WASTE MATERIALS IN CONSTRUCTION Putting Theory into Practice
WASTE MATERIALS IN CONSTRUCTION Putting Theory into Practice
This Page Intentionally Left Blank
Studies in Environmental Science 71
WASTE MATERIALS IN CONSTRUCTION Putting Theory into Practice Proceedings of the International Conference on the Environmental and Technical Implications of Construction with Alternative Materials, WASCON '97, Houthem St. Gerlach, The Netherlands, 4-6 June 1997
Edited by
J.J.J.M. Goumans ISCOWA The Netherlands
G.J. Senden ISCOWA The Netherlands
H.A. van der Sloot
Netherlands Energy Research Foundation (ECN) Petten, The Netherlands
1997 ELSEVIER Amsterdam
-
Lausanne
-
NewYork-
Oxford
-
Shannon
-
Singapore
-
Tokyo
ELSEVIER SCIENCE B.V. Sara Burgerhartstraat 25 P.O. Box 211, 1000 AE Amsterdam, The Netherlands
ISBN 0-444-82771-4 © 1997 ELSEVIER SCIENCE B.V. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Science B.V., Copyright & Permissions Department, P.O; Box 521, 1000 AM Amsterdam, The Netherlands. Special regulations for readers in the U.S.A. - This publication has been registered with the Copyright Clearance Center Inc. (CCC), 222 Rosewood Drive Danvers, Ma 01923. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside of the U.S.A., should be referred to the owner, Elsevier Science B.V., unless otherwise specified. No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. This book is printed on acid-free paper. Printed in The Netherlands
Studies in Environmental Science Other volumes in this series 1 2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Atmospheric Pollution 1978 edited by M.M. Benarie Air Pollution Reference Measurement Methods and Systems edited by T. Schneider, H.W. de Koning and L.J. Brasser Biogeochemical Cycling of Mineral-Forming Elements edited by P.A. Trudinger and D.J. Swaine Potential Industrial Carcinogens and Mutagens by L. Fishbein Industrial Waste Management by S.E. Jorgensen Trade and Environment: A Theoretical Enquiry by H. Siebert, J. Eichberger, R. Gronych and R. Pethig Field Worker Exposure during Pesticide Application edited by W.F. Tordoir and E.A.H. van Heemstra-Lequin Atmospheric Pollution1980 edited by M.M. Benarie Energetics and Technology of Biological Elimination of Wastes edited by G. Milazzo Bioengineering,Thermal Physiology and Comfort edited by K. Cena and J.A. Clark Atmospheric Chemistry. Fundamental Aspects by E. Meszaros Water Supply and Health edited by H. van Lelyveld and B.CoJ. Zoeteman Man under Vibration. Suffering and Protection edited by G. Bianchi, K.Vo Frolov and A. Oledzki Principles of Environmental Science and Technology by S.E. Jorgensen and I. Johnsen Disposal of Radioactive Wastes by Z. Dlouh~/ Mankind and Energy edited by A. Blanc-Lapierre Quality of Groundwater edited by W. van Duijvenbooden, P. Glasbergen and H. van Lelyveld Education and Safe Handling in Pesticide Application edited by E.A.H. van Heemstra-Lequin and W.F. Tordoir Physicochemical Methods for Water and Wastewater Treatment edited by L. Pawlowski Atmospheric Pollution 1982 edited by M.M. Benarie Air Pollution by Nitrogen Oxides edited by T. Schneider and L. Grant Environmental Radioanalysisby H.A. Das, A. Faanhof and H.A. van der Sloot Chemistry for Protection of the Environment edited by L. Pawlowski, A.J. Verdier and W.J. Lacy Determination and Assessment of Pesticide Exposure edited by M. Siewierski The Biosphere: Problems and Solutions edited by T.N. Veziro~lu Chemical Events in the Atmosphere and their Impact on the Environment edited by G.B. Marini-Bettolo Fluoride Research 1985 edited by H. Tsunoda and Ming-Ho Yu Algal Biofouling edited by L.V. Evans and K.D. Hoagland Chemistry for Protection of the Environment 1985 edited by L. Pawlowski, G. Alaerts and W.J. Lacy Acidification and its Policy Implications edited by T. Schneider Teratogens: Chemicals which Cause Birth Defects edited by V. Kolb Meyers Pesticide Chemistry by G. Matolcsy, M. Nadasy and Y. Andriska Principles of Environmental Science and Technology (second revised edition) by S.E. JQrgensen and I. Johnsen Chemistry for Protection of the Environment 1987 edited by L. Pawlowski, E. Mentasti, W.J. Lacy and C. Sarzanini Atmospheric Ozone Research and its Policy Implications edited by T. Schneider, S.D. Lee, G.J.R. Wolters and L.D. Grant
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70
Valuation Methods and Policy Making in Environmental Economics edited by H. Folmer and E. van lerland Asbestos in Natural Environment by H. Schreier How to Conquer Air Pollution. A Japanese Experience edited by H. Nishimura Aquatic Bioenvironmental Studies: The Hanford Experience, 1944-1984 by C.D. Becker Radon in the Environment by M. Wilkening Evaluation of Environmental Data for Regulatory and Impact Assessment by S. Ramamoorthy and E. Baddaloo Environmental Biotechnology edited by A. Blazej and V. Privarova Applied Isotope Hydrogeology by F.J. Pearson Jr., W. Balderer, H.H. Loosli, B.E. Lehmann, A. Matter, Tj. Peters, H. Schmassmann and A. Gautschi Highway Pollution edited by R.S. Hamilton and R.M. Harrison Freight Transport and the Environment edited by M. Kroon, R. Smit and J.van Ham Acidification Research in The Netherlands edited by G.J. Heij and T. Schneider Handbook of Radioactive Contamination and Decontamination by J. Severa and J. BAr Waste Materials in Construction edited by J.J.J.M. Goumans, H.A. van der Sloot and Th.G. Aalbers Statistical Methods in Water Resources by D.R. Helsel and R.M. Hirsch Acidification Research: Evaluation and Policy Applications edited by T.Schneider Biotechniques for Air Pollution Abatement and Odour Control Policies edited by A.J. Dragt and J. van Ham Environmental Science Theory. Concepts and Methods in a One-World, Problem-Oriented Paradigm by W.T. de Groot Chemistry and Biology of Water, Air and Soil. Environmental Aspects edited by J. T61gyessy The Removal of Nitrogen Compounds from Wastewater by B. Halling-Sorensen and S.E. JQrgensen Environmental Contamination edited by J.-P. Vernet The Reclamation of Former Coal Mines and Steelworks by I.G. Richards, J.P. Palmer and P.A. Barratt Natural Analogue Studies in the Geological Disposal of Radioactive Wastes by W. Miller, R. Alexander, N. Chapman, I. McKinley and J. Smellie Water and Peace in the Middle East edited by J. Isaac and H. Shuval Environmental Oriented Electrochemistry edited by C.A.C. Sequeira Environmental Aspects of Construction with Waste Materials edited by J.J.J.M. Goumans, H.A. van der Sloot and Th. G. Aalbers. Caracterization and Control of Odours and VOC in the Process Industries edited by S. Vigneron, J. Hermia, J. Chaouki Nordic Radioecology. The Transfer of Radionuclides through Nordic Ecosystems to Man edited by H. Dahlgaard Atmospheric Deposition in Relation to Acidification and Eutrophication by J.W. Erisman and G.P.J. Draaijers Acid Rain Research: do we have enough answers? edited by G.J. Heij and J.W. Erisman Climate Change Research. Evaluation and Policy Implications edited by S. Zwerver, R.S.A.R. Rompaey, M.T.J. Kok and M.M. Berk Global Environmental Biotechnology edited by D.L. Wise Municipal Solid Waste Incinerator Residues by A.J. Chandler, T.T. Eighmy, J. Hartlen, O. Hjelmar, D.S. Kosson, S.E. Sawell, H.A. van der Sloot and J. Vehlow Freshwater and Estaurine Radioecology edited by G. Desmet, R.J. Blust, R.N.J. Comans, J.A. Fernandez, J.Hilton, and A. de Bettencourt Acid Atmospheric Deposition and its Effects on Terrestrial Ecosystems in The Netherlands edited by G.J. Heij and J.W. Erisman Harmonization of Leaching/Extraction Tests edited by H.A. van der Sloot, L. Heasman and Ph. Quevauviller
vii
Dear colleague, The international society ISCOWA herewith presents the pro ceedings of the international conference WASCON'97, which has been held from June 4 till June 6, 1997 in Valkenburg the Netherlands. SCOPE OF THE C O N F E R E N C E Many western countries are still facing the problem of a growing burden of waste materials, accompanied by a shortage of primary materials. Serious problems with cleaning-up old landfills and pollution of the groundwater are currently making disposal of waste very difficult in many countries. The protection of soil and water, the limitation of waste production and the re-use of waste materials are key items in policy concepts, generally stated "Sustainable Development". With respect to waste materials, extensive research has been carried out to find a market for these materials, e.g. powder coals fly ash in concrete and incinerator slag in road cons truction. Beneficial use of products derived from waste materials can in fact contribute to sustainable development. However, the market for such waste-derived products mostly involves their re-use as construction materials, implying close contact with the soil. If not properly managed, this may result in pollution of the soil, or even of the groundwater, due to the uncontrolled release of contaminants. In order to predict and control potential contamination, laboratory leaching tests have been developed in several countries, e.g. the USA, Canada, Germany and the Netherlands. The knowledge gained from this research can be used to contol or eliminate possible contamination. A problem is the fact that the various tests being used are not comparable, but harmonization is on its way. The second theme of the conference addresses the state of the art in technical solutions and procedures to use waste materials for the production of construction materials such as concrete, calcium silicate bricks, artificial gravel and other products. Solidification is discussed broadly, as is the treatment and application of MSWI by-products. Various contributions regarding environmental policy and legislation complete the conference. The organizing committee hopes that the conference indeed contributed to the solution of the environmental problems concerning the re-use of waste materials and thus to sustainable development in building practice.
On behalf of ISCOWA Dr. J.M. Goumans
This Page Intentionally Left Blank
ix Committees
ISCOWA wishes to thank the members of the committees for their contribution to WASCON '97. Organizing Committee
Scientific Committee
G.J. Senden, ISCOWA, Chairman ir. J. de Castro, ISCOWA R. Fetlaar, Conference Manager L. Haverkort, ISCOWA ir. R.T. Eikelboom, Ministry of Housing, Spatial Planning and the Environment, The Netherlands dr.ir. N. Raemakers, University of Maastricht, The Netherlands
dr.J.J.J.M.Goumans,ISCOWA, Chairman prof.dr. J.Cabrera, University of Leeds, United Kingdom dr. H.A. van der Sloot, ECN, The Netherlands dr. J. Hartl6n, Sweden prof.dr.ing. P.Schieszl, IBAC, Germany Dr. D. Kosson, Rutgers University, USA prof.dr. Shin-ichi Sakai, Kyoto University, Japan
Finally ISCOWA wishes to thank the following organizations who gave financial support to WASCON '97: EC, DGXII, Brussels, Belgium Commission of the European Communities, Directorate General XII, Science, Research and Development, Directorate C : Industrial and Material Technologies, Measurements and Testing Ministry of Housing, Spatial Planning and the Environment.Director ate General for the Environment, The Netherlands GKE/Vliegasunie, De B ilt, The Netherlands Dutch Fly Ash Corporation CUR, Gouda, The Netherlands Center fir Civil Engineering, Research and Codes CROW, Ede, The Netherlands Center for Codes and Research in Civil Engineering Ministry of Transport and Watermanagement, Directorate General for Watermanagement, Delft, The Netherlands JWRF, Japan Waste Research Foundation Kyot 0 University, Kyoto, Japan Novem, Sittard, The Netherlands Netherlands Agency for Energy and the Environment
This Page Intentionally Left Blank
Preface .......................................................................................................................................
VII
Overview of MSWI residue recycling by thermal processes ...................................................... 1 Kyoto University Tokio, Kyoto, Japan S. Sakai and M. Hiraoka, Quality improvement of MSW bottom ash by enhanced ageing, washing and combination processes ................................................................................................................. Tauw Milieu bv and Waste Processing Association, The Netherlands J.J. Steketee, R.F. Duzijn and J.G.P. Born
13
Construction materials manufacturing by the technology of melting .................................... 25 Kubuta Corporation, Osaka, Japan S. Abe Producing permeable blocks and pavement bricks from molten slag ................................... 31 Takuma Corporation Ltd., Hyogo, Japan M. Nishigaki Investigation of sintering processes in bottom ash to promote their reuse in civil construction(Part 1) Element balance and leaching ...................................................................................... 41 (Part 2) Long term behaviour .................................................................................................... 51 ABB Corporate Research Ltd, Switzerland and Forschungszentrum Karlsruhe, Germany A. Selinger, V. Schmidt, B. Bergfeldt, H. Seifert, J. Vehlov and F.G. Simon
The acid extraction process ......................................................................................................... 59 H. Kawabata, Kobe Steel Ltd., Hyogo, Japan T. Inoue, Unitika Ltd, Osaka, Japan Pre-treatment of MSWI Fly ash for useful application ............................................................ 67 TNO Waste Technology Division, Apeldoorn, The Netherlands E. Mulder and R.K. Zijlstra Direct melting process for MSW recycling ................................................................................ 73 Nippon Steel Corporation, Tokyo, Japan M. Osada
The ABB dry ash concept: INRECTM ...................................................................................... 79 ABB Corporate Research Ltd, Switzerland A. Selinger and V. Schmidt Municipal Solid Waste Incinerator Bottom Ashes as Granular Base Material in Road Construction ....................................................................................................................... 85 Institut fur Technische Chemie, Institut fur Strassen - und Eisenbahnwesen, Germany G. Pfrang-Stotz, J. Reichelt Test Project Crushed Masonry 50/150 mm in the Ventjagersplaat River Dam .................. 91 Ministry of Transport, Delft, The Netherlands ' H. A. Rijnsburger
xii
Evaluation of treatment of gas cleaning residues from M S W I with chemical agents ........... 95
Environment Preservation Center, Kyoto University, Japan, ECN, The Netherlands S. Mizutani, H.A. van der Sloot, S. Sakai Recycling for road improvement .............................................................................................. 105
OECD, USA Ch.J. Nemmers Quarries reinforcement with stabilised bottom ashes ............................................................ 115
INERTEC and ADEME, France A. Bouchelaghem, M-C. Magnie and V. Mayeux The influence of monolith physical properties on diffusional leaching behaviour of asphaltic pavements constructed with M W S combustion bottom ash .... :............................. 125
University of New Hampshire, Durham, USA T. Taylor Eighmy, D. Crimi, I.E. Whitehead, X. Zhang and D.L. Gress Design and construction of a road pavement using fly ash in hot rolled asphalt ................. 149
University of Leeds, CEMU, Dept. Civil Engineering, England J. Cabrera Engineering properties of the coal ashes stored in the Valdeserrana Lagoon Andorra power plant .................................................................................................................................
167
Polytechnical University of Valencia, Spain P.A. Calderon Garcia, E. Peris Mora and J. Parrila Juste Mine tailings - practical experiences in rifling up harbours ................................................. 175
Public Works, Engineering Division, Rotterdam, The Netherlands J. van Leeuwen and K. Ratsma Upgrading the use of recycled material - UK demonstration project ................................... 185
Building Research Establishment, Watford, England Dr. R.J. Collins Beneficial use of contaminated sediments within the Meuse river system ........................... 193
IWACO and Ministry of Transport, Public Works and Water management, The Netherlands A.L. Hakstege, J.J.M. Heynen and H.P. Versteeg Integration of Testing Protocols for Evaluation of Contaminant Release from Monolithic and Granular Wastes ............................................................................................. 201
gutgers University, Dept. Chem. Biochem. Eng., USA ECN, The Netherlands D.S. Kosson and H.A. van der Sloot Development of a leaching protocol for concrete .................................................................... 217
ECN, IBAC, NMi, Research Institute of the Cement Industry I. Hohberg, G.J. de Groot, A.M.H. van der Veen and W. Wassing Use of a Chelating Agent to Determine the Metal Availability for Leaching From Soils and Wastes .........................................................................................................................
Rutgers University, Dept. Chem. Biochem. Eng., USA A.C. Garrabrants and D.S. Kosson
229
xiii
Leaching Characteristics of Communal and Industrial Sludges ........................................... 247 ECN, The Netherlands P. A. J. P. Cnubben and H. A. van der Sloot Influence of Concrete Technical Parameters on the Leaching behaviour of Mortar and Concrete ...................................................................................................................................... 253 IBAC, Germany I. Hohberg and P. Schiessl By-products management and quality control ........................................................................ 259 Dutch Fly Ash Corporation, The Netherlands J.W. van de Berg and A. Boorsma Maasvlakte Fly-ash processing plant ....................................................................................... 269 Dutch Fly Ash Corporation, The Netherlands J.B.M. Moret and J.W. van den Berg Fly ash as binder in concrete ..................................................................................................... 279 KEMA, The Netherlands L.J.L. Vissers Upgrading and quality improvement of PFA .......................................................................... 289 KEMA, The Netherlands H.A.W. Cornelissen The effect of the Dutch building materials decree on the by-products from coal fired power stations .................................................................................................................... 301 Dutch Electricity Generating Board, The Netherlands M.P. van der Poel Prediction of environmental quality of by-products from coal fired power plants ............. 311 KEMA, The Netherlands R. Meij Short leaching test compared to a column leaching test for internal quality control of coal bottom ash ........................................................................................................................... 327 KEMA, The Netherlands E.E. van der Hoek and F.J.M. Lamers Retention in mortars of trace metals in Portland clinckers ................................................... 339 LAEPSI - INSA Lyon, France I. Serclerat and P. Moszkowicz Study of cement-based mortars containing Spanish ground sewage sludge ash .................. 349 Polytechnical University, Valencia, Spain J. Monzo, J. Pay, M.V. Borrachero, A. Bellver and E. Peris-Mora Fly ash - useful material for preventing concrete corrosion .................................................. 355 IMS, Beograd and Faculty of Technology, Novi Sad, Yugoslavia S. Mileti, M. Ili, J. Ranogajec and M. Djuri
xiv
Fly ash as the basic material for inorganic binders production ............................................ Institute for Materials Testing, Belgrade, Yugoslavia M. Iliac, S. Miletic and R. Djuricic
365
A study of Potential of Utilising Electric Arc Furnace Slag as Filling Material in Concrete ...................................................................................................... 373 Royal Institute of Technology, Sweden C. B~iverman and F. Aran Aran. Properties of portland Cement Mortars Incorporating High Amounts of Oil -Fuel Ashes ............................................................................................................................ Universidad Polytecnica de Valencia, Spain J. Pay, M.V. Borrachero, J. Monz¢, M.J. Blanquer and E. Gonz lez-L¢pez
377
The use of fly ash to improve the chloride resistance of cement mortars ............................. 387 University of Leeds, CEMU, Dept. Civil Engineering, England J. Cabrera and G. Woolley
Low lime binders based on fluidized bed ash .......................................................................... 401 Moravia-Silesian Power Plant and Technical University of BRNO, Czech Republic J. Drottner and J. Havlica Structural performance of reinforced concrete made with sintered ash aggregates ........... 411 University of Leeds and Maunsell & Partners Consulting Engineers, England P.J. Wainwright and P. Robery Investigating waste/binder interactions by neural network analysis ................................... 421 Imperial College of Science, Technology and Medicine, London, England C. D. Hills, J.A. Stegemann and N.R. Buenfeld The use of MSWI bottom ash in hollow construction materials ........................................... 431 Net Brussel, Brussel, Belgium E. Jansegers
Using Chemfronts, a geochemical transport program, to simulate leaching from waste materials ........................................................................................................................... Royal Institute of Technology, Stockholm, Sweden C. B~iverman, L. Moreno and I. Neretnieks
437
Overview of geochemical processes controlling leaching characteristics of MSWI bottom ash ...................................................................................................................... 447 ECN, The Netherlands J. Meima and R.N.J. Comans.
Heavy metal binding mechanisms in cement based waste materials .................................... 459 Swiss Federal Institute of Environmental Science and Technology, Switzerland C. Ludwig, F. Ziegler and C. A. Johnson
ICPMS, Hydride-generation ICP-MS and CZE for the study of solidification/stabilisation of industrial waste containing Arsenic ........................................ 469 University of Leuven, Dept. Chem. Engineering, Belgium C. Vandecasteele, K. van den Broeck and V. Dutr,
xv
Application of computer modelling to predict the leaching behaviour of heavy metals from M S W I fly ash and comparison with a sequential extraction method .............. 481 Katholieke Universiteit Leuven, Belgium P. Van Herck, B. van der Bruggen, G. Vogels and C. Vandecasteele Models for leaching of porous materials ..................................................................................491 Polden, Insavalor and LAEPSI, INSA Lyon, France P. Moskowicz, R. Barna, F. Sanchez, Hae Ryong Bae and J. Mehu A generalized model for the assessment of long-term leaching in combustion residue landfills ...........................................................................................................................501 Royal Institute of Technology, Sweden J.N. Crawford, I. Neretnieks and L. Moreno Influence of the Type of Cement used on the Leaching of Contaminants Leached from Solidified Waste Containing Arsenic ........................................................................................513 Depart. Chem. Engineering, Kath. Universiteit Leuven, Belgie V. Dutr6 and C. Vandecasteele Verification of laboratory-field leaching behaviour of coal fly ash and M S W I bottom ash as a roadbase material ..............................................................................519 INTRON, ECN, Technical University Delft, The Netherlands J.P.G.M. Schreurs, H.A. van der Sloot and Ch. F. Hendriks Leaching of chromium from steel slag in laboratory and field tests -a solubility controlled process ? ....................................................................................................................531 Swedish Geotechnical Institute, Sweden A.-M. F~illman The application of incinerator bottom ash in road construction ...........................................541 Danish Road Institute, Denmark K. A. Phil Acid resistance of different monolithic binders and solidified wastes .................................. 551 Wastewater Technology Center .Corp., Canada J.A. Stegemann and C. Shi Research and Standardization Programme for Determination of Leaching Behaviour of Construction Materials and Wastes in the Netherlands ................................. 563 Van Heijningen Energie en Milieuadvies B.V. and ECN, The Netherlands R.J.J van Heijningen and H.A. van der Sloot Utilisation of flue gas desulphurisation by-products in the cellular concrete technology ....................................................................................................................571 University of Cracow, Dept. Mining and Metallurgy, Poland W. Brylicki and A. Lagosz State of the art of gypsum recovery for a Spanish power plant ............................................. 581 Polytechnical University, Valencia, Spain E. Peris-Mora, J. Monz¢, J. Paya and M.V. Borrachero
xvi
Fine grinding of hard ceramic waste in the rotary-vibration mill ........................................ 591 Technical University of Mining and Metallurgy, Cracow, Poland J. Sidor, A. Mariusz W6jcik and J. Kordek Influence of the Ca content on the Coal Fly Ash Features in Some Innovative Applications ............................................................................................................. 599 Universita di Messina, Universita di Reggio Calabria, Italy P. Catalfamo, S. Di Pasquale, F. Corigliano, L Mavilia Processing and application of phosphoric gypsum ................................................................. Intron, Kemira Agro, Hydro Agri, The Netherlands R. van Selst, L. Penders an W. Bos
603
Valorization of Lead-Zinc Primary Smelter Slags .................................................................. 617 Metaleurop Recherche, France, ECN, The Netherlands, Polden INSA-Lyon, France D. Mandin, H.A. van der Sloot, C. Cervais, R. Barna, J. Mehu The long term acid neutralizing capacity of steel slag ............................................................ 631 Royal Institute of Technology, Stockholm, Sweden J. Yan, C. B~iverman, L. Moreno and I. Neretnieks Reusing water treatment plant sludge as secondary raw material in brick manufacturing .................................................................................................................. 641 TNO, Reststoffenunie Waterleidingbedrijven, Boral Industry bv, The Netherlands L. Feenstra, J.G. ten Wolde and C.M. Eenstroom Assessment of chemical sensibility of Waelz slag .................................................................... 647 Polden, Insavalor, Laepsi, INSA, France, ECN, The Netherlands, Metaleurop Recherche, France Hae-Ryon Bae, R. Barna, J. M,hu, H.A. van der Sloot, P. Moskowicz and C. Desnoyers Immobilisation of heavy metals in contaminated soils by thermal treatment at intermediate temperatures ....................................................................................................... 661 IWACO, SCG, ECN, The Netherlands C. Zevenbergen, A. Honders, A.J. Orbons, W. Viane, R. Swennen R.N.J. Comans and H.J. van Hasselt Investigation strategies for contaminated soils in Finland ..................................................... 673 Geological Survey of Finland, Espoo, Finland H.L. Jarvinen Development of fast testing procedures for determining the leachability of soils contaminated by heavy metals .......................................................................................... 679 lWACO, ECN, SCG, The Netherlands J.J.M. Heynen, R.N.J. Comans, A. Honders, G. Frapporti, J. Keijzer and C. Zevenbergen Electrokinetic transport in natural soil cores .......................................................................... University of Leeds, England D.I. Stewart, L.J. West, S.R. Johnston and A.M. Binley
689
xvii
Re-use of sieve sand from demolition waste ............................................................................699 TNO Waste Technology Division, Apeldoorn, The Netherlands E. Mulder Organic substances in leachates from combustion residues ..................................................705 Link6ping University and Swedish Geotechnical Institute, Sweden I. Pavasars, A-M. F~illman, B. Allard and H. Bor6n
Leaching behaviour of PCDD/Fs and PCBs from some waste materials ............................ 715 Environment Preservation Center, Kyoto University, Japan S. Sakai, S. Urano and H. Takatsuki Environmental quality assurance system for use of crpshed mineral demolition waste to use in earth constructions ...........................................................................................725 VTT Chemical Technology, Finland M. Wahlstr6m, J. Laine-Ylijoki, A. M~ia~itt~inen, T. Luotoj~rvi and L. Kivek~s Environmental certification of bottom ashes from coal fired power plants and of bottom ashes from municipal waste incineration ....................................................................735 KEMA, Dutch Fly Ash Corporation and Waste Processing Association, The Netherlands F.J.M. Lamers, J.W. van den Berg and J.G.P. Born Quality assurance in the laboratory analysis of contaminated soils ..................................... 749 M.J. Carter Associates, England L. Heasman Dutch policy as incentive to environmentally controlled re-use of waste materials ............ 755 Ministry of Housing Spatial Planning and the Environment W.M.A.J. Willart, The Netherlands Evolution of regulations and standards for stabilized hazardous industrial waste management in France ..............................................................................................................757 SPDP Ministerede rEnvironnement, POLDEN-INSA Lyon Developpement, ADEME, France A.-F. Didier, J. M , h u , Valerie Mayeux Test methods and criteria for the assessment of immobilized waste ..................................... 765 INTRON, The Netherlands G.J.L. van der Wegen Inorganic immobilisation of waste materials ...........................................................................769 Delft University of Technology Faculty of Civil Engineering F. Felix, A.L.A. Fraaij and Ch. F. Hendriks Physical properties and long term stability of stabilized contaminated soil ........................ 781 Tampere University of Technology, Finland P. Kuula-V~iis~inen, K. Kumila and H-L. J~irvinen Evaluation of contaminant release mechanisms for soils and solidified / stabilized wastes .......................................................................................................787 Rutgers University, Dept. Chem. Biochem. Eng., USA A.C. Garrabrants
xviii
Response of various solidification systems to acid addition ................................................... 803 Wastewater Technology Centre Burlington Canada J.A. Stegemann, C. Shi and R.J. Caldwell Contaminated soil - cement stabilization in a demonstration project .................................. 815 Public Works, Engineering Division, Rotterdam, The Netherlands J. van Leeuwen, A. Pepels and G. van Ernst Stabilization of a galvanic sludge by means of calcium sulphoaluminate cement ............... 823 Univ. of Napels, Frederico II, Italy R. Cioffi, M. Lavorgna, M. Moarroccoli and L. Santoro Reuse of secondary building materials in road constructions ............................................... 831 Public Works, Environmental Engineering Department, Rotterdam, The Netherlands T. Berendsen MSWI residues in The Netherlands, Putting Policy into Practice ........................................ 841 Service Centre MSWI Residues and Waste Processing Association, The Netherlands J.G.P. Born and R.A.L. Veelenturf The Materials and Energy Potential method for the quantitative distinction between waste valorization and elimination in the cement industry ....................................................851 TNO Institute of Environmental Sciences, The Netherlands J.A. Zeevalkink Using environmental economics in decision making and policy formulation for sustainable construction ............................................................................................................859 University of East Anglia (CSERGE) and University College, England A.L. Craighill and J.C. Powell Application of secondary materials : a success now, a success in the future ........................ 869 Ministry of Transport, Public Works and Watermanagement, The Netherlands J. Th. van der Zwan Sustainable Building and the Use of Raw Materials in the Civil Engineering Sector ......... 883 RWS-DWW, The Netherlands H. Wever.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
Overview of MSWI residue Recycling by Thermal Processes Shin-ichi Sakai a and M a s a k a t s u Hiraoka b Environment Preservation Center, Kyoto University, Kyoto 606-01, Japan b Institute of Systems Engineering Research for Global Environment (ISERGE), Kyoto 600, Japan a
Abstract The melting technology reduces the volume of incinerator residues, bottom ash and fly ash, making the melted slag stable and non-toxic. Moreover, this type of treatment allows the melted slag to be used as a resource again. In Japan, the melting process was developed in the 1980's and has been in practical operation at around 24 municipal solid waste (MSVV) incineration facilities including scheduled ones. By the melting process, PCDDs/PCDFs in residues are decomposed at temperature of approximately 1,400~ in the furnace and heavy metals are concentrated in the fly ash of melting process. The drafting of an 'effective reuse manual' is introduced, aiming at promoting the safe reuse of incinerator residues, by setting reprocessing technologies, reuse standards and their evaluation methods.
1. Introduction The gross amount of municipal solid waste (MSVV) generated annually in Japan is approximately 50 million tons. Approximately 71.2% of this MSW is incinerated, producing approximately 6 million tons of residue which is then landfilled, with leachate control. Recently it has become more and more difficult to secure landfill locations, particularly in urban areas. Consequently, reducing the volume of incinerated MSW ash and looking for ways in which to reuse residues, are urgent targets to be developed. Fly ash produced during MSW incineration is classified as "general wastes requiring special controls." One of the following four treatment methods must be applied to the generated fly ash: 1) melting and solidification, 2) solidification with cement, 3) stabilization using chemical agents or 4) extraction using acid or other solvents. The melting technology reduces the volume of incinerator residues, bottom ash and fly ash, making the melted slag stable and non-toxic. Moreover, this type of treatment allows melted slag to be used as a resource again. The melting operation works by keeping the temperature at approximately 1,400~ in a hightemperature furnace by electricity or by the combustion of fuel. After the residues' physical and chemical state changes, they are cooled in order to solidify it again. In this way, the mass and volume of the residues is greatly reduced, producing a high-density melted product. By melting the residues at such a high temperature and with the change in physical and chemical state, it is possible to produce a melted slag with high stability. However, this technology needs to be improved in certain areas, e.g. reducing the rate of repairing refractory materials, and improving control technology to ensure stable operation of high-temperature melting. The melted-solidified slag can be used as construction material, such as for roads, and is also a useful material in land reclamation, since the bulk of the material is reduced by half to one-third of the original incinerator ash. Another advantage of this method relates to the fact that incinerator fly ash contain hazardous substances such as heavy metals, which can cause problems when they leach out into waterways. By this process of melting and solidification, metallic compounds are stabilized in the 'molecular' structure of the waste product, thereby preventing them from leaching out and dispersing into the surrounding environment. 2.
Melting Technology
2.1 Present Status of Melting Process Development 1,2) In Japan the sewage sludge melting process was developed in the 1980's and has been in practical operation at around 10 full-scale plants. 3).4) In some plants being operated MSW fly ash, along with bottom ash, is melted. The first melting plants used thermal surface melting furnaces,
electric arc-type and coke-bed type melting furnaces. Since then new melting technologies such as plasma melting furnaces, electric resistance melting furnaces and low frequency induction furnaces have been developed and put into practice. At present, 24 municipal solid waste incineration melting-treatment facilities (including scheduled ones) which use the system are shown in Table 1. Some of the systems are still at the trial stage of operation. Each company is, however, making efforts to proceed in their research and development and to bring their technology to the marketplace. Melting technology is almost at a feasible stage. Fusion or vitrification of MSW incinerator residues is not practiced in Europe and North America 5), but detoxification of thermal filter ash has been under development 6)
Table 1
Full-Scale Melting Plants of MSW Incinerator Residues in Japan
Municipalities 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
2.2
Numadzu City Kashima Town Eastern Saitama 2 Eastern Saitama 1 Isahaya City Sayama City Tokyo Ota Anan City Handa City Omiya City Matsuyama City Sakado City Shirane Regional Center Tokai City Abiko City Eastern Saitama, New 1 Kinuura regional center Sayama City Mima regional center Hachioji City Tamagawa regional Togane City Kamo regional center Yokohama City
Completion 08/1979 0611981 0311985 0311986 03/1987 0311991 0411991 10/1991 02/1993 0311993 03/1994 07/1994 10/1994 0311995 0311995 09/1995 09/1995 03/1996 03/1997 0311998 0311998 0311998 0311999 0312001
Capacity ton/d 20 6.5/8h 14.4 15 12.3 15 250 4.8 24 75 52 9.6 7/16h 15 10 80 15 15 5/16h 18 25 26 30 60
Unit No. 1 1 2 2 1 1 2 2 1 1 1 1 1 1 1 2 2 1 1 2 2 1 2 1
Manufacturer Kubota Takuma Takuma Takuma Kubota Kubota Daido Takuma Ebara Daido Ebara Takuma Kubota Nippon Steel Hitachi Zosen Daido I.H.I Takuma Kobe S t e e l NKK Daido Takuma Hitachi Zosen NKK
Furnace type Rotating surface Surface melting Surface melting Surface melting Rotating surface Rotating surface Electric arc Surface melting Plasma Electdc arc Plasma Surface melting Rotating surface Coke bed Surface melting Electric Arc Coke bed Surface melting Plasma Electric Joule Electric Arc Surface melting Plasma Electric Joule
Principles of the Melting Systems
At present there are a variety of furnace melting systems that have been developed and are being put into practice. These systems can be divided roughly into two categories: one uses fuels as an energy source and the other uses electricity. The systems can be further classified as follows: (1) o Surface melting furnaces o Swirling-flow melting furnaces o Coke-bed melting furnaces o Rotary kiln melting furnaces o Internal melting furnaces (2) <Electric melting systems> o Electric-arc melting furnaces o Electric resistance melting furnaces o Plasma melting furnaces o Induction melting furnace (High-frequency, Low-frequency) Some of the fuel-burning melting systems, e.g., coke-bed melting and rotary kiln melting, can not only melt the incineration residues, but can also directly melt MSW. Each of the nine kinds of systems listed has its own particular characteristics. At this stage it is
not possible to say clearly which system is the best. It is most important to use the most appropriate system for the particular conditions of each municipality, or to select a system according to a priority setting. Generally, in the case of a large incinerator with a power generation facility, the electric melting system, which can make use of the recovered electric power, can be selected. In case of a comparatively small incinerator without power a generation facility, the fuel-burning melting system will be selected. 2.3 Surface Melting This is one of the fuel burning-type melting systems. It uses heavy oil, kerosene or gas as the fuel. The structure of the furnace consists of an ash feeding device, main body and burner, as indicated in Fig. 1 7.8,9). One type of design has a pair of furnaces with the two systems facing each other. In another design, the furnace itself consists of an outer body and an inner body, with the outer body rotating. The surface melting furnace works in such a way that continuously-supplied incinerator residues melt from the surface by the heat of the fuel burning. It is then discharged via the outlet port. In this way, the melted slag hardly touches the furnace body directly, and the incinerator residues themselves act as an insulator to protect the furnace body. This type of furnace has a rather large exhaust gas volume and is more suitable for the comparatively small capacity range. INCINERATIONASH ' ~
BURNER
:::::~~.:::.:.:
ASH SUPPLY NELTING SLAG
~
SLAGCONVEYER
(a) Fixedbed type I A~ HOPPER
I NCI HEi~TI ON ASH BURliER R BODY
_._,
CONBUSTIONAlP, UTERBODY
Jl, ~J~ I ~
(b) Rotatingtype Fig. 1
_..,.., EXHAUSTGAS
(c) Fixed bed type I[ Structures of Surface Melting Furnace 7,B)
2.4 Electric Arc Melting The structure of an electric-arc melting furnace is shown in Fig. 2 lo). it consists of the furnace body, lined with refractory lining, an artificial graphite electrode which penetrates to the inside of the furnace, a power supply to feed electricity, an inlet for the entry of the residues, an exhaust and an outlet port. This type of melting furnace works by the application of alternating current to the electrode, which is arranged so as to generate an arc discharged inside the furnace. The heat produced by arcing causes the residues on the metal base to melt. The arcing generates such a high temperature that even residues containing metal can be melted evenly within a short time. The melted slag is removed continuously via the outlet port. It is quenched with water and taken out by conveyor. Any components in the incinerator residues are burnt completely in this type of furnace, and are then removed by the exhaust gas. The atmosphere in this furnace is oxidative. This technology has been applied in the field of steel making.
POWER SOURCE EOUIPI~ENT
I
'
~
.
'~J/
L---,EL.:rING slAGLAYE, Fig. 2
Structure of Electric Arc Melting Furnace lo)
2.5 Plasma Melting This is another type of furnace that uses electricity to melt the incinerator residues. The structure of the furnace is shown in Fig. 3 11, 12). It consists of the surface body, with refractory lining, plasma torches, and a power supply system. There are a variety of plasma torch designs in use, made by the different manufacturers, and each has its specific character. This type of furnace works as follows: first it makes an arc discharge at the electrode inside the plasma torches. This is then passed through the plasma formation gas (air or inert gas) to produce a high-temperature plasma. This plasma is then directed to the incinerator residues by being continuously supplied into the melting furnace. In this furnace there are two types of atmosphere, oxidation and reduction. The melted slag is continuously removed through the outlet port. .. I ~{ClIIERAT I ON ASfl
~ i . 9 ----14 r.,.. 9 PL~SH/,FORgATIO" GAS I I "
. ~ llOeffR~
~1" C(~BUSTIOfi(;8~SER /
i
:-i
I
!
PLAS~ I"OR~
,,~.p~ ~o~,.~ F.J
POffi~R SOURCE EOUI P~IENT
(a) Single torch .type
IdFITI ~ $1.,~G
(b) Twin torch type Fig. 3
Structures of Plasma Melting Furnace 11,12)
3.
Behavior of Heavy Metals and PCDDs/PCDFs in the Melting Process
3.1 Standard Leaching Tests Leaching of heavy metals from the slag was evaluated using the standard leaching tests defined in Notifications No.13 and No.46 of the Environment Agency (JLT13, JLT46) in Table 2. Some points about the standard leaching tests are discussed in the next section. A typical analytical result is shown in Table 3. All of the specified substances in the slag leachate were either nondetectable or below the detection limit, demonstrating that the slag satisfies the environmental standards. In addition, the very low leaching of lead, which has recently become a problem in the effective utilization of recycled materials, is one of the remarkable features of this process.
Table 3
Leaching Test Results for MSW Incinerator Residues and Melted Slag 13)
Sample Item pH Cadmium, Cd Lead, Pb Hexavalent chromium, Cr 6. Arsenic, As Mercury, Hg Cyanogen, CN Selenium, Se Alkylmercury, R-Hg Organophosphorus, Org-P PCB Thiram Simazine Thiobencarb Trichloroethylene Tetrachloroethylene Dichloromethane Carbon tetrachloride 1,2-dichloroethane 1,2-dichloroethylene Cis-1,2-dichloroethylene 1,1,1-trichloroethane 1,1,2-trichloroethane 1,3-dichloropropene Benzene Zinc, Zn Copper, Cu Chloride ion, CI Electric conductivity mS/m
Fluidized bed Stoker furnace furnace fly ash fly ash mg/I mg/kg mg/I mg/kg 12.3 6.8 0.01> 0.1> 33.5 335 28.3 283 10 100 0.04> 0.4> 0.2> 2> 0.01> 0.1> 0.01> 0.1> 0.0005> 0 . 0 0 5 > 0.0005> 0 . 0 0 5 > 0.1> 1> 0.1> 1> 0.01> 0.1> 0.01> 0.1> 0.0005> 0 . 0 0 5 > 0.0005> 0 . 0 0 5 > 0.01> 0.1> 0.01> 0.1> 0.0005> 0 . 0 0 5 > 0.0005> 0 . 0 0 5 > 0.006> 0.06> 0.006> 0.06> 0.003> 0.03> 0.003> 0.03> 0.02> 0.2> 0.02> 0.2> 9 * 0.03> 0.3> 9 9 0.01> 0.1> 9 9 0.02> 0.2> 9 9 0.002> 0.02> 9 9 0.004> 0.04> 9 9 0.002> 0.02> 9 9 0.004> 0.04> 9 9 0.03> 0.3> 9 9 0.006> 0.06> 9 9 0.002> 0.02> 9 9 0.001> 0.01> 5 50 850 8500 1> 10> 1> 10> 9530 95300 9500 95000 3580 3630 -
3.2
Behavior of Heavy Metals
(1)
Behavior and Mass Balance of Metals 1,s)
Minimum limit of Molten slag determination (hydropulping) mg/I mg/I mg/kg 9.3 0.01> 0.1> 0.01 0.01> 0.1> 0.01 0.02> 0.2> 0.02 0.01> 0.1> 0.01 0.0005> 0.005> 0.0005 0.01> 0.1> 0.01 0.01> 0.1> 0.01 0.0005> 0 . 0 0 5 > 0.0005 0.01> 0.1> 0.01 0.0005> 0 . 0 0 5 > 0.0005 0 . 0 0 6 > 0.06> 0.006 0 . 0 0 3 > 0.03> 0.003 0.02> 0.2> 0.02 0.03> 0.3> 0.03 0.01> 0.1> 0.01 0.02> 0.2> 0.02 0 . 0 0 2 > 0.02> 0.002 0 . 0 0 4 > 0.04> 0.004 0.02> 0.2> 0.02(0.002) 0.04> 0.4> 0.04(0.004) 0.03> 0.3> 0.03 0 . 0 0 6 > 0.06> 0.006 0 . 0 0 2 > 0.02> 0.002 0.01> 0.1> 0.01(0.001) 0.1 > 1> 0.1 0.1> 1> 0.1 2 20 1 4.48 -
Inorganic compounds like metallic elements, especially in fly ash, are redistributed after the melting treatment according to the boiling temperature. It is considered that metals with high boiling points like Si, AI and Ca, are converted into slag and substances with low boiling points like Cd and Pb are converted into fly ash or melting furnace exhaust gas. As shown in Table 4, the concentrations of heavy metals with low boiling points like Cd and Pb in ESP ash from melting furnaces are 5 to 10 times higher than those of fly ash. Based on this analysis, the flow rate of flue gas and the quantity of solids formation, the mass balance and the transfer rate of each constituent are shown in Table 5, assuming the input to be 100. SiO2, AI203, CaO, Fe, Mg, Mn, T-P, TiO2, T-Cr and Cu indicate high transfer rates into slag. In contrast, Cd and Pb volatilize into flue gas and are finally concentrated into ESP ash from melting furnace. Na, K, T - S , T-CI, As and
Table 2. Test name
Env~ronmental Agency Notification No. 13 (Note 1)
Environmental Agency Notification No. 46 (Note 1)
Leaching Test Methods
Leachmg vessel
Unspecified
Unspecified
Ministry of Health and Welfare Tentative draft of slag test (Note 2) Airtight bottle (CO2 method) or beaker (pH-static method), (1L polyethylenebottle or 1L glass beaker)
Sample
< 5 mm
< 2 mm
Sample mass
> 50 g
'509
Solvent
Distilled water (Adjusting Distilled water 1) pH 4. CO?saturated water (C02 pH 4 through the way of 20 minto pH 5.8-6.3 by HCl or (Adjustingto pH 5 8-63 3 method) bubbllng of deionized water by NaOH) by HCI) 2) Adding HNOl to deionized water, C02 gas. and keeping the 1st eldon pH 7 and the 2nd one pH 4. (pH-static method) 10. 1 10 1 10.1 ( 5 . 1 ~ 2 ) l0:l 1 1 1 (COZmethod). 1
US ratlo Leaching frequency
Horizontalshaklng (200 t~meslm. ampltude 4-5 cm)
Duration
6 hours
6 hours
Filtration
Ipm GFP
Temperature Ordinary (approx 20°C) Note 1
Note 2 Note 3 Note 4 Note 5
TCLP pH dependency test (Author et at's commonly (EPA Method 1311) used method) 1L beaker at this test Any mater~al compatible with waste, zero-head space conta~ner
Unspecified (1L glass beaker at this test)
1L beaker
10-30 mm
20-50 mm (C50 mm: uncrushed)
c 125 pm
Uncrushed (fly ash, hydropulped slag)
< 9 5 mm
> 50 g
'509
169
50 g at this test
100 g
1) Acetic ac~dbuffer Adding HNOsto deisonized - A t this test, distilled water, and keeping the 1st water and HNO3 or NaOH solution (pH 4 93) Using solvent different in 2) Acetic acid solution elub'on pH 7 and the 2nd (pH 2 88) (Note 4) acidity (alkalinity) or one pH 4. keeping the leachate a certain pH 10.1 20 : 1 1 0 0 : l (50:1x2) 1 1 2 (Note 5)
-
2 (pH-static method) (Note 5) HorizonMl shaking (200 timeslm. amplitude: 4-5 cm)
Agitation
Availability test (NEN 7341)
Ministry of Construction Tentative draft of C02 method
- Horizontal shaking, 200 timeslm.
Stirring and splashing (200 rpm)
Stirrer
Stirrer
Rotating and shak~ng (30 + 2 rpm)
24 hrs (C02 method). 3 hrs x 2 (pH-static method) After 20 mln centr~fugal 0 45 pm MF separation at 3000 rpm. 0.45 pm MF
24 hours
3 hours x 2
23 hours at thls test
18 hours
At this test, 0.45 pm MF
0 6 - 0.8 pm GFF
Ordinary (approx 20°C)
Ordinary (approx. 20°C)
Ordinary
22.3 i 39:
amplitude: 4-5 cm (C02 method) - Stlrrlng and splashing (pH-stabc method)
Ordinary (approx. 20°C)
0.45 pm MF After 20 mincentrrfugal separation at 3000 rpm. 0.45 IIIT MF Ordinary
dichloromethane,carbon tetrachloride, 1.2-di-, 1,l.l-tri- and 1, I ,2-trichlorwthane. 1,3dichloropropeneand benzene (volatile matters), an Erlenmeyer flask For trl-, tetra-, 1.241- and ~i~-1,2-di-chloroethylene. wlth screw cap (500 ml) was used. As for agitation. 4h-stirring by stirrer was implemented. Regarding filtration of elubon, the filtrate was extracted by syringe and filter paper was attached to the syringe. This IS the same method that is applied to the examinahon of volatile substances in sludge (Environmental Agency, nothication no.13) or soil (Environmental Agency, notification no 46) Test In C02 method or in pH-static method is selected. pH targek at this test were 2. 4. 6, 8, 10. 12 and 13. Dlstllled water is added to the sample of 5 g and they are shaken for 5 min, pH IS measured, the solvent of 1) is chosen if pH is over 5 If pH IS below 5, 1.0 N HCI of 3.5 ml IS added and 10 mln-shaklng 1s done at 50°C. If pH 5>, 1) is selected. and ~fpH 5c. 2) IS used. New solvent is added to filter residues and the leaching operation is repeated.
Table 4
Compositions of Solid Materials
Fly ash
Item* Moisture Heating Value Ash Combustible C H N Volatile-S Volatile-CI O TotaI-S TotaI-CI Si Ca AI Na K Mg Fe P TiO2 Mn Cd Pb Zn Cu As TotaI-Cr TotaI-Hg
0.56 58O 89.3 10.7 5.9 0.28
20%
Crushing ~-~1 Sieve
Grog ,.
Fret Mill
Matrix
Weighing[
3.0mm Mesh
~[ Crushing ~-*l Sieve
Ceramic Clay
Pigment
Fret Mill
1.0mm Mesh
Mix-MI 15 mind,
l ~l Weighing [ q B!endin ~ [ Molding 35%
Weighing Ceramic 25% Clay 20%
[Weighing [
~l Weighing[
l
,,,
Water& Binder
2%
Packing ~ ['H Final . . Product . . ~-~ Inspection ~-~ Sintering II Shrink Wrapping
~-
2OOtonF.P.
~._.._l
Tunnel Kiln 1200~176 about 80 hours
Figure 6 PavementBrick Manufacturing Process
[I Drying I~ 80~150~ abt.48 hours
39
5.3. Quality of the Products (1) Color The color is brownish, with delicate variances due to the location of the brick on the truck as it received different temperatures and oxygen of varied densities. Black dots appear on the surface caused by the oxides of metals remaining in the slag. Photograph 1 displays the appearance of the bricks as they are laid own on the floor.
Photograph 2
Pavement Brick Photograph 3
Pavement Brick
(2) Quality Table 6 shows the physical properties. These are the average values derived from 10 samples drawn at random from the products. The water absorption ratio at 4.1% and the compression strength at 1,278kg/m 2 meet the JIS standard of 200kg/m 2, respectively.
Table 6
Physical Properties of Pavement Brick Item
Unit
..A.p.parent Pore Ratio
%
Avcrag e 9.4
.......
! i i
Standard Deviation 1.06
~
i JIS R 1250 Standard i Brick No.3 i :!
i Ordinary ~ Pavement Brick i
-
t
-
Water Absorp.t.!o.n..Rate.................%...............:..5:.1........i .............9...5.............[.............!e.s.s..t..h.an...1..3..............i .....................-...................... 9
Apparent Specific Gravity
-
2.5
~
0.027
i
-
i
2.35
Bulk Sp..ecific Gravity.
-
2.26
i
0.018
i
-
i
2.29
......................................................
....C.omp.res.s.i.o.n...S.t.r.engt..h. Bending Strength
" .....................
.....k.~cm 2
.1.,..2..7..8.. ......i 134
!
; ..................................................
128.1
i
21.7
i
T ............................................
more than 200
'
-
-
!
-
mm
226.5 i
0.78
i
Standard Size
i
230 m m
i Width
mm
112.51
0.58
i
Standard Size
9
Measurement
T ...............................
i Length
9
kg/cm 2
~' .....................
~
i
:
114 m m
40
(3) Leaching Test In order The The
to determine
the safety
of the brick,
test specimens
were
bricks
test methods
were:
Publication
three
pH Method(controlled Table
to pH4
7 shows
Table 7
crushed
using
the leaching
elution
to sizes
tests have
smaller
No13,No46
than
been
carried
5mm,
out.
and the whole
of the Environmental
Agency
bricks. and
Low
HNO3).
test results.
Leaching Test Results of Pavement Bricks
Specimen
Crushed
Whole
Environmental Soil
' i t e m ............. U n i t ............. Nol3" ................ No4"6 .......... "pH4iHN'O3i ........... N o l 3 ................. No46" ........... p'H'4iH'N'O3i"
Standard
.~..~ .......................................... .7:.3 ..................... 7:.8 ....................... 4..1 ........................ 6..4. .................... .6:.4 ....................... .4..2 ......................... .-. ............ ..~-..rI.z. ........... m . ~ ................. - .................. . F e ~+ + C e ~+
The results are expressed in mMol 02 per gram solid material. Exp. 6 Diffusion tube tests) The diffusion tube is filled with untreated, wetted material to a length of 25 mm with a consistency similar to that of the labelled material. The Na-22-tracer labelled material is added up to a total length of 50 ram, and the second piston is put in position. The tube is stored in a saturated environment to avoid drying out during the experiment. After one day, the combined segments are cut into slices and are transferred to pre-weighed counting tubes, dry at 85 ~ weigh, and count in a sodium iodide crystal connected to a two-channel analyzer. The mass of the slices and the total mass of the tube contents are used to calculate the axial length of each slice. The effective diffusion coefficient were calculated based on the mobility of the Na-22-tracer, sample weight and the moisture content. From the pDe of Na and its free mobility in water the tortuosity of the granular matrix is derived. Results and discussions
Leaching concentration for JL 7"-13 Table 2 pH and metal concentration for JLT-13 The results of JLT- 13 are shown (unit: mg/L except pl-I) in Table 2. As the final pH is high, Pb pH Cd Pb Cu Zn . is leached from original fly ash in high original 12.6 ND 7.44 0.04 2.54 concentration due to its amphoteric ch 12.4 ND 0.01 ND 0.04 ph 11.0 ND ND ND ND character (Japanese standard of Pb for fe 10.6 ND ND ND ND waste disposal is 0.3 mg/L). All of the treated materials meet the requirements, because the release of the metals is controlled. However, it is a result of only one single batch test and other evaluations are needed to ensure that there are no long term environmental impacts. Availability of metals Comparing the availability of the Table 3 Release of metals in Availability Test (unit: mg/kg) materials, we can see a significant Cd Pb Cu Zn difference between the three treated original 66 168 139 4490 materials (Table 3), though there is ch 55 166 10 3500 almost no differences in JLT-13. For ph 14 0 23 1590 example, "ph" treatment decreases the fe 54 160 82 3520 availability of Pb significantly, though "ch" and "re" material show little decreasing of the availability. For the other metals similar changes are observed. When availability is reduced, this implies that the metals are incorporated in insoluble mineral phases. The availability reflects a leaching potential and as such is an important factor for evaluating long term environmental impact. pH and pe pe of the leachates are plotted versus pH (Fig. 1). Here pe is a value of negative logalithm of electron activity and there is a relationship pe = EH/59.2 (En means the standard hydrogen potential). For normal oxygenated water with pH adjusted to cover the range pH 4 - 13, there is an almost linear relation (slope is about 1) between pH and pe 6~. For leachates of fly ash or treated residues, there is a similar relation far from the
98 I level observed in oxygenated water. This points DW ( 0 2 = 0 . 2 a t m ) - - , - - o r i g i n a l - = - c h - - - - - p h - - ~ - f e I 18 at strong reducing properties for all. The intercept of y-axis is indicative of the redox behaviour of the materials. As "ch" treated material has the smallest intercept value, we can see that chelating agent has a high reducing capacity. This is in agreement with the results of 6 redox property test. The redox capacity values 2 derived from redox property test are shown in Table 4. The lower value for ph and fe treatment relative to the original material may 1 4 7 10 13 oH be due to some kinds of oxidation during the treatment process. On the contrary, chelating Fig. 1 Relationships between pH and pe agent have reduced the original Table 4 Redox capacity of samples (unit: mmol O2/kg) material in the treatment Sample original ch ph fe i process, redox capacity 387.8 448.8 235.0 188.1 I
14
"~-
-2
n
i
i
t
Release of metals Release of Cd, Pb, Cu, Zn versus pH from each materials are plotted (Fig. 2). As for cadmium, the leachability is reduced for all treatments. Particularly, the treatment :
original
. . . . Availability
-
ch
=
= fe ]
ph
1000
",,.
100
\ |
10
0.1
UD
E
2
5
8
11
2
14
1000
5
8
11
14
10000
_m
Q) n,
Pb
.
100
Cu
. ~
Zn
1000 100
10
10
1
1
0.1 2
5
8
11
14
0.1 2
5
pH Fig.2 Metal release vs. pH of lcachate
8
11
14
99 with chelating agent results in a significant reduction in release in the pH range 6 to 10 ( > 500). For the two other treatments ph and fe the reduction is not as large, although still very significant. In the pH range below 4 the difference between the different treatment options is not as large (factor 2 to 10). As for lead, there are two pH domains to be discussed in relation to treatment effects: one is the behaviour of under alkaline conditions and the other is the behaviour under acidic conditions. In the neutral pH range leachability of Pb is small for all materials including the original material. All three treatments methods reduce Pb leachability significantly in the pH range 5 to 12. The "ph" reduces release even down to pH 4. In the acidic range below pH 4 the difference becomes less until at pH 2 the difference amounts to a factor of 2 - 3 only. As for copper, "oh" treated material has a strong effect over the entire pH range from 2 to 13. Although "ph" and "fe" materials show similar behaviour, there is a difference when both change in release and change of ANC (acid neutralising capacity) are taken into account. This aspect is discussed in the next section. As for zinc, there is small difference among three treated materials, though they are all lower than 10 ~ original the untreated fly ash in the pH range 4 to 10. -=- ch 8
ANC and release of metals The ANC [meq/g] versus pH are plotted E4 in Fig. 3. There is a significant difference =~ < between the four materials. In the neutral pH o range 5 to 9, there is a significant difference -2 between original fly ash and treated materials. It 1 4 7 10 13 means that it is harder to decrease the leachate pH pH for original fly ash than treated materials. Fig. 3 ANC value vs. pH The sensitivity to externally imposed pH changes as reflected by the -----Cd-rebase . . . . . w a s t e standard . - A d d e d acid change of ANC in 1 100 . 10 100 10 relation to pH is an inal 8 important factor for 8 evaluation of treatment methods. In Fig. 4, the 4~ Cd release curve and added acidity curve q %t~f \ versus final pH are 0.1 , , .... "%....A 0 0.1 0 1 4 7 10 13 4 7 10 13 plotted. Japanese 10 100 10 standard for waste ,9 ch fe disposal for Cd (0.3 '| mg/L, it is equivalent 10 to 3 mg/kg of release.) are also drawn as a dot-dash .9. . . . i line in the same figure. ". _ -0.1 However, the pH3 1 4 7 10 13 oH #I release behaviour is Fig. 4 Added acidity and release of Cd not the only aspect to "& .
.
.
.
I
!1 ~"""'""".\~Ii
~
J1
,
100 be addressed here. The 25 25 amount of acid required to odginal ch 20 20 make a change in pH is 15 15 crucial in the sensitivity ===Ill 9 10 III 10 analysis. The buffeting 5 5 F 9 9 capacity that reflects the I I i i mj | | i i l~ iiii i 0 0 resistance of the material to 3 5 7 9 11 1 v 1 3 5 7 9 11 a change in pH due to 25 25 o acidification is higher for ,, fe 20 ph 20 "ph" than for "fe". In Fig. 5 15 15 this is illustrated by the 10 10 concentration change of Cd 9 a 5 II II 5 9 9 I I tlll relative to the amount of i i i l 0 ' ' ' " 0 acid required to make the 1 3 5 7 9 11 I 3 5 7 9 11 change. A larger value pH indicates that less acid is needed to bring about a Fig. 5 pH vs. dC/d(acid) of Cd for 4 materials significant change in concentration. In the graph for "ph" and "re" two maxima are observed; one at pH 7.5 8 and one at pH 4. In comparison with the untreated material the sensitivity for pH change has shifted from pH 8 to pH 4 for all treated materials. In the "ch" treated material, the peak at pH 8 is eliminated entirely. A pH of 4 will not be reached under landfill conditions easily provided there is no mixing with material containing degradable organic matter. Similarly, the role of changing the reducing conditions can be discussed. The information needed for such an evaluation is currently insufficient. However, as a general statement it is clear that oxidation of the sulfide treated materials must be avoided as significant changes in metal leachability can be expected upon oxidation. In a management option to be selected for disposing of the waste, this aspect must be addressed.
pH change of leachate under open condition and closed condition The pH change of the leachate versus L/S value of open and closed test are plotted (Fig. 6). There is a significant difference between open test and closed test. In open test, pH of the leachate decrease continuously. In the closed test the pH increases first and then decreases or open closed remains at the same pH 14 value. The final pH of the open test is always lower 12 12 ~ than that in the closed 11 11 vessel for all samples by 1 10 lO ~ o~in= to 2 units. It can be 9 -o-ch 9 explained by CO2 in the -'-r:' 8 i , ,J. l atmosphere" and it is an 8 0 20 40 60 80 100 0 20 40 60 80 100 important to consider the p.effect of CO2 during US Fig. 6 Change of pH of the leachate leaching test operations.
101
Redox potential of leachate under open condition and closed condition T h e EH (standard 800 8OO hydrogen potential) Chelate '- . odginal 6O0 60O values are plotted versus pH in Fig 7. Solid marks 400 4OO (QO l, etc.) mean 200 2OO closed tests' results, and 0 0 open marks ( O Q I--1, etc.) shows the open q , -2OO -200 9 10 11 12 13 14 8 9 10 11 12 13 14 tests' results. As the ORP "~ 800 is a function of the pH, it - .. ferrite phosphate is meaningless to 60O 6OO compare the ORP values 400 4OO directly. In order to 2O0 2OO compare the ORP values in a meaningful way, the 0 0 measured ORP values for -200 -20O the treated and untreated 8 9 10 11 12 13 14 8 9 10 11 12 13 14 material must be pH compared relative to the Fig.7 E. vs. pH for serial batch test dotted lines reflecting the ORP for oxidised conditions as a function of pH.In general, EH values of the leachate in the open test are higher than the closed ones. This is a result of oxidation by 02 in the air and exhalation of Hz from the reaction of the material with alkaline water. For original ash, the lowest points are the result of the closed test at L/S = 2. In the test the swelling of the closed bottle was obserbed. It may be the effect of the H2 gas generation from the contacting with distilled water. This gas generation can be explained by following reaction: 2 AI + 3 H20 + 8 OH" ---> 2 AI(OH)4" + 3 H2 The formation of hydrogen gas from original ash during the leaching test was confirmed by gas chromatography. The swelling of bottle was not observed for treated materials. The treated materials were mixed with water in treatment process at which state part of the H2 may escape to the atmosphere, whereas, the leaching test is the first opportunity to contact with water for untreated fly ash. The large difference between the original fly ash in the open test and closed test is attributed to the degassing of H2 from the open vessel. The "ch" treated materials are generally reducing and their leachate under open conditions are lower than the dosed test results of other materials. Furthermore, for chelating material, EH v a l u e s increase at the same pH. It means the material is oxidized during the serial batch test. This is attributed to the higher reducing capacity of this material. On the contrary, the plots of ph and fe materials are almost pararell to the dotted line. It indicates these materials are neither oxidized nor reduced during the serial batch test. 41"
It
,=,
tl
,h
lb.
=l
i
i
I
i
i
i
i
i
i
i
q,
41t
i
i
i
i
i
i
i
i
=l,
i
I=,
i
102
Release of metals under open condition and closed condition As for the I 9 sedalbatch Cor~ert ....... A~ilability 9 JLT-13 I metal concentration and release, almost 100001000I Pb(close) 100001000t" Pb(open) i Pb, Cd, Cu, Zn are 100 ""i ................................... 100 ""i ............. ~'~................ ," not detected from 1 0 ~ " treated materials, pH 1 of the leachate is one 0 20 40 60 80 100 0 20 40 60 80 1O0 of the most important 10000 t" ...................... factors. Therefore, ~ 1000 ............... 7_n(close) 100001000100t10...................................... 9 Zn(open) only metal release E lOO from original ash are h,.~ shown in the figures 1 I, , , , 1 (Fig. 8). Pb, Zn and " o 20 40 60 80 100 0 2o 40 60 80 1oo Cu in closed test 1000 tOO0 / ! leached out more 100 ..................................... 9 t 100 .... Cu(close) 1 _, ' . ' Cpen)u(o than open test. Na 10 10 and K leached out in 1 highly concentration o~ "-,---'-"~, , , :t 0.1 0 20 40 60 80 100 0 20 40 60 80 1O0 from all materials, and there is small US differences between Fig.8 Metal release vs. L/S in the serial batch test open and closed condition. The I -*-- original ---- ch --*- ph --~ fe release of salts from __~ 20 2O original fly ash and lS J 15 z~ treated materials is 1 1o 10 plotted in the figures 5 K(open) I 5 K(closed) (Fig. 9). Na a n d K 0 from treated 0 20 40 60 80 100 0 20 40 60 80 100 materials leached out t~ 20 20 more than original fly 15 ,'---~--~ )' t 15 ash. It implies that cr 1 0 .5 '~ --f-' " i . with the additives Na Na(ope.) S and K are added. The 0 salt release from 0 20 40 60 80 100 0 20 40 60 80 100 untreated and treated US materials is high. In Fig. 9 K and Na release vs. L/S in the serial batch test evaluating environmental impact, it is important to focus on a broader range of elements than only heavy metals. Several studies have shown that also oxyanions can be important as they often show a maximum leachability at neutral pH 8). 1
|
,I.
I
i
,I.
/
i
!
'
0
0
i
i
i
'
1
Effective diffusion coefficient and tortuosity of materials In order to be able to quantify release under confined conditions, where diffusion may be the release controlling mechanism, it is important to determine if the treatment influences the physical behaviour of the material. This is done by measuring the
103 tortuosity of the treated materials by carrying out diffusion tube tests with a Table 5 pDe value and tortuosity of materials constituent considered not to interact with material od~linal ch ph fe the matrix (here Ha is chosen). The pDe pDe 9.59 9.85 9.84 10.02 value for Ha, which is an negative tortuosity 10~ 10~ 10~ 10T M logarithm of an effective diffusion coefficient value, for each material are shown in the Table 5. The tortuosity is caluculated from Do/De. Do is the free mobility of Na in water and the value is pD0 = 8.88 at 22 ~ All treatments show a pDe value larger than the untreated. This is attributed to the formation of a denser matrix in part caused by additional precipitate formed. This is consistent with the observation that "re" treated material the effect most significant. As was shown in the batch extractions this positive physical effect does not necessarily mean a lower release as other factor come into play. Conclusion
In order to evaluate the treatment of MSWI residues with chemical agents, the performance of three kinds of treated materials are compared with the behaviour of the untreated MSWI fly ash. Each treated material shows good results for JLT-13 test, however the evlauation based on this single extraction test is too limited to assess the potential environmental impact of the materials in different disposal scenarios. The treatment using a chelating agent is based on complexation of metals with an organic sulfide and the treated material showed a very high reducing capacity and strong retention for metals over a wide pH range. The treatment with phosphate showed a significant decrease in availability, especially for lead, and a very low leachability over the entire pH range from 4 to 13. The treatment with ferrite showed an increased high physical retention as well as a good retention in the pH domain 5 - 12. In order to evaluate the performance of the treatment by chemical agents, the changes in leaching behaviour of elements as a function of pH were compared with the buffeting properties of the treated materials (Acid Neutralizing Capacity), that can be derived from the pH controlled leach test. Besides, the final ORP value of leachate are plotted versus pH of the leachate, the linear relationship are recognized for each material. In order to know the effect of the air during the leaching test, the serial batch test under open conditon and closed condition are performed. There are clear differences between open test and closed test in view of pH and ORP. It is a result of carbonation and oxidation by respectively CO2 and 02 in the air. Furthermore, the H2 generation from untreated ash explains the low ORP in a closed test. In terms of waste management the following conclusions can be drawn: - When materials are produced with very strong reducing properties either in-plant or afterwards through treatment, the management of the materials requires that measures are taken to ensure that the material will never be exposed to the atmosphere, as oxidation will ultimately lead to loss of retaining potential and could result in uncontrolled release. - The generation of H2 in treatment plants is potentially dangerous, if this aspect is not sufficiently recognised in the design and operation of such plants. - Based on the information generated in the more elaborate tests, better predictions of
104 the behaviour of materials can be made. Particularly, in relation to quantifying the changes resulting from changes in exposure conditions (failure of lining, top cover or exposure to the atmosphere). This information can be used to better manage these wastes.
Acknowledgement This paper is based on the results of the study during Mizutani's visit to ECN in the Netherlands. Patrick Cnubben, Dirk Hoede, Petra Bonouvrie and Marco Geuzebroek are gratefully acknowledged.
References 1) Environment Agency in Japan: Japanese Environment Agency Notificaition of No. 13 (1973) [in Japanese] 2) CEN TC 292 Working Group 2: Compliance test for Leaching of Granular Waste Materials and Sludges, Tenth Draft, (1994) 3) NEN 7341. Leaching characteristics of building and solid waste materials Leaching tests - Determination of the availability of inorganic components for leaching. Draft 1993 (previously part of NVN 2508), Netherlands Normalisation Institute, the Netherlands 4) Drait NVN 7348 Determination of reducing properties of materials. 5) van der Sloot,H.A., de Groot, G.J., and Wijkstra, J." Leaching characteristics of construction materials and stabilization products containing waste materials. In: Environmental aspects of stabilization and solidification of hazardous and radioactive wastes ASTM STP 1033, Cote, P.L. and Gilliam, T.M. Eds., American Society for Testing and Materials, Philadelphia, 1989,pp. 125-149 (1987) 6) Lindsay, W. L. : Chemical Equilibria in Soils, A Wiley-Intersicence publication (1979) 7) Schramke, J.A. : Neutralization of alkaline coal fly ash leachates by CO2 (g), Applied Geochemistry, vol.7, pp. 481-492 (1992) 8) H.A. van der Sloot: Developments in evaluating environmental impact from utilization of bulk inert wastes using laboratory leaching tests and field verification. 1996.Waste Management, 16 (1-3), 65-81. 9) IAWG: An International Perspective on Characterisation and Management of Municipal Solid Waste Incineration Residues, final document, Chap. 20 (1995)
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
105
RECYCLING FOR ROAD IMPROVEMENTS Presentation by Charles J. Nemmers, P.E. at the W A S C O N '97 Conference The Netherlands, June 1997
Einstein once said "The significant problems we face cannot be solved by the same level of thinking that created them." These words challenge us to think anew how we use and reuse materials for the construction and reconstruction of transportation facilit;es, especially highways and other roads. The Road Transport Research Program Steering Committee of the Organization for Economic Cooperation and Development (OECD) headquartered in Paris, formed a Scientific Expert Group to review the issue of recycling in road building and I will be reporting on what we did. But it is the need to do our work at a different level (i.e., Einstein quote) that made our collective work both challenging and productive. Fifteen nations plus two Asphalt associations were involved in our study group, a survey of recycling in methods, products, laws, and procedures in these countries was conducted and a state-of-the-practice report was prepared. Several of my colleagues on this report are with us here today and I trust that what I say will be close enough to what we wrote that they will be able to identify it. Recycling is now a well proven technology, often a preferred choice for construction, a backbone for an entire equipment industry, and a requirement, if not a necessity, in many countries. Since 1977 - the year of publication of the first OECD report on the Use o f Waste Materials and By-Products in Road Construction - both the volume and quality of the recycling of road by-products in the road sector have increased significantly in OECD countries. However, the generation of waste remains high as shown in the chart below.
106
Amounts of Municipal Trash in the World The world's most industrialized nations produced an estimated 450,000,000 metric tons of municipal trash in 1992. This figure shows kilograms of trash produced per person per year in the urban areas around the world in selected countries
Kilograms of Trash Produced Per Person In Municipalities of Selected Countries
U.So
9
Australia
J
. . . . . . . .
"Cur~ "
............ Lfa
Canada Iceland Norway Netherlands
i.g- .
.
.
.
.
Luxern bourg France Denmark
9 I
'i,
Japan Betglum Sw itzerland Turkey
n_.
Hungary
,,
Spain
,.
Italy
.,
Poland
.,
Portugal Greece Mexico
~u
~
......
o,.
.. . . . . - ! . . . . . . . . ,., . . . . . . . .
L
i .~ =
0
(Source: Washington Post, 23 November 1996)
200
400
|00
8OO
107 In 1995, the OECD Road Transport Research Program Steering Committee determined that the advances in recycling called for the Scientific Expert Group on Recyclingfor Road Improvements to take a new look at this important subject area. The primary goal of the Group was to generate information that would describe the state-of-the-art in recycling and promote recycling of waste and by-product materials, especially those generated in the road sector, in road construction. It was anticipated that the Group's efforts would help to: establish or change policy in Member countries; support improvements in current specifications; identify workable technologies; and identify needed research and knowledge transfer initiatives with the expectation of increasing recycling efforts and discovering more innovative and efficient solutions. SUCCESSFUL RECYCLING REQUIRES L E A D E R S H I P While addressing the recycling of both road by-products and non-road by-products in this report, it is agreed that recycling of road by-products takes precedence in order for t'_-z road industry as a whole to take responsibility and demonstrate leadership to the fullest extent possible. The principle of "cleaning up own house first" must apply. The responsibility for road by-products belongs to the road sector. The following general hierarchy for waste management was adopted (applicable to any industry; in this case, applied to the road construction industry): 9 9 9 9
Minimise waste production. Recycle in parent industry. Recycle in other industries. Incinerate. 9 Incinerate with energy recovery; 9 Incinerate to reduce volume. 9 Dispose of in a landfill. Following this hierarchy, an industry (i.e., road construction) not only sets an example for recycling in general, but takes responsibility for reducing its own by-product disposal problems by first attempting to recycle within the industry itself. In the case of the highway industry, it certainly makes sense to recycle road by-products in roads. Not only will this reduce costs of disposal during construction, but the material similarities suggest that such uses would be technically feasible as well. It also makes sense to use by-products from other industries, when appropriate. In fact, the report shows that in some cases (e.g., fly ash in cement concrete) certain by-products may actually benefit road construction by improving the materials.
108
From information supplied by the participating countries, the Group identified three areas that must be jointly considered for successful recycling efforts. They are: 9 engineering factors; 9 environmental factors; 9 economic factors; Each of these areas contain elements - ranging from the political context to specific engineering risks - which have a significant bearing on the potential application of by-products. SURVEYS ARE BASIS OF STATE OF THE PRACTICE Two major surveys were conducted as part of this study. 1) Survey to assess the extent of current use of various by-product materials in Member countries. This assessment was accomplished by distributing survey questionnaires to OECD Member countries. Materials generated by road construction were considered separately from those generated by other industries. a) For road materials, the information requested in the survey included extent of use in various road applications, amounts available, a summary of important properties, test methods and guidelines and description of any new techniques being developed. b) For non-road materials, the information requested also included extent of use in
109 various road applications, additional information was requested on amounts, material tests and acceptance criteria, construction equipment and procedures, quality control tests, standard specifications and factors used in evaluating environmental andeconomic suitability. 2) Survey to assess waste management and recycling policy in Member countries, especially in regard to road building. Information requested here dealt with official policy, organisation(s) responsible for policy, economic issues affecting policy, regulations, obstructions to implementation and technology transfer. These surveys gave us engineering environmental and economic input into our three circle model. The first survey showed that in the 20 years since the last OECD report on recycling in road construction more c-.untries are now recyclin~ many different materials and much more of these materials as part of road construction. Recycling of asphalt pavement is being done in all of the countries surveyed. Nearly 100 million tons of RAP are being produced annually. Other waste materials from non-road industries are also being recycled into road construction and over twenty of these by-products were identified. These products ranged from old tires to various slags and ashes, to glass, paper and plastic. It is clear that the engineering properties, the environmental consequences and the economic possibilities of these non-road by-products need to intersect for them to be a useful recycled material. Through our study we identified several road and non-road by-products that were clearly winners and these were: 9 reclaimed asphalt pavement in new asphalt pavements; 9 reclaimed concrete pavement in new concrete pavement; 9 blast furnace slag as supplementary cementing material in Concrete or stabilised base and subbase; 9 steel slag used in base courses and asphalt pavements; and 9 coal fly ash used in a highway embankment. Recycling materials considering the engineering properties and the environmental consequences is good business but this is not the only way to encourage recycling. This is where our second survey comes into play. Here we asked for the policies that were being utilized to increase recycling. It is clear that many countries are using policies (that is incentives and disincentives) to effect an outcome favoring recycling. It is possible for governments to influence the free market by tax enforcement or subsidy strategies that are designed to promote recycling and the use of recycled by-products. All market parties are encouraged to act in a market-conforming mariner by actions such as: providing sufficient information on the long-term performance of by-products, stimulating test or research projects, implementing a waste tax, providing recommendations and requirements for the use of by-products or subsidising recycling and reproduction facilities. Similarly "Restrictive"
110 regulations can be used to reduce the production of waste and to control its disposal. These regulations must often be balanced, however, with additional policies that encourage sorting, recycling, and reuse.
In several countries, government takes the responsibility for increasing recycling and reuse of by-products. It also works to develop a system of laws and regulations that will restrict the construction industry in how they deal with wastes and encourages the use of by-products. In order to monitor results, most countries set recycling goals for the future. Before a new byproduct is accepted for use, it may be necessary for this by-product to undergo research or a period of testing and evaluation that validates its quality and reliability. A combination of government incentives (subsidies, research) and disincentives (taxes, dumping fees) done co-operatively with industry seems to offer a good formula for success. To further the advancement of recycling world-wide, all countries - regardless ot their current good practices in recycling - should continue to explore new recycling opportunities, advance technical knowledge, and design increasingly effective programs. Keeping in mind that Einstein is calling us to move to a higher level of thinking, our group identified an Implementation Plan for Recycling and provided a model partnering agreement that illustrates the importance of co-operation and outlines the philosophies, roles, and broad-based implementation responsibilities of each partner in carrying out recycling.
111
Individual Partner
Implementation Item ,
By-Product Supplier
Environmental Authority
Engineering Authority (Road Administration Agency)
Owner Agency (Road Administration Agency) Contractor
,,
Specifies product names and characteristics. Develops quality plan to ensure product consistency. Produces product to meet specifications and quality plan requirements. Defines cost and sale parameters. Specifies general environmental regulations. Dcfines pollutants and risks of pollution. Defines appropriate contamination limits. Defines methods to study pollution and specifies associated tests. Proposes taxes or specific incentives to facilitate the use of by-products. Determines suitability of by-products which can be used in road construction and rehabilitation. Undertakes documentary research reviews and data base and supplier information. Considers technical/economical feasibility for each application. Proposes combination of by-product use, technical processes and cost considerations to owner agency and co-operating partners. Conducts any necessary research and demonstration sites on selected options with owner agency and co-operating partners. Develops specifications for project applications and proposes standards. Defines performance requirements or technical needs. Selects by-products and techniques according to program needs/economies. Reviews research results and selects demonstration sites in conjunction with engineering and environmental authorities. Adopts technical regulations and standard documents for contracts. Proposes specific techniques on projects in conjunction with suppliers. Carries out construction activities using by-products according to standards and specifications set forth by owner agency, environmental and engineering authorities. Defines techniques as incorporated in technical proposals and requirements for future applications.
112 SIX RECOMMENDATIONS This Recycling for Road Improvements report recognises that the use of by-products in road construction is contingent upon the technical, engineering, value of the material. However, this contingency is viewed not as an excuse so as to not recycle, but rather as a need to disseminate good information and examples as well as a larger call for continued research into the prudent use of by-products. There is a definite governmental role in this sector. It is suggested that measured public sector involvement can accelerate this adoption of much recycling technology. From the investigation and analysis of the current state of recycling, a series of recommendations are identified that could help to keep pace with increasing global stores of byproducts and decreasing space for landfills.
1. Test materials before recycling. The results of the survey on the use of recycled road materials, shows a significant growth in the quantity, diversity and quality of recycling over the past 20 years. The most commonly used road by-products continue to result from recycled asphalt pavements. The scale of using recycled concrete pavement by-products is less significant than that for asphalt concrete but is developing. The effectiveness of recycling clearly requires one to test the road material
before
recycling so as to: 9 9 9 9
Identify the engineering factors that are critical. Determine the highest level of recycling that is possible. Assure that recycling this time will not foreclose recycling options in the future. Avoid recycling materials that have serious health consequences - i.e., pavements, containing coal tar - or accommodating other materials that have health and safety consequences, i.e., lead-based paints, asbestos in demolition waste.
2. Ensure that recycled by-products are used wisely. Recycling is underway in all countries and will continue to increase as research and new technologies expand the opportunities. The OECD countries responding to the Group's survey clearly show that they are, in many ways, "cleaning up their own house first." It is clear that sometimes recycling (i.e., fly ash, slag aggregates) may offer a product that is better than virgin materials. But it is extremely important to review previous, related research and to test and evaluate as necessary to be sure that the by-product does perform acceptably. There are limitations in nearly all areas of by-product technologies and it is only worthwhile to recycle when you stay within the performance range of the material.
113 In using non-road by-products one must fully consider the acceptable boundaries of use. For instance, while steel slag is a good material in asphalt, it must not be used in concrete. Similarly, many of the applications of unbound by-products require that they not be used in areas close to a water table. Other by-products, such as scrap tires, offer some recycling potential. However, many countries are still researching the value of incorporating scrap tires into hot mix asphalt. Although the road offers good opportunities to accept by-products from non-road sources, it should not serve as a "longitudinal landfill." In other words, the road must first and foremost serve its transport function. When compatible with the recycling of non-road by-products, the road can offer good, reasonable opportunities for governments and industry to recycle byproducts from industries outside the road sector. Continued research into recycling non-road by-products, by both the producers and users, holds promise to reduce global waste disposal and also to reduce the resulting stress that this places on the environment.
3. Promote the increased use of proven recycling solutions. To show how recycling of road by-products and non-road by-products really works, "winners" were identified. Asphalt pavement and concrete pavement recycling are technologies that work, have significant environmental advantages and are economically attractive. The research, demonstration projects and equipment developed in response to these technologies shows that the road industry has done much to "clean up its own house first." These technologies are remarkably similar in all countries and their results are consistent. Similarly the recycling of non-road by-products, such as slags and coal fly ash~ottom ash, were explained through examples from several different countries. Interested individuals can now see a range of by-products from "very successful" to "offering good potential" and use this shared information to improve recycling techniques. We must share.
4. Support policies that foster recycling and discourage dumping. As important as the engineering factors are to the use and recycling of road and non-road materials, other non-technical factors are also significant. Governmental laws, policies, and regulations can set restrictions so as to reduce the production of waste, control disposal and limit the use of new materials - or governmental regulations can be promotional and subsidise the use of recycled by-products, fund research, testing, evaluation and demonstration of recycled materials. What becomes clear as one reviews the survey responses is that a balance of regulations and policies is paramount to the success of a recycling program. Within this regulatory environment there is a stronger emphasis on controls in those countries with denser populations and fewer available disposal options. Promotional options seem more effective in those environments where more options to recycling are available. It appears that increased support of research into better use of by-products was needed in all countries. This is especially true in countries that have established landfill restrictions. For
114 example, the European Community has strong landfill restrictions with a year 2002 compliance date. Most countries reported recycling goals, with many being over 50 percent recycling by the turn of the century. In addition to the need to balance restrictive and promotional policies, it is clear from the information received that the success of any recycling program also depends on the involvement of the public and private sectors. The construction industry, the academic community, the owners of the roads, the recycling industries, equipment manufacturers and others can solve many more recycling problems by working together rather than in isolation. Pilot studies and test sections are strongly recommended. 5. Balance engineering, environmental and economic factors.
The central theme of this entire report is captured in the three overlapping circles where the best potential for recycling is defined by that area of overlap that occurs when engineering (technical), environmental and economic factors are properly balanced. If these factors are not in balance, the result can be environmental requirements to use a recycled by-product that has not been proven, roads being built using structurally sound byproducts but creating unacceptable side effects, or a serious underestimate of the economic effects of any of these strategies. The best solutions are those that balance all three factors in developing an informed judgement. Trade-offs will always need to be made. What is important is that these trade-offs be made with an understanding of all three factors and a concern for "the big picture." 6. Increase research and knowledge transfer.
The lack of adequate information on the long-term performance of by-products, standard specifications and testing requirements will continue to slow the recycling movement. The study team encourages: an increase in recycling research and demonstration projects; an increase in sharing recycling technology among countries; an increased use of incentives for recycling; creation of better devices to encourage recycling; and an effort to convince sceptics (through documentation and publication) of effective recycling. This report should help in this regard.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
115
Abdelkrim BOUCHELAGHEM, INERTEC, Nanterre, France Marie-Claire MAGNIE, INERTEC, Nanterre, France Val~rie MAYEUX, ADEME, Angers, France
SUMMARY
With the new developments of the Beaulieu District in Caen (France), the reinforcement of the underground mine workings appears to be necessary : these mines have been worked since the Xlth century and are now beginning to fall in. In place of traditional material (scouring sand for example), they are filled in with a specific mortar, prepared with inertized bottom ashes, developed for this use by INERTEC. Iron has first to be removed from the bottom ashes, coming directly from the MSW incinerator. Bottom ashes are then passed through a sieve, in order to obtain an homogeneous material, and stored on a watertight area. Sieved bottom ashes are mixed with water and reagents, as defined in INERTEC process in order to have a pumpable product, and then pumped to mine workings. Final material features are tested by inner and outer controls, according to the french regulation. Besides, a specific study has been begun by INERTEC and ADEME on the long term behavior of inertized bottom ashes in underground mine workings, in application of the X 30-407 french methodology, which first results are given in this paper.
116 principles for waste management are integrated :
Introduction In 1995, France published its first standard about long term
the 9 ultimate waste (i.e. waste which cannot be reused,
leaching behavior. It is a guideline which describes the
recycled or treated in the current technical and economical
methodology to assess the long term behavior of waste in a
conditions) alone has to be stored,
given scenario.
reduction 9 of waste production and / or waste noxiousness,
Concerned about environmental protection, INERTEC, in
waste 9 valorization, limitation 9 of waste transport in terms of distance as much
collaboration with ADEME, has decided to adopt this new step to the innovating operation of inertized incineration
as duration.
bottom ashes utilization as mine backfill material.
On December 18, 1992, specific regulations about industrial
Before presenting the first results of this operation, we would
waste landfill dumping confirm for industrial waste the great
like to introduce the French regulation context.
principle of the 1992 act and modify noticeably these landfilling conditions.
1. First laws about w a s t e 1.1 How to manage waste disposal in general
1.2 Particular case o f bottom ashes
With its first law about wastes (1975) and a specific law
On May, 1994, a French Ministry Circular specifies criteria
about classified plants (1976), France owns an efficient
regarding incinerator bottom ashes for landfill dumping and
legislative systems based on the principle of producer's
reused in road basement.
responsibility and best available technology to manage their
Three classes of bottom ashes are defined : 9Class V, low leachable fraction : this class of bottom ashes
waste from production to disposal. Rapidly, a complete network of collective disposal plants for
can be reused directly but are subject to a few restriction
municipal and industrial waste is created.
concerning contact with water, 9Class M : intermediate bottom ashes, they are reusable
Judged heavy at first, these new laws contribute to the development of a new industry whose efforts are increasing.
after
maturation
or treatment
(12
months
maximum)
On July 13, 1992, the 1975 act is updated and new
provided after this the criteria for class V are respected.
Solubles (%) Parameter
Class V limit
Class M limit
Solubles % TOC (mg/kg) S04 (mg/kg) Cr 6 (mg/kg) As (mg/kg) Cd (_mg/kg) Pb (mg/kg) Hg (mg/kg)
5 1500 10000 1.5 2 1 10 0.2
10 2000 15000 3 4 2 50 0.4
.
.
.
.
Toe (mg/kg)
1
t[g (mg/kg)
..
/ .
S04 (rag?k-g)
Cr6 (mg/kg)
>---....
Pb (mg/kg)
.d (m~/kg)
Figure 1- Regulation Limits for Domestic Refuse Incinerator bottom Ash
[
(]lass M lirnit:]
117 9Class S : High leachable fraction, their disposal must be in
each mine was worked without controls, frequently to the
class II landfills.
detriment of the most elementary safety rules. Pillar widths
The test actually used for pollution potential is a compliance
may vary from 3m to lm on a side, from one room to the
test. It consists in three successive leachings after grinding
next. Each mine grew differently over time with different
the ash to 4 mm size, using standard X 31-210 test
pillar numbers and sizes and different faulting affecting the
procedure. Leachate pollutant contents are compared with
roof. Pillars that were too few in number or too small in size
the limits specified in the ministry circular (see fig. 1).
cracked and sometimes collapsed, bringing the roof down with them. Periodical
inspections by the Caen mines
2. An avoidable evolution
department indicates that the competent overlying strata of
These new conditions of waste disposal are not defined on a
massive limestone have so far prevented ground level
notion of impact on environment. It is probably because of
subsidence.
the lack of scientific data in this field.
The land was mainly agricultural, i.e. without building loads
This lack of scientific data was often made up for by the
and only intermittently occupied, and raised no safety
development
problems.
of
technologies
improving
always
more
However, plans to develop the Beaulieu area (22 hectares)
disposal waste. However it seems today that this approach of best available
made it necessary to provide support to all the workings to
technology can not constitute the only answer in term of
prevent future damage to buildings,
roads and buried
objectives and level to reach as far as environmental
services, since static loads applied
by buildings and
production is concerned.
static/dynamic loading from roads would put added stress on
In
this
content,
ADEME
(the
French
Agency
for
mine
roofs.
The
support
work
was
awarded
to
Environment) has been committed for over 5 years with
SOLETANCHE, with a range of support designs to suit pillar
several
at the
cracking and ground loads. The workings under the planned
integration of v
O
12 LUll_. 0
o
0
11
I
I
I
None
Medium
$ I
High
Degree of Microcracking Figure 6: Influence of Microcracking on Tortuosity (a), Rcl (b), and PDe,cn (c)
144
16000
a
I
O
I
I
12000 o
O
8000
"I::C 0 I--
4000 0
w
None
Medium
High
Degree of Microcracking c
70
c
60
(1)
50
O .m
b
O
O
40 E
v
ll) t--
(J
30 20 10
None
Medium
0
High
Degree of Microcracking 00
15
9 E D D
C
I
I
I
14 O
~.._~0 w~:
13
~ 8
12
O O I
I
I
None
Medium
High
Degree of Microcracking Figure 7: Influence of Microcracking on Tortuosity (a), Rca(b), and PDe,ca (c)
145
40000
A
a
30000
>E
. R
20000
10000
(J O
I
High
I
"~
Medium
w
Low
Degree of Granularization
60000
00
50000 _b
(D
O3
>E
Ern :3v
O
I
I
I
40000 30000 20000
|
10000 0
I
High
I
Medium
I
Low
Degree of Granularization
Figure 8: Influence of Granularization on Bt,c1 (a) and Bt, ca (b)
146
100 >,,,
80
I
a
I
I
o
60
0
40
0 !--
20 High
Medium
Low
Degree of Granularization to
I
I
I
b
.m
1--"
ID
0
E l--
I
I
I
High
Medium
Low
Degree of Granularization 12
(/) C~
I
.9 E (/) o,..-
I
(~-
as
11
~._~ ~: o
Ir
w~Oo
I
O
10
J
J
J
J
I
I
I
High
Medium
Low
Degree of Granularization Figure 9" Influence of Granularization on Tortuosity (a),
RcI (b),
and PDe, o (c)
147
100
I
I
I
80 o 1E 0 I--
O
60 40 20 High
Medium
Low
Degree of Granularization 300
c O .,...
250
c
_b
150 E
(-.
I
~
100
v
I
O
200
rr~
I
9 o
50
O O
I
I
I
High
Medium
Low
Degree of Granularization u)
c~"
.o E tl::
aj
o
13.0 12.8
I
C
O
I
12.6 12.4 12.2
~8
12.0
o O I
High
~
I
Medium
I
Low
Degree of Granularization Figure l 0" Influence of Granularization on Tortuosity (a), Rca(b ), and PDe,ca (c)
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
149
DESIGN AND CONSTRUCTION OF A ROAD PAVEMENT USING FLY-ASH IN HOT ROLLED ASPHALT
by S.E. Zoorob and J.G. Cabrera C.E.M.U., Department of Civil Engineering, University of Leeds, Leeds LS2 9JT, U.K.
ABSTRACT
Using fly ash to substitute the filler in bituminous mixtures is not only a way of disposing of this waste in a safe manner but it is a way of reducing the energy requirements for the preparation and placement of this composite in road pavements. This paper describes briefly the background work carried out in the laboratories of CEMU to develop the mix design and to evaluate the reductions of energy requirements for the production of a bituminous mixture, i.e., hot rolled asphalt (HRA) which is used in the UK to surface motorways and heavily trafficked roads. Hot rolled asphalt is prepared in site plants heating the aggregates and bitumen at 160 ~ C and is placed and compacted normally at not less than 125~ Thus the energy requirements for the preparation, placement and compaction of HRA is very high. The HRA designed at CEMU is a low energy mix which contains fly ash. It can be prepared at a temperature of i25 ~ C and placed and compacted at a temperature as low as 85 ~ This mix was used for the construction of the overlay of a vew heavily trafficked road. the A689 in the north of England. The construction of the overlay for the pavement is described and the results of the in situ pavement are assessed with results obtained during four years of monitoring the performance of the section paved with the new low energy HRA. The paper also discusses the design and development of a semiautomatic rut measuring devise which was especially designed to monitor the performance of the pavement. The environmental and cost implications of using fly ash in bituminous mixes is quantified using the results obtained during the laboratory design and construction of the trial HRA overlay.
150 INTRODUCTION One of the most important requirements of bituminous mixtures is that their compaction characteristics in the laboratory can be repeated during construction in the field. Poor performance of bituminous mixtures in road pavements is in many cases attributed to poor mixing and inadequate compaction. Mixes that can be mixed, handled and placed without difficulties are said to be workable. Workability is a parameter which indicates these attributes in a bituminous mix. Most bituminous mixes can be made workable if enough high temperature of compaction is maintained during the process, this is obtained by heating mineral aggregates, filler and bitumen to relatively high temperatures, and transporting and laying the mixes in short periods to avoid loss of temperature. Many mixes become unworkable when they reach temperatures of + i20 ~ The main objectives of the study reported in this paper were: To assess the effect of fly ash (FA) on the engineering and performance properties of hot rolled asphalt. To assess the influence of changes on the temperature of mixing and compaction in conventional and FA hot rolled asphalt. To validate any findings using a wide range of mineral aggregates and fillers. To conduct a full scale trial using one of the design mixes. -
-
The project on the design of low energy hot rolled asphalt (LEHRA) using FA was supported by the Energy Efficiency Office of the Department of the Environment U.K., Cleveland County Council U.K., National Power U.K. and Tilcon North Limited U.K. MATERIALS USED IN THE INVESTIGATION AND THEIR PROPERTIES The materials used in the laboratory investigation were representative of those used in road pavement construction in the North of England. Four coarse aggregates, four sands, three limestone powders and four fly ashes were selected for the study. The code used to distinguish them are : Coarse Aggregate (CA) Limestone Filler (L)
Sand (S) Fly Ash Filler (FA)
The symbol is followed by a number from 1 to 4, this distinguishes the type and origin of each component. Details of the origin of these materials have been reported in reference [1 ]. Relevant properties are shown in Tables 1, 2 and 3. The particle size distribution of the limestone and fly ash fillers were similar mainly on the silt size range. There is a marked difference between the two types of filler. Limestone fillers are on average finer than FA fillers and have a higher specific gravity. The shape factor number [2] which is a measure of the sphericity of a particle shows that FA is predominantly spherical in shape while limestone is not. This characteristic of FA allows it to function as a filler in a solidliquid or solid-plastic composite without unduly increasing the viscosity of the composite.
151
Table 1 IVlaterial
S1 $2 $3 $4 CA1 CA2 CA3 CA4
Table 2 Filler FA1 FA2 FA3 FA4 L5 L6 L7 Spec. limits
Table 3
Sand and Aggregate Properties Relative Density on Oven dried basis 2.489 2.383 2.534 2.579 2.861 2.893 2.728 2.753
Relative Density on a saturated and Surface dried basis. 2.494 2.402 2.547 2.595 2.910 2.923 2.756 2.763
Filler Properties Voids of Dry Relative Compacted filler Density 0.211 2.179 2.412 0.275 2.249 0.281 2.384 0.235 2.773 0.255 2.824 0.243 2.725 0.248 -
Apparent Relative Density 2.502 2.430 2.568 2.621 3.009 2.982 2.807 2.780
Bulk Density in Toluene (g/ml) 0.450 0.638 0.344 0.422 0.612 0.625 0.638 0.5 - 0.9
Water Absorption (% of dry mass) 0.210 0.808 0.512 0.611 1.719 1.032 1.030 0.353
i
% retained on 75gm sieve 14.64 12.77 9.16 11.81 3.30 5.72 8.48 < 15.00
Filler Mean Diameters, Specific Gravities and Surface Areas
Filler Type
Drax PFA Blyth PFA Thorpe Marsh PFA West Burton PFA Ballidon Filler Marsden Limestone Scottish Limestone
Mean Diameter (micron) 14 10 11 9 7 7.4 5.4
Specific Gravity g/cc 2.179 2.412 2.249 2.384 2.773 2.824 2.725
Surface Area (m2/g)
Calculated Surface Area (m2/g)
Shape Factor
0.196 0.248 0.242 0.279 0.309 0.287 0.407
0.215 0.213 0.218 0.233 0.217 0.171 0.180
2 3 3 3 3 3
The bitumen used was a straight run nominal 50 pen grade bitumen. The bitumen properties measured are shown in Table 4.
152
Table 4
Properties of Bitumen
Penetration 25 ~ (• Softening Point (R&B) ~ Relative Density Penetration Index
52 53 1.029 -0.20
P R E P A R A T I O N OF HOT R O L L E D ASPHALT MIXES The proportions of coarse aggregate, sand and filler required to produce size distributions within the specifications given in BS 594" Part 1"1985, were 9 Coarse aggregate" Fine Aggregate" Filler 34% 9 56% 9 10% The eight aggregate combinations labelled M1 to M8 were used to produce HRA mixes at various mixing and compacting temperatures. An example of the resultant particle size distribution for mix 2 together with BS 594 specification limits are shown in Figure 1. Specimens were compacted in the laboratory, using the gyratory testing machine (GTM) [3] The main characteristic of the GTM compactor is that it allows the application of an axial static pressure at the same time that the specimen is subjected to a dynamic 'kneeding motion' which resembles the mode of energy applied in the field by construction plant. For each combination of mixing-compaction temperature, four samples were prepared using the GTM for compaction and measurement of the Workability Index. The compaction conditions in the GTM : Vertical pressure 0.7 MPa, Angle of Gyration 1~ No. of revolutions 30. These conditions give an energy of compaction of the same order as the energy of compaction applied by 50 blows per side with the Marshall hammer. The mixing and compaction temperatures used in the project are given in Table 5.
Table 5 Code TI T2 T3 T4
Mixing and Compaction Temperatures Mixing Temp. ~ 140 140 130 130
Compaction Temp. ~ 125 115 115 105
Code T5 T6 T7 T8
Mixing Temp. ~ 120 120 110' 110
Compaction Temp. ~ 105 95 95 85
153 RESULTS F R O M THE L A B O R A T O R Y STUDY
Workability In this study the method used to assess workability as developed by Cabrera [3, 4], consists of monitoring the specimen height and hence volume reduction during compaction. Knowing the specific gravity of the mix, for any specimen, the porosity can be plotted against the number of compactive revolutions. The experimental line should approximate a linear relation of the form: P, = A - B log to (i) A and B are constants. where A - intercept with the y axis. B = slope of the line. i = number of revolutions. The Workability expressed by the "Workability Index" (W.I.) is defined as the inverse of the constant A. i.e. the porosity at zero revolutions multiplied by 100. W.I.=
/ l) ~-
x 100
As expected W.I. values decrease as temperature decreases due to the increase in bitumen viscosity as the softening point is approached. But most importantly, all FA mixes at all mixing and compacting temperatures exhibit higher W.I. values than conventional HRA mixes, an example is shown in Figure 2. This can be attributed to the fact the FA particles have more rounded and less angular texture aiding workability. There is clear evidence that FA mixes will compact better than limestone mixes even at the lowest temperature of compaction used in the laboratory.
Stability. and Flow Values Densities. Marshall Stabilities and Flows were obtained according to BS 598 [5]. In general, Stability values decrease as mixing and compacting temperatures decrease. In all cases, the Stability of the conventional hot rolled asphalt mixes were only slightly higher than their counterpart FA mixes. Nevertheless, the stability of all mixes satisfy the criteria for roads carrying medium traffic (up to 6000 vehicles/lane/day). See Tables 6 and 7 for the stability and flow design parameters.
Criteria for the Stability of laboratory designed asphalt. Table 6 BS 594" Part 1 91985 Marshall Stability of complete Traffic (colmnercial vehicles mix (kN). per lane per day) 2to8 Less than 1500 1500 to 6000 4 to 8 Over 6000 6 to 10 Table 7
Asphalt Institute Design Criteria Light Traffic Medium Traffic Compaction 2 x 35 2 x 50 Stability (kN) 3.33 5.33 Flow,(0.25mm) 8 - 18 8 - 16 Porosity(%) 3 - 5 3 - 5
Heavy Traffic 2x75 8.00 8 - 14 3-5
154 FA mixes exhibit consistently lower flow values than their counterpart conventional mixes at all mixing and compacting temperatures, nonetheless almost all the flow values measured were less than 4 mm.
Porosity and Voids in Mineral Aggregate (VMA) There is no marked change in VMA values as the temperature of mixing and compaction decreases. Also for each aggregate type, both conventional and FA mixes do not exhibit a great change in VMA values. Porosity values for all the mixes at all temperatures were below 6%.
Creep Stiffness The creep test is carried out on duplicate specimens at 40~ The test lasts two hours, and gives results which allow the characterisation of the mixes in terms of their long term deformation behaviour [6]. Analysis of creep test results carried out on the hot rolled asphalt containing FA filler show the normal variation in stiffness of mix values with respect to stiffness of bitumen. The stiffness of the bitumen was obtained from a Van der Poel nomograph. The nomograph gives values of stiffness as a function of the time of loading, the temperature difference between test conditions, the Softening Point temperature, and the Penetration Index. From the Smix - Sbit experimental values, regression lines were obtained. These regression equations are of the form" Log Smix = X Log Sbit + C and were used to obtain the Smi x values at one hour loading time. Figure 3 shows an example for mixes 1 and 2.
Determination of the Optimum Bitumen Content (o.b.c.) The Leeds Design Method [4], recommends that the optimum binder content should be obtained by averaging the binder contents corresponding to the following parameters" Maximum Stability, Maximum Density, Minimum voids in the mineral aggregate, Maximum compacted aggregate density, Maximum Stiffness. The optimum value obtained should lie within 3 - 5% porosity and below 4 mm Flow. The o.b.c's for the mixes prepared at different temperatures of mixing and compaction were then averaged and the results used as the o.b.c, for each mix combination independently of the temperature of mixing and compaction. On average the o.b.c's of all mixes were very close to 7% for all temperatures [ 1].
155
Fatigue Testing Programme In this part of the programme, an Instron 8033 Servo Hydraulic Dynamic Testing machine was used to produce a sinusoidally shaped loading pulse on the beam specimens. The machine was also equipped with an oscilloscope to aid in monitoring the shape of the applied load and the response pattern. The output data. consisting of the magnitude of the applied load, the approximate piston head position and the number of cycles, were updated and displayed on the visual display unit of the Instron. To measure the resultant strains on a beam, a PL-60 wire resistance strain gauge having 60 mm length was glued at two locations of one beam side. The central strain gauge was fixed at a location approximately 25 mm above the bottom of the beam. The strains produced by the repeated loading at the center of the beams were amplified by a Differential Strain Amplifier and then transmitted to a data logger which was connected to a computer. The fatigue test was performed by placing the rectangular beams on a 50 mm thick rubber foundation which had a modulus of elasticity value of 4.3 MPa. The dynamic load was applied to the centre of the beam via a rubber loading block 60 mm wide, see Figure 4. The whole assembly consists of the beam and rubber pad which rests on a very stiff 25 mm thick steel plate which is supported directly on the base of the loading frame.
Method of Analysis During fatigue testing, sudden failure due to detect propagation in the detected as the mode of failure in the test configuration, hence brittle toughness calculations are not applicable. Thermal fatigue or ductile deformation in the region of rupture due to the build up of heat is the mechanism.
brittle mode was not fracture and fracture t'ailure where plastic more realistic failure
By employing a dissipated energy approach the results of different types of dynamic tests, carried out under different sets of conditions and with several types of asphalt mixes, can be described by a single, mix-specific relation :the number of cycles to fatigue is related mainly to the amount of energy dissipated during the test. The occurrence of rest periods, the use of controlled-stress or controlled-strain tests, the effects of frequency and temperature do not significantly influence the dissipated energy relation. During a controlled strain test the stress amplitude and the phase angle change. This means that for the calculation of the total dissipated energy it is necessary to integrate the functions of stress and phase angle over the number of loading cycles concerned. This integration is approximated by a summation of the energy into fixed intervals of "constant" cycles, i.e. cycles in which it can be assumed that the stress and phase angle in that interval are nearly constant. From the energy relation it follows that at a higher fatigue life of the mix more total energy per unit volume can be dissipated [7].
156 By using the original strain versus the number of cycles, the cumulative energy dissipated for each beam was calculated up to the point where the test was terminated. Figure 5 shows the relation between the number of cycles to the cumulative dissipated energy value of 60x 106 J/m 3 for the initial strain at the four main load levels (95 N, 150 N, 260 N and 1400 N). It is clear that FA and ordinary mixes behave in the same manner and there is no distinction between their fatigue properties.
FULL SCALE ROAD T R I A L Following the successful outcome of the laboratory investigation, a full scale road construction trial to assess the performance of Low Energy Hot Rolled Asphalt (LEHRA) under real intense traffic loading was carried out. The selection of the mix for the construction of the road trial was indirectly determined by the geographical position in the U.K. of the road selected by Cleveland County Council. The road selected served high traffic volumes which provided realistic conditions for the evaluation of the performance of the LEHRA. It was also a requirement that the road structure should not have reached a service life requiring reconstruction, but one where a strengthening surface layer should be the most acceptable engineering improvement. A section of the A689, Eastbound slow lane, one mile West of Trunk Road A19, was made available for the construction of the trial section. The most convenient materials were those designated Mix 1 and Mix 2 of the laboratory study. Materials and Location of the Road Trial
The trial pavement was constructed on April 1991. The plant for the preparation of the bituminous mixes was made available by Tilcon North Ltd., and consisted of a Miller Batch Asphalt Plant of 240 ton/h capacity with delivery of 3 ton mix per batch, located in Blaydon, North Yorkshire. Two Hot Rolled Asphalt mixes were produced 9A control mix containing the conventional limestone filler and an experimental mix containing FA filler designed according to the Leeds Design Method. Both mixes conform to BS594, Part 1" 1985. Designation 30/14. Analysis of the mix composition produced in the batching plant showed that the mix was identical to the mix designed in the laboratory and described earlier. Production Sequence
The trial consisted of the production and laying of 70 tonnes of LEHRA and 80 tonnes of conventional HRA in the slow lane, Eastbound carriageway of the A689 Wolviston to Billingham road [8]. The length of the trial road was 320 m and the thickness of the layer was 40mm. The total work for the morning was scheduled to be 150 tonnes. Evaluation of the production process highlighted the following"
157 1. The PFA would not flow properly through the silo screw-feed system, flow was maintained by manually rodding the base of the silo. The use of aeration and vibration which is common in the concrete industry will solve this problem. 2. Attempts to match the slow rate of filler feed with an equivalent sand feed rate and the reduction of temperature, caused the drier burner control to become unstable below about 135oc. The final production temperature of the mixes were as indicated in Table 8.
Table 8 a) b) c) d)
Production Temperatures
Material Control HRA LEHRA Mix LEHRA Mix Control HRA
Target Temp. (oc) 160 130 140 160
Actual Temp. (~ 140 130 137.5 160
Quantity (Tonnes) 30 30 40 50
Mix temperatures at the plant were taken using electronic thermocouple type thermometers in accordance with the recommendations set out in BS 598: Part 109: 1990. Because of the long loading time and breezy conditions, there was a drop of around 8o(2 in the temperature of the material which was first loaded. A hand held thermal anemometer was used for the measurement of ambient temperature and wind velocity at the site. The results indicated a range of speeds varying between 5 and 7 m/s. Wind gusts of up to 10 m/s were also recorded. The temperature was 20oc + 1~
Ener~, Use During The Trial Analysis of gas meter readings on the drum dryer during the plant mixing stage indicated the following: Control Mix 257 Therms for 80 Tonnes - 3.21 Therms / Tonne. (338.6 MJ / Tonne.) LEHRA Mix 200 Therms for 70 Tonnes - 2.86 Therms / Tonne. (301.7 MJ / Tonne.) Data for the asphalt mixing plant to December 1990 show an energy cost of production (at a target efficiency of 85%) of s 1.01 p for asphalts, and 58p for macadams per tonne. Of the 43p difference, 70% is a consequence of the additional moisture content of asphalt sands when compared with crushed rock used in macadams. The other 30% is 'heating' energy; a consequence of the higher mixing temperatures needed for 50 pen. grade bitumens usually used with the asphalts. The substitution of limestone by PFA allowed a reduction in the temperature of the raw materials in the asphalt plant at the heating/mixing stage of production. This reduction in heating energy is equivalent to 30%. Therefore using the difference of heating requirements for HRA and macadams, it can be said that the savings at the heating / mixing stage are 30% of 43 p or 12.9 p / tonne of LEHRA produced. Table 9 shows the details of the trial road construction, including materials and temperatures of mixing and compaction.
Tnl)le 9
Details o f the T r i a l Ro21cl Construction.
A 689 Wolviston to Billingham HRA Trials, laid 11th April 91. Distance (m) 0 - 80 81 100 101 - 145 146 - 200 201 - 240 240 - 320
-
Materials Control HRA (Limestone Filler). HRA with FA Section I. HRA with FA Section 2. HRA with FA Section 3. IIRA with FA Section 4. Control HRA (Limestone Filler).
R.O.S. = Rate o f spread o f chipping (kg/m2). S.M.T.D. = Texture depth measuremerits by Laser Texture Depth Meter (mm). Note; the two rows o f data indicate the near side and o f f side measurements respectively.
159 A S S E S S M E N T OF T H E P E R F O R M A N C E OF T H E T R I A L P A V E M E N T The following control tests were carried out in the field 9 1- Traffic Control. 3- Surface texture measurements.
2- Rut Depth measurements. 4- Dynamic deflection measurement.
Cores were obtained from the different sections of the pavement and measurements of porosity, density and stability were carried out.
laboratory
Traffic Count for A689 Eastbound Cleveland County Council and Leeds University carried out measurements of vehicle flows, these were : Total number of Vehicles in both directions - 8877 Total HGV's in both directions - 2832 % of HGV' s - 31.9 % Total number of Vehicles Eastbound - 4371 % o f H G V ' s - 32 % Total number of Vehicles Westbound - 4506 % of HGV's - 31.7 %
Rut Depth Measurements Initially rut depth measurements for the entire trial length were taken on 5 dates initially using a Straight Edge (3 m in length). The average rut depth values showed that, is was not possible to draw any trends, since there is no evidence of road deterioration. Realizing the need for a more accurate means of monitoring rut depth profiles, which does not entail a worker having to lie on a wet road surface on a cold day taking readings using a small meter rule to an unrealistic accuracy of 1 mm, a rut depth measuring beam was developed at the laboratories of CEMU, University of Leeds. The Leeds Rut Depth Measuring Beam (LRDMB) automates the process of rut depth measurement. A measuring arm, follows the pavement rut depth profile, this is connected to a carriage that runs along the beam length. A hand held microcomputer then digitizes and stores the electrical signals sent off from the measuring arm at every 100 mm traveling distance along the beam. The signals are generated via an angular variable differential transformer that measures the change in angle that the arm makes as it follows the irregularities of the pavement surface. The digital values are then transferred to a template on a spreadsheet and this converts the readings to actual deflections in mm and plots the resulting profile. A longitudinal-section of the beam constructed is shown schematically in Figure 6. Figure 7 shows the longitudinal rut depth profiles along the entire trial obtained using the LRDMB. The experiment confirmed that the road sections made with LEHRA showed rut depth values which were comparable with those of sections containing the conventional HRA.
160
Surface Texture Measurements Texture depth measurements by Laser Texture depth meter, measured in accordance with the specifications for Roads and Bridges clause 929 [9],were also taken. An example of the results is shown in Table 9, and the second, a more recent set is presented in Table 10. D y n a m i c Deflection M e a s u r e m e n t s Deflection measurements taken on the trial road using a Deflectograph [10,11,12] are shown in Figure 8. The trends shown in this Figure lead to the clear conclusion tht L E H R A and H R A deflect to the same extent confirming the L E H R A is as good a material as the conventional HRA. T a b l e 10 Distance (m) 0-10 10-20 20- 30 30 - 40 40- 50 50 - 60 70 - 80 80- 90 90 - I00 100-110 110- 120 120- 130 130- 140 140 - 150 150- 160 160- 170
T e x t u r e Depth M e a s u r e m e n t s Near Side Off Side Distance (m) Wheel Track Wheel Track 0.87 0.94 170- 180 1.15 0.96 180- 190 1.14 0.75 190 - 200 1.12 0.83 200- 210 1.11 1.25 210- 220 1.27 1.2 220- 230 1.08 1.17 230- 240 1.27 1.24 240- 250 1.26 1.15 250- 260 1.13 1.23 260- 270 0.95 1.28 270- 280 1.07 1.23 280- 290 1.27 0.96 290- 300 1.09 0.93 30O- 310 1.11 0.98 310 - 320 1.16 1.06 320- 330 330- 340
Near Side Wheel Track 1.15 1.31 1.22 1.18 1.33 1.25 1.16 1.07 1.08 1.22 1.06 1.11 1.22 1.21 1.11 1.0 0.84
Off Side Wheel Track 1.19 1.04 1.15 1.02 1.02 1.07 1.07 1.15 1.12 1.02 0.87 0.82 0.9 1.01 1.2 1.02 0.95
Laboratory. Tests Cores taken at various times during the monitoring period of 4 years showed that the FA and conventional HRA performed satisfactorily. The details of measurements carried out on these cores have been reported in references [8 and 13].
161 CONCLUSIONS From the results obtained in the Laboratory the following conclusions are offerred 9
9 Fly ash is known to possess predominantly spherical particles when observed under an electron microscope, this enables them to improve the packing properties of HRA. FA particles tend to occupy more bulk volume per unit weight as compared with the more familiar irregular surfaced limestone powder particles. This can be observed from the lower bulk density in the Toluene test. 9 FA filler hot rolled asphalt has far higher workability index than conventional hot rolled asphalt for any of the aggregate combinations used. This finding implies that hot rolled asphalt containing FA can be mixed and compacted at temperatures as low as 110 ~ 85 ~ respectively without impairing its engineering and performance properties. 9 The savings in energy input are considerable and thus FA-HRA can be classed as a low energy material. 9 Replacement of limestone filler with FA does not affect the optimum bitumen content of hot rolled asphalt. 9 The Stability and Flow of the FA mixes satisfy the criteria for medium traffic (up to 6000 CVd), laid down by the Ministry of Transport and the Asphalt Institute of U.S.A. for the range of temperatures tested. 9 Using the Energy approach to eliminate the effect of temperature, it is shown conclusively that mixes containing F A replacement behaved in a similar manner to conventional HRA mixes under repetitive flexural loads. Under a particular stress level, both types of mixes needed the same number of cycles to dissipate a chosen quantity of energy. F r o m the results of full scale trials, the following conclusions are offerred Laying Operations It was clear that using FA as a filler has allowed the temperature of mixing and compaction to be lowered without affecting the optimum binder content. This not only has a significant benefit in terms of heating energy, but also encourages laying under adverse winter conditions. The net energy saving amounts to 12.9 p / tonne of the total energy expenditure in the preparation of the mix. Other savings regarding laying at low temperature and avoidance of waste due to cooling have not been quantified. Laboratory. Results Hot Rolled Asphalt with pfa fillers conformed to BS 594:1985 specifications and were able to perform as well as the materials containing the conventional Limestone fillers. Results from cores tested in the Laboratory were within the specifications for roads carrying up to 6000 CV's per day. Road Monitoring Monitoring of the road has been continuosly carried out : rut depths, surface texture and deflectogragh measurements have been analysed. There was no sign of any form of distress, and the trial sections have shown excellent performance.
162 Hot Rolled Asphalt with PFA as a filler has per~brrned as well as the conventional HRA containing Limestone filler. Cleveland County Council will continue to monitor the performance of the trial section so that information regarding its long term performance can be confirmed.
REFERENCES
l0 ll
12
13
Cabrera J.G. and Zoorob S.E., 'Design of Low Energy Hot Rolled Asphalt', Proc. Performance and Durability of Bituminous Materials, Ed. J G Cabrera and J R Dixon, pp. 289-308, Spon, 1996. Cabrera J.G and Hopkins C.J. 'The influence of PFA shape on the properties of concrete. Ash Tech 84, Second International Conference on Ash Technology, pp. .~v_~-398, London. Cabrera J.G. 'A new methd for the assessment of the workability of bituminous mixtures'. Journal of the Institution of Highways and Transportation No. 11, pp 17 to 23, 1991. Cabrera J.G. 'Hot bituminous mixtures: Design for performance'. 1st National Conference on Bituminous Mixtures and Flexible Pavements. University of Thessaloniki, Greece, pp 1-12, 1992. BS 598: Part 107: 1990, Sampling and examination of bituminous mixtures for roads and other paved areas. Cabrera J.G. and Nikolaides A.F., 'Creep performance of cold dense bituminous mixtures'. Highways and Transportation. Vol. 35, no. 10, pp. 7-15, 1988. Zoorob S.E. and Cabrera J.G., 'Laboratory investigation on the fatigue properties of low energy hot rolled asphalt', 2nd National Conference on Bituminous Mixtures and Flexible Pavements. University of Thessaloniki, Greece 1996, pp. 1 to 13. Rockliff D. 'The use of pulverised fuel ash as a filler in hot rolled asphalt mixturespractical aspects'. Proc. Performance and Durability of Bituminous Materials. Ed. J G Cabrera and J R Dixon, pp. 309-315, Spon, 1996. Department of Transport. Specification for Highway Works, Clause 929, HMSO, London. Kennedy C.K.and Lister N.W., 'Prediction of Pavement Performance and the Design of Overlays', TRRL LR 833. Department of Transport, Roads and Local Transport Directorate, Departmental Advice note: HA/25/83. Deflection Measurement of Flexible Pavements. Analysis, Interpretation and Application of Deflection Measurement. Lister N.W., Kennedy C.K. and Ferne B.W., 'The TRRL Method for Planning and Design of Structural Maintenance', Proceedings Fifth Intemational Conference on the Structural Design of Asphalt Pavements, University of Michigan Vol. 1, Delft 1982, pp 709 to 724. Cabrera J.G. and Zoorob S.E., 'Field performance of low energy hot rolled asphalt'. 2nd National Conference on Bituminous Mixtures and Flexible Pavements. University of Thessaloniki, Greece 1996, pp. 129 to 150.
163 Figure 1 Particle Size Distribution for Mix 2. Coarse & Fine aggregates - Birtley, Filler = Birtley Limestone. % Passing. 100
§
90
~,m,,
........
-I-Spec. Upper Limits
80
-~Mix
:-:-:~:':
:
2
"
"
;
: .
.
.
........
: : : :::! .
.
.
.
.
.
.
.
:
.
.
.
.
.
":Ji
:--:
.
.
!
i :~.!l/..!..
.
[.-
~
.
.
.
.
"
"
70 60 50 40
. . . . . . . . . . . . . . .
~~
2O 10
--
"-!--
,
o
.....
0.01
;
0.1
.
. ,., . . . .
,...,
. ,. . . . . . . . . . . . . .
::::::::::::::::::::::::::::::::::::::::::::::::::: ,
. . . . . . . .
,
;
,
.....
1
;
10
Sieve Size (mm). Figure 2 Workability index values v.s. bitumen content for mixes M1 and M2 at Temperatures T3 and T7
Figure 3 Stiffness of mix (MPa) v.s. bitumen c o n t e n t
Workability Index
;
17 16
Creep Stiffness (MPa). .
.
.
.
.
.
.
.
15
e
- - - ! - - - ! - - - !
14
5
---'~---
.
.
.
....
'
-:.
.
.
.
.
.
.
.
', . . . .
',---
....
,---
13 12 11 ~ , " " ~ ~ l 7
[ - " " ',. . . . '
5.5
6
"n'g te m p~.-$30- I
'," "T7 . mixin ~ temp'. 110, l 9 compaction 95. ; I 6.5
7
7.5
Bitumen Content %
8
8.5
5.5
6
6.5
7
7.5
Bitumen Content (%)
8
8.5
164
Figure
4
Fatigue
Insulated
Loading
Configuration
Wooden
Chamber "-x
]
Load Cell
/ Steel Loading Plate R[~~/Lbber
SCraps
loading pad
[
-4---
Bituminous beam-'--~
Signals from strain gauge -r to Data Logger
Rubber
Freezer Unit
base
circulating cold air
Steel base plate
~
Dynamic
Load
Figure 5 Number of cycles to dissipate 60 million J/m3 per unit volume v.s. initial strain for all beams >, o3
1.000E+07
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
tILl 13 (1) O (1) O_ O3
,
.
.
.
.
.
.
.
.
-
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
,
.
.
.
. ......
,
' :
:.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
:D .
.
.
.
.
.
~ ,
10,000
.
r
.
.
.
.
.
.
.
.
.
.
100
.
.
.
.
.
.
, L
.
.
.
.
.
.
.
. L ,
.
,
200
::
.
: : : : : : : : : : ~ ...... '
.
.
.
.
.
.
.
cyclic load = 1400N
.
.
.
.
.
.
.
.
.
.
..
,
.
.
Max. a p p l i e d
t
.
.
:..._.
' : : ; : : : : : : - : : : : .
.
.
~ ......
~
,
E
z
.
,--N-
cyclic load = 150N
: : : : : : " : : .
.
~ .............
Max. a p p l i e d
.........
.
.
.
: - - - - :
.T-- . . . . . .
o >,
.
.
,
:
:
.
.
.
,
=260N
(D .D
.
'
:
.............
100,000
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
load
: . : : : . . : . . .
00 (D
O
.
.
~ ......
O
o
.
r
cyclic =95N
/
......
.
-,
oe
4,
1,000,000
.
r
Max. a p p l l e a H
300
.
.
.
.
.
.
9
.
.
.
.
.
.
L
.
.
.
.
.
.
.
~ .
.
.
.
.
,
400
Initial Strain (us)
500
600
165 Figure 6
Schematic Diagramme of the Leeds University Rut Measuring Device. Direction of movement 9 ~
lS,gna,s recorded every 10cm
I i
il
[
__ / !
I
, j _
--
Psion microcomputer
~
_ Ii --'--~- 1
~
Lightweight Aluminium hollow beam 3.5m
/ 'enmh
5 ~ ~ ~i~-,,--.,--~--~-~L. ~-. --_~ 9
9
9
_
Wheel follows ______ road profile r
Rut depth (mm).
/ / ''7"/
Figure 7
~ Chang:j%;npglsi%n re(:
L o n g i t u d i n a l Rut D e p t h Profiles
] -)~2.5 m from curb edge. + 3 . 0 m from curb edge.
8
'~t 6
4
2
k
-10~.12 ~C~176 -14
L
0
,
20
,
40
HRA~ 60
80
HRA + PFA 100
I
120
,
140
160
180
~-~C~176 ,
200
,
220
240
260
HRA.r ,
280
,
300
320
o
o
Distance along road (m).
Figure 8 Offside Wheel Path Deflections A689, Eastbound Trial Area. Output from Deflec3 Programme at 2 4 / 0 5 / 9 1 . 6O
55 ~
A
~.~_~, ~
C ~ " r '~
D
~.~
E
~4
F
5O -
E o 45 -
~4o
~ 3o
o
o
o
,
o
o
o
0
,
o
95%.
Because of the effect of sensitivity of the k-value for slight differences in compressive strength, it is important to determine every measuring point accurately (three times).
284 K-VALUES OF DUTCH FLY ASHES KEMA performs the quality control of Dutch fly ashes by order of the Dutch Fly ash Corporation since 1988. Since 1989 also measurements of the k-value of fly ashes in combination with CEM I 32.5R are carried out at water/cement + fly ash ratio 0.60. In figure 4, k-values of Dutch fly ashes from several power plants after 28 days curing are given. As can be seen in figure 4 the efficiency factor k of Dutch fly ashes has increased since 1995 with about 35% to a value of 0.75. Also the standard deviation decreased from 0.20 to 0.11. This is the result of quality control of the Dutch fly ashes, the use of coal mixtures (blends) and better firing techniques.
7
CONCLUSIONS
The k-value of Dutch fly ashes, measured in combination with CEM I cement, has increased over the years. At this moment the k-value has a mean value of 0.75 with a standard deviation of 0.11 at a water/cement + fly ash ratio of 0.60. This is the result of quality control, the use of blend coal mixtures and better firing techniques. A review of research work performed by German and Dutch researchers showed that a uniform determination is needed. KEMA developed a determination method based on previous research being: -
the k-value applies to a given fly ash type in combination with a given cement type
-
the k-value is dependent on the water/cement ratio
-
the k-value should be measured at a constant binder content for fly ash and reference concrete
-
the k-value should be measured at a curing time of 28 days and a humidity of > 95%.
REFERENCES
IBAC, 1987. Untersuchungen an Mbrteln und Betonen mit Steinkohlenflugasche fQr eine erweitere Anrechenbarkeit der Steinkohlenflugasche. INTRON, 1989. De k-waarde van vliegas.
285 KEMA, 1988. K-waarde van vliegas in beton. KEMA, 1992a. Spreiding van de k-waarde van vliegas in beton; invloed van partij cement en monstergrootte. KEMA, 1992b. Schatting van een algemene k-waarde van vliegas in beton. KEMA, 1992c. Effect van de analysemethode op de k-waarde van vliegas in beton. KEMA, 1992d. K-waarde van gezeefde vliegassen. KEMA, 1994. Model voor het effect van de granulometrie van vliegas op de eigenschappen van mortel en beton. KEMA, 1996. Vergelijking van de berekeningsmethoden van de k-waarde van vliegas. NEVILLE. Properties of concrete.
286
Appendix A page 1
te" L-
.> t,..
E O O
co
co water/cement ratio
Figure 1
Principle of the k-value
(W/C+F) constant +
f/c=0.25
A
f/c=O
45 E E zr 40
-
C
"o
o0
D
t-
e
A
.>
.=3o Q.
E
8
B
25 240
270
I
I
I
3OO
33O
36O
binder content (c+f) kg/m3 Figure 2
Influence binder content concrete
3,9O
287 Appendix B page 1 K-VALUE DETERMINATION Table 1
Compressive strength reference at different water/binder ratios
water/binder ratios
compressive strength N/mm 2
WBF 0.40
59.8 - 59.7 - 61.2
WBF 0.50
46.1 - 43.8 - 45.6
WBF 0.55
3 7 . 2 - 36.3
WBF 0.60
3 2 . 9 - 3 1 . 7 - 32.4
Table 2
Compressive strength fly ash concrete at different water/binder ratios WBR 0.45
WBR 0.48 - 0.49
WBR 0.65
va-1
45.1
40.4
21.7
va-2
44.1
41.4
18.8
va-3
44.7
40.1
19.3
va-4
44.2
41.3
18.7
fly ash sample
288
28 days curing [!i!i!!i!!!!i!i!i!!i!i~i!ilil plant 1
plant 2
mean value
1.00 f 0.90 0.80 0.70 r
0.60
i
,> 0.50 0.40 0.30 0.20 0.10 0.00
1989
1990
1992
1993
1995
year Figure 4
K-values Dutch fly ashes
+
reference
+
fly-ash
concrete
70 tM 60 E E z 50 -
/
c-
E:
|
40
-
(/)
> 30 -
,m (/) (/) (I)
~. 20
-
I
F
E
o o lOo o.oo
I
0.50
,/" /
I
1.00
I I I I
II
1.50 WCR
Figure 3
K-value of fly ash 28 days curing
I
I
2.00
2.50
3.00
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
289
UPGRADING AND QUALITY IMPROVEMENT OF PFA
H.A.W. Cornelissen KEMA P.O. Box 9035 6800 ET Arnhem, the Netherlands
ABSTRACT An example of upgrading and quality improvement is micronization of fly ash (PFA). The usual fly ash particle size range is up to 200 micrometer. For the improvement of concrete properties, however, fine fly ash particles in the range up to 10 micrometer are preferable. Micronized fly ash is produced with a mean diameter of less than 5 micron. This very fine powder is an excellent high performance type II filler for concrete. 1
INTRODUCTION
In the Netherlands about 30% of the production of electricity is realised by coal-firing, which results in a collection of about 900.000 tons/year fly ash in the electrostatic precipitators. Nowadays, fly ash is fully accepted as a raw material. It is used for the production of cement (62%), light weight aggregate (20%), concrete, asphalt and other applications (18%). With respect to by-products management the approach is more and more focused on the reduction of costs and further economical optimization. The value of a product like fly ash is mainly determined by the value of the products that will be partly substituted, or by new attractive possibilities of the end products. If for instance cheap regular sand is replaced by fly ash, its value is relatively low. If fly ash can be used as cement, the value is higher. If, moreover, the resulting concrete properties are improved, its value will raise further. In general, quality control measures and upgrading will turn by-products into valuable resources as schematically indicated in figure 1. To improve the quality of fly ash, methods for benification are being introduced. The methods are focussed on various fly ash parameters like grain size and carboncontent (1).
290
A main reason for upgrading fly ash is to enhance its added value, which makes it more profitable. For the use in concrete, the particle size of the fly ash is very important. Especially fine particles are needed because they improve the packing of the combination of gravel, sand and cement particles. This results in a denser concrete, which is consequently stronger and more durable.
value valuable
resource resource
__ upgrading
w QC,
(upgrading)
by-product 0~__ waste
Figure 1
/
~.~ ///
/
/
utilization
_ ~t~:~c~li~on
storage
--D
niches
D
time
.......
Trends in by-products marketing
HIGH PERFORMANCE CONCRETE The use of high performance concretes (high strength and high durability) is becoming more and more common. At the same time there is a tendency towards concrete mixtures which need only minor compaction energy for optimal filling of the formwork. These concretes have a high slump and must therefore be very stable in order to assure homogeneity (see figure 2). For the proportioning of these concrete mixtures, special additives and ultra fine fillers like silica fume (particle size 0.1-1.0 microns), are needed. However, the availability of these ultra fine fillers is very limited and as a consequence the price is high. Because of the developments indicated, it is expected that the need for ultra fine fillers will strongly increase.
291 relative compressive strength 1.1
"
mortar f/c = 0.33
1.00.9 0.8
""--------1-- 41---- 9 0 d
0.7
_
A
"
-
28 d
~
0.6
2d 0.5 10
20 D-50 (10 -3 mm)
30
40
Figure 2
Concrete strength versus fly ash fineness (D-50) (6)
3
FLY ASH FINENESS
It is well known that fly ash fineness is a major parameter for its effect in concrete. The fine particles in the grain size distribution have a relatively high contribution to the strength development of concrete. This was shown in various research projects (2, 3, 4). Also by KEMA this effect was demonstrated (5). A typical result is shown in figure 3. Fly ashes were classified and mixed to given D-50 values (being the mean particle size of the size distribution). It can be seen that both mortal and concrete strengths increase, if finer fly ashes are added. The effect is stronger for concrete and is more pronounced for the finer particle size range. The usual fly ash particle size range is up to 200 micrometer. For the improvement of concrete properties, however, fine fly ash particles are preferable in the range up to 10 micrometer. These fine fraction can be separated from the bulk amount fly ash, or the coarser fly ash can be processed to finer particle sizes. Separation is cheaper but less effective because the amount of fine particles in the original fly ash is very limited.
292 strength (N/mm~) I
veryilstrong
very~durable I I I I I
60
normal 0
Figure 3
very liquid and stable
workability
Trends in concrete properties CLASSIFICATION OF FLY ASH
Two low-NO x fly ashes with lower and higher carbon content (indicated as B and M) produced by Dutch power stations were selected for the tests. The ashes were processed in an air classifier and, if necessary, mixed to obtain certain gradings. The particle size distribution was measured by Malvern 2600 C analyzer and the gradings were qualified in terms of characteristics of the size distribution, i.e. D10, D50 and D90, but also by the grading modulus G (4). G = (6/((1/dl) -
(lld2))lln(d21dl)
In the formula dl and d2 represent the diameters of the smallest and the largest size particles of a group between two successive sieves. Between these sieves the size distribution is assumed to vary linearly to a log scale. Table 1 gives the corresponing values. Concrete cubes (150x150x150 mm) were cast and stored in the fog room (20 ~ and RH > 95%). The cement content of the reference mixes was 320 kg/m 3, which was substituted with 20% (by weight) fly ash in the case of the fly ash mixtures.
293 For all mixtures normal hardening Portland cement was used. The maximum aggregate size was 31.5 mm, while the slump of the fresh concretes was adjusted at 70 mm (plus or minus 10 mm). The amount of water needed for a given workability (the slump of the concrete cone = 70 mm) was measured for all concrete mixtures. Then cube compressive strength was determined at ages of 7 and 28 days. The results are presented in table 2. The results prove that fly ash fineness is a major factor for concrete properties. MICRONIZATION OF FLY ASH 5.1
Processing
It was decided to process fly ash to particles in the one micrometer range, because this will result in an important market for fly ash as a high valuable resource. Besides by air classification fly ash fineness was increased by grinding to particle sizes down to between about 5 and 10 micron (6). The ultra-fine range was also reached in Japan, by vaporization at about 2400 ~
and condensation of fly ash (7). The present
project, investigates whether fly ash can be micronized economically to less than 5 pm and to see if the performance of this product in mortar and concrete is satisfactory.
5.2
Materials
The effects of the various types of fillers were tested with normal hardening Portland cement (PC-A). A typical fly ash was used (57.0% SiO2, 26.4% AI203, 4.4% Fe203, 4.2% C and 1.8% CaO). The fine fraction was collected in the bag-filter of an air classifier, whereas the ultra-fine fraction was obtained by grinding. It was found that the mean particle size being 21.6 l~m for the input fly ash was reduced to 9.9 l~m for the air classified fly ash and to 1.6 l~m for the ground one. A picture of micronized fly ash is given in figure 4. Particle size distribution curves are given in figure 5.
294
Figure 4
SEM-picture of micronized fly ash
100
undersize (%)
/
/
4
5O
J
/
/
/
/ /'
r
/
/
I
3 /
/
/
// //
/
/ / ,/
fly-ash 1 !"] Input
/
2 r-I classified 3 r-] ground 4 I-'1 silica fume
o. _ _ _ _ ~ . . . ~ 0.1
Figure 5
.~
1
10
100
Jill
1000
Particle size distributions of the classified and ground fly ashes in comparison with the input fly ash (percentage passing versus sieve opening in micrometers)
295 The silica fume involved, had an SiO2 content of 92% (m/m) and a mean particle size of 0.12 ~m. In the tests, a 50% to 50% (m/m) combination of silica fume and micronized fly ash was also applied. 5.3
Laboratory scale tests
In the concrete compositions, 360 kg/m 3 PC-A cement was used; the maximum grain size of the river gravel was 31.5 mm. Because a major objective was to realise highly fluid mixtures, the fresh concrete slumps were 230 plus-minus 40 mm. This resulted in a water to cement ratio of 0.32 for the mixtures containing fillers and 0.35 for the reference mixture (no filler added). An overview is given in table 3. If needed, chemicals called super plasticizers were added to improve the workability of the fresh concrete. The amounts of filler added were 5%, 10% and 15% (m/m cement). After 3, 7, 28 and 91 days curing, the compressive strength was determined (see table 4). The results indicate that the strength values are significantly higher for the mixtures with higher filler content, especially in the case of filler types micronized fly ash and silica fume. Note: In table 3 and 4 the reference mixtures are indicated as CREF 1, 2 and 3. The various types of fillers are AC (= air classified PFA), AG (= ground PFA), SF (= silica fume) and SA (= 50% AG plus 50% SF). The numbers 5, 10 and 15 indicate the amount of addition of filler (m/m cement). 5.4
Full-scale test
Based on the results of the laboratory tests two mixes were designed for the full-scale tests, being precast L-shaped elements (100 x 150 x 300 cm3). Special attention was given to the workability in order to realise a mixture which can be placed with minor compaction energy. So the PC-A cement content was raised to 410 kg/m 3 and about 2% (m/m cement) melamine-sulphonate super plasticizer was added. The maximum grain size in the mixes was 16 mm. In mixture 1 the water to cement ratio was 0.36 and in mixture 2, 0.38. The addition of micronized fly ash was 12% (m/m cement) in both cases (see figure 6).
296 The moulds were filled in one batch of 1.2 m 3 fresh concrete. After that compaction was needed for 2 minutes. One day later the elements were demoulded. Mixture 1 showed airencapsulations at the surface, while the element made from mixture 2 showed an excellent surface texture. At an age of 28 days, concrete compressive strength was determined from drilled cores. It was found that after one day hardening the cube strength was already about 55 N/mm 2. The 91 days strength values proved to be about 95 N/mm 2 for both mixes. .:
...
.....:.
9
..
.
.
.:... .
.
.
.
.
.
9
ill
Figure 6
High performance concrete element with micronized fly ash
6
CONCLUSIONS
It is a challenge to use the potentials of the by-products of coal combustion. Because of environmental, economical and technical reasons the present worldwide utilization rate of 35% will strongly be stimulated. In many countries like the Netherlands these by-products are fully accepted, which results in a 100% utilization, mainly in the building materials industry. By adequate measures like quality control and up-grading, fly ashes prove to be excellent raw materials for the building industry. Fly ash fineness proves to be an important parameter for the quality of the products.
297
By appropriate grinding it is possible to micronize fly ash to sizes under 5 l~m. By micronizing all fly ash can be processed, while in the case of air-classification the output of fine material is very limited (< 10%). The effect of addition of these ground fly ashes on properties of concrete was determined and compared to the effects of air classified fly ash, silica fume and combinations of these two types of fillers. It was found that in concrete, the fluidity was positively effected by these types of fillers. So, the irregular shape has no significant effect. Also high strength values were reached. Mixes with micronized fly ash behaved well during full-scale tests. It can be concluded that micronized fly ashes are excellent high performance fillers for concretes.
REFERENCES "Innovation for a sustainable future". Proceedings of the 12th International symposium on Management & Use of Coal Combustion Byproducts (CCB's). ACAA, January 1997. G. Wooley. "Effects of fineness and loss on ignition on concrete performance". Report of the Association of Quality PFA suppliers, UK, 1989. P. Schiessl and R. H~rdtl. "The change of mortar properties as result of fly ash processing." IBAC Mitteilungen, pp. 247-294, 1989. B.P. Hughes and AI-Ani. "PFA fineness and its use in concrete". Magazine of concrete research, no. 147, pp. 99-106, 1989. H.A.W. Cornelissen, C.H. Gast. "Upgrading of fly ash for Utilization in Concrete". Fourth Canmet/ACI Conference on Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, Istanbul, 1993. R. H~irdtl, 1991. Effectiveness of Fly Ash Processing Methods in Improving Concrete Quality. In: Waste Materials in Construction, ISBN 0-444-89089-0, pp. 399-406. Y. Matsufuji, et al., 1993. Study on Properties of Concrete with Ultra fine Particles Produced from Fly Ash. Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, Proceedings International Conference, ACI SP 132, vol. 1, Istanbul pp. 351-365.
298 Table 1
sample code
Size distribution characteristics and LOI
LOI*
D10
D50
D80
G
(~.m)
(~m)
(~.m)
(l/mm)
(%)
B0
7.4
39.2
133.5
225
4.56
B1
4.5
23.0
75.3
456
4.50
B2
6.6
17.4
46.1
462
4.14
B3
3.2
9.5
32.2
850
5.70
M0
6.5
24.2
96.8
299
8.31
M1
4.4
17.7
53.5
563
8.04
M2
4.2
15.7
45.3
600
7.46
M3
2.9
8.7
36.3
919
7.94
loss on ignition at 815 ~ for 10 minutes Table 2
sample code
Concrete test results
water content
(dm31m 3)
compressive strength (N/mm 2) 7 days
28 days
reference
163
30.3
38.3
B0
163
23.0
31.4
B1
160
25.2
33.3
B2
161
23.9
32.9
B3
156
28.2
38.7
M0
162
24.5
33.3
M1
160
27.0
36.8
M2
158
27.3
36.0
M3
157
27.9
38.7
299 Concrete compositions and workability data
Table 3
Sample
fly ash
SF
code"
(kg/m 3)
(kg/m 3)
w/c
SP
Slump
Spread
Spread
(%)**
(mm)
(static)
(jolting)
(mm)
(mm)
CREF1
0
0.32
2.75
170
320
420
CREF2
0
0.35
2.5
220
470
540
CAC05
18
0
0.32
2.5
190
330
420
CAC10
36
0
0.32
2.5
230
430
520
CAC15
54
0
0.32
2.5
250
500
570
CAG05
18
0
0.32
2.5
230
420
520
CAG10
36
0
0.32
2.5
240
540
600
CAG15
54
0
0.32
2.5
270
610
>700
CSF05
18
0.32
2.5
220
400
500
CSF10
36
0.32
2.5
220
350
470
CSF15
54
0.32
2.5
190
320
440
C = concrete; AC = type of filler; 0.5 = % filler (m/m cement) superplasticizer as weight percentage of cement plus 0.2 filler
300 Table 4
Properties of hardened concrete containing the indicated types and amounts of fillers
Sample code
filler type
filler
(%)
Compressive strength (MPa) 3 days
7 days
28 days
91 days
CREF1
no
56.0
68.7
81.8
90.6
CREF2
no
38.8
49.8
64.2
69.3
CAC05
AC
5
56.3
68.2
83.6
92.3
CAC10
AC
10
52.3
64.9
81.6
93.7
CAC15
AC
15
49.7
62.3
79.8
92.4
CAG05
AG
5
53.0
65.3
80.0
85.9
CAG10
AG
10
52.6
66.2
86.2
96.0
CAG15
AG
15
55.0
69.6
95.8
105.9
CSF05
SF
5
55.1
69.4
94.0
95.3
CSF10
SF
10
58.3
70.6
96.6
103.9
CSF15
SF
15
57.9
74.3
99.3
104.8
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
THE EFFECT OF THE DUTCH BUILDING MATERIALS DECREE ON THE BY-PRODUCTS FROM COAL-FIRED POWER STATIONS
M.P. van der Poel N.V. Sep- Dutch Electricity Generating Board Arnhem, the Netherlands
Summary The Dutch government published the Building Materials Decree in November 199~ This new legislation sets the boundaries for imissions of organic and inorganic components from building products into the soil. The immision of building material,, according to the Building Materials Decree is calculated from the results of regulatq leaching test on these building materials. This paper will highlight the effect of the Building Materials Decree on the by-products from coal-fired power stations.
1.
The Building Materials Decree
The Building Materials Decree, referred to as the Building Materials (Soil and surf~ Waters Protection) Decree is published in November 1995 and applies to the use q building materials in a work on or in the soil or in surface water. Building materials in the meaning of this decree are characterised as granular (stonelike) building materials used outside. If the total contents of silicon, calcium c aluminium (with the exeption of metallic aluminium) in a building material together amount to more than 10% m/m of the total building material, it is a building materi~ in the meaning of this decree. The aim of the decree is to set the environmental conditions from the point of view soil and surface waters protection for the use of primary and secondary materials, or in the terrestrial soil, in the surface water and on or in the soil beneath surface water. Although the decree primarily lays down rules with a view to protecting the quality soil and surface waters, it intends also to contribute towards achieving other
302
environmental policy objectives, such as reuse. Because of the continuity of reuse for some building materials special categories are incorporated. For the unmoulded use of fly ash from power plants no special category is being incorporated because the continuity of the present applications is not in issue.
2.1
By-products from coal-fired power stations
Three energy sources are mainly used for the large-scale generating of electricity in the Netherlands. In 1995 these are coal (45%), gas (47%) and nuclear fuels (8%). Most of the carbon from coal is converted to heat during combustion between 1300 and 1700 ~
depending on the type of burner. The inorganic fraction in the coal is
left as ash, produced by the fusion of clay minerals. Water loss combined with rapid cooling of the hot flue gas in the power station exhaust system prevents the elements forming crystal structures, and the minerals are turned into a glassstructure. The process is similar to that which takes place during a volcanic eruption. Fly ash therefore is about 70% glass. Fly ash and bottom ash from coal-fired power stations and products derived from these, are stonelike secondary building materials in the meaning of the Building Materials Decree. For the power industry fly ash and bottom ash from pulverised-coal power stations are the most relevant by-products in relation to this decree.
2.2
Fly ash
The hot fluegas is fed through electrostatic filters which remove remove 99% of the fly ash. In 1995 the Dutch power stations produced 878,000 tons of pulverised coal fly ash. The produced fly ash can all be utilised. The greatest part goes to the cement industry and concrete industry to be used as: -
partial substitute for cement
-
filler used in the manufacture of Portland fly ash cement raw material for clinker production filler and binding agent in concrete
-
substitute for cement in cellular concrete.
303
Fly ash is also utilised as: -
filler in asphalt mixes
-
raw material for the production of artificial gravel.
All the applications of fly ash in the Netherlands are so called moulded applications which means that leaching is determined by diffusion. 2.3 B o t t o m a s h
Bottom ash is produced by the melting and sintering particles of ash in the boiler. Because of the gravity, bottom ash falls down into collecting tanks with water underneath the boiler. Bottom ash is also a product that can be utilised very well. Bottom ash is mainly utilised in the application of: -
road construction
-
concrete block manufacture.
3.
Immission standards
The starting-point for determining a maximum permissible load on the soil is that the multifunctionality of the soil must be guaranteed. As it is not realistic to employ building materials without some burden on the soil, a marginal soil-load is accepted on the assumption that this does justice to the starting-point of maintaining multifunctionality. Marginal soil-load has numerically interpreted as: a load on the soil resulting from leaching from the building material which mathematically results in an increase in the solid phase of the soil of no more than 1% of the contents of pollutants in relation to the long-term targets for soil in 100 years, averaged over one metre of standard soil deemed to be homogenous.
It is assumed that in general this marginal soil-load will also protect the groundwater in a sufficient way. On the basis of the marginal soil-load concept and the average background ieveU in the Netherlands, the maximum permissible immission standards for inorganic
304
substances were determined. When determining these immission standards, it was necessary for a number of substances to rise the immission level to allow the continued use of building materials. That means that in general the immission standards are low, especially for mobile elements. For this mason it is not allowed to use unmoulded fly ash, for instance as a layer in road constructions, any longer. This particularly because of the initial leaching of Molybdeen. In appendix A and B leaching characteristics of fly ash and bottom ash are given. 4.
Classification in categories
Building materials are divided into category 1 and category 2 building materials. - category 1: the building materials referred to in this decree as category 1 may be used without taking isolation measures. However, control measures should be taken to prevent the building materials mingling with the soil and to allow removal. - category 2: building materials with a greater immission into the soil than the
marginal load of 1% may only be used as category 2 building material if the immision can be reduced to below the maximum immision standards belonging to this category. Permanently controlled isolation measures are necessary for materials belonging to category 2. Isolation measures are understood to be measures which virtually rule out any contact of that building material with rainwater or groundwater when that material is used. Regulations for these isolation measures and the control and monitoring measures appurtenant to these measures are stated in the decree and elaborated in a ministerial decision. 5.
Determining of standards
The composition standards for inorganic substances and the immission standards for inorganic substances in category 1 and in category 2 building materials must be determined by one of the laboratories designated by the Environment Minister and the Minister for Transport, Public Works an Water Management. With the aid of difficulty functions and correction factors immission into the soil as a result of emission from a building material will be calculated - after leachability from that building material has been determined in the laboratory in accordance with the
305
draft Dutch standard NEN 7340. Defining the functions and correction factors has been a very difficult process. This because the long-term behaviour over a period of 100 years by extrapolation from the results of laboratory tests is a rather theoretical approach. In this framework it can be told that studies by the Dutch Electricity Generating Companies have stated that the behaviour of fly ash in the laboratory is different from the behaviour of fly ash under field conditions. 6.
Directions for use
One of the departure points of the decree is that it is the task of the party using building materials in a work to ensure that the building materials in question do not become mixed with the soil or aquatic sediment during the life of the work, and that they are indeed removed when the work in which the building materials have been used is removed. These obligations apply in principle to all building materials - with the exeption of clean earth - whose use is permitted under the decree, irrespective of where they are used. In addition to the duty of removal a number of other directions for use formulated as general rules have been incorporated in the decree. In figure 1 the directions for use for the various categories of building materials when used on or in the soil are given.
306 Figure 1 Soil Protection Act (Landfill) Decree
Immission1~
Special category Slag's from waste incineration plants
Category 2 building materials
U2 >
-isolation measures: * 0,5 m above mean highest groundwater level * isolation (capped from above) - control measures in connection with duty of removal: * minimum quantity 10,000 tonnes (1,000 tonnes for foundation layers in road construction) - other control and monitoring measures (management and maintenance)
Category 1 building materials
Special
Building
- control measures in connection with duty of removal
category
Materials
(generally normal management and maintenance)
tar-holding
Ul >
Decree not
asphalt
enforced
aggregate s org
G
Composition G
:Composition value for standard soil
Sorg
:Limit-value for organic components
Ul
:Leaching limit-value for category 1 building materials
U2
:Leaching limit-value for category 2 building materials
7.
Approvals and other forms of proof
On 1 J a n u a r y 1988 the decree will c o m e into full force for all building materials. The user of a building material is than required to have available information on the
307
composition of that building material and the expected immisions as a result of that material being used. For nearly every building material whose application requires a report in advance the information on composition and immission must accompany that report. These materials are category 1 earth, all category 2 building materials and building materials containing tar-holding asphalt aggregate. For the remaining category 1 building materials the competent authority may request the mentioned information from the user, up to five years after the building material has been used. In effect, this duty means that the user should demonstrate or be able to demonstrate that he is employing a category 1 or a category 2 building material. The user can provide the proof concerning the quality of the building material by handing over an approval recognised by the Minister of Housing, Spatial Planning and the Environment and the Minister of Transport, Public Works and Water Management on the basis of certification of the product. As a recognised approval the product certificate and the attestation can be distinguished. The product certificate means a certificate stating a product conforms to certain product specifications. However this certificate does not mean that the building material will satisfy the requirements of the Building Materials Decree in every instance and every application, as this also requires a test of the specific application to the immission requirements. An attestation, in relation to the Building Materials Decree, indicates that a building material, which has a certain composition and leaching properties as described in the product certificate, if employed as stated in the attestation, satisfies the composition and immission requirements of the decree. The attestation will therefore in effect make a statement about a specific building material in a certain use in relation to the category in which that building material and use fall. The user may also attempt by other means - such as a manufacturer's declaration that the building material satisfies the composition and immission requirements of the decree. To do so, the decree imposes on the user the duty to engage an accredited
308
of equivalent foreign laboratory to access composition and immission. These assessment must be carried out in accordance with the rules set in or by virtue of this decree. A building Materials Decree Approvals Manual, in which guidelines and test criteria are laid down, is currently in preparation.
8.
Consequences for fly ash and bottom ash
With the publication of the Building Materials Decree for the power industry development toward achieving environmental certification of construction materials in relation to fly ash and bottom ash has reached an important phase. Because of the Building materials Decree and the leachability of fly ash, this building product has to be used in moulded applications. Leaching of moulded applications is determined by diffusion. For bottom ash selenium (Se) is a critical element. In thin layers however the application of not certificated bottom ash as a category 1 building material is almost 100%. Certificated bottom ash can be used in its entirety as a category 1 building material. In 1996 the first bottom ash under certificate (attestation) is supplied.
References [1]
Dutch decree of 23 November 1995, containing regulations on the use of
building materials {Building Materials (Soil and Surface Waters Protection) Decree on or in the soil or in surface waters}. [2]
Milieugienische kwaliteit van primaire en secondaire bouwmaterialen in relatie
tot hergebruik en bodem- en oppervlaktewaterenbescherming, RIVM-rapp.no. 771402006, RIZA-rapp. no. 93.042 [dec.1993]. [3]
Annual report of 1995 from NV GKENliegasunie.
APPENDIX A
man and ann'ronmsnt
EC-vliegas
Building material:
NV8052 wkl leachlng characteristics
~dentificat~on number: 17 D r
m
composition
US=10 columntest in rngkg
adjusted value granular materials .bmd
A.
U1
U2
S1
088
700
37500
N
mun
n
o m
.d(n I ) m n m m m u m m n,Ul o
s
~ o m
osa7 3200
h
550
5
21€6
1278
Om2
cd
o m
007
1000
17
o m
o m
o m
0011
C4
042
2W
25000
2
0018
0011
0010
0025
Ct
130
I200
129300
17
1728
1393
0160
4673
CU
072
350
37500
7
0080
0101
0007
0-5
He
o m
008
500
I
o
o m
o m
MO
028
091
12500
16
5810
5143
0350
15280
NI
110
3 m
~ ~ 0 0 0 10
0064
0089
0010
0878
58m75LQC.3
NA: No illormalion.rsllabb. ERR: ~land.rddsvt.lon
2.m
m
n>U2 bp(-1 I
9
kpbdfn 1))
0948
061s
2881
0803
0056
0483
wl*yn d d h
D D I6
I1
0450
0624
1357
0608
'
D
aqua regia in mgkg
APPENDIX B
m n and nrimnmnt
EC-bodemas
Building material: identfiicationnumber:
NV8053.wkl leaching characteristics
170~93
U S 4 0 columntest in mgkg
agustd v a l u ~
composition
granular materials .b""n(
As
Ul
In
31
066
700
37500
~1
s m
mm
7 m m
w
o m
007
iom
042
260
25003
Q
133
I200
Cu
072
350
N 61
mm
8dln.l)
0.108
0.m
54
2.857
41
0 UX
57
12Y)OO 37500
n*mum -hum
m u 1 mln bglmem) -1.5
lop(*qn.l)) 0478
2.398
0.24)
14.800
8
0.334
0.378
0.W4
0.-
0.050
3
118%
0.548
0.W
0.a
0.001
0.700
1
-1.782
0.524
'
81
0.024
0.030
0.001
0 240
.1.830
0.448
'
D
81
0.053
0.099
0.010
0.740
1
-1.550
0.482
'
D
-2.553
0.151
o m
ow
sao
54
0.W3
0.001
0.w
0005
Mo
028
091
12500
58
0.144
0.135
0.010
OUr,
7
-1 078
0.19
NI
110
370
25000
01
0.097
0.131
0.005
1.m
1
-1.M
0.581
Pb
1 m
em
1m00
9 ,
005
043
5000
s.
OM
0.10
Sn
027
240
V
1 e4
9200
125000
Zn
3 w
1d.m
12wm
Ma, 2YJ00
290
4.10
Cl
mOO
660000
CN mmp
007
0 1
CN rdl
001
o m
2500
13W
10000
4WW
F Iol
dd.lhr
0.520
Y
81
w I h
0.002
500 00
ma, 12500
NA: No inlormn(nn ndlmbb. ERR: s1.darddsvlat~n zem
'
' ,'
D
D
aqua regia in mg/kg
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
311
PREDICTION OF ENVIRONMENTAL QUALITY OF BY-PRODUCTS OF COAL-FIRED POWER PLANTS
ELEMENTAL COMPOSITION AND LEACHING Ruud Meij KEMA P.O. BOX 9035 6800 ET Arnhem, the Netherlands
ABSTRACT
In the Netherlands the elemental compositions of the various streams of coal-fired power plants are well recorded. The information of these studies is used to calculate enrichment factors for the trace elements in ash, the vaporization percentage of minor and trace elements in flue gases, the degree of removal of gaseous minor and trace elements from flue gases in flue-gas desulphurization installations and the leaching percentage of the elements in ash. These relative parameters combined with trace-element analyses of the coals are used to predict the concentrations of trace elements in the ash, in the leachate and in the flue gases in the gaseous phase. The model is also valid for co-firing with secondary fuels. 1
INTRODUCTION
In 1995 29% of electricity in the Netherlands was generated using coal. Only imported bituminous coal is fired. Coal is imported from all over the world. Major suppliers are Australia and the USA. Other suppliers are Colombia, South Africa, Indonesia, Poland and China. Today mostly blends are fired. In the Netherlands the only boilers installed are pulverized coal-fired dry bottom types. The flue gases are cleaned by high-efficiency coldside electrostatic precipitators (ESPs) and by flue-gas desulphurization (FGD) installations of the lime(stone)/gypsum process. Table 1 shows some typical values for a 600 MWe coal-fired power plant in the Netherlands. The by-products are bottom ash, collected ash, gypsum and sludge of the waste-water treatment plant. The collected ash from the electrostatic precipitators (ESPs) is called pulverized fuel ash (PFA) in the UK and fly ash in the USA. In this paper it will be called PFA.
312 The policy in the Netherlands is in principle not to produce waste, but to produce usable residues (for the environmental legislation in the Netherlands see the paper by Van der Poel (Van der Poel, 1997). The electricity generating companies in the Netherlands founded a special firm for the marketing of the coal-firing residues: "de Vliegasunie" (Dutch Fly Ash Corporation). This firm also stimulates research of and experiments with applications. Long-term disposal of coal-firing residues is impossible at present. So far the Dutch Fly Ash Corporation has realized almost 100% utilization of all by-products. For more information on this subject see the paper by Van den Berg (Van den Berg, 1997). Table 1
Averaged mass flows at a modern coal-fired power plant of 600 MWe
description
unit
net capacity
MWe
600
full load hours
h.a-1
6,000
,,
description
%
thermal efficiency
40.5
unit
ratio bottom ash
12/88
bottom ash
ton.a1
15,600
PFA
ton.a1
115,000
energy demand
Mj.a-1
3.2-101
gypsum
ton.a1
41,000
coal demand
ton.a-1
1.2-106
sludge
ton-a1
600
coal, ash content
% (w/w)
coal, caloric value
MJ.kg1
27
coal, sulphur content
% (w/w)
fly dust emission
desulphurization efficiency % collection efficiency ESP
%
MEASUREMENTS
11
ton.a1
60
92
process water FGD m3.h"1
100
99.75
limestonedemand
ton.a1
24,000
0.7
P E R F O R M E D AT C O A L - F I R E D P O W E R P L A N T S IN
THE NETHERLANDS
2.1
Introduction
In this chapter an overview is given of the research in the field of (trace) elements performed at coal-fired power plants in the Netherlands. It concerns complete mass balance studies, studies limited to some streams and leaching studies.
313
2.2
Mass balance studies at power plants in 1980-1992
Following the reintroduction of coal as a fuel for power plants, the environmental consequences for electricity generation have been thoroughly studied; for instance in the Dutch National Coal Research Programme (NOK). A fairly important environmental aspect is trace elements. The concentrations and distributions of trace elements in coal, ash, and in flue-gas ( in the vapour phase) were determined in sixteen mass balance studies in coalfired power plants. The first flue-gas desulphurization (FGD) system was installed in the Netherlands in unit 13 (CG-13) of the Gelderland power plant in 1985. Extensive testing was performed at this unit in the following year (1986) in order to study the fate of (trace) elements in a coal-fired power plant equipped with a wet flue-gas desulphurization facility of the limestone/gypsum type. This aspect was researched in detail (Meij, 1989). An important aspect in these mass balance studies are the relations between the various streams, from which relative parameters could be deduced. 2.3
Studies at some related streams at power plants in 1988 and 1993-1995
Trace elements are emitted into air in solid (fly ash) and gaseous states. The emissions in the solid state are low due to their high degree of removal in ESPs. The emissions in the gaseous phase are relatively more important. In a plant with FGD equipment both emissions are further diminished. Hence, from an environmental point of view, the gaseous emissions require further research. Consequently, in 1988 the removal of gaseous minor and trace elements in FGD plants was studied at all Dutch units equipped with FGD systems. The concentrations of the gaseous trace elements were measured in the flue gases both upstream and downstream of FGD installations together with the concentrations in the feed coal. The relative parameters which could be deduced were the vaporization percentage and the removal in the FGD installation.
314 In the years 1993-1995 26 samples of feed coal with the corresponding pulverized fuel ash were analyzed for their elemental composition. It concerns mostly blends and represents the recent Dutch policy of coal purchase. 2.4
Studies of the leaching behaviour of bottom ash and fly ash in 1991-1995
Leaching behaviour of by-products has been studied at KEMA since 1980. However old data are not useful, because different leaching tests were used. It was KEMA who took the initiative to standardize leaching tests. Nowadays the column test and the availability test are mostly used in the Netherlands. In 1991 and 1995 45 bottom ash samples were studied: elemental composition and leaching behaviour established by the column test. In 1993-1995 26 fly ash samples were studied: elemental composition and leaching behaviour. The elemental composition was fixed after an aqua regia digestion and after a total digestion or INAA (instrumental neutron activation analyses; method without digestion). The leaching behaviour was established by the column test and the availability test. 2.5
Studies of the leaching behaviour of fly ash under field conditions
The leaching behaviour of pulverized fuel ash under field conditions has been studied in large lysimeters (height 3m80 and 0m95) at the KEMA premises since 1993. This project should lead to a better understanding of the leaching process in field condition and therefore a lot of parameters are monitored: water balance, composition of pore water and leachate, including pH, Eh and speciation of As, Cr and Se (Meij, et al., 1994; Van der Hoek et al. 1995).
315
Studies at power plants during co-firing in 1993-1996
2.6
In the years 1993-1996 11 test series were performed at coal-fired power plants (7x) and at a test facility at KEMA (4x) during co-firing secondary fuels such as sewage sludge, paper sludge, wood and petroleum-coke. In these test-series all the relevant streams were monitored and compared with the situation without co-firing including leaching behaviour.
PARAMETERS DEDUCED FROM STUDIES AT COAL-FIRED POWER PLANTS 3.1
Introduction
The studies mentioned in chapter 2 yield typical parameters that provide the relations between the streams concerned. These parameters are independent of the situation at that particular moment and can be used in a general way in models for predicting the composition of the streams and the leaching behaviour of the by-products (Meij, 1994).
Relation between elemental coal composition and elemental
3.2
ash-composition After combustion of the coal, ash remains. In general the ash contains the same elements as were present in the coal, but enriched in the ash by a factor equal to 100/(ash content in %). Three types of ashes are to be considered: -
ash collected on the down side of the boiler and called bottom ash or slag
-
ash collected in flue gases by flue gas control devices, such as electrostatic precipitator (ESP), this type of ash is named pulverized fuel ash (PFA) in the UK and fly ash in the USA
-
ash that escapes the flue-gas control devices and will be emitted through the stack, called fly ash.
In this paper the three types of ash are called bottom ash, PFA and fly ash, respectively.
316 The enrichment in the ash depends on the type of ash and the particular element. The term "relative enrichment" was introduced to properly describe the behaviour observed (Meij et al., 1983). The relative enrichment factor (RE) is defined as:
RE =
(element concentration in ash) 9(% ash content in coal) (element concentration in coal) 100 Classification of elements based on their behaviour during combustion in
Table 2
boiler and ducts with their Relative Enrichment factor (RE) class
bottom ash
PFA
fly ash 1)
behaviour in installation
I
= 1
=1
= 1
not volatile
IIc lib Ila
> of the industrial clinkers were synthetised in an electrical laboratory furnace. Corrective additions of metals were made, so as to obtain identical metal contents in the industrial samples and in the laboratory replicates (table 1). The release behaviour of the mortars made from these first two sets of samples were compared, using the tests described below.
II ][ il
mg]k~ 'I Industrial '"
I Replicate II Industrial
'
Cr
Pb
ZH
101 98 58 62
32
2oo
6 2 13 24
192 224 228 246
H Replicate H I Industrial I 18 ....m Replicate 155 14 248 table I : Heavy metal contents of the industrial samples and their laboratory replicates
A third set of clinkers was made in laboratory furnace, from an industrial raw meal that have been enriched in chromium, zinc and lead before the clinkerisation. Three metal levels were chosen, up to ten times the maximum concentrations usually encountered in industrial samples (table 2). The levels of hexavalent chromium have also been checked afterwards.
mg/kg
II HI|
Crtotal
Pb
"
Zn
B ...... i'80 .... 150 230 M 1005 680 1090 H 1810 1805 1920 table 2 : Heavy metal contents of the enriched laboratory cements
Cr(W) io0 610 1120
These three sets of samples have been crushed after addition of gypsum, so as their final composition and their hydraulic properties are similar to industrial Portland cement. Mortars have been prepared with the usual Cement/Sand/Water ratio = 1/3/0.5, and were mould in cylindrical bars ; after 28 days of maturation at 20~
and 98% of relative humidity, they were cut
into disks of 1,4 cm high and 8 cm in diameter (volume 70,4 cm 3, surface 132 cm 2, weight 152+_2g).
341
These test samples were submitted to sequential leaching tests in deionised water during 100 days overall, according to the following: 9 Static, batch leaching tests in sealed polyethylene bottles 9 Ratio : volume of leachant / sample surface = 5cm (i.e. 660 dm 3 of leachant ; liquid to solid mass ratio = 4,33) 9 Immersions:
- either 10 sequential contacts of 1d - 1d - 1d - 4d - 7d - 7d - 7d - 14d - 28d - 30d (total 100d), - or 1 continuous contact, during which aliquots are withdrawn at the times above mentioned. 9 Leachant : deionised water excepted otherwise stated.
Complementa~ experiments 9 Sequential leaching tests had been carried out as described here above, though using an alkaline leachant (pH 12.7) which is typically non aggressive for the cement matrix, or in pH conditions regulated at 7 by nitric acid additions. 9 Extraction tests: Mortar samples crushed to 100~m were contacted till equilibrium with solutions of various pH (liquid to solid mass ratio - 10). The influence of the pH o f the leachant upon the effectiveness of the fixation has been established, by using contact solutions maintained at various pH values between 6 and 13.
Results The leachates have been analysed by ICP AES or graphite furnace SAA. The detection limits are 4lag/l for chromium, 10lag/1 for lead and 3 lag/] for zinc.
l-Comparision of industrial samples and laboratory replicates The table 3 presents the total release of metals by the mortar bars contacted with deionised water. Most o f the leachates exhibit metal levels below the detection limits.
cumulative amount leached' in 100d (lag) Zn Pb Cr I Industrial .....! Replicate II Industrial II Replicate HI Industrial HI Replicate
NS NS NS NS NS NS NS NS NS NS NS NS NS 29 NS 33 NS NS table 3 : Metal release from the industrial samples and the laboratory replicates (10 immersions, total duration 100d) NS 9non significant because too many leachates concentrations < detection limit The metal release being non measurable in deionised water, these two sets of samples have been tested in conditions more aggressive for the cement matrix, to make sure that the clinkerisation in
342 laboratory furnace provides samples whose leaching behaviour is representative of industrially-made clinkers. Lead and zinc
For the industrial samples and their laboratory replicates, the lead and zinc concentration in the leachates are consistently under or close to the detection limits, whatever the conditions applied: sequential leaching of monoliths in deionised water (table 3), in alkali, at pH 7 regulated, or even during the tests of extraction from crushed material at various pH values. Hence there is no measurable difference between the two sets of samples, considering the release of lead and zinc. Chromium
The chromium release is also quite low, and sometimes under the detection limits. The cumulative chromium extracted over the test duration can though be worked out. In the different chemical conditions tested, the metal release from the monoliths appeared to be directly linked to the chromate content of the solid. The clinkering conditions of the sample (industrial kiln or laboratory furnace) has no influence. This result has been confirmed by the extraction tests on crushed material, as shown in table 4 : IIdeionised water 0tga)
pH 7 (ptga)
791 I industrial 50 789 I replicate 31 355 H industrial 15 489 II replicate 22 1005 HI industrial 66 1036 .... !II replicate 84 table 4 : Solubilisationof chromium from crushed material
% Cr 6+ extracted at pH7
110 106 105 108 110
Concerning chromium, the samples made from laboratory clinkers present a leaching behaviour very similar to the corresponding industrial samples. Concerning lead and zinc, and within experimental accuracy, the laboratory samples do not exhibit any obvious discrepancy from the industrial ones. The study has therefore been pursued using solely the enriched laboratory clinkers.
2- Release mechanisms studied on enriched laboratory samples
The mortars made from enriched clinkers exhibit a measurable metal release (table 5), the first extracts being the more concentrated. The amount of lead and zinc leached are still very low ; the corresponding concentrations in the leachates are in the lag/l range. The chromium levels are somewhat higher. For the sample H (enriched to ten time the usual content in industrial samples), the concentrations in the leachates reach 150~tg/1, in the experimental conditions here applied.
343
Cumulative amount leached (ttg)
Enriched
Pb
Cr
samples
Zn
34 NS NS 245 34 26 M 459 101 23 H table 5 : Metal release from the laboratory enriched samples (10 immersions total duration 100 d) NS ' non significant because too many leachate concentrations < detection limit B
~nc
The figure 1 shows the extraction of zinc from crushed mortar at various pH values. Zinc appears to be insoluble for the pH higher than 8. Its concentrations in the leachates are lower by of orders of magnitude than the solubility of the common zinc compounds such as hydroxides or carbonates, thus indicating a chemical bounding in the solid phase. 25C
pg/I Zn F"
20C 15C 10C 50 0
I
7
8
9
10
11
pH
12
figure 1 : Extraction of zinc from crushed material - sample H Such a low solubility strongly limits the release of zinc by the mortar monoliths : The level of zinc in the leachates of mortars blocs are consistently in the ~tg/l range. The cumulated amount leached in 10 immersions (total length 100 days) are between 15 and 25 ~tg; there is no significant influence, neither of the chemical conditions of leaching, nor of the metal content in the solid. Lead The solubilisation curve of lead from crushed mortar is presented figure 2. Lead is partly extractable from the mortar in strongly alkaline conditions, but it is bound in the solid for pH values under 12.5. Just like for zinc, the equilibrium concentrations against pH are far lower than the expected solubility of lead
compounds 6,7.
344
5C) pg/I Pb
)C)
5C)
0
7
8
9
10
12
11
13 pH 14
figure 2 : Extraction of lead from crushed material at various pH - Sample H The various experiments undertaken have pointed out the proportionality between lead release and its level in the mortars. The results of lead released from monoliths are therefore expressed as percentages of the metal content of the solid (figure 3) 9 1,4%
% lead leached in 42 days
1,2% 1,0%
0,8% 0,6% 0,4% 0,2% 0,0%
!
pH 7
!
D-water
pH 12,7
figure 3 : Percentage of lead leached from mortar bars, in various leaching conditions The mortar bars exhibit a good retention of lead when they are contacted with neutral or moderately alkaline solutions (the leachant pH rose up to about 11 during the tests in deionised water); but when the leachant pH is very high, lead is partly released by the mortars. The comparison of the two previous curves shows that lead release is strongly influenced by a solubilisation process controlled by the pH value.
Chromium Whatever the leaching conditions, the release of chromium is directly proportional to the chromate content of the sample. The extraction test at various pH (figure 4) provided important results : - As foreseen with the non-enriched samples (w 2.), the chromium in the leachates is solely in its hexavalent form. Hence the trivalent chromium is never solubilised. - The total amount of chromate of the sample is extracted for pH values under 10. - Chromate, though usually soluble, is bound in the solid in the pH range 1 l-13 ; we point out the fact that this range corresponds to the pH domain in which the ettringite phase is stable according to Damidot and Glasser 8.
345
Cr extrait (rng/I) 30252015-
1050
6
I
I
I
I
7
8
9
10
11
12
13 pH 14
figure 4 : Extraction of chromium from crushed material at various pH - Sample H The release of chromium by the mortar bars is shown figure 5. Just like in extraction tests, only hexavalent chromium is leached. The mortars contacted with deionised water release less metal than in aggressive conditions such as pH 7 regulated. Surprisingly, the release is quite high in an alkaline solution of pH 12.7, although the extraction test proved that the solubilisation of chromium is minimal for a pH value of 12.5. Hence the release of chromium cannot be explain taking into account only the influence of the pH.
2,0%
chromium extracted in 42 days
1,5% 1,0% 0,5% 0,0%
. . . . . I
pH 7
l
D-water
I
pH 12,7
figure 5 : Percentage of chromium leached from mortar bars, in various leaching conditions
Discussion The release of zinc by the monoliths contacted with deionised water is very low (In our tests, less than 25~tg of zinc is extracted in 100 days of leaching, whatever the metal content of the mortars). It is due to the fact that this metal is bound in the solid as a compound which is insoluble in water. This result can be extended to the various chemical conditions applied in our tests, as zinc is chemically retained in the matrix in the pH range 8-13. It must be highlighted that the elevated alkalinity of cement ensure an elevated pH in the mortar bars even if the surrounding solution is fairly
346 aggressive : this remark explains why the zinc release by the monoliths is still low when the leachant is maintained at pH 7. The release of lead from the monoliths is controlled by a solubilisation process strongly dependant of the pH conditions applied to the material. This metal being bound in the matrix for the pH lower than 12.5, its release is limited as long as the alkalinity of the leachant is not too elevated, which is the case of leaching in deionised water or at pH 7 regulated. Very high values of pH would be reached only during a prolonged contact between the cement and the liquid, or when the mortar is submitted to an extremely alkaline solution. Such conditions of pH does not occur during the real utilisation of cement based materials. An other part of this s t u d y I has proved that the chromate is chemically bound in the ettringite structure, in substitution for sulfate. The interpretation of the leaching results requires to consider the specific chemical properties of such chromate-ettringite, and especially its solubilisation mechanisms. As an example, the figure 6 presents the evolution of chromium concentration during an unique continuous contact. The metal level rise up and becomes stable after a few days. This stationary value is directly proportional to the chromate content of the solid. The release of chromium is due to the partial dissolution of ettringite in which it is contained as an impurity. The stationary level of chromium is due to the quick saturation of the bulk with respect to ettringite. Depending upon the chromate level in the solid, the bulk is in equilibrium with ettringites containing different amounts of metal impurity. The partial dissolution of these phases liberates a corresponding amount of chromate into solution. 200 pg Cr 150
XX
X
X
l
x
100 ~ , 50
X
eB
*M
"
xH
poe
0 0
50
75 j o u r s 2 5 0
figure 6 : Amount of the chromium released in the leachant during one prolonged contact
Conclusions The first part of our study has shown that the traces metals occurring in Portland cement clinker are retained in the relevant mortars bars when they are submitted to deionised water. The metal concentration in the leachates are consistantly under or close to the detection limits. Furthermore, samples enriched in metals up to ten times the levels usually encountered in industrial samples also exhibit very low metal release.
347 To explain these results, the work presented in this paper focused on the understanding of the retention mechanisms of the trace metals, and on the identification of the parameters controlling their release: -
Zinc is bound in the solid and is nearly insoluble in the chemical conditions applied ; therefore its release by mortar bars contacted with deionised water is very low. This result has be extended to leachants in the pH range 7-13.
-
Lead is nearly not released in deionised water: we showed that this metal is insolubilised by the cement matrix, provided the leachant pH is under 12.5. It must be pointed out that in real conditions of use of cement materials, the contact water never reaches such elevated pH.
- The trivalent chromium is bound in the mortars. The soluble chromate ions are partly retained in the matrix owing to their fixation in the ettringite phase. Their release is linked to the dissolution mechanisms of this phase, and the quick saturation of the leachant with respect to ettringite appears as the limiting factor.
Acknowledgement
The authors would like to acknowledge the Association Technique de l'Industrie des Liants Hydraulique (,4 TILH) and the Agence De l'Environnement et de la Maitrise de l'Energie (ADEME) for supporting these studies.
References
1. Sercl6rat, I. Les m~taux traces clans le clinker de ciment Portland- R~tention dans les mortiers et integration dans les hydrates de ciment. Thesis ISAL 960140, INSA Lyon, France (1996). 2. Germaneau, B., Bollote, B. and Defosse, C. Leaching of heavy metals from mortar bars in contact with drinking and deionised water. Emerging technologies symposium on cement and concrete in the global environment, Chicago (1993). 3. Sprung, S. and Rechenberg, W. Einbindung von schwermetallen in Sekundarstoffen durch Verfestigen mit Zement. Betonteschnische Berichte 1986-88, Beton Verlag Ed., Dusseldorf (1989). 4. An analysis of selected trace metals in cement and kiln dust. R&D serial N~
Portland
Cement Association, Skokie, ILL (1992). 5. Kanare, H.M. and West, P.B. Leachability of selected chemical elements from concrete. Emerging technologies symposium on cement and concrete in the global environment, Chicago (1993). 6. Pourbaix, M. Atlas des ~quilibres ~lectrochimiques ~t 25~
Gauthier-Villar, Paris (1963).
7. Sanchez, F. E,tude de la lixiviation de milieux poreux contenant des esp~ces solubles. Application au cas de d~chets solidifids par liants hydraulique. Thesis ISAL 960118, INSA Lyon, France (1996). 8. Damidot, D. and Glasser, F.P. Thermodynamic investigation of the CaO-A1203-CaSOa-H20 system at 25~ and the influence ofNa20. Cem. Concr. Res. 23 : 221 (1993).
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997Elsevier Science B.V. All rights reserved. STUDY OF CEMENT-BASED SEWAGE SLUDGE ASH
MORTARS
349 CONTAINING
SPANISH GROUND
Monzo J., PayS, J., Borrachero M.V., Bellver A. and Peris-Mora E. Grupo de Investigaci6n en Quimica de los Materiales (GIQUIMA) Departamento de Ingenieria de la Construcci6n Universidad Polit6cnica de Valencia Camino de Vera s/n 46071-Valencia (Spain)
Abstract
A study of cement based mortars containing spanish ground sewage sludge ash is presented. The influence of original and ground sewage sludge ash on mortars workability and compressive strength has been studied. An initial decrease of workability is observed when 30 % of Portland cement is replaced by original ash. When ash grinding time increases a little increased of workability is observed. Mortars containing a 15 % of ash cured at 40~ for 14 and 28 days showed equal or higher compressive strength than control mortar. No significative differences sere observed among mortars containing ash with different grinding times.
Introduction
As a consequence of water treatment processes, a large amount of sewage sludge is obtained. A part of this sewage sludge is used in agriculture as organic fertilizer and soil amendment. But, depending on the origin of water treated (municipal, industrial, ...) chemical parameters change, and in some cases accumulation of heavy metals and other toxic compounds can be occur, producing adverse impacts on human health and the environment (1). The incineration of sewage sludge is one alternative to manage the excess of sewage sludge production, that some cities are using. This method permits to reduce the volume until 90%, and the sewage sludge ashes obtained can be deposit in controlled landfills. However, landfill sites space limitations and environmental problems have guided the investigations of altemative uses in construction. Sewage sludge ash (SSA) has been used to manufacture bricks (2), to incorporate into concrete mixtures (3,4),in asphaltic paving mixes (5) and mortars (6). In a previous research (7) some properties of cement-based mortars containing SSA were studied. The objective of the present work is to study the influence of grinding of SSA on workability and strength of cement-based mortars. SSA were obtained from sewage treatment plant of Pinedo (Valencia, Spain), that produces about 2,000 tons/year.
350 EXPERIMENTAL Materials. Portland cement used for mortar preparation was conforming to the specifications of ASTM type I. Fine aggregate was natural sand with 2.94 fineness modulus. SSA were obtained from sewage treatment plant of Pinedo (Valencia, Spain). Sikanol-M was used as plasticizer.
Apparatus and procedures. Samples of original SSA were ground using a laboratory ball-mill (Gabrielli Mill-2). SSA samples were introduced into the bottle-mill containing 98 balls of alumina (2 cm diameter) and were ground during 2.5, 5 and 10 minutes. Mortar specimens cast in square prismatic mortar molds with internal dimensions of (40x40x160) mm were used. Preparation of mortars was carried out according to ASTM C-305 test (8), mixing 450 g. of Portland cement, 1350 g. of natural sand and 225 mL of water for control mortar and the rest of mortars replacing by mass a 15% of Portland cement by original or ground SSA. Mortars were put in a mold for obtaining specimens, which were stored in a moisture room (20-x1~ for 24 hours. Afterwards the specimens were demoulded and cured by immersion in 40-x1~ water in order to activate the hydration process until testing at 3, 7, 14 and 28 days. Mortars for workability studies were prepared according to ASTM C-305 (8), mixing 450 g. of Portland cement, 1350 g. of sand and varying water volumes between 200-225 mL for control mortar. The rest of mortars were prepared replacing growing percentages of Portland cement by ground SSA and workability test were developed following ASTM C-109 (9) test. Some tests were developed using a mortar plasticizer (Sikanol-M) in a 0.1% in weight respecting SSA + cement. Freshly prepared mortars were placed into a conic mold which is centered on the flow table. Mortar was put on two layers and compacted with a wooden tamper (10 times). Afterwards, the mold was removed and the table was dropped 15 times (one per second). Flow table spread (FTS) was given as a mean of maximum and minimum diameters of the spread cone.
Results and discussion SSA obtained from water treatment plant was analyzed and the results obtained are presented in Table 1. From among these data can be emphasized the high concentration of sulfate in SSA (12.4 % expressed in SO3content ). High concentration of sulfate are due, chemical reagent used in water treatment. Workability (FTS). The influence of original and ground SSA on mortar workability has been studied. In Figure 1. Flow Table Spread (FTS) versus SSA grinding time is represented for mortars containing a 30% of SSA and 0.5 water cement ratio. In this figure is compared the platicizer influence on FTS. An initial decrease of workability is observed when a 30% of control mortar cement is replaced by original SSA (SSA 0), a more marked decrease is observed in mortar containing plasticizer. A different behavior is observed, for mortars with or without plasticizer, when SSA grinding time increases. Mortars containing plasticizer increase FTS when grinding time do. The most important increases is observed between SSA 0 and SSA 2.5. The absence of plasticizer shows a decrease of FTS when grinding time increase. In all cases workability of mortars containing plasticizer was higher than mortars without it.
351 Table 1. Chemical Composition of Original Sewage Sludge Ash and Portland Cement ~,~
,
~6~
~
.
~.~ . . . . .
,,
Moisture
0.5
....
Loss on ignition
5.1
3.02
Insoluble Residue
16.1
0.95
SO3
12.4
3.54
7.4
2.85
SiO2
20.8
21.00
CaO
31.3
62.87
MgO
2.6
1.05
AI20 3
14.9
4.94
P205
6.7
0.1
Fe20 3
200 175 A
E E
O0
0.1% sikanol 150 -
i-
U.
125
100 t Control
0% s i k a n o l i SSA 0
i S S A 2.5
i SSA 5
1 S S A 10
S S A grinding t i m e (min)
Figure 1. FTS values of mortars containing 30% of SSA versus SSA grinding time
In Figure 2. FTS versus volume of water for mortars containing a growing replacement of cement by ten minutes ground SSA and 0.1% (in weight) of plasticizer is represented. As could be expected, a increase of FTS is observed when water volume do, but this behavior is more pronounced when SSA percentage is slow (15 and 30%). Probably, the important adsorption of water on SSA particles surface determines the short increase of FTS when high SSA percentages (45 and 60%) are used. Compressive Strength (Re). Preliminary studies make clear, in first place, that SSA did not present autocementicious hardening, whereas, secondly, mixtures of Ca(OH)2 -SSA hardened in few days. This behavior indicated that SSA could present pozzolanic activity. The influence of original and ground SSA on mortars compressive strength has been studied. Mortars containing a 15% of ash and 0.5 water / (cement + SSA) ratio were cured at 40~ and tested at 3,7,14 and 28 days (Figure 3.). No plasticizer was used. The results obtained showed higher Rc in short
352 curing time (3 and 7 days) for control mortar (without ash). When curing time increases (14 to 28 days) mortars containing ash showed equal or higher Rc than control mortar. This fact confirm pozzolanic behaviour of SSA. No significatives differences were observed among mortars containing SSA with different grinding times. 175 -
~.
p.
"
15%
150
125 ,,
,,,
100 200
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
205
210
215
220
225
Volumeof water(mL) Figure 2. FTS values of mortars containing from 15% to 60% of SSA versus volume of water
50SSA2.5 45-
.,~
SSA10
control
c o n t r ~
7
40
A
A
n =E
/
1
v
0
35
i
30
25
0
. . . . . .
!
7
. . . . .
I
. . . . . .
14
Age (days)
v
21
. . . . . .
28
Figure 3. Compressive strength of mortars containing 15% of SSA with different grinding times
4
. . . . . . .
0
i
7
. . . . . .
T
. . . . .
14
i
21
. . . . . .
28
Age (days) Figure 4. Flexural strength of mortars containing 15% of SSA with different grinding times
353 Flexural Strength (Rf). The influence of original and ground SSA on flexural strength of mortars has been studied (Figure 4). The results obtained showed higher Rf for control mortar than ash mortars except for 28 days curing time that mortar containing 2.5 minutes ground SSA that gave same Rf than control mortar. No significative tendency between SSA grinding time and Rf is observed.
Conclusions 1. High concentration of sulfate are present in SSA, due to chemical reagents used in water treatment 2. A cement replacement by SSA in mortars produces a decrease of workability, being a more marked decrease when mortar contains plasticizer 3. A increase of workability in mortars containing plasticizer is observed when SSA grinding time do 4. Mortars cured f4 to 28 days at 40~ containing 15 % of Portland cement replaced by ground SSA gave equal or higher compressive strength than control mortars. No significative differences were observed among mortars containing SSA with different grinding times 5. Flexural strength for mortar was higher than 15 % SSA replaced mortars except for 28 days curing times
Acknowledgment We would like to express our gratitude to Mr GermAn Rodriguez and Mr Alejandro Mulet from Consell Metropolit/~ de l'Horta for providing us the samples of SSA, SIKA S.A. (Valencia office) and Cementos Asland (Puerto de Sagunto plant) for their support for this research projet.
References 1. Dean, R.B. and Suess, M.J. "The risk to health of chemicals in sewage sludge applied to land" Waste Manage. Res. 1985, 3, 251-278 2. Allenman, J.E. and Berman, N.A. "Constructive sludge management: biobrick" J.Environ. Eng. Div., ASCE, 1984, 110, 301-311 3. Tay, J.H. "Sludge ash as filler for Portland cement concrete" J. Environ. Eng. Div. ASCE 1987, 113,345-351 4. Tay, J.H. and Show, K.Y. "Clay blended sludge as lightweight aggregate concrete material" J. Environ. Eng. Div., ASCE 1991, 117, 834-844
354 5. A1 Sayed M.H., Madany I.M. and Buali A.R.M. "Use of sewage sludge ash in asphaltic paving mixes in hot regions" Constr. Build. Mater. 1995, 9, 1, 19-23 6. Bhatty, J.I. and Reid, J.K. "Compressive strength of municipal sludge ash mortars" ACI Mater. 1989, 86, 394-400 7. J.Monz6, J.Payfi, M.V.Borrachero and A.C6rcoles, Cement and Concrete Research, 1996, 26, 9, 1389-1398 8. ASTM C-305-80. "Mechanical Mixing of Hydraulic Cement Pastes and Mortars of Plastic Consistency". 9. ASTM C-109-80. "Standard Test Method for Compressive Strength of Hydraulic cement Mortars (Using 2-in. OR 50-mm Cube Specimens)".
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
355
FLY ASH - USEFUL MATERIAL FOR PREVENTING CONCRETE CORROSION
S. Mileti61, M. I!i61, J. Ranogajec 2 and M. Djuri6 2 IIMS Institute for Materials Testing, Beograd, Yugoslavia, 2Faculty of Technology, Novi Sad, Yugoslavia
Abstract Large quantities of fly ash is produced in our country every year. Most of fly ash is got from lignite that means that this material is not so useful for concrete production. This paper presents results of investigations of sulphate corrosion of concrete made from portland cements and portland cements with the fly ash addition. The addition of fly ash appears to be very useful for preventing sulphate corrosion of concrete even in the case of very strong ammonium-sulphate corrosion according to our results. The effect of ammonium sulphate solution on the durability of Portland cements (various C3A content) with partial replacement of 30% mass percent'fly ash was investigated. Results show that fly ash addition to Portland cement can improve resistance to ammonium sulphate attack. Key words- Portland cement, fly ash, corrosion, ammonium-sulphate
1. Introduction Concrete has been widely used as the most important constructional building materials in the world. More and more attention has been paid to the mechanical properties and durability of cement concrete. Generaly, the durability and the degradation coefficient of the concrete has been considered as a dominant factor in addition to the fact that the mechanical properties could satisfy the demand of the construction design. In practise, concrete buildings suffer simultaneously mechanical, chemical and physical attacks. Therefore, the effect of mechanical stresses must be taken into consideration when durability and corrosion resistance of a concrete are estimated, i.e., the study of stress corrosion of concrete is necessary and very important for durability. 1-4 Chemical degradation of concrete is the consequence of reactions between the constituents of cement stone, i.e. calcium silicates, calcium aluminates and above all calcium hydroxide etc., with certain substances from water, solutions of soil, gases, vapours, etc.5-8 The most important aggressive ions are: SO4 2", Mg 2+, NH4 § CI-, H § HCO3-. Primarily, the types of chemical corrosion of concrete can be devided into two groups, i.e., expansive corrosion and dissolving corrosion, with respect to the cause of failure of concrete. The attack of sulphate ions on cement stone can cause expansion, in general due to the formation of ettringite C3A.3CaSO4.32H20, in the shape of prismatic crystals. 9'1~ The consequences are damages to the concrete and destruction at worst. The concrete corrosion by ammonium sulphate, for example, covers the most aggressive corrosion on concrete, neither balancing nor creation of protective gel takes place. In this case concrete is damaged not only by expansion, but also by dissolving the cement stone.
356 In this investigation the method of Koch and Steinegger 8 'xs used to test the sulphate resistance of the cements. According to the authors criterion of the sulphate resistance was the quotient:
Rc=Flexural
Strength of the Sample Stored in the Sulphate Solution Flexural Strength of the Sample Stored in Water
/1/
The results show that there is a considerable influence of the mineral composition of Portland cement clinker and cement on the behaviour of concrete in the presence of aggressive sulphate and ammonium ions. For the manufacture of concrete resistant to the attack of aggressive ions special attention should be paid to the selection of cement. ~0
2. Experimental To investigate the resistance of cement to sulphate attack Portland cement and Portland fly ash cement manufactured in Yugoslavia were used" 9 Portland cement B (PCB)- according to the European cement standard EN 197-1: CEM-I 9 Portland fly ash cement B (cement clinker B) with 30% fly ash (PCBP)- according to the European cement standard EN 197-1: CEM II/B-V 9 Portland cement K (PCK)- according to the European cement standard EN 197-1: CEM-I 9 Portland fly ash cement K (cement clinker K) with 30% fly ash (PCKP)- according to the European cement standard EN 197-1: CEM II/B-V The potential phase analysis, chemical contents, physico-chemical and mechanical properties were determined for all starting materials used. Cement pastes were prepared by Koch-Steinegger method. 8 Specimens of 1x 1x6 cm were molded and compacted by vibration. After one day at 100% relative humidity the specimens were demolded and kept immersed in water for 21 days. After that, samples were immersed in the aggressive solutions of different concentrations for different periods of time. Control samples were prepared and stored in distilled water under the same conditions as reference. As aggressive solution, ammonium-sulphate concentrations 2.5%, 5%, 7.5% and 10% was used, but, results only for 10% ammonium-sulphate solution are presented. The mass change of samples, SO42- content change in solution and flexural strength were measured after 7, 14, 28, 56, 90, 180 and 270 days of storage in the aggressive solution. Other testing methods used in this work are: 1. Determination of standard strength (EN 196-1) 2. Chemical analysis (EN 196-2) 3. Determination of setting time (EN 196-3) 4. Determination of the sieve residue (EN 196-6) 5. Determination of specific surface (EN 196-6) 6. Calculating the potential phase analysis (ASTM C 150)
357 3. Results and discussion
The selected aggressive environment represents very strong aggressiveness to ensure fast results for the real conditions which can be present in underground waters in Yugoslavia. TABLE 1 Potential phase composition of Portland cement clinker Potential phase composition, %mass CaS C28 C~A C4AF
Portland cement clinker KB KK 57.5 67.0 13.5 12.7 13.3 6.6 8.7 9.1
The potential phase analysis of the Portland cement clinkers is given in Table 1. It can be seen that the cements have low and high C3A content in clinkers of 6.6% and 13.3% influencing the sulphate resistance. The ordinary Portland cement is not resistant to the attack of sulphates because it has a considerable content of tricalcium aluminate - C3A, whose hydrates react with sulphate ions, giving expansive compounds. Portland cement with increased resistance to sulphates must have a low content of C3A. According to the literature the difference in the C3S content could be significant regarding sulphate resistance too. TABLE 2 Fly ash chemical composition Chemical composition, %mass LOI SiOz A120~ Fe20~ CaO MgO SO~ N%0
K20 Hydrated water Insoluble residue
Fly ash 5.7 50.9 21.7 11.6 6.5 2.7 0.05 0.3 0.7 34.6 76.6
Table 2. presents the chemical composition of fly ash. According to the high content of SiO2, A1203 and Fe203 and the low content of CaO the fly ash is suitable for cement production though loss on ignition was relatively high.
358
TABLE 3 Chemical composition of cements Chemical composition, %mass
Cement PCB 19.7 7.0 2.7 62.0 0.1 0.8 0.1 2.0 2.2 0.4 0.4 0.07
SiO 2 A1203 Fe2Oa CaO Insoluble residue LOI CaO free SOa in CaSO 4 MgO Alkalies as NaaO K20 MnO
PCK 21.0 5.3 2.9 63.8 0.1 0.7 0.4 1.7 1.4 0.3 0.3 0.07
PCBP 14.0 6.2 2.7 44.7 20.2 3.0 0.0 2.0 2.4 0.4 0.2 0.04
PCKP 15.6 4.9 2.9 47.5 18.7 2.9 0.0 1.5 1.2 0.4 0.3 0.05
The chemical composition of the cements is presented in Table 3. All cements meet Yugoslav standard JUS B.C1.011. Portland fly ash cements have a higher loss on ignition and contain less free CaO than Portland cements. TABLE 4 Ph~,sico-chemical properties of cements Physico-chemical properties Sieve residue at 0.09 mm sieve, %mass Density, g/cm ~ Specific surface, cmZ/g Setting -standard consistence, %mass -initial time, min -final time, min Volume stability -Le Chatelier test, mm
PCB 1.8 3.1 3320
Cement PCK PCBP 2.6 5.2 3.2 2.9 3100 3720
25.8 165 225
23.8 165 225
28.0 240 330
27.5 255 360
1.0
1.5
1.0
1.0
PCKP 6.0 2.9 3710
Table 4. presents figures characterizing fineness, density, standard consistency, setting time and volume stability of the test cements. Obviously the addition of fly ash raises the water demand for standard consistency and sieve residue and extends setting time, but has no significant influence on other characteristics. All characteristics are in compliance with Yugoslav standard JUS B.C1.011.
359 TABLE 5 Standard strength of cements Strengths, MPa PCB Flexural: -2 days -3 days -7 days -28 days Compressive: -2 days -3 days -7 days -28 da~cs
]
PCK
Cement [ PCBP
PCKP
4.4 5.3 7.2 8.0
3.7 4.4 7.4 8.9
2.5 3.6 6.2 8.3
2.1 2.9 4.7 8.4
15.7 19.8 30.2 40.3
13.2 16.0 32.8 50.9
8.8 14.9 24.2 39.5
7.4 10.4 19.4 44.9
Table 5. gives values for flexural and compressive strengths of cements after 2, 3, 7 and 28 days. Due to the clinker phase composition, Portland cement PCK had lower initial strength but higher later strengths after 7 days. The Portland fly ash cements had lower compressive strength even after 28 days than the corresponding Portland cements. In this way, complete characterization was implemented regarding all the cements used in this investigation. Figs. 1. and 2. presented mass change of the samples immersed in mentioned aggressive solution. Generally, it is obvious that Portland ash cements had much lower mass change than Portland cements. Test samples from cement PCB lasted only 56 days due to expansion components formation. Capability of mass change for cements PCK and PCKP compared to cements PCB and PCBP was much higher. Those cements also lasted longer. The reason for this must be in clinker composition diferencies.
25
20
/
0 15 o10 o o~
~
_.____-Q
Q
j.. l j
I
l0
-//
--I--PCB
--e--
,
0 0
I
50
,
I
100
,
I
,
150
i
200
,
PCBP
i
250
,
300
Time (days) Fig 1. Mass change for Portland cement PCB and Portland fly ash cement PCBP
360
25
20
./ I f
o 15 Im
vl
- - 9 PCK 9
--V-- PCKP
I
50
,
I
100
,
I
,
150
I
200
,
I
250
300
Time (days) Fig. 2. Mass change for Portland cement PCK and Portland fly ash cement PCKP Figs. 3. and 4. presented SO4 2- content change of the aggressive solution where test samples were immersed. Generally, it is obvious that for Portland ash cements had much higher SO42content change than for Portland cements. Test samples from cement PCB lasted only 90 days due to expansion components formation. It is asumed that all SO42 content changes in solution was directly connected with SO42 bonding in test samples with aluminate components. Capability of SO42 content change for cements PCK and PCKP compared to cements PCB and PCBP was much higher. Those cements also lasted longer. The reason for this must be also in clinker composition diferencies.
361 1300 1200 II00
1000 900 --D--PCB l --o--PCBP
800 0 o
700
r.~
600
e-'
500
,
0
I
,
50
I
,
100
I
,
I
150
,
200
I
,
250
300
Time (days) Fig.3. SO42- c o n t e n t change for Portland cement PCB and Portland fly ash cement PCBP
1300 1200 I100 1000 [ --o--PCK I - - A - - PCKP
900
~
o 800
~
0
700 -
o* r.t3
6~176 f
500
0
I
50
,
I
100
,
I
,
150
I
200
,
I
250
,
300
Time (days) Fig.4. SO42" c o n t e n t change for Portland cement PCK and Portland fly ash cement PCKP
362
10
0.8
0.6
0.4
~O~o~
0.2
0
--i--PCB --o--PCBP
50
100
150
200
i 250
m u I 300
Tmae (days) Fig. 5. Sulphate resistance coefficients for Portland cement PCB and Portland fly ash cement PCBP, according to Koch-Steinegger
tO
0.8
0.6
--m-- PCK I --o-- PCKP
0.2
0.0
0
,
I
50
,
I00
,
150
-
I
200
I 250
w 300
Ttme (days) Fig. 6. Sulphate resistance coefficients for Portland cement PCK and Portland fly ash cement PCKP, according to Koch-Steinegger Koch-Steinegger figures with sulphate resistance coefficients are presented in Figs. 5 and 6. It can be seen from Koch-Steinegger method that the Portland cement with fly ash has better resistance to sulphate aggression for the both kind of Portland cements. Hence, no one
363 of the tested cements shows satisfactory resistance, what is understandable because 10% (NH4)2SO 4 solution was used instead of 4.4% Na2SO4 solution as aggressive medium. Used Portland cements with low content of C3A with and without 30% fly ash (PCK and PCKP) had better sulphate resistance than Portland cements with high content of C3A (PCB and PCBP). The results of sulphate susceptibility tests according to Koch-Steinegger characterized by degradation coefficients are presented in Figs. 5 and 6. From the diagrams, it can be clearly seen that cements with the addition of 30% of fly ash showed distinct higher resistance against the ammonium sulphate solution. The increase of corrosion in the very begining for all cements is a normal phenomenon, because the creation of expansive compounds closes the pores and makes cement paste impervious to aggressive ions. However, further increase in the volume within the paste very quickly results in cracking. For the Portland cement PCB with the high content of C3A this occured after 28 days only. Samples from Portland cement PCK and cements with fly ash addition have endured 90 days. This can be explained by the fact that fly ash in cements has formed a protective layer thus retarding corrosion process and increasing durability. Portland cements, on the other hand, showed, depending on C3A content, either linear or exponential type of degradation after initial period of forming the protective layer. This layer obviously became negligible due to the action of NH4+ ions thus opening new pores and accelerating corrosion process again. The investigations are evidently encouraging, because the addition of fly ash has pointed to realistic prospects for its positive effect. Therefore, cement PCBP, with the addition of fly ash, shows good resistance to the aggressive attack by sulphate solution, although this cement is, due to its phase composition, very unsuitable in that sense. It is evident that resistance of Portland cement to sulphate attack is directly related to its content of C3A. This was confirmed even in the case of complete elimination of physicochemical influence of fly ash on the properties of cement (bonding Ca(OH)2, filling pores, etc.).
4. Conclusion
The results of testing the attack by aggressive sulphate solutions allow the conclusions: 1. The resistance of cements to sulphate attack is higher with a lower content of aluminate in clinker PCK and especially with addition of fly ash to the cement. 2. The addition of 30 % of fly ash to Portland cement as a replacement improves the durability of Portland cement to a considerable degree. 3. Koch-Steinegger method shows that both Portland cements did not resist the strong attack of 10 % (NH4)2SO 4 solution.
following tricalcium of clinker extremely
Reference
Biczok,I.: Concrete Corrosion, Concrete Protection. 8th Edition. Budapest, 1972. Mehta,P.K.: Mechanism of sulfate attack on Portland cement concrete - Another Look. Cem.Concr.Res. 13 (1983) pp. 39-51. Mitrovid, N. and Du~id, V." Sulphate corrosion of concrete. Proc. of Yugoslav. Symposium, Split, 1985, pp. 59-75. Moskvin,V.M., Ivanov,F.M., Alekseev,S.V. and Guzeev,E.A.: Korozija betona i 2elezobetona. Moskva, 1980.
364
10. 11.
12.
Regourd,M.: Structure and behaviour of slag Portland cement hydrates. 7th Intern. Congr.Chem.Cem., Paris, 1980, Vol. I, pp 278-291 Taylor,H.F.W.: Crystal structure of some double hydroxide minerals. Mineral.Mag. 39 (1973) pp. 247-256. Miletid, S. and Ilid,M." Sulphate corrosion of portland cement with various mineral compositions. Proc. 15. Symposi. Corrosion and Protection of Materials, Beograd, 1995, pp. 255-262. Koch,A., Steinneger,H.: An rapid method for testing the resistance of cements to sulphate attack. Zement-Kalk-Gyps 13 (1960), No.7, pp. 317-324. Matsufuji,Y., Koyama,T. and Harada,S.: Service life predictive method of building materials. Proc. Durability of Building Materials and Components 7, Vol. 1, Published by E&FN Spon, London, 1996, pp. 45-53. Miletid, S. and Ilid,M." Cement stone corrosion in sulphate environment. NTP VJ, 46, No. 3 (1996) pp. 23-31. Miletid, S., Ilid,M., Ranogajec,J. and Djurid,M." Sulphate corrosion of Portland cement and Portland cement mixed with fly ash and slag as a function of its composition. Proc. XVI Symposi. on Nordic Concrete Research, Helsinki, 1996, pp. 339-340. Ranogajec,J., Ilid,M., Miletid, S., Lazar,S. and Milinkovid,Lj." Effect of sulphate and ammonium ions on the cement stone corrosion. Proc. XX Congr. JUDIMK, Cetinje, 1996, pp. 97-102.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
FLY ASH AS THE BASIC MATERIAL BINDERS
365
FOR INORGANIC
PRODUCTION
M. Ili6, S. Mileti6, R. Djuri~,i6
IMS Institute for Materials Testing, Bul.vojv.MiYiCa 43, Belgrade, Yugoslavia
Abstract Among several other possibilities for the fly ash utilization in building materials industry and civil engineering (cement production, concrete production, blocks production etc.) another way to utilize this material is to make inorganic binder of it. This paper presents results of our investigation of lignite fly ash from thermopower plant for the cement and the inorganic binder production. Results are satisfying regarding quality requirements for Portland and masonry cement. This kind of materials could be produced with the fly ash as main or replacing raw material. Key words- Fly ash, deposit, building materials
1. Introduction For several years, organizations connected with the building industry have been involved in research on energy conservation in Portland cement concretes. These organizations have been encouraging the use of less energy-intensive materials, specifically pozzolans such as fly ash as admixture or partial replacement for the relatively more expensive Portland cement. In addition to volume/mass replacement, these pozzolans can react chemically with the calcium hydroxide of the Portland cement to produce CSH gel, which is cementitious and contributes to the strength increase of the Portland cement TM. Today, all of the world have one problem yet. This is a problem about solid waste, great number of landfill with solid waste and problems about minimization of waste and recycling and utilization, too. Thermo-power stations, which are using coal as the fuel, mostly have installed very efficient equipment for preventing the emission of solid particles (fly ash) to the atmosphere. Electrostatic precipitators with 3 to 5 fields collect about 99% of solid particles. For our country it is about 10,000,000 t per year of this material. Wetted material is then transfered to deposit places which presents very big ecological problem 5-7. By definition, fly ash is a fine powder of mainly spherical, glassy particles having pozzolanic properties and consisting essentially of SiO2 and A1203. Fly ash is obtained by electrostatic precipitation of dust-like particles from the flue gases of furnaces fired with pulverized coal. Deposit places are mostly very profesionally arranged and protected, but there are still deposit places which are unconvenient for such aim. The aim of our investigations was to establish procedure for deposits of waste fly and bottom ash investigation for their safe utilization in building materials production. Reaching this aim means reducing enormous ecology problem so as the getting equal quality products (cement, mortar, concrete, bricks, etc.) 89 ' . Our basic9 aim was
366 to establish procedure for safe utilization of deposited waste fly and bottom ash due to their chemical composition, radioactivity, unburnt carbon, demands. If it is about the utilization in cement production e.g., it should be noted that the activity of fly ash depends not only on its own properties, but on the physical and chemical properties of the cement employed, even within the same cement type. The fly ash should therefore be tested with the cement intended to be used in practice in mortar and concrete. Unless otherwise specified, an ordinary Portland cement should be used to test basic activity. For years the IMS Institute has been investigating possible reuse of wastes in the building materials production. Waste is unavoidable associate of power processes, chemical processes and industrial and mining operations. The biggest waste generators in our country are the thermo-power plants. Hundred of square kilometers of good soil are replaced with deposit places for fly and bottom ash. Performed investigations were based on geological testing, chemical testing, radioactivity, heavy metals content, content of impurities, physical characteristics and pozzolanic activity of mixture of deposited fly and bottom ash. Those investigations were pert'ormed on large number of deposits.
2. Materials and methods The following materials were used in our investigation: 1. Cement Cement used in this research conformed with the European specification EN 197-1 for common cements. 2. Gypsum Powdered gypsum used in this research was waste phosphogypsum conformed with the Yugoslav standard JUS B.C1.032, as the dihydrate gypsum. Gypsum was added in cement for regulation of setting time in amount of 3 % mass. 3. Mineral admixtures The mixture of fly and bottom ash, produced as the waste material and conformed with the Yugoslav standard JUS B.C1.018. was used as mineral admixture. The unit sample should be representative for the test purpose. The taking of each sample of at least 4 kg for complete testing is recommended. From this sample a laboratory sample of at least 1 kg is obtained by subdividing, such as quartering. Performed investigations were based on the following methods: 9 9 9 9 9 9 9
geological survey chemical testing radioactivity heavy metal content content of impurities physical characteristics pozzolanic activity All the testing were performed according to the following testing methods:
367 1. Determination of strength (EN 196-1) 2. Chemical analysis (EN 196-2) 3. Determination of setting time and soundness (EN 196-3) 4. Determination of fineness (EN 196-6)
3. Results and discussion
Results of geological investigations of the deposit are presented in Table 1.
Number of coreholes 17
TABLE 1 Geological investigations of the deposit Total core Average drilling Maximum drilling length, m length, m length, m 149 9.22 12.0
Minimum drilling length, m 6.5
It could be concluded that the deposit is covered with sufficient number of coreholes and enables the right evaluation of deposit in such manner to establish the opencast mining in proper way. Our investigations were based on chemical testing, radioactivity, heavy metals content, content of impurities, physical characteristics, and pozzolanic activity of fly ash. Those investigations were performed within very long period so the variation could be estimated of some characteristics in the function of time and place of origin. We performed laboratory, semi-industrial and industrial tests with this material as the mineral admixture for cement production too. Composition of fly ash from deposit is given in Table 2. TABLE 2 F1~r ash composition Composition Silicon dioxide (SiO 2) Aluminium oxide (A120~) Ferric oxide (FelOn) Calcium oxide (CaO) Magnesium oxide (M~O) Sulphur trioxide (SO~) Sodium oxide (Na20) Potassium oxide (K20) Loss on i~nition Insoluble residue
Mass percent 28.44-37.93 8.25-12.37 7.11-9.20 20.42-24.48 2.01-3.40 1.29-3.56 0.20-0.35 0.30-0.50 13.68-24.44 26.84-52.36
The results summarized in Table 2. indicate that the fly ash constituents up to 100 % varies but the main parameters mostly satisfy the requirements of JUS B.C1.018. Loss on ignition is out of limits given in mentioned standard but it is due to hydrated water bonded to ash particles and not to unburnt carbon, and so it could be used. It is evident from the chemical analysis of the fly ash that the variations of some constituents (A1203, CaO, LOI) are large, but it could be said that besides those variations, chemical composition satisfies the requirements of JUS B.C1.018 for the most important constituents. No contamination was evident.
368 Radioactivity of fly ash was, also, measured because it is very important by point of view of ecology and health, and this value is 0.398-0.520 Bq/kg. It satisfies the national limits. The study therefore suggests that partial replacement of cement by fly ash enables utilization of deposits as waste materials. Pozzolanic activity is determined and presented in Table 3. TABLE 3 Pozzolanic activity of fly ash Pozzolanic activity, Flexural strength Compressive strength
MPa 0.4-2.4 1.1-9.2
It is generally known that fly ash addition in cement results in lower initial strengths and with constant increasing in time. Activity of this kind of fly ash is rather low but it doesnt affect the basic characteristics of cement produced with the addition of 30 % fly ash. Chemical composition of cement produced with 30 % of fly ash are given in Table 4. TABLE 4 Chemical composition of cement with 30 .% of fly ash Composition , Mass percent SiQ 15.98 5.15 A1203 2.05 Fe203 CaO 43.11 Insoluble residue 23.69 1.25 Moisture at 105 ~ C Loss on ignition 3.31 0.74 CO 2 in CaCOz CaO free 0.12 CaO in CaCO~ 0.94 1.07 CaO in CaSO 4 1.53 SO~ in CaSO 4 CaS 0.00 MgO 1.01 Alkalies as Na20 0.28 K20 0.27 MnO 0.05 FeO 0.71 0.06 P205 C1_
The results indicated that the composition of cement satisfies the requirement of JUS EN 197-1 for common cements. The content of the minor oxides such as P205, MnO, MgO, alkalies is so small that they do not affect the cement properties. Physical-mechanical properties of same cement are given in Table 5.
369 TABLE 5 Physical-mechanical properties of cement with 30 % of fly ash Properties Unit Value % Sieve residue at 0.09 mm 4.40 Specific surface (by Blaine) 3790 Density g/cm 3.00 % Standard consistence 28.70 Setting time min -initial 165 -final 330 Soundness (Le Chatelier) mm 1.0
cm~/w
Properties Flexural strengths Compressive strengths Shrinkage Heat of hydration
[
Unit MPa MPa mm/m J/g
[
3 days 3.5 12.5 -0.019 166.0
7 days 5.1 20.9 -0.125 250.0
TABLE 6 Chemical composition of masonry cement Composition, % mass BM SiO 2 + Insoluble residue 19.70 A120~ 3.22 Fe~O~ 2.97 CaO 53.67 P20~ 0.06 Moisture at 105 ~ C 0.82 Loss on ignition 0.00 15.84 CO2 CaO flee 0.53 SO~ 1.77 MgO 1.51 K20 0.17 MnO 0.04 0.14 Na20 S 0.00 Alkalies as Na20 0.25 Water soluble sulphates SO~ 1.43 1.72 SO 4 Water soluble alkalies Na~O 0.14 K20 0.17 Alkalies as Na20 0.25
28 days 7.1 35.0 -0.737 -
90 days 8.1 41.1 -0.812
NM 21.97 2.98 1.52 48.87 0.05 0.89 0.00 19.96 0.86 2.14 1.61 0.07 0.04 0.10 0.00 0.15 1.20 1.44
0.04 0.06 0.08
370 The properties of cement, which are presented in Table 5. indicate that this cement exibits lower density, small increase in water demands, good strengths and low heat of hydration due to fly ash addition. In other words, this cement satisfies the requirements of JUS B.C1.013 for the cement with the low heat of hydration. Flexural and compressive strengths were measured at 3 days, 7 days, 28 days and 90 days according to the procedure laid down in JUS. Three specimens were tested at the end of each curing period.The results obtained for compressive and flexural strength make it possible to formulate a hypothesis as to the role played by the different percent of fly ash. Masonry cements (signed as B M and NM) produced with the fly ash as main constituent are presented by characteristics presented in Tables 6 and 7. Chemical characteristics shown in Table 6. are usual for such kind of material. Soluble constituents are rather low. TABLE 7 Physical-mechanical and mechanical properties of masonr~r cement Properties Unit ] BM NM % Sieve residue at 0.09 mm 0.50 1.64 Specific surface (by Blaine) 5511 6060 Density g/cm 2.92 2.81 % Standard consistence 26.3 24.7 Setting time min -initial 210 105 -final 315 225 Soundness (Le Chatelier) mm 0.0 0.0 % Entrained air content 14.0 17.5 % Water retention 79.5 89.6
cm~/w
Water / binder 0.50 0.533
Strength, MPa 7 days 28 days flexural compressive flexural compressive
Flow, mm
BM
NM
BM
NM
BM
NM
BM
NM
BM
NM
164 180
158
3.9 3.9
2.0 -
15.7 14.0
6.6 -
5.5 5.1
3.2
21.4 19,9
10.9 -
Physico-chemical and mechanical characteristics of both masonry cements presented in Table 7. are usual for such kind of material, although fly ash is used as main constituent. Both samples of masonry cements produced with fly ash as main constituent satisfies the requirements of Yugoslav standard JUS B.C1.010 for such kind of material. Based on presented results, the maps of quality of some characteristics and the project for opencast mining were made. Environment protection project for opencast mining and dispatch stations were also made, concerning measures for air and noise pollution. The production of above mentioned cements started few years ago very sucessfully.
371 4. Conclusion
Results of large scale investigations on various deposit places of waste mixtures of fly and bottom ash led us to following conclusions and procedure: 1. 2. 3. 4.
Classification of wastes geared to their possible reuses as input material. Basic geological examinations. Detailed geological examinations. Chemical, physical, mechanical testings and examination of possible harmful ingredients (radioactivity, heavy metals, etc.). 5. Monitoring of some properties in function of time or space namely origin. 6. Laboratory, semi-industrial and full scale tests in users plants. 7. Design of opencast mining. 8. Design of environment protection during opencast mining and dispatch. 9. Design of production divisions for preparing waste material to be used in some production. 10.Mining and dispatch. 11.Quality assurance and constant technical surveillance. In this way all possible negative consequences of inadequate use of waste material are eliminated. It must be said that each deposit have to be examined by different procedure due to differencies between deposit places and origin of fly and bottom ash. That was our conclusion based on very large scale of examination on deposit places. Based on very large number of investigations from which only small part are shown in this paper, it could be concluded that the variations in quality of fly ash are large enough, so the safe utilization of this material could be done with permanent testing and separation.
References
1. Mileti6, S., Stefanovir, M., Djuri~ir, R., Proc. WASCON'91, Confon Environmental Implications of Construction with Waste Materials, NOVEM, Maastricht, Netherlands, (1991), p.4. 2. Davis, R., ASTM Spec. Techn. Publ. No.99, p. 3. 3. Zivanovi6, B., Djoki6, S., Mileti6, S., Ka~arevir, Z., Tehnika, 5, Beograd, (1984), p.6. 4. Ramakrishnan, V., Coyle, W.V., Brown, J., Tlustus, P.A., Venkataramananujam, P., Effects of Fly Ash Incorporation in Cement and Concrete, Materials Research Society, Pittsburgh, (1981), pp. 233-245. 5. Popovi6, K., Dimic D., Kameni6, N., Krstulovi6, P., Miletid, S. Se~un, Lj., Proc.RILEM Intern.Symp.Test Quality and Quality Assurance in Testing Laboratories for Construction Materials and Structures, Paris, France, (1989), p.5. 6. Dragirevir, Lj., Zivanovi6, B., Miletir, S., Zbornik radova, I Jugoslovenski simpozijum o keramici, II, SHD, JUDIMK, Beograd, (1981), p.5. 7. Berry, E.E., Malhotra, V.M., ACI Journal, Proceedings Vol.77 (2), (1980), p. 59 8. Pearson-Gallovay, "Civil Engineering", N.Y.7/8, 1950. 9. Sersale, R., Proc. of 7th Inter.Congress on Chemistry of Cement, Paris, Vol. I, IV-l/3, (1980)
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
373
A study of the potential of utilising electric arc furnace slag as filling material in concrete Catharina B~iverman and Francisca Aran Aran 1 Department of Chemical Engineering and Technology Royal Institute of Technology Stockholm, Sweden Abstract Sand in concrete has been substituted by steel slag from an electric arc furnace. The physical properties investigated were break load and compressive strength. The two materials showed similar results. The leaching properties of the materials were also investigated to study whether the steel slag concrete could be acceptable from an environmental point of view. The leaching properties of both materials showed similar results, except for chromium leaching. The release rate of chromium has been calculated, and the results show that the chromium leaching should be no problem.
Introduction and background Large quantities of waste material are produced annually in the steel industry. The use of these secondary materials in civil engineering applications is of great interest nowadays, both to decrease the amounts landfilled and to replace natural aggregates like sand and gravel. The utilisation of these materials is only interesting if there are no environmental effects caused by the substitution. Recycled scrap iron generates a slag with a high metal content. The slag used in this study comes from a scrapmetal-based steel industry, and is an electric arc furnace (EAF) slag. EAF slag is presumed not to have any binding properties useful in concrete production, and can, therefore, be used to replace natural aggregate in concrete. The aim of the study was to investigate whether steel slag can be used in concrete as filling material to substitute sand. The influence of steel slag in concrete was investigated and compared to normal concrete. The steel slag concrete was also compared to steel slag to study the influence on leaching of cement in contact with the slag.
Material and Methods The slag used in this study was an EAF slag from Fundia Steel AB, Sweden. The slag was crushed and sieved to give a particle size distribution similar to that of natural sand.Two types of concrete were prepared: a normal concrete with natural sand as aggregate, and a steel slag concrete with steel slag as aggregate. The preparation technique was the standard technique for preparing concrete specimens. The specimens were kept in a moisture chamber at 100% humidity for 7 days, the first day in the mould, and at 50% humidity for 21 days.
Physical properties The physical properties investigated were compressive strength and breaking load in order to investigate whether steel slag concrete could make a good product. The steel slag concrete had a higher density than the normal concrete. The compressive strengths and breaking loads of the two different types of concrete were similar, and they can both be considered as medium strength concrete. The steel slag concrete was, however, more brittle than the normal concrete.
Leaching tests The leaching properties of the steel slag concrete were studied to see whether the material is acceptable from an environmental point of view. Batch leaching tests were performed on crushed material with a particle size less than 0.16 mm, at constant pH, 9.5, 10.5, 11.5, 12.5 and 13.5, at liquid-to-solid ratio of 5 and at pH 9.5, 11.5 and 13.5 at a L/S ratio of 100. Samples were taken after one and seven days. A second series of experiments performed on the steel slag concrete and on the normal concrete involved batch leaching tests using slabs. This test can be used to study the diffusion, from deeper portions of the sample. Similar tests have earlier been used for diffusion measurements in granite (~. Concrete slabs, 6*40*40 mm, were made. Ten slabs were put in a Teflon holder and placed in a vessel, see figure 1. A volume of 300 ml of water was added to the vessel. The leachate pH was kept at 9.5, 11.5 or 13.5 by Resident in Barcelona, Spain.
374 the addition of nitric acid or sodium hydroxide. The vessels were closed and samples were taken after 1, 2, 4, 8, 16 and 32 days.
Figure 1.
Ten slabs of concrete were put in a Teflon holder and placed in a vessel.
Analyses The samples were analysed with ion chromatography. The ion chromatography system consists of a Dionex model DX-300, with both suppressed conductivity detection and post-column reaction with UV/VIS-detection. The content of Ca, K, Mg, Mn, S, Co, Cr, Cu, Zn, C1, Pb, Ni were determined.
Results and discussion
Crushed sample The natural pH of the leach water in contact with the samples was the same for the two concrete samples, whereas it was one pH unit lower for pure steel slag. This shows that the final pH of the leachate is controlled by the cement and that the substitution did not considerably affect the final pH. As the pH in our experiments was manually adjusted by the addition of acid, we observed that it was difficult to keep a constant pH, due to the high pH buffering capacity of the materials. This problem did not occur when the pH was kept at a higher pH than the initial pH. The buffering reactions are probably the dissolution of oxides, mainly lime (CaO). This effect was the same for both steel slag concrete and normal concrete. Chromium was the only element of those analysed for which the leach pattern differed significantly between steel slag concrete and normal concrete. The chromium concentration in the batch experiments is shown versus pH in figure 2. The results show that the chromium leaching from the steel slag concrete follows the leach pattern for steel slag at high pH, above 12, but that at a pH below 12 the leaching from the steel slag concrete is higher than that from both the steel slag and the normal concrete. The minimum chromium leaching from the steel slag concrete was observed at the natural pH of the material. The concentration at p is in the same range as that of the maximum level for Swedish drinking water, which is 0.9 ~M 12 Stage-A represents the initial alteration processes which take place when the dry bottom ash first contacts water, which is in the quench tank. Reactions include the hydrolysis of the oxides of Ca, A1, Na, and K, and the dissolution/reprecipitation of hydroxides and salts of these main cations [7,25,30]. The resulting bottom ash pH is strongly alkaline (12.4) and controlled by the solubility of portlandite (Ca(OH)2) [7]. (B) quenched/non-carbonated bottom ash, with pill 0-10.5 In stage B bottom ash pH has been decreased to 10-10.5 by the formation of ettringite (Ca6A12(SOa)3(OH)12.26H20), gibbsite (AI(OH)3), and gypsum (CaSOn.2H20) [7,25]. When the three minerals coexist, no degrees of freedom are left and pH is fixed at pH 10 [7,25]. Due to continuing hydrolysis secondary minerals such as amorphous Fe/Al-(hydr)oxides, hydrous aluminosilicates, and possibly zeolites begin to precipitate [4,5,7]. Soluble salts will be leached rapidly with percolating water [e.g. 4,5,7,11,30]. Biodegradation of residual organic matter and dissolution of reduced mineral phases may create a reducing environment [5,18]. (C) carbonated bottom ash, with pH8-8.5 In stage C bottom ash pH has further decreased to equilibrium values of 8-8.5 by absorption of CO2 and subsequent precipitation of calcite (CaCO3) [e.g. 5,7,8]. The CO2 required for this carbonation may infiltrate from the atmosphere or come from biodegradation of organic residues [5,26,30]. The neoformation of Fe/Al-(hydr)oxides and hydrous aluminosilicates continues. Similar to the weathering of volcanic ashes, these hydrous aluminosilicates are an intermediate reaction product in the transformation of glasses to clayminerals [6]. The 2:1 clay mineral illite seems to be the final product of glass weathering in MSWI bottom ash [6]. Weathering has been shown to have a significant effect on the leaching of trace elements from MSWI bottom ash [8,19]. The leaching of Cd, Pb, Cu, Zn, and Mo from C-type bottom ash, for example, is generally significantly lower than from more fresh bottom ash [8,19]. A potentially important mechanism is the sorption of trace elements to neoformed (amorphous) Fe/Al-minerals [8,19,20,29]. Furthermore, the neutralisation of bottom ash pH from > 10 to 8-8.5 and the formation of less soluble secondary minerals of trace-elements also contribute to reduce leaching [ 19]. Lower trace-element leaching from weathered bottom ash does not seem to be caused primarily by a prior release of these elements from the residues during storage [ 19,31 ].
4. LEACHING The rate at which an element is leached from the bottom ash is dependent on its abundance in the bottom ash, its availability to the solution, the dissolution kinetics of the primary solids containing the element, whether or not the element will reprecipitate as a secondary solid or will sorb to solid substrates, and the kinetics of these precipitation/sorption reactions [13]. Kirby and Rimstidt [13] have identified 3 basic types of solution behaviour duringbatch leaching of MSWI bottom ash: 1. availability, which means that there is a lack of concentration-change due to exhaustion of a phase. This type of behaviour is usually observed for soluble salts, such as Na, K, and C1 [5,7,13,18]. Furthermore, molybdenum may show this type of behaviour at strongly alkaline pH [7,18]. In general, the higher the Liquid to Solid (L/S) ratio, the more elements will show this type of behaviour.
452 2. kinetic, which means that the rate of mass transfer from the solid to the liquid phase or v.v. is the concentration-limiting step. The contact time between the solid and the liquid phase usually determines whether kinetics are important or not. In general, two steps can be observed in element leaching from bottom ash: a fast release of the element, which is generally completed within 24 h, followed by a slow release or re-binding which may continue for more than 1 week [5,8,13]. The leaching of silicon, for example, is strongly influenced by slow dissolution/precipitation kinetics of silicate-minerals [5,8,13]. Furthermore, the slow transformation of the primary high-temperature solids into stable secondary solids has been shown to affect the leaching of several other elements as well [7,19]. Little is known, however, about the kinetics of these weathering reactions. Alternatively, a slow release may also be the result of diffusion processes, which are believed to become important when the residues are monolithic in form (e.g. incorporated into asphalt pavement), when they are compacted to low permeability, or when they are overlain by an impermeable barrier [ 18,32]. 3. equilibrium, which means that the concentration of an element is controlled by a dissolution/precipitation equilibrium or by a sorption equilibrium. Various elements experience retention in the bottom ash matrix by these processes: Table 2 gives an overview of the proposed controlling-mechanisms for MSWI bottom ash at different stages of weathering. Below, we review underlying geochemical processes, such as complex formation, dissolution/precipitation, sorption and redox reactions, which control element leaching from MSWI bottom ash.
5. G E O C H E M I C A L PROCESSES C O N T R O L L I N G LEACHING
complexation processes Hydrolysis and complexation with carbonate are the dominant inorganic complexation reactions in bottom ash leachates. These reactions cause, for example, the solubility-curves of amphoteric elements such as Fe, A1, Zn, Cu, and Pb to follow V-shaped patterns as a function of pH [33]. Figure 2 illustrates the effect of hydrolysis on the solubility of Zn. As a result, pH is a dominant controlling parameter in element leaching from (waste) materials, which is in correspondence with experimental data [e.g. 7,12,16,19,30,34,35]. Other potentially important inorganic complexes include Cd-C1 complexes, which may become significant in leachates from fresh bottom ash [16]. MSWI bottom ash releases substantial amounts of dissolved organic carbon (DOC) originating from incomplete burning of the original waste and/or subsequent biodegradation processes [7]. Copper, which is known to have a very high affinity for organic material [36], has been shown to be bound for >90% to DOC in leachates of both fresh and 1.5-year old MSWI bottom ash [21 ]. The conditional stability constants of these Cu-DOC complexes have been determined using a competitive ligand exchange / solvent extraction technique [21]. Figure 3 illustrates the importance of this organic complexation on the leaching of Cu from fresh MSWI bottom ash, and shows that Cu-leaching under environmental conditions (pH>7) is dominated by this process.
453
1E-02 _1
g
c N
1E-03
-
1E-04
-
1E-05
-
1E-06
-
O
~
o
. o
....
Zn(;~,:
1E-07
I "
"
o.: "
I
5
'
-
7
"
- 1
"
I
".o ~i
,"
8' I
I
9
"z"(OH)/I
11
13
pH Figure 2. The effect of inorganic complexation on the solubility of zinc. The solid line represents the predicted total concentration of zinc in equilibrium with the mineral zincite (ZnO). Dashed lines represent concentrations of corresponding Zn-species. Symbols represent total dissolved Zn in type B bottom ash leachates at L/S=5. Data were taken from Ref. 19.
1 E-04
O
'
~Total
E-O5
r~ 1 E - 0 6
Cu
I1~,,
-
c u'§
1E-07
"
Cu-0rg
-
"
-
- ~ ----
" ~ . Cu-inorg ,
6
-
0
7
~
,"
,
,
8
9
10
11
pH Figure 3. The effect of organic complexation on the leaching of Cu (O) from fresh MSWI bottom ash at L/S=5. The solid line represents the predicted total concentration of Cu in equilibrium with the mineral tenorite (CuO). Dashed lines represent corresponding concentrations of Cu 2§ the sum of the inorganic Cu-complexes, and the sum of the organic Cu-complexes. (Modified after Ref. 21.)
precipitation/dissolution processes
Precipitation/dissolution processes control bottom ash pH (see above) and the leaching of in particular major elements from MSWI bottom ash (Table 2, Figure 1). In the case of major elements, solubility-controlling minerals indicated by geochemical modelling generally correspond to minerals detected by spectroscopic analysis of the bottom ash (Table 2). Precipitation/dissolution processes may also control the leaching of trace elements from Aand B-type bottom ash. Proposed controlling processes for trace-element leaching, however, are often indicated by geochemical modelling only (Table 2) because low bulk concentrations hamper the detection of trace-element species by means of spectroscopic techniques [2,19]. A
454 step-wise approach for the geochemical modelling of element-concentrations in equilibrium with potential solubility-controlling minerals is given in Meima and Comans [7].
sorption processes Sorption is a general term which refers to all processes, except the precipitation/dissolution of pure mineral phases, which remove a chemical species from the aqueous solution to a solid phase. Sorption processes are expected to be important when suspensions at equilibrium are undersaturated with respect to known solubility-controlling minerals. Potential sorbent minerals in MSWI bottom ash are amorphous or crystalline Feand Al-(hydr)oxides, hydrous aluminosilicates, and calcite [8,20]. Recent studies have shown that surface complexation reactions can successfully describe the leaching of trace-elements from combustion residues, such as MSWI bottom ash and coal fly ash [9,20,37-39]. In addition, trace-elements have been found to be associated with secondary and potential sorbent minerals in weathered MSWI bottom ash (Table 2). A step-wise approach for the modelling of surface complexation or surface precipitation processes in heterogeneous systems such as MSWI bottom ash is described by Meima and Comans [20]. This approach is based on (1) the database of surface complexation and surface precipitation reactions and associated equilibrium constants for sorption of ions on Hydrous Ferric Oxide [40], (2) 'selective' chemical extractions to obtain the available sorbent mineral concentrations, and (3) leaching of the bottom ash at pH-values unfavourable for sorption to obtain the available trace-element concentrations. The identification and modelling of sorption processes in heterogeneous solid systems such as MSWI bottom ash is, however, at its beginning. Because of their potential importance, these processes deserve considerably more attention in future research.
redox processes In fresh MSWI bottom ash the prevailing redox conditions are oxidizing [7,11]. During disposal or utilization of the bottom ash, however, the redox potential may decrease strongly by biodegradation of residual organic matter and/or by the presence of reduced mineral phases [5,18]. Relatively low redox potentials were recorded, for example, in percolate from landfilled combined MSWI bottom and fly ash [41] and in a 6-week old storage of fresh MSWI bottom ash [7]. Variations in bottom ash EH may affect metal mobilities by: 9 directly changing the oxidation states of redox sensitive elements to more soluble/insoluble species. The leaching of Cu [18,35,42,43], Cr [18,43], As [43], and V [43], for example, has been shown to increase toward more oxidizing conditions, whereas the leaching of Fe was decreased [7,35,43]. 9 changing the amount of redox sensitive metal surfaces (Fe/Mn-(hydr)oxides) available for sorption [35]. 9 changing the degree of (co)-precipitation or complexation with other redox sensitive cations and anions, e.g. the precipitation of heavy-metal sulphides [ 18,35,41,43]. The cited studies show that the influence of EH on metal solubilities in MSWI bottom can be significant and that further research on this topic is required.
455
CONCLUSIONS AND RECOMMENDATIONS
F O R F U T U R E RESEARCH
The leaching of major and trace elements from M S W I bottom ash has successfully been described on the basis of geochemical processes such as complexation, precipitation/dissolution, and sorption processes. For the prediction of the long-term behaviour of M S W I bottom ash in the environment the results imply that: 9 materials should be tested at pH 10 and/or pH 8.3, depending on whether or not the materials are used in contact with air and may become carbonated; 9 the concentrations of toxic elements in leachates are likely to be greatest in the earliest stages of disposal: the most soluble phases dissolve rapidly, while the capacity of secondary minerals to bind trace elements may not be large enough. Furthermore, m o l y b d e n u m is very mobile at alkaline pH. These problems may be overcome by (1) neutralising the pH of the bottom ash, and (2) by adding sorbent minerals to the bottom ash [44]. 9 on the long-term, the leaching of toxic elements is likely to be reduced by the neutralisation of bottom ash pH and by sorption to neoformed minerals. Future research should concentrate (a) on a further identification/modelling of these sorption processes and (b) on the kinetics of these weathering/sorption reactions, i.e. the period of time that is required to obtain a sufficient reduction in trace-element leaching. 9 Little is known about the processes that control element leaching from M S W I bottom ash in reducing environments, which, therefore, also requires further research.
REFERENCES 1. 2. 3. 4. 5.
6. 7. 8. 9. 10. 11.
Eary L.E., Rai D., Mattigod S.V., and Ainsworth C.C. (1990) Geochemical factors controlling the mobilization of inorganic constituents from fossil fuel combustion residues: II. Review of the minor elements. J. Environ. Qual. 19, 202-214. Kirby C.S. and Rimstidt J.D. (1993) Mineralogy and surface properties of municipal solid waste ash. Environ. Sci. Technol. 27, 652-660. Mattigod S.V., Rai D., Eary L.E., and Ainsworth, C.C. (1990) Geochemical factors controlling the mobilization of inorganic constituents from fossil fuel combustion residues: I. Review of the major elements. J. Environ. Qual. 19, 188-201. Kirby C.S. A Geochemical analysis of municipal solid waste ash, Ph.D. Dissertation, Virginia Polytechnic Institute and State University, 1993. Zevenbergen C. and Comans R.N.J. (1994) Geochemical factors controlling the mobilization of major elements during weathering of MSWI bottom ash. In Environmental Aspects of Construction with Waste Materials (Eds J.J.J.M. Goumans, H.A. van der Sloot, and Th. G. Aalbers), pp. 179-194. Elsevier Science B.V., Amsterdam. Zevenbergen C., van Reeuwijk L.P., Bradley J.P., Bloemen P., and Comans R.N.J. (1996) Mechanism and conditions of clay formation during natural weathering of MSWI bottom ash. Clays and Clay Minerals 44, 546-552. Meima J.A. and Comans R.N.J. (1997) Geochemical modelling of weathering reactions in municipal solid waste incinerator bottom ash. Environ. Sci. Technol. 31, 1269-1276. Meima J.A., van der Weijden R.D., Eighmy T.T., and Comans R.N.J. The effect of carbonation on traceelement leaching from municipal solid waste incinerator bottom ash. Submitted for publication. Theis T.L. and Richter R.O. (1979) Chemical speciation of heavy metals in power plant ash pond leachate. Environ. Sci. Technol. 13, 219-224. Fruchter J.S., Rai D., and Zachara J.M. (1990) Identification of solubility-controlling solid phases in a large fly ash field lysimeter. Environ. Sci. Technol. 24, 1173-1179. Theis T.L. and Gardner K.H. (1992) Dynamic evaluation of municipal waste combustion ash leachate. In: 5th International Conference on Ash Management and Utilization. pp27-65. Arlington, VA.
456
12.
13. 14.
15.
16.
17. 18.
19. 20. 21.
22. 23.
24.
25.
26. 27.
28.
29.
30. 31. 32.
33. 34.
Comans R.N.J., van der Sloot H.A., and Bonouvrie P.A. (1993) Geochemical reactions controlling the solubility of major and trace elements during leaching of municipal solid waste incinerator residues. In Municipal Waste Combustion Conference., pp. 667-679. Air and Waste Management Association, Williamsburg, VA. Kirby C.S. and Rimstidt J.D. (1994) Interaction of municipal solid waste ash with water. Environ. Sci. Technol. 28, 443-451. Eighmy T.T., Eusden jr. J.D., Marsella K., Hogan J., Domingo D., Krzanowski J.E., and St~npfli. D. (1994) Particle petrogenesis and speciation of elements in MSW incineration bottom ashes. In Environmental Aspects of Construction with Waste Materials (Eds J.J.J.M. Goumans, H.A. van der Sloot, and Th. G. Aalbers), pp. 111-136. Elsevier Science B.V., Amsterdam. Eighmy T.T., Eusden Jr J.D., Krzanowski J.E., Domingo D.S., St/impfli D., Martin J.R., and Erickson P.M. (1995) Comprehensive approach toward understanding element speciation and leaching behavior in municipal solid waste incineration electrostatic precipitator ash. Environ. Sci. Technol. 29, 629-646. van der Sloot H.A., Comans R.N.J., and Hjelmar O. (1996) Similarities in the leaching behaviour of trace contaminants from waste, stabilized waste, construction materials and soils. The Science of the Total Environment 178, 111-126. Van der Sloot H.A. (1996) Developments in evaluating environmental impact from utilization of bulk inert wastes using laboratory leaching tests and field verification. Waste Management, 16, 65-81. Chandler A.J., Eighmy T.T., Hartl6n J., Hjelmar O., Kosson D.S., Sawell S.E., van der Sloot H.A., and Vehlow J. (1997) Municipal solid waste incinerator residues. In: Studies in Environmental Science, 67, Elsevier: Amsterdam, The Netherlands. Meima J.A. and Comans R.N.J. The leaching of trace-elements from municipal solid waste bottom ash at different stages of weathering. Submitted for publication. Meima J.A. and Comans R.N.J. Application of surface complexation/precipitation modelling to contaminant leaching from weathered MSWI bottom ash. Submitted for publication. Meima J.A., van Zomeren A., and Comans R.N.J. The complexation of Cu with dissolved organic carbon in leachates from municipal solid waste incinerator bottom ash; determination of conditional stability constants. (manuscript in preparation) Theis T.L. and Gardner K.H. (1990) Environmental assessment of ash disposal. Crit. Rev. Environ. Control. 20, 21-42. Zevenbergen C., Vander Wood T., Bradley J.P., Van der Broekck P.F.C.W., Orbons A.J. and Van Reeuwijk L.P. (1994) Morphological and chemical properties of MSWI bottom ash with respect to the glassy constituents. Hazard. Waste Hazard. Mater. 11,371-382. Eusden Jr. J.D., Holland E.A., and Eighmy T.T. (1994) Petrology, bulk mineralogy, and melt structure of MSW bottom ash from the WASTE program. In Proceedings of the 16th annual Canadian Waste Management Conference, Calgary. Comans R.N.J. and Meima J.A. (1994) Modelling Ca-solubility in MSWI bottom ash leachates. In Environmental Aspects of Construction with Waste Materials (Eds J.J.J.M. Goumans, H.A. van der Sloot, and Th. G. Aalbers), pp. 103-110. Elsevier Science B.V., Amsterdam. Johnson C.A., Brandenberger S., and Baccini P. (1995) Acid neutralizing capacity of municipal waste incinerator bottom ash. Environ. Sci. Technol. 29, 142-147. Pfrang-Stotz G. and Schneider J. (1995) Comparative studies of waste incineration bottom ashes from various grate and firing systems, conducted with respect to mineralogical and geochemical methods of examination. Waste Management & Research 13,273-292. Johnson C.A., Kersten M., Ziegler F., and Moor H.C. (1996) Leaching behaviour and solubilitycontrolling solid phases of heavy metals in municipal solid waste incinerator ash. Waste Management 16, 129-134. Zevenbergen C., Bradley J.P., Van der Wood T., Brown R.S., Van Reeuwijk L.P., and Schuiling R.D. (1994) Natural weathering of MSWI bottom ash in a disposal environment. Microbeam analysis 3, 125135. Belevi H., St~impfli D.M., and Baccini P. (1992) Chemical behaviour of municipal solid waste incinerator bottom ash in monofills. Waste Management & Research 10, 153-167. Stegemann J.A., Schneider J., Baetz B.W., and Murphy K.L. (1995) Lysimeter washing of MSW incinerator bottom ash. Waste Management & Research 13, 149-165. Kosson D.S., van der Sloot H.A., and Eighmy T.T. (1996) An approach for estimation of contaminant release during utilization and disposal of municipal waste combustion residues. J. of Hazard. Mater. 47, 43-75. Stumm W. and Morgan J.J. (1981) Aquatic Chemistry (2nd edn). John Wiley, New York. Theis T.L. and Wirth J.L. (1977) Sorptive behavior of trace metals on fly ash in aqueous systems. Environ. Sci. Technol. 11, 1096-1100.
457
35.
36. 37. 38. 39. 40. 41. 42.
43.
44.
DiPietro J.V., Collins M.R., Guay M., and Eighmy T.T. (1989) Evaluation of pH and oxidation-reduction potential on leachability of municipal solid waste incinerator residues. In International Conference on Municipal Waste Combustion, pp. 2B.21-43. U.S. EPA and Environment Canada, Hollywood. FL. Buffle J. (1988) Complexation reactions in aquatic systems: an analitical approach; Ellis Horwood Series in Analitical Chemistry; Ellis Horwood: Chichester. Dzombak D. and Morel F. (1992) Modeling the leaching of metals from hazardous waste incineration ash. Proceedings of the Incineration conference; Albuquerque, New Mexico. Kersten M., Moor C., and Johnson C.A. (1995) Emissionspotential einer milllverbrennungsschlackenmonodeponie ftir schwermetalle. Mtill and Abfall, 11,748-758. Van der Hoek E.E. and Comans R.N.J. (1996) Modeling arsenic and selenium leaching from acidic fly ash by sorption on iron (hydr)oxide in the fly ash matrix. Environ. Sci. Technol. 30, 517-523. Dzombak D.A. and Morel F.M.M. (1990) Surface Complexation Modeling: Hydrous Ferric Oxide. John Wiley & Sons, New York. Hjelmar O. (1989) Characterization of leachate from landfilled MSWI ash. In International Conference on Municipal Waste Combustion; U.S. EPA and Environment Canada: Hollywood, FL, pp 3B. 1-19. Van der Sloot H.A., Hoede D., and Comans R.N.J. (1994) The influence of reducing properties on leaching of elements from waste materials and construction materials. In Environmental Aspects of Construction with Waste Materials (Eds J.J.J.M. Goumans, H.A. van der Sloot, and Th. G. Aalbers), pp. 483-490. Elsevier Science B.V., Amsterdam. F/~llman A-M and Hartl6n J. (1994) Leaching of slags and ashes - controlling factors in field experiments versus in laboratory tests. In Environmental Aspects of Construction with Waste Materials (Eds J.J.J.M. Goumans, H.A. van der Sloot, and Th. G. Aalbers), pp. 39-54. Elsevier Science B.V., Amsterdam. Comans R.N.J., Meima J.A., and Geelhoed P.A. (1997) Development of a technology to reduce the leaching of contaminants from MSWI bottom ash by the addition of sorbing components. Presentation at WASCON 1997 and submitted for publication.
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
459
HEAVY METAL BINDING MECHANISMS IN CEMENT-BASED WASTE MATERIALS
Christian Ludwig, Felix Ziegler and C. Annette Johnson Swiss FederalInstituteof EnvironmentalScience and Technology(EAWAG), CH-8600 DObendorf,Switzerland ABSTRACT: Field and laboratory experiments were carried out to elucidate the geochemical and hydrological mechanisms that are important to understand the binding mechanisms of heavy metals in landfills with cement based waste materials. The focus of the work was on Zn(II), firstly in sorption experiments with calcium-silicate-hydrate, and secondly as a component in the leachate from a field lysimeter experiment. The leachate of the lysimeter containing cemented electrofilter ashes was sampled during rain events in order to determine the leaching processes. It was found that most of the rainwater was in intimate contact with the waste material in this field system and that while hydrological factors caused changes in concentrations of up to 100% (8-161xM), the concentration range was controlled by geochemical factors. The residence time of the water was sufficiently long to be able to describe Zn concentrations by thermodynamic calculations. The pH values in the leachate ranged between 12.5 and 13.1 where Zn2SiO4(s) appeared to be the most stable phase. Comparison with laboratory experiments suggested that alternative mechanisms could be important. In the laboratory experiments Zn appeared to be incorporated into the calcium-silicatehydrate frame forming solid solutions that have varying solubilities depending on the Ca/Zn ratio in the Cal.x-Znx-silicatehydrates. The field data agreed well with this alternative model. Thermodynamic and kinetic factors are discussed and compared with respect to the geochemical and hydrological contributions.
INTRODUCTION Today, over 80% of Switzerland's municipal waste is incinerated. The heavy metal-rich flue gases produced during incineration are scrubbed with electrostatic precipitators followed by aqueous washing treatments to remove acidic gases and potentially harmful heavy metals including mercury. The resulting electroprecipitator ash and the solid residues of the aqueous treatment are mixed. This filter ash (FA) is predominantly inorganic and is rich in heavy metals ~, and therefore, has to be handled as hazardous waste 2. Though altemative processes designed to quantitatively separate heavy metals waste residues are in development 3'4, landfilling is at present a common practice. In Switzerland, FA is mixed with cement before it is disposed of in mono landfills. Laboratory tests have shown a reduction in the heavy metal concentration in the leachate of cement-stabilized FA by factors of 5 to 205. The mechanisms which reduce the mobility are still a matter of debate. Thus, the long-term leachability of the cemented FA cannot be predicted. In order to make such predictions, it is necessary to understand the geochemistry and hydrology of these systems. The hydrology of a landfill determines whether and how long water can come into contact with the solid material. The geochemical processes determine the reactions between the solid phase and the leachate. Depending on the composition of the landfill and the residence time of the water in contact with the solid phases, concentration of elements are kinetically or thermodynamically controlled. In recent work, Johnson et al. 6 and Kersten 7 have shown the usefulness of thermodynamic calculations for the interpretation of the heavy metal concentrations in the leachates of a landfill with municipal solid waste incinerator ash. However, kinetic factors play an important role for slow geochemical processes that cannot be estimated by thermodynamic calculations. For kinetically-controlled geochemical processes, the product concentrations may scale with the reactivity of the different reactants. Here, the known ligand exchange rates around hydrated metals can be used to estimate the reactivities, e.g. for dissolution 8'9 or adsorptionl~ processes. For our field studies, we have chosen a pilot landfill that was built for scientific research purposes 5'11. The advantage of this site is that the landfill is completely filled and that CO2 contamination of the leachate can be avoided. Of special interest was the investigation of the effect of preferential flow during rain events upon the concentrations of the dissolved cations and anions in the leachates. Our field studies were accompanied by laboratory experiments and theoretical investigations about the geochemical reactions of importance. In this paper we have chosen Zn(II) as an example for a trace
460 element. Cycling of Zn is of major interest due to its high concentration in the FA. Zn is also a suitable element for laboratory experiments because it is highly soluble and allows experiments above the detection limits of common methods of analysis. EXPERIMENTAL Materials. CaCO3 (p.a.), ZnC12 (p.a.), KC1 (p.a.), CaC12 (s.p.), NaC1 (s.p.), NaOH and HC1 titrisol at various concentrations, Si standard solution (1000ppm SiCI4 in 5M NaOH) and concentrated HNO3 (s.p.) were obtained from Merck. SiO2 (Aerosil 300) was purchased from Degussa. All solutions were prepared from 17Mg2 ultrapure water (Barnstead Nanopur) which was filtered through a 0.2 ~m in-line filter. For the laboratory experiments the ultrapure water was boiled under Ar. HDPE-flasks for the field sampling and for the sorption experiments were leached with acid (--0.6M diluted from concentrated HNO3). The pilot landfill. The lysimeter is located next to the old landfill "Teuftal" in Mfihleberg (Kanton Bern) and was constructed TM for scientific research. The landfill contains cemented FA in form of cubic blocks that have an edge length of 0.5m. The plant is approximately 1.5m deep, has a surface area of 16m 2, and is covered with clay-silt (0.2m), gravel (0.8m), and humus (0.3m) layers. Additional installations at the existing sampling station were made to prevent the samples from CO2 contamination and to prevent the drainage solutions from blocking the tubing. Sampling and field measurements. Only under wet conditions did we find enough discharge for sampling. A measuring cell with a rotating stirrer and a cell volume of 0.04dm 3 was connected to the drainage outlet to measure temperature, conductivity and pH with a testo 252 field equipment. It was not possible to perform on-line pH measurements because pH-electrodes become unstable in basic solutions. The pH-electrodes were calibrated using Merck titrisol buffer solutions (7, 10, and 13). The conductivity cell was checked with KC1 solutions. The conductivities and pH values were corrected for the temperature at 25~ The temperature corrections for the conductivity measurements were based on leachate samples. Concentrated samples were diluted and the conductivities were measured at different temperatures between 6 and 27~ The established framework was used to interpolate. HDPE-flasks (0.25dm 3) were filled with sample, sealed thightly, and stored for further investigations at about 1012~ which is close to the sampling temperature, to reduce the possibility of precipitation. Sample analysis. A1, K, Na, Si, and Zn were measured with ICP-OES (Spectro, Spectroflame). The samples were diluted with acid by a factor of 5 to give a pH value of about 2. Diluted samples were analyzed for SO42- and C1 using an IC (Sykam) equipment with a Sykam (A04) column. CO32- was measured without pretreating the sample solutions using a TOC (Shimazu 5050) analyzer. The samples were sealed before the measurements to minimize CO2 contamination. Laboratory experiments. The experiments were carried out in a glove box under Ar. Calciumsilicate-hydrates (C-S-H) with a Ca/Si ratio of about 1 was synthesized after Atkins et. al ~2 by mixing 12.11 g of CaO (prepared by heating CaCO3 at 900~ for 24 hours) with 12.89g SiO2 and suspended in 0.5dm 3 water in a l dm 3 HDPE bottle. This suspension was shaken for 7 days on a rotary shaker (Btihler, Swip SM 25) at 150rpm. The suspension was then centrifuged (10 minutes, 6000rpm). The separated solid was vacuum-dried. X-ray powder diffraction spectras of the product were in agreement with the spectra of C-S-H as obtained by Taylor 13. Sorption experiments were performed in presaturated solutions (S1) with respect to C-S-H with a composition of [Si(IV)]=0.1mM, [Ca(II)]=3mM, [OH-]=8.3mM, [C1-]=0.1M, [Na+]=0.1M resulting in a pH value of 11.7. For the Zn experiments a stock suspension ($2) was prepared by adding 1g C-S-H to 0.5dm 3 of the presaturated solution and equilibrated for 7 days on the rotary shaker at 150rpm. Then, 1cm 3 of $2 was added to 50cm 3 S 1 and was equilibrated, again for 7 days. Afterwards aliquots of a Zn stock solution was added to obtain final Zn concentrations of 0.96, 0.48, 0.19, 0.096, 0.048, 0.019, and 0.0048mM. The suspensions were equilibrated for 4, 28, 53 or 87 days. In a withdrawn sample the pH value was measured using a combined glass electrode (Metrohm 6.0202.100). The remaining sample
461
was filtered (0.451am nylon, Whatman) and acidified with 0.3cm 3 of concentrated HNO3. Zn concentrations below 0.1ppm were measured by anodic stripping voltammetry (DP-ASV, Metrohm VA-Stand 694, VA-Processor 693) and above this value with AAS (Perkin-Elmer 5000).
THEORETICAL ASPECTS Heavy metal solubility is controlled by a spectrum of very slow to fast geochemical processes. Most important are sorption, dissolution, and precipitation reactions at mineral surfaces and diffusion and transformation reactions in the solid phases. Reactions in solution are generally fast. The different processes can be classified in order of their rates as follows: SOLID transformation ~ diffusion
12 mol/kg, the amount of basic salts is insufficient to neutralise the added acid giving a logarithmic decrease in the pH.
L 8 -
6-4-
,o
e
a 9176 ** ~
2-
.
00,
L 2
L 4
~ 6
t
t
~
,
~
8
10
12
14
16
*
*t
18
J
20
acid d o s e (mol/kg)
Figure 1" Final pH as a function of the acid dose (L/S=10 1/kg, extraction time=3hrs)
3.4 Simulation 3.4.1 Theoretical solubility
Infigure 2, the leaching diagT'am for zinc is presented. In the figure the experimental data are presented along with the calculated concentrations (mg/g fly ash) as a function of the pH of the solution a~er leaching. The leaching diagram represents the quantitative partitioning between the different phases (15). The indicated minerals are the controlling solids at the specific pH value. At pH 6, for example, about one third of the total zinc concentration remains in solution (figttre 2), whereas one third is precipitated as smithsonite and one third as ZnO.SiO2. ~
-•
.A
10 ~'
8
8
6
Smithsonite (ZnCO3) / Zn,(O/H)sCl2z
~ ~
2 0
t
----+--------~-----I--
1
2
t
3
4
5
-
t
6
7
8
9
10
11
12
13
14
,n Figure 2" Leaching diagram of zinc (,%Experimental values; m calculated) Other solubility controlling solids are Zns(OH)sCI2 (pH 6 to 8) and ZnO (pH 7 to 13). At low pH the equilibrium concentration is restricted only by the available amount of Zn in the fly ash. At higher pH values, the concentration at equilibrium is determined by solubility
486 restrictions. At pH 8, different zinc minerals occur and no more zinc remains in solution. At pH > 13, zinc solubility increases again, due to the formation of hydroxide complexes. The theoretical predictions are in excellent agreement with the experimental data. When using the sequential extraction method the results of the first two steps can be compared with these calculated results. In the water soluble step zinc shows a very low leachability at a pH of 9.9 which is reflected by the low values in the leaching diagram. At a pH of 3.4 of the acid soluble step in the sequential extraction method the leaching of Zn is already completed. Thus the water and acid soluble steps give accurately water and acid soluble fraction. The simulation results for aluminium are given in figure 3 (15). Between a pH of 4 and 13, aluminium is precipitated as diaspore (AIOOH). The leaching diagram is in very good agreement with the experimental values. However, considering the high concentration of aluminium on the fly ashes, aluminium might also exist as an alumino-silicate compound. This was proven by the sequential extraction procedure where a high residue fraction for A1 was found. The small water soluble fraction is in good agreement with the low solubility at the pH of 9.9. At pH 3.4 the leaching for AI is not yet completed. Thus the fraction of AI and Ca in the acid soluble step is too low. It is also possible that if in the sequential procedure the pH slightly shifts, different results can be obtained. A final pH in this step of 1.5 would be more appropriate. 80
70 60 *~
~A A
50
~ 3o
Dklspore (AIOOH)
~ 20 ~ 1o 0 0
1
2
3
4
5
6
7 pll
8
9
10
11
12
,
I
13
14
Figure 3" Leaching diagram of aluminium
The leaching diagram for calcium is given in./igure 4 (15).
A
2O0 ~ ' ~ ,
>,~ E=
o~ 15o
~
t
'~~
"8 lOO .~
,3
Ca-silicates
2H ,
~
Hydroxyapatite (Ca3(PO,,,)2)
1~\ A,
50
.....
0
1
-
+
I ~__~
. . . . . .
2
Calcite (CaCO,j) ---~----- .... ' ~ .
.
3
4
5
.
.
.
Dolomite(CaCO3.MgCO3)
,
6
7 pH
8
9
10
11
Figure 4: Leaching diagram of calcium
12
13
14
487 Gypsum was found to be a controlling solid over the whole pH range; at high pH values hydroxyapatite, calcite and dolomite appear as controlling solids. The higher soluble fraction found for Ca compared to other metals can be explained by the observed leachability at pH 9.9. At pH 3.4 the leaching for Ca is not yet completed. Thus the fraction Ca in the acid soluble step is too low. And again a final pH of 1.5 in this step would be more appropriate. The leaching diagram for lead is given in figure 5 (15). Only one controlling solid occurs: chloropyromorphite at a pH lower than 9, and Pb(OH)2 at high pH values. When the fly ash is leached with water the fly ash shows a low leachability for lead as can be seen in the leaching diagram. The water soluble step in the sequential extraction method gives also a low leachability. At pH 3.4 the leachability according to this diagram is 0 %. According to the acid soluble fraction in the sequential extraction method it should be almost 100%. This difference can be explained by the absence of a high Cl-concentration in the acid soluble step where CH3COOH is used. Thus chloropyromorphite can not be formed and the leachability is higher. 3T.
A
2,5 =
2
1,5
Chloropyromorph (P~(P04)3Cl)
~'~-" 0,51 "~ a. 0
A
0
~ , 2
1
b(OH)2I
3
~A& I &.~l$ 4 5 6
t 7 pH
i 8
9
10
, 12
11
13
14
Figure 5 Leaching diagram of lead Theoretical calculations and experimental determination of the cadmium equilibrium are in good agreement Oqgure 6) (15). Under very basic conditions, Cd(OH)2 occurs. At a pH of 3.4 of the acid soluble step in the sequential extraction method the leaching of Cd is already completed. Thus the acid soluble step gives an exact view of the acid soluble fraction. The water soluble step gives a fraction of 7.2%. This step has a pH of 9.9 and at that moment the leaching of Cd already starts. 0,3
~"
_ r
0,25 0,2
/~A ,~
o,15
~
0,1
~
,if A
A
avite (CdCOj)
Cd(OHh
\
0,05 0 0
1
2
3
4
5
6
7
8
9
10
I1
Figure 6: Leaching diagram of cadmium
12
13
488
3.4.2 Equilibrium pH It is clear that the final pH of the leaching solution is very important for the leaching of the metals and is therefor important for the sequential extraction procedure. The pH is determined
by the acid dose and the dissolution of matrix elements. Therefore modelling was also applied to predict the equilibrium pH of fly ash leached with an acid solution of given AD. When the fly ash is brought in contact with the acid solution, basic oxides will dissolve causing an increase of the pH, giving the fly ash a certain buffering capacity. The equilibrium pH aider dissolution of basic oxides was calculated by MINTEQA2 for various values of AD. It was found that the variation of pH is mainly related to the AD and the dissolution of aluminium and calcium (CaCO3, CaO and A 1 2 0 3 ). 200
14 .! ~
180
~
,9 9 9 9
-~
160
exp pH
- ' - - - O - - - calculated pH
140
- ....
Ca in solution
120
~"
~
AI in solution
100
~=
80
~r
f
9
60
~
40
-
r 20
0
io
m -2
0
1
2
3
4
5
6
7 A D
8
9
10
11
12
13
14
15
16
17
(mol/k g)
Figure 7: Experimental and calculated pH, calcium and aluminium concentration as a function of the AD
Figure 7 compares experimental and calculated pH values that are in good agreement. The experimentally determined concentrations of calcium and aluminium are also shown. Up to a AD of 4 mol/kg, aluminium oxides do not dissolve and CaO is the main neutralising basic oxide. At a AD higher than 10 mol/kg, calcium is nearly completely dissolved and A1203 is the main neutralising basic oxide. Once the soluble amount of aluminium is in solution, the buffering capacity of the fly ash is consumed and increasing the AD leads to a logarithmic decrease of the pH. When the sequential extraction procedure is used for a fly ash with a different A1 or Ca composition the pH of the acid soluble step changes, leading to a different result. So the composition of the fly ash has an influence on the results of the sequential extraction procedure.
4. C o n c l u s i o n The main elements of the leaching process are solubility and precipitation of heavy metals. Leaching diagrams obtained by simulation calculations are a useful means to obtain a clear picture of the precipitation equilibria for different metals. At a given equilibrium pH of the leaching solution, the leaching diagrams indicate which metals are in solution, and which minerals are precipitated. By comparison with experimental data, the simulation program has proven to give reliable results. Because the results obtained by calculation are in good agreement with experimental results, it may be assumed that the simulation of the leaching behaviour can be extended to fly ashes with different compositions, or to other leaching conditions.
489 The leaching diagrams can be used to verify the accuracy of the sequential extraction procedure. By comparing the results of the water and acid soluble step of the sequential extraction procedure with the leaching diagrams, it is clear that the acid soluble step should be performed with a more concentrated acid in order to obtain a pH of 1.5. At this pH the leaching process of all of the investigated metals is finished. Otherwise it would be possible to obtain different resutls when the pH of the leaching solution in the acid soluble step shifts slightly. Also the choice of acid can be important for evaluating the leachability of the fly ash. Lead precipitates when HCI is used, but with acetic acid lead has a high leachability. The other metals show no influence of the choice of acid. The results of the water soluble step gives a good agreement with the leaching diagrams. During the leaching process, the pH of the leaching solution increases as basic metal salts dissolve. The equilibrium pH is determined by the AD, and the amount of CaCO3, CaO and A1203 on the fly ash. Thus the composition of the fly ash can influence the results of the acid soluble step because the pH changes.
References 1. Wille D., De Boeck G., hn,entarisatie Huishoudelijke AJi,alstoffen in Vlaanderen in 1994, Productie, hlzame#ng en Vetwerking, Openbare Afvalstoffenmaatschappij voor het Vlaamse Gewest, Publicatie nr. D/1996/5024/4, April 1996. 2. Senelle R., Dujardin J., van Damme M., VLAREM II, Die Keure La Charte, Brugge, 1995. 3. Vehlow J., Brown H., Horch K., Merz A., Schneider J., Stieglitz L., Vogg H., SemiTechnical Demons#'ation of the 3R-Process, Waste Management and Research, 1990, vol. 8, 461-672. 4. Gong Y., Kirk D.W., Behaviour of mm#cipal so#d waste incinerator fly ash, Journal of Hazardous Materials, 1994, vol. 36, 249-264. 5. Van Herck P., Vandecasteele C., Wilms D., Characterisation of fly ash from municipal waste incineration and study of the leaching #1 view of metal removal, Proceedings of Solid Waste Management: Thermal Treatment and Waste-to-energy Technologies, Washington, 1995, Air & Waste Management Association, Pittsburgh, 723-728. 6. Comans, R.N.J., Meima, J.A., Model#ng Ca- sohtbi#ty in MSW1 bottom ashes leachates, Elsevier, Proceeding of the international conference on environmental implications of construction materials and technology developments, Maastricht, The Netherlands, 1-3 June 1994, 103-110. 7. Eighmy T.T., Dykstra J., Krzanowski J.E., Domingo D.S., St/~mpfli D., Martin J.R., Erickson P.M., Comprehensive approach towards understanding element speciation and leaching behaviour in municipal so#d waste incineration electrostatic precipitator ash, Environmental Science & Technology, 1995, vol. 29, 629-646. 8. Clesceri L.S., Greenberg A.E., Trussell R.R., Standard methods for the examination of water and wastewater, American Public Health Association, Washington DC, 17th Edition, 1989, part 4, 177-178. 9. Allison J.D., Brown D.S., Novo-Gradac K.J., MINTEQA2/PRODEFA2, A geochemical assessment mode/for em,ironmental ~systems: Uset"s man,al, Environmental Research Laboratory U.S. EPA, Athens, GA, 1991. 10.Felmy A.R., Girvin D.C., Jenne E.A., MINT'EQ- A computer progT"am for calculating aqueous geochemical equilibria, U.S. EPA Project 600/3-84-032, US EPA, Washington D.C., 1994.
490 11.Kirby C.S., Rimstidt J.D., Mineralogy and Surface Properties of Municipal Solid Waste Ash, Env. Sc. Technol., 1993, vol. 27, 652-660. 12.Derie R., Les cendres volantes des incinOrateurs d'ordures mdnag~res. StructureReactivitY, Chimie Nouvelle, 1993, vol. 10, nr 37, 1091-1097. 13.International Ash Working Group, An international perspective on characterisation and management of" residues from municipal solid waste incineration, Summary Report, International Ash Working Group, 1994. 14.van der Hoek, E.E., Comans, R.N.J., Speciation of As and Se during leaching of fly ash, Environmental Aspects of Construction with waste materials, Elsevier, Proceeding of the international conference on environmental implications of construction materials and technology developments, Maastricht, The Netherlands, 1-3 June 1994, 467-476. 15.Van der Bruggen, B., Vogels, G., Van Herck, P., Vandecasteele, C., Simulation of the leaching behaviour of heavy metals from municipal waste incineration fly ash, submitted for publication in Journal of Hazardous Materials.
Goumans/Senden/van der S l o o t , E d i t o r s Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
MODELS
FOR LEACHING
OF POROUS
491 MATERIALS
Pierre M O S Z K O W I C Z (1), Radu BARNA 0)(2), Florence S A N C H E Z 0), Hae Ryong BAE 0), Jacques M E H U (2) (1) LAEPSI, INSA Lyon, 20 av A Einstein, 69621 Villeurbanne, France (2) P O L D E N , INSAVALOR, BP 2132, 69603 Villeurbanne, France
Abstract The release of soluble species contained in solidified/stabilized wastes are assessed by leaching tests. Interpretation of experimental results must be supported by precise modeling of the different phenomena involved hydrodynamics, dissolution, chemical interaction, diffusive transport. The models are presented, which can apply according to the leaching scenario (with or without advection). -
solubilization shrinking core model, diffusionnal model, coupled dissolution/diffusion model.
1. Introduction The Laboratory of Environmental Analysis of Industrial Systems and Processes (LAEPSI) of INSA, Lyon and the division P O L D E N INSAVALOR have consecrated a significant part of their research over the past few years to the study of stabilized waste leaching tests. Interpretation of the results leads to the evaluation of the environmental quality of the obtained materials. Solidification using hydraulic binders gives rise to porous monolithic materials. The pollutants initially contained in the wastes are confined and may even be stabilized within the solid matrix. Certain mechanisms of solidification/stabilization are well known, whereas others are still the object of research. There are numerous factors which govern waste solidification/stabilization. and their choice can be optimized in order to obtain materials with characteristics meeting the technical and environmental specifications of the considered scenarios : good mechanical strength, good leaching behaviour, etc... The release of soluble species contained in a porous monolithic cement-based block in contact with water is the result of complex and coupled phenomena (at the block surface and within the block: -
-
water transfer in the porous medium up to saturation, dissolution of the species in the porewater according to the local chemical context,
492
-
transport of species in solution due to the effect of concentration gradients, change of species solubility in the porewater (including possible reprecipitation) if the context of the latter has undergone certain modifications, due to the pH profile for example, following the release of pH controlling species (portlandite...).
The leaching studies of porous structures in different scenarios of liquid/solid contact lead to the use of several models to describe pollutant release. Two cases can be distinguished: 1- Leaching of monoliths without advection : Several models can describe mass transfer in the porous solid: solubilization shrinking front model, diffusional model, coupled dissolution/transport model. ; 2- Leaching of granular beds with advection : in this case the hydrodynamics of the system must be taken into account by the percolation-leaching model. If the leaching imposes particular conditions of leachant flow around the solid, the mass transfer mechanisms (solubilization, diffusional transport etc...) must be integrated in the hydrodynamic model..
2. L e a c h i n g m o d e l s w i t h o u t a d v e c t i o n
2. 1. Solubilization shrinking core m o d e l This model can describe the case of the main elements of the matrix without major chemical interactions (example: Ca, Na, K,...). It considers the coupling of the two phenomena: instantaneous solubilization (up to the saturation limit of the pore water solution) of the species present in the solid phase of the porous matrix saturated with water and its diffusional transport in the pore solution (characterized by the diffusion coefficient D), without chemical coupling with another species (the common ion effect is not taken into account). The main constitutive hypotheses of the model are : 1- Initially, at t = O, the solute has a uniform concentration in the solid equal to SO(kg/m3). 2- The solution is saturated by the solute with a constant concentration Csat, as long as the aqueous phase in the porous matrix is in equilibrium with the solid phase still containing the solute. d C d2C 0 < x < X(t)
d t
3x 2
d X
x=xm = So d t i
x > X(t)
C = Csat,
S = SO
It is therefore a problem of a "shrinking front": the mobility of the dissolution front is governed by the mass balance at the front position X(t).
493
The rate at which the dissolution front shrinks within the solid is proportional to the square root of time X ( t ) = Ks/i-. Between this front and the liquid interface, transport of pollutants in the pore water takes place by diffusion. O n the other side of the front towards the core there is no mass transfer. The parameters of the model are 9Csat, So and D. The concentration at the liquid/solid interface varies according to the leaching scenario. If renewal is sufficient, this concentration can be considered as zero. .Front
C=O Liq
I
x=O
x
I
x=X(t)
Fig. 1 9The shrinking front solubilization model (one species)
2.2. Diffusional m o d e l The diffusional model is widely used to interpret leaching tests (tank leaching test N E N 7345). Strictly speaking, this model describes the transport of one species initially present, completely dissolved in the pore water. This description can be extended to the case of very soluble elements whose solubility does not change according to the variable physico-chemical leaching context and which are instantaneously and quantitatively solubilized in the pore water: Na, K, C1...) Numerous experiments, in different scenarios, have shown that a diffusional model (based on Fick's law) can correctly describe the released flux J (kgs -1 m -2) of the very soluble elements which are not constitutive elements of the solid matrix : --+
f "
=-Dax
0c dn
where D a= apparent diffusion coefficient (m 2 s-1) C = volumetric concentration, (kg m -3) Generally used assumptions for applying this equation are : isotropic porous media, in which the structure is constant with time. The solubilization is considered to be instantaneous and not mass limited. The concentration at the solid/liquid interface is zero (which is the case if water renewal is sufficient). The apparent diffusion coefficient D a (in m2/s) is assumed to remain constant in time and space, which implies, in particular, that the solid is saturated with water from the beginning of the process and that no physical or chemical alteration disturbs the diffusion phenomenon.
494 The fundamental diffusion equation becomes: Of
(~2C
~2C
0 2 C "]
d t - Da x ~ , d x 2 + d y 2 + d z 2 ) For a solid of infinite length (x e [0, oo D in contact with the liquid via a normal plane surface of direction x, the flux of the leached material can be written :
J(t) =-Da d~xl x=0 =
Co x ~ a
if C o is the initial leachable concentration. The two parameters Co and Da can be identified by two distinct experimental tests (availability test and tank leaching test). We proposed an approach based on the simultaneous identification of the two diffusional model parameters, from only one tank leaching test if the leaching time is sufficient to reach depletion of the released species in the solid core of the leached sample. In figure 2 the values of the standard deviation z of the simulated and experimental sodium concentrations in the leachate are represented, according to the varying values of D a and C 0. The optimal values of the parameters C O and D a are used in figure 3 (continuous curve) to simulate the released sodium flux J (mgs -1 m -z) in comparison with the average experimental flux (pOints).
__/-
.....
---
z
3
4
5
6
7
8
~t Fig. 2 : Optimal values of C Oand D a
Fig. 3 : Simulation of Na release
Long term simulation for any scenario involving solid/leachate contact is then theoretically possible. The application limits of the model are reached when the physical characteristics of the material itself are modified (increased porosity, destruction of the porous structure...). We have also observed a 100 to 1000 times lower "residual flux": the release continues although the defined (theoretical) diffusional flux has stopped. A possible explanation of this phenomenon could be the continuation of release of a less soluble phase after depletion of the more soluble phase.
495
For the same experiment, the comparison between the solution of the diffusional model (parameters" C o and Da), and the shrinking core model (parameters Csat, So and D) shows that"
Co=A
IoD a =SO a 2
e
' Csat
and
'I
Da =1
-
t
D
=J
Co erf 2 , f ~
Based on the available experimental results (solute mass released in the leachate), it is not possible to distinguish between the two models : as long as the solid "remains" semi-infinite, the mass released is proportional to the square root of time. In conclusion, with respect to the initial hypothesis, the two models are equivalent for the modelling of the release of specific species. However, an important difference is that the concentration profile of the species in the pore water can be calculated from the shrinking core model.
2. 3. C o u p l e d d i s s o l u t i o n / d i f f u s i o n
model
This model must be considered in the case of elements whose solubility depends on the variation of the physico-chemical context, pH in particular (example" amphoteric metals). Experiments have shown that amphoteric metals release is controlled by solubility in the pore water context. The pH evolution within the solid and especially at the solid/liquid interface is a significant parameter. Lead is a typical case of such behaviour. Pb additive release in a mortar elaborated from Portland cement (sequential leaching of identical samples of the same monolithic material containing PbO) in contact with different chemical contexts : demineralized water (W), controlled pH 5 and pH 10, alkaline water at pH 12.5 (AW) is presented in the figures below. 6000
[ ........
300 ""
5000 1W 4000
9 pH5 ApHIO
2000
:
1ooo
~jm
0
-" ~
150
i
100 _A 9 F-
50
i ~
0
m
250
200
OAW
r
3000
s
1
~ 2000
t
9
~ 4000
9
OpH5
9
1
0 o
2000 time
tim e (hours)
4000 (hours)
Fig. 4" Leaching of Pb in different leachants The release of Pb is sensitive to the chemical context of the leachant and cannot be described by the simple diffusional or shrinking core model. A coupled dissolution/diffusion model can describe the release of chemically more complex species contained in a stable porous matrix in contact with water. In the case of a porous matrix containing two leachable components: calcium hydroxide and lead hydroxide, the modelling of release can be decomposed into several stages :
- release of portlandite, described by a shrinking front model ;
496
-
calculation of the induced pH profile, assuming that local chemical equilibrium takes place in the porewater ; - determination of local lead solubility (by calculation or from specific experimental determination); - description and calculation of lead transport by diffusion in the porewater and/or at the solid/liquid interface.
Different simulations were carried out and compared to the results obtained from leaching of cement-based matrixes containing lead. The coupled dissolution/diffusion model allows representation of the leaching tests results using demineralized water and confirms the interracial character of lead release. As long as lead in solid form is present in the matrix zone near the leaching surface of the matrix, its release is controlled by a solubilization phenomenon at the solid/liquid interface. In this case, the leaching model can be simplified: diffusional transport of lead within the matrix can be neglected. A model based on the shrinking core model to describe the calcium release from which the pH evolution at the solid/liquid interface is assessed (Figure 5) and variable lead solubility according to this pH (assessed in an Acid Neutralization Procedure) allows a good representation of the phenomena (Figure 6). The case of lead is specific:the release seems to be governed (on the time scale of our laboratory leaching tests) by a lead solubilization phenomenon at the solid/liquid interface, itself governed by the pH as the relevant parameter. 12,2 12 11,8 _ 11,6 11,4 ~ 11,2 ~" 11 10,8 10,6 i 10,4
0
i l l
f// ~[[ J ! ! ~!! I I
0
!
i
r
m
/
I I
I
f
_-1
/
r
#
=
~'-3
|
...............
1000 2000 3000 4000 5000 t (h)
Figure 5" Simulation of pH evolution near the solid/liquid interface.
-5 I . -6
0
.
.
2
.
. . . . . ...................
4
6
8
.
.
10
12
14
pH
Figure 6" Experimental solubility curve according to pH
The observed discontinuities on the simulation of pH evolution at the matrix cement/solution interface result from the sequential character of the leaching test carried out (periodical renewal of the leachant). Knowledge of the pH near the matrix cement/solution interface allows, from the experimental solubility curve, evaluation of the lead concentration at saturation in this zone. The quantity of lead released can be estimated via an interfacial transfer coefficient (figure 7).
497 2000
1500 l 9 Poin~exp. '~-'Simulation
1000
500 0 0
1000
2000
3000
4000
5000
t (h)
Figure 7 : Simulation of lead release. Leaching with demineralized water
3. Leaching models with advection Leaching of granulate materials is usually carried out by percolation tests. Transport by advection must then be taken into account. The 6rst level of modelling concerns characterization of the hydrodynamic regime of liquid through the column. Dispersion must be taken into account. The general equation governing solute transport in the mobile liquid phase is as follows (one dimensional model) :
c dt
c - D
/l ~ , d2xJ _ v d x
+ R
where v is the velocity of the liquid, D the hydraulic dispersion coefficient and R a source term (corresponding to dissolution flux). The phenomenon of advection-dispersion can classically be translated by a model consisting of n identical contactors of continuous stirred open reactor type each containing a mobile phase in contact with the solid phase. The presence of stagnant zones can also be taken into account. The porous solid is therefore in contact, in each reactor, with the mobile liquid as well as the stagnant liquid. The solute is exchanged between the solid phase and the liquid phase as well as between the mobile and immobile liquid phases. In the following figure, the hydrodynamic model of percolation-leaching is presented.
Q
Cm,j-1
Vm Cmj
V.
.... d,r; o
Cmj
Cim.i .........
Figure 8 : Hydrodynamic model of percolation-leaching of porous granular material
498 The experimental use of a tracer allows identification of model parameters (n number of reactors, mobile fraction fm of the volume and transfer coefficient K (between the two liquid zones). But problems of interaction between the porous medium and the tracer (diffusive transport in the porous system, surface sorption phenomena, etc...) may complicate interpretation of experimental results. The balance of tracer in the cell j for time i'At, while flowing and in the stagnant zone (Coats and Smith model), is : i i d V i W dCi,j QCm,j_I-QCm,j = ~ - ( Cj)=fm--~ n n dt
W dCiim,j
+
f
- - ~ dt
lm n
i dCim,j
dt
- K(Ci
i ) m,j -- Cim,j
where :
Cm,i: the volumetric concentration of tracer in the mobile fraction and in the cell j, kg/m 3, Cim,j : the volumetric concentration of tracer in the stagnant zone and in the cell j, kg/m 3,
V : the volume accessible to the fluid, m 3, fm : the mobile fraction in the liquid phase, fxm : the immobile fraction in the liquid phase, I(
~o rn
-1.5
pH
0 -0.5
r~
I
I
I
I
I
I
I
6
7
8
9
10
11
12
CaS(M.2H20 (Gypsum)
~ -1.5
I
5
/
/
~
~
~
*
~
_
0 +
9
0
-2 -2.5
•
o X
-3
+
IITO
cl rm
x
-3.5 -4
o
I
I
I
I
I
i
J
i +.
5
6
7
8
9
10
11
1 2 4+
1-
-0.5 -1 -1.5 r~
-2
I
13
t~
BaCO3 (witherite)
Batch tests
8000
+
O
6000 3 __.
o8
/
20
r--" .
.
.
.
4000 ~ 2000
,,.4--/--
+ +
v
-~.
>
8ooo ~ ~,
,0
w
PORTLAND C'EMENTI;ILICA FIJME i2 and 2W)
6000
5000 > 4ooo... n 3o
10 am
8
n,
6
Om~ Ua 4 2
0
CIID O 0
0
0
0
0
30o0 ~ ~
0
2000 '= o m o 1000
onnF~
0
ACTIVATED BLAST FURNACE SLAG (3)
120 ~
r 6000
100
5000
m
80
4000
=:
60
og a
/
+
3000
0 ,o; 40
>
1000
0
0
COAL FLY ASH/LIME (4 and 4W)
120
8g
o
2000 "= o
20
,e,
~
~5
6000
100
5000
80
400o
60
3000
o ,,=, 40
2o00 o oo
20
lO00
0
0
0
20
40
Figure corroded
60
T I M E (DAYS) HIGH ETTRINGITE (5 and 5W)
80
2. p H 3 a c e t i c a c i d :
depth (O no waste, + with waste) and
acid added
(M
n o w a s t e , -. . . . w i t h w a s t e )
100
557 4OO .E10
~
s
~
~
2
~
(
I
-
P:' ==~6
)
0 0
300
10o O r ,
.
.
9
.
.
.
9
.
,,
.
.
.
.
.
9
.
.
.
9
PORTLAND CEMENT (1 and l W )
12
~
10
C
C
.
OO
400 300
8 (no
"=m o
0
OD O00DB3:IXIIIXID
n, 6 oc o ~ 4 (,.)s
>.
r~ 2oo~ "B.
.~-:-_q.-+- -+-+
.. ~ .--:-
ou~4 o
m a
O0
::::::|
: UL:.UdL;: ~ u ~
++
> (->
30 2OOo3 >
+ .-I-_
-- - .---:'=':=
m ~
100
2 o
PORTLAND CEMENT/SILICA FUME (2 and 2W) 400
I0 uJ
8
n,
6
(3
300
( [ ll-]f]-~-x~J-- 0(3 (3(3D
o~4 ua
......... O0 . . _.._.__.,.~...--.
O0
> c)
200 o3 o> m
100
2 0
o
ACTIVATED BLAST FURNACE SLAG (3) 400
10 '" a
8
e,,
6
300
(3000
(o~4 ,.)a
..... z
2
,
~
~
_ '_.~.""',"'
120
.
a ~ r,, 80
204o~! 0
--
0
.
~ .
100
O0
X n ~ n R ' ~ F : T . ' 3 ;; . . . .
00
T + + ----
. . . . . . . . . . . COAL FLY ASH/LIME (4 and 4W)
.
.
o3 o> m
100
--
.
3 c:)
200
.
.
.
.
o
0
,
1600 m--
,., . . . . . . . . :::: ~ + ~
=" ~. s0
o~,
.
/:%
..
i11
o-
:
.........
.-,- .
:::':: :;,,,,::::::
.
.
.
++
.
.
120o
" "
"~
800 oB o~,
++
,
200
300 400 TIME (DAYS) HIGH ETTRINGITE (5 and 5W)
500
F i g u r e 3. p H 5 a c e t i c acid" c o r r o d e d d e p t h (O n o w a s t e , + w i t h w a s t e ) a n d acid added (--
n o w a s t e , - .... w i t h w a s t e )
> ('>
5
. . . .
400=O~ 0
S00
558
solution. An orange colour was also observed in the outer layer of the portland cement formulation suspended in nitric acid. Other authors attributed an orange colour change observed for portland cement-based solidified wastes immersed in acetic acid to oxidized iron (Cheng and Bishop, 1996a). This explanation is consistent with the absence of a colour change in the specimens not contammg portland cement. Addition of silica fume consumed the excess calcium hydroxide generated by the portland cement, but the portland cement/silica fume specimen crumbled suddenly and completely, after exhibiting no corrosion for 7 weeks, suggesting that expansion reactions as well as acid attack may have affected its structural integrity. Calcium depleted from the calcium-rich CSH of this formulation may have reacted with the acetate to cause this effect. The portland cement and portland cement/silica fume specimens containing waste did not suffer any noticeable expansion, and were gradually but quickly corroded. The activated blast furnace slag and fly ash/lime binders had not yet disintegrated when ~he experiment was discontinued after 80 days, but exhibited corrosion depths of 6 and 2 mm, respectively. Expansion did not appear to occur in these binders, and may have been averted by the lack of easily soluble calcium, as the Ca/Si ratio of these systems was 0.5. Addition of waste to the fly ash/lime system resulted in its complete disintegration within a week, showing again that the modified hydration products in this formulation had a poor acid resistance. The high alumina cement formulations also disintegrated within a week, due to the high solubility of ettrmgite and calcium chloroaluminate at low pH. 3.3 Corrosion by pH 5 Acetic Acid Comparison of Figures 1 and 3 shows that pH 5 acetic acid was also more aggressive than pH 3 nitric acid, to all but the fly ash/lime/waste formulation and the high alumina cement formulations, with corroded depths between 6 and 12 mm measured after 16 months of immersion. It is possible that the deterioration of specimens in this experiment was due to a combination of acid attack and expansion. Because of the lower concentration of acetic acid, a different mechanism occurred than in the pH 3 acetic acid experiments; rather than causmg massive expansion and failure, formation of smaller amounts of calcium diacetate lead to cracking at a smaller scale which facilitated acid attack. Specimens with less soluble calcium, i.e., with a lower Ca/Si ratio binder (activated blast furnace slag and fly ash/lime) and/or contaming waste, were not as subject to expansion and therefore deteriorated less. The high alumma cement specimens containing incinerator ash immersed in pH 5 acetic acid were the only ones of this bmder type which had not completely corroded by the end of the test. A sharp discontmuity is apparent in the corrosion and acid addition curves after 22 weeks, which may be indicative of formation of a protective layer of aluminum triacetate and/or calcium diacetate by reaction of the acetic acid with alummum and calcium liberated by ettrmgite and calcium chloroaluminate dissolution. This protective layer may not have had an opportunity to form at pH 3, because of the extreme rapidity of dissolution of the specimens. The discontmuity in the corrosion and acid addition curves for the r e m a m m g specimens after 25 weeks corresponds to a replacement of the leachant, and is unlikely to be related to calcium diacetate precipitation, as the amount of acetate added was insufficient to exceed solubility limits. For the same reason, formation of a calcium diacetate precipitate under natural conditions is unlikely. 3.4. G e n e r a l i z e d Factors in Acid Attack In general, the corrosion plots exhibited three stages: (1) a lag period before (2) acceleration of deterioration, followed by (3) a decrease in rate of deterioration. It is postulated that the lag period is attributable to depletion of easily soluble alkalinity from the surface layer of the specimens, which resulted in consumption of acid, but left a structurally stable
559
matrix. In the case of the portland cement formulations the soluble alkalinity was initially likely to have been mainly sodium, potassium and free lime; later on, once the pH of the corroding surface layer dropped below 12.5, and for the lower Ca/Si ratio products and those containing waste, it is expected that decalcification of calcium silicate hydrate (CSH) took place. For the CSH-based matrices, visible deterioration of the matrix structure would not be expected until the pH of the corroding layer decreased below 9.9, where CSH coexists with more soluble silica gel (see review in Stegemann et al., 1994). The lag period was longer for the lower Ca/Si ratio solidified products because the higher Ca/Si ratio products contained free lime. Free lime is more soluble than CSH, and leaves a higher porosity matrix as it dissolves, increasing exposure to acid attack. Cheng and Bishop (1996b) found a porosity of 0.8 in the decalcified layer of portland cement-based sohdified wastes. For the high alumina cement formulations, the lag period may have been caused by dissolution of alkalis and free lfine; deterioriation of the matrix structure would be anticipated to start when the pH of the surface layer dropped below approximately 11 (see review in Stegemann et al., 1994). The period of accelerated corrosion was linear as a function of time. In the case of the CSH-based matrices, it resulted from increased dissolution of the silica-rich CSH, alummosilicates and silica gel r e m a m m g after decalcification of the CSH, as the pH of the corroded layer dropped from 9 to below 5. Agam, a higher Ca/Si ratio increased the vulnerability to acid attack in all three series of experiments. The corrosion rate was highest for the high alumina cement matrices, in which ettrmgite or calcium chloroaluminate dissolved rapidly. In the portland cement and portland cement/silica fume systems, with and without waste, and in the activated blast furnace slag and fly ash/lime systems, it is postulated that a protective surface layer consistmg mamly of silica gel, and also containmg a l u m m a and iron compounds, gradually developed over time. F u r t h e r leaching of alkahs and calcium and inward movement of acid to the corrosion front then became controlled by diffusion through this layer (Pavlik, 1994). The benefit of the silica gel protective layer was most evident in the relatively low corrosion of the activated blast furnace slag and fly ash/lime specimens upon immersion in pH 3 acetic acid. The low free lime content of these matrices resulted in a particularly dense silica gel protective layer. This layer has an additional benefit in sohdified wastes, in that the silica gel can adsorb heavy metal contammants at pH values as low as 5 (Schmdler et al., 1976). A high proportion of free lime, or other soluble calcium, such as ettringite, can result in secondary precipitation of calcium salts, which may contribute to formation of a protective layer but can also lead to expansion, as was seen in the acetic acid experiments. The corrosion rate did not decrease over time for any of the high alumina cement specimens, except the one made with incinerator ash and immersed in pH 5 acetic acid discussed m 3.3, as dissolution of the ettrmgite and/or calcium chloroaluminate matrices in nitric acid was complete, leaving no residue nor depositing a protective layer on the surface of the monohth. Other authors have found that, at the same concentration, mineral acids are more corrosive to cements than weak acids (Pavlik, 1994 and Bayoux et al., 1990). Such was not the case in this series of experiments, because continued addition of acetate caused precipitation and expansion reactions which lead to increased matrix deterioration in the weak acid.
3.5. Acid C o n s u m p t i o n of Different Matrices A straight line relationship between depth of corrosion and acid addition would be expected for matrix deterioration by chemical dissolution. Plots of the depth of corrosion as a function of the amount of acid added are not shown here, but a straight line passing through the origin was fitted to the data for each specimen by the method of least squares, to determine slopes of mm of corroded depth per mmole of acid added. For the pH 3 nitric acid and pH 5 acetic acid experiments, most
560
correlation coefficients (R) were found to be between 0.8 and 1, indicating that the corrosion depth and acid addition data were highly correlated, except when the corroded depth was too small to be accurately measured (e.g., activated blast furnace slag corroded by nitric acid), or when deterioration mechanisms other than dissolution were a factor (e.g., fly ash/lime/waste formulation in nitric acid). For the pH 3 acetic acid data, correlation coefficients higher than 0.8 were determined only for the blast furnace slag and fly/ash lime systems. These formulations were the only ones which maintained their structural integrity in the pH 3 acetic acid; under aggressive attack by acetic acid other deterioration mechanisms, e.g., expansion, cracking and crumblmg, came mto play for the other formulations. For comparison with acid neutralization capacities measured for ground samples (Stegemann et al., 1994), the mm of corroded depth per mmole of acid added were converted to mmoles of acid added per gram of dry cement using the surface area exposed to the acid, and the solidified product densities and cement contents. Table 2 summarizes the resulting values for each specimen. The approximate amount of acid per gram of dry cement in the formulation, which was required to achieve complete matrix destruction, was read from the acid neutralization capacity (ANC) curves generated previously. It was assumed that the all matrices were completely destroyed at pH 5. Table 2. Amount of acid required for complete matrix destruction Batch mmol of acid/ lW 2 2w 3 4 4w 5 5w of dry cement 1 62 16 47 25 17 7.0 7.0 8.9 pH 3 HNO3 19 160 ** 55 360 800 41 39 20 pH 3 CH3COOH ...... 1.6.0 64 9.0 22 12 12 43 21 I1 pH 5 CH3COOH 21 23 16 20 13 5.0 ll 8.0 6.4 .~NC to pH 5* > 17 * based on Stegemann et al., 1994 ** slope not calculable because sudden crumbling followed a period of no corrosion Values for which the correlation between corrosion depth and acid addition was poor (i.e., R= 40
._
NIPS
4----
Natural (AJabaster) Landplaster
m
E
20
300
212
150
106
75
53
38
27
19
13
Particle Size (microns) Natural
"0-
Synthetic
9,4
6.6
4.7
3.3
2.4
585 4.4. Normative The parameters that indicate the quality of the desulphogypsum, because its utilization as by-product in European industry of the plaster are withdrawals in following Table I (Wirsching, 1993). The "United States Gypsum Company" (USG) has a long experience in the characterization and the marketing of the american plasters; It establishes a series of characteristics for the artificial plasters,(Henkels, Gaynor, 1995). Ojanpera and Cabbage (1993) (see Table II) offer a comparison between natural plasters and desulphogypsum in the one which are emphasized important advantages of these last with respect to several of the characteristic parameters.
Table IL ~ U t y of Gypsum FGD and natural
Table I. Quality criteria for Gypsum FGD Parameter
% weight
Water content
< 10
CaSO4.2H20
> 95
Soluble MgO
< 0.1
Propriety
Gypsum FGD
Free water %
6-10 !
Combinated water %
CaMg
,
Natural Gypsum
~
0.2-3
!
19-21
19-21
0--5
0.15-1.6
CI-
< 0.01
Carbonates %
Na20
< 0.06
caso,.~o %
91-99.8
97-98.5
SO2
< 0.25
SiO, ppm
1700-7000
5000--6500
5-9
AI20s ppm
1000-5000
10%); + = moderate (5-10 %)" +/- = traces (5%); - = non detectable (< 5%)
Selective dissolution with oxalate and dithionite The results of the selective dissolution experiments are given in Figure 6. The results indicate that the dithionite extractable heavy metals fraction and to a lesser extent the oxalate extractable fraction are affected by the thermal treatment. The dithionite extractable heavy metal fraction is in general significantly higher in the thermally treated samples than in the untreated sample. Moreover, the dithionite extractable heavy metal fraction generally increases when the samples are treated at longer residence times. With the exception of Cu, a relatively small amount of the heavy metals, compared to the amount released in the dithionite extraction, is extracted in the oxalate extraction. The increase of the dithionite extractable heavy metal fraction after thermal treatment most likely results from the transformation of amorphous iron hydroxides to more crystalline forms. This conclusion is consistent with the earlier noted TEMobservations that during the thermal treatment a significant fraction of these metals is scavenged by magnetite structures. In addition, the results reveal that Cu and to a lesser extent Cd, Pb, and Zn are also retained by thermal decomposition products of the clay minerals (allophane-like materials). Along similar lines, compositional mapping using electron microscopy recently indicated that these heavy metals were incorporated into newly formed hydrous aluminosilicate rims on glassy grains of weathered municipal solid waste incinerator ash (Zevenbergen et al., 1996 and references therein). In conclusion, the formation of magnetite and allophane-like materials may significantly reduce leaching of those metals from thermally treated soil.
668 0.15 ,-, 0.125 o~ o~ 0.1
E u
Cd
A O~
0.05
I t~ dithionite 9oxalate
I~ lO
x 0.025
5
untreated
15o
Pb
20
dithionite I oxalate
0.075
,-, 125
3O
-~ 25
10 min; 650 C
30 min; 650 C
untreated
40 35 o~30
Zn
A
I B dithionite
9oxalate
I
,
10 min; 650 C
v
30 rain; 650 C
10 min; 650 C
t
30 min; 650 C
Cu m dithionite
20 15
9oxalate
lO 5 untreated
t
0
untreated
i
10 rain; 650c
i
30 min; 650c
Figure 6. Amount of dithionite and oxalate extractable heavy metals (in mg/kg) from the untreated and the thermally treated sample III (10 min/650~ Leach testing Figure 7 summarizes the column test results derived form the untreated and treated samples. The results are presented as cumulative emission (mg/kg) at a L/S ratio of 10. As the temperature and/or residence time increase, the emission of the heavy metals generally decreases. This effect is most obvious for Cd, Cu, Pb, Ni, and Zn. As exhibits no systematic trend. In sample I a longer treatment time seems to enhance the leachability of this oxyanion, while the reverse is observed in samples II and III. The leachability of Cr appears to be slightly affected by the thermal treatment with the exception of the sample which has been treated at the highest temperature and longest residence time (sample II, 30 min/750~ In this sample the leachability of Cr is significantly higher than in the untreated sample. For comparison, CEC, pH, and leaching data of As, Cr, Cu, Ni, Pb, and Zn derived from different soils before and after treatment in thermal soil cleaning plants (closed symbols) are plotted together with the experimental results (open symbols) in Figure 8. These results reveal that the pH generally increases from values around 8 to values around 10, while the CEC decreases after thermal treatment. Both alterations are probably a result of the removal of organic matter by thermal destruction. The leaching data derived from the samples which have been treated in the thermal treatment plants are consistent with the above reported experimental leaching data. Recent studies have revealed that the leachability of heavy metals like Zn, Cd, Pb, and Cu from thermal residues (e.g. coal fly ash and MSWI ashes) and contaminated soils typically attains a minimum value in the pH range between 7 and 10 (Van der Sloot, 1996; Meima and Comans, 1997). Since the pH of both untreated and thermally treated soils roughly coincides with this pH-range, it is not likely to assume that the observed decrease in leachability of these contaminants after thermal treatment is exclusively due to a pH-effect.
669
sample I
~ A
[] untreated
0.1
E
[] 10 min; 550 C B 10 min; 650 C
"E 0"01
II 30 min; 550 C
0.001 As
i
Cr
Cu
Pb
Ni
Zn
sample II
,.,,,
E
Cd
[] untreated
0.1
930 min; 650 C B 10 min; 750 C
0.01
930 min; 750 C 0.001 As
Cd
Cr
Cu
Pb
Ni
Zn
sample III E
[] untreated
0.1
[] 10 min; 650 C 0.01
B 30 min; 650 C
0.001
I
As
Cd
Cr
Cu
Pb
Ni
Zn
Figure 7. Leachability (cumulative emission at L/S = 10 in mg/kg) of heavy metals from the untreated and thermally treated samples.
670
i
1o.ooo
I~ 9
. ..-- 9
E i ~9 1.000 .
.
.
0,010
s
....
,....
0 o ......
.---..o
0.010 l/I
0.001 0.001 0.010 0.100 1 .000 10.000 emission before thermal treatment (mglkg) 1o.oo0
...-"
...."
E
"
i .
"" J
l
~
0 "
"
.
. . t, 0.010 0,100 1 .000 10.000 emission before thermal treatment (mglkg)
10.000 I
0.100
9
9 0.001 . . 0.001
.
1.ooo Pb
0.010
O /
.-
e i
.. - " ....
...........
..... 9
.
lOO
-
..
"
""
Zn. .......
1.000
~ 0.100
.
1.000
9
As
.-
"
~0.100
,~
0.010
i
0.001 0.001 0.010 0.100 1 .000 10.000 emission before thermal treatment (mglkg)
S
I
0.001 J ,, --" ,, ,' 0.001 0.010 0.100 1 .000 10.000 emission before thermal treatment (mglkg) 1.000
"
0.100
0.010
1.000
t ~
0.100
0.010
| 0.001
0.001 0.01 0 O. 100 1 .000 emission before thermal treatment (mglkg)
0.0(!1
I,
t, '
0.001 0.010 O. 1O0 1 .000 emission before thermal treatment (mglkg)
Figure 8. CEC (meq/100g), pH and leachability of As, Cd, Cr, Cu, Ni, and Zn (cumulative emission at L/S = 10 in mg/kg) before and after thermal treatment (open symbols = thermal experiments" closed symbols = soil thermal treatment plants).
671 60
r
o
E
5o
"E
40
9~r
30
- 9 .,~
{
20
r w
10
,t., c
o
14
E
12
*"
10
0
tw
o
.1=
8
z 4
10
0
C EC
20
30
40
50
60
before thermal treatment (meqllOOg)
4
6
8
10
12
pH before thermal treatment
Figure 8. Continued 4.
C O N C L U S I O N S
Recent research has revealed that heavy metals exhibit a lower leachability in soils after thermal treatment at intermediate (550 to 750~ temperatures than the original materials. Our experimental data are consistent with these observations. However, we also observed that thermal treatment had little or even an inverse effect on the leachability of As and Cr. The experimental data reveal that thermal treatment results in a drastic change of the mineralogical an chemical nature of the soil. The thermal alterations which may contribute to the observed decrease in leachability of these heavy metals are: an increase of the pH; formation of crystalline magnetite structures from ferrihydrite and from Fe expelled by the clay minerals during the treatment. These magnetite structures appear to have scavenged heavy metals during the treatment; formation of reactive, amorphous aluminosilicates from clays. It should be noted that the pH of thermally treated soil will gradually decrease to approximately its original pH due to weathering and accumulation of organic matter. At present, it is unknown to what extent these processes will affect the leachability of these heavy metals. Further study is needed to assess the effects of weathering on leaching behaviour of thermally treated soils on the longer term. The results presented in this paper and other work suggest that thermal technologies in the intermediate temperature range (500 - 650~ offer an opportunity to simultaneously destroy organic contaminants and to immobilize heavy metals in soils in one unit operation. However, further work is necessary to determine the capabilities and limitations of these technologies for soils contaminated with both heavy metals and organic pollutants. The results of this study seem to justify emphasis on the role of iron and clay minerals in thermally treated soil and on the conditions of thermal treatment which affect the behaviour of these constituents in soils. R E F E R E N C E S
Bates, J.K., Bradley, J.P., Teetsov, A., Bradley, C.R., and Buchholtz ten Brink, M., 1992. Colloid formation during waste form reaction: Implications for nuclear disposal, Science, Vol. 256, p. 649-651. Eddings, E.G., Lighty, J.S., and Kozinski, J.A., 1994. Determination of metal behaviour during incineration of a contaminated montmorillonite clay. Envirn. Sci, Technol. 28, 17911800.
14
672 Heynen, J.J.M., Comans, R.N.J., Honders, A., Frapporti, G., Keijzer, J., and Zevenbergen C. (these proceedings). Development of enhanced testing procedures for the determination of leachability of heavy metal contaminated soils. Meima, J.A. and Comans, R.N.J., 1997. Geochemical modelling of weathering reactions in municipal solid waste incinerator bottom ash. Environmental Science and Technology (in press). Mizota, C. and van Reeuwijk, L.P., 1989. Clay mineralogy and chemistry of soils formed in volcanic material in divers climatic regions. Soil Monograph 2, ISRIC, Wageningen, The Netherlands, 186 p. Van Hasselt, H.J., 1996. Emission control system to meet Ducth stack emission standards on soil thermal cleaning plants. International Incineration Conference, Seatle, Washington, p. 177-181. Van der Sloot, H.A., Comans, R.N.J., and Hjelmar, O., 1996. Similarities in leaching behaviour of trace contaminants from wastes, stabilized wastes, construction materials, and soils. Sci.Tot.Env. 78, p. 111-126. Wei, Y., 1995. Leaching study of thermally treated cadmium-doped soils. Hazardous Waste & Hazardous Materials. Volume 12, No 3, p. 233-242. Zevenbergen, C., Van Reeuwijk, L.P., Bradley, J.P., Bloemen, P., and Comans, R.N.J., 1996. Mechanism and conditions of clay formation during natural weathering of MSWI bottom ash. Clays and Clay Minerals, Vol. 44, No 4, p. 546-552
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997Elsevier Science B.V. All rights reserved.
673
INVESTIGATION STRATEGIES FOR CONTAMINATED SOILS IN FINLAND Hanna-Liisa J~irvinen Geological Survey of Finland Espoo, Finland
ABSTRACT Geological Survey of Finland has completed a project, the objective of which was to find reliable investigation strategies and methods for contaminated soils. The applicability of the ISO-standard Draft, the Dutch prestandard and U.S. EPA recommendations to Finnish geological and hydrogeological environment was studied. During the preliminary assessment, standard drafts turned out to be complicated and expensive to carry out in practice. However, the basic principles in drafts were found applicable also in Finland. During the project a new Finnish practice was created. Investigations on contaminated sites are divided into three phases: preliminary survey, field investigation and additional investigations. Based on the preliminary survey the site is whether "probably uncontaminated" or "potentially contaminated". The distribution of the contaminants is always supposed to be heterogeneous in Finnish geological environment. Quality control samples reveal the quality of sampling and analyzes. The local baselines (background values) must always be verified in any assessment of contaminated sites. The sampling pattern is determined after the preliminary survey and it is designed primarily caseby-case. The sampling intensity is recommended for both probably uncontaminated or potentially contaminated sites.
1 INTRODUCTION This research project was carried out by Geological Survey of Finland and Technical Research Centre of Finland. It was partially financed by The National Environmental Geotehnics Program (organized by The Technology Development Centre of Finland). The objective of this project was to recommend reliable investigation strategies and methods for contaminated sites as well as provide quality specifications for various remediation methods. In the final project report the recommendations are given for a number of soil and water samples, their type and locations, quality control, analytical methods etc. The report includes a short review of geophysical methods, which are an increasing field of investigation. The recommendations are based on the experiences gathered from investigation projects made by Geological Survey of Finland and Finnish consultant companies. So far, widely differing approaches and research intensities have been used in the completed remediation projects. Investigations have generally been insufficient. Usually remediation decisions are based limited risk assessment if any, which may lead to unnecessarily intensive measures. The aim of this project was to unify research procedures to improve risk assessment and remediation solutions. The applicability of the ISO-standard Draft (ISO 10381-5 Version 6, draft) Ref.l., the Dutch prestandard (NVN 5740) Ref.2. and U.S. EPA recommendations Ref.3. to Finnish geological and hydrogeological environment was studied. However, they turned out to be complicated and expensive. Thus, the project published a new Finnish practice Ref. 4.
674 2 RESEARCH PHASES Investigations in contaminated sites are divided into three phases like in ISO-standard Draft Ref.l.: preliminary survey, field investigation and additional investigations. During the preliminary survey, information on the past and present activities on the site as well as basic information on the soil stratification and hydrogeology is gathered. The hypothesis of the situation at the site is formulated like in ISO-standard Draft Ref. 1. and the Dutch prestandard Ref 2. The site is whether "probably uncontaminated" or "potentially contaminated". However, the distribution of the contaminants is always supposed to be heterogeneous in Finnish geological environment. Soil and groundwater sampling is carried out during the field investigation phase. The chemical analyzes of samples will reveal if the site is contaminated. Additional investigations are applied if needed. The objectives of additional investigations are the following: to provide detailed information on the geological condition of the site and its impact on surrounding area, to give detailed information on contamination (3D) for risk assessment as well as for remediation design, cost estimate and performance. The intensity of additional investigations depends on geological and hydrogeological conditions, the reliability of earlier investigations, the future activity on the site and considered remediation technologies. After investigations the risk assessment is carried out to evaluate the hazards on the site and in the surrounding areas. Health risk and ecological risk assessment provide information to decision makers as the consequences of possible actions.
3 SUBSTANCES FOR CHEMICAL ANALYZES When potentially contaminated sites are investigated, soil and groundwater samples should be analyzed for the contaminants that are probable according to the preliminary survey. In addition, a small number of samples should be analyzed for a wider spectrum of substances. The most common contaminants should be analyzed when probably uncontaminated sites are in question. Ref. 1. Typically in Finland soil has been contaminated by the following inorganics in industrial or other activities: arsenic, chromium, copper and mercury in wood-processing industry; copper, nickel, zinc and cyanides at mines, metal smelters and shooting ranges; zinc and chromium in surface treatment; mercury and chromium in leather and fur industry. The typical organic contaminants are creosote oil, chlorinated phenols, dioxines and furans at wood preserving facilities; PCBs in chemical wood-processing industry; oils, solvents, dioxines and furans in chemical, textile, metal and machine building industry as well as at waste management facilities; oils, solvents, and gasoline at gas stations. Ref.5.
4 QUALITY CONTROL SAMPLES AND BASELINE SAMPLES Quality control (QC) samples are taken near to the point from which the original soil samples were taken. QC samples are divided into two parts which both are analyzed. The differences between an original sample and a QC sample reveal the quality of sampling. The differences between QC subsamples reveal the quality of analyzes. The recommended number of QC samples is 10% of the number of original samples. At least two QC samples should always be taken.
675 To get more reliable information on contamination in groundwater, samples can be taken from several groundwater pipes/wells and several times. Geochemical baseline samples give the natural concentrations of elements at the region. Natural concentrations of several elements exceed the guide values designated for contaminated soils in many areas in Finland. The local baselines (background values) must always be verified in any assessment of contaminated sites. Ref. 6. The baseline samples have to be taken near the site of concern, from the same kind of geological environment that is not contaminated.
5 SAMPLING PATrERN Samples should be taken from both probably uncontaminated sites and potentially contaminated sites. Sampling pattern is determined after the preliminary survey. It is designed primarily case-by-case. The locations of sampling points and sampling intensity are based upon the knowledge of site conditions, such as geological variabilit, the area of the site, contaminant concentrations and migration directions. Also systematic or random sampling can be used if there is a specific reason for the applicability. Groundwater monitoring pipes/wells should be located upgradient and downgradient of the site as well as at the contaminated site.
6 SAMPLING INTENSITY The number of samples to be taken depends on the area of the site, topography and geological conditions. Every other sampling point is extended deeper -- in other words at every other point both topsample and subsample are taken. Topsamples are taken from the topsoil layer which existed when contamination occurred. Topsamples are taken from the surface to one meter deep (0 - 1 m). Subsamples are taken at the level where human activities have not extended (natural soil layers). If there is a soil layer with very low permeability (clay, clay rich in organic material or silty till) it is recommended to take subsamples from the surface of this layer or from the soil layer above bedrock. It needs to be very careful not to penetrate soil layers with low permeability. In addition, the possibility of perched groundwater needs to be considered. Drilling holes can be filled up with bentonite slurry to avoid migration. It is economical to take several samples at the same time, although all of them would not be analyzed. During the field investigation primarily individual samples are analyzed. Mixed samples may be useful in same cases, but their applicability has to be considered carefully. In practice, it is possible to mix at the most five individual samples. Groundwater samples are always analyzed individually. Table 1 shows the recommended sampling practice when the site is probably uncontaminated. When the site is probably uncontaminated, groundwater samples are taken from the nearest groundwater wells. Surfacewater samples are taken toward the flowing direction from a ditch, lake, pond or river.
676 Table 1. Recommended sampling practice for probably uncontaminated sites.
~
S
o
i
l
samples S a m p l i n g grid in h o r i z o n t a l p l a n e e.g.
(ha)
Number sampling points
Number of a n a l y z e d topsamples
Number of a n a l y z e d subsamples
3 4-9 6-13 7 - 16 8-18
3 4-9 6-13 7-16 8-18
2 2-4 3-6 3-8 4-9
9 10 11 12 13
9 - 20 1 0 - 22 11 - 24 12 - 25 13 - 27
4-10 5-11 5-12 6-12 6-13
Number of QC samples
mxm 9
75 x 75 75 x 75 80 x 80 80 x 80 90 x 90 100 x 100
1
50
x
20 22 - 24 - 25 - 27 -
If the hypothesis of the site is potentially contaminated the investigations have to ensure the contamination of different subsites and strata. In addition, it is necessary to find the distance from where on no contaminants are detected or their concentration is lower than the trigger concentration. In the table 2 is shown the recommended sampling practice for potentially contaminated sites.
Table 2. Area
(ha)
9
Recommended sampling practice for potentially contaminated sites. Soil s a m p l e s
Sampling g r i d in horizontal p l a n e e.g. mxm 25 30 35 35 40 40 40 45 45 50
x x x x x x x x x x
25 30 35 35 40 40 40 45 45 50
Groundwater samples Number of sampling points
Number of a n a l y z e d topsamples
Number of analyzed
Number of QC samples
Number of monitoring pipes/wells
subsamples
16 25 28 32 36 39
-
25 36 43 50 56 61
42 - 66 45 - 71 48 - 75
16 25 28 32 36 39 42 45 48
-
25 36 43 50 56 61 66 71 75
8-12 1214 16 18 19 21 22 24 -
18 21 25 28 30 33 35 37
3-4 3-4 3-4 4-6 4-6 4-6 4-6 4-6 4-6 >6
677 Near "hot-spots" smaller grids are recommended, e.g. 10 m x 10 m. "Hot-spots" are usually located at the places where the polluting agents are used: near the cylinder and sink at wood impregnation plant, around oil tanks etc. When the site is very large, it is economical to use large grids at first. In addition, geophysical methods can be useful before locating sampling points. The geological conditions have to be considered carefully when deciding number and location of groundwater wells.
7 CROSS-CONTAMINATION DURING SAMPLING Cross-contamination is a severe problem during sampling process. Sampling should be started from the cleaner part of the site and then proceeded to more contaminated parts. To avoid contamination the sampling equipment can be cleaned mechanically or chemically. Usually mechanical cleaning (drying, brushing, blowing, rinsing) is adequate. When contaminants are not water soluble, special cleaning agents or solvents can be used. Organic contaminants can be cleaned with such chemicals as methanol, acetone, hexane or isopropanol. Chemical cleaning is recommended before a new sampling project is started. Usually mechanical cleaning is sufficient between the sampling points. Ref. 7.
8 DISCUSSION Even though it is difficult, almost impossible to give general instructions for investigations some guidelines have to be established. In the beginning there is a strong tendency to underestimate the need of detailed data. Later, during remediation the deficiencies turn out to be expensive. The applicability of these recommendations is reevaluated in future when more experience is gathered. Currently the Geological Survey of Finland is working on research project including intensive sampling on contaminated wood impregnation plants.
9 REFERENCES
1. Draft International Standard, ISO 10381-5 Version 6, 18 July 1994. Soil quality Sampling -Part 5: Guidance on Procedure for the Investigation of Urban and Industrial Sites with Regard to Soil Contamination. ISO/TC 190/2/PG 5 Working Draft. 2. NVN 5740. 1991. Bodem. Onderzoeksstrategie bij verkennend onderzoek. Delft: Nederlandis Normalisatie-Instituut. 3. Mason, B.J. Preparation of Soil Sampling Protocols: Sampling Techniques and Strategies. PB92-220532, U.S. Environmental Protection Agency, Las Vegas, Nevada (1992). 4. Mroueh, U-M., Jarvinen H-L. and Lehto O. Saastuneiden maiden tutkiminen ja kunnostus. Teknologiakatsaus 47/96, The Technology Development Centre of Finland, Helsinki (1996). 5. Puolanne, J., Pyy O. and Jeltsch U. (eds.). Saastuneet maa-alueet ja niiden k/isittely Suomessa. Muistio 5/1994, Ymp~iristOminister6, Helsinki (1994).
678 6. Salminen, R. Raskasmetallien luonnolliset taustapitoisuudet eri maalajeissa. Environmental geology applications, The Finnish Environment 71, Finnish Environmental Institute, Helsinki (1996). 7. Naturv~.rdsv~.rket. V~.gledning for miljOtekniska markunders6kningar. Del II: F~.ltarbete. Raport 4311. Naturv~.rdsv~irket, Lindk6ping (1994).
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
679
Development of fast testing procedures for determining the leachability of soils contaminated by heavy metals J.J.M. Heynen a, C. Zevenbergen a
R.N.J.
Comans b,
A.
Honders c,
G.
Frapporti a,
J.
Keijzer a,
IWACO B.V., Consultants for Water and Environment, P.O. Box 8520, 3009 AM Rotterdam, The Netherlands ECN, Netherlands Centre for Energy Research, P.O. Box 1, 1755 ZG Petten, The Netherlands SCG, Centre for Soil Treatment, Europalaan 250, 3526 KS Utrecht, The Netherlands
Abstract In the Netherlands, the use of mildly contaminated soils, both treated and untreated, in civil and public works is regulated by the Dutch Act for Building Materials ("Bouwstoffenbesluit") [1]. By law, the leachability of inorganic contaminants and heavy metals, as determined by the column test (NEN 7343), is not permitted to exceed certain values. Drawbacks of the prescribed test procedure are the relatively high costs and the long time needed for testing. The latter adds to the logistics of soil handling and utilization. Therefore, faster and less costly testing procedures for determining the leachability of soils are required. Over the last 3 years, an extensive study has been performed on the leachability of Cd, Cu, Zn, Pb, Hg, Ni, As and cyanide in treated and untreated soils and some dredged sediments. In this study and earlier studies 230 samples of untreated and treated soils and dredged sediments were analysed for specific soil parameters, contaminant contents and leachability. The analytical data were accumulated, statistically interpreted and evaluated. Some of these soil samples were selected for geochemical speciation modelling [2], some other soil samples were selected for studies on the effects of thermal treatment on leaching behaviour [3]; the results are reported in separate papers in these conference proceedings. The study resulted in: (1) a deeper level of understanding of the chemical and physical processes governing the leachability of contaminants from (re-usable) soils; (2) the description of a less time-consuming procedure for assessing the leachability of contaminants from (re-usable) soils. The latter result can reduce the need for temporary storage and consequently can lower the handling and logistic costs considerably.
680 1. INTRODUCTION In the Netherlands, the use of secondary materials like mildly contaminated soils, both treated and untreated, is regulated by the Dutch Act for Building Materials ("Bouwstoffenbesluit")[1]. About 1.5 to 2 million tonnes of mildly contaminated soil per year are re-used in the Netherlands and an additional 1.5 to 2 million tonnes of contaminated soils are treated by various processes (mainly wet processes, e.g. classification and flotation, and thermal processes). After treatment, the soil should be suitable for re-use. If a soil is to be re-used, the leachability of inorganic contaminants and heavy metals from that soil, as determined by the column test (NEN 7343), is not permitted to exceed certain values, prescribed by the Dutch Act for Building Materials [1]. Drawbacks of the prescribed test procedures are the relatively high costs and long time required for testing (about 5 weeks). The latter slows down soil handling and delays utilization, resulting in the need for costly temporary storage. Therefore, attemps are being made to develop faster and less costly testing procedures for determining the leachability of soils. Also more insight is required into methods for optimizing soil treatment with respect to leachability. The objectives of the study described in this paper are: 1. the development of fast testing procedures for determining the leachability of inorganic contaminants from treated and untreated contaminated soils. These testing procedures need to be evaluated against the testing procedures prescribed in the Dutch regulations; 2. the assessment of the effects of soil treatment processes on the leachability of the soil. Clearly, the second objective overlaps with topics of related papers in these conference proceedings on speciation modelling [2] and leaching behaviour of thermally treated soils [3]. Therefore, the second objective will not be discussed extensively in this paper. A large number of samples of treated and untreated contaminated soils and dredged sediments have been investigated using laboratory analyses (determination of leaching characteristics with the standard column leaching test and fast leaching tests and leachingrelated parameters). The laboratory data gathered have been combined with leachability data from earlier studies [4,5,6] and have been statistically analysed.
2. TESTED SOILS AND DREDGED SEDIMENTS 2.1. Selection of soils to be studied For this study, samples of suitable treated and untreated contaminated soils and dredged materials were provided by the Dutch Centre for Soil Treatment (SCG), the Dutch Development Programme for Treatment Processes for Contaminated Sediments (POSW) and several soil treatment contractors and soil distribution centres. About 250 soils to be treated via the Centre for Soil treatment were screened on certain intake parameters. According to a prescreening in 1993, the most frequently appearing inorganic contaminants were: Cu, Zn, Pb, Cd, Hg and cyanide. Initially the study was focused on sets of treated and untreated soils and dredgings which were contaminated with the above-mentioned contaminants (in dredgings Ni was analysed instead of cyanide). The soils studied were selected on the basis of their contaminant levels. Later on, in order to
681 fill in the leachability bandwidth to be studied, an effort was made to obtain soils with elevated leaching levels (around the legal limits for leachability). From then on, nontreatable soils as well as soils re-usable without any treatment were added to the study; soils were selected on the basis of the CEN (TC 292) two-step batch leaching test and arsenic was added to the set of parameters. An overview of the origin and number of samples included in this study is given in table 1. Table 1 Origin and number of samples included in the full research programme (one combination of untreated and treated soil samples) Technology Number Soil untreated/treated wet 19 sets (n = untreated/treated thermal 7 sets (n = re-usable soil 14 samples (n = non-treatable soil 6 samples (n =
set is a
38) 14) 14) 6)
Dredged sediments
untreated/treated wet 2 sets untreated/treated biological 1 sets untreated/ripened 2 sets ripened sediments 2 samples remark: 1 two outputs (coarse and fine granular) and one input (3 samples per set).
(n (n (n (n
= = = =
61) 2) 4) 2)
2.2 A d d i t i o n a l data from other studies
The data set obtained from selected soil samples and laboratory testing was expanded using data from other studies on leaching characteristics of different types of soils. These soils were: natural soils [4]; contaminated soils from the Rotterdam [5] and Amsterdam [6] region. the leaching database of the Centre for Soil Treatment (SCG). In this way a database containing data on contaminant concentration and leachability of a total of ca. 230 samples of soil and dredged material was compiled.
3. L A B O R A T O R Y TESTING The following parameters were determined for 86 samples investigated: contaminant concentrations: Cu, Zn, Pb, Cd and Hg. Cyanide was also determined in soils treated by wet techniques (classification, flotation, etc.). Ni was determined in dredged sediments; As was determined mainly in non-treatable and re-usable soils; leachability of the above-mentioned contaminants using the column test (NEN 7343). Leachability was determined only for contaminants with soil concentrations
-
682 above Dutch background levels; leachability of the above-mentioned contaminants using the CEN TC 292 two-step batch test; availability of the above-mentioned contaminants for leaching (NEN 7341). Availability was determined only for contaminants with soil concentrations above Dutch background levels; analysis of specific soil parameters: lutum, fraction < 63 #m, organic matter, cation exchange capacity (CEC), chloride, sulphate, carbonate, phosphate, iron, sulphide, pH and electric conductivity.
4. STATISTICAL EVALUATION AND RESULTS The data gathered were stored in a database-structure and were statistically analysed with SPSS 7.0 for Windows", using non-parametric statistics. Correlations between soil parameters and leachabilities (determined by several different leaching tests) have been calculated using the Spearman rank correlation coefficient. The use of non-parametric statistics was preferred, because of the possible existence of outliers and a large number of leachabilities below the detection limits. Most striking observations are: at near neutral soil pH-values and soil concentrations below Dutch intervention levels, heavy metal leachabilities are below Dutch legal re-use standards in more than 95 % of the cases; at pH < 5, the pH of the soil is strongly correlated with the percentage of the total amount of heavy metal leached (relative leachability of heavy metals); leachability of cyanide nearly always exceeds the Dutch legal maximum limits for re-use; there is no correlation between typical soil-characterizing parameters (e.g. calcite, organic matter) and leachability; wet soil treatment (classification/flotation) and, in particular, thermal soil treatment generally lower the leachability. leachability and contaminant soil concentration are not correlated within the relevant concentration bandwidth (background level to intervention level) [7] . Figure 1 shows the leachability (column test) versus contaminant concentration for copper. the results of the column-test (NEN 7343) and the 2-step batch test (CEN TC 292) show a positive correlation. This is illustrated in figure 2 which shows the correlation between these two leaching tests for copper, which has the highest correlation. However, this correlation has a very large bandwidth, which reduces the prediction precision of the CEN-test in relation to the column-test L/S= 10 (which is the reference test in the Netherlands). a short column test L/S= 1 is also positively correlated with the complete column test L/S= 10 (NEN 7343) (illustrated for copper in figure 3). Although the bandwidth is somewhat smaller, extrapolation from L/S= 1 to L/S= 10 introduces an additional inaccuracy in the predictability, because part of the leaching occurs in the interval between L/S = 1 and L/S= 10. Remark: the leachability or emission limit (U1) for the re-use of contaminated materials, -
-
-
-
-
-
683 as stated in the Dutch Building Material Act [1], differs according to the application depth of the contaminated material. Throughout this paper the value of U1 is related to an application depth of 0.7 m. 12.6 9
3"0 O)
1
._~6.7 374.0
i"EU I : ~
.2-
09
.1-
3.9
4.6
:~
.03.02-
9
,d.~l._ 9 ",,,,
3.6.
9
4.2
9
8.0
97.7
..,& 7 ( ~ ".~,~"18, .4
9 4,,al ~.~:~ ~.~,,.~.~..o~,i;~,.;_ " .=~ ~r.-~ .~ =,
3.8
9 ~
:8;~:
E
9
4.2
o
,-
5.6
9 8.4; " ~
4.s e.S,~149
9
9
-,,., 9
7.9
~.6
7.1 9
6.7
9
"
7.e ~ . ~ ,
9
.01-
|~2
-'7 S
9t~;.7,.~6
"
7.8
9
~
6..'~..~.7 s.6..
9
9
9
7.7
O
"
= 0
12.6
:88~:
.003~
10.2
J
.002
.001 4
.~
"2"
7
o"
"o
-Oo
~o if
,Oo%
intervention level
Cu soil (mg/kgdm)
Figure 1. Cu-leachability (column test L / S = 10) versus Cu soil concentration (labels are pH-values).
~"
1,
"I "
I:~ " U 1 - - -
,,,
,-
. . . . .
. ~ . . . .
E ~"
.2
9
:8~
"
9
-
-'"
"::"
.
""
.02
~
.01
. "
:88~
.003 .002-~" .001
,, ,
~
,
i
,
,
"
,
__
Q~
,
,
,
i
.
.
.
U* for CEN
.
,
U1
Cu CEN TC 292 test (mg/kgdrn) Figure 2. Column test ( L / S = 10) leachability versus two-step CEN (TC 292) batch test leachability for copper.
684 4 E ~UI
!
:2;
II Or)
.1-
"&
.o3.
o ~
0
--
'
v o
-,.-, E E
9
.02 .01
-88~ .003 .002 .001
9
9
/
J/." %
lww
.
it|: 4!"
~
9
,
m
. ,
,
,
,
%Oo, %0,
,~
,
69
U* for L / S = 1
2'
U1
Cu short column test L/S=1 (mg/kgdm)
Figure 3. Column test leachability for copper.
5.
(L/S=10)
leachability
versus
short column
test
(L/S=1)
FAST P R O C E D U R E S F O R D E T E R M I N I N G L E A C H A B I L I T Y
Because the legal maximum leaching limits in the Netherlands are based on the column test L / S = 10 (NEN 7343), a fast procedure can only be used as a tool for predicting the results of the column L / S = 10 test. Available options are: 1. a faster test; 2. an empirical model, based on the relation between soil parameters and leachability; 3. testing at maximum contaminant concentration and pH bandwidth.
5.1 Faster leaching test -
For this option, two tests are taken into consideration: the CEN TC 292 test; the column test L/S = 1. Both tests save about 3 weeks, as compared with the full ( L / S = 10) column test.
There is a risk that a fast leaching test result will be below the legal limit whereas the mandatory test result will exceed the legal limit (U1). This risk can be set at an acceptable level: e.g. 5 percent 9 The results of the fast test should consequently be compared to
685 newly adjusted, derived limits (U*) with a 95% reliability that the result of the mandatory test will not exceed the legal limit (U1). This is illustrated in figures 2 and 3, by the 5 % risk (upper limit) line of the linear regression (data are log-transformed to achieve normality). Table 2 gives the derived leaching limits, based on the available dataset, and the percentage of results below the derived leaching limit. Table 2 Derived leaching of the mandatory (U1). metal U1 (mg/kg)
limits (U*) for the CEN and column L/S= 1 test, below which the result column test with 95 % reliability will not exceed Dutch leachability limits results < U1
(%) As Cd Cu Hg Pb Ni Zn Remark: Material
U* CEN (mg/kg)
results < U* CEN
U* L/S= 1 (mg/kg)
(~)
0.88 92 0.025 41 0.032 94 0.00017 79 0.72 94 0.12 60 0.018 90 0.00035 57 1.9 99 0.23 86 1.1 93 0.017 68 3.8 94 0.7 71 U1 (and derived U*) are based on emission limits as Act [1] for an application depth of 0.7 m.
results < U* L / S = I
(%) 0.006 0.0012 0.056 0.00032 0.04 0.018 0.38 stated in the Dutch
44 69 83 62 90 73 82 Building
The percentages below U* are comparable for both datasets. Note that the percentage below U* is lower than the percentage below U1. This is due to the chosen risk limit of 5 %. Soils with a fast test leachability above U* have a risk higher than 5 % of exceeding U1 (Dutch leaching limit for re-use) and should be tested with the mandatory column test (L/S= 10). Sufficient data have now been gathered to justify the conclusion that both fast tests can be used for Cd, Pb, Zn and Cu. With regard to As, Hg and Ni and other metals not included in this study, more data are necessary to obtain a more precise prediction. There is no statistical preference for the one or the other studied option for this fast leaching test. The CEN-test uses the same L/S-10-ratio as the full L/S= 10 column test, which means that time-dependent leaching behaviour can be compared more easily. Further the CEN-test is probably better compatible with (developing) European regulations. However, if the result of the fast test does not comply with the derived limits, a full mandatory column test L/S= 10 (NEN 7343) should be carried out. The advantage of the short column L / S - 1 test is that it is the first step of the full column test and can simply be prolonged.
5.2. Empirical model The anticipated relation between contaminant soil concentrations and leachability could not be confirmed by statistically significant correlations. A combination of concentrations and soil specific parameters (calcite, organic matter etc.) did not yield significant correlations with leachability either. Therefore it is concluded that an empirical leaching model that uses the soil characterization parameters analysed in this study is not feasible.
686 5.3. Maximum
levels
An examination of the overall database (230 soils including earlier studies [4,5,6], containing As, Cd, Cr, Cu, Hg, Ni, Pb, Zn and cyanide) reveals that only 0 (Cr) to 9 (As) % of all observations on heavy metals and arsenic exceed the Dutch legal leachability limits (U1). This percentage can even be reduced when soils comply with two preconditions: the contaminant concentration is below Dutch intervention level; the pH of the soil is above 5. For all soils that complied with these two conditions this percentage decreased to 5 % or less. This implies that over 95% of all soils (that comply with the above-mentioned preconditions), meet the standards that Dutch legislation has set for re-use in civil and public works. This finding is in agreement with the results of other recent studies, e.g. a study on the leaching behaviour of zinc in contaminated Meuse sediments [8]. It would seem therefore that leaching tests on soils that comply with mentioned preconditions may be skipped, since in 95 % of the cases their leachability will meet Dutch legal standards. Exceptions to this rule may be: thermally treated soils with respect to Mo, Sb, Se; these metals may have a higher leachability in thermally treated soils [3]; ripened dredged sediments which may have elevated leaching of sulphate, chloride and bromide [9]. However, leachability of these anions is strongly correlated with their concentration, which probably gives a good indication of the result of the column L/S= 10 leaching test (NEN 7343). soils with cyanide concentrations above background level.
-
6. CONCLUSIONS In 95 % of the 230 cases the soil samples contaminated with the following heavy metals: As, Cd, Cr, Cu, Hg, Pb, Ni and Zn, with contaminant levels below Dutch intervention levels and pH above 5, showed leaching levels below the legally set Dutch limits for reusable contaminated soils. It would seem therefore that treated and untreated soils, with heavy metal concentrations below intervention level and pH above 5, in fact comply with the Dutch Building Material Act [1]. However, soils containing other contaminants or soils which do not meet the preconditions for pH and metal concentration must still be tested for leachability. A fast leachability test can be used to obtain a prescreening. This fast test could be: (1) the 2-step CEN TC 292 batch test or (2) the short column L / S = I test. The results of these fast tests can be translated into the mandatory column L/S= 10 (NEN 7343) test by stating an adjusted leachability level below which the mandatory column test with 95% confidence level will not exceed the legal limits. There is no statistical preference for the one or the other fast leaching test. From the practical and regulatory point of view, both tests have their own advantages and disadvantages. Apparently, in most cases, a fast leachability determination procedure is possible. There will then be no need for extended temporary storage while awaiting test results. This will result in significant cost reductions for handling and logistics.
687 7. RECOMMENDATIONS A leaching protocol for treated and untreated contaminated soils and dredged materials needs to be developed within the framework of a quality assurance system. In this protocol guidelines must be given for situations where no leachability tests are required. This protocol can be based on the results of this study. The precision of the observed relations and derived reliabilities can be further improved by continued systematic gathering of data of soils and dredged materials, that become available in the future. Special attention should be given to possible critical parameters: oxy-anions in thermally treated soils and sulphate, chloride and bromide in (ripened) dredged materials and cyanide. The data currently available are reliable enough for applying a fast leaching determination procedure, as described in this paper, to treated and untreated soils contaminated with Cu, Pb, Cd and Zn.
8. REFERENCES Bouwstoffenbesluit Bodem- en Oppervlaktewaterenbescherming, Staatsblad 1995, 567. (Dutch Building Material Act). R.N.J. Comans, Modelling processes controlling metal leaching from contaminated and remediated soils, these proceedings. C. Zevenbergen et al., Immobilization of heavy metal contaminated soils by thermal treatment at intermediate temperatures, these proceedings. P.G.M. de Wilde, J. Keijzer, G.L.J. Janssen, Th.G. Aalbers, C. Zevenbergen, Uitloogkarakteristieken en chemische samenstelling van referentiegronden, RIVM report nr. 216402001, August 1992. IWACO, Uitloogonderzoek aan verontreinigde grond afkomstig uit de Rotterdamse regio, IWACO report nr. 1028100, December 1992. IWACO, Uitloogonderzoek aan verontreinigde grond afkomstig Amsterdamse regio, IWACO report nr. 1037720, November 1993.
uit
de
J. van Leeuwen, A. Orbons, E. van Gent and Th.G. Aalbers, Uitloging van zware metalen uit stads- en natuurgronden, Bodem. Vol.4, nr.2, pp. 83-86, May 1995. A.L. Hakstege, J.J.M. Heynen and H.P. Versteeg, Beneficial use of contaminated sediments within the Meuse river-system, these proceedings. J. Heynen, S. Ouboter and P. Kroes, Chemische aspecten bij het rijpen en nuttig toepassen van verontreinigde baggerspecie, Rijkswaterstaat DWW reportnr. W-DWW-96-043, May 1996.
This Page Intentionally Left Blank
Goumans/Senderffvan der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997Elsevier Science B.V. All fights reserved.
689
E L E C T R O K I N E T I C TRANSPORT IN NATURAL SOIL CORES Douglas I. Stewart, L. Jared West, S. Richard Johnston and Andrew M. Binley Abstract
ElectroMnetic transport in natural soils has been investigated by applying a constant voltage across 500mm long by approximately 200mm diameter natural soil cores for periods of up to 8 weeks. Contaminant ions were circulated through afluidfilled reservoir between the anode and the soil and distilled water was circulated through a similar reservoir adjacent to the cathode. During the experiments electrical current, voltage along the core, water flow rate, and anolyte and catholyte p H were monitored at regular intervals. Periodically, the electrical supply to the power electrodes was switched off and detailed electrical measurements were made using 68 monitoring electrodes positioned around the soil core, in order to produce three dimensional electrical resistivity images of the cores. After testing the cores were dissected and analysed for contaminant content, pore fluid composition, and pH. Data are reported that show that zinc tracer transport is initially strongly retarded, with the zinc predominately sorbed to the soil. Initially zinc enters the anode region mainly by electromigration. This cannot change the pore fluid ionic strength due to charge balance constraints, and hence zinc influx is limited by initial ionic strength. However, after 8 weeks of testing, the ionic strength of the pore fluid in the anode half of the cores were significantly elevated by co-diffusion and electroosmotic advection of the anolyte into the core. Introduction Electrokinetic treatment is a developing technology for treating contaminated land. An electric current is passed through the soil causing migration of charged species towards collection wells. In fine grained soils pore water flow is also induced. Many laboratory studies have been performed to characterise electrokinetic transport, and the dominant processes have been identified (e.g. Eykholt and Daniel, 1994; Hamed et al., 1991; Yeung and Mitchell, 1993), and insitu field decontamination has been attempted (e.g. Lageman, 1993).
This paper reports a study investigating electrokinetic transport in natural soil cores. The aims were to characterise the interactions between natural soil and the contaminant, to investigate the influence of organic matter on electrokinetic transport, and to establish the usefulness of electrical resistivity tomography (ERT) for monitoring electrokinetic transport in the soil cores. ERT is a non-invasive technique where the spatial variation of resistivity is determined from measurements taken with an array of electrodes, which during insitu decontamination would be located in boreholes or along the ground surface. Methodology
Core preparation The cores used in this study were taken from the Lancaster University Hazelrigg field station. The sub-soil at this site is an orange/brown clayey loam containing rounded gravel particles up to 15cm in diameter. The mineralogy of this soil determined by quantitative XRD using a position sensitive detector (M. Batchelder, pets. comm.) is 46% quartz, 18% kaolinite, 14% montmorillonite, 11% illite, 5% feldspar, and 6% aluminium and iron hydroxides. The
690
Figure 1: Soil core during electrokinetic testing topsoil is brown/grey silty loam, 30-40cm deep, with most of the root material in the upper 10-20cm. The moisture content of these samples were all around 20% by weight. Soil cores were carefully excavated by hand. The upper horizon of top-soil was removed (1520cm, sufficient to remove visible plant mass). Each core was exposed by careful hand excavation, aiming for a sample diameter of 200mm (actual diameters were typically 240mm). Samples of the soil trimmings were taken for analysis every 10cm over the depth of the core. An electrode shell and geotextile were then placed on the top, and the entire core and electrode shell were coated with fibre-glass resin and fibre glass matting was applied. This was repeated twice more, and a final coat of resin was applied. The core was then left overnight for the resin coating to dry. The next day, it was under-cut and inverted, and sealed in polythene to prevent moisture loss. Once in the laboratory, each core was trimmed, a second reservoir shell was positioned and sealed with glass fibre and resin. Next, 68 monitoring electrodes were inserted by drilling through the fibre-glass. Finally, the power electrodes and reservoir packing (a drainage geogrid) were inserted into the electrode shells, and the core was saturated by upward flow under approximately 1m head. Prior to testing, each core was positioned horizontally and the reservoirs were filled with the electrolyte solutions (see Figure 1).
Electrokinetic testing During testing a constant voltage of 30V was applied between the power electrodes, and reservoir header-tank weights, electric current, and the voltage along the core were recorded at regular intervals. Test durations were nominally 2 weeks, 4 weeks, and 8 weeks. Periodically data were gathered for ERT image reconstruction (while electrical supply to the power electrodes was disconnected) using a computer controlled earth resistance meter. A total of 1860 'four-electrode' measurements were taken using combinations of the 68 monitoring electrodes, in each of four radial planes and two axial planes. Each 'four-electrode' measurement consists of driving current between two electrodes and measuring the resulting potential difference between the other two electrodes. Electrode polarisation is minimised by using a low frequency alternating current. ERT data acquisition took approximately 4 hours. After testing, the cores were sliced into 50mm thick layers. Each slice was then photographed from the top and bottom, before the slice was homogenised. The pH of each slice was
691 measured by inserting a pH penetration electrode directly into the soil. Samples of each slice were then oven dried at 105~ to determine the moisture content, and the samples were analysed for elemental composition using X-ray Fluorescence Spectroscopy. Samples for XRF analysis were ground and compressed into pellets for analysis. Pore water was extracted by centrifuge displacement (at approximately 50,000g) using an immiscible organic solvent (technical grade tetrachloroethylene). The pore water samples were passed through a 0.45 micron filter, and sub-samples to be analysed for metal content were acidified. The pore fluid pH and conductivity were measured. Pore fluid cation and silica concentrations were measured by Induction Coupled Plasma Atomic Emission Spectroscopy (ICP-AES), and anion concentrations were measured by Eluent Suppressed Ion Chromatography (using a carbonate eluent). Total carbonate concentrations in the pore fluids were measured separately using the flow injection analysis procedure described by Hall and Aller (1992).
Electrical resistivity tomography data processing. A resistivity distribution was found for each set of four-electrode measurements by numerical non-linear inversion. The specimen was represented by a three-dimensional finite element mesh of 2288 triangular prismatic elements. This includes additional layers of elements at both ends of the specimen to account for the change in resistivity in the reservoir electrolyte. The resistivity of each finite element is treated as a variable for the inversion process. The data inversion employs a weighted least squares approach with regularisation. No constraining of the procedure using other measurements such as reservoir fluid conductivity was employed. More details of the three-dimensional ERT image reconstruction can be found in Binley et al. (1996). Results Data summarising each test is presented in Table 1. Voltage profiles taken from two rows of monitoring electrodes showed that the voltage applied across the cores were dropped steadily along their length. Variations in core conductivities were not discernible from this data. After problems with electrode degassing had been resolved, the electrical current through the cores was between 50 and 65mA and did not vary much during testing.
The moisture content in the anode half of the cores did not change much from field value of 20%, but in the upper, organic rich horizon it increased to between 30-40%. This increase in the moisture content was probably the result of pre-test core saturation. Electroosmotic flow data for test HR1 were erratic during optimisation of the test conditions. After optimisation, flow rates in specimens HR2 and HR3 varied between 100 and 300ml/day and tended to reduce during the test. Table 1" Test details Test [[ Duration
U
(days)
Charge passed (kC)
HR1 16 32 HR2 29 124 HR3 57 277 Note: Applied voltage was 30V in all three tests
Energy (k J) 968 3,820 8,464
Average EO permeability (cm2/V.s) 1.6xl 0 .5 0.Sxl 0-5
692
[-.._~,__ HR1 ~
HR2
~
HR3
5000 _ "~ 4000_ O'}
E
3 0 0 0 __
.m (9
> 2000 u .c_ 1 0 0 0 . N 0
0
0.2
'" 0.4
Normalised
-
: -0.6
-
I'0.8
--
', 1
dist. f r o m a n o d e
Figure 2: Total zinc level after core testing Figure 2 shows total zinc profiles from tests HR1-3 (the background zinc level was below 40mg/kg everywhere). The anolyte was originally Zn(NO3)z at a concentration of 1,000mg/1 of zinc. Figure 2 shows that zinc was transported into the cores from the anolyte. After 2 weeks of voltage application, the zinc level had risen to over 2000mg/kg in the anode slice, with a much smaller increase in the next 50mm slice. After 4 weeks, the level in the anode slice had reached about 4000mg/kg, and zinc levels were also elevated in the next two slices. After 8 weeks, zinc levels were elevated to around 4000mg/kg in all but the cathode slice (where it had only increased to about 500mg/kg). The initial soil pH decreased with depth from about 6.5 to 5. The soil pH after testing was typically about 5, with some increase towards the cathode, reaching pH 8 in the cathode slice of the core tested for longest (HR3). The pH of the catholyte was about 11 at the end of all the tests. Any Zn 2+ approaching the cathode reservoir is therefore likely to have been precipitated. Figure 3 shows the conductivity of the pore fluid from the background samples, and that from cores HR1-3 after testing (the values for normalised distances of 0 and 1 represent the reservoirs). In the background soil, the conductivity was highest (1600~tS/cm) near the ground surface (corresponding to the cathode end of the cores), dropping over 250mm to a value of 200~tS/cm. Pore fluid conductivity was steady at about 200ktS/cm over the next 250mm (corresponding to the anode half of the cores). After 2 weeks of testing, there was little change in pore fluid conductivity. After 4 weeks, the pore fluid conductivity in the anode slice had increased to 700~tS/cm, but with little change over the rest of the core. After 8 weeks of testing, the pore fluid conductivity in the anode half of the specimen had risen to nearly 1500ktS/cm, with an even larger increase in the anode slice (3000~tS/cm). The conductivity near the cathode was below the background value although this may reflect natural variability. Figure 4 shows the pore fluid composition of the background samples, and at the end of tests HR1-3. In the background soil, the pore fluid has elevated concentrations of calcium and nitrate near the ground surface (corresponding to the cathode end of the cores). Other ions present in the pore fluid from background samples were sodium, potassium, chloride, sulphate
693 o 14000
~"
12000
bg
=
HR1 ~ H R 2
.__ HR3
t
10000 8000 6000
4000 o
2000 0
0
0.2
0.4
0.6
0.8
1
Normalised dist. from anode
Figure 3" Conductivity of the extracted pore fluid and carbonate. The elevated concentrations of Ca 2+ and NO 3- near the surface may reflect fertiliser application. The data from core HR1 show that, after 2 weeks of testing, the pore fluid compositions are similar to those from the background soil. There were small increases in Zn 2+ in the anode slice and of natural cations (K § and Na +) in the cathode reservoir (relatively high calcium and nitrate concentrations in core HR1 probably reflect natural variability). The carbonate present in the cathode reservoir was probably HCO 3- produced by interaction of electrolytically produced OH- with atmospheric CO 2. The data from core HR2 shows that, after 4 weeks of testing, the pore fluid compositions remained similar over most of the core, although there was an increase in Zn 2+ in the 100mm nearest to the anode, and there was a significant increase in Ca 2+ in the cathode reservoir. The data from core HR3 show that, after 8 weeks of testing, there were elevated Zn 2+ levels in the two-thirds of the core closest to the anode and relatively low calcium and nitrate levels near the cathode. However, the pore fluid zinc levels remained much less than the total levels, which shows that most of the zinc was sorbed on contact with the soil. Figure 5 is a plot of the Zn 2+ in pore fluid against sorbed zinc, for those parts of all three cores which were composed of orange brown clay loam (i.e. excluding the topsoil). Despite the scatter, it can be seen that Zn(II) exhibits Freundlich sorption behaviour where the sorbed concentration is less than 4,000 mg/kg. Sorbed concentration did not exceed this value for higher pore fluid Zn 2+ levels. Figure 6 shows ERT images of core HR2 at intervals during testing (the elements representing the fluid reservoirs are included on the images). The main feature that is evident from the images is an increase in the catholyte conductivity due to electrolytic decomposition of water. The zone of conductivity increase in the images does not correspond exactly with the reservoir. This smearing is typical of ERT reconstruction for zones outside the electrode
694 ;
Zn
=
Ca
=
Na
,t
K
e
NO3
o
CO2
o
CI
A SO4
(a) lO
T
O/
--
-
0
i
i
0.2
-
0.4
!
-
,
0.6
0.8
--.
I
1
(b)
lO 0
-1-
0
(c)
=-
A
,
~
~
,
A
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
7o 6o
E
~ 40 "~ 30 "~ 20 o
lO o
(d)
60 50 40 30 20 10 0
N o r m a l i s e d dist from a n o d e
Figure 4" Composition of pore fluid (a) background, (b) HR1, (c) HR2, and (d) HR3
695 10000 9
m
A 9
b
9
1000
o'J
E
G =
. , -
N o9
100 9
L_
o
10
9
, HR1 1 0.01
I 0.1
9HR2
9HR3
]
I
I
I
I
I
I
10
100
1000
10000
P o r e Fluid Zinc, mg/I
Figure 5" Pore fluid zinc concentration versus sorbed zinc for the clayey loam arrays. Apart from this effect, the images show that the conductivity of the soil changed little over four weeks of testing. Discussion
The measured pore fluid anions and cations in cores HR1 and HR2 are approximately in charge balance, indicating that all major ionic species have been identified. The measured ionic concentrations indicate a large excess of positively charged species in most of core HR3, and in the cathode reservoirs at the end of tests HR2 and HR3. The ion chromatograph results for these regions have unidentified peaks, the most significant of which had a relatively long residence time in the ion chromatography column. It is inferred that interaction of dissolved species with either the equipment or the soil released (as yet) unidentified anionic species, which migrate towards the anode. Ideal solution conductivities were calculated from the ionic compositions using the dilute solution ionic mobilities (e.g. see Reiger, 1994). The ideal solution conductivities were typically higher than the measured pore fluid values, with the difference being largest where the ionic strength of the pore fluid was highest. This is consistent with the non-ideal behaviour of real electrolytes. The fractional contribution of each ion to the extracted pore fluid conductivity (its transport number) was calculated by assuming that (i) deviation from ideal behaviour within the pore fluid from each slice was the same for all ions and (ii) that all the major ionic species had been identified (this was not the case in core HR3). Within the cores, other effects such as diffusion and electroosmotic flow will modify transport numbers from those calculated for the extracted pore fluid (Alshawabkeh and Acar, 1996). Figure 7 shows the major contributors to pore fluid conductivity (defined here as ions with transport numbers greater than 0.1) in the cores. Each sub-region indicated in Figure 7 has the same major contributors, although ionic concentrations varied substantially within subregions. These diagrams, although an approximation to the main charge carriers in the cores, are a useful for identifying the major processes during electrokinesis.
696
Anode reservoir
Cathode
Soil c o r e
......... ; L
.a~,,~P-ffs;r;oir
q o cm
t=0
t = 7 days
t = 28 days
t = 14 days
Resistivity (ohm m) 10
20
30
40
50 60
70 80
Figure 6: ERT images of core HR2 The pore fluid ionic strength after two weeks of electrokinesis had not changed much from background values, despite the increased ionic strength in the reservoir (anions produced by electrolysis and cations from the core raised the level in the cathode reservoir). The major processes that have occurred are electromigration of zinc from the anode reservoir, replacing the natural pore fluid calcium near the anode, and electromigration of HCO 3- from the cathode reservoir, replacing nitrate and sulphate near the cathode. After 4 weeks, the pattern remains similar, although increases in pore fluid strength are now apparent in both the anode and cathode slices. These local increases probably result from codiffusion of anions and cations from the reservoirs, and in the anode slice, electroosmotic advectiono Charge balance constraints prevent electromigration alone from causing these ionic strength variations (i.e. zinc can only enter the anode region by the electromigration mechanism at the same rate that natural cations are leaving it).
697 Background inn
Ca 2+
Ca2+
NO 3SO42-
NO 3-
Anode Res.
Soil
Cath. Res.
Core HR1 .|
ill
Zn 2+ Ca 2+
Ca 2+
NO3- ] NO3SO42-
SO42-
SO42- t NO 3" NO 3- SO42-
Zn 2+ | Zn 2+
Zn 2+
Zn 2+ Na +
Ca 2+
Ca 2+
Ca 2+
Ca 2+
NO 3- NO 3-
SO42C1-
SO42-
SO42-
NO 3SO42-
NO 3-
HCO 3-
Zn 2+
Ca2+! Ca 2+
Ca 2+
Ca 2+
K
+
Na +
NO 3-
NO 3- HCO 3-
Core HR2 Ca 2+
_
HCO 3-
Core HR3 Zn 2+
Zn 2+
NO 3- NO 39-
Zn 2+
?-
Zn 2+
Ca 2+
SO42- HCO 3HCO 3-
?- indicates a region where there are unidentified anions. Figure 7: Distribution of major contributors to extracted pore fluid conductivity
Ca 2+
9-
698 After 8 weeks, the ionic strength in the anode half of the core had risen substantially, but had reduced slightly in the cathode half. The ionic strength in the anode half rose as anions accumulating in this region (due to diffusion and electroosmotic advection) are not used up in precipitation or electrode reactions. This allowed higher concentrations of Zn 2+ to enter from the reservoir The pore fluid data show that the requirement for continuity of current and charge balance within the cores prevents electromigration alone from altering ionic strength (it can only substitute ions for others of equivalent charge). Changes in ionic strength at a specific location are therefore the result of advection or diffusion of electrolytes to that location, or dissociation/neutralisation and dissolution/precipitation reactions at that location.
Conclusions O Initially, most of the zinc tracer entering the core was sorbed to the soil, whereas after 8 weeks of testing the sorption capacity had been exceeded across 60% of the specimen, permitting larger pore fluid zinc concentrations. | Initially, there was little change in pore fluid conductivity. This is because the dominant transport process is electromigration, which alone cannot alter ionic strength. | Later, co-diffusion and electroosmotic advection of the anolyte into the cores progressively increased the ionic strength of the pore fluid. Acknowledgements The authors would like to acknowledge the support of the U.K. Engineering and Physical Sciences Research Council through grant GR/K57770. M Batchelder, Natural History Museum is thanked for the quantitative XRD analysis. Appendix. References Alshawabkeh, A.N. and Acar, Y.B. (1996). Electrokinetic remediation. II: theoretical model. ASCE Journal of Geotechnical Engineering, 122(3), 186-196. Binley, A, Pinheiro, P. and Dickin F., (1996), Finite Element based Three-Dimensional Forward and Inverse Solvers for Electrical Impedance Tomography, In: Proc. Colloquium on Advances in Electrical Tomography, Computing and Control Division, lEE, Digest No. 96/143, p6/1-6/3, Manchester, June, 1996. Eykholt, G.R. and Daniel D. E. (1994). Impact of system chemistry on electroosmosis of contaminated soils. ASCE Journal of Geotechnical Engineering, 120(5), 797-815. Hall, P.O.J. and Aller, R.C. (1992). Rapid small volume flow injection analysis for total CO2 and ammonium in marine and freshwaters. Limnology and Oceanography, 37, 1113-1119. Hamed, J., Acar, Y.B. and Gale, J.G. (1991). Pb(II) removal from kaolinite by electrokinetics. ASCE. Journal of Geotechnical Engineering, 117(2), 241-271. Lageman, R. (1993). Electroreclamation: applications in the Netherlands. Environmental Science and Technology, 27, 2648-2650. Reiger, P.H. (1994). Electrochemistry. Chapman and Hall, 2nd ed. Yeung, A.T. and Mitchell, J.K. (1993). Coupled fluid, electrical and chemical flows in soil. Geotechnique 43(1), 121-134.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
699
Re-use of Sieve sand from demolition waste E v e r t Mulder
T N O - Waste Technology Division P.O.Box 342, 7300 AH Apeldoorn The Netherlands Phone: +31 55 5493919, Fax: +31 55 5493287
Abstract Sieve sand originates from activities as sorting and/or breaking of demolition waste. In a breaking process the first step is a sieve step, to remove the fines. Next, the coarse material is broken and upgraded to a secondary raw material, which can be used as a coarse aggregate in concrete, or as a road construction material. The fine material (sieve sand) cannot be applied without due consideration. The sieve sand may be contaminated with Poly cyclic Aromatic Hydrocarbons (PAHs) to such an extent that, according to the Dutch Building Materials Decree, the sieve sand is not applicable as a granular building material. For this reason Van Bentum Recycling Centrale ordered TNO to carry out a research into the possibilities of stabilising/solidifying the sieve sand in such a way that the PAHs are fixed (immobilised). In the Building Materials Decree the organic contaminants are assessed on the basis of total concentration, in mg/kg. However, from an environmental point of view, not the total concentration, but the leaching is relevant. For this reason, the impact on the environment of the application of stabilised sieve sand has been assessed in terms of mg leached per square meter surface area by TNO. The results of the leaching tests performed, show that even highly contaminated sieve sand (containing up to 1,000 mg/kg PAils) can be sufficiently stabilised. Only 0.7 mg/m 2 (being 0.002 % of the total concentration) is being leached during the 64 days lasting Dutch diffusion test.
Introduction In The Netherlands most of the demolition waste is being upgraded to a secondary raw material and re-used as a coarse aggregate in concrete, or as a road construction material. To obtain a material of high quality the demolition waste is sieved first, to remove the fines. This fine material (sieve-sand) should not be disposed of, but utilised, both from an economic and environmental point of view. On the other hand, it cannot be applied without due consideration. The sieve-sand may be contaminated with Poly cyclic Aromatic Hydrocarbons (PAHs) to such an extent that, according to the Dutch Building Materials Decree, the sieve-sand is not applicable as a granular building material (as will be explained in the next paragraph). For this reason Van Bentum Recycling Centrale ordered TNO to carry out a research into the possibilities of stabilising/solidifying the sieve sand in such a way that the PAHs are fixed (immobilised). The intention is to use the stabilised sieve sand to heighten a piece of land for use
700 as an industrial area. The stabilised sieve sand has to be environmentally assessed as a monolithic material, on the basis of leaching. In this paper the characteristics of sieve sand will be described first. Then a description will be given of the stabilisation process, that was used to immobilise especially the PAHs. After that the results of a leaching test on the stabilised sieve sand are given and the stabilised material is environmentally assessed. Finally the paper ends with some conclusions and a recommendation.
Sieve sand and its leaching characteristics Sieve sand originating from a breaker of demolition waste, consists for the greater part of sand and small concrete and ceramic brick particles. Besides, it may also contain small particles of wood, roofing material or plastics. Though these kind of "physical" contaminants are relevant for the strength development of the stabilisation, environmentally the "chemical" contaminants are of more importance. The most critical chemical contaminants are PAHs and sulphate. The PAHs are present in tar-containing particles, originating from for instance roofing material, chimneys or tar containing asphalt concrete. The sulphate originates from mortar and plasterboard. For the research, described in this paper, a sieve sand sample was used containing high concentrations of contaminants, as a worst case. Only the PAH content was measured. The sample contained 1000 mg/kg PAH (the 10 of the Dutch Ministry of VROM). This is much more than is allowed by the Dutch Building Materials Decree. As for the environment, not the total content of contaminants in a material is important, but only that part of the contents that will leach out, in its application. For that reason in The Netherlands the environmental assessment of building materials is primarily based on leaching (at least for inorganic contaminants). Because of the fact that for organic components standardised leaching tests are not available yet, the assessment of organics is still based on total content. Nevertheless, in this paper also PAHs will be environmentally assessed on the basis of leaching. For this reason part of the sieve sand sample was leached in a column test for organic components, in accordance with the draft standard NVN 7344. In this column test, up to a liquid / solid (L/S) ratio of 10 1/kg, totally 0,26 mg/kg PAHs were leached. Even if the Dutch Building Materials Decree should assess the utilisation of granular (unbound) sieve sand on the basis of leaching (instead of total concentration), this sample of sieve sand would not be allowed to be utilised. The calculated immission of PAHs would be two times higher than the limit value of a category 2 application [1]. However, from an other investigation it follows that a sample with a lower total concentration of PAHs (490 mg/kg) has leaching characteristics that would allow it to be utilised as a category 2 construction material [2]. Not only the leaching of PAHs was determined, but also the leaching of inorganic contaminants. The leaching of the heavy metals that were investigated (for instance arsenic, barium, molybdenum and antimony) was far below the limit values of the Building Materials Decree (even the category 1 limit values). On the other hand, the leaching of sulphate was beyond the category 1 limit value, but below the category 2 limit value. From these investigations it can be concluded that highly contaminated sieve sand should not be utilised as such, from an environmental point of view. According to the Building Materials Decree, the total concentration of PAHs is too high (even though this is environmentally less relevant). Nevertheless, also an assessment based on leaching indicates that this sieve sand
701 should not be used as such, because of the leaching of PAHs. Besides, the leaching of sulphate exceeds the limit value of a category 1 application. Therefore stabilisation is a must.
Stabilisation of PAH-containing sieve sand A research has been carried out to find the best binding agent for the stabilisation / solidification of the sieve sand, described in the previous paragraph. Four different agents were tested in two addition percentages (5 % and 10 %): 9 blast furnace slag cement; 9 geo-cement (a cement produced from secondary raw materials); 9 a mixture of blast furnace slag cement and an additive; 9 a special cement, specifically developed to bind high concentrations of organic components. Three criteria were used in finding the best binding agent. First of all the mechanical strength after 28 days of hardening had to be beyond 5 MPa. The specimens with 5 % binding agent added, did not fulfil this criterion. Of the specimens with 10% binding agent 3 kinds had a pressure strength of more than 5 MPa. The only one that did not, was geo-cement. The second criterion was the leaching of PAHs from the stabilised material, determined by means of the diffusion test (tank leaching test). The leaching of PAHs of the remaining three types of specimens differed from 0.7 mg/m2 surface area (blast furnace slag cement with additive) to 3.6 mg/m2 (special cement). The third criterion was the price of the binding agent. The blast furnace slag cement was much less expensive than the special cement. Concerning the additive, this was little more expensive than the blast furnace slag cement itself, but the mixture performed better than the cement as such, so the mixture of blast furnace slag cement and the additive was chosen as the best binding agent. In the experiments mentioned in the following paragraphs, stabilised sieve sand was investigated, that was mixed up with 9% blast furnace slag cement and 1% of an additive. The material was compacted well in cubes of 10 * 10 * 10 cm. After 28 days of hardening the density of the material was 1900 kg/m3. The pressure strength was 6 MPa.
Leaching of PAHs from stabilised sieve sand The leaching characteristics of a stabilised, monolithic material are in The Netherlands determined by means of the diffusion test (or tank leaching test). This test is standardised for inorganic components (NEN 7345). In this standardised test the specimen (or product) is immersed in five times its volume leachant, consisting of demineralised water, beforehand acidified to a pH of 4 with nitric acid. This leachant is renewed at 0.25, 1, 2.25, 4, 9, 16 and 36 days after the start of the test. At 64 days the test is finished. The eight eluate fractions are filtered, measured (pH and conductivity), conditioned and analysed on relevant components. From the analysis results for each component the quantities leached at the eight different times are calculated (expressed in mg/m2 surface area). These emissions are then plotted against time (on a double logarithmic scale). If the leaching indeed is diffusion controlled, this should yield a straight line with a regression coefficient of 0.5. If so, a diffusion coefficient can be calculated from the leaching data. This diffusion coefficient can be used to predict a diffusion controlled leaching (emission) in course of time by extrapolating the leaching with time.
702 In the research, described in this paper, the leaching behaviour of all inorganic components, mentioned in the Building Materials Decree, was investigated. Only a small number of components could be detected (Ba, Cu, Mo, Ni, Sb, V, Zn, CI, SO4). The results (emissions) are given in table 1. In the Building Materials Decree the organic components are assessed on the basis of total concentration, in mg/kg. However, from an environmental point of view, not the total concentration of a contaminant in a building material is important, but its leaching from the building material. For this reason, the impact on the environment of the application of stabilised sieve sand has been assessed in terms of mg leached per square meter surface area by means of the diffusion test. In principal this test was performed in accordance with NEN 7345, but additionally, some pre-cautions were taken to prevent the degradation and/or the absorption of leaching PAHs. The pre-cautions were: 9 The leaching vessel was made of glass and covered to avoid evaporation of the more volatile PAHs. 9 The leaching vessel was packed in aluminium foil, to prevent degradation of PAHs by ultraviolet radiation of sunlight. 9 The eluates were filtered in teflon filter devices, by means of pressure filtration, to avoid absorption of PAHs in the device. 9 The eluates were put in brown flasks in between the time of sampling and the time of analysis, again to avoid degradation of PAHs. The leaching of PAHs, determined in this way (according to NEN 7345, with additional precautions), proved to be diffusion controlled, on the analogy of the inorganic components. The diffusion coefficient was very low (1,5 * 1015 mE/sec). The emission in 64 days was 0.68 mg/m 2.
Environmental assessment of PAH-leaching In table 1 not only the emissions (in terms of mg/m 2 product surface area) are given, but also calculated immissions, in terms of mg/m 2 soil surface area. These immissions are calculated in order to be able to compare the leaching test results with the limit values of the Building Materials Decree. These limit values (Maximum Allowable Immissions into the soil), are based on the principle of "Marginal Burdening" of the soil. This means that the upper meter of soil may not be contaminated by leaching from building materials in there application by more than 1% of the target values for soil. The Building Materials Decree distinguishes two categories of applications, one without provisions to prevent rain water from coming into contact with the building material (Category 1) and one with those provisions (Category 2). These provisions can be a non-permeable clay liner or a plastic liner.
703 Table 1" Leaching characteristics of stabilised sieve sand Component Emission Immission Immission (calculated) Cat. 1 application Cat.2 application in mg/m 2 in mg/m 2 in mg/m 2 As Ba Cd Co Cr Cu Hg Mo Ni Pb Sb Se Sn V Zn Br C1 CN F SO4 PAHs
< 0.33 16" < 0.07 < 0.67 < 0.33 0.6 < 0.66 1.4 1.1 < 0.67 0.38 < 0.66 < 1.3 5.1 10 < 66 2400* < 20 < 66 51000" 0.68*
< 1.5 170 < 2.2 < 7 < 3.4 6.3 < 1.5 3.6 11 < 7.0 4.0 < 1.5 < 3.0 49 100 < 370 4100 < 150 < 690 85000 7.1
< 0.5 54 < 0.7 < 2.2 < 1.1 2.0 < 0.5 1.1 3.6 < 2.2 1.3 < 0.5 < 1.0 15 33 < 120 1300 < 47 < 220 27000 2.3
Max. Allowable Immission in mg/m 2 435 6300 12 300 1500 540 4.5 150 525 1275 39 15 300 2400 2100 300 30000 75 14000 45000 15
* = calculation based on a diffusion coefficient From table 1 it can be learned that the calculated immissions for most inorganic components are below the maximum allowable immission values, even for a category 1 application, except for SO4. For the anions Br and CN this is no sure, because the analysis techniques are not able to determine such low concentrations yet. The sulphate immission for a category l application is higher than the limit value, whereas the calculated immission for a category 2 application is still below the limit value. So, the stabilised sieve sand should be considered a building material that can be utilised in category 2 applications only (because of the leaching of sulphate). Also for PAHs immissions have been calculated, even though the Building Materials Decree does not give a maximum allowable immission for PAHs (because the Building Materials Decree assesses PAHs on the basis of total content). So, to be able to assess PAHs on the basis of leaching, a "maximum allowable immission" for PAHs had to be derived. This was done, starting from a target value of soil for PAHs of 1 mg/kg and following the same route as was done in the Building Materials Decree for inorganic components. In that way a maximum allowable immission of 15 mg/m 2 was calculated for PAHs.
704 The leaching of PAHs if very low, compared with the total amount. Only 0.002 % of the PAHs present in the stabilised sieve sand are leached during the 64 days lasting diffusion test. From table 1 it can be learned that, if PAHs would be assessed on the basis of leaching, the stabilised sieve sand would not have any problems to fulfil the criteria, even for a category 1 application. The results of this leaching research show that the stabilisation process is capable to decrease the leaching of PAHs to such an extent that that it can be considered harmless to the environment. This is in contradistinction to the conclusion of an assessment on the basis of total content of PAHs. For this reason it is highly recommended to environmentally assess organic components in building materials on the basis of leaching, on analogy to inorganic components.
Conclusions 9 Highly contaminated sieve sand may not be utilised, neither if assessed on the basis of total content of Poly cyclic Aromatic Hydrocarbons (PAHs), nor if assessed on the basis of leaching of these PAHs. 9 This highly contaminated sieve sand (containing up to 1,000 mg/kg PAHs) can be stabilised well by adding 9% of blast furnace slag cement and 1% of an additive. Only 0.7 mg/m2 (being 0.002 % of the total concentration) is being leached during the 64 days lasting Dutch diffusion test. 9 If the stabilised material has to be assessed on the basis of total content of PAHs the material may not be utilised still (not the presence has been effected, but its mobility). 9 However, if the stabilised sieve sand would have to be assessed on the basis of leaching, the material could be utilised in a category 2 application (because of the relatively high leaching of sulphate). The leaching of all other components (PAHs inclusive) is below the limit values of category 1 applications. Recommendation It is highly recommended to environmentally assess organic components in building materials on the basis of leaching (on analogy to inorganic components) and not on the basis of total content. Literature [1]
Zijlstra, R.K., and E. Mulder, Comparison of the shake test (CEN) and the column leaching test (NEN), TNO report No. TNO-MEP - R96/450 (in Dutch), December 1996.
[2]
Zijlstra, R.K., and E. Mulder, Determination of the leaching of PAHs with the column leaching test, TNO report No. TNO-MEP- R96/400 (in Dutch), November 1996.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997Elsevier Science B.V. All rights reserved.
705
Organic substances in leachates from combustion residues I. Pavasarsa, A.-M. FAllmanb, B.
Allard a
and H. Bor6na
aDepartment of Water and Environmental Studies, Link6ping University, SE-581 83 Linkoping, Sweden bSwedish Geotechnical Institute, SE-581 93 Link6ping, Sweden
Abstract
The release of water soluble organic substances from combustion residues (bottom ashes from municipal solid wastes and from wood) has been followed in laboratory batch leaching experiments. Leachates have been sampled during 70 days (single-step procedure) or frequently during first 24 hours (starting after 10 min), followed by a change of leachate solution and continued sampling during 70 days (two-step procedure). The leachates were analysed for conductivity, pH, Eh, TOC and metals (Cu and Cr). High concentrations of dissolved organic matter were obtained already within 24 hours (up to 30 mg/1 of TOC for the solid waste bottom ash). However, a rapid loss of the TOC from the solution (half-life of 50 days) was also observed. The release of copper appeared to be governed by the simultaneous release of organic carbon, while the release of chromium was independent of TOC.
1. INTRODUCTION The technical use of ashes and slags from combustion of various materials (municipal solid waste, wood etc.) is limited by the potential environmental effects due to releases of hamafial components from these products as a result of leaching by percolating precipitation. Most efforts to characterize and quantify the leaching properties of ashes and slags have been focussed on the release of inorganic components under various conditions, not considering the simultaneous release of soluble organic agents. However, the total contents of organic carbon in solid ashes from combustion of e.g. mixed municipal wastes and wood are usually several percent, assessed from measurements of loss of ignition. Total concentrations of organics in leachates in the 10-100 mg/1 range have been reported [ 1, 2]. There are also reports on the composition of the hydrophobic fraction, leached by organic solvents [3]. The aim of the present study is to assess concentrations and eventually metal complexing properties of readily soluble hydrophilic organics in leachates from combustion residues (municipal solid wastes and wood). Effects on the release of copper and chromium related to the simultaneous release of organic matter will be quantified.
706
2. MATERIALS AND METHODS 2.1. Materials Bottom ash from municipal solid waste incineration (BA) and wood (primary wood chips and secondary wood materials) firing (WA) were obtained from energy production plants at LinkOping, Sweden. All ash samples were dried at 50~ and crushed to a size of
1000
~
b)
200 100 0
t
0,1
1
t
t
10 100 Time, h
t
1000
10000
Figure 1. Conductivity as a function of time in BA (top) and WA (bottom) leachates (a) Single-step leaching (b) Two-step leaching
3.3. Redox potential Redox potentials of-100 to -150 mV corresponding to pe + pH of 6.5 to 7.5 were obtained at the start of the BA leaching, but increased to around 200 mV after 24 hours (pe + pH of 12). A new low level of around -100 mV (pe + pH of 10) was observed after the change of leachate solution in the two-step procedure (see Figure 3). This level increased slowly with time to a stable level of around 400 mV (pe + pH of 15-16). The redox potential in the closed bottle of the single-step procedure was-450 mV after 24 hours (pe + pH of 2), increasing with time to the same level as in the two-step procedure (pe + pH of 1516). Initial potentials around 400 mV were obtained in the WA system with minor changes with time (constant pe + pH of 15-16). Potentials corresponding to pe + pH of 15-16 (in all WA leachates as well as in the BA leachates after a long time) simply reflect a system in contact with the atmosphere (E ~ of 0.95 V).
709 10,0
Bottom ash
9,5 -r" ct.
9,0 8,5 []
8,0
i
0,1
10,0
I
1
I
o
Q.
~
9,0
~
u
1000
~
....~ ~ - - . -
9,5 "I-
I
10 100 Time, h
Wood ash
(a)
~.,
10000
(b)
--
8,5 8,0
I
0,1
1
I
t
i
10 100 Time, h
1000
10000
Figure 2. pH as a function of time in BA (top) and WA (bottom) leachates (a) Single-step leaching (b) Two-step leaching 20
Wood ash
15"I-
Bottom ash
(a, b) ~. (b"
/~
Q.
+ 10
/
0
I
0,1
1
I
(a) I
10 100 Time, h
Figure 3. pe + pH as a function of time in BA and WA leachates (a) Single-step leaching (b) Two-step leaching
I
1000
10000
710 3.4. Release of organic carbon The accumulated release of carbon, determined from the measured TOC-concentration, is given in Figure 4. TOC-concentration in step two (change of leachate after 24 hours in the two-step procedure) is added to the final value after 24 hours in step one. There were immediate releases of organic carbon into the leachates of both systems, giving TOClevels of 20 and 4.5 mg/l for BA and WA, respectively, already after 10 minutes of exposure. The TOC-concentrations had reached levels of 30 and 5.5 mg/1 after 24 hours, corresponding to a release of around 150 and 30 mg/kg from the BA and WA, respectively. 180 Bottom ash ,~n 160 m 140 "~ 120 E 100 d 80 i Wood ash 60 ttj
~
t~
n-
40
20 0
~
~~-.'~r 0,1
1
~-~=~"='-~
+
*
..... ~
(b)
~a) (b)
=- - ~ ~
~~ " ~ ' ~ -
10 100 Time, h
1000
(a) 10000
Figure 4. Organic carbon release as a function of time in BA and WA (a) Single-step leaching (b) Two-step leaching The carbon release continued after the change of leachate in the two-step procedure, and leveled out at a total release of around 170 mg/kg for the BA system, based on the TOC-concentration. Changes in TOC-concentrations were minor in the WA system in the second step of the two-step procedure. The high levels of the released organic carbon in the BA system represented only some 0.8% of the maximum organic carbon inventory (assessed from the LOI). The released fraction from the WA system was only some 0.03%, and the concentrations much lower than for the BA system, despite the fact that the carbon content was almost four times higher in the WA. The TOC-concentration was decreasing with time, after the initial fast in-growth, in the singlestep leaching for both the BA and WA systems. Half of the maximum TOC-concentration was lost alter about 50 days in the BA leachate, but already after 4 days in the WA leachate. Preliminary results (BA only) indicate the following distribution of leachable (by 0.1 M NaOH) organic matter: 25-30% hydrophobic acids, 5-10% other hydrophobic agents, 10-15% hydrophilic acids and 45-55% other hydrophilic agents. 3.5. Release of Cu and Cr
The accumulated releases of copper and chromium, determined from the measured concentrations, are given in Figures 5 and 6. The release in step two (after change of leachate after 24 hours in the two-step procedure) are added to the values after 24 hours in step one.
711 1,4 ~
Bottom ash
1,2
~
9aa
a ~ [] (b)
13) ~'-~ 9 1,0 o3
~
E =- 0,8 o 0,6 "o (!.}
m n,'
'
~
~
(a)
Wood ash
0,4
o
0,2
.8-
o
A
o ~
(a, b)
o
0,0 0,1
1
10 100 Time, h
1000
10000
Figure 5. Copper release as a function of time from BA and WA (a) Single-step leaching (b) Two-step leaching 0,5 O3
0,4-
03
Wood ash
jm..-~ [] [ ] ~ ] = r ~ o
E 0,3t...-
o -o 0,2ttl
m
0,1-
n,'
Bottom ash
(a, b)
0,0 0,1
1
10 100 Time, h
1000
10000
Figure 6. Chromium release as a function of time from B A and WA (a) Single-step leaching (b) Two-step leaching There was an immediate release of copper into the leachates of the B A systems, giving concentrations of around 120 mg/1 already after 10 minutes and 190 mg/l after 24 hours. Concentrations significantly above the blank values (10-20 rag/l) were not observed in the WA leachates. The copper release continued after the change of leachate in the two-step procedure, and leveled out at a total accumulated release of 1.3 mg/kg for the BA system. This represents around 0.04% of the inventory. The copper concentration was decreasing with time, after the initial fast in-growth, in the singlestep leaching for the B A system. A reduction to half of the maximum copper concentration was observed after about 10 days in the B A leachate. There was an immediate release of chromium into the leachates of the WA systems, giving concentrations of 40 mg/1 already after 10 minutes and 50 mg/1 after 24 hours. Concentrations significantly above the blank values (2-4 rag/l) were not observed in the BA leachates. The
712 chromium release continued after the change of leachate in the two-step procedure, and reached a total release of 0.4 mg/kg for the WA system. This represents around 0.4% of the inventory. The chromium concentration was decreasing slightly with time, after the initial fast in-growth, in the single-step leaching for the WA system.
4. DISCUSSION
The increase of conductivity with time indicates a similar dissolution process for the two ashes. The high conductivity obtained already after 10 min indicates a rapid release of soluble salts. Notable is the fact that most of these salts were released during the initial 24 hours (above 90% of the conductivi'ty after 64 days in the single-step procedure), in fact, largely within the first 10 minutes (60 and 80% of the final conductivity for BA and WA, respectively). The pH-development of the WA system indicated a continued release of hydroxide which was not compensated by the generation of acids, including CO2 contribution from the atmosphere. In the BA system, however, a pH-maximum of 9.2 was obtained already within one hour. A pH decrease towards 8.5 and below in this system after 64 days reflects an inflow of CO2 from the atmosphere into the system and, probably, also a release of acidic organic material. There is an approach towards a CaCO3-dominated system at constant CO2-pressure, which would have an equilibrium pH of 8.3 8.4. The decrease in pH was more pronounced in the single-step leachate, where the readily released TOC was not removed as in the two-step procedure. The considerably higher release of TOC from the BA in comparison with the WA could be one of the reasons for the different pH-developments. The differences in other pH-controlling systems (Ca, Mg, as well as Na-K) indicate possibly a higher content of alkali hydroxides in the BA-systems, which otherwise would give a higher pH in the BA than in the WA leachates, disregarding the potential effects of organics. The low redox potential initially obtained in the B A leachate, particularly in the single-step procedure after 24 hours of exposure in a closed system, indicates the presence of elements in their reduced state in the solid BA. The lowest measured potential (-450 mV) corresponds to an E~ of around 0.1 V, which could be representative of an Fe(III)/Fe(II)-couple. A significant fraction of the iron in the BA would be Fe(II), while no reducing capacity is indicated for the WA systems (with considerably lower total content of iron than the B A). As a consequence, chromium would be expected to exist primarilly as Cr(III) in the BA, while a significant or dominant fraction of Cr(VI) can not be excluded in the WA. The reduction of the TOC-concentration in the single-step leaching after the initial fast release of carbon indicates either a loss of volatile organics or adsorption on solid surfaces or possibly a degradation to carbon dioxide. The presence of a large fraction of leachable volatile organics is not likely in high-temperature combustion residues. A substantial adsorption or binding of organic agents directly after a release during the leaching is not probable. A degradation through microbial processes seems to be a likely explanation to the rapid loss of TOC. This degradation would contribute to the pronounced pH-decrease observed in the single-step systems in contrast to the second step in the two-step procedure, where the initially released TOC-fraction is removed by change of leachate solution. It is evident that the organic carbon fraction is of different nature in the two materials, reflected by the differences in leachability and degradation rate. Only a minor fraction of the high organic content of the WA is leachable by water. The maximum copper concentration (280 mg/l, corresponding to 4.4x10 -6 M) is 1-2 orders of magnitude above expected total solubility, considering complexation and potentially solubility
713 limiting secondary solid phases (hydroxide and possibly hydroxy carbonates). The similarity in leaching behaviour and concentration change with time between copper and TOC is striking. The appearent over-saturation can be explained, assuming that around 1% of the TOC represents a strong complexing agent with a complexing capacity of 6-7 meq/g, which is not unreasonable. Similar enhanced releases related to the presence of organics have previously been claimed [7]. The reduction in appearent copper concentration with time in the single-step procedure can either be due to a decreasing solubility or a loss of copper due to adsorption. The decreasing pH (c.f. Figure 2) could actually lead to an enhanced adsorption of an organic complex. The loss of organic carbon assumed to be due to microbial degradation and a related reduced total solubility is, however, more likely as an explanation to the decreasing copper concentration. The leaching behaviour of chromium in the WA system has some similarity with the behaviour of copper in the BA system. The slightly decreasing concentration of chromium with time in the singlestep leaching could be due to interactions with organics analogous to the copper system, although less pronounced. The absence of significant chromium releases from the BA system, however, indicates a predominant dependence on the redox conditions (c.f. Figure 3) and the oxidation state of chromium in the solid matrix. The existence of chromium predominantly as Cr(III) of low solubility in the B A would lead to a slow leaching-rate and low over-all chromium concentrations. In the WA, which apparently has a minor reducing capacity, chromium may exist partly as Cr(VI), which would be more mobile and soluble as compared to Cr(III).
5. CONCLUSIONS An immidiate release of organic carbon compounds was observed by the exposure of the BA and WA to water. The readily released organic fraction was partly lost from the systems with time, possibly as the result of microbial degradation. The long-term leaching of the organic inventory of the ashes and subsequent TOC-degradation in solution should be further analysed. The potential metal solubilizing effects of organic matter released from the ashes by leaching has been demonstrated for copper. Further studies of the nature of these organic agents (composition, complexing properties and chemical stability) are required in order to allow an assessment of the over-all importance of the organic fraction for release and mobilization of metals. The possible existence of strong complexing agents that may significantly affect both leaching rates and solubilities of certain metals is of particular interest.
6. ACKNOWLEDGEMENTS Financial support from the Swedish Waste Research Council (AFR), as well as from the Swedish Association of Waste Management (RVF) is gratefully acknowledged. Mr. J. Rogbeck and Mr. L. Larsson, Swedish Geotechnical Institute, both participated in the early planning of this project.
7. R E F E R E N C E S
1 P.H. Brunner, M.D. Mueller, S.R. McDow and H. Moench. Total organic carbon emissions from municipal incinerators. WasteManagement & Research (1987) 355-365
714 2 B.S. Shane, C.B. Henry, J.H. Hotchkiss, K.A. Klausner, W H. Gutenmann and D.J. Lisk. Organic toxicants and mutagens in ashes from eighteen municipal refuse incinerators. Arch. Environ. Contain. Toxicol. 19 (1990) 665-673 3 H. Belevi, N. Agustoni-Phan and P. Baccini. Influence of organic carbon on the long-term behaviour of bottom ash monofills. In Proc. Forth International Landfill Symposium, Cagliari (1993) Environmental Sanitary Engineering Centre, Cagliari 4 A.-M. F~illman and J. Hartl6n. Leaching of slags and ashes - controlling factors in field experiments versus in laboratory tests. In J.J.J.M. Goumans, H.A. van der Sloot and T. Aalbers (eds), Environmental Aspects of Construction with Waste Materials, Elsevier, Amsterdam, (1994) 39-54 5 I. Pavasars, A.-M. Fallman, B. Allard and H. Bor6n. Work in progress 6 Standard Methods for the Examination of Water and Wastewater. American Public Health Association, American Water Works Association, Water Pollution Control Federation, Washington (1985) 7 R.N.J. Comans, H.A. van der Sloot and P.A. Bonouvrie. Geochemical reactions controlling the solubility of major trace elements during leaching of municipal solid waste incineration residues. In J Kilgroe (ed.), Municipal Waste Combustion Conference, Air and Waste Management Association, Pittsburg (1993) 667-679
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
715
Leaching Behavior of P C D D / F s and PCBs from Some Waste Materials
S. Sakai, S. Urano, H. T akatsuki Environment Preservation Center, Kyoto University, Yoshida Honmachi, Sakyo-ku, Kyoto-city, 606-01, Kyoto, Japan
Abstract Although it is well known that some waste materials and their incinerator residues contain persistent organic pollutants (POPs) such as PCDD/Fs and PCBs, little attention has been paid to the leaching behavior of these chemicals because of their low leachability. Due to the co-existence of surfactants in wastes, however, leaching concentration of POPs may increase. Therefore, leaching tests with and without those substances were conducted in order to understand the influence of surfactant-like substances on POPs leaching. In those tests, LAS (Linear Alkylbenzene Sulfonate) and humic acid was used as surfactant-like substances. Shredder residues from car/electrics recycling and fly ash from a municipal solid waste (MSW) incinerator were used in content analyses and leaching tests. Furthermore, an experiment was carried out to understand the influence of fine particles to the leaching concentration of POPs. The results of the leaching tests indicate that surfactant-like substances increase the leaching concentration of POPs, and fine particles related closely to the transporting behavior of POPs.
1. Introduction Although it is well known that some waste materials and their incinerator residues contain PCDD/Fs and PCBs, the leaching behaviors of these persistent organic pollutants (POPs) have not been studied well because they seldom solute with water due to their low solubility. It has been reported that the leaching behavior of POPs with co-existent substances, which are surfactant-like substances, solute salts, and so on, is not the same as their behavior without them ':). Clarifying the influences of these co-existences and gaining an understanding of the maximum leaching concentration, or so called availability of POPs, is therefore considered to be imPortant. Since surfactants increase solubilities of hydrophobic substances, it is thought that the leaching behavior of POPs will be affected by these chemicals at disposal sites. In this study, content analysis and leaching tests with LAS and humic acid solutions have been conducted in order to understand leaching behaviors and availabilities of PCDD/Fs and PCBs from automobile shredder residues and MSW incinerator residue. Furthermore, the influence of fine particles on POP leaching concentration has also been investigated. From these results, a leaching test procedure for POPs is discussed.
716
2. Experiments 2.1 Samples In this study, shredder residues and fly ash from a munid pal solid waste indnerator (MSWI) were used for content analysis and leaching tests. Shredder residues, the remains of valuables recovered from waste automobiles and discarded electronic goods, total more than one million tons per year in Japan. Most of them have been landfilled in non-controlled landfill sites where they stayed until March of 1996. About five hundred thousands tons of shredder residues have also landfilled illegally on Teshima Island in Kagawa Prefecture, some of which have been burned in the open. Shredder residues contain PCBs used in capacitors and some other materials until 1972 in Japan. It has been reported that openly burned shredder residues contain PCDD/Fs z~. Three types of shredder residues, two kinds from junked waste automobi le shredder residues and one discarded electronic goods shredder residue, were used in this study. Shredder residue K and Y, originated by automobiles, were sampled in 1988 and 1995, respectively. Shredder W, shredder residue from electronic goods, was sampled in 1995. Table 1 shows the results of physical fraction analysis. After classifying 1 kg of residue according to its physical fractions, the weight of each fraction was measured. Automobile shredder residues consist of high percentage of fine materials, mainly glass and others, and 90 percent of the electronic goods shredder residue was plastics. MSWI fly ash was a sample from a continuous type incinerator with an incineration capacity of 200 ton/clay, and equipped with a mechanical stoker and electric precipitator. Table 1 Physical Fractions of Shredder Residues Mass Percntage Fractions
Shredder residue K from automobile
Shredder residue Y from automobile
Shredder residue W from electronic goods
Hard Prastics
17.1
7.5
36.6
Soft Prastics
4.1
8.9
54.0
Rubber and Leather
11.6
10.2
0.0
Fine Materials (Soil, Glass, Wood, Paper, Texture)
61.6
51.6
0.01
Metals
5.6
21.8
9.4
2.2 Con tent Analysis From 1 kg of shredder residue, the weight of samples were reduced as uniformly as possible, and then PCDD/Fs and PCBs were measured. 20 g of MSWI fly ash was picked, and after one hour ultrasonic treatment in 2 m mol/l HCI, PCDD/F and Co-PCB congeners, PCB homologue were measured.
717
2.3 Leaching Test 2.3.1 Test Series Table 2 Solvent in Leaching Test Table 2 shows solvents in the leaching test. LAS and humic acid, leachate from an Solvent Concentration [mg/l] industrial waste landfill site, and distilled water Distilled Water as a control experiment were used in the 10 leaching test. LAS, Dodecylbenzene sulfonic LAS (DBS) acid sodium salt (DBS), one of the main 1000 ingredients used in a home-use synthetic 10 Humic Acid detergent, was also used in this study. Although 200 the biodegradability of LAS is relatively high, Leachate from Humic Acid 8.4 it has been detected widdy in the water Indutrial Waste environment. At landfill site, LAS may be LAS 0.33 Landfill Site contained in industrial waste and sewage sludge. On the other hand, humic acid, which is a very stable substance formed in nature by degradations of organic substances with high molecular weight, has been detected in landfill leachate in Japan ranging from tens to hundreds mg/1. Although humic acid doesn~ have a designated structural formation generally, it has many benzene rings and various function groups LAS : Tokyo Chemical Industry Co., Sodium Dodecylbenzenesulfonate (D1238), soft type, 65 % in water and humic acid : Tokyo Chemical Industry Co., Nitrohumic Acid (H0161 ) were used in this study. Experimental solvent conditions set up high and low concentrations in the leachant. The low one has a detectable concentration in the environment, and the high one is set up considering with Critical Micelle Concentration (CMC) . It is well known that if the micelle is formed, leaching concentration of hydrophobic substances increases surrounded by surfactants "~ 2.3.2 Leaching Test Method In the leaching tests for shredder residues, each residue was picked up according to its original mass percentages of fraction, and after cutting all samples under 10 mm in a diameter, each component was mixed for use in leaching tests. In order to prevent a sampling variation, twice the weight of each component was taken and cut it in half almost completely in a series of leaching tests. Furthermore, leaching tests were carried out three times, and after centrifuge of the leachates respectively, supernatants were mixed. Although the leaching test time for waste material is regulated to 6 hours in Japan, it was set up for 24 hours of horizontal shaking for the purpose of reaching equiliburium. In the leaching test for MSWI fly ash, 100 g of a sample, whose particle size was under 4 mm, was mixed with leachant on the condition of L/S (liquid per solute ratio)=l 0 with 200 mg/1 humic acid solution, with 24 hours of horizontal shaking. Separation of solid and liquid after shaking was carried out with a centrifugal separator in this study because a large amount of soil and fine particles obstructed filtration. Therefore,
718
after an obtained leachate was filtered with glasswool to remove large materials, the leachate was centrifuged at 700 G, 10 minutes. This conditi on was determined by the sedimentation velocity formula to remove over 0.45 ~m particles. Particle density was required for this calculation, and it was measured with a pycnometer.
2.3.3 Experiment of the Influence of Fine Particles This experiment was carried out in order to understand the influence of fine particles on POPs leaching concentration. Fine particles are suspended particles, not those sedimented with gravity. In this experiment, a leachate obtained from another leaching test was divided into two fractions. Each half of the leachate was centrifuged and filtered, respectively and then PCDD/Fs and Co-PCBs congeners, PCBs homologue, and particle distribution were measured. From these results, the influence of fine particles on POPs leaching concentration was studied. This leaching test was carried out by shredder residue Y under conditions of I/S= 10 with 500 mg/1 LAS solution. 2.4 Analytical Method 2.4.1 PCDD/Fs and PCBs PCDD/Fs and PCBs were measured by HRGC/HRMS. Clean-up procedure was followed according to the Analytical Manual of PCDDs/PCDFs in Waste Management in Japan '~. GC columns were Spelco SP-2331 for low-chlorinated PCDD/Fs (from tetra to hexa) and J & S Science Co. DB-5 for PCBs and high-chlorinated PCDD/Fs (hepta and octa). 2.4.2 Particle Size Distribution Measurement Particle size distributions were measured using a Shimazu Particle Distribution Meter with a Centrifugal Separator (SA-CP3). Since the leachate contained LAS, it was washed three times with distilled water, and measured with the Particle Distribution Meter on after the addition of a dispersion reagent. This meter has a variable velocity centrifugal separator, which can measure particle concentration based on absorbances.
3. Results and Discussion 3.1 Con tent Analysis Table 3 shows the contents of PCB homologue in shredder residues. PCBs concentration in shredder residue K and Y ranged from 1,800 to 11,000 ng/g and 15,000 to 24,000 ng/g, respectively. Even the same shredder residue contained a rather varied PCBs content due to a difference in fractions. On the other hand, PCBs concentration in the electronics shredder residue was 1,200 ng/g, which shows low concentration in comparison with automobile shredder residues. As for PCDD/Fs in shredder residues, those were all under quantity limit, though the detectable limit is 0.1 ng/g, due to oily chemicals. On the other hand, PCDD/Fs were detected in the previous leaching tests on the LAS 1,000 mg/l series, which shows shredder
residues contain PCDD/Fs, though these were not detected in the content analysis because high detectable limit. Our recent experiments on a different shredder residue, however, indicate that it contained 0.25 ng-TEQ/g of PCDD/Fs. Table 3 Results of PCBs Contents Analysis for Shredder Residues [ng/g] Shredder Residues S.R. K Homologue
first
S.R. Y second
first
S.R. W second
M~CBs
2.1
9.6
19
27
1.3
~CBs
200
800
1300
1600
80
T3CBs
720
4200
4600
6800
320
T4CBs
450
3100
3700
5300
310
PsCBs
280
1600
3400
6400
340
H6CBs
140
770
1600
3200
170
H~CBs
21
120
280
600
25
O~CBs
2.1
28
70
1.8
N9CBs
0.2
1.1
3.1
6.0
0.2
D,~---'B Total CBs
0.3 1800
N.D. < 0.1 11000
N.D. < 0.1 15000
N.D. 40 mm. Fly ashes from Municipal Waste Incineration are kept separate from the MWIbottom ash. In the Dutch situation it is forbidden to prepare mixed ashes from fly ash and bottom ashes. In 1996 800,000 tons of MWI-bottom ash were produced in the Netherlands. The last years MWI-bottom ash is utilized for 100%, primarily in granular form as embankment material up to a hight of 10 m or more or as a road base material.
744 MWI-bottom ashes are supplied to the market with a certificate for its technical and environmental behaviour. The environmental part of this certificate is based on "old" legislation. MWI-bottom ashes up to now always comply with the demands for environmental certification. The Building Materials Decree enforces more severe demands than the present regulations. Because of that a large part of the MWI-bottom ashes does not comply with the demands from the Building Materials Decree. To safeguard its outlet to the market the Dutch Ministry of the Environment has developed a "Special Category for MWI-bottom ashes". In this category MWI-bottom ashes can be utilized under a set of isolation measures. With the Municipal Waste Incineration sector the appointment has been made to pursue steady quality improvement of its byproducts so that MWI-bottom ashes can be utilized as Category 2 Building Materials in future. 4.2
Critical elements
Since 1987 MWI-bottom ashes have been subjected to a regular quality control from which the environmental part is based on the serial batch test NEN 7349. This test is however not the compulsory test for the Building Materials Decree. Since 1991 all the Dutch Municipal Waste Incineration plants have also carried out column tests (the compulsory test for the Building Materials Decree) on their MWI-bottom ashes. The purpose was to build up sufficient leaching data to be able to prepare environmental certification according to the Building Materials Decree and to show the extent of quality improvement that has been realized in the run of years. In table 3 the leaching data for 1996 are shown.
745 Table 3 Component
As Cd Cr Cu Mo Ni Pb Sb Zn Br CI SO4
Leaching data for 26 column tests from MWI bottom ashes during 1996 Average leaching (mg/kg) O.054 0.003 O.O77 1.96 2.09 0.17 0.16 0.20 0.16 10.4 3040 5360
Standard deviation (mg/kg) 0.195 0.003 0.22 1.27 2.91 0.30 0.42 0.15 0.24 6.0 B.a. B.a.
Limit Category 2 BMD (mg/kg) 7.0 0.061 11.7 3.27 0.84 3.5 8.2 0.42 14 4.0 8800 22000
Variation coefficient
(%)
361 105 291 65 139 176 260 72 149 58 B.a. B.a.
From table 3 it can be concluded that the following critical elements exist for MWI-bottom ashes: Cu, Mo, Sb and Br. Based on the average leaching of chloride > 0.25 * U2, chloride could also be considered as a critical element. However, during the total period 1991 - 1996 the leaching limit for category 2 has only once been exceeded. A significant reduction in copper leaching has been effected between 1991 and 1996. For the critical elements the distribution of the leaching data has been established. The data can both be described by a normal and a log normal distribution. Because the fitting for a normal distribution seemed slightly better, the testing criteria have been based on a normal distribution of the results. Based on the leaching behaviour of bromide, presently all MWI-bottom ashes should be considered as MWI-bottom ashes. For most Municipal Waste Incineration plants also the leaching of molybdenum exceeds the Category 2 limits.
746 4.3
Procedures
for certification
An assessment guideline for certification is under development. Two classes for certification are distinguished, viz. Category 2 MWI-bottom ash and Special Category MWI-bottom ash. A definite assessment guideline is expected in 1997. The assessment guideline exists of the following aspects: -
method of sampling
moving stream
-
sample size
180 kg
-
number of increments
2O
-
sample frequency
1 sample per 2, 3 or 6 weeks, depending on the yearly quantity of MWI-bottom ash per MWI plant
-
-
-
certification method category 2 MWI-bottom ash:
production certification
special category MWI-bottom ash
production certification
limit correcting measures category 2 MWI-bottom ash
Xmoving average, 8
critical elements
Cu, Mo, Sb, Br
~>limit Category
2
The chosen method of certification is production certification. Based on the historical percentage of lots that fully meet with the demands for Category 2, the MWl-bottom ash of a specific Municipal Waste Incineration Plant is integrally classified as Category 2 MWI-bottom ash or as Special Category MWI-bottom ash. More than 75% has to meet with the demands for category 2 to be classified as Category 2 MWI-bottom ash. After the first occasion that the moving average for a Category 2 MWI-bottom ash exceeds the limits of category 2, correcting measures are taken to improve the quality. If within a set number of successive analyses the moving average is not brought down below the category 2 limit, the production certificate for category 2 can be (temporarily) withdrawn. 4.4
Actual
status
of certification
All MWl-bottom ashes are supplied with a certificate according to the existing regulation. Certification according to the Building Materials Decree will be introduced after the enforcement of the leaching limits from the Building Materials Decree. This enforcement is set for 1998.
747 MWI-bottom ashes of all Municipal Waste Incineration plants should be considered then as Special Category MWI-bottom ash. This is based on the leaching behaviour of bromide (always) and molybdenum (for most Municipal Waste Incineration plants) during 1996. 4.5
Further development of certification
The following aspects for MWI-bottom ash certification are under development; -
development of a short test. The possibilities of the CEN TC 292 test as a short test for quality control are considered. A comparison of the results from the column test and the serial batch test (from which a large set of data is available) has been carried out in 1996. More than 50% of the datasets batch-test / column test were not fit for comparison. A translation serial batch - column seems possible, but appears not reliable enough in the case of data near a leaching limit
-
demonstrations with quality improvement measures are carried out (for instance accelerated aging).
5
DISCUSSION
Both CF-bottom ashes and MWI-bottom ashes show a leaching level for the most critical element around one of the leaching limits. Because of that aspect, production certification is preferred; lot by lot inspection may lead to a high percentage of rejection, whereas a production certificate safeguards a more continuous supply. Leaching levels exceeding the limit set by the Building Materials Decree are allowed, provided they are compensated by other parts of the production, showing a lower leaching level. The use of short tests as an alternative for the compulsory column test is under development. The inherent lower reliability that is caused by the translation of one test to another may be "smoothed out" in the case of a moving average method.
748 LITERATURE
LAMERS, F.J.M. and BERG, J.W. VAN DEN, 1995. Environmental certification of pulverized fuel ash and bottom ash. Proceedings Power Gen Europe '95, Penwell, pp. 581-598 MINISTRY OF HOUSING AND ENVIRONMENT, 1995. Building materials Decree for the soil protection and surface water protection (in Dutch). NNI, 1995. NVN 7301, Leaching characteristics of solid earthy and stony building and waste materials. Sampling. Sampling of granular materials from streams. NNI, 1995a. NEN 7343, Leaching characteristics of solid earthy and stony building and waste materials. Leaching tests. Determination of the leaching behaviour of inorganic components from granular materials with the column test (in Dutch). NNI, 1995b. NEN 7349, Leaching characteristics of solid earthy and stony building and waste materials. Leaching tests. Determination of the leaching behaviour of inorganic components from granular materials with the cascade test (in Dutch).
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
WASCON 97 June 4 - 6, 1997 Houthem
St G e r l a c h
The Netherlands
W o r k s h o p DF2 -Quality Control and Certification
Q u a l i t y A s s u r a n c e in t h e i.,,~u,,. o h . ~ .,.~,,, ~ - ~ . :, A n a l y s i s o f Soils
Leslie Heasman M J C a r t e r Associates Station House Long Street Atherstone Warwickshire UK. CV9 IBH Tel +44 1827 717891 Fax +44 1827 718507
749
750
Quality Assurance in t h e
Analysis of Soils
Land is a resource that is scarce in many countries. In the industrialised countries land which becomes available for development in or near to the industrial and business centres of cities and towns has usually been used before and land which remains under use has frequently been subject to a variety of previous uses. Historically little consideration was given to the potential environmental impacts of industrial activities hence the controls over processes and emissions were not as careful as they are today. In order to minhnise the use of green field land for new developments, companies are encouraged to redevelop previously used sites. This results in a need to measure the contamination present at the site and to assess the associated environmental risk. Based on an assessment of risk proposals are made for the remediation deemed necessary to return the site to beneficial use. The entire process of site assessment and remediation frequently involves a number of different professionals with very different backgrounds and experiences. These include: 9 9 9 9 9 9 9 9 9 9 9 9
landowners developers financial advisers bankers insurers land agents/surveyors legal advisers/lawyers planners architects/engineers regulators environmental advisers chemists/laboratories
The different backgrounds and degree of knowledge of each of the parties involved in the redevelopment of potentially contaminated sites can create a lack of confidence by each of the various parties in the capabilities, knowledge and skill of the others. It is only when some degree of confidence is achieved between the different professional interests and some degree of trust is established that the main objectives of brown land redevelopment are likely to be achieved. The area of the analysis of contaminated soil is one in which it is vital to ensure confidence in data as it is the baseline on which risk assessments and remediation programmes are founded. Little work has been carried out on methodology for the analysis of contaminated soils and there is generally a poor understanding by the users of laboratory generated data of the confidence which can be attached to the data generated by any one laboratory. Many examples are cited of extreme discrepancies between data generated by different laboratories asked to analyse the same samples and between duplicate samples analysed in the same laboratories. A number of initiatives are underway in the UK to improve confidence and understanding between the laboratories which analyse contaminated soils and those who use the data in an assessment of the environmental impact of the sites from which the soil was sampled.
751 A report has been prepared by the UK Environment Agency ~ on quality assurance in the analysis of soils from contaminated land. The report is intended for users of laboratory generated data and seeks to describe and explain the quality control procedures used in analytical laboratories. The report reviews the existing application of quality assurance in the area of contaminated soil analysis. The different quality assurance procedures are explained and their benefits outlined. Recommendations are made for best practice regarding quality assurance in this technical field. Aspects of the general process that are addressed include sampling, transportation of samples, sample preparation, analysis and reporting of data. A report is in preparation by the UK Environment Agency 2 on the available laboratory methods used for the analysis of contaminated soil. The report is intended to inform users of laboratory data of the existing choice available in analytical methods and the different data which is generated. The effect of the use of different extraction procedures is explained as is the effect of using different analytical techniques. The report identifies the benefits and limitations of each of the methods used currently in the UK for the analysis of priority contaminants in soils. There is much discussion and confusion regarding the use of different analytical methodologies for the analysis of the same parameter. It is considered by some that consistency and comparability can only be achieved by the selection of single analytical methodologies for specific contaminants. This is not generally the view of the laboratories as this approach would eliminate the potential to develop and improve methods and would reduce flexibility. It is generally agreed by laboratories and the infor,ned users of laboratory data that provided that a method can be validated according to an accepted protocol with specified acceptance criteria standard methods need not be used. Work has been carried out funded by the UK Environment Agency3 to develop a validation protocol for methods used in the analysis of contaminated soil. The protocol provides guidance on criteria which can be used to assess whether analytical method performance is satisfactory and on the approach laboratories can take to provide data which can be assessed against these criteria. The validation protocol will provide a means by which non standard methods can be validated or standard methods can be validated for different soil types. Work has been carried out funded by the UK Environment Agency4 to identify and coordinate research related to the laboratory analysis of contaminated soil. It is concluded that lhnited research and development work is carried out in a number of areas related to the laboratory analysis of contaminated soils however the work is disparate and not well coordinated. A report on the work considers the drivers for improved methods of analysis for contaminants in contaminated soils which include the need for reliable data which is fit for purpose: to inform policy to support regulation for effective management of contamination 9 to reduce the commercial risks associated with land transfer and development
752 The report identifies the key organisations involved in funding and managing research into the laboratory analysis of contaminated soils and in identifying and improving existing analytical methodologies. Existing or planned relevant research progrmrunes are identified. The report includes proposals for the communication of research outputs and for liaison and coordination between the different parallel research programmes. A further initiative which has developed in the UK to improve confidence in the assessment and redevelopment of contaminated land is the formation of a body, The Forum on Contamination in Land (FOCIL), which coordinates the approach of the main professional institutions to this technical area. The FOCIL mission statement is to enhance the
understanding of and facilitate improved coordination between professionals dealing with contaminated land thus benefiting the process of assessing and managing risks associated with contamination, both commercially and environmentally. Key objectives of FOCIL include: to expand awareness, knowledge and competence among professional advisers and develop an integrated approach to contamination in land to identify, disseminate and encourage the use of best practice to assist market confidence, enhance environmental benefits and encourage sustainable development The activities of FOCIL include: 9
dissemination of information on available best practice guidance co-ordination and communication meetings with other groups involved with contaminated land
9
involvement in development and preparation of best practice guidance
9
hosting seminars and workshops on relevant topics and issues provision of a unified professional response to Government and other bodies on contamination issues
9
maintenance of listings of specialists in contaminated land matters
In order to improve cormnunication and understanding between the users of laboratory data and those who request analyses and use the data in their assessment of potentially contaminated sites FOCIL produced the sheets presented in Annex 1 to ensure that both parties involved in the analytical process are aware of all the relevant facts and have all the information necessary to ensure that the analyses carried out are fit for purpose and that the resultant data are of a known quality.
753 References
Laboratory of the Government Chemist, M J Carter Associates, Acer Environmental, H B Berridge and Partners. Quality Assurance Associated with the Analysis of Soil from Contaminated Land. UK Environment Agency. In press 1997. .
.
.
Methods for the Chemical Analysis of Soils from Potentially Contaminated Land. UK Environment Agency. In preparation 1997. Hyder Environmental, M J Carter Associates, H B Berridge and Partners. Validation of Methods for the Analysis of Soils from Potentially Contaminated Land. Recommendations from the Working Group on Soil Chemical Analysis. UK Environment Agency. In preparation 1997. M J Carter Associates, Hyder Environmental, H B Berridge and Partners. Framework for the Identification, Prioritisation and Coordination of Research Related to the Laboratory Analysis of Contaminated Soil. UK Environment Agency. In preparation 1997.
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved. W . M . A . J . W i l l a r t , I n i s t r y of The Hague, T h e N e t h e r l a n d s .
Dutch policy materials
as
incentive
Housing,Spatial
to
environmentally
755 Planning
and
controlled
the
Environment,
re-use
of
waste
Introduction The b u i l d i n g i n d u s t r y has a l o n g h i s t o r y o n the r e u s e of o l d m a t e r i a l s . T h e f i r s t c h r i s t i a n c h u r c h e s w e r e o f t e n b u i l t o n the s i t e of o l d t e m p l e s , w i t h the b u i l d i n g materials of t h e s e t e m p l e s . In o l d c i t i e s buildings and b u i l d i n g m a t e r i a l s h a v e b e e n r e n o v a t e d a n d r e - u s e d . T h e u s e of i n d u s t r i a l w a s t e s as b y - p r o d u c t s is of a m o r e r e c e n t date. B l a s t f u r n a c e s l a g a n d f l y a s h are u s e d in cement, r o a d b a s e s o f t e n w e r e m a d e w i t h s l a g or b u i l d i n g rubble. A f t e r the s e c o n d w o r l d w a r r u b b l e was u s e d in G e r m a n y a n d T h e Netherlands as a g g r e g a t e for the b u i l d i n g of n e w h o u s e s . In t h i s p e r i o d f r e s h b u i l d i n g m a t e r i a l s w e r e s c a r c e a n d the r u b b l e h a d to be r e - u s e d or landfilled. E n v i r o n m e n t was n o t an i s s u e at t h a t time. S o m e t w e n t y y e a r s ago the reuse of waste materials became politically interesting. The a s p e c t s of l e s s e n i n g the a m o u n t of f r e s h l y d e l v e d m a t e r i a l s on one h a n d a n d the a r e a o c c u p i e d b y l a n d f i l l on the o t h e r h a n d w e r e the d o m i n a n t f a c t o r s . The f i r s t activities were directed to l a r g e w a s t e streams as b u i l d i n g r u b b l e a n d f l y a s h f r o m coal f i r e d p o w e r s t a t i o n s . T h e u s e of f l y a s h w a s s t i m u l a t e d b y a b a n on l a n d f i l l for f l y ash. T h e c e m e n t i n d u s t r y w a s p u t under pressure to u s e f l y a s h in the n e g a t i o n s for e x t e n s i o n of t h e i r m i n i n g c o n c e s s i o n . O n this p r o d u c t the p r o d u c e r s of the a s h a n d the user, the c e m e n t i n d u s t r y h a d a m u t u a l i n t e r e s t a n d all f l y a s h in The N e t h e r l a n d s is u s e d mow. B u i l d i n g r u b b l e w a s a t o u g h e r s t r e a m to t a c k l e . The number of p a r t i e s i n v o l v e d was l a r g e r a n d the q u a l i t y of the p r o d u c e d m a t e r i a l s v a r i e d l a r g e l y . '~ogether w i t h the i n d u s t r y the g o v e r n m e n t w o r k e d on standards, certification of different materials. A very positive influence had the p r e s c r i p t i o n to u s e the m a t e r i a l in g o v e r n m e n t and m u n i c i p a l p r o j e c t s . At this t i m e r o u g h l y 90% of the b u i l d i n g w a s t e s a r e reused a n d s i n c e j a n u a r y this y e a r we h a v e a b a n o n the l a n d f i l l i n g of reusable building wastes. Policy At f i r s t t h e r e was no i n t e g r a t e d p o l i c y t o w a r d s w a s t e m a t e r i a l s a n d t h e i r re-use, m a n a g e m e n t was j u s t s o l v i n g a c t u a l p r o b l e m s . Later a new policy was d e v e l o p e d in w h i c h a p r e f e r r e d s e q u e n c e of a c t i v i t i e s c a m e f o r w a r d . The "steps" in this p o l i c y are: - p r e v e n t i o n of w a s t e - r e u s e of p r o d u c t s -reuse of m a t e r i a l -burning with energy recovery -landfill. To a t t a i n the a i m s of the p o l i c y a n u m b e r of p u s h a n d p u l l m e a s u r e s w e r e taken. P u s h i n g m e a s u r e s u s e d are: -a l a n d f i l l b a n for d e s i g n a t e d c a t e g o r i e s of w a s t e - t a x a t i o n on l a n d f i l l e d w a s t e s - t a r g e t s in the e n v i r o n m e n t a l p e r m i t s of p r o d u c e r s - d e p o s i t s on n e w p r o d u c t s to p r o c e s s t h e s e p r o d u c t s
when
defunct
P u l l i n g f a c t o r s are: - p r o v i d e r e g u l a t i o n s for the u s e of s e c o n d a r y m a t e r i a l s - s t i m u l a t e s t a n d a r d i z a t i o n a n d c e r t i f i c a t i o n of s e c o n d a r y m a t e r i a l s - p r e s c r i b e the u s e of s e c o n d a r y m a t e r i a l s for g o v e r n m e n t b u i l d i n g s a n d roads - s u p p o r t for r e s e a r c h a n d d e v e l o p i n g n e w t e c h n o l o g i e s for p r o c e s s i n g w a s t e s
756 to u s e f u l
products
and
realisation
of p i l o t
projects.
Results There has n e v e r b e e n an e v a l u a t i o n of the i n f l u e n c e of the d i f f e r e n t m e a s u r e s t a k e n but I c a n g i v e y o u an o v e r v i e w of r e c e n t d e v e l o p m e n t s . i.
2.
3.
4
L a n d f i l l ban: was i n t r o d u c e d in 1995 for o v e r t w e n t y d e s i g n a t e d w a s t e c a t e g o r i e s r a n g i n g f r o m car w r e c k s to h o u s e h o l d w a s t e s . The a i m is to p r e v e n t the l a n d f i l l of u s e f u l r e s o u r c e s a n d if not r e - u s a b l e t h a n r e c o v e r the e n e r g y or u s e the o r g a n i c w a s t e s in a c o m p o s t p r o c e s s i n g plant.As new techniques of w a s t e processing become available new t y p e s of w a s t e are a d d e d to the l a n d f i l l ban. R e c e n t a d d i t i o n s are b u i l d i n g a n d d e m o l i t i o n waste, b l a s t i n g g r i t a n d w a s t e wood. S i n c e the i n t r o d u c t i o n of the l a n d f i l l b a n the q u a n t i t i e s of l a n d f i l l e d wastes have dropped sharply. Taxation on l a n d f i l l e d wastes: this taxation serves several p u r p o s e s . The f i r s t is s i m p l y to g e t money. S e c o n d a r y r e a s o n s are to d i m i n i s h the f i n a n c i a l g a p in c o s t s b e t w e e n l a n d f i l l a n d b u r n i n g in an M W C P a n d b y r a i s i n g the p r i c e of d i s p o s a l , the f i n a n c i a l i n c e n t i v e to p r o c e s s a n d r e - u s e w a s t e grows. Targets in e n v i r o n m e n t a l permits. B y a d d i n g t a r g e t s on w a s t e processing in e n v i r o n m e n t a l permits the p r o d u c e r s is f o r c e d t a k e a c t i o n w i t h r e g a r d to his waste. It is n o t s i m p l y a m a t t e r of costs, he has to p r e v e n t or r e - u s e his w a s t e to c o n t i n u e his activities. In s o m e c a s e s a g r e e m e n t s of the g o v e r n m e n t with branch-organisations are made aimed at reducing waste or r e p r o c e s s i n g waste, e.g. p a c k a g i n g a g r e e m e n t . S o m e c o m p a n i e s h a v e t h e i r o w n p o l i c y on e n v i r o n m e n t a l targets with respect to reuse. A large company in this region, aims at r e d u c i n g the w a s t e l e a v i n g t h e i r p r e m i s e s to zero. T h e y h a v e a s y s t e m to s o l i d i f y a n d r e - u s e their own wastes in r o a d s o n t h e i r site. Deposits: It is a f a i r l y r e c e n t d e v e l o p m e n t , f i r s t introd1~ced to s t a r t a s e l e c t i v e d e m o l i t i o n of o l d cars. On new cars a d e p o s i t is p a i d to p a y for the f i n a l d e m o l i t i o n . The s y s t e m is q u i t e a s u c c e s s a n d h e l p e d to i m p r o v e the q u a l i t y s t a n d a r d s in the car d e m o l i t i o n b r a n c h .
So far the p u s h i n g f a c t o r s , p u s h i n g a l o n e is not s u f f i c i e n t to r e a c h the goals. P u l l i n g f a c t o r s are a l s o n e e d e d to r e a c h a g e n e r a l l y a c c e p t e d r e - u s e of m a n y p r o d u c t s . The a b o v e m e n t i o n e d p u l l i n g f a c t o r s are: 1
2
3.
4.
Regulations. The p u t t i n g of n e w p r o d u c t s on the m a r k e t is n o t e n o u g h to c o m e to the u s e of them. A g g r e g a t e s n o w can c o n s i s t of s e c o n d a r y m a t e r i a l s , a n u m b e r of y e a r s a g o y o u w e r e o n l y a l l o w e d to u s e n a t u r a l s a n d or g r a v e l . J o i n t r e s e a r c h f r o m all i n v o l v e d p a r t i e s l e d to n e w r e g u l a t i o n s w h i c h i n c l u d e d s e c o n d a ry m a t e r i a l s . Standardisation and certification. R e - u s e is m a d e p o s s i b l e b y a d a p t a t i o n of r e g u l a t i o n s . S e c o n d a r y m a t e r i a l s h a v e to c o m p l y with functional and environmental specifications, standards h a v e to be d r a w n up so that p r o d u c e r s as w e l l as u s e r s k n o w the q u a l i t y of the p r o d u c t s t h e y d e l i v e r or buy. To e n s u r e the c o m p l i a n c e w i t h the s t a n d a r d s c e r t i f i c a t i o n s y s t e m s are used. U s e of s e c o n d a r y m a t e r i a l s . To p r o m o t e the u s e of s e c o n d a r y m a t e r i a l s it is e s s e n t i a l that g o v e r n m e n t an m u n i c i p a l a u t h o r i ties lay d o w n an e x a m p l e b y p r e s c r i b i n g s e c o n d a r y m a t e r i a l s in t h e i r own b u i l d i n g s a n d roads. Support in the d e v e l o p m e n t of n e w t e c h n i q u e s and pilot projects. T h e r e are p r o g r a m m e s to s u b s i d i z e n e w t e c h n o l o g i e s a n d to p r o m o t e p i l o t p r o j e c t s for n e w p r o d u c t s m a d e of waste. O n e of t h e s e p r o g r a m m e s was d i r e c t e d at the r e - u s e of d r e d g i n g spoils, a m a j o r w a s t e s t r e a m in this c o u n t r y . A n o t h e r p r o g r a m m is d i r e c t e d at s t a b i l i s a t i o n of wastes, techniques have been d e v e l o p e d , n o w the a i m is the r e a l i s a t i o n of p i l o t p r o j e c t s .
Goumans/Senden/van der Sloot, Editors
Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
757
EVOLUTION OF REGULATIONS AND STANDARDS FOR STABILIZED HAZARDOUS INDUSTRIAL FINAL WASTE MANAGEMENT IN FRANCE Anne-France Didier (SDPD Mmist~re de rEnvironnement) -Jacques Mfhu ( P O L D E N -INSA LYON Ddveloppement) - Valfrie Mayeux (ADEME)
1. M a i n p r i n c i p l e s
and definitions
1.1. D e f i n i t i o n o f H a z a r d o u s I n d u s t r i a l W a s t e s
The regulatory status of H1W is defined in the decree which transposes into French law, the European list of hazardous wastes including the definition of hazardous wastes (mainly based on the 14 criteria in appendix III of the Directive 91/689. The decree establishes the relationship between what were previously called in France "wastes with harmful effects" and then "special industrial wastes". Hazardous wastes therefore include : - Special Industrial Wastes (which are the object of regional disposal plans, and on whose
landfilling a tax is applied which contributes to finance remediation of contaminated sites), - Dangerous Wastes (from household waste), - Health Activity Wastes (with risk of infection). Since the beginning of 1997, France has actively been working on the development of practical modalities of application of hazardous criteria (H1 - H14) to allow extension of the hazardous waste list to new wastes from the European Waste Catalogue. For this, Association R E C O R D (cooperative research network on waste) and the Ministry of Environment have launched joint research programmes concerning the application of hazardous criteria to a wide range of wastes, including stabilized wastes. The organizations in charge of the research are INERIS (H1-H3 "physical hazards"), INRS (H4 -H12 "hazards for human health) and P O L D E N (H14 "environmental hazard"). The methods should be operational at the beginning of 1998. The French Ministry of Environment also intends to publish a text concerning the general classification of all wastes (non-hazardous or municipal like, inert) based on the European Waste Catalogue. Such a comprehensive text would therefore greatly facilitate the work of local authrities and industry on site.
758 1.2. T r e a t m e n t / d i s p o s a l of SIW (Special Industrial Waste) Annually, France produces about 150 million tons of waste including 7 million tons of SIW. Apart from possible valorization scenarios (recycling, reuse, extraction of materials, substitution of materials, energy recovery,...) SIW can either be treated prior to disposal: physico-chemical treatments or incineration (1 558 329 tons in 48 installations (ADEME-1994)), or "disposed of" in a class I landfill, if they can be considered as "final" and "stabilized" (727 696 tons in 13 landfills (ADEME- 1994)).
1.3. N o t i o n of Final waste The law of July 13th 1992 concerning all wastes (hazardous, municipal like, inert) determines that in 2002, only "final" wastes can be admitted for landfill. According to this law, a final waste is "a material which can no longer be treated under present day technical and economic conditions, particularly by extraction of the valorizable fraction or reduction of its hazardous orpollutant character". This notion can of course evolve. Furthermore, as it is not based on quantitative criteria, it is often the object of many debates concerning its application particularly on household waste.
1.4. N o t i o n if Stabilized Final Special Industrial Waste (SF-SIW) The bye-laws (18/12/92) define two categories of Final SIW which must be stabilized prior to landftllmg in class I. From March 30th 1995, for category A (APC residues, steel dusts, used catalysts...) From March 30th1998 for category B (industrial waste water treatment sludges, metallurgy slag (excluding those from salt baths-based process), waste from polluted soil treatment...) The SF-SIW is considered as such if it satisfies a certain number of physico-chemical criteria defined by the bye-laws (18/12/92 and 18/2/94) which concern the immediately leachable fraction obtained from compliance tests (X31-210, X31-211...) in the process of standardization on a European scale (WG2 of C E N / T C 292). Progressively, the notion of long term behaviour of wastes according to the exposure scenario will be introduced in the French regulations for landfill (it is already the case for studies in progress concerning inert waste in class III landfills. It is the object of the French standard X30407, now awaiting standardization on the European level (WG6 of the C E N / T C 292). These new requirements will probably modify the notion of stabilized waste and therefore what can be expected from stabilization process performances. Meanwhile, Final SIW can be considered as stabilized either as they stand, if they meet the requirements of the bye-laws (18/12/92) or after application of a stabilization process.
759
1.5. Stabilization
p r o c e s s e s
Three main techniques are now used or are being developed in France : stabilization using mineral binders (industrial stage), encapsulation in organic binders (pilot stage), - vitrification (under development, first industrial unit in 1997).
-
-
Stabilization using mineral binders represents all the industrial installations to date, either near or on the class I landfill site. 9 of the 13 Final SlW French landfills have a stabilization facility for an annual capacity of about 600 000 tons.
2. E v o l u t i o n of regulations and standards (Stabilized) Industrial Waste (SF-SIW) management in France
Final
Special
The application (first partial then total) of the new regulation concerning Final SIW landfilling has brought to light the radical increase in management costs (stabilization + landfillmg). This has led industry to reconsider waste generating processes (reduction at the source, pretreatment) in order to reduce either the quantity or the polluting character of the wastes (authorizing, for example, admission as they stand). In certain extreme cases, this has led to questioning the process itself, leading to the abandon of the considered waste. As far as management of the "unavoidable" Final SIW is concerned, 3 fields of reflection are now open to industry: 1 Delisting of the stabilized waste (either classified as "M" assimilated to household waste, allowing landfilling in class II sites, or potentially as "I" inert, allowing landfilling in class III sites). 2 Valorization of the stabilized waste, thereby economizing the cost of landfill and even fixing a sale price for the material. 3 Valorization of the waste as it stands, thereby further economizing the cost of stabilization.
2.1. Delisting of (S)F-SIW This alternative was first offered by the regulation within the framework of article 14 of the byelaw 25/01/91 concerning household waste incineration. A project for an application circular of this article stipulating acceptance conditions of stabilized APC from MSW incineration in class II was elaborated and discussed with industry. This perspective was at the origin of new projects for stabilization process development, a number of which had to be reconsidered due to the withdrawal of the circular. Even today, it is clear that certain stabilization techniques are situated in the delistmg perspective. This is the case for processes including an extraction. This position can be justified on the one hand by the fact that the composition (and therefore the potential pollution) of the waste is
760
greatly modified (by extraction of the most soluble fraction and part of the leachable metals) and on the other hand by the fact that, in certain cases, part of this salt fraction can be valorized. This is also the case for vitrification, whose cost alone is much greater than that of stabilization by mineral binders + landfilling. A comprehensive evaluation procedure for vitrification processes is now in the validation phase. Its application could allow delistmg or even banalization of vitrified materials under certain conditions and respecting certain thresholds which remain to be defined. This is apparently not the case for processes using mineral binders which are at present used in waste stabilization for class I landfilling. It must be noted here that as stabilized wastes are not considered hazardous in the European list, the French regulations which stipulate their landffllmg as hazardous waste (according to the typology of the Landfill Directive draft) could be reviewed on a case by case basis, by considering the technical and scientific aspects, in particular for wastes having undergone an extraction or a reduction of the polluting potential.
2.2. Valorization of (S)F-SIW The possibility of valorizing certain stabilized wastes in Civil Engineering is being considered mainly for vitrified materials. Association R E C O R D has funded a study on the general approach concerning banalization of materials elaborated from waste, leading to the definition of technical specifications (use criteria) and environmental specifications (present and future) to be taken into account. A certain number of conclusions can be drawn from this study: - I t is practically excluded that civil engineers envisage the use of monolithic materials elaborated by stabilization processes (blocks of solidified waste for example), -
To have a chance of being valorized, a waste must be able to substitute a material already used (sand, granulate, filler) and must respect the technical specifications like granulometry, mechanical strength and reactivity towards the other componants (e.g. problem of compatibility between hydraulic binders and vitreous matrixes, which has led several vitriflers to orient their technique towards slow cooling which enhances crystallization),
-Furthermore, the cost must be lower than the material to be substituted and the quantitative and geographical availability must be at least equivalent, - Environmental specifications are cruelly lacking on the French level. The development of the standard X30-407 should compensate for this in the long run (see later), -Valorization which consists of simply diluting the waste in a construction, the waste not participating by its properties in the specifications of the construction, is to be excluded, - T h e simple fact that a waste has cost too much to be stabilized for landffllmg does not confer any value to it as regards valorization in civil engineering. Valorization of stabilized waste is therefore to be studied on a case by case basis according to demand, technical and environmental specifications in a given scenario.
761
2.3. Valorization of waste as they stand This is a real problem at the moment for wastes which do not follow the normal disposal routes but are stored internally on site (in-situ landfill, mmmg waste, metallurgy slag heaps). In most cases, a strict respect of the regulations (stabilization + landfilling) is not economically compatible with the considered industrial activity. Maintaining the activity (and the associated employment) implies research for acceptable solutions (including for the environment). It may entail modification of the site and landfilling conditions on site or the valorization scenario. A certain number of these wastes are already valorized in civil engineering locally. This is sometimes carried out "unofficially" which does not necessarily mean it is a clandestine activity, but rather on the basis of a temporary authorization while waiting for further data allowing delivery of an official authorization. To emerge from this unsatisfactory situation, the metallurgy industry has launched research programmes with the aid of the European Commission concerning feasibility of valorization of different types of metallurgy slag in civil engineering. The most promising scenarios are in road construction. For such wastes; local use is a crucial issue given the cost of transport. As far as valorization conditions in civil engineering are concerned, the limits are the same as those described in 2.2. The main problem is still the environmental specifications to be defined according to the scenario. To compensate for this lack of specifications, the practically systematic reference of most promotors in civil engineering for the use of waste is the circular of May 9th 1994 concerning MSW bottom ash. The Ministry of Environment states clearly that the regulatory conditions proposed here do in no way allow appreciation of the behavioural parameters nor prediction in the long term. Even if the economical and social context for MSW Incineration has needed a short term regulatory text, its extension to all types of material, for which landfill wants to be avoided, is abusive and not relevant. The Ministry of Environment and the Ministry of Civil Engineering are preparing a charter to propose a joint evaluation procedure. At first, this charter would apply to non hazardous waste (particularly MSW incineration bottom ash) and then would perhaps be extended to certain (S)FSIW according to their long term behaviour and in a well-defined scenario.
3. E v o l u t i o n
of environmental specifications
The future of stabilization and its development towards other wastes or other scenarios is direcdy related to the evolution of the environmental specifications. These will be carried out on two complementary and successive levels : the verification of stabilization performances and the long term behaviour of the obtained material in the specified scenarios.
3.1. Performances The arrival of a new regulatory context with strict constraints concerning the waste streams (theoretically covering all the F-SlW) including new wastes, relatively difficult to stabilize but for which a budget exists (APC from MSW incineration) has led to the development of more ambitious and more effective processes.
762 The (logical) drawback is that, taking into account the considerable extra cost of application of these techniques and their practically statutory situation on class I landfill site entrance for those managed by site managers, the demands of the authorities and producer industries are justifiably greater. For this situation, the A D E M E has initiated studies to develop "Comprehensive Evaluation Procedures of S/S processes - (CEP)", specific to the techniques used (mineral binders (POLDEN), vitrification (CEA), polymers (LNE), bitumen (forthcoming). Theae comprehensive procedures, whose application would be a condition for obtaining funding for the industrial scale processes, are being implemented for mineral binders, and are in the final stages for the others. The objective of these CEP is to establish the nature of pollutant retention by the technique considered (encapsulation, micro-encapsulation, inclusion in the porous structure, integration in the phases of the matrix, change in mmerology), the mechanisms of leaching behaviour (washoff, congruent dissolution, coupling of dissolution and diffusion, shrinking front,...) its quality (resistance to physical, chemical and biological attacks) and its durability (resistance to ageing under constraints). Even if the objective here is process-oriented instead of material-oriented, therefore situated before the behavioural evaluation of material in specific, landfill or valorization scenarios, a certain number of experimental procedures will be the same.
3.2. L o n g term behaviour The draft Landfill Directive mentioned 3 levels of waste evaluation : level I : basic characterization, long term leaching behaviour level 2 : compliance tests, verification of long term behaviour parameters level 3 : rapid tests - control on site. Level 2 corresponds to compliance tests which exist in most European countries for landfill acceptance (X31-210, X31-211 in France). Working Group 2 of CEN/TC292 is in charge of standardization of these tests on the European level. The stage of enquiry is now completed for the first test concerning fragmented waste. Working Group 6 (whose French mirror is the AFNOR commission X30Y) is in charge of level 1, which is of strategic importance as it supposedly conditions the other levels. The field of application of these studies covers much more than just the problem of (S)F-SIW landfflling. It also concerns other types of disposal (the project in progress as regards regulations for landfilling of different types of inert wastes integrates this notion), valorization in civil engineering (bottom ash in road construction for example) and in a general manner any evaluation of pollutant release from a defined source in the environment under specified conditions (mechanical, geotechnical, climatic, biological, site use, risk factors...) at a given time scale.
763
To prepare these evolutions, France has elaborated a methodological standard (X30-407), whose outlines, after amendment, were adopted by the participants in the European workshop. Standardization of the tests to be applied to given couples waste (or materials)/scenarios, is in progress. One of the first priorities that the Ministry of Environment has fixed at the X30Y commission, concerns acceptance of stabilized waste in class I landfzll. Logically, regulations should integrate these notions in the more or less long term, and propose rules to be respected which will probably be in the form of thresholds, but corresponding to the nature and to the objective of the tests carried out for this evaluation of long term leaching behaviour. For example, it is probable that notions of pollutant availablity according to the chemical context (pH in particular) and pollutant flux per unit surface for monolithic waste will be introduced. We think that this will have three types of consequence on stabilization processes : - A certain selection, which will favour the best processes, i.e. those for which a real stabilization (control of pollutant flux) will be proposed ; - A selection within the different waste sources which can be potentially stabilized towards those which can reach a real stable state. For the others (bio-evolving, oxidizable, soluble and not retained...) pretreatments may be deemed necessary. We could therefore see S/S processes evolve towards waste streams which, due to their physico-chemical nature, logically correspond to them. We would therefore change from (( a market to conquer )) logic to a "treatment logic". The so-called secure landfill could then play its role to the full : protecting "unstable" stabilized waste from external attack. ; -Justified and supported opening up of other horizons to stabilized wastes based on appropriate measures and models ; stabilization of contaminated sites, old stocks or slag heaps, "inert" wastes in class III,...Here, the waste streams, techniques and especially the environmental specifications remain to be defined. The X30-407 (or its european equivalent) will not be sufficient. The evaluation of ecocompatibility i.e. the cross checking of pollutant flux emitted and the acceptable flux in the environment will be necessary. It will consist of defining adapted scenarios (geotechnical, hydrodynamic, bio-physico-chemical,..) for stabilized wastes, i.e. generating flux considered as being in exchange equilibrium with the environment. INSA Lyon is organizing an international congress in September 1998 : " W A S T E STABILIZATION AND ENVIRONMENT" which aims to demonstrate the necessary complementarity of the three consecutive stages of evaluation : - control of pollutant emission from stabilized wastes according to their intrinsic character and scenario conditions, - transport and evolution of pollutants from these wastes in the environment, - i m p a c t of these pollutants on man and the environment. The final objective is of course to provide information to the regulatory and industrial organizations responsible for stabilized waste management.
764
Main reference works
9
des d6chets : p r o g r a m m e de recherche pluriannuel de I'ADEME, coordonn6 par P O L D E N - W G 6 du CENTC 292 -Groupe de travail "Mise en d6charge des d6chets mertes" Mimst~re de l'Environnement, POLDEN -Etudes d'application des crit~res de danger de la Directive 91/689 : Mmist~re de l'Environnement, Association RECORD, INERIS, INRS, P O L D E N - Groupe de travail, valorisation des d6chets en BTP : Mimst~re de l'Environnement - PEAs de PSS (liants min6raux, vitrification, polym~res) : A D E M E , P O L D E N , CEA, LNE - Stabilisation des DIS, Etude prospective : ADEME, A L G O E management, P O L D E N -Eco-compatibilit6
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
765
T e s t m e t h o d s and criteria for t h e a s s e s s m e n t of i m m o b i l i z e d w a s t e Dr.lr. G.J.L. van der Wegen Intron, institute for materials and environmental research B.V. P.O. Box 5187, 6130 PD SlTTARD, The Netherlands on behalf of Centre for Civil Engineering Research and Codes (CUR) P.O. Box 420, 2800 AK GOUDA, The Netherlands Abstract The Dutch environmental policy is aimed at a drastic reduction of its wastes as well as the minimalisation of the use of natural raw materials. It focuses on prevention and re-use of wastes. Immobilization of (hazardous) wastes can contribute to this policy either by producing building materials from wastes or, if the former is not achievable, by producing less leachable wastes for landfilling. A reliable set of test methods and related criteria is indispensable to apply these techniques in an environmentally safe and justified manner. Moreover, these test methods are also an expedient to further development of immobilization techniques as well as the quality assurance during full-scale application of these techniques.
Introduction The Dutch environmental policy is aimed at a drastic reduction of its wastes as well as the minimalisation of the use of natural raw materials. It focuses on prevention and re-use of wastes. Immobilization of (hazardous) wastes can contribute to this policy either by producing building materials from wastes or, if the former is not achievable, by producing less leachable wastes for landfilling. It is important to know the effect of immobilization techniques on the leaching behaviour as well as on the durability of immobilized wastes. Test methods and related criteria to assess the characteristics of immobilized wastes have been selected and/or developed within the frame-work of CUR-committee D24. These instruments for assessment are indispensable for the development of the immobilization techniques, for proving their fitness for purpose and for quality assurance during full-scale application of these techniques.
Definition and aim of immobilization Immobilization is defined as a technological treatment in which the physical and/or chemical properties of a waste are changed in such a way that the spreading of the pollutants by leaching or erosion is reduced adequately on the short as well as on the long term. Immobilization is a treatment on a particle size level. Storing waste in big bags or steel drums will prevent spreading of the pollutants, at least on the short term, but is not considered as immobilization. Immobilization has two aims: 1 Treatment of (hazardous) waste in such a way that it can be applied as a building material.
766 Producing less leachable wastes which can be landfilled under less severe isolation measures for soil protection (i.e. changing a hazardous waste into a socalled controlled leaching waste). The first aim, application as a building material, is preferred because of much lower costs of removal of the waste, less trespassing on capacity of landfilling and the reduction in use of natural raw materials. However, if this first aim cannot be achieved the second aim is still of importance. The costs of removal will still be lower (although not as much as for the use as a building material) and the very limited capacity of landfilling hazardous wastes in The Netherlands will be spared.
Regulations In The Netherlands, the Building Materials Decree sets regulations under which building materials may be applied in or on the soil or in surface water in an environmentally safe and justified manner. This decree classifies building materials according to their content of organic pollutants and to their leaching of anorganic pollutants. Depending on their classification different regulations are applied [1]. Test methods for the determination of the leaching behaviour of anorganic compounds are the socalled diffusion test for monolithic materials (Dutch Standard NEN 7345) and the socalled column test for granular materials (Dutch Standard NEN 7343), which is a percolation test. For organic compounds in building materials no suitable leaching test have yet been developed. Therefore, composition values for organic pollutants are determined instead of immission values, which undervaluates most immobilization techniques. The Building Materials Decree requires that the building materials possesses an adequate durability. This property is not fully translated into test methods and criteria. If a waste cannot be upgraded to a building material, in most cases it has to be dumped, according to the Dutch regulations set forth in the Disposal of Solid Waste Materials Decree. This decree classifies waste materials according to their leaching behaviour. The test method used for this classification is the above-mentioned column test, however with a lower liquid/solid ratio (1 instead of 10). The material to be investigated in the column test must be crushed to a particle size less than 4 mm according to NEN 7343. Crushed monolithic materials will therefore result in a substantial higher leaching rate than in practice because of the substantial higher specific surface due to the crushing. Immobilization techniques based on monolithic products are therefore undervaluated. Some immobilization techniques require binders and additives. They contribute to an increase of mass and often in an increase of volume compared to the untreated waste. In the case of landfilling this is an undesired side-effect of the treatment. Dutch regulations require that such an increase in volume must be less than 25%. Durability of immobilized waste to be landfilled is less essential than in case of use as a building material. Usually, the waste is covered by other waste or a layer of sand within a few days or weeks preventing further influence of atmospheric conditions.
Additional test methods In addition to the Dutch regulations described above, CUR-committee D24 has developed test methods and criteria to fully assess immobilization techniques for hazardous wastes. With respect to environmental aspects the following properties of the immobilized wastes have
767 been considered: Water soluble salts At higher contents this might lead to (partial) solution of the matrix of the immobilized waste. This will influence the leaching behaviour as well as the durability in an adverse way. Digestible organic matter This might lead to enhanced leaching due to adsorption and complexation reactions with pollutants. Redox-conditions Under reducing conditions leaching of some heavy metals is substantially less than in an oxidized state (e.g. lead and copper in the presence of sulphide compared to sulphate). The reducing capacity of the immobilized waste is important with respect to the leaching behaviour on the long term. Efficiency of immobilization In order to assess the merit of immobilization techniques or to compare different techniques a diffusion test similar to NEN 7345 has been developed for untreated waste. Comparison of the leaching behaviour of the untreated and treated waste will give the environmental gain of the immobilization techniques. The test methods developed for the above-mentioned properties together with those postulated in Dutch regulations are summarized in table 1. The durability of immobilized wastes depends on its own structure as well as the climatic conditions it will be exposed to. In The Netherlands the most relevant mechanisms of degradation are: Wet/dry Under alternating wet/dry conditions cohesion failure might occur due to recrystallization of salts in and/or shrinkage of the immobilized waste.
solution
and
Freeze/thaw Above a critical degree of saturation with water every material is susceptible to freeze/thaw conditions. This critical degree of saturation depends on the pore structure and strength of the material. Erosion In order to prevent spreading of pollutants by wind erosion, the immobilized waste should have a certain resistance against this mechanism of degradation. Oxidation A combined interaction of oxygen, moisture and solar radiation (activation energy) can degradate organic binders usually leading to a brittle and 'dusty' surface appearance. Test methods for these four durability properties are shown in table 2.
768
Conclusions Immobilization of hazardous waste plays an important role in the Dutch environmental policy. It can contribute to the reduction of wastes to be landfilled, to an increasing re-use of wastes thereby saving natural raw materials. A reliable set of test methods and related criteria is indispensable to apply these techniques in an environmentally safe and justified manner. Moreover, these test methods are also an expedient to further development of immobilization techniques as well as the quality assurance during full-scale application of these techniques.
References
Hendriks, Ch.F. and Raad, J.S .... Principles and background of the Building Materials Decree in the Netherlands", Materials and Structures, vol. 30, January-February 1997, pp. 3-10. CUR-report no. 183, ,,Guideline for the assessment of immobilized wastes", 1995 (in Dutch, with English summary).
Table 1. 'Environmental' test methods
I
property
description of test method
standard
leaching
diffusion test (monolithic) column test (granular) = percolation test 'granular' diffusion test redox-potential/reducing capacity soluble salts at liquid/solid ratio of 10 extraction with 0.1N NaOH
NEN 7345 NEN 7343 NVN" 7347 NVN" 7348
efficiency redox soluble salts organic matter NVN = pre-NEN
Table 2. Durability test methods property
description of test method
wet/dry resistance
6 wet/dry cycles: 5 hours in water of 20~ + 42 hours drying at 70~ monolithic: critical degree if saturation granular: 20 freeze/thaw cycles sand blasting test, originally designed for wear resistance Xenon-test chamber
freeze/thaw resistance erosion resistance oxidation resistance
standard modified ASTM D559 RILEM 4CDC3 NEN 5924 modified NEN 2875 ISO 4892
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All fights reserved.
769
Inorganic immobilisation of waste materials F.Felix, A.L.A. Fraaij and Ch.F. Hendriks Delft University of Technology, Faculty of Civil Engineering ABSTRACT Inorganic immobilisation uses cement to solidify certain waste materials. Water is added to the cement in a ratio between 0.4 and 0.6. The calcium-silicates of the cement react with the water to a calcium-silicate-hydrate-gel (CSH-gel), lime and heat. This gel is able to cling aggregates together. The result is a hardened product, including some pores. Pores larger than 20 nm are called capillary pores, smaller than 20 nm are called gel-pores. The concrete is characterised by a high pH about 13. The effectiveness of immobilisation techniques is a relation of immobilisation and leaching mechanisms, degradation mechanisms, economic and environmental merit and application potentialities. Because of the high pH most heavy metals show good immobilisation DroDerties, except molybdene and chromium (VI). Other components of the waste material are hard to immobilise, such as anions and organic compounds. Therefore additives may be used. Oxidants can start a REDOX-reaction with the anions and organic compounds are used to bind the organic components. Once the product is put into practice, the effectiveness of the technique depends on leaching mechanisms: surface release, diffusion and dissolving. Especially diffusion is important for immobilisation. Diffusion is caused by the presence of water in the pores of the material and concentration variations between the material and the medium. Degradation mechanisms, important for inorganic immobilisation, are erosion, wet/dry-periods, temperature changes and freeze/thaw-periods [25]. These influences can cause the formation of cracks. Economic merit is not often given. It can be divided in benefits and costs of the used and produced materials, disposal, transport, labour, energy, aftercare and investments. Like economic merit ,the environmental merit is not simply given. An important factor is the priority list of Lansink. The benefit of the technique is the replacement of primary materials by secondary materials. Furthermore other aspects have to be considered, such as the use of energy, the release of components and the use of materials. The effectiveness of immobilisation is last but not least dependent on the aDDliCation potentialities. The application of immobilised products, both the disposal and useful application, needs after-care. This means, control of the product is required. Furthermore, the effectiveness of inorganic immobilisation depends on the used waste material. Some material have more opportunities than others. ~Desired properties of the waste material are: high content of heavy metals, low content of anions and organic compounds. Moreover, it is desirable to minimise the pretreatment of the waste. Currently used waste materials for inorganic immobilisation are: filter dust, coal ashes, MIP-fly ash, residues from burning of coal, sewage sludge, lime sludges, fluorgypsum waste, arsenical waste, slag and dust from steel making. It can be concluded that inorganic immobilisation can show good encapsulation properties for certain waste types. Especially these waste types, containing heavy metals. Degradation of the product is mostly caused by the formation of cracks. Therefore, after care of the product is required.
INTRODUCTION Inorganic immobilisation is based on cement as solidifying agent. In addition of water cement hydrates to a cement gel, that is able to cling aggregates together. The result is a hardened product, which can chemical and physical encapsulate certain hazardous waste
770 materials. This technique is widely developed in several countries, such as the United States of America and Japan. However, in the Netherlands, little applications of this technique are known due to legislative restrictions in the past. Current changes in legislation are helping to put developed processes into practice and thereby requiring a research to the advantages and disadvantages of inorganic immobilisation techniques. The scope of this article is a 'state of the art' of immobilisation based on cement. The scope of paragraph 1 are properties of concrete relevant to inorganic immobilisation. In paragraph 2 the effectiveness of inorganic immobilisation techniques is given. This paragraph contains encapsulation mechanisms, degradation mechanisms, application potentialities, costs and environmental aspects. Paragraph 3 deals with required properties for the waste materials. Chemical and physical properties of the waste material will be related to the effectiveness of the immobilisation. The last paragraph presents the advantages and disadvantages of inorganic immobilisation.
1 RELEVANT PROPERTIES OF CONCRETE
Composition
Concrete is made up of cement, aggregate, water, air and additives. Currently used cement types in the Netherlands are: Portland cement, blast furnace slag cement and Portland fly ash cement. The clinker consists of the following minerals: 3CaO.AIO 3 , 3CaO.SiO2 and 2CaO.SiO2. In the presence of water, CaSO4.2H20 (gypsum) and/or CaSO4, 3CaO.AIO3 reacts to CaO.AI203.3CaSO,.32H20 (ettringite). This ettringite forms a layer on the cement aggregates and thereby delays the hydration reaction of cement. Water is added to the cement to react with the calcium-silicates and form a calciumsilicate-hydrate-gel (CSH-gel) and lime. An example of the reaction of 3CaO.SiO 2 with water is: 2 (3CaO.SiO2)
+
6 H20
->
3 CaO.2SiO2.3H20
+
3 Ca(OH)2
During the reaction, the content of unhydrated cement decreases and the content of CSHgel increases, which leads to a decrease in volume. The second objective of the addition of water is the increase in the workability of the concrete. Water surrounds the cement particles and aggregates. In the final product, water can be present in the concrete under several conditions. In the capillary pores 1 water can be present as water vapour. Above the saturation pressure condensation will occur. The saturation pressure increases with decreasing pore size. Some water is not physically bound to the surface of the solid material. This water is called 'free water'. Free water exists in the larger gel pores and the capillary pores due to condensation. In addition, water is often adsorbed to solid materials in the smaller gel pores. Water can also form a layer between components of the CSH-gel. Aaareaates, such as river gravel and river sand are added to the cement paste. Important factors of the aggregates are the weight, shape and particle size distribution. In some cases, additives are used to improve the workability of the concrete or to decrease the Water/Cement-ratio. After casting of the concrete, the concrete will be vibrated for proper compaction. Air decreases the strength of the concrete, by forming pores in the concrete. However, in special cases air is consciously brought into the cement gel by means of air entraining agents. Concrete thereby becomes more resistant to freeze/thaw(salt)-periods, due to these pores.
1An explanation of capillary and gel pores will be given later in this paragraph.
771
Strength
Strength develops in time and continues to increase for years. This process will be accompanied by a gradual change in pore size distribution.
Pores
Pores in the CSH-gel take approximately 25% of the total volume of CSH-gel, the gel pores. They are defined as pores smaller than 20 nm. Pores larger than 20 nm are called capillary pores. During the hydration reaction the capillary pores diminish, due to the growth of cement hydrates. An increase in capillary pores is responsible for a decrease in strength and durability, as well as for an increase in permeability of the final product. However, capillary pores will always be present in cement. The pores are due to the presence of water.
W/C-factor
An important factor in the properties of cement is the Water/Cement-ratio. Especially, the strength depends on the W/C-factor. A higher W/C-factor includes a lower strength. The reason for this is that cement of W/C-factor above 0.38 will always consists of capillary pores. Besides that, too little water will make mixing difficult [17]. A typical W/C-factor lies between 0.4 and 0.6.
Hydration heat
The hydration reactions are exothermic and thereby the setting of cement is accompanied by the formation of heat. Heat is transported from the centre of the matrix to the surface. This results in temperature gradients, which can cause cracks in the case of bulk concrete.
pH
The pH of the cement product is high, approximately 13. The pH depends on the cement type. Furthermore, the development of OH-concentration in the pore water is a function of the W/C-factor [8]. A high W/C-factor results in a lower pH, as shown in the following figure. mg O H ' / l i t e r 11000.
L 9o00 [ 8800
,
, ,
~-
Cement p.c.-A wcf.: x OJ.O
/
oo.s6
II
7700 I6600
,
,~O.L5
. . . .
11111 ilili11~.3.
!1111 Ilill
11 ilIF~
pH
! !
l~~]J~ll.3 66 l 1LLLII
ssoo| I IIIIIIII I I !11~/~rllll.3s. .oo I I II111111 I I l~Z.ii/i i 1111111 I i i111111.32g 33ooi --I I l..llllll:.iI b ~1~ -1i?111 I I I111111 22001 i ! III~_~ IIII!! ! I l llllll.zs. .oo I ~ I - I oi, I l lllllll ,I I ii1iii ! !11 ii11 I
1o
--
1oo
1ooo
- Time (hours|
Figure 1: development of the OH-concentration in the pore water as a function of the W/Cratio
772 2 EFFECTIVENESS OF INORGANIC IMMOBILISATION The effectiveness of immobilisation can be measured by the encapsulation of components of the waste material. The more components are encapsulated, the more effective the technique is. Therefore, immobilisation and leaching mechanisms will be described. Furthermore, the environmental and economic merit and application potentialities contribute to the effectiveness of inorganic immobilisation. Immobilisation mechanisms The effectiveness of encapsulation depends on the immobilisation mechanisms during and after the cement production. Firstly, the physical mechanisms will be considered. Secondly, the chemical mechanisms are described and finally the use of special additives is related to the desired encapsulation of components.
Physical mechanisms can be divided in hardening of the concrete and porosity of the concrete. Fine particles (less than 74/Jm) weaken the bounding between waste particles and the cement by coating the larger particles [17]. This coating inhibits chemical binding of the contaminants. Pretreatment of the waste may be required to reduce the fine particle concentration. As said before, pores are always present in the final product. The leachability of components is increased by the increase of the porosity. Furthermore, a high content of soluble salt in the matrix may lead to an increase of the porosity of the matrix, due to the leaching out of these contaminants [6]. Chemical mechanisms of immobilisation of the waste are due to the chemical attraction of the surface atoms and of the precipitation of hydroxides. Most cations, such as heavy metals are insoluble at pH 8-10. They are immobilised by the precipitation of metalhydroxides. Above and below this pH range the solubility increases. Figure 2 shows the solubility of cations in relation to the pH. At the high pH of concrete, most cations are not in its least soluble state. Ions, such as molybdene and chromium (VI) are mobile at high pH, due to a low chemical retention, see figure 4. Although chromium and molybdene show high leachability, some heavy metals can be bound to the CSH-gel by adsorption and chemisorption. Metal ions may be incorporated into the crystalline structure of the cement [17]. Anions are difficult to immobilise, f.e. cyanide, chloride and bromide. They do not precipitate with hydroxide and often are not bound into the crystalline matrix [22]. Exceptions of anions that can be immobilised are sulfates and sulfites. Organic compounds may delay the setting time. Besides that, organic compounds show less fixation in the matrix [22]. Moreover, organic compounds can increase the mobility of other compounds [6]. Organic compounds may be decomposed at high pH. This will lead to soluble organic compounds. Trace elements, both organic and inorganic, can be bound to the organic compounds and thereby leave the matrix [6]. Volatile organic compounds, but also inorganic volatile compounds such as mercury, can release the matrix during mixing. Oil and grease cause the coating of waste particles and thereby weaken the bond between particle and cement [17]. Pretreatment may be required. Additives Because of the described immobilisation mechanisms, some cations, anions and organic compounds are hard to immobilise. Therefore, additives are used in order to optimise the immobilisation. In the following, the common additives are given and related to the desired encapsulation mechanism. In order to optimise the hydration reaction, pozzolanic materials, such as blast furnace slag and cement-kiln dust are added. Pozzolanic materials need an additional calcium source
773 100 -
,,,
ii~
.
.
.
.
.
.
.
0.001
~oo1!
0,0001
6
7
8
9
10
11
12
pit
Figure 2" typical solubility curve for metals over a range of pH in order to start the hydration reaction. Therefore and for the formation of ettringite, an external calcium source is required, like CaCO 3, CaCI2 or CaS04. NaOH and MgOH may be used for to optimise the hydration reaction. Ohter additives are useful for the immobilisation of inorganic compounds: soluble silicates, reducing agents, oxidants and clay. Soluble silicates, such as silica gel (Na2Si4.xH20), can chemically bind inorganic contaminants. However, a common disadvantage of these additives is an increase of volume [17]. Reducing and oxidising agents, such as Fe2§ Fe3+, Mn 4§ S042 are added in order to precipitate salts [14]. A REDOX-reaction will lead to less soluble salts and thereby diminish the leaching out of salts through the pores. An example is the reduction of ferric (11) to ferric (111)and the oxidation of chromium (VI) to chromium (111). ~P~L Leiil-OrlliLIl-I-I
IIIll IIIl/
:-I;~
I Iq~o
I..LI-;~uiUUi11:;
thc:lll
ch'---:"--iuiiiiuiii
il~llvi ) .
Pl.~..~..,ir
:'~h.~ r
4...~(q./ i[;;ll~.}q.ll'i~a"P~rkl~ i . . . .
~u,
. ~i .p . . . .
pounds and excessive free water. It is also used as absorbent of organic compounds. Organic compounds can be JmmobJlJsed in the cementious matrix by addition of certain absorbents, such as the previous mentioned clay and active coal [10,18]. By means of clay, organic compounds are absorbed in the matrix by non-ionic forces. Soluble organic compounds with carboxyl and hydroxyl groups, can be fixed at the Ca2+-ions. They thereby become insoluble. Organic, particularly aromatic, substances can be JmmobJlJsed by absorbJng on a layered clay mineral. The clay is therefore modified with alkyl ammonium, which increases the adsorption surface. The surface has no longer hydrophJlic but organophJlJc conditions. After adsorption of the organic compounds, the clay is mixed with a hardenable inorganic binder, f.e. Ca(OH) 2 or Ca2+-contaJnJng cement. Other techniques consists of the addition of an absorbent, others than clay [3,16]. A possibility is mixing the waste with a binder and then granulate. The granulates are contacted with solvent to extract hydrocarbons. Or the organic waste may be dispersed in water containing a cationic ammJne as an emulsifier. Subsequently, the emulsion is mixed with cement. The last technique for organic waste materials is to saturate the waste material with water and mixing it with an inorganic hydraulic binder [21 ]. Because of the high pH some organic compounds will vaporJse.
774
Leaching mechanisms
Once the immobilised product is put into practice, the isolation of contaminants from the environment can not totally be guaranteed. The mechanisms that contribute to the release of these contaminants are caused by the presence of water in the pores or on the surface of the product, due to the following mechanisms: surface release, dissolving and diffusion. Figure 3 shows these three mechanisms. ImmoblllsedproduGt I
~
Water Dissolving
--~ .......
9 o.,u.,..
.....
....
I Goomot,i(:alurfgioo
Figure 3: leaching mechanisms Surface release is characterised by a short leaching out of contaminants in the first period of the application of the product. High soluble components at the surface of the product dissolve into the surrounded media. However, this leaching mechanism is important in organic immobilisation techniques and vitrification, inorganic immobilisation shows little surface release. Over a longer period, the rate of dissolving of contaminants is almost constant. An example is the release of Ca 2. from stabilised gypsum. However, this mechanism does not contribute much to the total leaching out of contaminants. The most important mechanism is diffusion [6]. Diffusion is caused by the presence of water in the pores of the material and concentration variations between the material and the medium. As shown in the following formula, the diffusion capacity can be related to the tortuosity and the chemical retention.
L= L f Do R T
f2. Do R.T
flux (m2/s) availability (-) average diffusion coefficient (m2/s) chemical retention (-) tortuosity (-)
The tortuosity of an immobilised product indicates the path length of the ion through the pores [6]. A low tortuosity means a small diffusion path and thereby a high leaching rate of the component. Some immobilised products show low tortuosity compared to most building materials, such as concrete and bricks [6]. While tortuosity is a measure for the physical resistance of a component to diffusion, chemical retention is a measure of the hindrance of diffusion due to chemical interactions. Chemical retention is highly correlated to the pH in the pores. A typical relation between pH
775
and some metals is shown in figure 4. 10'
,~. ~o _~ *
,, ....
,,**o
10'
~.~....-~"--"
_...,t 10"
~
/
~
~
Z,
........
.o
F
.. ....... ..'" . . . . " " " '
........................................................ ...
10" 10"
". . . . . . . . . . .
ee . I . ~ ,~"
~, rr
, o ~ 1 7 6. . . . . . . ,'
I0"
Me
7
-
9-
-
, 8
. . . .
, 9
.....
, "'" 10
"
"
, " 11
"
"'"
, 12
pH
Figure 4: relation between chemical retention of certain elements and pH from the environment
Leaching test
To test the leaching behaviour of immobilised products, several test methods are suggested. Dutch legislation concerning building materials and landfill prescribes column tests, diffusion tests and availability tests. The first step in the column test is to fractionise the product to particles smaller than 4 mm and to put them in a column [11]. Acid water (pH =4) is lead from the bottom to the top of the column. After certain periods, the liquid is tapped of. These periods correspond to specific water/material ratios (L/S= 0.1, 0.2, 0.5, 1.0, 2.0,5.0 and 10.0). This column test gives insight into the leaching behaviour of a fractionised waste over a short period (5 year) and moderate period ( < 5 0 years). Another test, the cascadetest, gives information about the long term leachability. A column, water/material ratio of 100 kg/kg, is therefore shacked. The disadvantages of these tests is that the material has to be fractionised, while the aim of immobilisation is to encapsulate hazardous components in a compact matrix. The second test, diffusion test, is specific for materials larger than 40 mm, like monoliths of immobilised waste. The product is surrounded by water of pH =4. At certain time intervals, corresponding to L/S = 0.25, 1, 2.25, 4, 9, 16, 36, 64, the water is tapped of and purified. The concentrations of the percolates are determined. The results of the tests give insight into the contribution of diffusion to the total release of components. The aim of the third test, the availability test, is insight into the leaching of inorganic components of a material under extreme circumstances [24]. The material is first pulvuris~d until 95% of the material is smaller than 125pm. It is presumed that diffusion can take place through the whole material. During 3 hours, the material is washed with water (pH = 7) and after that washed with acid water (pH =4). The ratio between water and material is each time 50. The composition of the percolate is determined and translated to the amount of leached components. These test methods are not capable of measuring the leaching behaviour of organic compounds. Therefore, Dutch legislation is based on the organic composition of an immobilised product. However, these composition demands in Dutch legislation contradict the aim of immobilisation. Not the composition of the product but the leaching behaviour is a measure of the effectiveness of immobilisation.
Degradation mechanisms
Besides leaching mechanisms, other degradation mechanisms may effect the immobilised product. This gives an indication of the durability of the product. Durability is defined as the
776 resistance to chemical, physical and mechanical influences. A product is durable when no intolerable decrease of relevant properties is caused by the influences. Figure 5 shows these external influences. Inorganic immobilisation products are mostly sensitive to erosion, wet/dry-periods, temperature changes and freeze/thaw-periods [25]. Erosion is defined as the degradation of material due to moving media, such as wind, rain and rivers [5]. The wind contains fine particles, that deteriorate the surface [6]. Also, human activities can effect the surface, by walking and driving. The erosion sensitivity depends on the strength level of the product. Products with high strength have high resistance to erosion. Wet/dry p~riods are characterised by the transportation of water between the environment and the immobilised product [6]. This may lead to (re)crystallisation. In addition, wet/dry-periods causes shrinkage or swelling of the material. Shrinkage is due to the release of water from the pores and the release of adsorbed water. Because of increased surface tensions, due to the release of water, the surfaces attract each other. This leads to shrinkage. The extent of shrinkage highly depends on the W/C-factor and the cement contents. Shrinkage is from major importance for products with high amounts of clay. This can cause the formation of cracks on the surface of the product. The opposite mechanism appears in the presence of water, the wetting periods. Although this mechanism is much faster than shrinkage, shrinkage is the major cause for the formation of cracks during wet/dry-periods.
/ o
o o
u.v.
-....
Free!
.......
.....
/
9
9
..... ....... T .... t .... I .... N'
CO a
Oi
T/ el, f . , .
Figure 5: degradation mechanisms
Temperatur~ ~hanges has been mentioned before in relation to the formation of heat during hydration. The same argumentation can be followed here. Temperature changes of the environment causes temperature gradients in the immobilised material. This can lead to the formation of cracks. Freeze/thaw-periods can also contribute to micro- and macro-cracks formation. Due to freezing periods, the water in the pores can be frozen. Because of the increase of volume, the pores will expand. Water that is not frozen is pressed into smaller pores and causes press-tensions. Another mechanism is the crystallisation of the water vapour in the concrete
777 to form ice. Due to the lower vapour tension of ice in comparison to water, water vapour moves to the ice. This leads to the growth of the ice in the larger pores and the decrease of water in the smaller pores. This mechanism also leads to the formation of cracks in the material. A specific case of freeze/thaw periods is the presence of thawing salt. Thawing salts lead to the decrease of the freezing point of water, depended on the concentration of salts. In the centre of the material the concentration of salts is lower than on the surface. Temperature gradients and these concentration gradients may causes the surface and the centre to freeze. The isolated water, between the ice zones, presses on the surface, which will lead to degradation of the surface [2]. Application po ten tiafities Before a useful application can be chosen, the product has to be tested on legislation concerning building materials. The aim of legislation concerning building materials is to promote the recycling of secondary building materials combined with the aim to control the leaching out of hazardous components. In order to determine the suitability of building materials, the leaching behaviour of the inorganic components and the organic composition of the material must be determined. If application is allowed, the destination of the product partly depends on the strength of the product. Compared to 'traditional' concrete, compressive strengths of immobilised products have decreased by a factor 10 [25]. The immobilised products may be used in f.e. sound walls. However, most of the immobilised products are not allowed to be used as building materials. In that case, legislation concerning the disposal of waste becomes active. This legislation is also based on the leaching behaviour of inorganic components in the waste. C2-waste: waste that is very harmful. At this moment, there is one C2-1andfill and in the near future no new disposal will be built. Some C2-waste materials are or will be prohibited to be disposed of, such as fly-ash from Municipal Incineration Plants (MIP). With permission of the specific foreign country, the waste may be exported. C3-waste: waste that has high leachability of inorganic harmful components. The waste must be isolated from the soil. C4-waste: this waste has moderate leachability of inorganic components. Some adaptations are needed to minimise the leaching out of components to the soil. Both useful application and disposal require more or less after-care of the immobilised product. In case of disposal, isolation, control and managing activities are required. When a product is usefully applied, the demolish of the object needs to be controlled. Costs Economic merit depends on the benefits and costs of the used and produced materials,
ned if the immobilised products can be usefully applied. Costs of transport, energy and labour are almost the same as for primairy materials. The disposal of the immobilised products may give some benefits. The benefit of the promotion of the waste material from C2disposal to C3-disposal is approximately f75.-/ton, from C2- to C4-disposal f2OO,-/ton. The total cost of the inorganic immobilisation of MIP-fly ash is estimated between f60,- and f180,- /ton fly ash [27]. In most cases the economic merit for both disposal and useful application is low or negative. Environmental aspects Inorganic immobilisation techniques are meant to contribute to the environmental merit of the waste. Therefore, a comparison is needed between other possible treatments of the waste material and inorganic immobilisation. In Dutch legislation the priority list of Lansink gives a first indication of the environmental merit of immobilisation. This priority of waste managements is as follows:
778 1. prevention 2. recycling; 3. useful application; 4. treatment; 5. combustion; 6. disposal. Immobilisation techniques can be placed on the fourth level and in most cases have to be compared with disposal of the waste. Most hazardous waste materials, such as hazardous sewage sludge, can not be treated otherwise or directly usefully applied. They are placed on a so called C2-, or C3-1andfill. An useful application of this immobilised waste material therefore seems to contribute to the environmental merit. However, a few comments have to be made. If the immobilisation technique does highly effect the environment, it is not clear that immobilisation is the right waste management. In case of inorganic immobilisation, degradation of the environment is caused by the use of energy, the use of primairy materials, the release of hazardous components during the process and the leaching out of hazardous components. Benefits for the environment are the replacement of primairy materials, such as gravel, by immobilised products and upgrading the waste material on the previous mentioned priority list of Lansink. Inorganic immobilisation show little release of components during the process and some use of additives and energy. The environmental merit of the technique can be the replacement of primairy materials by secondary materials. So, from environmental point of view, inorganic immobilisation is a good option, when the leaching out of components can be minimised and the properties of the product fulfill demands of legislation concerning building materials. Otherwise, an environmental merit is not simple to give. 3 PROPERTIES OF RELEVANT WASTE MATERIALS In the previous, the effectiveness of inorganic immobilisation has been considered. The effectiveness partly depends on the used additives. Besides that the properties of the waste material are an important factor in the effectiveness. Some components are good, other are hard to immobilise. In this paragraph, desired properties of the waste material are given. Inorganic immobilisation is partly based on chemical reactions. Therefore, the following properties of waste materials are desired. Waste materials containing heavy metals, especially Pb, Cd and As are mentioned in patents. Figure 4 shows that Mo and Zn are hard to immobilise. In addition Cr 8+ is mobile at high pH. It is therefore desirable to immobilise waste materials with high content of heavy metals, apart from Mo, Zn and Cr 8§ Secondly, the presence of anions is in most cases not desirable. Exceptions of anions that can be immobilised are sulfates and sulfites. Thirdly, low concentrations of organic compounds are desirable. Especially Volatile Organic Compounds can not be fixed in the solidifying matrix and can cause degradation of the immobilised product. The effectiveness depends not only on the chemical reaction, but also on the physical encapsulation. Some properties of the waste have negative or positive influence on this encapsulation. F.e., the presence of pozzolanic or hydraulic components in the waste is preferable. On the contrary, particles with size less than 74pm, are not preferable. They weaken the bond between particle and cement. From environmental point of view, the so called C2-waste materials can contribute more to environmental merit than C3-waste materials. Moreover, some C~ -waste materials are or will be prohibited to be disposed of, such as MIP-fly-ash. Furthermore it is desirable to minimise the pretreatment of the waste. Pretreatment requires additional energy, additional primairy materials and/or may lead to a residue, that has to be treated. Pretreatment may be required f.e. when the waste material contains oil, grease and particles less than 74pm.
779 Other considerations for the effectiveness of immobilisation are: variations in composition, other treatments for the waste material and the amount of the production of waste. Currently, some inorganic immobilisation techniques have been developed. Accordance to patents, the following waste materials are common: filter dust, coal ashes, MIP-fly ash, residues from burning of coal, sewage sludge, lime sludges, fluorgypsum waste, arsenical waste, slag and dust from steel making.
CONCLUSIONS In this last paragraph the advantages and disadvantages of inorganic immobilisation will be mentioned, which are deduced from the description of the effectiveness of inorganic immobilisation in the previous paragraphs. Because of the high pH of the product most heavy metals show good results, except from Mo and Cr 8+. Other components of the waste material are hard to immobilise, such as anions and organic compounds. These elements can leave the immobilised product through diffusion. Therefore, most inorganic immobilised products do not fulfill the legislation concerning building materials. Additives or pretreatment may be required to improve the immobilisation. Furthermore, immobilisation requires severe after-care of the product. However, in special cases this may lead to a decrease of economic and environmental merit of the technique. Inorganic immobilised products are highly sensitive to the formation of cracks. Especially erosion, wet/dry-periods, temperature changes and freeze/thaw-periods [25] influence the product. Besides these disadvantages, inorganic immobilisation can be a good solution for the amount of certain hazardous waste materials. REFERENCES [1] Bolier, D., 'lmmobilisatie mag weer', in: Milieumarkt, l Oe-jaargang, nr.5, 1996, p.27-29 [2] Bijen, J.M.J.M., Fraaij, A.L.A., Rooij, M.R., 'Materiaalkunde, collegedictaat',Delft, Technische Universiteit Delft, 1995 [3] Chevron Research Company, "Cleanup of oily wastes', WO-C-91/00900, 1991 [4] Conner, J.R., 'Chemical fixation and solidification of hazardous waste', New York, Van Nostrand Reinhold, 1990 [5] Civieltechnisch Centrum Uitvoering Research en Regelgeving, 'Handleiding voor het beoordelen van
immobilisaten' Gouda, CUR, 1995
[6] Civieltechnisch Centrum Uitvoering Research en Regelgeving, 'Beoordeling van immobilisaten, een voorstel voor criteria en testmethoden', Gouda, Civieltechnisch Centrum Uitvoering Research en Regelgeving, 1993 [7] Ecoserdiana, "Process for stabilizing and solidifying wastes from aluminium processing by means of an inorganic matrix', EP-C-O561746, 1993 [8] Fraaij, A.L.A., "Fly ash a pozzolan in concrete', Delft, Technische Universiteit Delft, 1990 [9] "Grenswaardennotitie, storten gevaarlijk afval', mei 1993 [10] Haese, R., e.a., "Verfahren zur Bindung yon organischen und anderen Stoffen, verbunden ,it einer Absenkung der Eluat-Werte bezOglig TOC und anderer Stoffe bei der Deponierung von Abf~llen', DE-B4405323, 1995
780 [11] Hendriks, Ch.F., 'Dictaat Bouwstoffenbesluit', Delft, Technische Universiteit Delft, 1995 [12] Interstock Nederland, "Process for the preparation and lech-resistant solidification of filter dusts and reaction products of flue-gas purification of waste and seage sludge incineration plants', WO-C-92/22512, 1992
[13] Lea, F.M., "The chemistry of cement concrete', Londen, Edward Arnold Publishers, 1970 [14] Lopat Industries, inc., "Compositions to encapsulate chromium, arsenic, and other toxic metals in wastes', EP-B-0352096, 1990 [15] Means, J.L., Smith, L.A. (e.a.), 'The application of solidification/stabilization to waste materials', Boca Raton, Lewis Publisher, 1995 [16] Noakes, John, E., "Solidifation of organic waste materials in cement', WO-B-92/15098, 1992 [ 17] Noyes, R., 'Unit operations in environmental engineering', Park Ridge, Noyes Publications, 1994 [18] Pelt&Hooykaas B.V., "Process for immobilizing environmentally noxious metals and organic substan-
ces', EP-B-0398410, 1990
[19] Pelt&Hooykaas B.V., "Toxic waste fixant and method for using same', EP-A-0535758, 1992 [20] Pelt&Hooykaas B.V., "Method for fixing waste material', EP-C-0547716, 1993 [21] Pelt&Hooykaas B.V., ",4 method for the immobilisation of waste material contaminated with organic chemical compounds', EP-B-0590711, 1994 [22] Projectgroep Ontwikkeling Saneringsprocessen Waterbodems, 'lnventariserend onderzoek naar "state of the art" van immobiliseren', Lelystad, Directoraat Generaal Rijkswaterstaat, 1992 [23] Rheinische Baustoffwerke GMBH&CO.KG, "Kompaktierung von Industriest~ube und Deponie der
Kompaktate', EP-B-0380713, 1990
[24] Stichting Postacademisch Onderwijs Civiele techniek en Bouwtechniek, 'Cursus-map immobilisaten', Delft, PAO, 1996 [25] Technotrans, "Conferentie immobilisatietechnieken', Ede, 1997 [26] Tuijn, J. van,'VROM is om: immobiliseren is weer bespreekbaar', in: Milieumarkt, 7'-jaargang, nr.6, 1993, p.18-21 [27] Vereniging van Afvalverwerkers, "Beoordeling van Koude en Thermische Immobilisatietechnieken voor verwerken A Vl-vliegas', Nationaal Onderzoeksprogramma Hergebruik van afvalstoffen, Utrecht, 1996 [28] Wastech, "Treatment of hazardous waste material', WO-A-91/05586, 1991
Goumans/Sendergvan der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All fights reserved. PHYSICAL P R O P E R T I E S C O N T A M I N A T E D SOIL
AND
LONG
Pirjo Kuula-V~iis~inen
781
TERM
STABILITY
OF
STABILIZED
Keijo Kumila
Tampere University of Technology, Tampere, Tampere University of Technology, Tampere, Finland Finland Hanna-Liisa J~irvinen
Geological Survey of Finland, Espoo, Finland ABSTRACT By far, Finnish authorities have applied Dutch procedures when accepting stabilization as a remediation solution. However, the local conditions should play an important role when quality standards are given. In this study one of the main interests was to evaluate the long term properties of stabilized materials in the harsh climate conditions. The investigated material (As, Cr, Cu contaminated) was sampled from the surface layer of an old impregnation plant in Eastern Finland. The binding agents used in stabilization were ordinary Portland cement, fly ash and gypsum. The mixes with different binding agents were planned to achieve two different compressive strengths, 5 MPa and 1 MPa. The compressive strength results indicated a slower early strengthening speed with the samples containing fly ash or gypsum. The measured water permeabilities varied between the values of 10-7 and 104 m/s. The results from the pore size distribution measurements showed a rather large amount of gravitation pores 20...30 % which result supports the rather large water permeability. The long term stability of samples were tested with freeze-thaw test (ASTM D 4842). A Finnish standard (SFS 5447) for testing the freezethaw resistance of concrete specimens was also used. The ASTM-freeze-thaw test results indicated that the samples of low compressive strength didn't withstand the stress as well as the firmer samples. All the samples were partially destroyed in the SFS freeze-thaw test.
INTRODUCTION In Finland there are several old saw mills and wood impregnation plants where the soil is ,-,.,.~.,,.,+,-.,,,~;~,-.,++,.,I k,,:~+ " u] ,,.+. o ~ , l ~ . , . , ,
~ V l _ , 1.
,~nA
,+",,hPnm++vm
Thoeo ...........
o n n t n r n l n n t o r .t. + . ~ i t o q
nro . .
.
nflon . .
.
qltlmto.ct
nn
groundwater areas wherea grounwater is situated under postglacial gravel and sand deposits with large water permeability. There is a urgent need to isolate the contaminant from the surrounding environment either by stabilization or by other remediation methods. Stabilization with the cement based binder agents is the most common remediation method for inorganic contaminants in soils. Obviously the utilization of stabilization is based on physical encapsulation and chemical fixation of contaminants allthough there are several different theories discribing the binding mechanism of contaminants in cement matrix (Cockel). The utilization of fly ash and waste gypsum as binder agent in stabilization is can be done for economical reasons to replace the more expensive binder agent cement. But also there usage of fly ash can decrease the leachability of heavy metals from stabilized materials by decreasing the pH value of the cement matrix (Cote2). The most important physical properties of stabilized soils are low water permeability and sufficient compressive strength. The values of compressive strengths mentioned in the literature vary in the large range from 0.1 MPa to 10 MPa In Finland the weather conditions are rather hard compared to most parts of Europe. The utilization of stabilization in our climate conditions should be tested before the remediation
782 project is done. The purpose of this research project was to evaluate the physical properties of stabilized contaminated soils and also evaluate the long term resistance of stabilized samples against freezing and thawing using different standardized methods. One of the main interests was to evaluate the effect of different compressive strengths on physical properties and long term stability.
MATERIALS AND METHODS The contaminated soil was originally from a surface layer of an old impregnation plant in Eastern Finland. The soil was contaminated by a impregnation agent contaning arsenic, chromium and copper. The average amounts of these contaminants were 650 mg/kg for As, 480 mg/kg for Cr and 590 mg/kg for Cu. The average pH value of the material was 5.4. The total amount of soil delivered to our laboratory was 1000 kg. The equal quality of the material was guaranteed by controlled cross mixing method using a large laboratory mixer. During the mixing it was noticed that the contaminants were concentrated to a certain "hot spots" which couldn't be broken totally during mixing. The soil contained also small parts of wood. The grain size distribution of the soil sample is presented in Figure 1. The average grain size of the contaminated soil was 0.45 mm and the amount of fine fractions ( 1. 4 . 1?_0
.
.
.
.
.
~.o:
.
.
....... ~ ......... ~ ~
~.oi~ ..... 4
. ~ ~ - - ~
....
.........~ ........ i .................i...............,0% 4 ........... ~ ................i ..................i ....................- 8 o ~
8.01 ............................................
,o~ ..........i ..............i
40
loo%
............... ................................. ~
,oo;-~~ :Z:
.
silicate . .
~
.~_~ ~'
i ..................... 5O%
.... ~
~
........ .................... i......................I ' o ~
~
e-
.....2 ...............A ..........,,,,i .........A-.~....-.-........ ! ..........~...:.:.-~- ..... ~ ....................i .................... ~ o ~
3.o~,~';,,,, 0.0
2.0
,,,, 4.0
6.0
,,,,!,,,,! 8.0
......... 10.0 12.0
14.0
,,,!,,,,!o% 16.0 18.0
e q o f a c i d / k g of d r y c e m e n t
-
NO waste, < 0
o
NO waste, 56d
x
WITH waste
•
dissolved
Figure 4. Titration curve for the blast furnace slag and sodium silicate binder system, with waste after 56 days curing, and without waste before mixing and after 56 days curing, showing also dis~lution of the activated slag matrix as a function of acid addition.
810
Coal fly ash/lime, 56d
13.0~
i!
12.0 i 11.0~
90%
j - -
~o.o~
-80%
i
T
9.0i -1-
100%
i
70%
..........
8.0 -
.
.
.
6.o:.
. . . . . . . . I" ;..............-40% ] ! ~ i i ...... 4................ i ............ -,................... :.............. ~................... ~................... -30% ~" - - ~ i
5.o'~................ ~ ..........-~........i
.............. !..................... i ................ i............... !.................... !.....................
-20%
4.~I~~........... ~.i ~ ..... ;3
60% -50%
.
0.0
m
"10
r
._0 0
LL
..9.i.............10% ..........
4.0
2.0
~,
......................
i
7.0 .
10
6.0
8.0
0% 1 0 . 0 12.0 14.0 1 6 . 0 18.0
eq of acid/kg of dry cement o
NO waste
x
9 dissolved
WITH waste
Figure 5. Titration curve for the coal fly ash and lime binder system, with and without waste after 56 days curing, showing a ] ~ dissolution of the lime/fly ash cement matrix as a function of acid addition.
High alumina cement/lime, 56d 13.0=;,.~.~E~_ ~I
1 2 . 0 : i ~ ~ ' - - ~ Z
,oo.
i
"
.................7....
100%
i,
i
~
'
.=
i
i
9 ~
i
. . . . . ~ ...............90% i
i ........ i.......... i .......... i ............. ! ..................~ ................. ~o%
"~ .................!...............~ i-~ .... i ........ I r9................... ~ i ~................... ~0~ .........I ............... ~..................... ~ ................... 9
=
8.o~, .............. ~ %
,-o~'-i'~ ........ i-~4 ......... !..........4................ ....... i..........i..........4............~o% ......... o.o~..........!......~.....~. ~ ......~....... ~ ...... ~ ................ ~...........~................... ......... ~o%~
~.o~................ i.......... i....~~~
,.0 .................... i............~ i . 30
D
i,
0.0
,,
gypsum
......... ~............. ~..........i............i...............~o%~ ......... ~
. . . . . i........................................................, 0 ~
m --:.......................... 4.0 6.0 8.0 10.0 12.0 14.0 eq of acid/kg of dry c e m e n t
- = ....
2.0-'
x
hazwash
m gyp-dissolved
16.0
A
D%
18.0
"
hwa-dissolved
Figure 6. Titration curves and matrLx dmsolutlon as a tunctlon ol acIO aomtmn for the high alumina cement and lime binder system, with gypsum, and with hazardous waste incinerator ash after 56 days curing.
811 Table 3. Dissolution of various binder s },stems at pH 9 Matrix components Amount of acid to pH 9 tecl/k~ of dr~ cement) Portland cement 16.0 Portland cement/waste 12.6 Portland cement/silica fume 11.0 Portland cement/silica fume/waste 9.5 Blast furnace slag/sodium silicate 6.2 Blast furnace sla6/silicate/waste 0.5 Coal fly ash/lime 3.6 Coal fly ash/lime/waste 2.8 High alumina cement/lime/~]~sum 8.1 High alumina cement/lime/haz, ash 4.8 i
i
Percent dissolved 50% 15% 35% 15% 10% 5% 7% 5% 1% 45% i
i
As expected, the first two pH plateaus, at 12.3 and 11.9, coincide with low solubility of the matrix, as indicated by total dissolved solids measurements. Matrix solubility appears to increase below pH 11.9. The portland cement system containing plating sludge does not exhibit a pH plateau at 12.3; it appears that the waste has consumed the lime, or altered hydration so that lime was not generated. The pH drops directly from 12.3 to 8, with a slight flattening at pH 10.5. Interestmgly, the fractional solubility of this sample appeared to be much lower than that of the pure cement system. A detailed analysis of the plating sludge was not performed, but it may have contained a high proportion of insoluble components. 4.2. Portland C e m e n t with Silica F u m e Replacement of 20% of the portland cement with silica fume decreased the Ca/Si ratio from 3 to 1.4. Accordmgly, it was expected that the pH plateau at 12.3 would disappear when the lime produced by hydration of the portland cement reacted with the silica fume. Indeed, the pH 12.3 plateau was apparent in the sample without waste at 7 days of curing, as shown in Figure 3, but had disappeared by 28 and 56 days of curing.
From 28 days of curing, the titration curves for the portland cement and silica fume samples were very. similar, with and without addition of the plating sludge. These curves were also similar in shape to the portland cement and plating sludge sample (Figure 2), exhibiting a plateau between pH l0 and 11. The matrix solubility indicated by the total dissolved solid measurements was lower than for the pure portland cement samples, increasing beyond the pH plateau. Again, the solubility of the sample containing waste was lower than that of the sample without waste. 4.3. Activated Blast F u r n a c e Slag The binder system produced by activation of blast furnace slag with sodium metasilicate has a low overall Ca/Si ratio of 0.5. Based on the earlier discussion ofpH data for a pure system, this would predict the coexistence of low Ca/Si CSH and silica gel, but only a relatively small proportion of the slag hydrates, so the Ca/Si ratio of the CSH is higher, and silica gel is not formed.
The pH vs. acid addition data for the blast furnace slag samples for 7 and 28 days resembled the titration curves plotted at 56 days in Figure 4. The curve for the slag system without addition of waste shows an initial pH of 12.5, droppmg to a pH plateau at approximately 9. A titration curve for the unreacted mixture, calculated based on the acid neutralisation capacities of the individual components, has also been plotted in Figure 4. Comparkson of the unreacted and reacted systems clearly demonstrates that a reaction takes place, with the products having a lower initial pH, and a higher and more distmct pH plateau, than the reactants. The total dissolved solids measurements indicate that the solubility of this system is low, even under acidic conditions. However, it does appear that the solubility increases at the start of the pH 9 plateau. The pH 9 plateau may represent CSH
812
coexisting with silica gel, its theoretical position having been altered by the presence of other components in the slag, such as Mg. The plating sludge inhibited the set of the slag. These samples did not develop physical strength, and pH plateaus for reaction products associated with strength development are absent from the titration curve. It seems likely that the plating sludge reacted with the sodium silicate, and prevented activation of the slag. In spite of the lack of hydraulic reactions, the solubility of the slag/waste system was still low.
4.4. Coal Fly Ash and Lime As coal fly ash alone has very little acid neutralisation capacity, and lime would control the pH at 12.3, the pH plateau between 11 and 12 for the fly ash and lime without waste in Figure 5 again confirms that pozzolanic reactions, creating CSH, have occurred. The data shown is for the sample cured for 56 days; the pH plateau in the sample cured for only 7 days was closer to 12, while the titration curve at 28 days was the same as that shown. Again, the Ca/Si ratio was approximately 0.5, but, as was the case for the slag system, a significant proportion of the fly ash has remained unreacted, so that the Ca/Si ratio of the CSH formed is higher. In this case, the development of strength by the samples containing plating sludge indicates that the presence of waste in the system did not inhibit the set. However, the titration curve for the sample containing waste is considerably different, exhibiting a rapid drop from pH 12 to 10.5, and then a linear slope to pH 4, with no discernible plateaus. Examination of total dissolved solids data shows low solubility for this system, with and without waste, although it does appear that the solubility of the pure system jumps when the pH drops below 9, while that of the system containing waste increases steadily as the pH drops.
4.5. High Ettringite System The titration curves for the two calcium sulphoalummate-based systems at 56 days have been plotted in Figure 6. The curves for 7 and 28 days had similar features. Two distinct pH plateaus are visible in Figure 6. The pH plot for the system containing gypsum rather than hazardous waste incinerator ash levels off at pH 12.4 and 10.8. That for the system containing ash levels off at pH 12 and 10. The overall acid neutralisation capacity for the sample containing incinerator ash is lower because the ash contains components which do not participate in the reaction, whereas the gypsum reacted fully. The pH 12/12.4 plateau is attributed to tetracalcium alummate hydrate (4CaO.Al2Oa.13H~O) which would be expected to be the other main hydration product in this system, in addition to ettringite, which results in the plateau at pH 10/10.8. Unreacted lime may also contribute to this plateau. The presence of impurities in the hazardous waste incinerator ash clearly affects the position of the pH plateaus, as compared with the gypsum system. Chloride may be incorporated in the ettringite, or form calcium chloroaluminate (3CaO.AhOa.CaCl:. 10H20). Other researchers have found this product to be stable between pH 11 and 12.5 (Ben Yair, 1971). Total dissolved solids measurements show the solubility of the ettrmgite system containmg gypsum rather than incinerator ash to be extremely low. The solubility data plotted in Figure 6 show that the use of ash rather than g3.j)sum in the matrix increases iLs solubility drastically. If it is assumed that the solidification process does not affect the solubility of the ash (69%), a matrix solubility due to ash dissolution of 41% may be calculated, based on 60% of ash in the matrix. In fact, a slightly higher matrix solubility is observed.
813
5. CONCLUSIONS Except for the pH plateau at 12.3 caused by excess lime, the pH plateaus observed in response to acid addition for real binder systems do not exactly correspond to those observed for the pure CaO-SiO2-H20 system. Different hydraulic cements exhibit different pH plateaus, which probably reflect differences in the structure and composition of the CSH formed. In general, plateaus seem to occur at pH levels lower than those anticipated. The addition of waste to a cementing system also appears to lower the pH plateau. Solubility of all matrices was low above pH l l.5; depending on the binder system, matrix dissolution appeared to increase at pH values ranging from 9 to 11.5. The following conclusions are drawn for the specific systems studied: 9
For portland cement, the amount of acid required to achieve a pH of 9 was 16 eq/kg cement, but greater than 10% matrix dissolution was observed at an acid addition of 6 eq/kg cement, and a pH of less than 11.5. 9 Addition of silica fume to portland cement appeared to lower the pH at which significant matrix dissolution was observed to approximately 10.5, with an acid addition of 6 eq/kg cement. 9 The activated blast furnace slag binder system showed increased matrix dissolution at pH 9, with an acid addition of 4 eq/kg cement, but overall low solubility over the pH range from 12.5 to 5. 9 The solubility of the coal fly ash and lime system was also low over the pH range from 12.5 to 4. An increase in solubility was observed after addition of 4 eq/kg of acid/kg cement, below pH 9. 9 The solubility of a high ettringite cement system, using gypsum to form calcium sulphoaluminates, is very low at pH values above 10, however, use of a high sulphate and chloride waste in place of gypsum results in a matrix with very high solubility.
6. R E C O M M E N D A T I O N S AND S U G G E S T I O N S FOR F U R T H E R W O R K 9
9
9
9 9
Rather than using a fixed pH as the criterion for performance in an acid neutralisation capacity test, it may be advisable to consider the pH at which a particular waste/bmder system appears to undergo a significant increase in solubility. As waste materials are generally not alkaline, adjustment of binder formulations containing alkaline components such as lime or sodium silicate to compensate for their consumption by the waste material should be investigated. From tests on ground samples, the actual effect of acid addition on the structural integrity of a solidified matrix is not easily apparent, because disintegration of structural matrix components may occur without chemical dissolution. Also, the physical structure of the matrix, including density and porosity, as well as its chemical stability, will influence its acid resistance. Thus, the acid resistance of monolithic samples should be investigated. Chemical changes continue to occur in cements over long time periods, therefor, the acid neutralisation capacities of different binders at ages of several years should be investigated. While micromorphological studies may confirm inclusion of waste components in calcium sulphoaluminate phases, leaching studies should be conducted to evaluate the stability of these compounds in the environment.
7. R E F E R E N C E S
Atkinson, A., Goult, D.J., and Hearne, J.A. (1985), "An Assessment of the Long-term Durability of Concrete in Radioactive Waste Repositories", Material Research Society, proceedings of symposium, Vol. 50.
814
Bambauer, H.U., Gebhard, G., Holzapfel, T., Krause, C., and Willner, G. (1988), "Schadstoff-Immobilisierung in Stabilisaten aus Braunkohlenaschen und REA-Produkten - I. Mmeralreaktionen und Gefiigeentwicklung: Chlorid-Fixierung", Fortschritte der Mineralogie, Vol. 66, pp. 253-279. Ben-Yair, M. (1971), "Studies on the Stability of Calcium chloroaluminate", Israel Journal of Chemistry, Vol. 9, pp. 529-536. Cot6, P.L. (1986), Contaminant Leaching from Cement-based Waste Forms Under Acidic Conditions, Ph.D. Thesis, McMaster University, Hamilton, Ontario. Day, R.L. (1992), The Effect of Secondary Ettringite Formation on the Durability of Concrete: A Literature Analysis, University of Calgary, Department of Civil Engineering, Research Report No. CE 92-2. Deng, M. and Tang, M. (1994), "Formation and Expansion of Ettringite Crystals", Cement and Concrete
Research, Vol. 24, pp. 119-126. Ghorab, H.Y., and Kishar, E.A. (1986), "The Stability of the Calcium Sulphoaluminate Hydrates in Aqueous Solutions", Proceedings of the 8th International Congress on the Chemistry of Cement, Rio de Janeiro, Vol. 5. pp. 104-109. Greenberg, S.A. and Chang, T.N. (1965), "Investigation of the Colloidal Hydrated Calcium Silicates. I]. Solubility Relationships in the Calcium Oxide-Silica-Water System at 25~ Journal of Physical Chemistry, Volume 69, Number 1, pp. 182-188. Grutzeck. M., Benesi, A., and Fanning, B. (1989), "Silicon-29 Magic Angle Spinning Nuclear Magnetic Resonance Study of Calcium Silicate Hydrates", Journal of the American Ceramic Society, Vol. 72., No. 4, pp. 665-68. Hassett, D.J., Pflughoeff-Hassett, D.F., Kumarathasan, P., and McCarthy, G.J. (1989), "Ettringite as an Agent for the Fixation of Hazardous Oxyanions", Proceedings of the Twelfth Annual Madison Waste Conference on Municipal and Industrial Waste, Madison, WI, September 20-21. Kumarathasan, P., McCarthy, G.J., Hassett, D.J., and Pflughoefl-Hassett, D.F. (1990), "Oxyanion Substituted Ettringites: Synthesis and Characterisation; and their Potential Role in Immobilisation of As, B, Cr, Se and V", Materials Research Society Symposium Proceedings, Volume 178, pp. 83-104. McCarthy, G.J., Hassett, D.J., and Bender, J.A. (1992), "Synthesis, Crystal Chemistry and Stability of Ettringite, A Material with Potential Applications in Hazardous Waste Immobilisation", Materials Research Society Symposium Proceedings, Vol. 245, pp. 129-140. Solem, J.I~, and McCarthy, G.J. (1992), "Hydration Reactions and Ettringite Formation in Selected Cementitious Coal Conversion By-Products", Materials Research Society Symposium Proceedings, Vol. 245, pp. 71-78. Taylor, H.F.W. (1993), "Nanostructure of C-S-H: Current Status", Advanced Cement-based Materials, Vol. 1, No. 1, October. U.S. Federal Register (1986), Appendix I, Part 268, Toxicity Characteristic Leaching Procedure (TCLP), Vol. 51, No. 216, November 7. Wastewater Technology Centre (1991), Proposed Evaluation Protocol [or Cement-Based Solidified Wastes, Environment Canada Publication EPS 3/HA/9, Ottawa.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
815
C o n t a m i n a t e d soil - c e m e n t stabilization in a d e m o n s t r a t i o n project. J. van Leeuwen, A. Pepels and G. van Emst Gemeentewerken Rotterdam (Public Works), Engineering Division. P.O. Box 6633 3002 AP Rotterdam, the Netherlands
Abstract
Public Works Rotterdam has carried out a feasibility study on the application of contaminated soil-cement stabilization layers (ref. 1). Conclusion was that contaminated soil can be benificially used as a soil-cement stabilization in a road construction. As a follow-up, a full scale demonstration project (1.000 m 2) was realized in 1995: a . (sub)base of a soil-cement stabilization layer with soil that should have been landfilled (ref. 2). Samples of the soil-cement layer have been tested on both environmental and physico-mechanical aspects. Conclusions were that the base is environmentally safe and no isolating provisions are required. The compressive strength is sufficient according to the standard of 1.5 N/mm 2. However, the need for licences and an environmental impact assessment and the procedure time involved will be the bottle-neck for the on-site immobilization of (hazardous) waste materials in constructions.
Introduction
In the city of Rotterdam large quantities of polluted soil are excavated as a result of soil sanitation and/or infrastructural works. The policy aims of the government and the municipality of Rotterdam are in the first place the beneficial use, secondly to clean and thirdly to landfill the polluted soil, depending on the type of soil combined with the type and content of contaminants. In infrastructural works, the base is mostly constructed of primairy sand - cement stabilizations. The question was whether the (clean) sand could be replaced by polluted, sandy soils. The policy of beneficial use of the municipality of Rotterdam is being implemented by stimulating the use of contaminated soil in infrastructural works. Public Works Rotterdam has carried out a feasibility study on the application of contaminated soil-cement stabilization layers (ref. 1). Conclusion was that contaminated soil can be benificially used as a soil-cement stabilization in a road construction. The question was whether the immobilisation of strongly polluted soil is executable on a full-scale project. In 1993 the national government started a tender-program on immobilization: Technology 2000 - Immobilization. The goal of this program is to stimulate the development and implementation of immobilization techniques on 22 selected hazardous materials, including non cleanable polluted soil. As a follow-up of the feasibility study and within the framework of the program . Technology 2000 - Immobilization, a full scale demonstration project was realized in 1995: a (sub)base of a soil-cement stabilization layer with soil that should have been landfilled (ref. 2). The stabilization is the base of the temporary storage site of DOPNOAP, a landfill for polluted soil in Rotterdam. The goals of this project were: the demonstration and testing of the beneficial use of polluted soil immobilized with cement as a road base. to check the feasibility and practical aspects of the immobilization process.
816
to determine the performance of the soil/cement stabilization under practical use. Figure 1 shows a cross-section of the construction. AFPHALT-LAYER
\ SUBSOIL
CONTAMINATEDSOILC~k~_~E MENT STABILIZATION
Methods and materials
The project consisted of the following steps (see figure 2): 1) 2) 3) 4) 5) 6)
the selection of a batch of polluted soil, which has to be landfilled (physical and environmental criteria) the characterisation of environmental and physical quality of the soil *content of contaminants and leaching behaviour *grain-size, pH, organic matter the determination of the optimal cement addition. the determination of practical aspects of the construction *mixed in place *mixed in plant the sampling of cylinders after execution and hardening the determination of environmental and physical quality of the soil/cement stabilization *compressive strength, durability. *leaching behaviour
In order to landfill the polluted soil, a large number of batches are registered at the Soiland Residuebank of Rotterdam. Some of these batches are suitable for immobilization. As a result of the feasibility study some initial criteria were derived for the selection of batches of soil. * organic matter < 10% dry weight * fines (< 2 um)< 5% dry weight * moisture content < 10% The batch which best met these criteria, was selected and stored at the site. Then samples were taken in order to characterize the batch and several tests were carried out in the laboratory. The leaching of the soil has been determined according to NEN 7343. One of the most important questions was the optimal addition of cement and the question whether other additives had to be added. The physical tests were carried out according to the Dutch RAW-standards. After 28 days of hardening, the compressive strength of the proctors was determined. If they met the standard (5.0 N/mm2), the proctors were tested in the tank leaching tests (NEN 7345), to determine the leaching of the immobilized product.
I
batch 1
siteinvestigation mixed in place / mixed in plant
initial criteria
selection and storage
Figure 2: The scheme of the project
characterization and investigation of cement content
execution
afler 28 days
I
+
afler 8 months
testing of : environmental and -physic0 mechanical quality durability (only 23 days)
818
When a batch is qualified as suitable and a large scale infrastructural work is planned, the go/no go decision can be made. The local geology of the site determines the method of realisation ('mixed in place' or 'mixed in plant'). After the realisation samples were taken in order to determine the environmental and physico-mechanical quality of the immobilized product. After respectively 28 days and 8 months of hardening, samples (cylinders) were taken out of the base. The cylinders were tested on both physico-mechanical and environmental aspects. Compressive strength, durability and leaching of the cylinders were determined after 28 days. After 8 months the compressive strength and leaching of the samples were determined again. Durability is defined as the resistance agains wetting, drying and freezing and is investigated by wet-dry and freeze-thaw cycles. The cylinders were exposed to wet/dry and freeze/thaw-cycles, according to Dutch RAW-standards. After those cycles the compressive strength has been measured again.
Results and discussion Characterisation of the batch
A batch of about 500 tons of soil, contaminated with heavy metals, mineral oil and poly aromatic hydrocarbons (pah's) and supposed to be landfilled, was selected for the base. in table 1, the results of the most critical components are shown. In the national Dutch legislation for the use of raw and secundairy materials (Bouwstoffenbesluit, the Dutch Building materials Decree) criteria for the use of soil are included. For both granular and immobilized materials standards on content and leaching have been developed. Table 1:
environmental quality
component
content
Leaching
Range
Trigger value
L/S=10
Copper (mg/kg)
95,2- 181
87,5
0,0660
0,58
Lead (mg/kg)
46,5- 125
325,5
0,0500
1,6
Zinc (mg/kg)
124 - 274
278,0
0,327
3,3
Trigger value
According to the mentioned legislation for benificial use, the soil is not suitable because of the high content of copper and the presence of oil and pah's. However, for immobilized soil and/or other materials, there aren't any standards for the content of heavy metals (and other anorganic components). On the other hand, the products or materials have to meet the standards for leaching for these anorganic components. For the organic components, no leaching tests have yet been developed, so the standards are based on the content of these components. From table 1 can be concluded that the leaching of copper, lead and zinc meet the standards for unisolated application. The addition of cement should be sufficient to produce a product which is environmentally safe.
819
Next, the physico-mechanical aspects of immobilization were determined. The results are presented in table 2. Table 2:
physical quality and demand of cement
Moisture content Fineness modulus (test 18) pH-test (test 23.2) Suitability test (test 22.1 )
17,6 % 21,6 % 1,50 1,90
10% cement 12,11 12,13
10 % cement 10 % cement 10 % cement
1,6 N/mm = 1,6 N/mm = 1,5 N/mm 2
Determination content of cement (test 22.2) 8 % cement 10 % cement 12 % cement
2,7 N/mm = 5,0 N/mm = 3,7 N/mm 2
The optimal content of cement is derived from the results in table 2 and the fea.sibility study with several batches of polluted soil (ref. 1). In the feasibility study an emperical relation between moisture content and cement content has been observed. It is concluded that the water-cement ratio should be 1:1. The moisture content of the batch was in the range 17,6 - 21,6%, so, according to the feasibility study the cement-content should be about 19%. The standard for the compressive strength of proctovs is 5,0 N/ mm 2. From table 2 a cement-content of 10% was deduced. In the demonstration-project, the mean of these two contents, 14,5% of cement, was used.
Practical aspects of realisation
The most easy way to realize the soil-cement stabilization layer is the mixed-in-place technique. Because of the instable subsoil at the site, it was not possible to get a proper mixture and condensation of the soil-cement. By using the mixed-in-plant proces, it was possible to get a good mixture of soil, cement and water. Next, the mix was put into the site and rolled with a road-roller and covered with an asphalt layer. The area of the soilcement stabilization is about 1.000 m 2.
Physico-mechanical and environmental aspects
The results of the physico-mechanical tests are displayed in table 3. According to Dutch standards the compressive strength of cylinders should at least be 1,5 N/ram 2. From tabel 3 it can be concluded that the compressive strength of all samples meets this
820
standard. From our experiences with traditional sand-cement stabilizations, the distribution in the results is in the same order of magnitude. After wet-dry and freezethaw cycles, the cylinders keep their strength, so the sustainability is judged to be good. Table 3:
physico-mechanical aspects
physico-mechanical aspect
Volumic weigth (kg/m 3,)
average
median
minimum
maximum
28 d
8m
28 d
28d
8m
28d
2000
1972
1998
1933
1935 2079
2,8
1,6
3,0
8m 2000
(n=16) (n=3) Comp.strength (N/mm 2 )
3,1
4,7
6,2
7,1
(n=16) (n=3) After freeze/thaw-cycles (n=8) - comp. strength
3,5
2,9
1,6
8,2
- volumic weigth
2013
2007
1960
2090
After wet/dry-cycles (n=8) - comp. strength
4,5
3,2
1,9
7,8
- volumic weigth
1944
1919
1835
2043
The environmental research consisted of tank leaching tests. After 28 days and after 8 month 3 samples were tested. These tests should be carried out for a period of 64 days, according to NEN 7345. As the granular soil already met the leaching standards, the leaching was tested for control. Most of the leaching of the components takes place in the first 4 days. A test for the period of 4 days normally gives a sufficient indication of the leaching behaviour. In table 4 the results of the leaching tests are given. The leaching is much below the trigger values. The base is environmentally safe and no isolating provisions are required. Table 4:
environmental aspects, range of leaching in mg/kg
sample after 28 days (leaching during 4 days) sample after 28 days (leaching during 64 days) sample after 8 month (leaching during 4 days) triggervalue reuse (leaching during 64 d)
Cupper
Lead
Zinc
1,93- 2,66
1,92- 3,26
1,89- 1,92
4,96
3,26
3,24
3,10 -4,36
1,74- 1,81
1,74- 1,94
51
120
200
821
Legislation
In the Dutch legislation, the Environmental Management Act, a licence is required for the treatment of waste-materials. For the treatment of hazardous waste, an Environmental Impact Assesment has to be submitted. In that way, the on-site immobilization of waste materials in case of the beneficial use in infrastructural works, also needs a licence. In this demonstration project the environmental licence to perform the project was no obstacle, because it was part of a larger project which already demanded a licence. In following immobilization projects, permission can be an obstacle because of the long procedure time and the large amount of information which should be submitted. In many cases an environmental impact assessment study will be demanded.
Costs
The costs of the soil-cement stabilization are comparable to the costs of a standard sand-cement stabilization and a base of granular, broken debris (so-called 'Repak'). The positive gain of the benificial use of the contaminated soil are the costs of disposal in the landfill (about 80 Dutch Guilders a ton, which do not have to be paid). The negative gains are the costs of additional cement, additional handling, environmental and physico-mechanical research. The differences in costs of the soil-cement stabilization layer are about f 1 ,-/m 2 in relation to repak and f 40,-/m 2 in relation to sand-cement. Soil-cement is the economical most benificial option. The use of cement is a critical aspect for an economic advantage. A lower content of cement will give a substantial reduction of the costs. The cement content in the soilcement stabilization is 4 to 6% more than in a regular sand-cement stabilization. In case of a small batch of polluted soil the costs of environmental and physicomechanal research are relatively high. The use of contaminated soils for a cement stabilization in a (road-) base seems to be economical beneficial when the total amount of soil is at least 500 tons.
Conclusions
The conclusions of the project were: The leaching of metals is in accordance with Dutch standards. The base is environmentally safe and no isolating provisions are required. The compressive strength is sufficient according to the standard of 1.5 N/mm 2. The distribution in the results is equal to those of sand-cement stabilizations. The durability is judged to be good. Considering the costs of landfilling, a minimal batch size of 500 tons of soil is economical beneficial, when it is used in infrastructural works. The site, including its base, is still in use, and no problems are being encounterd. In future projects a critical review on the optimal content of cement is recommended. The results of standarized tests in a laboratory with each batch of soil are preferable to the use of the empirical relation of the water-cement ratio. The need for licences and an environmental impact assessment will cause problems for the on-site immobilization of hazardous materials in constructions. Certification of processes and contractors (quality management) are recommended as important alternatives.
822
References 1:
-
2:
-
Kroes P.J. and J. van Leeuwen, Contaminated Soil Cement Stabilizations for Application as a Construction Material in "Environmental Aspects of Construction with Waste materials" (WASCON-proceedings), (1994). Gemeentewerken Rotterdam, Ingenieursbureau Milieu, Immobilization of non-cleanable soil for the beneficial use as a foundation (in Dutch), (1996).
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997Elsevier Science B.V. All fights reserved.
823
S T A B I L I Z A T I O N OF A G A L V A N I C S L U D G E B Y M E A N S OF C A L C I U M SULPHOALUMINATE CEMENT R. Cioffi 1, M. Lavorgna 2, M. Marroccoli 3 and L. Santoro 2 I Dipartimento di Ingegneria dei Materiali e della Produzione, Universith di Napoli "Federico II", P.le Tecchio 80, 80125 Napoli, Italy. 2
Dipartimento di Chimica, Universit~ di Napoli "Federico II", via Mezzocannone 4, 80134 Napoli, Italy.
3
Dipartimento di Ingegneria e Fisica dell'Ambiente, Universith della Basilicata, via della Tecnica 3, 85100 Potenza, Italy.
ABSTRACT A solid waste containing heavy metals from galvanic treatment has been stabilized by means of a binding matrix containing [3-2CaO.SiO2, 4CaO-3AI203.SO 3 and CaSO 4 able to give calcium silicate and trisulphoaluminate hydrates upon hydration. Experiments have been carried out with mixtures containing up to 60% waste under three different points of view, as follows. The influence of the waste on the technical properties of the stabilized products, the leaching behaviour under four different conditions and the effect of the leaching medium on the binding matrix have been studied.
1. INTRODUCTION In the light of the most recent directives of European Community regarding solid wastes management, stabilization/solidification processes will play in the near future a more and more important role because it will be only allowed to dispose of inert or stabilized residues. These directives have been acknowledge by the Italian Government by means of a specific act dated 5 February 1997. This act gives high priority to the development of technologies addressed towards recycling and reuse of solid wastes as well as recovery of raw materials and energy from the wastes themselves. Among the available stabilization technologies, the most frequently applied are cement-based. They rely on the formation of a calcium silicate hydrate matrix and make use of a number of systems such as ordinary portland cement, blast furnace slag and mixtures of lime and coal fly-ash (or other pozzolanic materials). A wide range of residues is currently stabilized by means of these inorganic processes, that is industrial solid wastes containing heavy metals, nuclear wastes, municipal solid wastes incineration ashes, sludges from wastewater treatment
824 plants and so on. In addition, this technology can also be applied to the stabilization of contaminated soils and sediments. In these cases the addition of bentonite can be helpful to reduce the amount of binding/stabilizing matrix. Once a specific stabilization process has been selected as the most appropriate for a specific solid waste, a number of questions must be answered in order to assess its environmental feasibility. These questions belong to three different technical fields, as follows. First of all, some interactions will arise between the binding/stabilizing matrix on one side and the waste components on the other. These interactions must be properly studied in order to understand how the binder performance changes in response to the waste admixture. This is particularly important if the ultimate scope is the reuse of the final product. Furthermore, the process must be environmentally acceptable which means that the release of contaminants due to leaching must be studied in different conditions in order to get an understanding of what the long term behaviour of the stabilized product in the environment will be. The test conditions in which leaching should be carried out will have to be designed keeping in mind the ultimate scope of this aspect of the of the research. Last but not least, and in addition to what just stated, emphasis should be given to the interactions which may arise in the environment between leaching media and the binding matrix itself. This aspect has been generally disregarded by the researchers in the field, but is particularly important because the long-term exposure to leaching media may result in complete release of contaminants if the stabilizing matrix undergoes substantial modification. In this paper a solid waste from galvanic treatment containing heavy metals, mainly Cd, Cr, Cu, Ni, Pb and Zn, has been stabilized by means of a novel cementitious matrix based on calcium silicate fl-2CaO.SiO 2 and sulphoaluminate 4CaO'3AI203"SO 3. Upon hydration this matrix forms calcium silicate hydrate and calcium trisulphoaluminate hydrate (ettringite). It was tested in previous work for both physico-mechanical and stabilizing properties proving to be suitable for the application under study. The study referred to in this paper deals with physico-mechanical properties of stabilized samples, release of metals in different leaching tests and matrix behaviour during leaching.
2. EXPERIMENTAL The components fl-2CaO.SiO 2 and 4CaO.3AI203-SO 3 of the binding matrix were synthesized in the theoretical ratio 1:1.5 by firing a raw mixture of CaCO 3 (45.37%), bauxite fines (31.09%), zeolitic tuff (13.27%) and CaSOn-2H20 (10.27%) at 1200~ for 90 min. The binder was obtained by adding anhydrous C a S O 4 to the fired mixture in the ratio 1:2.5. The chemical composition of the waste, bauxite fines and zeolitic tuff is reported in Table 1. Binder-waste mixtures were prepared containing 0 (pure binder), 20, 40 and 60% of waste and hydrated at 25~ 100% RH, and water/solid ratio equal to 0.4, 0.46, 0.5 and 0.56 for the systems containing 0, 20, 40 and 60% waste, respectively. These values were chosen in order to get constant workability. The hydration time ranged between one hour and 28 days (672 hours). Small samples (about 3 g) of each system have been used to study the kinetics of hydration. This part of the study has been carried out by determining the amount of chemically combined water by ignition at 1000~ for the time required to reach constant weight. In addition, the formation of hydrated products has been monitored by differential thermal analysis (DTA).
825 Table 1 Chemical composition of waste, bauxite fines and zeolitic tuff (wt%) LOI*
Waste 34.60
SiO2 A1203 CaO
48.12 7.62
K20 Na20 SO3 Fe203 MgO Cr203 NiO CdO ZnO CuO PbO MnO *LOI = Loss on ignition
Bauxite fines 25.40 4.70 51.50 0.02 -
0.15 0.44 6.57 0.98 0 04 0.04 0.02 001 0.03
15.3 0.04
Zeolitic tuff 9.93 52.93 17.21 3.54 7.26 2.97 0.13 3.71 1.44
_
Samples of the three mixtures containing 20, 40 and 60% of waste have been submitted to the following three leaching tests: (a) the dynamic TCLP test [ 1],with pH 4.94 acetic acid/sodium acetate buffer, liquid/solid ratio equal to 20 ml/g and leachant renewals at 1, 3, 8, 14, 24, 48, 96, 168, 376, 672 and 1344 hours on monolithic cylindrical samples dxh=2x3 cm2; (b) the same test as (a) but with pH 3.86 CO2-saturated solution instead of acetic buffer and renewals up to 672 hours; (c) the static US-EPA test [2] with pH 5 acetic acid solution and ratio liquid/solid equal to 16 mug on granulated sample (size < 9.5 mm) and (d) the availability test [3] carried out on 3 g of pulverized sample (size < 180 ~tm) with 150 ml of pH 7 nitric acid solution for 3 hours plus additional 150 ml of pH 4 HNO 3 solution for 3 hours. Following these tests, the leaching solutions have been analysed by means of atomic absorption spectroscopy and the solids before and after leaching have been characterized by means of DTA and scanning electronic microscopy (SEM).
3. RESULTS AND DISCUSSION Figure 1 shows the amount of normalized chemically combined water. The absolute values have been divided by the fraction of binder present in each mixture to get the normalized values. It is seen that only the mixture containing 20% of waste behaves like the pure binder, while in the cases of the 40 and 60% mixtures the waste does not simply dilutes the binder but its components inhibit the binder hydration. This effect increases as the content of waste in the
826 mixtures increases. 40 The hydrated samples have been t~ submitted to DTA -~ 30 stowing that in all the c d~ systems the main E o hydration product is .L~20 ettringite. As in similar [] Pure binder systems studied O 20% waste previously [4-6], other A 4 0 % waste 4 1 0 V 60% waste hydration products are calcium silicate hydrate, aluminium hydroxide I 0 ~l I I I I I I gel and traces of 672 13 7 1424 72 168 336 calcium monosulphate Time of hydration, hours (root scale) hydrate. Significant Figure 1. Normalized chemically combined water for the differences were only pure binder and the three mixtures binder-waste. observed at short hydration times as shown by the thermograms of Figure 2, relative to 1 hour hydration. In this figure the thermogram relative to the pure binder shows the main endotherm at 82~ that reveals the presence of ettringite. The three minor endotherms at 128, 205 and 260~ are related to dehydration of CaSO4"2H20, calcium monosulphoaluminate hydrate and aluminium hydroxide gel, respectively. The thermograms / / '~J/ /~ 60% relative to the systems containing 20, 40 and 60% of waste show that neither ettringite, nor calcium monosulphate hydrate form at 1 hour hydration when the system contains the waste. In these thermograms the endotherms related to dehydration of CaSOn'2H20 show that the waste catalyses the conversion of / \ G 0% anhydrous CaSO 4 to CaSOn-2H20, according to previous findings relative to similar systems [7]. The endotherms relative to , I i I i 1 t 1 , I dehydration of aluminium hydroxide are of 0 100 200 300 400 500 Temperature, ~ higher intensity in the cases of the wastebinder mixtures because this compound is simultaneously a hydration product and a Figure 2. Thermograms of samples aged component of the waste. The particular 1 hour. E: ettringite G: gypsum; M calcium baseline shape of the thermograms relative to monosulphate hydrate; A: aluminium the waste-binder mixtures is due to the hydroxide.
827 presence of waste. The values of unconfined compressive strength are 25 MPa for the pure binder and 21, 16 and 2 MPa for the mixtures containing 20, 40 and 60% of waste, respectively. These values are such that recycling the stabilized products in the field of building materials is possible for waste content up to 40%. The value of 2 MPa observed for the mixture containing 60% waste exceeds the value of 0.44 MPa recommended by Stegemann and Cot6 for segregated landfill disposal [8]. Figures 3 and 4 report the results of the 100 dynamic leaching tests carried out with the ~ Ca 10 acetic buffer and CO 2saturated solution, respectively. Table 2 ---_-2_-2 ....... shows the results of US,r _. . . . . -o---N~ O EPA leaching test and 0.1 availability test. The [] 20% waste O 40% waste data are relative to the Zk 60% waste 0.01 metals Cd, Cr and Ni, rj whose presence is of greater environmental 1 i I I 0.001 concern, and are 0 500 1000 1500 Time, hours expressed as percentages of the initial Figure 3. Results of TCLP test. quantity present in the sample. These results show that the leaching 10 Cd behaviour depends strongly on the nature of the metal as Cr, Ni and ::::::::::::::::::::::::: ............[] ................... Cd are released in -~.. ]~: ..[]..... steeply increasing
~
.
.
.
.
.
.
--lk amount in each test. The 0 " I ~) IE, P ! .....O -,a- . . . . . _Nj. . . . . . . .. . . ~ . . _ .-ZX--. . . . . . _ ..... chemical nature of the leaching medium has ~0.01 ~ V1 20% waste also a strong effect ~' O 40% waste inasmuch as quite larger amounts are released in TCLP test compared to 0.001 0 150 300 450 600 750 CO2-saturated solution Time, hours test. To this regard, it is important to point out Figure 4. Results of CO2-saturated solution leaching test. that pH does not have the effect that one would expect. In fact each step in the CO2-saturated solution test starts with a pH value of 3.86 and ends at about 5. These values, compared to the value of 4.94 of the acetic buffer make clear that pH is not the most important factor that characterize the leaching
828 medium. Table 2 Results of US-EPA leaching test and availability test (wt%) Type of test and waste content US-EPA 20%
Availability
40%
60%
20%
40%
60%
Cd
2.29
1.78
3.54
58.93
56.96
55.66
Cr
0.16
0.05
0.05
0.70
0.47
0.56
Ni
0.30
0.22
0.42
17.47
10.76
10.37
The results of Figures 3 and 4 show that the percentage of metal released increases with the amount of waste in the mixtures and this means that the binder does not simply physically segregate the waste from the attack of the leaching media, but that chemical interactions take place between the binder and the waste components whose extent decreases as the binder content decreases. Finally, the data of Table 2 confirm that the results of a leaching test depend strongly on the chemical nature of both the metal and the leaching medium, and also on the physical nature of the solid sample. The effect of the leaching medium on the binder in the stabilized samples has been studied only in the two dynamic tests carried out with the acetic buffer and CO2-saturated solution. The monolithic cylindrical samples submitted to these tests underwent significant modification which could be evidenced by cutting the samples themselves orthogonally to the cylinder axis. In this way it was possible to distinguish between an external leached layer of increasing thickness and an internal unleached shrinking core. The thickness of the leached layer increased with time and with the amount of waste in the mixture and reached the values of about 2, 3, and 7 mm at the end of both tests in the cases of 20, ._._b_b~ 40 and 60% waste, C respectively. Figure 5 shows the results of DTA I i I I i I carried out on samples 0 200 400 600 800 Tempcrature,~ of the leached layer and the unleached inner core Figure 5. Results of DTA carried out on samples from in the case of the external layer (a) and inner core (b) of system containing mixture containing 40% 40% waste after CO2-saturated solution leaching test.
829 waste submitted to the CO 2 -saturated solution test. It is seen that ettringite undergoes substantial decomposition to gypsum, aluminium hydroxide and calcium carbonate. This observation can also explain why lower amounts of metals are leached in the CO 2 saturated solution test. The formation of metal carbonates can limit metal ions solubility even if the pH is lower than in the TCLP test. This chemical modification of the system implies also substantial morphological modification, as it can be seen in Figure 6 where two micrographs of samples of the external leached layer and the inner unleached core are shown. The system of Figure 6 is that containing 20% waste submitted to TCLP test. The needlelike ettringite crystals visible in micrograph (a) completely disappear after leaching, as seen in micrograph (b). This observation, relative to the sample submitted to TCLP test, is not in contrast with the results that ettringite decomposes in CO2-saturated solution giving calcium carbonate among the other products. In fact, in a previous work carried out on systems similar to those studied in this paper [5], the acetic buffer attack caused ettringite decomposition into calcium acetate pentahydrate, gypsum and Figure 6. SEM micrographs of samples of external layer (below, aluminium hydroxide. b) and inner core (above, a) of system containing 20% waste after TCLP test.
830 4. CONCLUSIONS The results of the experiments can be summarized as follows with respect to the three aspects of the study emphasized in the introductory part. From the point of view of binder performance, the addition of waste to an extent greater than 20% inhibits the hydration process. On the other hand, mechanical performance is reduced starting from 20% waste addition. The leaching behaviour is strongly influenced by the chemical nature of both the metal and leaching medium. Other parameters that greatly influence metal release are the physical nature of the stabilized samples and type of liquid-solid contact. Of the two tests carried out under the same conditions, namely TCLP an CO2-saturated solution, the latter gives rise to lower amounts of metals released. Finally, it has been found that the attack of the leaching medium to the stabilized samples causes severe modification of the binder. This modification has been characterized from both the chemical and morphological points of view, showing that ettringite decomposes to gypsum, aluminium hydroxide and a calcium compound. The possibility that metal carbonates precipitate during the CO2-saturated solution test can explain the lower metal ions release observed in this test in comparison to TCLP test.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8.
Toxicity characteristics leaching procedure, Federal register, Vol. 51, No. 261 (1990). U.S- EPA Test method for evaluating solid waste, SW-846. Washington, DC: Office of Solid Waste and Emergency Response (1986). NEN 7341 (formerly NWN 2508), Determination of leaching characteristics of inorganic components from granular (wastes) materials. NNI, Delft (1993). J. Beretka, B. de Vito, L. Santoro, N. Sherman and G.L. Valenti, Cement and Concrete Research, 23 (1993) 1205. V. Albino, R. Cioffi, M. Marroccoli and L. Santoro, Journal of Hazardous Materials, 51(1-3)(1996)241. R. Berardi, R. Cioffi and L. Santoro, Journal of Thermal Analysis, in press. V. Albino, R. Cioffi, L. Santoro, and G.L. Valenti, Waste Management and Research, 14 (1996) 29. J.A. Stegemann and P.L. Cot6, Science of the Total Environment, 178(1-3) (1996) 103.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
831
Reuse of Secondary Building Materials in Road Constructions T. Berendsen The Environmental Engineering Department of Public Works Rotterdam, P.O. Box 6633, 3002 AP Rotterdam, The Netherlands.
Summary
Slags, asphalt and coal-ash, have long been used as secondary building materials. They are released when roads are reconstructed. Their reuse is presently prescribed in the Provincial Rules 'Working with Secondary Building Materials'. Mid1998, a new national decree will become effective: the 'Building Materials Decree'. It lays down the guidelines for design, realisation and maintenance of constructions, aiming at avoiding contamination of the soil by secondary building materials. The implementation of this policy requires road constructors to adopt a new approach. They have to do environmental research, and pay attention to design, isolation measures, maintenance and possible monitoring measures. In practice this means that adhering to a schedule concerning environmental research is essential. The civil engineer needs to implement environmental aspects into his basic planning, design and realization process. The Environmental Engineering Division of Public Works Rotterdam has developed a working model which incorporates the Building Materials Decree into the current engineering planning process. This working model is an instrument used by the Port of Rotterdam.
1.
INTRODUCTION
The city of Rotterdam produces large quantities of waste. Limited space, not only in Rotterdam but also in the rest of the Netherlands, prohibits waste material landfilling. So when secondary building materials, such as slags, asphalt, coal-ash and fly-ash, are released, they are usually reused. The municipality of Rotterdam also stimulates the reuse of building materials, because it saves on new raw materials and on the increasing tariffs for landfilling. Project leaders ought to keep in mind this economic benefit. Technical road reconstructions will confront engineers and workers with secondary building materials from the past, both in foundations and asphalt layers. Handling of these materials is outlined in rules and regulations, such as the 'Building Materials Decree' and the memorandum 'Working with Secondary Building Materials'.
832
In this paper special attention will be paid to secondary building materials because nearly 10 % of road foundations in the city and port of Rotterdam contain secondary building materials, a quantity large enough to deserve special attention.
2.
PROBLEM ANALYSIS
How to put into practice the policy of reusing secondary building materials when reconstructing roads? When we studied the rules and regulations in view of the practice, we found a few bottlenecks. The most obvious one is that the quality of the secondary building materials released during road reconstructions is unknown. Furthermore, the necessary contamination- and leaching tests take a lot of time and cause stagnations, and in some cases involve unexpectedly high costs of extra measures. Another problem is how to deal with small-scale projects, in which the reuse of secondary building materials is not financially interesting. To make sure that the reuse of secondary building materials is successful in the whole region, it is important to set up a good organization with a good infrastructure and knowledge of the market. The Public Works Department in Rotterdam is aware of these problems. Therefore, in order to avoid the bottlenecks, its Environmental Engineering Division developed a working model called: Pragmatic application of secondary building materials in road constructions.
3.
STARTING POINTS
As starting points for this working model two aspects have been studied. In the first place the applicable rules and regulations and in the second place the current practice of road constructions in Rotterdam. The working model described in this paper integrates these two aspects.
4.
RULES AND REGULATIONS
Five official decrees form the basis of the working model. The three most important are: (1) the Building Materials Decree, (2) the Project Decision Building Materials Decree, and (3) the memorandum Working with Secondary Building Materials. Less important are (4) the Waste Dumping Ban and (5) the Installation and Licenses Decree. These rules and regulations in relation with the process of road reconstructions are summarized below: ad(1) The Building Materials Decree sets limiting conditions for using primary and secondary building materials in civil projects in land- and waterbottom. It aims at establishing a national, general protection level for soil and at stimulating the reuse of secondary building materials. It applies to granular (unmoulded)
833
or stony materials, such as ashes, slags, soil and matured harbour sludge, used in the open air. Wood, steel, clean soil, and materials with parameter concentrations below the prescribed leaching level are excluded The Building Materials Decree divides building materials into three categories on the basis of composition values for organic compounds and immission values for inorganic compounds in building materials. Category 1 are building materials that do not exceed the composition and immission values. The chance of diffusion is very low. Therefore they can be used without special conditions, such as isolation. Category 2 are building materials exceed the composition values, but exceed the immission values without isolation. These materials need to be isolated from percolation and groundwater to minimize the risk of diffusion of chemical pollution, and are also subject to specific management and maintenance rules. The third is a special category, such as bottom ash from waste incineration and tarry asphalt granulate. Reusing those requires more isolation constructions. ad(2) The Project Decision Building Materials Decree (effective as per 6 December 1995) is a practical elaboration of the former. It helps you to decide on the provisions and management rules for category 2 and the special category. It deals, for instance, with determining the distance between the average highest groundwater level and the secondary building material, and with the difference between moulded and unmoulded waste. It also contains guidelines for leaching tests and isolation constructions, and a checklist for inspection and maintenance. For example, the surface of the road could function as an isolation for rain percolation. It is necessary to keep this intact by means of inspections and maintainance works. ad(3) The Building Materials Decree will become effective in phases. Its full implementation is expected in 1998. Until then an interim policy of the associated Dutch provinces: 'Working with Secondary Building Materials' is valid. Its most important requirements are: Determine the application category by contamination- and leaching tests. Take special isolation measures (e.g. liner) and environmental control measures, in planning and realization. Registration of the nature of the materials and the exact location is essential. When the location gets another function, it is obligatory to remove the secondary building materials. In general this interim policy is quite similar to the Building Materials Decree. ad 4) The 'Waste Dumping Ban' of 27 June 1995 lists all the building materials that are not allowed to be dumped, for example fly-ashes, building debris, sieve sand, purification sludge, contaminated soil, household and industrial waste.
834
The memorandum of implementation further explains the features of the parameters. ad 5) The 'Installation and Licences Decree' became effective on 1 March 1993. It applies to installations using more than 50 M 3 building materials from outside in or at the soil. These applications are subject to a license from the competent authority, unless the building materials are directly used in a civil project in an environmentally acceptable setting.
In review: What are the requirements and duties in relation to the different categories CATEGORY
REQUIREMENT/DUTY 1
2
B o t t o m ash f r o m
TAG
w a s t e incineration d u t y of reporting to authorities for soil > 50 m3 d u t y of reporting to authorities other than for soil
*
*
making a plan of e n v i r o n m e n t a l m a i n t e n a n c e
*
*
d u t y of taking back
*
*
d u t y of removal
*
*
possible e x e m p t i o n of d u t y of r e m o v a l
-
*
W a t e r pollution act
*
minimum amount 1 , 0 0 0 t o n n e s (roads) 1 0 , 0 0 0 tonnes (large civil projects)
-
isolation measures
-
*
2x
*
supervision and m a i n t e n a n c e
-
*
*
*
-
*
*
*
m a x i m u m a m o u n t in surface w a t e r
111
~
C U R R E N T P R A C T I C E OF R O A D C O N S T R U C T I O N S IN THE M U N I C I P A L I T Y OF R O T T E R D A M
Apart from the limiting conditions, i.e. rules and regulations, the practical feasibility also influences our model. Therefore, it is important to have insight into the road constructing process. We used the process developed by the Management and Maintenance Department of the Harbour Authority of Rotterdam, the so-called Procedure of Programmed Maintenance d.d. 2. June 1992. It outlines the different stages of road reconstructions. The programming phase is dedicated to planning. The main task is to realize a cost prognosis expressed in an estimation. The participants will form a project team. The next step is the phase of initiative. The most important part is to collect the
835
necessary information and to discuss with the other participants the ingredients for the program of requirements. Also the drawings of initiative will be made, and attention can be paid to the influence on the budget because of global knowledge of the quality of the soil and the released building materials. In the design phase the definitive design of the project will be completed. Subactivities are: plan development, definitive program of requirements agreed with the project team, sending the technical program of requirements to all departments and people involved, make and check the concept drawings, make credit estimation and financial proposal. In the technical preparing phase the specification with plans will be made and submitted. The budget estimation will be prepared and the order will be placed out conform this specification. The realisation phase sees the supervision of the realisation process, and sometimes change the realisation order. In the final phase, called transfer, the civil project will be transferred to maintenance management. Environmental maintainance can be placed under the same management, but in practice it will be done by rational road mantainance (for instance inspections of road damages). The most important activities are: inspection of the object, is everything constructed conform the plans, considering the costs by transfer, storing all information, such as drawings, into the archive or data base. These aspects determine the further amount of money, which will be involved in the maintenance of the project. A one-year intake period of transfer is advisable. An adequate transfer contract containing a liability clause for the constructor is essential.
m
IMPLEMENTATION OF RULES AND REGULATIONS IN PRACTICE: THE MODEL "PASM"
The activities within the framework of active building material management, conform the policy of reusing secondary building materials, are based on the usual working process of road reconstructions and have been condensed in a model called PASM: Pragmatical Application of Secondary building Materials. The model is a process schedule including the usual stages of programming, initiation, planning and design, technical preparation, execution, and transfer.
INITIATION
Planning
I
Model PragmaticApplication of Secundairy Waste Materials.
1Collectinformation~ of secundary| waste materials ~
No
836 , Yes temporary landfill
characteristicsof materialsto be used
No
Yes ....
12 Weeks
(4 Carryingout---~ characterising ~
I I
Yes I I
t
Technical preparation
I
design I constructionas to requirements category2 J ./
9 Weeks
environmental f planl+PBT S
_
C0nstruction
.......
I
7
._
6i -- " ~-i Informcompetent~ L authority
Building ~, management
Building 1 management
nvironmental Transfer
I ~-
Transfer
environmental care
)
~
Sg
[-
V T.... fer J
,
10' Maintenance
Finish.
/
J
837
7.
E X P L A N A T I O N OF THE S C H E D U L E
The activities in the schedule are further explained below. They have been arranged according to the stages of regular road reconstruction. When it is proven that the building material belongs to category 1, the shortcut on the right-hand side is applicable
Initiation The program of requirements of the present and the desired situation is drawn up in this stage. In view of the design of the new road construction, the following is important: 1. Collect general information about the quality of the soil at the construction site, about the level of pollution, and about the categories of the building materials to be released. It is essential to have insight into the categories. Expected extra work and extra costs should be reported to your principal and your partners.
Planning/design In this phase a program of requirements will be made and the definitive design will be agreed on. Will material from the old construction be used? If No, continue with step 3. .
.
,
Determine whether released materials need to be dumped temporarily. Temporary landfill on location could be arranged with the authorities. Determine the need to examine secondary building materials for establishing the right category. In most cases a leaching test is necessary (furnace slag, asphalt granulate, sieve sand etc). Building materials certified by the Ministry of Housing, Physical Planning and the Environment need not be examined. In the latter case, proceed to step 6a. Define the application category of the final product. Generally, cement will be added to the released secondary building materials in order to obtain a goodquality final product. This final product should be subjected to leaching tests, and its composition must be assessed in order to determine the application category. These tests take about 12 weeks. If the result is category 1, the only required procedure is notifying the authorities.
838
Technical Preparation In this phase the construction drawings will be prepared in accordance with the functional program of requirements. The technical programm of requirements will be detailed in view of the project's realization. .
.
6a.
Design adaptations in conformity with the application requirements for category 2 secondary building materials. These materials must be isolated from rain- and groundwater. Submit licence (PBT) one month before the start of the project. Draw up the environmental maintenance plan phase 1, describing tasks and responsibilities for supervision, inspection, and maintenance. Report to the competent authority. For category 1 two days before starting will do. Continue to step 9.
Realization stage In this stage the project will be realized. ~
,
Building management is the main task. It is important to make correct drawings of the isolation constructions and the installation. Add last corrections to the environmental maintainance plan. The subsequent final plan will describe the exact location of isolation measures and the maintenance tasks.
Transfer In this stage the object will be transferred to the manager of the technical maintenance department. In view of long-term and current road maintenance, the following is essential: .
10.
Transfering of the object. Final inspection, and collecting the final construction drawings and other information such as the type, location, amount, and characteristics of the secondary building materials and the isolation measures. It is important to store this information in an adequate data base and keep it for at least five years. The technical maintenance manager is responsible according to the rules and regulations. This step involves environmental maintenance, such as inspections, substitution, and repairs, for which the responsibility can be transferred from the owner to the technical maintenance manager.
10a This step is only applicable to category 1 secondary building materials. Normal maintenance of the surface of the road is sufficient.
839
8.
CONCLUSION
To avoid infraction of rules and regulation, as well as stagnation and high unexpected costs it is important to plan environmental research of the secondary building materials and the soil at the location at an early stage. The civil engineer needs to implement environmental aspects into his basic planning, design and realization process. The Environmental Engineering Division of Public Works Rotterdam has developed a working model which incorporates the Building Materials Decree into the current engineering planning process. This working model is an instrument used by the Port of Rotterdam.
9.
DISCUSSION
Environmental research, and particularly leaching tests, takes usually a lot of time and can be the critical path of the project. In many cases it may cause much stagnation and high unexpected costs. Another bottleneck for managers developing road constructions, is uncertain estimates if information about the building materials is lacking. As we have seen, the quality of the secondary building materials will determine the final costs of the extra measures. A third bottleneck is the situation when less than 1000 tonnes of secondary building materials becomes available. The costs of environmental research and special isolation measures are then out of balance with the total costs of the whole road construction project. Further, stagnation of small-scale road constructions in a city centre, with its many underground mains and busy traffic, should be avoided. In the Public Works Department of Rotterdam we try to tackle these bottlenecks by using the described working model in the first place. Aiming at more accurate cost estimations we are undertaking a network research of secondary building materials by checking old archives and doing field reseach. For various reasons it is important to collect all this information in a data base. The main one is that the information can be used as a tool for the market: When is what and where becoming available? At the centre of collected information a special organization could function as a 'broker' in secondary building materials, and in fact is already operating in Rotterdam. Having gained much knowledge and experience about the characteristics of secondary building materials, it should be possible to come to agreements with the authorities over the necessity to speed up the procedures for small-scale applications of secondary building materials.
840
10. REFERENCES
1) 2) 3) 4)
5) 6) 7)
Dassen W.G, Piersma W, Schelwald R, Vries I.M.J, Re-use of waste metrials in constructional works; experiences in the city of Rotterdam,the Netherlands, Waste Materials in construction, Elsevier 1991; Berendsen T, Kooman J, Handleiding bij reconstructie en groot onderhoud van wegen, environmental engineering division of Public Works Rotterdam, 13 juni 1996; Berendsen T, Kooman J, Achtergronddocument Milieubeheer bij reconstructie en groot onderhoud van wegen, environmental engineering division of Public Works Rotterdam, 13 juni 1996; Bouwstoffenbesluit bodem- en oppervlaktewaterbescherming. Besluit van 23 november 1995. Staatsblad, 1995 567. Provincie Zuid-Holland. Nota 'Werken met secundaire grondstoffen;. Den Haag, mei 1995. Stortbesluit Bodembescherming. Besluit van 20 januari 1993, houdende regels inzake het storten van afvalstoffen. Uitvoeringsregeling Bouwstoffenbesluit. Directoraat-Generaal Milieubeheer, Directie Bodem. 's-Gravenhage, 20 december 1995.
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
841
MSWI RESIDUES IN THE NETHERLANDS PUTTING POLICY INTO PRACTICE
Jan G.P. Bom, Ralph A.L. Veelenturf Service Centre MSWI residues, c/o Waste Processing Association, P.O. Box 19300, 3501 DH Utrecht, The Netherlands
Abstract
The Dutch policy with regard to the residues of Municipal Solid Waste Incineration (MSWI) aims towards maximization of useful application and minimization of required volume for disposal of these residues. This policy has been put into practice successfully for MSWI bottom ash. During recent years, virtually all bottom ash has found a useful application in road construction and embankments. In the case of MSWI fly ash, the policy led to the use of this material as a filler in asphalt for road construction. The demand for asphalt fillers containing MSWI fly ash, however, is limited. As a result only 20 - 30 % of the MSWI fly ash produced has been usefully applied as an asphalt filler. For the residues from flue gas scrubbing, no feasible useful application has been found to date. As a result, the entire production has been disposed of in landfills. Long term policy, however, also aims towards the development of uses for flue gas scrubbing salts.
1. INTRODUCTION The principle goal of MSW incineration is reducing the amount of space required for disposal of wastes in landfills. Incineration of MSW is preferred in the Netherlands. This preference is also an integral part of the regulations currently applicable in this respect. Incineration of MSW results in residues, typically occupying about 10% of the original volume (and 25% of the weight) of the incinerated waste. Useful application of these residues further increases the waste reduction, and therefore reduced the amount of space required for landfills. The Dutch policy not only takes into account the limited space available for land-fill, it also aims towards preserving natural resources. The total annual consumption of building materials in this country amounts 140 million tons. This consumption - primarily sand, gravel, clay and marl - results in a considerable loss of natural resources. These two aims lead to an increase of the useful application of residues, including those from MSW incineration. Like virtually all residues that can be used as secondary building materials, MSWI residues are contaminated, posing a potential risk of soil pollution. We are therefore faced with a paradox: use of residues is beneficial to the environment in the sense that it preserves natural resources, while in itself this use poses the potential environmental threat of leaching contaminants into the soil. To resolve this dilemma, regulatory measures have been stipulated for the use of residues, including MSWI residues.
2. GENERAL POLICY FOR MSWI RESIDUES In the period from 1990 through 1997, environmental legislation has changed drastically in the Netherlands. This has also affected useful application and disposal of MSWI residues. Briefly, the high standard of re-use of MSWI residues must be consolidated despite increasingly stringent environmental legislation. To this end, the government and the other parties involved are attempting
842 to work together as much as possible. This cooperation resulted in the creation of a 'Policy Paper MSWI residues' (March 1995). In this document (in Dutch: 'Implementation-plan AVI-reststoffen' [1 ]) a common course is set for the future. In order to coordinate the 36 proposals laid down in this Policy Paper, the 'Service Center MSWI residues' (acronym in Dutch: ACR) was established in mid1995. The ACR is a temporary organization located at the office of the Waste Processing Association. The latter will be responsible for acting on most of the proposals. The objective of the ACR is to finalize all proposals as agreed by 1998.
3. POLICY F O R MSWI B O T T O M ASH
The 'Policy Paper MSWI residues' summarizes the policy for MSWI bottom ash as follows: 1. Continuation of the current re-use figure (almost 100%) despite increased production. 2. Application of MSWI bottom ash preferably as replacement for sand in embankments in projects of at least 10,000 tons. Incentive for application in projects of 100,000 tons or more. 3. Bottom ash that does not meet common regulatory environmental quality standards (N2-status of the Dutch Building Materials Decree) is only to be used in government-controlled projects. 4. Improvement of the environmental quality in order to achieve N2-status, preferably by 1997.
4. PRACTICE OF MSWI B O T T O M ASH 4.1 Continuation of the current re-use figure (almost 100%)
In the Netherlands, raw MSWI bottom ash is upgraded prior to useful application. This upgrading has been described in detail elsewhere [2]. In short, upgrading consists of magnetically removing scrap material, and crashing and sieving to remove all components with a diameter larger than 40 ram. The non-combusted material present in the fraction > 40 mm is normally recycled back to the incinerator. I000
800 t 600 '
. . .
o
-
400_ "Z6
200 0
-1988
"'" ~ " ~ 1989
[
1990
1991
~-~ Production ~
1992
1993
Utilisation
/
1994
1995
1996
Disposal
Figure 1 Production and use of MSWI bottom ash in The Netherlands By means of this upgrading, the resulting MSWI bottom ash is transformed into a secondary building material. All MSWI bottom ash that has been produced over the past ten years has therefore found
843 useful application, primarily in embankments and road-base layers. The yearly production and use of MSWI bottom ash is depicted in Figure 1. Quality control and quality assurance can be regarded as one of the reasons for the present success of MSWI bottom ash in The Netherlands. In recent years, certification has become an important activity, nowadays covering the majority of MSWI bottom ash (Figure 2).
T
800 700 600 500 400 300 200 100 0
,Certi.ef:. . . . certi.e
. . . . . . . . . . . . . . . . . . . . . . . . . . .
m mm 1991
1992
1993
1994
1995
1996
Figure 2 Certification rate of MSWI bottom ash Untreated MSWI bottom ash contains about 10% retrievable ferrous metals. Annually some 65,000 to 75,000 tons of recovered steel scrap is recycled in the steel industry. Because the retrievable ferrous metals give a positive revenue, some of the processing costs are compensated. Most Dutch MSWIs (will) recover non-ferrous metals by applying Eddy-Current techniques as well. In 1996 some 1,700 tons of non-ferrous metals were recovered. Within a few years, this figure is expected to rise to about 3,500 to 5,000 tons (based on an expected annual production of more than 1,000,000 tons of MSWI bottom ash), depending on the methods for non-ferrous retrieval to be implemented. It should be noted that non-ferrous separation produces a better quality bottom ash and that the revenues of non-ferrous are such that it is considered a worthwhile investment. It can be concluded that every effort is being made to upgrade MSWI bottom ash to a generallyaccepted secondary building material. Over the past then years this effort has been rewarded by the market, in the sense that virtually all upgraded bottom ash has found useful application. 4.2 Preference for use in large projects Until the early eighties, MSWI bottom ash was used in predominantly small projects. The majority of these projects consisted of road-base construction and simple use as a paving material for (farm) yards. The latter projects had an average quantity of only 30 tons. This type of project has always been regarded as unsuitable by the government. The proposed Building Materials Decree (to be fully implemented in 1998) therefore prohibits the use of MSWI bottom ash in projects involving less than 10,000 tons. In the mean time, preferential use of MSWI bottom ash in large projects has been stimulated by all parties involved. This policy presents logistical problems, however. MSWI bottom ash is continuously produced while large projects that are suitable for MSWI bottom ash occur only incidentally. Most MSWI are able to stockpile the amount of bottom ash that is produced in one year, which appears to be sufficient for a majority of the projects. Recently an embankment was constructed using almost 1,000,000 tons of bottom ash. The size of this project required an intermediate stockpile near the project itself. In order to verify whether the agreed preference for large projects is implemented in practice, the ACR monitors the projects in which MSWI bottom ash is used. Figure 3 depicts the market share of
844 large (> 100,000 tons), intermediate (between 10,000 and 100,000 tons) and small scale projects (< 10,000 ton). In recent years, more than 2/3 of the bottom ash has been used in large-scale projects. In addition, projects involving 10,000 ton or less now comprise less than 5% of the market (Figure 3). One of the targets of the policy for MSWI bottom ash has therefore been implemented in practice: a majority of the material is used in large-scale projects. Moreover, in the near future the use of MSWI bottom ash in projects smaller than 10,000 tons will be prohibited by the Building Materials Decree.
1000 ~
< 10 kto n
I
10-100kton
~>100kton
I
'
800
"= •
600 400 z
~
200
1990
1991
1992
1993
1994
1995
1996
Figure 3 Size of projects in which MSWI bottom ash is used. 4.3 Preference for use in projects by order of the government The Dutch governmental structure consists of three layers: national authorities (ministry of Public Works), provinces and municipalities, which all order projects for road construction and embankments. With respect to large projects, the ministry of Public Works is the most likely candidate for ordering these projects to be constructed with MSWI bottom ash. Although this ministry is a strong advocate of use of secondary building materials in large projects, it holds the view that the provinces and municipalities also should take their responsibility in using MSWI bottom ash in their projects. 1000
800 600 ~"
400
r
200
90 I
91
92
93
Non-Government ~ Unknown
r-n Provinces
94
95
96
~ Municipalities
c--n Ministrie of PW
Figure 4 Ordering of MSWI bottom ash. The ACR monitors the actual market for MSWI bottom ash itemized per category of Government. In Figure 4 the results are shown, indicating that over 75% of the bottom ash is used by municipalities
845 and the ministry of Public Works. This result approaches the objective of 100% that has been determined in the policy for MSWI bottom ash. Yet non-government users are still needed to ensure that a considerable fraction of the production need not be dumped in landfills. It should also be noted that off-take by the ministry of Public Works can vary widely from one year to the next. In short, additional effort is required from all parties involved at this point in time if the desired level of 100% sale to government-controlled users is to be reached.
4.4 Improvement of the environmental quality The current successful use of MSWI bottom ash will be continued in the future. The strict environmental demands as presented in the 'Building Materials Decree' require the development of new techniques in order to improve the environmental quality of MSWI bottom ash. The current environmental quality of bottom ash does not meet the requirements as formulated in the 'Building Materials Decree' with respect to the leaching behavior of Copper, Molybdenum, Antimony and Bromide. Although practical uses will remain possible within the boundaries of future regulatory demands, the required precautions (such as the use of polyethylene and sand-bentonite liners) will probably weaken the market competitiveness of MSWI bottom ash. By improving the environmental quality of MSWI bottom ash, its market-share can be secured. The benefits of improved environmental quality are recognized by the 'Policy Paper MSWI residues', which has adopted this as one of the objectives for MSWI bottom ash.
%
t m Copper "0
... Molybdenum
-. 0
.
o
r n ..... ~ - - . - - - - ~ - - - - - 4 ~ - - - - - - n
,,..,
_._ Antimony
..~iii..Bromide
I
..
....
.
.
. . . - u - . . . . m ..... ~ 7 . . / - . ~ .- ~ . . ~ I . .-~
.
.
" ....-.,~,\
.
.
,r . . . . . e- ....
............................................................................................................................................................................ i~--i.~
. ~ ..
.
e.
*- .~.-.......... .:.": ~:'"~~;
~,
~"
........................................................ ~ . ~ : z - ~ H . < : : .
..................................................................
"'M
~
.m .... g
...... m .....
.... I
i
0,1
i
1991
i
i.
i
i
1992
i
i
i
i
1993
i
i
1
i
1994
I
I
i
i
1995
i
i
i
i
i
i
i
1996
Figure 5 Leaching of MSWI bottom ash related to the demands for granular materials (N2) of the 'Building Materials Decree' Briefly, the leaching of Antimony, Copper, Molybdenum and Bromide must be reduced in order to meet the general quality standards as laid down in the Building Materials Decree (Figure 5). During the period from 1992 until 1996, laboratory research and studies have been performed in search of applicable techniques for improving the leaching behavior. At the end of 1996 it was decided that both scrubbing MSWI bottom ash and accelerated aging of the materials as well as the use of additives were techniques worth testing on a larger scale. In 1997 accelerated aging of MSWI bottom ash will be tested on a pilot-plant scale (batches of 50 tons). In addition, process-integrated scrubbing of MSWI bottom ash using the quench will be tested at two different MSWIs. Depending on the results, a decision concerning implementation of one - or both - of these techniques will be made
846 early in 1998. Based on this time schedule, the required quality improvement should be reached at the end of the year 1999. Reduction of the leaching is regarded as a tool for maintaining the high level of use of MSWI bottom ash; it is not an objective in itself. In fact, the 'Policy Paper MSWI residues' aims towards 100% use of MSWI bottom ash, even if it does not meet the desired environmental quality. Should the techniques explained above fail to achieve the desired quality, or if they succeed only against high financial or environmental costs, it can as yet be decided that the techniques will not be implemented. In any case, the results of the large-scale experiments will probably lead to some debate about the question of whether the advantages of an improved quality justify the financial costs and environmental side-effects (e.g. discharge of waste water in case of scrubbing).
5. POLICY FOR MSWI FLY ASH
The 'Policy Paper MSWI residues' formulates the policy for MSWI fly ash as follows: 1. Continuation of the application as a filler in asphalt pavements. 2. Re-use of the remaining fly ash fraction by applying solidification techniques 3. In the case of disposal, quality improvement to a C3-class hazardous waste. In addition to the 'Policy Paper MSWI residues', other regulatory factors affect the manner in which policy for MSWI fly ash is put into practice. Due to its leaching behavior, MSWI fly ash is categorized as a C2-class hazardous waste. Disposal of C2-class (i.e. untreated) MSWI fly ash in landfills will be prohibited as from 1998. Improvement of the environmental quality of fly ash is therefore required in order to increase its uses and to make it possible to continue to dispose of the remaining fraction in landfills.
6. PRACTICE OF MSWI FLY ASH 6.1 Continuation of the application as a filler in asphalt
After years of decline, the production of MSWI bottom ash rose once again in 1996. This increase was brought about by the raise of incineration capacity in The Netherlands. The gradual decrease of the amount of fly ash produced can be explained by the modernization program of the MSW incineration capacity over the past years. Not only the flue gas emissions were reduced as a result of this modernization, but the amount of fly ash per ton of waste was almost halved (1990: 3.0%, 1996 1.6%) as well. Since the early eighties, MSWI fly ash has been used as a filler in asphalt in the Netherlands. On average, 20 kton of fly ash is used this way every year. Due to the limited capacity of the market for asphalt - and thus for asphalt fillers - in the Netherlands, the re-use figure for fly ash cannot increase substantially unless other uses are developed. MSWI fly ash replaces part of the lime present in asphalt-filler. About 30% of the filler mixture consists of fly ash. The water repellent properties of bitumen ensure low leachability of contaminants. The resulting asphalt meets the environmental demands formulated in the 'Building Materials Decree'. This is a result of the fact that bitumen encapsulates the fly ash particles and only 2% of fly ash is present in the asphalt.
847
100
1988
1989
1990
1991
1992
1993
I ~ Produktion ~ Utilisation
1994
/Disposal
1995
1996
1
Figure 6 Production and use of MSWI fly ash in The Netherlands 6.2 Treatment of MSWI fly ash and subsequent use or disposal As mentioned above, landfilling with MSWI fly ash will be prohibited as from January 1998 unless the leachability is reduced. In order to develop alternatives for landfilling, an extensive research program has been launched. The objective of this program is to explore options for reducing the leaching behavior of MSWI fly ash so as to produce a secondary building material or a C3-class waste material. In addition, the possibility of use of fly ash as an additive in concrete has been investigated.
Solidification Solidification involves the fixation of heavy metals, usually by employing cement and additives, in order to reduce their leachability. This quality improvement would result in a shift in landfilling category: C3-status instead of C2. Under more favorable conditions, solidified waste materials could be used as building materials. In recent years numerous solidification techniques and formulae have been tested. Until now, however, none of these have produced the required class-C3 landfill quality. Application as a building material is also impossible. The problems are caused by insufficient fixation of the soluble salts, such as chlorides and bromides. It should be noted that the failure of solidification techniques in treating MSWI does not apply to other types of waste materials which may contain less soluble salts. Soluble salt can be removed easily by applying washing techniques. This option, however, introduces contaminated waste-water. Only a limited number of Dutch MSWIs are permitted to discharge waste water at all. Thus the combination of solidification and washing is a vimaally unfeasible option for most of the MSWIs in The Netherlands.
Melting of MSWIfly ash Melting is regarded as a suitable technique for transforming MSWI fly ash into building materials. And yet the high operating temperatures of these melting processes - at least 1300~ - require a great deal of energy and expense. It is therefore generally accepted that melting techniques are only costeffective when the products can be used in a practical way. The quality of the melting products is considered, therefore, in conjunction with the standards for building materials rather with than those for landfilling. Since 1993, the Waste Processing Association has commissioned several laboratory research projects and feasibility studies concerning melting of MSWI fly ash. However, it was ultimately concluded that larger scale experiments were essential before any definite conclusions could be made. On behalf of one of the members of the Waste Processing Association two pilot-scale
848 melting experiments have been performed using different melting techniques. The resulting granular products have been compared with the Dutch 'Building Materials Decree'. It appeared that the leaching of Antimony did not meet the standards for unrestricted useful application within the framework of the legislation mentioned. The members of the Waste Processing Association therefore believe that the over-all feasibility (quality of the product, energy consumption, resulting residues) do not justify continued development of this technique. Use as an additive in concrete
In 1995 and 1996 the 'Service Center MSWI residues' commissioned two studies concerning the feasibility of use of fly ash as a puzzolanic additive in concrete. MSWI fly ash can be used in concrete in three ways: 9 as filler to obtain high density concrete; 9 as partial cement replacement; 9 as a high value additive after further size reduction (micronized ash, wet grinding). The application of MSWI fly ash in concrete is similar to its use as a filler in asphalt. The technical specifications for concrete, however, are more stringent. Based on the technical, environmental en economical properties, this option appeared technically feasible providing that the fly ash is scrubbed prior to pretreatment (milling). Again, the cause for this inevitable scrubbing step is the presence of soluble salts in MSWI fly ash. Without scrubbing before use of the ash, the leaching of the resulting concrete exceeds the limits for bromide as formulated in the 'Building Materials Decree'. Filler containing MSWI fly ash can not be used in re-enforced concrete due to the corrosive action of the chloride it contains. Removal of chlorides is required when use in reinforced concrete is considered. As stated above, scrubbing the fly ash introduces the problem of waste water. Taking into account that the balance of revenues and total costs are not much better than the current practice of disposing of untreated MSWI fly ash in landfills, the over-all feasibility of use in concrete is probably marginal. Use in German salt or coal mines
In 1996 useful application of MSWI fly ash in German coal and/or salt mines emerged as an attractive option. Specifically, the application in concrete for construction of walls in coal mines is an economically and environmentally favorable option that suits the Dutch policy with respect to MSWI fly ash (useful application). Early in 1997 a request for an export permit for this option was granted by the Dutch government. However, the government regards the use of MSWI fly ash in salt mines as a material for filling obsolete mine galleries as disposal. Consequently, no permits for the export MSWI fly ash for use in salt mines in this manner will be granted.
7. POLICY FOR MSWI FLUE GAS SCRUBBING RESIDUES Again we refer to the 'Policy Paper MSWI residues', in which the following policy for MSWI flue gas scrubbing residues is formulated:
1. At present no options for practical use. 2. Research of the options for re-use of the salt fraction in APC residues. 3. Explore the options for quality improvement so that disposal in landfills as a C3-type waste is made possible. At the present time, no regulatory demand for quality improvement has been formulated. The high costs for disposal of C2 wastes, however, represents a financial incentive for improving the quality to C3 waste.
849 8. PRACTICE OF MSWI FLUE GAS SCRUBBING RESIDUES 8.1 Production and subsequent landfill In 1989 the Dutch incineration emissions guideline 'Richtlijn Verbranden '89' was formulated. Subsequently, this guideline was changed into the 'Decree Air Emissions Waste Incineration'. This Decree came directly into force for new MSWI facilities on February 21, 1993. Existing plants were given until January 1, 1995 to comply with the emission limits stated in this directive. In order to meet these regulations, the flue gas scrubbing system in a typical Dutch MSWI facility currently contains: 9 one or more dust removal systems (usually ESP, sometimes fabric filters), 9 a 2-stage wet scrubbing system, 9 an activated cokes or activated carbon system 9 either a catalytic or a non-catalytic DeNOx-process. As a result, heavy metals are washed out of the flue gases using water (and the chemicals dissolved therein) and end up in a filter cake: ca. 3 kg per ton of waste. Acid components in the flue gases such as HC1 and SO2 are neutralized and subsequently discharged into surface water. Alternatively, this waste water is spray dried, producing approximately 15 kg of dry salt. In other words: the implementation of flue gas scrubbing is directly reflected in the production of solid residues. This is depicted in Figure 7.
35 30 .................t [ ~ Filtercake 25
o
I
~
Spray dry salt
]].......................................................................................... ~
...........
.....................................................................................
.................................................................................................................. . . . . . . ...........
1988
1989
1990
1991
.....
1992
1993
.....
1994
..... N N i
1995
..........
1996
Figure 7 Production of flue gas scrubbing residues 8.2 Treatment of flue gas scrubbing residues To date, there is no practical use for flue gas scrubbing residues. This is in part a result of the fact that this relatively new material is unknown. Moreover, the residues contain high levels of leachable contaminants and lack a matrix that has the physical strength to be exploited. Because these residues are the precipitated reaction products resulting from water treatment (water discharged by the wet flue gas scrubbers), their leaching behavior is only moderate compared to that of MSWI fly ash. Flue gas scrubbing residues are therefore categorized partly as C2 and partly as C3 hazardous waste.With respect to treatment of MSWI flue gas residues, the following options are considered:
850
Solidification The current initiatives for treatment of MSWI flue gas residues aim towards reducing the leaching behavior of the fiker cake using solidification techniques with subsequent disposal at C3 landfill sites. A description of this technique has been given for fly ash. Currently, only the disposal of treated residues is considered.
Purifying and discharging the salts Those MSWIs that are not allowed to discharge their waste water make use of spray-drying techniques. Spray-drying results in a product that consists primarily of salts. Upon spray-drying, the waste water is injected into hot (raw) flue gas. The raw flue gases contaminate the resulting dry salt with heavy metals. A subject of current investigation is the manner in which this product could be separated into a clean salt fraction and a contaminated filter cake containing heavy metals. Materials with a high soluble salt content, such as flue gas scrubbing residues, can be desalted by washing. Two options are available: 9 washing of salts and simultaneous or subsequent removal of specific contaminants to allow discharge of a clean salt solution into the sea; 9 selective washing of salts and beneficial practical application of the salt (CaCl2). The feasibility of both of these options is not yet clear. In any case, it is crucial that an acceptable solution for the discharge of salt water be found. It also has become apparent that there is a global abundance of CaC12. Any production of purified GaG12 out of flue gas scrubbing residues will probably result in additional discharge of this salt into the sea somewhere else.
9 CONCLUDING REMARKS In the Netherlands, the policy for MSWI residues has been agreed upon by all parties involved. This policy has been formalized in a 'Policy Paper MSWI residues'. The current and future practice with respect to use, disposal in landfills and treatment of MSWI residues is monitored in order to verify that it develops in accordance with the policy as determined. The monitoring is performed by a specific project organization (ACR), which also facilitates frequent meetings of the parties involved. During these meetings, the parties responsible for acting on proposals laid down in the Policy Paper report their progress to the other parties. In summary, not only has a policy been put forward but also an organization has been founded to ensure that this policy is put in practice, or modified where necessary based on a possible new and common understanding of all the parties involved.
10 REFERENCES
1.
2.
3.
Implementatieplan AVI-reststoffen / Policy Paper MSWI residues (1995), Publikatiereeks Afvalstoffen, hr. 1995/22, Report edited by the Ministry of the Environment, Zoetermeer, The Netherlands (in Dutch). Jan G.P. Bom, Antonius C.G. van Beurden, Emile A. Colnot, Ruud H. Keegel (1997), High standard upgrading and utilization of MSWI bottom ash, financial aspects. Paper presented at the Fifth annual North American Waste-to-Energy Conference, April 22-25, 1997, Research Triangle Park, NC, USA. Development of new technologies for MSWI residues (1997), brochure edited by the Netherlands Agency for Energy and the Environment, Utrecht, The Netherlands.
Goumans/Senderffvan der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
851
The Materials and Energy Potential Method for the Quantitative Distinction Between Waste Valorization and Elimination in the Cement Industry J.A. Zeevalkink TNO Institute of Environmental Sciences, Energy Research and Process Innovation, PO Box 342, NL-7300 AH Apeldoom, the Netherlands, e-mail
[email protected] Abstract A quantitative method is proposed to distinguish between the valorization and elimination of waste in a cement kiln. Examples are presented to illustrate the consequences of the developed approach. These examples are related to the process conditions in the kiln in the dry- and the wet-cement process. The Materials and Energy Potential (MEP) method which is presented in this report is based on the recognition that a specific waste can contribute to the cement-making process as an alternative raw material and, at the same time, as a source of energy. The paper is based on a report prepared for Febelcem, the Federation of the Belgian Cement Industry.
Introduction A quantitative method is proposed to distinguish between the valorization and elimination of waste in a cement kiln. Examples are presented to illustrate the consequences of the developed approach. These examples are related to the process conditions in the kiln in the dry- and the wet-cement process. Valorization is defined as the processing of a waste in a cement kiln to substitute raw materials and/or fuels. In this case, the waste contributes, in a positive way, to the cement production process. Waste combustion in a cement kiln without any substitution or process improvement and with the sole purpose of final waste processing is defined as elimination. The differentiation between elimination and valorization is of importance as regulations distinguish between waste elimination and valorization. For instance, directives of the European Union allow the export of waste for the purpose of valorization.
Proposals to distinguish between valorization and elimination A review of earlier proposed methods to define valorization shows that most approaches are based on the comparison of the waste with a fuel and that a clear appreciation of both the energy and the raw material value of a waste does not yet exist. Examples of conditions on calorific value or raw materials content are: 9 In Germany, according to the "Kreislaufgesetz", the energy content has to be larger than 11 MJ/kg and the fuel efficiency must be at least 75 %. Conditions on raw materials content have not been published. The Ministry of the Environment (VROM) in the Netherlands sets a calorific value limit of 15 MJ/kg and states that only liquids can be processed (valorized) properly in a cement kiln (i.e. no sludges
852
and no solids). In a former paper, a limit had been proposed of 18 MJ/kg or a useful ash content exceeding 50%. 9 In France, based on EC Directive 94/67, energy recovery for the cement industry is recognized from 5 MJ/kg. 9 In a proposition to B U W A L and in an OVAM paper, it is proposition that processing of a waste can only be regarded as valorization if the calorific value exceeds 25 MJ/kg and the contaminants in the waste do not exceed the given concentration limits o___rthe calorific value exceeds 15 MJ/kg and the concentration of the contaminants in the waste does not exceed the limits, and the total concenti'ation of Ca, Si, A1 and Fe is larger than 10 %.
General conditions for waste processing In order to have an acceptable treatment of waste in a cement kiln, some general conditions have to be met: 9
permit conditions and emission standards must be met; the quality of the cement must fulfil limits with respect to its structural capabilities and its environmental compatibility;
9
the production process must not be impaired and the safety of the workplace must be ensured; an environmental assessment should show that the cement process must be the best way of handling the waste materials. In this assessment, the cement option should be compared with alternatives such as reuse, recycling, incineration in specialized waste combustion facilities or other facilities;
9
the waste materials should not be mixed in order to reach the maximum allowable limits of contaminants in the waste.
These requirements result in criteria which limit the quantity of secondary materials used or can exclude specific wastes entirely. Several criteria have been formulated in the literature and are related to gaseous emissions, cement quality, health standards, and reactor maintenance. These criteria are necessary conditions for the application of waste in general, but do not determine the difference between elimination of waste or valorization. When these conditions are not met, the waste considered cannot be treated in a cement kiln: processing is not acceptable. Valorization or elimination The method which is presented in this paper is based on the recognition that a specific waste can contribute to the cement-making process as an alternative raw material and, at the same time, as a source of energy. This is a specific advantage of waste processing in the cement process which is expressed in the assessment method: the Materials and Energy Potential (MEP) method. Essential steps in the development of the proposed method are: division of the waste in a raw materials fraction and the rest or energy fraction which is separately evaluated as a source of energy;
853 quantitative measures for the raw materials content and the value of the energy fraction are developed; based on these measures, an assessment of waste processing as valorization or elimination is proposed. Below, the decision scheme is shown to decide upon valorization or elimination of a waste in a cement kiln following the MEP method. Another essential aspect of the proposed method is the interpretation of the term "'source of energy". In this study, a "source of energy" is distinguished from a "fuel" with calorific values of 15 MJ/kg up to 40 MJ/kg (wood, coal, oil). The starting point chosen is that any energy contribution (to the cement process) is sufficient for the classification "energy source".
Definition
o f
r a w
materials fraction
First, the raw materials part is established. This fraction contains the components that are useful to (functional in) the cement process: CaO (CaCO3), SiO2, A1203, Fe203 and SO3. The other inorganic components (including water) in the waste are allocated to the raw materials fraction up to maximum values, mwaf and maif, by which the fraction functional components is allowed to contain an equivalent amount of water and non-functional components as occur in natural raw materials. If Ca occurs as CaCO3, the CaCO3 quantity is allocated to the raw materials fraction. The following expression is used to calculate the measure M for the raw materials value of the waste: M = usmf/(
1 - minw ) ( 1 - mini )
wherein: usmf fraction of useful materials in waste as such waf water fraction in waste as such inf -- fraction of inert, non-functional components in the waste mwaf maximum water fraction allowed in raw materials fraction maif maximum inert fraction allowed in raw materials fraction mini - minimum value of inf and maif minw - minimum value of war and mwaf. -
-
-
-
Example 1: For the dry-cement process, the raw materials fraction can contain up to 15 % water. In this report, a maximum o f 10% is used as an example f o r the non-functional part o f the raw materials. So, f o r the dry-cement process m w a f = 0.15 and maif = 0.10. Example 2: For the wet-cement process, up to 30 % water atwl the same percentage, 10, o f nonfunctional (inert and trace) elements are allocated to the raw materials fraction, comparable to the natural raw materials. Again, as an example a maximum o f 10% non-functional components is used f o r the non-functional part o f the raw materials fraction. So, for the wet-cement process m w a f = 0.30 and maif = 0.10.
854
A s s e s s m e n t o f w a s t e as a s o u r c e o f e n e r g y
Secondly, the energy value of the rest or energy fraction ( - waste minus raw materials fraction) is expressed in a measure E. It is proposed to consider the combustion of a material as energy valorization if the autothermal combustion temperature, calculated for the actual conditions in the cement kiln, exceeds a required minimum process temperature, For the measure for the energetic value of a material in a process, E, the following expression is introduced: E = ( Tcomb - T o )
/
(Tref - T o )
wherein: Tr~f -
an essential reference temperature level in the process to be reached (~
Wcomb --.
the autothermal combustion temperature of the considered material under the prevailing process conditions (~
To =
an initial temperature level in the process to be considered as the starting temperature for the heating process (~
E expresses relatively the extent to which the required temperature level, Tr~', is reached or exceeded by the combustion of the energy fraction. This being the case, the material is able to contribute to the energy needs of the process. To expresses a basic temperature to be used as the initial temperature for calculating Tcomb.For example, To could be the combustion air temperature at the inlet of the kiln. As a consequence of the above, a material with the composition of the energy fraction is valorized as a source of energy if:
Ell Example: For the dry- as well as for the wet-cement process Trey is set at 1500 ~ exceeding the minimum required temperature f o r clinker formation of 1450 ~ . The process conditions to calculate the combustion temperature are: an oxygen concentration o f 3 %, an inlet temperature o f the air of 800 ~ (= To) and an energy efficiency o f 75 %. Thus, the E measure is calculated as: E = (Tcomb- 800)/(1500- 800)
855
Decision scheme for waste valorization in a cement kiln.
R e m a r k s :
no
Calculate the raw materials fraction of the waste M waste M
Acceptable with regard to: - health risks - emissions - technical product quality - environmental product quality
Raw materials fraction M: sum of CaO, CaCO 3, SiO2, ./~203, Fe203 and S O 3 - corrected for moisture content and non-funcUonal components in natural raw materials -
Calculate composition of rest fraction
Calculate concentrations and heating value based on 100 % rest fraction
Calculate max. temperature attainable when combusting rest fraction: Tcomb ~
Process conditions in cement kiln: - process temperature min. 1500 ~ ( = Tref) - 75 % energy efficiency - 3 % oxygen content air inlet 800 ~ (To) - M = raw materials fraction
Calculate M = raw materials measure E = energy measure
- E = (Tcomb - To)/(Tref- To) - T O = initial temperature (e.g. air inlet) - Tref = reference temperature, required in process
856
Generalized assessment of a waste as a source of r a w materials and energy For the general assesment of processing a waste with a raw materials and an energy part, the Materials and Energy Potential of the waste, defined as the sum of M and E, is proposed as a measure. It follows from the starting points referred to above that processing a waste with E> 1 or M - 1 in the cement kiln is a case of valorization. It is proposed generally to consider processing of a waste in a cement kiln as valorization when
E+M>I This relation is the basis for the Materials and Energy Potential method presented in this study. E is calculated from the energy fraction, M from the raw materials fraction. Examples are presented to show the consequences of this method that enables a quantitative distinction between valorization and elimination. For wastes with an M value of nearly 1, the formulated condition may be too strict. The result of the appreciation of the raw materials aspect is that TNO's MEP method favours processing of wastes with a raw materials component in the cement kiln. The allocation of (part of the) water in the waste to the raw material fraction is favours the processing of wet wastes in the wet-cement process. Generally, however, from the results of the calculations for actually applied as well as for artificially composed wastes, it is concluded, that in many cases the conclusion is the same for the wet process as for the dry process. In the following table, some calculations are presented as example. $
--~ waste
characteristics LHV * water ash
(MJ/kg) (%) (%)
Organic
Filtration
Artificial
Filter
LD
solvent
earth
waste
cake
stag
25 20
-
12.5 20 50
3.4 50 20
6 50 20
0 5 95
1873 0 1.53 1.53 YES
1912 0.59 1.75 2.45 YES
1151 0.24 0.50 0.74 NO
1400 0.24 0.86 1.09 YES
1.0 1.0 YES
1873 0 1.53 1.53 YES
2023 0.70 1.75 2.45 YES
1212 0.29 0.59 0.88 NO
1476 0.29 0.96 1.25 YES
1.0 0 1.00 YES
Dry-cement process Tcomb (excl. raw materials fraction) (~ a (-)
E (-) E+M (-) Valorization Wet-cement process Tcomb (excl. raw materials fraction) (~ i (-) E (-) E+M (-) Valorization
*
Lower Heating Value of waste as such
857
Conclusions
9 The main types of criteria for waste treatment in a cement process discussed in literature are conditions for emission standards, limits on concentrations of contaminants in the waste and limits with respect to cement quality. These aspects do not distinguish between valorization and elimination; cement processes in which wastes are used have to respect these limits whether valorization or elimination is at stake. 9 Generally, it can be concluded that in Germany, Belgium and The Netherlands, the issue of valorization and elimination has not been worked out on process technological considerations only, which explains the widely different ranges of criteria. Proposed conditions are mainly based on limits to heating values. Raw material aspects are hardly discussed. 9 The MEP method is based on the recognition that a specific waste can contribute to the cementmaking process at the same time as an alternative raw material and as a source of energy. This is a specific advantage of waste processing in the cement process. 9 The MEP method favours processing of wastes with a raw materials component in the cement kiln. 9 Non-functional compounds (Mg, P, Na-, K components and trace elements) are allowed in the raw materials fraction up to a preliminary maximum of 10 %. A better justified value should result from a study of quantities occurring in natural raw materials.
This Page Intentionally Left Blank
Goumans/Senden/van der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
USING ENVIRONMENTAL
859 ECONOMICS
IN DECISION MAKING AND POLICY FORMULATION FOR SUSTAINABLE CONSTRUCTION
WASTE MANAGEMENT
A.L. Craighill and J.C. Powell Centre for Social and E c o n o m i c R e s e a r c h on the G l o b a l E n v i r o n m e n t ( C S E R G E ) , University o f East Anglia, N o r w i c h and University C o l l e g e L o n d o n , U K Abstract
The UK Government is aiming to increase the amount of construction and demolition waste that is reused and recycled as part of its commitment to a sustainable waste management strategy. Reusing and recycling construction waste reduces the need for raw materials and energy, with corresponding reductions in environmental emissions, aesthetic impacts and damage to natural ecosystems. However, the use of secondary materials also gives rise to environmental and social impacts, particularly in the transport and reprocessing stages. Lifecycle assessment can be used to compare alternative options for the sustainable management of construction waste.
Within the assessment, economic valuations of
environmental and social impacts provide weightings to enable this comparison.
In the past, waste management decisions have been based primarily on financial cost, and there has been no mechanism for taking environmental and social costs and benefits into consideration.
Industry and local authorities are increasingly having to take account of a
broader range of criteria, and the discipline of environmental economics provides a means by which these 'external' impacts can be quantified and included in decision making alongside financial costs.
860 Introduction
An estimated seventy million tonnes per year (17%) of the UK's waste arises from the construction and demolition industries.
Although a large proportion (63%) of the waste
created is already 'recycled' most is used for low grade purposes such as access roads within landfill sites, and only 4% is used to replace primary aggregates in more demanding construction uses. As part of a commitment to a sustainable waste management strategy the UK Government has recently introduced a target to increase the reuse and recycling of aggregates in England from 30 to 55 million tonnes per year by 20061.
A number of recent reports examine the potential for recycling construction industry wastel' 2, 3 These confirm that the disposal route taken is usually the one with the lowest financial cost.
There is much potential for increased recycling of construction waste, but there is
currently no economic or technical incentive to do so. If it could be demonstrated that the environmental and social benefits outweigh the costs of recycling, and if these benefits could be included in the decision making process, then this would provide a greater incentive to recycle. This effect has been illustrated by the UK landfill tax, which has already gone some way towards providing such an incentive. Industry and local authorities are increasingly turning towards waste reduction and recycling strategies in order to reduce their disposal costs.
Using secondary materials for construction in place of primary materials displaces the environmental and social impacts which would have arisen from primary material extraction and processing. The reduced need for raw materials and energy for extraction and processing correspondingly reduces environmental emissions. aesthetic impact and damage to natural ecosystems.
Reduced extraction also lessens the
861 Although local authorities and the construction industry are being encouraged to recycle, this is not without its own environmental and social impacts, which arise particularly from transporting the materials and in reprocessing them.
Transport can be over significant
distances, and gives rise to environmental emissions, road congestion and casualties from road traffic accidents.
A technique that can be used to compare the overall level of impacts created by reusing or recycling, with those from landfilling construction waste, is lifecycle assessment (LCA). LCA examines environmental impacts over the entire lifecycle; from obtaining the raw materials, manufacture, distribution, use, re-use/recycling, to final disposal.
The overall
environmental impact can be determined and alternative options can be compared. We are developing a lifecycle assessment model which takes into consideration the impacts discussed above.
The application of economic damage costs to impacts is explored, which allow
environmental and social costs and benefits to be included in the decision making process alongside financial costs.
Both public and private bodies are being encouraged to take
account of a broader range of criteria, and the economic valuation of impacts can thus aid policy formulation for sustainable waste management.
Lifecyele assessment methodology
Lifecycle assessment (LCA) has been used successfully for examining the environmental impacts of products and materials 4. A limited amount of work has been carried out in applying LCA to waste management 5'6. However, there has been very little work done in the application of LCA to construction and demolition waste.
862 The main stages of LCA are goal and scope definition, inventory analysis, impact assessment and the interpretation of results. The goal and scope definition sets the study's boundaries and aims.
Setting the appropriate boundaries is not straightforward because each process
within the lifecycle is connected to several other processes, which has some impact on the main system. Sensitivity analysis can be used to determine which of tiaese have a significant effect on the main lifecycle and should be included in the LCA.
The inventory analysis is a detailed compilation of all environmental inputs and outputs at each stage of the life cycle. These are presented in terms of quantities of materials and energy required, and outputs of gaseous emissions, liquid effluent and solid waste. In our study we are collecting this data from case studies. Qualitative information, such as visual impact, or raw materials scarcity is difficult to include at this stage.
The inventory data are further analysed in the impact assessment stage.
The impact
categories are chosen (e.g. greenhouse gases) and the data is aggregated.
The relative
contribution of each input or output to each environmental impact is then quantified using carbon dioxide equivalents, for example. This results in a 'balance sheet' of impacts for each lifecycle, which provides the basis for comparison.
Unless the outcome is obvious and one lifecycle is better than the other for every impact, it is necessary to apply a system of valuation to the results. In the valuation stage of the impact assessment relative weights are assigned environmental impacts. This enables a comparison of impacts that have been quantified in different units.
Valuation remains the most
contentious aspect of LCA because the weighting factors involve a subjective element and trade-offs are required between different environmental problems.
863
Valuation methodologies
There are various alternative methodologies available for weighting the impacts in the impact assessment stage. The four main approaches can be classified as distance-to-goal techniques, environmental control costs, economic damage approaches or scon.'ng approaches.
More
details on this can be found in Powell et al. 7.
In our study, we make use of economic damage costs. Economic values are available for a number of different impacts including gaseous emissions, road accident casualties and road congestion (Table 1). They are based on factors such as the number of working days lost through illness, or the cost of repair to acid-rain damaged buildings.
They also usually
include an assessment of cost based on contingent valuation. This is a technique used by environmental economists involving a questionnaire survey which asks people how much they would be willing to pay to avoid a particular impact occurring.
The valuation
methodology is then to multiply the economic value by the emission, or the nurnber of expected casualties for example, arising from the entire lifecycle.
Alternative options can
then be compared.
Conclusions
Government pledges to increase the sustainability of waste management means that it is necessary for industry and local authorities to base decision making on a broader range of criteria, including environmental and social costs in addition to financial ones.
For
construction waste, the impacts of alternative management options can be compared within the framework of a lifecycle assessment. Nevertheless, a valuation methodology is required to provide relative weights for the impacts, to be able to make a decision.
864 Economic valuation is a useful weighting technique, enabling the comparison of different types of impact, such as carbon dioxide emissions versus methane emissions, or even carbon dioxide versus road traffic accidents. Unfortunately, economic valuations do not yet exist for all impacts. Although the valuation of environmental impacts has been used for some time by environmental economists, this technique is new in the field of LCA.
However, this
economic valuation may appeal to industry because it places environmental costs and benefits on the same scale as financial ones, thus making it easy to include them in the processes of decision making and policy formulation.
Acknowledgements CSERGE is a designated research centre of the Economic and Social Research Council (ESRC). This research is being funded by the Engineering and Physical Sciences Research Council (EPSRC).
References
1. Department of the Environment. Guidelines of Aggregates Provision in England, Min Planning Guidance (MPG6). HMSO, London (1994). 2. Bnmner, P.H. and Stampfli, D.M. Material balance of a construction waste sorting p
Waste Management and Research 11 (1): 27-48 (1993). 3. Shaw, J.M. Recycling in theory and practice: the case of highways construction. Mira
Planning: 14-17 (1995). 4. Habersatter, K. and Widmer, F. Ecobalance of Packaging Materials, State of 1 BUWAL Report, Federal Office of Environment, Forests and Landscape, B (1991). 5. White, P.R., Franke, M. and Hindle, P. Integrated Solid Waste Management: A Lifec
Inventory. Blackie Academic and Professional, Glasgow (1995). 6. Powell, J.C., Craighill, A., Parfitt, J.P. and Turner, R.K. A Lifecycle Assessment Economic Valuation of Recycling. Journal of Environmental Planning
Management 39 (1): 97-112 (1996). 7. Powell, J.C., Pearce, D.W. and Brisson, I. Valuation for Life Cycle Assessment of W Management Options. CSERGE Working Paper WM 95-07, Centre for Social Economic Research on the Global Environment, University of East Anglia University College London (1995). 8. Fankhauser, S. Evaluating the Social Costs of Greenhouse Gas Emissions CSEF Working Paper GEC 94-01, Centre for Social and Economic Research on the Glq Environment University College London and University of East Anglia (1994). 9. European Commission. ExternE: Externalities of Energy. European Commission DG Luxembourg (1995).
866 10. Department of Transport. Highways Economics Note No.l, 1993 Valuation of Road Accidents. Department of Transport, London (1994). 11. Newbery, D.M. Pricing and Congestion: Economic Principles Relevant to Pricing Roads.
Oxford Review of Economic Policy 6 (2): 22-38 (1990).
867 Table 1. Economic Damage Costs
Emission
(s
Road
(s
Casualties CO2
0.40
Mortality
744,060
CO
0.60
Serious injury
84,260
CH4
7.20
Minor injury.
6,540
802
258.40
Road
(pence/PCUkm b /HGVUkm c)
Congestion 127.00
Motorway
0.26
0.52
N20
61.40
Non central
12.30
24.60
PMIO a
898.00
Rural
1.50
2.99
NOx
Notesl"a Particulates less than 10 lam diameter; b Passenger car unit kilometre; c Heavy goods vehicle unit kilometre. References: 8,9,10,11
This Page Intentionally Left Blank
Goumans/Senderffvan der Sloot, Editors Waste Materials in Construction: Putting Theory into Practice 9 1997 Elsevier Science B.V. All rights reserved.
869
Application of waste materials a success now, a success in the future ir. J. Th. van der Zwan Road and Hydraulic Engineering Division Directorate-General for Public Works and Water Management Ministry of Transport, Public Works and Watermanagement Delft, The Netherlands tel: +31152699391 fax: +31152611361 E-mail:
[email protected] Summary The recycling or reuse of secondary materials is a nowadays practice in the Netherlands. At this time more than 10% of all granular materials used in he building industry is replaced by secondary materials. Especially in infrastructurale works large quantities are being applied. The drive for he use of secondary materials is sustainable development. Thanks to the government's policy and entrepreneurship successes have been scored. At this time nearly all streams of granular waste streams or industrial byproducts (f.i. building and demolition waste, milled asphalt, municipal incineration bottom ash, coal fly ash, steel slag phosphorus slag, blast furnace slag) are being reused completely. Over the years there is a perceptible change in the questions related to the use of those materials. Where in the beginning the questions were mainly technical, today they are dealing with market forces, economics etcetera. From a governmental point of view, it is necessary to spot in time bottle necks that can be prohibitive for the successful application of those materials. The role of the government is very subtle. On the one hand in the Netherlands there is a strong believe in a free market economy, on the other hand the interests of the government are in having environmental acceptable applications and in striving for high grade use. This means that the government has to set coals and to create conditions for a free market that will achieve those coals. In the paper the way the Directorate-General of Public Works and Watermanagement deals with this subject will be explained. A special attention is paid to the succesfactors that have been decisive for the successful introduction of the secondary materials. A study into these succesfactors has been performed in order tot enable the government to increase its efficiency in the implementation of the policy. Also the question of sustainable use of materials will be dealt with. Not always reuse of a material is a synonym for or in line with sustainable development. Given the fact that the first use of the secondary material of course fulfils technical and environmental criteria, it is necessary to take into account the reuse and re-reuse of secondary materials. In this kind of life cycle approach not only technical or environmental conditions have to be set, bus also labor conditions, actual control of the material and other questions influence the acceptability of the first application. If these questions are not taken into account than it is possible that a solution now creates a larger problem in the future.
870 Application of waste materials: a success now, a success in the future.
1.
Introduction
Recycling of materials in construction has been a normal matter in the Netherlands for many years. In [1], the situation in the Netherlands a number of years ago has already been examined. In recent years, a further development has taken place whereby more insight into market tendencies has arisen. The market for secondary materials has been professionalized whereby, in a number of cases, the differentiation between primary and secondary materials is beginning to fade. However, watchfulness remains the precept. Changing social, economic, scientific and technical insights can influence the current recycling possibilities. The government, with responsibility for collective standards and values, has a task to intervene and/or make adjustments where necessary. This paper will explain factors which influence the application of secondary materials, and the manner in which the government can work as effectively as possible proceeding from policy responsibility. 2.
Umbrella policy lines
In brief, the policy framework in which the recycling of secondary materials occurs will be stated. The Netherlands is a densely populated and relatively affluent country. This means that there are many people living per surface unit who all have their requirements for space and comfort. In addition, there is a high level of activity in many economic sectors. All these factors are accompanied by a continual need for the use of the limited space available in the Netherlands. Every year, very large quantities of land are used for dumping wastes, on the one hand, and for extracting surface minerals such as gravel, sand, clay, and limestone, on the other hand. The realisation that this way cannot be continued any longer is deeply anchored in Dutch society. The closing of material cycles, resulting in the need to dump less, and less exhaustion of non-renewable raw materials, is a consequence of that. The application of secondary materials derives from the national administration's policy as articulated in, among others, the National Environmental Policy Plan (NMP). [2]. The NMP states: "This NMP contains the strategy for the environmental policy for the medium long term. The strategy was developed against the background of the wish to solve or to control environmental problems within the life of a generation." For environmental control, striving for a sustainable development is the starting point. A sustainable development is a development that supplies the needs of the current generation without endangering the possibilities for future generations to also supply their needs. "Sustainable development takes shape through feedback at the sources aimed at a combination of: closing of cycles in the chain of raw material - production process - product - waste, and the accompanying emissions; saving energy, together with increasing efficiency, and the deployment of durable energy sources; promotion of the quality (above quantity) of products, production processes, raw materials, waste and environment, with a view to longer usage in the economic cycle." The quotations above from the NMP outline a policy-directed framework within which the recycling of materials takes place. The NMP-plus [3] published later does not give any change of policy, but rather a speeding up of it. A further realisation of the general policy is taking place in about three (for this subject) relevant policy lines.
871 2.1 The waste materials policy [4] The waste materials policy, for a significant part, is based on the so-called Lansink ladder. (the Dutch Lower Chamber motion from 1979). This ladder indicates a priority ranking for the waste materials problem. Lansink * * * *
Ladder: prevention recycling burning dumping
This ranking shows that prevention, or avoiding waste in accordance with policy scores highest, followed directly by recycling (in this context, by recycling is also meant the reuse of materials). In the stream of priorities, there are a number of materials which, in terms of nature and quality, are suitable for application in civil engineering. 2.2 Soil protection policy Protection of the soil is another policy line which influences the deployment of secondary materials. In a general sense, it can be said that secondary materials, due to their composition, can have other environmental effects than conventional materials. In the framework of the Soil Protection Law, a General Administrative Order (AMvB) was published which states preconditions to the application of materials on or in the soil. The Soil and Ground Water Building Materials Degree (BSB) [5] is intended to give the environmental-hygienic preconditions proceeding from soil and ground water protection to the use of secondary and primary materials on or in the arable soil or in ground water or on or in the soil under surface water. The BSB limits its operational sphere to granular (stony) materials applied outside. In applying construction materials, the concept of marginal burdening of the soil is used. Marginal burdening of the soil entails: a very minor increase in the proportions of contaminated materials in the compact layer of the soil, and protection of the ground water at the level of ground water target values. Marginal burdening of the soil is numerically filled in as: a burdening of the soil as a result of the extraction of surface materials, which mathematically leads to an increase in the compact layer of the soil of at least 1% of the proportions of contaminated materials in comparison with the soil target values in 100 years, averaged over a meter of standard soil considered to be homogeneous. The BSB assumes an emissions model which can be determined based on the emission from the surface material in a specific application. In order to continue existing recycling of materials, the maximum permitted emission level for a few materials has been increased. The emission applies to inorganic components. NEN-standards have also been developed for this in order to be able to determine the emission. A good extraction test for organic components is still lacking; because of that, a composition requirement was assumed for organic components. The BSB will come into effect in phases, the first for soil. Mid-1998, the BSB should be fully in effect. 2.3 Surfaceminerals policy [6] In order to supply the need for raw materials for construction, a policy in this domain has been developed by the government. Long-last development implies, among other things, integral control of the chain of raw materials. This entails closing the chain of raw materials in construction as much as possible, preventing degradation of the quality of raw materials, and limitating the production of waste. For the supply of raw materials, this particularly means the frugal use of raw materials and the responsible recycling of waste materials as secondary raw materials. This leads to, among other things, less excavation and dumping of waste. The main objective for the supply of raw materials in construction is: "The policy of the national government with respect to the supply of raw materials for construction has as goal to supply the need of private individuals, businesses and government for construction raw materials (in a socially
872 responsible manner) by: encouraging the use of raw materials as sparingly as possible; \ stimulating the deployment of secondary raw materials in a responsible manner as often as possible; supporting more deployment of replaceable raw materials; and ensuring the timely excavation of an adequate portion of surface minerals from Dutch Soil in the total supply of construction raw materials." The policy lines above indicate that it can be a matter of a synergy aimed at the application of secondary materials.
3.
Generalpreconditions and assumptions
In order to get the application of secondary materials in civil engineering off the ground, it is necessary to satisfy a large number of preconditions and assumptions. The most elementary principle is that there should be a market; in short, a supplying party and a demanding party who can come to terms based on a transaction that is attractive for both of the parties. In the Netherlands, the principle of the free market economy rules, which is also the prevailing opinion for the application of secondary materials. This means that the government is of the opinion that a very important role has been reserved for private industry. However, only in a few instances is private industry the producer of the waste material to be recycled. In nearly all cases, the government itself is a producer of waste materials, resulting from its function or its policy. (E-fly ash as a result of the energy policy, municipal waste incineration bottom ash (AVI-bottom ash) as a result of the policy of burning waste materials, construction and demolition waste from infrastructural works, dredging spoils from maintenance of waterways, etc.). This means that if the government wants to use private industry as a resource for solving its own problems, it must be attractive for business to operate in this market. In the following section, a number of essential preconditions will be explained.
3.1
Unambiguous policy
3.2
Private Industry
3.3
Engineering-technical parameters and rules
For the application of secondary materials, it is necessary that there be an unambiguous government policy that has taken shape in unambiguous rules that are fixed for a long time. The application of secondary materials is surrounded by risks and uncertainties. The risks concern both material-technical risks and environmental-hygienic risks. As far as the environmental-hygienic risks are concerned, it is noted that in the past, the lack of an unambiguous frame of reference of what is permitted under which preconditions, and the various ways in which the competent authorities (particularly provinces) deal with this, have frightened off potential customers. The various ministries responsible for policy (and responsible management within ministries) have not always worked together in an exemplary way in years past. At the same time, an application now may not lead in the future to a situation in which the current user is punished for his use. To realise government objectives, an active business community is necessary. It is the business community that through investments and implementation, ultimately sees to it that the policy objectives are realised. The business community is always ready to invest if the prospects are sufficiently attractive. That means that the government must pursue an investment-friendly policy, and should be reliable as legislator and lawgiver. There are examples where the government has not appeared to be too reliable a partner. That the business community has become cautious after that should be clear. To be reliable also means that the government should be ready to take risks itself by, for example, stimulating, as a customer, the application of secondary materials in its works. In a country where nearly every application is set down in standards, it is practically impossible to achieve a general application if there is not a setting down of the application of secondary materials in
873 a standard or a similar document in private law. The inclusion of secondary materials in standards and rules is an important precondition. This means that in the general sense, the effects of the application of a secondary material on the engineering-technical parameters of the construction segment must be known. In sum, the price/performance ratio of a secondary material or of a product manufactured from it should be known. In general, it is a time-consuming affair. Working for years with a limited choice of primary materials has led to, on the one hand, empiricism being based completely upon it and, on the other hand, standards and the like from that empiricism being valid only for those materials. In addition, the civil sector seems to be conservative. It takes a long time before a new material has proven itself in the market.
3.4
Economicsof the application
It has already been stated above that the price/performance ratio of an application should be known. One may assume from this that in the Dutch market economy, a material will only get a chance via market-conforming mechanisms. This means that the price/performance ratio of a secondary material must at least be comparable with a traditional material (and preferably, more suitable still).
In a general sense, it concerns an "artificial" market. Recycling or useful application is often not cheaper. This is partly due to the fact that not all cost factors which determine the price of primary materials are charged. That is a question of time-dependence is shown by the fact that bitumen, once a waste product, is now a raw material that has a market value of approximately f 350 per ton. If many applications are looked at, then it appears that these only get off the ground if favourable measure are introduced from the government's side which influence the economics of the product. The following will make that clear. The application of granulates in road construction has become a successs because dumping fees have increased sharply. The granulates have certain engineering-technical qualities which make application as a base course material possible. However, in order to have a chance in the marketplace, those materials should at least be cheaper than traditional materials. The application of a base course leads to a reduction in asphalt thickness. This means that the granulate base course must at least be cheaper than the price for the quantity of asphalt saved. In other words, the asphalt price determines the maximum price of the granulates. This means that a demolition waste recycling plant (because of the fixed and variable costs) must ask an acceptance-price in order to be able to sell the finished product competitively. Only with a dumping fee higher than the acceptance-price of the plant will the flow of demolition waste be altered toward recycling. Also, making changes in a package of products occurs only if this is economically attractive. The demolition granulate materials for road construction are in principle also suitable for application in concrete. This, however, requires an extra treatment step (sifting and washing), whereby extra costs are incurred, but an extra waste product (silt) also remains. This silt will have to be dumped at a very high cost due to the pollution level. Despite the fact that the granulate can yield a higher price (comparable with gravel), the net result is negative due to the extra treatment and the costs of processing silt. This means that without the creation of preconditions by the government in a market-conforming manner (e.g., a special dumping charge for silt), no altering of the stream of waste can be obtained.
3.5
Chainconcept approach
An important concept connected to unambiguous policy is the chain concept approach. In a general sense, this means that application of materials now may not stand in the way of future recycling. It is noted that the policy on that subject has not yet fully taken shape in an operational sense. The application of the chain concept means that a designer or a materials expert should ask himself if application of a secondary material now stands in the way of future recycling. Many factors play a role in this. A number of these will be explained. In the first place, it certainly concerns a civil engineering evaluation. If an application now leads to the
874 conclusion that the material soon cannot be recycled, the cart is being put before the horse. Indeed, in the application, the secondary material is often mixed with new materials, whereby the waste problem then only grows larger over time. In addition to a civil engineering reason, the same can also apply for environmental-hygienic and labour-hygienic aspects. Another aspect concerns enforceability. Due to preconditions, it could be necessary to allow a specific secondary material only in a specific application. One can think of the application of materials with lightly increased radioactivity only in concrete in an outside environment. (as might be the case with phosphorus slag). Future recycling of this concrete, then, must take place in such a way that this material does not come directly into concrete that is applied in an interior environment. If this cannot be guaranteed, it should seriously be considered not trying to pursue the first application. A careful weighing of risks and application possibilities is necessary for this. Here, an important responsibility lies with the government to make these choices and to create the necessary preconditions.
3.6 Quality For a customer, it is important to obtain certainty about the quality of the materials to be applied. More than with primary materials, secondary materials are surrounded by a negative image. (secondary - second-hand and thus, (in the mind of some clients) inferior with respect to primary materials). Supplying materials with a good quality guarantee (certificate) is of importance in order to get the application further off the ground. It concerns, then, a certificate which covers both the materialtechnical and the environmental-hygienic parameters. 3.7 Market To sell materials, a market is needed. On a macro scale, there appears to be no problem. Every year in the Netherlands, there is a need for approximately 140 ml tons of granulated raw materials, while approximately 30 ml tons of secondary materials can be brought to the market. At this moment, some 15 ml tons of secondary materials is being sold. It appears that in some market segments, there is saturation of the market, whereby growth in sales remains behind. In the Netherlands, it seems that the market for granular (unbound) base courses is saturated, whereby substitution effects could arise by further growth in production [7]. It should be realisecl that the market for secondary materials is very inelastic. Indeed, the production is largely determined by the size of the flow of waste and no._.~tprimarily by the demand for the product. 3.8 Market acceptance Satisfying the conditions mentioned above is not yet sufficient for market acceptance. There appear to be multiple forces at work. The customer has an important role in this. In time it must be so that the differentiation between secondary materials and primary materials fades. Think of the example of bitumen given, but one can also think of blast-furnace cement or Portland fly ash cement. Before it comes to that, it is necessary for customers to take their responsibility and open their works for the application of secondary materials. At this moment, it is the large authorities in particular that are fulfilling an exemplary function (Rijkswaterstaat (Directorate-General for Public Works and Watermanagement), the municipality of Rotterdam). Putting forward experiences about this is a means by which to achieve further spreading. Of importance is that managements of organisations themselves form a clear policy for deploying secondary materials. At this moment, the decision to apply secondary materials or not is often taken at a low level without clear guidelines from management at the foundation. 4.
Current situation
Figure 1 shows which materials in what quantities are produced and recycled. It is good to state here that recycling has various degrees, from low-quality application such as embanking material to highquality application such as, for example, replacement of scarce primary materials (application in
8?5 concrete or asphalt recycling). Different countries use different definitions of reuse and have different objectives. This does not mean that overall recycling figures from different countries can be compared with each other just like that. The degree of high quality and application in various market segments is a better measure for this.
demolition waste sand polluted ground steel slag l Recycling IE~ blast-furnace slag3 Production waste materiel phosphorous gypsum mmm phosphorous slag EC fly ash concrete and brickwork dredge spoil 1) ~ ~ ~ ~ m dredge spoil 2) n u n n n un nn dredge spoil 3) m m m m m m mum mmnnnnn n n waste incineration bottom ash asphalt rubble others , ,
m////ons o f
tons p e r y e a r 0
1
2
3
4
5
6
7
8
9
10 11
12 13 14 15
1) heavily contaminated 2) moderately contaminated 3) lightly contaminated
Figure 1: production and recycled amounts o f waste materials
In this article, this aspect will not be explained in further detail. At this moment, suffice it to say that in the Netherlands, and particularly from the standpoint of surface minerals policy, there is an effort to apply secondary materials at as high a quality as possible. The table shows that it is, of course, clear that recycling is successful in the Netherlands. However, circumstances can change, whereby it could be possible that a current application would come under pressure. Although the objective in the Netherlands is to reach a market-conforming sale of secondary materials and, thus, there is a very important role reserved for private market parties to anticipate developments, it may have become clear from the previous paragraphs that certainly for secondary materials, the role of the government is great. Therefore, the government deemed it necessary to find out what the success factors were in the past which saw to it that the current market situation arose, in order to learn lessons for policy to be developed in the future. In what follows, a number of important learning points from that study will be explained. 5.
Definition of success
In order to recognise success factors, it is necessary to define the concept success in detail. The definition of success strongly depends on the point of view of the observer. Success for one is failure for another. From the standpoint of the government, the following definitions could be applied.
876 5.1 Degreeof market acceptance The national government feels the need to obtain more insight into the market acceptance of secondary raw materials. ~ In the last decade, partly due to an active policy of various authorities, a large number of bottlenecks were overcome in order to be able to apply secondary raw materials in construction. The market acceptance of the various secondary raw materials differs, though. Some, such as granulated blastfurnace slag, are completely established; for others, conversely, market acceptance and appreciation is (more) uncertain. (figure 2)
no no
acceptation ap
dredge spoil class II and III
cleaned ground
rket cep
waste incineration bottom ash
iation
concrete and brickwork
blast furnace slag
D E G R E E OF S U C C E S S
Figure 2: representation of success
Crudely put, there are three groups of materials to differentiate: Accepted and preferred: Many secondary materials for which the (domestic) demand is greater A. than the supply are characterised by the fact that these materials are (often) preferred above primary materials in their specific application. Acceptance is determined by the unique quality/product characteristics, whereby price is less important.
Accepted: For many secondary materials, supply and demand are not always "in balance"
(usually oversupply). Then, application often depends on the relative price ratios between the various suitable materials. The price instrument is, thus, the success factor here in terms of sales (large price sensitivity). The quality differences in the supply (differentiating capability) are minor. Demand is often stimulated directly or indirectly by government. The material does have sufficient specific characteristics that, due the influence of price, lead to application. C.
Not yet accepted: For the secondary materials belonging to this group, sales and application are especially a result of pressure; for example, mandatory application or the policy-driven, explicit choice for application by specific customers (e.g., Rijkswaterstaat), without leaving it to market forces. Sensitivity to policy, then, is also great. Despite pressure, it must be concluded that in a number of cases, application is still not a viable proposition (material is dumped).
877 It is particularly groups B and C for which it is important to have insight into the factors which determine and/or can lead to such market acceptance that in the long run, it will also be a matter of establishment (appreciation). Not all secondary raw materials, however, have the potential to become fully established (appreciation). The degree of market acceptance/appreciation, however, does indicate to what extent a market-conforming sale is achievable and has a relationship with the extent to which the government should be involved in the sale. This will be discussed further. 5.2 Percentageof utilisation The percentage of utilisation is a simple instrument for ascertaining whether recycling of secondary materials has been realised. It gives quantitative information which can also make a development over time more visible. Figure 1 is an example. However, without testing on other objectives (e.g., prevention, market conformity, closing of the cycle chain approach), the instrument is too limited to be able to determine the degree of success. 5.3 Realisation of wishes and objectives Here, the perception of the actor plays a very strong role. Indeed, objectives and wishes can vary greatly from actor to actor. Thus, policy authorities will often have abstract objectives (degree of building cycle closing, long-lasting development, reduced volume of dumping, etc.), and the business community, operational objectives (continuity, market volume, revenue, profit, etc.). 6.
Sales of secondary materials and success factors
Because success here is a combination of these three definitions and is partly dependent on the actor, giving a formula for success is an impracticable affair. Due to the dynamics of time as stated earlier, circumstances change and other success factors are determining. The study shows that it is also possible to make this visible over time. To be emphasised is that what follows is indicative of the Dutch market and the Dutch situation. It certainly does not have to be true that in other countries with other societal, economic and social structures, the same end-evaluation needs to be reached. 6.1 Evaluation in time at the macro level Over time, there are four periods to be differentiated at the macro level (figure 3). Each period has its own impediments, but also its own success factors. In the following diagram, the most important impediments and success factors are summarised. The last period, 1995 to 2005, is characterised by long-lasting development. In addition to sharpening standards for asbestos, dust and radiation, an expected impediment is a greater reluctance to invest in new recycling techniques; for some materials, a greater dependence on a few processors and purchasers (environmental rules too complicated) and the falling away of national boundaries.
878
Period in the recycling industry
Most important impediments
Most important success factors
1. Trade in valuable secondary raw materials to 1970
9lack of good reference projects 9unknown reprocessing technology + application 9dumping and discharging cheap
9utilising unique product characteristics 9saving transport costs 9little thought for environmental risks
2. Trade + (threatening) scarcity of primary raw materials 1970-1985
9insufficient processing capacity/infrastructure 9suppliers split 9specifications, rules and regulations, product requirements not adjusted 9sharpened environmental policy (various compartments sometimes inconsistent) 9cost/return ratio
9technology development, incl. improved construction 9adjustment of product rules/specification standards 9threatened shortage of primary raw materials 9building up of (logistical) infrastructure
3. Trade + control of waste flow 1985-1995
9poor management of quality/ quality systems (to about 1990) 9authorities reticent as customer (to about 1990) 9unclear/environmental policy/ enforcement/provincial borders 9no continuity of supply and/ or balance of supply and demand in the short term
9high dumping charges 9quality improvement + certification 9suppliers bundle together 9consistent market policy 9authorities (RWS) serve as example
4. Sustainable development (closing of cycle + functional utilisation or fitting in) 1995-2005
9sharpening standards, Occupational Health & Safety, radiation, asbestos 9(competent) authorities more decentralised 9harmonisation supply + demand middle long term 9falling away of national borders
9ban on dumping + charges 9clarity as to what is allowed (environmentalhygienic/long-lasting context) 9(special)large-scale projects + multi-year planning projects 9long-last designs e.g. material decision lists, new technologies
Figure 3: Overview of bottlenecks and success factors, 1970-2005
6.2
Evaluation at the meso level In practice, in addition to influence from social groups and attitudes of the end users/consumers, there are three groups of actors to be differentiated:
879 government as lawgiver suppliers of secondary raw materials users of secondary raw materials. These actors are initiators of success factors and/or take measures aimed at sales promotion/improvement of secondary raw materials. Government measures again influence the activity of suppliers or users and thus, indirectly, sales (see paragraph 3). It appears that just as for "ordinary products", secondary raw materials also have a Product Life Cycle (see figure 4). Some secondary raw materials that are not saleable now or difficult to sell were previously sold without problems; e.g., dredging spoils and AVI-bottom ashes.
Different
phases
of
development
amount
#
no
turi y'
intere
GlYl~m,Jr'n
c k~=Ig o c l m m ~ l l t l o n opoll vwm_.oto elnd
cllo~l q~'c~ I
paper and synthetic mixtures
was'L'o inclnoratlon bottom aoh
oonor~'L'e ~ brlol<worl