium and irradiating the material to be treated
(US EPA 1995d; Kalen 1992).
5.2 Process Description
High voltage electron beam treatment (HVEBT) treats aqueous streams ^ ;
contaminated with organic constituents or with pathogens by irradiating the ''
stream with a high-energy electron beam. Treatment is conducted & normal /
* temperature and pressure. Figure 5.1 is an elevation drawing of the 75 k-W
facility in Key Biscayne, Florida. Physically, the system is simple. Tftere are
no moving parts, except for standard pumps. The aqueous stream is passed
over a weir (influent spreader in Figure 5.1) which converts it intoli cascading
flat curtain of water. A relatively thin water flow is necessary since the electron
beam's treatment efficacy is reduced as the thickness of the water increases. An
electron gun similar to the electron gun found in common cathode ray tubes
(i.e., television tubes) rapidly scans across the flat curtain of water.
The high-energy radiation sterilizes the stream and destroys organic contami-
nants by chemical redox mechanisms. Inorganic constituents in the aqueous
stream can also be chemically modified, although the data on the behavior of
inorganic constituents is not well established. Organic chemical destruction
efficiencies exceeding 99% have been achieved. In cases of high organic con-
centrations in the influent, multiple passes through the electron beams might be
required to achieve the desired effluent standards. The irradiation chamber and
electron beam source used for treatment must be shielded, but the treatment
leaves no residual radioactivity in the aqueous stream.
The interaction between the electron beam and the waste stream being
treated raises the temperature of the waste less than five degrees. It appears
unlikely, that under normal operating conditions, interaction of the electron
beam with ambient air results in the formation of ozone and other trace con-
taminants (O, CO, and NO2). However, the small air stream flowing across
the titanium window of the electron gun requires some form of treatment
prior to discharge, mainly for the catalytic destruction of O3. Since the vol-
ume of air used to cool the titanium window is relatively low, E-Beam treat-
ment systems would require an air pollution abatement system much smaller
than those typically utilized by competing technologies.
5.2
-------
Figure 5.1
Elevation of the Electron Beam Research Facility, Key Biscayne, Florida
i
Vault Exhaust Fan
Window Exhaust Fan
Influent Spreader
\ Scanner
5MeV50mA
ICT Electron
Accelerator
Vault Exhaust Duct
Reproduced courtesy of High Voltage Environmental Applications, Inc.
o
Q
•
-------
Ex-Sltu High Voltage Electron Beam Treatment
A potential concern in the use of this treatment method is that the simple
flow of the aqueous stream across an open weir can cause a release of vola-
tile contaminants such as trichloroethylene (TCE) and trichloroethane (TCA)
into the air within the treatment vessel. This air is normally contained in the
treatment chamber; however, during the US EPA-SITE demonstration (US
EPA 1995d) of this technology it was found that the cooling air stream flow-
ing past the titanium window isolating the electron beam source was leaking
into the treatment chamber and trace amounts of the volatile organic com-
pounds in the wastewater stream were being removed by stripping. The
problem was traced to the waste delivery system which allowed waste
.stream interaction with the cooling air. Interaction between the halogenated
VOCs with the ozone and molecular oxygen in the cooling air, and with the
electron beam, released the VOCs and, in addition, formed small amounts
(low ppm) of HC1 and phosgene. | '
This problem, although environmentally not significant because of the
low flow rates and low contaminant concentrations of the leaking stream, has
been completely eliminated in newer designs in which, the waste delivery
system uses a completely enclosed (proprietary) means to distribute the flow
of waste. The new design also isolates the waste from the electron beam
source by a second titanium window so that the gas stream that cools the
electron source never comes in contact with the waste stream. The new
design also recirculates the cooling gas stream, so that the ozone produced
by the electrical discharge, is recirculated. According to the vendor, these
design modifications eliminate the need for added air pollution control
equipment in almost all cases.
Table 5.1 compares the E-Beam technology against a number of alterna-
tive technologies for treating water contaminated with low molecular weight
organic compounds.
5.3 Scientific Principles
The high energy electron beam passing through the aqueous stream
causes atoms in its path to achieve highly-excited electron states and form
free radicals. Organic destruction occurs because the electron beam, when it
interacts with water, generates both powerful oxidizing radicals such as the
5.4
-------
Chapter 5
hydroxyl radical (»OH) as well as reducing radicals.such as the aqueous
electron (e"tq) and the hydrogen radical (H»)(Spinks and Woods 1990).
These free radicals react with other constituents in the water. Unlike photo-
chemical reactions where one photon of light initiates one (molecular) reac-
tion, a high energy electron is capable of initiating several thousand reac-
tions as it dissipates its energy (Spinks and Woods 1990).
Table 5.1
Comparison of Technologies for Treating VOCs in Water
Technology
Advantages
Disadv intages
Air Stripping
Steam Stripping
Air Stripping with Carbon
Adsorption of Vapors
Air Stripping with Carbon
Adsorption of Vapors and
Spent Carbon Regeneration
Carbon Adsorption
Biological Treatment
Chemical Oxidation
E-Beam System
Effective for high concentrations;
mechanically simple; relatively
inexpensive
Effective for all concentrations
Effective for high concentrations
Effective for high concentrations;
no carbon disposal costs; product
can be reclaimed
Low air emissions; effective for
high concentrations
Low air emissions; relatively
inexpensive
No air emissions; no secondary
waste; VOCs destroyed
No secondary waste; multiple
mechanisms for VOC destruction;
no chemicals (such as O3 orH2Oj)
required
Inefficient for low concentrations;
VOCs discharged to air
VOCs discharged to air, high
energy consumption*
Inefficient for low concentrations;
requires disposal or regeneration of
spent carbon; relatively expensive
Inefficient for low concentrations;
high energy consumption
Inefficient for low concentrations;
requires disposal or regeneration of
spent carbon; relatively expensive
Inefficient for high concentrations;
slow rates of removal; sludge
treatment and disposal required
Not cost-effective for high
contaminant concentrations;
may require chemicals such as
O3 andH2O2
High electrical energy
consumption; not cost-effective for
high contaminant concentrations;
relatively expensive"
Vendor adds that stream stripping does not destroy the contaminant but only removes it from the wastewater into a
second stream that then must be disposed.
**This conclusion is given in the US EPA report, but it is questioned by the vendor.
Source: US EPA 1995d
5.5
-------
Ex-SItu High Voltage Electron Beam Treatment
The efficiency of conversion of a high energy electron beam to a chemical
process is defined as G, which is the number of chemically active moities
(radicals, excited atoms, or other reactive products) formed or lost in a sys-
tem absorbing 100 electron volts (eV) of energy absorbed. Because water is
by far the predominant molecule found in any system likely to be treated, it
will form the predominance of reactive moities in a typical system. How-
ever, data collected on pure water spiked with selected contaminants will not
be generally applicable to actual contaminated waters because other con-
stituents in the influent stream play a significant role in the chemical reaction' <
by scavenging free radicals. Carbonates, iron, other inorganic compounds', /
and natural and synthetic organic compounds (other than target materials)
are examples that may affect destruction efficiency by acting as radical scav-
engers. As a result, experimental data used for scale-up must be qbtained
from actual site samples, rather than by spiking a readily available! clean
water stream. ?
Because of the aggressive nature of the free radicals and other reactive
moities formed by the electron beam, the rates of reactions are rapid. Rate
constants (pseudo first-order) for the chemical reactions between the (e'aq),
(H>), and (OH«) free radicals and a variety of organic compounds are in the
range of 107 to 1010 moHsec-1 (Spinks and Woods 1990).
5.4 Potential Applications
The greater the initial concentration of the organic contaminants in the
influent, the greater the dose of high-energy electrons that are required to
achieve a given effluent concentration. Therefore, the costs associated with
treatment increase substantially as the concentration increases.
The technology is generally applicable to aqueous streams or flowable
slurries (3-5% solids). Furthermore, the systems have a high tolerance for
suspended solids (Cooper et al. 1992). In general, materials handling con-
straints and the high electron energies that would be required to penetrate
beds of solid materials (e.g., soils) limit the system's usefulness to flowable
aqueous matrices.
5.6
-------
Chapter 5
The greater the organics loading of the influent stream, the greater the
quantity of radiation required to achieve a level of treatment and, hence, the
greater the energy cost. As a result, cost of treatment increases rapidly as the
organic concentration in the influent stream approaches approximately 1%.
High voltage electron beam treatment has been demonstrated to success-
fully destroy a wide variety of organic compounds dissolved in waters from
many different sources. It has successfully treated chlorinated organic com-
pounds (US EPA 1995d). Recently, bromate ions were reduced to bromide
ions (Siddiqui et al. 1996) in drinking water, Table 5.2 displays some matri-/
ces and compounds that have been treated using this technology (Kurucz'eV
al. 1991b). The table also gives the radiation doses used to achieve this re-
moval. It is noted that recent work has demonstrated higher removal effi-
ciencies for many of these compounds.
-------
Ex-Sltu High Voltage Electron Beam Treatment
Table 5.2
Summary of Percent Removal of Various
Compounds
Drinking Water
Chloroform
Bromodichlorom ethane
Dibromochloromethane
Bromoform
Wastewater/Groundwater Treatment
Carbon Tetrachloride
Trichloroethylene (TCE)
Tetrachlorethylene (PCE)
Trans-1 ,2-Dichloroethene
Cis- 1 ,2-Dichloroethene
1 , 1 -D ichloroethene
1 ,2-Dichloroethane
Hexachloroethane
1 , 1 ,1 -Trichloroethane
1 , 1 ,2,2-Tetrachloroe thane
Hexachloro-1.3-Butadiene
Methylene Chloride
Groundwater Treatment
Benzene
Toluene
Chloro benzene
Ethylbenzene
1,2-Dichlorobenzene
1 ,3-Dichlorobenzcne
1 ,4-Dichlorobenzene
m-Xylene
o-Xylene
Dieldrtn
Total Phenol
by Treatment Application
Percent
Removal
83
>99
>99
>99
>99
>99
>99
93
'98
>99
60
>99
89
88
98
77
>99
97
97
92
88
86
84
91
92
>99
88
Organic
Area
Required Dose
(Krads)
I
650
80
80
•*> 80
I
i>50
fsoo
$ 500
800
800
800
800
800
650
650
800
800
650
650
650
650
650
650
650
650
650
800
800
These tests were conducted at 120 gal/min and one pass only. Larger electron beam dosages would result in a greater
destruction of the organic compound.
Reprinted from Advances in Nuclear Science and Technology, Volume 22, Kurucz et al., p 36,1991 with permission of
Plenum Press.
5.8
-------
Chapter 5
It should be noted that organic compounds that are ad- or absorbed by grit
or other solids in the wastewater will probably not be subjected to the high-
energy electron beam, since the inert portion of the solid will absorb some of
the beam. As a result, ad- or absorbed contaminants will, most likely, not be
destroyed to the same level as will dissolved organics. The vendor claims to
have demonstrated acceptable organics destruction on waters containing up
to 3% solids.
Waste streams containing concentrations of organic in excess of 1%, or
which contain organic materials in suspension, should be pretreated usingj /'
physical means. Possible pretreatments include gravity separation (oil-water
separators), floatation, or flocculation. Because the accelerated electrons
actually convert the water into a reagent, this process works best when it
only has to destroy dissolved organic compounds. ^
Posttreatment requirements for the treated effluent from t$is process de-
pend on the nature of the feedwater and of the discharge requirements for the
site. Even if the treated effluent from this process does not meet the
site-specific discharge requirements, the electron beam process will, in most
cases, "soften" the organic compounds, thus making the stream amenable to
biological treatment.
5.6 Design
5.6.1 Design Basis
The key parameter that must be considered in the design of an E-beam
treatment system is the energy required to destroy the contaminant in ques-
tion to the discharge limit. Some guidance on this matter is given in Table
5.1; however, it is necessary to conduct treatability studies in order to estab-
lish the necessary electron beam energy and the type and level of posttreat-
ment that might be required. The system's inherent simplicity makes such
treatability studies at all scales relatively inexpensive to conduct. The exist-
ing pilot-scale unit, for which a large amount of scale-up data is available, is
mounted on a trailer and it can be readily moved to the site. Treatability
studies using the pilot-scale system have been successfully conducted at a
number of sites.
5.9
-------
Ex-Sltu High Voltage Electron Beam Treatment
5.6.2 Design and Equipment Selection
Two categories of equipment are needed for this process — water transfer
and distribution equipment and irradiation equipment. The water transfer
and distribution equipment are simply pumps, a waste delivery system, and
other standard water handling equipment. These are readily available or they
can be quickly fabricated. The irradiation equipment consists of standard
electron beam guns that can also be readily purchased or fabricated. The
electron beam equipment used for the full-scale system was built in the earlyv, f
1970s and is still functional; testimony to its reliability and durability. , / /
j>
5.6.3 Process Modification f •
The only parts of the process itself that can be modified are the^lectron
source and the water distribution equipment. The electron source tan either
be an electron gun or a radioactive beta particle source. Cobalt-60, which is
radioactive, has been used as an electron (beta-particle) source in the
laboratory-scale system, but its use for commercial-scale systems is not
economically viable, according to the vendor.
The design of the waste delivery system and the thickness of the sheet of
water cascading from the weir is another possible process modification.
Reducing the depth of water through which the electrons must penetrate
improves the fractional destruction of the organics, although it will not nec-
essarily improve the overall system performance, since the electron utiliza-
tion rate will decrease. The optimum configuration is best established on the
.basis of treatability studies.
The removal efficiency for the E-Beam technology increases with mul-
tiple passes of the wastewater through the system. A modification of the
process that involves changes in the operating procedure is to recycle some
or all of the effluent from the process. For total recycle, using the same
equipment would require batch treatment of the wastewater. Two holding
tanks, one feeding the E-beam system and the second holding the treated
effluent, and then reversing their rolls would be needed for such an applica-
tion. Alternatively, multiple E-beam systems could be assembled in series or
in parallel, depending on whether the application required high levels of
irradiation (and hence, high destruction) or high flow rates. Another modifi-
cation for a continuous flow system is to recycle a portion of the effluent.
5.10
-------
Chapter 5
Using this procedure, removal efficiencies (REs) could be improved without
the use of multiple electron beam generators and their inherent energy costs.
5.6.4 Prefreatment Processes
Waste streams containing concentrations of organic in excess of 1%, or
which contain organic materials hi suspension, should be pretreated using
physical means. Possible pretreatments include gravity separation (oil-water
separators), floatation, or flocculation. Because the irradiation process actu-
ally converts the water into a reagent, this process works best when it only'
has to destroy dissolved organic compounds. ^
/
5.6.5 Posttreatment Processes ,;
Posttreatment requirements for the treated effluent from tfiis process de-
pend on the nature of the feedwater and of the discharge requirements for the
site. Even If the treated effluent from this process does not meet the
site-specific discharge requirements, the electron beam process will, in most
cases, "soften" the organic compounds, thus making the stream amenable to
biological treatment.
5.6.6 Process Instrumentation and Control
The processing instrumentation and controls consist of a voltage or
current regulator to maintain the electron beam at a constant power, and
controls for the electromagnets which are used to cause the electron
beam to scan in a controlled pattern across the water flowing through the
waste delivery system. This equipment is conceptually identical to the
electron beam controls of a simple cathode ray tube, but operate at
higher current levels. This control equipment has been commercially
available for well over fifty years.
5.6.7 Safety Requirements
The irradiation equipment is a source of ionizing radiation. As a result,
appropriate shielding must be incorporated into the design. Fortunately,
there is no lingering radiation in the "hot" radiation areas of the system;
radiation only occurs when the electron beam is operating. When the beam
is off, there is no risk of radiation exposure to personnel entering the irradia-
tion chamber.
5.11
-------
Ex-SItu -High Voltage Electron Beam Treatment
Interlocks must be installed on the access doors to the irradiation chamber
to prevent entry when the electron beam is energized. Also, an emergency
shut-off switch must be installed in the irradiation chamber to allow anyone
who is inadvertently trapped within to deactivate the system.
5.6.8 Specification Development
The key requirements that must be incorporated in specifications for an
electron beam treatment application depend on whether the bids are for
equipment, which is to be installed by others, or for a turnkey electron beany
system with performance guarantees. *
/
If a vendor is to provide a turnkey system, then the performance guarantees
must be based on quah'ty of the water to be treated. Competing venders should
be supplied with the complete results" of all treatability studies conducted.
Specification of individual equipment to be installed is fairly straightfor-
ward. However, the special equipment for this process is covered by patents
held by the developer and, in general, can only be acquired through this
sole-source.
It is important to remember that this process employs the application of
high voltage at a large scale and relatively high power under highly humid
conditions. Therefore, evaluation of vendor bids should include an evalua-
tion of the vendors' qualifications and experience as well as a comparison of
the cost data submitted.
5.6.9 Cost Data
The following analysis presents cost information for using the HVEA
E-beam technology to treat groundwater contaminated with VOCs. Cost
data were compiled during the Superfund Innovative Technology Evalu-
ation (SITE) demonstration at the Savannah River Sits (SRS) and from
information obtained from independent vendors and HVEA. -Costs are
presented in February, 1995, dollars and are considered to be
order-of-magnitude estimates with an expected accuracy within 50%
above and 30% below the actual costs.
Two models, based on different groundwater characteristics, are presented
and compared. In Case 1, the groundwater has an insignificant level of alka-
linity (<5 mg/L as CaCO3) and contains VOCs that are easy to destroy using
5.12
-------
Chapter 5
free radical chemistry. In Case 2, the groundwater has moderate-to-high
alkalinity (500 mg/L as CaCO3) and contains additional VOCs, a few of
which are more difficult to destroy. In Case 1, a 21 kilowatt (kW) system is
used to treat groundwater at 150 L/min (40 gal/min); in Case 2, the same
system is used to treat the groundwater at 75 IVrnin (20 gal/min).
Tables 5.3 and 5.4 present the costs compiled in this analysis for Case 1
and Case 2, respectively. Additional analysis is provided in these tables that
compares the costs of addressing both with a 45 kW system and a 75 kW ., f
system. In Case 1, the 45 kW system treats groundwater at 300 L/min (8,0
gal/min), and the 75 kW unit treats it at 490 L/min (130 gal/min). In Case 2,
the 45 kW system treats groundwater at 150 L/min (40 gal/nun), and the 75
kW unit treats it at 250 L/min (65 gal/min). , !
Site-specific factors can affect the costs of using the E-beanJ treatment
system. These factors can be divided into the following two^categories:
waste-related factors and site features.
Waste-related factors affecting costs include waste volume, contaminant
types and levels, treatment goals, and regulatory requirements. Waste vol-
ume affects total project costs because a larger volume takes longer to
remediate. However, economies of scale are realized with a larger volume
project when the fixed costs, such as those for equipment, are distributed
over the larger volume. The contaminant types and levels in Hie groundwa-
ter and the treatment goals for the site determine:
• the appropriate E-beam treatment system size, which affects
capital equipment costs;
• the flow rate at which treatment goals can be met; and
• periodic sampling requirements, which affect analytical costs.
Regulatory requirements also affect permitting costs and effluent moni-
toring costs.
Site features affecting costs include groundwater recharge rates, ground-
water chemistry, site accessibility, availability of utilities, and geographic
location. Groundwater recharge rates affect the time required for cleanup.
Groundwater alkalinity may increase or decrease E-beam technology REs
depending on the contaminant involved. Site accessibility, availability of
utilities, and site location and size all affect site preparation costs.
5.13
-------
Ex-SItu High Voltage Electron Beam Treatment
§5
Ouf
O D
O-J
O ^
Q> O
SuJuj
£
-------
I 1
Cn
Utilities0
Effluent Treatment and Disposal0
Residual Waste Shipping and Handling0
Analytical Services0
Equipment Maintenance0
•Site Dcmobilizationb
Total One-Time Costs"
Total Annual O&M Costs0
Groundwater Remediation
Total Costs**-'
Net Present Value*
Costs per 1,000 galh
25,700
0
6,000
24,000
25,300
15,000
1,057,600
92,700
2,764,000
1,626,600
5.16
52,600
0,
6,000
24,000
36,200
15,000
1,472,600
130,500
2414,400
1,963,700
6.23
87,500
0
6,000
24,000
43,000
15,000
1,718,600
172^00
2427,900
2^23,400
7.06
•Costs are in February 1995 dollars
fixed costs
c Annual variable costs
"Fixed and variable costs combined
•Future value using annual inflation rate ol 5%
'To complete groundwater remediation, it is assumed that the 21 kW unit will take 15 years, the 45 kW unit will take 7.5 years, and the 75 kW unit will take 4.6 years to treat 315 million gal
'Annual discount rate ol 7.5%
"Net present value
Source: US EPA 1995C *>sr>- h
-------
Cn
Table 5.4
Costs Associated with the E-Beam Technology — Case 2°
(Alkalinity 500 mg/L as CaCO3 — Organics (See Section 5.6.9.1 Assumptions))
Treatment System Configurations in Kilowatts (kW)
21 kW (20 gaVmin)
Cost Categories
Site Preparation1"
Administrative
Treatment Area Preparation
Treatability Study and System Design
Permitting and Regulatory b
Mobilization and Startup b
Transportation
Assembly and Shakedown
Equipment11
Laborc
Supplies*
Disposable Personal Protective Equipment
Fiber Drums
Sampling Supplies
Itemized
($)
35,000
117,600
23,000
10,000
10,000
600
100
1,000
Total
($)
175,600
5,000
20,000
842,000
10,000
1,700
45kW(40galAnin)
Itemized
($)
35,000
161,600
23,000
10,000
15,000
- "•**.
600
100
1,000
Total
($)
219,600
5,000
25,000
1,208,000
10,000
«.£, 1,700'
~
• *». ^
75kW(65gaIAnin)
Itemized
($)
35,000
183,600
23,000
10,000
15,000
600
100
1,000
Total
($)
241,600
5,000
25,000
1,432,000
10,000
1,700
-------
Utilities0
Effluent Treatment and Disposal0
Residual Waste Shipping and Handling0
Analytical Services0
Equipment Maintenance0
Site Demobilization11
Total One-Time Costsb
Total Annual O&M Costs0
Groundwater Remediation
Total Costs***
Net Present Value*
Costs per 1,000 gal"
°x
25,700
0
.6,000
24,000
25300
15,000
1,057,600
92,700
6,281,600
2,472^00
7.85
-
X
»
,t
52,600
0
6,000
24,000
' 36,200
15,000
1,472,600
130^00
3,994,600
2,350,700
7.46
' 1
£_
t •'
>
87^00
0
6,000
24,000
43,000
15,000
1,718,600
172^00
3,547.200
2,618,100
8.31
•Costs are in February 1995 dollars
"Fixed costs
c Annual variable costs
"Fixed and variable costs combined
•Future value using annual inflation rate of 5% *
•To complete groundwater remediation, it is assumed that the 21 kW unit will take 30 years, the 45 kW unit wHI take 15 years, and the 75 kW unit will tako 9.3 years to treat 315 million gal
"Annual discount rate of 7.5%
"Net present value
Source: US EPA 1995c
***»,
-------
Ex-SItu High Voltage Electron Beam Treatment
5.6.9.1 Assumptions
The assumptions used for this analysis of E-beam technology costs are
based on information provided by HVE A and observations made during the
SITE demonstration.
Site-specific assumptions include the following:
• for Case 1, the contaminants and their average concentrations are
TCE at 28,000 ug/L and PCE at 11,000 ug/L in groundwater
which has an insignificant alkalinity of <5 mg/L as CaCO3;
'/
• for Case 2, some of the additional contaminants are saturated
VOCs that are relatively difficult to treat. These, VCM?s are /
1,1,1-TCA, 1,2-DCA, chloroform, and CC14; their conceritrations
range from 370 to 840 ug/L. The other additional contaminants
are BTEX compounds present at concentrations ranging from
200 to 550 ug/L;
• the site is a Superfund site located near an urban area. As a re-
sult, utilities and other infrastructure features (for example, ac-
cess roads to the site) are readily available;
• the site is located in the southeastern United States. This region
has relatively mild temperatures during the winter months;
• contaminated water is located in an aquifer no more than 100 ft
below ground surface; and
• the groundwater remediation project involves a total of 1,200
million L (315 million gal) of water that needs to be treated.
This groundwater volume corresponds to the volume treated by a
21 kW unit operating continuously for 15 years at a flow rate of
150 L/min (40 gal/min).
Equipment assumptions include the use of the 21 kW unit treating con-
taminated groundwater at a rate of 150 L/min (40 gal/min) in Case 1 and 75
L/min (20 gal/min) in Case 2. The system is operated on a continuous flow
cycle, 24 hours per day, 7 days per week. The system can, therefore, treat
nearly 79 million L/yr (21 million gal/yr) in Case 1, and about 40 million IV
yr (10.5 million gal/yr) in Case 2. Because most groundwater remediation
projects are long-term projects, about 1,200 million L (315 million gal) of
5.18
-------
Chapter 5
water are assumed to be treated in both cases. Case 1 remediation will take
about 15 years to complete, and Case 2 about 30 years. In practice, it is
difficult to determine both the volume of groundwater to treat and the actual
duration of a project.
Neither depreciation nor salvage value is applied to the costs presented
because the equipment is not purchased by a customer. All depreciation and
salvage value is assumed to be incurred by the vendor and is reflected in the
ultimate cost of leasing the E-beam treatment equipment.
. '•> F
Operating parameter assumptions using a 21 kW system are listed belqw:
• costs for 45 kW and 75 kW systems are presented4n Tables 5.3
and 5.4; , { '•
• the treatment system is operated 24 hours per day, V^ays per
week, 52 weeks per year; £ '
• the treatment system operating at full power has a maximum
voltage of 500 kV and a maximum beam current of 42 mA.
• the treatment system operates automatically without the constant
attention of an operator and will shut down in the event of a mal-
function;
• modular components consisting of the equipment needed to meet
treatment goals are mobilized to the site and assembled by the
contractor;
• air emissions monitoring is not necessary; and
• E-beam equipment will be maintained by the contractor and will
last for the duration of the groundwater remediation project with
proper maintenance.
Total costs are presented as future values based on the following financial
conditions. The costs per 1,000 gal (3,800 L) treated are presented as net
present values and assumes a 5% annual inflation rate to estimate the future
values. The future values are presented as net present values using a dis-
count rate of 7.5% (using a higher discount rate makes the initial costs
weigh more heavily). Because the costs of demobilization will occur at the
end of the project, the appropriate future values of these costs were used to
calculate the totals at the bottom of Tables 5.3 and 5.4.
5.19
-------
Ex-SItu High Voltage Electron Beam Treatment
Additional assumptions include:
• costs are rounded to the nearest $100;
• contaminated groundwater is treated to achieve the removal effi-
ciencies (REs) observed in SITE demonstration Runs 3 and 13
for Cases 1 and 2, respectively;
• the E-beam system is mobilized to the remediation site from
within 500 miles of the site;
»t
• operating and sampling labor costs are incurred by the client. , /
The vendor performs maintenance and modification activities
that are paid for by the client; '.
• initial operator training is provided by the vendor; and,/
• four groundwater extraction wells already exist on-site! They are
assumed to be capable of providing the flow rates discussed in
this economic analysis.
Cost data are presented for the following categories:
• site preparation;
• permitting and regulatory;
• mobilization and start-up;
• equipment;
• labor;
• supplies;
• utilities;
• effluent treatment and disposal;
• analytical services;
• equipment maintenance; and
• site demobilization.
Each of these categories is discussed below.
5.6.9.2 Site Preparation Costs
Site preparation costs include administrative, treatment area preparation,
treatability study, and system design costs. For this analysis, site preparation
5.20
-------
Chapter 5
and administrative costs, such as those for legal searches, access rights, and
site planning activities, are estimated to be $35,000.
Treatment area preparation includes constructing a shelter building and
installing pumps, valves, and piping from the extraction wells to the shelter
building. The shelter building needs to be constructed before mobilization
of the E-beam system. A 37 m2 (400 ft2) building is required for the 21 kW
system. The 45 kW system requires 74 m2 (800 ft2) of building space, and
the 75 kW system requires 92 m2 (1,000 ft2). Construction costs are esti-
mated to be $l,184/m2 ($110/ft2), which covers installation of radiation ','"''
shielding materials. A natural gas heating and cooling unit and related '/
ductwork is estimated at $20,000 installed. The total shelter Building con-
struction costs for the 21 kW system are estimated to be $64,000. {
Four extraction wells are assumed to exist on-site which are'located 61 m
(200 ft) from the shelter building. Four 132 L/min (35 gal/rnjn), 1.5 hp,
variable-speed Teflon* pumps are required to maintain the flow rates neces-
sary for each case. The total installed cost for the pumps, including electri-
cal equipment, is $5,600. Piping and valve connection costs are $20/m ($6/
ft), which covers underground installation. The total piping cost is $48,000.
Thus, total site preparation cost is estimated to be $117,600.
HVEA estimates the treatability study to cost $18,000, including labor
and equipment costs. System design includes determining which E-beam
system will achieve treatment goals and designing the configuration. The
system design is estimated to cost $5,000.
5.6.9.3 Permitting and Regulatory Costs
Permitting and regulatory costs in this analysis include permit fees
for discharging treated water to a surface water body. The cost of this
permit is based on regulatory agency requirements and treatment goals
for a particular site. The discharge permit for each case is estimated to
cost $5,000. Costs of highway permits for overweight vehicles are in-
cluded in the costs of mobilization.
5.6.9.4 Mobilization and Start-up Costs
Mobilization and start-up costs include the costs of transporting the
E-beam system to the site, assembling the E-beam system, and performing
the initial shakedown of the treatment system. HVEA provides initial opera-
tor training to its clients as part of providing the E-beam equipment.
5.21
-------
Ex-Situ High Voltage Electron Beam Treatment
Transportation costs are assumed at $6.21/km ($10/mi) for 621 km (1,000
mi), or $10,000. The costs of highway permits for overweight vehicles are
~, MS.- _-- - included in this total cost.
***-* Assembly costs include the costs of unloading equipment from the trail- .
ers, assembling the E-beam system, and connecting extraction well piping
and electrical lines. A two-person crew will work three 8-hour days to un-
load and assemble the system and perform the initial shakedown. The total
start-up costs are assumed as $ 10,000, including labor and hookup costs.
For the 45 kW and 75 kW systems, completion of initial assembly and 'f
shakedown activities is expected to require the two-person crewto work
' '' about five 8-hour days. The start-up costs for these systems are about
$15,000, including labor and electrical hookup costs. Total mobilization and
start-up costs for each case are estimated to be about $20,000. ^, »
*
5.6.9.5 Equipment Costs
Equipment costs include the costs of leasing the E-beam treatment sys-
tem. HVEA provides the complete E-beam treatment system configured for
site-specific conditions. All E-beam treatment equipment is leased to the
client. As a result, all depreciation and salvage value is reflected in the price
for leasing the equipment. At the end of a treatment project, HVEA decon-
taminates and demobilizes its treatment equipment.
Equipment costs are determined by the size of the E-beam system needed
to complete the remediation project and are incurred as a lump sum; as a
result, even though the equipment is leased to the client, it is not priced at a
monthly rate. For this analysis, HVEA estimates that the capital equipment
for both cases will cost $842,000 for a 21 kW system; $1,208,000 for a 45
kW system; and $1,432,000 for a 75 kW system.
5.6.9.6 Labor Costs
Once the system is functioning, it is assumed to operate continuously at
the design flow rate except during routine maintenance, which HVEA con-
ducts. /One operator performs routine equipment monitoring and sampling
activities. Under normal operating conditions, an operator is required to
monitor the system about once each week.
5.22
-------
Chapter 5
It is also assumed that system monitoring and sampling duties is con-
ducted by a full-time employee of the site owner who is assigned as the pri-
mary operator. Further, a second person, also employed by the site owner,
will be trained to act as a backup operator. Based on observations made at
the SITE demonstration, it is estimated that operation of the system requires
about one-quarter of the primary operator's time. Assuming the primary
operator earns $40,000/year, the total direct annual labor cost for each case
is estimated to be $10,000.
5.6.9.7 Supply Costs ,/V
No chemicals or treatment additives are typically used to tteat the ground-
water using E-t>eam technology. Therefore, no direct supply costs are ei-
pected. Supplies that will be needed as part of the overall grouridwater
remediation project include Level D, disposable personal protective equip-
ment (PPE), PPE disposal drums, and sampling and field analytical supplies.
Disposable PPE for each case is assumed to cost about $600/yr for the
primary operator: Used PPE is assumed to be hazardous and needs to be
disposed of in 90 L (24 gal) fiber drums. One drum is assumed to be filled
every 2 months, and each drum costs about $12. For each case, the total
annual drum costs are about $100.
During the demonstration at Savannah River Site, the average pH level of
the influent was about 4.7; the average pH level of the effluent ranged be-
tween 3.0 and 3.5. Depending on discharge permit levels and influent and
effluent pH levels, the pH may require adjustment. In this event, additional
supplies will be necessary. The quantity of supplies needed is highly
site-specific and difficult to determine; therefore, this analysis does not in-
clude posttreatment pH adjustment costs.
Total annual supply costs for each case are estimated to be $1,700.
5.6.9.8 Utility Costs
Electricity is the only utility used by the E-beam system. Electricity is used
to run the E-beam treatment system, pumps, blower, and air chiller. Electricity
costs can vary considerably depending on the location of the site, local utility
rates, the E-beam system used, the total number of pumps and other electrical
equipment operating, the use of the air chiller, and whether electrical power
lines are available at the site or must be installed.
5.23
-------
Ex-Sltu High Voltage Electron Beam Treatment
%
This analysis assumes that power lines are available at the site, and a
constant rate of electricity consumption based on the electrical require-
ments of the 21 kW E-beam treatment system. The pumps, blower, and
air chiller are assumed to draw an additional 20 kW, which is based on
observations made during the SITE demonstration at the Savannah River
Site. Therefore, the 21 kW unit operating for 1 hour draws about 42 kW
hours (kWh) of electricity. The total annual electrical energy consump-
tion is estimated to be about 366,910 kWh. Electricity is assumed to
cost $0.07/kWh, including demand and usage charges. The total annual- -
electricity costs for each case are estimated to be about $25,700. The '•''
total annual electricity costs are estimated to be $52,600 fopthe 45 kW
"% system and $87,500 for the 75 kW system. { '
*
Water and natural gas usage are highly site-specific, but assumed to be
minimal for each case in this analysis. As a result, no costs for^these utilities
are included.
5.6.9.9 Effluent Treatment and Disposal Costs
At the Savannah River Site demonstration, the E-beam system did not
meet target treatment levels for about half of the VOCs. Depending on the
treatment goals for a site, additional effluent treatment may be required and
additional treatment or disposal costs incurred. Because of this uncertainty,
effluent treatment or disposal costs are not included.
The E-beam system does not produce air emissions because the water
delivery and cooling ah- systems are enclosed. As a result, no cost for air
emissions treatment is incurred.
5.6.9.10 Residual Waste Shipping and Handling Costs
The only residuals produced during E-beam system operation are fiber
drums containing used PPE and waste sampling and field analytical supplies,
all of which are typically associated with a groundwater remediation project.
This waste is considered hazardous and requires disposal at a permitted fa-
cility. For each case, it is assumed that about six drums of waste are dis-
posed annually. The cost of handling and transporting the drums and dispos-
ing them at a hazardous waste disposal facility is about $1,000 per drum.
The total drum disposal costs for each case are about $6,000/yr.
5.24
-------
Chapter 5
5.6.9.11 Analytical Services Costs
Required sampling frequencies and number of samples are highly
site-specific and are based on treatment goals and contaminant concentra-
tions. Analytical costs associated with a groundwater remediation project
include the costs of laboratory analyses, data reduction, and Quality Assur-
ance/Quality Control (QA/QC). The analysis assumes that one sample of
untreated water, one sample of treated water, and associated QC samples
(trip blanks, field duplicates, and matrix spike/matrix spike triplicates) will
be analyzed for VOCs every month. Therefore, monthly analytical costs are''
about $2,000. '/'
A •
5.6.9.12 Equipment Maintenance Costs j '
»
HVEA estimates that annual equipment maintenance costs afe about 3%
of the capital equipment costs. Therefore, the total annual equipment main-
tenance costs for each case are about $25,300 for the 21 kW system, $36,200
for the 45 kW system, and $43,000 for the 75 kW system.
5.6.9.13 Site Demobilization Costs
Site demobilization includes treatment system shut-down, disassembly,
and decontamination; site cleanup and restoration; utility disconnection; and
transportation of the E-beam equipment off-site. A two-person crew will
work about five 8-hour days to disassemble and load the system. It is as-
sumed that the equipment will be transported 1000 km (670 mi) either for
storage or to the next job site. HVEA estimates that the total cost of demo-
bilization is about $15,000 for each case. This total includes all labor, mate-
rial, and transportation costs.
5.6.9.14 Economic Analysis Conclusions
Total estimated fixed costs are about $1,057,600 for each case. Of this
total, $842,000, or about 80%, is for E-beam equipment costs. Over 16% of
the total fixed cost is for site preparation; this cost is not entirely attributable
to operating the treatment system, but rather is necessary for setting up the
system. Total estimated annual variable costs are about $92,700 for each
case. Of this total, analytical service costs comprise about 26%, equipment
maintenance costs about 27%, and utility costs nearly 28%.
5.25
-------
Ex-Sltu High Voltage Electron Beam Treatment
The analysis of the base-case E-beam technology (21 kW) reveals that
operating costs are strongly affected by the E-beam system and flow rate
used. The larger systems take less time to complete a groundwater
remediation project, but the higher equipment and utility costs result in a
higher cost per 3,785 L (1,000 gal) of groundwater treated. The base-case
assumes that the total amount of groundwater to be treated is 1,200 million L
(315 million gal). In Case 1,15 years would be needed to complete the
remediation project; in Case 2,30 years would be needed. The total esti-
mated cost of the project is $2,764,000 for Case 1 and $6,281,000 for Case .' >'
2. The estimated cost per 3,785 L (1,000 gal) of groundwater treated (net ' / '
present value) is $5.16 for Case 1 and $7.85 for Case 2. *
Table 5.5 presents only the direct costs associated with the E-beam treat-
ment system. This analysis is provided to segregate the direct cosf^ of pro-
curing and operating the E-beam system from the total costs of | gYoundwa-
ter remediation project. The direct costs are the same for both cases. Total
fixed costs are estimated to be $900,000, and total annual variable costs are
estimated to be $67,000. The analytical supplies cost has been excluded
because at $l,000/yr, it represents about 1% of the total annual variable
costs. The direct cost per 3,785 L (1,000 gal) of groundwater treated is esti-
mated to be $4.07 for Case 1 and $5.99 for Case 2.
In summary, the cost of treatment using an HVEA E-beam system de-
pends on may factors such as the initial concentrations of organic contami-
nants, treatment objectives, the dose required to obtain the desired destruc-
tion, the volume of waste to be treated, the size of the treatment facility, the
length of treatment, and the manner in which capital recovery is handled.
The cost of treatment using HVEA systems in various industrial waste and
groundwater applications has ranged from $52/1,000 L to $7.57/L ($2/1,000
gal to $0.50/gal).
5.6.10 Design Validation
Validation of the design of an E-Beam treatment system is accomplished
using treatability studies to establish the necessary electron beam energy
required and to ascertain the nature and extent of posttreatment, if any, that
is required. The simplicity of the system enables these studies to be accom-
plished economically.
5.26
-------
I j
Table 5.5
E-Beam Treatment System Direct Costs0
pi
fo
Treatment System Configurations in Kilowatts (kW)
Cost Categories
Site Preparation1'
Treatability Study and System Design
Mobilization and Startup"
Transportation
Assembly and Shakedown
Equipment1'
Labor0
Utilities'1
Residual Waste Shipping and Handling6
Equipment Maintenance'
Site Demobilization1*
Total One-Time Costsb
Total Annual O&M Costs'
Cost per 1,000 gal Treated — Case ld
Costs per 1 ,000 gal Treated — Case 2«
21 kW
Itemized Total
($) ($)
23,000
23,000
20,000
10,000
10,000
842,000
10,000
25,700
6,000
25,300
15,000
900,000
67,000
4.07
5.99
45 kW
Itemized Total
($) ($)
23,000
23,000
25,000
10,000
10,000
1,208,000
10,000
52,600
6,000
36,200
15,000
1,271,000
104,800
5 17
6.05
75 kW
Itemized Total
($) ($)
23,000
23,000
25,000
10,000
10,000
1,432,000
10,000
87,500
6,000
43,000
15,000
1,495,000
146,500
am
7.10
*Th« table presents direct costs associated with the E-beam treatment system segregated from the costs incurred as a result^"conducting a groundwater remediation project All
assumptions used vn this analysis apply. "*"~
"Fixed costs
'Variable costs
"Net present value using the same assumptions used in Table 5.3
•Net present value using the same assumptions used in Table 5.4 ' "* ^
"v »
Source: US EPA 1995c " >
9
-------
Ex-Sltu High Voltage Electron Beam Treatment
5.6.11 Permitting Requirements
The process will most likely require acquisition of one or more discharge
permits for the treated water. No air permits are required. In addition, the
electron beam source requires that the site of operation have a permit for the
use of an ionizing radiation source. This permit is usually issued by the
health department for the county or the state and is the same permit as re-
quired for x-ray equipment.
5.6.12 Performance Measures • /
The system performance is determined by sampling and analysing the t
quality of the treated water. I
5.6.13 Design Checklist £
Following is a list of elements to be considered during design.
1. Quality and flow rate of the stream to be treated.
2. Characteristics of stream to be treated: pH, carbonates, concen-
trations of organics, and dissolved solids.
3. Target limits for treated effluent BOD, COD, etc.
4. Site utilities: electricity, water (drinking, sanitary, process), and
telephone.
5. Availability of support services, such as fire fighting, emergency
medical, etc.
6. Site accessibility for equipment delivery by road or rail, restric-
tions on loading, noise, etc., and availability of access for system
operation and maintenance. Maintenance access roads.
5.7 Implementation and Operation
5.7.1 Implementation Strategies
The E-beam and its design are well developed; however, its implementa-
tion is highly site-specific. Implementation must begin with a good
5.28
-------
Chapter 5
understanding of the site and the contaminants. This information should be
coupled with a set of treatability studies performed on samples of the
groundwater from the site using equipment on which data has been obtained
and on which successful scale-up have occurred at other sites. Conducting
the treatability study on actual site samples is especially crucial for this tech-
nology since small differences in chemical composition can have a major
impact on the performance.
The vendor used for these treatability studies can be the supplier of the
equipment or an independent party; however, a strong background in the use
of electron beam treatment is necessary. The vendor should be capable of
interpreting the results of the treatability studies and creating It set of perfor-
mance specifications for the pieces of equipment. The equipment! itself is
covered by patents held by the developer and, in general, it can'jbnly be ac-
quired through High Voltage Environmental Applications, Ine. While com-
petitive bidding is desirable, one must recognize that this application of high
voltage at a large scale and relatively high power consumption under highly
humid conditions requires extensive field experience and the ultimate vendor
selection process should include assessment of the vendor's experience in
similar applications in addition to a cost comparison.
5.7.2 Operation
Startup of the full-scale system was observed by the author of this chapter
during a visit in April 1996. Startup consisted of clearing the irradiation
chamber of personnel, slowly warming up the power supplies for the elec-
tron beam by increasing the power level to the operating power over ap-
proximately 15 to 20 minutes, then starting the flow of the contaminated
stream. The absence of moving parts is apparent.
Operation involves checking and adjusting pH and alkalinity of the
feedwater, periodic checks of the electric systems, and sampling and analysis
of the influent and effluent from the process. Once operating, the system
requires only a part-time person to maintain and monitor its performance.
Successful installation requires that the pH and alkalinity of the influent
be maintained within design specifications. The pH monitoring and auto-
matic adjustment using acid or alkaline solutions is a highly desirable addi-
tion to the system. Otherwise, the system requires minimal monitoring.
5.29
-------
Ex-Sltu High Voltage Electron Beam Treatment
5.8 Cose History
The E-beam technology was extensively evaluated by the Superfund
vu"' "" " Innovative Technology Evaluation (SITE) program of the U.S. Environmen-
tal Protection Agency (US EPA), Risk Reduction Engineering Laboratory,
now named National Risk Management Research Laboratory (NRML),
Cincinnati, Ohio. The report (US EPA 1995d) presents detailed perfor-
mance and cost information for the process as well as discussing the case
history of the demonstration program. The demonstration was conducted at-
the U.S. Department of Energy (DOE) Savannah River Site (SRS) in Aiken/
South Carolina, during two different periods totaling 3 weeks in*September
* - and November 1994. I •
During the demonstration, about 265,000 L (70,000 gal) of M-4rea
groundwater contaminated with VOCs was treated. The principal groundwa-
ter contaminants were trichlorethylene (TCE) and perchloroethylene (PCE),
which were present at concentrations of about 27,000 and 11,000 ug/L, re-
spectively. The groundwater also contained low levels (40 jJg/L) of
cis-l,2-dichloroethene(l,2-DCE). Before treatment, groundwater was
pumped from a recovery well into a 28,000 L (7,500 gal) equalization tank
to minimize any variability in influent characteristics. Treated groundwater
was stored in a 38,000 L (10,000 gal) tank before being pumped to an
on-site air stripper, which was treating contaminated groundwater from the
demonstration area.
During a portion of the E-beam technology demonstration, the groundwa-
ter was spiked with VOCs not present in the M-area groundwater. The re-
sultant influent concentrations ranged from about 100 to 500 ug/L for the
following spiking compounds: 1,1,1 -trichloroethane (1,1,1 -TCA),
1,2-dichloroethane (1,2-DCA); chloroform; carbon tetrachloride (CC14); and
benzene, toluene, ethylbenzene, and p-xylene (BTEX). Saturated VOCs
(1,1,1-TCA, 1,2-DCA, chloroform, and CC14) were chosen as spiking com-
pounds because they are relatively difficult to destroy using technologies
such as the E-beam technology that involve free radical chemistry. BTEX
were chosen because they are common groundwater contaminants at
Superfund and other contaminated sites. For the SITE technology demon-
stration, TCE, PCE, 1,2-DCE, and the spiking compounds were considered
to be critical VOCs.
5.30
-------
Chapters
6.8.1 Demonstration Procedures
The technology demonstration was conducted in five phases. Thirteen
test runs were performed during these five phases to evaluate HVEA treat-
ment system performance. During each run, influent characteristics or oper-
ating parameters were changed to collect information in order to meet
project objectives. The demonstration approach is summarized below.
During Phase 1, beam current, one of the principal operating parameters,
was varied to observe how E-beam dose affects treatment system perfor- ..,,
mance at a constant flow rate of 150 L/min (40 gal/min). Three runs were
conducted during Phase I using unspiked groundwater. ^
During Phase 2, spiked groundwater was used to collect information on
treatment system performance in destroying VOCs other than those present
in the M-area groundwater. Two runs were performed using different beam
currents and a constant flow rate of 150 L/min (40 gal/min). -^Phase 2 also
included a zero dose run to identify reduction in VOC concentrations result-
ing from mechanisms other than VOC destruction by the E-beam (for ex-
ample, volatilization).
Phase 3 tested the reproducibility of HVEA system performance for treat-
ing spiked groundwater. Three runs were performed under identical operat-
ing conditions, which were determined based on preliminary treatment re-
sults from Phases 1 and 2.
Phase 4 consisted of one run to evaluate HVEA system performance
at the minimum limiting flow rate [57 L/min (15 gal/min)] of the system
used for the demonstration. The minimum flow rate was chosen because
preliminary results from Phases 1, 2, and 3 indicated that the HVEA
system did not meet effluent target levels at higher flow rates and using
maximum beam current.
Phase 5 began 4 weeks after Phase 4 was completed. The interval be-
tween Phases 4 and 5 gave HVEA time to evaluate preliminary results from
Phases 1 through 4 and conduct additional studies on spiked and unspiked
groundwater from M-area recovery well RWM-1 in order to determine Phase
5 operating conditions. Based on information from the test runs and addi-
tional studies, HVEA adjusted the influent delivery system to improve over-
all treatment system performance. This was accomplished by increasing the
dose without increasing the beam current or lowering the flow rate.
5.31
-------
Ex-SItu High Voltage Electron Beam Treatment
TO evaluate the effect of the improved delivery system in Phase 5, HVEA
selected the same flow rate (75.71 L/min [20 gal/min]) and beam current (42
mA [63 kW]) as used in the reproducibility runs. Of the three Phase 5 runs,
one used unspiked groundwater, one used spiked groundwater, and one used
•alkalinity-adjusted spiked groundwater. Alkalinity was adjusted in one run
because carbonate and bicarbonate ions scavenged «OH, potentially affecting
VOC removal efficiency. During this run, a sodium bicarbonate solution was
added to the influent in order to adjust alkalinity from <5 mg/L to about 500
mg/L as CaCO3, which is within the typical range of groundwater alkalinity ,.
levels hi the United States. ' /
*.
5.8.2 Sampling and Analytical Procedures f •
During the demonstration, groundwater samples were collectedpt E-beam
influent and effluent sampling locations, and cooling air sample^ were col-
lected before and after the carbon adsorber.
Each test run lasted about 3 hours, and four groundwater sampling events
were conducted at 45-minute intervals during each test run. Groundwater
samples for VOC analysis were collected during each sampling event so that
average influent and effluent concentrations could be calculated based on
four replicate data points. Groundwater samples for other analyses were
typically collected during two of the sampling events.
Groundwater samples were collected during all runs for VOC and pH
analyses. Groundwater samples were collected during selected runs for
analysis for SVOCs, haloacetic acids, aldehydes, H2O2 (effluent only), TOC,
purgeable organic carbon (POC), total inorganic carbon (TIC), total organic
halides (TOX), chloride, alkalinity, and acute toxicity. Influent and effluent
samples were analyzed using US EPA-approved methods, such as those .
found in Test Met hods for Evaluation Solid Waste and Methods for Chemical
Analysis of Water and Wastes (US EPA 1990; US EPA 1983) or other stan-
dard or published methods (American Public Health Association, American
Water Works Association, and Water Environment Federation 1992; Boltz
andHowell 1979).
During Runs 1 through 10, cooling air samples were collected and ana-
lyzed for VOCs, O3, and HC1 using an on-site Fourier transform infrared
(FTIR) interferometer. Cooling air samples were not collected during Runs
11,12, and 13 (Phase 5) because of high costs associated with maintaining
5.32
-------
Chapter 5
the FTIR interferometer in the field during the 4-week interval between
Phases 4 and 5. This approach did not affect project objectives because
cooling air was analyzed only for noncritical parameters to meet a second-
ary objective.
On-site measurements of flow rate, beam current, and power consumption
were recorded during all runs.
In all cases, US EPA-approved sampling, analytical, and QA/QC proce-
dures were followed to obtain reliable data. These procedures are describe.^.
in the QAPP written specifically for the E-beam technology demonstration
(PRC Environmental Management 1994) and are summarized in the TER,
which is available from the US EPA project manager. * /
i
5.8.3 Removal Efficiency jf
Table 5.6 presents the range of critical VOC concentrations in the influent
to the E-beam unit for unspiked and spiked test runs.
Most of the performance data are reported based on average values from
replicate sampling events. In some cases, samples were analyzed at two
dilutions; when this occurred, the results for the lower dilution were used to
calculate the average value. For influent samples with analyte concentra-
tions at nondetectable levels, the detection limit was used as the estimated
concentration when the average value was calculated. For effluent samples
with analyte concentrations at nondetectable levels, one-half the detection
limit was used as the estimated concentration when the average value was
calculated. If all replicate effluent samples had nondetectable concentrations
of any analyte, the detection limit was used as the average value, the removal
efficiency was reported as a greater than (>) value, and the 95% upper confi-
dence limit (UCL) was not calculated.
After the demonstration data were reviewed, it was determined that more
than one approach should have been used to handle nondetectable influent and
effluent values in order to calculate averages. For influent nondetectable values,
the detection limit was used in place of the nondetectable value. Although this
approach deviates from typical environmental engineering practice, which is to
use one-half the detection limit for nondetectable values, using the full detection
limit is more appropriate in this case because (1) there is less variability in influ-
ent concentrations than in effluent concentrations, and (2) 50% or more of the
influent samples had VOC concentrations above the detection limit. For
5.33
-------
Ex-Situ High Voltage Electron Beam Treatment
example, in about 50% of the influent samples (28 out of 52 samples col-
lected in 13 runs), the concentration of 1,2-DCE was reported as
nondetectable (the detection limit is 40 Ug/L); in the remaining samples,
this compound was present at concentrations of up to 50 ug/L. For ef-
fluent nondetectable values, however, using the typical practice for han-
dling nondetectable values is more appropriate. This is the case because
the effluent data have greater variability as a result of E-beam treatment
and because the data are more limited (the effluent data from all the runs
cannot be pooled together because the operating conditions generally .
varied from ran to run).
Table 5.6 *
VOC Concentrations' in Unspiked and Spiked Groundwater Influent
voc
Unspiked Groundwater
Qlg/L)
Spiked Groundwater
(Hg/L)
TCE
PCE
1,2-DCE
1,1,1-TCA
U-DCA
Chloroform
Cd«
Benzene
Toluene
Ethylbcnzene
Xylenes
25,000 to 30,000
9,200 to 12.250
40to43
ND»
ND*
ND»
ND»
ND»
ND»
ND»
ND»
25,000 to 37,000
9,200 to 14,000
<40to45
200to500
210 to 840
240to650
150 to 400
220to550
170 to 360
95 to 250
85 to 200
N0« Not dot»ct»d, detection limit not given, estimated to be about 1 jig/L
ND" Not datactod.da(«ction limit 40 }ig/L
Source: US EPA 19S5c
5.34
-------
Chapter 5
*''
Table 5.7 summarizes the VOC removal efficiencies for unspiked and
spiked groundwater runs conducted at different E-beam doses. HVEA con-
trols dose by adjusting the beam current, the flow rate, and the thickness of
the water stream impacted by the E-beam. In HVEA's system, the beam
current is controlled directly from the control panel, the flow rate is con-
trolled by manually adjusting the influent pump, and the thickness of the
water stream is controlled by the influent delivery system.
During Phase 1, unspiked groundwater was treated at three different
doses. For these runs, the dose was varied by changing the beam current/"'
while the flow rate remained constant. As shown in Table 5.7, removal 'effi-
ciencies for TCE, PCE, and 1,2-DCE increased when the beam current was
increased. A similar effect was observed during Phase 2, which involved
spiked groundwater. The dose was increased further during Phase 3 by low-
ering the flow rate from 150 to 75 L/min (40 to 20 gal/min) ani increasing
the beam current to the maximum level (42 mA [63 kW]); corresponding
increases in REs were observed, particularly for spiked compounds. Finally,
for Phase 5, HVEA adjusted the delivery system and these adjustments in-
creased the dose although the beam current and flow rate were set at the
same levels as were used for Phase 3. HVEA considers information regard-
ing the delivery system to be proprietary. Phase 5 results indicate that the
delivery system adjustments increased removal efficiencies for most VOCs.
In fact, the operating conditions during Phase 5 generally yielded the highest
removal efficiencies observed during the demonstration.
Table 5.7 also shows that for all spiked groundwater runs, removal
efficiencies for TCE, PCE, 1,2-DCE, and BTEX were much higher than
for 1,1,1-TCA, 1,2-DCA, chloroform, and CC14. The difference in sys-
tem performance for these two groups of VOCs is postulated to be due
to the presence of double bonds between carbon atoms in TCE, PCE,
and 1,2-DCE and aromatic bonds between carbon atoms in BTEX,
which makes these compounds more amenable to oxidation by free radi-
cals generated by the E-beam. Furthermore, the removal efficiencies of
saturated chlorinated compounds, typified by CC14, were consistently
higher than removal efficiencies for 1,1,1-TCA, 1,2-DCA, and chloro-
form. This effect may be a consequence of the relatively large number
of chlorine atoms in CC14. The four chlorine atoms facilitate CCl4desta-
.bilization and are good "leaving groups" in the presence of free radicals;
therefore, CC14 may be more amenable to E-beam destruction than simi-
lar compounds with fewer chlorine atoms.
5.35
-------
Ex-SItu High Voltage Electron Beam Treatment
UJ
o
&
o
§ti it
_, op:
ll H
o\
«0 11 II
'
ll II
u H
II II
S
00
tr
I
ui
cvi
""*S o> i-
111 2
<
£ Z M
5.36
-------
Chapter 5
Table 5.8 shows that the HVEA treatment system achieved the efflu-
ent target levels for 1,2-DCE, CC14, and BTEX. Effluent target levels
were not achieved for 1,1,1-TCA, 1,2-DCA, and chloroform when these
compounds were present at spiked levels (230,440, and 316 ng/L for
1,1,1-TCA, 1,2-DCA, and chloroform, respectively). Effluent target
levels were also not achieved for TCE and PCE when they were present
at existing levels in M-area groundwater (27,000 and 11,000 ng/L for
TCE and PCE, respectively).
Only effluent concentrations for Runs 11 and 12 are shown in Table 5.8 * >'
because the HVEA treatment system displayed the best overall performance,
in terms of removal efficiencies during these runs. However^ffluent target
levels were met for toluene, ethylbenzene, and xylenes during the{ other runs.
*/
Table 5.8
Compliance with Applicable Effluent Target Levels
voc
TCE
PCE
1,2-DCE*
1,1,1-TCA
1,2-DCA
Chloroform
ca<
Benzene
Toluene
Ethylbenzene
Xylenes'
Effluent Target Level (Hg/L)
5
5
54
54
5
46
5
5
SO
57
320
95 Percent UCL for Effluent Concentration (Hg/L)
i Run 11
190
100
4U
NA •
NA
NA
NA
NA
NA
NA
NA
Run 12
1.100
250
4U
83
180
130
4U
4U
4U
4U
4U
U analyte not detected in the treatment system effluent at or above the value shown
NA not applicable (because the analyfe was not detected in the treatment system influent)
• Influent concentrations for 1,2-DCE and xylenes were below the effluent target tevete
Source: US EPA 1995c
5.37
-------
Ex-SItu High Voltage Electron Beam Treatment
5.8.4 Effect of Treatment on Toxicity
Bioassay tests were performed to evaluate the change in acute toxicity of
the groundwater after treatment by the HVEA system. Two common fresh-
water test organisms, a water flea (Ceriodaphnia dubia) and a fathead min-
now (Pitnephales promelas), were used in the bioassay tests. The acute
toxicity was measured as the concentration at which 50% of the organisms
died (LC50) and was expressed as the % age of influent or effluent in the test
water. One influent sample and one effluent sample from each run were
tested; chronic toxicity was not measured.
/ /
Table"5.9 presents the bioassay test results of influent and effluent
samples from Runs 4 and 5 (the reproducibility runs) and Runs fl, 12, and,
13. These results show that some influent samples and all effluent samples'
were acutely toxic to both test organisms. The change in groundv^ter toxic-
ity resulting from treatment by the HVEA system was evaluated statistically
using data from the reproducibility runs. Specifically, the meaii difference
between the influent and effluent LC50 values was compared to zero using a
two-tailed paired Student's t-test. The null hypothesis was that the mean
difference between influent and effluent LC50 values equaled zero at a 0.05
significance level. The critical t value at this significance level with two
degrees of freedom is 4.303. The calculated t values for the water flea and
the fathead minnow were 1.47 and 31.6, respectively. These results indicate
that treatment by the E-beam technology statistically increased groundwater
toxicity for the fathead minnow, but not for the water flea.
As noted above, influent and effluent samples for bioassay testing were
collected during Runs 11,12, and 13, which were conducted after HVEA
adjusted the influent delivery system to increase the dose. Although toxicity
data for these runs cannot be statistically evaluated because the influent char-
acteristics were different, the data suggest that the difference between the
influent and effluent LC50 values decreased for fathead minnows. Fpr
fathead minnows, the increase in toxicity resulting from E-beam treatment
(the difference between influent and effluent LC50 values) in Run 12 was
less than the average increase in toxicity in Runs 8 and 9. This fact may be
related to the higher VOC REs and reduced byproduct formation achieved
when the dose was increased by adjusting the influent delivery system for
Run 12. However, for water fleas, the LC50 data could not be compared
because the increases in LC50 values resulting from E-beam treatment were
not observed to be absolute values (that is, all observations were > values).
5.38
-------
Chapter 5
JSfc,
Table 5.9
Acute Toxlclty Data
LC50 (%)
Run*
4
5
7"
8b
9b
11
12
13
Ceriodaphnia dubia
influent
35
68
>ioo -
17
16
37
37
>100
Effluent
8.8
18
17
<6.2
<6.2
<6.2
<6.2
12
Pimephales promelas
Influent
72
100
>100 .
>100
>100
89
83
>100
Effluent
8.6
15
8.8
16 ''*
. A
18 ,
ls.8
f 9.8
\ 54
•Runs 7, 8, and 9 were the reproducbtlity runs (Phase 3). Runs 11, 12, and 13 were conducted after HVEA adjusted the
influent delivery system (Phase 5). Run 11 was conducted with unspiked groundwater, Run 12 was conducted with
spiked groundwater, and Run 13 was conducted with alkalinity-adjusted spiked groundwater.
"Using data from the three reproducibility runs, a two-tailed paired Student's t-test with a 0.05 significance level was
performed for each organism. The null hypothesis was that the mean difference between the influent and effluent LC50
values equaled zero. For LC50 values shown as >100 and <6.2, 100 and 6.2 were used to calculate the mean
difference. The calculated t values were 1.47 and 31.6 for Ceriodaphnia dubia and Pimephales promelas, respectively.
Source: US EPA 1995c
Published reports indicate the H2O2 generated by technologies involving
free radicals may contribute to effluent toxicity (US EPA 1993). The aver-
age effluent H2O2 concentration was 8.0 mg/L during the reproducibility
runs and 8.9 mg/L during Phase 5 runs. Literature data indicate that the
LC50 for H2O2 for the water flea is about 2 mg/L. Because no statistically
significant increase in acute toxicity for the water flea was observed during
E-beam treatment despite high levels of H2O2 in the effluent, it is likely that
any increase in toxicity associated with HP2 was counteracted by a decrease
in toxicity resulting from VOC removal. The fathead minnow is less sensi-
tive to H2O2 than the water flea. The Connecticut Department of Environ-
mental Protection (CDEP) reported an LC50 value of 18.2 mg/L of H2O2
with 95% confidence limits of 10 and 25 mg/L for the fathead minnow
(CDEP 1993). Therefore, the statistically significant increase in acute toxic-
ity for the fathead minnow during E-beam treatment is more likely to have
been caused by residual VOCs or treatment byproducts than by H2O2.
5.39
-------
I j
Figure 5.2
VOC REs In Reproduclbllity Runs
£ a
o
TCE PCE 1,2-DCE 1,1,1-TCA 1,2-DCA Chloroform CC14 Benzene Toluene Ethylbenzene Xylenes
KB Run?
Ov\ Run 8
8868 Run 9
•NA - 1,2-OCE was not detected in Run 7
Reproduced courtesy of High Voltage Environmental Applications, Inc.
-------
Chapter 5
5.8.5 Reproducibility of Treatment System Performance
VOC removal efficiencies in the Phase 3 reproducibility runs (Runs 7, 8,
and 9) are shown in Figure 5.2. This figure indicates that the removal effi-
ciencies for all VOCs were reproducible. The maximum difference among
removal efficiencies for the three runs occurred for 1,2-DCA, for which REs
ranged from 60 to 65%, and 1,2-DCE, for which removal efficiencies ranged
from 85 to >91%. However, for other VOCs, the removal efficiencies dif-
fered by only 2 to 3% for the three runs. The ranges of VOC removal effi-
ciencies during the Phase 3 reproducibility runs are shown in Table 5.7. « ''
5.41
-------
I I
-------
Appendix A
EX-SITU ELECTROCHEMICAL
TREATMENT PROCESSES
Ex-situ electrochemical treatment processes are intended to treat aqueous
streams contaminated with metals, suspended solids, emulsions, and some
organic compounds. These processes fall into two broad categorids: electro-
chemical coagulation (electrocoagulation) and electrochemical $xidation/
reduction. This appendix provides case studies for three ex-sftu'electro-
chemical treatment processes using Electrochemical Coagulation and
Alternating-Current Electrocoagulation:
1. ACE Separator™ marketed by ElectroPure Systems Inc. — This
technology was the subject of a testing program under the emerg-
ing technology portion of US EPA's Superfund Innovative Tech-
nology Evaluation (SITE) program (Barkely, Farrell, and Will-
iams 1993).
2. Electrochemical Treatment System (no trade name) — Andco
Environmental Processes, Inc. evaluated a pilot-scale electro-
chemical treatment system at the Milan Army Ammunition Plant
(Laschinger 1992).
3. Electrochemical Oxidation/Reduction — Silver (IT) Process mar-
keted by AEA Technology (Oxfordshire, UK). This process is
based on the electrochemical cell used for chlorine production.
The technology is applicable only to the treatment of very
low-volume waste streams. It is being marketed for the treatment
of highly toxic materials only such as nuclear waste and chemical
munitions (Batey 1995).
A.1
-------
Ex-Situ Electrochemical Treatment Processes
A. 1 Electrochemical Coagulation
Chemical coagulation has been used for decades to destabilize colloidal
suspensions and to effect precipitation of soluble metal species as well as
other inorganic species from aqueous streams. Alum, lime, and/or polymers
have been the chemical coagulants used. These processes, however, tend to
generate large volumes of sludge with a high bound-water content that can
be slow to filter and difficult to dewater. The treatment processes also tend
to increase the total dissolved solids content of the effluent, making it unac- •
ceptable for reuse within industrial applications.
Electrocoagulation uses an electric field to achieve the same effect as /
chemical coagulation, but without some of the previously mentioned* draw-
backs. Rather than adding chemical agents to a wastewater, electrochemical
oxidation/reduction runs a current between two electrodes. The^electrodes
can be made of aluminum or iron or aluminum or iron pellets can be placed
between the electrodes. In either case, the current generates aluminum or
iron ions which, along with the electric potential, result in the coagulation of
the suspended solids. The current also causes dissolved metals to precipi-
tate. In its simplest form, an electrocoagulation system passes the aqueous
stream past energized electrodes that modify the surface potential of the
particulate, causing it to agglomerate into large particles that settle or filter
more readily.
The alternating current electrocoagulation (ACE) technology was originally
developed in the early 1980s to break stable aqueous suspensions of clays and
coal fines produced in the mining industry. Traditionally, these effluents were
treated with conventional techniques that made use of organic polymers and
inorganic salts to agglomerate and enhance the removal of the suspended mate-
rials. The ACE technology was developed to simplify effluent treatment, realize
cost savings, and facilitate recovery of fine-grained coal.
ACE is based upon colloidal chemistry principles — principles using
alternating electrical power and electrophoretic metal hydroxide coagulation.
The basic mechanism for the technology is electroflocculation wherein small
quantities (generally <30 mg/L) of aluminum hydroxide species are intro-
duced into solution to facilitate flocculation. Electroflocculation causes an
effect similar to that produced by the addition of chemical coagulants such
as aluminum or ferric sulfate. These cationic salts destabilize colloidal sus-
pensions by neutralizing negative charges associated with these particles at
A.2
-------
Appendix A
neutral or alkaline pH. This enables the particles to come together closely
enough to agglomerate under the influence of van der Waals attractive
forces. See Figure A. 1 for the ACE basic process flow.
Although the electroflocculation mechanism resembles chemical coagula-
tion in that cationic species are responsible for the neutralization of surface
charges, the characteristics of the electrocoagulated floe differ dramatically
from those of floe generated by chemical coagulation. An electrocoagulated
floe tends to contain less bound water, is more shear resistant, and is more
readily filterable. • *r
'/''
Application of. an AC electric field to the electrodes induces dissolu-
tion of the aluminum and formation of the polymeric hydroxide species.
Charge neutralization and particle growth are initiated within the elec-
trocoagulation cells and continue following discharge of the aqueous
medium from the apparatus. In this way, product separation into solids,
water, and oils may be achieved.
Figure A.1
Schematic of an ACE Separator™'Used in
Alternating-Current Electrocoagulation
»• Gas Outlet
Effluent Liquid
Product Separation
Plate Electrodes
Influent Liquid '
Gas Inlet
Source: Barktey. Fan-ell, and Williams 1993
A.3
-------
Ex-Sltu Electrochemical Treatment Processes
A. 1.1 ACE Technology
A two-year research effort was conducted by ElectroPure Systems, Inc.,
to evaluate the technical and economic feasibility of ACE for remediation of
aqueous waste streams at Superfund sites (Barkley, Farrell, and Williams
1993). The ACE technology introduces low concentrations of nontoxic alu-
minum hydroxide species into the aqueous media by the electrochemical
dissolution of alummum-containing electrodes or pellets. The aluminum
species that are produced neutralize the electrolytic charges on suspended
material and/or prompt the coprecipitation of certain soluble ionic species, * /
facilitating their removal. *
- i
A. 1.1.1 Process Description ,/
The ACE technology was tested using two designs of the ACE Separator™:
• a parallel-electrode unit in which a series of vertically-oriented
aluminum electrodes form a series of monopolar electrolytic cells
through which the effluent passes; and
• a fluidized-bed unit with nonconductive cylinders equipped with
nonconductive metal electrodes between which a turbulent fluid-
ized bed of aluminum alloy pellets is maintained.
Electrocoagulation operating conditions are highly dependent on the
chemistry of the aqueous medium, especially conductivity. Other character-
istics, such as pH, particle size, and chemical constituent concentrations will
influence operating conditions. Treatment generally requires application of
low voltage (<150 VAC) to the electrocoagulation cell electrodes; current
usage is typically 1 to 5 amp-min/L (4 to 19 amp-min/gal). The flow rate of
the aqueous medium through an electrocoagulation cell depends on the elec-
trical conductivity of the solution, the nature of the entrained suspension or
emulsion, and the extent of electrocoagulation required to achieve the treat-
ment objective. Retention times as short as 5 sec are sometimes sufficient to
break a suspension. Electrocoagulation may be accomplished in a single
pass or multiple passes (recycle mode).
In the fluidized-bed unit, compressed air is introduced into the electroco-
agulation cells to assist in maintaining the turbulent fluidized bed and to
enhance the aluminum dissolution efficiency by increasing the anodic sur-
face area. It also provides a mechanical scrubbing action within the electro-
coagulation cell that reduces buildup of impermeable oxide coatings on the
A.4
-------
Appendix A
aluminum pellets and the inherent loss of efficiency that would result. Typi-
cally, the fluidized-bed unit dissolves aluminum at least one order of magni-
tude more efficiently than the parallel-electrode unit. Depending on system
configuration, maintenance of the apparatus is limited to periodic replenish-
ment of the aluminum fluidized-bed material and/or electrodes. For most
applications, pellets for the fluidized-bed unit can be produced from recycled
aluminum scrap or beverage containers. Where sludge reclamation is the
objective, however, the use of higher quality pellets is required to reduce the
introduction of impurities in the sludge. , - /
V
A. 1.1.2 Technology Testing
ElectroPure Systems, Inc. tested the technology in both the (
parallel-electrode and fluidized-bed configurations on various Surrogate
wastes containing emulsified diesel fuel metals and clays. The'wastes were
prepared to resemble those from leaking from underground storage tanks and
soil washing operations. The primary objective of such testing was to estab-
lish operating conditions for the ACE Separator™ to break the oil/water
emulsion and achieve reductions in clay, suspended solids, and soluble metal
pollutant loadings.
The surrogate wastes were prepared by mixing 0.2 to 3.0% (by weight) of
the -230 mesh (clay and silt) fraction of the US EPA's synthetic soil matrix
(SSM) with the following:
• 0.5 to 1.5% (by weight) Number 2 diesel fuel;
• 0.05 to 0.10% (by weight) of an emulsifier (Titon-lOOX* or
Alconox soap); and
• 10 to 100 mg/L of one or more of the following contaminants:
copper, nickel, zinc orthophosphate, or fluoride.
The pH of each surrogate mixture was adjusted with either sodium hy-
droxide or calcium oxide to the desired value (5,7, or 9) and the conductiv-
ity was raised to roughly 1200 to 1500 uS/cm (3,000 to 3,800 uS/in.) with
sodium chloride to simulate values expected in nature.
Initially, bench-scale electrocoagulation experiments using the parallel
electrode unit were conducted on five aqueous-based systems that included a
metals mixture, clay suspension, diesel fuel emulsion, soluble organic salt,
and diesel fuel/soluble organic emulsion. Optimum treatment times were
established by examining the contaminant loadings as a function of
A.5
-------
Ex-S!tu Electrochemical Treatment Processes
treatment time. To compare the results with conventional treatment pro-
cesses, aliquots of each surrogate stock solution were treated with alum.
Sufficient alum was added to give the aluminum equivalent to that intro-
duced by the electrocoagulation equipment.
When the results of the bench-scale experiments were applied to flow
reactor testing during the second year of the program, the following operat-
ing difficulties were encountered:
• persistent electrode coating and fouling; and
• low efficiencies of aluminum generation.
"• The program was, therefore, modified to include testing with the '.
fluidized-bed electrocoagulation cell design. Three phases of laboratory
experiments were undertaken to evaluate both electrocoagulation ufiits:
(1) preliminary screening experiments to demonstrate the feasibility of re-
ducing the concentration of each metal, (2) matrix experiments to define the
most opportune retention time and current (or current density), and (3) opti-
mization experiments to define other ACE Separator™ operating parameters
to achieve the most cost-effective removal conditions. The pH was adjusted
to 5,7, or 9 and the conductivity raised to approximately 1200 uS/cm (3,000
uS/in.)with sodium chloride. The conductivity of some surrogate wastes was
increased to approximately 3000 uS/cm (7,600 uS/in.) and subjected to elec-
trocoagulation. Surrogate wastes subjected to these experiments included
the five aqueous systems listed above as well as surrogate wastes containing
individual constituents such as nickel, zinc, copper, fluoride, and phosphate.
' Optimum operating conditions for the parallel-electrode unit were devel-
oped from these studies (Table A.I). These conditions served as the basis for
the subsequent pilot-scale tests. Similarly, the optimum operating conditions
for the fluidized-bed unit were 2.54 cm (1 in.) electrode spacing, 8- to +16-
mesh aluminum pellets size, and 20 amp current.
The bench-scale experiment conducted on the US EPA surrogate wastes
led to the following findings:
• when compared with alum treatment, electrocoagulation formed
approximately 83% less sludge volume and the filtration rate
unproved to 76%;
• for the fluidized-bed configuration, aluminum or stainless steel
may be used as electrode material with comparable results; and
A.6
-------
Appendix A
• with both increased frequency for the AC and increased retention
time, the agglomerated particles tend to disaggregate.
Pilot-scale tests were performed using both the parallel and fluidized-bed
configurations of the ACE Separator™. A 12 hour experiment using the ACE
fluidized bed separator™ was conducted on 208 L (55 gal) batches of surrogate
waste solution containing 0-2% (by weight) SSM fines, 0.5% (by weight) diesel
fuel, 0.05% (by weight) Alconox surfactant, and 10 mg/L each of Cu2+, Zn2+,
PO4^, P, and Ni2+. The conductivity and pH of the solution were raised to
1,200 uS/cm (3,000 uS/in.) and 7, respectively. The surrogate was recycled ''''
through a 10.2-cm (4-in.)-diameter, Schedule-80 PVC pipe, 61,0-cm '/
(24-in.)-high pilot-scale ACE Separator™ that was equipped witrftwo Type 316
stainless-steel electrodes 61-cm-high, 6.4-cm-wide (24-in.-high, 2.5-in.-wide)
and whose interior was filled with 8-to +16-mesh aluminum pellet^. The unit
was powered at a constant 20 amp, and the voltage was allowed/toVary as the
electrocoagulation treatment progressed over the 12 hour period. In this experi-
ment, the flow of the surrogate solution through the ACE Separator™ was varied
from 3.8 to 22.7 L/min-(l to 6 gal/min) and the quantity of compressed air in-
troduced into the solution feed line was a maximum of 10 psig (0.07 MPa).
Samples of the surrogate solution were collected at various times throughout the
experiment to document the rate of aluminum ion generation and the reductions
in concentration of the metal contaminants, chemical oxygen demand (COD),
and total suspended solids (TSS).
Table A.I
Optimum Operating Conditions for Parallel
Electrode Unit Based on Bench-Scale Tests
Parameter Value
Current 4 amp
Electrode Spacing ' 1.27cm (0.5 in.)
Retention Time 3 to 5 min
Frequency 10 Hz
Submergence , Fully submerged
Source: Barkfey, Farrell, and Williams 1993
A.7
-------
Ex-Situ Electrochemical Treatment Processes
In a similar pilot-scale test using the parallel-plate unit, the surrogate
waste was composed of essentially the same constituents as for the
fluidized-bed experiment. The notable changes were that the conductivity of
the solution was increased to approximately 3,000 uS/cm (7,600 uS/in.)and
no fluoride salt was added. The other operating parameters were based on
results obtained from the bench-scale tests. The aluminum generation and
consumption rates and the electrical power required to effect acceptable
phase separation as well as contaminant reductions were monitored.
Throughout the various phases of the experimental program, samples of' /
the treated effluent were collected and allowed to settle for 30 minutes. The
supernate was removed and analyzed. The subnate, containing the settled •
floe, was filtered and the filtrate and filter cake were analyzed.
-------
Appendix A
an ACE Separator™. Electrocoagulation improved the filtration rate of tita-
nium oxide by 63%. Other examples (for an oily emulsion and for biologi-
cal sludge) indicate highly enhanced filtration rates for electrocoagulated
wastewaters. Shear strength of an electrocoagulation floe is generally much
greater than the shear strength of an alum floe. Both sonic treatments (used
to evaluate the structural integrity of the floe) and actual filtration tests dem-
onstrated high shear strength of the electrocoagulated floes.
Electrocoagulation of metal- and phosphate-bearing industrial solutions
indicates excellent nickel, copper, and phosphate reductions. More than 90%"'
(concentration basis) of phosphate and copper can be removed from such solu-
tions at low aluminum and electric power requirements. Reduction in the nickel
concentration varies between 75% and 55% (concentration basis), i
Electrocoagulation of synthetic laboratory solutions and industrial waste-
water also confirmed the feasibility of using electrocoagulation for phos-
phate removal. Treatment of effluent from a commercial laundry reduced
the phosphate concentration (PO42-) from 45 mg/L to 5.4 mg/L after
low-intensity electrocoagulation (0.36 kW, 0.75 min retention time). Elec-
trocoagulation of process water from a phosphate mining operation reduced
the phosphate level by 91%, from 160 mg/L to 14 mg/L (3.3 kW, 0.17 min
retention time). Finally, treatment of dilute phosphoric acid solutions with a
nominal 100 mg/L total phosphate concentration and a conductivity of ap-
proximately 2000 nS/cm resulted in greater than 95% reductions in soluble
phosphate over a range of acidities.
A.I.1.3 Costs
Based on bench-and pilot-scale testing, projected treatment cost estimates
were developed. Overall treatment operating costs (electricity, aluminum
pellets, operation, and maintenance) will vary upwards from $0.13/1,000 L
($0.50/1,000 gal), depending on emulsion strength, unwanted component
concentration(s)(e.g., emulsifiers) in the effluent, and effluent TSS. Addi-
tional cost considerations may be involved in full-scale operation. Operator
supervision and maintenance would be limited to periodic replenishment of
the aluminum pellets, chemical pretreatment systems (e.g., salt addition to
enhance conductivity X and electrode replacement. Estimated operating costs
are based on laboratory and limited pilot-scale testing of effluents; currently,
these costs exceed those for comparable traditional chemical treatment (alum
or polyelectrolytes). The lower maintenance and operator supervision
A.9
-------
Ex-Sliu Electrochemical Treatment Processes
r
•equired for ACE Separator™ operation and the capability to use ACE Separa-
tor™ treated water in closed-loop, zero-discharge applications adds to its attrac-
tiveness. Successful commercialization requires further research to signifi-
cantly improve aluminum dissolution efficiency. If the ACE Separator™ can be
engineered to regularly generate sufficiently high aluminum dissolution concen-
trations, the technology may be applicable to industrial effluent treatment trains,
as well as to some site remediation activities.
The capital cost for a standard ACE Separator™ with a nominal through-
put capacity of 190 L/min (50 gal/min) is estimated by the vendor at
$80,000; for a 946 L/min (250 gal/min) throughput, $300,000. ^
A. 1.1.4 Conclusions t;
The technology offers a promising alternative for treating waste streams
containing clays, certain metal constituents, and other soluble plollutants.
As an alternative to chemical conditioning, the technology appears to have
an advantage over chemical coagulation because it does not add extraneous
soluble solids and because the sludge has a lower water content. As a re-
sult, the sludge from an electrocoagulation process has better filtering char-
acteristics than a sludge from a chemical coagulation process. The effec-
tiveness of electrocoagulation as compared to alum addition and polymer
coagulation offers:
• TSS: Electrocoagulation treatment and the polymer treatment
yielded equivalent results for the reduction of TSS in the treated
superaate. TSS values for alum treatment were four to five times
greater than those for ACE Separator™ treatment or polymer
treatment.
• Chemical Oxygen Demand (COD): Electrocoagulation resulted
in the highest COD reductions of the three methods. Removal
efficiency for COD was from two to four times higher than re-
moval efficiency for either alum treatment or polymer treatment.
• Lead: At high concentrations of lead, electrocoagulation achieved
approximately 55% removal of lead whereas, polymer treatment
showed higher removal (71%). Because some difficulties were
experienced with the alum treatment, these test results were invali-
dated. Electrochemical treatment of slurries with low concentra-
tions of lead resulted in the highest removal (96%).
A.10
-------
Appendix A
• Copper: The removal efficiency for copper in the supernate was
very high using electrocoagulation; 90% reduction was observed
when the concentration was high and 99% reduction was ob-
served when the concentration of copper was low. For high con-
centrations of copper, however, polymer and alum addition
achieved greater removal — virtually 100% reduction.
• Chromium: Electrochemical treatment resulted in good removal
for total chromium (87% and 94% reduction for the high and low,
concentrations, respectively). Alum and polymer addition ac:/','
complished similar removal.
Jk
• Cadmium: Cadmium levels in the supernate dropped as a result
of electrochemical treatment —14% in when the conpentration
was high and 99% in when the concentration was lo^. The in-
consistency between these two sets of experiment!?, as well as the
high concentrations remaining in the supernate and filtrates,
raised questions about the accuracy of the results for the high
concentration tests. For the low concentration tests, the cadmium
concentrations in both filtrates were much lower than the concen-
trations for either the alum or polymer-treated waters.
The following generalizations on the effectiveness of the ACE treatment
are made:
• ACE Separator™ treatment consistently reduces the TS and TSS
loadings to a degree equivalent with polymer treatment and to
approximately one-quarter the level achieved through alum addi-
tion; and
• better reductions in soluble metal concentrations are achieved
with electrocoagulation treatment than with alum treatment.
In summary, electrocoagulation is a promising, technically simple method
for achieving solid-liquid separations in aqueous-based waste streams. The
majority of the nontoxic, aluminum ionic species introduced will be re-
moved in the coagulated solids phase. The technology may be particularly
suitable for zero-discharge applications where the addition of chemicals and
the buildup of residual dissolved solids would adversely affect effluent qual-
ity or inhibit effluent reuse. Other potential applications include: (1)
remediation of groundwater and leachates (metals, COD/BOD removal),
(2) enhancement of clay separation from aqueous suspensions or emulsions
A.ll
-------
Ex-SItu Electrochemical Treatment Processes
resulting from soil washing operations, (3) breakage of oil/water emulsions
produced in the pumping of hydrocarbon-contaminated groundwater, (4)
removal of TSS from stormwater runoff, and (5) separation of oils and con-
taminants from thermal treatment system condensate.
A. 1.2 Andco's Electrocoagulation Pilot Study
Andco Environmental Processes undertook a pilot-scale study of elec-
trocoagulation to remove heavy metals and suspended solids from the . -
Milan Army Ammunition Plant (MAAP) located in the City of Milan • /
about 32 km (20 miles) north of Jackson, Tennessee. The MAAP has
' packaged and manufactured ordnance for over 50 years. At the tjme otf
operation, it was acceptable practice to rinse the packaging facilities
with a deluge of water to keep the facilities free of spilled expfosives.
The rinse water, with high concentrations of explosives, was^sent to
lined ponds. The ponds have since been capped and a carbon adsorption
treatment system is now being used to treat any packaging plant process
water. This pilot study was initiated to find a suitable technology for
cleaning up the large volume of groundwater that became contaminated
as a result of the pond water leaching into the ground.
The water to be treated was contaminated with both metals and organics
at concentrations above standard discharge limits. The pilot study consisted
of two independent processes. A pretreatment step for removal of metals
and any suspended solids by electrocoagulation and a secondary UV ozone
process for oxidation of organics. This discussion focuses on the electroco-
agulation system that was supplied by Andco to remove heavy metals, pri-
marily manganese and mercury, and suspended solids.
A. 1.2.1 Process Description
The Andco process electrochemically generates iron hydroxide from steel
electrodes. Through coprecipitation and adsorption, the iron hydroxide acts
to remove manganese, mercury, and other heavy metals from solution by
forming an iron hydroxide/heavy metal matrix. Electrochemical treatment is
followed by pH adjustment, clarification, and filtration.
A.12
-------
Appendix A
The electrochemical cell used in the pilot study consists of a fiberglass
body which supports and maintains a small gap between the sacrificial steel
electrodes. The system consists of two cells, each with 16 electrodes. A
direct electrical current is applied to the end electrodes and passed from
electrode to electrode throughout the process water flowing through the cell.
The electrical current causes water to break down into hydrogen gas and
hydroxyl ions with the simultaneous generation of ferrous ion from the steel
electrodes. The net effect is the formation of ferrous hydroxide. Hydrogen
gas is generated as a byproduct of the reaction and is released through a venY
on top of the cells. The process water is then passed through a retention / '
tank to remove the remaining entrained hydrogen bubbles. *•
From the retention tank the process water flows to the pH adjustanent tank
where a sodium hydroxide solution is added to increase the pH |o between
9.0 and 9.3. The pH of incoming groundwater during the tesjfc was typically
5.8. An increase in the process water pH occurs in the electrochemical cell.
The ferrous hydroxide generated in the cells is a weak base. Upon exiting
the cells, the pH of the test water was approximately 6.
Following the pH adjustment tank, the water flows by gravity to a corru-
gated, inclined-plate clarifier. A small amount of polymer flocculent is
mixed with the process water in the flash mix chamber to improve the set-
tling characteristics of the precipitated solids. Next, the water flows to the
flocculator, where a picket fence-type mixer gently agitates the solids, caus-
ing collisions that form larger solids. The solids settle to the bottom of the
clarifier, and the clear water overflows the effluent weir. Due to the low
solid content of the process water entering the clarifier, a sludge recycle
pump is included. The sludge recycle pump draws settled sludge from the
cone of the clarifier and pumps it to the flash mix chamber. This provides
the clarifier with a higher solids content to improve floe quality and clarifier
performance.
The settled solids are occasionally pumped from the cone of the clarifier
to the sludge holding tank before being sent to the filter press. The filter
press takes the 1-2% solids sludge from the sludge holding tank and pro-
duces a 25-30% solids filter press cake. The overflow from the clarifier
flows to a surge tank and is pumped through a polishing, multi media filtra-
tion system. The process water is sent to a treated water holding tank. Here,
the pH is adjusted to neutral before being sent to the UV ozone system, or
discharged to the treated water holding pool.
A.13
-------
Ex-Sltu Electrochemical Treatment Processes
A. 1.2.2 Treatment Levels
The treatment levels chosen for the pilot study were based on the result of
bench-scale treatment tests. For days 3 through 5, the treatment goal was 25
mg/L of iron. On day 6, a higher treatment goal of 50 mg/L was used.
Samples were taken regularly just after the electrochemical cell and the iron
level was determined using a HACH DR2000 Spectrophotometer. Table A.2
contains the average treatment levels based on the daily average iron genera-
tion. The iron generation was compared to the theoretical generation based f
on the cell amperage and flow to determine the efficiency. * /
Table A.2
Eectrochemlcal Iron Treatment Levels
Day
2
3
4
3
*
Treatment Level
IfpmTe)
-
28
32
32
50
E/C Cell Efficiency
(%)
.
135
157
157
J58
Source: Barktey, Farroll, and WJIiams 1993
On day 2, the iron generation was erratic and inconsistent; it was not
possible to accurately determine the average iron generation for that day. In
general, the iron generation was well below the theoretical level for the grab
samples collected on day 2. It appeared that the process water chemistry
was passivating the electrodes and reducing the cell's efficiency. On days 3
through 6, a salt (sodium chloride) was added to change the process water
chemistry. A small amount of salt greatly increased cell performance.
A.14
-------
I i
Table A.3
Electrochemical Precipitation — Days 2 and 3 Pilot-Scale Treatablility Data
Parameter Analyzed
DAY 2
Untreated
Water
PRETRT2
(pg/L)
Electrochemical
Precipitation
PRECIP2
(Hg/L)
DAYS
Untreated Water
PRETRT3
(Hg/L)
PRETRT3A
(Hg/L)
Electrochemical Precipitation
PRECIP3
Qig/L)
PRECIP3A
(Hg/L)
Treatment Goals
Reinjection
(Hg/L)
Surface
Water
Discharge
Oig/L)
TAL Metals
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Cyanide
Iron
Lead
Magnesium
Manganese '
Mercury
Nickel
Potassium
Selenium
Silver
<141
<38.0
<2.54
89.7
<5.0
<4.01
16,300
<6.02
<25.0
8.60
13.4
<38.8
'1.8
5,700
820
05
<34.3
1,310
<3.0
<4.60
<141
<38.0
<2.54
55.5
. <5.0
<4.01
16,100
<6.02
<25.0
<8.09
10.1
2^70
1.7
5,630
196
-------
I I
Table A.3 cont.
Electrochemical Precipitation — Days 2 and 3 Pilot-Scale Treatabllllty Data
Parameter Analyzed
DAY 2
Untreated
Water
PRETRT2
Oig/L)
Electrochemical
Precipitation
PRECIP2
(pg/L)
DAY 3
Untreated Water
PRETRT3
(Hg/L)
PRETRT3A
(Hg/L)
Electrochemical Precipitation
PRECIP3
(Hg/L)
PRECIP3A
(Hg/L)
Treatment Goal*
Reinjection
(jlg/L)
Surface
Water
Discharge
(Hg/L)
TAL Metals
Sodium
Thallium
^ Vanadium
O» Zinc
6,470
<81.4
<11.0
47.1
19400
<81.4
•cll.O
<21.1
7,100
<81.4
<11.0
60.3
6,730
<81.4
<11.0
31.6
46,200
<81.4
<11.0
<21.1
46,200
<81.4
<11.0
<21.1
NTG-
-
-
2,000
NTG
-
-
59
TCLVolatiles
Acetone
Benzene
Bromodichloromethane
Bromoform
Bromomethane
(2-Butanone) Methyl
Ethyl Ketone
Carbon Disulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
2-Chloroethylvinyl Ether
Chloroform
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
<13
<0.50
<0.59
<2.6
<5.8
<6.4
<0.58
<0.50
<1.9
<0.71
<0.50
<13
<0.50
<0.59
-------
Chloromethane
Dibromochloromethane
Dichloro benzene
1,1-Dichloro ethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethylcne
1 ,2-Dichloroethcne
1 ,2-Dichloropropane
Cis- 1 ,3-Dichloropropene
Trans- 1 ,3-Dichloropropene
Ethylbenzene
2-Hexanone
Methylene Chloride
Methyl Isobutyl Kelonc
Styrcne
1 ,1 ,2,2-Tetrachloroethane
Tetrachlo roethene
Toluene
l,l,i-Trichloroethane
1 ,1 ,2-Trichloroethane
Trichloroethene
Trichlorofluoromcthanc
Vinyl Acetate
Vinyl Chloride
Xylene
Acrolein
Acrylonitrile
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA '
NA
NA
NA
NA
NA
NA
NA
NA
<3.2
<0.67
<10
<0.68
<0.50
<0.50
-------
I J
00
Table A.4
Electrochemical Precipitation — Days 4, 5, and 6 Pilot-Scale Treatabllllty Data
Parameter Analyzed
DAY 4
Untreated
Water
PRETRT4
(WS/L)
Electrochemical
Precipitation
PRECIP4
(Hg/L)
DAYS
Untreated
Water
PRETRT5
(Hg/L)
Electrochemical
Precipitation
PRECIP5
(Hg/L)
DAY 6
Untreated
Water
PRETRT6
(Hg/L)
Electrochemical
Precipitation
PRECIP6
(Hg/L)
Treatment Goals
Reinsertion
(Hg/L)
Surface
Water
Discharge
Oig/L)
TAL Metals
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Potassium
<141
<38.0
<2.54
90.9
<5.0
<4.01
16,300
<6.02
<25.0
<8.60
17.2
<38.8
1.7
5,750
832
OA
<34.3
1,820
<141
<38.0
<2.54
30.4
<5.0
<4;01
15,400
<6.02
<25.0
<8.09
20.5
203
<1.3
5330
23.1
<0.2
<34.3
1,620
<141
<38.0
<2.54
93.8
<5.0
<4.01
16,700
<6.02
<25.0 "
<8.09
8.99
<38.8
<1.3
5,870
863
05
<34.3
1,600
<141
<38.0
<2.54
31.6
<5.0
<4.01
15500
<6.02
<25.0
<8.09
21.4
191
3.8
5,420
15.2
<0.2
<34.3
1,840
<141
<38.0
<2.54
87.8
<5.0
4.60
15,900
<6.02
<25.0
<8.09
118
<38.8
"**"- <1.3
552»~-
794 -
05
<34.3
1,740
<141
<38.0
EX06/24
23.6
<5.0
<4.01
14,600
<6.02
<25.0
<8.09
17.2
232
y EX06/24
4,020
13.7
<0.2
<34.3
i,oio .
NTG
-
0.0175
1,000
-
-
NTG
50
-
1^00
200
NTG
15
NTG
NTG
1.1
100
NTG
NTG
-
0.0175
1,000
-
-
NTG
11
-
6.54
5.2
300
1.32
NTG
50
0.012
88
NTG
1
I
o
Q
i
8
-------
I j
Selenium
Silver
Sodium
Thallium
Vanadium
Zinc
<3.0
<4.60
6,590
<81.4
<11.0
40.8
<3.0
<4.60
56,200
<81.4
<11.0
<21.1
<3.0
<4.60
6,930
<81.4
<110
43.2
<3.0
<4.60
57,000
<81.4
<11 0
<2U
<3.0
<4.60
6,260
<81.4
11 0
48.8
EX06/24
<4.60
43,200
<81.4
2u
.
50
NTG
2,000
12
NTG
59
Explosives
1 ,3-Dinitrobenzene
2,4-Dinitrotoluene
2,6-Dinitrotoluene
HMX
Nitrobenzene
RDX
Tetryl
r 1 ,3,5-Trinitrobenzcne
O 2,4,6'Trinitrotoluene
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
615
136
<0.738
744
414
4,110
<15.6
1,340
11,000
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
- NA
NA
NA
NA
• NA
NA
NA
NA
NA
NA
NA
NA
1
05
0.0068
400
17.5
2
NTG
2
2
1
05
0.0068
400
17.5
2
NTG
2
2
Other Parameters
Nitrate/Nitrite
Carbon
Total Organic
Total Inorganic
Purgeablc Organic
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
this chemical is not considered a contaminant of concern
NTG no treatment goal for this chemical
NA analyte not analyzed
Source: Laschinger 1932
21,000
12,400
2,400
<300
HSP* v
—»-et.
—
NA
NA
NA
NA
10,000/1,000
?
NTG
?
-------
Ex-SItu Electrochemical Treatment Processes
For days 3 through 6, the iron generation efficiency was well above
100%. In systems with low pH process water, iron generation efficiencies
greater than 100% are sometimes observed. The added iron is due to the
process water corroding the steel electrodes and, hence, dissolving additional
iron that is added by the electric current in the cell. The observed iron gen-
eration rates were still considered unusually high, even considering the pos-
sibility of corrosion. It is doubtful that the corrosion effect alone would have
such a significant effect on efficiency; other factors in the process water
probably contributed. '"*'
A, 1.2.3 Test Results * f
The treatment results for the electrochemical treatment are showii in
Tables A.3 and A.4. Results were collected for days 2 through 6 (jhe first
day was spent setting up the system). The heavy metal of primary* concern,
manganese, was to be reduced from an initial concentration of approxi-
mately 850 ug/L to less than 50 ug/L. The results for days 3 through 6 were
very good, the residual manganese ranged from 13.7 to 23.1 ug/L. Only on
day 2, was the result, 196 ug/L above the treatment goal. On this day, the
residual iron was 2570 ug/L, which was also above the treatment goal of 300
ug/L. Day 2 was the first day of the pilot system operation, and iron genera-
tion difficulties were encountered. The overall treatment level was insuffi-
cient to form a good floe, and as a result, a portion of the iron stayed in a
colloidal form and passed through the clarifier and multimedia filter. Also
on the first day of operation, insufficient solids had built up in the clarifier to
allow for sludge recirculation and increased floe quality. Precipitated man-
ganese was probably carried along with the iron through the system. For all
treated water tests, mercury was below detection limits of 0.2 ug/L.
An unusually high lead level, 3.8 ug/L, could have been due to an unusu-
ally high lead concentration in the influent, 1.8 ug/L. It could also have been
caused by a short upset in the system operating pH that allowed the lead
adsorbed on the iron floe to be released into the process water and show up
as a spike in the effluent. However, if this were the case, other heavy metals
would be expected to follow the same trend and this was not observed. The
result may also have been due to inaccurate analysis.
Another unexpected trend in the results was that, in general, the cyanide
level increased in the treated process water. No reaction occurred in the
electrochemical cell that would produce cyanide. However, the cyanide level
A.20
-------
Appendix A
was not a concern of the electrochemical treatment because it was to be
destroyed in the subsequent oxidation process.
From days 3 through 6, the residual iron level ranged from 191 to 254 \ig/
L. Although this was below the treatment goal of 300 ug/L, it was still more
than expected and more than Andco typically encountered for ferric precipi-
tation with polishing filtration. The expected residual iron concentrations is
in the range of 50 ug/L or less.
During the pilot study, visible particles were regularly seen in the effluent
of the multimedia filter The particles were large enough to be easily visible"
and should have been removed by the polishing filter. On two occasions,'the
filter was inspected and appeared normal. It is believed that apportion of, the
process water was bypassing the filtration bed due to the design of the multi-
media filter valve body. Better performance will be achieved with a
full-scale filter due to the use of discrete flow control valves and a deeper
media bed. •
A. 1.2.4 Chemical Consumption
The pilot study used steel electrodes, sodium hydroxide, polymer floccu-
lent, hydrochloric acid, sodium chloride, and hydrogen peroxide. The flow
rate through the electrochemical cells was determined using an industrial
rotameter with a stated accuracy of 15% of full-scale (Omega Model FL75).
A total of 67,090 L (17,725 gal) of contaminated groundwater was pro-
cessed by the treatment system. Although the pilot system was initially
operated at a 57 L/min (15 gal/min) feed rate, after two days of operation,
the feed rate was cut to 51 L/min (13.5 gal/min) to match the maximum
capacity of the pump supplying the contaminated water to the tank. Match-
ing the treatment flow rate with the feed rate enabled for continuous opera-
tion of the pilot system.
The chemical consumption for a full-scale system was calculated based
on the operation parameters of the pilotstudy. Table A.5 contains these val-
ues calculated on a per-million gallons (3,785,344 L) treated basis.
The solids generated and precipitated by the electrochemical process were
removed in the form of a filter press cake. The filter press cake was com-
posed mostly of ferric hydroxide and also included manganese, mercury,
other heavy metals, suspended solids, and other components adsorbed by the
ferric floe. While no leaching tests were performed on the sludge, the
A.21
-------
Ex-Sltu Electrochemical Treatment Processes
vendor contends it would pass the TCLP, and be classified as a nonhazardous
waste. At a treatment level of 25 mg/L of electrochemicaily-generated iron
and assuming that 10 mg/L of other components are removed from the pro-
cess water, 0.6 m3 (22 ft3) of sludge will be produced per million gallons
treated. This value assumes the filter can press cake to be 30% solids and
have a density of 1,281 kg/m3 (80 lb/ft3).
Table A.5
Chemical and Electrical Power Consumption
Per-MIIIion Gallons Treated. AndcoSystems
Item
Steel Electrodes at 28 mg/L Fe
Cell Power at 28 mg/LFe
Sodium Hydroxide
Polymer Flocculent
Hydrochloric Acid
Hydrogen Peroxide
Sodium Chloride
Power Consumption for Pumping
and Controls
Total
Usage
291 to
326 kWh
385 to
10 &
581)
20fc
188 to
1150kWh
Unit Cost
($)
0.039/lb
0.065/kWh
0.065/lb@50%
1.45/lb emulsion
0.12/lb@31%
0.236/lb@50%
0.034/lb
0.065/kWh
! Total Cost
($)
11.35
21.19
25.03
14.50
6.96
4.72
6.39
74.75
164.89
Source: Laschingw 1992
The treatment level on day 3, 28 mg/L Fe, was the minimum treatment
level used during the pilot-study, and at this level, the pilot system showed
good results. It is probable a full-scale system operated at somewhat less
than 25 mg/L Fe would meet treatment goals.
The electrodes used in the electrochemical process consisted of a
commercial-grade cold rolled steel. The estimated electrode consumption
A.22
-------
AppencKx A
assumes that the electrodes would be 80% consumed before replacement.
The cell power represents the electrical power consumed by the DC power
supplies in putting the 25 mg/L of iron in solution.
Sodium hydroxide was used for pH adjustment and for neutralization of
the acid wash solution. In a full-scale system, nearly all of the sodium hy-
droxide consumed would be used for pH adjustment. Only an estimated
11% would be used for acid wash solution neutralization. The groundwater
being treated in the tests had a pH of approximately 5.8. The iron added in
the electrochemical cells caused the pH to increase to approximately 6.5: %
the pH adjustment tank, sodium hydroxide was added to increase the pA to
a range of 9.2 to 9.5. Increasing the pH increased the efficiency of manga-
nese removal. The sodium hydroxide used in the pilot system w^s
pre-diluted to 6.9% to allow for better pH control. In a full-sc^e system,
sodium hydroxide would be added from a 50% solution. The sodium hy-
droxide consumption is based on the total amount consumed for pH adjust-
ment during the entire pilot study divided by the total gallons treated.
The polymer floccuient used was Andco 2600 high-molecular weight,
high-charge density emulsion. The polymer was diluted from its emulsion
form to a 0.2% solution for metering into the process. The volume of 0.2%
solution prepared and the volume remaining after the pilot study were re-
corded to determine the total consumption. The overall polymer addition
was 1.24 mg/L.
Hydrochloric acid was used to make up the acid wash solution. The acid
wash was used to recirculate a dilute acid solution through the electrochemi-
cal cells to remove any sludge and scale buildup from the electrodes. Ex-
cessive buildup in a cell can adversely affect the electrochemical reaction.
At the end of the acid wash sequence, the acid solution was recovered for
use in subsequent washes. In a full-scale system, the acid wash cycle is
automated, and the acid solution strength is checked regularly and replen-
ished on a monthly basis. Spent acid solution is neutralized and treated by
the system. The pilot system acid wash solution was prepared by mixing 95
L (25 gal) of water and 15 L (4 gal) of 31% hydrochloric acid. A daily acid
wash was performed at the start of each day. The acid consumption shown
in Table A.4 describes the predicted consumption for a full-scale system
operating at 1,900 L/min (500 gal/min). It is based on preparing a monthly
1,100 L (300 gal) batch of 10% hydrochloric acid plus an additional 50% to
account for acid that would be added during the month to maintain the
solution's strength.
A.23
-------
Ex-Situ Electrochemical Treatment Processes
A small amount of hydrogen peroxide was metered into the pH adjust-
ment tank to increase manganese removal efficiency. Hydrogen peroxide is
a strong oxidizer; it reacts with soluble manganese ions to form insoluble
manganese dioxide. In initial lab tests, 10 mg/L of hydrogen peroxide was
added to samples to ensure that all the iron generated by the electrochemical
cell and the manganese would be completely oxidized. During the pilot
study, the peroxide addition was originally set for 10 mg/L. This resulted in
excessive residual peroxide which decomposed in the clarifier and caused a
portion of the sludge to float. Therefore, the addition rate was decreased, t '
resulting in a peroxide residual of 0.5 mg/L as measured in the pH adjust-
ment tank The residual peroxide was regularly checked using ElfiQuant ,
Peroxide Test Strips. . *
It was originally expected that peroxide would be consumed by pie iron
as it is oxidized from its +2 to +3 state. The iron added by the ejectrochemi-
cal cell was +2 or ferrous state. The pilot study revealed that there was
enough dissolved oxygen and oxidizers in the process water to oxidize the
iron to its ferric state before it entered the pH adjustment tank. This resulted
in a lower-than-expected consumption of peroxide. The peroxide addition
pump was calibrated to produce a 0.5 mg/L residual in the pH adjustment
tank and remained at this setting until the final day of the pilot study when
the dosing was increased to produce 2 mg/L peroxide residual. Based on the
results, 0.5 mg/L residual peroxide achieved the desired treatment goals.
The peroxide addition rate was determined to be 1.24 mg/L hydrogen perox-
ide based on a pumping rate calibration.
Sodium chloride was used to change the conductivity of the process water
and, thereby, increase the iron generation efficiency. During the first day of
system operation, the iron generation was erratic and inconsistent. Indica-
tions were that the electrodes were being passivated by the components in
the process water. This situation is almost never encountered — the current
applied to the electrochemical cells provides a strong driving force for the
electrochemical reaction. Only under rare conditions has Andco encoun-
tered process water that can overcome the electrical driving force. The addi-
tion of a small amount of sodium chloride affects the conductivity of the
water and the surface reaction at the electrodes. Sodium chloride is conve-
nient to use because of its low hazard and low cost. In days 3,4, and 5 of the
pilot tests, sodium chloride was added to the process water at a rate of 56
mg/L. On the last day of the pilot study, the salt addition was decreased to
23 mg/L. The addition of the salt greatly unproved performance of the
A.24
-------
Appendix A
electrochemical cells. The performance did not decrease when the addition
rate was decreased from 56 mg/L to 23 mg/L. The chemical cost informa-
tion is based on the 23 mg/L addition. In a full-scale system, it is feasible
that the minimum amount required is below 23 mg/L.
The power consumption estimate is based on the electrical power required
for the process feed pumps, multimedia filter feed pumps, the pH adjustment
tank, clarifier mixers, and system controls. The process feed pumps and the
multimedia feed pumps consume approximately 85% of the electrical power.
The power consumption for the process feed pump is based on a design t '-*''
pump discharge pressure of 2 atm (30 psi). The multimedia filter pump is
based on a design discharge pressure of 2.4 atm (35 psi). The^power con-
sumption for an effluent pump was not included in this estimate, bbcause a
full-scale system would most likely use a gravity-flow discharge. In deter-
mining the power consumption by the pumps, a motor efficiency of 80% and
a pump efficiency of 60% were used.
A.2 Electrochemical Oxidation —
Silver (II) Process
The Dounreay Silver (H) Electrochemical Oxidation Process for the destruc-
tion of organic wastes arose as a result of studies on the dissolution of intrac-
table plutonium oxide residues created by the dissolution of nuclear (U, Pu)
oxide fuel in nitric acid (Batey 1995). These intractable plutonium oxide resi-
dues could be taken into acid solution for eventual plutonium recovery, but to
do so necessitated the use of particularly aggressive acid mixtures.
Experiments were performed using a simple, divided electrochemical cell
where a solution of silver nitrate and nitric acid was placed in the anode
compartment and nitric acid was placed in the cathode compartment. These
experiments demonstrated that, with the passage of an electric current, in-
tractable plutonium oxide residues dissolved rapidly. The Ag2+ ions gener-
ated at the anode were able to quickly oxidize the solid plutonium oxide to
soluble PuO22* and, at the same time, these ions were reduced to Ag* ions.
The Ag ions could then be re-oxidized at the anode to Ag (E[)+ which could
then react with insoluble material. The silver ions appeared to act as
electron-transfer agents between the electric power being fed to the cell and
A.25
-------
Ex-Situ Electrochemical Treatment Processes
the insoluble plutonium oxide, but were not consumed. This continuous use
of the silver oxidant has permitted the development of a practical process
that only requires the presence of a small amount of silver.
Based on the experiments, it was suggested that the Silver (H) would
probably react with organic matter contaminated with plutonium, such as
cellulose tissues used to clean up spills. Trials were carried out in which
plutonium-contaminated tissues were placed in the anode compartment of an
operating electrochemical dissolution cell. There was an immediate reaction
as demonstrated by the disappearance of the dark brown Ag2+ ions, resulting '
in a clear solution. The process continued until all the tissues were con-
sumed, whereupon the brown color of the Ag2+ ions again appeared. The ,
cellulose tissues were completely oxidized to carbon dioxide and wafer.
It was then a relatively simple step to examine the possibility of using the
process to destroy radioactive waste contaminated with tributylphqiphate/odor-
less kerosene solvent from nuclear fuel reprocessing plants. The initial stages
of these experiments in which solvent was added to the stirred compartment of
an electrochemical cell were not encouraging. However, as the temperature in
the cell increased due to the passage of the electric current, a reaction between
the Ag2+ ions and the solvent was observed. At 55°C (131°F), the reaction with
the electrochemically-generated Ag2+ ions resulted in the destruction of both the
tributylphosphate and the kerosene. Oxidation of kerosene was surprising be-
cause it usually does not react with oxidizing agents.
The electrochemical cell used to produce Ag2+ ions is of the
two-compartment type, with a fluoropolymer cationic-exchange membrane
separating the anolyte and catholyte sections. The membrane is necessary
because the reduced chemical species formed at the cathode, principally
nitrous acid, would otherwise react with the silver (II) ions produced at the
anode and reduce the efficiency of the destructive process. The anolyte is
stirred or circulated to ensure that silver (I) ions are brought efficiently to the
anode surface for oxidation to silver (H) ions; this transport process is the
rate-limiting step. The silver (H) ions then either react directly with the
organic material, or more likely, react with the water in the anolyte to form
radical species such as «OH, which then in turn react with the organic mate-
rial. The silver (H) ions are reduced to silver (I) ions in parallel with this
reaction and must be oxidized at the anode for the destruction process to
proceed to completion. In the case of the tributylphosphate/odorless kero-
sene solvent destruction, the final reaction products in the anolyte
A.26
-------
AppencKx A
compartment are carbon dioxide, phosphate ions, and protons (i.e., water is
consumed in the anolyte).
At the cathode, the nitric acid is reduced to nitrous acid (HNO2), NOx, and
water; the precise chemistry is determined by the choice of electrode mate-
rial. The formation of nitrous acid is the preferred reaction route as any
further reaction results in gassing due to NOX formation and may cause op-
erational difficulties. The nitrous acid generated at the cathode can be con-
' verted back into nitric acid and recycled by a regenerative catholyte circula-
tion system included in the process. ' 'f
•7 '
Two cell types that are manufactured by ICI, the filter presg design and
internally manifolded design, have been used to carry out the bulk of the
studies performed. Small-scale studies employed the FMO1, a ,,l/35th scale
model of the commercial-scale FM21SP electrochemical electrjblyser and
used a 60 amp bench-scale rig. This rig was used for the ma^rity of the
toxic organic destruction studies because of the small organic inventory
required for operation. Process-scale studies employed the FM21 SF cell in a
2000 amp pilot rig. This latter rig was used to demonstrate the destruction
on long runs of 10 days for tributyl phosphate/odorless kerosene and (up to 6
days) for organic ion-exchange resins.
The chemistry of the Silver (H) Process is summarized as follows:
1. At the anode, the silver (I) ions are oxidized to silver (II) ions:
6Ag+-»6Ag2++6e-
2. In the anolyte solution, the silver (n) ions react with water to
form oxidizing species (OH, «HO2, -NO3) represented by (O):
6Ag2+ + 3H2O -» 6Ag+ + 3[O] + 6 JT
3. The oxidizing species then react with the organics in the waste
stream that is introduced into the anolyte, oxidizing them to car-
bon dioxide, carbon monoxide, and water:
"CH2" + 3[0]->C02+H20
"CH2" represents a generalized carbon unit in an organic mol-
ecule, or more generally:
A.27
-------
Ex-Sltu Electrochemical Treatment Processes
Organics + [O] -> CO2 + CO + H2O + Inorganic Compounds
When nitrogen, phosphorus, sulphur, or chlorine are present in an organic
compound, these heteroatoms are oxidized to the mineral acid ion (e.g.,
nitrate, phosphate, sulphate, or chloride ions).
4. The silver (I) ions are then returned to the anode for reoxidization
to silver (H) ions to enable the reaction to continue.
5. The protons in the form of hydronium ions (H3O+) migrate across
the porous membrane to the cathode compartment under the
influence of the applied voltage. The protons are consumed, in
the cathode reaction along with the nitrate ions to form (mainly)
nitrous acid: , '
The catholyte solution containing the formed nitrous a^id is re-
generated by reaction with oxygen. Thus, the overafi stoichiom-
etry of the process is:
Organics + O2 -» CO2 + H2O + (Inorganic Compounds)
Construction and operation of the Silver (II) process is illustrated by
the following steps which refer to the simplified schematic of the pro-
cess, Figure A.2.
(a) Chemical agent and makeup/feed chemicals are.added to the
nitric acid/silver (I) nitrate solution which forms the anolyte
circuit (2) of the electrochemical cell (4).
(b) The anolyte solution is circulated through the electrochemical
cell (4) where silver ions are transformed into silver (fl) ions.
These silver (H) ions attack the organic chemical agent and
convert the organic chemicals to carbon dioxide, oxygen, trace
NOX (nitrogen oxides), protons, sulphate ions, phosphate ions,
nitrate ions, and silver chloride. In this reaction, the silver (II) ,
ions are reduced to silver (I) ions, which are recycled through
the electrochemical cell to continuously generate silver (II)
ions. Silver (I) ions, protons, and water diffuse through a cation
exchange membrane within the electrochemical cell (4) to enter
the catholyte circuit (3). The electrochemical cell (4) is the
heart of the process and is a type used extensively in the
chloralkali industry worldwide.
A.28
-------
Figure A.2
Simplified Schematic of the Silver (II) Process
I J
Gas Scrubbers
Anolyte Gas Condensers
IO
O
Feed Tanks
Catholyte and
NOX Circuit
Silver Recovery Unit
Anolyte Circuit
Electrochemical Cell—
Hold Up Tanks'
X*
Reproduced courtesy of AEA Technology (Oxfordshire, UK)
-------
Ex-SItu Electrochemical Treatment Processes
(c) The catholyte circuit supports the balancing cathode reaction
where nitric acid and protons are reduced to nitrous acid, NOx,
and water. The nitrous acid and NOX are oxidized to nitric acid
through reaction with oxygen and water. Excess water is re-
moved by distillation and sampled to confirm the absence of
chemical agents before discharge.
(d) Offgas from the anolyte circuit passes through a condenser (5)
to remove water and nitric acid vapors. Condensate is returned
to the anolyte circuit. The dried off gas stream is mixed with th,
offgas from the catholyte circuit and passed througha series of
scrubbers (6) and an active charcoal filter to remove residual /
NOX prior to discharge (6). { '
(e) At the end of a campaign, all of the solutions are discharged
from the Silver (II) plant to a silver recovery plant f7). The
final solutions are further tested for residual chemical agent
prior to discharge.
While there appears little doubt that the Silver (n) process can totally
oxidize organic compounds, the present systems are of limited size. The
modules (as claimed by the vendor) use standard off-the-shelf chlorine in-
dustry cells and can be run in parallel to create a unit of any size desired. As
such, the number of modules can be increased or upgraded as each applica-
tion requires. Despite the drawbacks at present, this is one of the few pro-
cesses that can oxidize organic compounds electrochemically at ambient
pressures and low temperatures.
•If
A.30
-------
Appendix S
LIST OF REFERENCES
Acar, Y.B. 1992. Electrokinetic soil processing (a review of the state of the art). Grouting, Soil, •» r
Improvement and Geosynthetics, Geotechnical Special Publication. 2(30): 1420-1423. , / -
Acar,Y.B. 1993. dean up with kilowatts. Soils. October 38-41.
*J^
Acar, Y.B. and A.N. Alshawabkeh. 1993. Principles of electrokinetic remediation. Environ Set &
Tech. 27(13): 2638-2647. \
Acar, Y.B., A.N. Alshawabkeh, and R.J. Gale. 1993. Fundamentals of extracting species from soils
by electrokinetics. Waste Management. 13:141-151. > '
r
Acar, Y.B., H. Li, and R.J. Gale. 1992. Phenol removal from kaolinite by electrokinetics J
Geotech.Eng. 118(11): 1837-1852.
Acar, Y.B., J. Hamed, R.J. Gale, and G. Putnam. 1991. Acid/base distribution in electro-osmosis
Transp. Res. Record. (1288): 23-34.
Acar, Y.B., RJ. Gale, A.N. Alshawabkeh, R.E. Marks, S. Puppala, M. Bricka, and R. Parker. 1995.
Electrokinetic remediation: basis and technology status. /. Hazardous Materials. 40: 117-137.
Allen, H.E. and P.H. Chen. 1993. Remediation of metal contaminated soil by EDTA incorporating
electrochemical recovery of metal and EDTA. Environmental Progress. 12(4): 284-293.
Alshawabkeh, A. and Y.B. Acar. 1992. Removal of contaminants from soils by electrokinetics: a
theoretical treatise. /. Envir. Sci. and Health, Part A. 27(7): 1835-1861.
American Public Health Association, American Water Works Association, and Water Environment
Federation. 1992. Standard Methods for the Examination ofWater andWastewater. 18th edition
Washington, DC.
American Society of Civil Engineers. 1992. Design of Municipal Wastewater Treatment Plants-
Volume I and II. WEF Manual of Practice No. 8, Water Environment Federation and ASCE Manual
and Report on Engineering Practice No. 76.
American Society of Civil Engineers and American Water Works Association. 1990. Water Treat-
ment Plant Design. 2nd edition.
Amirtharajah,A.andC.R.O'Melia. 1990. Coagulation processes: destabilization, mixing and
flocculation. Water Quality and Treatment, A Handbook of Community Water Supplies. American
Water Works Association. 4th edition. F.W. Pontius (ed.). New York: McGraw-Hill.
Baillod, C.R., B.M. Faith, and O. Masi. 1982. Fate of specific pollutants during wet oxidation and
ozonation. Environmental Program. 1(3): 217-227.
Barkely, N.P., C. Farrell, and T. Williams. 1993. Emerging Technology Summary, Electro-Pure
Alternating Current Electrocoagulation. US EPA. USEPA/540/S-93/504. September.
B.I
-------
List of References
Batey, B. 1995. The application of the electrochemical silver (II) process to the demilitarization of
chemical munitions. Paper presented at AEA Technology Conference on Alternative Technologies
for Chemical Weapon Demilitarization. Reston,VA. September 25-27.
Beeman, B. 1995. Personal communication with B. Beeman, Naval Communication Station, Stock-
ton, CA. December.
Benefield, L.D. and J.S.Morgan. 1990. Chemical precipitation. Water Quality and Treatment, A
Handbook of Community. Water Supplies. American Water Works Association. 4th edition. F.W.
Pontius (ed.). New York: McGraw-Hill.
BeUinger, J.A., E.D. Ferland, W. Killilea. 1994. Demonstration of the MODAR SCWO process.
Paper Presented at Waste Management Meeting. Tucson Arizona. February 28-March 4.
Blaney, C.A., L. Li, E.F. Gloyna, and S.U. Hossain. 1995. Supercritical water oxidation of
pulp and paper mill sludge as an alternative to incineration. Innovations in Supercritical
Eluids. American Chemical Society. Symposium Series 608. K.W. Hutchenson and»N.R.
Foster (eds.). Washington DC. .
Boltz, David F. and James A. Howell. 1979. Colorimetric Determination ofNonmetals. NewjYork:
John Wiley & Sons. \
Bruell, C.J., B.A. Segall, and M.T. Walsh. 1992. Electro-osmotic removal of gasolfne hydro-
carbons and TCE from clay. J. Env. Eng., ASCE. 118(1): 68-83.
Buxton, G.V., C.L. Greenstock, W.P. Helman, and A.B. Ross. 1987. Critical review of rate constants
for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals («OH/*O) in aqueous
solution. Journal of Physical and Chemical Reference Data. 17:313-886.
Cabrera-Guzman, D., J.T. Swartzbaugh, and A.W. Weisman. 1990. The use of electrokinetics for
hazardous waste site remediation. J. Air Waste Manag. Assoc. 40(12): 1670-1676.
Chen, C.T. 1995. Iron enhanced abiotic degradation of chlorinated hydrocarbons. Proceedings of
the 21st Annual RREL Hazardous Waste Research Symposium. USEPA/600/R-95/012. April.
Cipollone, M.G., S.M. Hassan, N.L. Wolfe, D.R. Bums, and P.M. Jeffers. 1997. Reduction of
halogenated hydrocrabons with iron: I. Kinetic observations. EPA Draft Report. Athens, GA.
Clifford, D.A. 1990. Ion exchange and inorganic adsorption. Water Quality and Treatment,
A Handbook of Community Water Supplies. American Water Works Association. 4th edition.
F.W. Pontius (ed.). New York: McGraw-Hill.
Cooper, WJ. 1996. Personal discussions with Leo Weitzman. April 12.
Cooper, W.J., T.D. Waite, C.N. Kurucz, M.G. Nickelsen, and K. Lin. 1992. An overview of
the use of high energy electron beam irradiation for the destruction of toxic organic chemi-
cals from water, wastewater, and waters containing solids. Draft paper provided by Dr.
William Cooper, High Voltage Environmental Applications, Inc. 9562 Doral Boulevard, Mi-
ami, FL 33178.
Cooper, WJ., E. Cadavid, M.G. Nickelsen, K. Lin, C.N. Kunicz, and T.D. Waite. 1993a. Removing
THMs from drinking water using high-energy electron-beam irradiation. Journal of the American
Waterworks Association. September. 106-112.
Cooper, WJ., D.E. Meacham, M.G. Nickelsen, K. Lin, D.B. Ford, C.N. Kurucz and T.D. Waite.
1993b. The removal of tri- (TCE) and tetrachloroethylene (PCE) from aqueous solution using
high-energy electrons. Journal of the Air & Waste Management Association. 43:1358-1366.
B.2
-------
Appendix B
•
>-'' ',' , ' ', '
Comwell,D.A. 1990. Air stripping and aeration. Water Quality and Treatment, A Handbook of
Community Water Supplies. American Water Works Association. 4th edition. RW. Pontius fed )
New York: McGraw-Hill.
Grain, N. and E.F. Gloyna. 1992. Supercritical water oxidation of 2,4-dichlorophenol and
pyndine. Paper presented at the Separations Research Program Fall Conference. The Univer-
sity of Texas at Austin. October 14-15.
DeU'Orco, P 1991. The separation of particles from supercritical water oxidation effluents. Masters
ITiesis. Civil Engineering Department, The University of Texas at Austin.
DeU'Orco, P, E.F. Gloyna, and M. Buelow. 1991. The separation of solids from supercritical water.
Supercritical Fluids Engineering. American Chemical Society.
DeU'Orco P., L. Li, and E.F. Gloyna. 1993. The separation of particulates from supercritical
«"«« ^l""*""' J- SeParation Science and Technology for Energy Applications'. '
-j): 625-642.
!™d,CRWelch- 1958' H^'nuisferto supercritical water. Transactions of ttfe
. 80(3): 746-52. j
Electro-Petroleum, Inc. Date unknown. Electrokinetic treatment of contaminated ^ils, sludges, and
lagoons. DOE/CH-9206. • , j
EnviroMetals Technologies, Inc. Undated. The EnviroMetal Process. Technical release. Mailing
address: 42 Arrow Road, Guelph, Ontario, Canada, N1K1S6.
Evans, 0. 1990. Estimating innovative treatment technology costs for the SITE program. Journal of
the Air and Waste Management Association. Volume 40, No. 7. July.
Felmy, A.R., D.C. Girvin, and E. A. Jenne. 1984. MINTEQ-A. A computer program for calculating
aqueous geochemical equilibria. USEPA-600/3-84-032. Athens, GA: US EPA.
Focht R., J. Vogan, and S.' F. O'Hannesin. 1996. Field application of reactive iron walls for in-situ
degradation of volatile organic compounds in groundwater. Remediation. Summer: 81-94.
Foussard, J.N., H. Debellefontain, and J. Besombes-Vailhe. 1989. Efficient elimination of organic
liquid wastes: wet air oxidation. J. Environmental Eng. 1 15(2): 367-385.
Carrels, R.M. and C.L Christ 1965. Ion exchange and ibn sensitive electrodes. Solutions, Minerals
and Equilibria . San Francisco: Freeman Cooper and Co.
Gehiinger, P., E. Proksch, W. Szinovatz, and H. Eschweiler. '1988. Decomposition of trichloroethyl-
ene and tetrachloroethylene in drinking water by a combined radiation/ozone treatment Water
Research. 22:645.
. 1993' lncineral'°n°fHazardous,Toxic.andMixedWastes. Cleveland
OH: North American Mfg. Co.
Gillham, R W. 1995. Resurgence in research concerning organic transformations enhanced by zero-
valent metals and potential applications in remediation of contaminated groundwater. Paper pre-
sented to the Division of Environmental Chemistry, American Chemical Society conference Ana-
heim, CA. April 2-7.
GUlham, R.W. and F. O'Hannesin. 1994. Enhanced degradation of halogenated aliphatics by zero-
valent iron. Ground Water. 32(6): 958-967.
Glaze, W.H. 1990. Chemical oxidation. Water Quality and Treatment, A Handbook of Community
Wate 'r Supplies. American Watei Works Association. 4th edition. F.W. Pontius (ed.). New York:
McGraw-Hill.
B.3
-------
List of References
Gloyna, E.F. and L. Li. 1993. Supercritical water oxidation: an engineering update. Paper pre-
sented at EPRI-NSF Symposium. San Francisco, CA. February 22-24.
Gloyna, E.F. and L. Li. 19.95. Supercritical water oxidation research and development updates.
Environmental Progress. 14(3): 182-192.
Griffith, J. and E.F. Gloyna. 1992. Supercritical water oxidation of o-cresol and methyl ethyl ke-
tone. Paper presented at the separations research program Spring Conference. The University of
Texas at Austin. April 14-15.
Hamed,J.,Y.B.Acar, and RJ. Gale. 1991. J.GeotechnicalErig. New York: ASCE.
Haroldsen, B.L., D.Y. Ariizumi, B.E. Mills, B.C. Brown, and D. Griesen. 1996. Transpiring wall
supercritical water oxidation reactor salt deposition studies. Sandia Report SAND96-8255«UC-702.'
September. ' / '
Hazlebeck, D. A., K.W. Downey, and A. J. Roberts. 1994. Development of prototype-hydrothermal
, oxidation systems for propellants, chemical agents, and other hazardous wastes. Extended abstractf
presented at the American Chemical Society. Atlanta GA. September 19-21. •
Helfferich, F. 1962. Ion Exchange. New York: McGraw-Hill. |
» *
Helling, R.K. 1986. Oxidation kinetics of simple compounds in supercritical wateip carbon monox-
ide, ammonia and ethanol. Ph.D. Dissertation. Massachusetts Institute of Technology, Cambridge,
Massachusetts.
Helling, R.K. and J.W. Tester. 1987. Oxidation kinetics of carbon monoxide in supercritical water.
J. Energy & Fuel. 1:417.
Hoasain, S.U. and C.A. Blaney. 1991. Method for removing polychiorinated dibensodioxins and
polychiorinated dibenzofurans from paper mill sludge. US Patent No. 5,075,017. December 24.
Holden, W., A. Marcellino, D. Valic, and A.C. Weedon. 1993. Titanium dioxide mediated photo-
chemical destruction of trichloroethylene vapors in air. Photocatalytic Purification and Treatment of
Water and Air. D.F. Ollis and H. Al-Ekabi (eds.). Elsevier Science Publishers B.V. pp 393 and 404.
Holgate, H.R., P.A. Webley, J.W. Tester, and R.K. Helling. 1992. Carbon monoxide oxidation in
supercritical water the effects of heat transfer and the water-gas shift reaction on observed kinetics.
Energy and Fuels. 6:586-97.
Holser, R.A., S.C. McCutcheon, and N.L. Wolfe. Undated-a. Mass transfer effects on the dechlori-
nation of trichloroethene by iron/pyrite fixtures in continuous flow column systems. Submitted for
publication. Copy can be obtained from S.C. McCutcheon at Ecosystems Research Division, Na-
tional Exposure Research Laboratory, U.S. Environmental Protection Agency, Athens, GA 30605.
Holser, R.A., S.C. McCutcheon, and N.L. Wolfe. Undated-b. The application of a rate expression
for the reductive dehalogenation of TCE-contaminated groundwater to the design and performance
of an in situ reactive zone. Submitted for publication. Copy can be obtained from S.C. McCutcheon
at Ecosystems Research Division, National Exposure Research Laboratory, U.S. Environmental
Protection Agency, Athens, GA 30605.
Hong, M.S., L. He, B.E. Dale, and K.C. Donnelly. 1995. Genotoxicity profiles and reaction charac-
teristics of potassium polyethylene glycol dehalogenation of wood preserving waste. Em. Sci. Tech.
29(3): 702-708.
Hutchenson, K.W., and N.R. Foster. 1995. Innovations in Supercritical Fluids. Washington, DC:
American Chemical Society.
Hunter, RJ. 1982. Zeta Potential of Colloid Science. London: Academic Press.
B.4
-------
Appendix B
Idaho National Engineering Laboratory (INEL). 1995. Corrosion investigation of multilayered
ceramics and experimental nickel alloys in SCWO process environments. INEL-95/0017.
™™™ r' and AlL Bourhis' 1992< Utiliz*ri°n of phenolics in the design of the
MODAR SCWO reactor. AIChE Annual Meeting. Miami, PL. November 1-6.
James M. Montgomery, Consulting Engineers, Inc. 1985. Water Treatment Principles and Design
New York: John Wiley & Sons. '
Jeffers, P.M, L.M. Ward, L.M. Woytowitch, and N.L Wolfe. 1989. Homogeneous hydrolysis rate
constants for selected chlorinated methanes, ethanes, ethenes, and propanes. Env. Science & Tech.
*
Johnson, T.L., M.M. Sheerer, and P.O. Tratnyek. 1996. Kinetics of halogenated organic compound r
degradation by iron metal. Environ. Sci. Technol. 30(8). , / .
Kahn, LI. and M.S. Alam. 1993. Heavy metal removal from soil by coupled electric-hydraulic
gradient. J. Env. Eng., ASCE. 120(6): 1524-1543. • /
Kalen, David V. 1992. Destruction of chemical warfare agents simulants in water solutions by
cobalt-60 gamma irradiation. Masters Thesis. Florida International University, Mi&ii, Florida.
Kimea William. 1996. Personal correspondence. From William Killea of Modar, General Atomics
to Leo Weitzman. F
Killilea, W.R., K.C. Swallow, and G.T. Hong. 1992. The fate of nitrogen in supercritical waste
oxidation. J. Supercritical Fluids. 5(1): 72-78.
Kurucz,C.N.,T.D.Waite,W.J. Cooper, and M.G. Nickelsen. 1991a. Full-scale electron beam
treatment of hazardous wastes —effectiveness and costs. Proc. 45th Industrial Waste Conference
(May 8-10, 1990). Chelsea, MI: Lewis Publishers.
Kurucz C.N.,T.D.Waite,W.J. Cooper, and M.G. Nickelsen. 199 lb. Advances in Nuclear Science
and Technology. Volume 22. J. Lewins and M. Becker (eds.). New York: Plenum Press.
Kurucz, C.N., T.D.Waite and W.J. Cooper. 1995. The Miami electron beam research facility: a large
scale wastewater treatment application. Radiation Physical Chemistry. 45(2 ): 299-308.
Kurucz, C.N.. T.D. Waite, W.J. Cooper, and M.G. Nickelsen. 1995. Empirical models for estimating
the destruction of toxic organic compounds utilizing electron beam irradiation at full-scale Radio-
tion Physical Chemistry. 45(5 ): 805-816.
Lageman.R. 1993. Electroreclamation: applications in the Netherlands. Environ. Sci. & Tech.
27(13): 2648-2650.
Laschinger, M.N. 1992. Pilot study report milan army anmuntion plant o-line ponds area. Study
performed by Andco Environmental Processes, Inc., 595 Commerce Drive Amherst, NY 14228.
Lee.D.S. 1990. Supercritical water of acetamide and acetic acid. PhD Dissertation. Civil Engi-
neering Department, The University of Texas at Austin.
Leith, D. and W. Licht; 1972. The collection efficiency of cyclone type particle collectors: a new
theoretical approach. American Institute of Chemical Engineers Symposium Series No. 126. 68:196.
Levenspiel. 1962. Chemical Reaction Engineering. New York: John Wiley & Sons.
Li, L., P. Chen, and E.F. Gloyna. 1991. Generalized kinetic model for wet oxidation of organic
compounds. AIChEJ. 37(ll):1687-97.
Lindgren, E.R., E.D. Mattson, and M.W. Kozak. 1994. Electrokinetic remediation of unsaturated
soils Emerging Technologies in Hazardous Waste Management IV. ACS Symposium Series 554
D.W. Tedder and F.G. Pohland (eds.). Washington, DC: American Chemical Society.
B.5
-------
List of References
Lindgrcn, E.R., M.W. Kozak, and E.D. Mattson. 1991. Electrokinetic remediation of contaminated
soils. Report No.: SAND-91-0726C. Sandia National Labs.
Lyons, T. 1995. Personal communication with T. Lyons, Risk Reduction Engineering Laboratory.
US EPA. August
Lyons, T. 1996. Personal communication with T. Lyons, National Risk Management Research
Laboratory. US EPA. January.
Marks, R.E. 1995. Personal communication, with R.E. Marks, Electrokinetics, Inc. August
Marks, R.E., Y.B. Acar, and R J. Gale. 1994. In situ remediation of contaminated soils containing
hazardous mixed wastes by bio-electrokinetic remediation and other competitive technologies.
Remediation of Hazardous Waste Contaminated Soils. D.L. Wise and D.J. Trantolo*(eds.). New
York: Marcel Dekker, Inc.
Matheson, LJ. and P.O. Tratnyek. 1994. Reductive dehalogenation of chlorinated methanes by iron
metal. Environ. Sci.& Tech. 28(12): 2045-2053. '.
Mattson, E.D. and E.R. Lindgren. 1995. Electrokinetic extraction of chromate from unsaturated
soils. Emerging Technologies in Hazardous Waste Management. ACS Series 607. D.y. Tedder and
F.G. Pohland (eds.). Washington, DC: American Chemical Society. j '
McAdams, W.H., W.E. Kennel, and J.N. Addoms. 1950. Heat transfer to superheated steam at high
pressures. Transactions of the ASME. 72(5):421-428.
McCutcheon, S.C. 1996. Personal communication with S.C. McCutcheon. US EPA National
Exposure Research Laboratory, Athens, GA. January.
Michna, R. 1990. Heat transfer to water in countercurrent flow within a vertical, concentric-tube
supercritical water oxidation reactor. Masters Thesis. Civil Engineering Department, The University
of Texas at Austin.
Mills, W.B., D.B. PorceUa, M.J. Ungs, S A. Gherini, K.V. Summers, L. Mok, G.L. Rupp, and G.L.
Bowie, 1985 (Revised). Water Quality Assessment: A Screening Procedure for Toxic and Conven-
tional Pollutants in Surface and Ground Water - Part I. USEPA/600/6-85/D02a.
Nickelsen, M.G., WJ. Cooper, C.N. Kurucz, and T.D. Waite. 1992. Removal of benzene and se-
lected alkyl-substituted benzenes from aqueous solution utilizing continuous high-energy electron
irradiation. Environnmental Science and Technology. 26: 144-152.
O'Hannesin, S.F. and R.W. Gillham. 1992. A permeable reaction wall for in situ degradation of
halogenated organic compounds. Paper presented at the 45th Canadian Geotechnical Society Con-
ference. Toronto, Ontario, Canada. October 25-27.
Palmer, C.D. and P.R. Wittbrodt 1991. Processes affecting the remediation of chromium-contami-
nated sites. Environmental Health Perspectives. 92:25-40.
Pamukca, S. and J.K. Wittle. 1992. Electrokinetic removal of selected heavy metals from soil.
Environ. Prog. 11(3): 241-250.
Parks, G.A. 1967. Equilibrium concepts in natural water systems. Advances in Chemistry Series
67. Washington, DC: American Chemical Society. p!21
Perry, R.H., C.H. Chilton, and S.D. Kirkpatrick. 1963. Chemical Engineers' Handbook, New York:
McGraw-HilL
PRC Environmental Management, Inc.. 1994. HVAEA electron beam technology demonstnation
final quality assurance project plan. Submitted to US EPA Office of Research and Development
(ORD). Cincinnati, OH. September.
B.6
-------
AppenoKx 6
Probstein, R.F. 1989. Physicochemical'Hydrodynamics: An Introduction. Boston and London:
Butterworth-Heinemann.
Probstein, R.F. and R.E. Hicks. 1993. Removal of contaminants from soils by electric fields. Sci-
ence. 260: 498-503.
Refer, C.K. and G.E. Strcit. 1989. Oxidation of Hydrocarbons and Oxyenates in Supercritical
Water, Phase II Final Report. LA-11700-MS, DOE/HWP-90. Los Alamos National Laboratory, Los
Alamos, NM.
Rogers, C.J. 1994. Basic chemistry of the ease catalyze decomposition process and its applications
nationally and internationally. Presented at the I&EC Special Symposium, American Chemical
Society. Atlanta, GA. September 19-21.
.'••f
Rollans, S., L. Li, and E.F. Gloyna. 1992. Separation of hexavalent and trivalent chromium frpm,
supercritical water oxidation effluents. Paper presented at the Industrial & Engineering Chemistry
Special Symposium, American Chemical Society. Atlanta, GA. September 21-2*.
Rousar, D.C., Marvin F. Young, and Scott Sieger. 1995. Development of Componentsjfor Waste
Management Systems Using Aerospace Technology. 46th International Astronautical Congress.
Oslo, Norway. American Institute of Aeronautics and Astronautics Paper #IAA 95-iA A. 1.2.07.
October 2-6. , »
r
Runnells, D.D. and C. Wahli. 1993. In situ electromigration as a method for removing sulfate,
metals, and other contaminants from ground water. Ground Water Monitoring & Remediation.
Winter: 121-129.
Scherer, M.M. and P.O. Tratnyk. 1995. Dechlorination of carbon tetrachloride by iron metal: effect
of reactant concentration. Paper presented to the Division of Environmental Chemistry, American
Chemical Society. Anaheim, CA. April 2-7.
Segall, B.A. and C.J. Bruell. 1992. Electroosmotic contaminant removal processes. /. Env. Eng.,
ASCE. 118(1): 84-100.
Senzaki, T. 1991. Removal of chlorinated organic compounds from wastewater by reduction pro-
cess: III treatment of trichloroethylene with iron powder II. KogyoYosui. 391: 29-35 (in Japanese).
Senzaki, T. and Y. Kumagai. 1988. Removal of chlorinated organic compounds from wastewater by
reduction process: treatment of 1,1,2,2-tetrachloroethane with iron powder. Kogyo Yosui. 357:2-7
(in Japanese).
Senzaki, T. and Y. Kumagai. 1989. Removal of chlorinated organic compounds from waste-
water by reduction process: II treatment of trichloroethylene with iron powder. Kogyo Yosui.
369: 19-25 (in Japanese).
Shanableh, A.M. 1990. Subcritical and supercritical water oxidation of industrial, excess activated
sludge. PhD Dissertation. Civil Engineering Department, The University of Texas at Austin.
Shapiro, A.P. and R.F. Probstein. 1993. Removal of contaminants from saturated clay by electroos-
mosis. Environ. Sci. & Tech. 27(2): 283-291.
Shapiro, A.P., P. Renauld, and R.F. Probstein. 1989. Preliminary studies on the removal of '
chemical species from saturated porous media by electroosmosis. Physicochemical Hydrody-
namics. 11 (5/6): 785-802.
Shoemaker, S. H., J. F. Greiner, and R. W. Gillham. 1995. Permeable reactive barriers. Chapter 11
of Summary Report from the International Workshop on Containment Technologies for Environmen-
tal Remediation Applications, Baltimore, MD. New York: John Wiley & Sons.
B.7
-------
List of References
Siddiqui. M., G.L. Amy, WJ. Cooper, C.N. Kurucz, T.D. Waite, and M.G. Nickelsen. 1996. Bro-
mate removal by the high energy electron beam process. Journal of the American Water Works
Association. 88(10): 90-101.
Sivavec, T.M. and D.P. Homey. 1995. Reductive dechlorination of chlorinated ethenes by iron
metal. Paper presented to the Division of Environmental Chemistry, American Chemical Society.
Anaheim, CA. April 2-7.
Shaw, R.S., T.B. Brill, A.A. Clifford, C.A. Eckert, and E.U. Franck. 1991. Supercritical water: a
medium for chemistry. C &. EN. December 23.
Snoeyink, V.L. 1990. Adsorption of organic compounds. Water Quality and Treatment, A Hand-
book of Community Water Supplies. American Water Works Association. 4th edition. F. W. Pontius (
(ed.). New York: McGraw-Hill. f/
Snoeyink, V.L. and D.Jenkins. 1980. Water Chemistry. New York: John Wiley & Sons.
Spinks, J.W.T. and R.J. Woods. 1990. An Introduction to Radiation Chemistry. New York: John /
Wiley & Sons. , *
Spritzer, M. 1993. Personal communication with Michael Spritzer, General Atomics. <|ctober.
Starr, R.C. and J.C. Cherry. 1993. Ground water currents. USEPA. USEPA/542/N£93/006. June.
State of Connecticut Department of Environmental Protection (CDEP). 1993. Personal Communi-
cation. From A. Lacobucci to US EPA regarding the acute toxicity of hydrogen peroxide to fresh-
water organisms.
Stradtting, J. 1989. A review of the hazards, safety aspects, design concepts, and contamination
control considerations relating to the safe use of liquid and gaseous oxygen. Prepared by Jack
Stradling at the NASA White Sands Test Facility. March 21.
Tester, J.W., H.R. Holgate, F.J. Armellini, P.A. Webley, W.R. KiUilea, G.T. Hong, and H.E. Bamer.
1993. Supercritical water oxidation technology: process development and fundamental research.
Emerging Technologies in Hazardous Waste Management If 1. ACS Symposium Series 518. D.W.
Tedder and F.G. Poland (eds.). American Chemical Society. 0097-6156/93/0518-0035.
Timberlake, D. 1993. Personal communication. Risk Reduction Engineering Laboratory,
USEPA. August.
Tongdhamachart, C. 1996. Supercritical water oxidation of anaerobically digested municipal
sludge. Doctoral Dissertation. Civil Engineering Department, The University of Texas at Austin.
Tratnyek, P.G. 19%. Putting corrosion to use: remediating contaminated groundwater wuth zero-
valent metals. Chemistry and Industry. July, pp 499-503.
Turner, M. D. 1993. Supercritical water oxidation of dimethyl methylphosphonate and thiodiglycol.
ARPA project sponsored. Doctoral dissertation. University of Texas at Austin. December.
U.S. Department of Energy (DOE). 1988. Radioactive Waste Management Order.. DOE Order
5820.2A. September.
USEPA. 1983. Methods for chemical analysis of water and wastes. Environmental Monitoring and
Support Laboratory. Cincinnati, OH. US EPA/600/4-79/020. March.
US EPA. 1987a. Alternate Concentration Limit (ACL) Guidance, Part I: ACL Policy and Informa-
tion Requirements. US EPA/530/SW-87/017.
US EPA. 1 987b. Joint US EPA-Nuclear Regulatory Agency Guidance on Mixed Low-Level Radio-
active and Hazardous Waste. Office of Solid Waste and Emergency Response (OSWER) Directives.
9480.00-14 (June 29), 9432.00-2 (January 8), and 9487.00-8 (August).
B.8
-------
8
nSri PAR 1988aC ?™°Colfora Chemical Treatment Demonstration Plan. Hazardous Waste Engi-
neenng Research Laboratoiy. Cincinnati, OH. April. 8
US EPA. 19&9CERCLA Compliance with Other Laws Manual: Part II Clean Air Act and Other
Environmental Statutes and State Requirements. OSWER. US EPA/540/G-89/006 Aug^l
uS l^^St^eth.^/orE^l^insSolidWaste. Volumes 1A-IC. SW-846. 3rd Edition.
update l. U5WbR. Washington, DC. November
ffm^I992' ^ ^r°n^eam trea'menl f°r removaloft^oroethyleneandtetrachioroethylene
trom streams and sludge. Emerging Technology Bulletin. BPA1J40/F-92/009. October.
UL^A' I??;, EleTn beamjreatment for *e removal of bentene and toluene from aqueous / '
streams and sludge. Emerging Technology Bulletin. US EPA1540/F-93/302. April.
US EPA. 1994 Electrokinetics Inc. (Electro-Klean™ Electrokinetic soil processTng). SupeMnd
™™™ChnoloEvl 7th edition. USEPA!/54?/R £/
n9!5a' /W"5/r" Remediati°n Technol°gy Status Report: Treatment Walls. iUSEPA542-K-
^L^'}?95b' Met^-e"h^ced abiotic degradation technology. EnviroMetal Technologies, Inc.
SuperfundlnnovativeTechnology Evaluation (SITE) Demonstration Bulletin. In publication.
S^Jl9^' EpeCtr0kin^CS S°a Proc«sing (Electrokbetics Inc.). Superfund Innovative Tech-
nology Evaluation Program; Emerging Technology Bulletin. USEPA/540/F-95/504.
US EPA. 1995d National National Risk Management Research Laboratory (NRML), High Voltage
Environmental AppLcaUons Inc. - Electron Beam Technology, Innovative Technology EvaluS
Report. June. Draft Report Avadable from NRML/SITE Program, US EPA, Cincinnati, OH 45268
Ugaz, A., S Puppala, R.J. Gale, and Y.B. Acar. 1994. Electrokinetic soil processing. Complicating
featoes of electrokmet.c remediation of soils and slurries: saturation effects and the role of me
cathodic electrolysis. Chem. Eng. Comm. 129: 183-200.
Vogan John. 1996. Personal communication. John Vogan of EnviroMetal Technologies with B
Mm. January 19.
'R °'Hannesin' «* W.H.Mautulewic2. 1995. Site specific degradation
sed, u > EleCtro° ^^ systems for treatm^t of water and
slumed sods contaminated w.th tox.c organics and ordinance residues. Paper presented at 17th
" 5 ^ang' CN' KU™CZ' ^ WJ- Co°P<*- ^996. Electron beam treatment of bisolids:
enhancement- Submitted to Journal of Environmental Engineer-
hterecie^ce Jn '^ Physicochemical Processes for Water Quality Control. New York: Wiley-
B.9
-------
List of References
Webley, RA., J.W. Tester, and H.R. Holgate. 1991. Oxidation kinetics of ammonia and ammonia-
methanol mixtures in supercritical water in the temperature range 530-700°C at 246 bar. Ind. Eng.
Chem.Res. 30(8): 1745-1754.
Weismantel, G. 1996. Supercritical water oxidation treats toxic organics in sludge. Environmental
Technology. 6(5). September-October.
Weitzman, Leo, Kimberly Gray, Frederick K. Kawahara, Robert W. Peters, and John Verbicky. 1994.
Innovative Site Remediation Technology — Chemical Treatment. Annapolis, MD: American Acad-
emy of Environmental Engineers.
White, G.C. 1972. Handbook ofCMorinationfor Potable Water, Wastewater, Cooling Water, Indus-
trial Processes, and Swimming Pools. New York: Van Nostrand Reinhold. . •'
Wightman, T.J. 1981. Studies in supercritical wet air oxidation. Masters Thesis. Chemical Engf-/
neering Department, University of California, Berkeley, CA. . ^
Wilson, RK. 1995. Zero-valent metals provide possible solution to groundwater problems. Chemi-
cal & Engineering News. July 3:19-22. {
Yamagata, K., K. Nishikawa, S. Hasegawa, T. Fuji, and S. Yshida. 1972. Forced corrective Seat
transfer to supercritical water flowing in tubes. Inter. J. Heat and Mass Transfer. |5:2575-93.
Yamane, C. L., S.D. Warner, J.D. Gallinatti, F.S. Szerdy, T.A. Delfino, D.A. Hankins, and J.L. Vogan.
1995. Installation of a subsurface groundwater treatment wall composed of granular zero-valent
iron. Paper presented to the Division of Environmental Chemistry, American Chemical Society
conference. Anaheim, CA. April 2-7.
Zhang S. and J.F. Rusling. 1995. Dechlorination of PCB on soils and clay by electrolysis in a
bicontinuous microemulsion. Environ. Sci. Technol. 29:1195-1199.
B.10
-------
Appendix C
SUGGESTED READING LIST
Bechtold, J.K. and S.B. Margolis. 1992. The structure of supercritical diffu^
sion flames with arrhenius mass diffusivities. Combust. Sci. and Tech. , /,
83(257).
Jtk
Bergan, N.E., P.B. Butler, and H.A. Dywer. 1991. High pressure thermddy-
namics in supercritical water oxidation processes. Presented at {he 2nd In-
ternational Symposium on Supercritical Fluids. Boston, MA., ft|ay 20-22.
Bramlette, T.T., B.E. Mills, K.R. Mencken, M.E. Brynildson, S.C. Johnston,
J.M. Hruby, H.C. Feemster, B.C. Odegard, and M. Modell. 1991. Destruc-
tion of DOE/DP surrogate wastes with supercritical water oxidation technol-
ogy. Sandia Report SAND90-8229.
Brown, M.S. and R.R. Steeper. 1991. CO2-based thermometry of
supercritical water oxidation. Appl. Spectrosc. 46(1733).
Butler, P.B., N.E. Bergan, T.T. Bramlette, WJ. Pitz, and C.K. Westbrook.
1991. Oxidation of hazardous waste in supercritical water: a comparison of
modeling and experimental results for methanol destruction. WSS/CI91-8.
Presented at the 1991 Spring Meeting of the Western States Section of the
Combustion Institute. Boulder, CO. March 17-19.
Chan, J.P.C., C.A. LaJeunesse, and S.F. Rice. 1994. Experimental tech-
niques to determine salt formation and deposition in supercritical water oxi-
dation reactors. Presented at the 1994 International Mechanical Engineering
Congress and Exposition. Chicago, IL. November 6-11.
Croiset, E. and S.F. Rice. 1996. Hydrogen peroxide decomposition in
supercritical water. Submitted for publication to AJChE Journal. December.
Hanush, R.G., S.F. Rice, T.B. Hunter, and J.D. Aiken. 1996. Operation and
performance of the supercritical fluids reactor (SFR). Sandia National Labo-
ratories Report, SAND96-8203. Livermore, CA.
C.I
-------
Suggested Reading List
Haroldsen, B.L., D.Y. Ariizumi, B.E. Mills, E.G. Brown, and D. Greisen. 1996.
Transpiring wall supercritical water oxidation reactor salt deposition studies.
Sandia National Laboratories Report, SAND96-8255. Livermore, CA.
Hunter,T.B., S.F. Rice, and R.G. Hanush. 1996. Raman spectroscopic mea-
surement of oxidation in supercritical water n. Conversion of isopropanol to
acetone. Industrial and Engineering Chemistry Research. 35: 3984-3.990.
Johnston, S.C. and C.E. Tyner. 1990. Thermochemical waste processing
technology at Sandia National Laboratories. Sandia Report SAND90-8692.. <
LaJeunesse, C.A., B.E. Mills, and B.G. Brown. 1994. Supercritical water
oxidation of ammonium picrate. Sandia Report, SAND95-8202! /
LaJeunesse, C.A., B.L. Haroldsen, S.F. Rice, and B.G. Brown. 1$96. Hy-
drothermal oxidation of navy shipboard excess hazardous materi^s. Sandia
National Laboratories Report. December. In preparation. I
LaJeunesse, C.A., S.F. Rice, R.G. Hanush, and J.D. Aikeri. 1993. Salt depo-
sition in a supercritical water oxidation reactor (Interim Report). Sandia
Report, SAND94-8201.
LaJeunesse, C.A., J.P. Chan, T.N. Raber, D.C. Macmillan, S.F. Rice, and
K.L. Tschritter. 1993. Supercritical water oxidation of colored smoke, dye,
and pyrotechnic compositions. Final Report: Pilot Plant Conceptual De-
sign. Sandia Report, SAND94-8202.
LaJeunesse, C.A., S.F. Rice, J.J. Bartel, M. Kelley, C.A. Seibel, L.G. Hoffa,
T.F. Eklund, and B.C. Odegard. 1992. A supercritical water oxidation reac-
tor: the materials evaluation reactor (MER). Sandia Report, SAND91-8623.
Margolis, S.B. and S.C. Johnston. 1989. Multiplicity and stability of
supercritical combustion in a nonadiabatic tubular reactor. Combust. Sci.
andTech. 65(103).
Margolis, S.B. and S.C. Johnston. 1990. Nonadiabicity, stoichiometry, and
mass diffusion effects on supercritical combustion in a tubular reactor.
Symp. (Int.) Combustion (Proc.) 23rd. Volume 533.
Margolis, S.B. and S.F. Rice. 1991. On completeness of combustion in an
isothermal flow reactor. Combust. Sci. and Tech. 78(7).
Melius, C.F., N.E. Bergan, and J.E. Shepherd. 1990. Effects of water on
combustion kinetics at high pressure. Symp. (Int.) Combustion (Proc.) 23rd.
Volume 217.
C.2
-------
Appendix C
Rice, S.F., C.A. LaJeunesse, and R.R. Steeper. 1996. Optical cells for high-
temperature and high-pressure raman spectroscopic applications. December.
In preparation.
Rice, S.F., R.R. Steeper, and C.A. LaJeunesse. 1993. Destruction of repre-
sentative navy wastes using supercitical water oxidation. Saridia Report,
SAND94-8203.
Rice, S.F., R.R. Steeper, and C.A. LaJeunesse. 1993. Efficiency of
supercritical water oxidation for the destruction of industrial solvent waste..,,.
Paper #22.2. Presented at the I&EC Special Symposium of the American/
Chemical Society. Atlanta, GA. September 27-29.
Rice, S.F., T.B. Hunter, and R.G. Hanush. 1996. Oxidative reactivity of*
simple alcohols in supercritical water using in situ raman specte0scopy. Pre-
sented at the Second International Symposium on Environmental Applica-
tions of Advanced Oxidation Technologies. San Francisco, (*A. February
28-March 1. Paper to appear in EPRI proceedings.
Rice, S.F., T.B. Hunter, A.C. Ryden, and R.G. Hanush. 1996. Raman spec-
troscopic measurement of oxidation in supercritical water I. Conversion of
methanol to formaldehyde. Industrial and Engineering Chemistry Research
35:2161-2171.
Rice, S.F., C.A. LaJeunesse, R.G. Hanush, J.D. Aiken, and S.C. Johnston.
1994. Supercritical water oxidation of colored smoke, dye, and pyrotechnic
compositions. Sandia Report, SAND94-8209.
Rice, S.F., R.G. Hanush, T.B. Hunter, R.R. Steeper, J.D. Aiken, E.
Croiset, and C.A. LaJeunesse. 1996. Kinetic investigation of the oxida-
tion of naval excess hazardous materials in supercritical water for the
design of a transpiration-wall reactor. Sandia National Laboratories
Report. December. In press.
Schmitt, R.G., P.B. Butler, N.E. Bergan, WJ. Pitz, and C.K. Westbrook.
1991. Destruction of hazardous waste in supercritical water JJ: a study of
high-pressure methanol oxidation kinetics. Presented at the Fall Meeting of
the Western States Section of the Combustion Institute. Los Angeles, CA.
October 13-15.
Steeper, R.R. and S.F. Rice. 1993. Supercritical water oxidation of hazard-
ous wastes. AIAA-93-0810. 31st Aerospace Sciences Meeting. Reno, NV.
January 11-14.
C.3
-------
Suggested Reading List
Steeper, R.R. and S.F. Rice. 1994. Optical monitoring of the oxidation of
methane in supercritical water. Presented at the Spring Meeting of the West-
ern States Section of the Combustion Institute. Davis, CA. March 21-22.
Steeper, R.R. and S.F. Rice. 1994. Optical monitoring of the oxidation
of methane in supercritical water. Presented at the Twelfth International
Conference on the Properties of Water and Steam. Orlando, FL. Sep-
tember 11-16.
Steeper, R.R. and S.F. Rice. 1995. Optical monitoring of the oxidation of .
methane in supercritical water. Physical Chemistry of Aqueous Systems. • I
H.J., White, J.V. Sengers, D.B. Neumann, and J.C. Bellows (eds^. New
York: Begell House, p 652. {
Steeper, R.R. and S.F. Rice. 1996. Kinetics measurements of methane in
supercritical water. Journal of Physical Chemistry. 100: 184-119.'
Steeper, R.R., J.D. Aiken, and S.F. Rice. 1996. Kinetics of the water-gas
shift reaction in supercritical water and high pressure steam. December. In
preparation.
Steeper, R.R., S.F. Rice, M.S. Brown, and S.C. Johnston. 1992. Methane
and methanol diffusion flames in supercritical water. /. Supercritical Fluids.
5(262).
Steeper, R.R., S.F. Rice, M.S. Brown, and S.C. Johnston. 1992. Methane
and methanol diffusion flames in supercritical water. Sandia Report SAND
92-8474.
C.4
-------
I 1
\
-------
THE WASTECH® MONOGRAPH SERIES (PHASE II) ON
INNOVATIVE SITE REMEDIATION TECHNOLOGY:
DESIGN AND APPLICATION
This seven-book series focusing on the design and application of innovative site remediation
Tchnlgies follows an earlier series (Phase I, 1994-1995) which cover the process *^£™;
evaluations, and limitations of these same technologies. The success of that series «f nnhhcahon
suggested that this Phase II series be developed for practitioners in need of design i
and applications, including case studies.
WASTECH* is a multiorganization effort which joins in partnership the Air and Waste Manage-
ment Association, the American Institute of Chemical Engineers the American Society of Civil
Engineers, the American Society of Mechanical Engineers, the Hazardous Waste Action
Codition. the Society for Industrial Microbiology, the Soil Science Society of America, and <,
the Wate Environment Federation, together with the American Academy of Environmental ,
Engineers, the U.S. Environmental Protection Agency, the U.S. Department of Defense, and M
U.S..Department of Energy. / .
A Steering Committee composed of highly respected members of each participating organization
with expertise in remediation technology formulated and guided both phases, with groject
management and support provided by the Academy. Each monograph was prepared by a Task
SroulTo^recognized Vxperts. The manuscripts were subjected to extensive peerifeviews prior to
publication. This Design and Application Series includes:
Vol 1 - Bloremedlation
Principal authors: R. Ryan Dupont, Ph.D., Chair,
Utah State University; Clifford J. Bruell, Ph.D.,
University of Massachusetts; Douglas C. Downey,
Parsons Engineering Science; Scott G. Ruling,
USEPA; Michael C. Marley, Ph.D., Environgen, Inc.;
Robert D. Morris, Ph.D., Eckcnfelder, Inc.; Bruce
Pivetz. USEPA.
Vol 2 - Chemical Treatment
Principal authors: Leo Weitzman, Ph.D., LVW
Associates. Chair, Irvin A. Jefcoat, Ph.D., University
of Alabama; Byung R. Kim, Ph.D., Ford Research
Laboratory.
Vol 3 - Uquld Extraction Technologies:
Soil Washing/Soil Rushing/Solvent Chemical
Principal authors: Michael J. Mann, P.E., DEE,
Alternative Remedial Technologies, Inc., Chair,
Richard J. Ayen, Ph.D., Waste Management Inc.;
Lome G. Everett, Ph.D., Geraghty & Miller, Inc.;
Dirk Gombert II, P.E., LIFCO; Mark Meckes,
USEPA; Chester R. McKee, Ph.D., In-Situ, Inc.;
Richard P. Traver, P.E., Bergmann USA; Phillip D.
Walling, Jr., P.E., E. I. DuPont Co. Inc.
Vol 4 - Stabilization/Solidification
Principal authors: Paul D. Kalb, Brookhaven National
Laboratory.'C/jaJn Jesse R. Conner, Rust Remedial
Services, Inc.; John L. Mayberry. SAIC; Bhavesh R.
Patcl, Brookhaven National Laboratory; Joseph M.
Perez, Jr., Battellc Pacific Northwest; Russell L.
Treat, Foster Wheeler Environmental Corp.
Vol 5 • Thermal Desorption
Principal authors: William L. Troxler, P.E., Focus
Environmental Inc., Chair, Edward S. Alperin, IT
Corporation; Paul R. de Percin, USEPA; Joseph H.
Hutton, P.E., Canoniti Environmental Services, Inc.;
JoAnn S. Lighty, Ph.D., University of Utah; Carl R.
Palmer, P.E., Rust Remedial Services, lac.
Vol 6 - Thermal Destruction
Principal authors: Francis W. Holm, Ph.D., SAIC, Chair,
Carl R. Codey, Department of Energy; James J.
Cudahy, P.E., Focus Environmental Inc.; Clyde R.
Dempsey, P.E., USEPA; John P. Longwell, ScJ>.,
Massachusetts Institute of Technology; Richard S.
Magee, ScJ>., P.E., DEE, New Jersey Institute of
Technology; Walter G. May, ScJX, University of Illinois.
Vol 7 - Vapor Extraction and Air Sparging
Principal authors: Timothy B. Holbrook, P.E., Camp
Dresser & McKee, Chair, David H. Bass, Sc.D.,
Groundwater Technology, Inc.; Paul M. Boersma,
CH2M Hill; Dominic C. DiGuilio, University of
Arizona; John J. Eisenbeis, Ph.D., Camp Dresser &
McKee; Neil J. Hutzler, Ph.D., Michigan Technologi-
cal University; Eric P. Roberts, P.E., ICF Kaiser
Engineers, Inc.
The monographs for both the Phase I and Phase II
series may be purchased from the American Academy
of Environmental Engineers*. 130 Holiday Court. Suite
100. Annapolis. MD. 21401; Phone: 410-266-3390,
Fax: 410-266-7653. E-mail: aaee@ea.net •
------- |