S S Manual of Practice

Download as pdf or txt
Download as pdf or txt
You are on page 1of 603

Stabilization and Solidification of Contaminated Soil

and Waste: A Manual of Practice


hjui

Edward Bates and Colin Hills


Stabilization and Solidification of Contaminated Soil and Waste:
A Manual of Practice

Copyright 2015 Edward Bates & Colin Hills

All rights reserved.


Edward Bates and Colin Hills retain copyright to this document, but grant the
United States Environmental Protection Agency (USEPA) the rights to post this
electronic version of this document on the USEPAs CLU-IN website (www.clu-
in.org), with the understanding that it will be made available to the public without
charge, for viewing or download, in total or in part. The availability of this work on
CLU- IN is for informational purposes, consistent with the purpose of the central
reference library, under U.S. Title 42 9660 (b)(8), and does NOT constitute an
endorsement by the USEPA. Permission is granted by the authors to refer to or
quote from this publication with the customary acknowledgment of the source.

Although every precaution has been taken in the preparation of this document,
the authors assume no responsibility for errors or omissions. Neither is any
liability assumed for damages resulting from the use of this information contained
herein.

www.hyggemedia.com

1
Preface
Stabilization/Solidification (S/S) is a remedial technology that has attracted
widespread use around the world. S/S employs readily available cement-based
binders that can be formulated into a targeted system that cements soil or waste
into a hardened engineering material. The application of S/S results in the physical
and chemical stabilization of contaminated material over the long term, enabling
disposal or re-use on- or off-site.

In this second volume on S/S, the authors present the state of the art of the
application of S/S in the field. As such, this Manual of Practice is designed for use
by regulators, professionals and informed individuals. Considering the multinational
contributions to this work, the reader may note that sometimes words may use the
United Kingdom English spelling, for example stabilization rather than the
conventional American English spelling stabilization. However, in all cases, the
meaning should be clear.

This Manual has been prepared in close consultation with experienced practitioners
and vendors of S/S who apply this remedial technology on a day-to-day basis. As
such, this Manual will provide an authoritative reference source for all who are
involved in managing the risks associated with contaminated soil and waste.

The authors are indebted to the hard work and professionalism of the contributors to
this work, and it is their invaluable contributions that have made this Manual of
Practice possible.

The companion volume to this work entitled: Stabilization and Solidification of


Contaminated Soil and Waste: Science draws extensively from the scientific
literature and personal experience of the authors, and is intended to deliver a readily
accessible account of the science underpinning S/S. Both volumes should be seen
as complimentary, providing invaluable insight into why S/S is a versatile risk
management strategy, fit for the 21 st Century.

Edward Bates and Colin Hills


Cincinnati, Ohio University of Greenwich, UK

September 2015
2
Acknowledgments
The contributing editors wish to express their most sincere gratitude to all the expert
authors who have given their valuable time to the preparation of material for this
document.

Each author is gratefully acknowledged and is listed in the table below with the
subject matter of their contribution. Information on each contributing author can also
be found in Section 11 Meet the Authors, where their particular expertise,
experience and present professional affiliation are given.

The Contributing Editors particularly wish to acknowledge the substantial contribution


made by Dr Peter Gunning to the completion of this Manual of Practice. Peters
unselfish support and encouragement throughout was exemplary, and was a major
factor in the successful and timely completion of this work.

The authors would also like to offer special thanks to Miss Esme Hills for her steady
eye, and to Mr Stany Pensaert for his constructive comments and invaluable
insights gained as a practitioner of S/S of Contaminated Soil and Waste.

3
Summary
This manual of practice brings together the collective experience of leading
practitioners in the design and execution of remedial actions employing S/S.

This manual presents the state of the art in the application of S/S for the remediation
of contaminated land and treatment of waste.

The information presented is, to the contributing editors` knowledge, not available in
any single document anywhere at the current time.

The simple structure of this manual is designed to inform the reader on the
background of S/S, and then to the site assessment of risk, the design and selection
of binders and the choice of equipment, to field-execution and product quality
control/assurance, and finally to capping and post-remedial management of sites
that have been treated using S/S.

It is intended that this manual will provide an accessible reference source for the
planning and execution of S/S under most circumstances, and includes an inventory
of over 200 completed S/S operations in the USA and elsewhere, and more than 40
detailed case studies where S/S has been successfully used to treat both organic
and inorganic contaminated materials.

Each section of this manual has encapsulated the world experience of recognised
experts in his/her particular expertise. However, it should be noted that the subject
material presented here represents the considered views of the authors and should
not be taken to necessarily represent the views of the authors employer, or any
government or private organisation, or of the other contributors to this work.

The mention of trade names or commercial products are for illustrative purposes
only, and does not constitute an endorsement or recommendation for use in any
application, and no public or private funding was provided for preparing this
document.

4
Contributing Authors
Section Title Contributors

1.1-1.4 Introduction and overview Colin Hills and Ed Bates


1.5 Long-term performance Peter Gunning and Colin
Hills
1.6 History of use of cementitious Peter Gunning and Colin
binders Hills
1.7 External Information Sources Peter Gunning
2.1 Use of S/S in the USA (Superfund Ed Bates
and RCRA)
2.2 Use of S/S in the UK and Europe Colin Hills and Paula Carey
2.3 Use of S/S in Canada Colin Dickson and Cameron
Ells
2.4-2.5 Risk management by S/S, S/S and Kate Canning
the Risk Framework
2.6 Future trends Ed Bates and Colin Hills
3.0-3.3 Applicability of S/S Paul Lear
3.4 Geotechnical considerations Craig Lake
3.5 Ex-situ and in-situ application of S/S Paul Lear
4.0-4.6 Ex-situ S/S equipment and Paul Lear
application
5.0-5.1 In-situ auger mixing Dan Ruffing and Ken
Andromalos
5.2-5.3 Injection tillers and rotary drum Robert Schindler and Dan
mixers, In-situ bucket mixing Ruffing
6.0 Quality assurance and quality Roy Wittenberg
control
6.1 Quality assurance and quality Roy Wittenberg
control during S/S
6.2-6.4 Sample collection - the performance Ed Bates
sampling plan, Sample preparation,
curing and testing, variation and
failures

5
7.0 Performance specifications/tests Tom Plante and Ed Bates
7.1 Setting overall goals for treatment Tom Plante
7.2 Importance of site characterisation Tom Plante
and conceptual site model
7.3 Performance parameters, test and Tom Plante and Stany
methods Pensaert
7.4 Impact of other protective measures Ed Bates and Craig Lake
8.1-8.2 Common reagents, Reagent Paul Lear
sourcing
8.3 Bench-scale treatability study, Ed Bates and Paul Lear
design and execution
8.4 Pilot-scale tests Tom Plante
8.5 Selecting samples for treatability Paul Lear
testing
9.1 Overview of factors to consider for Roy Wittenberg
post-treatment management of S/S
material
9.2 Post-treatment management of S/S Roy Wittenberg
9.3 Capping technologies Roy Wittenberg
9.4 Capping strategies based on slope, Roy Wittenberg
drainage, and climate
9.5 Post-construction monitoring Ed Bates
Appendix S/S Completed Projects List Stany Pensaert, Aiman
A Naguib, Paul Lear, Ken
Andromalos, Steve Birdwell,
Tom Plante, Bob Garrett and
others
Appendix Selected S/S Case Studies Stany Pensaert, Paul Lear,
B Ken Andromalos, Steve
Birdwell, Dan Ruffing and
others.

6
Contents
1.0 Introduction and overview ........................................................................................... 30

1.1 Purpose ............................................................................................................... 30

1.2 Target audiences ................................................................................................. 31

1.3 Definition of terms................................................................................................ 32

1.4 Appropriate uses of S/S technology ..................................................................... 34

1.5 Long term performance........................................................................................ 37

1.6 History of use of cementitious binders .................................................................. 41

1.7 Information sources on S/S systems .................................................................... 44

1.7.1 General information .................................................................................... 44

1.7.2 Technical information.................................................................................. 44

1.7.3 Government departments and agency information resources ...................... 44

1.7.4 Research, committees and associations ..................................................... 44

1.7.5 Forums, networks, and discussion groups................................................... 44

2.0 History of S/S as a risk management tool .................................................................... 50

2.1 Use of S/S in the USA (Superfund and RCRA)..................................................... 50

2.1.1 The regulatory impetus for the remediation of sites in the USA .................... 50

2.1.2 Use of S/S in the EPA Superfund Program ................................................. 52

2.1.3 Conclusions ................................................................................................ 55

2.2 Use of S/S in the UK and Europe ......................................................................... 57

2.2.1 Treatment of waste and soil by S/S ............................................................. 57

2.2.2 Legislative background ............................................................................... 58

2.2.3 Public and private initiatives ........................................................................ 59

2.3 Use of S/S in Canada .......................................................................................... 61

2.3.1 Performance (risk)-based versus prescriptive regulations ............................ 61

2.3.2 Relative use of S/S in Canadian provinces .................................................. 62

2.3.3 Sydney Tar Ponds and coke ovens project Nova Scotia ........................... 62

2.4 Risk management by S/S .................................................................................... 63

2.4.1 Definition of risk .......................................................................................... 63

2.4.2 Risk management frameworks .................................................................... 63

2.4.3 Summary .................................................................................................... 65

2.5 S/S and the Risk Framework .............................................................................. 65

2.5.1 Water environment ..................................................................................... 65

2.5.2 Human health ............................................................................................. 67


7
2.5.3 Other receptors........................................................................................... 68

2.5.4 Durability and integrity ................................................................................ 68

2.6 Future trends ....................................................................................................... 69

2.6.1 USA ........................................................................................................... 69

2.6.2 UK and Europe ........................................................................................... 70

2.6.3 Conclusions ................................................................................................ 71

3.0 Applicability of S/S ...................................................................................................... 72

3.1 Metals ................................................................................................................. 72

3.1.1 Chemistry ................................................................................................... 72

3.2 Other inorganic compounds ................................................................................. 76

3.2.1 Chemistry ................................................................................................... 76

3.3 Organics.............................................................................................................. 76

3.3.1 Chemistry and physics of organic stabilization ............................................ 77

3.3.2 Physical aspects of organic stabilization ..................................................... 81

3.4 Geotechnical considerations ................................................................................ 82

3.4.1 Soil classification ........................................................................................ 82

3.4.2 Site Characterization of Soil and Groundwater ............................................ 87

3.4.3 Approaches to site variability ...................................................................... 89

3.5 Ex-situ and in-situ application of S/S .................................................................... 90

3.5.1 Technical considerations ............................................................................ 91

3.5.2 Physical material handling .......................................................................... 95

3.5.3 Economics.................................................................................................. 99

4.0 Ex-situ S/S equipment and application ...................................................................... 102

4.1 Mixing Chambers............................................................................................... 104

4.1.1 Pug-mill mixers ......................................................................................... 105

4.1.2 Screw Mixers ............................................................................................ 108

4.1.3 Ribbon Blenders ....................................................................................... 110

4.1.4 Quality control using mixing chambers ...................................................... 112

4.2 Excavator bucket mixing .................................................................................... 114

4.2.1 Quality Control using excavator bucket mixing .......................................... 118

4.3 Tillers and Other Ex-situ Mixers ......................................................................... 119

4.3.1 Rototillers ................................................................................................. 120

4.3.2 Asphalt millers .......................................................................................... 121

4.3.3 Paddle aerators/compost turners .............................................................. 123


8
4.3.4 Quality control using tillers and paddle mixers ........................................... 125

4.4 Mixing pits ......................................................................................................... 128

4.4.1 Earthen Pits.............................................................................................. 129

4.4.2 Open-top tanks ......................................................................................... 131

4.4.3 Enclosures ............................................................................................... 132

4.5 Ancillary equipment for ex-situ mixing ................................................................ 133

4.5.1 Screening equipment ................................................................................ 133

4.5.2 Crushing equipment.................................................................................. 135

4.5.3 Shredding equipment................................................................................ 136

4.5.4 Magnetic separation equipment ................................................................ 137

4.6 On-site placement of ex-situ treated material ..................................................... 138

4.6.1 Timing ...................................................................................................... 138

4.6.2 Placement ................................................................................................ 139

4.6.3 Compaction .............................................................................................. 139

5.0 In-situ S/S equipment and its application................................................................... 145

5.1 In-situ auger mixing ........................................................................................... 145

5.1.1 Equipment ................................................................................................ 145

5.1.2 Staffing requirements................................................................................ 150

5.1.3 Treatment metrics and considerations....................................................... 152

5.1.4 Treatment plan ......................................................................................... 154

5.1.5 Quality control .......................................................................................... 157

5.1.6 Operational Issues .................................................................................... 160

5.1.7 Summary of limitations, advantages, and disadvantages........................... 161

5.1.8 Costs ........................................................................................................ 162

5.2 Injection tillers and rotary drum mixers ............................................................... 163

5.2.1 Equipment ................................................................................................ 163

5.2.2 Staffing requirements................................................................................ 166

5.2.3 Treatment metrics and considerations....................................................... 167

5.2.4 Quality control .......................................................................................... 169

5.2.5 Operational issues .................................................................................... 171

5.2.6 Advantages and disadvantages of rotary tillers and injection drum mixers . 171

5.2.7 Costs ........................................................................................................ 172

5.3 In-situ bucket mixing .......................................................................................... 172

5.3.1 Equipment ................................................................................................ 172


9
5.3.2 Staffing requirements................................................................................ 173

5.3.3 Treatment metrics and considerations....................................................... 174

5.3.4 Quality control .......................................................................................... 175

5.3.5 Operational Issues .................................................................................... 177

6.0 Quality assurance and quality control ........................................................................ 178

6.1 Quality assurance and quality control during S/S................................................ 178

6.1.1 Quality assurance and quality control during S/S....................................... 178

6.1.2 The importance of CQA in the S/S design and construction process ......... 179

6.1.3 Development of an effective CQA program ............................................... 179

6.1.4 CQA objectives......................................................................................... 181

6.1.5 Roles and responsibilities in the CQA process .......................................... 181

6.1.6 General categories for the CQA process ................................................... 183

6.1.7 Standards and decision processes for S/S performance............................ 187

6.2 Sample collection - the performance sampling plan............................................ 193

6.2.1 Frequency of sampling.............................................................................. 194

6.2.2 Number of samples per sampling event .................................................... 195

6.2.3 Sample collection methods ....................................................................... 197

6.3 Sample preparation and curing .......................................................................... 200

6.3.1 Sample preparation .................................................................................. 200

6.4 Variation and failures ......................................................................................... 206

6.4.1 Normal variation ....................................................................................... 208

6.4.2 Recognising failure ................................................................................... 209

6.4.3 Handling failure......................................................................................... 211

6.4.4 Coring of in-place S/S material.................................................................. 214

7.0 Performance specifications for S/S ........................................................................... 216

7.1 Setting overall goals for treatment ...................................................................... 216

7.2 The importance of site characterisation and the conceptual model ..................... 222

7.3 Performance parameters, tests/methods ............................................................ 226

7.3.1 Performance parameters .......................................................................... 226

7.3.2 Performance tests and methods ............................................................... 232

7.3.3 Phases of performance testing.................................................................. 235

7.3.4 Performance criteria in specifications ........................................................ 236

7.4 Impact of other protective measures ................................................................... 237

8.0 Developing effective S/S formulations ....................................................................... 244


10
8.1 Common reagents ............................................................................................. 244

8.1.1 Portland cement ....................................................................................... 244

8.1.2 Quicklime and hydrated lime ..................................................................... 245

8.1.3 Cement kiln dust ....................................................................................... 247

8.1.4 Lime kiln dust ........................................................................................... 248

8.1.5 Fly ash ..................................................................................................... 249

8.1.6 Bottom ash ............................................................................................... 251

8.1.7 Magnesium oxide ..................................................................................... 252

8.1.8 Phosphates .............................................................................................. 253

8.1.9 Sulfides .................................................................................................... 254

8.1.10 Carbon-based reagents ............................................................................ 255

8.2 Reagent sourcing .............................................................................................. 257

8.2.1 Proprietary reagents ................................................................................. 257

8.2.2 Mineral and waste reagents ...................................................................... 258

8.2.3 Importance of location .............................................................................. 259

8.2.4 Reagent samples for testing ..................................................................... 260

8.3 Bench-scale treatability design and testing......................................................... 260

8.3.1 Bench-scale treatability study objectives ................................................... 261

8.3.2 Importance of the conceptual implementation plan .................................... 262

8.3.3 Reagent selection considerations ............................................................. 263

8.3.4 Bench-scale treatability testing approaches .............................................. 263

8.3.5 Initial sample characterisation ................................................................... 272

8.3.6 Laboratory procedures .............................................................................. 272

8.4 Pilot-scale tests ................................................................................................. 274

8.4.1 Pilot test objectives ................................................................................... 276

8.4.2 Confirming the conceptual implementation plan ........................................ 278

8.4.3 Implementing the pilot testing ................................................................... 278

8.4.4 S/S safety considerations ......................................................................... 286

8.5 Selecting samples for treatability testing ............................................................ 287

8.5.1 Option 1: Maximum Contamination Sample .............................................. 287

8.5.1 Option 2: high contamination sample ........................................................ 288

8.5.2 Option 3: average contamination sample .................................................. 289

8.5.3 Rationale for using the high contamination sample.................................... 290

8.5.4 Sites with multiple materials requiring treatment ........................................ 290


11
9.0 Post-treatment capping and monitoring ..................................................................... 291

9.1 Overview of post-treatment management of S/S material ................................... 291

9.1.1 Climatologic and geographic considerations ............................................. 292

9.1.2 Site geometry ........................................................................................... 292

9.1.3 Integration with existing site development ................................................. 293

9.1.4 Future site use .......................................................................................... 293

9.1.5 Regulatory requirements........................................................................... 294

9.2 Post treatment management of S/S ................................................................... 296

9.2.1 Key engineering parameters S/S post treatment management .................. 296

9.2.2 Engineering approach for preparing S/S material prior to capping ............. 297

9.3 Capping technologies ........................................................................................ 303

9.3.1 Overview of design considerations and general types of caps applicable

for S/S monoliths ...................................................................................... 303

9.3.2 Clay and/or other low permeability earthen materials ................................ 305

9.3.3 Geo-membranes....................................................................................... 308

9.3.4 Geo-synthetic clay liner (GCLs) ................................................................ 310

9.3.5 Clay with geo-membrane .......................................................................... 313

9.3.6 Geo-composite clay liner with geo-membrane ........................................... 314

9.3.7 Evapotranspiration (ET) caps .................................................................... 315

9.3.8 Asphalt and concrete pavement applications ............................................ 316

9.3.9 Comparison of capping technologies ........................................................ 318

9.4 Capping strategies based on slope, drainage, and climate ................................. 319

9.4.1 Slope and drainage considerations ........................................................... 319

9.4.2 Capping strategies based on climatologic conditions ................................. 324

9.5 Post-construction monitoring.............................................................................. 327

9.5.1 Regulatory impetus and guidance for post-construction monitoring ........... 327

9.5.2 Specific features common in monitoring programs .................................... 329

10.0 Consolidated Reference List .................................................................................... 333

11.0 Contributing author biographies ............................................................................... 345

Appendix A: Remediated sites employing S/S .................................................................. 386

Appendix B: Case studies employing S/S ........................................................................ 421

Case Study 1: Abex Superfund Site ............................................................................. 424

Case Study 2: American Creosote Superfund Site........................................................ 426

Case Study 3: Bayou Trepagnier .................................................................................. 431


12
Case Study 4: BHAD Chromium Pond.......................................................................... 435

Case Study 5: Cambridge ............................................................................................ 440

Case Study 6: Former Camden Gas Works Site ........................................................... 443

Case Study 7: Camp Pendleton Scrap Yard ................................................................. 448

Case Study 8: Chevron Pollard Landfill Site ................................................................. 453

Case Study 9: Carnegie ............................................................................................... 458

Case Study 10: Columbus ............................................................................................ 460

Case Study 11: CSX Benton Harbor Scrapyard ............................................................ 462

Case Study 12: Double Eagle Refinery Site .................................................................. 465

Case Study 13: Dundalk............................................................................................... 467

Case Study 14: East Rutherford ................................................................................... 469

Case Study 15: Foote Minerals Superfund Site ............................................................ 472

Case Study 16: Ghent .................................................................................................. 476

Case Study 17: Guernsey ............................................................................................ 481

Case Study 18: Hercules 009 Landfill Site .................................................................... 486

Case Study 19: Hoedhaar Lokeren............................................................................... 489

Case Study 20: Irving ................................................................................................... 492

Case Study 21: Johnston Atoll Solid Waste Burn Pit ..................................................... 495

Case Study 22: Kingston .............................................................................................. 498

Case Study 23: London Olympic Site ........................................................................... 502

Case Study 24: Martinsville .......................................................................................... 506

Case Study 25: Milwaukee, Wisconsin ......................................................................... 509

Case Study 26: Nederland ........................................................................................... 513

Case Study 27: New Bedford ....................................................................................... 517

Case Study 28: Nyack .................................................................................................. 519

Case Study 29: NYSEG Norwich NY MGP Site ............................................................ 522

Case Study 30: Obourg ................................................................................................ 526

Case Study 31: Perth Amboy ....................................................................................... 529

Case Study 32: Portsmouth.......................................................................................... 531

Case Study 33: Rieme ................................................................................................. 534

Case Study 34: Roma Street Station ............................................................................ 540

Case Study 35: Sag Harbor.......................................................................................... 544

Case Study 36: Sanford MGP Superfund Site .............................................................. 547

Case Study 37: Sderhamn ......................................................................................... 552


13
Case Study 38: Southeast Wisconsin ........................................................................... 555

Case Study 39: St Louis ............................................................................................... 559

Case Study 40: Sunflower Army Ammunition Depot ..................................................... 561

Case Study 41: Sydney Tar Ponds ............................................................................... 564

Case Study 42: Umatilla Ammunition Depot ................................................................. 577

Case Study 43: Valero Paulsboro Refinery ................................................................... 580

Case Study 44: Waukegan ........................................................................................... 584

Case Study 45: West Doane Lake Site ......................................................................... 586

Case Study 46: X-231B Pilot Study .............................................................................. 590

Case Study 47: Zwevegem .......................................................................................... 593

Index 598

14
List of Tables
Table Title Source
1.1 S/S applications examined in 2010 PASSiFy
(2010)
1.2 General information sources on S/S Gunning
1.3 Technical information sources on S/S Gunning
1.4 Information on S/S from government departments and Gunning
agencies
1.5 Information on S/S from research, committees and Gunning
associations

1.6 Information on S/S from forums, networks, and Gunning


discussion groups

3.1 Considerations for choosing test pits versus boreholes Lake


4.1 Advantages and disadvantages of pug-mill mixers Lear &
Gunning
4.2 Advantages and disadvantages of screw mixers Lear &
Gunning
4.3 Advantages and disadvantages of ribbon blenders Lear &
Gunning
4.4 Advantages and disadvantages of excavator bucket Lear &
mixing Gunning
4.5 Advantages and disadvantages of rototillers Lear &
Gunning
4.6 Advantages and disadvantages of asphalt millers Lear &
Gunning
4.7 Advantages and disadvantages of paddle aerators Lear &
Gunning
4.8 Ex-situ mixer manufacturers Lear &
Gunning
4.9 Advantages and disadvantages of earthen pits Lear &
Gunning
4.10 Advantages and disadvantages of open-top tanks Lear &
Gunning
4.11 Advantages and disadvantages of enclosures Lear &
15
Gunning
5.1 Examples of in-situ auger treatments including auger Geo-
diameter and depth treated Solutions
5.2 Quality control planning for auger-based S/S mixing Geo-
Solutions &
Gunning
5.3 Advantages and disadvantages of auger mixing Geo-
Solutions &
Gunning
5.4 Quality control planning for BOSS-based S/S mixing Geo-
Solutions &
Gunning
5.5 Advantages and disadvantages of rotary tillers and Geo-
drum mixers Solutions &
Gunning
5.6 Quality control planning for bucket-based S/S mixing Geo-
Solutions &
Gunning
6.1 Example of a simple decision making framework Wittenberg
agreed between the site engineer and contractor
6.2 Pb in leachates from construction performance Ed Bates
samples at 14, 21, and 28 days of curing, United
Metals Site (g/L)
6.3 Results from duplicate performance samples, Ed Bates
American Creosote site
7.1 Typical S/S Site assessment considerations T Plante
9.1 Minimum index properties for the use of clay Wittenberg
& Gunning
9.2 A comparison of capping technologies NRT
9.3 Capping technologies vs slope and drainage NRT
9.4 Capping technologies and climate NRT

16
List of Figures
Figure Title Source
Cover Photo of two crane mounted ISS rigs in Winter Ed Bates
1.1 Treating creosote, PCP, & dioxins in pug-mill Ed Bates
1.2 Installing a grout slurry blanket to reduce Ed Bates
emissions during in-situ S/S treatment
1.3 Conceptual model of a site with environmental After EA, 2004b
loading
1.4 Longevity of S/S materials proposed by selected Passify, 2010
authors
2.1 Cumulative source control technology selection After
over 30 years
USEPA 2013
2.2 Selection of S/S vs all other technologies after USEPA,
2013
2.3 Waste types selected for S/S After
USEPA, 2000
2.4 S/S selection of in-situ vs ex-situ After USEPA,
2013
2.5 Relative increase in selection of in-situ vs ex-situ After USEPA,
S/S 2013
3.1 Lead-contaminated soil and battery casings Paul Lear
from a former battery-recycling site,
subsequently chemically stabilized
3.2 Chromium paint pigment-contaminated soil Paul Lear
being chemically stabilized in-situ
3.3 A contaminated wastewater sludge being Paul Lear
excavated
3.4 Treatment facility to chemically stabilize and Paul Lear
solidify acid sludge tar
3.5 General description and identification of soils After BS5930
3.6 Description and identification of soils After BS5930
3.7 Description and identification of soils After BS5930
3.8 Ex-situ treatment in a roll-off box Paul Lear

17
3.9 Typical ex-situ S/S equipment, including pug- Ed Bates
mill, silos, reagent tank, water tank, stacker and
haulage
3.10 Ex-situ S/S treatment enclosure Paul Lear
3.11 In-situ S/S treatment of oily sludge Paul Lear
3.12 In-situ mixing equipment operating on top of S/S Paul Lear
treated material
3.13 Typical in-situ S/S equipment, batch plant with Ed Bates
reagent silos and mix tanks: crane, Kelly bar,
and auger
3.14 A 3 ft limestone boulder excavated to facilitate Ed Bates
in-situ treatment
4.1 Ex-situ processes Peter Gunning
4.2 A pug-mill with its ancillary equipment Paul Lear
4.3 Ancillary equipment of a pug-mill Paul Lear
4.4 Paddles inside a typical pug-mill Ed Bates
4.5 Screw mixer US Air Filtration,
Temecula
4.6 Ribbon mixer Rana Perforators
4.7 Control panel for waste feed conveyor Paul Lear
4.8 Calibrating reagent feed from a screw auger Ed Bates
4.9 Display panel for an automated pug-mill Ed Bates
4.10 Modified excavator bucket for improved mixing Paul Lear
4.11 Allu PMX Power Mixer attachment Allu, Pennala,
Finland
4.12 Rotating mixing head attachment Paul Lear
4.13 Rotating rake mixing attachment Paul Lear
4.14 Rototiller attachment Paul Lear
4.15 Asphalt miller Bomag
4.16 Paddle aerator mixing windrowed soil Paul Lear
4.17 Excavator mixing of Portland cement into Pb- Paul Lear
contaminated soil and debris

18
4.18 Bucket mixing hydrated lime and oily waste in Ed Bates
earthen pit
4.19 Stabilization of hexavalent chromium- Paul Lear
contaminated soil in a roll-off box
4.20 Grizzly screen for separation of large debris, Paul Lear
with over-sized material, to the left and
processed soil for ex-situ mixing, right
4.21 Grizzly screen (foreground) combined with a Paul Lear
trommel to remove battery casings and debris
from excavated material
4.22 A rotary shear shredder processing scrap metal Paul Lear
4.23 Transport and placement of ex-situ S/S material Ed Bates

4.24 Placed ex-situ S/S-treated material after multiple Ed Bates


passes of the dozer
4.25 Proctor compaction-testing results for ex-situ Paul Lear
S/S treated material
4.26 Sheeps-foot roller compacting S/S treated Ed Bates
material (note the oval pads on the roller)
4.27 Smooth roller compacting ex-situ S/S treated Ed Bates
material
5.1 In-situ processes Peter Gunning
5.2 A crane mounted soil-mixing rig R Schindler
5.3 A crane mounted soil-mixing rig R Schindler
5.4 An excavator-mounted rig D Ruffing
5.5 An excavator-mounted soil-mixing rig D Ruffing
5.6 Auger used at the USX Site, Duluth, MN Ed Bates
5.7 Auger used at a coal gas plant site in FL Ed Bates
5.8 An automated batch plant D Ruffing
5.9 Self-propelled dry storage silo Allu
5.10 Typical Column Layout Showing Overlapping Geo-Solutions
Columns to Achieve 100% Coverage
5.11 showing typical excavated overlapping columns Geo-Solutions

19
5.12 In-situ mixing of soil under a live fibre optic cable Ed Bates
5.13 In-situ mixing under a fibre optic cable Ed Bates
5.14 Backhoe (Excavator) operated soil stabilizer Geo-Solutions
(BOSS)
5.15 Backhoe (Excavator) operated soil stabilizer Geo-Solutions
5.16 Lang mixer- excavator, arm, and mixing head as Ed Bates
one unit
5.17 Allu mixer head (in red) attached to a standard Ed Bates
excavator
5.18 A batching plant for a large S/S project R Schindler
5.19 Example of bucket mixing Geo-Solutions
5.20 Bucket mixing of coal tar soils to depth of 15 ft Ed bates
(5 m)
6.1 Flowchart for sampling and analysis-related NRT
decision making
6.2 Flowchart for design-related decision making NRT
6.3 Flowchart for operational issues-related decision NRT
making
6.4 Collection of sample using piston tube inserted Ed bates
by excavator
6.5 A sampler employing a hydraulic gate Ed bates
6.6 Preparing 3 in x 6 in (75 x 150 mm) mold Ed bates
6.7 A 0.5 in (12.5 mm) screen to remove oversize Ed bates
debris
6.8 Labelling of specimen mold and cap Ed bates
6.9 Sample molds curing on-site in water bath Ed bates
6.10 Checking set/strength by a pocket penetrometer Ed bates
6.11 Detailed record keeping of properties of cured Ed bates
sample
6.12 A slump test on an S/S bulk sample Ed bates
6.13 Poorly prepared field-specimens T Plante
6.14 A well-prepared specimen with minimal air voids T Plante

20
6.15 Increase in strength by S/S soil over time T Plante
6.16 Using a field penetrometer to assess strength of Ed Bates
field placed S/S material
6.17 Excavating Top of In-situ Column Ed Bates
6.18 Cores obtained by sonic drilling of S/S material Ed Bates
7.1 Performance goals and performance Modified from
specifications in the S/S process ITRC
7.2 Strength-gain of bench vs field demonstration T Plante
(FD) data
7.3 A simplified conceptual site model
7.4 Installing a slurry wall, Whitehouse, Florida Ed Bates
7.5 Constructing a S/S sub-cap, Brunswick Ed Bates
7.6 Constructing a vertical wall using a Bauer Panel Ed Bates
Cutter, Brunswick
7.7 Oily water emulsion, Stauffer Ed Bates
7.8 Solidifying oily water emulsion, Stauffer Ed Bates
7.9 Installing FML at Stauffer Ed Bates
8.1 Hydrated lime added to waste in a mix pit Ed Bates
8.2 Fly ash stockpile on a treatment site with its Ed Bates
cover removed
8.3 Granular activated carbon in a super sack Ed Bates
8.4 Design of S/S formulations Peter Gunning
8.5 Unconfined compressive strength testing Paul Lear
8.6 Permeability Paul Lear
8.7 A limited bench-scale field treatability study Ed Bates
8.8 Reagents are measured on a weight/weight Paul Lear
basis to the untreated soil
8.9 Pocket penetrometer testing for approximate Paul Lear
strength
8.10 Eight-foot (2.5 m) In-situ auger and carrier Ed Bates
assembly

21
8.11 Pilot test using a 10 ft. (3 m) diameter in-situ Ed Bates
auger
8.12 Pilot test for in-situ bucket mixing Ed Bates
8.13 Ex-situ pug-mill with support equipment Ed Bates
8.14 Field quality control test specimen preparation T Plante
8.15 Field slump test performance T Plante
8.16 Grout density test by mud balance T Plante
8.17 Field check to assure homogenous mixing Ed Bates
8.18 Reagent silo electronic scale and calibration T Plante
weight
8.19 Trial pit in an S/S soil monolith T Plante
8.20 In-situ auger test producing steam, sulfur Ed Bates
dioxide emissions
8.21 Collecting a bulk treatability sample from auger Ed Bates
flights
9.1 Typical profile for a Subtitle D cap NRT
9.2 A typical profile for a Subtitle C cap NRT
9.3 A smooth roller being used to compact ex-situ Ed Bates
treated material
9.4 Direct placement of S/S swell by an excavator NRT
for final grading
9.5 Final shaping/contouring of in-situ S/S using a E Bates
dozer
9.6 Reconditioning and placement of in-situ S/S NRT
swell
9.7 Example single component cap with a NRT
compacted clay layer
9.8 Cover layer comprising 2 ft of compacted soil, Ed Bates
Schuylkill Metals, Florida.
9.9 Construction of a low permeability clay cap NRT
9.10 Final grading to direct surface drainage NRT
9.11 Installation of a geo-membrane showing the Ed Bates
welded seams between sheets

22
9.12 A single component cap with a geo-membrane NRT
9.13 A typical single component cap with a geo- NRT
synthetic clay layer
9.14 GCL and soil cover with drainage channel, Peak Ed Bates
Oil Site
9.15 Cover with a GCL, 2 ft of soil, and a gravel Ed Bates
surface
9.16 Profile of a composite cap with a geo-membrane NRT
over a compacted clay layer
9.17 A composite cap with a geo-membrane over a NRT
GCL
9.18 A monolithic evapotranspiration cap NRT
9.19 A capillary break evapotranspiration cap NRT
9.20 A coarse-grained soil-based ET cap over an S/S NRT
monolith
9.21 Groundwater monitoring wells around an S/S NRT
Monolith at the Sanford Gasification Plant Site,
FL
9.22 Groundwater monitoring wells around an S/S NRT
monolith at the Sanford gasification plant site

23
List of abbreviations and acronyms

Abbreviation/
acronym
ADEME Environment and Energy Management Agency
ANS/ANSI American Nuclear Society/American National Standards
Institute
ARAR Applicable, or relevant, and appropriate requirement
ASTM American Society for Testing and Materials
BC British Columbia
BCA British Cement Association
BDAT Best Demonstrated Available Technology
BGS Below Ground Surface
BS British Standards
BTEX Benzene, Toluene, Ethylbenzene and Xylene
BOSS Backhoe operated soil stabilizer
CASSST Codes and Standards for Stabilization/Solidification
Technology
CBR California bearing ratio
CERCLA Comprehensive Environmental Response, Compensation,
and Liability Act
CEMBUREAU The European Cement Association
CH Clay, high-plasticity
CKD Cement kiln dust
CL:AIRE Contaminated Land: Applications in Real Environments
COC Contaminant of Concern
COPA Control of Pollution Act
CPT Cone Penetration Testing
CQA Construction quality assurance

24
CQC Construction quality control
CL Clay, low plasticity
C-S-H Calcium silicate hydrate
CSM Conceptual site model
cy Cubic yard
DQO Data quality objective
DNAPL Dense non aqueous phase liquid
EA Environment Agency
EPA Environmental Protection Agency (United States)
ET Evapotranspiration
EU European Union
FL Florida
FML Flexible membrane liner
ft Foot
FY Fiscal year
GCL Geo-synthetic clay liner
GPS Global Positioning System
HDPE High density polyethylene
HSWA Hazardous and Solid Waste Amendments
IC Institutional controls
IEA International Energy Agency
ITRC Interstate Technology & Regulatory Council
kPa Kilopascal
L Litre
lb Pound
LDPE Low density polyethylene
LEAF Leaching Environmental Assessment Framework
LKD Lime kiln dust
LTRA Long-Term Response Action

25
m Metre
MGP Manufactured Gas Plant
MNA Monitored natural attenuation
MPa Mega Pascal
MPH Miles per hour
NAPL Non-aqueous phase liquids
NEN The Netherlands Standardization Institute
O&M Operation and maintenance
PAH Polycyclic aromatic hydrocarbon
PCP Pentachlorophenol
PCB Polychlorobiphenyl
PVC Polyvinyl chloride
PSI Pounds per square inch
QA Quality assurance
QC Quality control
RCRA Resource Conservation and Recovery Act
REACH Registration, Evaluation, Authorisation and restriction of
CHemicals.
RPM Revolutions per minute
SARA Superfund Amendments and Reauthorization Act
SC Sandy clay
SM Sandy silt
SOP Standard Operating Procedure
SP Sand, poorly-graded
SPLP Synthetic Precipitation Leaching Procedure
SPT Standard Penetration Testing
S/S Stabilization/solidification
STARNET Stabilization/solidification Network
SW Sand, well-graded

26
TCLP Toxicity Characteristic Leaching Procedure
TOC Total Organic Carbon
TPH Total Petroleum Hydrocarbons
UCS Unconfined Compressive Strength
USCS Unified Soil Classification System
USEPA United States Environmental Protection Agency
VOC Volatile Organic Compound

27
Table of Unit Conversions
Speed
mph to kph kph to mph
1 mph 1.61 kph 1 kph 0.62 mph

Length
centimeters to inches inches to centimeters
1 cm 0.39 in 1 in 2.54 cm

metres to feet feet to metres


1 m 3.28 ft 1 ft 0.30 m

Area
Hectares to Acres Acres to Hectares
1 ha 2.47 ac 1 ac 0.4047 ha

Weight
Short Tons to Tonnes Tonnes to Short Tons
1 Ton 0.91 Tonne 1 Tonne 1.1 Ton

Long Tons to Tonnes Tonnes to Long Tons


1 Ton 1.02 Tonne 1 Tonne 0.98 Ton

Pounds to Kilograms Kilograms to Pounds


1 lb 0.45 kg 1 kg 2.2 lb

Ounces to Grams Grams to Ounces


1 oz 28.3 g 1 g 0.03 oz

28
Volume
Cubic yards to Cubic metres Cubic metres to Cubic yards
1 yd3 0.76 m3 1 m3 1.31 yd3

US Gallon to Litre Litre to US Gallon


1 gal 3.78 l 1 l 0.26 gal

Imperial Gallon to Litre Litre to Imperial Gallon


1 gal 4.54 l 1 l 0.22 gal

Pressure
PSI to MPa MPa to PSI
1 psi 0.007 MPa 1 MPa 145 psi

Density
Pounds per cubic foot to Tonnes per Tonnes per m3 to Pounds per cubic
m3 foot
1 pcf 0.02 t/m3 1 t/m3 62.4 pcf

Torque
Kilograms force metre to Pounds force Pounds force foot to Kilograms force
foot metre
1 kgf-m 7.23 lbf-ft 1 lbf-ft 0.14 kgf-m

Currency (at time of writing, August 2015)


US Dollars to Euros Euros to US Dollars
1 USD 0.91 1 1.10 USD

29
1.0 Introduction and overview

1.1 Purpose
This book presents a synthesis of the practical experience in the remediation of
hazardous waste sites by stabilization and solidification technology (S/S)
professionals. The sites that typically require treatment are contaminated by industrial
residues and present a risk to human health and the environment.

The practical experience contained herein is intended to act as a reference source


on: the suitability of wastes for treatment, the design, application and quality aspects
of S/S, and other key issues that (together) capture how S/S is applied in the field
and validated as a risk-management strategy.

The issues facing the practitioner on how to implement S/S are discussed, and the
know how presented should provide valuable insight into the remediation of
contaminated land and treatment of waste by S/S. No decision regarding treatment
should be based solely on this document. Rather this document should be used in
conjunction with other references and the knowledge of skilled professionals to reach
reasoned decisions to fit the needs of any specific site.

Each contributing author is a recognised authority on the application of S/S and is


responsible for the material presented in their respective sections. Contributing
authors have experience gained from the application of SS in the USA, Canada, the
United Kingdom and Continental Europe experience that the editors believe is not
shared in any other published work.

30
1.2 Target audiences
This Manual of Practice has been designed as a practical reference for regulators,
site owners, engineering firms, and others involved in selecting, designing, bidding,
and providing oversight for the remediation of hazardous waste sites using S/S. This
book provides guidance on applicable contaminants, site characteristics, project
planning, equipment capabilities, production rates, performance specifications and
the quality assurance of S/S treated materials.

This manual should provide stakeholder reassurance on the appropriateness of S/S


as a viable and cost effective technology for managing the risks associated with
contaminated soil and waste. Included are references to numerous case studies and
an extensive reference list of completed projects that successfully employed S/S.

31
1.3 Definition of terms
The term Solidification/Stabilization has been defined somewhat differently in various
publications and promulgated regulations. This manual of practice follows the
definitions for solidification and stabilization as presented in the Interstate Technology
and Regulatory Council document Development of Performance Specifications for
Solidification/Stabilization (ITRC 2011):

Although solidification and stabilization are defined separately, they are often
implemented simultaneously through a single treatment process. The EPA defines
each as follows (EPA 2000):

Solidification involves the processes that encapsulate contaminated material to


form a solid material and restricts contaminant migration by decreasing the surface
area exposed to leaching and/or by coating the contaminated material with low-
permeability materials. Solidification is accomplished by mechanical processes,
which mix the contaminated material with one or more reagents. Solidification
entraps the contaminated material within a granular or monolithic matrix.

Stabilization involves the processes where chemical reactions occur between the
reagents and contaminated material to reduce the leachability of contaminated
material into a stable insoluble form. Stabilization chemically binds free liquids and
immobilises contaminated materials or reduces their solubility through a chemical
reaction. The physical nature of the contaminated material may or may not be
changed significantly by this process.

The following definitions for in-situ and ex-situ are used in this manual of practice. In-
situ and ex-situ - are defined based solely on the manner in which the treatment
mixing is accomplished. Various regulatory agencies and authors have defined these
terms differently and their definitions may have significance regarding specific
requirements that must be met. For this Manual of Practice the following definitions
are used.

In-situ is defined to mean that the contaminated soils/sludges are mixed with
treatment reagents without removing the contaminated soils from the ground. After
treatment the material may be left in its original location, placed in a different location
on-site, or be sent off-site.

Ex-situ is defined to mean that the contaminated soil/sludge was excavated from its
original location in the ground and then mixed with treatment reagents. After
32
treatment the material may be returned to its original location, placed in a different
location on-site, or be sent off-site.

As an example, reagents could be applied either as solids or as a grout directly to


contaminated soils and then be mixed with an excavator. This would be in-situ
treatment. If however the contaminated soils were removed from the ground and
placed in a nearby mix pit, then reagents were added and mixed with an excavator,
this would be ex-situ treatment.

These definitions are based on the mixing equipment and the approaches and are
used in this document. Regulatory definitions may, however, differ from these
definitions, and from agency to agency.

Other terms such as site remedial goals, material performance goals, specifications,
etc., are defined below and discussed in Section 7 and essentially follow the
definitions used by ITRC (ITRC, 2011).

Remedial Goals overall objectives of the remedy to address the identified


risk pathways
Material Performance Goals expected behaviours of the treated S/S
material to support meeting the remedial goals
Material Performance Specifications the collection of parameters, tests
and criteria to be utilised in developing a mix design and in evaluating the
ability of the mix design to meet the material performance goals
Construction Performance Specifications the collection of parameters,
tests and criteria to be utilised to verify that the treated material created
during implementation is consistent with the materials developed and
characterised during the treatability testing and that key performance
characteristics (e.g., strength, permeability, and possibly leaching reduction)
are consistently met as the treatment progresses

33
1.4 Appropriate uses of S/S technology
With the development of an effective binder formulation and careful implementation,
S/S can be used to treat a wide range of organic and inorganic contaminants in soils
and sludges. Excavated sediments are also treatable. Section 3 discusses potential
applications and considerations in detail, while the appendices provide numerous
examples.

Generally, S/S can be used to treat metallic compounds, longer chain petroleum
hydrocarbons and many of the larger chlorinated organics. Figure 1.1 is typical of a
site set-up, where a binder system employing several reagents is being applied to
excavated-material, in an ex-situ treatment step. The three reagent silos (cement, fly
ash, activated carbon) and a pug-mill were used to treat soils contaminated with
creosote, pentachlorophenol (PCP), dioxins, and metals, at the American Creosote
Site in Tennessee. One of the considerable strengths of S/S is its versatility of
application, using commonly available reagents in a formulation designed to manage
the risks from contamination at a specific site. The successful treatment of soils at the
American Creosote site is a good example of this (Bates et al, 2002).

Although dioxins can be treated quite easily with S/S, PCP can be difficult hence the
use of activated carbon in the treatment formula employed at American Creosote. It
should be noted that S/S often is not very effective for treating volatile organics or
volatile inorganics, nor is very effective for treating liquid fuels or organic solvents
such as benzene, toluene and xylene. Fortunately other technologies such as
bioremediation and vapour extraction can be very effective for such contaminants.

However, if contaminants for which S/S is not well suited are present in a soil or
sludge at very low concentrations, then often the material can be treated using a
modified binder, incorporating activated carbon or organophilic clays into the
formulation. Figure 1.2 depicts the installation of a grout slurry blanket over a bermed
cell to minimise emissions during subsequent in-situ S/S treatment employing such a
special binder.

34
Figure 1.1: Treating creosote, PCP, & dioxins in pug-mill

Figure 1.2: Installing a grout slurry blanket to reduce emissions during in-situ
S/S treatment

35
It should be noted that S/S is generally easier to implement in sandy, silty or gravely
soils, than in soils with high clay content as it is easier to achieve a uniformity of
mixing in the former, while the latter may tend to leave residual clay balls of unmixed
and untreated material. Sections 3.1 and 3.2 of this document provide a more
detailed discussion regarding appropriate contaminant types and soils for treatment
by S/S.

The use of S/S for managing the risk associated with contamination arising from
industrial processes has a long history, extending back more than half a century. The
successful application of S/S is however dependent upon a detailed knowledge of the
behaviour of contaminants in the environment, the soil-matrix to be treated, the
appropriate choice of binder systems and its application to produce a rock-like
monolithic or granular product in which contaminants are encapsulated.

The following sections of this document present the collective real-world experience
of leading practitioners on how S/S is applied and validated as a versatile risk
management strategy.

36
1.5 Long term performance
S/S stabilized soils and wastes are vulnerable to the same physical and chemical
degradation processes that affect any other cement-bound material. Any impact on
the binder system can affect its capacity to immobilise contaminants in the longer-
term.

Stabilized/solidified material will differ from conventionally bound materials in


significant ways. S/S systems incorporating cement as the primary binder are (to a
certain extent) analogous to concrete, which is widely used as a construction
material. Concrete is a carefully designed product with specific physical and
mechanical properties. An S/S system however is inherently a more heterogeneous
product, and therefore cannot be as precisely engineered as its final characteristics
are not normally accurately predicted.

Figure 1.3 is a conceptual model of an S/S system, which illustrates the complexity
of the processes and the impactors on performance with time. The key variables
underlying the efficacy of S/S can be described as the nature of the soil/waste
matrix, the contaminants, the interaction with the binder system and the transport
properties of the waste form and surrounding containment system, whether that be
engineered containment, or the inherent geology of the site of deposition/placement.

Acting upon the waste form after placement are the specific environmental loads,
operating in the environment of service. These may include a hydraulic gradient,
gaseous or saline infiltration, micro- and macro-biological activity, freeze-thaw etc.
Together, all of these variables will dictate the performance of the waste form with
time time that should ideally be measured in millennia.

Despite the very different design specifications, and nearly 60 years of the use of
S/S in the USA, it is encouraging to note that there are no reported major failures of
S/S waste-forms in that country. Similarly, this observation has been supported by
the findings of the PASSiFy project (PASSiFy, 2010), which examined samples of
S/S materials taken directly from remedial operations in the USA, UK and France
(see Table 1.1).

37
Figure 1.3: Conceptual model of a site with environmental loading

The PASSiFy study highlighted that a number of risk-indicators were present in the
samples examined, but these did not indicate the onset of deleterious reactions. As
waste forms appeared to behave much like cement-bound materials, the interactions
between the soil-fraction, the waste and the binder system could be explained.

The life expectancy (the time in service where contaminants are not significantly
released) of different S/S systems is predicted to extend from decades to thousands
of years. Performance is dependent upon the binders being employed, the
contaminants being treated and the environmental loads impacting upon the waste
form. Figure 1.4 presents data taken from several studies to predict the long-term
behaviour of S/S systems.

38
Table 1.1: S/S applications examined in 2010

Contaminants Age S/S Method

Proprietary Mix
Additive
Cement
Site Observations

Ex-Situ

Fly ash
Metals

In-Situ
BTEX
VOC
PAH
PCB

TPH (yrs)
American All chemical and physical
Creosote, targets met, apart from
3
Tennessee, permeability. Evidence of
USA natural weathering
All original targets met.
Pepper Steel,
14 Evidence of natural
Florida, USA
weathering
All original targets met,
South 8th
apart from compressive
Street, 4
strength. Evidence of
Arkansas, USA
natural weathering

Georgia Power All original targets met,


Company, apart from compressive
12
strength. Evidence of
Georgia, USA natural weathering
Selma, All original targets met.
5
California, USA No weathering data

All original targets met.


Halton, UK 5 Evidence of natural
weathering
All original targets met.
Caerphilly, UK 1 Evidence of natural
weathering
Quarry Dump, All original targets met,
10
USA apart from permeability

39
Figure 1.4: Longevity of S/S materials proposed by selected authors

40
1.6 History of use of cementitious binders
The history of modern cement begins with the endeavours of the British Engineer
John Smeaton. Charged with rebuilding the Eddystone Lighthouse in Cornwall,
England, during the mid-eighteenth century, his quest to find a durable water-
resistant material led to the discovery that the best mortars were made by calcining
limestone containing clay impurities (Miller, 2009; Hall, 1976). The resulting strong
binder according to Smeaton would equal the best merchantable Portland stone in
solidity and durability (MacLaren and White, 2003).

However, the story of cementitious binders actually spans 9 millennia, as described


below. The secrets of making a perfect binder were lost for well over a thousand
years after the fall of the Roman Empire. In the interim period leading up to the
Industrial Revolution, simple lime mortars were used. These materials had little or no
hydraulic properties (the ability to set underwater), and only fully set upon prolonged
contact with carbon dioxide from the atmosphere, and consequently, from the
surface inwards (Hall, 1976).

The Romans were accomplished at producing high quality hydraulic cements by


blending quicklime, sand, volcanic ash and aggregate, and even used ox blood as a
plasticising agent! Unlike modern concretes, the secret of these mixtures was the
use of the pozzolanic volcanic ash, named after the Pozzuoli, near Naples in Italy
where it is found.

Pozzolanic cements derive their strength from the rich silica and aluminate phases in
the ash, which react with lime to form hydrated calcium silicates and aluminates.
These compounds are similar to the cementing minerals in modern Portland cement
(MacLaren and White, 2003; Hall, 1976).

In his Ten Books on Architecture published around 25BC, Vitruvius advised on the
use of different types of aggregate and binder ratios for specific applications. By the
middle of the first century AD, Roman civil engineers had mastered the art of
underwater concreting and constructed the harbour at the city of Caesarea. The
importance of the raw materials quality, the detrimental effect of sand contaminated
with earth and the problems associated with the excessive use of marine sand were
then clearly understood (Delatte, 2001). The attention to detail, and use of specified
ingredient is key to the endurance of many Roman structures, including the
Pantheon, the Pont du Gard, and the Basilica of Constantinople (Wayman, 2011;
MacLaren and White, 2003).

41
It is likely, however, that the Romans inherited cement technology from their Greek
neighbours (Wayman, 2011). One of the best examples is a 3000 year-old cast
water tank in the ancient city of Kamiros, on the Greek island of Rhodes. Analysis of
this tank has shown that the Greeks had an excellent knowledge of concretes,
practiced careful blending of different grades and types of aggregates, and mixed
binders composed of volcanic earth and lime. The concrete on Rhodes still exhibits
excellent compressive strength, elastic properties and low porosity despite exposure
to weathering over millennia (US Department of Energy, 2000).

Whilst the Romans and Greeks were perfecting pozzolanic-based formulations, the
Chinese explored alternative additives. The pyramids of Shaanxi, built during the Qin
Dynasty around 200BC, contained conventional mixtures of lime and volcanic ash or
clay, whereas more unusual materials were used in other applications (Miller, 2009).
Lime was mixed with sticky rice to create a remarkable composite construction
medium employed in all manner of important buildings, including tombs, urban
constructions, and water conservancy facilities (Yang et al., 2010).

From the third millennium BC, the Egyptians developed their own skills in the use of
cementitious binders. The earliest construction utilised simple clay mortars to bind
stone blocks. As their technology advanced, gypsum or lime-based mortars were
adopted for the construction of the later pyramids. Like the Romans and the Greeks,
the Egyptians were able to effectively work with the raw materials available to them,
and formulated numerous different mortars from burnt lime and gypsum, in
combination with marly limestone, kaolinitic clay, natron salt, sand, and
diatomaceous earth.

An alternative theory proposed in the mid-1980s suggests the casing stones of the
pyramids of Giza were in fact cast in-situ, using a granular limestone aggregate and
an alkali alumino-silicate-based binder (MacKenzie et al., 2011; Jana, 2007). On the
Indian subcontinent, an equally wide variety of cement types were in use from the
third millennia BC. Gypsum and lime cements, and bitumen mortars were utilised for
wells, drains and building exteriors in early Bronze Age settlements such as the
Mohenjo-Daro (Mound of the Dead) (Mays, 2010).

Between the third and first millennia BC, the kingdoms of Mesopotamia were also
creating binders from local raw materials. Clay, bitumen or lime-based cements were
used to bond adobe bricks or stone blocks in structures from courtyard houses to the
immense ziggurats (Barbisan & Guardini, 2007; Rogers, 1900). Bitumen was also
employed for waterproofing. Blended materials (e.g. bitumen with clay, or lime with
bitumen) were routinely used for specific applications (Moorey, 1994; Johnson,
1987).

42
One of the earliest examples of the use of cementitious materials is at the early
Bronze Age village of Yiftahel in Northern Israel. Yiftahel is one of the oldest
permanent villages ever found, and contains a number of concrete floors. Carbon-14
dating of seeds embedded in the floor suggests that it was laid approximately 8850
years ago, in two layers, consisting of a roughly compacted lower layer, and a
careful troweled surface. Analysis of samples from the floor showed that it is
composed of nearly pure calcium carbonate and a small amount of silica which was
probably sand, and exhibits compressive strengths equivalent to modern structural
concrete (Kanare et al., 2009; US Department of Energy, 2000).

When it comes to the use of cementitious binders, it seems there is little we could
teach our ancient ancestors!

43
1.7 Information sources on S/S systems
There is a considerable amount of information in the literature about S/S, including in
excess of 2000 journal papers and other documents. However, to obtain a balanced
and authoritative view on the merits and difficulties of using S/S to treat soil and
waste, the authors have identified a number of key information sources, which
together provide a wealth of information on S/S technology. These include regulatory
and industry guidance, the scientific literature and from networks and discussion
groups, and are given in Tables 1.2 to 1.6 below.

1.7.1 General information


A number of general information resources on S/S in Table 1.2 are available to
provide insight into all aspects of S/S.

1.7.2 Technical information


The documents identified in Table 1.3 provide more technical information on specific
aspects of stabilization/solidification technology.

1.7.3 Government departments and agency information resources


A number of agencies are responsible for the regulation and enforcement of
environmental policies relating to contaminated land, pollution, and treatment.
Information pertaining to these is given in Table 1.4.

1.7.4 Research, committees and associations


Listed in Table 1.5 are organisations that represent major industries related to S/S
and/or conduct research into key topics connected to S/S.

1.7.5 Forums, networks, and discussion groups


The list given in Table 1.6 includes platforms, which allow contractors, planners,
developers and researchers to share information and knowledge.

44
Table 1.2: General information sources on S/S
Publisher/Authors Date Title/Publisher
Environment Agency 2004 Guidance on the use of
Stabilization/Solidification for the Treatment of
Contaminated Soil. Science Report
SC980003/SR2.
Al-Tabbaa, A. and 2006 UK Stabilization/Solidification Treatment and
Perera, A.S.R Remediation Parts 1-7. Advances in S/S for
Waste and Contaminated Land Proc. Int.
Conf. on Stabilization/Solidification Treatment
and Remediation. pp 367-485.
BCA 2001 Cement-based stabilization and solidification for
the remediation of contaminated land. British
Cement Association Publication 46.050.
BCA 2004 The essential guide to stabilization/solidification
for the remediation of brownfield land using
cement and lime.
BCA 2005 Remediating brownfield land using cement and
lime. British Cement Association.
Concrete Centre
Environment Agency 2004 Review of Scientific Literature on the use of
Stabilization/solidification for the Treatment of
contaminated soil, solid waste and sludges.
Science Report SC980003/SR2.
http://publications.environment-
agency.gov.uk/pdf/SCHO0904BIFP-e-
e.pdf?lang=e/pdf
USEPA 1999 Solidification/stabilization Resource Guide.
Report EPA/542-B-99-002. April 1999.
http://www.epa.gov/tio/download/remed/solidsta
b.pdf
Construction http://www.ihs.com/products/design/uk-
Information Service solutions/construction-information-service.aspx
Contaminated Site http://clu-in.org/remediation
Clean-Up Information
Portal for Soil and http://www.eugris.info
Water Management in
Europe
Soil Environmental http://www.soilenvironmentservices.co.uk
Services

45
Table 1.3 Technical information sources on S/S
Organisation/Authors Date Title/Publisher
Barnett, F., S. Lynn, 2009 Technology Performance Review: Selecting
and D. Reisman and Using Solidification/Stabilization Treatment
for Site Remediation. EPA 600-R-09-148.
Army Environmental 1998 Solidification Technologies for Restoration of
Policy Institute (AEPI Sites Contaminated with Hazardous Wastes.
Conner, J.R 1997 Guide to Improving the Effectiveness of
Cement-Based Stabilization/Solidification.
Portland Cement Association. PCA: EB211.
Conner, J.R 1990 Chemical Fixation and Solidification of
Hazardous Wastes. New York, New York: Van
Nostrand Reinhold.
Conner, J.R, and 1998 The History of Stabilization/Solidification
Hoeffner, S.L. Technology, Critical Reviews in Environmental
Science and Technology, 28 (4), pp 325-396.
USEPA 1997 Innovative Site Remediation Design and
Application, Volume 4:
Stabilization/Solidification. EPA 542-B-97-007.
USEPA 2013 Superfund Remedy (Technology) Selection
Reports, First thru Fourteenth Editions.
http://clu-in.org/asr
EPRI 2012 State-of-the-Practice Liners and Caps for Coal
Combustion Product Management Facilities,
Electric Power Research Institute, EPRI, Palo
Alto, CA, October 2012, Report 1023741.
Interstate Technology & 2011 Development of Performance Specifications for
Regulatory Council Solidification/Stabilization.
(ITRC) http://www.itrcweb.org/GuidanceDocuments/so
lidification_stabilization/ss-1.pdf
Ramboll Norge AS 2009 Cement Stabilization and Solidification (STSO):
Review of Techniques and Methods, Ramboll
Norge AS, Oslo, Norway. Rap001-Id01,
57.2009.
Paria, S. and P.K. Yuet 2006 Solidification/stabilization of organic and
inorganic contaminants using Portland cement:
A literature review. Environmental Reviews
14(4):217-255.
46
PASSiFy Project 2010 Performance Assessment of
Solidified/Stabilized Waste-forms, An
Examination of the Long-term Stability of
Cement-treated Soil and Waste (Final Report),
CL:AIRE, RP16.
http://www.claire.co.uk/index.php?option=com_
cobalt&view=record&cat_id=23:stabilization-
solidification&id=298:performance-
assessment-of-stabilizedsolidified-waste-
forms-passify&Itemid=61
Spence, R.D. (Editor) 1993 Chemistry and Microstructure of Solidified
Waste Forms. Lewis Publisher.
Taylor, H.F.W. 1997 Cement Chemistry. Thomas Telford
Publishing, London.

Table 1.4 Information on S/S from government departments and agencies


Organisation/Body Address
Environment Agency http://www.environment-agency.gov.uk/
Department for https://www.gov.uk/government/organisations/department-
Environment, Food and for-environment-food-rural-affairs
Rural Affairs
United Nations http://www.unep.org/
Environment
Programme
European Environment http://www.eea.europa.eu/
Agency

Agency for Toxic http://www.atsdr.cdc.gov


Substances and
Disease Registry
Army Environmental http://www.aepi.army.mil/
Policy Institute
UK Government https://www.gov.uk/contaminated-land
United States http://www.justice.gov/enrd/
Department of Justice:
Environment and
Natural Resources
Division

47
USEPA https://clu-
in.org/techfocus/default.focus/sec/Solidification/cat/Overview/
European Union http://ec.europa.eu/environment/index_en.htm
Environment

Table 1.5 Information on S/S from research, committees and associations


Organisation/Body Address
Portland Cement Association http://www.cement.org/
Environmental Protection UK http://www.environmental-protection.org.uk/
Contaminated Land: http://www.claire.co.uk/
Applications in Real
Environments (CL:AIRE)
The Concrete Centre http://www.concretecentre.com/
Construction Industry Research http://www.ciria.org/
and Information Association
Interstate Technology & http://www.itrcweb.org/
Regulatory Council
The Chartered Institution of http://www.ciwem.org/
Water and
Environmental Management
Chartered Institute of http://www.cieh.org/
Environmental Health
British Cementitious Paving http://www.britpave.org.uk/
Association
Cement Association of Canada http://www.cement.ca/en

Table 1.6: Information on S/S from forums, networks, and discussion groups
Organisation/Body Address
Network for Industrially http://www.nicole.org/
Contaminated
48
Land in Europe
Stabilization/solidification http://www-starnet.eng.cam.ac.uk/
treatment and
remediation network
Brownfield Briefing: News, http://www.brownfieldbriefing.com/
Views, Analysis
Common Forum on http://www.commonforum.eu/
Contaminated Land
Association for Environmental http://www.aehsfoundation.org/
Health & Sciences (AEHS)
Environmental Knowledge https://connect.innovateuk.org/web/sustainabili
Transfer Network tyktn

49
2.0 History of S/S as a risk management tool

2.1 Use of S/S in the USA (Superfund and RCRA)1


The United States has been a leader in the application of treatment technologies for
remediation of hazardous waste sites including the application of
Solidification/Stabilization (S/S). Sites are remediated under a number of regulatory
programs including the well-known US Environmental Agency (USEPA) Superfund
program. Others include the USEPA RCRA Corrective Action program, State led
clean-up programs, and private party voluntary clean-up actions (without USEPA or
State oversight).

A comprehensive compilation of all site clean-up actions, or a comprehensive list of


all sites remediated by S/S is not readily available. Perhaps the most authoritative
data base available on the selection of S/S for site remediation is that compiled by
the USEPA Superfund program which tabulated 280 S/S source treatment
technology selections spanning the 30 fiscal years (FY) 1982-2011 (USEPA 2013,
Appendix B).

2.1.1 The regulatory impetus for the remediation of sites in the USA
A strong driving factor for development of treatment technology and hazardous site
remediation in the United States has been Federal legislation mandating
responsibility for management of hazardous wastes and remediation for sites
contaminated by past operations. In 1976 the US Congress passed The Resource
Conservation and Recovery Act (RCRA) which gives EPA the authority to control
hazardous waste from the cradle-to-grave. This includes the generation,
transportation, treatment, storage, and disposal of hazardous waste (USEPA 2011).
In 1984 HSWA - the Federal Hazardous and Solid Waste Amendments focused on
waste minimization and phasing out land disposal of hazardous waste (USEPA
2011).

Of considerable importance under RCRA regulations, is that a generator of


hazardous wastes is not absolved of liability for proper management and disposal of
these wastes by contracting these services to another party. RCRA however was
designed to apply to current and future generators of hazardous wastes.

1
The information presented in this section is based upon the authors experience, and
is not intended to present the policy of the USEPA, nor has it been reviewed and
approved by the USEPA.

50
In 1980, the US Congress enacted the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA, or Superfund) to address the dangers of
abandoned or uncontrolled hazardous waste sites. CERCLA provides the USEPA
and other federal agencies the authority to respond to a release or a substantial
threat of a release of a hazardous substance into the environment, or a release or
substantial threat of a release of "any pollutant or contaminant, which may present an
immediate and substantial danger to public health or welfare (USEPA 2007, page 1-
1). This legislation has been of great importance since it provides for taking action to
remediate abandoned sites and/or compelling responsible parties to take such action
even though the creation of the problem may predate regulations that prohibited such
disposal. This retroactive provision within CERCLA has been responsible for
stimulating the clean-up of many sites.

The Superfund Amendments and Reauthorisation Act of 1986 (SARA) expressed a


preference for permanent remedies (that is, treatment) over containment or removal
and disposal, in remediation of Superfund sites (USEPA 2007, page 1-1). This
preference for treatment over containment or disposal has been a powerful driver for
the development and application of treatment technologies, including S/S. Oversight
of contaminated site clean-up is generally done under one, or more, of the following
authorities:

Federal USEPA
Superfund - CERCLA
RCRA
State
Delegated Federal Authority
State Hazardous Waste Programs
State Voluntary Programs
Private Sector
Unregulated Sites
Licensed (Delegated) Professionals

The combination of RCRA (especially RCRA corrective action), for active sites and
sites previously active under RCRA permits and which still have financially viable
responsible parties, and the Superfund program for orphan sites and sites
contaminated prior to the RCRA, has been a powerful stimulant for the development
of treatment technology. A compiled database indicating the frequency of S/S use at
RCRA sites is not available, though RCRA does recognise S/S as a BDAT (Best
Demonstrated Available Technology) for treatment of many waste streams containing
metals.

51
Although not specific to S/S, some appreciation of the magnitude of the RCRA
corrective action clean up activity can be obtained from a recent USEPA publication
(USEPA 2006 - 2008). This publication indicated that in 2008, there were 1968
RCRA high priority sites, with 96.2% controlling human exposure, 83.4% involved
controlling the migration of contaminated groundwater, 43% involved choice of final
remedy, and 34.6% reported the final remedy was under construction or had been
completed.

As of January 7, 2014, the USEPA Superfund Program reported 1320 current sites,
53 proposed new sites, and 374 sites that had been deleted
(http://www.epa.gov/superfund/sites/npl/index.htm). The Superfund program
continues to provide frequent status reports that contain information on the frequency
of remedy selection, including S/S, for Superfund sites, which can be accessed at
http://clu-in.org/asr/.

2.1.2 Use of S/S in the EPA Superfund Program


The USEPA Superfund program has issued a number of excellent Superfund remedy
reports, and other documents, describing treatment technologies for the remediation
of the NPL sites under this program. Many of these documents can be accessed
through the website http://clu-in.org/remediation/.

Remediation of hazardous waste sites in the USEPA Superfund program is based


upon a risk management approach and compliance with applicable, or relevant, and
appropriate requirements (ARARs); meaning the site clean-up and remediation is
managed in such a manner, that residual risk is reduced to an acceptable level and
meets ARARs. Thus, since S/S generally does not remove or destroy the COCs
(contaminants of concern), S/S accomplishes its risk reduction objective by blocking
the pathway between the COC and the receptor (human or the environment). S/S
does this by either reducing the solubility of the COC (stabilization) or by containing
the COC in a low permeability matrix (solidification), or by using both processes
(solidification/stabilization).

The Superfund Remedy Report, Thirteenth Edition, remarks


solidification/stabilization continues to be the most frequently selected ex-situ source
treatment technology (EPA 2010 page 9). In this same report, S/S is listed as the
second most frequently selected in-situ source treatment technology from 2005-2008,
exceeded only by in-situ soil vapour extraction.

The most recent Superfund Remedy Report, Fourteenth Edition, reported that for FY
2005-2008, and FY 2009-2011, ex-situ S/S was selected for 19% and 13% of the
source treatments respectively, and is still the second most commonly chosen ex-
52
situ remedial technology for sources (USEPA 2013, page 9). This same report
indicates that in-situ S/S was selected for 9% of the source treatments for each of the
periods FY 2005-2008 and FY 2009-2011. Also of interest is that during FY 2009-
2011, S/S was selected in 11% of the 56 decision documents expressing remedy
components for sediments (USEPA 2013, page 11, Table 2).

Appendix B of this report, (EPA 2013), provides a table showing treatment


technologies selected for each fiscal year from 1982 thru 2011. Based upon this data,
Figure 2.1 was prepared and displays the most frequently selected source treatment
technologies over this 30 year period. Figure 2.1 shows S/S as the second most
frequently selected technology at 22% of remedy documents selecting source
treatment, slightly behind soil vapour extraction at 24%. Figure 2.2 uses this same
data base to graph the frequency of selection for S/S verses all other source
technologies over this 30 year period. Figure 2.2 shows that S/S (combined ex-situ
and in-situ) has consistently been selected in about 20% of all remedy documents
selecting source treatment under Superfund.

In the absence of more recent data, Figure 2.3 gives the generalised contamination
found at sites for which S/S was selected under the Superfund Program. The
literature is lacking a study, evaluating over time, trends in the frequency of selection
of S/S for treatment of sites with organics. However, in the author`s experience, there
seems to be increasing acceptance that S/S can effectively treat many non-volatile
organics, alone or mixed with toxic metals.

In-situ S/S seems to be increasing modestly as a remedial technology, slightly


displacing ex-situ S/S. Figure 2.4 compares the relative frequency of selection of in-
situ S/S with selection of ex-situ S/S. Although selection of S/S overall, combined in-
situ and ex-situ has remained fairly constant since about FY 2000, in-situ S/S
appears to be increasing as a proportion of all S/S treatment selections. Perhaps this
is due in part to the ability of in-situ S/S to treat wastes to substantial depth, including
below the water table, without the need to de-water.

53
In-Situ Plus Ex-Situ
Soil Vapour Extraction
11
Solidification/Stabilisation
24

15 Physical Separation

Incineration

8
Thermal Desorption
22
12 All Other

8
Bioremediation
% of 1266 Treatment Technology selections (Excluding Groundwater)

Figure 2.1: Cumulative source control technology selection from 1982-2011


(After EPA-542-R-13-016, Appendix B)

350
1266 Source Remedy Decisions (Excluding groundwater)

300

250

200
All other
150 S/S

100

50

0
1982-86 1987-91 1992-96 1997-2001 2002-06 2007-11

Figure 2.2: Selection of S/S vs all other technologies 1982-2011 (After: USEPA
542-R-13-016, Appendix B)

54
2.1.3 Conclusions
Decades of experience in the application of S/S technology to a wide variety of
contaminant and soil types makes the USEPA Superfund Program experience a
valuable resource for anyone involved with S/S, or any treatment technology. For the
last 30 years, S/S has consistently been a technology of choice for the USEPA
Superfund program, being selected consistently in about 20% of all remedy
documents with source area treatments. This trend continues today with a number of
sites in active remediation employing S/S in 2014. The USEPA RCRA Program has
also designated S/S as a BDAT technology for many waste types containing metals.

In view of the large number of successful applications over extended time within the
USA, it is surprising that S/S has not been more widely used in other countries. This
is even more remarkable considering that in the USA, for S/S treated sites, as with
any site that leaves contaminants on the site, the Superfund program requires a
review every 5 years, to assure that the remedy is performing as required.

6 Organics only 6%
7

31
All Other 7%

Metals Only 56%

56
Metals & Organics 31%

Figure 2.3: Waste types selected for S/S (EPA-542-R-00-010)

55
280 S/S Remedy Decisions Over 30 Years

Figure 2.4: S/S selection of in-situ vs ex-situ 1982-2011 (After USEPA 542-R-13-
016, Appendix B)

56
2.2 Use of S/S in the UK and Europe
Although the use of S/S can be traced back more than 50 years in North America,
uptake of this technology (for treating contaminated soil) has been much slower in
Europe. This primarily arose because of the historical low cost and plentiful supply of
landfill space, and for much of this time, a pervasive immature legislative
environment.

The use of S/S in the UK can be traced back to the early 1970s for the treatment of
mixed wastes prior to co-disposal with domestic waste, in unlined quarry workings.
Two centralised S/S treatment sites operated in the UK, one in the English Midlands,
and the other East of London.

2.2.1 Treatment of waste and soil by S/S


By the late 1950s S/S was being used in France to treat low-solids containing waste
(Conner, 1990) and by 1978 was routinely used to treat hazardous waste
(Environment Agency, 2004a), having been originally developed for cementing
radioactive residues. The focus on wastes remained and by the early 2000s, S/S
was being used for waste treatment in France, Austria, The Netherlands and in
Portugal.

In France, in 2004, there were 12 central processing facilities employing S/S to treat
400,000 tonnes of waste per year (Pojasek, 1978).

In the English Midlands, a waste treatment plant started operation in the 1970s
primarily processing metal plating residues by a process called Sealosafe (Conner,
1990). Over 1 Million tonnes of materials were treated, before the operation was
closed amid concerns over product quality (ENDS, 1988). The operating company
was eventually successfully prosecuted in 1990 under Trades Descriptions
legislation, although it should be said that doubts on the efficacy of the S/S product
were reported as early as 1983 (ENDS, 1983). It is a widely held view that the
reported failures at this commercial plant held back the uptake and acceptance of
S/S in the UK for a number of years. Nevertheless, the second site, East of London,
successfully employed the Stablex process between 1978 and 1996, and up to
400,000 tonnes of hazardous wastes were treated each year and placed in a sanitary
landfill (Conner, 1990).

The first widely reported example of contaminated soil treated by S/S in the UK took
place at the former ICI explosives plant at Ardeer in Scotland in 1995. Here,
10,000m3 of soil were treated by in-situ mixing using a lime, cement and slag-based
binder system (Wheeler, 1995).
57
2.2.2 Legislative background
In the UK, the Public Health Act of 1848 recognised an accumulation or deposit that
was prejudicial to health or a nuisance. Attempts to regulate industrial waste (with
reference to waste produced from alkali manufacture) were made in 1863 (Kiefer,
2012). However, since 1973, the direction of environmental protection across the
European Community area has been articulated in the European environmental
action programmes, which were first implemented in that year.

European environmental law, through single-issue Directives, and more recently


Frameworks, shapes legislation in individual countries including the UK. This
facilitates an integrated approach to environmental law making across the
Community, but allows local variations in the way the laws are enacted.

Although the origins of European Environmental Law date back to the 1970s with air
and water directives, the 1987 European Treaty (The Single Environment Act, 1987)
was the first to include a policy in the sphere of the environment. The Polluter Pays
principle was subsequently established (The Sixth Environment Action programme of
the European Community 2002-2012) introducing liability and an incentive to
incorporate environmental considerations into the design of products or processes.

In the UK, the Control of Pollution Act (1974) dealt with a number of environmental
issues including waste disposal, atmospheric pollution and public health. The COPA,
as it is known, was augmented by the separate Special Waste regulations in 1980,
and amended in 1989 (Control of Pollution (Amendment) Act (1989)). The COPA was
also re-enacted by the 1990 Environment Protection Act, allowing for improved
control over emissions to land from industrial and other processes, and the
introduction of risk into the assessment of contaminated land and its condition.

In 1999, the Pollution Prevention and Control Act introduced the requirements of
European Council Directive 96/61/EC on Integrated Pollution, Prevention and Control
(IPPC) to ensure the best technical option is available to prevent emissions to air,
water and ground. The strong emphasis on risk reduction and minimisation was a
fundamental step-change over earlier legislation.

The European Council Directive 99/31/EC (The Landfill Directive) is also of great
relevance to S/S as it categorised waste on harmfulness and the division of landfills
into three classes:

58
Landfills for hazardous waste
Landfills for non-hazardous waste
Landfills for inert waste

Standard waste acceptance procedures were introduced, and the pre-treatment of


waste (prior to acceptance) was accompanied by a rigorous system of permitting.

With respect to S/S, its use is facilitated by:

The option to pre-treat waste/soil by S/S prior to landfilling


The accepted use of appropriate risk management option to prevent waste
going to landfill
A substantial increase in the costs of landfill disposal particularly in the UK
with the introduction of the Landfill Tax, promoting waste (and soil) recycling
and re-use

2.2.3 Public and private initiatives


With an emphasis on risk management and the diversion of waste from landfill there
was an increased interest in the use of S/S throughout the 1990s. In 2000, national
guidance (England and Wales) was being developed in a partnership between
industry and government. This project, called CASSST (Codes and Standards for
Stabilization/Solidification Technology), was initiated at the University of Greenwich,
and resulted in the publication of national guidance on S/S by the Environment
Agency (2004b).

In the first of its kind, this EA guidance was supported by an exhaustive review of the
science behind S/S (Environment Agency, 2004a), to allow UK stakeholders to be
conversant with the potential strengths and weaknesses of the technology, and to
facilitate the formulation of the best design solutions employing S/S.

Other UK initiatives in the early 2000s included a research council sponsored


academic-led network (STARNET, 2004), industry-led guidance BCA (2004) and an
EU-supported project on S/S waste form performance (PASSIFY 2010). The latter
project, called PASSIFY (Performance Assessment of Stabilized/Solidified Waste
Forms) also involved the EA, the EPA, academics and stakeholders in the USA, UK
and France, including ADEME. Under this initiative a number of S/S remedial
operations were examined in detail in each country, including several Superfund
sites.

59
In the years following the publication of Environment Agency Guidance, the uptake of
S/S in the UK has increased and example remedial operations are cited in the
literature (STARNET, 2004; Environment Agency, 2004b). The use of S/S is now
firmly established as a credible risk management option for both soil and waste, as
exemplified by its use at the Olympic Park in London (Penseart, 2008).

Currently, in Europe no comparable national (or Community) guidance on S/S exists


and applications of the technology to contaminated soil remain second to those of
waste treatment. In many EU member countries, landfill still remains a relatively
cheap option, and until the cost and availability of disposal become prohibitive, landfill
will remain as an attractive available alternative to risk management options such as
S/S.

It is known that in France, guidance on S/S is being considered and reports indicate
that within 3 years ADEME may publish in this respect (Chateau 2012, pers. com).

60
2.3 Use of S/S in Canada
The use of S/S in Canada has been primarily focussed on the treatment of
contaminated soil and management of mining spoil.

Examples of Canadian S/S projects since 1990 included remote military bases,
former industrial facilities, urban waterfront redevelopments, and mining sites.
Example remedial actions are documented by, for example, the Cement Association
of Canada (www.cement.ca), and sites treated by S/S include:

The Dockside Green redevelopment project in Victoria BC, for lead in soil to
meet provincial hazardous waste leachate standards
Western Steel Mill, Vancouver BC, for cadmium, lead, and zinc waste
A rifle range in Burnaby, BC for lead, zinc, copper and antimony to meet the
provincial hazardous waste targets
Glacier National Park, BC, for lead in soil
False Creek, Vancouver (Winter Olympic project) for zinc and pH in soil for a
residential property redevelopment
Canmore, Alberta, for mine waste recycling as on-site structural fill for a
mixed-use residential development
Swan Hills, Alberta, for spray dryer salts and bag house fly ash for disposal
St. Catherines, Ontario, for lead and PAH in soil to meet the Ontario Land
Disposal Regulations at a 5 ha (50,000m2) property development
Sydney, NS, for petroleum hydrocarbons in sediments at a former steel mill, to
meet strength and permeability criteria

2.3.1 Performance (risk)-based versus prescriptive regulations


A prescriptive approach may call for a compound concentration in soil or water to
be reduced to a specific value, so that the resulting risk or hazard by some assumed
exposure conditions are acceptable. Prescriptive based environmental project goals
were almost universally used in Canadian jurisdictions in the early 1990s.

A performance approach involves more site specific information, and may call for
changes in compound concentration, migrating characteristics, exposure route
conditions, receptor characteristics, any other variable, or combination thereof such
that the target acceptable resulting risk or hazard is not exceeded.

While an option for defining site specific performance goals existed in some
Canadian jurisdictions in the early 1990s, Nova Scotia was the first province where
performance goals were being consistently specified and used for remedial
operations. Within 5 years of the 1996 Nova Scotia Environments Guidelines for the
Management of Contaminated Sites, the other Atlantic Canadian provinces (New
Brunswick, Prince Edward Island, and Newfoundland and Labrador) revised their
own regulations.

In 2007, new guidelines were issued in Alberta to encourage the use of


performance-based environmental goals, followed by the same in British Columbia.
Subsequently, the Federal Government of Canada released revised guidance for
sites under federal jurisdiction. Nunavut, the Yukon Territory and the Northwest
Territories now incorporate federal guidance for the use of S/S within their own
jurisdiction.

Since 2005, S/S has been used in mining project developments in Canada to reduce
the production of waste material and to protect ground and surface water resources.
The result from the S/S of mine tailings is their accepted use as a sustainable
construction material for use as structural backfill.

2.3.2 Relative use of S/S in Canadian provinces


Canadian jurisdictions with the greatest experience in using performance (risk)-
based objectives tend to also be those where the most significant S/S projects have
occurred (including Nova Scotia). This was noted in an earlier review (Ells, 2010) of
the relative regulatory receptiveness among Canadian jurisdictions to S/S activities
as:

Relatively receptive (e.g. Nova Scotia, New Brunswick, Newfoundland and


Labrador, Alberta, and British Columbia)
Initially Receptive (e.g. Prince Edward Island, Ontario, Quebec, Government of
Canada, and Manitoba)
Not Yet Receptive (e.g. Northwest Territories, Nunavut, Yukon Territory, and
Saskatchewan)

2.3.3 Sydney Tar Ponds and coke ovens project Nova Scotia
When it was completed in 2013, the Sydney Tar Ponds project was the largest S/S
remedial operation in the world, and the most publically and politically prominent
contaminated site remediation project in Canada. At Sydney, over a million tonnes of
sediments and soil contaminated with oily residues and metals from the production
of steel and coke were treated using S/S.

62
The performance specification used at Sydney included hydraulic conductivity
(permeability), SPLP (leachability) and unconfined compressive strength. Treatment
protects the surface waters, the important fishery in Sydney Sound and provides for
land for recreational and light industrial activities. Because of the prominence of this
remedial action, a huge amount of technical information is publicly available. A case
study on this project is included in Appendix B.
2.4 Risk management by S/S
The principles of risk-based remediation are now embedded in contaminated land
legislation across the US and Europe. A suitable for use approach to contaminated
land management is now widely adopted, implementing remedial action where it is
demonstrated that unacceptable risks are being caused (or could be potentially
caused) by contamination on a site for a designated end use or for potentially
affected human or environmental receptors. This section will explain the basic
principles of risk management, within which S/S is applied.

2.4.1 Definition of risk


Risk can be defined as a combination of the probability, or frequency, or occurrence
of a defined hazard and the magnitude of the consequences of the occurrence. In
the context of land contamination, this relates to the potential for contaminants to
harm human health, impact water resources and ecological receptors, and damage
buildings and infrastructure.

Assessment of risk is founded on the source-pathway-receptor concept, identifying:

Source - the pollutant hazards associated with the site


Receptor - possible receptors at risk from the identified hazards (e.g.
human, water resource, flora and fauna, buildings and infrastructure)
Pathway - a route or means by which a source can impact a receptor

For risks to be present at a site, all three elements (source-pathway-receptor) of a


plausible pollutant linkage must be present.

2.4.2 Risk management frameworks


Whilst factors such as acceptable levels of exposure and exposure models vary with
individual nations policies and legislation, most countries share a common risk-
management framework structure based on site characterisation, environmental risk
assessment, and evaluation of remedial measures, remediation implementation and
validation (e.g. see Rudland et al., 2001).

63
A tiered approach to risk assessment is adopted, with assessments increasing in
detail where risks are potentially unacceptable:

Preliminary risk assessment: a desk based, qualitative assessment of potential


risks. Information on the site history, ground conditions and environmental setting is
collated and used to develop the conceptual model for the site.

Generic quantitative risk assessment: a comparison of contaminant


concentrations with generic assessment criteria. Intrusive ground investigations are
undertaken to obtain representative soil, water and soil-gas data to develop the
conceptual site model. Generic assessment criteria are conservative, applying to
limited defined land end uses and pollutant pathways.

Detailed quantitative risk assessment: the derivation of detailed assessment


criteria using site-specific data. Comparison of soil, water and gas data site-specific
assessment criteria or target levels to determine if unacceptable risks are present.

Phasing the risk-assessment process permits action to be taken rapidly to resolve


obvious problems, whilst more detailed assessment in other scenarios may
demonstrate no unacceptable risks are posed. A phased risk-based approach allows
financial resources to be allocated most effectively.

Remediation using a risk-based methodology is an alternative approach to the total


clean-up of a contaminated site. A risk-based approach can be used to establish that
land use is neither technically or financially feasible, nor sustainable, or to establish
that land use is possible and that the costs of treatment are acceptable.

However, by adopting a risk-based approach the ground conditions, environmental


setting, and current and proposed site usage can be considered. Furthermore, soil
type, geology, hydrogeology, the water environment, local ecosystems all impact on
the risk posed by contamination. Similarly, the form, intensity and frequency of a
receptors exposure to contaminated soil, groundwater and soil gas are dependent
on the end use of the site. As an example, as residential end use is more sensitive
than an industrial end use, a greater degree of remediation would be required.

Where the level of risk justifies corrective action, risk-management can take the form
of managing the receptor, breaking the pathway or reducing or removing the source
material.

64
S/S does not remove or destroy contaminants, but immobilises them by a
cementitious reaction involving the soil matrix, contaminants and binder reagents to
promote sorption, precipitation or incorporation into crystal lattices, and/or by
physically encapsulating the contaminants.

2.4.3 Summary
Treatment by S/S therefore breaks the pollutant linkages by both breaking the
pathway and reducing the source material. However, the release of contaminants will
still occur from treated material, and the objective of the design process is to produce
a durable waste form from which release is controlled, to levels that do not pose an
unacceptable risk to receptors.

It should be noted that dermal and volatile exposure pathways are not necessarily
broken by S/S. Further thought as to the use of the technique in conjunction with
others is needed, or possibly the careful reuse of treated materials as part of an
engineered solution. These concepts are discussed in more detail in the following
section.
2.5 S/S and the Risk Framework
To understand how S/S fits into the risk framework, it is important to consider how
S/S breaks pollutant linkages to different receptors. By assessing different potential
receptors and pathways in turn, it is possible to identify issues specific to S/S that
must be considered as part of the environmental risk assessment process and
stabilization design.

2.5.1 Water environment


The chemistry involved in S/S is complex, but the mechanisms considered to reduce
the mobility of contaminants include:

Sorption to soil and binder / additive materials


Precipitation as a result of pH modification, reducing solubility
Incorporation within the crystal lattice
Encapsulation within the CSH gel formation

Infiltration and leaching pathways are reduced as a result of decreased permeability


and reduced mobility of the contaminants. Mass flux of contaminants to groundwater
and surface water receptors is therefore also reduced.

65
A robust hydrogeological model is an essential part of understanding the potential
risks and the applicability of S/S. It is important not only to consider potential
pollutant linkages, but also to be able to anticipate the groundwater flow regime and
chemistry. No mass reduction in contaminants occurs; rather the risks are managed
via a controlled very slow release of contaminants. Factors that are needed to
develop a conceptual site model are discussed in more detail in Section 7.2.
However, in the context of the risk-management framework, factors that can affect
the potential durability and long-term leaching performance of treated materials
require early consideration, to ensure they are adequately addressed throughout the
design and approval process.

Considering the risks to water environment receptors (surface waters and


groundwater) must first involve understanding the likely leaching behaviour from the
S/S treated material. Key to this is the development of an appropriate testing regime
representative of the likely end-use scenario. Potential exposure to aggressive (e.g.
acidic and sulphate rich soils and groundwaters) or saline ground conditions or
environments where a high degree of carbonation could be envisaged should be
identified at the conceptual model development stage. The identification of
aggressive conditions may require additional performance testing where necessary,
and modified standard testing procedures to represent the site-specific conditions
highlighted.

Following this, a comparison of likely release levels with remediation targets is then
needed.

Although remediation targets are usually established by risk assessment, and are
site-specific, they can be based on drinking water standards or health advisory levels
for contaminants of concern. As such, the appropriate compliance points are
selected at which environmental standards must be met to protect the receptors.
These may be the aquifer or surface waters or some point nearer to the source.
Target concentrations will reflect background concentrations, current and future use
of the water resource and environmental standards.

A tiered approach to assessment is then usually undertaken, with an initial


assessment determining whether the target concentration is exceeded at the
compliance point, then subsequent tiers of assessment consider dilution, dispersion,
retardation and degradation by biotic or abiotic processes at increasing levels of
complexity. The level of assessment undertaken is usually dependent on the level of
risk posed, or the relative sensitivity of the receptor.

66
When comparing leaching test results with remediation targets, care needs to be
applied to ensure that the comparison is relevant. Groundwater models can require
comparison with pore water concentrations, and consideration needs to be given to
exactly what data is being compared to what target, and if a direct comparison is
appropriate. As discussed in later sections, a wide array of leaching test methods is
available, all reflecting different leaching behaviour of the treated materials. An
understanding of what the leaching data is telling you is needed and the risk
assessment and testing regime should ideally be developed in conjunction.

Potential surface water run-off pathways also need consideration. On a short-term


basis, the potential risks to surface water receptors during construction (from
material storage or partially treated materials) will require addressing during
development of the environmental mitigation measure strategy. However, on a
longer timeframe, the environment of service of the waste form, and what
environmental loads may impact upon it (if any) will need to be assessed so there
are no unforeseen adverse impacts resulting from the remedial action.

2.5.2 Human health


The key potential exposure pathways to human health receptors can be summarised
as:

Ingestion of soil, dust, home grown produce and waters


Dermal contact with soil, dust and waters
Inhalation of soil, dust and vapours from soil and waters

Whilst S/S can feasibly reduce generation of dust via production of a monolith, other
pathways are not addressed by S/S treatment alone. Treatment may not reduce
exposure to volatile contaminants, although the resulting reduction in permeability
may restrict vapour movement to some extent. Treated material will also have a high
pH, which would cause problems via the dermal pathway. Considerable attention
therefore needs to be given to other potential pollutant linkages, and whether a
treatment-train approach needs to be adopted.

The principal of a treatment-train is simple the use of multiple treatment methods


used either sequentially or simultaneously to enable the treatment objectives to be
met. Common techniques may include bioremediation, chemical oxidation, soil
vapour extraction and even incineration, depending on the range of contaminants
present. Where other techniques are to be used in conjunction with S/S, the
treatability trials must be designed to reflect this, to ensure testing is appropriate.

67
Another option is to use a low permeability-capping layer (Section 9). The placing of
treated material at depth on a site can sometimes be sufficient by itself, supported by
a detailed human health risk assessment to demonstrate that the risk from volatile
contaminants is acceptable.

A final consideration with respect to human health receptors is whether the treatment
process itself can increase the volatilisation of contaminants. The heat of hydration
from binder components, particularly with a lime binder, may result in vapour
emissions. This should be considered in the environmental risk assessment, so that
appropriate mitigation or control measures can be put in place if needed.

2.5.3 Other receptors


The potential for adverse impacts on other receptors such as vegetation, ecological
receptors and buried services must also be considered as part of a risk-based
remediation solution. The reduced permeability, high pH and high contaminant
concentrations in treated material could have an adverse impact on plant growth and
wildlife, and reuse of stabilized materials within soft landscaping areas should not be
recommended. Mitigation measures such as a capillary break layer may be needed
to prevent alkaline groundwater causing problems.

With respect to buried services, considerations are more practical, such as whether
the placement of the hardened treated material restricts access to services, and if it
could be necessary to dig through it as part of later maintenance activities.
Mitigation typically comprises a thorough validation of the extent and location of
treated material, and inclusion within operation and maintenance plans to ensure
these risks are communicated to future maintenance workers so that appropriate
precautions can be taken. Sometimes a clean Corridor is constructed through the
treated material and utilities are placed within this clean corridor.

2.5.4 Durability and integrity


In understanding the environment of service of a waste form, it may be necessary to
incorporate additional protection measures. These may include a capping layer to,
for example, restrict infiltration, or measures to mitigate the effects of freeze-thaw.
The thorough investigation of likely environmental and other impacts after treatment
is a key part of the development of the conceptual model and ensures an appropriate
testing regime is adopted to test the S/S formulation adopted to provide confidence
in future behaviour.

In ensuring the durability and integrity of the waste form, S/S can be part of a risk-
based remediation framework. A good environmental risk assessment will consider
68
the risk factors for degradation of S/S materials as the conceptual site model is
developed, and in this respect a comprehensive review of relevant factors is
provided by the Environment Agency (EA 2004b) and (Hills et al., 2013).
2.6 Future trends
As was discussed in Section 2.1, S/S has been a workhorse technology for
remediation of hazardous waste sites in the United States since at least 1982. S/S
has been selected in ex-situ and in-situ applications in nearly one quarter of
Superfund remedy documents that selected source treatment. Figures 2.1 and 2.2
(Section 2.1) indicated that the selection of S/S has remained nearly constant over
30 years from 1982-2011.

The specific reasons for the popularity of S/S are, in the authors opinion, a
combination of the following factors, as S/S:

Is applicable to most inorganic and a wide selection of organic contaminants


(see Section 3.1)
Is competitive on cost for many applications
Can be deployed using readily available equipment (and materials) with a
minimum set-up time
Can be completed faster than most other technologies
Has an extensive tract record of success on a wide range of site and
contaminant types

2.6.1 USA
Given the long history of selection and successful use, the authors expect that S/S
will continue to be a popular risk-management strategy available well into the future.
To this end in the USA, the 2011 publication of guidance on S/S by the Interstate
Technology Regulatory Council (ITRC 2011) provides authoritative reference
material to the regulatory community, and this will enable greater acceptance and
use of this technology.

However, the authors believe there will be changes going forward, with S/S being
employed more frequently at sites containing non-volatile organic contaminants,
either alone or in combination with inorganic contaminants. This expanded
application of S/S is supported by the increased ability/availability of experienced S/S
experts in developing treatment formulations that successfully immobilise organic
contaminants, especially non-volatile organics. The successful treatment
formulations have included, for example, organophilic clays, activated carbon, and
ground blast furnace slag, and will enable more sites (containing a wider variety of
organic contaminants) to be successfully treated.

69
Historically S/S has been employed far more often in ex-situ than the in-situ
applications. More recently, however, in-situ applications are increasing in frequency,
as illustrated in Figure 2.5, which is based upon a recent USEPA Superfund remedy
selection report (EPA, 2013). This trend may reflect the increased depth of treatment
possible with higher power in-situ augers and the use of longer Kelly bars. It could
also be due to greater availability of in-situ treatment equipment other than augers
(described in Section 5).

Figure 2.5: Relative increase in selection of in-situ vs ex-situ S/S (USEPA 542-
R-13-016, Appendix B)

2.6.2 UK and Europe


In the UK and Europe one of the key drivers for using S/S is the availability and cost
of landfill space, and the restrictions placed upon the digging and dumping of waste
and soil through the EU Landfill Directive (Directive 99/31/ED) and associated
legislation, in different EU member countries. As options for managing soil and waste
become more expensive and restricted in nature, S/S can be expected to become
more widely appreciated and credible as a low-cost management option.

70
In the Authors experience, the reluctance to treat organic or mixed contamination is
becoming less and, by way of example, gas-works sites (MGP) are now routinely
treated by S/S in England and Wales.

Development work also continues on the use of novel binding systems employing,
for example, geo-polymers (Fernndez Pereira et al. 2009; Guo and Shi, 2012)),
and the use of CO2 activated systems (Lange et al. 1996,1997; Gunning, 2011) to
produce carbonate-cemented S/S products.

2.6.3 Conclusions
For over two decades S/S has consistently been a favoured technology for use by
the USEPA Superfund program. S/S has been selected consistently for over 20% of
all sites treated, and this trend continues today, not only within the Superfund but
also with sites remediated under different jurisdictions.

The authors believe that there will be an increase in use of S/S for sites with organic
contaminants as the effectiveness of applying S/S to organic contaminants has
improved and acceptance by the remediation industry and regulatory authorities for
treating organics with S/S has increased. There is also an increasing trend for
applying in-situ S/S, as available equipment continues to improve and its capability
of working at greater depths, including below the water table and without having to
de-water.

71
3.0 Applicability of S/S

One of the strengths of S/S is that it can be applied to metals, other inorganic
compounds, and most organic compounds. This section briefly discusses the many
inorganic and organic contaminants for which S/S has been successfully applied.
However, for a detailed review of the chemistry of individual inorganic and organic
compounds, and how these interact (with the binder) in an S/S system the reader is
directed to the informative book written by Conner (1990).

Further discussion of the mechanisms involved with specific compounds can be


found in the reference material identified in Section 1.7, and in the accompanying
science volume of this work.

3.1 Metals
Early in the development of S/S, the treatment technology was primarily applied to
metals, as the chemistry utilised to remove metals from solution was already
established and practiced for water treatment. Since metals cannot be destroyed,
S/S treatment focused on producing a less mobile and less toxic form of the metals
in waste materials.

Though any metal can be chemically immobilised by S/S, the primary metals of
environmental concern are antimony (Sb), arsenic (As), barium (Ba), beryllium (Be),
cadmium (Cd), cobalt (Co), copper (Cu), chromium (Cr), lead (Pb), mercury (Hg),
nickel (Ni), selenium (Se), silver (Ag), Thallium (Tl) and zinc (Zn). These metals are
those listed in the EPAs Universal Treatment Standards (40CFR 268.48). Figure
3.1 and 3.2 shows typical metals-contaminated waste requiring S/S treatment.

3.1.1 Chemistry
The chemistry involved in S/S is complex, and still remains poorly understood for
some contaminants. For both metals and (some) organic contaminants, sorption
processes, precipitation and incorporation into solid cementitious phases may occur
during S/S. In order to present key aspects of the nature of metal contaminants, the
properties of important metals are discussed below.

Antimony, Sb, is a Group V element, has valence states of +3, +5, and -3, which
complicates its chemistry. Antimony has definite cationic chemistry, but only in its
trivalent state. It forms both oxides and sulfides readily in its trivalent state.
Reduction of antimony to the trivalent state often simplifies its chemical fixation.
72
Arsenic, As, is a Group V element with principal valence states of +3, +5, and -3.
The valence state can be changed readily and reversibly. It combines with other
metals to form arsenides, and forms both oxides and sulfides. Arsenic chemistry is
complicated by its variety of valence states and ability to be present in both cationic
and anionic species in solution. Often the redox potential must be adjusted to cycle
the arsenic into only one valence state prior to chemical immobilisation.

Figure 3.1 Lead-contaminated soil and battery casings from a former battery-
recycling site, subsequently chemically stabilized

Beryllium, Be, is a Group II element with a predominantly covalent chemistry,


though it forms a cation with a valence state of +2. Beryllium forms a low solubility
hydroxide at near normal pH values. Beryllium hydroxide is amphoteric, creating a
beryllate (Be(H2O)42-) anion in alkaline solution.

Cadmium, Cd, is a Group IIB element, with one valence state of +2. Cadmium forms
low solubility compounds with carbonate and hydroxide, and can be chemically
fixated by alkaline precipitation. Cadmium also forms a low solubility phosphate

73
compound. Because cadmium forms stable complexes with ammonium, cyanide,
and halides, it will not precipitate in the presence of these complexing agents.

Cobalt, Co, is a Group VIII transition metal with primary valence states of +2 and +3.
Cobalt forms low solubility phosphate, sulfide, and arsenosulfer compounds. Cobalt
forms stable hydroxides, which are amphoteric at high pH.

Copper, Cu, is a Group IB metal, with a primary valence state of +2. Copper forms
low solubility carbonate, hydroxide, phosphate, sulfur, and arsenosulfur compounds.

Figure 3.2: Chromium paint pigment-contaminated soil being chemically


stabilized in-situ

Chromium, Cr, is a Group IVB metal. It has three valence states (+2, +3, +6), but
+3 and +6 are the most prevalent. While Cr +3 is poisonous, Cr+6 is highly toxic and
carcinogenic. In the +3 valence state, chromium is cationic and forms low solubility
carbonate, hydroxide, phosphate, and sulfide compounds. In the +6 valence state,
chromium is predominantly present as an anion (CrO 4-2 or Cr2O7-2), and can form
insoluble compounds with barium or lead. Typically, Cr6+ compounds are reduced to
their trivalent state and then immobilised, reducing both the toxicity and solubility.

74
Lead, Pb, is a Group IVA element, with a primary valence state of +2. Lead forms
low solubility carbonate, chromate, halide, phosphate, sulfate, and sulfur
compounds. Lead forms stable hydroxides, which are amphoteric at high pH.

Mercury, Hg, is a Group IIB metal, existing mainly in three valence states: 0, +1, and
+2. While Hg0, elemental mercury, is insoluble, it often has a layer of HgO on its
surface giving it an appearance of solubility. Therefore, the surface of Hg 0 is often
reacted with either sulfides to form insoluble mercuric sulfides or metals (e.g. copper)
to form amalgams. Cationic mercury species form low solubility halide, hydroxide,
phosphate, and sulfides.

Nickel, Ni, is a Group VIII transition metal, with a primary valence state of +2, though
valence states of -1, 0, +1, +3, and +4 are also known. Cationic nickel species form
low solubility carbonate, hydroxide, phosphate, and sulfide compounds.

Selenium, Se, is a Group VIB element, with principal valence states of -2, 0, +4, and
+6. The valence state can be changed readily and reversibly. It combines with other
metals to form selenides, and forms both oxides and sulfides. Like arsenic, selenium
chemistry is complicated by its variety of valence states and ability to be present in
both cationic and anionic species in solution. Adjustment of the redox potential can
be used to adjust selenium into only one valence state prior to chemical
immobilisation.

Silver, Ag, is a Group IB metal with a primary valence state of +1. Silver forms low
solubility halide, hydroxide, and sulfur compounds.

Thallium, Tl, is a Group III metal, with valence states of +1 and +3, though it is
typically found in the +1 valence state. As a cation, thallium forms low solubility
compounds with halides (except F), hydroxide, and sulfide.

Zinc, Zn, is a Group IIB metal, with a primary valence state of +2. Zinc forms low
solubility compounds with carbonate, hydroxide, phosphate, and sulfide.

75
3.2 Other inorganic compounds
As with metals, many inorganic compounds cannot be destroyed, or are recalcitrant
in nature. Treatment of metals and other inorganic compounds by S/S is focussed
on producing a less mobile, less toxic form of the compounds involved, or through
physical entrapment in the cementing phase or pore structure of the waste form.

Anions that are often treated by S/S include: fluoride (F -), cyanide (CN-), nitrate
(NO3-), phosphate (PO4-3), and sulfide (S-2); however, chloride can be readily
incorporated in a hardening cement-based system, but that exceeding the binding
capacity of the cement involved will be readily lost by dissolution/diffusion on
exposure to a hydraulic gradient.

3.2.1 Chemistry
Fluoride, F-, is easily immobilised as calcium fluoride.

Cyanide (soluble) CN-. Soluble cyanides, including loosely-complexed cyanides


such as nickel or cadmium cyanides, readily form insoluble compounds with iron.

Nitrate, NO3-. All nitrate compounds are readily soluble. Therefore, S/S treatment of
nitrate in waste materials is often futile.

Phosphate, PO4-3. Soluble phosphates form insoluble compounds with many


cations, including calcium, iron, and lead.

Sulfide, S-2. Soluble sulfides form insoluble compounds with most metal cations,
particularly iron. This property makes sulfides a valuable reagent for the S/S
treatment of metals, particularly as a sulphide-based system is a reducing system
and can be used to manage hexavalent Cr, by reducing it to the trivalent species.

3.3 Organics
Traditionally, wastes contaminated with non-volatile and semi-volatile organics such
as dioxins, explosives, lube oil range (>C28) and asphaltic petroleum hydrocarbons,
pesticides, polyaromatic hydrocarbons (PAHs) polychlorinated biphenyls (PCBs),
were treated via on-site thermal treatment technologies. Problems with the public
and regulatory acceptance of thermal treatment technologies have limited the
application of these technologies on such wastes.

76
S/S has become an acceptable low cost treatment alternative to reduce the mobility,
and thereby manage the risk by minimising exposure pathways, of these recalcitrant
organics. S/S is typically not applied to volatile organics.

Figure 3.3 shows wastewater sludge being excavated at a former automobile plant.
The sludge contained an organo-nickel complex from plating operations at the plant.
The sludge was chemically stabilized for disposal in an on-site RCRA landfill.

Figure 3.4 shows a S/S treatment system set up at an acid sludge tar pond.

3.3.1 Chemistry and physics of organic stabilization


The S/S of organic-contaminated waste may involve the alteration/transformation of
the organic compounds themselves, or their participation in physical processes, such
as adsorption and encapsulation. The S/S of organic compounds has recently been
reviewed (Hills et al., 2010).

The outcome of S/S is to retard the movement of the hazardous constituents within
prescribed safe limits, defined by leachate quality. However, the mechanisms of
organic stabilization involved are very difficult to distinguish apart, and this tends to
reflect the fact that S/S of organic compounds rarely involves the formation of solid
organic salts.

The immobilisation assumed to be due to adsorption might involve some sort of


chemical bonding or change in the speciation of a compound, or the formation of
degradation products. The pervasive chemical environment within S/S treated
hazardous waste is highly alkaline with a high number of soluble components, and
may promote both short and longer-term changes in the nature of certain organic
contaminant species.

In practice, however, inorganic stabilization systems operating at ambient


temperatures and pressures in non-exotic aqueous environments can induce
stabilization through hydrolysis, oxidation, reduction and compound formation.

Hydrolysis refers to the reaction of a compound with water. This usually results in
the exchange of a hydroxyl group (-OH) for another functional group at the reaction
centre.

77
Figure 3.3: A contaminated wastewater sludge being excavated

Figure 3.4: Treatment facility to chemically S/S acid sludge tar

78
Hydrolysis may be catalysed by acids or bases (e.g. H +, OH-, or H3O+) and many
involve intermediate species (Dragun, 1988). Metal ions such as copper and calcium
may act as catalysts for certain organic structures, and adsorption on surfaces such
as clay and activated carbon may also facilitate reactions involving organic
molecules.

Many organic compounds are resistant to hydrolysis, including some esters, many
amides, all nitriles, some carbamates, and alkyl halides. Those less resistant to
hydrolysis include alkyl and benzyl halides, poly-methanes, substituted epoxides,
aliphatic acid esters, chlorinated acetamides, and some phosphoric acid compounds.

Oxidation and hydrolysis are the most common pathways for the reaction of
organics in stabilization systems. Oxidation of organics occurs via two pathways
(Dragun and Heller, 1985). In one, an electrophilic agent attacks an organic
molecule and removes an electron pair; in the other, only one electron is removed,
forming a free radical. The former is heterolytic, the latter homophilic.

Free radial formation reactions have lower energy barriers than the oxidation of a
polar compound or cleavage of a covalent bond. It is worthy of note that organic
oxidation reactions in the chemical industries are typically catalysed by crystalline
aluminosilicates at elevated temperatures and pressures. It has been recognised
that this also occurs at ambient temperatures with clay and soils, not only in
oxidation, but in reduction, hydrolysis, and neutralisation reactions (Dragun, 1988).
Iron, aluminium and trace metals within layered silicate minerals have been identified
as specific catalysts (Garrido-Ramirez et al., 2010), though not all clays exhibit this
property.

Based on work by Dragun and Heller (1985), two generalities can be made
concerning the oxidation of organics by soils and clay minerals:

Many substituted aromatics undergo free-radical oxidation, e.g. benzene,


benzidine, ethyl benzene, naphthalene, phenol
Chlorinated aromatics and polynuclear organics are unlikely to be oxidised

Water content may be one of the more important constituents of soils, and possibly
also in wastes. Partially saturated systems are more likely to undergo oxidation
reactions than saturated ones. The above comments apply to "natural" oxidation by
reagents normally used in stabilization systems, or characteristics of the waste itself.
The deliberate addition of strong oxidisers such as potassium permanganate or
79
hydrogen peroxide is a different matter. These reagents have been used in
stabilization for the oxidation of phenols and other organics, as well as inorganics
and treatment of metal complexes.

An important consideration in all organic reaction schemes involving hazardous


constituents is the product of the reaction, which may be equally or more hazardous
than the original reactant. This is especially true when oxidation processes are
employed, as their use in stabilization systems, as in addition they may also create
hazardous species from other organics in the waste which were previously non-
hazardous or less hazardous.

The use of strong oxidants on wastes, which contain chromium, might result in the
formation of Cr6+, necessitating a subsequent reduction step to Cr 3+.

Reduction of organic compounds may be defined as either:

An increase in hydrogen content or a decrease in oxygen content, or


A net gain in electrons

Reduction can occur in clay systems with the clay acting as a reducing agent.
Reductive alteration of organic compounds in waste has been poorly studied and is
not well understood.

Compound formation between organic compounds and metals or other cationic


species can form less soluble and/or less toxic compounds. For example, the
solubility of oxalic acid is 95,000 mg/L, compared with 6 mg/L for the calcium salt.
Organic acids of environmental concern might be effectively immobilised in the
calcium-rich environment of commercial stabilization systems. In addition to the
formation of salts, a number of other direct reactions are possible between organic
contaminants and organic or inorganic reagents under ambient conditions.

Boyd and Mortland (1987) developed the concept of a two-step reaction scheme
where the organic molecule is first adsorbed on a clay surface, then reacted via
surface catalysis. They claimed to have achieved polymerisation and de-chlorination
reactions in this manner, including detoxification of dioxin analogues. Ortego (1989)
discussed the formation of stable organo-metallic complexes, such as those formed
between Cu2+ and benzene, in the interlayer of laminar-structured clays. Zinc, Cd,
and Hg co-ordination compounds with polyamines have been also reported, as has

80
the decomposition of acid-sensitive organics by acid adsorbed water at the clay
surfaces (Kostecki, 1992).

As a subset of compound formation, several investigators have postulated the


formation of linkages between organic compounds and insoluble substrates such as
stabilization binders like cement (e.g. Montgomery et al., 1983). It seems likely that
the instances of strong "sorption" at these surfaces may really be the result of
exchange reactions, similar to those known to occur during the production of organo-
clays from clay minerals by treatment with organic cations of the form [(CH 3)NR]+.

3.3.2 Physical aspects of organic stabilization


Work on "physical" immobilisation of organics has been primarily focused around
several materials and mechanisms. Cote (1987) has shown that a variety of organics
can be sorbed fairly effectively by cement-based stabilization processes,
incorporating activated carbon and bentonite additives, whereas fly ash and soluble
silicates are less effective. Other conclusions from this work were:

Volatile organics were not well immobilised


Water soluble contaminants were not well immobilised
Organics with low water solubility were well immobilised

Kyle et al. (1987) compared a number of lime, kiln-dust and fly ash-based mixtures
with organic reagent (vinyl ester, acrylic, epoxy, polymer cement) on several
industrial wastes spiked with various priority pollutant organics. They found that the
organic reagents produced poorer results, as measured by total organic carbon
(TOC) in the leachate, than did the inorganic reagents. The addition of activated
carbon to lime/fly ash systems lowered the TOC. Work by Lear and Conner (1992)
found a similar reduction in the leachability of organics when activated carbon and
other adsorbents were added to Portland cement systems. The efficacy of oxidisers
was also noted in that study.

Christenson and Wakamiya (1987) found that Kepone leaching was increased in
highly alkaline systems such as cement/soluble silicate, but decreased by
encapsulation in either an organic polymer or a proprietary molten-sulfur blended
binder. Co-precipitation in ferric hydroxide precipitation systems was found to
remove chlorendic acid, humic acid, PCBHs and other compounds from landfill
leachate (Pojasek, 1980).

81
3.4 Geotechnical considerations
The suitability of soil for S/S is somewhat different than for typical geotechnical
projects. There is little choice but to utilise the soil type that is contaminated at a
given site. Given this inevitability, the type of soil presented for S/S and its spatial
variability will have an influence on the choice of equipment used and the selection
of the binder system/additives used for treatment.

It is therefore essential that site conditions are well understood to enable treatment
to be applied in a thorough, timely and cost-effective manner. A proper
geotechnical/hydrogeological site investigation will limit risks to project owners and
consultants. Following is an introduction to soil classification, site characterisation,
and site variability as it applies to suitable applications of S/S technologies.

3.4.1 Soil classification


The classification of soil can be a relatively easy procedure for experienced
geotechnical lab and field personnel. However, a site investigation for an S/S project
that uses incorrect or inconsistent soil classifications creates confusion for third party
consultants, owner consultants, regulators and contractors and could potentially lead
to litigation. There are many different forms of proper soil classification depending on
the country where the work is being performed.

None of these classification systems is necessarily more correct than the others.
However, what is important from a soil classification standpoint is that the site
investigation report is systematic and clear and uses a recognised classification
system. Depending on the location of the work, this decision will be governed by the
state of practice in the geographical region. Figures 3.5, 3.6, and 3.7 show the
classification system provided in BS 5930 (code of practice for site investigations). A
Unified Soil Classification System (USCS) can also be found in ASTM D 2487.

Most soil classification systems provide an indication of particle size distribution,


colour, organic content and the physical behaviour of the soil in the classification.
These soil classifications found in borehole and test pit logs are often performed on a
visual basis and laboratory basis using tests such as particle size and Atterberg
Limits.

Although following the proper soil classification protocol will assist in ensuring clarity
in S/S planning and design, site investigations incorporating soil classifications
should also include indications of compactness/consistency, colour (e.g. Munsell
system; Goddard, 1979), and the presence of materials such as debris, organic-

82
based plant matter (i.e. peat), free-phase organic contaminants, etc. Also, man-
made fill material should be distinguished from naturally occurring soil deposits.

Where debris is present, efforts should be made to provide as much description as


possible as related to the nature of the debris (brick, concrete, steel rebar etc.) as
well as the size of this debris, if known.

Particle size information will be required to address requirements of pre-processing


of the soil prior to S/S activities. Depending on the size of the debris, the amount of
cobbles or boulders present, or the density/consistency of the material on a potential
S/S site, the type of construction method used to provide S/S treatment may vary.
For example, it may be necessary to screen out oversize material prior to processing
through a pug-mill, excavate debris prior to treating soil with an in-situ auger, etc.

In some cases conventional ex-situ and in-situ S/S approaches may not be practical,
given the presence of a large amount of boulders. In this case a less conventional
approach, such as in-situ jet grouting, may be selected. As a worst-case scenario,
portions of the site may be non-treatable with soil stabilization if excessive debris or
large quantities of cobbles and boulders are present.

83
Figure 3.5: General identification and description of soils (modified from British Standard, BS 5930)
84
Figure 3.6: Identification and description of soils (modified from British Standard, BS 5930)
85
Figure 3.7: Identification and description of soils (modified from British Standard, BS 5930)

86
3.4.2 Site Characterization of Soil and Groundwater

If possible, previous knowledge of the history of the site development (e.g. past land
use, original topography prior to site development) should be used during the desk
study phase to inform and develop a conceptual model of the soil, rock and
groundwater conditions (see Section 7.2). Often this understanding of site
conditions needs to extend to neighbouring properties, especially if the site
boundaries are close.

The nature of soil, rock and groundwater conditions must be delineated vertically and
laterally with test pits and/or boreholes, in the contaminated area and outside of its
borders. For large projects, test pits and boreholes are usually both performed to
take advantage of both types of investigations and reveal the most information about
the site. The type of investigation (i.e. borehole versus test pit) will be governed by
many different factors but below is a list of some of the considerations in selecting
the type of investigation for potential S/S sites.

87
Table 3.1: Considerations for choosing test pits versus boreholes
Test Pits Boreholes

The presence of debris can be seen Enables sampling at greater depth


Bulk samples for testing of binder compared to test pits
formulation are available Allows in-situ testing (SPT, CPT) to
Excavation-related issues can be obtain quantitative data
Advantages

quickly identified Allows installation of monitoring wells


Construction is relatively quick and and results from pump tests, k-tests
inexpensive to perform over a wide Not as intrusive or disruptive as test
range of geographical/geological pits
environments) Appropriate for a variety of
groundwater conditions

Limited by excavator reach (15-30 ft; Sophisticated interpretation of ground


5-30 m), may be shallower for dense conditions needed to inform site
layers at depth conceptual model
Can cause significant site More expensive, especially for remote
disturbance, especially for deep test sites
pits Small sample sizes may not allow
Large excavations may require binder-formulation trials
remedial actions prior to development Identification of extent of buried debris
Difficult to obtain useful information and oversize material not possible
Disadvantages

for non-cohesive soils below the


water table
In-situ testing (e.g. Standard
Penetration Testing (SPT), Cone
Penetration Testing (CPT)) is not
performed due to sample disturbance
(i.e. limitations are placed on the
amount of quantitative information
available)
Does not provide information for
seismic design
Monitoring wells cannot be properly
installed where a test pit is
constructed

As for ground engineering projects, prior to remediation, a site investigation (SI) is


required to examine the lateral and vertical properties of soil, rock and groundwater
across a site. The SI is also used to obtain samples to enable the nature and extent
of contamination to be characterised, however there is normally a trade-off between
88
the amount of material collected during the SI and the cost. If ground conditions
prove difficult, or the extent and nature of contamination is unforeseen, then more
detailed and focussed investigative work is carried out at a later stage.

Thus, these studies are performed early in the project-development process and are
used to inform the decision-making process leading to the choice of remedial design.
The SI supplies fundamental information on groundwater flow regimes, the nature of
contamination (distribution, type, level) and the host soil/rock characteristics.
Information obtained from the SI is also used in the design of any dewatering
scheme, the choice of complementary treatment technologies, including as part of a
treatment train, as well as preliminary selection of S/S equipment that will be most
suitable for use at a particular remedial site.

However, generally if S/S is selected as the method to manage contamination at a


site, additional investigations may have to be performed to increase the amount (and
reliability) of data available on:

The variability of contaminants present


Soil and rock conditions
The geochemical properties of the soil and groundwater, as pertains to S/S
treatment (i.e. pH, sulphate, etc.)

Additional factors such as the presence of free-phase organics may need to be


established to avoid problems with respect to their interaction with available
binder/additives. The appointed remediation contractor may often undertake these
additional investigations.

For smaller projects, the costs associated with more detailed phases of investigation
may be prohibitive. However, with larger projects, the acquisition of key data that can
facilitate a higher degree of certainty in the remedial design reduces the residual
risks to both owners and contractors. With larger projects, a phased programme of
ground investigation is often undertaken to better understand soil, rock and
groundwater conditions, allowing the sequential development of a site conceptual
model, enabling refining both the model and remedial options to be employed at the
contaminated site.

3.4.3 Approaches to site variability


Ground conditions at any site will be variable; the spatial variation of contaminants is
no exception. The approach taken will assist in reducing the risk-potential to
89
acceptable levels. As previously mentioned (Sections 2.4 & 2.5) it is impossible to
eliminate risk entirely.

In increasing the intensity of data acquisition from the site investigation (i.e.
increasing the number of test pits, boreholes etc.) it will be possible to reduce the
risks associated with ground variability to an acceptable level. However, given that
ground investigative techniques are expensive, as the intensity of the site
investigation increases so too will the total cost of the investigation. It is a judgment-
based decision (by the regulator, consultant or site owner) as to what constitutes an
adequate amount of intensity, and the level of risk that is acceptable.

For large projects however, the increased cost of the SI can save money in the
construction phase of the project. For smaller projects, the added cost of an
enhanced SI may outweigh the cost-benefits associated with better defining site
variability. This paradox means that contractors enter a bidding process with more
uncertainty that can ultimately lead to cautious (i.e. expensive) bids for the project.

For S/S projects, the level of variability that is present on a site can influence the
type of equipment selected for the remedial project and often results in a binder
formulation that is conservative. By way of example, if ground and other site
conditions show that 25% of a site can be treated with 8% cement addition, and the
remainder with 10% cement addition, it is likely that a contractor will apply 10% to
the whole site.

Industrial sites that present predominately man made fill or made-ground have
potentially large variability in soil properties over small distances and hence
conservative binder formulations are chosen. This conservatism may inevitably add
additional costs to the project, but allows a consistent approach to the whole site,
limiting potential mistakes and reducing the chance of failing performance
specifications, which can ultimately delay the project, diminish contractor confidence,
and escalate the final cost of the remedial operation.

3.5 Ex-situ and in-situ application of S/S


There are two major methods of applying S/S treatment to contaminated soil: ex-situ
and in-situ.

The ex-situ method of applying S/S involves the excavation of contaminated media
(soil, sludge, sediment, etc.) and its transfer to another location, either on- or off-site
90
for subsequent mixing with reagents/binders. In-situ treatment by S/S involves
mixing binders/reagents into the contaminated media in the place where it is found
and thus does not involve excavation and transport to another location.

Although the definitions given above are quite distinct and are used in this document,
the reader should be aware that there are a number of other definitions of a
regulatory nature, and these may be different.

3.5.1 Technical considerations


The major technical advantage to ex-situ S/S treatment is that the properties of the
media (e.g. moisture content, texture, contamination level, degree of
aggregation/agglomeration, amount and type of debris) are directly observable in the
media to be treated, and this facilitates the production of a quality controlled
product.

If the properties of the contaminated media deviate from those used in the design
criteria the process can be quickly adjusted to compensate, and to maintain
treatment goals. This is particularly important where, for example, drum carcasses
are encountered where not expected, or there are localised and discontinuous (but
significant) hot-spots of contamination, not foreseen in the original SI.

Ex-situ S/S has the advantage that both liquid and solid reagents can be used to
treat the contaminated media, using commonly available mixing equipment.
Furthermore, water, which is often critical for thorough mixing, can be easily added
where necessary to maintain mix properties.

An additional advantage is that samples of the treated material are also easily
obtained during processing, such as at the mixer discharge point (see Figure 3.8),
enabling rapid evaluation of S/S product as it is being formed.

Furthermore, the application of ex-situ S/S involves more commonly available


equipment, and there are more vendors available that can apply this method of S/S.
In addition, it is possible to identify more quickly any inadequate treatment and to re-
route this material back into the system to ensure the maintenance of treatment
goals, or disposal by another method.

91
Ex-situ treatment also has the advantage of being suitable for shallow soils, located
above the water table, and this facilitates selective materials removal, and a lower-
cost treatment option.

The major technical disadvantages for ex-situ S/S include primarily space
requirements and ease of materials handling. However, there must be space
available to stockpile materials to be treated and to apply S/S and to hold the treated
product until verification of performance criteria is received. It should be noted that
ex-situ S/S involves significant logistical issues, related to the transport of the
contaminated media to the site of S/S treatment and then onto the location of final
disposition (which may be off- or on-site). Figures 3.9 and 3.10 are examples of
typical ex-situ S/S treatment-systems.

If the final disposition of the treated material is on-site, the placement and
compaction of the treated material becomes another requirement for ex-situ S/S
treatment. Another technical disadvantage of the ex-situ S/S is the increased design
required for deep (greater than 15 feet or 4.5 meter) contamination. Excavations this
deep often involve sheeting and shoring, increasing the complexity of the excavation.
Also, excavation below the water table will require management of the groundwater
that may infiltrate the excavation.

The major technical advantage of in-situ S/S treatment is that the contaminated
media is treated in place (Figure 3.11). There is no need for a separate on- or off-
site treatment area and no need to consider the logistics of transport prior to
treatment, and transport only becomes a concern if the final disposition of the treated
material is off-site.

Other major advantages, depending on the site, are that treatment below the water
table is feasible without dewatering and emissions from in-situ treatment are far less
than from ex-situ excavation and treatment.

92
Figure 3.8: Ex-situ treatment in a roll-off box

The major technical disadvantage for in-situ S/S is that properties of the media must
be inferred from or interpreted from how the in-situ S/S equipment reacts during
treatment. Skilled operators are often required to effectively operate the equipment.
In-situ S/S treatment often requires the treated material to develop enough
compressive strength to support the equipment as treatment progresses. Figure 3.12
shows in-situ S/S equipment operating on previously treated material.

On some sites, it may be possible to sequence the S/S work so that the in-situ S/S
equipment never has to operate on top of treated material. However, many sites will
require the in-situ S/S equipment to operate on top of treated material.

In-situ S/S ancillary and equipment involves track-mounted cranes or carriers (see
Figure 3.13 and Section 5), which can require treated material to attain compressive
strengths up to 50 psi (0.35MPa) before the equipment can safely operate on it.

Many of the in-situ S/S rotary mixers and augers require that S/S reagents be liquid
or slurry, as these equipment types use the reagent slurry as drilling fluid to ease the
movement of the equipment into the subsurface and to mix the reagents into the
media.

Figure 3.13 shows a batch plant for slurried S/S reagents. Reagents that are difficult
to slurry, such as zero-valent iron, can pose a problem for in-situ S/S. Specialised

93
sampling devices are required to collect samples of the in-situ S/S treated material
and the sampling is often complicated by the viscosity of the treated material.

Figure 3.9: Typical ex-situ S/S equipment, including pug-mill, silos, reagent
tank, water tank, stacker and haulage

Figure 3.10: Ex-situ S/S treatment enclosure

94
3.5.2 Physical material handling
One of the advantages of in-situ S/S treatment is the limited material handling
required to prepare the media for treatment. The presence of debris
(foundations/footers, drum carcasses, piping, etc.) in the subsurface can be
detrimental to the in-situ S/S equipment (see Section 5), causing damage to the
augers and/or penetration refusal at depths less than the desired S/S treatment
depth. Figure 3.14 shows a 3 ft (0.92m) limestone boulder excavated in order to
facilitate in-situ treatment.

Figure 3.11: In-situ S/S treatment of oily sludge

95
Figure 3.12: In-situ mixing equipment operating on top of S/S treated material

Figure 3.13: Typical in-situ S/S equipment: batch plant with reagent silos and
mix tanks; crane, Kelly bar, and auger

96
Figure 3.14: A 3 ft limestone boulder excavated to facilitate in-situ treatment

Dense, compacted lithology, such as layers of boulders, compacted gravel, or glacial


till, may also hinder penetration rate and/or cause penetration refusal at depths less
than the desired treatment depth.

If the site information includes boring logs with n-values from standard penetration
testing, n-values of greater than 50 for a 6 depth (15cm) may be indicative of a
dense compacted lithology. If these impediments are known or considered likely to
occur at a site, material handling in the form of pre-excavation of the media is often
employed to remove or loosen the impediments. If the lithology and depth requires
this pre-excavation under a bentonite-slurry, the added bentonite often improves the
chemical and physical properties of the final treated material.

Ex-situ S/S mixers (see Section 4) require debris such as rock and brick to be
removed or separated from the media and the media size-reduced if necessary
before it can be introduced into the mixer. This handling is necessary to prevent
pinching of the media between the mixing paddle, auger or screw and the side of the
mixer vessel. This not only can damage the paddle or screw, but also impacts the
gear drive and motor which turn the paddle or screw. This is an easily avoidable
issue for ex-situ S/S treatment.

97
The major types of material handling equipment commonly utilised for the
preparation and/or separation of media for ex-situ mixers are: screening equipment,
crushing equipment, shredding equipment, and magnetic separation equipment (see
Section 4).

Mixing energy
The mixing energy imparted to the media during S/S treatment varies depending on
S/S equipment, whether in-situ or ex-situ. Typically, the ex-situ S/S mixers, such as a
pug-mill, exert the highest mixing energy on the materials to be treated.
Hydraulically driven in-situ S/S rotary mixers, many of which are mounted on
excavators, are close to ex-situ S/S mixers in mixing energy, but impart less mixing
energy to the media as it is treated. The in-situ S/S auger mixers, driven by large
hydraulic turntables or gearboxes, are moderate in their mixing energy. Excavators,
whether used for in-situ or ex-situ mixing, impart the lowest mixing energy to the
media during treatment.

Mixing energy is an important factor in the production of a consistent, reproducible,


homogenous treated material. But it is not mixing energy alone, but the combination
of mixing energy, mixing speed, and mixing time that is required to produce a
homogeneous treated material. This combination can be adjusted or tailored for the
specific media being treated, regardless of the specific mixing equipment selected
for S/S treatment, to produce a consistent, reproducible, homogenous treated
material. Therefore no type of S/S mixing equipment, whether in-situ or ex-situ, can
be deemed superior to the others on the basis of mixing energy.

Site considerations
Treatment depths of less than 15 feet (4.5 meters) are often an advantage for ex-
situ S/S over in-situ, as these shallower treatment depths are typically amenable to
excavation without the need for sheeting or shoring. Deeper treatment depths are
advantageous for in-situ S/S treatment.

The presence of the water table within the treatment depth is advantageous to in-situ
S/S. Not only does one avoid the need to remove and manage the water, but the
water reduces the amount of reagent grout needed to act as a drill fluid while it
increases the penetration rate. The excess moisture in the media complicates the
excavation of the media, and its material handling, for ex-situ S/S.

98
3.5.3 Economics
Mobilisation/demobilisation
The in-situ rotary mixers and excavators are also less expensive to mobilise and set
up. The mobilisation typically involves just excavator, mixing heads, and a reagent
batch plant. Only the batch mixing plant requires set-up, which can typically be done
in less than a day.

Though more complicated, the ex-situ mixing and material handling equipment is
typically only slightly more expensive to mobilise and set-up. Most of the ex-situ S/S
equipment is transportable and can be off-loaded and set up for operation in less
than a day.

The in-situ S/S auger is the most expensive to mobilise and set-up. The cranes,
though transportable when disassembled, require re-assembly on-site. The
gearboxes or turntables then need to be attached, followed by the auger. Typically
this takes three to four days to complete.

Equipment
The daily equipment costs for the ex-situ mixing equipment are typically on the order
of $1000/day and are the least expensive when compared to the in-situ mixing
equipment. The daily costs for the excavator-mounted in-situ S/S equipment is only
slightly more expensive. However, the crane-mounted in-situ S/S equipment is
significantly more expensive, on the order of $3,000/day.

The daily cost for most of the ex-situ material handling equipment is typically less
than $1,000/day. Ex-situ S/S also requires excavators or loaders to feed the media
to the mixing or material handling equipment and then remove and stockpile the
treated material. In-situ S/S also requires an excavator to handle swell from the S/S
treatment.

Overall, the total daily equipment costs for ex-situ S/S are typically less than those of
in-situ S/S. However, daily equipment cost for certain in-situ S/S configurations (e.g.
excavator-mounted mixers with the ability to accept dry reagent addition) can be
comparable to those for ex-situ S/S. In-situ S/S using crane-mounted mixers and
batch plants have the highest daily equipment costs.

99
Labour requirements
Most in-situ mixing systems with dry reagent addition require two workers to manage
the reagent storage and delivery and operate an excavator-mounted in-situ mixer.
Ex-situ S/S systems typically require three to four workers to feed the system,
operate the mixer (and material handling equipment, if necessary), manage the
reagent storage and delivery, and stage the treated material. In-situ S/S involving
crane mounted equipment mixers can require five to six workers to handle the
reagent storage and grout preparation, operate the mixer, and manage swell.

Reagent storage and delivery


Silos for reagent storage typically cost on the order of $500/day, whether used to
supply ex-situ mixers or in-situ mixing equipment set up for dry reagent addition or
batch plants. Batch plants typically run about $2,000/day, including the cost for all of
the tanks, mixers, pumps, and hoses needed to prepare and deliver the reagent
grout to the ex-situ or in-situ mixer.

Dry reagent addition, whether for ex-situ or in-situ S/S, typically requires only a
reagent silo to store the reagent on-site and deliver that reagent to the ex-situ or in-
situ mixing equipment. This is the least expensive reagent storage and delivery
option. Reagent delivery in the form of a reagent grout typically requires silos to store
the dry reagents and batch plants (Figure 3.13) to prepare the reagent grout, making
this a more expensive option for reagent storage and delivery.

For the in-situ mixing equipment that can receive dry reagent feed, the reagent is
delivered pneumatically, which makes it difficult to consistently and accurately deliver
the reagents.

For the in-situ mixing equipment, reagent grout addition, though more expensive,
results in more accurate, consistent, and reliable reagent delivery.

Production
Ex-situ S/S systems typically treat 350-1,000 yd3 (270-760 m3) of media per day, per
mixer. Site-specific production rates vary based on media properties (e.g. moisture
content, texture, plasticity), reagent addition level (i.e. productivity decreases with
increasing reagent addition), and site considerations (e.g. space available for
equipment or to stockpile untreated and treated materials).

100
In-situ S/S systems typically treat 400-1,200 yd3 (300-920 m3) per day, per mixer.
Site-specific production rates vary based on media properties (e.g. moisture content,
texture, plasticity), reagent addition level (i.e. productivity decreases with increasing
reagent addition), and site considerations (e.g. depth of treatment, depth to water
table).

Overall costs
Typically the cost per cubic yard for ex-situ S/S processing, excluding reagent costs,
ranges from $15 to $30/yd3. This cost does not include the cost to excavate,
transport, and stockpile the untreated media or the cost to stockpile, transport, and
place the treated material.

For in-situ S/S, the processing costs, excluding reagent costs, can range from
around $20/yd3 (0.8m3) for using excavator-mounted equipment with dry reagent
addition to $50/yd3 (0.8m3) for crane-mounted equipment with reagent grout addition.

In general, the costs for ex-situ S/S and in-situ S/S processing cover a similar range.
However, site-specific factors (media properties, depth of treatment, depth of water
table, and site space considerations) may give one method an advantage over the
other.

101
4.0 Ex-situ S/S equipment and application

The nature of a soil or waste and the reagents to be mixed with it is an important
factor on the choice of mixing equipment to be used on site. Site characteristics such
as its layout, the space available, size and geographical location, are also important.
There are various types of ex-situ mixing equipment including pug-mills, excavators,
screw mixers, ribbon blenders, tillers, asphalt mixers, and paddle aerators which are
discussed below. In addition, design and application of mixing pits is also presented.
Figure 4.1 outlines the processes involved in ex-situ application.

102
Figure 4.1: Ex-situ processes

103
4.1 Mixing Chambers
Mixing chambers are designed to mix, or blend different materials to produce a
homogenous product. The body or casing of the mixer contains or restricts the
movement of the materials being combined during the blending process. Within the
mixing body, rotating paddles, screws, ploughs, or ribbons lift, throw, fold, or knead
the material. The mixing action and energy input (into the forming mixture) from the
rotating paddles, screws, ploughs, or ribbons influence:

What materials can be treated/mixed


How long it takes to mix the materials being presented
The thoroughness of the mixing action (i.e. how homogeneous is the product)
The cost of the mixing process

The arrangement of the mixing chamber can be horizontal or vertical. Typically,


horizontal mixing chambers are used for continuous mixing, while vertical mixing
chambers are used for batch mixing. Continuous mixing chambers receive the
material to be mixed at one end and discharge the mixed product at the other.

Batch mixing chambers have their feed materials introduced before mixing begins,
and when mixing is completed the product is then removed (from the mixing
chamber).

Continuous mixing allows for high materials throughput. Mixing time is set by the rate
at which material is fed into the mixer, its volume and mixing action. Batch mixing
has the advantage that the mixing time can be controlled, but processing capacity is
related to the size of the mixing body and its cycle time, including loading of
materials and unloading of product.

During batch mixing, the final product is usually discharged (by dumping) requiring a
homogeneous product that is free flowing or fluid enough to flow out of the mixer by
gravity. Many continuous mixers use conveyors to move the discharged product,
which must be viscous enough to be conveyed. Thus, the rheology of the freshly
mixed S/S product is very important, and needs controlling for effective mixing and
conveyance to the point of use.

Mixers can receive both liquids and solids. The solids can be dry powders, thin
slurries, or pastes. However, the types of solids to be processed will need to be
mixed in such a way as to produce a homogeneous product with a minimum of
processing time. As an example, ribbon blenders and screw mixers cannot be used
104
effectively with cohesive soil or waste and are not appropriate for blending these
materials with binders.

There are environmental considerations for mixing chambers. Since materials are
conveyed into and out of mixers, secondary containment is required to collect and
contain any spillage, especially when one of the materials being introduced is a
waste material; this is typically achieved by using a bermed or lined mixing platform.
Often the materials to be mixed are dusty in nature and require effective dust control
measures, such as a water-mist/spray within the mixer body.

However, a shroud with dust control measures may be required under some
circumstances. This may include dust filtering or a bag-house, to effectively control
any dust emissions. This is especially important when operations are within a
populated urban area and where sensitive receptors are within close proximity (tens
of meters) of the site. Similarly odour control during mixing can often be addressed
by supplying a mist or spray, or an odour suppressant polymer or foam within the
mixer body. However, mixer shrouding and use of an activated carbon filter or a
reactive scrubber may be suitable mitigating measures when sensitive receptors
may be close to the site of operation.

4.1.1 Pug-mill mixers


Pug-mill mixers are horizontal mixers that can be used for the continuous mixing of
dry free-flowing powders, thin slurries, and thick pastes. A typical pug-mill consists of
a horizontal box-like or trough mixing chamber with a top inlet on one end and a
bottom discharge at the other end, 2 shafts with opposing mixing paddles, and a
drive assembly to rotate the shafts.

A continuous pug-mill is an effective mixing platform that can achieve a thoroughly


mixed, homogeneous product within a few seconds. The action of the pitched
paddles moves the material from the bottom of the trough, up each side, and forces
the material back down between the shafts. This both kneads and folds the S/S
mixture. Mixing efficiency and product residence time are influenced by the size of
the paddles, paddle swing arc, overlap of left and right swing arc and the dimensions
(and volume) of the mixing chamber, and the nature of the material being mixed.

Pug-mills are often supplied as part of a pug-mill mixing system including a feed
hopper and feed conveyor to ensure a consistent supply of materials. The feed
hopper can employ either belts or augers to control the transfer rate of the waste
material onto the transfer conveyor, which often includes a belt-scale to allow the
operator to accurately determine feed rate and quantity of material to be treated.
105
Since the reagents are typically added on a weight-for weight basis, the waste feed
rate from the belt-scale can be used to control the reagent addition rate.

A discharge conveyor to stack the treated product exiting the pug-mill can be
included, but this is often a separate piece of equipment. Similarly, the reagent silos
and feed-system can also be part ancillary equipment supplied. Figures 4.2 and 4.3
show a typical pug-mill system with support equipment. Figure 4.4 shows the trough
of a pug-mill and its paddle design. The advantages and disadvantages of pug-mill
mixers are listed in Table 4.1.

A list of pug-mill suppliers by country is shown in Table 4.8.

Figure 4.2: A pug-mill with its ancillary equipment

106
Figure 4.3: Ancillary equipment of a pug-mill

Table 4.1: Advantages and disadvantages of pug-mill mixers


Advantages Disadvantages

Pug-mills are high capacity (200 Throughput is dependent on the


tonnes of materials per hour) characteristics of the waste media
High shear/mixing-energy mixers. The (moisture content, plasticity, particle
twin shafts in a pug-mill can rotate at size distribution)
up to 500rpm The waste/soil to be treated must be
Very homogeneous and well-mixed screened to remove debris greater
S/S product than 2in (50mm). This may require
Pug-mills are very tolerant of debris additional handling/processing steps,
and will typically tolerate small stones including the drying of waste materials
and pieces of bricks, concrete, wood, with moisture contents above their
and metal debris plastic limit, as is needed to allow
Can effectively process soil or waste of effective screening
a widely variable consistency, handling For waste volumes less than 2,000 m
3

effectively viscous fluids, free-flowing costs can significantly impact the


powders, and plastic and/or sticky economics of on-site treatment
solids
Transportable and often mounted on a
wheeled frame for road transportation
Can be ready for operation in less than
half a day
Are relatively inexpensive to mobilise,
set-up, operate and demobilise

107
Figure 4.4: Paddles inside a typical pug-mill

4.1.2 Screw Mixers


Screw mixers are horizontal or near-vertical mixers for the continuous mixing of dry
free-flowing powders and solids, with moisture contents below their plastic limit. A
screw mixer consists of a circular-shaped mixing body (usually at least ten times
longer than it is wide), a screw auger mounted on a shaft, and a drive assembly to
rotate the shaft.

A screw mixer is usually loaded at one end (typically the lower end when used in a
near vertical configuration). Mixed material is typically discharged from the other end.
The rotation of the screw lifts and moves the material to be blended, creating a
folding effect as it moves up the auger flight. The screw is set at a fairly close
clearance to the mixer body so no material remains along the mixer body. The pitch,
rotational speed, and length of the screw can be used to set the mixing time.

Screw mixers cannot be used for sticky or plastic solids, as they tend to ball up and
bind in the mixer. Figure 4.5 shows a screw mixer with its lid removed. The
advantages and disadvantages of screw mixers are listed in Table 4.2.

A list of screw mixer suppliers by country is shown in Table 4.8.

108
Table 4.2: Advantages and disadvantages of screw mixers
Advantages Disadvantages

The high rotational speeds thoroughly Throughput is dependent on the waste


blend waste media with reagents materials characteristics (moisture
Screw mixers can process in excess of content, plasticity, particle size
100 tonnes per hour distribution) and screw design (which
Screw mixers are simple to operate controls the residence time in the
and have few moving parts to maintain mixer)
The waste material must be screened
to remove debris larger than 1 inch, as
screw mixers are not tolerant of hard
debris, e.g. concrete, bricks, metal
Other handling/processing steps may
be required to prepare the waste
media, including, for example, drying of
waste materials with moisture contents
above their plastic limit (to allow
screening and subsequent processing)
Are not capable of handling waste
media that are above their plastic limit
or are sticky or tacky in nature
Generally not suited for mixing liquids
Screw mixers are typically designed for
fixed facilities

109
Figure 4.5: Screw mixer

4.1.3 Ribbon Blenders


Ribbon mixers are also horizontal mixers for the continuous mixing of dry free-
flowing powders, thin slurries, and solids with moisture content below their plastic
limit.

A typical ribbon blender has a trough-shaped mixing body (usually two to three times
longer than it is wide) with a semi-circular bottom, a horizontal longitudinal shaft
upon which are mounted arms supporting a combination of ribbon blades and
paddles. A drive assembly is used to rotate the shaft.

Loading a ribbon mixer with materials is usually done at (the top of) one end while
discharging is done at the bottom of the other end. A long and complex single paddle
is mounted axially on the shaft and is used to disperse the material to be mixed to
the outer ribbon. Compared to a pug-mill, the paddles on a ribbon blender move
relatively slowly, feeding the material to be blended into the ribbon.

110
The most effective ribbon design is the double spiral in which the outer ribbon moves
the product in one direction and the inner ribbon moves it in the opposite direction.
This opposed movement creates axial flow of the material though the mixer and
prevents build-up of materials in the mixer. The pitch of the ribbons are designed to
mix material slowly, resulting in long residence times.

To improve the dispersion of materials in ribbon mixers, the ribbon blades are built to
be close to the cylindrical wall to provide high shear mixing. Rubber wipers can also
be fitted on the ribbons to lift material from near the wall into the middle of the ribbon.
Figure 4.6 shows a ribbon mixer with its cover removed to display its internals. The
advantages and disadvantages of ribbon blenders are listed in Table 4.3.

A list of ribbon blender suppliers by country is shown in Table 4.8.

Figure 4.6: Ribbon mixer

111
Table 4.3: Advantages and disadvantages of ribbon blenders
Advantages Disadvantages

The high rotational speeds thoroughly Throughput is dependent on the waste


blend waste media with reagents materials characteristics (moisture
Ribbon mixers can process in excess content, plasticity, particle size
of 20 tonnes per hour distribution) and ribbon design (which
Simple to operate and have few controls the residence time in the
moving parts to maintain mixer)

Many fixed facilities that treat sludges The waste material must often be
and liquids utilise ribbon mixers due to screened to remove debris larger than
their simplicity of operation and this offset, which is typically less than 1
thoroughness of mixing cm
Extremely intolerant of debris,
especially hard debris
Not capable of handling plastic, sticky
or tacky waste media
Are typically designed for fixed facilities

4.1.4 Quality control using mixing chambers


Typically, the S/S mix design determined during the treatability study (see Section 8)
is expressed on a weight/weight ratio, or the percentage of reagent to waste material
to be treated. For example, a 0.05 mix ratio or 5% additive addition rate, both equate
to 5 tonnes (5.1 tons) of reagent added to every 100 tonnes (102 tons) of waste
treated.

For pug-mills and screw mixers, the quality control revolves around calibrating the
waste and reagent feed-rates to the mixer. The waste feed rate for pug-mills or
screw mixers is typically determined by a belt-scale on the feed conveyor belt to the
mixing box. This belt-scale should be calibrated at least weekly, preferably daily,
using the manufacturers recommended calibration procedure. Figure 4.7 shows the
control panel for a belt scale.

The reagent feed can be conveyed to the pug-mill mixing box either by a conveyor
belt or screw auger. If the reagent feed is conveyed by a belt, a belt-scale can be
used to measure the feed rate and this belt-scale can be calibrated at the same time
as the waste feed belt scale. If a screw auger is used to convey the reagents to the
mixing box, a calibration between the screw auger rotation speed and the weight of
reagent delivered needs to be completed. This calibration should be verified at least
112
weekly, preferably daily. Figure 4.8 shows a pug-mill crew calibrating the reagent
feed from a screw auger.

The waste and reagent feed rates are typically manually controlled in older pug-mills,
but are automated in newer pug-mills. Figure 4.9 shows the control screen from an
automated pug-mill.

Similar to pug-mills and screw mixers, the quality control of ribbon blenders involves
calibrating the waste and reagent feed rates to the mixer. However, most ribbon
blenders are operated in a batch (rather than continuous) mode. Ribbon blenders
are typically mounted on load cells, which should be periodically calibrated on the
manufacturers recommended schedule. During operation, a known weight of waste
along with a known weight of reagents are charged into the ribbon blender and then
mixed.

Figure 4.7: Control panel for waste feed conveyor

113
4.2 Excavator bucket mixing
Standard excavators can be used to lift, sift, turn, and mix materials. A conventional
bucket, a bucket modified to improve efficacy with sifting and mixing (Figure 4.10), or
a specialised mixing head (Figures 4.11, 4.12, and 4.13) can be used for excavator
mixing.

Excavators can be used to treat waste materials either in-situ or ex-situ. For in-situ
S/S with an excavator, the waste to be treated is divided into treatment cells. The dry
or liquid reagents are spread over the surface of the treatment cell, and the
excavator operator uses the bucket or mixing head attached to the boom and stick to
mix the reagents into the waste, until homogeneous. The treated waste is then
either left in place or removed and stockpiled for transportation to the disposal
location.

For ex-situ S/S with an excavator, the waste to be treated is laid out on a concrete
pad or lined area (see Section 4.4) and the dry or liquid reagents are spread over
the surface of the waste. The excavator operator uses the bucket or mixing head
attached to the boom and stick to mix the reagents into the waste, until
homogeneous.

Figure 4.8: Calibrating reagent feed from a screw auger

114
Figure 4.9: Display panel for an automated pug-mill

Figure 4.10: Modified excavator bucket for improved mixing

115
The treated waste is then removed and stockpiled for transportation either to its
original, or an alternate, on-site location for placement, or to an off-site disposal
location. The next batch of waste is then placed on the treatment area and amended
with reagent for treatment. The advantages and disadvantages of excavator bucket
mixing are listed in Table 4.4.

A list of excavator bucket suppliers by country is shown in Table 4.8.

Table 4.4: Advantages and disadvantages of excavator bucket mixing


Advantages Disadvantages

Common construction equipment and It must be possible to spread the


requires minimal training for operation reagents on top the waste to be treated
Conventional or modified buckets are Specialised mixing attachments are not
typically unaffected by the presence of tolerant of debris and debris should be
debris (wood, concrete, or metal) in the removed prior to treatment
waste Excavator buckets are typically not
Up to 1,200 yd (920 m ) per day can designed for thorough mixing of wastes
3 3

be treated using an excavator with a and reagents and the treated material
conventional bucket may not be thoroughly homogeneous
unless sufficient mixing time is
employed
Plastic soils are not adequately mixed
Specialty mixing attachments can be
expensive, may not be available
locally, require more training to operate

116
Figure 4.11: Allu PMX Power Mixer attachment

Figure 4.12: Rotating mixing head attachment


117
Figure 4.13: Rotating rake mixing attachment

4.2.1 Quality Control using excavator bucket mixing


Excavator mixing (either ex-situ or in-situ) is a batch process, so the quality control
revolves around determining the weight of waste and reagent added during the
process. For all in-situ excavator mixing and much ex-situ mixing, the weight of the
waste is calculated using the in-place volume and density of the waste material. The
in-place volume is that of the treatment cell during in-situ excavator mixing and is
typically the excavated volume for ex-situ excavator mixing. Occasionally, the waste
material is transported over truck scales to obtain the waste weight for excavator
mixing.

For ex-situ and in-situ excavator mixing, the reagent is typically conveyed either by
conveyor belt or screw auger or pneumatically. If the reagent feed is by belt, a belt
scale can measure the weight of reagent conveyed to the treatment cell or treatment
pad. If a screw auger is used to convey the reagents, the screw auger rotation speed
and duration and the weight of reagent delivered needs to be calibrated.

118
Alternatively, reagents are pneumatically conveyed to the treatment cell. In this case,
the weight of waste in the treatment cell or on the treatment pad is managed in full
truck loads of reagent, for convenience..

4.3 Tillers and Other Ex-situ Mixers


Tillers employ rotating shafts with tines, teeth, or paddles designed to cut, fracture,
fold, lift, and/or shear soils. Tillers are typically not stand-alone equipment but are
coupled with other construction equipment such as excavators, front-end loaders,
tractors, or skid steers. The motor driven rotating shaft is typically mounted on the
tiller. The hydraulic power supply provided to the equipment is also attached to the
tiller.

For ex-situ S/S, the waste media are placed in lifts, ranging from 1-3 ft (0.3-1 m) in
depth in a treatment area, often a concrete pad or lined area (see Section 4.4). The
reagents (dry or liquid) are spread on top of the soil lift; typically fertiliser spreading
equipment is used for this task. The tilling equipment is then used to mix the soil and
the reagents. Multiple (three to four) passes are typically used to mix the waste
media with the reagents. When the waste and reagents are thoroughly mixed, the
treated material is then removed from the treatment area and stockpiled awaiting
confirmation of successful treatment.

There are environmental considerations associated with tillers. The treatment area
often used is a concrete pad or lined and berm-enclosed area designed to contain
the waste media and any spillage. Reagents and waste materials are sometimes
dusty in nature, and dust control, involving sprays or water mists, are often a
consideration. However, shrouding the treatment area within a sprung structure or
building and pulling air through the building and into a bag-house may also be
necessary to control dust emissions, especially when the mixing will occur in an
urban area, where sensitive receptors may be close (tens of meters) to the site
boundary.

Similarly, odour control during mixing can often be addressed by supplying a mist or
spray of an odour suppressing polymer or foam within and/or over the treatment
area. However, enclosing the treatment area within a sprung structure or building
and pulling air through the building and into an activated carbon canister or a
reactive scrubber may be necessary. Again, this level of odour control is likely only
when the mixing will occur in a populated area where sensitive receptors may be
close to the site boundary.

119
4.3.1 Rototillers
Rototillers use a rotating shaft with curved tines designed to cut, fracture, loosen,
and lift soil as the equipment rolls over the surface. Rototillers are typically mounted
on the three-point hitch on the back of backhoes, front-end loaders, or tractors and
use the hydraulic system of the host equipment to power its motor and rotate the
shaft. Figure 4.14 shows a rototiller attachment for a small tractor.

Rototillers can be used to treat soil either in-situ or ex-situ. For in-situ S/S using a
rototiller, the dry or liquid reagents are spread over the surface of the waste to be
treated. The rototiller is then driven over the waste, mixing the reagents into the top 8
to 12 inches of the waste. The treated waste is then removed and stockpiled for on-
site placement after all of the waste is treated, or for transportation to an off-site
disposal location. The next 8-12 inch (20-30cm) lift of waste is then amended with
reagent and treated, until the depth of contamination is reached.

Figure 4.14: Rototiller attachment

For ex-situ S/S with a rototiller, the waste to be treated is laid out in 8-12 in (20-
30cm) lifts on a concrete pad or lined area (see Section 4.4) and the dry or liquid
reagents are spread over the surface of the waste. The rototiller is then driven over
the waste, mixing the reagents into the waste. The treated waste is then removed
and stockpiled for transportation either to its original, or alternate, on-site location for
120
placement or to an off-site disposal location. The next lift of waste is then placed in
the treatment area and amended with reagent for treatment. The advantages and
disadvantages of rototillers are listed in Table 4.5.

A list of rototiller suppliers by country is shown in Table 4.8.

Table 4.5: Advantages and disadvantages of rototillers


Advantages Disadvantages

Can easily treat up to 1 ft (30cm) thick It must be possible to spread the


lifts of soils, either ex-situ or in-situ reagents on top of the waste to be
For in-situ S/S, up to 2,500 yd (1,912 treated
3

m3) per day can be treated using a The spreading operation involved in
rototiller ex-situ S/S limits the daily treatment
rate to closer to a maximum of 1,000
yd3 (765 m3) per day
They are typically not designed for
thorough mixing of soil and reagents
and the treated material may not be
thoroughly homogeneous unless
multiple passes are employed
Plastic soils may not be adequately
mixed
Not amenable to brick, concrete, or
metal debris greater than 2 in (5 cm) in
diameter
Have difficulty operating on un-even
soil surfaces
Are not standard roadway construction
equipment and may not be available
locally

4.3.2 Asphalt millers


Asphalt grinders, cold planers, or millers employ a rotating drum with cutter teeth to
loosen and remove worn asphalt as the equipment rolls over the asphalt surface.
Bomag supplies asphalt millers, which can scarify to a depth of up to 12 in (30 cm)
over a treatment width from 39-78 in (1-2 m). This equipment can be used to mix soil
with reagents for S/S. Figure 4.15 shows a Bomag asphalt miller.
Asphalt mixers can be used to treat waste materials either in-situ or ex-situ. For in-
situ S/S with an asphalt miller, the dry or liquid reagents are spread over the surface
121
of the waste to be treated. The asphalt miller is then driven over the soil, mixing the
reagents into the top 12 in (30 cm) of the waste, removing the treated waste, and
conveying the treated waste to dump trucks.

The treated waste is then stockpiled for on-site placement after all of the waste is
treated, or for transportation to an off-site disposal location. The next 12 in (30 cm)
lift of waste is then amended with reagent and treated, until the depth of
contamination is reached.

For ex-situ S/S with an asphalt miller, the waste to be treated is laid out in a 12 in
(30cm) lift on a concrete pad or lined area (see Section 4.4) and the dry or liquid
reagents are spread over the surface of the waste. The asphalt miller is then driven
over the waste, mixing the reagents into the waste, removing the treated waste, and
conveying the treated waste to dump trucks. The treated soil is then stockpiled for
transportation to its original location for placement, or to an off-site location. The next
lift of waste is then placed and amended with reagent for treatment. The
advantages and disadvantages of asphalt millers are listed in Table 4.6.

A list of asphalt mixer suppliers by country is shown in Table 4.8.

Figure 4.15: Asphalt miller


Table 4.6: Advantages and disadvantages of asphalt millers
122
Advantages Disadvantages

Can easily treat up to 1 ft (30cm) thick It must be possible to spread the


lifts of soils, either ex-situ or in-situ reagents on top the waste to be treated
Are standard roadway construction The spreading operation involved in
equipment and can typically be found ex-situ S/S limits the daily treatment
worldwide rate to closer to a maximum of 1,000
Little technical expertise is required to yd3 (765 m3) per day
operate the equipment They are typically not designed for
Capable of handling non-metallic thorough mixing of soil and reagents
debris and the treated material may not be
For in-situ S/S, up to 3,500 yd3 (2,680 thoroughly homogeneous
m3) per day can be treated Non-cohesive or plastic soils may not
be adequately moved onto the
conveyor
Have difficulty operating on soil
surfaces

4.3.3 Paddle aerators/compost turners


Paddle aerators consist of triangular paddles mounted on a rotating shaft. Paddle
aerators can be mounted in the front of front-end loaders, bulldozers, tractors or skid
steers, using the equipments hydraulic system to power the motor, which turns the
rotating shaft. The paddles are curved to lift and turn material. As the equipment
moves forward, the rotating paddles lift and turn the material, providing a mixing
action. Paddle aerators vary in width from 6-11 ft (1.8-3.4 m) and can mix material to
a depth of 5 ft (1.5 m). Figure 4.16 shows a paddle aerator mixing soil in a windrow.

Paddle aerators can be used to treat waste either in-situ or ex-situ. For in-situ S/S
with a paddle aerator, the dry or liquid reagents are spread over the surface of the
waste to be treated. The paddle mixer is then driven over the waste, mixing the
reagents into the waste, to a depth of up to 3 ft (0.9 m). If greater treatment depths
are required, the treated waste can be removed and stockpiled for later on-site
placement, or for transportation elsewhere. The next layer of waste is then amended
with reagent and treated, until the depth of contamination is reached.

For ex-situ S/S with a paddle aerator, the waste to be treated is laid out in a lift of up
to 3 ft (0.9 m) deep on a concrete pad or lined area (See Section 4.4) and the dry or
liquid reagents are spread over the surface of the waste. The paddle aerator is then
driven over the waste, mixing the reagents into the waste. The treated waste is then
removed and stockpiled for transportation either to its original location for placement
or to an off-site disposal location. The next lift is then placed and amended with
123
reagent for treatment. The advantages and disadvantages of paddle aerators are
listed in table 4.7.

A list of paddle aerator suppliers by country is shown in Table 4.8.

Table 4.7: Advantages and disadvantages of paddle aerators


Advantages Disadvantages

Can treat lifts of soils up to 2-5 ft (0.6- It must be possible to spread the
1.5 m) in depth either ex-situ or in-situ reagents on top of the waste to be
For in-situ S/S, up to 2,500 yd3 (1900 treated
m3) per day can be treated using a The spreading operation involved in
paddle aerator ex-situ S/S limits the daily treatment
rate to closer to a maximum of 1,000
yd3 (765 m3) per day
Cohesive and plastic materials may not
be thoroughly mixed
Not amenable to brick, concrete, or
metal debris greater than 4 in (10 cm)
in diameter
Paddle aerators are specialised
equipment produced by a limited
number of suppliers and may not be
available locally

124
Figure 4.16: Paddle aerator mixing windrowed soil

4.3.4 Quality control using tillers and paddle mixers


Tiller mixing, either ex-situ or in-situ, is a batch process, so the quality control
revolves around determining the weight of waste and reagent added during the
process. For all in-situ rototiller mixing and during ex-situ mixing, the weight of the
waste is calculated based on the in-place volume and density of the waste material.
The in-place volume is that of the treatment cell during in-situ rototiller mixing and is
typically the excavated volume for ex-situ rototiller mixing. Occasionally, the waste
material is transported over truck scales to obtain the waste weight for rototiller
mixing.

For ex-situ and in-situ rototiller mixing, the reagent is typically conveyed by conveyor
belt, screw auger or pneumatically. If the reagent feed is conveyed by belt, a belt-
scale can be used to measure the weight of reagent delivery to the treatment cell or
treatment pad. If a screw auger is used to convey the reagents, the screw auger
rotation speed (and duration) and the weight of reagent delivered needs to be
calibrated. Alternatively, reagents can be pneumatically conveyed to the treatment
cell, and in this case the weight of waste (in the treatment cell or on the treatment
pad) is often controlled so as to require full truckloads of reagent to simplify reagent
delivery.

125
Table 4.8: mixer manufacturers
Location Paddle Aerator Asphalt Miller Rototiller Ribbon Blender Screw Mixers Pug-mill
Aran International
Australia
Pty
Austria Andritz Group Andritz Group
Belgium Menart Sprl
Shandong Sunco Weifang General Shandong Yuntai Shanghai Senfan Yangzhou Nouya
Agricultural Machinery Machinery Co. Machinery Co. Machinery
Equipment Wuhan Kudat Yucheng Dadi Yangzhou Nouya Company
Technology Industry and Machinery Co. Machinery Co. Jinan Xucheng
China Zhenzhou Repale Trade Zhenzhou Co.
Machinery Co. Whirlson Trade Double Crain
Co. Ltd Machinery
Manufacture
(Leling) Co., Ltd.
Steelcraft, Inc UniTrak
Canada Engineered
Products
Denmark Baltic Korn A/S
Finland Allu Group
Backhus Wirtgen BmbH Backhus J. Engelsmann AG
Kompost- Kompost- CATS GmbH
Germany
Technologie Technologie Doppstadt
Doppstadt Putzmeister
Ambica Shitla Road Krishna Rana Perforators Leo Road Pvt Ltd
Engineering Equipment Engineering Atlas Industries
India
Works Shiv Shakti Road Rana Perforators S.P. Enterprises
Equipments Shitla Road
126
Equipment
WAMGROUP
Italy
IMER Group
Switzerland Gericke AG
Winkworth Mixer BG Europa Ltd
UK
Co.
Midwest Bio- Roadtec BEFCO Aaron Process US Air Filtration DustMASTER
Systems Maddock Rotomec Equipment Auger Enviro Systems
Brown Bear Equipment Terex Servis-Rhino Applied Chemical Manufacturing Eagle Iron Works
Cedarapids Technology Specialists Excel Machinery
Bomag America National Bulk Acrison JW Jones
Equipment Kohlberg/Pioneer
Jaygo Maxon Ind.
United Peerless
States Conveyor and
Manufacturing Co.
McLanahan Corp.
Pug-mill Systems,
Inc
Rapid
International, Inc
Terex Cedarapids

127
4.4 Mixing pits
Mixing pits are earthen pits, open-topped tanks of metal or concrete, or three-side
concrete (though can also be metal or wood) enclosures in which ex-situ S/S
treatment can take place.

The waste media are placed into the mixing tanks, and the dry reagents are added
pneumatically or via super-sacks (e.g. 1 to 2 ton bags or 0.9-1.8 tonnes); liquid
reagents are pumped into the pit. Typically an excavator is used to mix the waste
media with the reagents, though tillers are sometimes employed. When the waste
and reagents are thoroughly mixed, the treated material is then removed from the
mixing pits and may be stockpiled, to await confirmation of successful treatment.

There are environmental considerations for mixing pits. Since materials are
conveyed into and out of mixers, secondary containment is typically required to
collect and contain any spillage. For tanks and enclosures, secondary containment
may involve double-walled structures, or placement within a bermed and lined area.
Often the reagents, and sometimes the waste materials, to be mixed are dusty in
nature and dust control via water mist or spray within and/or over the mixing pit may
be necessary. However, shrouding the mixing pits within a sprung structure or
building and pulling air through the building and into a bag-house may be necessary.
Similarly, odour control during mixing may be addressed by an odour suppressant
polymer mist or spray or foam within and/or over the mixing pit.

However, enclosing the mixing pit within a sprung structure or building and pulling air
through the building and into an activated carbon canister or a reactive scrubber may
be necessary. This level of odour control is likely only when the mixing will occur
close to a populated area or sensitive receptors. Figure 4.17 shows an excavator
with a standard bucket mixing Portland cement into lead contaminated soil and
debris on a concrete pad.

128
Figure 4.17: Excavator mixing of Portland cement into Pb-contaminated soil
and debris

4.4.1 Earthen Pits


Earthen mixing pits may be excavated into contaminated soil or established within
impoundments and lagoons. Waste media are transferred into the pits using
excavators or loaders, and the reagents (wet or dry) are added to the pit. An
excavator is then used to mix the waste and the reagents until a visually
homogeneous product is formed.

The treated waste is sampled to verify successful treatment, and then may be
removed and stockpiled for off-site disposal or later placement back on-site;
alternatively, the product can be left in-place, on-site. Figure 4.18 shows an
excavator with a standard bucket mixing hydrated lime into an oily waste in an
earthen mix pit. The advantages and disadvantages of earthen pits are listed in
Table 4.9.

129
Table 4.9: Advantages and disadvantages of earthen pits
Advantages Disadvantages

Simplicity and flexibility Limited by the reach of the excavator


The size can be varied as necessary to arm
contain the mixing volume, which can Excavators or loaders may need to
range from 20 to 1,000 m3 (26-1300 track onto the treated material to
yd3) access other pit areas
No need for secondary containment Sprung structures or buildings are
Many simple options to place the required to contain dust or odour
waste media into the pits controls - these structures would need
No feed preparation is required to be mobile if the treated material is
Typically, only an equipment operator left in place within the mixing pit
and perhaps a labourer are required to
run a mixing pit and only an excavator
is needed for the mixing operation

Figure 4.18: Bucket mixing hydrated lime and oily waste in earthen pit

130
4.4.2 Open-top tanks
Open-top tanks are metal or concrete structures in which the waste/soil are placed
for mixing. The waste material is conveyed to the open-top tank by conveyor,
excavator, or loader and the reagents, whether dry or wet, are added directly to the
tank.

An excavator is then used to mix the waste and the reagents until the product is
visually homogeneous. The treated waste is then sampled to verify treatment, and
the product is removed and stockpiled for off-site disposal or placement on-site.
Figure 4.19 shows an excavator bucket mixing reagents into hexavalent chromium-
contaminated soil in a roll-off box. The advantages and disadvantages of open-top
tanks are listed in Table 4.10.

Table 4.10: Advantages and disadvantages of open-top tanks


Advantages Disadvantages

Simple metal containers such as roll- The lateral extent of a mixing tank is
off boxes or inverted sea-land only limited by the reach of the arm of
containers (with the bottoms removed) the mixing excavator
can be used for tanks. Tanks can The metal bottoms of the tank require
range in size from 10 to 1,000 yd 3 (7.7- careful consideration by the excavator
765 m3) operator to avoid puncturing with the
Many simple options to place waste excavator bucket during mixing
into the tank, and little to no feed
preparation is required.
Minimal staff required

131
Figure 4.19: Stabilization of hexavalent chromium-contaminated soil in a roll-
off box

4.4.3 Enclosures
Enclosures often have three concrete sides and a concrete bottom. The concrete
bottom of the enclosure is sloped to allow the enclosure to drain away from the open
side.

The waste media can be transferred into an enclosure using conveyers, excavators
or loaders. Reagents are added directly to the tank. An excavator, loader, or
bulldozer is then used to mix the waste and the reagents thoroughly, until visually
homogeneous. The treated waste is then sampled to verify successful treatment has
been carried out. The treated waste is often removed and stockpiled for off-site
disposal or placement on-site. The advantages and disadvantages of enclosures are
listed in Table 4.11.

132
Table 4.11: Advantages and disadvantages of enclosures
Advantages Disadvantages

Enclosure may comprise a concrete The footprint of an enclosure is


slab with the sides made of portable generally only limited by the reach of
concrete barriers (e.g. K-rails or Jersey the excavators mixing arm, which must
barriers) be able to treat all of the waste
The size of the enclosure can range material presented
from 20 to 1,000 yd3 (14.4-765 m3) Liquid or self-levelling wastes and
Little to no feed preparation is required treated materials are unsuitable for
prior to waste placement treatment in mixing enclosures as even
personnel requirements are generally with a profiled/sloping base (away from
limited to an equipment operator and the opening) the enclosure will not
labourer; an excavator is needed for often adequately contain these types of
mixing materials

4.5 Ancillary equipment for ex-situ mixing


Except when excavator mixing is employed, the wastes often require pre-processing
to remove debris and to reduce their particle size to enable adequate blending with
the S/S reagent to be achieved. This section will discuss some of the material
handling equipment requirements when preparation of waste material is needed
before ex-situ mixing.

4.5.1 Screening equipment


Screening equipment is often employed to classify or separate waste material by
size. Screening equipment may vary on the size classification (or cuts) required,
the dimensional opening of the screen material, and the material throughput.

Grid screeners or grizzlies consist of a parallel grid of inclined iron or steel bars.
Hydraulic vibration or an oscillation of the grid is often employed to improve the
separation efficiency and throughput of the grid. Grizzly screens (Figure 4.20) are
often used to remove large material, greater than 6 in (15 cm) in diameter from the
waste material.

Screening plants employ inclined vibrating screen-decks for size separation. Screen
decks are 4 ft wide by 8 ft long (1.23 x 2.46 m), with screen openings ranging from 4
in x 4 in (10 x 10 cm) down to 100 mesh (149 m), and a screen plant can have one
or multiple screen decks. The screen decks are vibrated to provide maximum
screening efficiency and the speed of the vibration can be adjusted.

133
Trommels are horizontal rotating cylindrical screens with up to 500 ft 2 (46.5 m2) of
screening area (Figure 4.21). Their large screening area produces highly efficient
action and high throughput rate. The screen dimension is commonly 1 in x 1 in (2.5 x
2.5 cm), but can be varied as necessary. The pitch or incline of the trommel can also
be varied to improve screening efficiency.

For all of the screening equipment, the undersized material passes through the
screen, while the oversized material moves across the screen and is discharged off
the end of the screen. The oversized materials from multiple screen decks can be
combined or kept separate, depending on the screen plant design. For most ex-situ
S/S mixers, the final screened material is often needed to be less than 1 in (2.5 cm)
in diameter.

Grizzly screens are often used to separate large material (greater than 6 in (15 cm)
in diameter) from the media, while screen plants or trommels are used to provide a
less than 1 in (2.5cm) in diameter material for S/S treatment (Figure 4.21).

Figure 4.20: Grizzly screen for separation of large debris, with over-sized
material, to the left and processed soil for ex-situ mixing, right

134
Figure 4.21: Grizzly screen (foreground) combined with a trommel to remove
battery casings and debris from excavated material

4.5.2 Crushing equipment


Crushing equipment is employed to size-reduce hard waste material and debris,
often in combination with screening equipment. Crushing equipment typically relies
on impact (collision of the material with moving surfaces), attrition (intra-aggregate
abrasion between two hard, moving surfaces), shear (cleaving action which occurs
as aggregates are pinched between two surfaces), and compression (crushing
material between two surfaces) to fragment the waste material.

Impactors use fixed hammers attached to a rotating axis and breaker bar lining the
internal radial surface, while hammer mills employ free-swing hammers attached to a
rotating axis with screening bars located along the radius of the hammers.

The offset of the breaker bars from the hammer radius determines the size of the
crushed material. The screen bars in the hammer mill also produce attrition of the
material before it exits the crusher. Single roll crushers employ breaker bars
attached to a rotating drum above an impact plate.

135
A combination of shear, impact, and compression fragments the material as it
passes between the breaker bars and the impact plate. The offset distance between
the radius of the breaker bars and the impact plate determines the size of the
crushed material. Jaw crushers use a swinging jaw to compress material against a
fixed jaw. Impactors handle wet, sticky material best, as they are open-bottomed.
Hammer mills and single roll crushers handle hard materials, but hammer mills are
not tolerant of plastic or sticky materials as they blind the screen bars.

Jaw crushers are best at handling very hard materials, but cannot be used with
sticky or abrasive materials. Jaw and single roll crushers produce coarse (greater
than 1.5 in (3.75 cm) in diameter) crushed material, while impact hammers and
hammer mills can produce a smaller material, down to less than inch (1.25 cm) in
diameter, depending on the offset.

4.5.3 Shredding equipment


Shredding equipment is often employed to size-reduce soft shear-able waste
material and debris, including wood, vegetation, plastics, and some metals.
Shredding is often used in conjunction with screening equipment, and size-reduction.

Shredding equipment typically relies on shear to fragment waste material and debris.
Shredding equipment varies in the mechanics of the shredding and the size of the
shredded product.

Rotary shear shredders (Figure 4.22) employ counter-rotating shafts or cutter


blades. Material is shredded as it is drawn between the interfaces of the two counter-
rotating cutter blades. The close tolerance of the cutter blades performs the shearing
action.

Tub grinders use an inclined tub or rotary screen to feed a rotor or hammer mill. The
tumbling action of the tub screens the material prior to shredding.

Rotary shear shredders can process wet and plastic material. Tub grinders are more
suited to handle hard and abrasive materials, due to their efficient use of impact
forces. Rotary shear shredders produce a coarser shredded product than tub
grinders, since the tolerance of the cutter blades is greater.

136
Figure 4.22: A rotary shear shredder processing scrap metal

4.5.4 Magnetic separation equipment


Magnetic separators are used to remove ferrous metal debris from waste or soil.
After removal, the metal is often taken off-site for recycling. Additionally, separating
the ferrous metal prior to further processing (such as crushing or shredding
operations) protects equipment and reduces maintenance. Magnetic separators are
typically suspended over the body or head pulley of a conveyor.

Stationary overhead magnets can be suspended using a cable sling-support and


periodically must be swung to the side for the ferrous metal to be removed from the
surface of the magnet. Continuous, self-cleaning magnets involve a belt with a
movement that is used to clean the separated ferrous metal from the belt.

There are two configurations for magnetic separators: in-line or cross belt. In-line
continuous, self-cleaning magnets are installed with the cleaning belt running parallel
to the conveyor movement. Therefore, the on-line configuration is only available
above the head pulley. Cross-belt continuous, self-cleaning magnets are installed
with a cleaning belt running perpendicular to the direction of conveyor movement,
and can be placed over the body or head pulley of the conveyor. The In-line
configuration requires less magnetic force than the cross belt, but the discharge of
the cross belt is easier to implement.

137
4.6 On-site placement of ex-situ treated material
Waste materials treated by ex-situ mixing are often either disposed of in an off-site
disposal facility (e.g. landfill) or are disposed of on-site. When the material is
disposed of off-site, it is typically stockpiled until it is confirmed to have met the
required performance criteria (discussed in Section 7) and then loaded into trucks,
or railcars, for transport to the off-site disposal facility. When the material is disposed
of on-site, it often must be properly placed, spread, and compacted. This section will
discuss the procedures for placing and compacting the treated material, and the
material properties that facilitate the placement and compaction.

4.6.1 Timing
When the performance criteria for the ex-situ S/S treated material include specified
properties related to durability, UCS strength and/or permeability, the treated
material must be placed and compacted within 24 hours (preferably 12 hours) of
treatment.

As the development of physical performance criteria relies on cementitious or


pozzolanic reactions, initial set generally occurs within the first 24 to 48 hours. It is
therefore necessary to have the treated material placed and compacted before the
initial set has occurred. If the initial set is disrupted or placement and compaction
occurs after the initial set, the treated material will become impaired and unlikely to
achieve the required performance goals. Therefore, ex-situ S/S treated material
should be placed and compacted on-site within the same day as it is treated.

It is important to note that setting of S/S materials is generally much slower than that
of normal concrete on account of the nature of the materials being treated and the
interaction of waste and soil matrices, and contaminants with cementitious reactions.
That said, the rate of setting is an important parameter to S/S as it is to concrete,
and should be monitored to ensure compliance with performance targets established
by bench-scale testing.

If the performance criteria for ex-situ S/S treated material does not include the
development of physical performance criteria, the timing of the on-site placement
and compaction is not as critical. Under these circumstances, the treated material
can be stockpiled after mixing until it is operationally advantageous for its placement
and subsequent compaction.

138
4.6.2 Placement
Typically, ex-situ S/S treated material is transported on-site in dump trucks or off-
road dump trailers. The treated material is transported to a placement area, dumped
and spread using a dozer in loose lifts that tend not to exceed 18 inches (45 cm) in
depth. Figure 4.23 shows the transport and spreading activities for pug-mill-treated
S/S material, involving a dozer making multiple passes over the loose lift to remove
any large voids in the placed material. Figure 4.24 shows the surface of material
after multiple passes of the dozer.

For ex-situ S/S treated material with physical performance criteria, the treated
material needs to have sufficient moisture to allow cementitious and pozzolanic
reactions to occur. Therefore, the treated material may have to be placed with a
higher-moisture content than the optimum required for maximum density.

Additional water may also be applied to such treated material during placement to
ensure that it does not dry out to the point where cementitious reactions cannot
proceed. The avoidance of water starvation involves careful consideration by the
field engineer overseeing materials placement.

4.6.3 Compaction
Compaction of treated, placed material involves maximising the density of the
product and thereby minimisation of its volume.

Compaction requirements
Compaction can be based on achieving greater than specified maximum dry density,
as defined by its moisture-density relationship, or degree of effort needed in the
field.
For the first case, the compaction requirement is based on attaining a greater than a
specified percentage (typically 90% or 95%) of the maximum dry density from
Proctor compaction testing. The Proctor method (ASTM D698 or D1557) is a
laboratory determination of the moisture content-density relationship for a compacted
soil, or soil-like material, and moisture content at which the material has its maximum
compacted density.
In Figure 4.25, the moisture content range for compaction to 95% of the maximum
Proctor density is shown. Compaction of placed ex-situ S/S treated material with
moisture content between 6.8% and 22.3% should result in a dry density of greater
than 103.4 pcf (0.016 tonnes/m3) (95% of the maximum dry density). Ex-situ S/S
treated material with a moisture content below 6.8% would have to be wetted or a

139
moisture content above 22.3% would, depending on the material, need to be either
wetted or dried before it could be compacted.
Figure 4.25 shows the moisture/density relationship from the Proctor compaction
testing of an ex-situ S/S treated material.

Figure 4.23: Transport and placement of ex-situ S/S material

Figure 4.24: Placed ex-situ S/S-treated material after multiple


passes of the dozer

140
This first case is only applicable when the performance criteria for the ex-situ S/S
treated material do not include the physical performance criteria (durability, strength
and/or permeability).

As discussed (Section 4.6.1), physical performance criteria rely on cementing


reactions within the treated material, with placement and compaction being
completed before the initial set is complete. Changes to the moisture content of the
placed material will disrupt the initial set leading to failure to meet performance
criteria. Therefore, the compaction requirement for ex-situ S/S treated material that is
needed to achieve physical performance criteria, is typically expressed as a
specified number (typically at least two) of passes with specified compaction
equipment (see below).

Compacting equipment
The following discussion on compaction equipment is based on the Caterpillar
Compaction Manual (Caterpillar Tractor Company, 1989).

Pneumatic tire compactors: Pneumatic tire compactors are used on small to


medium sized compaction jobs, primarily on bladed, granular base materials.
Pneumatic tire compactors are not suited for high production, thick lift compaction
projects. The compaction forces (pressure and manipulation) generated by the
rubber tires work from the top of the lift down to produce density. The amount of
compaction force can be varied by altering the tire pressure (the normal method) or
by changing the weight of the ballast (done less frequently). The kneading action
caused by the staggered tire pattern helps seal the surface.

One advantage of pneumatic compactors is that there is little bridging effect between
the tires, thus they seek out the soft spots in the fill and compact them. For this
reason, they are sometimes referred to as "proof" rollers.

Sheeps-foot roller: Sheeps-foot rollers are named after Roman road builders, who
used to herd sheep back and forth over base material until the road was compacted.
The sheeps-foot roller (Figure 4.26) consists of pads attached to a cylindrical drum.
These pads penetrate through the top lift and actually compact the lift below. When a
pad comes out of the soil, it fluffs the material at the surface, resulting is a loose-
layer of surface material, and when the next lift is placed this becomes compacted,
and the next layer is fluffed up.

141
Because the top lift of soil is always being fluffed, the process helps aerate and dry
out treated materials with excessive moisture contents. However, the top lift at the
end of each day is loose and not compacted, and this has the disadvantage of acting
like a sponge, should it rain, slowing the compaction process.

By their very nature, sheeps-foot rollers only exert pressure to effect compaction
they do not provide impaction or vibration. As such, multiple-passes (six to ten in
total) may be necessary to achieve the required density in 8 to 12-in (20-30 cm) lifts.

Figure 4.25: Proctor compaction-testing results for ex-situ S/S treated material.
Note the moisture content range defining greater than 95% maximum dry
density. (pcf = pounds per cubic foot)
142
Figure 4.26: Sheeps-foot roller compacting S/S treated material
(note the oval pads on the roller)

Tamping foot compactors: Tamping foot compactors are high speed, self-
propelled, non-vibratory rollers. They usually have four steel padded wheels and are
equipped with a dozer blade. Their pads are tapered with an oval or rectangular
face.

Like the sheeps-foot, a tamping foot compactor compacts from the bottom of the lift
to the top. But because the pads are tapered, the pads can walk out of the lift without
fluffing the soil. Therefore, the top of the lift is compacted and the (top lift) surface is
relatively smooth and sealed.

Because tamping foot compactors are capable of speeds near 15-20 mph (24-32
kmh), they develop and exert all four forces in compaction: pressure, impaction,
vibration and manipulation to the lift of S/S material. This not only increases their
effective compaction (of a lift) but the production rate is relatively much higher.
Generally two to three passes will achieve the desired density in 8 to 12 in (20-30
cm) lifts, though four passes may be needed in materials that are plastic in nature.

The main limitation to the use of tamping foot compactors is that they are best suited
for large projects, as long uninterrupted passes at a high speed are key to a high
production rate. Tamping foot compactors are considerably more expensive than
vibratory compactors.
143
Vibratory compactors: Vibratory compactors work on the principle of the
rearrangement of S/S particles to decrease voids and increase density. Vibratory
compactors come in two types: smooth drum and padded drum varieties. Smooth
drum vibratory compactors generate three compaction forces: pressure, impaction
and vibration. Padded drum units also generate manipulative force. Compaction is
generally assumed to be uniform throughout the lift subject to vibratory compaction.

The forces generated by a vibrating drum hitting the ground are large, resulting in a
densification of the lift. The amount of compaction achieved is a function of the
frequency of these blows, the force of the blows and the time period over which the
blows are applied. This frequency/time relationship accounts for slower working
speeds and a speed of 2 to 4 mph (3.2 to 6.4 kph) provides the best results.

Smooth drum vibratory compactors (Figure 4.27) were introduced first and are most
often used on granular materials, with particle size ranging from large rocks to fine
sand.

They are also used on semi-cohesive soils with up to 10% cohesive soil content. The
thicknesses of a lift will vary according to the size of the compactor but, generally,
the lift thickness of granular material should not exceed 24 in (60 cm).

When padded drum machines were introduced, the materials that could be treated
were extended to include soils with up to 50% cohesive material and a greater
percentage of fines. When the pad penetrates the top of the lift it breaks the bonds
between the particles within the cohesive soil and improves compaction results. This
is a result of the geometry of the pads which are involute (i.e. curled inward), self-
cleaning, and able to walk out of the lift without fluffing the surface. The typical lift
thickness for padded drum units on cohesive soil ranges from 12 to 18 inches (30-45
cm).

Figure 4.27: Smooth roller compacting ex-situ S/S treated material


144
5.0 In-situ S/S equipment and its application

5.1 In-situ auger mixing


Auger mixing is the most widely used technique of in-situ soil mixing and is
especially useful for deeper applications, greater than 15 feet (4.6 m).

Auger mixing is suitable for a wide range of soil types to depths in excess of 100 ft
(30.8 m), although specialised methods are necessary for depths beyond 60 ft (18.5
m) below ground surface (BGS). Quality control is easier in comparison to all other
types of in-situ soil mixing (including e.g. rotary drum mixing and bucket mixing).

5.1.1 Equipment
Generally the equipment used for in-situ soil mixing consists of augers and their
support carriers to conduct the injection and mixing, a batch plant to prepare the
reagent slurry and pump it to the auger assembly, and miscellaneous support
equipment.

Figure 5.1: In-situ processes

145
Augers and their support carriers
There are two main types of drilling rigs used for in-situ soil mixing using an auger:
crane- and excavator-mounted units. Crane-mounted units consist of mechanically
driven rotary drilling heads fitted on a crawler crane. Excavator-mounted units
consist of hydraulically driven, rotary drilling heads fixed to an excavator base.

Crane-mounted units can generally accept slightly larger diameter augers (10-12 ft
or 3-3.6 m) and, in theory, have deeper mixing depth capabilities (by increasing the
amount of boom on the crane and installing a longer Kelly bar). On the other hand,
excavator-mounted rigs are generally best suited to smaller diameter augers (9-10 ft
or 2.8-3.1 m) and have stroke lengths limited to the mast height of the equipment
(generally less than ~60 ft or 18.5 m). However, recent advances in excavator-
mounted equipment have made these systems comparable and (in many ways)
more advantageous to use, than crane-mounted systems. This results from their
superior mobility, higher operating torque and lower mobilisation costs. Examples of
crane-mounted rigs are shown in Figures 5.2 and 5.3, while excavator mounted rigs
are shown in Figures 5.4 and 5.5.

The mixing head of the crane-mounted system is mechanically powered by an


engine/transmission combination, whereas the mixing head of the excavator-
mounted system is hydraulically powered using hydraulic pumps running off the
engine system of the host machine. Crane mounted systems generally have
maximum torque outputs (typically 250,000 350,000 ft-lbs or 34.56-48.39 kgf-m)
that far exceed the listed maximum torques of their excavator mounted counterparts
(typically 100,000 250,000 ft.lbs or 13.83-34.56 kgf-m). However, the operating
torque of crane-mounted systems is much less than the maximum torque, which is
only achieved during small portions of the engine power curve. Higher torque
requirements result in lower rotation speeds and therefore less thorough mixing. The
operating torque of excavator-mounted systems is very close to maximum available
torque and often exceeds the operating torque of much larger crane-mounted
systems. This allows higher rotation speeds at higher torque, resulting in a better
mixing.

146
Figure 5.2: A crane-mounted soil-mixing rig

In both the crane and excavator-mounted systems, reagents are normally pumped
through a hollow Kelly bar and out of the auger ports. Reagents are most commonly
added in a liquid or grout form that acts as both the drilling lubricant and the final
stabilization reagent, but reagents can also be added in a dry-powder form.
Occasionally, for shallow/smaller applications, reagents are added at the surface, by
spreading.

Auger mixing provides the highest quality in-situ mixing available. The high torque
available in auger-mounted rigs makes them ideal for mixing dense sands and stiff
clays. A number of manufacturers produce drill rigs that can be used for soil mixing,
but most commercially available rigs require modification through additional specialty
equipment to be fully utilised in this application.

Auger mixing involving both crane and excavator mounted systems has been in use
for 2530 years, however, in recent years the remediation industry has increasingly
tended to use excavator-based systems, due to their high level of mobility and
consistent torque output.

147
The actual soil cutting/mixing component, the auger bit, has been constructed in
many styles, though they all operate essentially the same. Figure 5.6 and Figure 5.7
illustrate two styles of auger bits. Both were used to complete in-situ auger treatment
on soils contaminated with coal tar wastes. Note that Figure 5.6 shows the use of
aiming stakes to locate column placement. This has largely been replaced by use of
very accurate GPS systems supplemented with depth control from a nearby survey
crew.

Figure 5.3: A crane-mounted soil-mixing rig

Figure 5.7 illustrates the auger often used today. In this case it is a 10ft diameter
auger and one can see the cutting teeth on the forward side of the blade and the
reagent injection ports on the trailing edge of the blade. Below the blade is a
projection, commonly called a stinger which helps guide the auger bit into the soil.

148
Batch plant equipment
Batching plants consisting of silos, pumps, and mixing tanks are critical to the
success of most in-situ soil mixing projects, relying on reagent delivery through ports
located on the mixing head of an auger. Batching plant configurations vary widely, as
their make-up/configuration includes the practitioners preference, reagent type and
quantity, the number of reagents to be used and the reagent pumping distance, and
siterelated constraints.

For applications where the reagent(s) is being added as a grout, the batch plant can
become quite extensive. An example of an automated batch plant is shown in Figure
5.8.

The majority (>95%) of in-situ soil mixing projects use pre-mixed reagents in a fluid
grout or slurry form prior to injection. With the increased acceptance of this S/S
technology, it is not uncommon to blend two or three dry reagents in a batching plant
that is automated.

Figure 5.4: An excavator-mounted rig


149
Automated plants use weigh-scales to accurately measure mix components in a
process that improves efficiency and quality control.

For applications where the reagent is being delivered from dry, pressurised storage
tanks, pneumatic pumps are required. An example of a self-propelled dry pneumatic
reagent hopper is shown in Figure 5.9.

Ancillary equipment
In addition to the batching plant, a variety of ancillary support equipment is
necessary for the successful completion of an auger-based soil-mixing project. The
support equipment may include excavators, dozers, loaders, forklifts, man-lifts,
pumping systems, hoses, survey equipment and data loggers.

5.1.2 Staffing requirements


Typically an auger-based soil-mixing project requires a supervisor, drilling rig
operator and support labour, a batch plant operator and support labour, and
QC/engineering staff. However, staff requirements vary from project to project,
depending on ancillary work required and chosen batch plant configuration.

Figure 5.5: An excavator-mounted soil-mixing rig

150
Figure 5.6: Auger used at the USX Site, Duluth, MN

Figure 5.7: Auger used at a coal gas plant site in FL


151
Figure 5.8: An automated batch plant

5.1.3 Treatment metrics and considerations


The maximum treatment depth, the optimum auger diameter and reagent addition
methods and production rate are highly variable, depending on site-specific
conditions and the equipment employed. Obstructions in the sub-surface, such as
concrete slabs, pipes, rocks, and disused cables can substantially slow production,
break augers, and significantly increase cost. Such sub-surface objects should be
removed prior to starting auger-treatment to avoid potential costly delays. Additional
treatment metrics and considerations follow below.

152
Figure 5.9: Self-propelled dry storage silo

Depth of treatment
The depth of treatment is dependent on a number of factors, including soil-type and
relative density, the auger diameter and its configuration, the torque available to the
Kelly-bar, mast length and downward force capability (drill crowd and tool weight).

As previously mentioned, auger systems are generally limited to depths less than
60ft (18.46 m), unless specialty high-torque power units are used with smaller
diameter augers. Crane-mounted systems tend to have slightly deeper maximum
treatment depths, but mast extensions for excavator-mounted rigs are available to
match the depth capabilities of almost any crane-mounted unit.

Most of the excavator-based systems currently in use are limited to maximum depths
in the 4555ft (14-17 m) range. It is important therefore, that practitioners should
comment on the depth limitations of their respective equipment in the soil conditions
anticipated at the project site. The effective maximum working depth can be derived
from geotechnical data, such as from CPTs, SPTs, and other common site
investigation-derived data. A decrease in the auger diameter (other factors being
equal) will generally allow for treatment at greater depth, however smaller augers
have a major adverse impact on production rates and cost. Table 5.1 lists some

153
remedial operations employing in-situ auger mixing, including the depth treated and
auger diameter employed.

Anticipated production rates


Generally, auger-based systems are capable of treating between 200-600 yd3 (153-
460 m3) of soil or sludge per working day (based on an 8 hour shift). However, this is
somewhat dependent on a variety of factors including relative soil geotechnical
properties, maximum treatment depth and reagent dosage, etc. At the optimum
production rate, with treatment depths of 10-40 ft and 8-10 ft diameter augers (3.1-
12.3 m and 2.5-3.1 m respectively), it is not unusual to treat over 800 yd 3 (612 m3)
per auger in an 8-10 hour shift. However, due to maintenance requirements and the
occasional (expected) equipment breakdown, a lower average production rate
should be anticipated.

Reagent addition methods


Reagent addition is normally supplied by pumping/injecting through a wet Kelly-bar/
auger, or by adding the reagents at the soil surface (and then mixing in). However,
adding the reagents at the surface significantly limits the effective treatment depth
that can be achieved.

Reagent addition through the mixing tool provides improved delivery distribution and
therefore improved quality control by comparison. Depending on the equipment and
batch plant make-up, both dry and wet reagent addition are possible, with the former
being advantageous on projects with very high moisture content soils. Wet reagent
delivery is however, better for an even vertical reagent distribution within each S/S
column. Bench (Section 8.3) and pilot scale (Section 8.4) treatability tests are
critical for determining the appropriate reagent formula and field application methods.

5.1.4 Treatment plan


A site-specific and detailed treatment plan (otherwise known as a work-plan) is
produced by the contractor, prior to starting the remedial operation/treatment.

The treatment plan should include the specific equipment to be used, staffing,
proposed work schedule, reagent addition mixing and dosage rates and the plan for
sample collection/curing/testing, site safety and reporting requirements.

For in-situ auger mixing, a critical component of the plan is the precise layout and
planned depth of each column, so as to achieve the desired area of treatment with
154
overlapping columns. By employing overlapping columns, complete treatment is
achieved without leaving any untreated void spaces.

The degree of overlap will vary depending on project-specific needs. Overlapping


columns will cause some portions of the soil to be mixed and treated two or possibly
three times. Increased overlapping will provide greater assurance that all the soil is
being treated, but in a slower production rate and at a higher treatment cost.

Figure 5.10 illustrates a portion of a typical column layout showing overlapping of


adjacent columns.

155
Table 5.1: Examples of in-situ auger treatments including auger diameter and depth treated
Auger Volume
Constructor Date Max. Depth,
Site Name Diameter, Ft Treated, cy Considerations
Completed Ft (m)
(m) (m3)
Confidential Geo- Fall 2011 10 (3) 41 (12.5) 58,000 Dense glacial till with
Solutions Spring 2012 (44,340) cobbles, high torque drill
Inner Slip Site Remediation Geo- Fall 2011 3 and 8 (0.9 32 (9.7) 6,500 Loose dredge sediments,
Solutions and 2.4) (4,970) tight access and poor
subgrade
SAR Levee Repair Geo- Fall 2010 9 (2.7) 54 (16.5) 5,500 High torque drill rig, dense
Solutions (4,205) sandy soils
MW-520 Site Remediation Geo- Spring 2010 9 (2.7) 20 (6.1) 15,200 Clayey silts
Solutions (11,620)
Ameren Site Remediation Pilot Geo- Fall - Winter 5 (1.5) 37 (11.3) 500 (382) Dense glacial tills, test
Study Solutions 2010 program (small volume)
Joachim Creek South Alignment Geo- Winter 9 (2.7) 40 (12.2) 2,500 High plasticity clay
Bearing Capacity Improvement Solutions Spring 2010 (1,911)
OMC Plant 2 Site Remediation Geo- Fall 9 (2.7) 25 (7.6) 8,900 Dense sand & gravel
Solutions Winter 2011 (6,805)
Former Municipal Wastewater Geo- Summer 8 (2.4) 26 (7.9) 8,600 Very dense clays
Treatment Lagoon Stabilization Solutions Fall 2012 (6,575)
P&G Site Remediation Geo- Spring 2012 9 (2.7) 30 (9.1) 19,500 Lagoon sediments
Solutions (14,910)
Front and T Street Site Geo- Summer 10 (3) 33 (10) 40,000 High torque drill
Remediation Solutions Fall 2012 (30,580)
Former Hanley Area Site Geo- Spring 2012 5 (1.5) 30 (9.1) 1,400 Tight access, small drill,
Remediation Solutions (1,070) dry mixing
156
Figure 5.10: Typical column layout showing overlapping columns to achieve
100% coverage

For the constructor, over-lapped columns requires careful planning of the sequence
of column construction, as cutting into previously treated and solidified columns can
be difficult, if too much time has passed since initial treatment. A pragmatic approach
involves treating every other column, then returning and cutting the skipped
overlapping columns on the second or third day. Figure 5.11 shows S/S columns that
have been excavated, illustrating that the overlapping of the columns facilitates the
complete treatment of the target interval.

5.1.5 Quality control


The level of quality control available for auger systems is very high in comparison to
other in-situ soil mixing methods. The quality control procedures available vary
based on the equipment being used, the reagent type (liquid or dry), and the
preference of the practitioner.

157
Table 5.2 gives key features of a typical quality control program from an auger-based
S/S mixing application.

Figure 5.11: showing typical excavated overlapping columns

158
Table 5.2: Quality control planning for auger-based S/S mixing
Quality Control Plan
Information Details
Lines of communication
Key personnel & responsibilities
Methods and procedures for
verifying reagent addition at
depth
Column layout showing 100% coverage of the
Project staging
treatment area
Layout procedures GPS, Total Station, Triangulation
Sampling procedures
Non-conformance procedures
Daily monitoring requirements
Information Details
Grout consistency (wet Density, Viscosity, Temperature, pH
applications)
Dimensions (effective treated area & treated
depth)
Column centre-point locations
Treated Columns Unique column identification
Target reagent weight based on effective
treated area & treated depth
Number of mixing strokes
Total volume of grout added via flow meter
Reagent Addition (dry and wet
(wet)
applications)
Weight of reagent via weigh-scales (dry)
Rotary head (RPM)
Mixing energy Lift rate (if applicable)
Grout pressure/flow rate
Sample collection and curing
Information Details
Mold, store, transport, and testing
Completion of QC reports
Information Details
Daily report: site activities, problems, safety
issues, progress map (what has been
Daily Report
completed), total daily volume treated,
cumulative volume
QC report: reagent usage total and per
Reagent Usage treated volume, effective area calculations
(treated volume calculations), start stop time
Quality assurance of operational reporting by an independent engineer

159
5.1.6 Operational Issues
Equipment
The essential maintenance of equipment is an important part of the management of
any construction project, including S/S by in-situ soil mixing. Production can be
adversely affected by equipment that is poorly maintained and where wearable parts
are not regularly inspected and kept in an operational condition. Both the grout and
soils being treated are abrasive, and can have adverse effects on exposed portions
of the equipment, due to wear and chemical degradation.

The compatibility between the equipment being used and exposed to grout and site-
based contamination should be reviewed and a mitigation strategy implemented
where necessary.

Obstructions
Auger-based mixing equipment is sensitive to the presence of large sub-surface or
overhead obstructions. Given the deeper soil treatment depth-limitations of auger-
based systems, some obstructions can be too deep and too costly to effectively
remove.

Shallow obstructions are more easily dealt with when the treated soil is in a liquid
state, such as immediately after mixing. Site constraints may limit the removal of
obstructions during the mixing operation, and so the best approach might be to
stabilize all the soils around the obstruction and then remove the obstruction (from
between stabilized soils).

Spoil
In general, 15% to 30% of the volume of the treated soil becomes spoil, sometimes
called swell, slop or float. This spoil is formed above ground, as treated material,
which accumulates due to the mixing process and addition of reagents.

Above the water table some of the spoil becomes subsumed into the S/S column,
filling the pore spaces between soil particles. However, below the water table spoil is
displaced upwards to the surface of the column.

The spoil requires removal and/or disposal where possible. It generally has a high
slump value, being composed of soil, liquid grout, and groundwater. Spoils can be
moved and channelled in their liquid state or allowed to take an initial set, before
selective removal and transport/placement for disposal.
160
5.1.7 Summary of limitations, advantages, and disadvantages
In comparison to the other types of in-situ soil mixing, auger-mixing systems have
few limitations. All methods of in-situ soil mixing are heavily influenced by the
presence of sub-surface obstructions, and auger mixing of soil is no different.

Occasionally, a skilled operator can navigate the auger around small obstructions,
but in general, unforeseen sub-surface obstructions stop a soil-mixing project in its
tracks. Buried utility lines must be located and cleared if they are in the area to be
treated.

A rare exception to this is shown in Figure 5.12, which shows a 10 ft (3.1 m)


diameter in-situ auger treating soil immediately under an active fibre optic cable. The
fibre optic cables are located within the PVC pipes (just to the right of the Kelly bar)
and were temporarily exposed and supported while the auger bit was located
underneath the lines and then into the soil below.

Figure 5.13 is a video showing the mixing taking place in the location of the fibre
optic cable mentioned above.

Figure 5.12: In-situ treatment of soil under a live fibre optic line

161
Figure 5.13: In-situ mixing under fibre optic line

Auger mixing can be difficult in extremely dense soils or very fat clays (liquid limit
greater than 50) as dense soils tend to cause accentuated wear of the mixing
equipment components and its difficult to achieve a consistent mix in fat clays.

The advantages and disadvantages of auger mixing are summarised in Table 5.3.

5.1.8 Costs
Typical costs for stabilization using the auger mixing system are indicated below.
However, it should be noted that the costs involved are very sensitive to the depth of
treatment, the types of soils being treated, obstructions to mixing and the chemical
reagents being used. In addition, the specifics of performance sampling
requirements may also be important. As a rule of thumb, current prices for
mobilisation are $75,000 to $250,000, whereas the application of treatment is in the
range $30 to $60 / yd3 (0.7 m3) + reagent costs.
Table 5.3: Advantages and disadvantages of auger mixing
Advantages Disadvantages

High production rate/reduced schedule Sensitive to obstructions


time Requires specialty expertise and
Possible in difficult drilling conditions, equipment
stiff clays / dense sands Less efficient than other types of in-situ
Deeper depth capabilities or ex-situ mixing for shallow depths (<
Treat below water table without 5 ft)
dewatering
162
5.2 Injection tillers and rotary drum mixers
Injection tillers and rotary drum mixers are suitable for the in-situ mixing of a wide
range of soil types to depths up to about 12 ft (3.7 m), although deeper mixing is
sometimes achievable in certain sludge-like materials.

For the purpose of clarity, injection tillers and rotary drum mixers are considered to
be an attachment to a standard excavator and can also be referred to as a Backhoe
Operated Soil Stabilizer, or BOSS unit. Quality control is limited in comparison to
auger mixing, but is better than that achieved with bucket mixing.

5.2.1 Equipment
Mixing equipment generally consists of some form of rotating mixing head through
which reagent is injected either dry, or more commonly, as grout slurry and mixed in
place at the target depth. In addition a batch plant is required to prepare the reagent
slurry according to the formula developed during bench scale treatability tests
(Section 8.3) and refined during the field pilot test (Section 8.4). Ancillary equipment
may include pumps, hoses, support excavators and/or front loaders, a dozer, and
survey equipment.

Mixing head types


In-situ soil mixing with injection tillers and rotary drum mixers is typically
accomplished using attachments to standard construction equipment, i.e.
excavators, dozers, and front-end loaders. Excavator arm attachments have the
greatest depth capabilities and are often referred to as BOSS systems for short.
BOSS systems replace the digging bucket of an excavator assembly with a mixing
arm that has a rotary drum mixer at the end. Figures 5.14 and 5.15 show
commercially available BOSS systems used by Geo-Solutions.

With a BOSS system, the mixing head(s) is powered by the hydraulic system of the
host machine, or a separate hydraulic power pack can be mounted on the host
machine. Reagents may be pumped through the mixing arm or added at the surface.

Most of the BOSS units have the capability of pumping reagents through piping
which discharges just above the mixing head and this is of critical importance if
mixing to depths of more than 2-3 ft (0.6-0.9 m) is required.

163
If reagents are applied to the surface, it is very difficult to mix them uniformly to a
depth of more than 2-3 ft (0.6-0.9 m). Thus, the BOSS systems that inject the
reagents at the point of mixing have the capability to achieve thorough mixing to
depths of 12-15 ft (3.6 to 4.6 m), sometimes more, depending on the length of the
mixing arm and how this arm is articulated.

The BOSS-type unit works well in sludges and soft soil applications. If soils are stiff
or of a clay type material, the mixing heads require a much higher torque, which is
only available on a few of the currently available BOSS models.

Figure 5.14: Backhoe (excavator) Operated Soil Stabilizer (BOSS)

The BOSS units, which are available in a variety of forms, have been available to the
remediation industry for over 20 years. Commonly available commercial systems
include the Lang and the Allu Mixer.

The Lang mixer (langtool.com) has frequently been used in the USA and is available
in several models (e.g. Figure 5.16) with somewhat different depth limitations
depending on the length and articulation of the arm.

164
Figure 5.15: Backhoe (excavator) operated soil stabilizer

Figure 5.16 Lang mixer- excavator, arm, and mixing head as one unit

165
Reagent is injected just above the mixing head, which is sealed and custom fitted to
the excavator body, in one complete unit. This facilitates mixing in wet soil.

The Allu mixer (allu.net) has been less available in the USA. The Allu mixer is
provided as an attachment to be placed on a standard excavator body. Several
models with varying depth capability are available. However in the model recently
used on a coal tar site in Florida, it was observed that the top of the Allu attachment
was not sealed and thus could not be immersed in the treated soil. Figure 5.17
illustrates one model of the Allu mixer.

Batch plant equipment


Batch plants consisting of silos, pumps, and mixing tanks are critical to the success
of most in-situ soil mixing projects, namely those relying on reagent delivery through
ports at the mixing head such as the BOSS systems.

Batch plant configurations vary widely, due to plant makeup and configuration
and the practitioners preference. The type and quantity of reagents, as well
as the pumping distance and site constraints are also important. Where the
reagent is being added as a grout, the batch plant can be extensive. An
example of a large soil-mixing grout production plant is shown in Figure 5.18.
When the reagent is being delivered as a dry powder, pressurised storage
tanks and pneumatic conveyance pumps are required. An example of a self-
propelled dry pneumatic hopper is provided in Figure 5.9.

Ancillary equipment
In addition to the batch plant, a variety of support equipment is necessary for the
successful completion of a soil-mixing project using the BOSS system. Supporting
ancillary equipment may include excavators, dozers, loaders, forklifts, man-lifts,
hoses, pumps and surveying apparatus.

5.2.2 Staffing requirements


Typically a rotary drum mixing or injection tilling remedial project requires a
supervisor, mixing apparatus operator, mixing apparatus support labourer, batch
plant operator, batch plant support labourer, and QC/engineering staff. Labour

166
requirements vary from project to project depending on ancillary work and batch
plant configurations.

5.2.3 Treatment metrics and considerations


Some treatment metrics and considerations are briefly discussed below. However, in
practice these are very site- and equipment-specific, and the reader should consult
with an experienced practitioner regarding the application to any specific site.

Treatment Depth
As previously discussed, rotary drum mixers and BOSS systems are typically limited
to depths shallower than about 12-15 ft (3.69-4.62 m). Applications of this technology
to depths 15 ft (4.62 m) or deeper are possible, but mixing quality and quality control
become limited at increased depth. Note however that equipment designs and
capabilities are evolving and newer equipment may achieve good mixing at greater
depths. Practitioners should prepare and submit digging charts to illustrate the full
extent of their equipments capabilities, i.e. maximum treatment depth in relation to
the machine body.

Figure 5.17: Allu mixer head (in red) attached to a standard excavator

167
Figure 5.18: A batching plant for a large S/S project

Due to the limited power and penetration capacity of these mixers at greater depths,
auxiliary equipment may be necessary to pre-excavate or loosen the soils prior to
treatment. In these cases, BOSS treatment may not be considered purely as an in-
situ treatment.

Anticipated production rates


BOSS systems are capable of treating between 150-600 yd3 (115-459 m3) of soil or
sludge per working day (assumed 8-hour shift). The production rate is highly
dependent upon the site soils/wastes, formula addition rate, depth of treatment, and
operator skill. The production rate will significantly slow for treatment depths over 10
ft (3.07 m).

Reagent addition methods


Reagent addition may be delivered by pumping through the mixing tool or by adding
the reagents at the surface. Reagent addition through the tool provides improved
mixing and quality control in comparison to reagent addition at the surface.
Depending on the equipment and batch plant configuration, both dry and wet reagent
addition are possible options. Dry reagent addition can be advantageous on projects
with high moisture content soils.

168
Pre-construction investigations
Prior to commencement of field operations, it is important to thoroughly investigate
and delineate the treatment area and depth of treatment. Bench scale studies
(Section 8.3) are usually employed to determine efficient (or worst case) reagent
application rates and the mode of reagent delivery (grout or dry powder). In addition,
bench scale studies can be used to estimate spoil volume. Generally pilot-scale field
tests are conducted to refine the reagent dosage and the planning for execution of
the desired treatment option at full-scale.

5.2.4 Quality control


The level of quality control for BOSS systems is much less than that available on the
auger-based soil mixing systems. The quality control available for BOSS systems
varies based on the equipment used and the method of reagent delivery (wet or dry).

The components typical of a quality control program on a BOSS application are


given in Table 5.4.

For all in-situ S/S treatments, performance samples are collected and cured in a
similar manner to that described in Sections 6.2 and 6.3.

The correct care and attention should be given to sample storage and transportation
of quality control samples, which should be stored in a temperature-controlled
environment. During storage, samples should not be subjected to movement or
vibration, particularly during the initial 24 to 72 hours when undergoing initial set.
Samples should not be transported until they have initially set and achieved a
reasonable strength, usually within the period 3 to 7 days. When being shipped by
courier, samples should be properly packed to minimise movement and damage
during transportation.

169
Table 5.4: Quality control planning for BOSS-based S/S mixing
Quality Control Plan
Information Details
Lines of communication
Key personnel &
responsibilities
Lines of communication
Methods and procedures for If available, GPS can be used to assist the operator in
verifying reagent addition at verifying complete mixing of the entire treatment block
depth
Grid map or daily treatment area.
Development of a grid to determine reagent application
Project staging zones. Each zone should be sized for a proportion of the
treatment area that can be completed in a few hours
(certainly less than 1 day).
Layout procedures GPS, Total Station, Triangulation
Sampling procedures
Non-conformance
procedures
Daily monitoring requirements
Information Details
Grout consistency (wet Density, Viscosity, Temperature, pH
applications)
Dimensions (length, width, depth)
Location
Treated panel or volume per
Unique identification
stroke
Target reagent weight
Number of mixing strokes
Reagent Addition (dry and Total volume of grout added via flow meter (wet)
wet applications) Weight of reagent via weigh-scales (dry)
Rotary drum (RPM)
Mixing energy Lift rate (if applicable)
Grout pressure/flow rate
Sample collection and
curing
Information Details
Mold, store, transport, and
testing
Completion of QC reports
Information Details
Daily reporting of site activities, problems, safety issues,
Daily Reporting progress map (what has been completed), total and
cumulative volume treated
QC reporting of reagent usage (total/per treated volume,
Reagent Usage effective area calculations (treated volume calculations),
start stop times etc.
Quality assurance reporting by Independent engineer

170
5.2.5 Operational issues
Equipment
Equipment maintenance procedures are an important part of every construction
project, but can be especially important on in-situ soil mixing projects. Rotary drum
mixers have a limited mixing depth due to the equipment configuration, being
constrained by the length of the mixing arm.

Mixing at extended depth requires that the entire mixing arm, machine joint, and
machine boom are beneath the soil/sludge surface, and the grout and soil
undergoing mixing can have adverse effects on mechanical joints leading to
excessive equipment downtime and additional maintenance requirements.

Obstructions
The configuration of mixing arms allows them to more easily move around
obstructions than an auger-based unit, but they are still limited by the presence of
sub-surface and overhead obstructions. Given the shallow soil treatment limitations,
most obstructions in these applications can be easily removed using a support
excavator.

Obstructions are more easily removed when the treated soil is in a liquid state
immediately after mixing. Site constraints may, however, limit removal of
obstructions during the mixing operation. The best approach is then to stabilize all
the soil around the known obstruction and then remove this by excavation between
stabilized soil-units.

Spoil
In general, 15% to 30% of the treated soil volume becomes spoil due to bulking by
the addition of reagents and the mixing process. The spoil materials require handling
and/or disposal by channelling whilst in their liquid state or after an initial set, when
they can be treated and disposed of as a soil.

5.2.6 Advantages and disadvantages of rotary tillers and injection drum


mixers

The advantages of rotary tillers and injection drum mixers are listed in Table 5.5.

171
Table 5.5: Advantages and disadvantages of rotary tillers and drum mixers
Advantages Disadvantages

Lower cost than auger mixing Limited to a maximum depth of about


High production rate 15 ft (4.62 m) below the working
Reduced schedule surface
Can readily utilise reagent in a wet Applicable only in sludge or soft soils
(grout) or dry state Reduced level of quality control in
Reagent injection can be applied just comparison with auger mixing
above the mixing drum(s) Obstructions require removal for
complete mixing

5.2.7 Costs
The costs of treatment with the BOSS system are very sensitive to the depth of
treatment desired, the soils being mixed and the reagents being used. As a guide,
current prices for BOSS mobilisation are $50,000 to $150,000, whereas the
application of treatment is in the range $15 to $30 / yd 3 (0.7 m3), excluding reagent
costs.

5.3 In-situ bucket mixing


In-situ bucket mixing (excavator mixing) refers to using a standard excavator that
may have an extended reach, and an excavator bucket to mix reagents into the soil
in place.

Reagents may be added dry or as a slurry, but the effective depth for mixing is
however quite limited compared to other options. This approach to mixing is the
simplest form (in terms of equipment) of in-situ S/S treatment and has been
successfully used for decades. A video clip showing in-situ bucket mixing can be
viewed at http://youtu.be/bUFg2siBXd4

5.3.1 Equipment
Bucket mixing is generally carried out with excavators and standard excavation
buckets or specialty buckets (see Figure 5.19) designed to facilitate high slump soil
mixing. Shallow (1-2 ft or 0.3-0.6 m) mixing applications may be completed with bull-
dozers or front-loaders, but this application is uncommon and is limited.

172
Although the equipment used is universally available, it should not be assumed that
anyone with an excavator can successfully accomplish in-situ bucket treatment. The
mixing techniques require experience and the quality control needs are rigorous.

Figure 5.19: Example of bucket mixing

Ancillary Equipment - A variety of support equipment is necessary for the


successful completion of a soil-mixing project utilising bucket mixing. Support
equipment may include excavators, dozers, loaders, forklifts, man-lifts, etc. If
reagents are added as slurry, then a batch plant like that used for in-situ auger or
BOSS mixing will be required. Whether reagents are added dry or as slurry,
equipment is required to accurately measure reagent addition per unit of soil.

5.3.2 Staffing requirements


Typically a soil mixing project completed using bucket mixing requires a supervisor,
mixing excavator operator, mixing excavator support labour, and QC/engineering
staff. Labour requirements vary from project to project depending on ancillary work
and reagent addition procedures.

173
5.3.3 Treatment metrics and considerations

Treatment depth
Bucket mixing is limited to treatment depths less than 8 ft (2.5 m), as the uniform
mixing of surface-added reagents beyond this then becomes progressively more
difficult to achieve. However in some cases, bucket mixing may be extended to
about 15-20 ft (4.6-6.2 m) if specific procedures are carefully followed:

The first 5-6 ft (1.5-1.8 m) are initially mixed uniformly with reagent slurry. It is
necessary that the treated soil have a high slump. The excavator then carefully
removes several buckets of soil from below the previously mixed and slurried soil,
depositing the fresh soil above that previously mixed, and then proceeds to uniformly
mix this soil whilst adding fresh reagent. The void created by excavating below the
previously mixed soil is immediately filled by the mixed soil-slurry. This process is
repeated as necessary. Due to restrictions on excavator reach, often a portion of the
targeted cell is mixed to full depth, then the excavator re-positioned to mix another
vertical slice.

This technique will not work with all soils/sludges and requires constant quality
control monitoring to assure that the treated soils are homogenously mixed with
reagent. Quality control is already challenging with bucket mixing and even more so
at greater depth. Figure 5.20 depicts successful bucket mixing of coal tar
contaminated soil to a depth of about 15 ft (4.6 m). Note that the picture shows
clumps of deeper soil excavated to the surface, but not yet mixed so as to be
homogenous. Part of the quality control process is to frequently extract an excavator
bucket of mixed material from various depths, and visually check for clumps of
unmixed soil.

Anticipated production rates


Its possible to treat 200-700 yd3 (153-536 m3) of soil per working day using in-situ
bucket mixing (assumed 8-hour shift). However this is very site-specific depending
on soil type, equipment, depth, and operator skill. Bucket mixing can often break up
and treat clays and other hard soils that are difficult to treat with BOSS systems.

Reagent addition methods


Pumping through the mixing head is not possible in soil mixing completed using
bucket mixing. Reagents are typically spread over the surface of the mixing area
and mixed into the soils as the bucket mixes the soils. Reagents may be added dry
or as a slurry.
174
Spoil
As with other in-situ treatment methods, a significant soil bulking may take place and
a large amount of spoil created. The amount of spoil will vary to 15% or more,
depending on the reagent type and dosage, and whether the untreated soil/waste
has air-filled pore space or is saturated. The spoil will have the same characteristics
as the rest of the treated material.

Pre-construction investigations
Prior to starting the remedial operation, it is important to thoroughly delineate the
treatment area and the depth of treatment. Bench-scale studies are used to
determine most effective reagent type and its application. The bench-scale studies
can also be utilised to estimate spoil volume.

5.3.4 Quality control


The level of quality control required for bucket mixing is considerably less than that
available in auger or BOSS applications. The components typical of a quality control
program on a bucket mixing application are given in Table 5.6. As mentioned, it is
important to inspect material from various points and depths in the treatment cell to
assure that no clumps of untreated soil remain. The maximum acceptable size of a
clump of untreated soil should be declared in the S/S specifications. As an example,
a 4 in (10 cm) size has been specified on several projects.

Figure 5.20: Bucket mixing of coal tar soils to depth of 15 ft (5 m)


For all in-situ S/S treatments, performance samples are collected and cured in a
similar manner to that described in Sections 6.2 and 6.3. The correct care and
attention should be given to storage and transportation of quality control samples,
which should be stored in a temperature-controlled environment.

175
Table 5.6: Quality control planning for bucket-based S/S mixing
Quality Control Plan
Information Details
Lines of communication
Key personnel & responsibilities
Grid map or daily treatment area.
Development of a grid to determine reagent
application zones. Each zone should be
Project staging
sized for a proportion of the treatment area
that can be completed in a few hours
(certainly less than 1 days production).
Layout procedures GPS, Total Station, Triangulation
Sampling procedures
Non-conformance procedures
Daily Monitoring Requirements
Information Details
Dimensions (length, width, depth)
Location
Treated panel or volume per
Unique identification
stroke
Target reagent weight
Number of mixing strokes
Total volume of grout added via flow meter
Reagent Addition (dry or wet
(wet)
applications)
Weight of reagent via weigh-scales (dry)
Rotary drum (RPM)
Mixing energy Lift rate (if applicable)
Grout pressure/flow rate
Sample Collection and Curing
Information Details
Mold, store, transport, and testing
Completion of QC Reports
Information Details
Daily reporting of site activities, problems,
safety issues, progress map (what has been
Daily reporting
completed), total and cumulative volume
treated
QC reporting of reagent usage (total/per
treated volume, effective area calculations
Reagent Usage
(treated volume calculations), start stop times
etc.
Quality Assurance Reporting by Independent Engineer

176
During storage, samples should not be subjected to movement or vibration,
particularly during the initial 24 to 72 hours after manufacture, when undergoing
initial set. Samples should not be transported until they have initially set and allowed
to cure to a reasonable strength, usually within the period 3 to 7 days. When being
shipped by courier, samples should be properly packed to minimise movement and
damage during transportation.

5.3.5 Operational Issues


Equipment
The maintenance and operational requirements are similar to a normal excavation
operation, except that the excavator arm and bucket are subjected to continuous
contact with stabilization agents and or contaminated groundwater. This contact can
result in an increase in equipment maintenance and repair.

Obstructions
Small obstructions can be removed using the mixing excavator during the soil
mixing. Larger obstructions may require a breaker or other means of sizing the
obstructions prior to removal.

Spoil
In general, about 15% to 30% of the treated volume becomes spoil material requiring
handling and disposal. Spoil generally displays a high slump as it is composed of a
soil/grout/groundwater mixture. Spoil can be moved by channelling while liquid or
allowed to take an initial set, after which it can be handled and disposed in a similar
way to soil.

177
6.0 Quality assurance and quality control

6.1 Quality assurance and quality control during S/S


A great deal of general information on project quality assurance and quality control
can be found on the USEPA website using the following link:
http://www.epa.gov/quality/qapps.html. The discussion in this section is specific to
the use of S/S technology as commonly practiced. Additional information relevant to
QA/QC can be found in the following sections: sample collection, including frequency
and number of replicates 6.2, sample preparation and curing 6.3, test methods 7.3,
bench-scale treatability testing 8.3, pilot field tests 8.4, and selecting samples for
testing 8.5.

6.1.1 Quality assurance and quality control during S/S


An important part of any remedial action involving S/S is the construction quality
assurance (CQA) and construction quality control (CQC) monitoring activities. The
CQA is the task of the responsible engineer and/or the site owners representative,
whereas the CQC is the primary responsibility of the construction contractor with
oversight by the responsible engineer:

CQA is the independent monitoring and verification by the responsible


engineer to verify that the remedial works are meeting the agreed
technical specification and performance objectives.
CQC is the system of measurement and monitoring activities conducted
by the construction contractor to assure that construction will meet the
agreed upon technical specifications.

During an S/S project, CQC involves monitoring and documentation of all aspects of
the S/S operation from preparation of the reagent through mixing of reagent with the
target media (i.e. soil or sediment).

To maintain effective CQC the contractor establishes operational metrics (e.g.


reagent addition/mixing times etc.) for each aspect of the S/S operation to ensure
compliance with the agreed technical specification and performance objectives. The
CQA process reviews the CQC to provide the assurance that the operation conforms
to the technical specification. This process includes sampling and analysis by an
independent laboratory to verify that the performance objectives (e.g. minimum
strength, maximum permeability, and leaching parameters) are being met.

178
6.1.2 The importance of CQA in the S/S design and construction process
Verification of S/S performance involves a comprehensive CQA program that
examines each phase of the S/S project, from bench-scale testing through to the full-
scale field operation.

An effective CQA program will help control costs by highlighting failure to meet the
agreed specification, especially when there is (as there often is) a time-lag between
S/S sample collection and receipt of laboratory data. A key component, therefore, is
the rapid assessment of construction performance criteria that can indicate problems
early on (and which allow for timely field adjustments or corrections), thereby
reducing potential costs and minimising potential risks.

The transition from bench to pilot-scale testing is a critical juncture for assessing the
metrics from CQC and CQA effectiveness. The careful consideration of full-scale
reagent preparation and application is critical in meeting the design objectives and
for managing project costs. For example, during in-situ S/S, important reagent
properties, such as accurate reagent densities, are critical for defining the
appropriate water:reagent ratios, binder delivery rate and required mixing effort.

The effectiveness of in-situ mixing is dependent upon factors that include the mode
of application, e.g. auger or bucket mixing, and operational parameters, such as the
number of auger passes through a column, the rate of auger rotation and duration
of mixing required. For ex-situ operations, verification of accurate reagent addition
rates calibrated to the unit weight of soil being treated and constant
monitoring/adjustment of water addition rates to adjust for varying field moisture of
soil being treated are necessary to achieve desired product properties.

6.1.3 Development of an effective CQA program


In general, a successful S/S operation is measured by its ability to meet the agreed
technical specification. Furthermore, a consistent performance data set, received
from both field and the laboratory testing showing achievement of treatment goals,
will provide a high level of confidence in the work carried out, and facilitate public
and regulatory acceptance of the treatment.

The monitoring of the complex performance factors involved during S/S is dependent
on full understanding of the technology and the issues concerning field
implementation. A well-designed and successful CQA program will ensure that the
project meets its remedial targets, is timely and within budget.

179
The first step in designing a CQA program involves establishing clear data collection
objectives for bench-scale testing to elucidate the key issues faced during full-scale
implementation of S/S. This is particularly important as the bench-scale testing of
potential binders for S/S often serves as the basis for contractor bidding and
payment. Thus, the CQA program must encompass the entire S/S design process
from pre-design data collection through full-scale application. Every step of the
process must be scrutinised to ensure the work being carried out meets the agreed
remedial performance targets.

Pre-bench-scale data collection objectives involve a careful assessment of the sub-


surface soil conditions. The soil type encountered, its variability, moisture content
and density data will all impact the amount (and type) of reagent(s) to be used during
bench-scale testing and during the full field-scale operation. An inadequate
assessment of field conditions can therefore have significant impacts on
performance and cost if:

Field soil densities are under or overestimated at bench-scale, then reagent


application at full-scale may be too low or high. Low dosages may lead to S/S
performance failure, whereas high dosages may lead to unnecessarily higher
project costs and potential leaching failures. As the field density of the
treated S/S waste form is determined from the design proportions derived
from bench-scale studies, this criterion is directly linked to contractor
payment. Thus, great care is needed when obtaining representative field
samples for subsequent evaluation.
Dry densities of the soils and reagents are not accurate since they are used
as a basis for estimating the amount of binder addition during in-situ
application. As this is also related to moisture content, it is important that
variations in moisture are fully determined to ensure soil dry density is not
under or overestimated.
The variation in untreated soil moisture content is not accurately understood
since this will influence the effective water/binder ratios used during full-scale
mixing.
The percentage and variability of clay in the soil is not accurately understood
as this will directly affect the amount of reagent used and mixing effort
required as well as the volume increase due to treatment. This can negatively
impact projected cost or cause site-based space logistical issues.

180
6.1.4 CQA objectives
The CQA/QC objectives for S/S can be grouped into two principal phases, design
and construction.

Design
The design objectives ensure that the contractor performs S/S in accordance with
the technical specifications and contract drawings. This covers the preparation and
delivery of the reagents to achieve the agreed performance standards, extending
from reagent delivery on-site to the examination of S/S-treated soil. A key aspect of
the verification process involves documenting that the approach at pilot-scale is
brought forward to full-scale and consistently implemented.

Construction
The construction objectives involve verifying the consistent application of the mix
design at full-scale. A robust data set from the laboratory testing of samples will
show both consistency and compliance with the design assumptions established
during bench- and pilot-scale testing.

The CQA/CQC objectives for S/S sampling are interdependent and require a
continuous assessment of S/S performance. This ultimately will provide a high
degree of confidence that performance standards are met, particularly when the data
generated from testing is only normally available 7 to 28 days (or longer) after the
mixing/placement of S/S material in the field.

Monitoring construction objectives during the initial stages of the S/S operations will
be critical pending development of a full-scale geotechnical data set indicating
consistent and acceptable performance. It is desirable during bench and field pilot
testing to obtain data on how critical parameters such as strength, permeability, and
leaching, developed over time. Then by preparing a few additional performance
molds during remediation, one can compare results at shorter time periods to assure
that desired parameters are developing as expected so as to achieve proper results
at the designated time period.

6.1.5 Roles and responsibilities in the CQA process


The successful implementation of CQA/CQC and the maintenance of the agreed
design parameters is dependent upon key staff appointments:

181
Engineer or owners representative
The CQA program is the responsibility of the site engineer responsible for the S/S
remedial design and/or the designated site-owners representative and may involve
the following:

Reviewing the contractors CQC program and the quality procedures


identified for each aspect of the S/S remedial process
Coordinating the collection and testing of CQA samples, and their transfer to
off-site laboratories
Preparing treated S/S sample molds for onsite curing
Reviewing the geotechnical and chemical performance data for
completeness and ensuring the results obtained meet the required CQA
performance criteria
Reviewing the geotechnical and/or chemical laboratory analytical data
Documenting the construction and CQC/CQA monitoring/testing process
Preparing CQA sampling reports (for the contractor)
Evaluating CQA/CQC testing performed and recording any relevant
observations
Reviewing the results of CQA/CQC laboratory testing
Evaluating the testing results obtained that do not meet the agreed
performance objectives

Contractor
The contractor is responsible for all aspects of CQC in strict accordance with the
technical specifications, including the following:

Coordinating activities with the site engineer and CQA team leader to meet
the agreed schedule and requirements of CQA testing
Implementing design changes through engineer-approved modifications
(based on the assessment of the CQA program)
Coordinating site surveys and material testing requirements
Providing recommendations to and/or consulting with the lead project
engineer

Geotechnical laboratories
A key part of the CQA program involves third-party laboratory testing by a laboratory
that is independent from the contractors CQC program. Independent testing is used
to verify that S/S is being conducted in accordance with the agreed technical
specification and the S/S products comply with the performance objectives.

182
Geotechnical testing is conducted during full-scale operation, and two geotechnical
laboratories, independent of the contractor`s laboratory, are often selected to provide
CQA. One laboratory can serve as the primary geotechnical laboratory for routine
analyses, whereas the second undertakes additional CQA by analysing replicate-
samples (e.g. 5% of samples) being supplied to the primary laboratory. Generally not
all analyses by the contractor`s laboratory are repeated by the CQA laboratory.
Often replicates for about 10% of the samples analysed by the contractor`s
laboratory are selected for analysis by the CQA laboratory for the purpose of
validating analyses by the contractor`s laboratory.

6.1.6 General categories for the CQA process

Batch plant operations


CQA procedures for the batch plant operations should address the following:

Reagent delivery: this involves inspection of receipts for dry reagents delivered to
the batching plant to verify that the correct reagents are used in the agreed mix
designs.

Batch plant calibration: including CQA calibration data (from the contractor) to
verify precision and accuracy of methods for calculating reagent densities.

Reagent densities: are required for grout-based binders, and are calculated from
the water:reagent (or binder) ratio required for pumping. Reagent densities require
independent examination as part of the CQA program, via sampling of the production
grout from the batch plant and/or by verifying the mix weights/volumes prior to
delivery to the S/S mixing equipment. Batch plants often have their mix tank placed
directly on a load scale allowing for reagent and water addition based directly on
weight. With this approach, measuring grout densities is not necessary. For ex-situ
mixing, reagents are added on a weight basis to the untreated soil. With a pug-mill
for example, the untreated soil is added via a weigh belt and reagents are added via
weigh belts or through use of calibrated screw feed from silos.

Water to reagent/solids ratios (W/S): are directly related to reagent density and
can be independently evaluated as part of the grout preparation process. During in-
situ application, the W/S is kept as low as possible to reduce volume increase (the
S/S swell), but can be adjusted during application in response to changes in ground

183
or other conditions encountered during mixing. For ex-situ mixing, water is added to
achieve the desired moisture content for placement/compaction.

Grout pumping rates: can be compared with the results from the pilot-scale tests to
verify that the rate of grout delivery is correct and of consistent quality.

S/S mixing operations


During in-situ S/S, the contractor is required to provide CQC documentation covering
the field application of the binder system. The CQC documentation required may
include the following:
Grout injection pressure and rate of delivery
Auger rotation speed and pressure
The vertical speed of auger advancement
Mixing tool type(s) and/or diameter(s)
The number of vertical passes through the entire treated soil column
The duration of mixing
The depth of mixing
The controls over both horizontal and vertical (auger) alignment and S/S
column overlap

For ex-situ mixing with a pug-mill, CQC documentation may be required for:
Logs of hourly (sometimes quarter hourly) and daily feed rate of incoming
untreated soil
Daily calibration logs for reagent feed rate
Transit time for soil in the pug-mill
Logs of water addition rates
Horizontal and vertical survey to precisely locate the placement of each days
treated soil

For other types of in-situ or ex-situ mixing, similar CQC data should be provided. The
objective here is to document the reagent addition(s) and mixing operation(s), and to
provide a precise record of where the treated material is located.

The parameters controlling S/S mixing will be periodically (and independently)


verified for comparison with the contractors CQC data. As part of this evaluation,
discrete freshly treated S/S sample material may be collected from a designated
depth within a given S/S column or specified mixing location.

184
During ex-situ application, samples may be selected spontaneously and at pre-
determined time intervals during the mixing process, and subjected to visual
inspection for key qualitative parameters, including sample colour and homogeneity
(presence of poorly-mixed media) and contamination. These samples can also be
cured on-site, and inspected as for the freshly treated material.

The qualitative data obtained will provide supporting data for the overall
effectiveness of the operation and for any proposed modifications to the agreed
operating parameters, or the CQC program. A detailed discussion of ex-situ mixing
methods is provided in Section 4 while a similar discussion on in-situ mixing
methods is provided in Section 5.

Post S/S surveying and record documentation


The completed S/S columns or cells are typically the responsibility of the contractor
for surveying. The results are provided for CQA review and evaluation to assure that
the columns, or cells, were completed in accordance with the technical specification.
The survey data obtained are used to verify the following:

The number of columns or cells completed


The locations of columns or completed cells
The location/dimensions and overlap of columns or cells
The total volume of soil treated by S/S
The location of columns or cells with different mix designs (if applicable)

The post-S/S survey data may also be used for an analysis of the S/S operation
including the following:

The rate of S/S product production, including unforeseen change to


production rates and potential impacts on the agreed time schedule, and/or
conflicts with other elements of the construction
A comparison with the CQC documentation and the identification of potential
impacts on operational production efficiencies
A comparison with engineering estimates and contractor applications for
payment: to confirm if the S/S undertaken requires revision of project costs/
payments to the contractor

185
Assessment of geotechnical laboratory performance-related data
The geotechnical data established during pilot testing can be used to cross-
reference the initial full-scale data obtained (e.g. for UCS and permeability). At full-
scale the results obtained can be referenced to each mix design used, the soil type
treated and any other major aspect of the S/S operation, and applies to both in-situ
and ex-situ S/S application.

The data trends established can be used to monitor compliance with performance
standards and to establish the average data set for the S/S treatment. In addition to
confirming maintenance of performance standards, these data can also be used to
support the CQA defined construction and design objectives, including:
A comparison of geotechnical testing data with defined operating parameters
to verify that S/S meets the agreed technical specifications and performance
standards required
An evaluation of data trends to indicate any change in subsurface conditions
that may require modification to the mix design
The re-treatment and/or re-testing of S/S material as the forward projection
(e.g. strength development) of performance data indicates a pending failure
to meet agreed standards

The secondary laboratory generated geotechnical testing data can be used to


corroborate the results from the primary laboratory. The comparison of results can
highlight issues (e.g. internal QA/QC protocols or procedures), whereby the following
specific actions can be undertaken:

The QA/QC data from one, or both, laboratories indicates failing treatment
properties, or failure to properly follow testing protocols. Additional testing
can be performed (by the failing laboratory) on archived samples that are
cured for a longer period of time (e.g. 28 days), or by the immediate repeat
testing of replicates samples.
If no laboratory testing errors are identified, then replicate samples should
be tested by both laboratories at the next sampling interval. If this involves
28-day old samples, then both laboratories will perform the test, on the first
available day that both laboratories are able to do so.
If a confirmatory test from the same laboratory fails as a result of this
procedure, then additional testing data will be requested from the primary and
secondary laboratories, and may be used to identify any systematic
laboratory procedural errors.
If confirmatory testing meets the agreed specification, then the failing
sampling event will be considered resulting from a random error in one of the
sample testing results.

186
6.1.7 Standards and decision processes for S/S performance
Performance standards
The development of appropriate site-specific S/S performance standards is complex
and is covered in detail in Section 7.

Decision making framework


In the event of the need to change the specification for S/S, or for any other
eventualities where change is required to the remedial action, a pre-defined decision
making process should be agreed between the engineer and contractor. This is
particularly important when:

Retreatment is required as performance standards are not met


Changes are required to the contractors CQC program, to monitor
operational metrics more effectively
Modification is required to the full-scale S/S operation to meet the
performance standards

A framework/process for decision-making can be established in the form of


conditional if/then statements, designed for implementation at specific times during
testing. This approach provides a structured decision-making process for both the
engineer and the contractor, and an example is provided in Table 6.1 below.

Evaluation of the decision making framework


An evaluation of the decision-making process is required as part of the CQA
process. This will evaluate whether the appropriate corrective actions meet with the
agreed decisionmaking process when specified performance criteria are not being
met. Three general categories for potential S/S performance-related issues arise:

Sampling and analysis: particularly errors during laboratory geotechnical testing


resulting in false negative or false positive test results.

Design: substantively different sub-surface conditions from those encountered


during pilot testing.

187
Operational: operating parameters that are not consistent with those demonstrated
during pilot testing, or not effective for delivery or mixing of binder to meet agreed
performance standards.

Each issue should be reviewed to assess whether the mix design and operational
parameters identified will meet the agreed performance standards during full-scale
implementation of S/S.

Particular attention is required to potential random and/or systemic errors, such as


irregular curing of molded S/S samples, or improperly calibrated measuring
equipment, respectively. Then the full-scale data set obtained can be compared with
the base-line data (established during pilot testing), to enable a review of the
changes to agreed procedures, including:

Table 6.1: Example of a simple decision making framework agreed between the
site engineer and contractor
Outcome Decision
Met at 7 days Considered effective at 28 days
Not met at 7 days Option to re-treat the failed column/section or re-test samples at
14 days
Met at 14 days Considered effective at 28 days
Not met at 14 Option to re-treat the failed column/section or re-test samples at
days 28 days
Not met at 28 Contractor to remove or re-treat failed column/section or, at
days engineers option, allow more time for curing

An evaluation of field sample compliance with SOPs (for sample collection


and preparation) and field inspection of archived S/S molds for compromises
of sample integrity
Submission of replicate S/S samples to the second geotechnical laboratory
for a comparison with the results from the primary geotechnical laboratory
Collection of untreated soil samples in areas where S/S fails to meet the
agreed performance standard, to evaluate if subsurface conditions have
deviated from the baseline assumptions

188
Consideration of the effects of variations in operational parameters on
geotechnical laboratory test results
Evaluation of contractor CQC procedures to determine if the established
inspection and monitoring procedures should be revised to address
deficiencies

Figures 6.1 to 6.3 are example decision-making flow charts to address sampling and
analysis, design and operational issues.

189
Failure to meet performance
criteria due to sample integrity

Sample collection
and preparation
meet specifications?

Yes (i.e, sample


prepared properly No (e.g., inadequate
and appears hydration)
acceptable)

CQA lab in Retest with duplicate or


compliance with compare with Contractor
ASTM methods (e.g., collected sample
review laboratory
QA/QC data)
Assess
operational and
design parameters
Yes No

Evaluate
Retest with alternate
operational and
sample using second lab
design parameters

Retest with Meets


performance
alternate sample
critera?

No Yes

Evaluate Test second set of samples


Operational hydrated for longer number of
design parameters days (e.g, 28 days)

Figure 6.1: Flowchart for sampling and analysis-related decision making

190
Failure to meet performance criteria due to
incorrect mix design

Operational CQC data meets


technical specification (e.g.,
proper reagent densities,
water:reagent ratios)?

No Yes

CQA lab in compliance


Re-treat with ASTM methods?

Yes No

Change in subsurface Refer to sampling and


conditions? analysis decision making
framework

Yes No

Reassess design basis for reagent Assess modifcations to


mix and percentages operational processes (e.g.,
mixing duration and techniques)

Figure 6.2: Flowchart for design-related decision making

191
Failure to meet performance criteria due to batch plant prep
and/or mixing

CQC data for batch plant meets technical


specification (e.g., correct reagent densities,
water:reagent ratios)?

No Yes

CQC data for mixing meets technical


Re-treat specification (e.g., correct delivery,
mixing rate)?

No
Yes

Re-treat Refer to design and sampling


and analysis decision making
frameworks

Figure 6.3: Flowchart for operational issues-related decision making

192
Corrective action
Potential corrective actions can be identified on the basis of an evaluation of
decisions and may include the following modifications:

Design mix(es): change could include increasing and/or decreasing the amount of
reagent and/or reagent ratios (e.g. ratio of Portland cement to slag or bentonite)

S/S operational procedures: these may include increasing the duration of mixing,
the number of vertical auger passes through the column, and adjusting the water to
binder ratio(s).

CQA/CQC procedures: can include modification to the frequency of inspections,


the types of inspection carried out, on data reporting, and ways in which key S/S
parameters are measured by the Contractor.

No Action: a no action result can be considered if a failed S/S column or section


will not interfere with meeting the overall project objectives, and appropriate
corrective measures have been implemented for future S/S operations.

6.2 Sample collection - the performance sampling plan


The performance-sampling plan is critical to the effective execution of an S/S
remedial action and should be carefully prepared and reviewed (and endorsed) by all
stakeholders in the project. It should contain, at a minimum, the procedure for
collecting samples, the frequency of the sampling events, the test methods that will
be employed, the number of sample coupons to be prepared per sampling event,
preparation of the specimens (or coupons) prior to testing including their curing and
storage, and the agreed procedure for handling sample failures.

Quality Assurance is discussed in Section 6.1, whereas performance test methods


are discussed in Section 7.3. Sample preparation and curing are discussed in
Section 6.3, and handling performance sample failure is addressed in Section 6.4.
The frequency of sampling events, the appropriate number of sample coupons to
prepare, and the methods for collecting the bulk performance samples are discussed
in this section.

Performance sample collection during S/S is not simple and requires careful thought
to assure that an adequate number of samples are collected at sufficient frequency
193
to meet the project quality assurance objectives. The precise method of collecting
the samples depends on both the equipment used for treatment, and in the case of
in-situ treatment, the depth of soil mixing. Failure to carefully consider these factors
ahead of time can result in collection of too few performance samples to document
treatment performance or the introduction of bias that jeopardises validation that the
treatment met the S/S project objectives.

6.2.1 Frequency of sampling


The frequency of performance sampling depends on the overall size of the S/S
project, the daily treatment rate, and the variability of soil or COC properties
observed, as there is no single answer! However the following guidance is
suggested:
At least once each day of significant production
Every 500-1000 cubic meters (or cubic yards)
A large enough number of sampling events to be able to demonstrate
statistical significance

Good practice will generally lead to 1 or 2 sampling events during a typical 10-hour
work-day, since generally 500-1000 yd3 (383-765 m3) will have been treated if there
was no significant downtime. During a typical treatment day when 500-1000 yd3
(383-765 m3) are mixed, the authors prefer 2 sampling events, the first scheduled
about two hours into the treatment day and the second about two hours prior to
stopping treatment for the day. This is to help assure that treatment operations were
consistent. It is suggested that the exact collection times be varied to both suit
convenience and to avoid the treatment vendor knowing in advance when they will
occur or in which cell, column, or batch they will occur. It should be noted that
anytime the oversight engineer suspects that the treatment may be questionable
(soil property change, apparent non-homogenous mixing, issue with reagent addition
rate, etc.) an additional sample should be collected.

The frequency of sampling is also influenced by the need to have had an adequate
number of sampling events by the end of the project to be able to draw a statistically
valid conclusion regarding the overall effectiveness of the treated material properties.
This is not an issue for large projects, say over 100,000 yd 3 (765,000 m3), but may
become important for small projects.

All stakeholders should agree ahead of time regarding what would be a minimum
number of samples to conclude that the project was successful. For small projects,
194
the authors have suggested 20 sampling events as a default minimum. By way of
reference, an unnamed creosote-contaminated site using a pug-mill for treatment
recorded 131 performance-sampling events during S/S of 81,000 tons (74,000
tonnes) of soil; whereas a former MGP site contaminated with coal tar, employing in-
situ augers for the treatment of 136,000 yd 3 (104,040 m3), carried out 211 sampling
events.

6.2.2 Number of samples per sampling event


The rule of thumb is to reasonably determine the minimum number of samples
needed per sampling event, and then increase these by 50-100%. The minimum
number of samples should be adequate to provide individual samples for each
performance test to be carried out, such as for strength, permeability, and leaching.

Multiple samples will also be needed for testing at different cure times as for the
determination of strength by a standard method (typically at 7, 14, and 28 days of
age). For the determination of permeability testing this commonly takes place at 14
and 28 days, to allow for maturation.

Leach testing is commonly carried out at 7, 14, and 28 days of age to document
improving retention of contaminants within the S/S waste form matrix. It has often
been the practice to discontinue testing of longer cured samples once the required
specification has been met. However it should not be assumed that early
achievement of design properties will be achieved and it is important to ensure that
an adequate number of samples for the full extended testing programme are
collected, adequately stored and available, including reserve samples (which should
always be collected as insurance against unforeseen events).

Additional samples are inexpensive to collect and prepare initially, but there is no
option to go back to obtain them at a later date. Thus it is recommended that an
additional 50-100% more samples (reserve samples) be collected than the minimum
(to accommodate planned testing).

Occasionally a target property may not have been met after the full cure period, yet
the data obtained indicates that the material is maturing slower than expected, but is
likely to meet the required specification at a later date. If the property is met at the
longer cure time, then the treated material in place is likely just fine. In such a case it
may make sense to continue curing from e.g. 28 to 56 days to enable the target
criteria to be met. This pragmatic approach to maturation requires additional samples
from the reserve to accommodate this extended testing.

195
It is not unusual for some performance samples to produce results that are outliers.
The cause may be apparent, such as when a sample develops a crack and records
a high permeability, and sometimes the cause is unknown. It is then desirable to
repeat the test on two replicate reserve samples to either confirm the result or to
show that it was anomalous and not representative of the S/S material being treated.

An example of a performance-sampling plan might look like the following. However


each case is different and should be agreed upon by all stakeholders in advance:

Strength testing at 7,14, 28 days, 3 samples plus 3 reserve = 6 samples


Permeability testing at 14, 28 days, 2 samples plus 2 reserve = 4 samples
Leach testing at 7,14, 28 days, 3 samples plus 3 reserve = 6 samples
With a total of 16 samples per sampling event, two events/day = 32
samples/day

Some economy regarding the number of samples and sample preparation time can
be realised by utilising the same sample mold (container) size for multiple tests.
Often the same size/shape sample can be used for strength, leaching, and
sometimes permeability testing. This then allows for any individual reserve sample to
be a reserve sample for any of the criteria tests. In such a case, 50% reserve
samples should be quite adequate.

A large number of samples can be produced very quickly during S/S and appropriate
facilities are required on site for their curing and storage (see Section 6.3 and 6.4)
until they can be shipped to the testing laboratory. It is important that no sample be
excessively handled or shipped until it has taken a hard set. This will prevent
damage to samples when they are immature. The result of damage to the samples
at this stage may mean that the sample fails to meet the design criteria even though
the in-place treated material is fine (see Section 6.4 for a discussion on variation
and failures). In most cases, setting will occur in 3 to 5 days, but in the authors
experience, there have been exceptions that required a much longer cure time to
achieve a hard set, yet the samples met the treatment criteria. Once treatment
criteria have been met, the samples may be disposed, though it may be wise to hold
them until the end of the project if feasible.

196
6.2.3 Sample collection methods
There is no one best method for collecting performance samples during full-scale
treatment by S/S. A number of successful collection methods exist and these share
common key features:

S/S material must be fresh from the treatment process before any field
setting of material commences. Ideally, samples should be collected
immediately after mixing, as from a pug-mill discharge or from the cell or
column as soon as mixing is completed.
Samples must be collected in sufficient bulk for all the anticipated testing
required, plus contingency reserve samples.
S/S material should be collected when the S/S treatment process is operating
normally.
Fresh S/S material should be examined immediately to assure it appears
homogenous.
Staff and equipment are in place to immediately prepare the
samples/coupons/molds as required for testing.

The quantity of fresh S/S material to be collected should be at least 150% of what is
needed to prepare the samples, and preferably 200% to provide for rejected material
and waste (Section 6.3.4). It should be noted that during sample preparation, the
last 20% of the material obtained may have undergone change in moisture content
or particle size due to drying, draining, or during holding and sample preparation.
Generally one or two 5 gallon buckets (19 litre) are used to temporarily hold the bulk
sample while coupons are prepared.

The method for collecting the fresh samples for performance testing depends on the
equipment being used for S/S treatment. With a pug-mill, the sample is usually
obtained at the point of discharge. Some pug-mills employ an open discharge box
and treated material is dumped and removed by front loaders to the location of
disposal or into trucks for onward transport. Pug-mill mixers may also discharge onto
a stacker belt that lifts and discharges the treated material into a truck for transport.
In either case, the bulk performance sample should be carefully collected
immediately after discharge into buckets (approximately 70% full) using a shovel. It
is also acceptable to collect the fresh sample from a dump truck at the disposal area,
providing the time between discharge and collection does not exceed 15 minutes.

If ex-situ mixing by an excavator in a mix pit is being carried out, then a sample can
often collected by the excavator bucket, immediately following mixing. If the mixing is

197
in-situ using an excavator or a rotating tiller (and the depth of mixing is shallow - say
about 5 ft or 1.54 m), the excavator can be used to collect the sample from the
treated cell.

For in-situ mixing below 5 ft, it is preferable to use a sampler designed for collecting
a sample from a specific designated depth. This approach assures that the fresh S/S
sample does originate from a designated depth, and is commonly used when in-situ
treatment uses augers, injection tillers (such as the Lang), or for excavator mixing.
For example, if in-situ auger mixing is used to treat material to a depth of 30 ft (9.2
m), then it is common practice to collect a performance sample from designated
intervals, e.g. 5-10 ft, 15-20 ft, or 25-30 ft (1.5-3.1, 4.6-6.2 and 7.9-9.2 m,
respectively).

For any specific sampling event, the sample should be collected at the completion of
mixing to full depth. Typically one sample is collected with the designated depth
specified just before sampling. Across subsequent sampling events, the collection
depth will be varied randomly, though a specific depth may be selected more often if
it presents a greater concern. Under special circumstances such as at project start-
up or a full-scale field test, samples may be collected from multiple depths to assure
uniform treatment to full-depth. Figure 6.4 shows a hydraulically operated piston tube
sampler being used to sample a freshly mixed auger column. The excavator both
inserts and retrieves the sampler, which in Figure 6.5 is fitted with a hydraulically
controlled gate that is opened at the desired sampling depth.

A number of sample collecting equipment designs have been employed to gather


samples from different depths. However the most successful designs all have
common features, which are:
The sampler is rugged and suitably engineered for sampling at depth in
abrasive soil slurry
A sampling chamber is fitted that is capable of holding a suitable quantity of
sample for testing/evaluation
The sampler can be pushed into, and retrieved from, the S/S material using an
excavator
The sampling chamber remains closed until target depth is reached
At the target depth, a piston or gate can be opened hydraulically or
pneumatically to allow the sample to be collected
Regardless of the method used to obtain the discrete sample, it should be closely
inspected to assure that the mixing appears uniform. If clumps of untreated soil
larger than a predetermined size (generally 2-4 in (5-10 cm) in diameter are

198
observed, additional mixing of the column/waste form may be required prior to
making another attempt to obtain a performance sample.

Figure 6.4: Collection of sample using piston tube inserted by excavator,

Figure 6.5: A sampler employing a hydraulic gate

199
6.3 Sample preparation and curing
Once the required quantity and quality of bulk samples has been collected, it is
necessary to prepare the designated number of individual replicate specimens for
the sampling event.

The size and shape of the specimen molds is often specified by the test method. For
example, a commonly used method to assess strength is: ASTM D 1633 (ASTM),
which requires the use of right cylindrical molds with a length equal to twice the mold
diameter.

Although this provides choice as to actual size used,2 in, 4 in or 6 in (5 cm, 10 cm,
15 cm) diameter molds, the authors advise that molds smaller than 3 in (7.5 cm)
diameter may produce lower strength values, while molds larger than 6 in (15 cm)
diameter are cumbersome to handle, and do not yield significantly better results.
Thus, molds in the range of 3 to 6 in (75-150 mm) are often the best choice, whereas
specimens for permeability or leach testing may, or may not, demand different sizes
of specimen.

6.3.1 Sample preparation


The bulk sample obtained from the S/S waste form or from the mixer will require
careful placement into the designated sample molds for curing. The objective in
loading the sample molds is to try to replicate, as closely as practical, the conditions
of the treated material curing in place in the field.

If the material curing in place in the field is moist but not sloppy (as from a pug-mill),
it is most likely being placed in the field using compaction equipment such as dozers
and perhaps rollers. Thus the performance sample is placed into the molds, in lifts
and compacted using a rod so as to mimic, as closely as feasible, the field
compaction effort, and avoid loose material and voids. If the material in the field is
quite fluid and sloppy (as is often the case for in-situ treatment), then it is placed into
the molds and a rod may need to be gently used to eliminate air bubbles (Figure
6.6). In both cases, it is necessary to screen the material placed into the molds to
eliminate oversize material For example, a 2 in (5 cm) piece of debris in a 3 in
diameter mold will result in a high permeability value from the lab test. However this
same oversize debris in the field S/S waste form will have little effect on permeability
of the monolithic waste form. A common practice is to screen the bulk samples
through a 0.5 in (1.25 cm) screen to remove oversize debris (Figure 6.7). In this
image, note that all molds are carefully labelled, including the top cap, to indicate the
sampling event, location, time and project (Figure 6.8).

200
Capping cylindrical molds is essential to prevent loss of moisture during curing and
to avoid having one end of the cured specimen becoming overly dry and brittle,
leading to erroneous test results.

Figure 6.6: Preparing a 3 in x 6 in (75 x 150 mm) mold

Once the specimens have been filled, they need to be cured to allow S/S reactions
to take place. The intent is to mimic as far as is possible, the conditions experienced
by the waste form in the field, as curing progresses slowly over days and weeks. If
the specimens are allowed to sit out in the field, or if placed in a storeroom under
atmospheric conditions they will dry out and not cure properly, and will not be
representative of the field-cured material. To avoid this, the freshly prepared samples
are placed in a containment environment with very high or saturated atmospheric
moisture, but at ambient temperatures.

201
Figure 6.7: A 0.5 in (12.5 mm) screen to remove oversize debris

Figure 6.8: Labelling of specimen mold and cap

202
A specially controlled room can be used for storage for curing, but most often a
standard picnic cooler containing water saturated towels, or (if the molds are sealed
on the bottom), approximately an inch of water is placed in the bottom of the cooler
(see Figure 6.9). Plastic caps should be placed over the top of the molds to protect
them. The literature provides at least one standard method for curing (ASTM D1632-
07).

Stored samples should be checked every day to determine if they have set. Stored
samples should not be transported off-site for testing until they have hardened, as
unset S/S material can be damaged producing unrepresentative results when tested.
Reserve replicate samples can be used to track setting either by physical
examination/handling or by the use of a pocket penetrometer (Figure 6.10). Once
they have set, the molds can be carefully wrapped and sent to the laboratory for
testing at the designated time.

Figure 6.9: Sample molds curing on-site in water bath

It is very important to ensure that detailed records are kept of all sample collection
and preparation activities (Figure 6.11). These records become critical if apparent
failures are reported later in the performance sample-testing program.

203
Details should include: date, time, location and a detailed visual description of the
sample, its apparent moisture content and the presence of debris, etc. A slump test
(Figure 6.12) is sometimes carried out on a fresh bulk sample as a quantitative
descriptor of sample rheology.

It is also good practice to record the condition of cured samples in the field or in the
laboratory prior to testing, and to photo-document the samples prior to testing. This
is a good assessment of the quality of the prepared cylinders, as poor preparation
will induce air voids (Figure 6.13) and other imperfections. A suitably prepared
cylinder is shown in Figure 6.14.

Figure 6.10: Checking set/strength by a pocket penetrometer

204
Figure 6.11: Detailed record keeping of properties of cured sample

Figure 6.12: A slump test on an S/S bulk sample

205
Figure 6.13: Poorly prepared field-specimens

6.4 Variation and failures


The testing of cement-based S/S soils contains many similarities with the testing of
civil/geotechnical engineering materials, including commonality of test methods (see
Section 7.3). As is normal practice, replicate samples are tested to obtain a
statistically valid result. It is normal for performance samples to exhibit variation,
which can be caused by a number of factors, including:

206
Figure 6.14: A well-prepared specimen with minimal air voids

The natural variability of soils and the COCs


The length of time the sample has been cured before testing
A non-homogenous matrix, caused by inadequate mixing during treatment
An irregular rate of reagent application
Variations in sample moisture content (and the consequent degree of
hydration of binder experienced)
Variations in the way the sample was placed into the mold
The inclusion of debris in the sample
Cracking of the sample, due to curing or handling
Variability in the application of the testing method
Errors in the reporting of data
Poor quality control in the testing laboratory
Differences introduced by methods applied by different testing laboratories

With so many potential factors introducing variability in sample test results, it is


important to differentiate between normal variability that does not affect the viability
of the remedy and those variations that indicate treatment failure.

207
6.4.1 Normal variation
As discussed in Section 7 (Developing Performance Specifications) for material
treated and disposed on-site, an overall performance goal is established. For
example, this might include achieving an average overall permeability value 1x10 -6
cm/sec ( 1x10-8 m/sec) for the solidified mass of treated material. However this
does not mean that every result obtained from a suite of samples of treated S/S
material must individually meet this criterion, in order to meet the treatment goal.

With this in mind, the variability that can normally be expected without undue
concern that something is going wrong is summarised from the authors experience
below:

Strength 20% low


Permeability half order of magnitude more permeable
Leaching 20-50% target value

With respect to leaching in particular, the amount of normal variation is very sensitive
to how low the target was set. For example, if the target number was 282 g/L (as it
was for one Pb contaminated site) then a 20% variation means 56.4 g/L, which is
not unreasonable. However if the target number was 15 g/L (as experienced at
another Pb contaminated site) then 20% equals a variation of only 3.0 g/L.
Variability this small is likely to be exceeded given all the possible factors affecting
the test result, and would not necessarily indicate remedy failure.

Table 6.2 presents some construction performance leaching results (SPLP) from
remedial actions at United Metals, a battery-recycling site, which was heavily
contaminated with Pb. The performance target was Pb 15 g/L, which is quite
restrictive, and the same as the USEPA drinking water standard.
Table 6.2: Construction performance samples - United Metals site

Days Pb in Leachate (g/L)


Curing/sample 1 2 3 4 5 6
14 1.2 1.1 18 7.9 15 15
21 0.6 0.6 NA 7.9 5.3 0.6
28 0.6 0.6 9.1 2.5 3.4 7.8

208
All performance samples met the target value, but there was scatter in the data
obtained, with the increase in Pb leaching values at 28 days vs 21 days for sample
number 6 explained by micro non-homogeneity within the sample(s) selected for
testing. From a 3 in x 6 in (75 x 150 mm) mold, only 100 grams were selected and
used for determining the SPLP lead content. Hence micro non-homogeneity,
unimportant in the large monolith, impacted the laboratory result.

An example of what is considered normal variation in physical properties is illustrated


in Table 6.3, which shows the results from replicate samples from the successfully
remediated American Creosote Site in Tennessee. The data show not only variation
between replicate samples, but also within the range of values from multiple
sampling events across the entire course of the remediation. More information about
the successful remediation of this site contaminated with creosote,
pentachlorophenol and dioxins is available in Bates (2002). Acceptable ranges for
variation from target values are often stated in remediation contracts. Development
of acceptable, but pragmatic, ranges for variations should be developed during
design with input from the key stakeholders in the project.

6.4.2 Recognising failure


If the performance specifications include a target value derived from an average of
all treated values, it is especially important to recognise failure of the treatment as
early as possible.

Performance specifications are often written to include a value to be achieved after


28 days of curing, or longer. However, it is desirable to conduct early testing so as to
recognise impending failure. Being able to recognise that failure is occurring from
early testing allows one to intervene and apply corrective action long before the 28-
day deadline passes, thus reducing the amount of off-spec material produced.

209
Table 6.3: Results from duplicate performance samples, American Creosote
site

UCS at 7 Days (psi) Permeability, cm/sec (m/sec)


Sample Replicate Sample Replicate
224 125 6.3x10-6 (6.3x10-8) 2.3x10-5 (2.3x10-7)
201 210 7.9x10-7 (7.9x10-9) 9.9x10-8 (9.9x10-10)
160 178 1.4x10-6 (1.4x10-8) 8.2x10-7 (8.2x10-9)
284 399 1.3x10-7(1.3x10-9) 1.3x10-7 (1.3x10-9)
445 267 8.5x10-6 (8.5x10-8) 8.4x10-6 (8.4x10-8)
245 223 4.0x10-6 (4.0x10-8) 2.9x10-6 (2.9x10-8)
285 235 2.7x10-7 (2.7x10-9) 1.2x10-6 (1.2x10-8)
135 193 1.6x10-6 (1.6x10-8) 3.6x10-6 (3.6x10-8)
260 258 6.5x10-7 (6.5x10-9) 4.4x10-7 (4.4x10-9)
315 383 4.5x10-7 (4.5x10-9) 7.2x10-7 (7.2x10-9)

Since some normal variation in data is to be expected, with some data falling below
the overall average target value, it may be difficult to determine that treatment is
failing. One of the best methods is to compare the values being obtained during the
early stages of remediation against those achieved during the bench or pilot-scale
treatability study.

During the later stages of the treatability study (Section 8.3), replicate samples
should have been produced to demonstrate the viability of the treatment formula by
testing at, for example, 7, 14, and 28 days of age. This database can serve as the
reference template to identify anomalous results. For example, the strength of
performance samples at 7 days of age can be directly compared with the 7 day data
recorded by the treatability samples. Once a significant database has been built
using performance samples from the remediation, then this replaces reliance on the
treatability database as the template for comparison. Figure 6.15 illustrates how the
strength increases over time for typical S/S treated material.

210
Figure 6.15: Increase in strength by S/S soil over time

6.4.3 Handling failure


For S/S treatment, handling failure is neither easy nor straightforward. The best
approach is to place substantial emphasis on construction quality control and on
early detection of any treatment problems.

With some remedial technologies, it is possible to re-treat material that is considered


to have failed. However for S/S, re-treatment is often not viable, especially if leaching
was one of the performance criteria, because S/S processes usually employ cement
or other alkaline reagents and these binders will have irreversibly changed the
properties of the soils, especially with respect to pH.

As the solubility of many metals (and some organics) are pH dependent,


retreatment, with the addition of more alkaline reagents, may cause the soil to exhibit
increasing leaching of, for example, amphoteric metals like Pb and As.

If however there are no concerns with leaching criteria, and the failure is one of
strength or permeability, it may be possible to re-treat the material. However, even in
these cases, the re-treatment formula likely will need to be changed from the original
formula and designed to treat the cause of failure. This may require a new bench-
scale treatability study and a consequent delay in completing the remedial action.

211
If the testing results indicated that it did not, or is likely not ever to, meet
specifications, then the failing test result must be examined to see if it is valid and
not simply an inaccurate value.

As mentioned, factors inducing erroneous data include sample cracking, poor


compaction, and the inclusion of debris. Thus, an investigation of the laboratory test
procedures used, the quality of samples tested, record keeping, and calculations
used, especially dilution factors applied during leachate analysis, is necessary. Then
one may have the laboratory undertake repeat testing, using reserve replicate
samples that have been kept for such eventualities. If the reserve samples meet the
remedial targets and samples from other sampling events have been fine, then it is
likely that the earlier poor result was due to a poorly prepared or non-representative
sample.

If the performance testing does reveal a treatment failure, then it is important to know
exactly where the failed material is located and how much was produced on the
same day of production so as to judge the potential significance of this event. For
example, if the permeability specification is that the average of all treated material
should be 1 x 10-6 cm/sec ( 1 x 10-8 m/sec) and that no single sample exceed 1 x
10-5 cm/sec (1 x 10-7 m/sec), then a sample reporting 2 x 10 -5 cm/sec (2 x 10-7
m/sec) is outside the specifications and a failure. However, if this represented only
half a days production, out of 60 days of treatment, and is located in the interior of
the monolith, this significance is not great. If, however, it is located on the edge of
the monolith, it may be more significant. Thus position may weigh into the decision
regarding whether to take an action, considering that it may be necessary to rip out a
lot of good treated material to reach a piece that is slightly off-spec.

If it is determined that an action is warranted, there are several options. The off-spec
material could be removed and sent off-site for disposal. The off-spec material could
be removed and retreated, but note earlier discussion regarding potential difficulty in
retreating. With either of these options, it may be necessary to remove and manage
a significant amount of treated material that does meet specifications, to gain access
to the target material. In this case, an alternative option is to leave the material
where it is, but take a compensating action. A compensating action may take the
form of providing additional isolation from the environment, for example if the
material is above the water table a GCL might be added as part of a cap design that
did not previously include it. This was done on one of the authors` sites.

If the treated but failed material is readily accessible then one should examine the
S/S material before beginning expensive or difficult removal and/or retreatment
options. Sometimes the bulk treated material may prove to be within specification,
and it was the performance samples that were not representative.
212
With reference to the remediation of a Pb-contaminated site, performance samples
from 2 sampling events indicated a serious failure in strength. The S/S-treated
material was accessible and field inspection revealed that some of the material was
out of specification, leading to removal and retreatment. However it was also found
that a significant portion appeared well cemented and within specification. This
material was tested in place using a field cone penetrometer to measure penetration
resistance as a surrogate for compressive strength (Figure 6.16).

Although unconfined compressive strength in the laboratory and penetration


resistance/compressive strength in-place in the field, are not directly comparable, the
in-place strength results were sufficiently impressive, that it was agreed that this
material did not require removal and retreatment. Likewise if the performance
sample failure had been of a leaching nature, samples from the actual field placed
material could have been obtained and tested.

Figure 6.16: Using a field penetrometer to assess strength of field placed S/S
material

In another example, in-situ augers were used to treat soils contaminated with coal
tars from operation of an MGP. Performance samples indicated that the top parts of
a small number of columns (out of the hundreds treated) did not meet requirements
for strength. Since the tops of these columns were accessible, it was decided to
excavate the suspect columns until obviously well cemented competent material was
encountered, then retreat and replace the poor quality material (Figure 6.17).

213
Handling failure is difficult and expensive in all cases, however the best approach is
to avoid failures in the first place by implementing strict construction quality control
measures and by using a binder formulation that provides a safety margin, above
minimum treatment criteria. If performance samples do indicate a failure, the use of
reserve samples to verify the result, augmented by field investigation of the affected
material, is encouraged. If failure is verified, then be imaginative in looking for
possible solutions. Retreatment is sometimes, but not always, a viable solution.

6.4.4 Coring of in-place S/S material


Coring in-place material has sometimes been proposed for collecting performance
samples. It has also been tried in order to conduct evaluations of in-place treated
material several years after the remediation. However, this approach can introduce
significant bias in the results obtained, and is not recommend, except in special
circumstances.

The physical aspect of cutting a core can introduce micro-fractures that increase and
may invalidate permeability testing results. These fractures will also have a negative
effect on strength. Furthermore, if during the coring water is used as a lubricant and
coolant, then the core will be pre-leached before recovery. If coring is dry, then
substantial heat is generated which adversely affects the core properties.

Figure 6.17: Excavating top of in-situ column

214
The authors have attempted the field extraction of cores several times using differing
drilling equipment, but the result is almost always the retrieval of cracked, shattered,
bits and pieces of material (Figure 6.18).

This is partly due to the fact that S/S soils generally are weakly cemented materials
with typically 2 to 10 % of the strength of structural concrete. Treatment
specifications for S/S usually require about 50-100 psi (0.3-0.7 MPa), compared to
2000 psi (14 MPa) or greater for concrete applications such as sidewalks and
driveways.

Figure 6.18: Cores obtained by sonic drilling of S/S material

215
7.0 Performance specifications for S/S

The development of performance specifications for S/S is a critical step in the


planning, design, and implementation of this technology.

As a technology class, S/S:

uses different binders, reagents, and additives (Section 8)


employs varying methods of implementation, utilising different equipment
(Sections 4 and 5)
can treat waste/contaminants with contrasting behaviour (Section 3)
relies upon different chemical and physical mechanisms to achieve treatment
including: encapsulation/containment, source control, solubility and/or
leaching reduction, etc.

Thus, the setting of performance expectations for S/S materials is not a one size fits
all approach but requires thoughtful site-specific planning. One should avoid simply
adopting performance criteria, test methods, and material properties from other
sites/projects, from case studies, or misappropriating regulatory tests (such as the
Toxicity Characteristic Leaching Procedure, TCLP, (EPA method 1311)) without
regard for their original purpose.

The inherent dangers in adopting others` approaches or defaulting to promulgated


regulatory tests without due consideration of site-specific conditions, risk pathways,
and objectives can lead to failure to treat, or over-treatment resulting in excessive
cost or failure of the remedy, and the prospect of litigation.

To date the majority of available guidance emphasises the technical aspects of S/S
technology and appropriateness for use at a given site and waste/contaminant types.
More recently, the ITRC published guidance on developing performance
specifications for S/S (ITRC, 2011) that includes useful flowcharts illustrating how
material performance goals and specifications are developed and implemented to
achieve successful S/S. The basic concepts and logic flow can also be applied to
both in-situ and ex-situ S/S with some modifications of the considerations column.
This ITRC flowchart, adapted for use in this Manual of Practice, is shown in Figure
7.1. Key concepts are used as the basis for this section.

7.1 Setting overall goals for treatment


The key terms for setting goals for treatment were introduced in Section 1.2 and are
summarized here as follows:

216
Figure 7.1: Performance goals and performance specifications in the S/S process (Modified from ITRC, 2011)

217
Remedial goals: overall objectives of the S/S remedy to address the
identified risk pathways

Material performance goals: expected behaviour of the treated S/S material


to meet the remedial goals

Material performance specifications: the collection of parameters, tests


and criteria for developing a mix design and the testing of the mix design to
meet the material performance goals

Construction performance specifications: the data/criteria to be used to


verify that the S/S treated material is consistent with the findings of the
treatability testing phase, and that key performance characteristics (e.g.
strength, permeability, leaching behaviour) are consistently met during
treatment

In addition, the following components of performance specifications (based


on ITRC, 2011) are useful to define here:

Performance parameters: the materials properties that enable the S/S-


treated product to be fit for its intended purpose (e.g. leaching, hydraulic
conductivity etc.)

Performance tests: testing of the S/S-treated material that can result in one
or more representative measurements (e.g. leaching evaluation by ANS/ANSI
16.1, hydraulic conductivity by ASTM D5084)

Performance criteria: design values used to demonstrate that acceptable


performance has been achieved

Performance criteria may be established for the remediated contaminated site


based on regulatory criteria (e.g. groundwater concentrations derived from
water quality criteria or other risk-based standards), or a material
parameter/value considered suitable for meeting established remedial or
regulatory goals, such as e.g. a maximum acceptable hydraulic conductivity
value, or a maximum acceptable leachate concentration for COCs to achieve
the remedial goals/targets.

The process of developing and implementing performance specifications for


S/S begins with setting site remedial goals. These are developed from the

218
Conceptual Site Model (CSM, see Section 7.2) and provide meaningful and
appropriate performance goals to be achieved by S/S.

The remedial goals are intended to address the impacted media, risk
pathways and exposure endpoints. The role of S/S is site-specific, ranging
from a sole technology (single treatment option) to a component of an overall
remedial strategy/treatment train. Examples of the way S/S has been utilised
include:

Ex-situ S/S combined with a composite surface cap and a perimeter


slurry wall as a source control to limit mass flux of a mixed-contaminant
site (i.e. VOCS, PAHs, pesticides, PCBs and metals), to meet
groundwater quality standards at a down-gradient well point of
compliance (Peak Oil Site described in Appendix C of ITRC [2011]).
In-situ S/S to encapsulate soils containing coal tar and non-aqueous
phase liquids (NAPL) mixed with soil at the former Columbus Georgia
manufactured gas plant (MGP site). The outer perimeter of this site was
adjacent to a river, where a binder-rich mix was developed to function
as a barrier/containment wall facilitating construction of a riverfront park
(EPRI, 2003; USEPA, 2009).
A number of former MGP sites where in-situ S/S encapsulated soils
containing coal tar/NAPL to reduce the mass flux to groundwater (EPRI,
2009). Some of these sites used S/S as the sole remedy to manage
impacted soil and groundwater, while others included additional
components, such as covers.

S/S can be used alone or in combination with other technologies (see Section
7.4) and the material performance parameters and tests will reflect the site-
specific approach adopted.

By way of example, at sites where the source material is accessible for


treatment, S/S can be the primary remedial remedy. Here, a source term
defined from leaching tests coupled with a groundwater transport model can
verify that the (groundwater-related) remedial goals can be achieved by S/S
alone. At other sites where the contaminated material is not completely
accessible, S/S may be used as a containment system, whilst recognising that
some groundwater treatment may be necessary. In this scenario, a
groundwater model and a detailed leaching evaluation may not be necessary,
as the impacted groundwater will be managed by other technologies.

However, when using S/S in a treatment train, it is important to consider


compatibility between different technologies. For example, if the raising of
groundwater pH may result from S/S, the effect of this change on down-
gradient treatment design, e.g. chemical oxidation/bioremediation etc. will
need to be considered.
219
Having established site remedial goals and how the S/S technology will be
used and implemented, the material performance goals can then be
established. Examples of material performance goals include:

A bearing capacity to support overlying soil cover, structures, or


equipment
Chemical treatment of contaminants to achieve lower soluble species
and limit mass flux to groundwater (i.e. stabilization)
Containment via reduced permeability (i.e. solidification) limiting
groundwater contact
Containment to prevent migration of NAPL
Reduced leaching rates (i.e. mass flux) to meet groundwater remedial
goals at the designated point of compliance at some distance from the
treated material

Most S/S projects will utilize one or more material performance goals similar
to those examples presented above to develop a mix design to meet the site
remedial goals. The use of other supplemental technologies in conjunction
with S/S (e.g. barrier walls, caps), as described in Section 7.4, may also
influence the degree to which source materials need to be treated,
considering that secondary barriers may be in place as well.

An important concept to recognize in developing a mix design for S/S is that


the material performance goals are goals. That is, they are desired
properties of the treated material. The material performance goals will be
used to identify the material performance specifications which will include the
performance parameters to be evaluated, the performance tests to be
conducted, and a preliminary set of performance criteria.

As various mix designs are developed and evaluated and the performance
tests (and other material property tests as desired) are performed, it may not
always be possible to achieve the performance criteria (e.g. a desired
leachate concentration of a specific contaminant) or the reagents or specialty
additives required may be too expensive or in limited supply for the
quantities/proportions needed based on the treatability testing. The cost and
availability of reagents is often taken into account during the choice of mix
designs selected for testing to assure that the formulas tested are
economically viable for full-scale treatment. Failure of economically viable
formulations to fully achieve the desired property does not have to mean that
the treatability testing has failed, or that S/S will not work for the site, rather
this may be a point in the treatability testing where the material performance
goals and how S/S fits into the overall remedial strategy are re-evaluated.

220
At this point, the introduction of supplementary technologies (e.g. capping,
vertical barrier walls, down-gradient groundwater treatment) might also be
deemed necessary to meet site-specific goals. For example, a more cost-
effective S/S mix design (that does not meet the preliminary leaching criteria)
may be selected and used in conjunction with a barrier wall or an outer ring
(i.e. at the interface between the S/S-treated material and groundwater) as a
more cost effective solution to achieve the site remedial goals.

The material performance specifications developed in conjunction with the


treatability testing will ensure that the S/S mix design achieves the material
performance goals. However, the testing of cured specimens must recognise
that under some circumstances extended curing times beyond 28 days are
appropriate, especially where contaminated sediments and other
contaminated materials require treatment.

The concept of testing materials at 28 days of cure time comes from the
concrete industry where it has important construction considerations, but
during S/S, this time limit does not fully account for development of desirable
properties. Under some circumstances, therefore, the performance evaluation
should take many months, e.g. when difficult materials are being treated or
diffusion controlled leaching of COCs is being evaluated.

However, it is often desirable for a quicker turn-around of results to meet


deadlines or construction productivity targets, whilst also maintaining the
material performance goals. Under these (or changed) circumstances, the
tests to be used may differ from the material performance specifications
developed during the treatability testing. By way of example, the evaluation of
the strength development during the treatability study will allow for shorter
specimen-cure times in the field. It is possible to predict (from the rate of
strength-gain achieved) if the treated material is on-track to meet the target
strength. An example strength-gain plot of bench vs pilot-test data is shown in
Figure 7.2. Although the plot shows some variability in pilot-test data
(compared to the bench testing), the data was used to make periodic
adjustments to the quantity of reagents applied at the particular site in
question.

For the evaluation of leaching, it is often not practical to perform leaching tests
to monitor construction progress and compliance with the performance
specifications. In some cases, therefore, strength and permeability are used
to demonstrate consistency with the data obtained from treatability testing,
with the leaching evaluation being undertaken later. In other cases, batch
leach test or a single-point test at a relatively short curing time (e.g. 7 days)
can be useful. However, it is critical that shorter-duration performance tests
also be evaluated during bench-scale testing (Section 8.3) to determine what
the construction performance criteria should be, as they may need to be
different from the material performance criteria as, for example, contaminant
221
flux rate decreases with increasing S/S curing time. ITRC provide example
applications in this respect (ITRC, 2011).

Other tests that may be useful during construction include slump (ASTM
C143), moisture content and grout density. Performance parameters/test
methods are discussed further in Section 7.3.

Figure 7.2: Strength-gain of bench vs. field demonstration (FD) data

7.2 The importance of site characterisation and the conceptual


model
The development of a remedial strategy and choice of technologies requires a
thorough understanding of COC release, the distribution and fate of these
contaminants and their likely exposure pathways, and the risks posed to
human health and the environment.

A site characterisation program based on specific data quality objectives


(DQOs) should be developed in accordance with USEPA and/or state-specific
regulatory programs (see: www.itrcweb.org for other appropriate guidance,
including accelerated site characterisation: Publications ASC-1 to ASC-4;
sampling, characterisation and monitoring: SCM-1 to SCM-3; and for DNAPL:
DNAPL-1 and DNAPL-4).

222
A conceptual site model (CSM) is a vital tool in understanding the site
conditions and the applicability of S/S. CSMs typically describe key site
features including:

Hydrogeology
Flow regime(s) (granular media, karst, or fractured bedrock)
Stratigraphy (simple or complex with preferential flow zones)
Aquifer characteristics (hydraulic conductivity, depth of confining units,
state groundwater classification)
Contaminants
Type, concentration, mass released, toxicity and mobility
Contaminant distribution in soil, bedrock, groundwater, surface water,
sediments, and soil vapour
Presence of NAPLs
Groundwater plume geometry, direction, and rate of movement
Site topography, structures, utilities, the surrounding land use and
potential receptors

A sample conceptualisation of a contaminated site and its relationship to


identified receptors is illustrated in Figure 7.3.

223
Figure 7.3: A simplified conceptual site model

224
The CSMs should be carefully prepared in accordance with regulatory (or
other) guidance such as ASTM E1689 Standard Guide for Developing
Conceptual Site Models for Contaminated Sites (ASTM, 2008).

The development of a CSM also helps determine if there are data gaps in the
site characterisation, which could affect the site characterisation or the
remedial technology evaluation, and what data gaps need to be addressed
(e.g. additional site characterisation, pre-design investigation, treatability
evaluation).

Once a detailed understanding of the site is obtained, the applicable or


relevant and appropriate requirements (ARARs) from a regulatory perspective
should be developed, the present and potential future site risk analysis can be
performed, and the potential clean-up standards that may apply can be
identified.

The risk analysis will identify potential receptors, what COCs and risk
pathways are present, and those that may pose unacceptable risk to human
health and the environment. The level of detail required is however, site-
specific and dependent on the regulatory setting. For guidance one can
consult the ITRC Web site (www.itrcweb.org) for appropriate guidance
documents concerning contaminated site risk assessment (ITRC Publications
RISK-1, RISK-2, and RISK-ALL).

Following the detailed analysis and risk assessment of the site to be treated,
the remedial action objectives can be formulated. Then one can identify what
technologies should be considered (and how they should be used) and the
remedial goals for the site can be determined (see Section 7.1).

An important consideration at this stage is the assessment of site geology and


hydrogeology, contaminant-related factors, and the limitations of remedial
technology options to meet the remedial targets identified. Sites with DNAPLs
and complex geology have been recognised by the USEPA as often
impracticable to fully manage, and these should be evaluated as such
(USEPA, 1993).

Alternate remedial strategies for complex sites where restoration is technically


impracticable or cost prohibitive, may include actions such as exposure
control (through institutional controls), source control by removal, treatment,
or containment, and where practicable, groundwater restoration by in-situ or
ex-situ treatment or natural attenuation.

225
S/S is a commonly used source control remedy with the treated materials
either left in-place, consolidated into an on-site disposal cell, or disposed off-
site at a licensed landfill facility. S/S can be used as a stand-alone remedy or
in combination with other remedial actions (see Section 7.4).

In order to support the selection of S/S there are a number of factors that
should be considered in the site investigation, CSM, and risk analysis phase
of the project. Table 7.1 summarises the primary site assessment
considerations required, addressing soil, groundwater and contaminant
properties and treatment effectiveness. In addition, the method of S/S
implementation will have limitations that may be due to soil type (e.g. clay
content, rocks, cobbles, debris), presence of debris, obstructions or rubble
(which may require removal through pre-processing), or aggressive soils or
contaminants (e.g. acidic conditions).

The behaviour of S/S materials in the environment and site conditions which
influence treatment by S/S are described in several useful documents
including: ITRC (2011), EPRI (2009) and Environment Agency (2004a and
2004b). Waste/solidification reagent interactions and interferences are
addressed in PCA (1997 and 1998), Spence and Shi (2005) and Environment
Agency (2004b).

7.3 Performance parameters, tests/methods


7.3.1 Performance parameters
As described, the performance/properties of S/S treated materials are
characteristic of the ability of the waste form to meet its intended purpose.
These parameters are derived from the material performance goals
established for a site, and can be grouped as follows:

226
Table 7.1 Typical S/S site assessment considerations
S/S Evaluation Project
Typical Analyses/Observations Significance
Factor Phase
Soil Classification/ Site Gradation Soil physical properties and their variability at a site have a
Physical Investigation, USCS Classification significant impact on the number of mix designs
Characteristics Treatability Atterberg Limits evaluated/required and the choice of equipment and
Study, Pre- Moisture Content reagents to perform solidification.
Design Unit Weight Some physical characteristics such as cobbles and
Investigation Debris content boulders, significant debris content, and very dense soils
Organic content may eliminate in-situ S/S from consideration.
Porosity
Density
Permeability
Soil and Pre-Design Salinity Affects cement hydration reactions and long term durability.
Groundwater Investigation, pH Additives may be required to overcome interference
Geochemistry Treatability Sulfate mechanisms.
Study Other cement hydration interference parameters Ettringite (a form of calcium aluminum sulfate) formation due
as needed to high sulfate concentration causing excessive swell.
Contaminant metals
Solidification reagent compatibility with site
contaminants (adverse reactions)

227
S/S Evaluation Project
Typical Analyses/Observations Significance
Factor Phase
Contaminant Site History of contamination and location of source Affects applicability of S/S and possible selection of
Characterisation Investigation, areas additives for contaminant attenuation or mix-ability
and Treatability Classes of contaminants improvement.
Distribution Study Presence and distribution of NAPLs ISS generally limited to 60-80' below ground surface
Distribution of contaminants in geologic strata depending on geology. Grout rheology may need to be
Contaminant concentrations and mass released modified to reduce mixing shear resistance.
Contaminant properties (phase, solubility, Off-site impacts require access agreements, possible
volatility) temporary loss of use, and/or utility reconstruction and
On- or off-site impacts relocation.
Extent of soil and groundwater impacts Waste acceptance criteria for off-site disposal of treated
NAPL physical properties material.
Leaching behaviour Leaching behaviour of untreated soils provides a baseline
against which to judge results of treatability studies and full-
scale application.
Water Table Depth Site Investigation Water Table depth and seasonal variability Amount of water available for cement hydration affects grout
Vadose zone and/or saturated zone contaminant design.
and/or NAPL impacts Up-gradient groundwater mounding due to in-situ solidified
Presence of perched water table soils may be a concern at shallow water table sites.
Penetration of perched water table may create new
migration pathways for contaminants.
Hydrogeology Site Investigation Geologic strata including geometry of geologic Affects contaminant distribution and accumulation zones.
units Affects amount of solidification reagent to reduce hydraulic
Hydraulic conductivity of impacted zones conductivity.
Groundwater flow direction & gradients

228
S/S Evaluation Project
Typical Analyses/Observations Significance
Factor Phase
Land Use & Site Site and surrounding land use (e.g. residential, Affects ability to leave contaminant mass in the ground.
Infrastructure Investigation, industrial, commercial, etc) Active infrastructure and/or buildings can limit accessibility
Feasibility Study, Site access to impacted soils.
Remedial Design Ability to impose institutional controls Buried remnant structures and utilities may require
Site ownership demolition and removal prior to S/S.
Archaeological/historic features Groundwater pH and alkalinity changes due to ISS need to
Ecological status be considered.
Site activity (actively used or abandoned?)
Presence of active or abandoned buildings and
subsurface infrastructure, utilities, foundations,
tanks, etc.
Proximity to water bodies
Groundwater use
Future Site Use Site Potential for future contact with solidified soils Future intrusive work and future building construction affect
Investigation, Geotechnical properties of solidified soils mix design, thickness of clean soil buffer above solidified
Feasibility Study, soils, and need for a geo-membrane barrier over solidified
Remedial Design soils or a vapour intrusion barrier.
Creating clean corridors may facilitate future underground
utility installation needed for site redevelopment and reuse.

229
S/S Evaluation Project
Typical Analyses/Observations Significance
Factor Phase
Receptors/Risk Site Investigation, Routes of exposure Affects S/S implementation, mix design, need for additional
Pathways Risk Assessment, containment or separation of solidified soils, and
Feasibility Study, Compounds or contaminant classes driving the vapour/emissions controls during remediation.
Remedial Design risk

Severity of the risk

Ability to implement engineering and/or land


use controls

Remedial action objectives

Aquifer/Waterbody Site Investigation Regulatory classification of groundwater and Affects remedial goals, clean-up criteria, and determination
Status surface waters of practicability to achieve clean-up criteria

230
Environmental performance (i.e. release of constituents to the
environment through leaching)
Characteristics affecting leaching (e.g. permeability to reduce water
contact, pH when chemical stabilization is a desired mechanism)
Characteristics affecting placement (e.g. strength, bearing capacity,
durability, etc.)

The key performance parameters most often used to evaluate S/S materials
are strength, hydraulic conductivity, and leaching. The importance of these is
summarised below, and discussed in ITRC (2011), EPRI (2009), Spence and
Shi (2005), USEPA (1989), and EA (2004b).

Strength: Strength is the ability of the treated material to withstand an applied


load. For S/S materials, strength may be used to assess suitability for
placement of ex-situ treated material in a defined disposal area (i.e. adequate
bearing capacity), to assess the creation of a monolithic mass (i.e. unconfined
compressive strength), or as a surrogate parameter to ensure that the
chemical reaction of the binder (e.g. Portland cement), and water has
occurred and has not been subject to significant interference by the waste
constituents or other site-specific geochemical reactions (e.g. sulfate
inhibition). Strength can also be used as a surrogate measure of durability,
considering that, in general, it would be expected that higher strength
materials would be more resistant to deterioration over time.

Hydraulic conductivity: Hydraulic conductivity provides a measure of how


easily water can pass through a material. For S/S materials where
containment and encapsulation of contaminants (i.e. solidification) is the
desired treatment mechanism, reducing the hydraulic conductivity reduces the
ability of water to come in contact with the contaminants, and therefore
influences the rate of leaching. A reduction of two to three orders of
magnitude in hydraulic conductivity through treatment of S/S materials is
typically needed to change the mode of water contact from a flow-through
scenario to a flow-around scenario (Atkinson, 1985).

Leachability: Leachability is the materials ability to release a contaminant


from a solid phase into a contacting liquid. The leachability of S/S materials
can be influenced by contact with an eluent and the chemical reactions
involving the S/S binders, and contaminant solubility. Leaching into
groundwater is a principal pathway for contaminant release into the
environment, and is the key target parameter addressed by S/S.

231
7.3.2 Performance tests and methods
Numerous test methods exist for evaluating the performance of S/S materials.
Some useful references on methods and their limitations include ITRC (2011),
Spence and Shi (2005), and EA (2004b). Two categories of assessment
testing that are considered in S/S (Cajun and Shi [2005]) are:

Basic information or index tests, which measure fundamental material


properties such as gradation, moisture content, dry density, plasticity,
contaminant concentrations, etc., typically applied to untreated material
characterisation
S/S material performance tests which relate to the performance of the
treated material

For S/S where solidification is desired, testing is performed on samples cured


in cylinder molds, over enough time for waste-binder-water reactions to
progress satisfactorily, typically at least 14 days and preferably 28 days, or
even longer where necessary (see Sections 6.3, 6.4, for treated sample
preparation and curing considerations). When chemical stabilization is the
only desired treatment, curing and testing of physical properties such as
strength and hydraulic conductivity may not be necessary.

Strength: Unconfined Compressive Strength (UCS), (ASTM D1633; BS


1924:4; ASTM 2166-06), is the most common test method used in the US for
evaluating the strength of molded soil-cement cylinders. This test is the
measure of the ability of a monolithic specimen to resist an applied load
without breaking, and is only applicable to cemented or otherwise cohesive
specimens.

If the treated S/S material is a granular product (e.g. resulting from ex-situ
metals stabilization) determination of a UCS test may not be appropriate. A
soil bearing capacity test, such as the Cone Penetration Test (ASTM D3441),
may be a better test to determine a treated materials ability to support a load
after placement. In either scenario, the S/S material should be able to support
the loads that will be present after treatment, whether as overlying soil/cover,
or future construction or structures.

Other tests that may be used include BS-1924-2 1994, ASTM D 2166-
06(cohesive soils), and CBR according to ASTMD1883-99 with prior
compaction via ASTM D0698-00 or D1557-00.

A value of 50 psi (0.3 MPa) UCS is often used as a minimum acceptable


value in practice, but is really a site-specific determination. A minimum
strength value should be selected based on the anticipated loads the treated
material will encounter where it is created or placed and to demonstrate if it is
232
a desired property that the solidification reactions occurred. In some
situations, post S/S construction may require driving piles through treated
material, in which case a maximum strength may be desirable. Also in some
cases a much lower strength may be acceptable based upon site-specific
considerations.

Hydraulic conductivity: ASTM Method D5084 is the most common US


method for the determination of hydraulic conductivity using a flexible wall
permeameter. The selection of constant-head or falling-head procedures will
be based on the expected hydraulic conductivity range (see discussions in
Cajun and Shi [2005] and EPRI [2009]). Typical hydraulic conductivities of
solidified soils range from 1x10 -4 cm/sec to 1x10-8 cm/sec (1x10-6 m/sec to
1x10-10 m/sec), comparable to that of clay materials. The USEPA
recommends that the permeability of solidified soils should be at least two
orders of magnitude below that of the surrounding soil (USEPA, 1989). For
many sites, a hydraulic conductivity criteria of 1x10-5 cm/sec to 1x10-6 cm/sec
(1x10-7 m/sec to 1x10-8 m/sec) is used. In Europe this is often site-specific:
leachability and permeability are combined in a model, as the effects of both
influence the mobility of the pollutant. Sometimes very low permeabilities (<
10-7 cm/s or <10-9 m/s) are specified to compensate for the leaching potential
of a pollutant.

Leachability: Leachability tests are typically performed on solidified or


stabilized materials to evaluate mass transfer from a solid to a liquid.
Extraction-based or equilibrium tests evaluate the mass of material that can
be leached under a specific set of test conditions. Mass transfer, or flux-based
testing, evaluates the rate of leaching over time from a mass or surface area
of treated materials. Detailed discussions of the various test methods are
provided in Cajun and Shi (2005), ITRC (2011), and EA (2004b) among
others.

Common US extraction tests include the Toxicity Characteristic Leaching


Procedure (TCLP, EPA Method 1311) and the Synthetic Precipitation
Leaching Procedure (SPLP, EPA Method 1312). TCLP has limited use with
S/S materials, unless they are destined for disposal in a landfill and
hazardous waste characterisation is necessary. SPLP has been used for S/S
materials. However the test crushes the specimen and provides a releasable
amount of the waste constituent which provides no information on the time
rate of leaching. SPLP may be most useful when evaluating the effectiveness
of chemical stabilization treatment mechanisms as opposed to solidification
treatment mechanisms.

In Europe various leaching protocols exist, based on granular or crushed


samples, or based on monolithic samples (diffusion testing). Many countries
adopt these procedures in their own legislation.

233
Simple and short time batch leaching tests, as the EN 12457 series, are often
applied in S/S validation, although the tests are based on granular or crushed
material, and in that respect do not simulate the in-situ situation. Moreover,
these tests do not give information about the kinetics of the leaching process.
More complex tests, as the column test NEN 7343 (for granular materials) or
the tank test NEN 7345 (monolithic materials) are longer time, expensive (up
to 8 eluates are analysed), but yield information on the kinetics of the
leaching, and can be used to predict the long term leaching behaviour.

Since S/S most often relies on the formation of a solidified monolithic material
such that the majority of the groundwater or infiltration flows around the
treated material, diffusional release of contaminants from the treated material
over time is the controlling behaviour. Therefore, flux-based leaching tests are
considered more appropriate as they evaluate the time-dependent release of
contaminants. The American Nuclear Society (ANS) test method 16.1,
Measurement of the Leachability of Solidified Low-Level Radioactive Wastes
by a Short-Term Test Procedure (ANS 2003), is a commonly used flux-based
test that was initially designed for cement-stabilized low-activity nuclear
waste, but has been adopted for use with other cement-stabilized wastes.

Some emerging EPA leaching methods for S/S materials are being developed
and validated currently. These methods characterise leaching behaviours as a
function of pH, liquid-solid ratio, and mode of water contact. These emerging
leaching methods were initially developed as part of Vanderbilt Universitys
Leaching Environmental Assessment Framework (LEAF;
www.vanderbilt.edu/leaching), and a discussion of these LEAF methods is
provided in ITRC (2011). References for many of the European Methods can
be found in Cajun and Shi (2005), EA (2004b), Kosson et al. (2002) and
Perera (2004).

How to determine an acceptable level of leaching is a challenge many


projects face and has been addressed in many different ways. Three common
approaches are:

Demonstrating a reduction in leaching through treatment


Determining an acceptable attenuation between the treated material
and a down-gradient point-of-compliance
Use of hydrogeology and contaminant fate and transport modelling
using an S/S material source term derived from leach testing and
simulating time and distance-dependent groundwater concentrations

Relating leaching performance goals and various methods to assess


compliance are provided in ITRC (2011) and the reader is referred to that
document for more detail.

234
Durability testing is sometimes performed as a relative comparison among
mix designs. The two most common durability tests, Wetting and Drying
Testing of Solid Wastes (ASTM D4843) and Freezing and Thawing Test of
Solid Waste (ASTM D4842), are applicable only to materials that will
experience these cyclic conditions.

For most S/S applications where material is below the frost line or below the
water table, these cyclic conditions do not apply. As described in USEPA
(1989), no standards exist for whether stabilized material has passed
durability testing, although 15% weight loss is often considered an acceptable
amount. According to USEPA (USEPA, 1989) if the durability test results
show relatively low loss of materials and retention of physical integrity after
testing, the mix design is likely adequate for long-term stability. If the test
results show a large loss of material and loss of physical integrity, a different
mix design should be considered to provide long-term stability. Further
discussion of durability tests is provided in USEPA (1989). However, it is
important to note that such durability tests apply only to strongly cemented
materials and not to cohesive S/S-treated material. European methods for
durability testing can be found in EA (2004b).

7.3.3 Phases of performance testing


Performance testing is conducted in three phases from initial mix design
development through implementation verification as follows:

Bench-scale treatability testing


Pilot-scale field demonstration
Full-scale implementation

Bench-scale testing determines one or more mix designs capable of treating


the contaminated material and meeting the material performance goals. In this
phase, the selected mix design is tested on replicate samples of waste
material to ensure that material variability is accounted for. Both physical and
chemical containment can be achieved by the S/S using a cementitious
reagent, and further discussion of the bench-scale treatability study is
provided in Section 9.3.

Pilot-scale demonstrations determine if the mix design applied at the field-


scale (using the full-scale mixing equipment) will produce results consistent
with those obtained in the bench-scale testing. This phase also allows for
further assessment of the physical and chemical variability of the waste
material and how the selected mix design performs, and is used to further
develop and optimise the mix design (if needed) and the construction
methods for full-scale implementation.

235
Tests other than performance tests may be used to determine if treated
materials are consistent with those produced in the laboratory. These
consistency tests are typically established in the bench and pilot phases for
use during the construction phase, and can be performed on freshly mixed
material or cured specimens.

The real-time testing of freshly mixed material is used to identify significant


variations in material properties that can affect performance test results, such
as grout density, slump, and homogeneous mixing. Short-term tests on
specimens as they cure may include strength-gain rate using a pocket
penetrometer or UCS tests at intervals such as 3, 7, 10, 14, 21, and 28 days
to evaluate strength-gain rate, and/or visual observation of bulk field-cured
samples (e.g. 5-gallon (19 L) bucket). Abbreviated leaching tests, or leaching
tests performed with shorter cure durations (i.e. 7 or 14 days) may also be
considered in this phase.

Full-scale implementation of S/S uses a combination of performance and


consistency tests to monitor treatment progress in real-time and conformance
with performance criteria.

As mentioned, leach testing may be limited to the bench-scale and pilot-scale


phases with strength and permeability being used as surrogate measures for
leachability during full-scale S/S. As long as the target strength and
permeability are met, then the leaching performance should be met as well,
although an exception to this may be for metals stabilization, for which short-
duration extraction tests, such as SPLP, may be needed to demonstrate
chemical stabilization has been achieved. Other indicative parameters (for
chemical stabilization) that can easily be measured in the field include the pH
of the treated material.

7.3.4 Performance criteria in specifications


Performance criteria statements should be written to recognise that some
limited variability in test results is often acceptable without compromising the
overall success of the remedy (see Section 6.4). This approach relies on an
assessment of failures in light of the overall aggregated performance of the
treated material and the potential contaminant flux from the entire S/S mass to
the environment, and not just from one sample point.

Incorporating tolerance intervals in the performance criteria specification can


provide flexibility to field personnel, whilst recognising that variable field
conditions are the rule, rather than the exception at S/S sites. An example of
tolerance-intervals for strength, for use in performance criteria follows:

236
An average of all performance samples must not be less than 50 psi (0.3
MPa), no individual sample shall be less than 40 psi, and no more than 20%
of the performance samples shall be less than 50 psi (0.3 MPa).

The site owner/representative/designer and regulator should develop a


consensus on what is acceptable based on the agreed remedial objectives
whilst considering the location of failing test results, among other factors.

For example, a few failing test results that meet the lower tolerance limit that
are randomly scattered throughout the S/S area may be less of a concern
than a cluster of failures in one general area (possibly indicating differing
soil/contaminant conditions, inadequate reagent batch preparation, or
inadequate mixing) or along the outer edge (where groundwater or infiltration
exposure may be greatest).

Retreatment can be expensive, is difficult to effect on previously cured


materials, and may not result in significant improvement due to pH increases
and/or solidifying broken cemented material. Thus the value of replicate
specimens for retesting if a failing result is obtained cannot be under-
estimated at this juncture (Sections 6.2 and 6.4).

7.4 Impact of other protective measures


It is increasingly common to combine S/S treatment with other protective
measures as part of a package of technologies to remediate sites at lower
cost while still achieving acceptable levels of protection. The objectives of
these protective measures are often to isolate the S/S material from surface
and groundwater.

Isolation from surface water can be achieved via low hydraulic conductivity
caps (e.g. flexible membrane liners (FMLs), geo-synthetic clay liners (GCLs)),
whereas for groundwater, vertical cut-off walls keyed into an underlying low
permeability strata promote groundwater flow around the site (e.g. bentonite
slurry walls) or vertical drains to capture groundwater flow coming into the site
(i.e. vertical drains).

Most S/S remedial projects to date have used S/S to treat all of the
contaminated media of concern. Incorporating multiple barrier systems with
S/S can affect the required performance specifications for the S/S treated
component. By incorporating such measures, groundwater flow through the
237
site will be very low and any potential off-site migration will be due to slow
diffusion limited transport. Methods to assess the impact of multi-barrier
systems on contaminant migration are outlined by Lake (2013).

Examples exist on how both the extent of contaminated media and the S/S
performance specifications can change for a multi-barrier system. The
Whitehouse Waste Oil Pits site in Florida, USA (EPA, 2011), was remediated
by using the following approach:

Use of shallow in-situ S/S to treat the top two lifts of soil and oily acid
sludge (approximate the upper 3 ft [0.9 m]) to create an S/S sub-cap of
low hydraulic conductivity
The installation of an RCRA cap (containing both a GCL (geo-synthetic
clay liner) and an FML (flexible membrane liner))
Construction of vertical confining walls, bentonite slurry walls, around
the contaminated media, keyed through multiple underlying clay layers
at depths of approximately 65 ft (19.8 m)

According to EPA (2011) All substantial elements of the physical construction


of the remedy were completed on May 4, 2006, and initial groundwater
monitoring results indicate the remedy is effective in isolating groundwater
contamination. Since the remedy incorporated a multi-barrier approach via a
HDPE/GCL RCRA cap and vertical confining walls that were keyed into a
confining clay (see Figure 7.4) the only leaching specification was that the S/S
treated material did not leach contaminants of concern (COCs) in excess of
the untreated material. A permeability of 1x10 -6 cm/sec (1x10-8 m/sec), and a
minimum strength of 0.3 MPa (50 psi) for the S/S treated material were
retained as specifications.

Another such example of utilising a multiple barrier approach is the Brunswick


Wood Preserving Site in Georgia, USA. This disposal area contained several
lagoons in which wood preserving wastes such as creosote and
pentachlorophenol (PCP), with dioxins, were disposed.

The remedial approach called for excavation and ex-situ S/S treatment of the
top 3 - 4 ft (1-1.2m) of contaminated soil and sludges, and the return of these
S/S treated materials as a sub-cap over the lagoons (shown in Figure 7.5).
The formula used was 10% Portland cement and 10% fly ash, by weight, to
the excavated soil/sludge.
238
Figure 7.4: Installing a slurry wall, Whitehouse, Florida

Vertical confining walls keyed into the underlying confining clay at a depth of
about 60 - 70 ft (18.3 - 21.3 m) were constructed using both trench/backfill,
like those at Whitehouse and by using a Bauer panel cutter (Figure 7.6). The
cap over the S/S treated material contained a GCL, but in this case no FML.
The S/S treated sub-cap contained a hydraulic conductivity specification of
1x10-6 cm/sec (1x10-8 m/sec) and a strength specification of 100 psi (0.7
MPa).

Due to the 3 - 4 ft (1-1.2m) of low hydraulic conductivity sub-cap and the


vertical confining walls, the leaching specification for the S/S treated material
was that the S/S treatment did not significantly increase the leaching potential
for COCs in the S/S treated material.

Although the strength specification of 100 psi (0.7 MPa) may seem high, it
was easily achieved by the formula designed to meet the permeability
requirement, since the site soils were high in sand and silt content.

A third and very challenging example can be found at the Stauffer Chemical
site in Florida, USA. This site formulated insecticide and herbicide products
into dusts, granules, pellets, and liquids that were packaged for commercial
239
distribution. Compounds such as kerosene, xylene, clay, and diatomaceous
earth were also handled on site.

The primary remedial approach was on-site excavation of contaminated soils


and placement into an RCRA equivalent lined and capped containment cell
located on-site.

Figure 7.5: Constructing a S/S sub-cap, Brunswick

240
Figure 7.6: Constructing a vertical wall using a Bauer Panel Cutter,
Brunswick

During the remedial operation, a substantial amount of groundwater was


collected and treated to facilitate the excavation. Upon preparing to
decommission the temporary water treatment system, it was revealed that a
substantial amount of pesticide and herbicide contaminated oily water
emulsion had collected as a residual in some of the water treatment tanks
(Figure 7.7). By employing treatability studies, it was determined that these
contaminated emulsions could be solidified using an excavator mixing on an
existing concrete pad.

The formula used was (by volume): 100 parts of oily water, 200 parts site
sand, 25 parts proprietary oil absorbent, 5 parts cement, and 2 parts water
absorbing polymer. In this case, solidification alone was used without
stabilization, to change the oily contaminated water emulsion into a soil-like
material. It was not necessary (though it would have been possible) to turn the
waste into a monolithic material.

However by converting the emulsions into a soil-like material (Figure 7.8), it


was feasible to dispose this waste, as a solid, in the RCRA C-equivalent
containment cell that was already being constructed on-site for other
contaminated soils.

241
Figure 7.9 shows installation of the FML as part of the construction of the
RCRA C-equivalent disposal cell. There were no specifications for hydraulic
conductivity, strength, or leaching required for the solidified treated material,
due to its being placed into a fully containing RCRA cell.

Figure 7.7: Oily water emulsion at Stauffer

Figure 7.8: Solidifying oily water emulsion Stauffer

242
Figure 7.9: Installing an FML at Stauffer

These three examples illustrate how the integration of multi-barrier systems


with S/S can substantially change the performance specifications required
from the S/S treated material. The level of change to the performance
specifications will, however, vary depending on the given project and
regulatory environment and the overall performance required for the site.

243
8.0 Developing effective S/S formulations

8.1 Common reagents


Common non-proprietary products used for S/S include Portland cement,
quick and hydrated lime, cement kiln dust, limekiln dust, fly ash, bottom ash,
magnesium oxide, phosphate in various forms, sulfides, and carbon-based
reagents.

Combinations of these reagents are available under various proprietary


names, but this section will discuss the basic generic reagents. Information on
sourcing reagents is provided in Section 8.2.

8.1.1 Portland cement


Portland cement is made by heating a mixture of limestone with clay, shale,
sand, iron ore, bauxite, fly ash and/or slag to about 1450C in a kiln. The raw
materials are sintered to form a clinker which contains 70-80% calcium
silicates, with over 60% of these silicates as tricalcium silicates, and trace
amounts of tricalcium aluminate, tetracalcium aluminoferrate, magnesium
oxide, calcium oxide, and alkaline sulfates (e.g. Na 2SO4 and K2SO4).

The cooled clinker is mixed with calcium sulfate (gypsum) and pulverized. It is
considered a hydraulic binder as it sets and hardens by chemical interaction
with water. Portland cement retains its strength and stability after hardening,
even under water.

Portland cement relies on both pH control and chemical reactions to


chemically immobilise metals. pH control is important because many metals
form low solubility hydroxides and stabilized materials which have a leachate
pH in the range of 8 to 11 and typically have low levels of leachability. The
efficacy of Portland cement as a binder for use in S/S is discussed in the
companion volume of this work (Hills et al., 2014).

Many of the metals can also isomorphically substitute into the calcium silicate
hydrates (C-S-H) and other phases, which form when Portland cement reacts
with water. The incorporation of the metals into these compounds decreases
their availability. Furthermore, the cemented matrices can encapsulate free-
phase organics or organics sorbed to waste particles, and there is evidence
244
that in some circumstances solidorganic salts are precipitated and
encapsulated in the hardened S/S waste form.

Portland cement is delivered as a dry powder, in bulk pneumatic trailer, bags,


or 1-ton jumbo bags. Due to its reaction with water, Portland cement must be
stored in a manner to prevent contact with water in liquid or vapour form. The
interaction with water vapour means that Portland cement will degrade over
time and its storage on-site is time restricted (less than 1 month). Portland
cement that has become lumpy has lost a significant portion of reactivity and
should not be used for S/S.

In bulk, the Portland cement powder is pneumatically transferred to silos,


equipped with bag-houses to contain dust during the transfer, for storage. The
silo should be well sealed to prevent contact with liquid water. The surface
area to volume ratio for Portland cement in the silo is low, minimising
exposure to water vapour. Portland cement delivered in bags or jumbo bags
should be stored on wooden boards above a plastic barrier and covered with
plastic.

Portland cement can cause health effects by skin contact, eye contact, or
inhalation. Risk of injury depends on duration and level of exposure and
individual sensitivity. Portland cement contains traces of calcium oxide (that
is corrosive to human tissue), crystalline silica (that is abrasive and can
damage lungs), and hexavalent chromium (that can cause allergic reactions).

8.1.2 Quicklime and hydrated lime


Quicklime (CaO, or burnt lime) is produced by the calcination of limestone
above 825C. Due to aluminium, iron and/or silicate impurities in the
limestone, the quicklime is often clinkered during the heating process, and
thus grinding is often employed to produce a dry powder. Hydrated lime
(Ca(OH)2, or slaked lime) is produced by the reaction of quicklime with water
(otherwise known as slaking). The slaked lime is then thermally dried at
approximately 200C and ground to a fine powder.

Quicklime and hydrated lime (Portlandite) impart pH control to chemically


immobilise metals. The calcium hydroxide gel, formed by reaction with water,
can also provide a medium for the sorption of organics of low water solubility.

245
Quicklime is delivered in bulk as both pebble and dry powder, by pneumatic
trailer, 50 or 90-pound bags (23 or 45kg), or 1-ton bags. Due to its exothermic
reaction with water, quicklime must be stored dry. The interaction with liquid
water can be both exothermic and violent, and the reaction with water vapour
means that quicklime will degrades with time, and its storage on-site should
be brief (less than 1 month). However, unlike Portland cement, quicklime that
has reacted with water can be used for S/S purposes, though its efficacy is
slightly reduced.

Hydrated lime is delivered as a dry powder, in a bulk pneumatic trailer, 50 or


90 pound (23 or 45kg) bags, or 1-ton bags. Although hydrated lime contains
water, it should be stored in a manner to prevent contact with liquid water.
This is because water will degrade its handling properties. However, unlike
quicklime or Portland cement, hydrated lime can be stored on-site for several
months, and when in contact with water, can still be used for S/S purposes,
though its handling properties as a powder will be affected.

In bulk, powdered quicklime or hydrated lime are pneumatically transferred to


silos, equipped with baghouses to contain dust emissions during the transfer
to storage. The silo should be well sealed to prevent contact with liquid water.
Bulk pebble quicklime can be delivered in dump trailers and should be used
immediately upon delivery. Quicklime (powder or pebbles) and hydrated lime
delivered in bags should be stored on wooden boards above a plastic barrier
and covered with plastic.

Quicklime can cause serious harm to health by dermal or eye contact and
inhalation. Its very alkaline nature and energetic reactivity with water can
pose a serious risk of injury. Similarly, hydrated lime is hazardous due to its
alkalinity. Figure 8.1 shows hydrated lime being added to waste in a mix pit.

246
Figure 8.1: Hydrated lime added to waste in a mix pit

8.1.3 Cement kiln dust


Cement kiln dust (CKD) is the fine-grained powder removed from the exhaust
gas of a cement kiln by its air pollution control devices. The CKD contains
unreacted raw materials from the Portland cement production and is largely
recycled back into the production process.

The CKD that is not recycled is typically disposed of to landfill; some is also
used beneficially for S/S. CKD contains calcium silicates (though more
dicalcium silicate than tricalcium silicate) and calcium carbonates, with trace
amounts of tricalcium aluminate, tetracalcium aluminoferrate, magnesium
oxide, calcium oxide, and alkaline sulfates.

As with Portland cement, CKD relies on both pH control and chemical


reactions to immobilise metals. The formation of low solubility hydroxides
along with isomorphic substitution into the calcium aluminosilicates forming on
contact with water decreases the leachability of certain metals. The
cementitious matrices can also encapsulate free-phase organics or organics
sorbed to waste particles.
247
CKD is delivered in bulk as a dry powder in bulk pneumatic trailer, or 1 ton
bags. Like Portland cement, CKD must be stored dry to prevent degradation
on contact with water in liquid or vapour form. Site storage should be brief
(less than 1 month), and CKD that has become lumpy will have lost a
significant portion of reactivity and should not be used for S/S purposes.

As for Portland Cement CKD is pneumatically transferred to silos, equipped


with dust control and sealed to prevent contact with water. CKD delivered in
bags should be stored on wooden boards above a plastic barrier and covered
with plastic.

CKD can cause serious harm via skin and eye contact, or by inhalation. The
risk of injury depends on the duration and level of exposure, but as CKD
contains traces of calcium oxide (that is corrosive to human tissue), crystalline
silica (that is abrasive and can damage lungs), and hexavalent chromium (that
can cause allergic reactions), exposure should be minimised.

In the United States, CKD is categorised by EPA as a "special waste" and has
been temporarily exempted from federal hazardous waste regulations under
Subtitle C of the Resource Conservation and Recovery Act (RCRA). The US
EPA is in the process of developing standards for the management of CKD
and has published a set of proposed Subtitle D (i.e. non-hazardous, solid
waste) regulations to govern CKD management. In Europe, CKD falls under
REACH regulation, and can be re-incorporated into cement products or be
disposed appropriate to its hazardous/non-hazardous nature (see: www.
Defra.gov.uk/environment/chemicals/reach/).

8.1.4 Lime kiln dust


Lime kiln dust (LKD) is the fine-grained powder removed from the exhaust gas
of a lime kiln by its air pollution control devices. LKD is largely unreacted or
partially reacted limestone; it is generally recycled back into the quicklime
production process. The LKD not returned to the production process is sold
for beneficial reuse, such as a reagent for S/S. LKD is a combination of
calcium carbonate and calcium oxide, with traces of aluminium, iron, and
silicate minerals.

LKD relies primarily on pH control to chemically immobilise metals. The


calcium hydroxide gel formed by reaction with water can also facilitate
sorption of organics with low water solubility.
248
LKD is delivered as a dry powder, in bulk pneumatic trailer or 1 ton bags. It is
reactive with water, and should be dry-stored. LKD that has come in contact
with water can be used for S/S purposes, though its handling properties as a
powder will be affected.

In bulk, LKD is pneumatically transferred to silos, equipped with baghouses to


contain dust during the transfer, for storage. The silo should be well sealed to
prevent contact with water. LKD delivered in bags should be stored on
wooden boards above a plastic barrier and covered with plastic.

Similar to quicklime or hydrated lime, LKDs highly alkaline nature can cause
serious harm to health.

8.1.5 Fly ash


Fly ash is produced in boilers burning pulverised coal and is removed from the
boiler exhaust gases by electrostatic precipitators, baghouses, or scrubber
systems. Fly ash is formed either by the agglomeration of mineral inclusion
from the coal into hollow spheres (cenospheres) or by the condensation
outside of the combustion zone of mineral inclusions vapourised in the
combustion zone.

Fly ash is a heterogeneous mixture of silicon, iron, and aluminium oxides,


carbon, and calcium oxide. Because it is collected from the gas phase, fly ash
particles are typically smaller than 200 m in diameter and the mean particle
diameter for fly ash is often less than 50 m.

Class C fly ash (produced from the burning of lignite or subbituminous coal)
relies on both pH control and chemical reactions to chemically immobilise
metals. pH control is provided by the CaO in the Class C fly ash. But many of
the metals can also isomorphically substitute into the calcium aluminosilicates
which form as the fly ash particles interact with calcium hydroxide, either
produced from the reaction of Class C fly ash with water or by calcium
hydroxide added in conjunction with the Class C fly ash.

Class F fly ash (produced from the burning of anthracite or bituminous coal)
relies primarily on chemical reactions to chemically immobilise metals. Class
F fly ash typically contains less than 20% CaO and little to none of this is
actually present as CaO (free lime). Metals can be isomorphically
249
substituted into the calcium aluminosilicates which form as the fly ash
particles interact with calcium hydroxide that is added in conjunction with the
Class C fly ash.

High carbon fly ash (with greater than 10% loss on ignition) can be used for
the stabilization of organics. The carbonaceous component of these high
carbon fly ashes can sorb free-phase organics or organics sorbed to waste
particles, reducing their leachability.

Fly ash is delivered as a dry powder, in bulk pneumatic trailer or dump trailers.
Pneumatic trailers can be off-loaded to silos or pneumatically transferred to
covered stockpiles. Dump trailer loads are typically dumped into stockpile
areas and covered or dumped at the point of use and used immediately.

Due to its fine nature, fly ash is very dusty and may be reacted with water to
reduce its dusting. In bulk, fly ash is often pneumatically transferred to silos,
equipped with baghouses to contain dust during the transfer, for storage. The
silo should be well sealed to prevent dust emissions during transfer.

The fine crystalline silica within the fly ash can cause damage to lungs if
inhaled.

A temporary fly ash stockpile is shown in Figure 8.2. The cover was
temporarily removed so that a front loader could access the stockpile.

250
Figure 8.2: Fly ash stockpile on a treatment site with its cover removed

8.1.6 Bottom ash


Bottom ash is also produced in boilers burning pulverised coal. It consists of
spherical or angular agglomerated mineral inclusion from the coal, the
particles of which were too large to be carried in the flue gases.

Bottom ash is a heterogeneous mixture of silicon, iron, and aluminium oxides,


carbon, calcium sulfate, and calcium oxide.

Bottom ash produced from burning coal in the presence of limestone relies on
both pH control and chemical reactions, but primarily the latter, to chemically
immobilise metals. pH control is provided by the CaO in the bottom ash, but
many of the metals can also isomorphically substitute into the calcium
aluminosilicates. These form as the bottom ash particles interact with calcium
hydroxide, either produced from the reaction of bottom ash with water or by
calcium hydroxide added in conjunction with the bottom ash. Bottom ashes
typically contain less the 20% CaO and little to none of this is actually present
as CaO (free lime).
251
High carbon bottom ash (with greater than 10% loss on ignition) can be used
for the stabilization of organics. The carbonaceous component of these high
carbon bottom ashes can sorb free-phase organics or organics sorbed to
waste particles, reducing their leachability.

Bottom ash is delivered as a dry solid in dump trailers. The trailers are
typically dumped into stockpile areas and covered or dumped at the point of
use and used immediately. Due to its sandy particle size, bottom ash has little
dust associated with it.

The fine crystalline silica within the bottom ash can cause damage to lungs if
inhaled.

8.1.7 Magnesium oxide


Magnesium oxide (MgO) is found naturally as periclase, but is more
commonly produced by the thermal decomposition of magnesium carbonate
or magnesium hydroxide. Due to the presence of aluminium, iron and/or
silicate impurities, MgO is often pelletised and then ground to produce a dry
powder.

The effective use of MgO relies on both pH control and chemical reactions to
immobilise metals. The control of pH is important as it promotes the formation
of low solubility metal hydroxides. Stabilized materials with pore solution pHs
in the range of 8 to 11 typically have low levels of metal leaching. The use of
MgO stabilizes pore solution pH near 10. Furthermore, many metals can
isomorphically substitute into magnesium aluminosilicate, which often forms
when MgO is added to soil, with a consequent reduction in leaching.

Magnesium oxide is delivered as a dry powder, in bulk pneumatic trailer or in


1-ton bags. Magnesium oxide should be stored in a manner to prevent contact
with water, to prevent degradation of its handling properties. In bulk, MgO is
pneumatically transferred to silos, equipped with dust-control facilities.
Storage silos should be well sealed to prevent contact with water, and when
delivered in bags these should be stored on wooden boards above a plastic
barrier and be completely covered with plastic sheeting.

Magnesium oxide can cause health effects by skin contact, eye contact, or
inhalation, due to its highly alkaline nature.

252
8.1.8 Phosphates
Phosphates form low solubility salts with many metals, including aluminium,
cadmium, iron, and lead. This fact makes phosphates a useful reagent for the
chemical fixation of metals. There are a number of different phosphate
materials that are typically used for stabilization treatments. They are listed
below, along with information on the material, and its delivery and handling.

Phosphoric acid
Phosphoric (or orthophosphosphoric) acid (H3PO4) is a mineral acid. It is
typically available in aqueous solutions of 75-85%. At these concentrations, it
is a clear, odourless, syrupy liquid.

Phosphoric acid is delivered as an aqueous solution in chemical totes or by


bulk chemical tanker. Bulk chemical tanker loads are typically transferred by
pump to a holding tank or vessel for storage.

Because it is a concentrated acid, phosphoric acid is corrosive and can cause


health effects by skin contact, eye contact, or inhalation.

Calcium phosphates
Calcium phosphates are a name given to a family of compounds containing
calcium and orthophosphate (PO43-). These include monocalcium phosphate,
dicalcium phosphate, tricalcium phosphate, hydroxyapatite (Ca 5(PO4)3(OH)),
and apatite (Ca10(PO4)6(OH, F, Cl, Br)2). Triple superphosphate, often called
TSP, is a concentrated calcium monophosphate used as a common
phosphate fertiliser and is typical of a calcium phosphate used for stabilization
treatment.

Calcium phosphates are typically delivered in 1-ton bags. Calcium


phosphates should be stored in a manner to prevent contact with liquid water,
as water will degrade its handling properties. These bags should be stored on
wooden boards above a plastic barrier and covered with plastic or their
contents transferred into silos for storage.

Calcium phosphates can cause health effects by skin contact, eye contact, or
inhalation.

253
Sodium phosphates
Sodium phosphates are a name given to a family of compounds containing
sodium and orthophosphate (PO43-). These include monosodium phosphate,
disodium phosphate, trisodium phosphate, and sodium aluminium phosphate
(Na8Al2(OH)2(PO4)3). Typically, trisodium phosphate has been utilised for
stabilization treatment.

Like calcium phosphates, sodium phosphates are typically delivered in 1-ton


bags. Sodium phosphates should be stored in a manner to prevent contact
with liquid water, as it will degrade its handling properties. These bags should
be stored on wooden boards above a plastic barrier and covered with plastic
or their contents transferred into silos for storage.

Sodium phosphates can cause health effects by skin contact, eye contact, or
inhalation.

Bone phosphates
Bone phosphates are bone residues that have been treated with caustic and
acid, then neutralised with lime and dried.

Bone phosphates are typically delivered in 50-pound (23kg) bags or 1-ton


bags and should be stored in a manner to prevent contact with liquid water, as
water will degrade its handling properties. The bags should be stored on
wooden boards above a plastic barrier and covered with plastic or their
contents transferred into silos for storage.

8.1.9 Sulfides
Like phosphate, sulfides form low solubility salts with many metals, including
cadmium, iron, lead, and mercury. This makes sulfides useful as a reagent for
the chemical fixation of metals. There are a number of different sulfide
materials that are typically used for stabilization treatments. They are listed
below, along with information on the material, and its delivery and handling.

254
Sodium sulfide
Sodium sulfide commonly refers to the hydrated version of Na 2S and is
commercially available as an unspecified hydrate (Na 2SxH2O) with the
weight percentage of Na2S specified. The commercially available grades are
produced by the reduction of Na 2SO4 with carbon and typically contain
approximately 60% Na2S by weight.

Sodium sulfide is typically delivered in 50-pound bags (23kg), or 1-ton (0.9


tonne) bags and should be stored in a manner to prevent contact with liquid
water, as water will degrade its handling properties. The bags should be
stored on wooden boards above a plastic barrier and covered with plastic or
their contents transferred into silos for storage.

Sodium sulfide is highly alkaline and can cause health effects by skin contact,
eye contact, or inhalation. Reaction of sodium sulfide with acids can liberate
highly toxic hydrogen sulfide (H2S) gas.

Calcium polysulfide
Calcium polysulfide typically refers to a family of calcium sulfides (CaS x)
produced by reacting calcium hydroxide with sulfur, leading to its common
name of lime sulfur. Commercially, calcium polysulfide is available as a
solution with a specified weight of CaSx, typically 29% by weight.

Calcium polysulfide is delivered as an aqueous solution in chemical totes (an


IBC container) or by bulk chemical tanker. Bulk chemical tanker loads are
typically transferred by pump to a holding tank or vessel for storage.

Calcium polysulfide is highly alkaline and is corrosive. It can cause health


effects by skin or eye contact. Calcium polysulfide has a distinct smell of
rotten eggs. Reaction of calcium polysulfide with acids can liberate highly
toxic hydrogen sulfide (H2S) gas.

8.1.10 Carbon-based reagents


For metals, chemical fixation (stabilization) often involves the precipitation or
re-precipitation of soluble metal species as less soluble species, as
hydroxides or sulfides. The immobilisation of the hazardous metals in less
soluble species slows the potential release of the hazardous constituent into
the environment and lessens the material's impact on the environment. For
255
organic-contaminated wastes, reactions that alter the organic compound or
physical processes such as adsorption and encapsulation are used to retard
the movement of the hazardous constituents. Carbon-based materials can
often be used as chemical reagents to provide organophilic surfaces for the
adsorption and/or reaction of organic contaminants.

Organoclay
Organoclay is a name given to a family of naturally occurring clay material
intercalated with organo-cations, typically quaternary alkylammonium ions.
Intercalating these organo-cations to the ion exchange sites on the internal
surfaces of the clay platelets produces an organophilic layer between clay
platelets. Organic compounds can partition into and be adsorbed within this
organophilic layer, allowing organoclays to sequester organics. This property
makes organoclays useful reagents for the chemical fixation of organics. The
specific capacity for partition and adsorption of organics depends both on the
organic compound to be adsorbed and the organo-cation (s) employed by the
specific organoclay.

Organoclay is delivered as a dry powder, in 1-ton bags. Organoclay should be


stored in a manner to prevent contact with liquid water, as water will degrade
its handling properties. Organoclay bags should be stored on wooden boards
above a plastic barrier and covered with plastic.

Organoclays can cause limited health effects by inhalation or by skin or eye


contact.

Activated charcoal and carbons


Activated charcoal and carbons are produced from natural materials such as
coir (coconut fibre), lignite coal, nutshells, peat, and wood, by a combination
of physical and chemical processes. These starting materials may be
pyrolysed in the absence of oxygen (carbonisation), and then activated in an
oxidising environment above 250C. Often, the raw material is amended with
mineral acids or salts to reduce the carbonisation temperature. The activation
process produces a final material that is extremely porous and has a high
surface area (>500 m2/g) for adsorption. Since activated charcoals and
carbons are organophilic and have a high surface area for adsorption, these
materials are useful reagents for the chemical fixation of organics.

Activated charcoal or carbon is delivered as a dry powder, or granules, in bulk


pneumatic trailer or 1-ton bags (Figure 8.3). In bulk, charcoal and activated
256
carbon is pneumatically transferred to silos, equipped with baghouses to
contain dust during the transfer, for storage. Activated charcoal or carbon
delivered in bags is typically stored on wooden boards above a plastic barrier
and covered with plastic, though these reagents are not adversely affected
when contacted with water.

Activated charcoal or carbon can cause limited health effects by inhalation or


by skin or eye contact.

Figure 8.3: Granular activated carbon

8.2 Reagent sourcing


Locating reagent sources is an important step in the design process for S/S
treatment. Based on the performance goals for the S/S treatment, potentially
applicable reagents need to be established.

8.2.1 Proprietary reagents


Proprietary reagents can only be purchased from their manufacturer or its
distributer(s). However, non-proprietary reagents such as Portland cement,
lime products, and fly ash can be sourced from multiple vendors. Literature
and Internet searches can provide information on proprietary reagents that
may be of interest.
257
Information on manufacturers of most chemicals (e.g. sulfides, phosphates,
etc.) can also be garnered on-line from sites such as Alibaba.com
(http://www.alibaba.com/), ChemNet (http://www.chemnet.com/), or
ThomasNet (http://www.thomasnet.com).
.

8.2.2 Mineral and waste reagents


For most mineral and waste reagents, commercial associations or
environmental agencies offer contact or listing information. These include:

Portland cement and cement kiln dust:

The Portland Cement Association in the United States


(http://www.cement.org/)
the Cement Association of Canada (http://www.cement.ca/)
the Mineral Products Association in Britain
(http://www.mineralproducts.org/)
the CEMBUREAU in Europe (http://www.cembureau.be/)

all provide information on the plant and sales office locations of their
members. Since all Portland cement manufacturers create cement kiln dust,
contacting the nearby cement plants can help identify locally available
sources.

Quicklime, hydrated lime and lime kiln dust:


The National Lime Association in the United States and Canada
(http://www.lime.org/index/)
the British Lime Association (http://www.britishlime.org/)
the EuLA in Europe (http://www.eula.eu/)

provide information on the plant and sales office locations of their members.
As all lime manufacturers create lime kiln dust, contacting the nearby lime
plants can help identify locally available sources.

258
Fly ash and bottom ash these materials are produced in coal-fired power
plants:
For the United States and Canada, the respective national
environmental agencies (US Environmental Protection Agency and
Environment Canada) maintain lists of the coal-fired power plants
operating within those countries
The IEA Clean Coal Centre (http://www.iea-coal.org.uk/site/2010/home)
maintains a list of coal-fired power plants in its member countries
(Australia, Austria, Canada, EC, Germany, Italy, Japan, Poland, South
Korea, South Africa, the United Kingdom and the United States)

Wikipedia also contains similar lists of the operating coal-fired power plants in
most countries (http://en.wikipedia.org/wiki/List_of_coal_power_stations),
though care should be taken to verify the reliability of the information posted
on this site.

Bentonite is produced worldwide in Argentina, Brazil, Canada, Cyprus,


Germany, Greece, Hungary, India, Italy, Japan, Mexico, Poland, Romania,
Spain, Turkey, and the United States:
The Wyoming Mining Association
(http://www.wyomingmining.org/minerals/bentonite/bentonite-members/)
the European Bentonite Producers Association
(http://www.ima-europe.eu/about-ima-europe/associations/euba) have
membership lists, which provide points of contact
Information on bentonite manufactures can also be garnered on-line
from sites such as www.Alibaba.com or www.thomasNet.com

8.2.3 Importance of location


Transportation costs are often on the order of 20% of the total delivered
reagent pricing for mineral manufactured products such as Portland cement
and lime. For waste products, such as fly ash or cement or limekiln dusts,
transportation can account for over 50% of the delivered pricing. Therefore,
the identification of local or nearby sources of applicable mineral or waste
reagents is necessary to minimise reagent costs for the S/S treatment. It is
important to emphasise that the specific sources of waste products
contemplated for use in the S/S project be identified, and only these sources
be used for waste product reagents used in the treatability testing (Sections
8.3 and 8.4).

259
8.2.4 Reagent samples for testing
Many vendors will supply a sample of their material for testing purposes.
While this is not as important for commercial chemical or manufactured
materials (e.g. Portland cement or lime products), this is very important for
waste (e.g. fly ash, bed ash, cement or lime kiln dust) or off-specification (e.g.
spent carbon), as the properties of these materials vary widely from source to
source. Testing with the specific waste or off-specification material is
necessary to ensure that these materials will perform as anticipated with the
specific waste material to be treated.

8.3 Bench-scale treatability design and testing


Treatability testing for S/S is conducted at both the bench (laboratory)-scale
and in the field at full-scale (pilot testing). This section describes bench-scale
testing procedures, which are conducted before field- or pilot-scale testing.

Several excellent references for conducting S/S treatability tests are available
in the literature (ITRC 2011, Environment Agency 2004, USEPA 1992,
USACE 1995, EPRI 2009). The remainder of this section will focus on
conduct of bench-scale treatability tests. The reader may want to refer to
Section 8.4 for a discussion of field- pilot- or full-scale tests, Section 8.5 for a
discussion on selecting the best sample for treatability testing, Section 6.3 for
a discussion of sample preparation and curing, Section 7.1 regarding the
setting of overall goals for treatment, Section 7.2 regarding the importance of
the site conceptual model, and Section 7.3 for a discussion of test and
analytical methods

A site specific bench-scale treatability test is needed to obtain essential


information to evaluate the feasibility of S/S to treat the contaminated
material, support selection of the remedy, establish treatment design
parameters, develop the most cost effective formula and obtain information
needed for scale-up to field tests.

Treatability testing for S/S may require 3-6 months, or more, to collect
samples, select reagent formulas, perform the mixing, allow the treated
samples to cure, analyse the results, and repeat the process at several
tiers/levels of testing to derive the best and most cost effective treatment
formula. Treatability testing is a systematic process to assure that the data
generated is fit for the evaluation, identification of information for further
testing, and development of full-scale specifications.

260
Analytical testing during bench-scale testing can be quite expensive if the
practitioner is not experienced in selecting appropriate formulas to test, and in
designing the sequence of testing activities. In addition, the collection of
representative samples for bench-scale treatability testing is highly critical for
the success of an S/S project, as selection and testing of the wrong materials
from site will result in misleading results and the possibility of costly failure
during full-scale treatment (see Section 8.5 regarding selection of the correct
site material for treatability testing).

S/S bench-scale treatability studies usually include the following steps (after
ITRC 2011):

1. Prepare a work plan


2. Collect test samples
3. Characterize the Initial Sample
o Homogenize raw materials
o Perform physical testing
o Perform chemical testing
4. Perform treatability testing
o Identify appropriate reagents
o Conduct testing by mixing reagents with contaminated material
and prepare formulations for further testing
o Optimize mix design
o Selection of mix design verification phase
o Prepare final mix design and test
5. Analyse, Assess and Validate Data
6. Prepare treatability study report

8.3.1 Bench-scale treatability study objectives


Prior to performing a bench-scale treatability test, the objectives of the testing
should be identified. Typical objectives include:

Demonstrate that the technology can meet remedial performance


objectives
Selection of appropriate reagent(s)
Determine the most economical treatment formula (reagent(s) and
dosage)
Determine the impact of selected reagent(s) on contaminants and
chemicals present in the waste material
Develop treatment parameters and level of process control required
Identification of contaminant/chemical emissions
261
Identification of material handling issues and associated safety
problems
Assessment of physical and chemical uniformity of the treated material
Estimate volume increase due to addition of reagent(s)
Establish full-scale performance evaluation parameters and criteria
Develop more accurate cost estimate
Assure that the formula(s) developed can successfully treat all
variations of soil and contaminant mixtures on the site

8.3.2 Importance of the conceptual implementation plan


A conceptual implementation plan should be prepared prior to conducting the
treatability testing. This should address the proposed means of achieving the
remedial objectives, including the areas for treatment, type of contamination
to be treated, depth of treatment, type of equipment to be used for mixing
reagents, etc.

The concept of how the remedial operation will be implemented is critical to


designing the bench-scale treatability study. For example if the concept is to
use in-situ augers to carry out S/S, then a candidate slurry-based reagent
formula should be prepared, with a viscosity that is suitable for that purpose.

Thus, bench-scale testing will employ slurries (not solid reagents) to be mixed
with the contaminated materials taken from site. However, if the concept is to
mix using a pug-mill, then bench-scale testing should employ dry reagents,
and supplemental water as necessary to achieve the desired final mix
consistency.

The desired final mix consistency, especially moisture content, may depend
on the transport and placement of the treated material. By way of example, if
the treated material exits a pug-mill into on-site dump trucks (for transport to
and compaction at the disposal area), then the material should not exit the
pug-mill as a wet slurry (or it will not be capable of transportation and
compacted as desired).

The treated material should however contain sufficient moisture to facilitate


the chemical hardening reactions and facilitate compaction. Thus, the
objective of the bench-scale treatability testing is to replicate as closely as
possible the full-scale treatment, mixing, and placement of contaminated
materials. Thus, development of a conceptual field implementation plan
should precede the bench-scale treatability testing process.
262
8.3.3 Reagent selection considerations
The site remedial investigation data should be reviewed and potential
reagents identified prior to treatability testing. The selection of potential
reagents for testing depends on several factors, including the:

contaminants to be treated
concentration of contaminants detected
geotechnical properties of the waste material
agreed performance parameters
minimum acceptable performance criteria for the site

The identification of potential reagents often relies on a practitioners


experience. In the absence of previous experience, a survey of the technical
literature is a good starting point, where a number of reagents may be
identified. The narrowing down of possible binders (reagents) will result in a
lower treatability testing cost, over a shorter timescale. The successful
selection of candidate reagents benefits greatly from experience. Selection of
reagents relies upon knowledge of (after ITRC, 2011):

Previous successful use of a reagent (a track record)


Interference and chemical incompatibilities
Organic and physical chemistry potential reactions
Compatibility or reagents with the disposal site or reuse environment
Availability of the reagent
Cost of the reagent

A more detailed discussion on the above considerations is provided by the


USEPA (EPA 1993), but it should be noted that the reagent used in the
treatability testing should be exactly the same as planned for use during the
full-scale remediation. For example a sample of F fly ash from plant X
collected historically and archived in a laboratory stockroom may not perform
the same as a currently produced F fly ash from the same source, or from a
different source!

8.3.4 Bench-scale treatability testing approaches


Bench-scale testing is most often conducted in the laboratory, though
occasionally a limited test is conducted in the field. Relatively small amounts
of site-contaminated material are treated using an assortment of candidate
reagents and dose rates in order to determine the most cost effective
263
formulation to achieve performance goals. Bench-scale testing can be
conducted using a tiered approach (often the best), a shotgun approach, or a
focused problem approach, as described below.

The tiered approach to bench-scale testing is a step wise approach, and is


often the best for producing the most cost effective formula with the lowest
treatability testing cost. It does takes longer, however, and usually requires at
least 3 to 6 months or more for difficult to treat wastes. The results from each
test tier are used to determine the subsequent steps and the next set of
formulations to be evaluated. There are several different ways to define the
tiers, but the following example is one that the authors have found usually
works well. The four tier design process is illustrated in Figure 8.4.

264
Figure 8.4: Design of S/S formulations

265
Tier 1: Reagent screening involves the candidate reagents being tested
individually and/or combined as initial formulations.

The objective is to assess the beneficial effect of the reagents and to establish
upper and lower boundaries on the % of reagent needed for strength
development. Generally, the focus is on achieving strength. Permeability may
sometimes be evaluated on selected formulations that meet strength
requirements.

Leaching is normally considered in Tier 2 or 3 testing. Tier 1 testing is


specifically intended to highlight the most promising reagents, and determine
the minimum and maximum dosages to be evaluated in step 2.

For example in Figure 8.4, Tier 1, if we assume the data from strength testing
of formulas 1, 2, 4, and 6 have failed to achieve performance objectives, and
formula 8 has achieved 5 times the strength objective, it is possible to
extrapolate that 5-8% w/w cement addition alone is not sufficient, and the use
of Class F fly ash did not make a significant difference, this latter reagent can
be eliminated from further consideration.

However the inclusion of slag with cement did make a difference and formula
8 represented an excessive reagent dosage. Thus, the inclusion of
permeability testing of formulas that met strength requirements might be
beneficial, or (alternatively) one can delay all permeability testing to Tier 2.

Figure 8.5 shows UCS testing on cured cylindrical samples. Filter paper is
sometimes added above/below the coupon/sample to determine if any free
liquids were released whilst the specimen was subject to testing under
pressure.

Tier 2: Formulation refining involves refining the reagents/formulations


found to be effective in Tier 1 to ensure that all physical properties can be
achieved and leaning (reducing binder content) of the formulations to reduce
reagent costs. The most economical formulations that meet all the physical
requirements can be leach tested and examined to determine whether
additional reagent dosage, or possibly that new reagents, are need to achieve
leaching targets.

266
In Figure 8.4, Tier 2, formulas 3 and 5 were carried forward from Tier 1, while
formulas 9, 10, and 11 were modifications of 3 and 5. The original formula 7
without fly ash was formula 5, and since, in our example, fly ash was
determined not to be beneficial, formula 7 was not carried into Tier 2. Formula
12 is the original formula 4 (also without the fly ash) but with the amount of
slag being doubled.

Figure 8.5: Unconfined compressive strength testing

In our example, we can assume all mixes meet the strength criterion but
formulas 3 and 9 fail on permeability (a typical permeability testing-rig is
shown in Figure 8.6). We can conclude that slag needs to be used (with
cement) in the mix design to meet target physical properties. Those mixes
267
that passed our physical criteria (formulas 5, 10, 11, and 12) are then leach
tested, in which formulas 5 and 11 fail. Formulations 10 and 12 both meet
leaching criteria.

On analysis, formulas 10 and 12 have lower cement contents (and


consequently, lower pHs which can sometimes be important with
immobilising amphoteric metals), but also contain higher slag contents.
Although both formulas 10 and 12 meet all our criteria and could go forward to
Tier 4 testing, it may be possible to reduce reagent dose rates, save treatment
cost, and still meet our criteria. To explore this further, testing at Tier 3 can be
used to determine the most cost effective formula.

Figure 8.6: Permeability testing


268
Tier 3: Formula optimisation
It is tempting to accept formula 10 as the cheapest available. However,
significant cost savings can be made if formula 10 can be adjusted to maintain
performance but with a reduced cement and/or slag content. Tier 3
formulations are based upon mix design formula 10, but with reduced cement
and/or slag content, hopefully achieving all criteria with considerable cost
savings on treatment at full-scale.

As formula 10 has already proven successful it is not necessary to fully test it


again in Tier 3. It is however shown in Figure 8.4, Tier 3, for reference as it
has already passed all our criteria. The mix variations of formula 10 are first
strength tested, where formula 17 fails. Formulations 13-16 are examined for
permeability, where formula 14 fails. Formulas 13, 15, and 16 are then leach
tested, and all pass the established criteria. The four mix designs passing Tier
3 are formulas 10, 13, 15, and 16. The lowest cost formulations are mix
designs 13 and 16, and one of these is chosen and carried forward to Tier 4
testing.

Tier 4: Verification
At Tier 4 the apparent lowest cost successful formula (i.e. meets all our
treatment criteria) is re-mixed and tested 3 times to ensure reproducibility.
Thus, repeated mixing, curing, and testing this formula using three replicate
actions and sampling each mix is undertaken, not mixing once and then
taking 3 sample sets. Should the results not meet all criteria, then it is
necessary to return to Tier 3 to verify the second lowest-cost formulation that
met the established performance criteria.

At some sites, where contaminated soil is variable, it is possible to determine


that there exists more than one reasonable worst case soil/contaminate type
to be treated (see Section 8.5). Under these circumstances, in Tier 4, the
successful, selected formula is applied to these other soil/contaminant types
to verify performance criteria are met. If not, then it may be necessary to
develop a new formulation for these soils. The conclusion of Tier 4 testing is
a successful, low-cost binder formulation for full-scale field pilot testing.

269
The Shotgun approach.
The shotgun approach describes a mix designing process that is used when
treatability testing is required at short notice. A large number of formulations
are mixed, cured and tested, all at once. This approach takes less time, but
generally costs are higher (more mixes are tested and analysed). It should be
noted that this approach might not produce the most cost effective formula for
field application, and one not suitable sometimes for difficult to treat wastes.
However, if time is not available for a tiered approach, the skill of the
practitioner can be used to produce an effective binder formulation in about 6
weeks.
With 24 or more binder formulations to test against performance criteria,
strength, permeability, and leaching tests could be conducted concurrently, or
sequentially (i.e. with strength testing first, as it is the fastest and cheapest).

With a sequential approach, those formulations that passed the required


strength would be subjected to the leaching evaluation, with those passing
both (strength and leaching) submitted for permeability testing. If the tiered
approach discussed previously was used, either 5% cement with 6% slag or
4% cement with 7% slag would meet all of our criteria. In the shotgun
approach illustrated in Figure 8.4, formulas 21, 22, and 23 met all the criteria
(formula 24 failed leaching due to high leachate pH). It is noteworthy that
despite the shorter testing timescales and larger numbers examined, the
resultant formulation is not optimised for cost effectiveness.

The focused-problem approach: applies to special conditions where a


formula has been developed, used with success, but then generates failures,
usually due to difficult field conditions and/or more challenging materials.

In this case one should first try to determine what changed in the field, then
conduct a focused test to determine how the current formula can be modified
to compensate. The first step is to quickly identify the changes in materials
being treated. These could be physical changes or changes in contaminant
concentrations. Often resolution is achieved by using more binder, but if this
does not work, then a targeted treatability study can be undertaken, including
adding another reagent (that experience has shown may be beneficial). For
example, the addition of bentonite to reduce permeability, or a reagent to
reduce contaminant solubility (see Section 8.1) is required. Occasionally a
bench-scale treatability study may be conducted in the field during S/S. Figure
8.7 shows the development of a formula to solidify (not chemically stabilize)
an oily pesticide-contaminated waste-water for disposal in an on-site RCRA C
landfill.

270
Figure 8.7: A limited bench-scale field treatability study

Regardless of the approach used, once an acceptable mix design has been
formulated and verified through replicate testing, the mix design will need to
be applied at field-scale in a pilot-demonstration phase (see Section 8.4).

In preparation for a field demonstration, it is useful to develop additional data


in the laboratory to support the field efforts. For example, determining the
slump of the lab-treated material will provide insight into mixing of reagent and
contaminated material at full-scale. A very low slump e.g. <2 in (5 cm) may be
difficult to mix in the field and require additional water. However, the slump to
be achieved depends on the mixing equipment and method of disposal used
at full-scale. If a wetter mix is needed in the field, this could have a detrimental
effect on strength or permeability of cured materials. It may also be useful to
evaluate strength and permeability as a function of time up to 28 days, to
establish if shorter curing times will be possible, or if full curing is necessary to
obtain the required testing results. In some cases a cure time longer, or
shorter, than 28 days may be appropriate.

271
8.3.5 Initial sample characterisation
Once a sample is collected from the site, it must be carefully homogenised to
ensure that as aliquots (of the sample) are selected for treatment by different
candidate formulations, there is no significant difference in the untreated soil
aliquots that would affect the treatment results. A common approach is to
homogenise the bulk sample, and obtain three random sub-samples/aliquots
for analysis. If the three untreated aliquots all produce similar results for key
contaminants (totals or leaching), then the bulk sample can be considered
homogenised.

An initial (baseline) characterisation of the samples can then be performed,


including as a minimum, the physical and leaching characteristics, using the
same leaching test method to be employed during treatment. Section 8.5
provides guidance on selecting samples for treatability testing. Prior to initial
characterisation, oversized material should be removed by screening through
a 0.5 in (1.25 cm) mesh sieve, as larger particles will cause failure of small
test specimens used in treatability testing (even though oversize material of
this nature would have little, if any, effect during full-scale treatment).

8.3.6 Laboratory procedures


Generally it is difficult to homogenise samples while being collected in the
field. Field samples are collected and transported in two or three 5 gallon (19-
20 litre) shipping buckets, and at the laboratory these are combined in a
large mixer or drum and blended until they appear to be homogenised.

A rotary drum-mixer or commercial paint stirrer is sometimes used for this


purpose. However if the materials contain volatile contaminants, then to avoid
loss of these compounds a zero head-space mixer may be required.
Following homogenisation, sample characterisation is conducted (see
Section 8.3.5).

Once a bulk sample has been collected and homogenised it is ready for
bench-scale treatability testing. The reagent addition rate for each test is
specified in the treatability test plan as a percentage (weight/weight) of the
untreated soil, including its soil field moisture content.

The materials are carefully measured as shown in Figure 8.8 ensuring there is
enough mixed material to fill all the sample/coupon molds. The size/shape of
these coupon/molds is often specified by the analytical method used. Water is
also slowly added as a measured volume to the mixing bowl to facilitate the
mixing process.
272
The reagents are sometimes blended with water to form slurry prior to mixing
with the soil, if that is the approach envisioned at full-scale treatment. If the
treatment approach involves in-situ injection of a binder-slurry followed by
mixing, there will be limits set for viscosity of the slurry. Thus, the bench-scale
test should also add the reagents as slurry at the viscosity compatible with the
planned equipment to be used at full-scale.

If ex-situ mixing is planned, the reagents are added dry, but the amount of
water added is limited by the handling properties required for the treated soil.
For example, if the treated soil exiting the mixing will be transported by truck,
followed by spreading/compaction by a dozer, then the water added during
bench-scale testing must be capable of replicating the properties required for
this scenario.

Samples produced for testing should be free of bubbles/air voids, and be


allowed to cure in an undisturbed manner. In the field, treated material forms
a large moist mass that cures over days and weeks. In the laboratory, the
sample coupons are capped and placed in a humid environment at ambient
temperature, with samples removed as and when necessary for testing,
including one on a daily basis, to assess setting and initial strength
development. Figure 8.9 shows this being done with a pocket penetrometer.

Figure 8.8: Reagents are measured on a weight/weight basis to the


untreated soil

273
Figure 8.9: Pocket penetrometer testing for approximate strength

8.4 Pilot-scale tests


As described in Section 8.3, treatability testing for S/S is conducted at both
the bench- (laboratory) scale and then in the field, either prior to, or as the
initial phase of the full-scale implementation (i.e. the pilot testing). This section
describes the role of the pilot testing phase in implementing an S/S remedy.

There is no one-size-fits all approach to pilot testing as each site and waste
has its own unique characteristics and challenges. As such, the authors have
not found any single comprehensive guide to pilot phase testing for S/S.
Rather, there are a number of useful references which touch on the various
aspects of pilot testing including: Environment Agency (2004a), USEPA
(1986), USEPA (1989), Fleri and Whetstone (2006), ITRC (2011), Shi (2004),
Perara, et al. (2004), and Butler, et al. (1996).

The remainder of this section will focus on the importance of, planning for,
and implementing a field pilot test for S/S. The reader should refer to Section
8.3 for a discussion of bench-scale treatability studies for S/S, Sections 4
and 5 on S/S implementation equipment, and Section 6 on quality assurance
and quality control in S/S applications as well the sampling plan, sampling
methods, and curing of samples.

274
The properties of S/S-treated wastes may be different in the field, compared
to the laboratory, due to variation in composition of a waste, the accuracy of
the reagent dosage application, variation in site soil characteristics, and the
type of S/S mixing and placing equipment and processes.

Pilot-scale testing should be performed using the equipment proposed for full-
scale and the most successful mix design(s) determined during the bench-
scale testing program. Pilot-scale testing will demonstrate how the mix(s) will
perform at field-scale with relatively large volumes of waste, and be based on
ability to meet performance criteria, ease of implementation and cost. The
pilot-scale test provides the opportunity to:

Refine the mix design, based on field-scale observations and testing


Reduce construction risk
Optimise S/S equipment processes and operational parameters
Reduce variability in the treated product
Determine the operational parameters under which the S/S equipment
will be working
Verify the thoroughness of mixing
Verify the swell generated by S/S treatment
Evaluate variability in site and waste characteristics
Demonstrate the ability to meet performance criteria using the
selected equipment/processes on a large scale under field conditions
for comparison against the results of the bench tests.

Although pilot testing can be costly and time consuming, it can be used to
assess site safety considerations, reduce work stoppages, and increase
product consistency and process reliability. Pilot-scale tests can also be used
to train equipment operators (on the characteristics of the waste and the
solidified product), and help a contractor optimise construction and process
efficiency. An opportunity to develop or refine working practices and quality
control procedures exists potentially resulting in cost savings during full-scale
S/S, and reducing the risk of failure or the production of variable quality
treated materials.

The pilot test phase is ideally implemented as a separate project phase


between the bench-scale treatability study and full-scale implementation. This
will provide the highest level of confidence in the selected mix designs/mixing
process for field application and allow time to assess QC test results to
reduce the overall construction risk. In practice, however, it is not uncommon
for the pilot testing to be performed as an initial phase of full-scale
implementation, after the S/S contractor has been awarded the contract. The
275
decision to perform a separate pilot test or to perform the pilot as part of full-
scale implementation is generally driven by project cost, timescales, project
complexity or a combination of these factors.

The designer and owner, and in some cases the contractor (where the latter is
tasked with developing the mix design/implementation method) will need to
make this determination. In general, for sites or wastes where the potential for
variability is low and previous experience indicates, or where pilot phase
testing involves optimisation of the mix design only (i.e. the mix design has a
proven track record with the wastes), pilot-scale testing is often part of the full-
scale implementation program.

If, however, there is any doubt as to equipment suitability, waste variability,


treated product consistency, site safety concerns related to reagent-waste
reactions or S/S equipment, mix design scale-up to field conditions, cost
escalation potential, or regulator acceptance of the S/S process, then a
separate field pilot test is essential.

Another important consideration in determining how and when to implement


the pilot test is how the owner desires to allocate project risk. The more
information that is available to S/S contractors, the lower the risk of
implementation issues arising, and consequently, more certainty over
construction costs.

Additionally, the selection of performance-based contracting, design-build


delivery approaches, or design-bid-build approaches, may influence the pilot
testing approach. However, it is important to recognise that regardless of the
contracting or risk-allocation approach, the pilot test often results in some
modification of mixing procedures, reagent proportioning, and safety
protocols. This in turn influences the cost of implementation, through
optimised processing (lower cost) or through increased variability in site or
waste characteristics (higher cost).

276
8.4.1 Pilot test objectives
The objectives of performing an S/S pilot test will vary slightly from project to
project depending on the complexity of the site or wastes, the uniqueness of
the reagents used, and the requirements of working within specific regulatory
programs or regulatory jurisdictions. Typical objectives for the pilot test phase
include one or more of the following:
Enable visual observation of field-scale treatment processes
Enable visual observation of treated materials
Determine if debris, rocks, cobbles are present of sufficient size to
impede in-situ mixing or require pre-excavation or pre-processing
Evaluate treated material handling and placement characteristics and
issues
Assure that the formula(s) developed can successfully treat all
variations of soil and contaminant mixtures on the site
Demonstrate implementation of the construction quality control plan
and adequacy of the selected tests and observations to control the
treatment process
Refine estimated production rates and overall S/S project schedule
Evaluate the consistency of mix design performance between bench-
scale testing and full-scale construction techniques
Identify scale-up factors that will be useful in full-scale such as
strength-gain rate and freshly treated material (uncured) consistency
Assess the applicability and reliability of compliance criteria identified
during the bench-scale testing and establish full-scale performance
evaluation parameters and criteria
Evaluate the consistency of treated product using full scale equipment
Enable trouble-shooting and optimisation of S/S equipment, grout
plants and grout pumping, calibration, process controls, material
handling, and data collection systems
Validate that the technology can meet remedial performance objectives
under field conditions
Refine reagent selection, proportioning, or sequencing of application
Evaluation of contaminant/chemical emissions and controls
Identification of material handling issues and safety problems in
handling the waste
Estimate volume increase due to addition of reagent(s)
Develop a more accurate cost estimate
277
8.4.2 Confirming the conceptual implementation plan
As described in Section 8.3.2, a conceptual implementation plan should be
developed prior to bench-scale testing to identify how the wastes will be
treated, by what means will the wastes be mixed with reagents, and, for ex-
situ treatment, how the treated materials will be placed/disposed.

The pilot phase provides the opportunity to validate and/or modify the
conceptual implementation plan. For example, if the concept was to use in-
situ augers to mix reagents, as a grout slurry, with the waste in-place, the field
pilot may result in identifying difficulties with mixing thoroughness, penetration
depth or grout viscosity for the batch plant and grout pumps.

The pilot phase also presents the opportunity to assess different auger
diameters or configurations, vary mixing speed, penetration rate, or number of
strokes of the auger in each column, or to modify grout properties to
overcome field difficulties (see Section 5.1 on Auger Mixing). If the concept is
to perform ex-situ mixing with a pug-mill or other suitable equipment, the
material handling characteristics for hauling and placement may require
modification of the moisture content to optimise these unit operations. In
addition, one can vary and optimise the feed rate, which in turn controls the
mixing time. Excavating a large volume of waste at the site for ex-situ
treatment may also reveal the need for pre-processing such as screening or
size reduction to avoid handling issues within the mixing equipment (Section
4).

The pilot test should also identify the optimum processing rate and any dust,
emission, or safety issues. For example it may be necessary to add shrouds
to lower dust emissions.

8.4.3 Implementing the pilot testing


The pilot-scale test phase should be guided by a work plan prepared for the
site, identifying the objectives, selection of the pilot test location(s) and
treatment volume, types of equipment to be utilised and evaluated, reagent
mix(s) to be evaluated, quality control and quality assurance testing, process
parameters to be monitored, and QA/QC recordkeeping.

It is also essential to have a safety plan in place for the pilot test. As
mentioned previously, the pilot test should be performed with the same full-
scale equipment planned for the remediation. Figure 8.10 shows equipment
that might be used for an in-situ auger pilot test, and includes an 8 ft diameter
278
auger (2.5 m), and support equipment. Not shown, are the slurry batch plant,
pumps and hoses that would also be necessary. Figure 8.11 is a video
showing a 10 ft (3.0 m) in-situ auger conducting a pilot test. Note that
although it was a pilot test, all equipment and procedures are the same as
would be used for the actual remediation. Also, that a grout slurry blanket was
laid down first to reduce release of volatile emissions. Figure 8.12 is a video
showing a pilot test for in-situ bucket mixing. Here also the equipment and
procedures are the same as would be used for the actual remediation.

Figure 8.13 shows a full-scale ex-situ pug-mill system including reagent silos
and a stacker for loading treated soils into trucks for transport to the disposal
area. Pilot tests should not be conducted with scaled-down equipment or
equipment differing from that actually planned for use in the remediation.

In general, it is good practice to collect samples for testing at a higher


frequency (including more cure times) in the pilot phase than is anticipated for
the full-scale in order to further evaluate variability and to have replicates
available for potential additional testing to assess results that are significantly
different from what was expected. It is important to view the pilot phase not as
a pass/fail on S/S implementation, but rather an opportunity to further assess
site conditions and treated waste behaviour and characteristics, and to make
modifications to the mix design or application methods to address identified
scale-up issues.

Quality control test results from pilot-scale may vary from results obtained in
the bench-scale testing. The pilot-scale test is the opportunity to perform
additional tests with longer curing times than the performance criteria, to
understand if apparently failing results (if encountered) will improve with
additional curing time, or other process modifications.

The selection of the amount of waste to be treated during a pilot


demonstration and the location to treat the waste or soil are determined on a
project-specific basis. Typically, at least 500-1,000 yd3 (380-765 m3) of
material should be treated.

The volume selected should be sufficient to evaluate materials variability,


cover at least one full-day of operation, achieve the maximum depth of S/S
anticipated (for in-situ mixing), and allow for mixing of multiple
batches/cells/columns. For ex-situ mixing, the pilot test should produce a
sufficient quantity of material to evaluate preferred methods for field
placement of the freshly mixed material.
279
Figure 8.10: Eight-foot (2.5 m) in-situ auger and carrier assembly

Figure 8.11: Pilot test using a 10 ft (3 m) diameter in-situ auger

280
Figure 8.12: Pilot test for in-situ bucket mixing

Figure 8.13: Ex-situ pug-mill with support equipment

281
The pilot-scale trials also present an opportunity to acquire treatment data at
the field-scale to validate impacts due to minor alterations to the chosen mix
design. For example, if the conceptual implementation plan involves
excavator bucket-mixing in cells using a 10% binder, then treating several
cells and varying the mix over 9%, 10% and 11% will provide insight into the
performance of the S/S material, if (during full-scale implementation) deviation
from the desired 10% binder-content occurs. It may also indicate whether the
mix developed in the bench-scale testing, over or under performs at field-
scale.

Pilot-scale tests provide the opportunity to assess the adequacy of the quality
control program and to ensure that the correct field-specimen preparation
(see Figure 8.14) procedures are being followed. Field tests, including
evaluation of consistency (via the slump tests - see Figure 8.15), and grout
testing (via e.g. the mud balance test Figure 8.16) can be evaluated to ensure
the results are within the desired range. Figure 8.17 illustrates a visual field
QA/QC check to assure adequate homogenisation and mixing. The
requirement here is that no clumps of unmixed soil exist that exceed fist size
(about 4 in/10 cm). The excavator removes several buckets of material from
various locations and depths. Reagent metering and weighing systems (see
Figure 8.18) are also critical to ensure that the right dosage levels are being
applied. The reader is referred to Section 6 for a more detailed discussion of
QA/QC monitoring and data collection.

Figure 8.14: Field quality control test specimen preparation

282
The pilot test should also include post treatment evaluation of the treated
material in-place after curing which affords the opportunity to view mixing
homogeneity, overlap between treatment columns or cells, identify if
segregation of materials and reagents occurs, as well as verify that material
has cured adequately under field conditions.

Figure 8.19 illustrates a test pit excavated through solidified contaminated


soils. Test pits are also useful, in shallow groundwater conditions, to
demonstrate the absence of free water (visual) in solidified materials,
providing confirmation that groundwater flow is essentially excluded from
within the solidified soils.

Figure 5.11 (Section 5.1) illustrates exposure of solidified in-situ soil columns
to evaluate cured conditions.

Test pits also afford the opportunity to verify the strength and durability of the
treated mass through observation of relative excavation effort, and in cases
where future excavation ability is desired, to assess whether that goal has
been achieved.

Figure 8.15: Field slump test performance

283
Figure 8.16: Grout density test by mud balance

Figure 8.17: Field check to assure homogenous mixing

284
Figure 8.18: Reagent silo electronic scale and calibration weight

Figure 8.19: Trial pit in an S/S soil monolith

285
8.4.4 S/S safety considerations
The pilot-scale testing phase will enable the designer and the contractor to
assess safety considerations that may not have been obvious or observable
during bench-scale testing. As described by the USEPA (1986), safety
considerations upon scaling-up to field-scale treatment can involve fuming,
heat development, and volatilisation of organic compounds.

When vapour emissions are moderate, collection using a shroud, followed by


treatment with activated carbon for example, may be sufficient. Figure 5.2
includes a picture of an in-situ auger rig with a shroud in place.

In more severe cases, pre-treatment, low-heat reagents, or alternate means


of reagent sequencing or blending may be required, as the addition of large
volumes of reactive alkaline reagents can result in excessive or explosive
release of VOCs and the potential for excessive heat generation from
exothermic hydration reactions.

An example of this is provided in Figure 8.20, when a pilot test (conducted as


the initial phase of remediation) injected large quantities of alkaline reagents
into the sulfuric acid-laced waste oil sludge, resulting in a violent exothermic
reaction. The steam generated rose about 30 ft (10 m) into the air, and
dangerous concentrations of sulfur dioxide gas were released. The issue was
resolved by pre-treating the acid sludge with ground agricultural limestone
(calcium carbonate) to raise the pH above 4.5 before injection of the alkaline
reagents was carried out.

Figure 8.20: In-situ auger test producing steam, sulfur dioxide emissions
286
8.5 Selecting samples for treatability testing
Selecting a representative sample for treatability testing is an important part of
the development of the mix design for use in S/S. The result of the treatability
testing is the mix design (or formulation), which will be used for full-scale
treatment. Therefore, the samples used for the treatability study will directly
influence the mix design, and through it, production rate, production
sequencing, and cost of the full-scale treatment.

Following the site investigation, the concentrations of the COCs and their
variability (both horizontally and vertically) will be known and the area(s)
requiring treatment by S/S delineated. The contamination levels are also often
used to determine where to obtain samples for S/S treatability testing. The
following sections summarise the 3 most common schemes for determining
where to obtain treatability samples.

Figure 8.21 shows a bulk-sample being collected from the flights of an auger.
Great care and attention are needed to collect a representative sample at this
particular location.

8.5.1 Option 1 : maximum contamination sample


Under this option, the samples to be evaluated are collected from the area(s)
of the site where the contamination levels are considered to be at a maximum.
The objective of the sampling is to obtain a worse case sample which
represents the most difficult material to treat on-site. Mix designs that achieve
performance criteria using these samples will likely be applicable to any
material encountered that requires treatment.

287
Figure 8.21: Collecting a bulk treatability sample from auger flights

Advantages: The mix design developed during the treatability study should
be applicable to any material encountered on the site requiring treatment.
Therefore, the risk of failure (and/or retreatment) during full-scale S/S will be
minimal.

Disadvantages: Typically, higher levels of COC require greater binder usage.


A mix design developed for the maximum contaminant levels will often require
more reagents thus be more costly to apply. Since the full-scale S/S treatment
utilises stockpiles, cells or columns typically involving hundreds of cubic yards
(cubic metres) of material, the act of treatment normally imparts some
homogenisation to the material and thus, moderation of the contamination
levels. Therefore, it is unlikely that the maximum contaminant levels will be
encountered during full-scale treatment. Thus, higher reagent addition levels
than are actually needed will be applied.

8.5.1 Option 2: high contamination sample


This approach involves sample collection at the area(s) of the site with
approximately 75% of the maximum contamination levels. The objective of the
sampling is to obtain a sample representing the actual most difficult-to-treat
material that will be encountered on-site, taking into account that excavation
288
and stockpiling for ex-situ treatment or in-situ treatment of cells or columns
will somewhat homogenise the materials contained within them. Mix designs
that will meet the agreed performance criteria for this sample will likely be
applicable to all of the material encountered during the remedial operation.

Advantages: The mix design developed will likely be applicable to all of the
material encountered on-site. Therefore, the risk of failure (and/or
retreatment) during full-scale S/S will be low (typically less than 2%). The mix
design developed for these high contaminant levels will often have lower
reagent addition levels compared to Option 1, therefore the mix design will
have a lower cost to implement.

Disadvantages: The risk of failure and retreatment under this option is


typically in the range of 2% to 10%. Therefore, some treated S/S cells in the
most contaminated area(s) of the site may require retreatment. Although
retreatment of an S/S stockpile, cell, or column is only slightly more costly
than the original treatment, the additional costs involved need to be
recognised, but may not be significant.

8.5.2 Option 3: average contamination sample


Under this option, samples are collected from the area(s) of the site with
average contaminant levels. The objective of the sampling is to obtain a
sample representing the typical material that will be encountered on-site,
taking into account that some homogenisation will occur from stockpiling or
the treatment process. Mix designs that will meet the design performance
criteria using this sample will most likely be applicable to the bulk of the
material encountered on-site.

Advantages: The mix design developed will most likely be applicable to much
of the material on-site. The risk of failure (and/or retreatment) during full-scale
S/S will be low to moderate. This approach will generally use the least
binder/reagent of the three approaches, hence the lowest reagent cost to
implement.

Disadvantages: In practice, the risk of failure and retreatment under this


option is typically within the range of 10% to 25%, and in practice the most
contaminated area(s) of the site will require retreatment by S/S and the
additional costs may be significant.

289
8.5.3 Rationale for using the high contamination sample
The first option, Option 1 for the selection of treatability samples will produce
a very robust mix design, which can be implemented during full-scale S/S with
high confidence. However this mix design will likely be costly. Option 3
represents the other end of the spectrum, and although the mix design is
less robust (and less costly), there could be up to 25% of the site requiring
retreatment.

The second option (highly contaminated sample) typically provides the most
economical mix design to implement by reducing reagent addition levels while
reducing the retreatment rate to acceptable levels. Typically, the savings in
reagent usage for Option 2 more than off-set retreatment cost providing the
most economical choice for the selection of samples for treatability testing.

8.5.4 Sites with multiple materials requiring treatment


Many sites have more than one type of material to be treated (e.g. soil,
sludge, paste, etc.), and within these there are many possible permutations
(e.g. soil and DNAPL-saturated soil and sludge materials from different
processes that were co-disposed). Typically each material type should be
sampled for the treatability study. As discussed in Section 8.3 (bench-scale
treatability tests), good practice involves the collection of multiple (two or
three) 5-gallon buckets of each material type to assure an adequate quantity
for all tests that may be desired. All the materials requiring remediation are
included.

Conducting treatability testing on all the material types encountered at a site


provides information on whether all material types can be treated with one
common mix design or that separate mix designs will be required. The
samples obtained for each material type should have high (approximately
75% of maximum) contaminant levels (Option 2 above).

290
9.0 Post-treatment capping and monitoring

The general strategy for post treatment management of S/S material should
be considered early in the design process and include collaboration with
critical stakeholders and regulatory agencies.

The long-term performance and management objectives should be


established with respect to future land use, post remediation operation and
maintenance, required engineering and institutional controls, and the need for
long term monitoring etc. Accordingly, an appropriate post-remediation
management strategy requires an integrated approach where the long-term
performance and management objectives are evaluated at various stages of
the design process and during full-scale construction. These stages include:

Pre-design data collection


Assessment of reagent mix designs during bench-scale testing
Verification of full-scale performance objectives during pilot testing
Performance verification during construction
Final as built configuration, and
The requirements for post construction monitoring

The following discuss the factors most important for the selection and design
of a cap for S/S treated material, and for post construction monitoring. The
considerations for selecting a cap design and for post construction monitoring
are the same for either in-situ or ex-situ S/S treated sites.

9.1 Overview of post treatment management of S/S material


A major design objective is to develop a capping layer that will divert surface
water away from the S/S monolith to prevent ponding. Key issues to be
addressed during the design process include:

The reconciliation of the estimated quantity of S/S material following


completion with actual quantities arising during treatment
The modifications necessary to finalise geometry of the cap arising
from the field conditions experienced
Ensuring the cap meets future land use objectives established during
the design phase

291
During the pre-design stage, careful consideration must be given to variables
that will impact on the final site conditions and long-term operation and
maintenance requirements. For an S/S project, this can be difficult, as it may
require interpolation of the final site conditions in the absence of key data.
Nonetheless, by carefully assessing potential factors of importance a
management strategy for post-remediation conditions is possible and will help
define data requirements from bench- and pilot-scale testing.

The following factors require consideration, and are discussed below:

Climatologic and geographic considerations


Site geometry
Integration with existing site development
Future site use
Affected property owners
Regulatory requirements

9.1.1 Climatologic and geographic considerations


Key design parameters can be obtained by considering surface water
sources, and whether the site is located in a cold weather region, where
freeze/thaw effects might impact on the long-term stability of the S/S monolith.

Surface water management requirements involve an assessment of a number


of parameters such as the incidence and severity of rainfall, seasonal
fluctuations in precipitation and the likelihood of wetting and drying cycles
impacting on the monolith surface. The impact of cold weather requires an
evaluation of the thickness of the cap, the final geometry and elevation of the
S/S monolith relative to final site grading, and how any excess S/S material
(swell from in-situ treatment) is managed and placed.

9.1.2 Site geometry


The geometry of the site will influence where and how S/S-swell material is
placed, and how placement is sequenced within the full-scale S/S operations.
Ultimately, the final S/S monolith elevations will need to take into
consideration the pertinent design parameters such as surface water drainage
and management that will be directly influenced by the site logistics.

292
9.1.3 Integration with existing site development
The post remedial plan needs to consider all existing structures, the site
services including underground and above ground utilities and roadways, as
these components are integral to the remediated site and influence the final
cap configuration.

Site design plans will need to include specific requirements for completing the
S/S monolith near existing structures/foundations and the maintenance of
adequate cover for landscaping and surface water management. Clean utility
corridors may need to be established across the S/S monolith to
accommodate existing and/or new underground utilities, but do pose
engineering challenges (for their integration into the final cap configuration
and management of any surface waters) which must not be underestimated. If
not properly designed, they may serve as containment vessels for standing
water that could adversely impact on the long-term performance of the S/S
monolith. For example, even if the monolith performs as designed with
respect to leachability of target constituents, the presence of reagents such as
Portland cement could lead to elevated pH levels in ponded contact water. In
turn, this could cause the leaching of key contaminants, leading to surface
water discharge standards not being met.

9.1.4 Future site use


Under some circumstances, the long-term development objectives for a site
remediated by S/S may not be known at the time of treatment. However,
where they are known early design consideration for the final cap
configuration may be critical to achieving final post S/S management
objectives. In addition to managing the amount of S/S material to be treated, a
number of engineering parameters may require evaluation during bench-scale
testing to meet the development objectives, including:

Mix designs enabling lower UCSs for subsequent excavation of S/S


material for foundations, infrastructure and surface re-grading
Mix designs with higher UCSs for enhanced bearing capacity and
shear strength
Mix designs specific to soil types for management of bearing capacity
and potential settlement induced by future load bearing structures

293
A number of future use requirements may also need to be considered that
could range from mildly intrusive to major modifications to the cap and the S/S
monolith, including:

Landscaping, including the planting of trees and/or shrubbery


Re-grading/modification of cover thickness and/or removal or
repositioning of portions of the monolith (to accommodate new grading
and drainage)
Removal of monolithic material to accommodate new foundations
and/or infrastructure

Flexibility in the design of the cap needs to be considered to allow the


property owner(s) to accommodate future changes including:

Reducing the elevation of the monolith to enable shallow building


foundations
Improved conditions for landscape plantings (facilitated by earthen
materials in the cap)
Shallow gradients to permit future access and usability

Additional assessments will also be required to establish agreed engineering


and institutional controls with which the property owner must comply. These
are normally incorporated as restrictive covenants and may include the
following:

Restrictions on modifications of the cover system and/or the monolith


Planning and notification requirements for proposed significant
modifications/new development
Minimum technical specifications and performance specification
compliance upon cover system reconstruction
Management of regulated materials, such as disturbed S/S material
and treatment requirements for excavation dewatering (if applicable)

9.1.5 Regulatory requirements


Requirements for the design, construction and maintenance of post
remediation cap systems for S/S monoliths are regulated by local, regional
(State) and national (Federal) regulatory agencies responsible for approving
remedial design plans and specifications under various clean-up programs.

For example, projects under the USEPAs Superfund program require


submission of a cap design for approval by the USEPA Project Manager that
294
would comply with the established remedial performance objectives. The
assessment of applicable design requirements could include comparison with
current Subtitle D (municipal solid waste) and C (hazardous waste)
specifications. Performance objectives meeting Subtitle D specifications
include the following:

Use of barrier materials that will have a permeability of no greater than


1 x 10-5 cm/sec (1 x 10-7 m/sec)
Incorporate a barrier-layer of earthen materials of >12 in (>30 cm) in
thickness
Have an erosion control/topsoil layer constructed of earthen materials
with a thickness of >6 in (15 cm), capable of supporting native
vegetative growth and for frost protection
Use of a geo-membrane (if the liner construction includes a geo-
membrane)

However, under more rigorous Subtitle C specifications, requirements may


include:

A cover-soil layer in addition to an erosion control/topsoil layer


A drainage layer e.g. a free draining sand and/or gravel material, or a
geo-composite consisting of geo-net-bonded with a geotextile (on one
or both sides)
A barrier layer of >24 in (90 cm) with a maximum hydraulic conductivity
of 1 x 10-7 cm/sec (1 x 10-9 m/sec)

Typical example profiles for Subtitle D and C landfill cover are illustrated
below:

Figure 9.1: Typical profile for a Subtitle D cap

295
Figure 9.2: A typical profile for a Subtitle C cap

The primary objectives of capping an S/S monolith are to prevent surface


water ponding and maintain a direct contact barrier. Accordingly, there are a
number of different types of caps that have been approved by regulatory
agencies. These are dependent upon site-specific conditions, site location
and the agreed site-specific requirements of the remedial action. Approved
caps range from a simple single layer of soil to more complex multiple layer
systems, combining earthen and geo-synthetic materials.

9.2 Post-treatment management of S/S


9.2.1 Key engineering parameters S/S post-treatment management
A major consideration for site management after S/S is the assessment of the
total amount of material that will be generated during the treatment. This is
influenced by the volume increase from addition of the reagent and water, and
in the case of in-situ S/S treatment, the volume of swell that will be produced.

The final volumes of S/S material will directly affect the site grading and
monolith surface configuration, and future surface water management
requirements. An assessment of the anticipated volumetric changes should
be estimated early in the design process (i.e. during bench-scale testing),
from the total absolute volume(s) of the materials involved in the treatment
(soil + reagent + water). This figure can then be compared to the volume
experienced during treatment.

296
The volume increase due to ex-situ mixing is most affected by:

In-place vs excavated volumes, where expansion occurs during


excavation as voids are introduced thus reducing the density of the
material being treated. (This volume expansion is further impacted by
the addition of reagent and mix water, and will need to be reconciled
with the space available for replacement.)
The in-situ (or field) moisture content, which can be used to partly
make-up mix water needed for the reagent mixing/activation needs
The required reagent quantities that must be added

The amount of swell that will be generated during in-situ S/S is most affected
by a several key parameters that include the following:

A variation in the moisture content of soils to be treated will directly


impact on the full-scale water to reagent ratios applied, and the mixing
performance experienced upon reagent injection. Thus the amount of
native water present along with water injected with the reagent slurry
will affect the amount of swell that will be generated
The percentage of fine drained soil (clay and/or silt) will directly impact
on swell, with higher percentages of fines resulting in a higher swell.
The finer grained materials often require higher reagent and water
additions to achieve uniform mixing. In addition, highly heterogeneous
soil conditions will make the estimation of swell volumes difficult, and
swell volumes exceeding 30 % have been recorded at a number of S/S
sites
The reagent density is an important parameter, which relates directly to
the quantity of reagent needed for treatment. This is based on the
density of the soil to be treated and the estimated volumes and the
specified mix design. As different reagent formulations/densities may
be required for different soil types, the characterisation of sub-surface
conditions is extremely important
The minimum water to reagent ratio is typically in the range of 1:1 and
is directly related to reagent density and viscosity. Maximum viscosity
limits are needed to prevent problems, such as the clogging of reagent
injectors during mixing. Backhoe-mixing for example can tolerate more
viscous slurries, and thus produce lower swell volumes upon treatment

9.2.2 Engineering approach for preparing S/S material prior to capping


Preparation of the S/S monolith prior to capping will require interim and post-
construction strategies for managing the S/S treated material, including any
swell that has arisen from treatment, to achieve the final grading requirements

297
for surface water management and to conform with site-specific logistical and
structural constraints.

Typically, during ex-situ treatment, the treated material is placed and


compacted in lifts, with the final shaping (of the monolith) accomplished
concurrently with placement. Dozers, and sometimes rollers, as shown in
Figure 9.3, are used for this purpose.

Figure 9.3: A smooth roller being used to compact ex-situ treated


material

During typical in-situ S/S utilising large-diameter augers, excess material from
binder injection is managed as each column is formed, using a backhoe or
excavator. Swell material is graded or removed prior to starting the next S/S
column.

As discussed previously, if this excess material needs to be moved and/or


stockpiled for final grading, additional reagent may be required to recondition
the disturbed, partly set swell in order to meet performance targets. This is a
key consideration as the volume of swell material may be large, if fine grained
soils are involved.

298
Two basic approaches for managing swell material for final placement/grading
prior to capping are as follows:

Direct wet placement and grading is suitable for sites that are not
constrained by structural or topographical features, such as buildings or
waterways. This form of management is preferred by contractors, as the swell
material does not require multiple handling, enabling the S/S operation to be
completed as one continuous process. Although less flexible for modifying the
contours of the cover system in response to S/S treated volume changes, a
key advantage of this method is that the material is placed under the same
conditions and within the timeframe of the parent S/S material.

Staging and Reconditioning of swell material involves sequential handling


by stockpiling and reallocation to other areas of the site. Disturbance of the
S/S material results in a loss of treatment integrity and produces properties
that deviate from those of the parent material. Consequently, the swell
material will need to be re-conditioned (with additional reagent) to meet
performance characteristics consistent with the undisturbed treated material.
The multiple handling of material further increases the overall volume of S/S
material that will need to be managed, but does provide greater flexibility for
modifying final cap geometry.

The grading of the S/S monolith can be completed using excavators or dozers
with no forced compaction requirement. Figures 9.4 and 9.5 show examples
of final grading/contouring applying a direct placement approach using an
excavator and a dozer, respectively.

299
Figure 9.4: Direct placement of S/S swell by an excavator for final
grading

It is usually desired that the characteristics of the swell material are consistent
with the undisturbed treated material, and meet the same performance
requirements. A smooth homogeneous appearance of the swell material is to
be expected following mixing, providing a high degree of confidence that the
final surface of the monolith will perform similarly to the underlying
undisturbed treated material.

Figure 9.6 is an example of placement and final grading of swell material by


reconditioning using grout slurry with a low percentage of binder/reagent in
lifts. A dozer is used to construct an embankment and access ramp to an
industrial area for use by semi-tractor trailers.

300
Figure 9.5: Final shaping/contouring of in-situ S/S using a dozer

Reconditioning involves re-mixing stockpiled swell material in a shallow pit


with the reagent grout. The dozer then pushes bucket loads of the swell
material through the reagent grout (located in the pit behind the dozer) and
immediately places this freshly reconditioned material onto the embankment.

301
Figure 9.6: Reconditioning and placement of in-situ S/S swell

Assessment of the volume of S/S material involves the quantity of additional


reagent needed to return the dried and disturbed swell to a consistency and
appearance similar to the freshly treated material from the full-scale S/S
operation. This approach can be evaluated at bench-scale level by drying and
then reconditioning S/S material with different reagent additions, followed by
performance testing for properties such as UCS and permeability and then
comparing with long term performance objectives.

302
9.3 Capping technologies

9.3.1 Overview of design considerations and general types of caps


applicable for S/S monoliths
As mentioned, two of the primary objectives of a capping strategy are to
prevent surface waters ponding on top of the monolith leading to degradation
and/or leaching, and to provide a long-term direct contact barrier.

As also previously discussed, the S/S treated material will be subject to


regulatory approvals, depending on the nature of the remedial program
involved. Accordingly, the cap design will be site-specific and dependent on
the intended function of the site (especially site-specific performance goals).
Site-specific design considerations for capping the monolith include:

Compatibility with required grading and drainage contours/final


monolith geometry
The required thickness of the cap ensuring protection of the monolith
from environmental loads e.g. due to climate, such as freeze/thaw or
wet/dry exposure, and to meet the regulatory requirements from a
direct contact barrier
The availability of acceptable earthen materials for constructing the cap
(e.g. a low permeability clay)
Erosion control requirements for the management of surface water
runoff
Support for native vegetative growth and landscaping restoration

There are several types of caps/capping systems that can be used on an S/S
monolith, including:

Earthen caps comprised of clay and/or other low permeability geo-


materials
Geomembrane systems involving e.g. high (or low) density
polyethylene (HDPE, LDPE) or polyvinyl chloride (PVC) with fabrication
options such as smooth or textured surfaces
Geo-synthetic clay liner (GCL) systems that involve a bentonite clay
layer bonded by geotextile fabric or geo-membrane
Evapotranspiration (ET) covers using fine- and coarse-grained layers
of earthen materials with different grain sizes to create a capillary break
to prevent the build-up of saturated conditions over the monolith
Asphalt and concrete pavements that can be used as the primary cap
or in combination with other capping materials/technologies
303
These capping technologies may be used alone or in combined (composite)
approaches at a specific site. They may, but do not always, include drainage
and cover, and surface or topsoil-layers above the barrier layer. A general
description for each of these specific layers follows.

The drainage layer is located directly above the barrier layer and may consist
of granular free-draining sand and/or gravel material, or a geo-composite
consisting of a geo-net bonded with a geotextile on one or both sides.
Construction using a free-draining soil may include the placement of a non-
woven geotextile filter fabric between the barrier layer and drainage layer and
between the drainage layer and upper materials, to prevent potential piping
and/or migration of fines that could lead to clogging of the drainage layer. The
typical recommended minimum design thickness for the drainage layer is 12
in (30 cm), as thinner layers are more difficult to construct uniformly and
quality assure, and may facilitate damage to the barrier layer during
placement.

The cover layer is located directly over the drainage layer, and is designed to
meet site-specific conditions and the location of the site. Considerations may
include protection from vegetative root penetration and/or animal intrusion into
either the drainage or barrier layers, and protection from freeze/thaw and/or
desiccation and erosion. In some applications, the cover layer may be
designated as an erosion layer, with a design thickness highly dependent on
site-specific requirements. Different soil-type materials may be used,
depending on site location and availability. Under some circumstances, the
cover layer may also be a vegetative layer, if it is integrated with the surface
layer and/or contains sufficient organics to support vegetative growth.

The surface layer or topsoil sustains adequate vegetative growth to


minimise erosion, and protects the cover and provides an enhanced overall
visual appearance. Topsoil is the most commonly used material but in some
locations where it may be difficult to support growth (such as arid/desert-like
climatic regions) other materials such as sand, gravel or cobbles may be more
suitably used. Different types of geo-synthetic applications are also available
which include the use of synthetic turf layers that eliminate the need for
maintaining vegetation, reduce long-term concerns for cap stability and
provide the appearance of a manicured landscape.

Evapotranspiration (ET) caps present an alternate approach to single and


composite cap systems and differ substantially in the engineering
mechanisms for managing surface water infiltration and precipitation. In
contrast to relying on barrier and drainage layers, which direct surface water
into a drainage layer to promote dissipation of infiltration off the cap, the ET
relies primarily on storage within a soil layer. Removal of the stored
304
infiltration would rely primarily on evapotranspiration and transpiration
mechanisms. Hence, these types of caps are more suited to arid regions,
although some have been installed at sites in Florida, Maryland,
Pennsylvania, Georgia, Illinois, Michigan and Wisconsin (USEPA, 2003).

A further examination of each of these technologies and their applicability to


S/S post-construction conditions follows.

9.3.2 Clay and/or other low permeability earthen materials


The use of clay and/or other low-permeability earthen materials as cover over
S/S-treated material has met regulatory approval at a number of remediated
sites. This approach offers flexibility for specifying locally available material at
lower cost. Generally two earthen materials are used in two layers:

An initial low permeability layer of a thickness depending on site-


specific requirements, but typically 1-2 ft (30-60 cm) to meet the direct
contact pathway (promotes surface water runoff and negates standing
or pond water)
An upper vegetative layer consisting of 6-12 in (15-30 cm) of
topsoil/suitable material to support vegetative growth and aid
landscaping

Depending on site characteristics, the cap may not include a drainage layer. A
typical cap comprising a single-component clay, or low permeability earthen
materials is shown in Figure 9.7.

A variety of different materials may be used to construct these types of caps.

Acceptable materials can include those classified as: SW, SP, SM, SC, CL,
CH or a combination of these group symbols specified under the Unified Soil
Classification System (USCS).

Unspecified soil types are often referred to as unclassified fill, and the
recommended minimum thicknesses should be 2 ft (60 cm) to provide
sufficient thickness for a rooting zone. Such caps are generally placed in lifts
and compacted.

305
Figure 9.7: Example single component cap with a compacted clay layer

Construction may or may not include a drainage layer and/or placement of a


cover soil depending on the nature of the cap. An example of a successful
capping layer comprising 2 ft of soil material is the Schuylkill Metals Site in
Florida (Figure 9.8), where the S/S monolith and cap are located in the
foreground. The lake in the background is not on the monolith.

Performance of an earthen cap will depend highly on the types of material and
specified methods for construction. Low hydraulic conductivities may be
possible with compaction, near optimum moisture contents and maximum dry
densities based on standard or modified proctor test results. Some general
design considerations include:

306
Figure 9.8: Cover layer comprising 2 ft of compacted soil, Schuylkill
Metals, FL

Compaction should be evaluated against additional performance


criteria, such as infiltration and suitability for long-term sustainable
vegetative growth
The capacity of the cap to effectively store surface water infiltration
whilst promoting surface water drainage is a consideration (as in the
design for ET caps) but is dependent on locally available materials/site
location. In this respect, a lower level of compaction may provide a
superior performance
The cap thickness should be designed with storage capacity, long-term
stability and potential for erosion as key design criteria
Minimum index properties for the use of clay in low permeability caps
include those indicated in Table 9.1 (Benson et al. (1994)

Table 9.1: Minimum index properties for the use of clay

Property Minimum Index


Liquid limit 20
Plasticity index 7
Percentage fines (i.e. particles passing the No. 200 sieve) 30
Percentage of clay 15
Activity 0.3

307
However, there may be considerable variation in these properties depending
on the availability of local materials, site-specific design requirements and
flexibility in regulatory acceptance. Other considerations for selection of an
appropriate material include the following:

Greater cap thicknesses may be required where potential for


freeze/thaw conditions exist, such as in the northern United States
where frost can penetrate 6 ft below the ground surface, even if bench-
scale studies indicate the monolith exhibits freeze/thaw resistance
Lower permeabilities may be required where repeated wet/dry cycling
are a concern, and/or site-specific conditions will not permit sufficient
surface grades to promote effective drainage off the monolith
Use of alternative materials with higher percentages of sand and/or
gravel to provide subgrade conditions for potential future development,
as plastic soils may not have the bearing strength for
foundations/roadways

The engineering of soils to achieve low hydraulic conductivities may involve


compaction on the wet side of the optimum moisture content. Conventional
specifications may include a minimum dry unit weight and a range of water
contents >90% of the maximum dry unit weight, and within 0-4% (wet of
optimum, modified Proctor), respectively. Figure 9.9 shows the construction of
a 12 in (30 cm) thick, low permeability cap using locally available clay.

Figure 9.10 shows the final grading operations to promote positive drainage
towards a waterway.

9.3.3 Geo-membranes
Geo-membranes or flexible membrane liners (FMLs) provide a relatively
impermeable layer with readily available materials.

Geo-membranes are delivered in rolls that can be easily joined/seamed


together, and include high density polyethylene (HDPE), low density
polyethylene (LDPE), and poly vinyl chloride (PVC) although there are many
other options.

308
Figure 9.9: Construction of a low permeability clay cap

Figure 9.10: Final grading to direct surface drainage


309
These membranes are more flexible than compacted clay and can
accommodate irregularities in an S/S monolith surface better than a clay cap.
Water infiltration through a geo-membrane will occur through punctures, so
care must be taken during the various stages of installation. Figure 9.11
illustrates the installation of a typical geo-membrane.

A consideration for using geo-membranes is the stability of the final cover


slopes, as the smooth membrane surfaces have a low interface friction that
can cause overlying soil layers to slide. Thus, textured membranes can be
used on higher slope angles to reduce the possibility of sliding. A typical
single component geo-membrane cap is shown in Figure 9.12.

The preparation of the subgrade is an important consideration during


installation. Water infiltration through a punctured geo-membrane typically
occurs before upper capping layers are in place. If the surface of the S/S
monolith is used as the subgrade, it is recommended that particle sizes no
larger than 3/8 in (1 cm) are allowed at the monolith surface to prevent
puncturing of the liner. If subgrade conditions exceed this then an engineered
fill of <3/8 in (1 cm) can be placed over the monolith. Alternatively, a thick (e.g.
12 ounce/380 g) non-woven geotextile could be used below the membrane.

9.3.4 Geo-synthetic clay liner (GCLs)


GCLs are readily available from manufacturers and can be easily installed.
Performance of GCLs compares favourably to the applications using
compacted clay with the advantage that it requires considerable less space
for installation. A typical configuration for a single component cap using a
GCL is illustrated in Figure 9.13.

An example of a GCL and soil cover over the S/S monolith is illustrated in
Figure 9.14. This Figure also shows the use of a concrete flume to direct
surface runoff across the cap.

A key design consideration is the very low internal shear strength of bentonite
leading to instability on capped slopes. This can be compensated using
fabrics, such as needle-punched geotextiles, which typically yield the highest
shear strength followed by stitch bonded. However, the use of needle-
punched backing will require a non-woven geotextile on one side.

310
Thus, the use of GCLs needs careful design consideration involving cap
geometry, anticipated slope(s), interface friction angles and available cap
materials used for layers above the GCL.

Cap design should include an acceptable safety factor to prevent sliding of the
cover material over the liner system. GCLs using either needle-punched or
stitch bonding can be used in this respect, as unreinforced GCLs are not
recommended for slopes >10:1 (horizontal to vertical). In contrast, reinforced
GCLs (needle-punched) have been successfully installed on slopes >3:1. A
typical acceptable factor of safety for veneer stability is 1.5.

Figure 9.11: Installation of a geo-membrane showing the welded seams


between sheets

311
Figure 9.12: A single component cap with a geo-membrane

An example of a cover using a GCL and 24 in (60 cm) of soil, incorporating an


added gravel surface for wear-resistance is illustrated in Figure 9.15, the
American Creosote Site, Jackson, Tennessee. The gravel surface was added
to facilitate site re-use as an equipment storage yard.

Figure 9.13: A typical single component cap with a geo-synthetic clay


layer

312
Figure 9.14: GCL and soil cover with drainage channel, Peak Oil Site

9.3.5 Clay with geo-membrane


When a geo-membrane is used over compacted clay, the clay provides a
smooth base that minimises the potential for punctures during installation.
Additionally, the geo-membrane can be quickly installed over the clay,
protecting the clay from desiccation and cracking. The interface between the
smooth clay and the geo-membrane can have low friction and may be
susceptible to sliding. A textured membrane can be used to increase friction
between the two layers. A typical composite cap using clay with a geo-
membrane is shown in Figure 9.16.

313
Figure 9.15: Cover with a GCL, 2 ft of soil, and a gravel surface

Figure 9.16: Profile of a composite cap with a geo-membrane over a


compacted clay layer

9.3.6 Geo-composite clay liner with geo-membrane


Geo-membranes and GCLs can also be used together to provide a hydraulic
barrier that is more flexible and resistant to differential settlement than

314
compacted clay. A typical configuration for a composite cap using a GCL with
a geo-membrane is illustrated in Figure 9.17.

Figure 9.17: A composite cap with a geo-membrane over a GCL

9.3.7 Evapotranspiration (ET) caps


ET caps rely on a soils water storage ability instead of low permeability layers
to prevent the infiltration of water and ponding onto the underlying S/S mass.

There is an increasing use of these alternative covers but currently limited


performance data to support their design guidance (USEPA, 2003). ET caps
use soil layers to store water until the water is removed via
evapotranspiration. Functionality is dependent upon the balance between
surface runoff, infiltration, soil storage, and evapotranspiration. ET caps are
separated into two categories: monolithic, constructed in a single soil layer,
and capillary break caps, constructed with two soil layers of differing grain
size.

ET caps are generally only suitable for use in arid and semi-arid regions (e.g.
the western United States) but have been occasionally used in areas with
more humid conditions. ET caps may be lower cost than conventional caps if
local soils are available to minimise transportation costs.

Monolithic ET caps rely on the water storage properties of a single soil layer,
constructed to a thickness to promote storage at the peak rainfall times of the
year. The soil layer used needs to have a storage capacity greater than the
peak infiltration volume, to prevent infiltration into the underlying S/S material.

315
A typical configuration for a monolithic ET cap is illustrated below in Figure
9.18.

Capillary break ET caps rely on the unsaturated hydraulic properties of soil to


create a capillary break. This is facilitated by placing a fine-grained soil layer
directly over a coarse-grained soil layer, with the former serving as a
monolithic barrier for storing water, while the coarse layer acts as the capillary
break. Essentially the water prefers to remain in the small pore spaces of the
fine-grained material.

Figure 9.18: A monolithic evapotranspiration cap

This system will prevent percolation as long as the coarse layer remains
unsaturated, and therefore it is important to know the water balance of the
area to ensure sufficient thickness of the fine grained storage layer. For
capillary break ET caps, the fine-grained layer can range from 1.5-5 ft (45-
150 cm) and the coarse-grained layer can range from 0.5- 2 ft (15-60 cm)
(USEPA 2003). A typical configuration for a capillary break ET cap is given in
Figure 9.19.

An example of the placement of the coarse grained layer for an ET cap


constructed over an S/S monolith at a project site in the south eastern region
of the United States is illustrated in Figure 9.20.

9.3.8 Asphalt and concrete pavement applications


The integration of an asphalt and/or concrete pavement into a capping
strategy can be important for S/S projects located in urban areas.
316
Urban applications include parking areas and/or streets that may need to be
restored following completion of remedial construction. Design parameters
that may need to be considered include:

Final grades/slopes and elevations for the S/S monolith to meet


required pavement structure subgrade and final grade requirements
Utility corridors to accommodate surface water drainage for
reconstructed streets or parking areas
Integration of pavement structures with other components of the cap
Transition of new pavement structures with existing structures at the
limits of the S/S monolith

Figure 9.21 is an example of an asphalt roadway that was reconstructed over


an S/S monolith. The landscaped areas on either side of the roadway consist
of a capillary break ET cap. Prior to construction, detailed communications
with the municipality were necessary to fully identify the reconstruction and
restoration requirements, and detailed S/S designs were prepared for street
alignment, profiles, grades, and the location of storm water collection points
and conveyance. The final slope grading and construction requirements were
also evaluated for the ET cap to meet final grades for the roadways.

Figure 9.19: A capillary break evapotranspiration cap

317
Figure 9.20: A coarse-grained soil-based ET cap over an S/S monolith

9.3.9 Comparison of capping technologies


Table 9.2 provides a general comparison of capping technologies (excluding
asphalt and concrete pavement) with respect to key criteria, consisting of the
following:

Constructability: the ease or difficulty of construction which will be directly


influenced by the types of materials selected and type of cap (i.e. single vs
component or ET cap)

Permeability: of different capping scenarios is dependent upon site-specific


drainage and local materials properties and availability

Freeze/Thaw Durability: is the susceptibility of capping materials to


degradation on exposure to repeated cycles of freezing and thawing, and is
highly influenced by geographic location, the type of materials used and final
thickness of the cap.

318
Figure 9.21: Reconstruction of an asphalt roadway over an S/S monolith
in Sanford, FL

Desiccation: is the susceptibility of the capping materials to the loss of


moisture necessary to maintain the desired low hydraulic conductivity.
Repeated cycles of wetting and drying can alter the properties of clay or
GCLs. Design considerations include the available capping material, the sites
geographic location and the engineering controls necessary to prevent
moisture loss.

9.4 Capping strategies based on slope, drainage, and climate

9.4.1 Slope and drainage considerations


Selection of the most appropriate capping strategy to meet anticipated slope
and drainage conditions will be highly dependent on a number of site-specific
factors and the intended function for the cap that include the following:

Design limits for surface water infiltration


Availability and type of local materials for cap construction
Limitations for allowable grades to promote effective drainage
319
Logistical constraints based on surrounding topography and existing
land use

Potentially applicable capping strategies based on site-specific logistical and


surface water drainage conditions are summarised in the Table 9.3.

320
Table 9.2: A comparison of capping technologies

Type of Cap Constructability Permeability Freeze/Thaw Durability Desiccation

Single Component Caps


Placement and compaction in lifts to
Silty or clayey soils (SM, SC). Varies but may achieve less than
meet specified compaction Moderate durability Moderate resistance
Unclassified fills 1x10-5 cm/sec (1x10-7 m/sec)
requirements
Higher level of placement and Varies typically between 1x10-5 Low durability if not Low resistance if not
Compacted clay (CL, CH) compaction QA/QC required than for (1x10-7 m/sec) to less than 1x10-7 provided sufficient cover provided sufficient cover
-9
earthen caps cm/sec (1x10 m/sec) for frost penetration for moisture loss
HDPE or PVC liners may require -7 -9
Less than 1x10 cm/sec (1x10
Geomembrane placement of appropriate fill to prepare High durability High resistance
m/sec)
subgrade prior to placement
Low resistance if not
GCLs could be placed directly over Less than 1x10-7 cm/sec (1x10-9
Geosynthetic clay liner (GCL) High durability provided sufficient cover
prepared ISS monolith surface m/sec)
for moisture loss
Composite Caps
Clay layer would be placed directly
Less than 1x10-7 cm/sec (1x10-9
Clay with Geomembrane over ISS subgrade followed by High durability High resistance
m/sec)
geomembrane
Geosynthetic clay liner (GCL) Use GCL bonded to geomembrane Less than 1x10-7 cm/sec (1x10-9
High durability High resistance
with Geomembrane and place directly over ISS subgrade m/sec)
Evapotranspiration (ET) Caps
Place fine grain layer in lifts with Not applicable because performance Low durability generally Low to moderate
Monolithic limited compaction (e.g. rubber tyred relies on storage of surface water not applicable for resistance depending on
or tracked equipment) infiltration colder/humid climates soil type
Not applicable because performance
Place fine and coarse grained layers Low durability generally Low to moderate
relies on infiltration and storage of
Capillary Break in lifts with limited compaction (e.g. not applicable for resistance depending on
surface water with capillary break for
rubber tyred or tracked equipment) colder/humid climates soil type
drainage of excess water

321
Gently graded slope conditions will require significant reliance on barrier or ET
layers to prevent surface water infiltration through the cap, and pose the most
significant concern for ponding or standing water over the monolith.

Under most barrier design applications, it is not recommended to have cap


design geometry with less than 3% slope. In addition, dependent on
geographic location, inadequate drainage may lead to extended saturated
conditions leading to long-term operation and maintenance (O&M) challenges,
for maintaining acceptable cap vegetation and landscaping.

As indicated in Table 9.3 moderately graded slopes may provide optimal


conditions for a range of capping technologies to meet performance
objectives. In contrast to gently graded conditions, the presence of steeply
graded slopes may impact on soil veneer stability and/or erosion of cover
materials, precluding the use of materials such as GCLs with low interface
friction angles.

Construction using earthen materials may be applicable for steeply graded


slopes if the material can be adequately compacted to minimise erosion.
Erosion can be a concern for ET covers on steeper slopes due to the low
compaction used to enhance water storage capacity.

322
Table 9.3: Capping technologies vs slope and drainage

`Applicability of Capping Technologies Based on Slope and Drainage Conditions


Type of Cap Condition 1 Gently Graded Condition 2 Moderately
Condition 3 Steeply Graded
Slopes (less than 5%) with Graded Slopes (5 to 10%)
Slopes with rapid drainage
poor or restricted drainage positive drainage
Single Component Caps
Silty or clayey soils (SM, SC,
ML, CL-ML). Unclassified fill Less Applicable1 Applicable2 Applicable
materials
Compacted clay (CL, CH) Applicable Applicable Less Applicable
Geomembrane Applicable Applicable Less Applicable
Geosynthetic clay liner (GCL) Applicable Applicable Potentially Applicable3
Composite Caps
Clay with Geomembrane Applicable Applicable Less Applicable
Geosynthetic clay liner (GCL)
Applicable Applicable Potentially Applicable
with Geomembrane
Evapotranspiration (ET) Caps
Monolithic Potentially Applicable Applicable Applicable
Capillary Break Potentially Applicable Applicable Applicable
[1] Less Applicable = significant engineering, construction, or maintenance issues that must be resolved for effective long-term
implementation of the technology. [2] Applicable = can be applied effectively using standard engineering and construction
practices. [3] Potentially Applicable = advanced engineering or construction practices may be needed to effectively implement the
technology.

323
9.4.2 Capping strategies based on climatologic conditions
The prevailing climatic conditions at a site are a key factor in determining the
most appropriate capping technology and the types of materials that can be
used (Table 9.4). Depending on the location, seasonal conditions will
influence a number of design factors including:

Desiccation of low permeability clay layers compromising cap integrity


and increasing erosion
Loss of cap integrity due to repeated cycles of freeze/thaw
Slope instability and erosion due to cap materials becoming saturated
Ability to establish adequate vegetative growth

324
Table 9.4: Capping technologies and climate
Applicability of Capping Technologies Based on Climate Conditions 1
Type of Cap Humid and Moist
Humid (Dfa) Dry, Subhumid (Dfb) Semi Arid (Bsk)
Subhumid (Cfa)
Single Component Caps
Silty or clayey soils
Applicable3 Applicable Potentially Applicable4 Applicable
(SM,SC). Unclassified fills
Compacted clay (CL, CH) Applicable Applicable Potentially Applicable Less Applicable2
Geomembrane Applicable Applicable Applicable Applicable
Geosynthetic clay liner
Applicable Applicable Potentially Applicable Less Applicable
(GCL)
Composite Caps
Clay with Geomembrane Applicable Applicable Potentially Applicable Potentially Applicable
Geosynthetic clay liner
(GCL) with Applicable Applicable Potentially Applicable Potentially Applicable
Geomembrane
Evapotranspiration (ET) Caps
Monolithic Less Applicable Less Applicable Potentially Applicable Applicable
Capillary Break Less Applicable Less Applicable Potentially Applicable Applicable
[1] The climatic zones are based on Thornthwaite (1948): humid, sub-humid, semiarid; with supplemental Koppen-Geiger
classifications by Rubel and Kottek (2010) where Bsk=arid, steppe, cold; Cfa=warm temperate, fully humid, hot summer; Dfa=snow,
fully humid, hot summer, and Dfb=snow, fully humid, warm summer.
[2] Less Applicable = significant engineering, construction, or maintenance issues that must be resolved for effective long-term
implementation of the technology. [3] Applicable = can be applied effectively using standard engineering and construction
practices. [4] Potentially Applicable = advanced engineering or construction practices may be needed to effectively implement the
technology.

325
The two critical design parameters most affected by geographic location are
freeze/thaw durability and desiccation. Therefore, careful consideration should
be given to the type of barrier selected in single component cap applications
where extreme fluctuations in temperature and/or precipitation occur.
Composite component cap applications using geo-membranes enhance long-
term performance.

Freeze/thaw durability and desiccation are less likely for monolithic or


capillary break ET caps because their performance is based upon water
storage potential and the evapotranspiration of surface water infiltration rather
than serving as a simple barrier layer.

326
9.5 Post-construction monitoring
Once construction of the S/S remedy, including the cap and any ancillary
features, is complete, then an extended period of monitoring is often required.
The objectives/requirements of the monitoring program are often very site-
specific, and dependent on the requirements of the site owner and regulatory
agencies. It is best to define the objectives and general requirements for post-
construction monitoring (including the frequency of monitoring events,
longevity of monitoring, and general reporting requirements) during the initial
design phase, before construction of the S/S remedy begins.

Details of the monitoring system, such as exact placement and design of


monitoring wells, are best determined at the close of remedy construction as
the precise quantities of soil treated and its final placement, including the cap,
often vary in detail from the original pre-construction design. Considerations
for post-construction monitoring of S/S sites are further discussed below.
However, this section is not intended as a definitive reference for developing
post-construction monitoring programs

9.5.1 Regulatory impetus and guidance for post-construction


monitoring
Key factors for the development of a post-construction monitoring program
are requirements of the responsible regulatory agencies. These may vary
greatly depending on jurisdiction and upon site-specific factors.

An example is the USEPA guidance on post-construction completion activities


at Superfund sites (USEPA, 2005 and 2012). USEPA 2005 lists activities as:

Long-Term Response Action (LTRA) generally applies to the first 10-years


of fund financed ground and surface water restoration.

Operation and Maintenance (O&M) includes the activities required to


maintain the effectiveness and integrity of the remedy. Also includes
continued operation of ground and surface water restoration remedies after
LTRA.

Five-Year Reviews are required by statute to assure protectiveness for any


remedial action that leaves hazardous substances on a site above levels that
allow for unlimited use and unrestricted exposures. Five-year reviews are also
conducted as a matter of policy in other situations.

327
Institutional Controls (IC) using non-engineered instruments, such as
administrative and/or legal controls, that typically minimise the potential for
human exposure to contamination and/or protect the integrity of the remedy
by limiting land or resource use.

Remedy Optimisation involves performing reviews to improve the


performance and/or reduce the annual operating cost of remedies without
compromising protectiveness.

NPL Deletion is the removal of sites or portions of sites from the NPL
(National Priority List) because no further response action is appropriate.

Reuse involves working with the parties seeking to redevelop Superfund sites
to ensure that their activities do not adversely affect the implemented remedy.

Except for NPL deletion, these activities may apply to any S/S remediation
under any jurisdiction, since the S/S process leaves contaminants, though
immobilised, in place within the S/S treated matrix.

Therefore, activities to assure the continued protectiveness and maintenance


of the remedy are important, and include regular reviews such as five-year
reviews, mentioned above.

For Superfund sites where contaminants are left in place after construction
completion, a review is required every 5 years to assure that the remedy is
still protective and is well maintained. Guidance for conducting such 5-year
reviews has been prepared by the USEPA (USEPA, 2001). Although
designed specifically (for the USEPA Superfund program) the guidance,
which includes check-lists is a valuable resource for post-construction
inspection, monitoring, and reporting, at any site where S/S has been
employed.

A number of completed 5-year review reports are available on the USEPA


website (USEPA, http://cumulis.epa.gov/fiveyear/index.cfm).

328
Additional guidance is available in the Environment Agency guidance
document on the use of S/S (Environment Agency, 2002), especially Section
5 and Appendix 4. The EA document (page 56) suggests that Specific
objectives for long-term monitoring for a re-use scenario may be:

To demonstrate whether S/S remains effective


To provide a basis for implementing mitigation measures
To identify detrimental changes in the re-use scenario (e.g. water table
rise)
To provide a basis for ceasing monitoring.

The EA document also discusses what topics should be included in a


monitoring report and the potential decision basis for discontinuing monitoring.

The ITRC (Interstate Technology & Regulatory Council) has also recently
published guidance on Development of Performance Specifications for S/S,
including in Section 7, post-construction monitoring, referred to as
stewardship in their document.

The ITRC document states (ITRC, 2011, page 52): Long-term stewardship of
a completed S/S remedy may include monitoring of environmental media in
contact with and potentially affected by the remedy, monitoring of institutional
controls, monitoring and maintenance of engineering controls, financial
assurances, and periodic review(s) by the controlling environmental agency.

9.5.2 Specific features common in monitoring programs


Monitoring of remediated S/S sites is done for the following purposes:

To assure that the S/S treated material continues to meet its original design
performance property of reducing the release rate of contaminants to
groundwater or surface water bodies through low permeability and low
leachability. As discussed in Section 6.4.4, it is impractical and generally not
necessary to obtain representative samples of cured materials in the field long
after treatment is completed. Rather, monitoring the effects of treatment
through groundwater quality monitoring adjacent to and/or down gradient of
the treatment zone is fairly standard practice.

329
Demonstration that release of contaminants to groundwater is adequately
controlled over the long-term is usually made by locating groundwater wells
immediately adjacent to the placed S/S material, and sampling for a selected
list of COCs, along with field measurements for pH, Eh, and other selected
indicator parameters.

Immediately adjacent is a relative term as it is influenced by the rate of


groundwater movement and physical access including avoidance of
compromising the cap. For example a distance of 10 -100 ft (3-30 m) from the
edge of the monolith can be considered as normal. Often a subset of the
COCs present in the S/S treated material is selected with preference for
those most likely to be detected (most soluble) should the S/S material fail to
maintain its control over release.

Although sampling/monitoring is often conducted quarterly (or seasonally) for


the first couple of years, the frequency may be reduced thereafter to semi-
annually or annually if no issues are detected. It should be noted that
immediately after S/S treatment, especially in-situ S/S treatment, there will be
a slightly elevated pH in the groundwater contacting the treated material. This
is normal, should dissipate in a few months, and does not indicate a failure of
the treated material. An example of a monitoring well network around an S/S
monolith is indicated in Figure 9.22.

To document that groundwater quality is improving after remediation:

Often where S/S is implemented as the source control technology to treat


contaminated soils, groundwater has also been impacted through the release
of COCs from the contaminated soils. Thus, once the source has been
successfully managed by S/S, it is reasonable to expect the groundwater
quality to improve with time.

For some sites, this has led to selection of MNA (monitored natural
attenuation) as either the primary, or secondary, method for remediating
impacted groundwater. Therefore, monitoring wells are installed and a
monitoring plan developed to assess groundwater improvement over time and
eventual achievement of the agreed groundwater quality goals.

To document that engineering controls are functioning properly and


maintenance is being conducted:

Engineering controls can include fencing, the cap itself, surface water
diversion/runoff controls, and vertical impermeable walls/barriers.

330
Figure 9.22: Groundwater monitoring wells around an S/S monolith at the Sanford gasification plant site

331
Caps, surface water diversion controls and other surface engineering are inspected
during the periodic site visits, for groundwater sample collection events, and after
unusual precipitation events. Subsurface engineering controls like vertical
impermeable walls are monitored indirectly through groundwater monitoring in the
same manner as the S/S treated material. If the site has a vegetative cap, this is
inspected and any necessary repair carried out. It is important that invasive trees
and shrubs be removed from caps as their roots can not only cause damage, but
may over time invade the S/S material.

To document the continued application of institutional controls:

Institutional controls include deed restrictions and other prohibition of excavation


within the S/S treated material or construction of wells through the treated material
without prior approval from the regulatory agencies. They may also restrict how the
surface can be used. Controls are applied by regulatory agencies and often enforced
at a local level. Any inappropriate site use is noted during periodic site inspections
and a check can be made to assure that institutional controls are being recorded on
property transfer deeds.

To document the exit-strategy:

This is highly variable and dependent on the nature of regulatory oversight and site-
specific characteristics.

Under the USEPA Superfund program, S/S treated soils, which still contain the
COCs, despite being treated to control release, fall into the same category as
containment cells, and 5-year reviews are required in perpetuity.

Sites remediated under State programs in the USA follow the requirements of
individual states, which are highly variable. The EA guidance document on S/S, as
previously mentioned, provides information on the potential decision basis for
discontinuing monitoring. In any case, the site owner, regulatory agency(s), and
other stakeholders will need to reach a consensus on what aspects of monitoring
can be discontinued and on what basis. The authors believe that if the S/S treatment
is still performing well after 15-20 years, it will likely continue to do so indefinitely.
Support for this is provided in the PASSiFy project report (PASSiFy 2010) evaluating
the long-term performance of S/S treated material at a number of sites in the USA,
France and the UK.

332
10.0 Consolidated Reference List

ASTM International, http://www.astm.org/DIGITAL_LIBRARY/index.html

ASTM D422, Standard Test Method for Particle-Size Analysis of Soils, American
Society for Testing and Materials, Vol. 4.08, D422, 2007

ASTM D698, American Society for Testing and Materials, Philadelphia,


Pennsylvania

ASTM D1557, American Society for Testing and Materials, Philadelphia,


Pennsylvania

ASTM D1633, Standard Test Methods for Compressive Strength of Molded Soil-
Cement Cylinders, American Society for Testing and Materials, Philadelphia,
Pennsylvania

ASTM, Standard Guide For Developing Conceptual Site Models for Contaminated
Sites, Standard E1689-95 in Book of Standards Volume 11.06, ASTM
International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA
19428, 2008, www.astm.org

ASTM E2081 - 00(2010)e1, Standard Guide for Risk-Based Corrective Action


Environment Agency, Model Procedures for the Management of Land
Contamination, CLR 11, 2004

ASTM D2487, Practice for Classification of Soils for Engineering Purposes


(Unified Soil Classification System), American Society for Testing and Materials,
Vol. 4.08, D2487, 2011

ASTM D2488, Standard Practice for Description and Identification of Soils (Visual-
Manual Procedure), American Society for Testing and Materials, Vol. 4.08, D2488,
2009

ASTM D4318, Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity
Index of Soils, American Society for Testing and Materials, Vol. 4.08, D4318, 2010

333
ASTM D5084, Standard Test Methods for Measurement of Hydraulic Conductivity
of Saturated Porous Materials Using a Flexible Wall Permeameter, American
Society for Testing and Materials, Philadelphia, Pennsylvania

Barbisan, U., Guardini, M., (2007). Reinforced concrete: a short history.


Tecnologos, Cavriana, Mantova, Italy

Bates, E.R., Akindele, F. and Springle, D., (2002). American Creosote Site Case
Study: Solidification/Stabilisation of Dioxins, PCP, and Creosote for $64 per Cubic
Yard. Environmental Progress 21, 2, pp 79-84

Benson, C.H., Zhai, H., and Wang, X., (1994). Estimating Hydraulic Conductivity of
Compacted Clay Liners. Journal of Geotechnical Engineering, 120, 2, pp 366-387

Boyd, S.A., and Mortland, M.M., (1987). Use of Modified Clays for Adsorption and
Catalytic Destruction of Contaminants, in 13th Ann. Research Symposium at
Cincinnati. U.S. Environmental Protection Agency, Cincinnati, OH, (July 1987)

Bozkurt, S., Moreno, I. and Neretnieks, I., (2000). Long Term Processes in Waste
Deposits. The Science of the Total Environment 250, 1-3, pp 101-121

British Cement Association, (2004). The Essential Guide to


Stabilisation/Solidification for the Remediation of Brownfield Land using Cement
and Lime. BCA Report 46.112

Butler, S. M., Barth, E. F., and Barich, J. J., (1996). Field Quality Control
Strategies for Assessing Solidification/Stabilisation, Stabilisation and Solidification
of Hazardous, Radioactive, and Mixed Wastes: 3rd Volume, ASTM STP 1240, T.
(Michael Gilliam and Carlton C. Wiles, Eds.), American Society for Testing and
Materials

Caterpillar Compaction Manual, (1989). Caterpillar Tractor Company, Peoria,


Illinois

Chateau, Laurent, personal communication, (November 2012)

Christenson, H. and Wakamiya, W., (1987). In HazTech News, (June 18)

334
CFEM, Canadian Foundation Engineering Manual, Canadian Geotechnical
Society, (2006)

Conner, J.R., (1990). Chemical Fixation and Solidification of Hazardous Waste.


Van Nostrand Reinhold

Control of Pollution Act (1974): http://www.legislation.gov.uk/ukpga/1974/40

Control of Pollution (Amendment) Act (1989):


http://www.legislation.gov.uk/ukpga/1989/14. Accessed August 8, 2012

Cote, P.,(1987). Assessment of Solidification Technologies for the Immobilisation


of Organic Compounds, Draft by Wastewater Technology Centre, Environment
Canada, (April 1987)

Delatte, N.J. (2001). Lessons from Roman cement and concrete. Journal of
Professional Issues in Engineering Education and Practice 127, 3, pp 109-115

Dragun J., (1988). The Soil Chemistry of Hazardous Materials. Hazardous


Materials Control Research Institute, Silver Springs, MD

Dragun, J., and Heller, C.S., (1985). Physicochemical and structural relationships
of organic chemicals undergoing soil and clay catalyzed free radical oxidation. Soil
Science 139, 2, pp 100-111

Ells, C., (2010). In: Proceedings of the International Solidification/Stabilisation


Technology Forum. (Eds. Craig B. Lake and Colin D. Hills). June 14th -17th, Cape
Breton University, Sydney, NS, Canada.

Emerick, K., (1995). The Survey and Recording of Historic Monuments. Quarterly
Journal of Engineering Geology 28, pp 201-205

ENDS, (1988). Report 158, P8


http://www.sciencedirect.com/science/article/pii/S0166111697802257

ENDS, (1983). Report 120, pp 10-13

335
Environment Agency, (2004a), Guidance on the Use of Stabilisation/Solidification
for the Treatment of Contaminated Soil, Codes and Standards for Stabilisation and
Solidification Technologies (CASSST) Project, Science Report SC980003/SR1,
September 2004,
www.environment-agency.gov.uk

Environment Agency, (2004b), Review of Literature on the use of


Stabilisation/solidification for the Treatment of Contaminated Soil, Solid Waste and
Sludges. Codes and Standards for Stabilisation and Solidification Technologies
(CASSST) Project, Science Report SC980003/SR2, September 2004,
www.environment-agency.gov.uk

Environment Agency, (2004c). Model Procedures for the Management of Land


Contamination, CLR 11

EPRI,(2009). An Integrated Approach to Evaluating In-Situ


Solidification/Stabilization of Coal Tar Impacted Soils, Electric Power Research
Institute, EPRI, Palo Alto, CA, March 2009, Report 1018612

European Landfill Directive (1999). Directive 1999/31/EC. http://eur-


lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31999L0031:EN:NOT

Fernndez Pereira, C., Luna, Y., Querol, X. D. Antenucci, D. and Vale, J., (2009).
Waste stabilisation/solidification of an electric arc furnace dust using fly ash-based
geo-polymers. Fuel 88, 7, pp 1185-1193

Fleri, M. A., and Whetstone, G. T., (2007). In-situ Stabilisation/Solidification:


Project Lifecycle, Journal of Hazardous Materials, 141, pp 441-456

Garrido-Ramirez, E.G., Theng, B.K.G. and Mora, M.L., (2010). Clays and oxide as
catalysts and nanocatalysts in Fenton-like reactions A Review. Applied Clay
Science 47, 3-4, pp 182-192

Gunning, P.,(2011). Accelerated carbonation of hazardous wastes (PhD Thesis)


oai:gala.gre.ac.uk:7135.
http://gala.gre.ac.uk/7135/1/Peter_John_Gunning_Accelerated_carbonation_2011.
pdf

Goddard, E.N., (1979). Rock Colour Chart, Geological Society of America, Boulder
Colorado

336
Guo, X. and Shi, H., (2012). Self-Solidification/Stabilisation (S/S) of Heavy Metal
Wastes (HMWs) of the Class C Fly Ash (CFA)-Based Geopolymers. Journal of
Materials in Civil Engineering doi:10.1061/(ASCE)MT.1943-5533.0000595

Hall, C., (1976). On the History of Portland Cement after 150 Years. Journal of
Chemical Education 53, 4, pp 222-223

Hills, C.D., Antemir, A., Leonard, S.A. and Carey, P.J. (2010). Applications of
Stabilisation/Solidification for the Treatment of Organically Contaminated Soil and
Waste. In: Proceedings of the International Solidification/Stabilisation Technology
Forum. (Eds. Craig B. Lake and Colin D. Hills). June 14th-17th, Cape Breton
University, Sydney, NS, Canada.

Hobbs, D.W and Taylor, M.G., (2000). Nature of the Thaumasite Sulfate Attack
Mechanism in Field Concrete. Cement and Concrete Research 30, pp 529-533

Integrated Pollution Prevention and Control: http://www.defra.gov.uk. Accessed


August 18, 2012

ITRC, (2011). Development of Performance Specifications for


Solidification/Stabilisation, Interstate Technology Regulatory Council, ITRC, 50 F
Street, NW, Suite 350,Washington, DC 20001,
http://www.itrcweb.org/gd.asp

Jana, D., (2007). Evidence from Detailed Petrographic Examinations of Casing


Stones from the Great Pyramid of Khufu, a Natural Limestone from Tura, and a
Man-made (Geopolymeric) Limestone. Proceedings of the 29th Conference on
Cement Microscopy. Quebec City, Canada, May 20 -24, 2007

Johnson, R., (1987). History and development of modified bitumen. Proceedings


of the 8th Conference on Roofing Technology, pp 81-84

Kanare, H., Milevski, I., Khalaily, H., Getzov, N., Nasvik, J.,(2009). How Old is
Concrete? Concrete Construction, January 2009

Kiefer, D.M. Its all about Alkali, in Chemistry Chronicles:


http://pubs.acs.org/subscribe/archive/tcaw/11/i01/html/01chemchron.html
Accessed August 8, 2012

337
Klich, I., (1997). Permanence of Metal Containment in Solidified and Stabilized
Wastes. Unpublished Ph.D. Thesis University of Texas A and M

Kostecki, P.T., Calabrese E.J., and Bonazountas M., (Eds.), (1992). Hydrocarbon
Contaminated Soils (Volume 2). Proceedings 5th Conf on Hydrocarbon
Contaminated Soils, Amherst, Mass. Sept. 24 -27, 1990. CRC Press

Kyles, J.H., et al., (1987). Solidification/Stabilisation of Hazardous Waste: A


Comparison of Conventional and Novel Techniques, in Toxic and Hazardous
Wastes: Proceeding of the Nineteenth Mid-Atlantic Waste Conference. Bucknell
University, Lewisburg, PA.

Lake, C.B., (2013). Assessing Geo-Environmental Performance of Cement-Based


Containment Systems, 2011 Canadian Geotechnical Society Colloquium Address,
to be submitted to the Canadian Geotechnical Journal

Landfill of Waste: http://ec.europa.eu/environment/waste/landfill_index.htm


Accessed August 8, 2012

Lange, L.C., Hills, C.D. and Poole, A.B., (1997). Effect of Carbonation on
Properties of Blended and Non-blended Cement Solidified Waste Forms. Journal
of Hazardous Materials 52, pp 193-212

Lange, L.C., Hills, C.D. and Poole, A.B., (1996). Preliminary Investigation into the
Effects of Carbonation of Cement-Solidified Hazardous Wastes. Environmental
Science and Technology 30, pp 25-30

Lear, P.R., and Conner, J., (1992). Immobilisation of Low Level Organic
Compounds in Contaminated Soil, in P. Kostecki et al. (Eds.). Hydrocarbon
Contaminated Soils, Volume II, Lewis Publishers, Chelsea, MI.

MacKenzie, K. J. D., Smith, M.E., Wong, A., Hanna, J.V., Barry, B. J., Barsoum,
M. W., (2011). Were the casing stones of Senefrus Bent Pyramid in Dahshour
cast or carved? Materials Letters 65,2, pp 350-352

MacLaren, D.C., White, M.A., (2003). Cement: Its Chemistry and Properties.
Journal of Chemical Education 80, 6, pp 623-635

338
Mays, L.W., (2010). A Brief History of Water Technology During Antiquity: Before
the Romans. In: Mays, L.W. (Eds.) Ancient Water Technologies, Springer Science

Miller, R., (2009). Diamonds and Cement: A Review of Concrete Over the Ages.
International Polished Concrete Institute. Available at http://www.ipcionline.org/.
Accessed 14.08.12
http://www.ipcionline.org/index.cfm?fuseaction=journal.showArticle&mainArticle=E
E5D6012-D9BE-42B5-B04F-027A3C7068DD&categoryID=CS

McConnell, J.D.C., (1955). The Hydration of Larnite (B-Ca2SiO4) and Bredigite


(A1-Ca2SiO4) and the Properties of the Resulting Gelatinous Plombierite. Mineral
Magazine 30, pp 672-680

Montgomery, D.M., Sollars, C.J., Perry, R., Tarling, S.E., Barnes, P. and
Henderson, E., (1991). Treatment of organic-contaminated industrial wastes using
cement-based stabilisation/solidification 1. Microstructural analysis of cement-
organic interactions. Waste Management and Research 9, 2, pp 103-111.

Moorey, P.R.S., (1994). Ancient Mesopotamian Materials and Industries: The


Archaeological Evidence. Oxford University Press

Mottershead, D.N., (2000). Weathering of Coastal Defensive Structures in South-


West England: A 500 Year Stone Durability Trial. Earth Surface Processes and
Landforms 25, pp 143-1159

Ortego, J.D.,(1989). New Chemical Technology for the Containment of Hazardous


Substances, personal communication, (June, 1989)

PASSIFy, (2010). Performance Assessment of Stabilized/Solidified Waste Forms,


Final Report. CL:AIRE Project RP16.
http://www.claire.co.uk/index.php?option=com_cobalt&view=record&cat_id=23:sta
bilization-solidification&id=298:performance-assessment-of-stabilizedsolidified-
waste-forms-passify&Itemid=61

Penseart, S., De Groeve, S., Staveley, C., Menge, P and De Puydt, S., (2008).
Immobilisation, Stabilisation, Solidification: A New Approach for the Treatment of
Contaminated Soils. Case Studies: London Olympics and Total Ertveldt. 15th
Innovation Forum Geotechnique

339
Perara, A.S.R., Al-Tabbaa, A., and Johnson, D., (2004). State of Practice Report,
UK Stabilisation/Solidification Treatment and_ Remediation Part_VI: Quality_
Assurance__and__Quality_ Control_
www-starnet.eng.cam.ac.uk, Cambridge University, December 2004

Pojasek, R., (1978). Stabilisation, Solidification of Hazardous Wastes.


Environmental Science and Technology 12, 4, pp 382-336

Posasek, R. B., (1980). (Ed.) Toxic and Hazardous Waste Disposal, Vol. 4. Ann
Arbor Science Publishers, Inc., Ann Arbor, MI.

Portland Cement Association, Guide to Improving the Effectiveness of Cement-


Based Stabilization/Solidification, PCA Publication Number EB211, 1997

Pollution Prevention and Control Act, (1999):


http://www.legislation.gov.uk/ukpga/1999/24 Accessed August 8, 2012

Reid, J.M. and Clarke, G.T., (2001). The Processing of Contaminated Land in
Highway Works. TRL Report 489

Rogers, R.W., (1900). A History of Babylonia and Assyria: Volume 2. Assyrian


International News Agency

Rubel, F. and Kottek, M., (2001). Observed and Projected Climate Shifts 1901-
2100 Depicted by World Maps of the Kppen-GeigerClimate Classification.
Meteorologische Zeitschrift_19, pp 135-141. DOI: 10.1127/0941-2948/2010/0430,
2001

Rudland, D. J., Lancefield, R. M., Mayell, P. N., (2001). Contaminated Land Risk
Assessment A Guide to Good Practice, CIRIA Report: C552

Shi, C., (2004). Quality Assurance/Quality Control (QA/QC) for Waste


Stabilisation/Solidification, in Stabilisation and Solidification of Hazardous,
Radioactive, and Mixed Wastes, (R. D. Spence and C. Shi, Eds.). CRC Press,
Boca Raton, FL.

STARNET: Stabilization/Solidification Treatment and Remediation Network:


http://www-starnet.eng.cam.ac.uk. Accessed August 8, 2012

340
Sweeting, M.M., (1960). The Caves of the Buchan Area, Victoria. Zeitschrift fur
Geomorphologie 2, pp 81-91

Terzaghi, K., Peck, R.B. and Mesri, G., (1996). Soil Mechanics in Engineering
Practice, John Wiley and Sons, 3rd Edition

The Single European Act:


http://europa.eu/legislation_summaries/institutional_affairs/treaties/treaties_singlea
ct_en.htm. Accessed August 8, 2012

The Sixth Environment Action programme of the European Community 2002-2012:


http://ec.europa.eu/environment/newprg/intro.htm. Accessed August 8, 2012

Thornthwaite, C.W., (1948). An Approach toward a Rational Classification of


Climate. Geographical Review 38, pp 55-94

USACE (U.S. Army Corps of Engineers), (1995). Engineering and Design


Treatability Studies for Treatability Studies for Solidification/Stabilisation of
Contaminated Material. Technical Letter No. 1110-1-158

USEPA, (2001). Comprehensive Five-Year Review Guidance, OSWER 9355.7-


03B-P, EPA 540-R-01-007, June 2001,
http://www.epa.gov/superfund/accomp/5year/index.htm

USEPA, (1989). Stabilisation/Solidification of CERCLA and RCRA Wastes:


Physical Tests, Chemical Testing Procedures, Technology Screening, and Field
Activities, Office of Research and Development. EPA/625/6-89/022

USEPA, (1992). Guidance for Conducting Treatability Studies under CERCLA.


EPA/540/R- 92/071a, 1992, United States Environmental Protection Agency,
Office of Research and Development

341
USEPA, (1993). Technical Resource Document: Solidification/Stabilisation and Its
Application to Waste Materials. Office of Research and Development. EPA/530/R-
93/012.,
http://nepis.epa.gov/Exe/ZyNET.exe/2000DYKB.TXT?ZyActionD=ZyDocument&Cl
ient=EPA&Index=1991+Thru+1994&Docs=&Query=&Time=&EndTime=&SearchM
ethod=1&TocRestrict=n&Toc=&TocEntry=&QField=&QFieldYear=&QFieldMonth=
&QFieldDay=&IntQFieldOp=0&ExtQFieldOp=0&XmlQuery=&File=D%3A%5Czyfile
s%5CIndex%20Data%5C91thru94%5CTxt%5C00000008%5C2000DYKB.txt&Use
r=ANONYMOUS&Password=anonymous&SortMethod=h%7C-
&MaximumDocuments=1&FuzzyDegree=0&ImageQuality=r75g8/r75g8/x150y150
g16/i425&Display=p%7Cf&DefSeekPage=x&SearchBack=ZyActionL&Back=ZyAct
ionS&BackDesc=Results%20page&MaximumPages=1&ZyEntry=1&SeekPage=x&
ZyPURL

US Department of Energy, (2000). Ancient Cementitious Materials. Available at:


http://www.wipp.energy.gov/PICsProg/documents/Ancient%20Cementitious%20M
aterials.pdf Accessed August 14, 2012

USEPA, (1986). Handbook for Stabilisation/Solidification of Hazardous Wastes,


Office of Research and Development, EPA/540/2-86/001, June 1986

USEPA, (2008). 2005 and 2008 Corrective Action Baselines - Final Results,
Summary of Environmental Indicator and Final Remedy results: 2006-2008 (PDF),
http://www.epa.gov/epawaste/hazard/correctiveaction/baseline.htm

USEPA, Five Year Reviews Online, undated, USEPA, Five Year Reviews Online.
http://cumulis.epa.gov/fiveyear/index.cfm

USEPA, (1993). Guidance for Evaluating the Technical Impracticability of


Groundwater Restoration, U.S. Environmental Protection Agency, OSWER
Directive 9234.2-25, 1993, EPA/540/R/93/080

USEPA, National Priorities List (NPL),


http://www.epa.gov/superfund/sites/npl/index.htm, Accessed January 7, 2014

342
USEPA, (2005, 2012). National Strategy to Manage Post Construction Completion
Activities at Superfund Sites, OSWER 9355.0-105,
http://www.epa.gov/superfund/cleanup/postconstruction/

USEPA, (2000). Solidification/Stabilisation Use at Superfund Sites, EPA-542-R-


00-010, Page 5, http://www.clu-in.org/download/remed/ss_sfund.pdf

USEPA, (2011). Summary of the Resource Conservation and Recovery Act,


March 2011, http://www2.epa.gov/laws-regulations/summary-resource-
conservation-and-recovery-act

USEPA, (2010). Superfund Remedy Report, Thirteenth Edition, September 2010,


EPA-542-R-10-004, http://clu-
in.org/download/remed/asr/13/SRR_13th_MainDocument.pdf

USEPA, (2013). Superfund Remedy Report, Fourteenth Edition, 2013, EPA 542-
R-13-016, http://www.clu-in.org/download/remed/asr/14/SRR_14th_2013Nov.pdf

USEPA, (2007). Treatment Technologies for Site Clean up: Annual Status Report
Twelfth Edition, 2007, EPA-542-R-07-012
http://clu-in.org/download/remed/asr/12/asr12_main_body.pdf

USEPA, (2009). Technology Performance Review: Selecting and Using


Solidification/Stabilisation Treatment for Site Remediation, U.S. Environmental
Protection Agency, National Risk Management Research Laboratory, Office of
Research and Development, Cincinnati, OH, November 2009, EPA/600/R-09/148,
https://clu-in.org/download/techdrct/600r09148.pdf

USEPA, (2011). Whitehouse Oil Pits, Site Summary Profile, updated on March 22,
2011
http://www.epa.gov/region4/superfund/sites/npl/florida/whthowsopfl.html

van Zomeren A., van Wetten, H., Dijkstra, J.J. van der Sloot, H.A. and Bleijerveld,
R., (2003). Long Term Prediction of Release From a Stabilised Waste Monofill and
Identification of Controlling Factors. In: Proceedings Ninth International Waste
Management and Landfill Symposium, 6-10th October, 2003, S. Margherita di
Pula, Cagliari, CISA, Italy
343
Wayman, E., (2011). The Secrets of Ancient Romes Buildings: What is it about
Roman concrete that keeps the Pantheon and the Colosseum still standing?
Available at http://www.smithsonian.com. Accessed 14.08.2012

Wheeler, P., (1995). Leach Repellent. Ground Engineering 28, 5, pp 20-22

Yang, F., Zhang, B., Ma, Q., (2010). Study of Sticky Rice - Lime Mortar
Technology for the Restoration of Historical Masonry Construction. Accounts of
Chemical Research, 43, 6, pp 936-944

344
11.0 Contributing author biographies

345
AUTHOR PROFILE: KEN ANDROMALOS

CURRENT POSITION: Vice President of Engineering

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Pittsburgh B.S. 1978 Civil Engineering


University of Pittsburgh M.S. 1984 Civil Engineering
Professional Engineer 1990 Pennsylvania

RELEVANT PROFESSIONAL EXPERIENCE

Has 35 years experience in engineering,


construction management and operations in the
specialty geotechnical and environmental
remediation markets.

Has acquired extensive project experience,


across the United States and Canada.

Remediation experience includes: off-site


removal activities, in-place closures, sludge stabilization, dredging, slurry walls and
other types of groundwater cut-off walls, in-place soil mixing, soil vapor extraction,
groundwater pump and treatment systems, and in-situ solidification/ stabilization
(ISS).

Kens remediation experience includes working at various active and in-active


facilities, including complete steel making facilities (including coking operations),
wood treating sites, manufactured gas plant (MGP) sites, Superfund sites, former

346
Chemical Warfare facilities, petroleum refineries, manufacturing facilities, landfills,
chemical plants, and numerous disposal sites.

His ISS and soil mixing experience includes:

25 years of direct soil mixing project experience


20 years of direct ISS project experience
Numerous treatability studies for solidification/stabilization

He has co-authored over two dozen technical papers and articles on specialty
geotechnical techniques and environmental remediation.

Ken is a regular presenter at various remediation and geotechnical conferences.

347
AUTHOR PROFILE: EDWARD R. BATES

CURRENT POSITION: Consultant for Remediation of Hazardous


Waste Sites

E MAIL: [email protected]

EDUCATION / REGISTRATION

Michigan State University B.S. 1968 Geology


Michigan State University M.S. 1970 Geology

RELEVANT PROFESSIONAL EXPERIENCE

Retired from the U. S. Environmental


Protection Agency in 2009 after 32 years of
service , including over 20 years working as
an expert technical advisor on site
characterization, remedy design, and remedy
construction for nearly 100 CERCLA and
RCRA sites including more than :

25 Solidification/Stabilization Sites
22 Treatability Studies for S/S
20 Site-specific sets of Specifications for S/S

19 Invited lectures/symposia presentations


14 Publications on technology and remediation of hazardous waste sites

Contributor to the ITRC Guidance Document: Development of Performance


Specifications for Solidification/Stabilization, http://www.itrcweb.org/gd.asp

Member of the International PASSiFy Project Team for Preparing the Document:
Performance Assessment of Solidified/Stabilized Waste-forms, An Examination of
348
the Long-term Stability of Cement-treated Soil and Waste (Final Report), (2010).
http://www.claire.co.uk/index.php?option=com_phocadownload&view=file&id=308:R
esearch%20Project%20Reports&Itemid=25

SELECTED AWARDS AND HONORS

U.S. EPA 9 Bronze medals


Special Commendation from the Assistant Administrator of the Office of Research
and Development For more than a decade of outstanding technical support and
assistance to EPA Regions and Programs in their mission to manage and clean
up waste sites
EPA Scientific and Technological Achievement Award for Advancing the state of
the science for cost effective remediation of contaminated wood preserving sites
EPA Scientific and Technological Achievement Award for chemical stabilization of
mixed organic and metal compounds
U.S. EPA Administrator`s 2008 National Notable Achievement team Award for
Cross Program Results for Land Revitalization, Camilla Wood Preserving Site

349
AUTHOR PROFILE: STEVEN R. BIRDWELL

CURRENT POSITION: President and CEO for Remedial Construction


Services, LP

E MAIL: [email protected]

EDUCATION / REGISTRATION

Baylor University B.S 1986 Engineering Geology


Missouri-Rolla University M.S 1987 Geological Engineering

RELEVANT PROFESSIONAL EXPERIENCE

Mr. Birdwell has over 27 years of experience in the


management of environmental remediation and
geotechnical construction projects.

He began his career in 1987 and then founded


Remedial Construction Services, L.P. (RECON) in
1989 to meet the demand for environmental services
created by new regulation.

Mr. Birdwell has presented at 4 International conferences on environmental


remediation technologies, including in-situ bioremediation, stabilization of ash
products, and the remediation of chlorinated solvents in groundwater.

Mr. Birdwell currently holds two patents dealing with solidification and stabilization of
sludge:

Patent US 8,193,115 B2 For Solidifying High Moisture Sludge


Patent US 7,897,831 B2 Method for Solidifying High Moisture Sludge,
Solidified Sludge Therefrom.
350
In addition to his executive duties, Mr. Birdwell assists Program and Project
Managers with project planning, resolution of technical operational issues, contract
management, client management and strategic planning.

Mr. Birdwell also leads RECONs team of professionals tasked with taking new
technologies to market.

351
AUTHOR PROFILE: DONALD (DONNIE) BURKE

CURRENT POSITION: Executive Project Director, Nova Scotia


Lands Inc.

E MAIL: [email protected]

EDUCATION / REGISTRATION

University College of Cape Breton Diploma 1992 Environmental


Technology

Technical University of Nova Certificate 1994 Environmental


Scotia Management

University College of Cape Breton Bachelor 1999 Environmental


Science
Nova Scotia Public Service Certificate 2012
Commission Executive Leadership

Saint Mary`s University Certificate 2014 Financial


Management

RELEVANT PROFESSIONAL EXPERIENCE


Mr. Burke is an Executive Director with the Province of Nova
Scotia with more than 20 years experience in administrative
and project management in the government (both municipal
& provincial) and private sector particularly in environmental
remediation and civil construction. During his career he was involved in or
responsible for schedule, budget, quality control, environmental management,
regulatory compliance, stakeholder management and risk for large projects.
Most recently:

Sydney Tar Ponds and Coke Ovens Cleanup Project Project Director-
Oversaw all aspects of engineering design, contract/consultant
management, construction oversight, project quality, environmental
management. Directly engaged in community engagement, health and
safety, public relations and stakeholder management.

352
Cape Breton Regional Municipality - Solid Waste Manager; managed the
design and implementation of a $30m overhaul of the CBRM Waste
Management System inclusive of siting and construction of a new central
composting facility, waste transfer station and the addition of a new waste
collection stream for commercial and curbside collection.
Cape Breton Regional Municipality - Public Works Manager Central
Division - managed the day to day operations of public works departments
for the former City of Sydney and surrounding areas. Departments included
Street Department, Traffic Department, Water Department and Sewer
Department. Duties also include capital works management (design,
contact management and construction oversight).

5 Invited lectures/symposia presentations on environmental management and


project management
9 Conference presentations on remediation management and/or brownfield
redevelopment

MEMBERSHIPS

The Project Management Institute


Tech Nova Certified Technology Professionals
ECO Canada

SELECTED AWARDS AND HONORS

Project Management Professional (PMP)


Environmental Professional (EP)
Governor General`s Award for Excellence in Public Service Recipient

353
AUTHOR PROFILE: KATE CANNING

CURRENT POSITION: Geoenvironmental Consultant, Arup

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Newcastle Upon Tyne M.Eng 1996 Civil & Environmental


Engineering
University of Surrey PhD 2000 Civil Engineering

RELEVANT PROFESSIONAL EXPERIENCE

Geo-environmental consulting engineer


specialising in brownfield development, with
over 15 years experience. Following a PhD
evaluating the use of synthetic zeolites for the
treatment of heavy metal contaminated
effluents, Kate undertook a research
fellowship investigating the use of novel
cement stabilization methods for the
remediation of contaminated soils and wastes at the University of Greenwich.
Funded by Blue Circle Cement, the project included establishing field and
bench-scale trials for a range of stabilization/solidification techniques, and
supporting a Mobile Plant Licence application.

Kate joined Arup in 2003, and has become an experienced environmental risk
assessor and brownfield development specialist. Her involvement with
contaminated land remediation crosses all aspects of the work, from
contamination investigations, risk assessments, design of remediation works,
gas protection measure design, waste classification and the development of
sustainable earthworks strategies through to site supervision and validation of
remediation works. Most recently, her work has included outline remediation
design and abnormals costings for many complex regeneration projects. She

354
has worked on a diverse range of contaminated sites, ranging from infilled
quarries and manufacturing facilities to gasworks sites.

Kate has worked on a number of full-scale cement stabilization schemes in


the UK, including the First Street project on a former gasworks site in
Manchester.

SELECTED AWARDS AND HONORS

2010 Ground Engineering Awards: First Street Remediation Scheme -


Sustainability Award - Highly Commended
2010 Ground Engineering Awards: First Street Remediation Scheme
Geotechnical/Geoenvironmental UK Project over 1m - Highly
Commended
2009 Brownfield Briefing Remediation Innovation Awards: First Street
Remediation Scheme - Best Conceptual Model Award - Runner-up
2009 Brownfield Briefing Remediation Innovation Awards: First Street
Remediation Scheme - Best Communication Stakeholder Engagement
Award - Runner-up

355
AUTHOR PROFILE: PAULA CAREY

CURRENT POSITION: Managing Director, Carbon8 Systems Ltd

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of St. Andrews B.Sc 1978 Geology


University of Keele Ph.D 1982 Geology

RELEVANT PROFESSIONAL EXPERIENCE

Has 25 years as a university lecturer in a


department of Geology/Environmental
Science teaching up to Masters level.

Subjects include: petrography, structural


geology, metamorphic geology, waste
management, contaminated land remediation,
site investigation, and natural materials for the
construction industry.

Was co-director of the Centre for Contaminated Land Remediation at the


University of Greenwich and carried out research in the durability and testing
of construction materials and the use of accelerated carbonation in the
treatment of waste and contaminated soils.

From 2010 Dr Carey has been Managing Director of a research and


development company commercialising the use of accelerated carbonation in
waste management.

356
Has over 11 publications on accelerated carbonation.

Was a contributor to the Science Review of Solidification and Stabilization for


the Environment Agency, accompanying national guidance on S/S, published
by the Environment Agency.

Member of the International Project team for the PASSiFy Project:


Performance Assessment of Solidified/Stabilized Waste Forms, a Chartered
Geologist and Fellow of the Geological Society.

357
AUTHOR PROFILE: COLIN DICKSON

CURRENT POSITION: President of Industrial Engineering Systems Ltd.,


Halifax, Nova Scotia

EMAIL: [email protected]

EDUCATION:
Dalhouse University DipEng 1985 Engineering
Technical University of Nova Scotia BEng 1988 Engineering

RELEVANT PROFESSIONAL EXPERIENCE

Colin is a certified Professional Engineer with the


Association of Professional Engineers of Nova
Scotia (Engineers Nova Scotia in 1991). Colin
completed his Executive Master of Business
Administration at Saint Marys University in 2008 and
received the Gold Medal of Academic Excellence for
the highest academic achievement in the program.
He has enjoyed continuing his professional
development in fields related to marine,
environmental, industrial, civil and structural
engineering.
In 2010, Colin was invited by Dalhousie University to join the Faculty of
Engineering in an advisory basis as an Engineer-in-Residence to Civil and
Resource Engineering Department and lecturing on Engineering Ethics.

Colin served in the Canadian Forces as a Marine Systems Engineering


Officer (from 1983 to 1996) retiring with the rank of Lieutenant (Navy) and
receiving a Canadian Decoration (CD) in 1995. Following appointments in the
private sector he became Vice President (Engineering) for the MacDonnell
Group of companies prior to joining the Cement Association of Canada (CAC)
as the Director Business Development in 2003, He became Vice President-
Atlantic Region between July 2008 and April 2011. Until May 2013 Colin was
Project Manager of EastLinks Wireless Implementation Project, and then
358
became Zone Leader for Bell Mobility responsible for Newfoundland and
Labrador.

During his 8 years with the CAC Colin (with colleague Charles Wilk of the
Portland Cement Association) developed expertise in cement-based S/S,
throughout Canada and the United States. He attended conferences on S/S in
the UK, United States and Canada, regularly presented on S/S (e.g.
Americana, SETAC and REMTECH) and partnered with Dalhousie University
on research and development of S/S involving treatment of contaminated soil
and sediment.

Colin managed three multiday S/S science and technology conferences in


Atlantic Canada involving international experts in S/S technology, thereby
building capacity for the largest Canadian S/S project the Sydney Tar Ponds
Project.

During the Canadian Environmental Assessment Agencys Full Panel Review


for the Sydney Tar Ponds Project Colin led the technical team who prepared
and presented the performance and capacity of the cement-based
solidification/stabilization technology to the panel responding to their technical
inquiries for public review.

During the implementation of the S/S for the Sydney Tar Ponds Colin was
active in responding to technical questions from the Sydney Tar Ponds
Agency and their consultants AECOM/CBCL. In 2009-2010 Colin project
managed, led and hosted the S/S-Tech International Science and Technology
Conference in Sydney, NS. Since 2010 Colin has been maintaining the S/S
technical information network through the establishment of the LinkedIn
Group S/S-Tech.

359
AUTHOR PROFILE: CAMERON ELLS

CURRENT POSITION: Project Engineer/Manager with Cameron


Consulting Inc.

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of New Brunswick B.Sc 1987 Civil Engineering

ELEVANT PROFESSIONAL
EXPERIENCE

A Professional Engineer (P. Eng.);


environmental consultant; and
entrepreneur, based in Halifax NS.

A remediation specialist providing risk


assessment, regulatory response support, and other services, to public and
private sector clients.

Provided responsible party laboratory testing recommendations regarding


Solidification/Stabilization.

Conceived and directed bench-scale laboratory testing programs, to identify


preferred S/S mixes for site-specific project conditions.

Supported contractor teams in preparing competitive S/S application


proposals.

360
Gave conference presentations recommending specific rationales for
determining S/S technical project performance goals.

Provided industry with recommendations of preferred jurisdictions for their


next efforts at encouraging S/S applications.

Prepared a paper and presented on the relative regulatory receptiveness


among Canadian jurisdictions to potential S/S activities and applications. This
was included in the peer reviewed conference proceedings of the S/S Tech
Conference, in Sydney NS Canada (June 2010).

361
AUTHOR PROFILE: ROBERT GARRETT

CURRENT POSITION: President and Owner of Garrett


Consulting, Inc. (GCI)

E MAIL: [email protected]

EDUCATION

Auburn University 1972, B.S. Fisheries


Management

RELEVANT PROFESSIONAL
EXPERIENCE

Mr. Garrett is an experienced


professional in the assessment and
remediation of environmental
contamination. He has served as a consultant since 1972, beginning his
career in aquatic environmental assessments of both estuarine and fresh
water ecosystems. He has both participated and managed multidisciplinary
projects. His interests led him into the hazardous waste field where he has
provided site environmental assessments, laboratory analyses, and
development and implementation of remedial alternatives. Most recently, he
has been responsible for project management of hazardous waste
remediation projects, including bioremediation, soil fixation/solidification and
building decontamination. Mr. Garrett formed Garrett Consulting, Inc. in 1990.

REPRESENTATIVE PROJECT EXPERIENCE

GCI conducted remedial activities at the Brunswick Wood Preserving


Superfund Site in Brunswick Georgia. The four year project involved site
remediation, Solidification/Stabilization (S/S) of 140,000 tons of
soils/sediments (including bench- and pilot scale studies) and construction of
final subcaps and caps. GCI was able to identify and implement a more
efficient stabilization process that saved the client $5 million.

GCI conducted a soil and ground-water biodegradation project for Federal


Paper Board. The 5-acre site was contaminated with MEK, Toluene, and
other light solvents due to a tank-farm release. The site was initially
investigated using geophysics and drilling to determine the extent of
contamination. When the extent of the plume was determined, the site was
cleared of trees and approximately 60 vacuum extraction wells were installed.
Contaminated ground-water was recovered, treated biologically in above-
ground pools, and discharged to the site through spray irrigation. Above-

362
ground and in situ biological treatment resulted in 98% reduction of
contaminants within 3 months.
GCI conducted a soil and ground-water biodegradation project for Denny's,
Inc. / Proficient Food Company (PFC) in Los Angeles, California. The site
was contaminated with diesel and gasoline released from underground
storage tanks (USTs). After the USTs were removed by GCI, approximately
10,000 cubic yards of heavily contaminated soil were excavated.
Approximately 50,000 gallons of groundwater were pumped and added to the
soil. This soil was innoculated with bacteria and nutrients, and treated on site
until it met acceptable regulatory criteria. Then, the soil was backfilled and
compacted in the original excavation. On-site remediation saved PFC in
excess of $1,000,000.

GCI conducted a lagoon closure for Goodyear Tire and Rubber Company in
Pasadena, TX. The initial approach by the engineer was to solidify the
hazardous sludge that contained hydroquinone, and other VOCs, cover, cap,
and conduct 30-year post closure care. GCI demonstrated through bench
and pilot tests that biodegradation would destroy the hazardous constituents.
The initial phase of the project involved the addition of 30,000 gallons of
concentrated sulfuric acid to lower the pH from 14 to 9. Next, the liquid
fraction was drained and the sludge was solidified with ground rice hulls.
Nutrients were added, and the material was aerated with track excavators.
The material was temporarily removed from the lagoon. A clean bottom was
obtained, and a drainage layer was placed over the bottom. After the material
had biodegraded, it was replaced in the lagoon, graded, seeded and mulched.
Benzene concentrations were reduced from 3,000 ppm to less than 5 ppm.
This eliminated the need for our client to conduct 30 years of post closure
care and saved hundreds of thousands of dollars.

SAMPLE CERTIFICATIONS

Qualified Consultant of Environmentally Sensitive Property


Hazardous Materials Instructor
Licensed UST Contractor
Certified Phase I Auditor

363
AUTHOR PROFILE: PETER GUNNING

CURRENT POSITION: Technical Project Manager, Carbon8 Systems Ltd


Visiting Lecturer, University of Greenwich, UK

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Greenwich BSc. 2004 Geology


University of Greenwich MSc 2006 Chemical Science
University of Greenwich PhD 2011 Chemical Science

RELEVANT PROFESSIONAL EXPERIENCE

Dr Gunning joined Carbon8 in 2008 specialising in


the treatment of contaminated soil and waste.

Since his appointment, he has been responsible


for laboratory research including product and
process development on industrial waste recycling,
treatment of contaminated materials, and mineral
carbon capture.

He has co-ordinated numerous pilot-scale proof-of-


concept trials including the design and construction
of novel carbon capture and waste/soil treatment
plants.

364
Dr. Gunning has continued to contribute to research at the University of
Greenwich, and has an active role in the supervision of the research team at
the Centre for Contaminated Land Remediation.

In 2011 he became a Visiting Lecturer at the University of Greenwich (Centre


for Contaminated Land Remediation).

He has published numerous journal papers and articles on waste


management, soil remediation, mineral sequestration of carbon, and
secondary aggregates and recycled construction materials.

365
AUTHOR PROFILE: COLIN D. HILLS

CURRENT POSITION: Professor, Director, Centre for Contaminated


Land Remediation, Faculty of Engineering and
Science, University of Greenwich, Chatham
Maritime, Kent ME4 4TB, UK
E MAIL: [email protected]

EDUCATION:

Queen Mary, University of London BSc 1980 Geology


Queen Mary, University of London MSc 1981 Industrial Petrology
Imperial College, London PhD 1993 Environmental Engineering
Imperial College, London DIC 1993 Environmental Engineering

RELEVANT EXPERIENCE
Professor Colin Hills has over 30 years
experience in geo-materials and cement-
based systems for the treatment of soil and
waste, including periods of working in
Europe, the Middle East and Africa. For the
past 25 years he has been extensively
concerned with S/S, has published over 100
journal papers, and has authored guidance
on S/S for the Environment Agency
(England and Wales). His work has
attracted international recognition, has won
national and regional awards and has led to
innovative treatments for the management
of difficult wastes.

PROFESSIONAL MEMBERSHIPS AND AFFILIATIONS:

Chartered Scientist (CSci)


Fellow of the Geological Society (FGS)
Member of the Institute of Materials, Minerals and Mining (MIMMM)
Member of the ISCOWA
366
Member of the European Science Foundation Review Panel
Academic Lead, CO2-Chem KTN (Utilisation, Cluster)
Adjunct Professor, Civil and Resource Engineering, Dalhousie University,
Canada

SELECTED AWARDS (ACADEMIC and COMMERCIAL) and HONOURS


Kent Innovative Climate Change Technology Award (2006)
Kent Environment Business of the Year (2006)
IChemE Green Chemical Technology Award (Chemistry Innovation) (2007)
National Winner (and winner of South-East final), Shell Springboard (2008)
Times Higher Award for Outstanding Contribution to Innovation and
Technology (2008)
UK National Recycling Awards: Best Recycled Product (2013)
CIWM, (Murphy Innovative Practice SME Award) (2013)

367
AUTHOR PROFILE: DIANE V. INGRAHAM

CURRENT POSITION: Senior Project Manager, Stantec


Consulting Ltd

E MAIL: [email protected]

EDUCATION / REGISTRATION

Dalhousie University B.Sc. 1974 Physics


Massachusetts Institute of Technology S.M. 1976 Civil Engineering,
Hydrology
University of British Columbia Ph.D 1980 Civil Engineering,
. Hydrology

RELEVANT PROFESSIONAL EXPERIENCE

Dr. Ingraham is a senior project manager with


substantial (30 years) experience in management of
project quality, schedule, budget, and risk for large
projects, specifically those involving heavy civil
construction and environmental regulatory compliance
or remediation. Most recently:

Nalcor Energys Lower Churchill Muskrat Falls Hydroelectric Project


Regulatory Compliance Labrador as Stantecs Senior Project
Manager
Sydney Tar Ponds and Coke Ovens Cleanup Project Quality
Contracts Manager, including the Solidification/Stabilization
Remediation of the Tar Ponds; management of quality and
development of SOPs and training for consultants and contractors
Technical Training Consultant Exxon/LearnCorp International 4
New LNG Plant Projects

368
5 Invited lectures/symposia presentations on management and quality for
hazardous site remediation
7 Conference presentations on remediation quality management
6 Standard Operating Procedures for solidification/stabilization (mixing,
sampling, specimen preparation, testing, storing, and disposal) and 1
Qualified Pass Procedure (UCS, Permeability/Conductivity, and SPLP) for In-
Situ Solidification/Stabilization

Member of

The Project Management Institute


The Canadian Society of Civil Engineers
The American Society for Quality (Quality Management Division)

SELECTED AWARDS AND HONORS

Project Management Professional (PMP)


Killam Post-Doctoral Fellow (Civil Engineering), University of British
Columbia, 1980-1
Killam Whos Who

369
AUTHOR PROFILE: CRAIG B. LAKE

CURRENT POSITION: Associate Professor, Department of Civil


and Resource Engineering, Dalhousie
University, Halifax, Canada

E MAIL: [email protected]

EDUCATION / REGISTRATION

Dalhousie University BEng 1995 Civil Engineering


University of Western Ontario PhD 2000 Geotechnical Engineering

RELEVANT PROFESSIONAL
EXPERIENCE

35 years experience in engineering,


construction management and
operations in the specialty geotechnical
and environmental remediation
markets. Extensive project experience
across the United States and Canada.

Remediation experience includes off-


site removal activities, in-place
closures, sludge stabilization, dredging,
slurry walls and other types of
groundwater cut-off walls, in-place soil
mixing, soil vapor extraction, groundwater pump and treatment systems, and
in-situ solidification/stabilization. This remediation experience includes
working at various active and in-active facilities including complete steel
making facilities (including coking operations), wood treating sites,
manufactured gas plant (MGP) sites, Superfund sites, former Chemical
Warfare facilities, petroleum refineries, manufacturing facilities, landfills,
chemical plants, and numerous disposal sites.

370
ISS and soil mixing experience includes:

25 years of direct soil mixing project experience


20 years of direct ISS project experience
Numerous treatability studies for solidification/stabilization

Co-authored over two dozen technical papers and articles on specialty


geotechnical techniques and environmental remediation. Regular presenter at
various remediation and geotechnical conferences.

371
AUTHOR PROFILE: PAUL R. LEAR

CURRENT POSITION: Senior Technical Director, Envirocon

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Nebraska-Lincoln B.S. 1982 Soil Science


University of Illinois M.S. 1984 Soil Chemistry/Mineral
University of Illinois Ph.D. 1987 Soil Chemistry/Mineral

RELEVANT PROFESSIONAL
EXPERIENCE
Dr. Lear has over 25 years of
experience in hazardous waste
treatment, laboratory management,
and chemical process development.
His experience includes selecting and
evaluating treatment alternatives,
providing data for preliminary design
activities and project equipment
specifications, assisting project design
teams, and implementing the final
design.

Dr. Lear has hands-on experience with full-scale remediation activities and
specializes in process troubleshooting. He has provided technical operational
support to bioremediation, dewatering, soil washing, stabilization, thermal,
and wastewater treatment activities at toxic, hazardous, and radioactive waste
remedial sites. He has been involved with projects involving state of the art
treatment of hazardous waste, including:

First application of in-situ chemical oxidation using auger mixers,


First CAMU for waste treatment in the State of California and Region
IX,
First full-scale implementation of S/S treatment for dioxin,
First application of in-situ thermal treatment using auger mixers,
First full-scale project for the S/S of explosives and explosive-
contaminated soil,
First S/S of organic wastes in Australia,
First ISS project at a manufactured gas plant site in New Jersey, and
First in-situ mixed hydraulic barrier wall.

Dr. Lear had been involved with S/S remedy implementation at over 100
CERCLA and RCRA sites.

12 Invited Lectures/symposia presentations


21 publications on ISS technology for remediation of hazardous waste sites
372
AUTHOR PROFILE: JEROME MACNEIL

CURRENT POSITION: Contract Manager, Sydney Tar Ponds


Agency

E MAIL: [email protected]

EDUCATION / REGISTRATION

Carleton University B.Sc. 1988 Biology


University of Engineering. 1996 Environmental
Guelph

RELEVANT PROFESSIONAL EXPERIENCE

Mr. MacNeil is a Senior Project Manager and Environmental


Engineer with more than 17 years of experience. Over the
course of his career he has developed specific expertise in the
areas of contract management, environmental management
and contaminated site remediation. Mr. MacNeil has managed
in excess of 130 million dollars worth of environmental
remediation and heavy civil projects. He also has experience
in environmental site assessments, asbestos and mold
abatement, air quality assessments, solid waste management, OH&S plan
development and implementation, and development of environmental training
programs.

5 Invited lectures/symposia presentations on project and environmental


management
4 Conference presentations on remediation management

Member of

The Project Management Institute


Engineers Nova Scotia

373
AUTHOR PROFILE: AIMAN M. NAGUIB

CURRENT POSITION: Managing Partner/ Project Coordinator,


ENTACT, LLC

E MAIL: [email protected]

EDUCATION / REGISTRATION

United States International


University BSc. 1988 Civil Engineering
Washington University
in St. Louis MSc 1989 Construction Management

RELEVANT PROFESSIONAL
EXPERIENCE

Mr Naguib has more than 24 years of


experience in general remediation, and
specialty environmental and
geotechnical construction.

He has worked extensively on complex environmental remediation and


geotechnical projects involving in-situ S/S, slurry walls, permeable reactive
barriers and cap/containment construction.

As a Project Coordinator and Technical Advisor at ENTACT, he is responsible


for managing and coordinating S/S projects. This includes cost estimates,
proposal preparation, work planning, mix designs/treatability studies and the
management of field operations, including resource allocations.

Mr Naguib has been responsible for the execution of over 26 in-situ S/S
projects employing auger mixing, and 37 slurry walls (including design and
build) throughout the United States, and most significantly, has developed
374
extensive and specialized expertise in the area of in-situ S/S having
successfully treated millions of tons of impacted materials.

His in-situ S/S experience involves on-site remedies for S/S of metal
contaminants (lead, arsenic, chromium), and organics (petroleum
hydrocarbons, NAPL, etc.), using backhoes, horizontal rotary mixers, and
single/multi-vertical auger mixing techniques.

Mr Naguib's technical expertise is recognized throughout the industry and


includes treatability studies, mix design programs using proprietary additives
and common binders (e.g. Portland cement, bentonite, slag etc.) with
contaminated soil, sludge, and sediments.

PUBLICATIONS
On-Site Mixing of Dry and Slurred Reagents in Soil and Sludge using
Shallow Soil Mixing, Air & Waste Management Association's Annual
meeting, June 1992
Design and Construction of Deep Soil Mix Retaining Wall for the Lake
Parkway Freeway Extension, Geo-Congress 1998, Boston, MA, October
1998
Ground Improvement for Large Above-Ground Petroleum Storage Tanks
Using Deep Mixing, Geo-Denver 2000, Denver, CO, August 2000

375
AUTHOR PROFILE: STANY PENSAERT

CURRENT POSITION: Manager Department Research &


Development, Process Technology and
Remediation Design, DEME
Environmental Contractors

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Ghent (Belgium) M.S. 1990 Civil Metallurgical Engineer

University of Ghent (Belgium) M.S. 1994 Biomedical Engineer

RELEVANT PROFESSIONAL
EXPERIENCE

Stany has 20 years service as head of


the R&D Department at DEC-DEME
Environmental Contractors NV.

He is responsible for the scientific and


process support for all environmental
works at: site remediations, soil
treatment, environmental dredging and
sediment treatment.

His work is world-wide supporting the DEME group.

For about the last 10 years, he has been very active in innovative approaches
for the stabilization/solidification applied on various sites in Europe.
376
He is a Lecturer for University of Ghent on Environmental courses, has
delivered lectures to the Environment Agency in the UK, Belgium, Sweden on
various soil treatment techniques.

Has published over one hundred publications, case studies and technology
appraisals with respect to soil remediation and environmental dredging.

377
AUTHOR PROFILE: THOMAS R. PLANTE

CURRENT POSITION: Program Manager/Senior Remediation


Specialist at Haley & Aldrich, Inc.
E MAIL: [email protected]

EDUCATION / REGISTRATION

University of New Hampshire B.S 1987 Civil Engineering


University of Massachusetts at Amherst M.S 1990 Environ. Engineering

RELEVANT PROFESSIONAL EXPERIENCE

A engineering consultant with over 25 years of


experience in site characterization, remedy
evaluation and selection, remedial design and
remedial construction, primarily focused on
municipal, industrial, and utility clients
throughout the United States.

His primary environmental remediation focus


area is the investigation and remediation of
former manufactured gas plant (MGP) sites
containing coal tar, cyanide, ammonia, various metals, and other
contaminants.

Mr. Plante has conducted research for the Electric Power Research Institute
(EPRI) on a number of investigation and remediation topics related to former
MGP sites, including solidification/stabilization, vertical barriers, product
mobility, and technical impracticability demonstrations.

Mr. Plante was the principal investigator for the following EPRI research
reports on solidification: In-Situ Solidification/Stabilization (ISS) Bench-Scale

378
Testing of Manufactured Gas Plant (MGP) Impacted Soils, EPRI, Palo Alto,
CA, 2004. Technical Report 1010949, 2004.

An Integrated Approach to Evaluating In-Situ Solidification/Stabilization of


Coal Tar Impacted Soils, EPRI Technical Update 1018612, March 2009.

Mr. Plante has performed in-situ solidification at fly ash lagoons, fuel oil
release areas, and coal tar sites. Mr. Plante has led solidification treatability
studies for 11 sites and has been involved in the design, preparation of
technical specifications and quality control/quality assurance plans, and
implementation of in-situ solidification remedies at five sites, utilizing
excavator buckets, large diameter augers, or in-situ rotary blenders.

Mr. Plante has been a frequent presenter on solidification and other site
investigation and remediation topics at environmental symposia. He was a
contributor to the ITRC Guidance Document and ITRC trainer for web-based
training: Development of Performance Specifications for
Solidification/Stabilization S/S-1, http://www.itrcweb.org/gd.asp.

379
AUTHOR PROFILE: DANIEL G. RUFFING

CURRENT POSITION: Project Manager at Geo-Solutions, Inc.,


Pittsburgh
E MAIL: [email protected] (business)
[email protected] (personal)

EDUCATION / REGISTRATION

Bucknell University B.S 2008 Civil and Environmental Engineering

Bucknell University M.S 2010 Civil and Environmental Engineering

RELEVANT PROFESSIONAL EXPERIENCE

As a Project Manager at Geo-Solutions,


Daniel is responsible for generating estimates,
preparing proposals, managing bench-scale
studies, and the management of projects,
involving slurry trench cut-off walls, soil
mixing, jet grouting, and other specialty
geotechnical contracting techniques. Daniels
experience is founded in research conducted
during his university education supplemented by field engineering experience,
project supervision, and project management at Geo-Solutions.

Daniels field engineering and management experience on S/S projects spans


5+ years. S/S specific experience includes:
Manager of 7 field applications of S/S using in-situ soil mixing and/or jet
grouting
On-site project engineer for 4 field applications of S/S using in-situ soil
mixing
Manager of 16 Treatability (Bench-Scale) Studies for S/S

380
Daniel has delivered 7 Invited Lectures/symposia presentations, and has 10
publications on the theory and application of slurry trench cut-off walls and soil
mixing.

PROFESSIONAL AFFILIATIONS
Member, GeoInstitute of ASCE (2008 - Current)
Asst. Secretary/Treasurer of Pitts. GeoInstitute Board of Directors (2013 -
Current)
Associate Member, ASCE, Pittsburgh Chapter (2009 - Current)
Voting Member, American Society of Testing and Materials (ASTM),
Committee D34, (2010 - Current)
Student Member, International Association of Foundation Drilling (ADSC)
(2008 - 2009)
The 2009 ADSC Industry Advancement Scholarship
Member and President, Chi Epsilon Civil Engineering Honor Society,
Bucknell University Chapter (2007 - 2009)
Member, Tau Beta Pi Engineering Honor Society, Bucknell University
Chapter (2008 - 2009)
Student Member, American Society of Civil Engineers (ASCE), Central PA
Chapter (2005 - 2009)

381
AUTHOR PROFILE: BOB SCHINDLER

CURRENT POSITION: President & CEO of Geo-Solutions Inc.


E MAIL: [email protected]

EDUCATION / REGISTRATION

Gannon University B.S. 1987 Engineering

RELEVANT PROFESSIONAL
EXPERIENCE

Mr. Schindler is President/CEO of Geo-


Solutions Inc., an established
international, environmental
remediation and geotechnical
contracting company specializing in
slurry walls, bio-polymer drains,
reactive barriers, in-situ soil
solidification/stabilization, soil mixing, jet grouting and other related
techniques.

As President of Geo-Solutions, he oversees all company functions,


operations, finance, administration, health and safety, quality assurance,
marketing, and business development.

He has managed numerous S/S projects and treatability studies. Quantities of


in-situ S/S treated materials exceed one million cubic yards, using a variety of
application methods and reagents.

He has actively participated in many cutting-edge projects, for example he


managed the first permeable reactive barrier project installed using the
Biopolymer trenching technique at the Oak Ridge National Laboratory in Oak
Ridge, Tennessee. He also managed the first permeable reactive barrier
project installed using the Deep Soil Mixing method for NASA in Cape
Canaveral, Florida.

382
In December of 2012, he led the efforts to acquire Geo-Con and subsequently
managed the merger of the two companies in 2013. From an operations
standpoint, the merger was completed in the Spring of 2013.

An addition to Geo-Solutions facilities in New Kensington was completed in


November of 2013 and all facets of the companies were completely merged
by December of 2013.

Mr. Schindler has co-authored over 10 publications relating to specialty geo-


environmental construction.

383
AUTHOR PROFILE: ROY E. WITTENBERG, PE

CURRENT POSITION: Vice President / Principal Engineer

E MAIL: [email protected]

EDUCATION / REGISTRATION

University of Illinois B.S. 1987 Sociology


University of Colorado B.S. 1985 Civil Engineering
University of Colorado M.S. 1990 Civil Engineering

RELEVANT PROFESSIONAL EXPERIENCE

Mr. Wittenberg has over 27 years of experience


performing project engineering and management,
technical supervision, design engineering and
analysis, construction oversight and regulatory
interface.

His project management and construction experience


includes a number of site remedial restoration
projects along major waterways.

His environmental experience includes conducting


remedial alternatives evaluations, and development of risk-based clean-up
approaches.

His technical experience includes environmental engineering for soil/sediment and


groundwater treatment, bench- and pilot-scale testing and civil engineering
applications for site restoration and redevelopment.

A significant portion of his professional project experience is associated with former


manufactured gas plants using in-situ solidification/stabilization as a primary
remediation technology.

384
He has authored or co-authored numerous publications and presentations on the
subjects of solidification/stabilization, innovative remedial approaches for complex
sites and restoration for former industrial properties.

Areas of expertise include:

Environmental and civil engineering for site restoration and redevelopment


Construction management for civil and environmental related projects
Regulatory negotiation and permitting
Feasibility studies
Geotechnical testing and design
Bench and pilot scale treatability studies
Solidification/stabilization design and implementation

PROFESSIONAL REGISTRATIONS
Professional Engineer in: Colorado, Florida, Illinois, Michigan and Wisconsin.

385
Appendix A: Remediated sites employing S/S

The following list of sites employing S/S in the USA and Europe was compiled
from data supplied by S/S construction contractors and engineers.

The data is presented as it was received from these sources and has not
been independently verified by the editors. However each project listed in the
following Table identifies the person that submitted data along with the name
and E-mail contact (or in a few cases the relevant publication), should
additional information be required.

This extensive list of over 200 completed S/S projects illustrates the
widespread successful application of S/S to a wide variety of contaminants
and site types.

386
Sites Outside The United States
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

ENTACT, LLC Sydney Tar Soil; Metals, PAHs, In-situ Remediation; Cement ENTACT, LLC Sydney Tar
2013 Ponds 570,000 VOCs, PCBs, Hydraulic treatment Ponds Agency
Solidification/Sta cubic TPH Excavators www.tarpondscle
bilization meters of anup.ca
Sydney, Nova impacted
Scotia CANADA sediment
DEC nv Hoedhaar 10,000 Hg Excavator Comply to non- Fe + iron Stany Pensaert Stany Pensaert
Belgium Lokeren Belgium tons of soil haz landfill hydroxides Pensaert.Stany
2012 @deme.be
DEC nv Obourg 170,000 none Lime blending Reuse as Paper ash + CKD Stany Pensaert Stany Pensaert
Belgium tons machine engineered 11 MPa plate test Pensaert.Stany
2011 sediment backfill @deme.be
DEC nv Sderhamn 31,000 As/Cu/Cr Mobile pug-mill Comply to non- Fe @ 2% w/w Stany Pensaert Stany Pensaert
Belgium Sweden tons soil mixer haz landfill Pensaert.Stany
2010 @deme.be
DEC nv London Olympics 60,000 Heavy metals, Mobile pug-mill Reuse as Fe @ 2% w/w Stany Pensaert Stany Pensaert
Belgium tons soil PAH, TPH mixer + allu engineered Biochar @ 1% w/w Pensaert.Stany
2007-2010 bucket on backfill @deme.be
excavator
DEC nv Total 250,000 Various Fixed plant Comply to haz Various fly ashes + Stany Pensaert Stany Pensaert
Belgium Ertvelde Belgium tons acid hydrocarbons based on pug- landfill cement Pensaert.Stany
2005-2008 tar + soil mill mixer CBR > 11% @deme.be
K < 10-9 m/s
DEC nv Bekaert Belgium 50,000 Cu/Pb/Cd/Zn Mobile pug-mill Reuse as backfill Paper ash @ 2% Stany Pensaert Stany Pensaert
Belgium tons soil mixer w/w Pensaert.Stany
2007 @deme.be

387
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

DEC nv La Floridienne 60,000 As Mobile pug-mill Comply to haz Fe @ 2% w/w Stany Pensaert Stany Pensaert
Belgium Belgium tons waste mixers landfill Pensaert.Stany
2004 @deme.be

DEC nv La Floridienne 75,000 Cyanides Mobile pug-mill Comply to haz OPC + GGBS Stany Pensaert Stany Pensaert
Belgium Belgium tons waste mixers landfill Pensaert.Stany
2004 @deme.be
DEC nv Guernsey UK 31,000 TBT Allu rotary Reuse for port Biochar @ 2% w/w Stany Pensaert Stany Pensaert
Belgium tons bucket extension Pensaert.Stany
2003 sediment @deme.be

Sites In The United States


WRScompass Wilmington Coal 45,000 cy PAHs, coal tar, Excavator Redevelopment 1.5-3% Portland + Paul Lear Paul.Lear
2010, Gasification Site, soil DNAPL 4.5-9% slag plear@envirocon.
2011,2013 Wilminton, DE >50 psi UCS com
-6
<1x10 cm/s
permeability
Geo-Solutions Wood Preservers 58,454 Wood Treating In-situ Stabilization/ Portland Cement K. Andromalos K. Andromalos
2012 Remediation cycy Chemicals Auger Solidification 1.00x10-6 kandromalos@ge
Warsaw, VA Bucket 50 psi o-solutions.com
Geo-Solutions West End 88,000 MGP impacted In-situ Stabilization/ Cement K. Andromalos K. Andromalos
2012 Remediation cycy soils Auger Solidification Bentonite kandromalos@ge
Cincinnati, OH o-solutions.com
Geo-Solutions PG&E Front and 42,500 BTEX In-situ Stabilization Portland cement K. Andromalos K. Andromalos
2012 T Street Site cycy PAHs Auger Regenerated kandromalos@ge
Sacramento, CA Granular Activated o-solutions.com
Carbon

388
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula by: Or Contact
of treatment
Geo-Con Atlantic Wood 47,000 DNAPLs In-situ Stabilization/ Portland Cement K. Andromalos M. Kitko
(trade name of Industries cycy Excavator/ Solidification Slag Cement mkitko@geo-
Geo-Solutions) Superfund Site rotary blender Organophilic Clay solutions.com
2012 Portsmouth, VA system 4.0x10-6cm/sec50psi
-6
Geo-Con MGP Site 54,300 PAHs, BTEX In-situ remediation 1x10 cm/sec K. Andromalos D. Payne
(trade name of Remediation cycy Large diameter UCS 50 psi dpayne@geo-
Geo-Solutions) Homer, New auger solutions.com
2012 York
Geo-Con Former 42,000 Creosote, In-situ Superfund Cement K. Andromalos G. Maitland
(trade name of Texarkana Wood cycy PCPs Large diameter remediation Powdered activated gmaitland@geo-
Geo-Solutions) Preserving auger carbon solutions.com
2012 Company Site
Texarkana,
Texas
Geo-Solutions Indian Head 1,300 cycy Chlorinated In-situ Treatment/ ZVI K. Andromalos K. Andromalos
2012 Naval Base Solvents Auger Chemical kandromalos@ge
Indian Head, MD Reduction o-solutions.com
Geo-Solutions Former Hanley 1,400 cycy Impacted Soils In-situ Stabilization ZVI K. Andromalos K. Andromalos
2012 Area Ordnance Auger kandromalos@ge
Site o-solutions.com
St. Louis, MO
Geo-Solutions Former 4,500 cycy Creosote In-situ Stabilization Powdered K. Andromalos K. Andromalos
2012 Columbus Wood Auger Activated Carbon kandromalos@ge
Treaters o-solutions.com
Columbus, IN
Geo-Solutions Former 7,500 CM Chlorinated In-situ Chemical ZVI K. Andromalos K. Andromalos
2012 Municipal Solvents Auger Reduction kandromalos@ge
Wastewater o-solutions.com
Treatment
Lagoon

389
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula by: Or Contact
of treatment
WRScompass, Valero NRP 18,000 cy, TPH, Metals In-situ Lagoon Closure, 2% Portland + 6% Paul Lear Paul.Lear
2012 Closure Project, sludge/sedi Excavator Redevelopment slag + 2% Ferrous plear@envirocon.
Paulsboro, NJ ment sulfate, >50 psi UCS com
-6
<1x10 cm/s perm-
eability. Leachability
SPLP < NJ Class II
GWQS
WRScompass PSEG, Camden, 19,000 cy PAHs, coal tar, Excavator Redevelopment 3% Portland + 9% Paul Lear Paul.Lear
2012 NJ soil DNAPL slag, >50 psi UCS plear@envirocon.
-6
<1x10 cm/s com
permeability
RECON Pond 9 48,000 cy Acid tar, with In-situ bucket Cap Patented LSS Remedial S. Birdwell
2012 Marcus Hook, PA tarry H2S and SO2, mixing Construction steven.birdwell@r
sludge benzene, Services, L.P. econservices.com
asbestos
RECON Bayou 50,000 cy Hydrocarbons In-situ bucket Strengthen for Patented LSS Remedial S. Birdwell
2012 Trepagnier sediments mixing cap construction Construction steven.birdwell@r
Remediation, St. Services, L.P. econservices.com
Charles Parish,
LA
RECON Refinery Pond 40,000 cy Hydrocarbons In-situ bucket Impoundment Patented LSS Remedial S. Birdwell
2012 Closure, sludge mixing closures Construction steven.birdwell@r
Norco, LA Services, L.P. econservices.com
RECON Pond 9 48,000 cy Acid tar, H2S & In-situ bucket Cap Patented LSS Remedial S. Birdwell
2012 Marcus Hook, PA tarry SO2, benzene, mixing Construction steven.birdwell@r
sludge asbestos Services, L.P. econservices.com
RECON Bayou Trepagni- 50,000 cy Hydrocarbons In-situ bucket Strengthen for Patented LSS Remedial S. Birdwell
2012 er Remediation, sediments mixing cap construction Construction steven.birdwell@r
St. Charles Services, L.P. econservices.com
Parish, LA

390
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

RECON Refinery Pond 40,000 cy Hydrocarbons In-situ bucket Impoundment Patented LSS Remedial S. Birdwell
2012 Closure, sludge mixing closures Construction steven.birdwell@r
Norco, LA Services, L.P. econservices.com
Envirocon, Manufactured 35,000 Coal Tar In- Site remediation, Cement 2% T. Plante T.Plante
2011 Gas Plant Site, cycy (BTEX, PAHs) situ/excavator risk reduction GBFS 6% tplante@haleyaldr
Ohio bucket UCS 50 psi ich.com
-6
1X10 cm/sec 207-482-4622

ENTACT, LLC WE Energies 33,361 cy Benzo(a)pyrene In-situ 10 ft Remediation Cement-Slag grout ENTACT, LLC Wisconsin
- 2011 Gaslight Point of MGP (PAHs); Coal; diameter Department of
MGP Site impacted Tar; BTEX mixing auger Natural
Remediation soils mounted on Resources
Racine, WI drill rig.
WRScompass, NYSEG Elmira 10,000 cy PAHs, coal tar, In-situ auger Containment
4% Portland + 0.5% Paul Lear Paul.Lear
2011 MGP Site, soil DNAPL bentonite plear@envirocon.
Elmira, NY >50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass Sanford 125,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2011 Gasification soil DNAPL slag plear@envirocon.
Facility, Sanford, >50 psi UCS com
FL <1x10-6 cm/s
permeability
Geo-Con F-Area Seepage 1,400 lf Tritium In-situ Groundwater Impermix K. Andromalos G. Maitland
-6
(trade name of Basin Multiple Auger barrier 1x10 cm/sec gmaitland@geo-
Geo-Solutions) Aiken, South UCS 50 psi solutions.com
2011 Carolina

391
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Geo-Con (a Rosemont Levee 4,900 cycy N/A In-situ Reinforcement 50 psi K. Andromalos G. Maitland
trade name of Improvements Auger Cement gmaitland@geo-
Geo-Solutions) Lowell, MA solutions.com
2011
Geo-Con (a Perimeter Wall 560,000 Coal Fly Ash In-situ Containment 250 psi K. Andromalos S. Artman
trade name of Stabilization cycy Excavator Cement sartman@geo-
Geo-Solutions) Kingston, TN Bentonite solutions.com
2011
Geo-Solutions MW-250 Site 7,626 cycy TCE In-situ Oxidation Potassium K. Andromalos K. Andromalos
2011 Remediation Auger Permanganate/ kandromalos@ge
East Rutherford, Cement o-solutions.com
NJ
Geo-Solutions OMC Plant Site 8,900 cycy TCE In-situ Chemical ZVI K. Andromalos K. Andromalos
2011 Remediation Auger Reduction/ Bentonite kandromalos@ge
Waukegan, IL Treatment o-solutions.com
Geo-Solutions Inner Slip Site 6,457 cycy MGP impacted In-situ Solidify K. Andromalos K. Andromalos
2011 Remediation sediment Auger kandromalos@ge
New Bedford, o-solutions.com
MA
Geo-Solutions Former Miller 2,778 cycy Pesticide In-situ Chemical Sodium Persulfate/ K. Andromalos K. Andromalos
2011 Chemical Site Xylene Auger Oxidation/ Lime kandromalos@ge
Robbinsville, NJ Stabilization o-solutions.com
RECON Cell 11 250,000 cy Various In-situ bucket Stabilization for Cement/ash blend Remedial S. Birdwell
2010 Port Arthur, TX soil and hydrocarbons mixing construction Construction steven.birdwell@r
sludge foundation Services, L.P. econservices.com
RECON Fly Ash Ponds 115,000 cy Petroleum coke In-situ bucket Impoundment Lime kiln dust, Remedial S. Birdwell
2010 Delaware City, fly ash and vanadium mixing closure Portland cement and Construction steven.birdwell@r
DE soil Services, L.P. econservices.com

392
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

RECON Soil Mixing Demo 1,500 cy fly Fly ash In-situ soil Construct barrier 289 pounds of Remedial S. Birdwell
2010 Project, ash mixing Portland cement per Construction steven.birdwell@r
Kingston, TN cubic yard Services, L.P. econservices.com
Geo-Solutions SAR Levee 6,100 cycy In-situ Stabilization K. Andromalos K. Andromalos
2010 Repair Auger kandromalos@ge
Newport Beach, o-solutions.com
California
Geo-Solutions MW-520 Site 7,626 cycy TCE In-situ Oxidation/Solidifi Potassium K. Andromalos K. Andromalos
2010 Remediation Auger cation Permanganate/ kandromalos@ge
East Rutherford, Portland Cement o-solutions.com
NJ
Geo-Con Former Mfg 15,000 Chlorinated In-situ Oxidation Potassium K. Andromalos G. Maitland
(trade name of Facility cycy solvent Excavator Solidification Permanganate/ gmaitland@geo-
Geo-Solutions) Parsippany, NJ mounted Portland Cement solutions.com
2010 blender head
Geo-Con Shermerhorn 10,800 PCBs In-situ Stabilization Portland Cement K. Andromalos M. Kitko
(trade name of Creek cycy Chlorinated Backhoe mkitko@geo-
Geo-Solutions) Schenectady, NY Solvents solutions.com
2010
WRScompass, NYSEG Norwich 25,000 cy PAHs, coal tar, In-situ auger Containment 8% Portland + Paul Lear Paul.Lear
2010 MGP site, soil DNAPL 0.75% bentonite plear@envirocon.
Norwich, NY >50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass, Newell- 32,000 cy Lead, arsenic, Ex-situ pug-mill Remediation 1.5% - 4.5% Paul Lear Paul.Lear
2010 Rubbermaid soil and cadmium EnviroMag plear@envirocon.
Palmieri Site, glass SPLP metals < UTS com
Monaca, PA

393
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

WRScompass, United Metals 36,000 cy Lead Ex-situ Pug-mill Remediation 8% Portland cement Paul Lear Paul.Lear
2010 Inc, Marianna, FL soil + 8% TerraBond plear@envirocon.
SPLP lead < 0.015 com
mg/L
>50 psi UCS
-6
<1x10 cm/s
permeability
WRScompass ConEd, White 30,000 cy PAHs, coal tar, In-situ auger Redevelopment 2.5% Portland + Paul Lear Paul.Lear
2009 Plains NY soil DNAPL 7.5% slag plear@envirocon.
>50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass, Orkin Atlanta 5,000 cy Chlordane In-situ Closure 5% Portland cement Paul Lear Paul.Lear
2009 Site, Atlanta, GA soil excavator + 0.05% carbon plear@envirocon.
TCLP Chlordane com
<0.03 mg/L
WRScompass, West Doane 18,000 cy Pesticides, In-situ Impoundment 18% Portland + 3% Paul Lear Paul.Lear
2009 Lake IRAM Sludge/sed metals Excavator Closure Bentonite + 3% plear@envirocon.
Project iment organoclay + 3% com
carbon
>50 psi UCS
<1x10-6 cm/s
permeability
WRScompass, ArvinMeritor 17,500 cy TPH, metals In-situ Impoundment 12% cement kiln Paul Lear Paul.Lear
2009 Sludge Lagoon sludge Excavator Closure dust plear@envirocon.
closure Project, >50 psi UCS com
Granada, MS

394
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Geo-Con Vandenberg Air 22,000 Chlorinated In-situ Permeable BOS-100 K. Andromalos G. Maitland
(trade name of Force Base, vertical Solvents Auger Reactive Barrier gmaitland@geo-
Geo-Solutions) California wall square solutions.com
2009 feet
Geo-Con MGP Site 16,290 BTEX In-situ Stabilization Slag Cement K. Andromalos M. Kitko
(trade name of Macon, GA cycy PAHs Large Diameter Portland Cement mkitko@geo-
Geo-Solutions) Auger Bentonite solutions.com
2009
RECON Site Preparation 425,000 cy Trace metals In-situ bucket Stabilization for Patented LSS/25 psi Remedial S. Birdwell
2009 Port Arthur, TX dredge mixing construction UCS Construction steven.birdwell@r
spoil and foundation Services, L.P. econservices.com
marsh
sediment
RECON Site Preparation, 650,000 cy Trace metals In-situ Stabilization for Patented blend of fly Remedial S. Birdwell
2009 Sabine Pass, TX dredge excavator construction ash Construction
spoils foundation Services, L.P. steven.birdwell@r
econservices.com
RECON Cell 9 172,000 cy Various In-situ bucket Impoundment Cement/ash blend Remedial S. Birdwell
2009 Port Arthur, TX sludge hydrocarbons mixing closures Construction steven.birdwell@r
Services, L.P. econservices.com
RECON Soil Stabilization, 21,000 cy Tetrachloro- In-situ soil Site closure Granular ground Remedial S. Birdwell
2009 Tuscumbia, AL soil ethene mixing blast furnace slag Construction steven.birdwell@r
and Portland cement Services, L.P. econservices.com
RECON Bayou 675,000 cy Hydrocarbons In-situ bucket Cap Cement and fly ash Remedial S. Birdwell
2009 Stabilization, Port sediments mixing blend/25 psi UCS Construction steven.birdwell@r
Arthur, TX Services, L.P. econservices.com

395
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

GCI Brunswick Wood 100,000 T Wood Ex-situ Produce Soil Cement 10% R. Garrett Robert Garrett
2008 Preserving Site Preserving ARAN Pug-mill Cement Fly ash 10% bobgarrett417@b
Brunswick, GA Waste ellsouth.net

WRScompass, Central Hudson 5,000 cy PAHs, coal tar, In-situ Containment 3% Portland + 9% Paul Lear Paul.Lear
2008 Gas & Electric soil DNAPL Excavator slag plear@envirocon.
MGP, >50 psi UCS com
Poughkeepsie, <1x10-6 cm/s
NY permeability
Geo-Con Camp Lejeune, 30,000 TCE/PCE In-situ Remediation ZVI K. Andromalos G. Maitland
(trade name of North Carolina cycy Large diameter gmaitland@geo-
Geo-Solutions) auger solutions.com
2008
Geo-Solutions MGP Site 13,000 BTEX, In-situ Stabilization Cement/Slag/ K. Andromalos K. Andromalos
2008 Remediation cycy napthalene Large diameter Bentonite kandromalos@ge
Rochester, New auger UCS 50 psi o-solutions.com
York 1.0E-6 cm/sec
Geo-Con Linskey Way 1,200 cycy MGP impacted In-situ Stabilization Cement/Bentonite K. Andromalos M. Kitko
(trade name of Cambridge, MA soils Jet Grouting mkitko@geo-
Geo-Solutions) solutions.com
2008
Geo-Con (a Hunter Ferrell 8,428 VSF N/A In-situ Seepage Barrier 1.00x10-6 K. Andromalos M. Kitko
trade name of Landfill Auger Bentonite mkitko@geo-
Geo-Solutions) Irving, TX solutions.com
2008
Geo-Con (a ISS Perimeter 7,200 cycy BTEX In-situ Barrier Wall 1.00x10-6 K. Andromalos M. Kitko
trade name of Wall PAHs Auger Excavation 50 psi mkitko@geo-
Geo-Solutions) Sag Harbor, NY Support Cement solutions.com
2008

396
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Geo-Solutions Former Rocky 1,917 lf CWM impacted In-situ Barrier Wall Bentonite K. Andromalos K. Andromalos
2008 Mountain Arsenal soils Multiple auger kandromalos@ge
Closure o-solutions.com
Commerce City,
CO
WRScompass Manufactured 22,000 Coal Tar In- Site remediation, Interior- 7.4% ; T. Plante T.Plante
2008 Gas Plant Site, cycy (BTEX, PAHs) situ/excavator risk reduction Exterior - 20%. Paul Lear tplante@haleyaldr
Waterville, ME bucket Reagent consisted ich.com
of 1 part Type I/II 207-482-4622
Portland cement to 3
parts slag. UCS 30 Paul.Lear
psi, 1X10-6 cm/sec plear@envirocon.
Interior and com
1X10-7 cm/sec
edge cells
Geo-Solutions Utica, New York 15,000 PCBs In-situ Stabilization Cement K. Andromalos K. Andromalos
2007 cycy Large diameter kandromalos@ge
auger o-solutions.com
Geo-Con (a CKD 31,400 Cement Kiln In-situ Containment 1.00x10-6 K. Andromalos S. Artman
trade name of Groundwater VSF Dust Excavator 50 psi sartman@geo-
Geo-Solutions) Remediation solutions.com
2007 Metaline Falls,
WA
Geo-Con (a Lowes 7,134 cycy Saturated In-situ Soil 100 psi K. Andromalos M. Kitko
trade name of Heidelberg Unconsolidated Excavator Improvement Cement mkitko@geo-
Geo-Solutions) Carnegie, PA Soils solutions.com
2007

397
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
-6
Geo-Con (a Stabilized Soil 45,750 Coal tar In-situ Barrier Wall 1.00x10 K. Andromalos M. Kitko
trade name of Barrier VSF VOCs Auger 40 psi mkitko@geo-
Geo-Solutions) Plattsburgh, NY Cement solutions.com
2007 Bentonite
WRScompass Niagara Mohawk, 45,000 cy PAHs, coal tar, In-situ auger Containment 8% Portland + 1% Paul Lear Paul.Lear
2007 Sarasota soil DNAPL bentonite plear@envirocon.
Springs, NY >50 psi UCS com
-6
<1x10 cm/s
permeability
WRScompass, Cambridge 20,000 cy PAHs, coal tar, In-situ auger Containment 2% Portland + 6% Paul Lear Paul.Lear
2007 Creek MGP, soil DNAPL slag plear@envirocon.
Cambridge, MD >50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass, Foote Minerals 220,000 cy Lithium In-situ Auger Redevelopment 7-10% slag Paul Lear Paul.Lear
2007 site, Exton, PA >50 psi UCS plear@envirocon.
-6
<1x10 cm/s com
permeability
WRScompass, Motiva Crude 120,000 TPH In-situ Refinery Portland cement or Paul Lear Paul.Lear
2007 Expansion cy, sludge Excavator Expansion cement kiln dust plear@envirocon.
Project, Port >10 psi UCS com
Arthur, TX
RECON Site 15,000 cy Chlorinated In-situ soil Risk reduction 10% cement, UCS of Remedial S. Birdwell
2007 Remediation, soil solvents, mixing 50 psi and 1x10-7 Construction steven.birdwell@r
Tarrant City, AL hydrocarbons cm/sec perm Services, L.P. econservices.com
& trace metals
RECON Section 7 1,550,000 Various In-situ bucket Impoundment Fly ash/20 psi Remedial S. Birdwell
2007 Port Arthur, TX cy sludge hydrocarbons mixing closures Construction steven.birdwell@r
Services, L.P. econservices.com

398
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

RECON Site Preparation 825,000 cy Trace metals Ex-situ bucket Stabilization for Patented LSS/25 psi Remedial S. Birdwell
2007 Cameron Parish, dredge mixing construction UCS Construction steven.birdwell@r
LA spoil foundation Services, L.P. econservices.com
RECON Site Preparation 800,000 cy Trace metals In-situ Stabilization for Patented blend of fly Remedial S. Birdwell
2007 Sabine Pass, TX dredge excavator construction ash Construction steven.birdwell@r
spoils foundation Services, L.P. econservices.com

RECON Purity Superfund 45,000 cy Low pH, In-situ with Superfund Cement/25 psi UCS Remedial S. Birdwell
2007 Site, soil hydrocarbons rotating mixing remediation and leachability Construction steven.birdwell@r
Fresno, CA head Services, L.P. econservices.com
RECON Oxidation Basin 35,000 cy Hydrocarbons In-situ Strengthening Fly ash Remedial S. Birdwell
2006 Baytown, TX soil and excavator for cap Construction steven.birdwell@r
sludge construction Services, L.P. econservices.com
RECON Cell 4 52,000 cy Various In-situ bucket Stabilization for Cement/ash blend Remedial S. Birdwell
2006 Port Arthur, TX sludge hydrocarbons mixing construction Construction steven.birdwell@r
foundation Services, L.P. econservices.com

Geo-Con Peerless Photo 10,500 Silver In-situ Stabilization Cement/Bentonite K. Andromalos M. Kitko
(Geo- Shoreham, NY cycy Cadmium Jet Grouting mkitko@geo-
Solutions) solutions.com
2006
Geo-Con MGP Site 11,400 BTEX In-situ Solidification Cement/Bentonite K. Andromalos M. Kitko
(trade name of Nyack, NY cycy PAHs Large diameter 1.0E-5 cm/sec mkitko@geo-
Geo-Solutions) auger 50 psi solutions.com
2006
Geo-Con (a Greys Landfill 11,702 N/A In-situ Ground 85 psi K. Andromalos M. Kitko
trade name of Sparrows Point, VSF Auger Improvement Cement mkitko@geo-
Geo-Solutions) MD solutions.com
2006

399
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
Geo-Con (a Walmart 8,400 VSF N/A In-situ Sub-grade 194 psi K. Andromalos G. Maitland
trade name of Supercenter Auger Improvement Cement gmaitland@geo-
Geo-Solutions) Scarborough, ME solutions.com
2006
WRScompass, West side and 32,000 cy PAHs, coal tar, In-situ auger Redevelopment 2%Portland+5% slag Paul Lear Paul.Lear
2006 Racine MGP soil DNAPL +Rheomag, >50 psi plear@envirocon.
-6
sites, Racine, WI UCS, <1x10 cm/s com
permeability
Shaw, 2006 Goodfellow Air 11,000 cy Lead, PAHs En situ Remediation 5% Portland cement Paul Lear Paul.Lear
force Base Small excavator TCLP lead <5 mg/L plear@envirocon.
Arms Range, com
San Angelo, TX
WRScompass, Kane County 19,500 cy, Lead, copper In-situ Impoundment 5-10% lime kiln dust Paul Lear Paul.Lear
2006 Remediation & sludge Excavator closure TCLP lead <5 mg/L plear@envirocon.
Storm Water com
Improvement, St
Charles, IL
WRScompass, Larimer County 3,000 cy Lead Ex-situ Pug-mill Closure 3% Triple Paul Lear Paul.Lear
2006 Landfill Firing soil Superphosphate plear@envirocon.
Range TCLP Lead <5 mg/L com
remediation, Fort
Collins, CO
WRScompass, Walter J Heinrich 1,000 cy Lead Ex-situ Closure 3% Triple Paul Lear Paul.Lear
2006 Training Facility, soil Excavator Superphosphate plear@envirocon.
Tampa, FL TCLP lead <5 mg/L com
WRScompass, Normandy Park, 24,000 cy Lead Ex-situ Remediation 3%Triple Paul Lear Paul.Lear
2006 River Hills & soil and excavator Superphosphate plear@envirocon.
Residential sediment TCLP lead < 5 mg/L com
Cleanup, Tampa,
FL

400
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Shaw, 2006 FMC Atvex 7,000 cy Lead, arsenic, Ex-situ Closure 3% Triple Paul Lear Paul.Lear
Fibers Fine soils, antimony excavator Superphosphate + plear@envirocon.
Fractions sludges 0.5% ferrous sulfate com
Stabilization, SPLP metals < UTS
Front Royal, VA
Shaw, 2006 City of Arlington 8,000 cy Lead, PAHs Ex-situ Closure 7.5% cement kiln Paul Lear Paul.Lear
Skeet Range, excavator dust plear@envirocon.
Arlington, WA TCLP lead <075 com
mg/L
TCLP PAHs <0.010
mg/L
Shaw, 2005 FUSRAP 75,000 cy Lead Ex-situ pug-mill Closure 1% phosphoric acid Paul Lear Paul.Lear
Colonie Site soil TCLP lead <5 mg/L plear@envirocon.
Remediation com
Project, Colonie,
NY
WRscompass, Augusta 15th 125,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2005 Street, Augusta, soil DNAPL slag, >50 psi UCS plear@envirocon.
GA <1x10-6 cm/s com
permeability
WRScompass, Americus MGP 28,000 cy PAHs, coal tar, In-situ auger Redevelopment 8% Portland + 1% Paul Lear Paul.Lear
2005 Site, Americus, soil DNAPL bentonite plear@envirocon.
GA >50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass, Vandalia Road 29,000 cy PAHs, coal tar, In-situ Containment 3% Portland + 9% Paul Lear Paul.Lear
2005 Facility, Pleasant soil DNAPL Excavator slag, >50 psi UCS plear@envirocon.
Hill, IA <1x10-6 cm/s com
permeability

401
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Shaw 2005 Fort McClellan 21,000 cy Lead Ex-situ Closure 5% Portland cement Paul Lear Paul.Lear
Iron Mountain soil excavator TCLP lead <5 mg/L plear@envirocon.
Ranges 12 and com
13, Anniston, AL
WRScompass, Former Scrap 52,000 cy Lead Ex-situ Redevelopment 5% Portland cement Paul Lear Paul.Lear
2005 Yard Lead soil Excavator TCLP lead <5 mg/L plear@envirocon.
Removal, Benton com
Harbor, MI
RECON Impoundment 20,000 cy RCRA metals Ex-situ bucket Impoundment Fly ash Remedial S. Birdwell
2005 Closure, sludge mixing Closure Construction steven.birdwell@r
Houston, TX Services, L.P. econservices.com
RECON Tex Tin 167,000 cy Acid and heavy In-situ and ex- Superfund Fly ash/20 psi Remedial S. Birdwell
2005 Superfund, sludge, metals situ remediation Construction steven.birdwell@r
Texas City, TX sediment, bucket mixing Services, L.P. econservices.com
soil
RECON Landfill 67,000 cy Various Ex-situ bucket Stabilize existing Strength and paint Remedial S.Birdwell
2005 Emergency sludge hydrocarbons, mixing berms around filter Construction steven.birdwell@r
Action drilling muds, lagoons Services, L.P. econservices.com
Southern, CA tank bottoms

402
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

RECON SWB Closure 125,000 cy hydrocarbons Excavator Impoundment Fly ash and Portland Remedial S. Birdwell
2005 Nederland, TX soil and bucket mixing closures cement Construction steven.birdwell@r
sediment Services, L.P. econservices.com
RECON Site Preparation 1,172,000 Trace metals In-situ Stabilization for Patented blend of fly Remedial S. Birdwell
2005 Cameron Parish, cy dredge excavator construction ash Construction steven.birdwell@r
LA spoils foundation Services, L.P. econservices.com
RECON Cell 3 80,000 cy Various In-situ bucket Stabilization for Cement/ash blend Remedial S. Birdwell
2005 Port Arthur, TX sludge hydrocarbons mixing construction Construction steven.birdwell@r
foundation Services, L.P. econservices.com

Geo-Con Refining Facility 20,000 Petroleum In-situ / Stabilization Bentonite K. Andromalos B. Buccille
(trade name of Lima, OH cycy Backhoe UCS 100 psi bbuccille@geo-
Geo-Solutions) solutions.com
2005
Geo-Con (a Funnel & Gate 9,921 VSF Impacted In-situ Funnel 1.00x10-7 K. Andromalos M. Kitko
trade name of Muskegon, MI Groundwater Excavator System Bentonite mkitko@geo-
Geo-Solutions) solutions.com
2005
RECON Basin 15 Closure 300,000 cy hydrocarbons In-situ bucket Impoundment Class C fly ash, Remedial S. Birdwell
2004 Roxana, IL sludge and rake closure cement and bed ash Construction steven.birdwell@r
mixing Services, L.P. econservices.com
RECON Laydown Facility, 85,000 cy Hydrocarbons In-situ Strengthening Fly ash and cement Remedial S. Birdwell
2004 Richmond, CA soil and excavator for cap Construction steven.birdwell@r
sludge construction Services, L.P. econservices.com
RECON Remedial Action, 5,400 cy Nitrobenzene In-situ jet Remedial action Potassium Remedial S. Birdwell
2004 Buffalo, NY soil grouting Permanganate 600 Construction steven.birdwell@r
points per column Services, L.P. econservices.com

403
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

IT Corporation, B.F. Shaw 5,000 cy lead In-situ Remediation 3% Triple Paul Lear Paul.Lear
2004 Connex Facility, excavator Superphosphate plear@envirocon.
Troutville, VA TCLP lead <5 mg/L com
WRScompass, Augusta MGP 15,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2004 Off-sites, soil DNAPL slag plear@envirocon.
Augusta, GA >50 psi UCS com
-6
<1x10 cm/s
permeability
IT Corporation, Sunflower Army 15,000 cy Lead, Ex-situ Closure 5% Portland cement Paul Lear Paul.Lear
2004 Ammunition soil and explosives, excavator TCLP lead < 5 mg/L, plear@envirocon.
Depot SWMU sludges propellants No reactivity com
22, De Soto, KS
IT Corporation, Former Volunteer 13,000 cy Lead Ex-situ Remediation 5% Portland cement Paul Lear Paul.Lear
2004 Army excavator TCLP lead <5mg/L plear@envirocon.
Ammunition com
Plant CFI Lease
Site,
Chattanooga, TN

IT Corporation, Texas American 13,000 cy Lead, TPH In-situ Closure 10% EnviroBlend Paul Lear Paul.Lear
2004 Oil Site, sludge excavator SPLP lead <0.015 plear@envirocon.
Midlothian, TX mg/L, >15 psi UCS com
<1x10-6 cm/s
permeability
IT Corporation, Former Volunteer 16,000 cy Lead Ex-situ Remediation 6% Portland cement Paul Lear Paul.Lear
2004 Army soil excavator TCLP lead <5 mg/L plear@envirocon.
Ammunition com
Plant scrap Yard,
Chattanooga, TN

404
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
IT Corporation, Fernald Silos 1 & 4,000 cy Lead, Ex-situ pug-mill Closure 8% Portland cement Paul Lear Paul.Lear
2004 2 Stabilization, sludge radionuclides Pass paint filter test plear@envirocon.
Fernald, OH com
WRS Lake Wire Lake Pb Ex-situ State Action EnviroBlend 80/20 D. Wheeler Dave Wheeler
Infrastructure & Lakeland, FL Sediment, Excavator Risk Reduction 3% dwhee87@yahoo.
Env. 9,000 cycy mixing ZapSorb 1% com
2003
WRScompass, Augusta MGP 25,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2003 North Parcel, soil DNAPL slag plear@envirocon.
Augusta, GA >50 psi UCS com
<1x10-6 cm/s
permeability
WRScompass, Appleton MGP 32,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2003 Fox River Canal soil DNAPL slag plear@envirocon.
Riverbank, >50 psi UCS com
Appleton, WI <1x10-6 cm/s
permeability
WRScompass, Sumter MGP 22,000 cy PAHs, coal tar, In-situ auger Redevelopment 3% Portland + 9% Paul Lear Paul.Lear
2003 Site, Sumter, SC soil DNAPL slag plear@envirocon.
>50 psi UCS com
<1x10-6 cm/s
permeability
IT Corporation, Andersen Air 13,000 cy Lead, antimony In-situ Remediation 2% - 4% Triple Paul Lear Paul.Lear
2003 Force Base, soil Tiller Superphosphate plear@envirocon.
Guam TCLP lead <5 mg/L com
IT Corporation, Yorktown Naval 6,000 cy Lead, cadmium In-situ Remedaition 5% Portland cement Paul Lear Paul.Lear
2003 Air Station Site 4 tiller TCLP lead <5 mg/L plear@envirocon.
Burn Pit com
Remediation,
Yorktown, VA

405
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

IT Corporation, Former Lee Field 7,500 cy Lead In-situ Closure 4% Portland cement Paul Lear Paul.Lear
2003 Naval Air Station soil excavator TCLP lead <5 mg/L plear@envirocon.
Small Arms com
range, Green
Cove, FL
RECON CAMU Construct 125,000 cy Various Bucket mixing Leachability Fly ash/20 psi Remedial S. Birdwell
2003 El Paso, TX soil hydrocarbons requirements Construction steven.birdwell@r
Services, L.P. econservices.com
RECON 250 Canal, 15,000 cy Hydrocarbons In-situ Strengthening Fly ash and cement Remedial S. Birdwell
2003 Richmond, CA sludge excavator for cap Construction steven.birdwell@r
construction Services, L.P. econservices.com
RECON Surface 70,000 cy Ferric chloride In-situ bucket Impoundment Federal Safe Remedial S. Birdwell
2002 Impoundment sludge sludge mixing closures Drinking Water Act/ Construction steven.birdwell@r
Closure, 20% hydrated lime, Services, L.P. econservices.com
Edgemoor, DE 10% Portland
cement and 10%
water
RECON Oxidation Pond, 110,000 cy Hydrocarbons In-situ Strengthening Fly ash and cement Remedial S. Birdwell
2002 Richmond, CA sludge excavator for cap Construction steven.birdwell@r
construction Services, L.P. econservices.com
Geo-Con 2002 Flyash Pond, Over Metals In-situ, long- Flyash pond Cement 10% T. Plante T.Plante
Yarmouth, ME 10,000 reach closure tplante@haleyaldri
cycy excavator ch.com
207-482-4622
ENTACT, LLC Browns Battery 43,000 Lead Ex-situ Pug- Superfund Phosphate based ENTACT, LLC USEPA Region 3,
- 2002 NPL Superfund cubic yards mill remediation; risk reagent PADEP, US
Site of reduction Department of the
Pennsylvania impacted Interior Fish &
soil Wildlife Services

406
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

IT Corporation, Docklands 60,000 cy PAHs In-situ Redevelopment 10% Portland Paul Lear Paul.Lear
2002 Redevelopment, Sludge, tar, excavator cement + 10% plear@envirocon.
Melbourne, soils cement kiln dust or com
Victoria,Austraila 10% fly ash
>25 psi UCS
WRScompass, Athens 60,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2002 Riverbank, soil DNAPL slag plear@envirocon.
Athens, GA >50 psi UCS com
-6
<1x10 cm/s
permeability
IT Corporation, Chevron Perth 45,000 cy TPH, lead In-situ, Remediation 5%-10% Portland Paul Lear Paul.Lear
2002 Amboy Tank of tank Excavator cement plear@envirocon.
Cleaning bottoms >50 psi UCS com
Impoundments, TCLP lead <5 mg/L
Perth Amboy, NJ
WRScompass, Macon MGP 125,000 cy PAHs, coal tar, In-situ auger Redevelopment 2% Portland + 6% Paul Lear Paul.Lear
2002 Site, Macon, GA soil DNAPL slag, >50 psi UCS plear@envirocon.
<1x10-6 cm/s com
permeability
IT Corporation, American Home 24,000 cy Tar Ex-situ Closure 15% Portland Paul Lear Paul.Lear
2002 Products Tar, soil pug-mill cement plear@envirocon.
Impoundment 26 >50 psi UCS com
Closure, <5% strain
Boundbrook, NJ
IT Corporation, Sunflower Army 55,000 cy Lead, Ex-situ pug-mill Closure 5% Portland cement Paul Lear Paul.Lear
2002 Ammunition soil and explosives, TCLP lead < 5 mg/L, plear@envirocon.
Depot SWMU sludges propellants No reactivity com
10 & 11, De
Soto, KS

407
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
WRScompass, Savannah MGP 25,000 cy PAHs, coal tar, In-situ auger containment 2% Portland + 6% Paul Lear Paul.Lear
2001 Site, Savannah, soil DNAPL slag, >50 psi UCS plear@envirocon.
Georgia <1x10-6 cm/s com
permeability
WRScompass, Waycross Canal, 50,000 cy PAHs, coal tar, In-situ auger Redevelopment 3% Portland + 9% Paul Lear Paul.Lear
2001 Waycross, GA soil DNAPL slag,>50 psi UCS plear@envirocon.
-6
<1x10 cm/s com
permeability
WRScompass, Former Police 32,000 cy Lead Ex-situ Pug-mill Closure 3% Enviroblend Paul Lear Paul.Lear
2001 Pistol Range, soil TCLP lead < 5 mg/L plear@envirocon.
Tampa, FL com
IT Corporation, Former Henry 20,000 cy Chromium In-situ Closure 2% ferrous sulfate + Paul Lear Paul.Lear
2001 Woods & Sons soil excavator 2% calcium plear@envirocon.
Paint Factory polysulfde com
Paint Pigment TCLP Cr <5 mg/L
Disposal Area,
Wellesley, MA
IT Corporation, Former Kaiser 9,000 cy TPH, sulfuric In-situ Remediation 5% Portland cement Paul Lear Paul.Lear
2001 Steel Plant Tar acid excavator + 10% Class C fly plear@envirocon.
Pits ash com
Remediation, >25 psi UCS
Fontana, CA
IT Corporation, York Oil 12,000 cy PCBs Ex-situ Remediation 25% Portland Paul Lear Paul.Lear
2001 Superfund Site excavator cement plear@envirocon.
OU-2 Wetlands >50 psi UCS com
Soils, Moira, NY
IT Corporation, Barbers Point 41,000 cy Lead In-situ Closure 1-4% Triple Paul Lear Paul.Lear
2001 Scrap Yard soil Dozer Superphosphate plear@envirocon.
Remediation, TCLP lead <5 mg/L com
Oahu, HI

408
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
IT Corporation, Camp Allen 9,000 cy Cadmium In-situ Remediation 5% Portland cement Paul Lear Paul.Lear
2001 Naval Station excavator TCLP cadmium <1 plear@envirocon.
Salvage Yard, mg/L com
Norfolk, VA
RECON Peak Oil and 55,000 cy Lead and Ex-situ pug-mill Superfund Portland cement and Remedial S. Birdwell
2001 Bay Drums soil chlordane remediation TSP Construction steven.birdwell@r
Superfund, 282 g/l SPLP and Services, L.P. econservices.com
-5
Tampa, FL 1x10 cm/sec perm.
RECON Power Plant 2,100 cy Wastewater In-situ bucket Disposal as Portland cement and Remedial S. Birdwell
2001 Stabilization, sludge and cooling mixing Class I quicklime Construction steven.birdwell@r
Denton, TX tower sludge Services, L.P. econservices.com
RECON SWIB/WWTS 35,000 cy Hydrocarbons In-situ bucket Levee Patented LSS Remedial S. Birdwell
2001 Cleanout Project, sludge mixing construction Construction steven.birdwell@r
Norco, LA Services, L.P. econservices.com
RECON Chemical 10,000 cy Radionuclides, In-situ soil Risk reduction Cement, fly ash, Remedial S. Birdwell
2000 Oxidation, Aiken, soil and heavy metals, mixing bentonite, and Construction steven.birdwell@r
SC sludge organic and zeolite Services, L.P. econservices.com
inorganics
Geo-Solutions, South 8th St, 40,000 Pb, acid oily In-situ/auger Superfund limestone 16% E. Bates USEPA
Four Seasons West Memphis cycy sludge Ex-situ/ remediation cement 13%, Fly ash 2009
2000 Arkansas bucket mix risk reduction 6.5%, UCS 100 psi
-6
1X10 cm/sec
SPLP Pb 15 ug/L
IT/OHM American 45,000 creosote, Ex-situ/pug-mill Superfund cement 5%, fly ash E. Bates USEPA
2000 Creosote, cycy PCP,dioxins, remediation 4.5 %, carbon 1.3% 2009
Jackson, TN furans risk reduction UCS 100 psi
-6
1X10 cm/sec
SPLP PCP 200
ug/L, SPLP
Dioxin30 pg/L

409
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
IT, 2000 Seneca Army 6,000 cy Lead Ex-situ Remediation 5% Portland cement Paul Lear Paul.Lear
Depot Firing Soils and excavator TCLP lead <5 mg/L plear@envirocon.
Range sediment com
Remediation,
Romulus, NY
IT Corporation, Drake Chemical 1,000 cy Arsenic, Ex-situ Waste treatment 5% Portland cement Paul Lear Paul.Lear
2000 Superfund site ash barium, excavator + 2% ferrous sulfate plear@envirocon.
Incinerator Ash cadmium, TCLP metals < com
Stabilization, chromium, lead Drinking Water
Lock Haven, PA Standards
IT Corporation, Roma Street 16,000 cy PAHs, TPH Ex-situ Redevelopment 5% Portland cement Paul Lear Paul.Lear
2000 Parklands soil, coke, excavator >50 psi UCS, plear@envirocon.
Redevelopment, ash, <1x10-6 cm/s com
Brisbane, cinders permeability,
Queensland, SPLP PAHs <0.01
Australia mg/L
IT Corporation, ABEX Superfund 20,000 cy Lead Ex-situ pug-mill Remediation 5% Portland cement Paul Lear Paul.Lear
2000 Site, Portsmouth, TCLP lead <0.75 plear@envirocon.
VA mg/L com
WRScompass, Ashley River 14,000 cy, PAHs, creosote Ex-situ Remediation Portland cement Paul Lear Paul.Lear
2000 Site, Charleston, sediment Excavator >50 psi UCS plear@envirocon.
SC com
IT Corporation, Fort Benjamin 15,000 cy Lead In-situ Remediation 4% Triple Paul Lear Paul.Lear
2000 Harrison Small tiller Superphosphate plear@envirocon.
Arms Firing TCLP lead <5 mg/L com
Ranges, Indiana
Earth Tech Southwire, Soil, Pb Ex-situ State Action, EnviroBlend 80/20, D. Wheeler Dave Wheeler
1999 Carrolton, GA 15,000 Excel Port-a- risk reduction 3%, TCLP Pb 5 dwhee87@yahoo.
cycy Pug mg/L com

410
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
IT Corporation, Willow Run 380,000 cy TPH, PCBs Ex-situ Pug-mill Closure 5% Cement kiln dust Paul Lear Paul.Lear
1999 Creek Site + 10%-15% Class F plear@envirocon.
Remedial Action, fly ash com
Washetaw >50 psi UCS
-6
County, MI <1x10 cm/s
permeability
OHM, 1999 Exxon Bayway 120,000 cy TPH In-situ Closure 5% Portland cement Paul Lear Paul.Lear
Refinery excavator + 5% - 10% cement plear@envirocon.
Impoundment kiln dust com
Closure, Linden, >10 psi UCS
NJ
IT Corporation, Former Blackhills 15,000 cy Chromium Ex-situ Closure 1% ferrous sulfate, + Paul Lear Paul.Lear
1999 Army soil/sludge excavator 1% calcium plear@envirocon.
Ammunition polysulfide + 5% com
Plant, Igloo, SD Portland cement
TCLP Cr <5 mg/L
RECON Section 9 126,000 cy Various In-situ bucket Strengthen for Patented LSS/25 psi Remedial S. Birdwell
1999 Port Arthur, TX soil and hydrocarbons mixing cap construction UCS Construction
sludge Services, L.P. steven.birdwell@r
econservices.com
RECON Track F 300,000 cy Various In-situ bucket Strengthen for Patented LSS/25 psi Remedial S. Birdwell
1999 Port Arthur, TX soil and hydrocarbons mixing cap construction UCS Construction steven.birdwell@r
sludge Services, L.P. econservices.com
RECON Pond Closure, El 23,000 cy Heavy metals In-situ Risk based Fly ash and Portland Remedial S. Birdwell
1998 Paso, TX sludge bucket mixing closure cement Construction steven.birdwell@r
Services, L.P. econservices.com

411
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

ENTACT, LLC Schuylkill Metals 265,000 Lead; metals; Ex-situ - Pug- Superfund cement 10% E. Bates USEPA
- 1998 NPL Superfund tons of battery casings mill remediation; risk TSP 2% 2009
Site - impacted reduction
Plant City soils and UCS 50 psi
Florida sediments; 1X10-6 cm/sec
battery SPLP Pb 1 mg/L
casings TCLP Pb 5 mg/L
Earth Tech ByPass 601 NPL Soil Pb, from Ex-situ Superfund Cement 5% D. Wheeler Dave Wheeler
1998 Concord, NC 10,420 battery Excel Port-a- remediation EnviroBlend 80/20 dwhee87@yahoo.
cycy recycling pug risk reduction 5% com

OHM, 1998 NL Taracorp 4,000 cy Lead Ex-situ pug-mill Remediation 5% Portland cement Paul Lear Paul.Lear
Residential TCLP lead <5 mg/L plear@envirocon.
Cleanup, Granite com
City, IL
OHM, 1998 Double Eagle 60,000 cy TPH, sulfuric In-situ Closure 15% Class C fly ash Paul Lear Paul.Lear
Refinery Acid acid tar acid, hydrogen excavator + 5% Portland plear@envirocon.
Tar Ponds, sludge sulfide, sulfur cement com
Muskogee, OK dioxide >50 psi UCS
pH>7
IT Corporation, Fort Gillem Small 5,000 cy Lead Ex-situ Remediation 5% Portland cement Paul Lear Paul.Lear
1998 Arms Range soil excavator TCLP Pb <5 mg/L plear@envirocon.
Remediation. com
Atlanta, GA
OHM, 1998 Fayette 10,000 cy Lead Ex-situ Remediation 8% lime kiln dust Paul Lear Paul.Lear
Equipment and soil pug-mill TCLP lead <5 mg/L plear@envirocon.
Salvage Site, com
Uniontown, PA

412
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
OHM, 1997 Camp Pendleton 44,000 cy DDT, DDD, Ex-situ Pug-mill Closure 5% Class C fly ash + Paul Lear Paul.Lear
IR Sites 3 and 6, soil DDE, dioxins, 1% carbon plear@envirocon.
Oceanside, CA PCBs, TPH, SPLP com
lead, cadmium organochlorine
pesticides <0.001
mg/L
SPLP PCBs <0.001
mg/L
IT Corporation, Shaler/JTC Soil 60,000 cy BTEX In-situ auger Remediation 10% Portland Paul Lear Paul.Lear
1997 Remediation soil cement, >50 psui plear@envirocon.
Project, Bruin, UCS, <1x10-6 cm/s com
PA permeability
RECON French Limited 30,000 cy Organics - In-situ bucket Stabilization Soil and fly ash Remedial S. Birdwell
1996 Superfund Site, sludge biosolids mixing Construction steven.birdwell@r
Crosby, TX Services, L.P. econservices.com
GNB Env. Yellow Water Rd 4,472 cycy PCBs Ex-situ, Excel Superfund Cement 27% D. Wheeler Dave Wheeler
Svcs. NPL soils Port-a-Pug remediation UCS 50psi dwhee87@yahoo.
1996 Baldwin FL 1X10-6 cm/sec com

OHM, 1996 Hercules 009 30,000 cy Toxaphene In-situ Closure 15% Portland Paul Lear Paul.Lear
Landfill, excavator cement, >50 psi plear@envirocon.
-6
Brunswick, GA UCS, <1x10 cm/s com
permeability
TCLP Toxaphene
<0.050 mg/L
OHM, 1996 NL Taracorp 15,000 cy Lead Ex-situ pug-mill Remediation 5% Portland cement Paul Lear Paul.Lear
Superfund Site, TCLP lead <5 mg/L plear@envirocon.
Granite City, IL com

413
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

IT Corporation, MOTCO Facility 14,000 cy Mercury In-situ Closure 15% lime kiln dust Paul Lear Paul.Lear
1995 Impoundment sludge, Excavator TCLP Hg <0.02 plear@envirocon.
Closure, salts mg/L, >25 psi UCS com
LaMarque, TX
IT Corporation, Mobil Refinery 600,000 cy TPH In-situ Closure Portland cement, Paul Lear Paul.Lear
1995 Stormwater excavator cement kiln dust, fly plear@envirocon.
Pond Closures, ash, bed ash (mix com
Beaumont, TX design varied
throughout project)
>10 psi UCS
OHM, 1995 Johnston Atoll 12,000 cy Lead Ex-situ Closure 5% hydrated Lime Paul Lear Paul.Lear
Ash Pit soil, ash, Pug-mill TCLP lead <5 mg/L plear@envirocon.
Stabilization metallic SPLP lead <5 mg/L com
Project, debris
Johnston Island
IT Corporation, BabCOCk & 46,000 cy Metals Ex-situ Remediation 15% Portland Paul Lear Paul.Lear
1994 Wilcox Koppell Electric arc pug-mill cement plear@envirocon.
Facility, Koppel,
furnace TCLP Pb <5 mg/L com
PA dust, soil, >50 psi UCS
sludge
Chemical GM Fisher Guide 370,000 cy Metals Ex-situ Pug-mill Closure 10% Lime Kiln Dust Paul Lear Paul.Lear
Waste Landfarm Soil, + 5% Class C Fly plear@envirocon.
Management/O Closure, Flint, MI sludge, Ash com
HM, 1994 sediment TCLP Cd <0.11
mg/L, Cr <0.6 mg/L,
Pb <0.75 mg/L, Ni <
11 mg/L, Ag <0.14
mg/L
>50 psi UCS

414
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Chemical DOE LEHR 2,500 cy Radionuclides Ex-situ Remediation 20% Portland Paul Lear Paul.Lear
Waste Facility, Davis, soil, pug-mill cement plear@envirocon.
Management, CA sludges, ANS 16.1 Leach com
1994 lab Indices >10 for
residuals radionuclides
OHM, 1994 Umatilla Army 25,000 cy Lead, TNT, Ex-situ pug-mill Closure 5% Portland cement Paul Lear Paul.Lear
Ammunition soil DNT, TNB, TCLP lead <5 mg/L, plear@envirocon.
Depot Burn Pits, RDX, HMX TNT < 0.20 mg/L, com
Hermiston, OR RDX <0.13 mg/L,
TNB < 0.18 mg/L,
RDX <0.2 mg/L,
HMX ,18 mg/L
Geo-Con Georgia Power PAHs, BTEX, In-situ/auger remediation cement 10-25% E. Bates USEPA
1993 Gas Plant, cyanide Site reuse UCS 60 psi 2009
Columbus 1X10-5 cm/sec
Georgia Interior and
1X10-6 cm/sec
edge columns

OHM, 1993 Chevron Pollard 150,000 cy TPH, sulfuric Ex-situ Pug-mill Redevelopment 10% calcium Paul Lear Paul.Lear
Pond acid sludge acid, hydrogen carbonate + 5% plear@envirocon.
Remediation, tar, bay sulfide, sulfur Portland cement + com
Richmond, CA mud dioxide 5% Class C fly ash
>50 psi UCS
1% alkalinity reserve
5>pH <12

415
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

Chemical Selma Pressure 11,500 cy Arsenic, Ex-situ pug-mill Remediation 12% Portland Paul Lear Paul.Lear
Waste Treating soil copper, cement + 4% plear@envirocon.
Management, Superfund Site chromium, organoclay com
1993 Remediation, pentachlorophe >50 psi UCS
Selma, CA nol, dioxins
OHM, 1993 Beals Battery 15,000 cy Lead Ex-situ Remediation 15% Portland Paul Lear Paul.Lear
Site soil pug-mill cement plear@envirocon.
TCLP Pb <5 mg/L com
>50 psi UCS
Silicate Selma Pressure 13,000 PCP, As, Cu, Ex-situ Superfund cement E. Bates USEPA
Technology Treating, cycy Cr, dioxins paddle mixer remediation activated carbon 2009
Corp. Selma California furans risk reduction USEPA
1993 2004
OHM, 1993 Kassouf- 14,000 cy Lead Ex-situ pug-mill Remediation 10% Portland + 1% Paul Lear Paul.Lear
Kimerling Battery hydrated lime + 10% plear@envirocon.
Disposal class C fly ash com
Superfund Site, TCLP <5 mg/L
Tampa FL >50 psi UCS
<1x10-6 cm/s
permeability
Modified ANS 16.1
Lead Leach Index
>10
Vendor Pepper Steel, 85,000 Pb, As, PCBs Ex-situ Superfund cement E. Bates USEPA 2009
unknown Medley Florida cycy equipment remediation fly ash
1993 soils unknown risk reduction UCS 21psi
1X10-6 cm/sec
EP Toxicity

416
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs
GCI Private Client 20,000 T Sewage Cake Ex-situ Produce Potting Lime 20% R. Garrett Robert Garrett
1992 Baltimore, MD ARAN Pug-mill Soil Soil 20% bobgarrett417@b
ellsouth.net

OHM, 1992 Henkel 52,000 cy Lead, arsenic, In-situ Closure 5% Portland cement Paul Lear Paul.Lear
Corporation TPH, PAH, excavator + 15% fly ash plear@envirocon.
Impoundment PCBs >50 psi UCS com
-6
Closure, <1x10 cm/s
Carlstadt, NJ permeability

OHM, 1992 Lees Farm 12,000 cy Lead Ex-situ Remediation 15% Portland Paul Lear Paul.Lear
Superfund Site, soil pug-mill cement plear@envirocon.
Woodville, WI TCLP Pb <5 mg/L com
>50 psi UCS
<1x10-6 cm/s
permeability
Chemical Portsmouth 1,000 cy TCE, In-situ Remediation TCLP TCE <0.5 Paul Lear Paul.Lear
Waste Gaseous soil radionuclides auger mg/L plear@envirocon.
Management, Diffusion Plant >50 psi UCS com
-6
1992 Site X-231, <1x10 cm/s
Portsmouth, OH permeability
ANS 16.1 Leach
Indices >10 for
radionuclides
Chemical Port of Los 75,000 cy Lead Ex-situ pug-mill Redevelopment 15% limestone + 5% Paul Lear Paul.Lear
Waste Angeles Berths soil sodium plear@envirocon.
Management, 212 and 215 polycarbonate com
1991 Remediation, Cal WET lead <5
Los Angeles, CA mg/L
TCLP lead <5 mg/L

417
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

OHM, 1991 New Hampshire 27,000 cy metals Ex-situ Remediation 15% Portland Paul Lear Paul.Lear
Plating Sludge and excavator cement plear@envirocon.
Superfund Site, soil TCLP Cd <0.11 com
Merrimack, NH mg/L, Cr <0.6 mg/L,
Pb <0.75 mg/L, Ni <
11 mg/L, Ag <0.14
mg/L
WRScompass, Columbus MGP 80,000 cy PAHs, coal tar, In-situ auger Redevelopment 3% Portland + 9% Paul Lear Paul.Lear
1991 FCRC, soil DNAPL slag plear@envirocon.
Columbus, GA >50 psi UCS com
<1x10-6 cm/s
permeability
Chemical Solvay Animal 5,000 cy Arsenic In-situ Closure 10% Class c fly ash Paul Lear Paul.Lear
Waste Health RCRA excavator TCLP As <5 mg/L plear@envirocon.
Management, Impoundment >50 psi UCS com
1991 Closure, Charles
city, IA
GCI Johnson Soil Pb Ex-situ State Action CKD R. Garrett Robert Garrett
1991 Controls 3,000 T ARAN Pug-mill Property 15% bobgarrett417@b
Atlanta, GA Transaction ellsouth.net

GCI BFI 10,000 T Liquid Waste Ex-situ Produce Solids Lime 15% R. Garrett Robert Garrett
1991 Atlanta, GA ARAN Pug-mill Soil 20% bobgarrett417@b
ellsouth.net

GCI Johnson Soil Pb Ex-situ State Action Proprietary R. Garrett Robert Garrett
1990 Controls 12,000 T ARAN Pug-mill Property 12% bobgarrett417@b
Atlanta, GA Transaction ellsouth.net

418
Treatment Site Name/ Quantity/ Contaminant In-situ/Ex-situ Purpose Reagent Submitted Reference/
Vendor/date Location Media & Equipment Formula/ by: Or Contact
of treatment treat specs

GCI U.S. Navy 10,000 T Pb Ex-situ Shooting Range CKD R. Garrett Robert Garrett
1990 San Francisco, ARAN Pug-mill Remediation 15% bobgarrett417@b
CA ellsouth.net

Chemical Lack Industries 20,000 cy Metals In-situ Closure 15% Lime Kiln Dust Paul Lear Paul.Lear
Waste Impoundment sludge Excavator TCLP Cd <0.11 plear@envirocon.
Management, Closure, Grand mg/L, Cr <0.6 mg/L, com
1990 Rapids, MI Pb <0.75 mg/L, Ni <
11 mg/L, Ag <0.14
mg/L
Geo-Con (a Dundalk Marine 350 LF Chromium Ore In-situ Strain Relief Horticultural Grade K. Andromalos M. Kitko
trade name of Terminal Processed Auger Peat Moss mkitko@geo-
Geo-Solutions) Baltimore, MD Residue3 solutions.com

unknown New York Harbor over metals, dioxins, Ex-situ/ pug- dredge spoils cement 8 % E. Bates USEPA
unknown and Port 145,000 PAHs, PCBs mill disposal, 2009
Authority, cycy Brownfield
New York redevelopment
Greenleaf Env. Edgewood Ave. Soil Pb Ex-situ State Action EnviroBlend 80/20 D. Wheeler Dave Wheeler
Svcs Site 4,500 cycy Excavator Property 3% dwhee87@yahoo.
Atlanta, GA mixing transaction <5 mg/L TCLP Pb com

419
Appendix A References
USEPA, Technology Performance Review: Selecting and Using Solidification/Stabilization Treatment for Site Remediation,
EPA/600/R-09/148, November 2009, http://www.epa.gov/nrmrl/pubs/600r09148.html

USEPA, Remedial Action Report Soil Remedy: Selma Pressure Treating Site, Selma, California, EPA CERCLIS ID Number
CAD029452141USEPA Region 9, San Francisco, California, September 29, 2004.

420
Appendix B: Case studies employing S/S

Appendix B presents a collection of S/S case studies from the USA and
Europe. Each case study is presented as a separate file named for the
location of the project. The files are presented in alphabetical order. These
case studies are presented as received from leading S/S vendors and the
editors have not independently verified the information. However for each
case study the name of the person submitting the case study and their E mail
address is provided should anyone wish to obtain more information regarding
that project.

No. Case Study Author Contact


Abex Superfund,
1 Paul Lear [email protected]
VA
American
2 Creosote Paul Lear [email protected]
Superfund Site
Bayou Steven R.
3 [email protected]
Trepagnier, LA Birdwell
BHAD
4 Chromium Pond, Paul Lear [email protected]
SD
5 Cambridge, MA Mark Kitko [email protected]
Former Camden
6 Gas Works Site, Paul Lear [email protected]
NJ
Camp Pendleton
7 Paul Lear [email protected]
Scrap Yard, CA
Chevron Pollard
8 Paul Lear [email protected]
Landfill Site
9 Carnegie,PA Mark Kitko [email protected]
Ken
10 Columbus, IN [email protected]
Andromalos
CSX Benton
11 Paul Lear [email protected]
Harbor,MI

12 Double Eagle Paul Lear [email protected]


Refinery Site,
421
OK
Dundalk,
13 Mark Kitko [email protected]
Baltimore, MD
East Rutherford, Ken
14 [email protected]
NJ Andromalos
Foote Minerals
15 Superfund Site, Paul Lear [email protected]
PA
Stany
16 Ghent, Belgium [email protected]
Pensaert
Guernsey, Stany
17 [email protected]
United Kingdom Pensaert
Hercules 009
18 Paul Lear [email protected]
Landfill Site, GA
Hoedhaar
Stany
19 Lokeren, [email protected]
Pensaert
Belgium
20 Irving, Tx Mark Kitko [email protected]
21 Johnston Atoll Paul Lear [email protected]
22 Kingston,TN Steve Artman [email protected]
London, United Stany
23 [email protected]
Kingdom Pensaert
24 Martinsville, VA Mark Kitko [email protected]
Milwaukee, Roy E.
25 [email protected]
Wisconsin Wittenberg
Steven R.
26 Nederland, TX [email protected]
Birdwell
New Bedford, Ken
27 [email protected]
MA Andromalos
28 Nyack, NY Mark Kitko [email protected]
NYSEG Norwich
29 Paul Lear [email protected]
NY MGP Site
Stany
30 Obourg, Belgium [email protected]
Pensaert
31 Perth Amboy, NJ Paul Lear [email protected]

422
32 Portsmouth, VA Mark Kitko [email protected]
Stany
33 Rieme, Belgium [email protected]
Pensaert
Roma Street
34 Paul Lear [email protected]
Station
35 Sag Harbor, NY Mark Kitko [email protected]
Sanford MGP
36 Paul Lear [email protected]
Superfund Site
Soderhamn, Stany
37 [email protected]
Sweden Pensaert
Southeast Roy E.
38 [email protected]
Wisconsin Wittenberg
Ken
39 St Louis, MO [email protected]
Andromalos
Sunflower Army
40 Ammunition Paul Lear [email protected]
Depot
Jerome
MacNeil [email protected]
Sydney Tar
41 Diane [email protected]
Ponds
Ingraham
[email protected]
Donnie Burke
Umatilla
42 Ammunition Paul Lear [email protected]
Depot
Valero
43 Paulsboro Paul Lear [email protected]
Refinery
Ken
44 Waukegan, IL [email protected]
Andromalos
West Doane
45 Paul Lear [email protected]
Lake Site
X-231B Pilot
46 Paul Lear [email protected]
Study
Zwevegem, Stany
47 [email protected]
Belgium Pensaert

423
Case Study 1: Abex Superfund Site

Location: Portsmouth, VA, USA

History: Former foundry

Contaminants: Lead, cadmium, zinc

28,354 tons
Scale:
Paper sludge ash
Reagents:
Ex-Situ
Method:
Housing
End-Use:

Site Description

This project included a multi-faceted removal action at Former Abex Foundry and the
adjacent Portsmouth Redevelopment Housing Authority's (PRHA's) Washington
Park Homes area. The project had extensive regulatory oversight, and involved:

Excavation, on-site staging, and classification of 85,000 tons of lead-impacted soil


and debris
Pug-mill stabilization using lime/Portland cement for 28,354 tons of hazardous
lead-contaminated soil to UTS Standards
Removal and disposal of 6,000 tons of construction debris
Transportation and disposal of 87,600 tons of non-hazardous waste and 1,200
tons of hazardous waste

Excavation included removal of 1 to 5 ft of soils determined by future land usage as


mandated by the Consent Decree involving Pnuemo Abex and the USEPA.
Relocation of the residents and securing of the PRHA apartments was performed
prior to excavation. In addition, a video survey was performed to capture the pre-
remediation condition of structures and ground cover.

424
Areas were excavated and direct-loaded into 16 yd3 dump trucks. Excavation grade
was maintained utilizing a laser level operated by a ground man who worked closely
with the equipment operator. Materials were transported and staged on a soil
storage and treatment pad in 100-ton delineated bins for waste classification.

The classification phase was performed to define which materials could be disposed
of as a non-hazardous waste without first requiring stabilization. The material
requiring stabilization (greater then 5 ppm TCLP lead) was then treated using a
Portec 53 Pug-mill. The reagent admixture rate was defined through treatability
testing and applied to 23,354 tons of material which achieved the criteria of 0.75 ppm
TCLP lead (UTS Standard) at a 99% efficiency pass rate. Following post-treatment
verification sampling, the material was loaded and transported and disposed of at a
local non-hazardous landfill.

Drivers

Public housing was constructed on top of soil contaminated from the past foundry
activities. The contaminated soils required removal to prevent contact with the
contaminated soil by the residents.

Objectives

Remove the contaminated soil


Treat (if applicable) the soil to remove its RCRA hazardous characteristic D008
(lead), dispose of the soil at an off-site landfill
Restore the site

Method

Material with greater than 5 mg/L TCLP-leachable lead was treated with Portland
cement to reduce the TCLP-leachable lead concentrations to less than 0.75 mg/L.

A pug-mill-based system, consisting of a feed hopper with conveyor, pug-mill,


reagent silo, and stacking conveyor was used to treat 150-200 tons of contaminated
soil per hour. The treated soil was segregated into 200 ton stockpiles for
confirmation testing.

425
Case Study 2: American Creosote Superfund Site

Location: Jackson, TN

History: Former wood preserving site

Contaminants: PAHs, PCP, arsenic, dioxins

22,500 tons
Scale:
Cement, fly ash & activated carbon
Reagents:
Ex-situ
Method:
Brownfield redevelopment
End-Use:

426
Site Description

The site consists of a former wood preserving site where wood was pressure-treated
with creosote and pentachlorophenol. Wastewater sludges from the wood
preserving operations were land applied, contaminating the site with polyaromatic
hydrocarbons (PAHs), pentachlorophenol (PCP), dioxins, and arsenic. The primary
goals of the remediation were to limit the solubility of the chemicals of concern to
minimize their leachability. The treated waste was then to be buried in an on-site
excavation area, covered with a clay liner and topsoil, and seeded with grass

OHM Corporation (OHM) completed remediation work at the American Creosote Site
in Jackson, Tennessee under contract with the State of Tennessee and the US EPA
Region IV. The project scope involved the remedial design, construction of a slurry
cut-off wall around the impacted soils, excavation of the impacted soils, collection
and treatment of perched water and DNAPL, stabilization treatment of the impacted
soils, compaction and placement of the treated soil back into the excavation area,
and capping the treated soil with clay and topsoil.

A 2,400 ft slurry cut-off wall was constructed around the perimeter of the impacted
soil. This cut-off wall was filled with a soil-bentonite mixture which had a permeability
of less than 1x10-6 cm/s and was keyed into the native clay layer underlying the site
at a depth of 3 to 5 ft below ground surface.

The impacted sandy soil inside of the slurry wall was excavated to the underlying
clay layer. Visibly stained clay material was also removed during the excavation.
The perched water and DNAPL in the excavation area were collected and removed
by a series of drainage trenches. The perched water and DNAPL were treated on-
site in a wastewater treatment plant designed, built, and operated by Shaw. In the
wastewater treatment plant, an oil/water separator was used to separate the water
from the DNAPL. The water was then passed through a sand filter and activated
carbon before it was discharged to a local POTW. The collected DNAPL was
containerized and shipped off-site for treatment and disposal.

The excavated material was stockpiled and screened through a screen plant
consisting of a grizzly screen and a screen deck. The oversized material (>2 in) from
the screening operation was stockpiled for placement with the treated soil. The
material passing through the screen deck was suitably sized for treatment and was
subsequently transported to the pug-mill feed hopper for stabilization treatment. The
soil was mixed with the reagents and water inside the pug-mill. The pug-mill system
was operated 12 hours per day, 5 day per week, treating 1,200 tons of soil per day.

427
The treated soil exited the pug-mill on a conveyor, and was placed in temporary
stockpiles until daily sampling was complete. The treated material exiting the pug-
mill had a moist soil-like consistency. Confirmatory testing for the leachability of
PAHs, PCP, dioxins, and arsenic, permeability, and unconfined compressive
strength was performed on the sample from each temporary stockpile. All of the
confirmatory samples met the performance requirements for leachability,
permeability, and unconfined compressive strength. After confirmatory sampling
results were received for a treated soil stockpile, the treated soil was placed back
into the excavation area in 12-in lifts and compacted. Daily surveys of the excavated
area defined the placement position of each stockpile of the treated soil.

After placement of all the treated soil, OHM graded the placed and compacted soil to
establish the required 1-2% slopes for the clay/topsoil cap. The graded soil was
covered by a Claymax geosynthetic clay liner (GCL) in accordance with the
manufacturers instructions. The GCL was then covered with 18-in of common fill
and 6-in of topsoil. The common fill was placed in 1-ft loose lifts and track-walked
into place. The topsoil was then hydromulched and seeded.

Risk Drivers

The site was slated for brownfield redevelopment by a local utility. The
contaminated soils required treatment to prevent off-site migration of the
contaminants into the surficial aquifer.

Objectives

Remove the contaminated soil


Treat to reduce the leachability of the contaminants and produced a high strength,
low permeable treated material
Place and compact the treated material into the excavation area
Restore the site

428
Method

The contaminated soil was treated with a mixture of Portland cement, Class F fly
ash, and activated carbon to produce a treated material which met the performance
criteria listed in Table 1.

Table 1: SPLP Results for the Wood Preserving Site Soil

Portland Cement 0.05


Class F Fly Ash 0.06
Activated Carbon Untreated 0.02
Water 0.05
Parameter SPLP (ug/L) SPLP (g/L)
Pentachlorophenol 9,600 <100
Benzo(a)pyrene Potency Equivalence 155 <1
2,3,7,8 TCCD Toxicity Equivalence 0.25 <0.03

Validation

Temporary 500 ton stockpiles of the treated soil were created. Confirmatory testing
for the leachability of PAHs, PCP, dioxins, and arsenic, permeability, and unconfined
compressive strength was performed on the sample from each temporary stockpile.
All of the confirmatory samples met the performance requirements for leachability,
permeability, and unconfined compressive strength.

Equipment Used

The stabilization treatment involved the use of a pug-mill system to mix the soil with
activated carbon, Class F fly ash, and Portland cement. OHM's pug-mill system
consisted of a variety of feeders, conveyors, silos, and a pug-mill mixer integrated
into a complete system for the continuous mixing of wastes and reagents. The
screened material was fed to an 8-yd3 feed hopper. The hydraulically-driven belt,
located in the bottom of the feed hopper, fed the material onto a 40-ft long by 24-in
wide belt conveyor. The conveyor belt was equipped with a Ramsey single idler belt
scale for the accurate, real-time determination of the rate, in tons per hour, of soil
being treated. The conveyor belt conveyed the material into the pug-mill for blending
429
with the stabilization additives. The pug-mill mix box was 4 ft wide by 3 ft high by 10
ft long. Paddles were bolted onto structural steel shafts with replaceable shafts
flanged on both ends for ease of maintenance. The mixer was V-belt driven by a
125-hp motor.

Dry stabilization additives were stored on-site in vertical cement silos. The silos
were self-leveling and had a capacity of 200 barrels of material. The silos were also
equipped with a top mount baghouse for dust control during silo filling. The feed
from each silo was controlled by a 14-in diameter screw auger powered by a 5hp
motor. The dry stabilization additives were introduced from silo feeders which are
attached to the pug-mill through a central feed auger. The soil was mixed with the
reagents and water inside the pug-mill. The pug-mill system was operated 12 hours
per day, 5 day per week, treating 1,200 tons of soil per day.

Specific issues

For this site, the EPA Superfund Technical Assistance and Response Team
(START) promulgated risk-based SPLP leaching criteria for PAHs, PCP, and dioxins.
Based on the successful completion of the American Creosote Site remediation, the
US EPA used these leachability criteria for PAHs, PCP, and dioxins for use at other
wood-preserving sites contaminated with creosote.

430
Case Study 3: Bayou Trepagnier

Location: St. Charles Parish, Louisiana

History: Petrochemical plant

Contaminants: Hydrocarbons

50,000 cy
Scale:
Paper sludge ash
Reagents:
In-situ bucket mixing
Method:
N/A
End-Use:

Site Description

The Bayou Trepagnier was the original drainage canal of two local petrochemical
plants, sandwiched in between the Mississippi River and Lake Pontchartrain. Over
the course of many years, residual petroleum impacted the sediment and banks of
431
the Bayou. The scope of this remediation project was to immobilize the petroleum
impacted sediment in the first 6,000-ft of the Bayou and build a cap over it, protecting
human health and receptors in the wetland/swamp.

In addition, the work included removal and chipping of trees, access road
construction and installation of a FlexiFloat Bridge over the new drainage canal,
construction of storm water holding cells for bayou dewatering activities, solidification
of bayou sediments, placement of a clean clay cap over solidified sediments, and
seeding of disturbed areas. A clean zone was also required that would entail
excavating stabilized sediments in a 500 ft segment, and reuse these to bulk-out
impacted sediment downstream that required stabilization, and backfill the clean
zone with imported fill material so a future diversion canal could be installed through
non-impacted soil.

Characterisation

RECON was involved early in the design phase, undertook treatability studies and
constructability reviews, and a field pilot demonstration project. The final design was
completed after incorporating data generated from the treatability study in the
Feasibility Study Phase, and a successful pilot field study project. During the design
phase of the project, RECON ensured that the data gathered and lessons learned
(during the Feasibility Study) were incorporated into the successful full-scale
project. Waste composition, including reagents were as follows:

Sediment ~ 20%
Moisture Content ~ 62-67%
Oil & Grease ~ <3%
Cement/Lime ~ 10-15%

Risk Drivers

Unstable petroleum-impacted sediment was at risk of cross-contaminating the


LaBranche Wetlands. The site was already impacted by saltwater intrusion from
Lake Pontchartrain, as a result of storm surge flooding, leading to die-back of the
wetland and swamp vegetation.

432
Objectives

The objectives of the project were to:


Access the site, minimizing the disturbance of wetlands
Install dams throughout to divide the bayou into several reaches
Stabilize the sediment in the bottom of the bayou to be supportive of a cap
Flip and reuse the access road material as the cap
Vegetate the cap and access with native plants

Method

Stabilization, not simply solidification, was the chosen remedy to mitigate risk of
chemically-impacted sediment transfer off-site. As with any chemically-impacted
material that is left in-place, stabilization binds both chemical and sediment together
to form a matrix that greatly reduces the leachability of the chemicals into the
environment. Therefore, no waste material required off-site disposal.

RECON utilized hydraulic excavators to perform mixing of reagents into the


sediment. Tandem dumps were used to import access road/cap material. Tracked
dump trucks were used to transport bank spoils to reaches throughout the bayou for
incorporation into the mix. A flexi-float bridge was installed at the egress point in
Engineers Canal to allow for one-way traffic of dump trucks. Long-reach excavators
were used to take advantage of reaching across and mixing grids the width of the
bayou. Other ancillary equipment included equipment mats, pumps, etc.

Validation

RECON used a pocket penetrometer to gauge the strength of stabilized material,


and performed Unconfined Compressive Strength (UCS) testing to verify. It was
determined that by creating a stabilized matrix, it would not require testing for
chemicals.

433
Issues

The project was a significant challenge and a great accomplishment, as the


stabilized cap over sediment within a bayou that meandered through a pristine
swamp, was deemed to have returned all project access to pre-existing conditions.

The Bayou Trepagnier Team worked closely together with regulatory and other
agencies, including the U.S. Army Corps of Engineers (USACE), State of Louisiana
Department of Environmental Quality (LDEQ), Pontchartrain Levee District (PLD),
State of Louisiana Coastal Protection Restoration Authority (CPRA) and its division
the Office of Coastal Protection and Restoration (OCPR), the Louisiana Department
of Wildlife and Fisheries (LDWF), and the Louisiana Department of Natural
Resources (LDNR) Office of Coastal Management.

Most risks were mitigated in the Feasibility and Design phases of the project by
having Owner, Engineer, Consultants and Contractor-RECON working together. The
project ended up approximately $1,500,000 under budget, and was on schedule
despite being shut down 4 weeks during high water levels in the Mississippi River
that required the USACE to open the Bonnet Carre Spillway gates.

434
Case Study 4: BHAD Chromium Pond

Location: Igloo, SD

History: Ordnance manufacture

Contaminants: Hexavalent chromium

13,400 tons
Scale:
Reducing agent + alkaline reagent
Reagents:
Ex-situ
Method:
N/A
End-Use:

Site Description

The former Blackhill Army Depot (BHAD) is located in Fall River County, Igloo, South
Dakota. The BHAD consists of approximately 21,100 acres and was a former
defence site, at which many forms of ordnance were stored, manufactured, and
destroyed from 1942, until its closure in 1967. Since closure, much of the property
has been used for livestock grazing and most of the former facilities show varying
stages of decay.

The chromium pond site was located in the ammunition workshop area and was
adjacent to the clean and paint building (Building 3038). The soil contained up to
8,000 mg/kg of chromium, primarily in the form of hexavalent chromium.

The remediation of the chromium pond areas entailed:

Conducting a soil stabilization treatability study


Work plan preparation
Mobilization of necessary personnel and equipment

435
Site setup including placement of project offices and installation of process
equipment and staging areas
Monitoring well abandonment
Excavation and treatment of 15,000 tons on chromium-contaminated soil
Transportation and disposal of the wastes
Confirmatory sampling and analysis
Surveying
Restoration, re-vegetation, and site clean-up.

Soil exceeding the risk-based remediation criteria for hexavalent chromium of 5.2
mg/kg for soil for 0 to 1 ft below ground surface, and 7.0 mg/kg for soil below 1 ft
below ground surface, was excavated and stockpiled. Processing took place in two
lined 150 ft x 100 ft bermed areas: the first used for reduction and interim storage,
and the second for stabilization, and final storage during confirmatory testing.
Each bermed area was graded to a sump to collect storm and decontamination
water for re-use in the stabilization treatment process. The process area contained
two 50 yd3 mixing boxes, interim storage of sixteen 40-ton batches awaiting
stabilization, and storage for thirty-six 40-yd3 batches awaiting confirmatory testing
and off-site disposal.

Characterisation

The soil was a silty clay soil with up to 8,000 mg/kg hexavalent chromium.

Risk drivers

The soil contained up to 8,000 mg/kg of hexavalent chromium. Crystals containing


hexavalent chromium were also found on the surface of the chromium pond. The
contaminated soils required removal to prevent contact with workers under the
potential commercial/industry redevelopment of the site.

Objectives

Remove the contaminated soil

436
Treat (if applicable) the soil to remove its RCRA hazardous characteristic D007
(chromium)
Dispose of the soil at an off-site landfill
Restore the site

Method

The treatment of the contaminated soil involved two stages: (1) mixing with reducing
agents to reduce the chromium to its trivalent oxidation state; (2) then treatment with
an alkaline reagent to immobilize the trivalent chromium. The stabilization treatment
of the metals-contaminated soil involves four process steps:

Reduction
Interim Staging
Stabilization
Confirmatory Testing

For the reduction step, an excavator or loader transferred 40 tons of soil from the
untreated stockpile to a mixing box, where 380 gallons of calcium polysulfide solution
and 2,850 gallons of water were added. Two supersacks, each containing 1 ton of
ferrous sulfate, were then added and the reagents and water mixed into the soil.

The equipment operator mixed the reducing agents, water and soil until it was visibly
determined that the mixture was homogeneous and was green in colour. The
reduced soil was then removed from the mixing box and transferred to one of the
interim storage bins. A sample of the treated soil from each treatment batch was
evaluated on-site to ensure no hexavalent chromium was present.

437
For the alkaline stabilization step, a loader transferred soil from the interim storage
bin to the stabilization mix box, where 2 tons of Portland cement was added. The
equipment operator then used an excavator to produce a visibly homogeneous final
product, which was removed to a final storage bin, prior to final sampling verification.
Following verification, the treated soils were transported to a licensed disposal facility
for disposal as non-hazardous waste/soils.

Validation

Samples of the treated soil were obtained at a frequency of every 40 tons treated
and subjected to TCLP testing for metals. All of the treated samples met the
performance criteria of less than 0.6 mg/L TCLP-leachable chromium on the first
pass.

Specific Issues

438
Residents of the nearby community were concerned with the disposal of the treated
soil in the local municipal landfill. Public meetings were held to explain how the
treatment reduces both the toxicity and mobility of the chromium contamination.

439
Case Study 5: Cambridge

Location: Cambridge, Massachusetts

History: Manufactured gas plant

Contaminants: MGP Waste (BTEX, PAHs)

1,200 yd3
Scale:
Portland Cement, Bentonite
Reagents:
In-situ, Jet Grouting
Method:

440
Summary

Geo-Con was contracted by BMR-Kendall Development to stabilize approximately


1,200 yd3 (cy) of MGP impacted soils underlying Linskey Way in Cambridge,
Massachusetts by the jet grouting method.

The impacted soils were the result of a residual contamination plume left over from
the MGP site that occupied the parcel from the mid-1800s. The plume could not be
stabilized during the in-situ remediation of the main site in 2000, because of the
number of major utilities existing beneath Linskey Way (this earlier work received
several awards).

Jet grouting uses the injection of grout at ultra-high (3,000 to 5,000 psi) pressures to
effectively cut and mix the soil, thereby creating soil-mix columns, forming a
monolith in which soils are solidified and stabilized.

The jet grout column locations were drilled with a non-destructive vacuum method
and cased to safely bypass the utilities located in the upper 6 ft of the treatment
area. A total-station-based movement monitoring system was implemented to
monitor and protect utilities and adjacent structures. Vapor handling equipment with
carbon treatment was utilized to minimize odors from the grouting operation,
supported by real-time monitoring of air quality.

A 2-phase test program established the jet grouting parameters, column spacing,
and reagent addition rates to meet the project solidification goals. Completed test-
column locations were cored to verify effective column diameter, and sampled and
subjected to centrifuge testing to verify that the cement addition was sufficient to
immobilize contaminants within the soil-cement matrix.

This project represents the first time jet grouting has been used for the remediation
of an MGP site. The nature of the site has presented numerous challenges to the
project team including:

Utilities Numerous, highly sensitive, subsurface utilities including an antiquated,


but active 36-in natural gas main, 18-in high pressure natural gas force main, 6-in
water line, electrical duct bank, storm sewer lines, and overhead power lines.

441
Public Protection The site is located in a very congested, high-traffic area. Site
is in the heart of Kendall Square in downtown Cambridge Massachusetts
bordered by business offices, parking garages, and a public skating rink.

Project Coordination There were numerous entities involved in the project


including Public Utilities, City Departments, adjacent building owners, multiple
oversight engineers and other contractors. Permitting and coordination of all
interests involved has been paramount throughout the course of the project.

Site Logistics Maintaining a full-scale jet grouting operation including batch


plant, cement/bentonite deliveries, jet grouting rig, vacuum pre-drilling, and spoils
solidification and handling, on a 7500 ft 2 parcel with as little impact to the publics
day-to-day activities as possible proved to be a major challenge.

Geo-Con installed approximately 680 columns (approximately 7200 ft 2) to complete


the stabilization of the contaminated soil under Linskey Way. The columns were
approximately 18 ft deep and were grouted from the top of clay later at
approximately 18 ft to 13 ft below ground surface. Geo-Con used a cement-bentonite
grout mix to complete the jet grouting. Grouting was completed using Geo-Cons C-7
jet grout rig and jet pump.

442
Case Study 6: Former Camden Gas Works Site

Location: Camden, NJ

History: Gas works

Contaminants: PAHs, DNAPL

Scale: 24,000 yd3

Reagents: Portland cement & blast furnace slag

Method: In-situ

End-Use: N/A

443
Short Site History and Description

The former Camden Gas Works is a former MGP facility located on five parcels of
land (Parcels 1 through 5) comprising approximately 12.5 acres in Camden, New
Jersey. The parcels are bounded by Front, Second, Locust, Spruce, Cherry, Walnut,
and Chestnut Streets in the City of Camden.

This project addressed soils on Parcel 2 (approximately 1.5 acres in size), which is
currently an unused lot owned by the City of Camden. Historical information
indicated that Parcel 2 was used for MGP operations from 1891 to the late 1950s.
Former MGP structures on Parcel 2 included 2 former gas holders (Gas Holder Nos.
3 and 4), tar and oil storage tanks and appurtenances, a governor house/laboratory
and a valve house.

Risk drivers

A utility acquired the site from the City of Camden and planned to construct a 69-13
kV Class H substation on the property. The contaminated soils required
removal/treatment to prevent contact with the contaminated soil by workers and to
prevent migration of PAHs off-site.

Objectives

Removal and disposal of the unsaturated soil


Treatment of the saturated soil down to 30 ft bgs
Restore the site

The work performed by WRScompass consisted of the following activities:

Site preparation activities were performed that included pre-work topographic


surveying and existing conditions surveys of adjacent facilities, locating existing
utilities using a combination of utility locator services and pre-trenching the site
perimeter using soft dig techniques since the site was surrounded by public
streets with numerous known and suspect active and abandoned utilities,
installing erosion controls, removing the existing perimeter fence where necessary
and installing temporary fence and concrete barriers around the site perimeter,
setting up a support zone and contamination reduction zone, performing clearing

444
and grubbing, constructing a material staging area, and installing a 200 gallon/min
water treatment system.
Demolition activities were completed to remove existing above grade structures at
the site. Most notable was the demolition of a large steel gas holder and
demolition of two small structures (including the associated asbestos abatement).
Concurrently, any abandoned utilities identified via the trenching discussed above
were removed and capped and any concrete pads, structures, and other
obstructions were removed from the work areas for ISS. WRScompass also
removed the gas holder walls and floor slab and a notable quantity of wood piles
used to support these structures. All resulting concrete and wooden debris was
loaded out and disposed of off-site.
Steel sheeting was installed around the perimeter of the site once the pre-
trenching was completed to clear the sheeting alignment and once certain active
overhead utilities were relocated. The sheets were installed as permanent
sheeting to protect a fragile ductile iron gas main located along the south side of
the project site.
Excavation of contaminated soil for off-site disposal was performed in three areas
of concern. Area 1 was only excavated to a depth of 3 ft. Since the planned
construction of the site required imported fill with certain soil qualities to a depth of
12 ft below grade, Area 2 was excavated that depth. In Area 3, the ISS-treated
material was required to be no higher than 12 ft below grade after treatment. Area
3 was excavated to between 13 and 14 ft to accommodate the swell resulting from
ISS treatment. This minimized the disposal of the denser ISS-treated material.
Excavated soil was loaded into trucks and transported to the pre-approved
thermal disposal facilities.
ISS was performed in Area 3 to the required depth of 30 ft using an excavator-
mixing method. A grout plant was set up and the required reagent grout was
produced in an on-site batch plant then conveyed to the ISS treatment cell, where
it was added on a per weight basis using a pre-determined mix design of 9% by
weight for granulated blast furnace slag and 3% by weight Portland cement. ISS
was performed on over 24,000 cy of soil.
Once ISS was completed, the site was backfilled as required, using primarily an
engineered fill material with the properties specified for the ensuing substation
construction project. A 4-inch layer of crushed stone surfacing was installed above
the engineered fill. Backfill material was installed in controlled lifts and compacted.
Other site restoration activities included the removal of temporary fence and
barriers and installation of permanent fence where specified as well as
replacement of concrete sidewalks and curbing. All equipment and temporary
facilities were decontaminated and removed from the site and support areas were
restored.

Method

445
Portland cement and ground granulated blast furnace slag were used to chemically
immobilize the PAH contamination and to improve the unconfined compressive
strength to >50 psi and reduce the hydraulic conductivity to <1x10 -6 cm/sec.

Validation

Samples of the treated soil were obtained at a frequency of every 250 tons treated
and subject to SPLP, UCS and permeability testing. The ISS performance criteria
were : PAHs concentrations in the SPLP leachate less than 10 times the New Jersey
Class II Groundwater Quality Standards, UCS > 50 psi, and hydraulic conductivity
<1x10-6 cm/sec. All of the treated samples met the performance criteria on the first
pass.

Equipment

The ISS operations were conducted using an excavator and a batch plant. The
batch plant was used to prepare the reagent grout. The appropriate amount of water
was metered into an initial 5 yd 3 batch tank (equipped with a high-speed, high-shear
mixer) and recorded. The reagents (Portland cement and ground granulated blast
furnace slag) were transferred from the silos to the batch tank using the internal
screw conveyor to deliver the specified volume of reagent. Each reagent was added
separately to the mix tank and the scales on which the mix tank sets were tared
before each reagent was added to verify that the correct amount of reagent had
been added. When the correct grout composition was achieved, the blended grout
was transferred to the excavator. The batch plant was also equipped with a second
storage tank to allow for temporary storage of a blended batch to allow for
uninterrupted production of batches. A high speed mixer in this second tank was to
ensure the blended batch does not separate. The pre-determined grout volume was
pumped to the treatment area based on the soil density, reagent admixture ratio, and
the treatment cell dimensions.

Specific Issues

This was the first ISS project at an MGP site in New Jersey. WRScompass and the
client met with the NJ Department of Environmental Protection and negotiated
leachability performance criteria for the site. The performance standards were the
leachate criteria determined as part of the risk-based development of site-specific
impact to groundwater standards and were equal to 10 (default dilution-attenuation
factor (DAF)) times the Class II NJ Groundwater Quality Standards for the PAHs.

446
This manner of calculating ISS leachate performance criteria has now been used at
multiple sites in New Jersey, though the DAF has been increased to 20.

447
Case Study 7: Camp Pendleton Scrap Yard

Location: San Diego County, CA

History: Scrap yard

Contaminants: PAHs, PCP, arsenic, dioxins

55,000 tons
Scale:
Fly ash & activated carbon
Reagents:
Ex-situ
Method:
Restoration
End-Use:

448
Site description

Marine Corp Base Camp Pendleton Site 6 was previously referred to as the DPDO
scrap yard. The scrap yard operated from the early 1950s to 1979 as a storage,
processing, and disposal area for scrap metal, salvage items, hazardous materials,
and transformer fluids. The yard was divided into four areas:

Polychlorinated biphenyls (PCB) spreading area


Road burning area
Battery electrolyte disposal area
Hazardous waste drum storage area

Approximately 1,000 to 2,000 gallons of dielectric fluid from transformers was


reportedly spread in the area for dust control. Immediately east of this area,
flammable liquids such as fuels, solvents, and paint thinners were used as igniter
fluids for burning wood debris during the 1950s and 1960s.

The scope of work issued by the Navy included the following requirements pertaining
to remediation at Site 6:

Demolition and Excavation. Demolition was required to remove existing


pavements and light structures and clearing and grubbing of vegetation in the area of
contamination.

Soil Treatment System Site Development. Site development included


construction of the treatment systems concrete foundation; utility connections (i.e.
water and electrical); support facilities; and shelter structures. Adequate laydown,
storage, and equipment access areas were constructed to ensure efficient operation
of the treatment system.

Transportation. Transportation was required to haul contaminated soil from


excavation areas to the treatment system. Upon satisfactory treatment, treated soil
was transported to an on-base disposal site.

Soil Treatment System Pilot Test. Pilot testing with 250 tons of contaminated soil
from Site 6 was performed to optimize the treatment system.

449
Operation of the Soil Treatment System. The soil treatment system was being
operated to remediate contaminated soil from Site 6 to achieve nondetectable values
for pesticides, PAHs, PCBs, and dioxins in the leachate from the Synthetic
Precipitation Leaching Procedure (SPLP). OHM treated soils with a PORTEC Pug
Mill Model 53 using Carbon and Type C fly ash to meet required specifications.

Sampling and Analysis. Confirmation sampling was performed to verify that all
contaminated soil was removed from Site 6 and remaining soils met or exceeded
cleanup criteria.

Disposal. Prior to disposal, the Navy approved an on-base disposal location which
was prepared by clearing an appropriately sized area, excavating that area to a
uniform depth below existing grade, and stockpiling those excavated soils for later
use as a cover material for the treated waste. The treated waste was placed in the
excavation and covered with a minimum of 1 ft of the stockpiled material.

Soil Treatment System Demobilization. Upon completion of soil treatment


operations, all equipment and temporary facilities were removed from the site.

Site Restoration. Site 6 was restored to accommodate a wetlands habitat.

Risk drivers for the remediation

A preliminary human health risk assessment and ecological risk assessment was
conducted and indicated that chemicals of concern at Site 6 included:

Antimony
Arsenic
Beryllium
Chromium (as Cr VI)
Zinc
dichloro-diphenyl-trichloroethane (DDT)
DDE
DDD
450
Arochlor 1260 (PCB)
Polyaromatic hydrocarbons (PAH)
OCDD (dioxin).

Objectives

Remove the contaminated soil


Treat to reduce the leachability of the contaminants and produced a high strength,
low permeable treated material
Place and compact the treated material into the excavation area
Restore the site

Method

The contaminated soil was treated with a mixture of Class C fly ash, and activated
carbon to produce a treated material with nondetectable values for pesticides, PAHs,
PCBs, and dioxins in the leachate from the Synthetic Precipitation Leaching
Procedure (SPLP).

Validation

Temporary 500 ton stockpiles of the treated soil were created. Confirmatory testing
for the leachability of pesticides, PAHs, PCBs, and dioxins was performed on the
sample from each temporary stockpile. All of the confirmatory samples met the
performance requirements for leachability.

Equipment

The stabilization treatment involved the use of a pug-mill system to mix the soil with
Class C fly ash activated carbon. OHM's pug-mill system consisted of a variety of
feeders, conveyors, silos, and a pug-mill mixer integrated into a complete system for
the continuous mixing of wastes and reagents. The screened material was fed to an
8-yd3 feed hopper. The hydraulically-driven belt, located in the bottom of the feed
hopper, fed the material onto a 40-ft long by 24-in wide belt conveyor. The conveyor
belt was equipped with a Ramsey single idler belt scale for the accurate, real-time
determination of the rate, in tons per hour, of soil being treated. The conveyor belt
conveyed the material into the pug-mill for blending with the stabilization additives.
The pug-mill mix box was 4 ft wide by 3 ft high by 10 ft long. Paddles were bolted

451
onto structural steel shafts with replaceable shafts flanged on both ends for ease of
maintenance. The mixer was V-belt driven by a 125-hp motor.

Dry stabilization additives were stored on-site in vertical cement silos. The silos
were self-leveling and have a capacity of 200 barrels of material. The silos were
also equipped with a top mount baghouse for dust control during silo filling. The feed
from each silo was controlled by a 14-in diameter screw auger powered by a 5-hp
motor. The dry stabilization additives were introduced from silo feeders which are
attached to the pug-mill through a central feed auger. The soil was mixed with the
reagents and water inside the pug-mill. The pug-mill system was operated 12 hours
per day, 5 day per week, treating 1,200 tons of soil per day.

452
Case Study 8: Chevron Pollard Landfill Site

Location: Richmond, CA

History: Landfill site

Contaminants: Pb, H2S, sulphuric acid

150,000 yd3
Scale:
Portland cement & limestone
Reagents:
Ex-situ
Method:
Restoration
End-Use:

453
Site description

OHM developed, designed, and operated an on-site neutralization and stabilization


process to treat 150,000 yd3 (more than double the amount originally slated for
stabilization) of waste at a former landfill located at a major northern California
refinery. The landfill was located immediately adjacent to Richmond Bay.

The waste consisted primarily of the following three materials:

Acid Sludge Tar (AST): A layer of soft, acidic tar produced as a sulfonated
lubrication oil by-product.

Dredged Bay Mud (DBM): A layer of natural, dredged soil originally placed on top
of the AST material. Over time, the DBM settled below the AST and transferred
residual sulfuric acid from the AST to the DBM.

Interface Layer: This was the most variable of the three wastes in its chemical and
physical characteristics, as there was little correlation between the vertical thickness
of the mixed layer, the relative amounts of AST or DBM, and either total depth or
location.

454
Characterisation

The AST was the waste by-product of sulfuric acid treatment of refinery stocks. The
material had a pH range of 0.5 to 1.5 and a total acidity of 12% to 25%. The DBM
was dredged material that was placed in the landfill as a matter of convenience and
had become commingled with the AST. Substantial portions of all three waste
groups were considered hazardous due to their corrosive characteristics. The DBM
and the interface layer contained significant levels of sulfur compounds, which were
a health and safety concern during excavation and processing. Additionally, the
surface of the landfill was unstable and would not support heavy equipment.

Objectives

Remove the AST and DBM from the landfill area


Neutralize the AST
Solidify the DBM and neutralized AST to produce a suitable structural fill material
Place and compact the treated material on an adjacent parcel as structural fill
Backfill the landfill area

455
Method

A 2-step neutralization/solidification process was utilized. Limestone was used to


neutralize the AST. The DBM and neutralized AST were solidified using Portland
cement.

Validation

Samples of the treated soil were obtained at a frequency of every 100 tons treated.
The samples were tested to ensure that the treated material met:

Final pH 5 to 12
96-Hour Bioassay Test pass
Excess Alkalinity greater than 1 wt%
Geotechnical less than 1% creep in 48 hours at 500 psf applied pressure
Particle size greater than 97% passing a No. 4 mesh
Paint Filter Test pass
TCLP RCRA non-hazardous

Over 96% of the treated samples met the performance criteria on the first pass.
Those 100 ton stockpiles not meeting the performance criteria were retreated.

The client also implemented a rigorous QA/QC program on the stabilized material
that included on-site testing for pH, paint filter test, particle size, and creep/bearing.

Equipment

The stabilization process started with the excavation and transport of the DBM and
AST from the landfill. The DBM was pumped from the landfill to a holding tank,
which was designed to hold enough DBM for 2 days of processing. The AST was
transported by truck to a clay pit, which also held a 2-day supply of material. The
AST was fed to a hopper that continuously supplied a conveyor. The AST was fed
from the conveyor to a crusher, designed to accept sticky, plastic material. Crushed
rock was fed continuously into the crusher with the AST to act as an abrasive agent
to keep the crusher clear. The crusher produced an AST product that was 97%
passing the No. 4 screen. This material was then fed into a pug-mill. In the pug-mill,
the AST was mixed with limestone at a ratio of approximately 33% of the total
acidity. This resulted in AST particles encapsulated with limestone to meet the
456
neutralization requirements. The treated AST material was conveyed to a second
pug-mill unit.

Concurrent with AST treatment, DBM was pumped over a shaker screen to remove
any AST particles greater than No. 4 mesh in size. The screened-out AST particles
fell into the AST day pit. The DBM was fed into a mixing tank for equalization. The
final stabilization step was performed in the second pug-mill. DBM, treated AST, and
cement were mixed in the second pug-sealer and the final stabilized product was
transported to a second landfill. The stabilized material was spread and allowed to
cure. Once the material had cured, it was compacted to 90% standard proctor
density. The total system throughout was approximately 60 tons per hour.

457
Case Study 9: Carnegie

Location: Carnegie, Pennsylvania

History: Natural soil

Contaminants: Saturated Unconsolidated Soils

7,134 yd3
Scale:
Portland cement
Reagents:
In-situ
Method:
N/A
End-Use:

458
Summary

Geo-Con was contracted by Frank J. Zottola Construction (Zottola) to solidify 7,134


yd3 of organic, saturated, unconsolidated soils at the site of the future Heidelberg
Lowes Home Improvement store in Carnegie, Pennsylvania.

The problematic soils were located under a crucial corner of the structure foundation
that would eventually house the electrical service for the building. The soils extended
from the existing ground surface to a clay/till-confining layer located as deep as 18 ft
below ground surface. The difficultly of the project was increased as the majority of
the unsuitable soil was beneath the groundwater table.

The objective of the project was to solidify the unsuitable soils beneath the site and
provide a suitable foundation for building floors, roadways and parking areas.

459
Case Study 10: Columbus

Location: Columbus, Indiana

History: Wood treating plant

Contaminants: Creosote

4,500 yd3
Scale:
Cement & activated carbon
Reagents:
In-situ
Method:
N/A
End-Use:

Summary

This project was performed at the former


Columbus wood treating plant in Columbus,
IN.

In-Situ Stabilization (ISS) work was


performed on soils that were impacted by
creosote. The treatment zone covered
approximately 12,000 ft2 of surface area
with depths ranging from 3 to 17 ft.

The project goal was to mix the


contaminated soils in-place with a cement
and powered activated carbon grout, with
the subsequent mixture meeting the
treatment criteria of a minimum 50 psi
unconfined compressive strength @ 14
days. Laboratory testing indicated that all of
the soils mixed columns were in compliance

460
with the treatment goals.

Geo-Solutions utilized state-of-the-art equipment and materials including soil mixing


using a rig with a 9 ft diameter mixing tool, automated grout mixing plant, and all
required carbon materials.

Quality control was the responsibility of Geo-Solutions and was largely provided
through the electronic monitoring of the Soil Mixing rig and supplementary daily grout
quality control reports, which documented that the specified carbon and cement
addition rates were applied.

461
Case Study 11: CSX Benton Harbor Scrapyard

Location: Benton Harbor, MI

History: Battery recycling

Contaminants: Lead

99,800 tons
Scale:
Reducing agent + Alkaline Reagent
Reagents:
Ex-situ
Method:
Restoration
End-Use:

Site Description

This 7-acre site was adjacent to a CSX switchyard and the Paw-Paw River near
downtown Benton Harbor, MI. Contamination at the site resulted from scrapped
metal processed for resale and recycled batteries, from 1959 to the mid-1980s.
Scrap metal and battery casings were used to fill in low-lying areas of the site. Lead
was the primary contaminant found at the site in soil.

462
Characterisation

The waste as a combination of soil, battery casings, scrap metal and debris used to
fill wetlands on the site.

Drivers

The site and adjacent areas were identified for redevelopment, and the contaminated
soil was required to be removed from the site. Treatment of the excavated material
was necessary to allow disposal in a municipal landfill.

Objectives

Remove the contaminated soil


Treat (if applicable) the soil to remove its RCRA hazardous characteristic D008
(lead)
Dispose of the soil at an off-site landfill
Restore the site

Method

The excavated contaminated soil and debris were mixed with Portland cement to
immobilize the lead. Each soil stockpile was sized to require one load (nominally 25
tons) of Portland cement.

An excavator was used to mix the contaminated soil and debris with the Portland
cement, normally over 2 hours, until visibly homogeneous. Sampling and analysis in
support of off-site disposal of the treated soils as non-hazardous waste/soils was
performed.

Once the analysis confirmed full-treatment, the treated soil was loaded onto trucks
and transported to licensed disposal facility for disposal as non-hazardous
waste/soils. Loaders removed soil from the designated final storage bins and
transferred the soil into the transport trucks. Load-out skirts contained any soil that
may be spilled during load-out.
463
An excavator was used to mix the contaminated soil and debris with the Portland
cement.

Validation

A 5-point composite sample was obtained from each stockpile after treatment. The
samples of the treated soil were subjected to TCLP testing for metals. Over 95% of
the treated samples met the performance criteria of less than 0.75 mg/L TCLP-
leachable lead on the first pass. Those that did not were retreated until passing.

Issues

The contaminated material continued below the groundwater level. Since the site
was adjacent to the Paw Paw River, dewatering was not possible and the excavation
had to occur in the wet. Visible confirmation of the removal was required. Backfilling
also occurred below the groundwater level.

464
Case Study 12: Double Eagle Refinery Site

Location: Muskogee, OK

History: Refinery

Contaminants: PAHs, TPHs, hydrogen sulfide

60,000 yd3
Scale:
Portland cement & fly ash
Reagents:
In-situ
Method:
Restoration
End-Use:

Site description

Impoundments at the Double Eagle Refinery in Muskogee, Oklahoma, contained


acid sludge tars. These tars, a byproduct of the sulfuric acid cracking of lube oils and
asphaltenes have a low pH (<2) and contain hydrogen sulfide. The recommended
remedial action was neutralization and solidification of these acid sludge tars.

Risk drivers

Residential housing was located at the limits of the refinery property. The
remediation had to be conducted without perimeter air exceedences or odour
complaints from the nearby residents.

Objectives

Treat (if applicable) the tar to increase its pH and improve its physical properties
Cap the treated material
Restore the site

465
The full-scale neutralization was accomplished as a 2-step process. In the first step,
the fly ash was added to the acid sludge tar and homogenized thoroughly using an
excavator. The minimal area of acid sludge tar was disturbed during the
neutralization and the neutralized material was covered with Rusmar foam, to
minimize odour and hydrogen sulfide emissions. In addition, a perimeter misting line
was suspended from existing power poles and an odour-neutralizing agent was
misted to prevent any odour emissions to nearby residents. The next day, the
neutralized material was mixed with the Portland cement and recovered with Rusmar
foam. After 5 days, the excavator could operate on top of the neutralized and
solidified material, allowing the work to progress from the circumference of each
impoundment out toward the middle. The neutralized and solidified material was
capped with fill, clean soils and topsoil. Final grade was established to allow proper
drainage over the impoundments and the area was seeded.

The combination of the 2-step treatment and the use of odor/vapour suppression
allowed WRScompass personnel to complete the impoundment solidification and
capping work in a cost effective manner and within the designated schedule. Just as
important, the odour and vapour suppression systems employed eliminated any
perimeter air quality exceedences and complaints from nearby residents

Method

The tar was neutralized using Class C fly ash. The neutralized material was
solidified using Portland cement.

Validation

Samples of the treated soil were obtained at a frequency of every 250 yd 3 treated.
The treated material was tested to meet pH (>7), hydrogen sulphide (<5 ppm) and
unconfined compressive strength (>50 psi at 28 days). All of the treated samples
met the performance criteria on the first pass.

Equipment

An excavator was used for both the neutralization and solidification steps.

466
Case Study 13: Dundalk

Location: Baltimore, Maryland

History: Natural soil

Contaminants: Chromium ore processed residues

350 lineal feet


Scale:
Peat moss/bentonite slurry
Reagents:
In-situ
Method:
Marine terminal
End-Use:

467
Summary

Geo-Con was contracted by CH2M Hill Constructors, Inc. to install approximately


433 lineal feet of Augered Strain Relief Trench (SRT) at the 1800 Area of the
Dundalk Marine Terminals in Baltimore, Maryland.

Chromium Ore Processed Residue (COPR) was used extensively as backfill for the
Marine Terminals during their construction. During long-term exposure to water, the
COPR exhibited significant expansion, resulting in problematic movement of the
Marine Terminal soils and damage to utilities and surface structures. The five
segments of SRT were intended to relieve lateral ground stress and provide strain
relief by permitting lateral expansion of subsurface layers of COPR.

The SRT was installed as a series of overlapping 2 ft diameter soil-mix columns,


placed on 1.4 ft centers to provide the specified continuity and effective trench
thickness.

The in-place soils were mixed with horticultural-grade peat moss at a rate of 8-10%
by weight of soil. The peat moss was slurried in a 6% bentonite/water solution in
Geo-Cons custom high-speed/high-shear batch plant and delivered through the
mixing rigs wet-kelly system and mixed with the in-place soils using Geo-Cons
custom made Shallow Soil Mixing (SSM) auger. The columns ranged from 10 to 15 ft
in depth.

The SRT was part of an on-going Pilot Study at the marine terminals. The project
was completed within budget and with no health and safety related incidents.

Geo-Con was also able to accelerate the SRT installation schedule and provide the
client a valuable schedule gain.

468
Case Study 14: East Rutherford

Location: East Rutherford, New Jersey

History: Petrochemical plant

Contaminants: TCE

7,626 yd3
Scale:
Potassium permanganate & Portland
Reagents:
cement

In-situ
Method:

469
Summary

The purpose of this project was to use single auger soil mixing to treat TCE
contaminated soils, using a 2-step approach involving: 1) In-situ oxidation with
Potassium Permanganate (PP), and 2) In-situ solidification with Portland Cement
(PC).

Geo-Solutions (GSI) was contracted by ERM to undertake soil mixing, site


preparation and clean-up and excavation, including stockpiling of 1500 cy (1148 CM)
of clean overburden, installation of 14 trench bedding plugs (to minimize PP
migration off-site), erosion and sediment control, installation of monitoring wells,
utility abandonment and site landscaping.

Soil mixing involved the installation of 242 soil columns using a Delmag RH-18 drill
rig, of 9 ft (2.75 m) diameter to depths of 17 ft-19 ft (5.2 m 5.8 m) below work-pad
elevation. A column layout for the soil mixing work is shown below:

In the course of the soil mixing work, GSI encountered over fifty 14 (356 mm)
diameter deep foundation concrete piles. Obstructed soil columns were isolated
and the obstructions excavated to the maximum possible depth, followed by re-
treatment. Pictures of the obstruction removal and obstructions are shown below.
This approach worked as planned and was executed without significant delay to the
original schedule.

470
471
Case Study 15: Foote Minerals Superfund Site

Location: Malvern, PA

History: Lithium production

Contaminants: Lithium tailings

220,000 yd3
Scale:
Portland Cement & blast furnace slag
Reagents:
In-situ
Method:
Restoration
End-Use:

Site description

The Foote Mineral Co. site is


located on a 79-acre property in
East Whiteland Township, PA.
Starting in 1941, the Foote
Company operated a variety of
process buildings for the
manufacture of lithium metal and
lithium chemicals and inorganic
fluxes for the metal industry.
Ores and minerals were also
crushed and sized there. When
the plant closed in 1991, the site
had two quarries, a pit used to
burn solvents, a lined basin and
more than 50 buildings and
process area.

The final site remedy included 1)


excavation and off-site removal of
radiation-contaminated soils; 2)
stabilization of process tailings

472
located in one of the two on-site quarries; 3) consolidation of other site wastes,
debris and contaminated soils into the quarry area; 4) capping of the quarry area at
the surrounding grade. WRScompass was contracted for the ISS of the process
tailing in the on-site quarry.

Risk Drivers

Several nearby private wells were found to be impacted by contamination and a


public water supply well about a mile downhill from the site had been shut down due
to contamination.

Objectives

Approximately 220,000 yd3 of lithium tailings were stabilized in-situ. The overburden
material from the ISS areas was excavated and hauled to the soil management area.
Once the pre-ISS excavation elevations were created, a grout mix consisting of
ground granular blast furnace slag (GGBFS) and water was mixed with the tailings.
The mixing depth was up to 70 ft with an average depth of 40 ft.

Prior to treated soil curing, representative samples were collected and placed into
appropriate molds for quality control tests. Molded specimens were cured for the
prescribed duration, then removed from their molds and tested.

Method

Portland cement and ground granulated blast furnace slag were used to improve the
unconfined compressive strength to >50 psi and reduce the hydraulic conductivity to
<1x10-6 cm/sec.

Validation

Prior to treated soil curing, representative samples were collected at a frequency of 1


sample per 500 yd3 treated and placed into appropriate molds for quality control
tests. Molded specimens were cured for the prescribed duration, then removed from
their molds and tested for UCS and permeability testing. Over 98% of the treated
473
samples met the performance criteria on the first pass. Those columns not meeting
the performance criteria were retreated.

Equipment

The ISS operations within the site were conducted using a 4000-series Manitowoc
crane equipped with an attached Hain Platform. The crane/platform assembly is
supplemented with a swivel-mounted, top-feeding Kelly Bar capable of reaching a
depth of 75 ft bgs. Augers were attached to the bottom of the Kelly Bar. A 9 ft
diameter auger was utilized on this project.

A batch plant was used to blend prescribed proportions of reagent and water to
produce a grout. This grout was pumped from the batch plant to the auger and
mixed with the soil. Depending on the soils physical properties, an excavator was
used to loosen the soil and remove debris before treatment.

The tailings were treated by mixing grout into overlapping columns of soil. A
prescribed quantity of grout was mixed into each column. Each column of soil was
treated using the following procedure:

When work began each morning and again whenever the grout storage/mixing
tanks were empty, at least one batch of grout was prepared to create a small
reservoir of grout for the work.
The horizontal coordinates of each column were located with surveying equipment
and marked using grade stakes. Each column top elevation was also marked with
the specified treatment depth. The required grout for each column was determined
based upon the specified mix design and column volume.
The mixing drill rig was initially positioned atop the column to be treated. An
operator lowered the Kelly bar and auger system through the soil and mixed the
grout and soil in-situ to the required depth for three passes per column.
Immediately after the soil was mixed with grout, samples of the grouted soil were
collected and placed into appropriate molds for testing.

After completing one column, the crane crawled to treat another column, with
adjacent columns overlapped to ensure complete coverage. This process was
repeated until a strip of soil was treated. Actual locations of the panels were field
verified and documented.

474
After completing ISS work, equipment that contacted contaminated soil was
decontaminated and demobilized.

475
Case Study 16: Ghent

Location: Ghent, Belgium

History: Chemical/metallurgical industry

Contaminants: Metals, cyanides

135,000 tonnes
Scale:
Zero-valent iron, cement/slag blend
Reagents:
Ex-situ
Method:
Not specified
End-Use:

Site Description

476
The company La Floridienne was located in the port of Ghent (Belgium) and was
active in the first half of the 20th century, producing metal salts and iron cyanide
pigments from residues collected from other chemical and metallurgical industries.

The site was abandoned and contaminated, apart from the former production area,
where two big stockpiles of production residue were found, called the red waste and
the grey waste.

Characterisation

The so-called red waste consisted of about 60,000 tons of roasted pyrite ashes.
Consequently the material was very acidic. Its texture was silty but was not cohesive.
The grey waste consisted of 75,000 tons of very cohesive alkaline gypsum
precipitate from water treatment operations employing excess lime. Both
composition and leachability are outlined in Tables 1 and 2, below.

Table 1: Composition range of the red and grey waste materials


Red waste tip Grey waste tip
pH 3.8 12.5
As (mg/kg DM) 35,000 40,000 500 1000
Pb (mg/kg DM) 2000 10,000 10,000 25,000
Zn (mg/kg DM) 500 5000 500 2000
Cyanides (mg/kg DM) 100 - 500 500 8000

Table 2: Leaching of the red and grey waste (EN 12457-4 leaching test)
Red waste tip Grey waste tip
pH 3.8 12.5
As (mg/l) 68 0.01 0.05
Cd (mg/l) 0.1 -1 n.d.
Cu (mg/l) 0.1 8 n.d.
Pb (mg/l) 0.1 0.5 0.1 10
Zn (mg/l) 0.1 0.5 0.05 0.5
477
Cyanides (mg/l) 0.01 0.1 10 300
n.d ~ no data

Objectives

Due to the large volume of material to be treated, the option to keep the material on-
site was preferred. Therefore, the wastes had to be treated to reduce leaching (and
in order to fulfil the European Waste Acceptance Criteria for hazardous waste
landfills) stored above groundwater and underneath an HDPE liner, to prevent rain
percolation.

The chemical (leaching) objectives are given in Table 3, below:

Table 3: Leaching criteria for hazardous waste landfills in Flanders (EN 12457-
4 leaching test)
Maximum leachability (mg/l)
pH 4 13
As 1
Cd 0.5
Cu 10
Pb 2
Zn 10
Cyanides 1

As the site would later serve as a container terminal, the S/S material had to a high
bearing capacity, i.e. a compressibility modulus of 17 MPa as measured in a plate
test. Specific durability criteria were not pre-defined, but were provided later, based
on a theoretical basis.

Drivers

Serious groundwater pollution from arsenic and cyanides resulted from the leaching
of both waste piles. A first step in the site remediation was, therefore, the elimination

478
of these piles as a pollution source, followed by the pumping-and-treating of
impacted groundwater.

Method

The acidic red waste had sufficient geotechnical properties, and only a reduction of
arsenic leaching was required. This was achieved by adding 1 w/w% of zero valent
iron. An alternative approach, based on cement-based solidification also decreased
arsenic leaching, but strongly increased the release of cyanide to unacceptable
levels.

The alkaline grey waste had to be geotechnically improved, so a cementitious


binder was required, but the use of cement and lime was not possible, and a strong
impermeable matrix was achieved by adding 40 w/w% of a binder consisting of
Portland cement and GGBS (ground granulated blast furnace slag). An interesting
chemical stabilization of cyanide in the form of Prussian Blue, was also investigated
but did not work sufficiently.

The whole project had to be carried out in a very short time-frame of only 8 months.
Therefore, both wastes were mixed in parallel in two different plants, each with a
throughput of around 150 tons an hour.

The non-cohesive red waste was mixed with the iron powder by means of a
continuous liming mixer (pictured left). The cohesive grey waste was mixed in a
batch concrete mixing plant (double axis pug-mill system, pictured right).

479
Challenges

Dust control was extremely important, since the waste materials on-site were highly
toxic, and the approach taken involved suppression spraying, using a minimum
amount of water.

Validation

Both treated materials were validated after at least one month after treatment, by
means of the EN 12457 4 leaching test, and checked against the criteria in Table 3
above.

480
Case Study 17: Guernsey

Location: Guernsey, Channel Islands, UK

History: Harbour (ship maintenance)

Contaminants: TBT (tributyltin)

25,000 m3
Scale:
Bio-char based reagent
Reagents:
Dredging & ex-situ
Method:
Marina
End-Use:

481
Site Description

The Port of St. Sampsons in the channel island of Guernsey was to be upgraded to a
yacht marina.

The dry (at low tide) inner harbour required dredging by dry earth moving equipment
followed by the construction of a sill and gate across the harbour entrance to
impound water at + 4,5m above CD. Dredging was to +3,5m above CD in order to
maintain a constant water level of at least 1 m within the harbour.

The work involved the removal of 25000 m3 of sediments contaminated with TBT
(tributyltin) at concentrations between 640 g/kg and 1770 g/kg, making sea-
dumping of these sediments impossible. The TBT resulted from ship hull
maintenance, primarily sand blasting and painting, and work was successfully
completed in 2003.

Characterisation

The sediments in the harbour contained a high percentage of sand and very little
organic material so they were very prone to leaching (average concentrations in
leachate were 1.8 g/l).

Risk Drivers

Since the sandy material was suitable for land reclamation south of the port, the
main risk involving re-use was the leaching of TBT to the groundwater.

Objectives

As no standards for TBT leaching existed at the time, a target of non-detectable


levels of TBT (< 0,005 g/l) in the leachate was preferred by the authorities.

Method

482
During the remedial design process, various laboratory-scale tests were carried out
to determine the appropriate additive for immobilising the TBT. Some experience
was already gained from testing of TBT contaminated sediments from the port of
Zeebrugge, Belgium (Table 1).

The latter showed that stabilization with cement is not working at all, as expected
from literature as TBT becomes very soluble at high pH. A commercial product, E-
clays (Envirotreat, UK), although being on previous TBT stabilization projects, did not
show any positive result.

The proprietary product Organodec, which is based on biochar, gave promising


results, but when combined with cement the positive effect was diluted (Table 2). It
was therefore decided to only use the addition of 2 % of Organodec in the Guernsey
project.

The sediment was treated by means of a rotary mixing bucket (brand Allu) mounted
on an excavator.

The production rate was about 30 tons an hour, and dosing was done by spreading
500 kg of Organodec (after soaking to prevent dust) over a batch of 16 m (about 25
tons) of sediments. Mixing was carried out 3 times to ensure a sufficiently uniform
distribution of additive.

483
Table 1. Lab scale stabilization tests on Zeebrugge sediment.

TBT leachablity of
Initial TBT treated material
Additives
leachability (g/l)
(g/l)

4.5 5% OPC2 105.08

4.5 5% E-clays3 4.4

4.5 5% E-clays + 1,752.09


12%OPC

4.5 2% OrganoDEC1 0.11

1
OrganoDEC = patented additive; 2 OPC = Ordinary Portland Cement; 3 E-clay =
environmental clay

Table 2. Lab scale stabilization tests on Guernsey sediment.

TBT leachablity of
Initial TBT treated material
Additive
leachability (g/l)
(g/l)

1.8 2% OrganoDEC1 <0.005

1.8 2% OrganoDEC+2% 0.55


OPC2

1.8 5% OrganoDEC+5% 1.6


OPC

1
OrganoDEC = patented additive; 2 OPC = Ordinary Portland Cement

Validation

484
Every batch of treated sediment (representing about 200 tons) was sampled and
tested according to EN 12457-4, with the majority of results below the detection limit
of 0.005 g/l.

485
Case Study 18: Hercules 009 Landfill Site

Location: Brunswick, GA

History: Chemical waste landfill

Contaminants: Toxaphene

88,150 yd3
Scale:
Portland Cement
Reagents:
In-situ
Method:
Restoration
End-Use:

Site description

The Hercules 009 Landfill Superfund Site is located on 16.5 acres near the City of
Brunswick, Georgia. The property was used as a borrow pit during construction of
Georgia State Highway 25 (Spur 25), which borders the property on the west.
Hercules Incorporated was issued a permit in 1975 to use 7 acres at the northern
486
end of the site, known as the 009 Landfill, to dispose of waste from the production of
an agricultural pesticide, toxaphene.

Hercules Incorporated began producing toxaphene in 1948 and continued production


through 1980. Toxaphene was one of the most heavily used insecticides in the
United States until 1982, when the United States Environmental Protection Agency
(EPA) cancelled the registrations of toxaphene for most uses. A registration is a
license allowing a pesticide product to be sold and distributed for specific uses in
accordance with specific use instructions, precautions, and other terms and
conditions. All uses of toxaphene were banned by the EPA in 1990.

Between 1975 and 1980, Hercules Incorporated operated the 009 Landfill under a
permit issued by the Georgia Environmental Protection Division (GEPD). The permit
allowed the Brunswick, Georgia, Hercules plant to dispose of waste water sludge
from the production of toxaphene. Part of the Hercules 009 Landfill was also used for
disposing empty toxaphene drums and toxaphene-contaminated glassware, rubble,
and trash. The landfill was constructed as 6 cells divided by subsurface berms
reportedly lined with a soil-bentonite clay mixture across the bottom and along the
bermed walls. The landfill was closed in 1983. The EPA added the Hercules 009
Landfill to the Superfund National Priority List in 1984. Hercules designed and
implemented the remedy for the site.

Objectives

Treat the landfill contents and soil to remove its RCRA hazardous characteristic
for toxaphene
Cap the treated material
Restore the site

Method

The remedy included Portland cement-based solidification/stabilization treatment.


The contents of the landfill were treated by S/S while the material remained in-situ.

The performance standards for the S/S treated material were at least 50 psi (0.34
MPa) unconfined compressive strength (UCS) and toxaphene leaching of less than
0.5 mg/L, as determined using the Toxicity Characteristic Leaching Procedure
(TCLP), on 28-day cured samples.

487
Validation

During treatment composite samples were made from S/S-treated material collected
at one third and two thirds the total treatment depth of each subcell. Cement addition
rates were verified based on written logs used for each subcell, which recorded
subcell volume and weight of cement used. The average dosage of Portland cement
applied to the 009 Landfill materials was approximately 14.8%, judged within project
tolerances. UCS was determined by pocket penetrometer on composite samples.
Selected cylinders made from treated material from the remedial action start-up
period were tested for UCS by ASTM D 2166 in order to correlate pocket
penetrometer results to UCS. Compressive strength of the treated material
increased over time and generally exceeded the 50 psi (0.34 MPa) requirement
within 3 to 5 days cure time. TCLP testing was conducted on composite samples of
blocks of four subcells. TCLP testing never revealed any presence of toxaphene in
the leachate of the composite samples tested.

Equipment

An excavator was used to mix the landfill contents with the Portland cement. The
landfill was divided into 25 x 25-ft (7.6 x 7.6-m) square cells for treatment. The total
wet weight of the untreated soil and sludge in the cell was determined using a
density of 100 lb/ft3 (1600 kg/m3) and the depth of untreated soil or sludge in the cell.
Based on the total wet weight of the soil and sludge in the cell, the amount of
Portland cement required for treatment was calculated. Up to 6 subcells were treated
at one time. The remedial action contractor used an excavator to mix dry cement into
the contaminated material while the material remained in place. Water for hydration
of the cement was added as needed. Records were kept including depths of
treatment of the as-treated (as-built) subcells. The depth of treatment extended
below the bottom of the landfill contents sludge zone adding to the total volume of
material to be treated. The treated depth of the majority of subcells extended into the
regional groundwater table.

The remedial action construction was completed by regrading and revegetating the
site. The primary intent of this activity was to establish an adequate vegetative cover
over the soil-cement cap, the stabilized landfill contents, and other disturbed areas of
the site resulting from remedial action activities. Rough grading involved adding
some selected fill from a nearby borrow area. During this fill placement, these areas
were rough graded and compacted to promote positive drainage. A vegetative cap
was placed on top of the graded area comprised of 6-in of loose fill, which was
fertilized and seeded.

488
Case Study 19: Hoedhaar Lokeren

Location: Hoedhaar Lockeren, Belgium

History: Hat-making industry

Contaminants: Mercury

10,000 tonnes
Scale:
Iron-based reagent
Reagents:
Ex-situ
Method:
Redevelopment/residential
End-Use:

Site Description

During two centuries the city of Lokeren in Belgium was known for its production of
high quality hats, made from rabbit fur felt. In addition to many hat-making
489
workshops, the city had more than 40 felt production sites, using fur treatment by
mercury nitrate brine.

Soil and groundwater pollution resulted from fur processing and the largest of these
facilities in Lokeren was Hoedhaar (Dutch for Hat-hair) and was situated near the
city centre. When the City Council decided to re-develop the site as a residential
area, the controlled demolition of the contaminated buildings, the removal of
asbestos and soil from the site and excavation of a contaminated brook took place.
From the 35,000 tonnes of soil excavated, 15,000 tonnes was soil-washed off-site,
and 10,000 tonnes were chemically stabilized to immobilize mercury, prior to
disposal to landfill. A further 10,000 tonnes were sent directly to landfill. The
groundwater was pumped and treated by a physico-chemical treatment plant
including microfiltration. The remediation took place from 2010 until 2012.

Characterisation

The soil consisted of loamy sand, contaminated with mercury, up to concentrations


of 1000 mg/kg dry matter (DM); however, the average concentrations were between
100 and 300 mg/kg DM. As the soil was low in both clay and organic matter, mercury
was highly leached, ranging between 0 and 40 mg/kg DM.

Risk Drivers

The presence of mercury in old buildings, soil and groundwater was the driver for the
remediation of this derelict site.

Objectives

The site-specific remediation target was 120 mg/kg DM for mercury. This level is
very high for a residential area, and so the site was covered with clean topsoil.

The target for the stabilization of the soil was a residual leachability of 1 mg/kg DM of
mercury, a criterion also used at the landfill where the soil was disposed of.

490
Method

Various chemical stabilization trials, targeted at known mercury speciation (e.g. as a


sulphide) were carried out, but without much success. A new approach employing
zero-valent iron and ferric hydroxides was successful, and was limited to 5 %.

As the amount of soil was limited, the use of specialised mixing plant could not be
justified. Therefore, a simple mixing approach using bucket mixing (see images
above) was used. The spreading of soil, additive dosing (one powder and one liquid),
was followed by excavator mixing. During early stages of the work the quality of
mixing, which was proportional to mixing time, was improved.

Validation

Batches of treated soil (representing about 250 tons) were sampled and tested
according to EN 12457-4.

491
Case Study 20: Irving

Location: Irving, Texas

History: Landfill

Contaminants: Mercury

744 linear feet


Scale:
Slurried bentonite
Reagents:
In-situ
Method:
N/A
End-Use:

492
Site Description

Geo-Con was contracted by The City of Irving, Texas, to install a Soil Mixing Barrier
Wall that tied into the existing slurry walls at the Hunter Ferrell Landfill. The soil-mix
barrier wall was overlapped into the existing slurry walls on the Middle and East
Tract cells. The site presented unstable/flowing sands and barrier wall installation
necessitated using soil mixing, as traditional slurry-trenching was not possible.

Objectives

The barrier wall was installed to complete the seepage barrier along the south side
of the future Middle Tract of the landfill. The work also included berm construction
over the barrier wall alignment, to provide 25-year flood prevention.

Method

The length of this soil mixing wall was 744 linear feet with an average depth of 45.1 ft
for a total of 8427.5 vertical ft2. The average depth to top of the shale key-in layer
was 40 ft.

The Barrier Wall was mixed with a Delmag RH20 Drill Rig with a 4-ft diameter soil
mixing auger attached. The overlapping 4ft diameter columns provided a minimum
36 inch thick wall. Geo-Con keyed into the underlying shale layer approximately 6 ft
and no less than 2 ft throughout production.

Slurried bentonite was used to create the soil mix material necessary to meet the
permeability requirement of 1x10 -7 cm/sec. The slurried bentonite was pumped
through the mixing tool while drilling and mixing the columns that created the
continuous barrier wall.

Geo-Con exceeded every permeability requirement on all samples sent to the lab,
with an average permeability being 1.79x10 -8 cm/sec Field samples of the freshly

493
mixed material were taken once a day from differing depths of a finished column
using Geo-Cons custom designed sampling device.

Samples were tested for permeability at Geotechnics Laboratory in Pittsburgh,


Pennsylvania. All other quality control tests were done on-site by the Geo-Con
Project Engineer. These tests included unit weight, viscosity, pH, and filtrate on the
bentonite slurry.

Challenges

A significant challenge on the project was the presence of buried obstructions


encountered during the mixing of the wall, including boulders, timbers, and steel
cables. The obstructions required selective removed from the alignment with an
excavator, so that all columns could be installed.

Ambient air in work zone air-monitoring was carried out on this site every one to two
hours of production, because of the potential for encountering buried trash at the
site. The project was completed in Level D at all times.

494
Case Study 21: Johnston Atoll Solid Waste Burn Pit

Location: Pacific Ocean

History: Hazardous waste incineration

Contaminants: Lead

22,000 tons
Scale:
Portland Cement
Reagents:
Ex-situ
Method:
Restoration
End-Use:

Site description

Johnston Atoll is located approximately 700 nautical miles southwest of the Hawaiian
Islands. It is an unincorporated territory of the United States under the operational
495
control of the Defense Nuclear Agency (DNA). Johnston Atoll is also managed by
the U.S. Fish and Wildlife Service (USFWS) as a National Wildlife Refuge. The atoll
is comprised of four small islands: Johnston, Sand, North, and East Islands, which
are surrounded by a coral reef.

The Solid Waste Burn Pit (SWBP), located on the northwest end of Johnston Atoll,
was constructed in 1978 and utilized to burn refuse generated during the daily
operation of the island. Hazardous materials, such as batteries, paints, and
solvents, were burned previously in the inactive portion of the SWBP. The inactive
portion of the SWBP contained ash material considered to be RCRA-characteristic
hazardous waste, based on its TCLP lead levels. OHM Remediation Services
(OHM) designed and mobilized an ash/residue treatment system to chemically
stabilize and solidify the lead-contaminated SWBP ash and reduce the leachable
lead levels to below the TCLP criteria.

The project was performed on an aggressive schedule due to the detection of lead in
the adjacent lagoon and the potential for continued contaminant migration during the
1995 hurricane season. Mobilization of equipment to Johnston Atoll was initiated in
April 1995, in advance of formal approval by EPA of the RCRA Class III Permit
Modification. The permit modification request designated the SWBP and
surrounding area as a Corrective Management Unit (CAMU), to allow treatment
without triggering Land Ban Disposal Restrictions. The Johnston Atoll CAMU was
the first CAMU approved by EPA Region IX.

This project required materials and equipment to be transported by ocean-going tug


and barge from Seattle to Johnston Atoll via Honolulu. Over 1,000 tons of hydrated
lime were containerized and transported, along with equipment including office,
laboratory and decontamination trailers; 17 heavy construction vehicles and trucks;
assorted materials handling equipment; and maintenance supplies and materials
required for the entire project. Eighteen Shaw project personnel were assigned to
the site for the entire project.

Risk drivers

Lead-contaminated particles from the SWBP were affecting the blood lead levels of
monk seals and migratory fowl that frequented Johnston Atoll. The contaminated
soils required treatment and capping to prevent contact with the contaminated soil by
wildlife.

496
Objectives

Remove the contaminated soil


Treat the soil to remove its RCRA hazardous characteristic D008 (lead)
Place and compact the treated material back into the excavation area
Cap the treated material with 3 ft of coral sand
Restore the site

Method

Material with greater than 5 mg/L TCLP-leachable lead was treated with hydrated
lime to reduce both the TCLP- and SPLP-leachable lead concentrations to less than
0.75 mg/L.

Validation

Samples of the treated soil were obtained at a frequency of every 200 tons treated
and subject to TCLP and SPLP testing for metals. Over 99% of the treated samples
met the performance criteria on the first pass. Those stockpiles not meeting the
performance criteria were retreated.

In 2003, samples of the treated material were collected by drilling through the cap.
The treated material still had less than 0.75 mg/L in the TCLP and SPLP leachate.
The treated material had less than 50 g/L in the Physiology Based Extraction Test
(PBET).

Equipment

A pug-mill system, consisting of a feed hopper with conveyor, pug-mill, reagent silo,
and stacking conveyor were used to treat 150-200 tons of contaminated soil per
hour. The treated soil was segregated into 200 ton stockpiles for confirmation
testing.

497
Case Study 22: Kingston

Location: Kingston, Tennessee

History: Natural soil

Contaminants: Coal fly ash

560,000 yd3
Scale:
Blast furnance slag, cement, bentonite
Reagents:
In-situ
Method:
Industrial spill amendment
End-Use:

498
Site Description

A dike failure at the Kingston Fossil Plant in 2008 released an estimated 5.4 million
cubic yards of coal ash from a dredge cell rupturing a natural gas line, and cutting
power and transportation and contaminating the Emory River. Three homes were
rendered uninhabitable, and a nearby residential area was evacuated.

The removal of coal ash from the Emory River was completed in May 2010. The
reconstruction of the failed dike, along with reinforcement of the remainder of the
containment cells will prevent future movement of impounded ash. The installation of
the Perimeter Wall Stabilization (PWS) is a key component of dike reinforcement.

The PWS is divided into 8 segments with differing design criteria in each. The design
includes both continuous in- and out-board walls, constructed with a series of
connecting shear walls perpendicular to the in-board wall lateral wall segments,
designed to withstand shear loads.

In some segments, the shear walls extend beyond the out-board perimeter wall to
form a buttress to the out-board wall. Each segment has different treatment widths
and area-replacement ratios to account for differing failure modes in the most
economical manner. The diagram below shows the typical design layout:

499
Method

Geo-Con is installing the PWS using the slurry wall trenching method. The slurry wall
is installed using a custom designed long reach boom and long stick excavators
fitted with 4 ft wide trenching buckets. The trenching bucket is specifically designed
for keying the wall into the shale bedrock formation at the site.

The trench is excavated using Cement-Bentonite (CB) slurry, which acts as hydraulic
shoring during excavation and cures in the trench to become the permanent backfill
material.

The self-hardening slurry includes a combination of Ground Granulated Blast


Furnace Slag cement, Portland cement, and bentonite slurry. The proportions of the
mixture were developed during an extensive laboratory testing program that
evaluated over 70 candidate mixtures.

Multiple mix recipes have been used in production segments to accommodate


different needs in the design. The cured CB slurry will reach an unconfined
compressive strength (UCS) of 200-400 psi.

The CB slurry wall is keyed into the local shale bedrock formation which occurs at
depths ranging from 45 to 65 ft below ground surface (bgs) and bedding planes that
dip from 15 to 20 from horizontal.

The self-hardening slurry wall trenching method employed on the PWS Project offers
a number of distinct advantages over other methodologies including:

A homogeneous wall is installed, creating a continuous perimeter wall system with


more consistent strengths throughout
No in-situ materials are used
A high capacity batch plant produces a consistent product
No cold joints: panels are overlapped for tight tie-in to previous work
Able to follow sloping rock and more accurate ability to measure bedrock key-in
than is inherent with in other geotechnical construction methods such as soil
mixing
The process is efficient with no wasted mixing: the entire width of the wall is the
effective width of the wall.
500
The PW footprint will form a continuous boundary of approximately 11,500 ft around
the property, and when completed, Geo-Con will have installed more than 12 linear
miles (560,000 yd3) of trench, making it the largest Cement-Bentonite slurry wall
installation in US history.

Geo-Con is working for TVA, which acts as the General Contractor. Construction is
scheduled to be completed by 2014.

Issues

Key project challenges include concurrent design and construction.

Geo-Con works closely with the design engineer on upcoming segments of the
design to ensure constructability and smooth transition between the segments.

An additional challenge is the strata through which excavation occurs. Generally, the
top 15 ft of the soil profile consists of the coal ash which is highly liquefiable. This
requires care in both the excavation and stability of work platform. The remainder of
the profile consists of sand and silt underlain with the shale bedrock key-in layer.
Because of the displacement during the ash slide, occasional non-native type
materials have also been encountered.

501
Case Study 23: London Olympic Site

Location: London, UK

History: Former industry

Contaminants: Metals, PAHs

60,000 tonnes
Scale:
Biochar & zero-valent iron
Reagents:
Ex-situ
Method:
2012 Olympic Games complex
End-Use:

Site Description

A large derelict site in East London was selected as the venue for the Olympic
games of 2012. The site was a former marsh adjacent to the Lea River that had
been reclaimed in Victorian times by backfilling of demolition waste and coal clinker.
These and subsequent industrial activities caused contamination.

502
The soil was a mixture of natural and made ground, was found to be contaminated
by heavy metals (e.g. arsenic), PAHs, TPH, sulphate, cyanide and ammonium. The
remediation took place between 2007 and 2010.

Characterisation

Although the majority of the soil, over 1 million tons, was treated by soil washing,
about 60,000 tons was too fine to treat with this technique. Therefore chemical
stabilization was proposed to the client. This soil varied from fine grained sandy loam
to stiff clay. Some of the soils showed elevated leachabilities of heavy metals, in
particular arsenic, others showed elevated leaching of organics (mainly PAHs), and
for some soils both metals and organics were leachable.

Objectives

Tables 1 and 2 below show some typical site-specific leaching target values for
reuse of soil on the site, for organic and inorganic contaminants, respectively. It
should be noted that some of the target values, such as for PAH, were lower than
the analytical methods levels of detection, and it was, therefore, agreed with the
Environment Agency to reduce the leaching to below the level of detection. No
geotechnical improvement of the soil was required.

Table 1: Leachability (mg/L) according to EN 12457-4 of relevant organic


pollutants before and after treatment by various additions of OrganoDEC for
hydrocarbon immobilization

1.5% % Target
Contaminant Input Average
OrganoDEC A Reduction value

Naphthalene 0.050 <0.00092 >98 0.012

Acenaphthylene 0.0014 <0.00012 >91 0.00014

Acenaphthene 0.0068 <0.00036 >94 0.00012

Fluorene 0.0049 <0.0002 >96 0.00014

Phenanthrene 0.0026 <0.0003 >88 0.00011

Anthracene 0.0013 <0.00014 >89 0.00002

Fluoranthene 0.0019 <0.00038 >80 0.00002

503
1.5% % Target
Contaminant Input Average
OrganoDEC A Reduction value

Pyrene 0.0013 <0.00046 >64 0.0001

PAH (Sum of EPA 16) 0.072 <0.003 >96 -

Table 2: Leachability (mg/L) according to EN 12457-4 of relevant heavy metal


pollutants before and after treatment by various additions of FeDEC, additive
for heavy metal immobilization.
1% % Target
Contaminant Input Average
FeDEC Reduction value

Arsenic 0.0082 <0.005 >39 0.05

Copper 0.0086 <0.005 >42 0.028

Zinc 0.088 0.018 79 0.125

Drivers

The Olympic site is criss-crossed by rivers and canals, and the environmental risk
driver was the leaching of pollutants via groundwater and surface waters. This risk
was translated into residual leachate concentrations of the treated soils, and
although the leachate targets varied slightly over the site, they were all very low.

Method

Both for the organic and the metal contaminants, chemical stabilization was applied.
For the organics a proprietary biochar (organodec) was added to the soil at very low
dosages (order 1 %). For the metals a zero valent iron powder was used, also at
very low dosages.

Most of the soil was treated by means of a batch mixing plant with a throughput of
about 120 tons an hour (a double axis pug-mill system, pictured below, right). It was

504
however, technically impossible to treat the very cohesive clay soils with this system,
due to fouling, and for this reason a rotary bucket was applied (pictured below, left).

Challenges

Due to the public interest in this site, and the fact that the Environment Agency was
not familiar with chemical stabilization, this project was seen as a demonstration
project and was strictly followed-up.

Validation

All treated soils were validated immediately after treatment, by means of the EN
12457 4 leaching test, and checked against the site-specific leaching criteria.

505
Case Study 24: Martinsville

Location: Martinsville, Virginia

History: Chemical Manufacturing

Contaminants: Carbon Tetrachloride

Scale: 5,000 yd3

Reagents: ZVI, Kaolin Clay

In-situ, LDA
Method:

506
Summary

Geo-Con was contracted by the DuPont de Nemours and Company, to stabilize


contaminated soils at the DuPont Martinsville Unit I ISM Remediation site using In-
situ Stabilization. Treatment involved the application of zero valent iron and Kaolin
clay.

Zero valent iron was selected based on the primary contaminant of concern, carbon
tetrachloride. Kaolin clay was used to minimize groundwater flow through the
treatment zone. Elevated levels of carbon tetrachloride were identified on-site in the
existing soil and groundwater with shallow soil concentrations as high as 30,000
mg/kg. Other contaminants detected included chloroform, methylene chloride,
dichloroethylene, barium and chromium.

Geo-Cons scope consisted of in-situ treatment of 5,000 yd3 of soil by the Shallow
Soil Mixing technology (SSM), without excavation of removal.

In this application, Geo-Con used a Cassagrande Model CM15 crane-mounted drill


rig to support the soil mixing operation. A wet or hollow Kelly bar connects the mix
auger with the drill turntable and carries slurry from the plant to the auger.

An 8-ft diameter mixing-auger was used to produce a homogeneous soil mix column.
The zero valent iron/kaolin clay slurry was applied as specified to a depth of 35 ft.
The 8-ft diameter columns were spaced to provide complete coverage of the
treatment areas.

Kaolin clay slurry was prepared in Geo-Cons on-site mixing plant and was injected
as the mixing auger was advanced downward to the maximum treatment depth of 35
ft to create the appropriate soil mix proportions within the column. The SSM columns
were laid out to provide full coverage for the 3 rates of reagent dosage required at
the site.

Once the Kaolin clay was injected and the column thoroughly mixed, iron was added
by driving in a steel casing, filled with the prescribed amount of iron for the column,
then pulling out leaving the sacrificial drive point and the iron in place. By proper
sizing of the casing for the specific iron application rate, the iron was distributed
evenly over the length of the column. The column was then remixed from top to
bottom. Cement was incorporated into the upper 20 ft of treated soil to improve the
workability of the treated soil so that it could be graded and capped.
507
A total of 78 columns were successfully installed to treat the designated area. A
sampling program was implemented to verify the amount of iron installed. Selected
columns were sampled at 10-ft intervals with Geo-Cons SSM sampling tool.

Soil mix samples were tested for iron content using a wet wash and iron separation
test. Results of sampling indicated the average iron content of each column was
greater than required and no individual samples were more than 20% less than the
required iron concentration.

Control of both carbon tetrachloride emissions and migrant nuisance odours to


prevent activation of off-site alarms both real and perceived were a significant
concern. Controls implemented included staged mixing and excavation, plastic
liners, tarps and application of latex foam. Other work related to the Unit I ISM
Remediation included demolition and removal of buried utilities and concrete
structures, abandonment of monitoring wells and construction of an asphalt cap.

508
Case Study 25: Milwaukee, Wisconsin

Location: Milwaukee, Wisconsin, USA

History: Manufactured Gas Plant (MGP)

Contaminants: MGP coal tar residuals

25,000 yd3
Scale:
Portland cement and blast furnace slag
Reagents:
In situ auger and bucket mixing
Method:
Property redevelopment and waterway
End-Use:
restoration

509
Summary

Natural Resource Technology supported all phases of planning and design through
implementation and remedial construction of the former manufactured gas plant
(MGP) facility located along a major waterway in Milwaukee, Wisconsin. Project
planning required extensive permitting and coordination with the City of Milwaukee,
Wisconsin Department of Natural Resources, state health department, property
owner and local businesses. The site contains a number of historic brick MGP
structures that have historic significance for the City, and are occupied by different
business concerns.

This high profile project was conducted in 2 phases. Phase I consisted of the
excavation and on-site thermal treatment or off-site disposal of approximately 58,000
tons of MGP-impacted soil and debris. Phase II involved the relocation of the main
truck access ramp to the property, installation of an environmentally-sealed sheet
pile system along the river, demolition of approximately 650 ft of historic dock wall
structure, in situ stabilization/solidification (ISS) of approximately 25,000 yd 3 of MGP-
impacted soil, and restoration of the shoreline with riprap. ISS operations were
performed directly along the river and in upland areas.

Remediation focused on maximizing removal or solidification of MGP impacted soil


in areas previously identified to contain the heaviest MGP impacts. Thermally-treated
material was beneficially reused for backfill in excavated areas. Excess ISS material
(swell) was beneficially used on-site to rebuild the river bank and for site grading.

Characterisation

Site Characterisation
Remedial Alternatives Analysis
Remedial Design
Permitting/Agency Negotiation
Bench- and Pilot-Scale Testing
Construction Quality Assurance
Construction Management

Risk Drivers

Potential contributions of MGP residuals to groundwater and surface water


Unstable historic dock wall structures and erosion of potentially MGP impacted
materials to the river
510
Protection of historic MGP structures

Objectives

Minimize the potential for future contributions of MGP residuals to groundwater


and surface water
Restore the site for future property redevelopment
Improve the waterfront to minimize potential for future erosion and degradation

Method

Extensive ISS bench-scale testing was conducted to evaluate a number of different


mix designs. Final mix designs consisted of combinations of Portland cement and
ground granular blast furnace slag. Specialized mix designs were developed to
address specific strength and stability requirements for the river and roadway
embankments. To reduce concerns for possible washout along the river, a
commercial stiffening agent suitable for submerged concrete applications was also
added. To access the MGP impacts, the existing truck access ramp was demolished
and reconstructed in another area of the site. Reconstruction included partial
demolition and reconstruction of a major city bridge that involved detailed
coordination with city engineers. Demolition of historic dock wall structures was also
required along the waterway to provide access for the ISS operations. Pilot- and full-
scale ISS operations were completed by the contractor (WRScompass) using a
combination of 8 and 12-ft diameter augers, and bucket mixing.

Validation

Discrete samples were collected from the top, middle or bottom of the selected ISS
columns or grids by WRScompass according to the following frequency:

One discrete sample collected every 1,000 yd3 or once per day of ISS production
One discrete sample collected every 100 linear feet around the perimeter of the
ISS treatment area

Samples were tested for unconfined compressive strength and hydraulic conductivity
to demonstrate compliance with the performance objectives.

511
Benefits of Solution

Use of ISS substantially reduced concerns for potential direct contact exposure to
the community and river during remediation
Application of ISS reduced long-term liability and cost in lieu of off-site disposal of
MGP impacted materials
ISS was effective for river embankment restoration and reconstruction

512
Case Study 26: Nederland

Location: Nederland, Texas

History: Hazardous waste impounds

Contaminants: Hydrocarbons

65,000 yd3
Scale:
Fly ash
Reagents:
Excavator bucket mixing
Method:
N/A
End-Use:

513
Site Description

The project involved the closure of 4 hazardous waste impoundments of


approximately 80 acres, and the placement of over 300,000 yd 3 of imported soils,
1.5m ft2 of geosynthetic clay liner (GCL), 1.0m ft 2 of geogrid, and the stabilization of
over 65,000 yd3 of excessively soft sediment. Work also included installation of a
vinyl sheet pile wall to control the migration of NAPL-impacted groundwater from
going off-site.

Other project features included performing a Resource Conservation and Recovery


Act (RCRA) closure on one of the waste impoundments, installation of Articulated
Block Mattress (ABM) at a storm water inlet that was constructed in the final erosion
control phase, and creation of 20-acres of wetlands within the confines of the former
Storm Water Impoundment (SWI).

Characterisation

The only sludge that required treatment via stabilization within the SWI was along
the existing drainage sloughs. The sludge in most of the SWI was in solid form due
to continual pumping of storm water from the basin, which allowed the sludge to
build a crust. The sludge, through many years of investigation, was characterized to
be non-hazardous, but still had a high hydrocarbon presence. Waste composition of
the stabilized material, including reagent, is shown below:

Solids ~ 35% to 40%


Moisture ~ 50% to 55%
Oil & Grease ~ <5%
Fly Ash ~ 5%

Risk Drivers

The risks associated with this project included the sludge and contaminant being
transferred off-site as a result of groundwater migration, and flooding or a hurricane
storm surge on the Neches River.

514
By stabilizing and capping the impoundment areas (including the installation of a
RCRA cap and cutting-off the flow of groundwater with a barrier wall) these risks
were mitigated requiring no further action from the agencies.

Objectives

The objectives of this remedial action were to:


1. Reduce sediment transfer off-site via surface water or groundwater
2. Reduce hydraulic conductivity from permeating through cap materials to get to
the waste
3. Meet closure regulations stipulated by the Texas Commission on
Environmental Quality (TCEQ)
4. Mitigate the damage to wetlands by creating wetlands on top of a portion of
the cap
5. Enhance storm water quality that discharged to the Neches River

Method

It was not necessary to stabilize but rather solidify since the waste requiring
strengthening was being classified as non-hazardous and groundwater was not at a
high elevation. The sludge required little reagent dosage to gain the strength it would
need to support a cap and eventual wetlands that was constructed on top the cap.
Had there been any chance of the treated material being immersed in water, then
stabilization would have been the preferred approach.

RECON utilized an excavator mounted on


swamp pontoon tracks to install several
drainage ditches in soft sediment/sludge.
This, coupled with simple sump
dewatering techniques, proved to lower
the water elevation in the sludge, forming
a 2-ft thick crust.

515
This formation of the crust allowed low-ground pressure dozers to traverse most of
the area. Other very excessively soft areas within the impoundment were stabilized
using fly ash, a by-product generated from a local power plant. Long-stick excavators
were used to mix the fly ash with the sludge. By excavating drainage ditches,
RECON used 50% less fly ash during the project. In addition, RECON was able to
use a standard bi-axle geogrid in lieu of high-strength.

Validation

RECON solidified the waste to support a cap and added the necessary small
dosages of high available lime fly ash to reduce the water content in the sludge to
produce a strength of approximately 10 lb/in2 (psi). This was monitored by use of
pocket Penetrometer and Torvane field vane shear.

Issues

In order to speed-up the construction schedule and receive a minimum of 200 loads
of soil per day, RECON developed and operated a borrow pit and purchased 17
dump trucks. With the help of a local hauler, RECON was able to haul in excess of
400 loads per day. This action allowed RECON to expedite the amount of GCL
installed per day that required daily cover to protect it from wet weather.

516
Case Study 27: New Bedford

Location: New Bedford, MA

History: Manufactured gas plant

Contaminants: MGP contaminated soil

6,457 yd3
Scale:
Not specified
Reagents:
In-situ
Method:
N/A
End-Use:

517
Summary

This project involved using a single auger soil mixing tool to solidify MGP impacted
dredge sediments. Geo-Solutions (GSI) was subcontracted by Charter
Environmental to complete all of the soil mixing work.

The soil mixing required the installation of 178 x 8-ft (2.44 m) diameter and 32 x 3-ft
(0.91 m) soil mixed columns to depths of 15 ft - 32 ft (4.6 m - 9.6 m) below work pad
elevation. A column layout for the soil mixing work is shown below.

All of the soil mixing work was completed using GSIs Delmag RH-18 drill rig, batch
plant, and silos. Prior to GSIs arrival, the dredge sediments were placed within sheet
pile cells.

The stabilization took place within the sheet pile cells using timber mats for rig
stability.

518
Case Study 28: Nyack

Location: Nyack, New York

History: Manufactured gas plant

Contaminants: BTEX, PAHs

11,400 yd3
Scale:
Portland Cement, Bentonite
Reagents:
In-situ
Method:

519
Summary

A Manufactured Gas Plant (MGP) operated at this site from 1852 until 1965. The site
covers a total land area of approximately 4 acres, is in an urban setting, with
adjacent properties used for a mix of commercial and residential purposes.

It is believed that gas was made from the coal carbonization process from 1852 until
1887, and then until 1889 the plant used oil instead of coal. From 1890 until 1938 the
plant used both coal and oil as feedstock for the carbureted water gas (CWG)
process. From 1938 until 1965 the site was used as an oil gas facility only during
times of peak demand, a practice known as peak shaving.

Coal tars are present at this site in the form of a dense oily liquid, which does not
readily dissolve in water. Materials such as this are typically found at MGP sites, and
are referred to as non-aqueous phase liquids or NAPL.

Specific volatile organic compounds of concern are benzene, toluene, ethylbenzene


and xylenes. These are referred to collectively as BTEX.

The specific semivolatile organic compounds of concern in soil and groundwater are
the following polycyclic aromatic hydrocarbons (PAHs):
acenaphthene acenaphthylene
anthracene benzo(a)anthracene
benzo(a)pyrene benzo(b)fluoranthene
benzo(g,h,i)perylene benzo(k)fluoranthene
dibenzo(a,h)anthracene chrysene
fluoranthene fluorene
indeno(1,2,3-cd) pyrene 2-methylnaphthalene
naphthalene phenanthrene
pyrene

Excavation was used at the upper terrace, whereas in-situ S/S was used for the
lower terrace of this site. Geo-Con was subcontracted by Sevenson Environmental
(Sevenson) to perform the in-situ treatment of the lower terrace soils.

520
In-situ treatment involved Geo-Cons custom-made Calweld Drill Platform mounted
on a Manitowoc 3900 crawler crane. The Calweld Drill is capable of over 400,000 ft-
lbs of torque and can turn a 10 ft diameter mixing auger at up to 21 ft below ground
surface (fbgs) to the bedrock surface.

A mixture of water, Portland cement and Sodium Bentonite (Montmorillonite Clay)


grout was prepared on-site and pumped through a hollow-stem Kelly bar to the
mixing auger. Overlapping columns provided 100% coverage of the treatment area
involving 11,400 yd3 (cy). Treated soil was collected every 500 cy and tested for
Unconfined Compressive Strength and Flex-Wall Permeability to ensure
immobilisation of contaminants. Treatment met a UCS of >50psi and a permeability
of approximately 1 x 10-6 cm/s.

Geo-Con also performed high-pressure jet grout mixing to ensure the interface
between the S/S columns and the site bedrock was sealed. The cement/bentonite
grout was delivered through ports at 3500 psi using a custom made sub-assembly
on a 4 drill string, and 755 lineal feet of jet grouted columns were installed. An
additional 311 cy of soil were also stabilized in this project that was completed within
budget and ahead of schedule and with no health and safety related incidents.

521
Case Study 29: NYSEG Norwich NY MGP Site

Location: Norwich, NY

History: Manufactured gas plant

Contaminants: PAHs, DNAPL

52,000 yd3
Scale:
Portland cement & bentonite
Reagents:
In-situ
Method:
Shopping plaza
End-Use:

522
Site description

The NYSEG Norwich Former MGP site is located at 24 Birdsall Street, in the City of
Norwich, Chenango County, New York. The exact starting date of MGP operations at
the site is unknown. However, Sanborn fire insurance maps suggest that the plant
operations started sometime between 1863 and 1887. By 1887 the site began
supplying manufactured gas to the City of Norwich under the name Norwich Gas
Lighting Company. Little is known about the generation and disposal practices of
residues from the MGP. However, 2 tar storage vessels existed in the subsurface
prior to their removal in 1997. In addition, a potential purifier waste disposal area was
identified in 1990 through an interview with a former employee of the MGP.
Manufactured gas was produced at the site using the coal gasification and
carbureted water gas processes. In 1892 the name of the facility operator was
changed to Norwich Illuminating Company, which was later changed to Norwich
Gas and Electric Company in 1917. Coal gas was produced on-site until 1917 and
then carbureted water gas produced from 1917 to 1953. NYSEG acquired the
property in 1939.

The site previously occupied approximately 1 acre of land located at 24 Birdsall


Street. In the years following cessation of gas production, former MGP structures
were razed and subsequently NYSEG used the site for equipment storage.
Presently, much of the property is paved with asphalt or covered with compacted
gravel. A NYSEG electric substation exists on the eastern portion of the site.

The northern part of the site has been developed as a shopping plaza with retail
shops. NYSEG purchased the former Aero Products facility located to the south and
used the building for storage for several years. During the summer of 2006, NYSEG
demolished the former Aero Products building. The off-site area that extends to the
south of the former Aero Products building is comprised of mostly residential
housing. NYSEG has purchased property at 37 and 41 Front Street and razed the
structures located on these properties to allow for the ISS of the underlying soils.

WRScompass performed this in-situ stabilization (ISS) project for NYSEG at the
Norwich Former MGP Site located in Norwich, New York. The scope of services
consisted primarily of ISS treatment of 52,000 yd 3 (cy) of soil using an auger mixing
method, with other activities necessary to prepare for ISS work and to restore the
site. The site was adjacent to an active shopping area that remained open during
remediation thus the work was highly visible to the public.

523
Risk Drivers

The contaminated soils required removal/treatment to prevent contact with the


contaminated soil by the residents and to prevent migration of PAHs off-site toward
the Chenango River.

Objectives

Removal and disposal of the unsaturated soil


Treatment of the saturated soil down into the clay aquitard
Restore the site

Site preparation included abandoning monitoring wells located in the work area,
performing a detailed utility location survey, exposing and retiring known gas mains
in the work area, installing erosion controls, removing the existing perimeter fence
where necessary and installing a temporary fence with privacy screen around the
site perimeter, setting up a support zone and contamination reduction zone, and
establishing a remote stockpiling area.

The site was pre-excavated in preparation for ISS work such that the treated material
would be at least 4 ft below the final grade as required by the NYSDEC. This
included the removal of asphalt pavement and concrete slabs, removing of over
6,000 cy of expectedly non-impacted soil, and removal and off-site disposal of known
hot spots of impacted soil. The soil that was potentially non-impacted was
transported to a stockpiling site several miles from the site and at another property
owned by the client, where it was sampled and verified to either be suitable for reuse
as backfill material at the site or disposed of off-site.

ISS was performed using a 10-ft diameter auger. A grout plant was set up and the
required reagent admixture was produced on-site then conveyed to the auger rig,
where it was added on a per weight basis using a pre-determined mix design of 8%
by weight for Portland cement and 1% by weight for bentonite. As the work
progressed, the reagent admixture was refined to reduce the amount of bentonite
required as the bentonite addition rate was hampering the ability to productively
complete the work under adverse winter weather conditions. ISS was first performed
at a 10-ft wide perimeter that was keyed 4 ft into the clay layer, followed by the
interior ISS keyed 2 ft into the clay layer.

524
Once ISS was completed, the site was backfilled to the required elevations. Suitable
excavated and stockpiled soil was returned from the remote stockpile location and
was supplemented by imported clean fill soil. Backfill material was installed in
controlled lifts and compacted. Once the required grades were established a
demarcation layer was installed atop the backfill material. Then the site restoration
was completed by installing paving sub-base stone and pavement in a majority of the
site, and topsoil and seeding in other specified areas.

Method

Portland cement and bentonite were used to improve the unconfined compressive
strength to >50 psi and reduce the hydraulic conductivity to <1x10-6 cm/sec.

Validation

Samples of the treated soil were obtained at a frequency of every 500 cy, treated
and subject to UCS and permeability testing. Over 98% of the treated samples met
the performance criteria on the first pass. Those columns not meeting the
performance criteria were retreated.

Equipment

The ISS operations within the site were conducted using a 4000-series Manitowoc
crane equipped with an attached Hain Platform. The crane/platform assembly is
supplemented with a swivel-mounted, top-feeding Kelly Bar capable of reaching a
depth of 75 ft bgs. ISS was performed using a 10-ft diameter auger attached to the
bottom of the Kelly Bar.

Specific Issues

The ISS work was conducted adjacent to an operating shopping plaza, with ISS
columns installed within 1 ft of the footer for a grocery store. Extensive vibration,
odor, and dust control was required to eliminate any impact of the operation on the
shopping plaza.

525
Case Study 30: Obourg

Location: Obourg, Belgium

History: Former manufactured gas plant

Contaminants: Coal Tar (PAHs, DNAPLs, LNAPLs)

110,000m3
Scale:
Paper sludge ash
Reagents:
Soil liming equipment
Method:
Sediment treatment centre
End-Use:

Site Description

The Walloon authority SPW (Service Publique de Wallonie) wanted to install a new
sediment treatment centre along the Canal du Centre, which is known for its four
large ship elevators (UNESCO world heritage). Two sediment storage lagoons that
526
were present on the site located near Obourg, were filled with soft poorly
consolidated sediment.

The proposal was to solidify the soft sediments (110,000 m in total) and use the
product in the construction of base layers and in new dykes. The picture above
shows the excavation works in progress (top middle) and the two newly constructed
lagoons (bottom middle) being lined with freshly solidified sediment.

Objectives

The objective of the work was to solidify the soft sediments and turn it into a
construction material that could achieve a compressibility modulus of 11 MPa (or 17
MPa at surface layer). The execution method chosen by the contractor, DEC
imposed an extra boundary condition: after treatment with binder the material had to
be stockpiled for several weeks before being compacted in its final place, ruling out
the use of cement as an additive.

Formulation

The clayey-silty sediment present at the site varied between 40% dry matter content
(very soft) to 65% dry matter content (plastic).

The sediment was not solidified with a traditional cementitious binder, as this would
result in a short term curing. Therefore a specific additive was applied, which
absorbs the sediments pore water, improves the structure, and flocculated the clay
particle. These properties result from using paper sludge ash, a meta-kaolin based
fly ash, produced from incineration of paper mill waste. This product can be found in
sufficient large quantities in most countries. On this site an average dosing of 20 %
w/w was used, resulting in the use of about 35,000 tons of paper ash.

Method

Although the initial plan was to use a batch-wise pug-mill mixing plant, a less
complex and cheaper method was applied. Standard soil liming equipment (dosing
and rotary mixing) was used on layers (or lifts) of 0.5 m (20 inches), and then
scraped and stockpiled by a bulldozer.
527
Testing

Validation of the solidification treatment was carried out on both laboratory- and field-
scale. Freshly treated sediment was sampled and compacted in a proctor mold,
tested immediately, at 14 days and after 28 days by means of CBR (Californian
Bearing Ratio). The results can be correlated to field plate tests, which were also
carried out to confirm the target values were achieved.

References

Bujulu P., Sorta A., Priol G., Emdal A. Potential of wastepaper sludge ash to
replace cement in deep stabilization of quick clay.

Characterization and Improvement of Soils and Materials, Session of the 2007


Annual Conference of the Transportation Association of Canada, Saskatoon,
Saskatchewan.

528
Case Study 31: Perth Amboy

Location: Perth Amboy, NJ

History: Landfill

Contaminants: Petroleum hydrocarbons

52,000 tons
Scale:
Cement
Reagents:
In-situ
Method:
N/A
End-Use:

Site Description

The Perth Amboy refinery operated from the 1920s to 1983, producing gasoline and
heating oil, before changing to asphalt refining. In the late 1990s, cleaning of the
storage tanks at the refinery was undertaken, and the oil sludge was placed into 2
impoundments the 8-acre Northfield Basin and the 2-acre Surge Pond.

Characterisation

The sludges were an emulsion of lubrication oil range (C28 C60) petroleum
hydrocarbons, asphaltic tars, and water.

Objectives

Chevron, the site owners, wanted to utilize the 8-acre Northfield basin for storm
water collection, and to enable this, in-situ solidification of the oily sludge in both
impoundments was proposed. This was followed by removal of material from the
Northfield Basin to the Surge Pond, capping and closure.

529
Method

The oily sludge material was encapsulated in a cementitious matrix using treatment
cells of 500 yd3. Solidification treatment started along the edge of the impoundments
and then continued inward. Portland cement was pneumatically conveyed to each
cell and was then mixed into the oily sludge using a long-reach excavator. The
solidified material could support the long-reach excavator within 2 days, enabling the
solidification process to continue to completion. Upon successful treatment of both
sites, excavation and removal of solidified sludge to the Surge Pond for final
disposal, was followed by covering with an impermeable clay cap.

530
Case Study 32: Portsmouth

Location: Portsmouth, Virginia

History: Wood Treating Facility

Contaminants: Creosote, Pentachlorophenol (PCP)

Scale: 47,000 yd3

Reagents: Portland Cement, Ground Granulated Blast


Furnace Slag Cement, Organophillic Clay

Method: In-situ, Excavator-based

531
Summary

The Atlantic Wood Industries (AWI) site consists of approximately 48 acres of land
on the industrialized waterfront of Portsmouth, Virginia, and 30 to 35 acres of
contaminated sediments in the Southern Branch of the Elizabeth River. From 1926
to 1992, a wood-treating facility operated at the site using both creosote and
pentachlorophenol (PCP).
At one time, the US Navy leased part of the property from AWI and disposed of
waste on-site, including used abrasive blast media from the sand blasting of naval
equipment and sludge from the production of acetylene.

Polynuclear aromatic hydrocarbons (PAHs), PCP, dioxins and metals contamination


(mainly arsenic, chromium, copper, lead and zinc) have been detected in soils,
groundwater, and sediments. The EPA issued a Record of Decision in 1995 for the
clean-up of the site, which was amended in 1997. The AWI Site was added to the
National Priorities List of most hazardous waste sites in 1990.

Geo-Con was contracted by the Norfolk District of the US Army Corps of Engineers
to perform Phase 1B of the cleanup effort which included the stabilization and
solidification of over 47,000 cy of DNAPL-impacted soils.

The Treatment Area, or Historic Disposal Area (HDA), consists of 29 separate


sections ranging in depth from 8 ft to 27 ft below ground surface. Due to the amount
of debris expected throughout the HDA, Geo-Con selected the excavator/rotary
blending SSM system to perform the stabilization of the DNAPL-impacted soils. This
system combines the use standard hydraulic excavators with an excavator-mounted
rotary-blending unit to homogenously blend the subsurface soils with the chosen
reagents.

The impacted soil was blended with a cement-based liquid grout consisting of:
Portland Cement
Ground Granulated Blast Furnace Slag Cement
CETCO Organophillic clay

Two separate mix designs were selected for the project depending on the section
being mixed. Each mix design was to attain a minimum UCS of 50 psi and a
maximum permeability of 4 x 10-6 cm/s.

532
Geo-Con served as the prime contractor for the project. The scope of work for the
project also included:

Relocation of electrical utilities for the neighbouring Portsmouth Public School


District Property to accommodate soil mixing
Perimeter air monitoring
Pre-excavation, removal, handling and stockpiling of over 3,000 cy of
miscellaneous debris
Final grading
Installation of an impermeable clay cap
Restoration

533
Case Study 33: Rieme

Location: Rieme, Belgium

History: Chemical industry

Contaminants: Acid tar

170,000 m3
Scale:
Cement/pozzolan/filler blend
Reagents:
Ex-situ
Method:
Not specified
End-Use:

534
Site Description

In the first half of the 20th century Fina (now Total) has produced white medicinal oil
from crude oil by means of concentrated sulphuric acid treatment. This process
typically generated a viscous residue, called acid tars, and a clayey residue, called
Fullers Earth.

At the production site in Rieme, located in the port of Ghent (Belgium), these
residues have been stored in large open unlined lagoons, containing about 80,000
m acid tars, varying from liquid to solid, 20,000 m of Fullers Earth, and 70,000 m
of contaminated soils.

Characterisation

The liquid acid tars found in the smallest lagoon were actually emulsions, with up to
50 % water content, a pH of around 2, and a sulphur content of about 1.7 %. The
more viscous to solid acid tars also contained high water contents (even up to 60 %),
were far more acidic (pH 0 to 1) and contained very high sulphur levels between 5
and 10 %. Due to the high water content the calorific value of these tars was limited.
An attempt to reveal the complex hydrocarbon composition of the tars was not
carried out.

Objectives

Due to the large amount of contaminated materials, and due to the absence of
external treatment facilities, it was decided to treat the materials by
solidification/stabilization, and store them in a confined disposal facility on the site.

The Flemish Environment Agency (OVAM) demanded that the treated materials
meet the European Waste Acceptance Criteria for hazardous waste landfills (except
for the parameter TOC). The disposal facility was constructed following the
European rules of a hazardous waste landfill (double bottom liner system, leak
detection, double cover system). On top of the chemical criteria Total set out
geotechnical and durability criteria. The most relevant criteria are summarized in
Tables 1 and 2, below.

535
Table 1: Main geotechnical criteria of S/S.
Parameter Unit Objective

Volume increase % < 30

CBR-value (ASTM D1883-99) % > 11

Compressibility modulus field static plate loading test MPa > 11

Hydraulic permeability m/s < 10-7

Strength loss by ageing (wet-dry according to ASTM 558 % < 15


and freeze-thaw according to ASTM 559)

Table 2: Main chemical criteria


Parameter Unit Initial value Maximum
range

pH 03 4 13

Water soluble part % on dry matter 5 20 10

DOC (leached according


mg/l 3,000 28,000 90% reduction
to EN 12457-3)

Drivers

The groundwater in the vicinity of the site was heavily contaminated by hydrocarbons
and sulphuric acid, extending into residential gardens near the site. Total decided to
stop the source of this pollution first by treating the acid tar lagoons. The next phase
of work( yet to be done) involved treating the groundwater.

Method

The main challenge for the S/S of the acid tars was achieving the high mechanical
strengths while staying within the volume increase boundary condition of 30 %. In
order to establish the right mix formulations, various commercially and locally
available binders and additives were chosen:
536
Cements: OPC (CEM I, ordinary Portland cement), blast furnace slag cement
(CEM III)
Alternative binders or pozzolans: CKD (cement kiln dust), GGBS (ground
granulated blast furnace slags), paper ash
Alkaline minerals: quicklime, steel slag, flue gas desulphurization gypsum
Filler materials: PFA (pulverized fly ash)

About 90 mix formulations were laboratory tested on the 6 types of lagoon-derived


materials, cured for 28 days and tested for CBR, the main trigger criterium. Mixes
that passed were then further tested for the chemical criteria, and a specific mix
formulation was chosen, based on the same 4 additives (but applied at different
dosages dependent on the requirements of the specific lagoon material).

Generally, the more liquid the acid tars were, the more filler/binder material was
required, relative to the neutralizing additive. Conversely, the solid acid tars required
less solidification agent, but as they tended to be more acid and leached greater they
needed greater amounts of neutralizing additive.

A bespoke mixing plant, with a 60 tons an hour throughput was designed and
constructed, and set up in a closed hall in order to be able to extract the emissions
from the acid tars (see picture below, right). The mixing unit itself was a batch
concrete mixing plant (double axis pug-mill system), and is shown schematically
below (image, left).

Challenges

The acid tars emitted extremely high amounts of sulphur dioxide. For safety and
nuisance reasons, the emissions had to be controlled and monitored during all
phases of the work and the process, as were various preventive measures to protect
the equipment (excavators, mixing plant) against corrosion.

537
Another specific issue was the presence of unexploded ordnance in the lagoons
from WW II, which required attention during excavation of the lagoons.

Validation

Internal validation of the treated materials was achieved using the CBR (Californian
Bearing Ratio ASTM D1883-99) after 1 month of curing.

Contractually, a more extended list of geotechnical tests was employed. The cured
materials were also WAC tested based on the EN 12457 4 leaching test, and
checked for several parameters (see Table 2) that were the most relevant.

The same mechanical tests were carried out after durability testing on the treated
materials (freeze-thaw and wet-dry ageing tests). A maximal strength loss of 15 %
of the required values was allowed.

References

Pensaert, S., De Puydt, S., Janssens, T., Vanpe, N., Vander Velpen, B., De Cock,
C., Goorden, G.,(2008). The Remediation of the Acid Tar Lagoons at Rieme,
Belgium. Consoil 2008, Milano, Italy.

538
Pensaert, S., De Puydt, S., Janssens, T.,(2010). The Remediation of the Acid Tar
Lagoons in the Port of Ghent. 2010 International Solidification/Stabilization
technology forum,Sydney, Nova Scotia, Canada, June 2010.

539
Case Study 34: Roma Street Station

Location: Brisbane, Queensland, AUS

History: Railway station

Contaminants: PAHs

7,000 m3
Scale:
Fly ash & waste ash/coke
Reagents:
Ex-situ
Method:
Parkland
End-Use:

Site description

Roma Street Station is a major railway station in Queensland and has been in
operation since 1874. Railway operations have affected the environmental quality of
the land underlying and adjacent to the station. Part of the area adjacent to the
stations was slated to be redeveloped into parkland. Contaminated soil on this
parcel needed to be addressed before the redevelopment could be initiated.

Characterisation

Contamination at the site comprises both in-situ and stockpiled soils containing
heavy metals, polycyclic aromatic hydrocarbons (PAHs) and total petroleum
hydrocarbons (TPH) in concentrations above regulatory threshold values. The
existing stockpiled soils originated from work associated with the adjacent Roma
Street Station upgrade.

The majority of in-situ contamination at the site is associated with a shallow ash/coke
soil layer to approximately 300mm (12 inches) depth, on average across the site.
This layer comprises heavy metals and PAHs above regulatory threshold values.

540
Risk Drivers

The Roma Street Parkland redevelopment was to create an attractive


garden/parkland of high standard where community and visitor patronage are the
final success factors. The key objective in proactive management of PAH-impacted
material at the site was to satisfactorily reduce the mobility of the PAH contamination
and eliminate exposure pathways so that the associated hazard is reduced to an
acceptable level.

Objectives

The remedial approach for the site was to separate clean from contaminated
material. Contaminated material was then treated on-site to reduce leachable
contaminant concentrations prior to containment within an on-site cell. The
stabilization process incorporated the beneficial reuse of waste ash and ash/coke
from on-site in the treatment process. The approach of maximising reuse of
materials on-site is consistent with the EPA remediation hierarchy which gives
preference to on-site solutions. The reuse of on-site waste also reduced treatment
costs, limited the volume of reagents brought on-site, and minimised the volume of
the on-site cell.

The treated material was placed back into the excavation inside a minimum design
cap and containment system. This placement of the treated material on-site was
consistent with, and did not interfere with, the following redevelopment of the Roma
Street Parklands. The resulting gardens and parkland provide the citizens of
Brisbane with a world-class botanical experience in a former industrial setting.

This remediation was the first organic stabilization project completed in Australia.

Method

The stabilization treatment involved the use of fly ash and waste ash/coke and was
designed to reduce the PAH leachability. The target post-remediation PAH
leachabilities were:

Naphthalene <50 ug/L


541
2-Methylnaphthalene <50 ug/L
Acenaphthylene <10 ug/L
Acenapthene <10 ug/L
Fluorene <10 ug/L
Phenanthrene <10 ug/L
Anthracene <5 ug/L
Fluoranthene <5 ug/L
Pyrene <5 ug/L
Benz(a)anthracene <2 ug/L
Chrysene <2 ug/L
Benzo(b) & (k)fluorene <2 ug/L
Benzo(a)pyrene <2 ug/L
Indeno(1,2,3-cd)pyrene <2 ug/L
Dibenz(a,h)anthracene <2 ug/L
Benzo(g,h,i)perylene <2 ug/L

The target leachable concentration for total PAHs in treated materials was 0.1 mg/L

Validation

Samples of the treated soil were obtained at a frequency of every 100 tonnes treated
and subject to SPLP testing for PAHs. Over 98% of the treated samples met the
performance criteria on the first pass. Those 100 tonne stockpiles not meeting the
performance criteria were retreated.

Equipment

A pug-mill system, consisting of a feed hopper with conveyor, pug-mill, reagent silo,
and stacking conveyor were used to treat 150-200 tonnes of contaminated soil per

542
hour. The treated soil was segregated into 200 tonne stockpiles for confirmation
testing.

543
Case Study 35: Sag Harbor

Location: Sag Harbor, New York

History: Manufactured gas plant

Contaminants: BTEX, PAHs

7,200 yd3
Scale:
Potassium permanganate & Portland
Reagents:
cement
Method:
In-situ

544
Summary

The Sag Harbor Manufactured Gas Plant operated from 1859 to 1930 and originally
produced gas from coal or wood resin. It was switched to a water-gas process in
1892. The by-products of gas production that may have either spilled, leaked, or
were intentionally disposed of on this site are responsible for contamination of the
environment.
The New York State Department of Environmental Conservation (NYSDEC) and the
New York State Department of Health (NYSDOH) announced their decision for a
remedy, which was signed on March 31, 2006.

The chemicals of concern at this site are residues of the former MGP process and
include volatile organic compounds, semi-volatile compounds, and cyanide. The
volatile organic compounds of concern are benzene, toluene, ethylbenzene, and
xylene (BTEX). The semi-volatile organics of concern are polycyclic aromatic
hydrocarbons (PAHs), and the principle waste material at this site is coal tar, a thick,
black, oily liquid which was a by-product of the gas production process. Coal tar has
been found under most of the site and mostly located in the upper 12 ft. The
groundwater, contaminated by contact with the coal tar, moves to the north, almost
as far as Sag Harbor Cove.

Geo-Con teamed with Sevenson Environmental Services and was contracted by


National Grid to provide a total remediation solution for this MGP site. The complete
clean-up effort included:

Installation of 7,200 yd3 of In-Situ Soil Mix Wall (SMW)


Water Table Drawdown (Dewatering)
Water Treatment and Discharging
Impacted Soil Excavation and Disposal
Backfilling

Geo-Con installed over 7,200 yd3 of vertical barrier wall around the perimeter of the
site using the Large Diameter Auger Shallow Soil Mixing (SSM) method. The
purpose of the wall was to provide structural support and a groundwater barrier
during excavation of the impacted soil. Soil was mixed in place using Geo-Cons
custom made Calweld Drill Platform mounted on a Manitowoc 3900 crawler crane.
Soil was mixed with Type 1 Portland Cement/water slurry, and was produced in Geo-
Cons batch plant. A combination of 7, 8, and 10-ft augers were used to complete the
SMW to depths up to 23 ft deep.

545
Due to the nature of MGP waste and this particular sites location (in the heart of the
Village of Sag Harbor), Sevenson/Geo-Con faced a unique set of challenges in the
completion of the project. The site was immediately adjacent to many active
businesses and occupied residences. Health and Safety, odour, and noise control
were paramount throughout the project. Odor control foaming was used continuously
to control MGP odours during pre-trenching and SMW installation, and a series of
structures equipped with vapor collection systems were used during excavation.
Perimeter real-time air monitoring was performed throughout the duration of the
project.

The small size and geographical location of the site provided several logistical
challenges requiring careful coordination of day-to-day construction activities.
Specific trucking routes were established and strictly enforced for all deliveries and
trucks hauling waste for disposal.

The project was completed within budget and ahead of schedule and with no health
and safety related incidents.

546
Case Study 36: Sanford MGP Superfund Site

Location: Sanford, FL

History: Manufactured gas plant

Contaminants: PAHs, DNAPL

125,000 yd3
Scale:
Portland cement & blastfurnace slag
Reagents:
In-situ
Method:
Restoration
End-Use:

Site description

The former Sanford Gasification Plant, a manufactured gas plant (MGP), was
located between 830 and 901 West Sixth Street and between Holly Avenue and

547
Cedar Avenue in Sanford, Florida. The remediation area included the former Sanford
Gasification Plant and a number of properties downstream from the site and
immediately adjacent to Cloud Branch Creek. Cloud Branch Creek which traverses
the remediation work areas discharges to Lake Monroe past the confluence of Mill
Creek located at the northern limit of OU3.

Soil remedial activities planned for the Sanford site by Natural Resources
Technology, Inc. and WRScompass included the demolition of 3 abandoned
structures, excavation of the first 2 ft (20,000 yd3) of unsaturated soils, in-situ
solidification (ISS) of 125,000 yd3 saturated soils, extensive utility relocates as well
as major improvements to Cloud Branch Creek in the form of the installation of
nearly 1,000 ft of 7 ft x 7 ft and 11 ft x 7 ft culverts, realignment of the creek as well
as 450 ft of open channel improvement of the creek located OU3 North terminating
at the confluence of Mill Creek.

Risk Drivers

The contaminated soils required removal/treatment to prevent contact with the


contaminated soil by the residents and to prevent migration of PAHs to Cloud Branch
Creek and Lake Monroe.

Objectives

Backfill the landfill area; removal and disposal of the unsaturated soil
Treatment of the saturated soil down to the clay aquitard
Realignment and improvement of Cloud Branch Creek
Restore the site

Method

Portland cement and ground granulated blast furnace slag were used to improve the
unconfined compressive strength to >50 psi and reduce the hydraulic conductivity to
<1x10-6 cm/sec.

Validation

548
Samples of the treated soil were obtained at a frequency of every 250 tons treated
and subject to UCS and permeability testing. Over 98% of the treated samples met
the performance criteria on the first pass. Those columns not meeting the
performance criteria were retreated.

Equipment

The ISS operations within the site were conducted using a 4000-series Manitowoc
crane equipped with an attached Hain Platform. The crane/platform assembly is
supplemented with a swivel-mounted, top-feeding Kelly Bar capable of reaching a
depth of 75 ft bgs. Augers were attached to the bottom of the Kelly Bar. A 10-ft auger
and a 12-ft auger were utilized on this project.

The appropriate amount of water was metered into an initial 5 yd 3 batch tank
equipped with a high-speed, high-shear mixer. The reagents (Portland cement and
ground granulated blast furnace slag) were transferred from the silos to the batch
tank using the internal screw conveyor to deliver the specified volume of reagent.
The water was added to the mix tank first and the volume of water recorded. Each
reagent was added separately to the mix tank. The scales on which the mix tank
sets were tarred before each reagent was added to verify that the correct amount of
reagent had been added. WRScompass periodically tested each batch being
prepared using a mud balance to insure the proper mix design was being met. The
batch number, volume of water used, and the weight of each reagent added were
recorded on a grout log by the batch plant operator.

When the correct grout composition was achieved, the blended grout was
transferred to the auger. The batch plant was also equipped with a second storage
tank to allow for temporary storage of a blended batch to allow for uninterrupted
production. A high speed mixer in this second tank was to ensure the blended batch
does not separate. The pre-determined grout volume was pumped to the treatment
area based on the soil density, reagent admixture ratio, and the work area
dimensions (i.e. column diameter or panel dimensions).

The ISS Treatment was performed in a series of overlapping columns as per the
schematic below. Columns along the perimeter of the ISS area had a 1 ft overlap in
the area known as the triple treatment triangle, while the interior columns had a neat-
line overlap.

549
WRScompass used a TOPCON 3000 Series Total Survey Station during ISS
operations. This instrument insured the proper overlaps of the columns, the locations
of the columns, the vertical extent of the treatment, and the rate of advancement of
the tool. Using the predetermined column locations, WRScompass placed stakes at
the center point of each column slated for treatment for the days production.

The crane operator would set the auger


tip immediately over the center stake
insuring the proper location. Once the
auger was set up at said location, ISS
personnel verified the proper column
designation and location. Prior to the
initiation of the drilling/ISS operations,
ISS personnel verified the key
parameters for the column (i.e. total
anticipated depth, grout volume needed,
etc.). This information was recorded in
the ISS Master QA/QC Log.

The grout mixing plant personnel were in


constant radio contact with the crane
operator and the QC personnel to insure
proper grout volumes are dispensed and
incorporated into the column.
WRScompass Quality Control Officer
(QCO) communicated with the operator
to verify that the Kelly bar is plum at the start of each column. The QCO verified
vertical depth by surveying the elevation of the top of the Kelly bar (known length)
when the auger is at the top of the mixing area and at the bottom of the mixing area.
When the terminating depth has been reached and the overall grout volume for the
column injected, the auger was extracted and reintroduced to the same column to
complete at least 3 mixing passes total per column to achieve a homogeneous
mixture. The auger was then moved to the next column unless the finished column
was slated for sampling.

In-situ treated material sampling was performed utilizing WRScompass in-situ


sampler. Upon the completion of the ISS column slated for sampling, the in-situ
sampler (depicted below) was lifted by the excavator and advanced to the vertical
midpoint of the column. Once the in-situ sampler reached the sampling depth, the
sampling chamber was opened using a hydraulic actuator. The sample would then
enter the sampling chamber. Once the chamber was filled, it was hydraulically
closed and the in-situ sampler was retrieved.

550
WRScompass ISS Swell Management Plan was to incorporate the ISS swell into the
sites final contours and grades. To the extent practical, all ISS swell was managed
on-site and within the ISS treatment limits. WRScompass would begin grading the
ISS to the sites final contours and grades before the ISS treated material had
started to set. This allowed an on-going determination as to whether or not all of the
ISS swell could be managed on-site and within the ISS treatment limits.

Specific Issues

One of the most sensitive and complex elements of the work was the stabilization of
approximately 250 lineal feet of roadway which carried banks of fiber optic lines.
These lines could not be moved which required WRScompass to stabilize the soils
underneath these lines. WRScompass designed and implemented a method that
successfully treated soils while supporting the fiber optics lines with no loss of
service.

551
Case Study 37: Sderhamn

Location: Sderhamn, Sweden

History: Wood treatment site

Contaminants: Metals, PAHs

31,000 tonnes
Scale:
Zero-valent iron
Reagents:
Ex-situ
Method:
Public park
End-Use:

Site Description

The former wood impregnation site at Sderhamn (Sweden) operated between 1937
and 1997, and was found to be contaminated by PAH and heavy metals (As, Cu,
Cr). This resulted from the use of creosote and CCA (Chromated Copper Arsenate
552
salts) for the impregnation of the wood. The contamination was found both in the soil
and the groundwater, as well as the sediments in the adjacent river.

The presence of various hotspots of heavy metal and creosote contamination was
the driver for the remediation.

191,000 tons of the site were excavated, of which 76,000 tons were sent off-site as
hazardous waste, 31,000 tons treated by chemical stabilization to be reused as
capping material on a local landfill, and the remainder of the soil could be reused on
site. The site was turned into a public park space.

The remediation works started in September 2009 and ended in October 2011.

Objectives

The main objective of the remediation of this site was to remove the contaminant
sources, and to optimise the beneficial reuse of materials excavated from the site.
The soils had to be remediated to below the Swedish standards MKM for metals and
PAHs, as indicated in the Table below.

Remediation standards for metals and PAHs at the Sderhamn site

Pollutant As Cu Cr PAH total


MKM-standard (mg/kg DM) 40 200 250 80

All soils above the remediation standards were excavated and brought to a landfill.
The soils not complying to the landfill leaching standard of 2 mg/kg DM for arsenic
were chemically stabilized.

Method

31,000 tons of soil were chemically stabilized in order to reduce the leaching of
metals, in particular arsenic. Zero-valent iron powder was added at a dosage of 1.5
%. This dosage was determined after an extensive preliminary lab testing
programme, assured a robust process, and avoided reprocessing.
553
The soil was treated by means of a batch mixing plant (double axis pug-mill system,
picture above), with a throughput of about 40 tons an hour.

Validation

Every batch of treated soil (representing about 1,000 tons) was sampled and tested
according to EN 12457-4. The results were compared to the target levels in the
Table above.

554
Case Study 38: Southeast Wisconsin

Location: Southeast Wisconsin, USA

History: Former Manufactured Gas Plant (MGP)

Contaminants: MGP coal tar residuals

33,000 yd3
Scale:
Portland cement and blast furnace slag
Reagents:
In situ auger and bucket mixing
Method:
Future redevelopment (not specified)
End-Use:

555
Summary

Natural Resource Technology provided remedial alternatives analysis, design, and


environmental construction management for a former manufactured gas plant (MGP)
facility located near Lake Michigan in a high profile, residential and commercial
downtown area. Project planning required extensive permitting and coordination
with the City, Wisconsin Department of Natural Resources, state health department
and the community.

A solution was designed to maximize remediation of the MGP impacted material,


both above and below the water table, by in situ stabilization/solidification (ISS) and
excavation. Both auger and bucket mixing techniques were used. ISS allowed cost-
effective treatment of soils to depths greater than what could be achieved with
traditional excavation due to site constraints. All ISS swell material was managed
on-site, as designed. The site remediation area is approximately 3.5 acres in size.

Characterisation

Remedial Alternatives Analysis


Remedial Design
Bench- and Pilot-Scale Studies
Construction Oversight and Documentation

Risk Drivers/Project Constraints

This project presented multiple spatial and logistical constraints including:

Multiple condominiums (high-rise and townhome), hotel, restaurants, and marinas


are located adjacent to the project site.
A 4-story vacant 60,000 ft2 commercial building was located on the site next to
where the remediation was performed.
A parking deck structure required demolition before remedial construction could
begin.
Former MGP structures (large foundations, piping, steel and concrete gas
holders) required removal prior to ISS.
MGP impacted soil and groundwater extended more than 30 ft below ground
surface near the on-site building and adjacent City streets.
Underground electrical, and City storm and sanitary sewers required rerouting
prior to project start.
556
Objectives

Restore the property for future development


Minimize potential for direct contact exposure to the community
Reduce contributions of MGP residuals to groundwater

Method

Over 33,000 yd3 were solidified using auger and bucket mixing techniques near
the on-site building to 30 ft below ground surface.
The contractor (ENTACT) utilized a DELMAG RH-32 caisson-type drill rig
equipped with a 10-ft diameter auger to complete overlapping ISS columns.
A conventional long reach excavator was utilized by ENTACT to complete bucket
mixed ISS grids.
Reagent grout mixing was conducted using an on-site batch plant.
Reagent consisting of a 3:1 mix (slag and Portland cement) mixed in-situ between
8 and 10 % relative to the dry unit weight of soil.
Approximately 21,000 tons of MGP impacted debris and soil was excavated and
landfilled.
Comprehensive perimeter air monitoring was performed.

Validation

Discrete samples were collected from the top, middle or bottom of the selected ISS
columns or grids by ENTACT according to the following frequency:

One discrete sample collected every 1,000 yd3 or once per day of ISS production.
One discrete sample collected every 100 linear feet around the perimeter of the
ISS treatment area.

Samples were tested for unconfined compressive strength and hydraulic conductivity
to demonstrate compliance with the performance objectives.

Benefits of Solution

557
Cost effective solution when compared with other technologies
Reduced vapor phase and fugitive particulate emissions during remediation
Reduced need for long-term management and environmental concerns for future
development
ISS monolith suitable as structural foundation for future construction

558
Case Study 39: St Louis

Location: St Louis, Missouri

History: Ordnance plant

Contaminants: Impacted soils

1,400 yd3
Scale:
Zero valent iron
Reagents:
In-situ
Method:
Not specified
End-Use:

Summary

This project was performed at the former St.


Louis Ordnance Plant in St. Louis, MO.

In-situ stabilization (ISS) work was


performed on soils that were impacted by
contaminated groundwater.

The treatment zone covered approximately


1,400 ft2 of surface area with depths from
the surface to bedrock ranging from 20 to
30 ft.

The project goal was to mix the


contaminated soil with zero-valent iron
(ZVI), and involved an innovative
application and mixing process that was
performed in the dry, i.e. without using
drilling fluids.

559
Treatment involved using a minimum 0.6% ZVI addition, with CH2M Hill confirming
that all soil mixed columns were in compliance with the treatment goals. Dry soil
mixing using ZVI at this site is the first now full-scale application of this type carried
out.

Geo-Solutions performed all the soil mixing activities utilizing specialty equipment,
including a soil mixing rig with a 5 ft diameter mixing tool, vapor
shroud/extraction/treatment system, and ZVI injection tooling.

560
Case Study 40: Sunflower Army Ammunition Depot

Location: DeSoto, KS

History: Ammunition depot

Contaminants: Explosives, propellants, lead

70,000 tons
Scale:
Portland cement
Reagents:
Ex-situ
Method:
Restoration
End-Use:

561
Site description

The 9065-acre Sunflower Army Ammunition Plant (SFAAP) was a government-


owned, contractor-operated military installation. Solid Waste Management Units
(SWMU) 10 and 11 consisted of the Blender and Roll House Area, F-Line Press
Buildings, and the F-Line ditches and settling/blender ponds. The F-Line served as
the final production area for N-5 propellant. The F-Line area had 21 ditches, which
originate at the production buildings and terminated into 3 pairs of settling ponds.
The 21 ditches are generally 10 ft deep with relatively steep banks. The 3 pairs of
settling ponds were unlined earthen impoundments equipped with stand pipes to
permit settling of solids and decanting of water. Two blender ponds were located
northwest of the production area. The ditches and ponds received wastewater from
the manufacturing process. The wastewater was contaminated with lead salts,
propellant waste, nitro-glycerine and nitrocellulose from the manufacturing process.

Objectives

Remove the contaminated soil


Treat to reduce the leachability of lead and reduce the reactivity of the explosives
and propellants
Stockpile the treated soil and transport to a local landfill for use as daily cover
Restore the site

A total of 46,000 tons of material was excavated from SWMUs 10 and 11. 22,000
tons of soil was removed from the 62 Roll/Blender House building foundations in
Areas A through H. A total of 19,000 tons of soil was removed from 22 F-Line
drainage ditches and ditches A through F. 5,000 tons of soil was removed from the
bottoms and sides of 8 settling/blending ponds. Post-excavation sampling and
analysis performed at the designated sample locations verified that the excavation
attained the remediation cleanup goals established for this site.

The concentration of propellant per unit volume of excavated material was sufficient
to propagate detonation only at some locations along the ditches, so the physical
nature and potential extent of energetic hazards associated with propellant
warranted special consideration. Therefore, the remedial action was worked at a
productive and efficient pace, while maintaining a safe working environment for site
personnel. The excavator cabs were outfitted with a specialized lexan blast
protective shield, customized to protect the glass windshield and prevent injury to the
operator. A 2,000-gallon water truck was readily available to provide continuous
water mist to the excavation area in an effort to minimize the potential for ignition.

562
Method

The contaminated soil was treated with Portland cement, producing a treated
material that did not exhibit a hazardous characteristic for leachability (primarily lead)
or ignitability (due to the presence of explosives and propellant).

Validation

Temporary 500 ton stockpiles of the treated soil were created. Confirmatory testing
to verify that the treated material that did not exhibit a hazardous characteristic for
leachability (primarily lead) or ignitability (due to the presence of explosives and
propellant) was performed on the sample from each temporary stockpile. All of the
confirmatory samples met the performance requirements for leachability.

Equipment

Contaminated soils were blended using a Findlay 393 Hydrascreen with a pre-cutter
pulverization unit. Once the material was screened to remove >2-in debris, the
homogeneous material was stabilized in the RapidMix 400 pug-mill using Portland
cement. Production rates through the pug-mill of over 1,000 tons per day were
routinely achieved

563
Case Study 41: Sydney Tar Ponds

Location: Sydney, Nova Scotia, Canada

History: Scrap yard

Heavy metals, PAHs, PCBs, VOCs


Contaminants:
700,000 tonnes (PAH affected)
Scale: 50,000 tonnes (PCB affected)

Portland cement, fly ash & slag

Reagents: In-situ

Method: Restoration

End-Use:

564
Background

The Sydney Tar Ponds are centrally located within Sydney, Nova Scotia, as shown
in Figure 1. The Tar Ponds cover an area of approximately 33 Hectares (81 acres).
They are what remain of the Muggah Creek tidal estuary after nearly 100 years of
steel and coke production activities. The sanitary sewage of the surrounding
communities also drained into this estuary prior to the cleanup. Preliminary works by
the municipal government now collect and treat the roughly 13 million liters a day
that once discharged into the ponds, through a modern treatment facility, which
discharges into Sydney Harbour. Two significant tributaries Wash Brook and Coke
Oven Brook drain the urban watershed and discharge into the Tar Ponds (AECOM,
2008-B).

The Sydney Tar Ponds consists of the North Tar Pond and the South Tar Pond). The
dividing line between these 2 ponds is the causeway and bridge at Ferry Street.

Figure 1: Location of the Tar Ponds site in Sydney Nova Scotia, Canada
(AECOM, 2008-B)

Over 100 years of steel making and coking industries in Sydney resulted in
deposition of coal tar, fine coal, and sediment within the ponds. The majority of the
sediments in the Tar Ponds were transported by Coke Oven Brook from the Coke
Ovens site. Contaminants in the sediments include heavy metals, polycyclic aromatic
hydrocarbons (PAHs), volatile organic compounds (VOCs), and polychlorinated
biphenyls (PCBs).

565
Figure 2: Contaminated Sediment in South Tar Pond

The overall work required to remediate the Sydney Tar Ponds and Coke Ovens Sites
was extensive, complex and distributed across a large area. Prior to this
remediation project, there had been 2 failed attempts to design and implement a
remedial solution. In order to make this project manageable, the project was split
into approximately 13 sub-projects or elements. The Solidification/Stabilization
(S/S) of the Tar Ponds was the largest element in both scope and value ($75m).

Project Funding, Stakeholders and Management

On May 12, 2004, after decades of study and 2 failed cleanup attempts the
governments of Canada (federal) and of Nova Scotia (provincial) provided $400m in
funding to clean up the sites (MOA 2004). The funding for the clean-up was split
between the Canadian federal government (70%) and the Nova Scotia provincial
government (30%). A number of major stakeholders maintained an active interest in
the progress of the project and provided the necessary checks and balances to
ensure its success.

As project implementer, the Sydney Tar Ponds Agencys (STPA) fundamental role
was to direct, procure, and implement the services needed to complete the project,
while protecting the interests of the funding partners. Other key project team
members included:
Independent Engineer (CRA) - validate the technical merits of the project, and
to report on costs incurred, cost to complete, and contract compliance
Stakeholder groups such as the Environmental Management Committee
(EMC), Community Liaison Committee (CLC), Operations Advisory
Committee (OAC), Remediation Monitoring Oversight Board (RMOB)
Design Engineer (AECOM) - design the project, and oversee the execution of
the design
Independent Quality Assurance Consultant Stantec Consulting Ltd

566
Design of Solution

The Design Engineer (AECOM) was engaged in October 2006 and immediately
began work on selecting a remedial option for the North and South Tar Ponds, as
well as for the other project elements. The remedial option chosen for the Tar Ponds
sediments was in-situ Solidification/Stabilization (S/S) with cement. In addition, an
engineered channel would be constructed through the Tar Ponds to permit the Coke
Ovens Brook and Muggah Creek to flow through the remediated Tar Ponds to
Sydney Harbour. After S/S was completed the area would be covered with an
engineered cap. To permit S/S in the dry an extensive water control management
system would be constructed to control any water intrusion from the ocean and
capture/pump incoming streams sequentially around the working areas (AECOM,
2008-B).

Remediation

Pre-S/S Work

Prior to S/S of the North and South Tar Ponds there were a number of tasks that had
to be carried out including the construction of access roads along the perimeter of
the ponds. The two key tasks that were required prior to S/S activities were as
follows:

Bench-Scale Testing and Pilot Study

As part of the Bench-Scale Testing, the Design Engineer (DE) collected numerous
sediment samples from both the North and South Tar Ponds. In the laboratory, the
DE mixed sediment samples with potential S/S reagents such as cement, slag
and/or flyash. The goal of the Bench-Scale Testing was to find reagent(s) and
sediment mixtures that could meet the project S/S performance criteria in the
laboratory. Utilizing the information gained from the Bench-Scale Testing, a Pilot
Program was carried out. The Pilot Project utilized an in-situ S/S approach using
interlocking steel sheet pile cells in the North and South Tar Pond. Seven cells were
constructed in the South Tar Pond and 6 cells were constructed in the North Tar
Pond, each with an approximate surface area of 27 m2. After the cells were
homogenized using an excavator, known quantities of various combinations of
cement, slag and/or flyash were mixed with the cell sediments. After mixing, samples
were taken and analyzed with respect to meeting the desired performance
567
requirements (see Tables 1 and 2). The Pilot Program indicated that there were a
number of reagent variations that had the potential of meeting the performance
criteria. A report of the Pilot Program was prepared and made available to the
potential bidders on the S/S project.

Dewatering of the S/S Work Areas

The Tar Ponds are a tidal estuary with 2 large brooks (Coke Ovens and Wash
Brook) flowing into the South Tar Pond. To carry out S/S it was necessary to
dewater the work areas prior to the S/S contractor beginning work in the area. To
achieve this pre-S/S dewatering, a contractor (Beaver Marine/MB2 Joint Venture)
was engaged by STPA to implement the Design Engineers dewatering design. To
create dry work areas in the South and North Tar Ponds a multi-stage sequencing of
pump stations was used. The pumping stations were comprised of inlet works,
pumping equipment, 48-in pipelines for conveyance of the flows and discharge
facilities with energy dissipation structures. The dewatering design initially involved
pumping the 2 brooks around the South Tar Pond with the discharge introduced into
the North Tar Pond. Once the flows had been diverted, pumping of the remaining
water above the sediments was the responsibility of the S/S contractor. After
completing the S/S of the South Tar Pond and construction of the channel, the
pumping infrastructure would be moved to drain the upper portion of the North Tar
Pond and then again to drain the lower portion of the North Tar Pond. The pumping
stations were a combination of electric and diesel pumps and had to be able to
handle a combined high peak flow in the range of 14 m3 per second or 3,698 gallons
per second (AECOM, 2008-A).

Figure 3: South Tar Pond Dewatering

568
Solidification/Stabilization (S/S)

The S/S tender was issued in early 2009 and the successful contractor Nordlys
Environmental LP was awarded the contract to carry out the work in June of 2009.
Nordlys Environmental LP was a limited partnership between a local contractor J&T
Van Zutphen Construction Inc. and ECC of Massachusetts. The contract was a
Performance Based Contract (PBC) in that the contractor only received a payment
for a specified S/S volume of sediment (or cell) when it successfully passed the 3
performance criteria (see Table 1). The contractor engaged experts to carry out
Quality Control (QC) on all the cells while STPA engaged an Independent Quality
Assurance Consultant, Stantec Consulting Ltd, to perform Quality Assurance (QA)
verified through random testing of approximately 10% of the cells. Stantec followed
the same procedures as the contractor for acquiring sample carry out as their QC
consultant. Failure of either the quality control (QC) or the quality assurance (QA) of
any of the performance criteria meant the cell failed and had to be remixed prior.
The remixed cell would have to pass re-testing in order for the contractor to receive
payment. Table 2 lists the 46 parameters tested using the Synthetic Precipitation
Leaching Procedure. Table 3 compares the frequency of QC and QA tests for each
test type for each cell.

Table 1: Contractual Performance Criteria for S/S Work by Contractor


(Ingraham, 2011)

Property Test Method Criterion


Unconfined
0.17 MPA
Compressive Strength ASTM D1633 Method B (modified)
(25 psi)
(UCS)
Hydraulic Conductivity 1 x 10-6
ASTM 5084 (Flex Wall)
(Perm) cm/sec
Modified Synthetic Precipitation
Site specific
Leaching Procedure 1312 (monolithic
Leachability (SPLP) leachate
structural integrity procedure) to check
criteria (SSLC)
48 site-specific compounds

569
Table 2: Parameters tested using Synthetic Precipitation Leaching Procedure
(SPLP) during the remediation

SPLP Parameters

Modified
Copper Strontium Dibenzo(a,h)anthracene
TPH (Tier 1)

Aluminum Iron Thallium Fluoranthene

Antimony Lead Tin Fluorene

Arsenic Lithium Uranium Indeno(1,2,3-cd)pyrene

Barium Manganese Vanadium Naphthalene

Beryllium Mercury Zinc Perylene

Boron Molybdenum Benzo(b)fluoranthene Phenanthrene

Cadmium Nickel Benzo(g,h,i)perylene Pyrene

Chromium Selenium Benzo(k)fluoranthene Total PAH

Cobalt Silver Chrysene Total PCB

Table 3: Sampling and Testing Requirements for Quality (initial tests at 28


days; retests at 56 days if needed) (Ingraham, 2011)

Cell Thickness & Sample


Test QC Frequency QA Frequency Frequency
Type (Minimum) (Minimum) Sediment 2m Sediment > 2m
2 finite location 1 homogeneous
All (~250m3) cells; Every 10th cell; 8 homogeneous
UCS sample from top
8 cylinders/cell cylinders/cell samples
& 1 from bottom
All (~250m3) cells; Every 10th cell; 2 1 composite 1 composite
Perm sample
2 cylinders/cell cylinders/cell sample
Cells 1-20: every Cells 1-20: every
SPLP cell 10th cell;
(Pre- & Cells 21-40: every Cells 21-40: every 1 composite 1 composite
Post- 2nd cell 20th cell sample sample
mixing) Cells 41+: every Cells 41+: every
4th cell 40th cell

570
Mixing the Cells

Each cell was mixed following the method agreed upon by the contractor, the Design
Engineer and the STPA. The contractor used hydraulic excavators to mix the
solidifying agents (Portland cement) and gravel/slag, if required, according to an
approved recipe. The following summarizes the steps utilized for in-situ mixing:

Contractor provides location of a cell to be mixed


Design Engineer (DE) confirms predicted bottom of cell using contractor provided
center of cell and Digital Terrain Model
DE visually confirms with the contractor the Field Confirmed Bottom of Sediment
Elevation while In-Field
Cell volume sized to received 1 tanker load of cement
Contractors excavator operator mixes the cement into the impacted sediments in
the cell. Cells are interlocking
While each cell is being mixed the DE observes operations to ensure contractor
compliance with the contract documents and standard operating procedures (i.e.
Daily Diary, Mixing Oversight Checklist)
Contractor takes bulk samples from which specimens are taken and tested
Where any non-compliance is found a Request for Action (RFA) or Non-
Conformance Report (NCR) is issued to ensure the Contractor corrects the issue
In a timely manner the DE responds to any Request for Information (RFI) the
contractor may have to ensure compliance to design
DE confirms cell quantity and compliance to contractual requirements
Additional Oversight Tasks Include: Confirming odour observations reporting,
water management observations, dust control, noise, safety and environmental
protocols, quality control observations, contract administration, change order
observations, noise and dust management observations

571
Figure 4: In-situ S/S with dust control conveyance boxes and odour control
foam (white) and pressure washer application of bio-solve

Special Challenges

As one might expect there were a number of special challenges associated with
carrying out a large S/S project in a tidal estuary in the center of a city. Below is a
summary of some of the challenges encountered and the techniques utilized to
overcome these challenges:

1. Water Management Issues associated with high groundwater flows and


close proximity of ocean
Although an issue throughout the project, this was a major challenge in the final
phase (Phase 3) which located adjacent to the ocean. A combination steel sheet
pile wall and armour stone barrier provided a physical separation and fairly
impermeable barrier but the tidal water moved relatively freely through the existing
slag area bordering the east side of the pond. The contractor strategically placed
clay and S/S berms in areas to aid dewatering efforts. This enabled S/S and
channel work to begin in specific locations while effective water management was
being carried out in other locations.

2. Constitution of impacted sediments varied throughout the North and South


Tar Ponds
The type, location, and amount of contaminant as well as the water content of the
sediment varied throughout the north and south ponds. In order to overcome
these issues, the contractor was constantly observing and analyzing the sediment
for these changes. This permitted the contractor to recognize issues prior to
mixing and make approved adjustments to mix deign and duration of mixing.

572
3. Sediment depths deeper than predicted in some areas
As previously stated, the contractors primary mixing methodology was bucket
mixing. In deeper areas (> 16 ft) bucket mixing began to be problematic resulting
in some cells not meeting the criteria on the first S/S attempt. After discussions
with stakeholders, the contractor began utilizing an ALLU type mixer in deeper
areas. The Allu mixer proved to be very effective in deeper areas.

4. Odours
Odour generation during S/S became a significant issue due to a combination of
sediment disturbance/drying, the exothermic cement/sludge reaction and general
warm weather mixing. Given that the work was being carried out in the middle of
Sydney, significant odour generation was not acceptable. Excavation of S/S
material for channel construction was especially odorous. A comprehensive
odour monitoring and management program was developed which involved
utilizing odour suppression materials such as foams, liquid sprays and surface
cover materials. In addition, work practices were altered to minimize odour
generation. For example, sediments were permanently relocated out of the
channel footprint prior to S/S.

5. Dust
As with odour, significant dust generation within the city was not acceptable.
Cement dust can pose a potential health and safety risk on-site and off-site. As
part of the dust control methodology, the contractor had a dust budget with pre-
defined dust generation limits. As the dust budget was approached or exceeded,
the contractor would stop work and alter work procedures to lessen dust
generating practices. In a conscious effort, the contractor utilized a cement
conveyance box which was fitted on top of the sediment to be mixed and then
powder cement was blown into the box prior to mixing. The contractor then
utilized significant wetting techniques during the mixing process.

6. Cell Sediment Volume and Cement


Initially each cell was of the same spatial size, however in the field it was soon
realized that mixing a fraction of a cement tanker load into a cell was not practical.
The procedure was modified to a volume-based approach so that the volume of
the cell required 1 tanker load of cement per cell.

Documenting Results

573
The contract required a cradle to grave document trail. At completion of the project
it was estimated that over 70,000 pages of quality control documentation alone was
generated and reviewed. Each of the 3,486 cells has a unique set of data associated
with it (See Table 4).

Table 4: Documentation captured for each cell (Ingraham, MacNeil, McNeil,


MacDonald, MacCormack, & Francisco, 2011)

Documentation: Data Contained:


Mixing Oversight Checklist DE verification of Key Cell Data and conformance to
Contract and SOPs
Cell Profile Center Co-ordinate Wet Density
Cell Co-ordinates Sediment Thickness
Bottom Elevation Cell Surface Area
Pre-Top Elevation Wet Volume
Post-Top Elevation Additional Soil Volume
In-situ S/S (ISS) Report Mix Time Hours of Equipment Used
for Mixing
Reagent Quantities
Cement Receipt Cement quantity
Test Reports UCS (x4), PERM and SPLP (x46 if applicable) test
results

Summary and Future Site Use

Stabilization of the final phase of the S/S project (phase 3) was completed on
January 16, 2013, with a total of 253,671 m3 mixed. Overall for all 3 phases of the
project, 679,016 m3 of sediment were treated and over 100,000 laboratory quality
control/assurance tests were carried out. The S/S component of the project was
completed ahead of schedule and under budget at a cost of $73.5m.
574
After completion of the S/S component of the project, the S/S material was capped
with a low permeability clay cap and then a Future Site Project was carried out which
created what is now called Open Hearth Park. As seen in Figure 5, Open Hearth
Park includes a common area, an outdoor concert venue and skating area, an all
weather multi-use sports field, natural turf field, a playground, a bike training facility,
off-leash dog park, public art displays and a trail network complete with bridges that
once again provides connectivity between downtown Sydney and the communities of
Whitney Pier, Ashby and Sydney`s North End.

Figure 5: Open Hearth Park located on the former Sydney Tar Ponds site

Below is a quote from the Honourable Diane Finley, Minister of Public Works and
Government Services in 2013 on the $400m Sydney Tar Ponds Cleanup:

"This has been the most successful contaminated site remediation project in
Canada's history and we're proud that it was completed on time and on
budget."

Bibliography

575
AECOM., (2008-A). Remediation of the Tar Ponds and Coke Ovens Sites Design
and Construction Oversight Services 100 Percent Design Report TP6A.
Sydney, NS: AECOM available through Sydney Tar Ponds Agency.
AECOM., (2008-B). Remediation of the Tar Ponds and Coke Ovens Sites Design
and Construction Oversight Services 100 Percent Design Report TP6B.
Sydney, NS: AECOM available through Sydney Tar Ponds Agency.
Ingraham, D., (2011). Managing Quality on the $400M Sydney Tar Ponds and Coke
Ovens Cleanup Project. 3rd International/9th Construction Specialty
Conference (pp. CN-061-1 to CN-061-10). Ottawa, ON: Canadian Society of
Civil Engineers.
Ingraham, D., (2011). Sydney Tar Ponds and Coke Ovens Remediation Update -
Phases I & II and Moving Forward. Presentation to Randle Reef
Contaminated Sediment Remendiation Project Members and Environment
Canada. Sydney, NS.
Ingraham, D., & McNeil, W., (2010). Quality on a Multi-Year Remediation Project.
2010 S/S-Tech: International Solidification/Stabilization Technology Forum.
Sydney, NS.
Ingraham, D., & McNeil, W., (2013). A Strong Quality Program - Imperative to
Environmental Remediation Success. Remediation Technologies Symposium
2013 - RemTech(TM) 2013. Banff, AB.
Ingraham, D., MacNeil, J., McNeil, W., MacDonald, J., MacCormack, S., &
Francisco, R., (2011). 100% Performance Demands Collaborative Quality
Management. RemEast 2011 Managing Challenging Environment
Remediation Projects Case Studies. Halifax, NS.
MacNeil, J., Ingraham, D., Gangopadhyay, S., & Van Zutphen, V., (2012). Managing
Program and Project Requirements During Sediment Stabilization of the
Sydney Tar Ponds. Real Property Institute of Canada - Federal Contaminated
Sites National Workshop. Halifax, NS.
Parker, M., Mcneil, W., Ingraham, D., & White, D., (2012). Sydney Tar Ponds and
Coke Ovens: Integrating Broad Socio-Economic Considerations into a
Contentious Remediation Project. 37th CLRA / ACRSD Annual General
Meeting "Seeding Change Cooperative Reclamation". Sydney, NS.

576
Case Study 42: Umatilla Ammunition Depot

Location: Hermiston, OR

History: Explosive disposal area

Contaminants: Explosives, metals

32,800 tons
Scale:
Portland cement & activated carbon
Reagents:
Ex-situ
Method:
Restoration
End-Use:

577
Site description

Five sites at the Umatilla Army ammunition depot were selected by the U.S. Army
and the U.S. EPA for remedial action. Three of the sites were contaminated with
explosives. One of the sites was the TNT Sludge Burial and Burn Area where TNT-
containing sludges from the Explosive Washout Plant may have been dumped and
burned. The other site consisted of open burning trenches and pads where a variety
of debris, ordnance waste, and explosives sludges were burned.

The remediation contaminants of concern for the sites were antimony, arsenic,
barium, beryllium, cadmium, chromium, lead, cobalt, thallium, copper, nickel, silver,
zinc, 1,3,5-TNB, 2,4-DNT, RDX, 2,4,6-TNT, HMX, and Tetryl. The remedial action
selected for the contaminated soil at these sites was stabilization, after
bioremediation of the explosives was demonstrated to be cost-prohibitive.

Objectives

Remove the contaminated soil


Treat to reduce the leachability of metals and explosives
Place and compact the treated soil back into the excavation area
Cap the site with a low permeability clay
Restore the site

Method

The contaminated soil was treated with Portland cement and activated carbon to
produce a treated material that met the performance criteria shown below.

Compound Leaching Requirement


(g/L)

Barium 100,000

Chromium 5,000

578
Copper 140,000

Lead 5,000

Zinc 1,100,000

1,3,5-TNB 180

2,4-DNT 130

RDX 200

2,4,6-TNT 200

HMX 40,000

Validation

Temporary 75 ton stockpiles of the treated soil were created. Confirmatory testing to
verify that the treated material met the leachability performance criteria was carried
out. 8 (1.8%) of the 437 production lots failed for TCLP-leachable explosives. This
material was reprocessed to meet the TCLP-leachable explosives criteria.

Equipment

A pug-mill system, consisting of a feed hopper with conveyor, pug-mill, reagent silo,
and stacking conveyor were used to treat 150-200 tons of contaminated soil per
hour. The treated soil was segregated into 75 ton stockpiles for confirmation testing.

579
Case Study 43: Valero Paulsboro Refinery

Location: Paulsboro, NJ

History: Refinery

Contaminants: Pb, As, Cd, V, Zn, TPH

18,000 yd3
Scale:
Proprietary binder
Reagents:
In-situ
Method:
Restoration
End-Use:

580
Site description

Valero Energy Corporation performed closure of the North Recycle Pond (NRP) area
located on the Paulsboro Refining Company LLCs refinery in Paulsboro, New
Jersey under an Administrative Consent Order (ACO) with the New Jersey
Department of Environmental Protection (NJDEP). The overall project scope not only
involved the closure of the NRP area through in-situ soil stabilization with chemical
reagents, but also excavation of sediment in the West Recycle Pond A (WRP-A) and
stabilization of this material in the NRP.

Characterisation

The total volume of NRP and WRP-A sediments requiring stabilization were 10,000
and 8,000 yd3 (cy), respectively. The sediments primarily had elevated levels of
volatile organics and metals above the NJDEP Soil Cleanup Criteria (SCS) and/or
Groundwater Quality Standards (GWQS).

Risk Drivers

Public housing was constructed on top of soil contaminated from the past foundry
activities. The contaminated soils required removal to prevent contact with the
contaminated soil by the residents.

Objectives

The first phase of work involved excavation of the WRP-A sediment and relocating
the material to the NRP. To accomplish this, it was necessary to pre-condition the
WRP-A sediment with reagent (Calciment).

The WRP-A also required dewatering using a 1,200 gallon per minute well point
system to lower the water table to target depths. Since the WRP-A received the
majority of the refinery runoff, a storm water by-pass system was installed consisting
of 300 linear feet of 34-in x 54-in elliptical RCP piping to intercept storm water flow
from the A Sump and divert it to adjacent pond WRP-B.

The conditioned WRP sediment was transported over 1 mile on refinery roads and
deposited into the NRP. Excavators then blended this relocated sediment with
581
existing NRP sediment. Once the sediments were homogenized, the surface was
surveyed. This top surface along with the pre-determined bottom defined the
stabilization depth, which was uploaded into GPS equipment and GPS controls were
mounted on excavators to provide quality control during mixing. Also, prior to
stabilization, the NRP was dewatered, which was accomplished using sumps and
dewatering trenches.

Stabilization was preformed typically in 700 cy cells using the reagent mix design
that was established via bench-scale testing performed by WRScompass prior to
mobilization. The stabilization reagents used included moist ferrous sulfate, Portland
cement, and granular blast furnace slag, which were placed directly over a treatment
cell then mixed with an excavator to homogenize the reagent with the sediment
throughout the cell. Post-treatment samples from each cell were analyzed to verify
achieving the treatment criteria for unconfined compressive strength and
permeability as well as SPLP testing for select metals of concern to satisfy stringent
NJDEP impact to groundwater cleanup criteria.

Post-treatment activities included importing and placing 23,000 cy of common fill


over the treated sediment in the NRP and installing a stone cover. Finally, a storm
water inlet was installed within the graded pond with a 15-in RCP gravity pipe, which
discharged to a nearby reconstructed lift station.

Method

The sediment material in the WRP-A was solidified with CalcimentTM (a proprietary
cement kiln dust source) to allow excavation and transport of the solidified sediment
to the NRP. The combined solidified WPA sediment and the NRP sediment was
stabilized with a combination of Portland cement, ground granulated blast furnace
slag, and ferrous sulphate to chemically immobilize the metals so that the SPLP
leachate of the final treated material met the New Jersey Class II Groundwater
Quality Standards.

Validation

Samples of the treated soil were obtained at a frequency of every 700 cy, treated
and subject to SPLP testing for metals along with UCS and permeability testing.
Over 90% of the treated samples met the performance criteria on the first pass.
Those treatment cells not meeting the performance criteria were retreated.
582
Equipment

An excavator was used to mix the WRP-A sediment with the Calciment. The same
excavator was used to mix the combined solidified WRP-A sediment and NRP
sediment with the Portland cement, ground granulated blast furnace slag, and
ferrous sulphate to produce the final treated material.

583
Case Study 44: Waukegan

Location: Waukegan, Illinois

History: Not specified

Contaminants: TCE

8,900 yd3
Scale:
Peat moss/bentonite slurry
Reagents:
In-situ
Method:
Not specified
End-Use:

Summary

The purpose of this project was to use single auger


soil mixing to treat TCE impacted soils in-situ. Geo-
Solutions (GSI) was subcontracted by CH2M Hill to be
the prime contractor for this phase of the remediation.

GSI completed all portions of the work relating to soil


mixing and subcontracted some of the general site
preparation and earthwork items.

584
The soil mixing required the installation of 224 9-ft (2.74 m) diameter soil mixed
columns to depths of 18 ft25 ft (5.5 m 7.6 m) below workpad elevation. Each
column was mixed with zero valent iron (ZVI) and bentonite slurry. The ZVI is meant
to reduce the TCE to less harmful constituents, whereas the bentonite slurry lowered
permeability, thereby reducing groundwater flow through the contaminated zone. A
representation of the column layout for the soil mixing work is shown above.

All of the soil mixing work was completed using GSIs Delmag RH-18 drill rig, batch
plant, and silos (see photo).

The mix area post-mixing (shown above) was almost completely inaccessible to
equipment. The material was allowed to consolidate over a few months before the
final geotextile cap was placed.

585
Case Study 45: West Doane Lake Site

Location: Portland, OR

History: Chemical works discharge pond

Contaminants: PAHs, DNAPL

22,000 yd3
Scale:
Portland cement & additives
Reagents:
In-situ
Method:
Restoration
End-Use:

586
Site description

West Doane Lake is a long slender lake, approximately 1,000 ft long, that is oriented
north-south, adjacent and parallel to the BNSF embankment and adjacent to a
former pesticide manufacturing facility. West Doane Lake has received soil,
sediment, surface water, and groundwater from the former pesticide manufacturing
facility. The southern portion of the lake is approximately 60 ft wide, and the
northern portion of the lake is approximately 40 ft wide. The southern portion of the
lake is deeper than the northern portion, with typical water depths of 1 to 2 ft. The
northern portion of the lake is often dry during the summer months.

Characterisation

The West Doane Lake sediments primarily consist of very soft to soft, black to gray,
overbank silt deposits. Permeability test results indicate a range of 10 -5 to 10-8
cm/sec for particle sizes ranging from clayey sands to high plasticity silts. Coarse
materials, believed to have sloughed off the railroad embankment, were often
encountered on the northern edge of West Doane Lake, along with black and gray
sands, believed to be foundry sands from a nearby property. The impacted
sediment depth was approximately 11 ft. Debris (brick, gravel, wire, concrete, and
battery casings) was observed on the eastern edge of West Doane Lake from
historic filling activities. Discontinuous non-aqueous phase liquid (NAPL) blebs were
observed in multiple borings completed within West Doane Lake.

Contaminants of interest (COIs) include chlorinated pesticide, chlorinated herbicides,


PAHs, VOCs, and metals.

Risk Drivers

West Doane Lake may be a potential continuing source of constituents of interest to


potential human and ecological receptors at the site. Both Oregon DEQ and DFW
consider West Doane Lake to be an important resource for migratory birds and water
fowl and consider the maintenance of a healthy aquatic ecosystem to support
migratory birds and waterfowl as an endpoint for evaluating West Doane Lake. ISS
at West Doane Lake will significantly reduce the potential contribution of COIs via
groundwater because the treatment will effectively bind the COIs within the stabilized
monolith to minimize leaching.

587
Objectives

Removal of the surface water


Treatment of the sediment down to 14 ft bgs
Restore the water level

The work performed by WRScompass consisted of the following activities:

Site preparation activities were performed that included pre-work topographic


surveying and existing conditions surveys of adjacent facilities, locating existing
utilities using a combination of utility locator services, installing erosion controls,
setting up a support zone and contamination reduction zone, performing clearing
and grubbing, and constructing a material staging area and a batch plant area.
A dam was placed on the inlet side of West Doane Lake and the water level
lowered until the contaminated sediment was exposed.
ISS was performed to the required depth of 14 ft using an excavator-mixing
method. A grout plant was set up and the required reagent grout was produced in
an on-site batch plant then conveyed to the ISS treatment cell, where it was
added on a per weight basis using a pre-determined mix design of 18% by weight
for Portland cement and 3% bentonite. Activated carbon and organoclay were
added dry to the surface of each treatment cell at 2% and 3% by weight,
respectively. ISS was performed on over 22,000 cy of soil.
A multilayer capping system of imported earthen and geosynthetic material with
storm water collection was installed on the treated sediment. A structural fill layer
was placed over the stabilized sediment and WDL banks to seal the surface and
provide a working subgrade. The structural fill was overlain by a geosynthetic clay
liner (GCL) and a 60-mil (60/1000 inch) HDPE geomembrane, a 1-ft thick layer of
sand, and a layer of filter geotextile. Structural fill was placed over the geotextile
in compacted layers to the desired final cap grades.
All equipment and temporary facilities were decontaminated and removed from
the site and support areas were restored.

Method

A combination of Portland cement, bentonite, activated carbon, and organoclay were


used to chemically immobilize the chlorinated pesticides, chlorinated herbicides,
PAHs, VOCs, and metals contamination and to improve the unconfined compressive
strength to >50 psi and reduce the hydraulic conductivity to <1x10 -7 cm/sec.

588
Validation

Samples of the treated soil were obtained at a frequency of every 500 yd 3 treated
and subject to UCS and permeability testing. The ISS performance criteria were >50
psi UCS and <1x10-7 cm/sec permeability. All of the treated samples met the
performance criteria on the first pass.

Equipment

The ISS operations were conducted using an excavator and a batch plant. The
batch plant was used to prepare the reagent grout. The appropriate amount of water
was metered into an initial 5 yd 3 batch tank (equipped with a high-speed, high-shear
mixer) and recorded. The bentonite and organoclay were added to the batch tank
from supersacks. Each reagent was added separately to the mix tank and the
scales on which the mix tank sets were tarred before each reagent was added to
verify that the correct amount of reagent had been added. When the correct grout
composition was achieved, the blended grout was transferred to the excavator. The
pre-determined grout volume was pumped to the treatment area based on the soil
density, reagent admixture ratio, and the treatment cell dimensions to add 3% by
weight each of bentonite and organoclay.

Portland cement was pneumatically transferred onto the surface of each treatment
cell, along with supersacks of activated carbon to achieve 18% and 2% by weight,
respectively.

When both the grout and the dry reagents had been added to a treatment cell, the
excavator mixed the sediment and reagents until visibly homogenous.

589
Case Study 46: X-231B Pilot Study

Location: Portsmouth, OH

History: Oil biodegradation unit

Contaminants: TCE, 1,1,1-TCA

210 yd3
Scale:
Portland cement & activated carbon
Reagents:
In-situ
Method:
Pilot-scale trial
End-Use:

Site description

The X-231B Oil Biodegradation Unit is located in the Portsmouth Gaseous Diffusion
Plant (PORTS), a U.S. Department of Energy production facility in Piketon, Ohio.
The X-231B Unit encompasses 0.8 acres and was reportedly used for the treatment
and disposal of waste oils and degreasing solvents from 1976 to 1983. From 1989 to
1990, efforts were made to close the X-231B Unit in compliance with RCRA
requirements. Site characterization activities revealed the presence of several VOCs
[e.g. trichloroethylene (TCE) and 1,1,1-trichloroethane (TCA)] in fine-textured soils
from the ground surface to a depth of 25ft.

Risk Drivers

TCE at levels higher than the Federal drinking water standard was found in the
shallow groundwater directly beneath and 750 ft down gradient from the X-231B
Unit. Concerned over the continuous release of contaminant VOCs into the
groundwater, the Ohio Environmental Protection Agency (Ohio EPA) required that
soil remediation be included in the closure of the X-231B Unit.

590
Objectives

The objective of the pilot test was to demonstrate that in-situ S/S could immobilize
the leachable VOCs while improving the physical strength and permeability of the
treated material.

Method

A combination of Portland cement and activated carbon were utilized to chemically


immobilize the VOCs, while producing >50 psi unconfined compressive strength and
<1x10-6 cm/s hydraulic conductivity within 28 days of curing.

Validation

During treatment of each column, the temperature, pressure, and VOC headspace
was determined within the column. Post-treatment samples were obtained from the
top, middle, and bottom of each column. All samples met the TCLP leachate
(<0.005 mg/L TCE), UCS (>50 psi) and hydraulic conductivity criteria (<1x10-6 cm/s)
criteria.

Equipment

A mechanical system was employed to mix unsaturated or saturated contaminated


soils while simultaneously injecting treatment or stabilization agents. The main
system components include the following:

a crane-mounted soil mixing auger


a treatment agent delivery system
a treatment agent supply
an off-gas collection and treatment system

The mixing system used in the demonstration was manufactured and operated by
Millgard Environmental Corporation, Livonia, MI. It comprised a track-mounted crane
with a hollow kelly bar attached to a drilling tool, known as the MecTool TM, consisting
of one 5-ft long horizontal blade attached to a hollow vertical shaft, yielding an

591
effective mixing diameter of 10 ft. Depths of 40 ft can be achieved with this
equipment.

A grout of the treatment reagents were injected through a vertical, hollow shaft and
out into the soil through 0.50-in diameter orifices in the rear of the soil mixing blades.
Treatment is achieved in butted or overlapped soil columns.

The ground surface above the mixed region was covered by a 14-ft diameter shroud
under a low vacuum to contain any air emissions and direct them to an off-gas
treatment process. The off-gas treatment system consisted of activated carbon filters
followed by a HEPA filter.

592
Case Study 47: Zwevegem

Location: Zwevegem, Belgium

History: Various industrial activities

Contaminants: Metals

20,000 tons
Scale:
Alkaline fly ash
Reagents:
Ex-situ
Method:
N/A
End-Use:

Site Description
Contamination occurred on a site in the South-West of Belgium as a result of
various industrial activities such as electrochemical plating. Two separated
areas could be distinguished, each underneath former production halls. In the
so-called area B the main problem was Zn, Pb, Cu and Ni found in the
shallow groundwater (up to 4 m BGL), together with some cyanide. Area C

593
on the other hand, showed mainly Zn, Pb and Ni. Traces of Cd, Cr and Ni
were also present. At both areas the groundwater was moderately acidic.

The spreading of the contaminated groundwater plume was defined as the


driver for remediation.

Characterisation

Table 1 below shows the composition of one of the groundwater hotspots at


area C.

Table 1: Groundwater concentrations at area C.


Pollutant pH Cd Cr Cu Pb Ni Zn
Concentration
3.73 0.011 0.173 0.952 87.7 1.14 282
(mg/l)

Groundwater contamination resulted from leaching of contaminants from soils


as illustrated in Table 2.

Table 2: Leaching of soils from areas B and C (expressed in mg/l


leachate, EN 12457-4).

AREA B AREA C
pH 4.04 4.95
As < 0.01 < 0.01
Cd < 0.001 < 0.001
Cr < 0.01 < 0.01
Cu 0.11 0.021
Hg < 0.0005 < 0.0005
Pb 0.57 8.14
594
Ni 0.097 0.06
Zn 1.49 14.3

The texture of the soil is loamy sand, which makes a classic pump-and-treat
remedial approach difficult, and ex-situ treatment of the soil by washing
economically not feasible (due to the high fines content).

Objectives

The remediation target for the site was derived via a risk assessment, with
groundwater pollution in the vicinity as the risk driver. This resulted in a
groundwater quality target in and around the remediated site x10 the
Bodemsaneringsnorm (Intervention Value) for each of the heavy metals.

These targets are listed in Table 3 below. The same target values were
applied to evaluate the leachability of the stabilized soils.

Table 3: Target groundwater concentrations for heavy metals.


Pollutant As Cd Cr Cu Hg Pb Ni Zn
Target
concentration 0.2 0.05 0.5 1 0.01 0.2 0.4 5
(mg/l)

Method

Initially, the environmental consultant for the project proposed stabilization of


the heavy metals by means of adding lime milk via in-situ soil mixing, to
precipitate heavy metals by increasing pH. However, this in-situ technique has
disadvantages, such as uncertainty over product durability over time resulting
in re-acidification), sensitivity to obstructions (stones, foundations, debris),
negative impact on geotechnical soil stability, and the difficulty in accurate
dosing of the lime.

595
For all these reasons an alternative method based on ex-situ mixing was
selected and approved in joint agreement with OVAM (Environment Agency of
Flanders), the consultant and the problem owner. The additive used was an
alkaline fly ash applied at a dosage of 2.5 %.

The soil was treated by means of a continuous mixing plant, with a throughput
of 200 tonnes an hour.

Validation

Every daily batch of treated soil (representing about 1000 tons) was sampled
and tested according to EN 12457-4. The results were compared to the target
levels in Table 3 above.

As can be seen, the stabilized soils complied with the site-specific reuse
criteria. In addition the VLAREA-criteria were met, which are the criteria for
free reuse of the soil as secondary building material in Flanders.

596
Table 4: Overview of the full-scale stabilization results (20 validation
samples for area B, 21 validation samples for area C). Values are
expressed in mg/l (leachate or groundwater)

AREA B AREA C
Reuse criteria

Average Max Average Max


Initial Initial 10XBSN Vlarea
stabilized stabilized stabilized stabilized

pH 4.0 11.9 12.4 5.0 11.9 12.4


As < 0.01 < 0.01 0.012 < 0.01 < 0.01 0.012 0.2 0.08
<
Cd < 0.001 < 0.001 0.001 < 0.001 0.001 0.05 0.003
0.001
Cr < 0.01 < 0.03 0.047 < 0.01 < 0.03 0.04 0.5 0.05
Cu 0.11 < 0.023 0.046 0.021 < 0.024 0.116 1 0.05
< <
Hg < 0.0005 0.0005 < 0.0006 0.0028 0.01 0.002
0.0005 0.0005
Pb 0.57 < 0.005 0.008 8.14 < 0.005 0.005 0.2 0.13
Ni 0.097 < 0.005 0.033 0.06 < 0.002 0.002 0.4 0.075
Zn 1.49 < 0.02 0.031 14.3 < 0.021 0.031 5 0.28

The treated soil was compacted during backfilling, and achieved a


compressibility modulus of 11 MPa, as tested by the field plate test.

597
Index

A cement, 34, 39, 45, 57, 76, 81, 211, 259, 260,
266, 267, 270
Acid Sludge Tar. See AST Cement kiln dust. See CKD
activated carbon, 34, 69, 79, 81, 286 chemical fixation, 72, 253, 254, 255
additives, 42, 216, 227, 228 chromium, 72, 74, 80
Allu Mixer, 164, 166 CKD (Cement kiln dust), 244, 24748, 258
aluminium, 79, 245, 248, 252, 253 clay, 42, 79, 80, 162, 174, 240, 244, 303, 305,
antimony, 61, 72 307, 313, 319, 321, 323, 325
arsenic, 72, 73, 75 compacted, 306, 310, 313,-315, 321, 323, 325
asphalt, 303, 316, 318 coal tar, 174, 195, 213, 219
asphalt millers, 121, 12223, 126 cobbles, 83, 156, 226, 227, 277, 304
AST (Acid Sludge Tar), 77, 78 COC, (Contaminant of Concern) 52, 207, 218,
auger, 93, 9597, 99, 105, 14546, 149, 151, 221, 238, 239, 287, 288, 330, 332
154, 156, 161, 175, 179, 184, 278, 279 columns, 154, 157, 160, 179, 185, 193, 194,
197, 213, 278, 288, 289, 298
B compliance, 52, 138, 178, 181, 190, 191, 219,
220, 221, 234
Backhoe, 163, 164, 165 Composite Caps, 31415, 321, 323, 325
barium, 72, 74 compressive strength, 39, 43, 93, 213
barrier wall, 220, 221 Cone Penetration Testing. See CPT
batch plant, 93, 96, 99, 100, 145, 149, 163, 166, construction quality assurance. See CQA
173, 183, 192, 278 construction quality control. See CQC
bench-scale testing, 179, 180, 235, 236, 260, contaminants, 34, 36, 37, 3839, 51, 55, 6568,
261, 262, 26364, 273, 275, 277, 278, 279, 89, 220, 22728, 231, 234, 263, 328, 329
291, 293 30
bentonite, 97, 193, 259, 270, 310 secondary, 105, 128, 130
benzene, 34, 79, 80 volatile, 67, 68, 272
binder system, 34, 36, 37, 38, 184 Contaminants of Concern (see COCs)
boreholes, 82, 8788, 90 copper, 61, 72, 74, 75, 79
BOSS systems, 163, 164, 166, 167, 168, 169, CPT (Cone Penetration Testing), 88, 153
172, 174 CQA (construction quality assurance), 178-183,
bottom ash, 244, 25152, 259 186, 187
boulders, 83, 97, 227 CQC (construction quality control), 178, 182,
bricks, 83, 97, 107, 109, 121, 124 211
BTEX, 39 CQC documentation, 184, 185
bucket mixing, 130, 145, 163, 172, 17375, 179, creosote, 34, 209, 238
279, 281 CSMs (conceptual site model), 64, 66, 69, 219,
223, 225, 226
C curing, 170, 176, 178, 188, 193, 195, 196, 200,
201, 203, 207, 209, 269, 271, 274
cadmium, 61, 72, 73, 253, 254 cyanides, 74, 76
calcium aluminosilicates, 247, 249, 250, 251
calcium hydroxide, 24951, 255 D
calcium oxide, 244, 247, 248, 249, 251
cap construction, 319 desiccation, 304, 313, 319, 321, 326
capping materials, 31819 dioxins, 34, 35, 76, 209, 238
capping strategies, 303, 316, 319, 324 DNAPL, 222
capping technologies, 303, 304, 322, 323 DQOs (data quality objectives), 222
carbon, 249, 251, 255, 25657, 260 drainage, 294, 304, 318, 319, 321, 323
CASSST (Codes and Standards for Stabilization drainage layer, 295, 304, 305, 306
and Solidification Technologies), 59 dust, 67, 130, 239, 245, 249, 250, 252, 257, 278

598
E H

earthen pits, 128, 12930 hazardous wastes, 46, 50, 57, 59, 295
Electric Power Research Institute. See EPRI hexavalent chromium, 245, 248
enclosures, 128, 13233 hydrated lime, 129, 130, 244, 245, 24647, 249,
engineering controls, 319, 329, 330, 332 258
Environment Agency, 45, 47, 57, 59, 60, 69, hydraulic conductivity, 218, 223, 228, 231, 232,
226, 260, 274, 329 233, 242
Environmental Protection Agency. See EPA hydrogeology, 64, 223, 225, 228, 234
EPA (Environmental Protection Agency), 32, 45,
46, 50, 52, 53, 59, 70, 238, 248, 263 I
EPRI (Electric Power Research Institute), 46,
219, 226, 231, 233, 260 infiltration, 68, 3045, 307, 315, 321
erosion, 304, 307, 322, 324 information sources, 44, 45
evaluation, 63, 180, 184, 185, 186, 187, 188, infrastructure, 63, 229, 293, 294
193, 221, 260, 282, 292, 293 inhalation, 245, 246, 248, 252, 253, 254, 255,
evapotranspiration, 303, 304, 305, 315, 321, 256, 257
323, 325, 326 injection drum mixers, 171
excavator mixing, 114, 118, 131, 133, 172, 177, in-situ S/S, 3435, 53, 92, 9395, 98, 99, 100,
198, 241 101, 120, 121, 122, 123, 124, 219, 296
explosives, 76 IPPC (Integrated Pollution, Prevention and
ex-situ mixing, 98, 99, 118, 125, 133, 134, 138, Control), 58
162, 183, 184, 197, 273, 278, 279, 297 iron, 76, 79, 245, 248, 249, 251, 252, 253, 254
ex-situ S/S, 52, 53, 91, 92, 97, 98, 99, 101, 119, ITRC, 32, 33, 216, 217, 218, 219, 222, 231, 232,
120, 122, 123, 13842, 216, 219 233, 234, 260, 261, 263, 329

F L

Feasibility Study, 22930 lagoons, 129, 238


flexible membrane liners. See FMLs landfill, 58, 59, 60, 138, 233, 247, 270, 295
fly ash, 34, 39, 81, 238, 244, 24950, 257, 259, erosion control/topsoil, 295
260, 263, 266, 267 leaching/leachability, 32, 63, 81, 231, 233, 236,
FMLs (flexible membrane liners), 237, 238, 239, 244, 247, 250, 252, 293
24243, 308 leaching tests, 219, 221, 234, 236, 270
formulations, 34, 59, 88, 220, 261, 264, 266, lime, 41, 42, 45, 57, 81, 249, 251, 254, 259
268, 269, 270, 287 lime kiln dust, 248, 258, 260
FY (fiscal years), 50, 5253 limestone, 244, 245, 251
loaders, 99, 119, 120, 123, 129, 130, 131, 132,
G 150, 163, 166, 173
Long-Term Response Action (LTRA), 327
GCL (Geosynthetic clay liner), 212, 237, 238,
239, 303, 310, 311, 31215, 319, 321, 322, M
323, 325
geo-membranes, 295, 303, 308, 310, 31115, manufactured gas plant. See MGP
326 material performance goals, 33, 216, 218, 220,
Geo-Solutions (GSI), 156, 163 221, 226, 235
Geosynthetic clay liner. See GCL materials
groundwater, 64, 65, 66, 87, 88, 89, 219, 220, abrasive, 136
221, 223, 229, 230, 231, 237, 32930 available, 305, 308
grout, 33, 149, 159, 160, 166, 169, 170, 171, cement-bound/cemented, 37, 38, 215, 235
172, 176, 228, 282 crushed, 13536, 234
GSI. See Geo-Solutions cured, 237, 271, 329
GWQS. See Groundwater Quality Standards difficult, 221, 287
excavated, 135
excess, 298
failed, 212

599
granular, 144, 234 organic contaminants, 69, 71, 72, 80, 256
hard, 136 organics, 53, 55, 76, 79, 80, 81, 211, 244, 247,
local, 308, 319 250, 252, 256, 304
mix/mixed, 108, 111, 114, 123, 174, 236, 272, free-phase, 89, 244, 247, 250, 252
279 organic Stabilization, 77, 81, 250, 252
multiple, 290 organoclay, 256
natural, 256 overlapping columns, 155
off-spec, 209, 212 oversize material, 134, 83, 88, 200, 272
placed, 139, 141, 213 oxidation, 77, 79, 80
plastic, 124, 136
preparing S/S, 297 P
soil-type, 304
solidified, 283 paddle aerators, 102, 12324, 126
stabilised, 68, 233, 235, 244, 252 PAHs (polycyclic aromatic hydrocarbons), 39,
sticky, 136 61, 76, 219
PCB, 39, 76, 219
mercury, 72, 75, 254 PCP, 34, 35, 238
metal debris, 107, 121, 124 pentachlorophenol, 34, 209, 238
metals, 55, 72, 73, 74, 7576, 128, 136, 137, performance criteria, 92, 138, 141, 191, 211,
244, 249, 251, 252, 253, 254, 255 216, 218, 220, 236, 269, 270, 275, 279, 287
immobilise, 244, 245, 247, 248, 249, 251, 252 permeability, 39, 195, 196, 200, 208, 210, 211,
trace, 79 231, 233, 266, 267, 268, 269, 270, 271
MGP (manufactured gas plant), 71, 213, 219 pesticides, 76, 219, 241
MGP Site, 219 phosphates, 7476, 244, 253, 254, 258
mix designs, 33, 185, 186, 188, 218, 220, 227, pilot-scale, 179, 181, 260, 275, 276, 279, 286,
229, 230, 235, 267, 275, 276, 28790, 293 292
mixers, 97, 100101, 104, 105, 108, 109, 111, pits, 88, 128, 129, 130, 301
112, 113, 128, 168, 200
mixing plasticity, 100101, 107, 109, 112, 232
batch, 104 plastic limit, 10710
complete, 170, 172 pneumatic trailer, 245, 246, 248, 249, 250, 252,
continuous, 104, 105, 108, 110 256
mixing equipment, 33, 98, 102, 163, 271, 278 polycyclic aromatic hydrocarbons. See PAHs
mixing head, 99, 114, 146, 149, 163, 164, 165 Portland Cement (PC), 41, 46, 48, 81, 12829,
66, 174 193, 231, 238, 24448, 257, 258, 259, 260,
mixing pits, 102, 128, 130 293
mixing tanks, 128, 131, 149, 166 post-construction monitoring, 327, 329
monitoring, 178, 179, 222, 291, 327, 328, 329, pug-mill, 34, 35, 94, 102, 105, 1067, 110, 112,
330, 332 113, 183, 184, 195, 197, 200, 262
post construction, 291
monolith, 67, 212, 293, 294, 298, 300, 303, 306, Q
308, 310, 322, 330
QA. See Quality Assurance
N QC. See Quality Control
Quality Assurance (QA), 31, 159, 170, 176, 178,
NAPL (non-aqueous phase liquids), 219, 220, 193, 194, 274, 278
223, 228 Quality Control (QC), 112, 113, 118, 125, 157
nickel, 72, 75, 76 59, 167, 168, 169, 170, 172, 173, 174, 175
non-aqueous phase liquids. See NAPL 76, 178, 282
quicklime, 41, 24546, 249, 258
O
R
off-site disposal, 116, 120, 121, 122, 124, 129,
131, 132, 228 RCRA (Resource Conservation and Recovery
oily sludge, 95 Act), 50, 5152, 240, 248
organic compounds, 72, 77, 79, 80, 81, 256, 286

600
reagents, 100, 11214, 116, 11825, 128, 147, sampling, 88, 94, 178, 183, 187, 189, 19092,
16364, 166, 174, 180, 183, 26163, 266, 194, 198, 200, 222, 269, 287, 288, 289
273, 29697 sand, 41, 42, 43, 239, 244, 295, 304, 308
addition of, 99101, 106, 152, 154, 159, 168, screw auger, 108, 112, 114, 118, 125
170, 174, 176, 179, 184, 194, 272, 302 screw mixers, 102, 104, 10810, 112, 113, 126
calibrating, 114 sediments, 53, 61, 62, 90, 178, 223
cementitious, 235 Shredding equipment, 98, 136
delivery, 100, 125, 149, 166, 169, 183 silica, 41, 43, 249, 251
dosage, 154, 169 silos, 94, 100, 149, 166, 183, 245, 246, 248,
grout, 98, 100, 101, 301 249, 250, 252, 253, 254, 255, 257
injection, 172, 297 site investigation, 82, 88, 90, 226, 228, 230, 287
inorganic, 80, 81 slag, 193, 244, 266, 267, 270
organic, 81 sludge, 34, 45, 77, 90, 112, 154, 164, 168, 172,
ratios, 179, 191, 192, 193, 297 238, 290
selection, 263, 277 slurry, 93, 172, 173, 174, 262, 273
receptors, 52, 63, 64, 65, 66, 68 soil, 34, 36, 61, 67, 79, 82, 8490, 123, 14144,
regulatory agencies, 32, 291, 294, 296, 327, 332 155, 161, 171, 174, 180, 31214
remedial action, 61, 63, 67, 88, 178, 187, 193, clayey, 321, 323, 325
208, 211, 226, 296, 327 cohesive, 105, 144, 232
remedial design, 89, 182, 22930 compacted, 139, 307
remedial goals, 33, 21821, 225, 230 dense, 162, 227
remedial strategies, 220, 222, 225 high moisture content, 154, 168
retreatment, 187, 211, 213, 214, 237, 288, 289 plastic, 116, 121, 123, 308
ribbon blenders, 102, 104, 110, 11113, 126 soil classification, 82, 227
rig soil types, 55, 64, 82, 145, 163, 174, 180, 186,
crane-mounted soil-mixing, 147, 148 226, 293, 297, 321
excavator-mounted soil-mixing, 150 soil vapour extraction, 53, 54, 67
risk assessment, 63, 64, 66, 67, 225, 230 solidification, 32, 46, 52, 220, 227, 231, 232, 241
Risk Framework, 65 Solidification and Stabilization, See S/S, 32, 46,
risk management, 59, 63 50, 52, 54
risk reduction, 52, 58 Solid Waste Burn Pit. See SWBP
risks, 30, 31, 34, 36, 58, 61, 62, 63, 64, 66, 68, SPLP (Synthetic Precipitation Leaching
77, 90, 222, 230 Procedure), 63, 208, 209, 233, 236
rotary drum mixers, 163, 167, 171 SPT (Standard Penetration Testing), 88, 97, 153
rotary tillers, 17172 S/S (Stabilization and Solidification), 30, 34, 36
rototillers, 12021, 126 37, 4448, 50, 5155, 57, 5863, 65, 6872,
77, 21516, 21821, 226, 32830
S employing, 55, 57, 59
full-scale, 236, 275, 288, 289, 290
samples, 169, 177, 183, 185, 186, 190, 19399, in-situ soil mixing, 160
204, 207, 212, 260, 272, 273, 287, 28890 columns, 154, 157, 160, 184
additional, 194, 195 contractor, 275, 276
average contamination, 289 equipment, 89, 98, 275, 276, 277
collection, 159, 176, 178, 188, 190, 193, 199, Evaluation, 227, 228, 229, 230
203, 288 S/S materials, 37, 40, 197, 198, 212, 215, 216,
cured, 195, 2045 23134, 237, 291, 293, 296, 299, 330, 332
fresh, 197 S/S mixing equipment, 98, 183
high contamination, 288, 290 S/S monolith, 291, 292, 293, 294, 296, 297, 299,
maximum contamination, 287 303, 310, 316, 317, 318, 319, 330, 331
minimum number of, 194, 195 S/S operations, 178, 181, 185, 186, 193, 299
replicate, 186, 206, 209, 210, 235 S/S projects, 82, 90, 178, 179, 194, 220, 259,
representative, 261, 287, 329 261, 292, 316
re-test, 188 S/S reagents, 93, 133
stored, 203 S/S selection, 50, 53, 54, 56, 69, 226
treated, 260 S/S soil monolith, 285
S/S soils, 211, 215

601
cement-based, 206 W
S/S systems, 37, 38, 44, 72
S/S-treated material, 140, 213, 218, 221, 305 waste, 30, 57, 59, 76, 79, 80, 11214, 116, 118,
treatment, 67, 72, 76, 83, 89, 92, 98, 99, 134, 119, 12025, 133, 275, 276, 27779
237, 239, 257, 259, 330, 332 oily, 129, 130
treatment of nitrate in waste materials, 76 organic-contaminated, 77, 256
Stabilization, 30, 32, 47, 52, 77, 80, 162, 220, remediation of contaminated land and
241, 255 treatment of, 30
chemical, 231, 232, 236 treated, 114, 116, 120, 121, 122, 123, 129,
Standard Penetration Testing. See SPT 131, 132
STSO. See Stabilization and Solidification
sulfides, 7276, 244, 254, 255, 258 water, 64, 67, 98, 18384, 191, 192, 193, 203,
Superfund, 50, 51, 53, 71 231, 24457, 27273, 296, 297, 315, 316
Superfund Sites, 51, 52, 59, 327, 328 water infiltration, 310, 315
swell, 99, 100, 160, 183, 275, 292, 296, 297 water table, 53, 71, 88, 92, 98, 101, 160, 162,
Synthetic Precipitation Leaching Procedure. See 212, 228, 235, 329
SPLP perched, 228
water vapour, 245, 246
T waterway, 299, 308

TCLP (Toxicity Characteristic Leaching


Procedure), 216, 233 Z
test pits, 8788, 90, 283
tillers, 102, 119, 125, 128 zero-valent iron. See ZVI
TOC (total organic carbon), 81 zinc, 61, 72, 75, 80
topsoil, 304 ZVI (zero-valent iron), 93
total organic carbon (TOC), 81
total petroleum hydrocarbons. See TPH
Toxicity Characteristic Leaching Procedure. See
TCLP
TPH (total petroleum hydrocarbons), 39
tractors, 119, 120, 123
trommels, 13435

UCS (Unconfined Compressive Strength), 63,


210, 213, 231, 232, 266, 267
Unconfined Compressive Strength. See UCS
Unified Soil Classification System (USCS), 82,
305
USACE (U.S. Army Corps of Engineers), 260
U.S. Army Corps of Engineers. See USACE
USCS (Unified Soil Classification System), 82,
305
USEPA, 5054, 56, 219, 222, 225, 231, 233,
235, 260, 263, 274, 315, 316, 327, 328
USEPA Superfund program, 50, 52, 55, 71, 328,
332

Validation, 63, 68
VOCs (volatile organic compounds), 39, 219,
286
volatile organic compounds. See VOCs

602

You might also like