UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Polybrominated diphenyl ethers in biosolids-amended soils Gorgy, Tamer George Alexan 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_gorgy_tamer.pdf [ 8.4MB ]
Metadata
JSON: 24-1.0063029.json
JSON-LD: 24-1.0063029-ld.json
RDF/XML (Pretty): 24-1.0063029-rdf.xml
RDF/JSON: 24-1.0063029-rdf.json
Turtle: 24-1.0063029-turtle.txt
N-Triples: 24-1.0063029-rdf-ntriples.txt
Original Record: 24-1.0063029-source.json
Full Text
24-1.0063029-fulltext.txt
Citation
24-1.0063029.ris

Full Text

Polybrominated Diphenyl Ethers in Biosolids-Amended Soils by Tamer George Alexan Gorgy A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (CIVIL ENGINEERING)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  May 2011  © Tamer George Alexan Gorgy, 2011  Abstract Polybrominated diphenyl ethers (PBDEs) are added to many consumer products as flame retardants. Their hydrophobic characteristics and high n-octanol-water coefficients make them partition in organic media such us sludge and biosolids, by-products of wastewater treatment which are commonly applied to agricultural soils to promote crop growth or discarded in landfills. Biosolids-amended soils have been found to contain up to 7×106 pg PBDEs/g dry weight, whereas leachates from biosolids and flame-retarded products in landfills, contained up to 4,000 pg PBDEs/litre. PBDEs in the environment could potentially cause serious health effects. Research was conducted to determine the concentration and mobility of PBDEs in biosolids, biosolids-amended soil, and clay liners used to retain PBDEs. A field study investigated the degree of PBDE contamination due to the application of biosolids at an agricultural site near Kamloops and an agricultural field in Totem Field at the University of British Columbia in Vancouver. PBDEs were found to migrate downwards to depths of at least 0.85 m. Laboratory experiments determined leachability of PBDEs from biosolids. PBDEs sorbed on fine particles suspended in the leachate, allowing PBDEs to exceed their aqueous phase solubilities. Concentrations were much higher on ultra-fine than on fine particles. Leaching column tests demonstrated that PBDEs leached from biosolids-amended soils and migrated through the soils. PBDEs in soils upgradient and downgradient of solid waste facilities in Northern Canada varied widely from location to location. There was evidence that PBDE contamination in Iqaluit is due to long-range atmospheric transport, whereas that found at Yellowknife is mainly from the solid waste facility. Laboratory experiments showed that sand-bentonite partially retained PBDEs. The ii  hydraulic conductivity decreased with leaching, and then gradually increased. The decrease is attributed to swelling, whereas the increase is due to shrinkage of the clay interlayer, owing to the hydrophobicity of the permeant. The research may be helpful in establishing regulations on land application of biosolids, regulating waste disposal and landfill design requirements.  iii  Preface This thesis is based on a series of field and laboratory investigations. Chapter 3 is based on a field sampling program designed and executed by the author, at the Kamloops Range Research Station of Agriculture and Agri-Food Canada. Analyses were carried out at the Institute of Ocean Sciences, Department of Fisheries and Oceans Canada in British Columbia by the author, assisted by personnel of the Institute. Chapter 5 is based on an experimental field investigation conducted at UBC’s Totem Field, Faculty of Land and Food Systems. The author was responsible for designing and implementing the experimental work which included fencing the area of study, applying the biosolids, monitoring, sampling of the soil and analysis. Mr. Sean Trehearne provided landscaping and the necessary tools at Totem Field. The sampling program described in Chapter 7 was designed and executed by the author at Iqaluit Solids Waste disposal Facility and the City of Yellowknife Waste Disposal Facility, under the supervision of the personnel at the two sites. Laboratory experimental work in Chapters 4, 6 and 8 involving design, set-up, sample collection, extraction and cleanup, analyses using Thermo Jarrell Ash Video 22 atomic adsorption spectrophotometer, tabulation of data, statistical analysis of data and preparation of manuscripts were performed by the author with guidance of institute of Ocean Sciences staff members. The authors performed the sample extraction and cleanup for analyses of PBDEs at the Institute of Ocean Sciences. The work was done under the supervision and direction Mr. Norman Crewe and Dr. Michael Ikonomou. Analyses using the VG-Autospec high-resolution mass spectrometer was performed by Ms. Maike Fischer at the Institute of Ocean Sciences. iv  A version of Chapter 4 has been published. Gorgy, T., Li, L. Y., Grace, J. R. and Ikonomou, M. G. (2010). Polybrominated diphenyl ether leachability from biosolids and their partitioning characteristics in the leachate. Water Air and Soil Pollution Journal 209(1-4): 109-121. A version of Chapter 6, PBDE mobility in biosolids-amended soils using leaching column tests, has been accepted for publication in Water Air and Soil Pollution Journal on March 31st, 2011.  v  Table of Contents Abstract ..................................................................................................................................... ii Preface...................................................................................................................................... iv List of Tables............................................................................................................................ xi List of Figures......................................................................................................................... xiii List of Abbreviations and Symbols ........................................................................................ xvii Acknowledgments ...................................................................................................................xxi Dedication ............................................................................................................................. xxii 1  Introduction...........................................................................................................................1 1.1 Problem Statement .........................................................................................................1 1.2 Scope and Objectives......................................................................................................4 1.3 Research Plan .................................................................................................................5 1.3.1 Review of Studies on PBDE Contamination in Biosolids and Agricultural Soils ......6 1.3.2 Preliminary Investigation of the Extent of PBDE Migration in an Agricultural Field........................................................................................................................6 1.3.3 Leachability of PBDEs from Biosolids and their Distribution between the Aqueous and Suspended Solids Phases in the Leachate ..........................................................6 1.3.4 PBDE Mobility in Biosolids-Amended Soil: Field Experiment ................................8 1.3.5 PBDE Mobility in Biosolids-Amended Soil: Laboratory Tests.................................8 1.3.6 PBDEs in Soils Surrounding Northern Canadian Solid Waste Disposal Facilities.....8 1.3.7 PBDE Retention and its Effect on the Hydraulic Conductivity of Sand-Bentonite Liner Material .........................................................................................................9 1.3.8 Conclusions .............................................................................................................9 1.4 Research Novelty and Contributions...............................................................................9  2  Review of PBDEs in Biosolids and Agricultural Soils .........................................................11 2.1 Introduction..................................................................................................................11 2.2 Extraction, Cleanup and Analytical Techniques............................................................15 2.2.1 Extraction ..............................................................................................................15 2.2.2 Cleanup .................................................................................................................16 2.2.3 Analysis.................................................................................................................23 2.3 PBDEs in Biosolids ......................................................................................................25 2.3.1 Europe...................................................................................................................29 2.3.2 North America.......................................................................................................31 2.3.3 China.....................................................................................................................34 vi  2.3.4 Middle East ...........................................................................................................34 2.3.5 Australia ................................................................................................................34 2.4 PBDEs in Biosolids-Amended Soils .............................................................................35 2.4.1 Europe...................................................................................................................35 2.4.2 North America.......................................................................................................36 2.5 Discussion ....................................................................................................................41 2.5.1 Biosolids................................................................................................................41 2.5.2 Soils ......................................................................................................................45 2.6 Prediction of Total PBDEs in Soil ................................................................................49 2.7 Recommendations ........................................................................................................50 2.8 Conclusion ...................................................................................................................51 3 A Preliminary Field Investigation on the Mobility of Polybrominated Diphenyl Ethers in Biosolids-Amended Soil ...........................................................................................................53 3.1 Introduction..................................................................................................................53 3.2 Material and Methods...................................................................................................56 3.2.1 Sampling Site ........................................................................................................56 3.2.2 Soil Sampling ........................................................................................................57 3.2.3 Physical and Chemical Characterization ................................................................58 3.2.4 PBDE Determinations............................................................................................58 3.2.5 Data Analysis ........................................................................................................60 3.3 Results and Discussion .................................................................................................61 3.3.1 Soil Properties .......................................................................................................61 3.3.2 PBDEs Concentration and Vertical Distribution.....................................................62 3.4 Conclusions..................................................................................................................71 4  PBDE Leachability from Biosolids and their Partitioning Characteristics in the Leachate....72 4.1 Introduction..................................................................................................................72 4.2 Material and Methods...................................................................................................75 4.2.1 Biosolids, Leachate and Suspended Solids Characterization...................................75 4.2.2 Leaching Column Experiments ..............................................................................77 4.2.3 Filtration Experiments ...........................................................................................77 4.2.4 Sample Clean-up and Analysis...............................................................................79 4.2.5 Data Analysis ........................................................................................................80 4.3 Results and Discussion .................................................................................................80 4.3.1 Biosolids, Suspended Particles and Leachate Characteristics..................................80 4.3.2 Leachability of PBDEs from Biosolids ..................................................................81 vii  4.3.3 PBDE Association with Suspended Particles and Leachate ....................................83 4.4 Conclusions..................................................................................................................91 5 Mobility of Polybrominated Diphenyl Ethers in Biosolids-amended Soil-Controlled Field Experiment ...............................................................................................................................92 5.1 Introduction..................................................................................................................92 5.2 Material and Methods...................................................................................................94 5.2.1 Soil Sampling ........................................................................................................95 5.2.2 Physical and Chemical Characterization ................................................................96 5.2.3 Sample Clean-up and Analysis...............................................................................97 5.2.4 Data Analysis ........................................................................................................98 5.3 Results and Discussion .................................................................................................98 5.3.1 Climate Conditions ................................................................................................98 5.3.2 Biosolids and Soil Properties .................................................................................99 5.3.3 PBDE Concentration in Reference Soils ..............................................................102 5.3.4 PBDE Concentrations in Biosolids.......................................................................103 5.3.5 PBDEs in Biosolids-Amended Soils ....................................................................104 5.4 Conclusions................................................................................................................113 6  PBDE Mobility in Biosolids-Amended Soils using Leaching Column Tests ......................115 6.1 Introduction................................................................................................................115 6.2 Materials and Methods ...............................................................................................120 6.2.1 Sample Characterization and Preparation for Leaching Column Tests..................120 6.2.2 Leaching Column Experiments ............................................................................121 6.2.3 Sample Clean-up and Analysis.............................................................................123 6.2.4 Data Analysis ......................................................................................................125 6.2.5 Mole balance .......................................................................................................125 6.3 Results and Discussion ...............................................................................................126 6.3.1 Physical Properties of the Biosolids Mixture and Agricultural Soil ......................126 6.3.2 PBDEs in Biosolids-soil and Agricultural Soil Layers..........................................126 6.3.3 Leachate PBDE Concentrations ...........................................................................132 6.3.4 PBDE Mole Balance............................................................................................134 6.4 Conclusions................................................................................................................136  7  PBDEs in Soils Near Northern Canadian Waste Disposal Sites .........................................138 7.1 Introduction................................................................................................................138 7.2 Methods .....................................................................................................................140 7.2.1 Site Background ..................................................................................................140 viii  7.2.2 Soil Sampling ......................................................................................................142 7.2.3 Storage and PBDE Analysis.................................................................................143 7.2.4 Soil Characterization............................................................................................144 7.2.5 Data Analysis ......................................................................................................144 7.3 Results and Discussion ...............................................................................................145 7.3.1 Soil Properties .....................................................................................................145 7.3.2 Vertical and Lateral Variations of PBDE Concentration.......................................145 7.3.3 Distribution Trends of Specific Congeners...........................................................151 7.3.4 Comparison to Soils from Different Regions........................................................157 7.4 Conclusion .................................................................................................................160 8  PBDE Retention in Sand-Bentonite Liner Material............................................................162 8.1 Introduction................................................................................................................162 8.2 Materials and Methods ...............................................................................................166 8.2.1 Leachate Preparation............................................................................................166 8.2.2 Barrier Material ...................................................................................................167 8.2.3 Free Swell Test ....................................................................................................169 8.2.4 Leaching Column Test.........................................................................................169 8.2.5 Sample Clean-up and Analysis.............................................................................173 8.2.6 Data Analysis ......................................................................................................174 8.3 Results and Discussion ...............................................................................................175 8.3.1 Physical and Chemical Properties ........................................................................175 8.3.2 Swelling Test Results...........................................................................................177 8.3.3 PBDEs Concentrations in Leachates ....................................................................177 8.3.4 PBDE Concentrations in Sand-Bentonite .............................................................178 8.3.5 PBDE Mass Balance............................................................................................182 8.3.6 Hydraulic Conductivity and Adsorption Capacity of Sand-Bentonite ...................183 8.4 Conclusions................................................................................................................186  9  Conclusions and Recommendations ..................................................................................187 9.1 Conclusions................................................................................................................187 9.2 Recommendations ......................................................................................................191  References ..............................................................................................................................193 Appendices .............................................................................................................................207 Appendix A: Additional Details and Data Related to the Preliminary Field Investigation (Chapter 3)..............................................................................................................................207 ix  Appendix B: Additional Details and Data Related to the Leachability from Biosolids and their Partitioning Characteristics in the Leachate Experiment (Chapter 4) .......................................219 Appendix C: Additional Data Related to the Controlled Field Experiment (Chapter 5)............228 Appendix D: Additional Details and Data Related to the Agricultural Leaching Column Test (Chapter 6)..............................................................................................................................238 Appendix E: Adtional Data Related to the Northern Canadian Waste Disposal Facilities Sites Investigations (Chapter 7) .......................................................................................................260 Appendix F: Additional Information and Data Related to the Sand-Bentonite Leaching Column Test (Chapter 8) ......................................................................................................................278  x  List of Tables Table 1.1. Major PBDE commercial products in use in 2001 (BSEF, 2004). ...............................1 Table 2.1 Technical products of PBDEs and their different compositions (Sellstrom et al., 2005; La Guardia et al., 2006).....................................................................................................12 Table 2.2 Estimated global demand for BDE commercial mixtures in 2001 (Bromine Science and Environment Forum, 2004).........................................................................................12 Table 2.3 Averaged properties of selected PBDEs (Palm et al., 2002). ......................................13 Table 2.4 Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples.......................................................................................................................17 Table 2.5 Advantages and limitations of different detection methods for PBDEs in biosolids and soil samples (Covaci et al., 2003) ......................................................................................24 Table 2.6 Concentration of PBDE congeners (pg/g dw) reported in published studies on biosolids............................................................................................................................26 Table 2.7 PBDEs in biosolids collected from a German wastewater treatment plant (Hagenmaier et al., 1992). Concentrations are in pg/g dw. ......................................................................29 Table 2.8 PBDE concentrations (pg/g dw) in soils amended with sludge and biosolids..............37 Table 2.9 Total concentration of PBDEs (pg/g dw) in biosolids-amended soils at different depths (Xia et al., 2010). ..............................................................................................................40 Table 2.10 Concentrations (%, w/w) of PBDEs in the PeBDE, OcBDE, and DeBDE commercial products (Sellstrom et al., 2005; La Guardia et al., 2006)...................................................45 Table 2.11 Concentration (pg/g dw) of PBDE congeners reported in the published studies and the organic matter content of the tested soils plotted in Figure 2.6 (Matscheko et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008; Andrade et al., 2010).......................................48 Table 3.1 Moisture content, cation exchange capacity and organic matter content of Kamloops field reference soils sampled on August 2005. ...................................................................61 Table 3.2 Moisture content, cation exchange capacity and organic matter content of Kamloops biosolids-amended soils.....................................................................................................62 Table 3.3 PBDE concentrations (pg/g dw) from single samples, on a dry weight basis (dw) basis, in three layers below reference soil surface samples in August 2005 and biosolids application surface from December 2004 sampling............................................................65 Table 4.1 PBDE concentration (pg/g dry weight) in biosolids prior to and after leaching with deionized water, and percent reduction due to leaching. Three samples were analyzed in each case. ..........................................................................................................................82 Table 4.2 Estimated water solubilities (pg/L) and concentrations (pg/L) of selected PBDE congeners in the leachate, Filtrate A and Filtrate B; three samples were analysed in each case. ..................................................................................................................................85 Table 4.3 Concentrations (pg/g dw) of selected PBDE congeners on the retained particles and ultrafines/fines ratios; three samples were analysed in each case........................................85  xi  Table 4.4 Mass balance comparison of PBDE masses (in pg) in the leachate determined based on analyses by three different methods. ..................................................................................87 Table 4.5 Ratios of BDE47 to 209 and BDE 47 to 99. ...............................................................91 Table 5.1 Details of the sampling program. ..............................................................................96 Table 5.2 Temperature and precipitation at Totem Field during August 2006-August 2007.......99 Table 5.3 Moisture content, cation exchange capacity and organic matter content of biosolids, reference and biosolids-amended soils............................................................................100 Table 5.4 Permeability (k) of Totem Field soil. .......................................................................102 Table 5.5 PBDE concentrations in (pg/g dry weight) in biosolids (derived from sewage sludge). ...........................................................................................................................104 Table 5.6 Exponential functions describing tetra, penta, hexa, hepta, octa, decaBDE and total PBDE concentration as a function of time. ......................................................................112 Table 6.1 Molecular weight of PBDE congeners in the same homologue groups. ....................125 Table 6.2 Concentration of eight major PBDE congeners in the biosolids-soil layer during the leaching tests in pg/g dry weight basis (dw).....................................................................129 Table 6.3 Estimated water solubilities (pg/L) and concentrations (pg/L) of selected PBDE congeners in the leachates after 1, 2 and 4 week/s of leaching. ........................................134 Table 6.4 Moles (µmoles) of PBDEs lost or gained in the leaching column tests .....................135 Table 7.1 Sampling locations and elevations (m) above sea level at Iqaluit Waste Disposal Facility and Yellowknife Waste Disposal Facility............................................................143 Table 7.2 Chemical and physical characteristics of Iqaluit (IQ) and Yellowknife (YELL) soil samples. ..........................................................................................................................146 Table 7.3 Mean concentrations (pg/g dw) of five PBDE congeners which are major contributors to the total PBDE concentration in soils from Iqaluit (IQ) and Yellowknife (YELL)........152 Table 7.4 Percent of BDE-47, -99, -100, -154, -153, -183 and -209 to total PBDE concentration including BDE-209. ........................................................................................................153 Table 7.5 Percent of BDE-47, -99, -100, -154, -153, and -183 normalized to the their sum of concentration...................................................................................................................154 Table 7.6 Average BDE47 and BDE99 concentrations (pg/g) in different environmental matrices in the Arctic used to calculate BDE47/99 ratio in Figure 7.8............................................160 Table 8.1 Chemical and physical properties of Ottawa sand and Na-Bentonite. .......................176 Table 8.2 Moisture content and organic content of the sand-bentonite layers during leaching column test. For layer locations, see Figure 8.3. ..............................................................176 Table 8.3 Concentrations (pg/L) of eight principal PBDE congeners in the permeant and leachate after 8, 16 and 21 days of leaching through sand-bentonite columns. ...............................178 Table 8.4 Concentrations (pg/g dw) of eight principal PBDE congeners in sand-bentonite preand post-leaching. ...........................................................................................................179  xii  List of Figures Figure 1.1 Flow chart of research. ...............................................................................................7 Figure 2.1 General structure of PBDEs containing up to 10 bromine atoms...............................11 Figure 2.2 Concentrations of BDE47, 99 and 209 and total concentrations of all reported PBDEs in biosolids analyzed between 1988 and 2008....................................................................42 Figure 2.3 Average concentration of BDE47, 99, 209 and total of all reported PBDEs in North American and European biosolids. ....................................................................................43 Figure 2.4 Percent contribution of BDE47, 85, 99, 100, 153, 154, 183 and 209 to their total concentration in North American and European biosolids..................................................44 Figure 2.5 Percent contribution of BDE47, 85, 99, 100, 153, 154, 183 and 209 to their total concentration in amended soils. (NA: Biosolids loading was not available) .......................46 Figure 2.6 Relationship between the PBDE congeners reported in the published studies and the organic matter content of the tested soils (Matscheko et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008; Andrade et al., 2010). BDE85 was not plotted since the majority of the concentrations were reported as non-detectable. ................................................................47 Figure 2.7 Comparison of measured average total concentration of PBDEs in biosolids-amended soils with simple predictions based on the assumption that PBDEs are mixed (without degradation or loss by transfer) into the top 0.2 m deep layer of soil..................................50 Figure 3.1 Location of Experimental Farm in Kamloops, British Columbia (Source: atlas.gc.ca). .......................................................................................................................56 Figure 3.2 Locations of sampling areas in December 2004 and August 2005. The area was divided into 5×2 m cells onto which different biosolids loadings were applied. .................57 Figure 3.3 PBDE concentrations from single samples, on a dry weight basis (dw) basis, in three layers below reference soil surface samples in August 2005 and biosolids application surface from December 2004 sampling: Layer 1: 0.05-0.15 m; Layer 2: 0.15-0.25 m; Layer 3: 0.25-0.35 m below soil surface. Note that “Others” represent the total concentration of BDE85, 100, 153, 154 and 183. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3). ...........................................................................................64 Figure 3.4 Comparison of PBDE concentrations in the 0.05-0.25 m layer soil samples collected from Kamloops in August 2005 with PBDE concentrations reported in published investigations from 0.0-0.30 m layers of biosolids-amended soils (Matscheko et al., 2002; Eljarrat et al., 2008; Andrade et al., 2010). Entries without concentration bars denote that these congeners were not reported. ....................................................................................66 Figure 3.5 Concentrations of BDE47, 85, 99, 100, 153, 183 and 209 in 0.05-1.05 m Kamloops 80 t/ha biosolids-amended soil sampled in August 2005. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3)............................................................67 Figure 3.6 Exponential functions describing (a) tetraBDEs; (b) pentaBDEs; (c) hexaBDEs and (c) total PBDE and BDE209 distribution with soil depth. Note that the concentration axes have different scales. .........................................................................................................69  xiii  Figure 3.7 Exponential functions describing (a) tetraBDEs; (b) pentaBDEs; (c) hexaBDEs and (c) total PBDE and BDE209 distribution due to the cation exchange capacity (CEC) of the soil. Note that concentration axes are not the same scale....................................................70 Figure 4.1 Schematic diagram of the apparatus used for biosolid leaching and the leachate filtration experiments. .......................................................................................................78 Figure 4.2 Cumulative PBDE concentration (pg/g dw) in biosolids before (CInitial) and after (CFinal) leaching. Note: error bars cannot be seen due to the logarithmic scale used for the vertical axis. ......................................................................................................................83 Figure 4.3 Cumulative PBDE concentration (pg/L) in the pre-filtered leachate (CTotal) and after sequential filtration using 1.75 (Filtrate A) and 0.45 (Filtrate B) µm filters........................84 Figure 4.4 Cumulative PBDE concentration (pg/g dw) on fine (>1.75 µm) and ultrafine (1.75 to 0.45 µm) particles. ............................................................................................................86 Figure 4.5 Mass balance diagram illustrating comparison of PBDE masses (in pg) in the leachate determined based on three different methods. ....................................................................88 Figure 4.6 Ratios of BDE47, 85, 99, 100, 153, 154, 183 and 209 in Filtrate A + fines to that in the leachate; and in Filtrate B + ultrafines to that in Filtrate A. ..........................................88 Figure 4.7 %-Ratio of BDE47, 85, 99, 100, 153, 154, 183 and 209 to ΣPBDE in the biosolids, leachate and particulate samples. .......................................................................................89 Figure 4.8 Concentrations of BDE47, 85, 99, 100, 153, 154, 185 and 209 in Filtrate B plotted against their Log KOW values.............................................................................................90 Figure 5.1 Totem Field test area arrangement where 80 t/ha of biosolids were applied and the time of sampling of each cell. A: Samples collected December 2006; B: Samples collected April 2007; C: Samples collected August 2007. Outer grey shaded cells were not sampled, but provided a buffer. ........................................................................................................95 Figure 5.2 PBDE congener concentrations in Totem Field reference soils at different depths. Where there is no bar, the measurement level was non-detectable. ..................................103 Figure 5.3 PBDE concentration in biosolids-amended soil from Totem Field at depths of (a) 0.00-0.05 m, (b) 0.05-0.25 m and (c) 0.25-0.45 m. Cumulative concentrations for 0.45-0.65 and 0.65-0.85 m depths intervals were negligible. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3). Note that the concentration axes do not have the same scales. ...............................................................................................................106 Figure 5.4 Ratio of PBDE concentration biosolids-amended soil (S) and reference soil (R) in Totem Field biosolids-amended soil between 0.05 m and 0.85 m depth. ..........................107 Figure 5.5 Cumulative PBDE concentration (pg/g dw) in Totem Field biosolids-amended Soils between August 2006 and July 2007 (8/07) in (a) 0.00-0.05 m, (b) 0.05-0.25 m and (c) 0.250.45 m. Note that cumulative concentrations in 0.45-0.65 and 0.65-0.85 m were negligible; error bars cannot be seen due to logarithmic scale used for the vertical axis.....................108 Figure 5.6 Change of (a) tetra, (b)penta, (c) hexa, (d) hepta, (e) octa and (f) decaBDE concentration (pg/g dw) in each soil layer with time (days). See Table 5.6 for exponential functions. ........................................................................................................................109 Figure 5.7 Change of total PBDE concentration (pg/g dw) in each soil layer with time (days)......................................................................................................................................112 xiv  Figure 6.1 Schematic of apparatus used for leaching column tests. For simplicity the diagram shows the connection between one reservoir and three leaching columns. However, the experiment had three reservoirs, each connected to three leaching columns, with each leaching column connected to a 4 L amber glass bottle....................................................122 Figure 6.2 BDE homologue group concentrations (pg/g dw) in (a) biosolids-soil layer (Layer 1; 0-14 mm) and (b) agricultural soil layer (Layer 2; 14-28 mm). Homologues without bars indicate that they were nd. Agricultural Layers 3 (28-42 mm) and 4 (42-56 mm) had a total PBDE concentration between nd to13 pg/g dw. ...............................................................130 Figure 6.3 Percent contribution of BDE homologue groups to ∑PBDE in (a) biosolids-soil layer (Layer 1; 0-14 mm) and (b) layer 2 (14-28 mm) of the agricultural soil. ..........................130 Figure 6.4 BDE homologue group concentrations (pg/L) in leachate after 1, 2 and 4 weeks of leaching deionized water through soil columns, corresponding to volume ratios of leachate to soil of 11, 20 and 34, respectively................................................................................133 Figure 6.5 Percent of PBDEs in biosolids-soil mix layer (Layer 1), agricultural soil layers (Layers 2,3 and 4) and leachates to that in the biosolids-soil mix prior to leaching. Note that the percentages are plotted on a logarithmic scale............................................................136 Figure 7.1 Site plans and sampling locations for (a) Iqaluit waste disposal facility, and (b) Yellowknife solid waste disposal facility. Sampling locations for Iqaluit waste disposal site are designated IQ-1, IQ-2, IQ-3 and IQ-4 and for Yellowknife solid waste disposal facility YELL-1, YELL-2, YELL-3 and YELL-4. .......................................................................141 Figure 7.2 Cumulative PBDE concentrations (pg/g dw) at Iqaluit Waste Disposal Facility. .....147 Figure 7.3 Cumulative PBDE concentrations (pg/g dw) at Yellowknife Waste Disposal Facility. ........................................................................................................................................148 Figure 7.4 Cumulative PBDE concentrations (pg/g dw) in soil upgradient and downgradient of: (a) Iqaluit Waste Disposal Facility, and (b) Yellowknife Waste Disposal facility. Errors bars portray 95% confidence intervals.....................................................................................148 Figure 7.5 Correlation between ΣPBDE and: (a) median particle diameter (D50), (b) total organic matter (%), (c) cation exchange capacity (CEC) and (d) moisture content of soil samples from Iqaluit and Yellowknife waste disposal facilities. ....................................................150 Figure 7.6 BDE47/99 ratio for soil samples from (a) Iqaluit waste disposal facility and (b) Yellowknife waste disposal facility. Error bars show 95% confidence intervals based on 2 to 6 replicates. .................................................................................................................156 Figure 7.7 Comparison of ΣPBDE concentrations (pg/g dw) in the soils downgradient of Iqaluit and Yellowknife waste disposal facilities and soils from the Canadian Arctic. RI: Rankin Inlet; AF: Apex Flats; CB: Cambridge Bay; MR: Mount Relly; DI: Devon Island (DanonSchaffer, 2010)................................................................................................................158 Figure 7.8 Mean BDE47/99 ratios calculated from data published in studies on PBDE levels in different environmental biota and matrices. Average BDE47 and BDE99 concentrations are presented in Table 7.6. (Lindstrom et al., 1999; Herzke et al., 2001; RAIPON/AMAP/GEF 2001; van Bavel et al., 2001; Christensen et al., 2002; Chernyak et al., 2003; Herzke et al., 2003; Law et al., 2003; Malmquist et al., 2003; Sellstrom et al., 2003; AMAP 2004; Fangstrom et al., 2004; Jaward et al., 2004b; Mariussen et al., 2004; Remberger et al., 2004; Savinova et al., 2004; Vives et al., 2004; Vorkamp et al., 2004a; Vorkamp et al., 2004b; Herzke et al., 2005; Muir et al., 2006). ............................................................................159 xv  Figure 8.1 Schematic of glass leaching column used to obtain biosolids leachate and in the leaching column tests. .....................................................................................................170 Figure 8.2 Schematic of leaching column test apparatus. .........................................................172 Figure 8.3 Layer definition of the cell in each glass column. ...................................................173 Figure 8.4 PBDE cumulative concentration (pg/L) as a function of degree of bromination in the permeant and leachates....................................................................................................179 Figure 8.5 Cumulative concentrations (pg/g dw) of PBDEs in (a) layer 1: 0-14 mm, (b) layer 2: 14-28 mm, (c) layer 3: 28-42 mm, and (d) layer 4:42-56 mm of the sand-bentonite columns after 8, 16 and 21 days of leaching. Error bars are 95% confidence intervals (n=3)..........181 Figure 8.6 Cumulative PBDE migration profile in sand-bentonite columns. ............................181 Figure 8.7 Cumulative PBDE concentrations pg/g dw in the sand-bentonite columns (concentrations of all four layers combined). Error bars are 95% confidence intervals (n=3). ........................................................................................................................................182 Figure 8.8 Mass of PBDEs in the leachate post leaching, retained in the sand-bentonite to PBDEs in the leachate pre leaching..............................................................................................183 Figure 8.9 Sand-bentonite hydraulic conductivity and PBDEs in leachate and sand-bentonite during leaching tests........................................................................................................185  xvi  List of Abbreviations and Symbols θ  Soil porosity  ρ  Soil density, kg/m3  v  Absolute viscosity, Pa•s  AASHTO  American Association of State Highway Transportation Official  ASE  Accelerated solvent extraction  ASTM  American Society of Testing and Materials  ATSDR  Agency for Toxic Substances and Disease Registry  AWA  Australian Waste Association  BSEF  Bromine Science and Environmental Forum  CFinal  Final (post-leaching) concentration, pg/g dry weight basis  CInitial  Initial (pre-leaching) concentration, pg/g dry weight basis  CB  Chlorobenzene  CEC  Cation exchange capacity  CFR  Code for Federal Regulations  CO2  Carbon dioxide  CWWA  Canadian Waste Water Association  d10  Grain diameter for which 10% of the sample grains are finer than  d60  Grain diameter for which 60% of the sample grains are finer than  DCM  Dichloromethane  DeBDE  DecaBDE commercial mixture  DNAPL  Dense non-aqueous phase liquid  DOM  Dissolved organic matter  dw  Dry weight basis  ECD  Electron capture detection xvii  EGME  Ethylene glycol monoethyl ether  EI  Electron ionization  Filtrate A  Filtrate passing through 1.75 µm filter  Filtrate B  Filtrate passing through 0.45 µm filter  g  Acceleration due to gravity  GC-HRMS  Gas chromatogram-high resolution mass spectrometry  GCL  Geosynthetic clay liners  glc  Glass leaching cells  HOC  Hydrophobic organic compound  HRGC  High-resolution gas chromatography  HRMS  High-resolution mass spectrometry  IR  Infrared  IPCS  International Programme on Chemical Safety  IUPAC  International Union of Pure and Applied Chemistry  k  Hydraulic conductivity (cm/s)  Kd  Distribution coefficient, (mol•kg-1 )/(mol•L-1)  KOA  Octanol-air partitioning coefficient  KOC  Organic carbon normalized partitioning constant, (mol•kg-1 organic carbon)/ (mol•L-1 water)  KOW  N-octanol-water partitioning coefficient, (mol•L-1 octanol)/ (mol•L-1 water)  LCT  Leaching column tests  LRAT  Long-range atmospheric transport  LRMS  Low resolution mass spectrometry  LRRI  Land Resource Research Institute  MA  Moles gained from the three agricultural soil layers  MB  Moles gained from the biosolids-soil layers  xviii  ML  Moles gained from the leachate  (MTOC)Leachate  TOC in leachate, mg  (MTOC)Filtrate A  TOC in leachate after passing through 1.75 µm filter, mg  (MTOC)Filtrate B  TOC in leachate after passing through 0.45 µm filter, mg  MFines  Mass of particles retained on 1.75 µm filter, g  MUltrafines  Mass of particles retained on 0.45 µm filter, g  meq/100g  Millieqivalents per 100 g  Na2SO4  Sodium sulphate  NCI  Negative chemical ionization  nd  Non-detectable  OcBDE  OctaBDE commercial mixture  OM  Organic matter content, %  PBDEs  Polybrominated diphenyl ethers  PCBs  Polychlorinated biphenyls  PeBDE  PentaBDE commercial mixture  POPs  Persistent organic pollutants  RSD  Relative standard deviation  SW  Estimated water solubility  St Error  Standard Error  t  Tonnes  t/ha  Tonnes per hectare  TOC  Total organic carbon, %  TOCFines  TOC per unit mass of particles retained on 1.75 µm filter, mg/g  TOCUltrafines  TOC per unit mass of particles retained on 0.45 µm filter, mg/g  TS  Total solids, g/L  U  Coefficient of grain uniformity xix  US EPA  United States Environmental Protection Agency  UBC  University of British Columbia  V  Volume passed through the cell, L  v/v  Volume basis  WWTP  Waste water treatment plant  xx  Acknowledgments I express my sincere gratitude and appreciation to Dr. Loretta Li, and Dr. John Grace, my supervisors, for accepting me as their student and for their excellent guidance, patience, and support during the course of my studies. I value their encouragement, wisdom, and advice. This research project would not have been possible without the support and help of Dr. Michael Ikonomou, member of my supervisory committee. I am deeply indebted to him for his guidance, help and support during the course of this research and my entire PhD program. Our conversations and his expert advice have shaped and refined my research. My sincere thanks to Dr. Kenneth Hall and Dr. James Atwater, members of my supervisory committee, for their valuable input. I am also deeply grateful to Ms. Susan Harper and Ms. Paul Parkinson of the Department of Civil Engineering for their time and effort for providing the necessary equipment at the Environmental Laboratory. I extend my sincere appreciation to Mr. Sean Trehearne at the Faculty of Land and Food Systems for his assistance at Totem Field. A special endearing gratitude to my wife Raquel for her unconditional understanding, patience, encouragement, help and emotional support during the course of this study. Special thanks are owed to my parents, who have supported me throughout my years of education, both morally and financially. I also convey my profound gratitude to NSERC (Natural Sciences and Engineering Research Council of Canada) and to Environment Canada for providing financial support.  xxi  Dedication  To
my
wife
and
companion
 Raquel

 and
my
newborn
baby
 Benjamin
George
  xxii  1 Introduction 1.1 Problem Statement Polybrominated biphenyl ethers (PBDEs) are incorporated as flame retardants in plastics, textiles, electronic circuitry and insulation materials. There are 209 PBDE congeners classified by the number and placement of bromine atoms. They are commercially prepared by synthesis, resulting in a product or fraction containing a mixture of PBDE congeners with various degrees of bromination (Rahman et al., 2001; D'Silva et al., 2004). Six (BDE46, 99, 100, 153, 154 and 209) of the 209 PBDE congeners are typically found within the three commercial mixtures: PentaBDE (PeBDE), OctaBDE (OcBDE) and DecaBDE (DeBDE). The estimated world market demand for PBDEs in 2001 is given in Table 1.1. According to the Bromine Science and Environmental Forum (BSEF) (2004), annual worldwide productions of the three commercial products in 2001 were 8,500, 3,825, and 54,100 metric tonnes (t), respectively. Table 1.1. Major PBDE commercial products in use in 2001 (BSEF, 2004). Region North America Europe Asia Total  PeBDE (t) 8,290 210 0 8,500  OcBDE (t) 1,375 450 2,000 3,825  DeBDE (t) 24,300 7,500 23000 54,800  PBDEs are ubiquitous due to their relative stability and have been detected in biotic and abiotic matrices including air, sediments, fish, birds, marine mammals and human plasma. PBDEs are hydrophobic and lypophilic. Their presence in human milk, with levels rising over time, has raised concern about their introduction into the environment (de Wit, 2002; Hites, 2004). Another reason for concern is that toxicological data show serious health effects such as  1  thyrogenic, estrogenic, and dioxin-like activities (Bergman and Urika, 2001). PBDEs also show behaviour and toxicity similar to polychlorinated biphenyls (PCBs) (Gouin and Harner, 2003). Levels of PBDEs in biosolids have increased over the past two decades (Nylund et al., 1992; Rayne and Ikonomou, 2005; Knoth et al., 2007; Eljarrat et al., 2008; Andrade et al., 2010). In Germany, total PBDE levels ranged from 1.1×105 to 2.5×106 pg/g on a dry weight basis (dw) (Knoth et al., 2007). Eljarrat et al. (2008) reported that total PBDE levels from Spanish biosolids were between 2.0×105 and 1.2×106 pg/g dw. In British Columbia, total PBDE concentrations ranged from ∼2.4×106 to 946×106 pg/g dw (Rayne and Ikonomou, 2005; Gorgy et al., 2010). Agricultural soils receiving biosolids have also been found to be contaminated (due to external sources) with PBDEs. In Europe, biosolids-amended soils contained 63 to 2.2×106 pg/g dw (Matscheko et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008). In North America, total PBDEs ranged from 3.1×103 to 1.3×107 pg/g dw (Reick, 2004; Andrade et al., 2010; Xia et al., 2010). Biosolids are treated sewage sludge from the treatment of domestic sewage in wastewater treatment plants (WWTPs) (Canadian Waste Water Association [CWWA], 2010). They are commonly applied as soil amendments in agriculture production fields (Australian Waste Association [AWA], 2008). Land application of biosolids is generally considered a sustainable practice because it recycles the nutrients to assist in growing crops (AWA, 2008). However, concerns exist with respect to the fate and bioavailability of persistent organic pollutants (POPs) present in biosolids. In the United States, biosolids are considered in Section 40 of the Code for Federal Regulations (CFR) Part 503 for the disposal of sludge and biosolids (United States Environmental Protection Agency [USEPA], 2004). Biosolids applied to land must meet strict  2  regulations and standards for metals in biosolids, pathogen reduction standards, site restriction, crop harvesting restrictions and monitoring. In Canada, the regulations and guidelines for use of biosolids are at the provincial/territorial level, rather than at the federal level (CWWA, 2010). Biosolids regulations typically stipulate limits on constituents, including trace elements and pathogens, relevant to the environment and human health. Biosolids are normally classified as high or low quality, based on meeting prescribed trace element and pathogen quality criteria, as well as process requirements for pathogen and vector attraction reduction. Biosolids quality has impacts on the range of end uses for the material, land use following biosolids application, and post-application monitoring. Quality criteria are also specified for the receiving soil before and after biosolids application. Since PBDEs are released into wastewaters from consumer products, and since these chemicals partially degrade and are hydrophobic in nature, they tend to adhere to the organic solids during wastewater treatment. Therefore, PBDE residues are present in biosolids as the material is applied to soils. The potential effects on ecosystem health and on the long-term fate of the PBDEs in soil are generally unknown due to the limited investigations and indefinite findings. No long-term studies have been carried out to examine PBDE concentrations and persistence in biosolids-amended soils where details of the sequence and durations of application are provided. Hence, limits of PBDEs in biosolids are currently not stipulated in regulations addressing biosolids application. It is estimated that ~ 19% (~2.0×106 t/year) and 47% (~3.0×106 t/year) of the biosolids produced in North America and Europe, respectively, enter municipal landfills (AWA, 2008). Osako et al. (2004) found that PBDE concentrations in leachate samples from Japanese landfills ranged from non-detectable (nd) to 4,000 pg/L. Mean total PBDEs in municipal landfill leachates from  3  Canada and the US were 232,000 (Danon-Schaffer, 2010) and 83,000 pg/L (Oliaei, 2005), respectively. This is likely due to the PBDE-containing waste, including biosolids, disposed in landfills. The efficacy of clay, used as a barrier material in landfills, in containing PBDEs in leachates and how its hydraulic conductivity evolves over time have not been investigated. Extensive research has been performed on PBDE levels and their trends in environmental matrices such as sediments, air, human and animals. The portion of studies on PBDE levels in soils, especially in biosolids-amended soils, and clay barrier materials is relative minute compared to studies on other environmental matrices. As a result, there is a lack of knowledge on the behaviour, fate and transport of PBDEs in agricultural soils and the effectiveness of clay barriers in retaining PBDEs, which are a vital component for government regulations and standards on the application of biosolids to agricultural soils and the disposal of solid wastes in landfills. The present study focuses on the extent and mobility of PBDEs in biosolids, soils and a clay barrier material. Investigations were based on field and laboratory experiments on the leachability of PBDEs from biosolids, soils and clay, and the temporal and spatial distribution of PBDEs in soils in biosolids-amended agricultural fields and near solid waste disposal facilities.  1.2 Scope and Objectives The present study was undertaken to investigate: - the leachability of PBDEs from biosolids; - PBDE mobilization in biosolid-amended soils; - PBDE distribution in biosolids-amended soils and soils near waste disposal facilities; and  4  - PBDE retention by, and effect on, the hydraulic conductivity of a clay material, conventionally used for containment of landfills. The following tasks were undertaken in support of these objectives: 1. Confirmation that PBDEs are mobile in biosolids-amended soils. An investigation was conducted on the vertical distribution of PBDEs in soils from a plot that had received biosolids in an agricultural field managed by Agriculture and Agri-Food Canada. 2. Determination of leachability of PBDEs from biosolids and their distribution between the aqueous and suspended solid phases of biosolids leachate in laboratory experiments. 3. Laboratory and field experiments on the leachability of PBDEs from biosolids and their mobility and distribution in biosolids-amended soil as a function of time and depth. 4. Field investigation on the degree of PBDE contamination in soils near solid waste disposal facilities in Northern Canada. 5. Laboratory experiments on the retention of PBDEs by sand-bentonite, a clay barrier material commonly used in containment of landfills, and its change in hydraulic conductivity as a result of leaching PBDE-contaminated leachate.  1.3 Research Plan Figure 1.1 illustrates the field- and laboratory-scale investigations undertaken to support the research objectives. The following sections summarize the investigations undertaken, and also describe the layout of the thesis.  5  1.3.1 Review of Studies on PBDE Contamination in Biosolids and Agricultural Soils Information on PBDE levels in biosolids applied to agricultural soil are discussed in Chapter 2. Studies on PBDE levels in biosolids and biosolids-amended agricultural soils were reviewed to compare the PBDE levels in different countries and/or regions. The review also covered the different methods of sampling and analyses, as well as identified the data gaps, which are key to understanding the fate of PBDE in biosolids-amended soils. 1.3.2 Preliminary Investigation of the Extent of PBDE Migration in an Agricultural Field A preliminary field investigation was carried out at an agricultural field, managed by Agriculture and Agri-Food Canada, to determine the concentration of PBDEs in the 0-0.30 m interval of biosolids-amended agricultural soils to which 20 and 80 tonnes per hectare (t/ha) biosolids had been applied in an agricultural field in Kamloops, British Columbia. PBDE concentrations at 0.05-1.05 m depths from the soils which had received 80 t/ha biosolids were investigated to determine the extent of vertical contamination due to the biosolids application over 8 months from December 2004 to August 2005. Information from this study provided insight for formulating the research program for the subsequent studies. A research paper based on this initial work is presented in Chapter 3. 1.3.3 Leachability of PBDEs from Biosolids and their Distribution between the Aqueous and Suspended Solids Phases in the Leachate A laboratory investigation was carried out to determine PBDE congener profiles in biosolids from an undisclosed1 WWTP in British Columbia and their leachability by water. PBDE fractionation in water and suspended solids of different sizes were also examined to determine the mobility of these contaminants within soils. Leachates from leaching column tests (LCTs)  1  The wastewater plant made confidentiality a condition in order to provide biosolids. 6  (glass column experiments) were passed through a series of filters, and the filtrates and retained particles were analyzed for PBDEs. This study helped to identify the mode of PBDE transport through biosolids and biosolids-amended soils. The paper based on these tests appears in Chapter 4. Objectives as stated in Section 1.2 -  Tasks as described in Sections 1.2 -  Chapter 3: Preliminary site investigation on soils amended with 20 and 80 t/ha biosolids  Confirm PBDE contamination in biosolidsamended soil and vertical mobility  -  Chapter 4: Leachability of PBDEs from  biosolids  -Determine PBDE congener profiles in biosolids. -PBDE fractionation in water and suspended solids  Chapter 5: Field experiment for PBDE mobility in agricultural soil amended with 80 t/ha  Determine PBDE temporal and vertical distribution in 5 depth intervals; 0.00-0.05 m, 0.05-0.25 m, 0.25-0.45 m, 0.45-0.65 m and 0.650.85 m at four month intervals over one year.  Chapter 6: Leaching column tests for PBDE mobility in biosolidsamended soil  Determine PBDE mobility in controlled environment to reduce external factors over 4 weeks.  Chapter 7: PBDEs in soils surrounding Northern Canadian solid waste disposal facilities  Chapter 8: Leaching column tests using sandbentonite  Examine PBDE levels in soils upgradient and down gradient solid waste disposal facilities in Iqaluit, Nunavut and Yellowknife, north Western Territories, and to determine whether PBDE contamination originated from the solid waste disposal facilities or external sources.  Examine PBDE retention by bentonite, and change in hydraulic conductivity of bentonite as a result of adsorption of PBDEs from a permeant consisting of biosolids leachates.  Provide a better understanding on PBDE mobility in biosolids-amended soils. Provide an indication in the evolution of bentonite behavior due to PBDE retention, and the effectiveness of bentonite in containing PBDEs. Aid in establishing policy and more informed regulations on land application of biosolids, and regulating waste disposal and landfill design.  -  Figure 1.1 Flow chart of research. 7  1.3.4 PBDE Mobility in Biosolids-Amended Soil: Field Experiment A controlled field experiment was conducted to determine the temporal and vertical distribution of PBDEs in biosolids-amended soils. The experiment included applying 80 t/ha biosolids on a 5×5 m area divided into 1×1 m cells. The biosolids were spread uniformly over each cell and mixed with the top 0.05 m of soil. Soil samples were then collected from an area which had not received biosolids and from cells where the biosolids were applied. Samples were collected at four-month intervals over one year from 5 depth intervals down to 0.85 m below the soil surface. Chapter 5 is a research paper based on the results. 1.3.5 PBDE Mobility in Biosolids-Amended Soil: Laboratory Tests LCTs were conducted to determine PBDE mobility in biosolids-amended soils in a controlled environment to reduce external factors that enhance volatilization and photodegradation of PBDEs. Deionized water was passed upwards through a glass column containing a 14 mm layer (210 g) of biosolids-amended soils under 42 mm (600 g) of agricultural soil. The latter was divided into three 14 mm layers to determine the PBDE distribution along the flow path of the infiltrating water. The collected leachate, biosolids-amended soil and agricultural soil layers were analyzed for PBDEs after 1, 2 and 4 weeks of leaching. The results of this study are provided in Chapter 6. 1.3.6 PBDEs in Soils Surrounding Northern Canadian Solid Waste Disposal Facilities Field investigations were conducted on PBDE levels in soils upgradient and downgradient of solid waste disposal facilities in Iqaluit and Yellowknife. The investigations addressed the distribution patterns of selected PBDE congeners as a function of distance from the waste disposal facilities and depth from the surface. The results, also detailed in Chapter 7 helped to  8  determine whether PBDE contamination originated from the solid waste disposal facilities or as a result of external sources such as long-range atmospheric transport (LRAT).  1.3.7 PBDE Retention and its Effect on the Hydraulic Conductivity of Sand-Bentonite Liner Material The adsorption and hydraulic conductivity of bentonite mixed with sand, used as a clay barrier for landfill, were investigated to determine whether bentonite can effectively contain and/or immobilize PBDEs in landfill leachate. The LCTs used a permeant consisting of biosolids leachates diluted to 50% by volume and spiked with 50 µg/mL pentaBDE mixture solution. The outcome of the study helped to determine the compatibility of sand-bentonite to PBDEcontaminated leachates especially through examining the time-variation of the hydraulic conductivity of sand-bentonite. Chapter 8 is a research paper giving the details and results of this study. 1.3.8 Conclusions Chapter 9 consists of general conclusions and recommendations for the various studies.  1.4 Research Novelty and Contributions The research presented in this thesis is the first to examine PBDE fractionation in water and suspended solids of different sizes, temporal changes of PBDE concentration in biosolidsamended soils over depths greater than 0.30 m and the ability of bentonite to retain PBDEcontaminated leachate. The results provide vital information on the degree of PBDE contamination in biosolids and their leachability as a result of infiltrating water. The results also provide insight on the transport of PBDEs in biosolids-amended soils and the extent of mobilization by examining PBDE vertical and temporal variations. The data can aid in 9  establishing policy and more informed regulations on land application of biosolids. In addition, the project provides insight on the ability of sand-bentonite to retain PBDEs from landfill leachates, thus assisting in waste management and regulating waste disposal and landfill design.  10  2 Review of PBDEs in Biosolids and Agricultural Soils 2.1 Introduction PBDEs have been incorporated as fire retardants in many commercial products, such as plastics, computers, textiles, and upholstery since about 1980. These brominated compounds act as flame retardants by releasing bromine ions, which scavenge OH and H radicals formed during combustion (Rahman et al., 2001). PBDEs are similar in chemical structure to polychlorinated biphenyls (PCBs) and are identified with the same International Union of Pure and Applied Chemistry (IUPAC) nomenclature and numbering system. There can be up to 10 bromine atoms attached to the two carbon rings. The nomenclature and numbering system are based on the number and position of bromine atoms attached to the molecule as shown in Figure 2.1. The PBDE chemical structure allows 209 congeners, divided into 10 homologue groups according to the number of bromine atoms in the molecule. Commercially, PBDEs are mixtures of various congeners (Rahman et al., 2001; D'Silva et al., 2004). These three commercial mixtures are Penta-BDE (PeBDE), Octa-BDE (OcBDE), and Deca-BDE (DeBDE) (Table 2.1). The PeBDE mainly contains penta and tetraBDEs (∼40 and 60%, respectively) (Sellstrom et al., 2005). OcBDE mixtures are mainly composed of hepta- and octa-BDEs (∼44 and 35%, respectively) and, to a lesser extent, hexaand nona-BDEs (∼10% each) (Sellstrom et al., 2005; MacDonald et al., 2009), whereas DeBDE, consists of almost 100% BDE209 (La Guardia et al., 2006). Br  O  Br  Br  Br Br  Br Br  Br  Br Br  Figure 2.1 General structure of PBDEs containing up to 10 bromine atoms. 11  Table 2.1 Technical products of PBDEs and their different compositions (Sellstrom et al., 2005; La Guardia et al., 2006). Technical Product PeBDE OcBDE DeBDE  tetraBDEs 24-38  pentaBDEs 50-60  hexaBDEs 4-8 10-12  Congener % heptaBDEs  octaBDEs  nonaBDEs  decaBDEs  44  31-35  10-11 <3  <1 97-98  Annual worldwide productions of PeBDE, OcBDE, and DeBDE technical products in 2001 were 8,500, 3,825, and 54,800 t, respectively (Bromine Science and Environment Forum [BSEF], 2004). The estimated world market demand for PBDEs in 2001 is given in Table 2.2. Since that time, PeBDE and OcBDE products have been banned in a number of countries. By the end of 2004, Great Lakes Chemical Corporation, the only American manufacturer of PeBDE and OcBDE products, voluntarily phased out their production (United States Environmental Protection Agency [US EPA], 2007). Ten U.S. states have banned PeBDE- and OcBDE including California, Illinois, Hawaii, Maine, Maryland, Michigan, Montana, New York, Oregon, and Rhode Island. In 2008 Canada banned the manufacture of all PBDEs and the use or sale of three PBDE congeners (tetra-, penta- and hexa-BDE), targeting the PeBDE and OcBDE commercial mixtures, but not DeBDE, which is proposed to be banned by 2013 (MacDonald et al., 2009). In Europe, the European Union banned PeBDE and OcBDE in all products in 2004 (USEPA, 2007). Both PeBDE and OcBDE were added to the Stockholm convention covering persistent organic pollutants in May 2009. Table 2.2 Estimated global demand for BDE commercial mixtures in 2001 (Bromine Science and Environment Forum, 2004). Continent North America Europe Asia Total  PeBDE (tonnes) 8,290 210 0 8,500  OcBDE (tonnes) DeBDE (tonnes) 1,375 24,300 450 7,500 2,000 23,000 3,825 54,800  12  PBDEs are quite resistant to physical, chemical, and biologic degradation making them ubiquitous. The boiling point of PBDEs is between 310 and 425°C and their vapour pressure is low at room temperature (Darnerund et al., 2001). PBDEs have low solubility in water, especially for the higher brominated compounds, and high n-octanol-water partitioning (KOW) coefficients, which have logarithmic values between 4 and 10. Table 2.3 lists the water solubilities and high KOW values of some congeners. These properties allow PBDEs to be persistent, bioaccumulating in fish, animals, and humans. PBDEs also undergo long-range atmospheric transport, depositing in remote environments, such as the Arctic, well-removed from their sources (Gouin and Harner, 2003). This is of concern due to the toxicological data indicating that these compounds may cause serious health effects such as thyrogenic, estrogenic, dioxin-like activities and with chemical properties similar to those of PCBs (Bergman and Urika, 2001). Table 2.3 Averaged properties of selected PBDEs (Palm et al., 2002). Compound BDE11 3,3'-di (C12H8Br2O) BDE30 2,4',6-tri (C12H7Br3O) BDE35 3,3',4-tri (C12H7Br3O) BDE37 3,4,4'-tri (C12H7Br3O) BDE47 2,2',4,4'-tetra (C12H6Br4O) BDE85 2,2',3,4,4'-penta (C12H5Br5O) BDE99 2,2',4,4',5-penta (C12H5Br5O) BDE116 2,3,4,5,6-penta (C12H5Br5O) BDE119 2,3',4,4',6-penta (C12H5Br5O) BDE128 2,2',3,3',4,4’-hexa (C12H4Br6O) BDE181 2,2',3,4,4',5,6-hepta (C12H3Br7O) BDE185 2,2',3,4,5,5',6-hepta (C12H3Br7O) BDE171 2,2',3,3',4,4',6,6'-octa (C12H2Br8O)  M (g/mol) 328 406.9 406.9 406.9 485.8 564.7 564.7 564.7 564.7 643.6 722.5 722.5 801.4  Solubility (mg/L) 0.088 0.026 0.00507 0.00507 0.00922 3.975 × 10-5 8.265 × 10-4 3.975× 10-5 3.975× 10-5 4.15 × 10-6 2.16 × 10-7 2.16 × 10-7 1.11× 10-8  Log Kow 5.43 5.70 5.70 6.12 6.32 6.98 6.97 7.19 7.19 7.97 9.44 9.44 9.98  PBDEs have been found in blood, milk and tissue samples of people, mammals, fish, birds and sediment and in air samples. The presence of increasing levels of PBDEs in humans has caused concern with respect to their introduction into the environment. The total concentration of PBDEs over three decades has risen exponentially in humans, with a doubling time of 5 years (Hites, 2004). The levels in blood, milk and tissue samples in Japan have been found to range 13  between 23 and 3,530 pg/g lipid, while corresponding levels in Europe and North America were higher, 70-13,400 pg/g lipid and 23-193 ×103 pg/g lipid, respectively (Hites, 2004). PBDE concentrations in ringed seal blubber increased exponentially between 1981 and 2000 from 572 to 4622 pg/g lipid (Ikonomou et al., 2002). Hites et al. (2004) showed that some marine mammals are contaminated with PBDEs (~2 ×106 pg/g lipid) doubles in ~ 5 years (Hites, 2004). The total level of PBDEs in guillemot eggs have increased rapidly over a 30 year period from 124 ×103 to 7,510×103 pg/g lipid. Studies on birds between 1984 and 1998 showed widely varying levels, from <10×103 pg/g lipid in chickens to > 1×106 pg/g lipid in cormorants and ospreys (Allchin et al., 1999). Fish from Europe have ~10 times lower PBDE concentrations than those from North America (Christensen et al., 2002), with mean total PBDE concentrations of 120×103 and 1,050×103 pg/g lipid, respectively. The high KOW of PBDEs led to their partitioning into high organic-matter-content media, such as wastewater effluent discharges, sludges and biosolids. Biosolids are dewatered sludge from wastewater treatment plants. Reported PBDE levels in influents, effluents and activated sludge are in the range of 420×103-4,300×103 pg/L, 0.004-900×103 pg/L, and 1,320×103-3,800×103 pg/L, respectively (Rayne and Ikonomou, 2005; Rayne and Ikonomou, 2005; Anderson and MacRae, 2006). It is believed (Gevao et al., 2008) that PBDEs introduced into wastewater originate from different sources including households, atmospheric deposition runoff and disposal of hazardous and industrial wastes. PBDE congeners detected in sewage sludge are closely related to those in commercial mixtures (Clarke et al., 2010). Congeners representative of the PeBDE formulations (BDE47, 99, 100, 153, and 154) are often present in similar proportions, whereas BDE209, the predominant congener of the DeBDE formulation, is consistently the main PBDE congener identified in sewage sludge (Hale et al., 2001a). 14  PBDE contamination of wastewater sludge and effluents may have implications for disposal and reuse alternatives. Sewage sludge and biosolids are often applied on agricultural lands and used for land reclamation, due to being rich in organic nutrients. This reduces waste, minimizing the cost of disposal. However, sludge and biosolids are sinks to hydrophobic organic contaminants like PBDEs, thus potentially contaminating the soil onto which they are applied. This chapter reviews existing data from different countries on PBDE concentration in biosolids and soils onto which these byproducts have been spread. Different extraction, cleanup and analytical methods used in determining PBDE concentrations are compared. Recommendations are made regarding data at greater depths and greater standardization in the analysis and reporting of data.  2.2 Extraction, Cleanup and Analytical Techniques Analytical methods (Table 2.4) for the determination of PBDEs are diverse and widely applied to different matrices such as sediments and biota samples (Eljarrat and Barcelo, 2004). However, few investigations have been reported on analytical methods for biosolids, and biosolids-amended soils. 2.2.1 Extraction Extraction of PBDEs from biosolids and biosolids-amended soils mainly utilizes Soxhlet extraction as the preferred technique.  Typical solvents used in the extraction are toluene,  hexane, and mixtures of toluene:acetone (4:1 volume basis (v/v)), hexane:dichloromethane (DCM) (1:1 v/v) or hexane:acetone (3:1 v/v) (Hagenmaier et al., 1992; de Boer et al., 2003; North, 2004; Reick ,2004; Rayne and Ikonomou, 2005; Anderson and MacRae, 2006; Eljarrat et al., 2008; Gevao et al., 2008). Soxhlet extraction involves heating the solvent to reflux, and then  15  flooding the sample placed in a thimble to extract the compounds of concern. Extraction times vary from 12 to 24 h. An alternative extraction method, pressurized liquid extraction, commonly known as accelerated solvent extraction (ASE), is also used, reducing the amount of solvent and the time of extraction. In this method the solvent is delivered to an extraction cell holding the sample, and the pressure and temperature are then increased to ∼70-100 bar (7,000-100,000 kPa) and ∼100-150oC, respectively (Hale et al., 2001a; Webster et al., 2008; Gorgy et al., 2010b). The extraction can take place in 30-45 minutes, using approximately 30 mL of solvent (Hale et al., 2001a; Webster et al., 2008). Samples have also been extracted with traditional techniques such as centrifugation (Nylund et al., 1992; Sellstrom et al., 1998; Oberg et al., 2002; Sellstrom et al., 2005). For PBDE extraction, acetone and acetone:hexane mixtures have acted as solvents. Comparisons have only been done on the recoveries of PBDEs using ASE and Soxhlet extraction from sediment samples. Studies involving biosolids and biosolids-amended soil samples have not been reported. Recoveries of PBDEs from sediments using ASE ranged from 22 to 82%, whereas those obtained by Soxhlet extraction were in the range of 42 and 81% (de la Cal et al., 2003). 2.2.2 Cleanup Multi-step cleanup methods are required for complex matrices such as biosolids and biosolidsamended soils (Eljarrat and Barcelo, 2004; Gorgy et al., 2010). The high organic content of biosolids necessitates their removal prior to gas chromatography to reduce any disturbance to the purification process. This is achieved by treating the extract with sulphuric acid and adsorbents such as Florisil silica and alumina. Furthermore, these matrices contain sulphur, which interferes with the analysis. The sulphur is removed by copper powder and multi-layer silica, modified with silver nitrate. No studies have been reported comparing different clean-up methods. 16  Table 2.4 Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Extraction  Cleanup  Chromatography  Analysis  Column Length  Temperature Program  Ionization Method  Hagenmaier et al., 1992  Biosolids  50 g dried sample is Soxhlet extracted with toluene for 18 h.  Extract was concentrated then diluted with hexane. The mixture was cleaned with H2 SO4 until no discoloration occurred. Further cleanup included silica gel, alumina and biobeads permeation.  Not specified  Not specified  Nylund et al., 1992 and Sellstrom et al., 1998  Biosolids  15-20 g sample centrifuged in centrifuge tube with acetone.  Extract mixed with NaCl solution (0.2 M) in phosphoric acid (0.1 M). Extraction continues using hexane/acetone (3:1,v/v), and re-extracted with a (90:10:2, v/v/v) mixture of hexane/DCM/decane. The combined organic phases are concentrated and mixed with hexane, 2-propanol and TBA-sulfite reagent. Water is then added. The hexane portion is transferred and mixed with H2SO 4 for clean up.  25 m  180oC for 2 ins, 20oC /min ramp to 220oC, then 5oC /min ramp to 310oC  Matscheko et al., 2002  Soil  16 g sample Soxhlet extracted using toluene.  Extract was cleaned by eluting through a mixed acid-alkaline silica column using n-hexane. Extract was purified on a 9 mm i.d. 0.5 g AX-21 carbon/celite column. Eluate was concentrated with nhexane:DCM (1:1, v/v).  30 m  From 180 oC (held for 2 minutes) to 300 oC in 45 minutes.  Not specified  Ion source temperature (oC) Not specified  Electron Energy eV Not specified  NCI  150  70  EI  250  30  17  Table 2.4 (continued) Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Extraction  Cleanup  Chromatography  Analysis  Column Length  Temperature Program  30 m except for decaBDE and 15 for decaBDE. 50 m .except for decaBDE. and 15 m (for decaBDE.  NS  90oC for 3 min and increased at 30 oC /min to 210 oC then increased by 5 oC/min to 290 o C  ECNI  200  70  All BDEs except BDE209: 90oC for 2 min and then to 320 oC at 4 oC /min held for 10 min. BDE209: 80 oC for 2 min and then to 320 oC at 15oC /min and held for 3 min, then increased at 15oC/min to 350oC. For tri-heptaBDEs: 90oC for 2 min then to 220oC at 50oC/min and to 300oC at 5oC/min and held for 20 min. BDE209: 90oC for 2 min then to 220oC at 30oC/min and to 300oC at 5oC/min and held for 8 min.  ECD  Not specified  Not specified  NCI  150  Not specified  Oberg et al., 2002  Biosolids  Same as Nylund et al., 1992  Same as Nylund et al., 1992  De Boer et al., 2002  Biosolids  Samples mixed with Na2SO4 and Soxhlet extracted using hexane/acetone (3:1,v/v).  Gel permeation chromatography over 300×25 mm, 10um pore columns using DCM. Extract concentrated under nitrogen, mixed with isooctane, and further cleaned with H2 SO4. After separation of iso-octane phase. H2SO4 is separated and washed with pentane to extract all PBDEs. Isooctane and pentane mix was then concentrated and eluted on silica gel.  Hale et al., 2002  Biosolids  Extraction using ASE: two 5-min extraction cycles with DCM at 100 o C and 1,500 psi.  Sample concentrated and cleaned using 2 g, silica gel, solid-phase extraction columns  60 m (except for decaBDE) and 15 (for decaBDE)  Christensen et al, 2003  Biosolids  5g mixed with Chem Tube-Hydromatrix (~5 g) and Soxhlet extracted with hexane:acetone (4:1 v/v) for 24 hours.  Column of alumina, silica gel activated, acidified silica gel and Na2SO4  60 m for tri-heptaBDEs and 15 m for decaBDE.  Ionization Method NCI  Ion source temperature (oC) 150  Electron Energy eV 133  18  Table 2.4 (continued) Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Extraction  Cleanup  Chromatography Column Length  Temperature Program  Analysis Ionization Method ECD  Ion source Electron temperature Energy eV o ( C) Not Not specified specified  Reick, 2004  Biosolids and soil  US EPA 3540 C: Sample mixed with Na2SO 4 followed by Soxhlet extraction with hexane.  Cleanup was with H2SO4 and Florisil according to USEPA Methods 3665 A and 3620 B  30 m  Not Specified  North, 2004  Biosolids  ~10 g of sample mixed with Na2SO 4 and Soxhlet extracted with DCM.  Sample cleaned with silica, Florisil and alumina chromatographic columns.  30 m  100 °C for 3 min, increase to 320 °C at 5 °C/min, and then held isothermal at 320 °C for 5 min.  Not specified  Not specified  Not specified  Rayne and Ikonomou, 2005  Biosolids  Samples were Soxhlet extracted for 16 h with toluene-acetone (5:1 v/v).  60 m  100oC for 1 min, 2oC/min ramp to 140oC, then 4oC /min ramp to 220oC, 8oC /min ramp to 330oC, hold for 1.4 minutes.  EI  315  Not specified  Ciparis and Hale, 2005  Biosolids  5 g mixed with 50 g Na2SO4 and ASE extracted with DCM.  Washed with KOH, high performance liquid chromatography grade water and H2SO4 and reduced in volume. DCM:hexane (1:1 v/v) were added to extract. Cleanup included passing extract through multi-layer silica column packed with basic, neutral, acidic and neutral silica gel layers. Sulphur and residual water removed by glass column filled with copper filings and Na2SO4. Sample eluted through an activated alumina column capped with anhydrous Na2SO4 using hexane followed by DCM-hexane (1:1 v/v). Silica gel.  15 m  For: BDE49, 47, 100, 99, 85, 154, and 153 Initial column temperature was held at 75oC for 1 min, then to 350oC at 10oC/min, and then held for 4.5 min. For BDE209: Initial column temperature was held at 120oC for 3 min, then ramped at 10oC /min to 350oC  ECD  950  Not specified  19  Table 2.4 (continued) Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Extraction  Cleanup  Chromatography  Analysis  Column Length  Temperature Program  Ionization Method  80oC for 2 minutes, 20oC/min ramp to 200oC, for triheptaBDE, 3oC /min ramp to 315oC, hold for 24 minutes; for octa-decaBDE, 6oC /min ramp to 315oC, hold for 10 minutes. 110oC for 1 min, 15oC /min ramp from 110oC to 180oC, then 2oC /min from 180oC to 280oC, and a final 10 min hold at 280oC.  Sellstrom et al., 2005  Soil  Same as Nylund et al., 1992.  Extraction same as Nylund et al., 1992  For tri-heptaBDE: 30 m; for octa-decaBDE: 15  Anderson and MacRae, 2006  Biosolids  Samples loaded in 300 mm ×23 mm chromatography columns with a plug of glass wool at the bottom. The samples were eluted with 300 ml of DCM at a rate of 1–2 ml per minute.  Acidic silica gel cleanup.  30 m for BDE71, 85, 99, 100, 138, 153, 154 and 183.  Knoth et al., 2007  Biosolids  2g freeze dried then Soxhlet extracted  Multi layer SiO 2–AgNO3, H2SO 4, NaOH; Macro Al2O3; GPC; Biobeads SX3; Mini Al2O3  15 m  Not specified  Gevao et al., 2007  Biosolids  15-20 g of sample mixed with Na2SO 4, then Soxhlet extracted for 1216 h using hexane. Cu was added during extraction to remove sulphur.  The extract was washed with H2 SO4 and NaCl (5% w/v) solution. The extract was then eluted through silica gel and alumina.  For tri-nona: 30 m For BDE 209: 15 m  For all PBDEs except BDE209: 80oC for 1 min and ramped at 12oC/min to 155oC, then 4oC/min to 215oC, and finally ramped at 3oC/min to 300oC and held for 10 minutes. For BDE 209: Held at 80oC for 2 min. and then ramped at 25oC/min to 220 oC, then ramped at 5 o C /min to 315 oC and held for 10 min.  Not specified  Ion source temperature (oC) Not specified  Electron Energy eV Not specified  Not specified  Not specified  Not specified  EI  Not specified  Not specified  NCI  230  Not specified  20  Table 2.4 (continued) Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Arnold et al., 2008  Biosolids  Eljarrat et al. 2008  Biosolids & soil  Wang et al, 2008  Peng et al., 2009  Extraction  Cleanup  Chromatography Column Length  Temperature Program  Analysis Ionization Method  Ion source temperature (oC) 300-325  Electron Energy eV Not specified  Sample dried for 24 hours at 24 hr at 50–60 o C, and extracted either by ASE or microwave extraction using hexane:acetone (1:1 v/v). ASE extraction involved three 5hr cycles at 1,500 psi and 150 oC; microwave extraction at 25 psi for 5 min using CED. Samples mixed with copper to remove sulphur and soxhlet extracted with DCM:hexane (1:1 v/v).  Florisil column  30 or 5 m  For 3m column: From 80 to 250oC at 5 oC/min, and from 250 to 315 oC at 3 oC min. For 5 m column: Ramps of 100–250oC at 6oC /min and 250–320 oC at 7 oC /min, then held constant for 5 min.  ECD  Treatment with concentrated H2SO4 and purified with alumina cartridge.  For tri-heptaBDE: 30 m; for octa-deBDE : 15 m  NCI  250  Not specified  Soil  Freeze dried and Soxhlet extracted for 24 h using DCM and hexane (1:1 v/v).  Acidic silica to remove lipid. Then filtered through anhydrous Na2SO4 and mixed with copper to remove sulphur. Extract then passed through multilayered silica gel column containing, from bottom to top, activated silica gel, basic silica gel activated silica gel, acid silica gel, activated silica gel, AgNO3 silica gel and anhydrous Na2SO4.  For tri- to heptaBDE: 30 m For BDE209: 15 m  EI  320-340  Not specified  Biosolids  Freeze dried and Soxhlet extracted for 24 h using hexane:acetone (1:1 v/v)  Activated copper to remove sulphur.  For tri- to heptaBDE: 30 m For octa- to decaBDE: 12.5 m.  For tri-heptaBDE: From 110oC to 180oC at 2oC/min, and held for 5 min then from 240 to 265 at 2oC/min and held for 6 min. For octa to decaBDE: 140oC for 1 minute and then ramped at 10 o C /min to 325oC, and held for 10 min. For tri- to heptaBDE: Initial oven temperature was 90oC, which was held for 2 min and increased to 210oC at 25oC /min, held for 1 min, then increased to 275oC at 10oC /min and held for 10 min and finally ramped to 330oC at 25 C /min and held for 10 min. For BDE209: Initial oven temperature was 90oC, which was held for 1 min. It was then increased to 340oC at 20oC /min and held for 2 min. Not specified  NCI  Not specified  Not specified  21  Table 2.4 (continued) Extraction, cleanup and analytical methods for PBDE determination in biosolids and soil samples. Reference  Sample Type  Extraction  Cleanup  Chromatography Column Length  Temperature Program  Andrade et al., 2010  Biosolids  10g mixed with 30g Na2SO4 and then centrifuged with DCM.  Alumina column  15 m  48oC for 3 min, 20 oC /min to 210 oC, 25 oC /min to 310oC, 310oC for 5 min.  Gorgy et al., 2010  Soil  5 g of sample mixed with 10 g Na2SO 4 and ASE extracted with toluene:acetone (4:1 v/v).  Same as Rayne and Ikonomoou, 2005.  Same as Rayne and Ikonomou, 2005.  Same as Rayne and Ikonomou, 2005.  Xia et al, 2010  Soil  2 g of freeze-dried sample extracted using ASE with DCM using two extraction cycles of 1,500 psi 100oC.  Deactivated silica gel column  30 m  Clark et al, 2008 & 2010  Biosolids  Freeze dried samples extracted using ASE with toluene  Activated copper and AgNO3 followed by acid and base modified silica gels and basic alumina column chromatography.  10 m  50oC and held for 2 min then increased to 200 oC at 25 o C/min, held for 1 min, increased to a final temperature of 300 oC at 6 o C/min, and then held for 10 min Initial temperature of 120°C held for 2 min, a ramp rate of 15°C/min from 120 to 230°C followed by a 5°C/min increase from 230°C to the final temperature of 320°C that was held for 5 minutes  Analysis Ionization Method NCI  Ion source temperature (oC) 300  Electron Energy eV Not specified  Same as Rayne and Ikonomou , 2005.  Same as Rayne and Ikonomou , 2005.  Same as Rayne and Ikonomo, 2005.  ECD  300  Not specified  EI  240  70  Na2SO4 : Sodium Sulphate AgNO3: Silver Nitrate v/v: Volume to Volume basis DCM: Dicholormethane m/m: Weight basis min: minute NCI: Negative Chemical Ionization EI: Electron Ionization ECD: Electron Capture Detector  22  2.2.3 Analysis Analytical techniques to determine congener-specific PBDE compounds usually employ gas chromatogram-high resolution mass spectrometry (GC-HRMS).  Depending on the PBDE  congeners investigated, chromatographic separation columns differ in length. Columns of length 25-30 m columns are common for the separation of tri- to hepta-BDEs, except for a few investigations where the lengths were 50 or 60 m (Hale et al., 2002; de Boer et al., 2003; Rayne and Ikonomou, 2005). For octa- to deca-BDE, 15 m columns are used because BDE209 can degrade at high temperatures (Hale et al., 2002; Oberg et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008; Gevao et al., 2008). In addition, gas chromatography temperature programs for BDE209 differ from those for tri- to hepta-BDEs, due to the sensitivity of BDE209 to higher temperatures. PBDE analyses are mostly based on negative chemical ionization (NCI) or electron ionization (EI) low-resolution mass spectrometry (LRMS). NCI-LRMS of PBDEs is dominated by the ion [Br]-, where the two ions corresponding to m/z= 79 and 81 are monitored. For EI-LRMS, the two most abundant isotope peaks for mono- to tri-BDEs and for tetra- to hepta-BDEs are selected (Eljarrat and Barcelo, 2004). Both ionization methods are subject to interference, especially for biosolids and biosolids-amended soils samples, due to undergoing extensive extraction. This allows them to co-elute a variety of matrix components and disturb the quantitative analysis. EIhigh resolution mass spectrometry (HRMS) provides higher accuracy, while requiring sophisticated, expensive instruments, paired with highly trained personnel and frequent maintenance (Eljarrat and Barcelo, 2004). Other analytic methods are based on electron capture detection (ECD). Table 2.5 summarizes the advantages and drawbacks of alternative detection modes.  23  Table 2.5 Advantages and limitations of different detection methods for PBDEs in biosolids and soil samples (Covaci et al., 2003) Detection ECD  NCI EI  Advantages Purchase and maintenance costs less than EI and NCI Easy to operate Good sensitivity Good selectivity for PBDEs Good sensitivity Good selectivity  Limitations Fairly sensitive to PBDEs Instability of linear range Low selectivity Frequent maintenance required High purchase cost High maintenance cost Difficult to use Higher “down-time”  Studies investigating how the different instrumental analytical methods compare in determining PBDEs levels from biosolids and biosolids-amended soils are lacking. However, Eljarrat et al. (2003) compared NCI and EI in determining PBDEs from sediment samples. In general, EI provided better structural information and is affected by different chlorinated interferences, whereas NCI eliminated these interference and showed better sensitivity, but provided less structural information. NCI is not able to resolve potential co-elution of BDE congeners with different degrees of bromination and also presented different brominated interferences, which were resolved using EI. Studies comparing different laboratory methods involving extraction, cleanup, and analysis of PBDEs specifically from biosolids and biosolids amended-soil samples are also lacking. This gap needs to be addressed in order to better review and compare PBDE levels in such samples. De Boer et al. (2002) reported the results of an inter-laboratory study on PBDEs, organized between November 1999 and April 2000, involving five biological samples, two sediments and two standard solutions. Results from 18 laboratories showed that BDE47 measurements were in reasonable agreement, with a range of relative standard deviations (RSDs) of 17–40%. Results for the BDE99 (RSD 25-77%), 100 (RSD 19-48%), 153 (RSD 30-48%) and 154 (RSD 25-43%) showed that further improvement of these analyses is needed to minimize interference, 24  especially for BDE99. Comparison of BDE209 analysis was not possible for the participating laboratories. The authors have indicated that the performance of the laboratories can be substantially improved by ensuring better calibration, good blanks, better GC resolution, and protecting BDE209 from incoming daylight and ultraviolet light (de Boer and Cofino, 2002). Also, long exposure and high temperatures in the GC should be avoided, and better internal standards should be used. These findings also apply to studying PBDE in biosolids and biosolids-amended soils.  2.3 PBDEs in Biosolids Biosolids are byproducts from the wastewater treatment processes. Sludge, usually in the form of liquid or semi-solids liquid, containing 0.25-12% solids by weight (Tchobanoglous et al., 2003), is treated by stabilization and composting to produce biosolids that can be used for agriculture and land reclamation (Tchobanoglous et al., 2003). On average, 50% of the biosolids are applied to land in the United States and Canada compared with 37% in Europe (Tchobanoglous et al., 2003). PBDE concentrations can vary by one to four orders of magnitude within WWTP, from ∼26×103 pg/L in treated effluents to 413×106 pg/L in biosolids (Rayne and Ikonomou 2005a; Clarke et al., 2010). In some investigations PBDE concentrations have been found to remain unchanged when converting sludge into biosolids (Knoth et al., 2007; Clarke et al., 2010). PBDEs in biosolids have been investigated since 1988 in Europe, North America, Australia, China and the Middle East. The concentrations of the different congeners are compiled in Tables 2.6 and 2.7. Table 2.6 summarizes the concentrations of BDE47, 85, 99, 100, 153, 154, 183 and 209 in biosolids, which are commonly reported. However, in some studies not all of these congeners are reported. 25  Table 2.6 Concentration of PBDE congeners (pg/g dw) reported in published studies on biosolids. Country  Origin Gothenburg (1988)  BDE-47 15,000  BDE-85 19,000  BDE-99 NA  BDE-100 NA  BDE-153 NA  BDE-154 NA  BDE-183 NA  BDE-209  ΣPBDEs  Study  NA  Nylund et al., 1992)  22 WWTPs (2000)  7,000*  420*  10,000*  1,700*  860*  720*  NA  11,000*  Germany  11 WWTPs in Rhine-Main area (2003)  47,900  NA  53,900  9,300  6,300  4,700  3,400  429,000  Denmark  Not specified (2002)  96,800  3,100  86,200  19100  7,800  6,100  2,000  248,000  37,500 (sum of BDE47, 85 & unspecified pentaBDE) 31,700* (sum of BDE47,-85,-99, 100, 153, -154 & 209 555,000 (sum of BDE28, 47, 99, 100, 153, 154, 183 & 209 238,000 (sum of tri- to heptaBDEs  Netherlands  Ameland (1999) St. Annaparochie (1999) Ameland (1999) Barcelona, Burgos, lieid Pamplona and Tortosa (2005) Mid-Atlantc, USA (2000)  9,500  nd  11,000  nd  nd  nd  11,000  nd  14,000  nd  190,000  NA  40,000  nd  38,000  4,800  8600  Sweden  Spain  USA & Canada  Palo Alto, California (2002) British Columbia, Canada (2002) Maine, USA (2002)  27,000  544,000  NA  Oberg et al., 2002 Knoth et al., 2007  Christensen et al., 2003 de Boer et al., 2003  7,000  11,000  4,000  3,000  9,000  484,000  1,311,600  Eljarrat et al., 2008  725,000  266,000  NA  NA  NA  1,470,000  2,810,000 (sum of BDE47, 99,100 & 209) 3,381,000  Hale et al., 2002  2,429,000 (sum of 61 congeners) NA  Rayne and Ikonomou, 2005 Anderson and MacRae, 2006  757,000  34,000  944,000  165,000  88,000  68,000  NA  118,3000  607,500  60,750  729,000  121,500  121,500  60,750  24,300  730,000  457,000  88000  2,130,000  319,000  102,000  108,000  NA  NA  North, 2004  26  Table 2.6 (continued) Concentration of PBDE congeners (pg/g dw) reported in published studies on biosolids. Country  Origin  BDE-47  BDE-85  BDE-99  BDE-100  BDE-153  BDE-154  BDE-183  BDE-209  ΣPBDEs  Study  Lake Superior, USA (2003) Lake Michigan, USA (2003) Ontario, Canada (2003)  767,000  1,327,000  NA  NA  NA  NA  NA  510,000  NA  Hale et al., 2003  507,000  NA  706,000  NA  NA  NA  NA  466,000  NA  NA  NA  NA  NA  NA  NA  NA  6,930,000  637,000 (sum of BDE-47, -99, 153, and -154)  2,050,000a 1,700,000b 1,200,000c 200,000d 950,000e 760,000f 240,000g 100,0000h 225200  NA NA NA NA NA NA NA NA NA  2,400,000 a 1,100,000 b 1,200,000 c 260,000 d 1,100,000 e 860,000 f 320,000 g 120,000 h 166400  405,000 a 355,000 b 200,000 c 43,000 d 18,0000 e 130,000 f 48,000 g 190,000 h 52400  235,000 a 240,000 b 127,000 c 29,000 d 110,000 e 77,000 f 33,000 g 120,000 h 20200  155,000 a 195,000 b 87,000 c 25,000 d 100,000 e 67,000 f 26,000 g 92,000 h 15400  37,000 a 59,000 b NA NA NA NA NA NA NA  4,250,000 a 2,150,000 b NA NA NA NA NA NA NA  USA (2005)  400,000  30,000  600,000  100,000  60,000  50,000  NA  350,000  9,760,500 a 88,37,500 b 2,850,000 c 850,000 d 3,600,000 e 5,410,000 f 960,000 g 4,100,000 h 483,000 (sum of BDE47, 99 100, 153 and 154) NA  US MidAtlantic (2005)  201,500  210,500  81,050  63,350  61,800  NA  870,000  1,540,000  US MidAtlantic (2006)  160,000  NA  170,000  73,000  60,000  57,800  nd  861,000  Arizona (2006)  430,000  NA  612,000  127,000  58,500  46,100  9900  804,000  British Columbia, Canada (2006)  231,000  10,000  268,000  55,000  26,000  24,000  8,000  324,000  Washington (2004)  ` Illinois, USA (2005)  1,426,400 (sum of BDE28, 47, 99, 100, 153, 154 183 and 209) 1,440,000 (sum of BDE28, 47, 99, 100, 153, 154 183 and 209) 2,087,500 (sum of BDE47, 99, 100, 153, 154 183 and 209) 946,000 (sum of 51 congeners)  Reick, 2004  Xia et al., 2010 Ciparis and Hale, 2005 Andrade et al., 2010  Arnold et al., 2008  Gorgy et al., 2010  27  Table 2.6 (continued) Concentration of PBDE congeners (pg/g dw) reported in published studies on biosolids. Country  Australia  China  Origin  BDE-47  BDE-85  BDE-99  BDE-100  BDE-153  BDE-154  BDE-183  BDE-209  ΣPBDEs  Study  US MidAtlantic (2007)  151,000  NA  156,500  69,300  57,000  55,500  31,800  945,500  Andrade et al., 2010  US MidAtlantic (2008)  152,000  NA  160,000  80,100  70,400  694,00  nd  12,20,000  Urban WWTPs (2006)  120,000  4,900  131,000  27,000  13,300  10,900  9,700  530,000  1,500,000 (sum of BDE28, 47, 99, 100, 153, 154 183 and 209) 1,820,000 (sum of BDE28, 47, 99, 100, 153, 154 183 and 209) 1,225,000 (Sum of 34 congeners)  Rural WWTPs (2006)  160,000  5,000  150,000  36,000  13,800  12,000  6,100  490,000  Perth (2008)  47,000  1,700  51,000  9,000  4,600  3,700  1,900  160,000  26 Cities (2005)  5,000  300  4,500  1,000  1,800  2,200  2,100  68,500  911,000 (Sum of 34 congeners) 30,000 (Sum of 34 congeners) 94,000 (sum of BDE17, 28, 47, 66, 71, 85, 99, 100, 138, 153, 154, 183 & 209)  Clarke et al., 2008; Clarke et al., 2010  Clarke et al., 2010 Wang et al., 2007  Guangzhou 14,000 NA 17700 2,850 1000 2,300 2,750 6,586,000 47,000 (sum of Peng et al., (2007) tri-heptaBDE) 2009 Middle East Jahra (2005970 NA 1,950 290 190 310 210 48,500 52,500 Gevao et (Kuwait) 06) al., 2008 Umm4,160 NA 8,400 1,340 820 1,180 440 360,400 376,700 Haylaman (2005-06) Riqqa (20051,860 NA 3,800 610 400 580 330 136,500 14,410 06) ΣPBDEs: Sum of the listed 8 congeners unless otherwise noted; nd: Non-detectable; NA: not available; *wet weight basis; a: biosolid sample 1 collected near Tacoma, WA; b: biosolid sample 2 collected near Tacoma, WA; c: Activated biosolid sample from STP1 along Yakima river, WA; d: Activated biosolid sample from STP2 along Yakima river, WA; e: Activated biosolid sample from STP3 along Yakima river, WA; f: Dewatered biosolid sample from STP1 along Yakima river, WA; g: Dewatered biosolid sample from STP2 along Yakima river, WA; h: Dewatered biosolid sample from STP3 along Yakima river, WA  28  Table 2.7 PBDEs in biosolids collected from a German wastewater treatment plant (Hagenmaier et al., 1992). Concentrations are in pg/g dw. Sample triBDE number 1 100 2 880 3 970 4 na 5 na 6 650 7 na 8 na 9 na Average 650 na: Not available  tetraBDE  pentaBDE  hexaBDE  heptaBDE  170 940 1,240 7,520 4,330 1,430 4,980 3,920 3,010 3,060  220 380 630 7,110 2,680 760 7,510 3,710 4,170 3,019  nd 20 30 1,210 430 70 1,010 880 720 486  nd nd nd 410 270 nd 380 430 460 390  Total triheptaBDEs 490 2,220 2,860 16,250 7,700 2,910 13,840 8,940 8,370 7,064  2.3.1 Europe One of the first European studies (Nylund et al., 1992) was carried out in1988 in Gothenburg, Sweden. Concentrations of BDE47, 85 and another unidentified penta-BDE were reported to be 15×103, 19 ×103 and 4×103 pg/g dw, respectively in biosolids samples obtained from WWTP (Table 2.6). A later Swedish study (Oberg et al., 2002) analyzed 114 samples from 22 WWTP for single PBDE congeners. The results were reported on a wet weight basis (ww). The average concentration of the BDE congeners were between 420 and 11×103 pg/g ww, with BDE209 having the highest concentration (Oberg et al. 2002). It might be expected that the sum of PBDE concentrations in the biosolids tested by Oberg et al. (2002) would be higher than those of Nylund et al. (1992) based on the increased use of PBDEs with time. However, this was not the case. It is also worth mentioning that Nylund et al. (1992) reported the sum of three congeners, none of which was BDE209, whereas Oberg et al (2002) reported the sum of 7 congeners, including the ubiquitous BDE209. Hagenmaier et al. (1992) examined PBDEs in biosolids used in agricultural land application from 13 WWTPs. PBDE concentrations are reported according to homologue groups in Table 2.7. The concentrations of tri-, tetra-, penta-, hexa- and hepta-BDE are in the range of 100 to 970 29  pg/g dw, 170 to 7,520 pg/g dw, 220 to 7,510 pg/g dw, not detected (nd) to 1,210 pg/g dw, and nd to 460 pg/g dw, respectively (Hagenmaier et al., 1992). In March 2002 to June 2003, biosolids samples from 11 municipal WWTPs in Germany were collected. The sum of BDE congeners 28, 47, 99, 153, 154 and 183 ranged from 12.5×103 to 288×103 pg/g dw, whereas BDE209 concentrations varied from 97.1×103 to 2,217×103 pg/g dw (Table 2.6). The two German studies reported PBDE concentrations in biosolids differently. BDE28, 47, 99, 153, 154 and 183, reported by Knoth et al. (2007), were qualitatively equivalent to tri- to heptaBDE concentrations reported by Hagenmaeir et al. (1992). The average sum of BDE28, 47, 99, 153, 154 and 183 was 126×103 pg/g dw from the biosolids samples collected in 2002, while the average sum of tri-hepta-BDE was 7,064 pg/g dw, suggesting that PBDEs in German biosolids increased by least 18 times in 15 years, an indication of the increased use of PBDEs. Biosolids samples from two Dutch WWTPs (de Boer et al., 2003) were analyzed for PBDEs as part of a national study on estrogenic contaminants in aquatic environments (Table 2.6). The concentrations of BDE47, 85, 99, 138, 153 and 209 in the sludge from the Ameland WWTP were 25 ×103, <500, 25×103, <1.4×103, <2.4×103, and 8.6×103 pg/g dw, respectively, whereas those from St. Annaparochie, were 11×103, <700, 14×103, <800, <2.6×103 and 90×103 pg/g dw, respectively. A Danish study (Christensen et al., 2003) examined 12 PBDEs in biosolids from a provincial plant receiving both urban and agricultural wastewater. Concentrations of BDE17, 28, 47, 49, 66, 85, 99, 100, 153, 154 and 183 ranged from 1.7×103 to 96.8×103 pg/g dw, whereas that for BDE209 was much higher, 248×103 pg/g dw (Christensen et al., 2003).  30  In Spain, a study was conducted on biosolids from five wastewater treatment plants (Eljarrat et al., 2008) as part of a study to determine the degree of PBDE contamination in both biosolids used for land application and biosolids-amended soils. Four of these plants treated urban wastewater, and one treated both urban and industrial wastewater. PBDE concentrations in biosolids ranged from 197×103 to 1,185×103 pg/g dw, with BDE209 the predominant congener and the highest for a European biosolids sample (Clarke et al., 2008). 2.3.2 North America Samples from different regions in the USA and Canada have confirmed the presence of PBDEs, at levels up to two orders of magnitude higher than the European levels. Eleven biosolids samples were collected prior to application on land from four U.S. states: Virginia, Maryland, New York and California (Hale et al., 2001a). This study found PBDEs containing 4 to 6 bromine atoms (BDE47, 99, 100, 153 and 154) were present in significant amounts in the biosolids (Table 2.6). The total PBDE concentrations in these biosolids were 1,100 ×103 to 2,290×103 pg/g dw. Concentrations of BDE209 varied widely from 85×103 to 4,890×103 pg/g dw (Hale et al., 2001a). Hale et al. (2002) examined PBDE levels from biosolids samples from a mid-Atlantic WWTP. Biosolids contained 1,370 ×103 pg/g dw of tetra- to hexa-BDEs (sum of BDE47, 99, 100, 153 and 154; Table 2.6), an order of magnitude greater than for European counterparts (Hale et al., 2002), reflective of the greater consumption of PBDEs in North America. In a study (North, 2004) to determine the breakdown of 41 PBDE congeners in a sewage treatment plant in Paolo Alto, California, the total concentration of PBDEs in biosolids samples collected in 2002 ranged from 61×103 to 1,440×103 pg/g dw (Table 2.6). Congeners BDE47, 99, and 209 accounted for 85% of the total PBDEs.  31  Biosolids applied on land in three watershed communities near Lake Superior were reported to have average concentrations of 767×103, 1,327×103 and 550 ×103 pg/g dw for BDE47, 99, and 209 respectively (Table 2.6). For biosolids in 11 watershed communities along Lake Michigan, the same congeners had average concentrations of 507×103, 706×103 and 466 ×103 pg/g dw respectively (Hale et al., 2003). A single Ontario biosolids sample was reported to have a pentaBDE concentration of 637×103 pg/g dw, lower than reported for American samples. However, the BDE209 concentration was 6,930 ×103 pg/g dw, this being higher than for other North American samples (Hale et al., 2003). Concentrations and patterns of the mono- through deca-BDE were found in all major unit operations and processes within a tertiary-level WWTP located in Kelowna, B.C., equipped with post-filtration ultraviolet light disinfection (Rayne and Ikonomou, 2005). Wastewater treatment processes, such as anaerobic, anoxic, and aerobic biological treatment, anaerobic digestion, dissolved air flotation and sand–anthracite filtration, did not substantially degrade or remove PBDEs. However, wastewater sludges were observed to provide an overall removal efficiency of 93% for PBDEs due to sorption on the solids. Hence, the resulting biosolids were observed to contain a total PBDE concentration of ∼2.4×106 pg/g dw, with BDE47 and 209 contributing 25 and 30%, respectively, of this total. Biosolids samples from an undisclosed WWTP in B.C. were analyzed for 51 PBDE congeners in 2006 by Gorgy et al. (2010). The average total PBDE concentration was 946×103 pg/g dw, 4 orders of magnitude less than reported by Rayne and Ikonomou (2005a). BDEs 47, 99 and 209 contributed 25, 28 and 34%, respectively to the total concentration of PBDEs. The percent contributions of BDE47 and 209 were similar to those reported by Rayne and Ikonomou (2005a), although the origins of the samples differed.  32  Activated biosolids from a WWTP in Orono, Maine servicing residential, university and commercial areas were analyzed (Anderson and MacRae, 2006) for nine PBDE congeners (BDE47, 71, 85, 99, 100, 138, 153, 154 and 183 to evaluate potential sources of brominated flame retardants in the community. The average concentrations of selected PBDE congeners were between non-detectable (nd) and 2,130×103 pg/g dw, with BDE99 having the highest concentration. Xia et al. (2010) evaluated levels of PBDEs in biosolids from 16 WWTPs in Illinois field plots which had received annual applications of biosolids for 33 years. The sum of the concentrations of BDE47, 99, 100, 153, and 154 congeners ranged from 71×103 to 1,020×103 pg/g dw. The relative abundance of these congeners followed the order BDE47 > BDE99 >BDE100 > BDE153 and BDE154. Biosolids from Washington State WWTPs were highly contaminated with PBDEs (Reick, 2004). The concentrations from biosolids from two WWTPs in the Tacoma region ranged from 37×103 to 4,250×103 pg/g dw. Concentrations in biosolids from activated and dewatered sludge from three WWTPs located along the Yakima river were 25×103 to 1,700×103 and 26×103 to 1,000×103 pg/g dw, respectively. Arnold et al. (2008) investigated seven PBDE congeners (BDE47, 99, 100, 153, 154 and 209) in biosolids derived from a conventional WWTP in Tuscon, Arizona. The congener concentrations ranged from 430×103 to 804×103 pg/g dw (Arnold et al., 2008). BDE47, 99 and 209 were found to contribute 21, 29 and 40%, respectively of the total concentration of these six congeners. Andrade et al. (2010) examined PBDE levels in biosolids from a large WWTP in the midAtlantic U.S. for 32 months between 2005 and 2008. The sum of the BDE27, 47, 99, 100, 153, 154, 183 and 209 concentrations increased from 202×103 to 182×104 pg/g between 2005 and 2008. 33  2.3.3 China In 2005 biosolids from 31 WWTPs in 26 cities in China were analyzed for BDE17, 28, 47, 66, 71, 85, 99, 100, 138, 153, 154, 183 and 209 (Wang et al., 2007). The sum of the concentrations of these congeners, excluding BDE209 ranged from 6.2×103 to 57×103 pg/g dw. The concentration BDE209 varied from below its limit of detection (1×103 pg/g) to 1,109 ×103 pg/g dw, and averaged 55% of the total (Wang et al., 2007). These levels are one to two orders of magnitude lower than those found in Europe. In 2006 seventeen congeners (BDE28, 47, 99, 100, 153, 154, 183, 196, 197, 203, 205, 206, 207, 208, and 209) were analyzed in biosolids by Peng et al. (2009) to characterize the occurrence, fate, and transport of PBDEs in two sewage treatment plants in the Pearl River Delta, South China. The total concentrations of these 17 congeners were 158×103 to 24×106 pg/g dw. BDE209 was most abundant, providing >90% of the total. 2.3.4 Middle East Gevao et al. (2008) investigated PBDE concentrations in biosolids from three WWTPs in Kuwait. Concentrations of BDE47, 100, 99, 153, 154, 183 and 209 and their sum are reported in Table 2.6. Concentrations were between 470 and 182×103 pg/g dw, with the sum for these seven congeners ranging from 53×103 to 380×103 pg/g dw. BDE209 again had the highest concentration, more than 90% of the total PBDE contamination in the biosolids. 2.3.5 Australia Two studies investigated PBDEs in sludge and biosolids in 2006 from Australian WWTPs (Table 2.6). Clarke et al. (2008) collected sludge samples from each State and the Northern Territory of Australia (Clarke et al., 2008). The mean total concentration of 34 PBDE congeners was 1,137 ×103 pg/g dw, with individual totals between 5×103 and 4,230×103 pg/g dw. The mean 34  total concentrations of PBDEs in sludge samples from urban and rural WWTPs were 1308×103 and 911×103 pg/g dw, respectively. In the other study, Clarke et al. (2010) quantified PBDEs released into the environment (biosolids, effluent) from a conventional activated sludge WWTP in Perth in 2008. The average total PBDE concentration in the sludge was 340×103 pg/g dw. PBDEs in sludge samples were lower than for those collected in 2006 (Clarke et al., 2008). The mean biosolids concentration was 300×103 pg/g dw. Hence PBDEs did not significantly change from converting the sludge into biosolids .  2.4 PBDEs in Biosolids-Amended Soils Few studies have investigated PBDE concentrations in soils which have received biosolids, and these were only from Europe and North America. Results are summarized in Table 2.8. 2.4.1 Europe Samples collected from soils from five different agricultural sites in southern Sweden were examined by Matscheko et al. (2002) and Sellstrom et al. (2005). Both studies focused on the bioavailability and bioaccumulation of PBDEs from soils to earthworms. Matscheko et al. reported the concentrations of BDE47 to BDE183, whereas Sellstrom et al. reported BDE47 to BDE209. Concentrations in the earlier case ranged from nd to 41×103 pg/g dw, whereas those from the later study were between nd and 2,200×103 pg/g dw. Some of the difference may be because the methods of extracting the compounds differed, as well as the use of alternative analytical instruments. In Spain, biosolids-amended soil samples from four different sites used to cultivate winter crops were examined for PBDE contamination by Eljarrat et al. (2008). Biosolids application ranged 35  from 15 to 20 t/ha, and they were ploughed to a depth of 0.15-0.20 m. Concentrations of 11 congeners were between 1×103 and 334×103 pg/g dw. Concentrations of seven of these congeners are reported in Table 2.8. 2.4.2 North America Reick (2004) obtained three samples from a hops field in Washington State, which had received an undisclosed amount of dewatered biosolids. One was a reference sample from an area that had not received any biosolids. One of the other two samples was from disced land, and the third from another location where no disking had taken place, but adjacent to a supporting pole (Reick 2004). The concentrations of PBDEs in the disked soil and adjacent to the supporting pole were 3.1×103 to 39×103 pg/g dw and 120×103 to 1,300×103 pg/g dw respectively (Table 2.8). Andrade et al. (2010) examined the PBDE trends in surface soils in 30 agricultural fields in Virginia. Ten fields had not received any biosolids, while ten had been subjected to one biosolids application 3.5 years prior to 2006, and the remaining ten fields had received more than one biosolids application for 5-15 years until 2006. The average loadings for single and multiple biosolids applications were ∼2.1 and 4.4 t/ha. The biosolids were applied by commercial spreaders that delivered a biosolids cake on top of the soil as a mixture of chunks of various sizes, resulting in an extremely heterogeneous and random layer (Andrade et al., 2010).  36  Table 2.8 PBDE concentrations (pg/g dw) in soils amended with sludge and biosolids. Country or State  Sweden  City  Total Biosolids (t/ha)  Date of Sampling  Soil OM (%)  Condition of field at time of sampling  BDE 47  BDE 85  BDE 99  BDE 100  BDE 153  BDE 154  BDE 183  BDE 209  Total  Ref.  Igelosa  0  6-Apr-00  4.7  33  NA  33  9  5  3  10  na  95  Matscheko et al., 2002  6-Apr-00  4.8  140  NA  170  52  20  17  29  na  430  6-Apr-00  5.4  Plowed autumn 1999, not harrowed or sowed.  300  NA  350  100  48  35  89  na  930  Petersborg  4 t/ha every 4 years from1981to 1997 (Total of 20 t/ha) 12 t/ha every 4 years from 1981 to 1997 (Total of 60 t/ha) 0  Plowed autumn 1999, not harrowed or sowed. Plowed autumn 1999, not harrowed or sowed.  6-Apr-00  2.6  27  NA  20  7  2  1  5  na  63  6-Apr-00  2.8  90  NA  96  26  13  9  10  na  250  6-Apr-00  2.8  Plowed autumn 1999, not harrowed or sowed.  80  NA  98  27  12  10  8  na  240  Lamna  4 t/ha every 4 years from1981to 1997 (Total of 20 t/ha) 12 t/ha every 4 years from 1981 to 1997 (Total of 60 t/ha) 0  Plowed autumn 1999, harrowed and sowed Plowed autumn 1999, not harrowed or sowed.  3-Apr-00  3.4  Sampled before sowing  22  NA  10  2  nd  nd  nd  na  79  34  NA  35  9  nd  nd  nd  na  79  3-4 days after harvest  8  NA  8  2  nd  nd  10  na  29  26  NA  31  8  nd  nd  9.3  na  75  39  NA  50  2  6  4  nd  na  110  230,000  NA  410,000  9  28,000  49,000  1,300  na  840,000  2.3 t/ha in 1998 0  Bjorketorp  2.3 t/ha in 1998 0 Total of 25 t/ha from 1978 to 1982  3.7 19-Sep-00  3.3 3.5  20-Sep-00  5.7 4.9  Ploughed grassland Clover cover  37  Table 2.8 (continued) PBDE concentrations (pg/g dw) in soils amended with sludge and biosolids. Country or State  Sweden  Spain  City  Total Biosolids (t/ha)  Date of Sampling  Soil OM (%)  Condition of field at time of sampling  BDE 47  BDE 85  BDE 99  BDE 100  BDE 153  BDE 154  BDE 183  BDE 209  Total  Ref.  Igelosa  0  6-Apr-00  3.19  15  na  26  6.4  4.9  nd  nd  71  120  Sellstrom et al., 2005  52  na  82  17  nd  nd  nd  410  580  59  na  9.5  18  14  13  nd  61  840  Petersborg  4 t/ha every 4 years from1981to 1997 (Total of 20 t/ha) 12 t/ha every 4 years from 1981 to 1997 (Total of 60 t/ha) 0  Plowed autumn 1999, not harrowed or sowed.  31  na  36  8.9  nd  nd  nd  76  150  18  na  240  48  27  21  nd  620  1,200  310  na  420  78  50  39  nd  1,000  2,100  Lamna (L)  4 t/ha every 4 years from1981to 1997 (Total of 20 t/ha) 12 t/ha every 4 years from 1981 to 1997 (Total of 60 t/ha) 0  Sampled before sowing  6.8  na  11  nd  nd  nd  nd  15  33  16  na  20  nd  nd  nd  nd  28  63  Bjorketorp  2.3 t/ha in 1998 0  Ploughed grassland Clover cover  31  na  58  nd  nd  nd  nd  750  840  450,000  na  59,000  120,00 0  52,000  46,000  nd  22  39×105  Tornabus Hostalets de Pierola Piera+Sabaell Riu Sec Pujalt  Total of 25 t/ha from 1978 to 1982 25-33 t/ha per year 2032005 15 t/ha per year in 2003 and 2005 NM 0  3.3  3.3  6-Apr-00  5.62 6.16  Plowed autumn 1999, harrowed and sowed  6.16  3-Apr-00  3.38 3.77  20-Sep-00  7.22 6.54  29-Apr-05  1,230  1,440  570  1510  1,330  3,190  33,2000  349,000  25-Oct-05  590  540  600  Nd  770  2,890  24,600  30,000  25-Oct-05  33,100  26,900  62,900  3,680  3,320  29,600  25-Oct-05  690  630  1,080  940  930  1,870  Eljarrat et al., 2008  1,082,000 1,185,000 14,600  20,700  38  Table 2.8 (continued) PBDE concentrations (pg/g dw) in soils amended with sludge and biosolids. Country or State  City  Total Biosolids (t/ha)  Date of Sampling  Spain  Pujalt  15 t/ha per year from 1997 to 2004 15-20 t/ha per year in1999, 2000 and 2001 15-20 t/ha in 2004 NM  25-Oct-05  NA  NM  NA  nd  NM  NA  NM  Between hops Next to supporting pole Planted with corn  33,000  39,000  5  5  Washington State  Control  Virginia  Soil OM (%)  BDE 99  BDE 100  BDE 153  BDE 154  BDE 183  BDE 209  Total  20,400  1,610  1,600  1,130  2,800  8600  655,000  689,000  25-Oct-05  920  1,080  2,150  nd  1,080  2,730  71,700  79,600  25-Oct-05  1,300  1,150  1,370  1,190  1,280  15,500  161,000  184,000  150  120  NA  NA  NA  NA  NA  6,700  3,800  NA  NA  NA  NA  NA  260,000  140,000  NA  NA  NA  NA  NA  NM  NA  NM  Reference Soils  Spring 2006  5  Single application of 21 t/ha in 2002 44 t/ha per application between 1996 and 2006  Spring 2006  5.4  Spring 2006  4.4  Condition of field at time of sampling  BDE 47  13×10  BDE 85  140  15×10  nd  nd  nd  nd  nd  nd  nd  nd  nd  Planted with corn  3,500  na  3,830  783  nd  nd  nd  14,000  22000  Planted with Corn  9,100  na  15,000  3,530  2,300  2,200  nd  32,400  64,000  Ref.  Reick, 2004  Andrade et al., 2010  Ref.: Reference NM: Not mentioned na: Not Available nd: Non-detectable  39  The dominant congeners in the agricultural soils were BDE47, 99, and 209 (Table 2.8) contributing 75% of the total concentration in surface soil collected from fields without any biosolids application, and the BDE mean total concentration was 3.0×103 pg/g dw. Fields subjected to one application had a mean total PBDE concentration of 22 ×103 pg/g dw, with BDE47, 99, and 209 together comprising 86% of this total. Fields with multiple applications had a mean total PBDE concentration of 64×103 pg/g dw, with BDE47, 99, and 209 combined contributing 88% of the total. Xia et al. (2010) investigated PBDE in soils from field plots in Illinois which had received annual applications of biosolids for 33 years. The cumulative biosolids loadings were 0, 555, 1,109, and 2,218 t/ha, with the latter 100 times higher than the USEPA-recommended agronomic rate of biosolids application (Xia et al., 2010). Samples were collected as 0-1.2 m soil cores, with each core divided into four depth intervals (0–0.15 m, 0.15–0.30 m, 0.30–0.60 m, and 0.60–1.20 m). Table 2.9 presents the total concentration of PBDEs in each interval. Total concentrations in the surface soil layer (0–0.15 m) increased with increasing cumulative loadings of biosolids. However, the influence of cumulative loading became less significant below a depth of 0.30 m. The total concentration of PBDEs dropped from 150×103 to nd, 300×103 to nd and 650×103 pg/g dw to nd between 0 and 1.2 m for the amended soils, with cumulative loadings of 555, 1,109 and 2,218 t/ha, respectively. Table 2.9 Total concentration of PBDEs (pg/g dw) in biosolids-amended soils at different depths (Xia et al., 2010). Depth (m) 0-0.15 0.15-0.30 0.30-0.60 0.60-1.20 nd: Non-detectable  554 15,0000 75,000 25,000 nd  Biosolids Loading (t/ha) 1,109 30,000 50,000 25,000 nd  2,218 650,000 125,000 55,000 nd  40  2.5 Discussion 2.5.1 Biosolids Results from the published studies summarized above vary widely. Each study includes different congeners, with the two most common being BDE47 and 99. Some do not report heavily brominated congeners, especially BDE209, owing to the difficulties in analyzing it, and also due to it being ubiquitous. This can result in background readings exceeding or similar to the concentrations analyzed (de Boer and Wells, 2006). Figure 2.2 presents results for BDE47, 99 and 209 the dominant PBDE congeners reported in biosolids samples and the corresponding total PBDE concentration. The data are depicted chronologically for Europe and North America, where most studies originated, to facilitate determination of a temporal trend. In general, North American samples contained one to two orders of magnitude more PBDEs than European samples. This is consistent with the amount of PBDEs produced globally (Table 2.2). BDE47 and 99 concentrations were generally highest in North American samples, while samples from one Chinese WWTP (Wang et al., 2007) had the highest BDE209 and total PBDE concentration. The reported concentrations of BDE47, 99, 209 and the total PBDE concentrations were averaged for each sampling year reported by the European and North American studies in Figure 2.3. Levels of BDE47 in biosolids from North America increased from ∼544×103 to 101×104 pg/g dw between 2000 and 2004, followed by a sharp decrease to 152×103 pg/g dw between 2004 and 2008 (Figure 2.3a). Levels of BDE47 in European biosolids were approximately 25 times less than their North American counterparts, but followed a similar pattern, increasing from ∼25×103 to 100×103 pg/g dw between 1999 and 2002, and then decreasing to ∼25×103 pg/g dw from 2002 and 2005.  41  Concentration (pg/g dw)  Figure 2.2 Concentrations of BDE47, 99 and 209 and total concentrations of all reported PBDEs in biosolids analyzed between 1988 and 2008. BDE99 in biosolids from North America fluctuated from ∼706×103 to 127×104 pg/g dw between 2000 and 2004, followed by a sharp decrease to 160×103 pg/g dw between 2004 and 2008 (Figure 2.3b). Levels of BDE99 in biosolids from Europe were approximately 13 to 76 times less than in North American. Levels increased from ∼21×103 to 86×103 pg/g dw between 1999 and 2002, and then decreased to ∼7000 pg/g dw from 2002 to 2005. Levels of BDE209 in North American and European sludge and biosolids followed a different temporal trend (Figure 2.3c). BDE209 decreased from ∼150×104 to ∼100×104 pg/g dw between 2000 and 2002, increased to ∼320×104 pg/g dw between 2002 and 2004, sharply decreased to ∼50×104 pg/g dw between 2004 and 2006, and finally increased to ∼125×104 pg/g dw between 2006 and 2008. The difficulties in obtaining accurate analyses of the BDE209 concentration may be responsible, to some extent, for the apparent fluctuations. 42  (b) BDE99  (c) BDE209  (d) Total PBDEs  4  Concentration (× 10 pg/g dw)  (a) BDE47  Year  Year  Figure 2.3 Average concentration of BDE47, 99, 209 and total of all reported PBDEs in North American and European biosolids. The total concentration of all measured PBDE in North American biosolids increased from ∼300×104 to ∼550×104 pg/g dw between 2000 and 2004, sharply decreased to ∼110×104 pg/g dw between 2004 and 2006, and finally increased to ∼182×104 pg/g dw between 2006 and 2008. Levels of BDE209 in European sludge and biosolids increased from ∼163×103 to ∼545×103 pg/g dw between 1999 and 2005. As shown in Figure 2.4, the congener profile (% of total) for BDE47, 85, 99, 100, 153, 154, 183 and 209 is dominated by BDE209. This profile provides a better indication of the dominance of BDE47, 99 and 209 in biosolids: BDE47 and 99 constituted 5- 25% and 2-32%, respectively of the total PBDEs found in European biosolids, and 10-20% and 10-40%, respectively in the North 43  American biosolids, whereas BDE209 constituted 35-90% and 30-70% in the European and North American sludge and biosolids, respectively. The temporal change of the percent contributions of BDE47, 99 and 209 is similar to their temporal change in concentrations described in Figure 2.3. The transition in the percent contribution of BDE47 and 99 is similar to the relative amounts in the PeBDE commercial mixtures (Table 2.10), indicating increased consumption of this commercial mixture from 1998 to 2004. The subsequent decreases in concentrations and percent contributions of these two congeners coincide with the period when European and North American industries voluntarily phased out the sale of PeBDE.  Figure 2.4 Percent contribution of BDE47, 85, 99, 100, 153, 154, 183 and 209 to their total concentration in North American and European biosolids.  The percent contribution of BDE209 to the total PBDE concentration in the biosolids is lower than in the commercial product DeBDE, the latter being more than 90%. Soderstrom et al. (2004) confirmed that BDE209 is unstable; it can debrominate due to temperature increases and  44  photolysis. It is therefore likely that BDE209 in wastewater influent decomposes to lower congeners as the wastewater and sludge undergo treatment. Table 2.10 Concentrations (%, w/w) of PBDEs in the PeBDE, OcBDE, and DeBDE commercial products (Sellstrom et al., 2005; La Guardia et al., 2006). Technical Product PeBDE DE-71 Bromkal 70-5DE OcBDE DE-79 Bromkal 79-8DE DeBDE Saytex 102E Bromkal 82-ODE  BDE47  BDE99  BDE153  BDE209  38.2 42.8  48.6 44.8  5.44 5.32  nd nd  nd nd  nd nd  8.66 0.15  1.31 49.6  nd nd  nd nd  nd nd  96.8 91.6  nd: Not-detectable 2.5.2 Soils The percent contributions of BDE47, 85, 99, 100, 153, 154, 183 and 209 to their total concentration in soil (Figure 2.5) differ from those in the biosolids. BDE85 and BDE209 could not be included where it was not reported (Table 2.6). Percent contributions of BDE47, 99, 100, 153, 154 and 183 were 2-40%, 1-50%, 3-10%, 3-5%, 3-5%, and 2-16%, respectively. The wider ranges of BDE47 and 99 indicate their mobilization, as well as degradation of higher-brominated compounds (Matscheko et al., 2002; Sellstrom et al., 2005; Andrade et al., 2010). BDE209 was the dominant congener, 50-90% of the total, when reported, and had the highest percent contribution in the reference soils, consistent with the ubiquitous nature of this congener. Its increased percent contribution in the amended soils likely reflects the fact that lower brominated congeners are more mobile, and thus have been transported deeper into the soil, away from where the biosolids were applied. Another factor could be that the lower congeners are more volatile, and hence may be preferentially lost to the atmosphere (Gouin and Harner, 2003; Wania and Dugani, 2003). However, these transfers are bound to be dependent on the time between biosolids application and sampling of the soil, as well as climatic conditions.  45  Further investigations are essential to determine how PBDEs introduced to the soils, debrominate to lower congeners, mobilize in the soil, and escape to the atmosphere over time.  Figure 2.5 Percent contribution of BDE47, 85, 99, 100, 153, 154, 183 and 209 to their total concentration in amended soils. (NA: Biosolids loading was not available)  Only one published study (Xia et al., 2010) has investigated the PBDE distribution at depth exceeding 0.20 m. These authors investigated the distribution of the sum of PBDEs over a depth of 1.2 m in fields to which biosolids had been applied. PBDEs decreased sharply with increasing depth. The authors did not report on specific congeners, nor did they investigate the variation of concentrations with time. There is a lack of vital information on the fate of PBDEs in biosolidsamended soil with time starting from when the biosolids were applied. This information could shed light on the PBDE transformation, leachability and fate in biosolids-amended soils. The effect of soil properties on PBDE concentration in biosolids-amended soils has not been discussed in published studies. Properties such as organic matter content (OM) are important in 46  the sorption of PBDEs by soils and their mobility in the soil. PBDE sorption increases with increasing soil organic mater content, hence retarding its transport. The OM of the soils tested in the published studies ranged from 2.6 to 7.2% (Table 2.8). Concentrations of the different BDE congeners are plotted against the reported organic matter content in Figure 2.6 and are summarized in Table 2.11. Although there is a clear indication of the increased PBDE concentration with increasing OM, the trend is difficult to express numerically.  Figure 2.6 Relationship between the PBDE congeners reported in the published studies and the organic matter content of the tested soils (Matscheko et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008; Andrade et al., 2010). BDE85 was not plotted since the majority of the concentrations were reported as non-detectable.  47  Table 2.11 Concentration (pg/g dw) of PBDE congeners reported in the published studies and the organic matter content of the tested soils plotted in Figure 2.6 (Matscheko et al., 2002; Sellstrom et al., 2005; Eljarrat et al., 2008; Andrade et al., 2010). OM BDE 47 BDE 99 BDE 100 BDE BDE BDE BDE Total (%) 153 154 183 209 nd 2.6 27 20 7 2 1 5 63 nd 2.8 90 96 26 13 9 10 250 nd 2.8 80 98 27 12 10 8 240 nd 3.2 15 26 6 5 nd 71 120 nd 3.3 8 8 2 nd 10 nd 29 nd 3.3 52 82 17 nd nd 410 580 nd 3.3 59 10 18 14 13 61 840 nd nd 3.4 7 11 nd nd 15 33 nd nd nd 3.4 22 10 2 nd 79 nd nd 3.5 26 31 8 nd 9 75 nd nd nd 3.7 34 35 9 nd 79 nd nd 3.8 16 20 nd nd 28 63 nd 4.4 9,100 15,000 3,530 2,300 2,200 32,400 64,000 nd 4.7 33 33 9 5 3 10 95 nd 4.8 140 170 52 20 17 29 430 nd 4.9 230,000 410,000 9 28,000 49,000 1,300 840,000 nd 5.0 nd nd nd nd nd nd nd nd 5.4 300 350 100 48 35 89 930 nd nd 5.4 3,500 3,830 783 nd 14,000 22,000 nd nd 5.6 31 36 9 nd 76 150 nd 5.7 39 50 2 6 4 nd 110 nd 6.2 18 240 48 27 21 620 1,200 nd 6.2 310 420 78 50 39 1,000 2,100 nd 6.5 450,000 59,000 120,000 52,000 46,000 22 3,900,000 nd 7.2 31 58 nd nd nd 750 840 BDE85 was not included since the majority of the concentrations were reported as non-detectable. nd: Non-detectable  Available studies differ in many aspects such as: (1) method of extraction and cleanup of samples, affecting the percent recovery of each congener since some extraction methods, such as accelerated solvent extraction vs. Soxhlet extraction or even simple mixing; (2) instruments used in the PBDE analysis, which affect detection limits, and the range of congeners analyzed, such as the ability to determine BDE209; (3) PBDE congeners reported; and (4) methods of characterizing the biosolids and soil samples when present. These differences all affect the accuracy and precision of the data reported, the basis for comparison, and the possibility of establishing trends and correlations.  48  There is also a lack of information on the temporal variability of PBDEs in the biosolids applied to agricultural soil. Studies monitoring the vertical distribution of PBDEs at depths greater than 0.20 m in soils that have received biosolids are vital to identify the fate and transport of these contaminants in biosolids and soils.  2.6 Prediction of Total PBDEs in Soil Major factors controlling the initial concentrations of PBDEs in soils are the biosolids loading and the method of application. Based on the application rates tested in the studies which have investigated PBDEs in biosolids-amended soil, we have estimated the initial total PBDE concentrations for European and North American biosolids (Andrade et al., 2010) as:  ( " %+ " % PBDE$ pg dw'- . Application Rate$ g 2 ' . Area m 2 * " pg % ) # g &, # m & Csoil$ dw' = " % " % # g & Soil Volume m 3 . / $ kg 3 ' . $1000g ' kg & # m & #  ( )  (Equation 2.1)  ( )  !  Some assumptions were needed to perform the calculations: (a) The average total concentration of PBDEs in European and North American biosolids applied were 545×103 and 175×104 pg/g dw, respectively. (b) The soil density, ρ, was 1,300 kg/m3. (c) The depth of the soil into which the biosolids were blended was assumed to be 0.20 m. The predicted total concentrations in the biosolids-amended soil compared with the observed average concentrations are presented in Figure 2.7. The predicted concentration for the European biosolids-amended soils are 8 and 32 times less than the measured values for biosolids loadings of 25 and 33 t/ha. The other values for both European and North American biosolids-amended soils are over-predicted. The overprediction is no doubt due to not considering volatilization, degradation and mobilization of PBDEs in the biosolids-amended soils. More investigation is required to determine the rate of 49  volatilization, degradation and mobilization of PBDEs in the soil so that the degree of PBDE  Total Concentration (pg/g dw)  contamination in biosolids-amended soils can be modelled and predicted more accurately.  Biosolids Loading (t/ha)  Figure 2.7 Comparison of measured average total concentration of PBDEs in biosolidsamended soils with simple predictions based on the assumption that PBDEs are mixed (without degradation or loss by transfer) into the top 0.2 m deep layer of soil.  2.7 Recommendations As PBDE levels have increased at alarming rates in different matrices and in humans, researchers have focused on methods to improve the analysis of these contaminants. Although numerous studies have been published on monitoring PBDE levels in aquatic environments, marine life, birds, animals and humans, there is a lack of published work on the fate and transport of PBDE in soils. Given that 40-60% of the biosolids produced by European and North American, contaminated with PBDEs, are applied to land, studies of this nature are important since soils can become a pathway for contamination of different organisms. 50  Although two of the three commercial BDE products are no longer being manufactured or sold, the third is still being made and deployed, and household and commercial items containing all three are still in widespread use, awaiting ultimate disposal. It is important to establish guidelines for the reporting and analysis of PBDEs in biosolids themselves and in soils which have received biosolids. Uniformity in reporting would facilitate comparison of the results from different studies, helping to establish trends for different congeners in different samples. These guidelines should include reporting the chemical and physical properties of the samples, such as the organic matter content, cation exchange capacity, and water content of the soil and biosolids. This information would be useful in determining the effect of these properties on the fate and transport of PBDEs when biosolids are applied to soils. The available published studies on PBDE levels in biosolids-amended soils have been reviewed. More research of similar nature is needed. In addition, it is recommend that the levels of PBDEs at different depths below the surface be measured after biosolid application to identify the vertical distribution of PBDEs in the soil, as well as to monitor their levels over time. Information from such studies will elucidate the temporal trends of these contaminants and also their mobility in soils. This will aid in assessing risk of contamination to groundwater and, ultimately, surface waters.  2.8 Conclusion Biosolids from wastewater treatment plants tend to be significantly contaminated with PBDEs. Soils which have received biosolids are also contaminated with PBDEs. PBDEs present in biosolids likely originate from the PeBDE and DeBDE commercial products used as flame retardants in consumer goods, whereas the percent contribution of PBDEs does not indicate any  51  significant influence of OcBDE. The results also indicate that these congeners can break down to lower congeners. The ratios of concentrations of BDE47 and 99 to the total PBDE concentrations in soils were found to be less than that of BDE209, and also less than the corresponding ratios in biosolids. This is attributed to heavier congeners such as BDE209 being less mobile in the soil and less volatile than their lighter counterparts. Greater consistency is needed in data reporting to facilitate further research, establish trends, and determine the fate and transport of PBDEs in biosolids and soils.  52  3 A Preliminary Field Investigation on the Mobility of Polybrominated Diphenyl Ethers in Biosolids-Amended Soil 3.1 Introduction Biosolids are byproducts from municipal and industrial wastewater treatment plants. The treatment of wastewater produces residual biosolids, which are of high organic matter contents, ranging from 40-70% dry matter depending on the type of digestion used during wastewater treatment (Laturnus et al., 2007). Biosolids are applied to agricultural soils to improve their vegetation yield. The organic matter content, which can be as high as 60% and high in nutrients, such as phosphorus and nitrogen (Laturnus et al., 2007) bolsters the capacity of the soil. Thus their application helps to maintain the organic content of agricultural soil, as well as improving its quality by providing nutrients necessary for plant growth. In North America and Europe, approximately 50% and 37%, respectively, of the biosolids produced are applied to farmland. In British Columbia, Canada, it was estimated that ∼ 67% (16,300 tons) of biosolids are applied to farmland and reclaimed lands (Environment Canada, 1999). Applying biosolids leads to several concerns. One issue is PBDEs, which are found at elevated levels in biosolids. In European biosolids, PBDEs ranged from 3×103 to 248×103 pg/g dw (Allchin et al., 1999; Christensen et al., 2002; Matscheko et al., 2002; de Boer et al., 2003). PBDEs in sludges and biosolids produced in North American wastewater treatment plants (WWTPs) were found to be 30 times higher than in Europe, with concentrations between 460×103 and 7,000×103 pg/g dw (Hale et al., 2001b; Rayne and Ikonomou, 2005a; Andrade et al., 2010; Gorgy et al., 2010b).  53  PBDEs have been widely added to a variety of consumer products as flame retardants to impede the initiation of fire. There are thee commercial PBDE products: PeBDE, OcBDE and DeBDE mixtures. These commercial mixtures have been used extensively since the 1980s (Birnbaum and Staskal, 2006) although the PeBDE and OcBDE products have been phased out in many jurisdictions. Due to the fact that they can readily leach from consumer products, PBDEs are widely present in different environmental matrices including: air, water, sediment, human milk and blood samples, and in birds and marine mammals (de Wit, 2002; Ikonomou et al., 2002; Hites, 2004). PBDEs are ubiquitous because they are not readily degradable. They have n-octanol-water partitioning coefficients (KOW), ranging between 104 and 1010. These high KOW values and the incorporation of PBDEs in many products have led to their partitioning into high organic-mattercontent media, such as wastewater effluent discharge, sludge and biosolids. In addition to PBDEs, a wide range of other organic and inorganic contaminants appear in biosolids (Bright and Healey, 2003). Organic compounds, which have been studied extensively include chlorinated organic compounds such as chlorobenzenes (CBs), PCBs, dioxin and furans, i.e. hydrophobic compounds, which qualitatively have similar physical and chemical properties as PBDEs (Eljarrat et al., 2008). Once introduced into agricultural soil, these contaminants can partition to dust particles, some of which enter the atmosphere or suspend in surface water by erosive action caused by rain and wind. The degradation of organic compounds in the soil can lead to the production of more toxic compounds (Beck et al., 1996; Bright and Healey, 2003; Laturnus et al., 2007). Introducing organic compounds to agricultural soil can lead to the transport of these compounds through the soil to groundwater, thus endangering drinking water sources (McGechan and Lewis, 2002). Hence, there is concern that PBDEs can behave in a similar manner in agricultural soils. 54  The transport parameters of contaminants from soil are strongly influenced by their water solubility. The mobility in groundwater and soil is generally inversely related to the hydrophobicity (Persson et al., 2008). Hydrophobic compounds are expected to strongly sorb to OM (Chiou et al., 1983) and mineral surfaces of the soil (Schwarzenbach et al., 2003). The mineral surface of the soil can be described by its cation exchange capacity (CEC). Particle size of soils also affect the mobility of these compounds, however, OM and CEC are generally considered to have greater influence on their mobility (Schwarzenbach et al., 2003). Soils high in OM and CEC will tend to sorb more hydrophobic organic compounds (HOCs) than soils with low OM and CEC. Hydrophobic compounds can also move considerable distances through the subsurface if they sorb or attach to mobile colloidal particles (Flury and Qiu, 2008). Enhanced mobility of these contaminants may cause contamination of groundwater. Previous literature on PBDEs in biosolids-amended agricultural soil has focused on the top 0.30 m layer. Studies in Europe and North America have reported PBDE levels of 580 to 840×103 (Matscheko et al., 2002; Sellstrom et al., 2003; Sellstrom et al., 2005; Eljarrat et al., 2008) and 140×103 to 7,600×103 pg/g dw (Reick 2004; Andrade et al., 2010), respectively in the top 0.30 m layer. However, investigations on PBDE-contaminated soils at depths beyond 0.30 m are not available. Such measurements are important as they provide a more holistic view on the vertical mobilization of PBDEs in biosolids-amended soils at depth and on potential leaching into groundwater resources. This study was conducted to examine PBDE contamination in agricultural soils due to biosolids application, to investigate the extent of PBDE vertical mobilization over time and the effect of soil CEC on PBDE distribution. In this chapter, we present the results of a preliminary investigation on PBDE mobility in biosolids-amended soil at an experimental farm in Kamloops, 55  British Columbia, Canada examining soils of different biosolids loadings and experimental conditions. In British Columbia, bioslids can be applied on forest, landfill, reclaimed mine, range or agricultural lands (Sylvis Environmental, 2008). To our knowledge, this is the first study that has examined PBDE mobilization in agricultural soils treated with biosolids at depths greater than 0.30 m. An intended outcome of this work is to realize that PBDEs do mobilize vertically in such environments and could eventually end-up in groundwater reservoirs.  3.2 Material and Methods 3.2.1 Sampling Site Kamloops (Figure 3.1) is located in the South-Central Interior of British Columbia, a dry region, with an average rainfall of 260 mm/year and an average snowfall of 770 mm/year. The average temperatures in winter and summer are -6.9 and 20.9°C, respectively.  0  300 km Kamloops Vancouver  Figure 3.1 Location of Experimental Farm in Kamloops, British Columbia (Source: atlas.gc.ca).  Kamloops was chosen because it is the site of an Agriculture and Agri-food Canada test farm where biosolids had been applied under a wide range of application conditions. The field sloped slightly towards the west, and was divided into four sectors, where 0, 20, 60 or 80 t/ha of 56  biosolids, obtained from an undisclosed urban WWTP had been applied. When applied in April of 2002, the biosolids were mixed with the top 0.03 m of the soil, and the soil was then seeded with alfalfa. 3.2.2 Soil Sampling Sampling was performed in two stages. The first took place in December 2004 to provide a preliminary indication of the degree of PBDE contamination in the biosolids-amended soil. Core samples were collected from areas which had received 20 t/ha and 80 t/ha biosolids (Figure 3.2). Details of soil sampling are provided in Appendix A. Each core sample was divided into three 100 mm layers. Sub-samples from each layer were stored in hexane-washed amber glass jars at 30ºC until analysis. 5m  2m  Locations of 1.05 m deep excavated pits for August 2005 sampling. The side under the 80 t/ha loading was divided into five 0.2 m layers, and samples were obtained from each.  Locations of 1.05 m deep excavated pits for August 2005 sampling. The side under the 80 t/ha loading was divided into five 0.2 m layers, and samples were obtained from each.  80 t/ha Locations of sampling in December 2004, at the center of the 5 × 2 m cell  80 t/ha  N  20 t/ha  Background sampling location 2 m away from the plot amended with biosolids, taken in August 2005.  Figure 3.2 Locations of sampling areas in December 2004 and August 2005. The area was divided into 5×2 m cells onto which different biosolids loadings were applied.  The second sampling stage in August of 2005 with samples collected from five locations: one from an area, referred to as reference soil, where no biosolids had been applied to determine the 57  background levels, and the remaining four from areas that had received 80 t/ha biosolids (Figure 3.2). The objective of the second sampling program was to determine the PBDE distribution in the biosolids-amended soils over a depth of 1 m. In each of these locations an open pit was excavated to a depth of 1 m and divided into six horizontal layers from which samples were taken corresponding to intervals of 0.00-0.05 m; 0.05-0.25 m; 0.25-0.45 m; 0.45-0.65m; 0.650.85 m; and 0.85-1.05m. The 0.00-0.05 m layer was excluded from sampling since information on surface treatment was not available. The open pit excavations allowed samples to be obtained from different depths without cross-contamination. 3.2.3 Physical and Chemical Characterization The moisture content, CEC and OM content of soil samples were determined according to the procedures outlined by the Land Resource Research Institute (LRRI) (1965). Briefly, moisture content was determined by the weight loss after drying the samples at 100±5ºC to a constant weight. The organic matter content was determined by loss-on-ignition using a muffle furnace for 2 h at 550ºC. CEC was measured by the modified sodium acetate replacement method, described in Appendix A. The concentrations of the exchangeable cations were measured by determining the Na concentration using Thermo Jarrell Ash Video 22 (Thermo Fisher Scientific Inc.) atomic adsorption spectrophotometer. 3.2.4 PBDE Determinations Approximately 10 g soil subsamples were used for PBDE determinations. These were spiked with a suite of  13  C-labeled PBDEs (13 C BDEs 3, 15, 28 47, 77, 118, 99, 100, 153 and 183)  internal standards, and  13  C-labeled BDE209 obtained from Cambridge Isotope Laboratories  (Andover, Massachusetts). The sub-samples were then homogenized with anhydrous sodium sulphate with a mortar and pestle, and then transferred into 33 mL stainless steel cells of a  58  Dionex ASE 2000 accelerated solvent extraction system (ASE). The ASE used a solvent mixture of 80:20 (v/v) toluene:acetone to extract the PBDEs at a pressure of 2,000 psi (1,370 kPa) and a temperature of 180°C.  Two static extractions were performed and 40 mL of extract was  collected for each sample. The sample extracts were then passed through a multilayer silica column packed with successive layers of silica gel (basic, neutral, acidic, neutral) and eluted with dichloromethane (DCM)hexane (1:1 v/v). Sulphur and residual water were then removed from the extracts by eluting them in a column filled with copper filings and Na2SO4, with 1:1 v/v DCM-hexane. Finally, the extract was passed through a neutral (pH 7) activated alumina column capped with anhydrous Na2SO4 and washed with hexane, followed by DCM–hexane (1:1 v/v) to elute the analytes of interest. Eluants from the alumina column were concentrated to <10 µL and spiked with a suite of  13  C-labeled PBDEs, (Cambridge Isotope Laboratories, Andover, Massachusetts) prior to  congener-specific PBDE analyses by high-resolution gas chromatography high-resolution mass spectrometry (HRGC/HRMS).  The instrument was a VG-Autospec high-resolution mass  spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas chromatograph and a CTC A200S autosampler (CTC Analytics, Zurich, Switzerland). For all analyses, the mass spectrometer was operated at 10,000 resolution in the positive ion mode at 39 eV energy and data were acquired in the single ion resolving mode. Analytes were identified by comparing the retention time with authentic calibration standards. Concentrations were calculated by the internal standard isotope dilution method using mean relative response factors determined from calibration standards, prior to and following sample analyses. Details on the composition of the internal standards, sorbents, solvents and conditions during sample extraction and in all of the cleanup steps, as well as the instrumental analyses used and the criteria used analyte identification and quantification are described in detail by Ikonomou et al. (2001). 59  Samples were analyzed in batches of 12 samples, including 9 actual samples, one of which was analyzed in duplicate, a procedural blank sample and an in-house reference sample. Method blanks, consisting of Na2SO4, were extracted following the same procedure as for environmental samples to check for potential contamination during sample extraction and cleanup.  The  recoveries of all surrogate internal standards were between 55 and 110% and the accuracy of determining PBDEs in spiked samples was between ±15-20%. 3.2.5 Data Analysis The concentrations of the samples collected in December 2004 were each based on single samples obtained from 20 and 80 t/ha areas, whereas the concentrations of the samples collected in August 2005 were each determined from the four samples from each layer. Analysis resulted in measuring 51 of the 209 PBDE congeners. In each batch, concentrations of PBDEs in the soil samples were corrected by subtracting the procedural blank concentration. For each sample a total PBDEs concentration, referred to hereafter as ΣPBDE, was calculated by adding all of the measured concentrations of the 51 PBDE congeners. To determine the congener distributions in the biosolids and soil samples, the cumulative sums of specified homologue groups were calculated. Error bars indicate 95% confidence limits, obtained from repeated measurements on the same samples. Of the 51 congeners measured, BDE47, 85, 99, 100, 153, 154, 183 and 209 were the most abundant, and were also detected in almost all samples examined.  The  concentrations of these eight principal congeners are considered in the discussion below. Another reason for focusing on these congeners discussions below is that these have also been reported in other published studies, facilitating comparison with our data.  60  3.3 Results and Discussion 3.3.1 Soil Properties Table 3.1 shows the moisture content, OM and CEC of Kamloops reference (un-amended) soils, sampled in August 2005, which were 18-20%, 3-4% and 5-14 milliequivalents per 100 g (meq/100 g), respectively. Results for the amended soils appear in Table 3.2. The addition of biosolids increased the OM and CEC of the soil. The OM was ~1.7 to 4 times greater in the samples collected from the 20 and 80 t/ha biosolids-amended areas in December 2004, respectively, than in the top 0.25 m layer of the reference location. There was also an increase in OM content of the 0.05-0.25 m samples collected from the 80 t/ha biosolids-amended areas in August 2005, which was ~2.7 times greater than the same layer of the reference soil. However, the four underlying layers were similar in OM and significantly lower than the overlying layers. Table 3.1 Moisture content, cation exchange capacity and organic matter content of Kamloops field reference soils sampled on August 2005. Moisture CEC(2) OM(4) Content(1) (3) (meq/100g) (%) (%) 14 0.05-0.25 20 3 5 0.25-0.45 20 4 5 0.45-0.65 20 3 6 0.65-0.85 18 3 8 0.85-1.05 18 4 (1) Moisture content based on weight (2) Cation Exchange Capacity (3) Milliequivalents per 100 g, which is equivalent to cmol/kg. Na concentrations are reported in Appendix A. (4) Organic Matter content based on weight Depth below surface (m)  Number of Samples 1 1 1 1 1  The CEC in the 0.05-0.25 m layer soil samples collected in December 2004 from the 20 and 80 t/ha biosolids-amended areas were ∼ 1.8 and 2.9 times greater than in the 0.05-0.25 m layer of the reference soil, respectively. In the 0.05-0.25 m biosolids-amended soil samples layers collected in August 2005, the CEC was ∼ 1.4 times greater than in the 0.05-0.25 reference soil 61  layer, and ∼ 1.8-2.8 times greater in the other underlying layers. The CEC in the 0.05-0.25 m soil sample layers were lower than their December 2004 counterparts possibly due to leaching of the minerals in the surface layer as a result of snow melt and rainfall surface runoff. Table 3.2 Moisture content, cation exchange capacity and organic matter content of Kamloops biosolids-amended soils Soil  20 t/ha 80 t/ha  Number of samples  1 1  Depth Moisture below CEC(2) Content(1) surface (meq/100g)(3) (%) (m) Samples collected December 2004 29 25 0.05-0.35 29 40.0 0.05-0.35 Samples collected August 2005  4 0.05-0.25 22 20 4 0.25-0.45 22 8 80 t/ha 4 0.45-0.65 22 7 4 0.65-0.85 22 9 4 0.85-1.05 19 11 (1) Moisture content based on weight (2) Cation Exchange Capacity (3) Milliequivalents per 100 g, which is equivalent to cmol/kg. Na concentrations used to determine CEC are presented in Appendix A (4) Organic Matter content based on weight  OM(4) (%)  5 11 8 3 3 3 3  3.3.2 PBDEs Concentration and Vertical Distribution Mean concentrations of the principal PBDEs in the three layers (100 mm in thickness) of the core samples taken in December 2004 are depicted in Figure 3.3. BDE85, 100, 153, 154 and 181 are represented as their total labelled “Others” in Figure 3.3. Table 3.3 presents the concentration of each of the principal congeners. BDE47, 85, 99, 100, 153 and 154 and 181 in the samples collected in August 2005 were non-detectable (nd) in the 0.05-0.25 m layer of the Kamloops reference soil. The levels of PBDEs in the 80 t/ha biosolids-amended samples were ∼18-36, 1232 and 1-4 times greater in the 0.05-0.15, 0.15-0.25 and 0.25-0.35 m layers, respectively than their corresponding values in the 20 t/ha biosolids-amended samples, and were 66-1380 times greater than in the reference soils. 62  PBDE levels in the biosolids-amended soils were broadly similar in magnitude to levels reported in published studies on biosolids-amended soils from Sweden, Spain and the US. A comparison of PBDE levels for specific congeners in the top 0.30 m layers of soils which have received biosolids from Sweden (Matscheko et al., 2002), Spain (Eljarrat et al., 2008) and Maryland USA (Andrade et al., 2010) and our study is provided in Figure 3.4. The samples in these studies were collected from the 0 to 0.35 m interval of the soils which received single loadings of 20-60 t/ha biosolids. The BDE47, 99, 100, 153, 154, 183 and 209 congeners were in the range of 11027,000, nd-9,300, 40-1,900, 20-4,000 15-3,300, 450-9,000 and 1,000-500,000 pg/g dw, respectively (Matscheko et al., 2002; Eljarrat et al., 2008; Andrade et al., 2010). It is worth noting that these investigations did not report BDE85. PBDEs reported by Andrade et al. (20080 and Eljarrat et al. (2008) were ~ 4-200 and ∼ 20 times greater, respectively than these in the 0.05-0.25 m layer from Kamloops in August 2005. A possible explanation for this could be the fact that we did not include the top 0.05 m of soil in the samples we examined. As described below this layer is enriched in biosolids and thus high concentrations of PBDEs. Conversely, our PBDE concentrations were 2-12 times greater than the levels reported by Matscheko et al. (2002). Although samples in the latter case also included the 0.05 m layer, the sampling took place almost 2 years after application of the biosolids. Based on findings from other studies it is possible that the amount of PBDEs in the top layer could be reduced due to a number of factors including leaching owing to rain and snowfall, volatilization (Harner and Shoeib, 2002; Gouin and Harner, 2003) and photodegradation (Sellstrom et al., 1998).  63  Figure 3.3 PBDE concentrations from single samples, on a dry weight basis (dw) basis, in three layers below reference soil surface samples in August 2005 and biosolids application surface from December 2004 sampling: Layer 1: 0.05-0.15 m; Layer 2: 0.15-0.25 m; Layer 3: 0.25-0.35 m below soil surface. Note that “Others” represent the total concentration of BDE85, 100, 153, 154 and 183. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3).  64  Table 3.3 PBDE concentrations (pg/g dw) from single samples, on a dry weight basis (dw) basis, in three layers below reference soil surface samples in August 2005 and biosolids application surface from December 2004 sampling. Reference Soil  20 t/ha 0.05-0.25 m 0.05-0.15 m 0.15-0.25 m depth depth depth BDE47 nd 4,219 3,788 BDE99 nd 5,281 4,728 BDE209 112 5,720 1,616 BDE85 nd 227 231 BDE100 nd 1,000 868 BDE153 nd 501 473 BDE154 nd 448 414 BDE183 nd 136 128 Total nd 2,313 2,115 Note that total is the sum of BDE85, 100, 153, 154 and 183 concentrations. nd: Non-detectable. Congeners  Biosolids Loading 0.25-0.35 m depth 250 219 276 nd 43 21 15 29 107  0.05-0.15 m depth 133,366 156,761 103,619 7,733 31,354 15,949 13,138 4,970 73,144  80 t/ha 0.15-0.25 m depth 70,898 83,020 50,337 2,794 15,491 8,338 7,243 2,268 36,134  0.25-0.35 m depth 775 905 524 42 162 76 60 23 364  65  Figure 3.4 Comparison of PBDE concentrations in the 0.05-0.25 m layer soil samples collected from Kamloops in August 2005 with PBDE concentrations reported in published investigations from 0.0-0.30 m layers of biosolids-amended soils (Matscheko et al., 2002; Eljarrat et al., 2008; Andrade et al., 2010). Entries without concentration bars denote that these congeners were not reported.  Concentrations of BDE47, 85, 99, 100, 153, 154, 183 and 209 in the biosolids-amended soils sampled in August 2005 are presented in Figure 3.5. BDE47, 85, 100, 153 and 154 and 181 were nd in Kamloops reference soil. BDE99 was 21 pg/g dw in the 0.25-0.45 m layer, while BDE183 was 28 and 42 pg/g dw in the 0.05-0.25 and 0.25-0.45 m layers. BDE209 ranged between 25 and 112 pg/g at depths between 0.05 and 1.05 m. The total concentration of the 8 principal congeners was between 28-129 pg/g dw in the reference soil. PBDEs in the 0.05-0.25 m layers from the August 2005 samples were 2-3 orders of magnitude lower than from December 2004. This decrease in PBDE concentrations from the top layer could be caused by a number of factors including the leaching of PBDEs from the biosolids and downward transport with the infiltrating water from precipitation and irrigation during the 866  month period between samplings. The decrease can also be caused by the photochemical decomposition and volatilization of PBDEs (de Wit 2002; Soderstrom et al., 2004). Microbial degradation of PBDEs can also be a factor, however it is considered to minutely contribute to the decrease of PBDEs because these compounds are known to be resistant to biodegradation. Further investigations in their microbial degradation in soil environments are required since most investigations have only focused on controlled and isolated cultures (Gerecke et al., 2001; He et al., 2006; Robrock et al., 2008; Robrock et al., 2009).  Figure 3.5 Concentrations of BDE47, 85, 99, 100, 153, 183 and 209 in 0.05-1.05 m Kamloops 80 t/ha biosolids-amended soil sampled in August 2005. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3).  An attempt to describe the PBDE dependence on depth resulted in exponential decay relations with coefficients of correlation (R2) below 0.58. Figure 3.6 presents the fitted exponential functions with depth for homologue groups tetraBDEs (Figure 3.6a), pentaBDEs (Figure 3.6b), hexaBDEs (Figure 3.6c), decaBDE and total PBDEs (Figure 3.6d), which led to R2 values with  67  depth of 0.58, 0.57, 0.47, and 0.51, respectively. The modest R2 values provide further evidence of the inconsistent distribution of PBDEs with depth. Exponential functions of the remaining homologue groups are not presented since the R2 were below 0.3. The concentration of PBDEs changed over time in biosolids-amended soil. Total PBDE concentration in the 0.05-0.25 m layer decreased from approximately 309×103 to 10×103 pg/g dw between December 2004 and August 2005. PBDEs generally decreased with increasing depth for the soil samples collected in August 2005 (Figure 3.6). However, PBDEs were slightly greater in the 0.85-1.05 m interval than in the 0.65-0.85 m layer. The decrease in PBDEs in the 0.05-0.25 m biosolids-amended soil from the August 2004 concentrations is mostly attributed to transport via suspended and/or colloidal matter. Gorgy et al. (2010) observed an increased partitioning of PBDEs to the aqueous phase due to the suspended particles. In addition, studies on hydrophobic contaminant of similar physical and chemical properties as PBDEs have shown the these contaminants can be mobilized via suspended and colloidal matter in the infiltrating water (Flury and Qiu, 2008). Volatilization and photochemical decomposition of PBDEs could have also occurred (Andrade et al., 2010). The concentrations of the PBDEs at depths below 0.05 m were greater than those in the reference soil, confirming downward migration of PBDEs via the infiltrating water. Except for the 0.65-0.85 and 0.85-1.05 m intervals, the concentrations decreased with depth. PBDEs varied by up to 2 orders of magnitude in the 0.05-0.25 m, 0.25-0.45 m and 0.85-1.05 m intervals (45 to 1,360 pg/g dw, 11 to 230 pg/g and 4 to 120 pg/g dw, respectively), but varied over a smaller range in the 0.45-0.65 m, and 0.65-0.85 m intervals (3 to 33 pg/g dw and nd to 32 pg/g dw).  68  Concentration (pg/g dw)  (a) TetraBDEs  -3.91x  y = 1607.54e 2 R = 0.58  Concentration (pg/g dw)  (c) HexaBDEs  (b) PentaBDEs  -4.12x  y = 2301.76e 2 R = 0.57  (d) DecaBDE & Total PBDEs  Total PBDE = 9441.7e 2 R = 0.51 y = 321.76e  -3.19x  2  R = 0.53  Depth (m)  -3.57x  BDE209 = 955.74e 2 R = 0.47  -3.15x  Depth (m)  Figure 3.6 Exponential functions describing (a) tetraBDEs; (b) pentaBDEs; (c) hexaBDEs and (c) total PBDE and BDE209 distribution with soil depth. Note that the concentration axes have different scales.  The increase in the PBDE concentrations in the 0.085-1.05 m interval over those in the 0.65-0.85 interval could be due to the soil properties at that depth. The transport of contaminants in soils is greatly affected by the OM content and CEC of the soil and the KOW of the contaminants (Schwarzenbach et al., 2003). High OM and/or CEC and high KOW cause contaminants to accumulate in the soil rather than mobilize. PBDEs, as mentioned above, have high KOW. Also the 0.85-1.05 m interval has a higher OM and CEC (4 % and 11 meq/100 g, respectively) than the 0.45-0.65 m and 0.65-0.85 m intervals (3%, 7-9 meq/100 g, respectively), allowing PBDEs to accumulate more in the 0.85-1.05 m interval.  69  PBDE dependence on the CEC of the soil is presented in Figure 3.7. TetraBDEs (Figure 3.7a), pentaBDEs (Figure 3.7b), hexaBDEs (Figure 3.7c), decaBDE and total PBDEs (Figure 3.7d) have an approximately exponential fashion with increasing CEC. R2 values are slightly greater than those derived from the variation with soil depth, 0.64, 0.67, 0.67, 0.68 and 0.73 for tetraBDEs, pentaBDEs, hexaBDEs, decaBDEs and total PBDEs respectively.  Concentration (pg/g dw)  (a) TetraBDEs  (b) PentaBDEs  (c) HexaBDEs Concentration (pg/g dw)  0.27x  0.25x  y = 8.07e 2 R = 0.68  y = 7.70e 2 R = 0.68  (d) DecaBDE and Total PBDEs Total PBDE = 55.57e 2 R = 0.73 0.22x  y = 3.75e 2 R = 0.67  0.26x  23x  BDE209 = 10.07e0. 2 R = 0.68  CEC (meq/100 g)  CEC (meq/100 g)  Figure 3.7 Exponential functions describing (a) tetraBDEs; (b) pentaBDEs; (c) hexaBDEs and (c) total PBDE and BDE209 distribution due to the cation exchange capacity (CEC) of the soil. Note that concentration axes are not the same scale.  The error bars in Figure 3.5, with 95% confidence limit and one standard deviation (Appendix A) and low R2 values in Figures 3.6 and 3.7 indicate wide variability among the four August 2005 sampling locations. Likely causes of this non-uniformity are:  70  - Non-uniform application of the biosolids, causing the same layers at the four locations to have different properties such as CEC, organic matter content and PBDE content; and/or - Heterogeneity of the soil within the area where the biosolids were applied, causing different soil porosities at the four locations, allowing water to infiltrate and carry the PBDEs at different rates.  3.4 Conclusions The application of biosolids to agricultural soil caused an increase in PBDE concentration over depths up to at least 1 m. The results indicate that PBDEs are mobile and migrate downwards below the soil surface to depths of at least 1.05 m. This can be attributed to their desorption from the soils and leaching with infiltrating water, likely due to suspended and/or colloidal matter carried by the water, causing increased partitioning of PBDEs to the aqueous phase. There was difficulty, given the limited number of samples which could be analyzed, in establishing a general trend for the vertical distribution of PBDEs in the soil due to uncontrolled factors such as non-uniform application of biosolids onto the surface and soil heterogeneity. Measured PBDE levels were less than in the American and Spanish studies, but greater than for a Swedish study where biosolids were applied to agricultural soils. Plotting tetraBDEs, pentaBDEs, hexaBDEs, decaBDE and total PBDEs against soil depth showed an exponential decay function. R2 values were between 0.41 and 0.58. Plotting the same concentrations with the CEC of the soil provided an exponential growth function, with R2 values between 0.67 and 0.73.  71  4 PBDE Leachability from Biosolids and their Partitioning Characteristics in the Leachate2 4.1 Introduction Polybrominated diphenyl ethers (PBDEs) are used as flame-retardants in plastics, textiles, electronic circuitry and insulation material. There are 209 PBDE congeners divided into 10 homologue groups, classified by the number of bromine atoms. Commercially, these flame retardants are available as technical mixtures dominated by congeners BDE47 (tetraBDE), 99 (pentaBDE), 100 (pentaBDE), 153 (hexaBDE), 154 (hexaBDE), 183 (heptaBDE) and 209 (decaBDE) (D'Silva et al. 2004). These compounds can be released from products, entering the environment and leading to human and wildlife exposure. Increasing levels of PBDEs in humans and wildlife are causing concern (Hites, 2004). These compounds generally have physico-chemical properties similar to polychlorinated biphenlys (PCBs), polybrominated biphenyls (PBBs), DDT, and to the human hormone thyroxin (Talsness, 2008). PBDEs have been shown to have serious toxicological effects in a wide range of tested organisms (Zuurbier et al., 2006). It is essential to understand the mechanisms by which PBDEs are mobilized in different environmental media and to identify the pathways by which humans and animals are exposed. The fate and transport of PBDEs in soils have not been extensively investigated. Soils are considered to interact with, and act as sinks for, PBDEs in the environment, since these contaminants are lypophilic, with high n-octanol-water partitioning coefficients (KOW), in the range of 104-1010, and thus adsorb onto particles having high organic content. Under certain 2  A version of Chapter 4 has been published in Water Air and Soil Pollution Journal: Gorgy, T., Li, L. Y., Grace, J. R. and Ikonomou, M. G. (2010). Polybrominated diphenyl ether leachability from biosolids and their partitioning characteristics in the leachate. Water Air and Soil Pollution Journal 209(1-4): 109-121. 72  environmental conditions such as seasonal increases in atmospheric temperature, PBDEs can be released from surface soils to the atmosphere due to volatilization, and thus be re-introduced into the environment (Gouin et al. 2005). It is common practice in Europe and North America to dispose of sludge and biosolids produced from sewage treatment plants onto agricultural lands. PBDE levels in European sludge and biosolids have been found to range from 3 ×103 to 248×103 pg/g dry weight (dw) basis (Allchin et al., 1999; Christensen et al., 2002; de Boer et al., 2003; Matscheko et al., 2002), whereas North American concentrations were 460 ×103 to 7,000 103 pg/g dw (Hale et al., 2001; Rayne and Ikonomou, 2005). Hence, the application of sewage sludge and biosolids onto agricultural soils is a source of PBDE contamination. Studies in Sweden and North America have reported PBDE levels of 580 to 840 ×103 (Eljarrat et al., 2008; Matscheko et al., 2002; Sellstrom et al., 2003; Sellstrom et al., 2005) and 140 ×103 to 7.6 ×106 pg/g dw (Reick, 2004), respectively, in agricultural soils that had received biosolids. Contaminant mobility/transport in soils is often studied as a two-phase system in which the contaminants partition between an immobile soil phase and the mobile aqueous phase (McGechan and Lewis, 2002). Hydrophobic compounds such as PBDEs sorb on the immobile soil phase. McCarthy and Zachara (1989) showed that particulate matter, suspended in the aqueous phase, acts as a third mobile phase. The particulates can sorb contaminants in a similar fashion as the immobile solid phase (i.e. soil), thereby transporting the contaminants in a similar manner as in the aqueous phase. The mobilization of PBDEs in soil can lead to suspension of particles containing PBDEs and passage of dissolved PBDEs into groundwater, thus becoming bio-available to humans and wildlife. This can occur in agricultural soils containing macropores due to wormholes and passages created by withered roots. Jarvis et al. (1999) and Kretzschmar  73  et al. (1999) have suggested that suspended matter can reach groundwater by passing through these macropores in the unsaturated zone. The sorption of PBDEs on particles can also affect the accuracy of the analytical determination of the concentration of these compounds in aqueous samples. Extracting these compounds from aqueous  samples  usually  involves  liquid-liquid  extraction  with  either  toluene  or  dichloromethane. Studies of PBDEs in wastewater influent and effluent samples (de Boer et al., 2003; Rayne and Ikonomou, 2005; Song et al., 2006) have used filters of 1.2 or 1.6 µm pore size to separate the dissolved and solid fractions from such samples prior to extraction. PBDEs in the retained solid fractions are commonly extracted using vigorous procedures such as Soxhlet extraction or accelerated solvent extraction (ASE). The determination of particles in leachate samples can vary depending on the method of sampling. The level of particulate matter in the leachate samples has a significant impact on the concentrations of these compounds in the corresponding samples, as these compounds are known to bond heavily onto organic particles. Special caution needs to be undertaken and strict protocols devised when leachate samples are examined in order to obtain consistent data. Both the liquid phase and the particulate matter need to be characterized separately. The use of different pore size filters results in inconsistency in the determination of dissolved fraction of PBDEs in aqueous samples. Large-pore filters lead to the exclusion of PBDEs sorbed onto suspended particles smaller than those captured by the filter, resulting in a measured concentration less than the actual concentration in the aqueous sample. No published laboratory studies have focused on the effect of different pore size filters on the determination of PBDE levels in aqueous samples. The present study investigates the overall concentration and congener profiles of PBDEs in biosolids applied to agricultural soil, and their leachability from the biosolids due to water 74  percolation. In addition, PBDE fractionation in the water and suspended particles of different size ranges were examined to determine the potential for enhanced mobility of these contaminants within soils. To our knowledge, this is the first study to determine the leaching of PBDEs from biosolids, and their distribution between the aqueous phase and different size fractions of suspended particles. The findings are intended to contribute to the understanding of the fate of PBDEs in subsurface environments and to the formulation of transport models for soils. The findings also shed light on the importance of the aqueous filter size in determining PBDEs in aqueous samples.  4.2 Material and Methods The leachability of PBDEs from biosolids was examined in glass column experiments, i.e. leaching column tests (LCTs). The biosolids were derived from an urban WWTP in British Columbia, which cannot be named at its request. At the WWTP, the biosolids were dewatered and underwent thermophilic and aerobic sludge digestion. They were expanded by 10% in volume in the laboratory to facilitate permeation of water. The leachates from the glass columns were filtered through a series of filters, and the biosolids, filtrates and retained particles were analyzed for PBDEs. 4.2.1 Biosolids, Leachate and Suspended Solids Characterization The moisture content and organic content of the biosolids and suspended particles were determined according to the procedures outlined by Chapman (1965). The organic contents, as total organic carbon (TOC), in the leachate and filtrate samples were determined according to Standard Method 5310B (Standard Methods, 1995). The sample was homogenized and a microportion injected into a heated reaction chamber packed with an oxidative catalyst, barium chromate. The water was vaporized and the organic carbon and inorganic carbon oxidized to 75  carbon dioxide (CO2) and water. The CO2 was transported in the carrier-gas stream and measured by an infrared (IR) analyzer (Standard Methods, 1995). The concentrations of suspended particles of two size ranges were determined by filtering 125 ml of the leachate through 1.75 and 0.45 µm pore size glass fibre filters in series. The particles retained on these successive filters are termed “fine” and “ultrafine”, respectively, whereas the filtrates passing through the 1.75 and 0.45 µm filters are denoted “Filtrate A” and “Filtrate B”, respectively. The masses of the retained particles were calculated after subtracting the dry weight of the filters from the dry weight of suspended particles + filter. If it is assumed that the TOC in the leachate and filtrates is due to suspended particles, the TOC per unit mass of suspended particles is:  TOCFines =  !  (MTOC )Leachate " (MTOC )Filtrate A  TOCUltrafines =  (MFines ) (MTOC )Filtrate A " (MTOC )Filtrate B  (M Ultrafines )  (Equation 4.1)  (Equation 4.2)  where:  ! TOCFines = TOC per unit mass of particles retained on 1.75 µm filter (mg/g); TOCUltrafines = TOC per unit mass of particles retained on 0.45 µm filter (mg/g); (MTOC)Leachate = TOC in leachate (mg); (MTOC)Filtrate A = TOC in leachate after passing through 1.75 µm filter (mg); (MTOC)Filtrate B = TOC in leachate after passing through 0.45 µm filter (mg); MFines= Mass of particles retained on 1.75 µm filter (g) MUltrafines= Mass of particles retained on 0.45 µm filter (g)  76  4.2.2 Leaching Column Experiments The biosolids were divided into three 850 g batches and inserted in cylindrical solvent-washed glass columns commonly known as glass leaching cells (glc), capped at both ends by aluminium end-plates (Figure 4.1). Each column was 56.5 mm in height and 100 mm in diameter. Porous ceramic discs of average diameter 100 mm and pore size 60 µm were attached to each end of the cell to ensure well-distributed water flow through the cell. In addition, 100 mm glass fibre filter paper of 5 µm average pore size was inserted on the inner sides of both porous discs. Each glc was connected to a solvent-washed aluminium reservoir, which held the deionized water, by 6.3 mm copper tubing. The water which passed through the biosolids in the glc (referred to as leachate hereafter) was collected in a 4 L solvent-washed amber bottle. Each glc was connected to a bottle by 6.3 mm copper tubing from opposite ends of the cell. The aluminium reservoir was filled with deionized water, pressurized with compressed air controlled by pressure regulators, and monitored by pressure gauges. The head supplied by the air pressure was equivalent to 35 kPa to provide sufficient flow (∼0.4 mL/min) that the experiment could be completed in 7 days. After terminating each experiment, the leachates were stored in a dark room at 4°C. The biosolids in the glcs were stored in hexane-washed amber jars at -30°C until analysis. 4.2.3 Filtration Experiments To determine the distribution of PBDEs in the aqueous and particles portions of the effluent, 250 ml sub-samples from each of the effluents collected in the amber glass bottles were filtered twice: (a) through a 1.75 µm filter, and (b) through a 0.45 µm pore size filter. After 250 mL had passed through the 1.75 µm pore size glass fibre filter, 125 mL of this filtrate was stored at 4°C in a solvent-washed amber bottle for PBDE analysis. The remaining 125 mL were next filtered  77  through the 0.45 µm glass fibre filter, and this filtrate was collected in solvent-washed amber bottle and stored at 4°C (Figure 4.1). To determine the fractionation by particle size, further 250 mL sub-samples from each of the effluents collected in the amber glass bottles were also filtered through a 1.75 µm filter, and through a 0.45 µm filter. After 250 mL had passed through the 1.75 µm glass fibre filter, the filters retaining the particles were oven-dried at 95°C for 24 hours and then weighed. The weights of particles on the filters were determined by subtracting the original filter weights from the weights after oven drying. This process was performed three times for each of the effluents collected from the glcs.  Porous discs with filter paper on inner sides  56cm  Discharge Leachate (4 L collected) end Glass leaching cell 100mm in diameter  Deionized water  Biosolids leachate  Fines retained on 1.75 µm filter were analyzed for PBDEs  1.75 µm filter Filtrate A  Ultrafines retained on 0.45 µm filter were analyzed for PBDEs  125 ml sub-sampled from filtrate A for PBDE analysis  0.45 µm filter  Filtrate B analyzed for PBDEs  Figure 4.1 Schematic diagram of the apparatus used for biosolid leaching and the leachate filtration experiments.  In order to determine the total particles in the effluents, 100 mL subsamples from the three effluents were transferred to three ceramic bowls, and then dried in the oven at 95 °C until constant weights were obtained. 78  4.2.4 Sample Clean-up and Analysis Biosolids and suspended particles samples were stored at -30°C and leachate samples at 4°C until analysis. The biosolids and suspended particles samples were processed in batches of 12, each batch consisting of a procedural blank, a certified reference sample, and nine samples, one of which was in duplicate. Prior to extraction, all samples were spiked with 25 µL of the EO5100 standard, a suite of  13  C-labeled PBDEs, and  13  C-labeled BDE209 (Cambridge Isotope  Laboratories, Andover, Massachusetts). Five grams of biosolids were homogenized with a mortar and pestle with dried anhydrous sodium sulphate (Na2SO4), based on 1:1.5 (sample: Na2SO4 m/m), and then transferred into 33 mL stainless steel cells of a Dionex ASE 2000 accelerated solvent extraction system. The solvent system used for extraction was 80:20 (v/v) toluene:acetone. The particles retained by the filter paper were weighed, and the same extraction procedure followed as for the biosolids. The leachate and filtrate samples were transferred to separatory funnels and extracted three times with approximately equal volumes of toluene. The sample extracts underwent cleanup using a three-step procedure. In the first step, extracts were passed through a multilayer silica column packed with successive layers of silica gel (basic, neutral, acidic, neutral) and eluted DCM-hexane (1:1 v/v). In the second, a column filled with copper filings and Na2SO4 removed sulphur and residual water, with samples again being eluted with 1:1 v/v DCM-hexane. The third step involved a neutral activated alumina column capped with anhydrous Na2SO4 and washed with hexane, followed by use of DCM-hexane (1:1 v/v) to elute the analytes of interest. Eluants from the alumina column were concentrated to <10 µL and spiked with 5 µL of the EO5101 standard, a suite of  13  C- labelled PBDEs, (Cambridge Isotope  Laboratories, Andover, Massachusetts) prior to congener-specific PBDE analyses by HRGC/HRMS. The HRGC/HRMS was a VG-Autospec high-resolution mass spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas 79  chromatograph and a CTC A200S autosampler (CTC Analytics, Zurich, Switzerland). Details on the composition of the internal standards, sorbents, solvents and conditions used during sample extraction and in all of the cleanup steps and the quality assurance/ quality control protocols followed were reported in detail by Ikonomou et al. (2001). 4.2.5 Data Analysis Data compilation and analysis resulted in measuring the concentrations of 51 of the 209 PBDE congeners. Congener concentrations measured in the procedural blanks were subtracted from the measured concentrations. All corrected congener concentrations were added to provide a total concentration, referred to hereafter as ΣPBDE. To determine the congener distribution trends in the biosolids, leachate and particles, the cumulative sums of specified homologue groups were calculated. Error bars indicate 95% confidence limits, obtained from repeated measurements on the same sample. Concentrations of the measured PBDE congeners are presented in Appendix B. Of all the congeners measured, BDE47, 85, 100, 99, 154, 153, 183 and 209 were the most abundant congeners, and were also detected in almost all samples examined. The concentrations of these eight principal congeners and their ratios to ΣPBDE are considered in the discussion below. Another reason for selecting these congeners in corresponding discussions was that these have also been reported in many previous studies, allowing for inter-comparison of our data.  4.3 Results and Discussion 4.3.1 Biosolids, Suspended Particles and Leachate Characteristics The average moisture content of the biosolids was 78%, whereas the organic content was 34% on a dry weight basis. TOC was reduced from 43 to 26 to 23 mg/L by means of the two levels of filtration, indicating that the particles were contributing greatly to the measured organic content  80  beyond dissolved organic matter in the leachate. Following the 0.45 µm filtration, TOC levels approached those of the deionized water. There were 32 mg/L measurable particles in the leachate, with ~71% retained on the 1.75 µm filter and a further ~5% retained on the 0.45 µm filter, meaning that ~7 mg/L of the particles passed through the 0.45 µm filter. The calculated TOC per unit mass of particles retained on the 1.75 µm filter was 800 mg/g, increasing to 1600 mg/g for particles passing through the 1.75 µm, but intercepted by the 0.45 µm filter. 4.3.2 Leachability of PBDEs from Biosolids The biosolid concentrations of the eight major PBDE congeners monitored are reported in Table 4.1. The PBDE levels measured from the analyses of replicate samples approached 1 µg/g dry weight (946×103 pg/g actual, see Table 4.1). These concentrations are 1.5 to 2 times higher than PBDE levels obtained in European biosolids applied to land (Allchin et al., 1999; Christensen et al., 2002; de Boer et al., 2003; Matscheko et al., 2002), but three to four orders of magnitude lower than in biosolids applied on land in the United States (Christensen et al., 2002; de Boer et al., 2003; North, 2004; Reick, 2004). The major contributors to the PBDE measured levels were the data from the tetra, penta and decaBDE homologue groups. The major contribution of these homologue groups to the biosolids could be due to the domestic wastewater treated by the WWTP from which the biosolids were obtained. DeBoer et al. (2003) and Oberg et al. (2002) reported that these compounds arise from widespread household use of PBDE-containing foams, plastics, and textile-based consumer products.  81  Table 4.1 PBDE concentration (pg/g dry weight) in biosolids prior to and after leaching with deionized water, and percent reduction due to leaching. Three samples were analyzed in each case. CInitial Congener BDE47 BDE85 BDE99 BDE100 BDE153 BDE154 BDE183 BDE209 ΣPBDE  Mean 231×103 10×103 268×103 55×103 26×103 24×103 8×103 324×103 946×103  Standard Error 28×103 2×103 35×103 7×103 5×103 4×103 2×103 120×103 7×103  Mean 28×103 1.5×103 33×103 7×103 3×103 3×103 1×103 292×103 369×103  CFinal Standard Error 930 90 980 220 90 100 30 31×103 1.2×103  % reduction 88 85 88 87 88 87 87 10 61  The results from the leachability experiments using distilled deionized water are summarized in Table 4.1. Both the initial (pre-leaching) and the final (post-leaching) concentrations, CInitial and CFinal respectively, are reported for all major BDE congeners measured in the biosolids. Each of these concentrations represents the average for three biosolids samples inserted in the glcs. The initial and final concentrations were in the ranges of 8×103 to 324×103 pg/g dw basis and 1×103 to 292 ×103 pg/g dw, respectively. The percentage reduction due to leaching, i.e. 100% (CInitial CFinal)/ CInitial, is also given in Table 4.1. Note that the passage of water through the columns resulted in removal of more than 85% of each of the congeners, except for BDE209, where only 10% was removed. As discussed below, BDE209 is the least soluble congener, so its lower removal is not surprising. ΣPBDE concentrations in the biosolids were reduced by more than 60% from 946 ×105 to 368 ×105 pg/g dw. The cumulative PBDE concentrations as a function of bromination level are presented in Figure 4.2. It is evident from the data that the tetra, penta and decaBDE homologue groups contributed most to ΣPBDE in both the pre-leaching and postleaching biosolid samples.  82  Concentration (pg/g dw)  Number of Bromines  Figure 4.2 Cumulative PBDE concentration (pg/g dw) in biosolids before (CInitial) and after (CFinal) leaching. Note: error bars cannot be seen due to the logarithmic scale used for the vertical axis.  4.3.3 PBDE Association with Suspended Particles and Leachate Leachate PBDE concentrations ΣPBDE increased from non-detectable to 48×106 pg/L in the deionized water after leaching through the biosolids. This concentration, after filtration through the 1.75 µm (Filtrate A) and 0.45 µm (Filtrate B) filters decreased to 95 ×104 and 38×104 pg/L, respectively (Figure 4.3). The concentrations of the individual PBDE congeners in the leachate, Filtrate A and Filtrate B are presented in Table 4.2. In all cases except for BDE47, the leachate concentrations exceed their estimated solubilities (SW) (Palm et al., 2002). Even essentially insoluble congeners with a high degree of bromination, such as BDE209, appear in the leachate. However, after filtration, the concentrations of BDE85, 99, 153 and 154 in the corresponding filtrates were well below their SW values, indicating that the particles retained by the filters were the main source of the PBDEs found in the leachate. From the data, it appears that more than 80% of the ΣPBDE in the leachate was sequestered in the fines and ultrafine particles captured by the corresponding filters.  83  Individual PBDE congeners were reduced by 97-98% except for BDE 47 and 100, which were  Concentration (pg/L)  reduced by 65% from their initial values by 1.75 µm filtration, as shown in Table 4.3.  Number of Bromines  Figure 4.3 Cumulative PBDE concentration (pg/L) in the pre-filtered leachate (CTotal) and after sequential filtration using 1.75 (Filtrate A) and 0.45 (Filtrate B) µm filters.  PBDEs in Suspended Particles The distribution of PBDE congeners in the fine and ultrafine particles are provided in Figure 4.4. The PBDE concentrations were 4×106 and 21×106 pg/g dw, respectively (Table 4.3), indicating that the ultrafine particles can sorb more than twice as much PBDEs per unit mass as the fines. This could be related to higher organic carbon content in the smaller particles as noted above and/or the higher surface area of ultrafine particles. As shown in Table 4.3, the concentrations of the selected congeners in ultrafines were 2.8-12.3 times the corresponding concentrations in the fines, with the highest ratio corresponding to BDE209, the least soluble of the congeners. The higher concentrations for PBDEs associated with suspended particles, in particular the ultrafine particles, agree with published studies on particle transport of hydrophobic contaminants. Several studies have noted the decrease in distribution coefficient, the ratio of the sorbed phase concentration to the solution phase (Gschwend and Wu, 1985; Katsoyiannis and 84  Samara, 2005; Voice et al., 1983). The non-settling particles in the aqueous phase increased the amount of hydrophobic contaminants, such as PCBs and other persistent organic pollutants, in the aqueous phase (Katsoyiannis and Samara, 2005). Voice et al. (1983) and Gschwend and Wu (1985) have described particles and organic molecules of sizes between 0.1 and 10 µm as carriers for organic pollutants in the aqueous phase.  In addition, column experiments on  radionuclides and pesticides (McGechan and Lewis, 2002) showed that they were transported attached to colloidal matter, 0.01-1 µm in size. Table 4.2 Estimated water solubilities (pg/L) and concentrations (pg/L) of selected PBDE congeners in the leachate, Filtrate A and Filtrate B; three samples were analyzed in each case. Congener  n  SW (pg/L)  BDE47 BDE85 BDE99 BDE100 BDE153 BDE154 BDE183 BDE209  3 3 3 3 3 3 3 3  9×106 4×104 8×105 5×105 4×103 4×103 2×102 Insoluble  Leachate Mean (a) St Error 7.81×106 27×103 4 47×10 2×103 4 975×10 34×103 4 186×10 6×103 3 115×10 3×103 3 91×10 24×103 4 33×10 1×103 4 925×10 208×103  Filtrate A Mean (b) St Error 1.77×105 2×103 3 11×10 70 220×103 1400 43×103 600 3×103 140 2×103 140 10×103 30 290×103 2×103  Filtrate B Mean (c) St Error 16×104 810 5000 150 13×104 4500 40×103 200 13×103 400 16×103 100 4×103 100 2×103 15  n: Number of samples; S W: Estimated water solubility (Palm et al., 2002); SW: values for BDEs 100, 153 and 183 were based on mean values established from corresponding homologues; Filtrate A: Filtrate passing through 1.75 µm filter; Filtrate B: Filtrate passing through 0.45 µm filter; St Error: Standard Error  Table 4.3 Concentrations (pg/g dw) of selected PBDE congeners on the retained particles and ultrafines/fines ratios; three samples were analyzed in each case. Congener  n  BDE47 BDE85 BDE99 BDE100 BDE153 BDE154 BDE183 BDE209 ∑ PBDE  3 3 3 3 3 3 3 3 3  Fines Mean 750×103 42×103 250×103 180×103 88×103 67×103 31×103 1×106 4×106  St Error 300×103 15×103 380×103 79×103 4×103 3×103 14×103 400×103 31×103  Ultrafines Mean 2×106 130×103 890×103 500×103 240×103 190×103 160×103 12×106 21×106  St Error 160×103 9×103 180×103 36×103 20 10 58×103 40×103 200×103  Ultra-fines/ fines 2.9 3.1 2.8 2.8 2.8 2.8 5.1 12.3 5.3  n: Number of samples; Fines: Particles >1.75 µm; Ultrafines:: Particles passing through 1.75 µm filter but retained on 0.45 µm filter; St Error: Standard Error  85  Concentration (pg/g dw)  Number of Bromines  Figure 4.4 Cumulative PBDE concentration (pg/g dw) on fine (>1.75 µm) and ultrafine (1.75 to 0.45 µm) particles.  The higher PBDE concentrations in the ultrafines compared with the fines help to explain the effect of filtration when analyzing PBDEs in aqueous samples. Based on the experimental measurements, a mass balance is performed in the following section on the PBDEs retained on the fines and ultrafine particles and those dissolved in the leachate and filtrates. The data presented in Table 4.4 and Figure 4.5 compare the mass of PBDEs in the leachate estimated from three different approaches: By subtracting the PBDEs in the biosolids after leaching from those measured before leaching (denoted below as fraction (a)); by direct measurement of the volumetric concentration of the PBDEs in the leachate, multiplied by the volume of leachate (referred to as fraction (b)); and by summation of the PBDEs determined in Filtrate B and those in the particles retained on both the 1.75 and 0.45 µm filters, with the first and third of these doubled to account for the fact that only half of Filtrate A passed through the 0.45 µm filter (referred to as fraction (c)). If errors were negligible, all three of these masses should be equal. While there are significant variations among the congeners due to experimental scatter, it is clear 86  from Table 4.4 that the mass ratio of (b) to (a) is always 0.4 or less, whereas that from (c) divided by (a) is closer to 1 in each case, though still <1. The differences suggest that filtration is able to recover PBDEs which one misses when measuring PBDEs dissolved in the liquid. However, the mass balance also indicates that the measurements in Filtrate B are underestimated. It is noteworthy that BDE47, with the highest solubility and therefore affected least by particulates in solution, gives the best match between the calculated values. Considering all the variables associated with these experiments this is a significant finding lending support to the role of particulates in introducing errors into these kinds of mass balance calculations. Table 4.4 Mass balance comparison of PBDE masses (in pg) in the leachate determined based on analyses by three different methods. BDE100  BDE153  BDE154  BDE183  BDE209  PBDEs in Biosolids prior to leaching  BDE47 1.96×108  BDE85 8.30×106  BDE99 2.28×108  4.63×107  2.21×107  2.03×107  7.06×106  2.76×108  PBDEs in Biosolids after leaching (a) PBDEs in leachate by difference. (b) PBDEs measured in leachate (c) PBDEs in leachate from summing that in Filtrate B, fines and ultrafines (PBDEs measured in leachate)/ (Calculated PBDEs lost due to leaching), i.e. (b)/ (a) PBDEs in Filtrate B, fines & ultrafines)/ (Calculated PBDEs in leachate), i.e. (c)/ (a)  1.20 ×108  1.23×106  2.82×107  5.85 ×106  2.87×106  2.35×106  9.33×105  2.48×108  7. 60×107  7. 07×106  1.99 ×108  4.05×107  1.92×107  1.80×107  6.13×106  2.80×107  3.12×107  1.87×106  3.90×107  7.43×106  4.62×106  3.63×105  1.33×106  3.70×106  6.66×107  3.76×106  7.87×107  1.59×107  1.10×107  5.87×106  3.13×106  1.47×107  0.41  0.27  0.20  0.18  0.24  0.02  0.22  0.14  0.88  0.53  0.39  0.39  0.57  0.33  0.51  0.54  The underestimation may be due to the PBDEs adsorbed on particles finer than 0.45 µm in diameter, not captured in the filtration steps. In addition, the liquid-liquid extraction procedure used to extract the PBDEs from Filtrate B may not be rigorous enough to bring into solution (organic solvent layer) the PBDEs attached to particles smaller than 0.45 µm. This can be inferred from the data analysis presented in Figure 4.6. Here the ratios of PBDEs in Filtrate A plus those in fines to PBDEs in the leachate and the ratio of PBDEs in Filtrate B, plus those 87  attached to ultrafines, to PBDEs in Filtrate A are summarized. Both ratios are greater than 1, providing support to the hypothesis that aqueous samples with large quantities of suspended high organic- content particles sequester PBDEs, which cannot be extracted effectively by conventional liquid- liquid extraction techniques.  Mass of PBDEs leached from Biosolids  Mass of PBDEs retained on fine particles Mass of PBDEs in Filtrate A Mass of PBDEs retained on ultrafine particles  Mass of PBDEs Filtrate B  Ratio of BDE Concentrations  Figure 4.5 Mass balance diagram illustrating comparison of PBDE masses (in pg) in the leachate determined based on three different methods.  Figure 4.6 Ratios of BDE47, 85, 99, 100, 153, 154, 183 and 209 in Filtrate A + fines to that in the leachate; and in Filtrate B + ultrafines to that in Filtrate A. 88  The mass balance also shows that the two filters in series yielded a greater recovery of PBDEs than not filtering, or using a single 1.75 µm filter. The two filters together captured 33-57% of the PBDEs leached from the biosolids. This indicates that filtering the particles could lead to more efficient recovery of PBDEs in aqueous samples enriched with finite solid particles. Distribution of PBDE Congeners in the various streams The BDE congener distributions in the aqueous and solid samples were examined by calculating the ratio of each of the selected congeners to ΣPBDE. Figure 4.7 reveals that the ratios are mostly similar for the various streams (leachate, filtrate and particles). The apparent high concentration of BDE100 in Filtrate A is not explainable, except in terms of experimental error. However, the high concentration of BDE209 in the ultrafines and its low concentration in Filtrate B are consistent with the ultra-low solubility of this congener which appears to favour adsorption on the finest particulates as the leachate is filtered through the two different sizes of  %-Ratio  filters.  Figure 4.7 %-Ratio of BDE47, 85, 99, 100, 153, 154, 183 and 209 to ΣPBDE in the biosolids, leachate and particulate samples. 89  Comparison of the %-ratios for the fine and ultrafine particles shows that the initial congener distribution of the biosolids only matches reasonably that of the particles carried by the particles retained on the 1.75 µm filter. The %-ratio of BDE47 decreased from 20 to 10% from the fines to the ultrafines, and increased by a similar factor from Filtrate A to Filtrate B. Alternatively, BDE209 increased from 30 to 60% between the fines and ultrafines and was reduced by 500 times in Filtrate B from its ratio in Filtrate A.  BDE47 was found to have the highest  concentration in Filtrate B compared to all the other congeners measured (Figure 4.8), consistent with the relatively higher water solubility of this congener. Figure 4.8 confirms that the light BDE congeners such as BDE47 are more likely to partition to the aqueous phase due to their higher water solubility and lower KOW, whereas more heavily brominated congeners like BDE209 partition into finer particles, laden with higher organic content. The greater BDE47 partitioning to the aqueous phase suggests that lower PBDEs might be more readily bioavailable, eventually contaminating groundwater and exposing end-users to PBDEs when water is ingested or used.  Figure 4.8 Concentrations of BDE47, 85, 99, 100, 153, 154, 185 and 209 in Filtrate B plotted against their Log KOW values.  90  Ratios of BDE 47 to BDE 209 in all matrices are provided in Table 4.5. Ratios from the biosolids, fines, leachate and Filtrate A are similar and in the range of 0.6-0.8. However, the ratios from the ultrafine particles and Filtrate B are substantially different, 0.2 and 80, respectively. This parallel data analysis complements the discussions in previous sections of this chapter, demonstrating clearly that the BDE47 congener greatly favours partition into the aqueous phase rather than onto the particles. Ratios of BDE47 to BDE99 are also summarized in Table 4.5. These ratios are between 0.8 and 0.9 in all matrices, but the ratio from Filtrate B differs greatly, again confirming the affinity of BDE47 to the aqueous phase. Table 4.5 Ratios of BDE47 to 209 and BDE 47 to 99. Initial Biosolids Fines Ultrafines Leachate Filtrate A Filtrate B  BDE47/209 0.7 0.7 0.2 0.8 0.6 80.0  BDE47/99 0.9 0.8 0.9 0.8 0.8 1.2  4.4 Conclusions PBDEs can sorb to fine and ultrafine particles of nominal minimum sizes 1.75 and 0.45 µm suspended in leachate, thereby allowing PBDEs to be present at concentrations exceeding their solubilities in the aqueous phase.  Double filtering with different pore sizes assists in  determining the PBDE levels in the leachate. Low PBDE congeners like BDE47, due to their higher solubility relative to more highly brominated congeners, tend to partition to the aqueous phase, and thus have the potential of becoming bioavailable to living organisms. The grade of filter used when analysing PBDEs influences the measured concentrations.  91  5 Mobility of Polybrominated Diphenyl Ethers in Biosolidsamended Soil-Controlled Field Experiment 5.1 Introduction PBDEs have been extensively used in plastics, textiles, electronic circuitry and insulation material due to their flame retardancy properties since the early 1970s (Birnbaum and Staskal, 2006). There are 209 PBDE congeners named according to the number and placement of bromines attached to their molecular structure. Commercially, these flame retardants are available as technical mixtures/products dominated by BDE47 (tetraBDE), 99 (pentaBDE), 100 (pentaBDE), 153 (hexaBDE), 154 (hexaBDE), 183 (heptaBDE) and 209 (decaBDE). The PeBDE commercial mixture mostly contains BDE47 and 99, with BDE100, 153 and 154 as minor components (Sellstrom et al., 2005). After bans on the use of the less brominated congeners tetra and pentaBDEs since 2006, the DeBDE commercial mixture is the major technical product still in use today. It consists predominantly of BDE209, but also contains small amounts of octa and nonaBDEs (Sellstrom et al., 2005). PBDEs may cause liver toxicity, thyroid toxicity, and neuro-developmental toxicity (United States Environmental Protection Agency [US EPA], 2007). Environmental monitoring programs have found traces of several PBDEs in human breast milk, fish, aquatic birds and even in remote regions such as the Arctic (Hites, 2004). The similarity of PBDE molecular structure, physical and chemical properties to PCBs and dioxins, which are known to have toxic environmental effects and resistance to degradation, cause concern that these emerging contaminants have similar environmental effects (Eljarrat et al., 2008). This concern is reflected by the inclusion of PBDEs as one of the nine new persistent organic pollutants (POPs) by the Stockholm Convention in May 2009.  92  PBDEs have n-octanol-water partitioning coefficient (KOW) from 104 to 1010 showing that partitioning occurs to high organic content substances, such as biosolids or sludge produced from wastewater treatment plants (WWTPs). PBDE levels in European treated sludge and biosolids have been found to range from 3×103 to 248×103 pg/g dw (Allchin et al., 1999; Christensen et al., 2002; Matscheko et al., 2002; de Boer et al., 2003), whereas concentrations of 460×103 to 7,000×103 pg/g dw have been measured in North American sludge and biosolids (Hale et al., 2001b; Rayne and Ikonomou, 2005; Andrade et al., 2010). Studies in Europe and North America have reported PBDE levels of 580 to 840×103 (Matscheko et al., 2002; Sellstrom et al., 2003; Sellstrom et al., 2005; Eljarrat et al., 2008) and 140×103 to 7,600×103 pg/g dw (Reick, 2004; Andrade et al., 2010), respectively, in the top 0.00-0.30 m of agricultural soils that had received biosolids. These studies confirm that application of sewage sludge and biosolids onto agricultural soils is likely to be a source of PBDE contamination. Based on the earlier investigation of biosolids-amended soil at an experimental farm in Kamloops, British Columbia, Canada (Chapter 3), it is confirmed that biosolids application is a source of PBDE contamination for agricultural soils. PBDEs were found to migrate downwards to at least 1 m over a period of 18 months, and varied widely and inconsistently from sample to sample making it impossible to establish definite conclusions regarding the dependencies of PBDE on various factors. Hence, as described in this chapter, a more systematic and controlled field experiment was undertaken to address these issues and provide more definitive results with respect to the mobility of PBDEs in biosolids-amended soils. To the best of our knowledge, this is the first controlled field experiment monitoring PBDEs in agricultural soils at depths greater than 0.30 m, with the impact of environmental factors either limited or controlled. The objectives of the study were: (i) to investigate the overall concentration and congener profiles of PBDEs in agricultural 93  soils onto which biosolids had been applied; (ii) to determine the effects of key factors on PBDEs; and (iii) to find the mobility of PBDEs in the biosolids-amended soil over a significant depth over a one-year period. The results are intended to provide reference information for evaluating exposure pathways for human and ecological risk assessment from PBDEs deposited on soils, especially those from biosolids. This study also allows the examination of PBDE distribution over depth and time through several soil sampling events, covering 0.85 m deep soil after the application of biosolids. Sampling three times provides more data to determine specific trends for PBDE mobilization and distribution, if any, as opposed to only a single sampling event in Kamloops (Chapter 3).  5.2 Material and Methods The field, referred to hereafter as Totem Field at the University of British Columbia, receives an average rainfall of 1400 mm/year and an average snowfall of 90 mm/year. Average temperatures in winter 2007 and summer 2006 and 2007 are 5 and 20°C, respectively. Air temperature and precipitation as rainfall and snowfall were monitored by an on-site meteorological station. The biosolids were from a WWTP in British Columbia where they were dewatered and underwent thermophilic and aerobic sludge digestion. These biosolids were applied on a 5×5 m area divided into 1×1 m cells (Figure 5.1). Prior to applying the biosolids, the area was cleared of vegetation and graded to ensure that the surface was flat and horizontal to minimize surface runoff and cross-contamination. Biosolids were applied to the area at a loading of 80 t/ha in August 2006. This loading was chosen to be consistent with the loading at Kamloops field (Chapter 3). To ensure uniform application of the biosolids on all cells, each cell was divided into four quadrants with each quadrant receiving 2 kg of biosolids. The biosolids were spread  94  uniformly over each quadrant and mixed with the top 0.05 m of soil. The area was then fenced to prevent access to the biosolids-amended area and to minimize cross-contamination.  5m  5m  C  B  A  A  C  B  B  A  C  1m 1m Figure 5.1 Totem Field test area arrangement where 80 t/ha of biosolids were applied and the time of sampling of each cell. A: Samples collected December 2006; B: Samples collected April 2007; C: Samples collected August 2007. Outer grey shaded cells were not sampled, but provided a buffer.  5.2.1 Soil Sampling Table 5.1 summarizes the sampling program. Soil samples (referred to as reference samples) were collected from an area which had not received biosolids and from the inner nine 1x1 m cells in the area where the biosolids were applied (Figure 5.1). The outer cells acted as buffer areas and were not sampled. Samples were collected at four-month intervals over one year (120 days, 240 days and 360 days), with three different cells sampled on each occasion (Figure 5.1). Sampling involved excavating an open pit and sampling from 5 layers: 0.00-0.05 m; 0.05-0.25 m; 0.25-0.45 m; 0.45-65 m; and 0.65-0.85 m below the soil surface. For each layer, 0.02-0.03 m of the vertical surface was scraped off using a trowel. Using a different trowel, a soil cube covering the entire layer was cut-off and placed in a hexane-washed amber jar. The samples 95  were taken from bottom to top, cleaning the trowels with distilled water and acetone before sampling each layer. The samples were then stored at -30ºC until analysis. Table 5.1 Details of the sampling program. Site  Cells Sampled  Sampling date  Sampling conditions  Sampling depths (m)  -  Biosolids Loading (t/ha) -  Reference soil  August 2006  Ploughed soil  Biosolidsamended soil  A  80  November 2006  Weed plants at ground level.  Biosolidsamended soil  B  80  March 2007  Biosolidsamended soil  C  80  July 2007  Weeded plants approximately 0.30 m in height Weeded plants approximately 0.60 m in height  0.05-0.25; 0.25-0.45; 0.45-0.65; 0.65-0.85 and 0.85-1.05 0.05-0.25; 0.25-0.45; 0.45-0.65; 0.65-0.85 and 0.85-1.05 0.05-0.25; 0.25-0.45; 0.45-0.65; 0.65-0.85 and 0.85-1.05 0.05-0.25; 0.25-0.45; 0.45-0.65; 0.65-0.85 and 0.85-1.05  5.2.2 Physical and Chemical Characterization  The grain size distribution for the soil samples was determined using laser refraction (Mastersizer 2000, Malvern Instruments, United Kingdom). This distribution is used to estimate the hydraulic conductivity, k (cm/s) of the soil based on the Slitchers equation (Odong 2007):  k=  g 2 #1#10$2 n 3.287d10 "  " = 0.255(1+ 0.83U )  (Equation 5.1) (Equation 5.2)  !  !  where g = acceleration due to gravity (m2/s); v = absolute viscosity (kg/s·m); θ=soil porosity; d10= grain diameter in (cm) for which 10% of the sample grains are finer than; 10-2 is a conversion factor from m to cm; and U is the coefficient of grain uniformity determined by the Mastersizer2000 as the ratio of d60 to d10. The moisture content, CEC and OM of the biosolids and soil samples were determined based on the procedures described in Chapter 3. Briefly, moisture content was found by drying soil 96  samples at 100±5°C to a constant weight, and the difference between the initial and final weights of each sample determined the moisture content. CEC was determined by the modified sodium acetate replacement method described in Appendix A, where the concentrations of sodium was measured by an atomic adsorption spectrophotometer, Thermo Jarell Ash Video 22 (Thermo fisher Scientific Inc.). The organic matter content was analyzed by loss on ignition in a muffle furnace for 2 h at 550°C. 5.2.3 Sample Clean-up and Analysis Sample clean-up and analysis followed the procedure recommended by Ikonomou et al. (2001). The biosolids and soil samples were processed in batches of 12, each batch consisting of a procedural blank, a certified reference sample, and nine samples, one of which was in duplicate. 5 g of biosolids or 10 g of soil were homogenized with Na2SO4 using a mortar and pestle to give a mass ratio of 1.5 parts of Na2SO4 to 1 part sample. The homogenized samples were then transferred to 33 mL stainless steel cells and spiked with 25 µL of the EO5100 standard, a suite of  13  C-labeled PBDEs, and  13  C-labeled BDE209 (Cambridge Isotope Laboratories, Andover,  Massachusetts). The cells were then placed in a Dionex ASE 2000 accelerated solvent extraction system for PBDE extraction. A 80:20 (v/v) toluene:acetone solution constituted the extraction solvent. The sample extracts underwent cleanup using a three-step procedure and analyzed HRGC/HRMS (Ikonomou et al., 2001). In the first step, extracts were passed through a multilayer silica column packed with successive layers of silica gel (basic, neutral, acidic, neutral) and eluted with DCM-hexane (1:1 v/v). In the second, a column filled with copper filings and Na2SO4 removed sulphur and residual water, with samples again being eluted with 1:1 v/v DCM-hexane. The third step involved a neutral activated alumina column capped with anhydrous Na2SO4 and washed with hexane, followed by addition of DCM-hexane (1:1 v/v) to 97  elute the analytes of interest. Eluants from the alumina column were concentrated to <10 µL and spiked with 5 µL of a suite of 13C-labeled PBDEs, (Cambridge Isotope Laboratories, Andover, Massachusetts)  prior  to  congener-specific  PBDE  analyses  by  high-resolution  gas  chromatography high-resolution mass spectrometry (HRGC/HRMS). 5.2.4 Data Analysis 51 of the 209 PBDE congeners were analyzed in this study. To account for background contamination congener concentrations measured in the procedural blanks were subtracted from the measured concentrations. All corrected congener concentrations were added to provide a total concentration, denoted as ΣPBDE. To compare the congener distributions in the biosolids and soil samples, the concentrations of the congeners with the same number of bromines (i.e., of the same homologue group) were added. Error bars indicate 95% confidence limits, obtained from repeated measurements (n=3) on the same samples. Concentrations of the individual PBDE congeners are presented in Appendix C. Of the 51 measured congeners, BDE47, 85, 99, 100, 153, 154, 183 and 209 were most abundant, were detected in almost all samples have, and been reported in most previously published studies. Hence, we focus on these congeners in our discussions below. The degree of PBDE contamination in biosolids-amended soils relative to reference soils is characterized below by the ratio of PBDE concentration in the biosolids-amended soil (S) to that in the reference soils (R) (S/R).  5.3 Results and Discussion 5.3.1 Climate Conditions Meteorological conditions at Totem Field during August 2006-July 2007 are summarized in Table 5.2. The average temperature during the experimental period was 10.1°C, and ranged from 98  -9.2 to 30.5°C. Total precipitation during this period was 1,511 mm, 1,424 mm as rainfall and 87 mm as snowfall. Table 5.2 Temperature and precipitation at Totem Field during August 2006-August 2007. Average Minimum Temperature Temperature (oC) (oC) August, 2006 17.0 10.1 September, 2006 14.9 8.4 October, 2006 10.1 -1.2 November, 2006 5.5 -9.2 August-November, 2006 11.9 -9.2 December, 2006 4.6 -1.6 January, 2007 3.1 -6.0 February, 2007 5.7 -1.3 March, 2007 7.1 -1.9 December, 2006-March, 2007 5.1 -6.0 April, 2007 8.7 0.9 May, 2007 12.3 4.4 June, 2007 14.5 8.1 July, 2007 18.3 10.8 April-July, 2007 13.4 0.9 August, 2006-July, 2007 10.1 -9.2 *Total snowfall is represented as snow water equivalent. Month  Maximum Temperature (oC) 25.1 26.1 16.1 14.7 26.1 11.2 11.3 12.3 15.8 15.8 16.6 23.0 24.0 30.5 30.5 30.5  Total Rainfall (mm) 7 51 72 328 459 165 194 122 214 696 77 56 79 58 270 1,424  Total Snowfall* (mm) 0 0 0 43 43 30 14 0 0 44 0 0 0 0 0 87  5.3.2 Biosolids and Soil Properties The moisture content, CEC and OM of the biosolids and reference soils are summarized in Table 5.3. The moisture content of the biosolids applied in Totem Field was approximately 69% (w/w), and had OM of 72% w/w (dry weight basis). The CEC of the biosolids was approximately 55 meq/100 g of biosolids. The soil at Totem Field is characterized as sandy loam soil with low clay content (Vellend et al., 2009). Table 5.3 presents the moisture content, OM and CEC of the reference soil; these were 13-16%, 2-8% and 1-7 meq/100 g, respectively, considerably lower than in the biosolids. Table 5.3 also shows that the moisture content, OM and CEC of the biosolids-amended soils were 1326%, 1-12% and 2-59 meq/100 g, respectively.  99  Table 5.3 Moisture content, cation exchange capacity and organic matter content of biosolids, reference and biosolids-amended soils Moisture CEC(2) Content(1) OM(4) (%) (meq/100g)(3) (%) Biosolids 69 55.1 72 Totem Field Reference Soils 0.00-0.05 15 7 8 0.05-0.25 16 1 6 0.25-0.45 13 2 3 0.45-0.65 14 3 2 0.65-0.85 15 3 3 Biosolids-amended Soils August-November, 2006 (1-4 Months) 0.00-0.05 26 59 12 0.05-0.25 20 26 11 0.25-0.45 17 2 8 0.45-0.65 17 3 1 0.65-0.85 17 4 2 December, 2006-March, 2007 (4-8 Months) 0.00-0.05 20 51 11 0.05-0.25 17 34 10 0.25-0.45 14 3 6 0.45-0.65 17 4 1 0.65-0.85 17 2 2 April-July, 2007 (8-12 Months) 0.00-0.05 15 43 10 0.05-0.25 16 34 9 0.25-0.45 13 4 5 0.45-0.65 14 4 1 0.65-0.85 15 3 2 (1) Moisture content based on weight (2) Cation Exchange Capacity (3) Milliequivalents per 100 g. Na concentrations used to determine CEC are presented in Appendix C (4) Organic Matter content based on weight Depth below surface (m)  The moisture content in the 0.00-0.05 and 0.05-0.25 m intervals changed directly with rainfall precipitation. The moisture content of these layers were highest in the first four months, when rainfall precipitation was highest, and decreased in the precipitation decreased as the later 8 months of the experiment. Moisture contents of the rest of the soil in the 0.25-0.85 m interval did not change according to the rainfall precipitation. The OM and CEC changes did not seem to respond to the climatic conditions. 100  Comparison of the reference soil and the biosolids-amended soil properties shows that the addition of biosolids changed the physical and chemical properties of the soil. Over a one-year period, CEC and OM increased approximately 8-26 and 1.5-2 times, respectively, in the top 0.25 m of the soil, and changed slightly in the 0.25-0.85 m interval. These results are similar to the changes in the CEC and OM of the Kamloops agricultural biosolids-amended soil (Chapter 3), which also increased, across a 0.05-1.05 m interval, by 2-6 times. Mixing biosolids with the surface soil increased both CEC and OM. These changes also affected the underlying soil, as the water infiltrates the soil mobilizing organic matter and minerals in the top biosolids-amended soil, and transporting them to greater depths. The increases in OM and CEC with depth are probably due to the nitrification of organic nitrogen and ammonia added with the biosolids, production of hydrogen cation displacing cations from the soil CEC and mobility of excess nitrate anions generated by nitrification as mobilizing anions for the CEC cations (Harrison et al., 1994). The permeability, k, of the soil was estimated from the size distribution of the soil, and specifically the d10 (10% finer by weight) grain sizes summarized in Table 5.4. Permeability provides an indication of how PBDEs can be transported downwards, as water infiltrating through the soil will also carry suspended matter which has been associated with the transport of PBDEs (Chapter 4). The permeability of the soil was between 3.1×10-3 and 1.4×10-2 m/s. The lowest permeability was calculated at a depth 0.65-0.85 m below the surface, and the highest in the 0.25-0.45 m depth interval.  101  Table 5.4 Permeability (k) of Totem Field soil. Depth (m) 0.00-0.05 0.05-0.25 0.25-0.45 0.45-0.65 0.65-0.85  Number of Samples 3 3 3 3 3  Coefficient of Uniformity Standard Average Deviation 2.98 0.42 2.96 0.53 1.85 0.73 4.00 0.64 4.75 1.91  d10 Average (mm) 8.9E-03 7.9E-03 1.1E-02 5.3E-03 5.1E-03  Standard Deviation 5.2E-04 1.2E-03 1.6E-03 1.1E-03 1.1E-03  k(m/s) 9.2×10-3 7.2×10-3 1.4×10-2 3.2×10-3 3.1×10-3  d10: 10% finer by weight 5.3.3 PBDE Concentration in Reference Soils PBDE concentrations are expected to be negligible or very low in reference soils. Mean concentrations of individual PBDEs in the reference soil are provided in Figure 5.2. BDE85, 100, 153, 154, 181 and 183 could not be detected in the reference soil. However, BDE 47 was present in the 0.05-0.25 m and 0.65-0.85 m layers at concentrations of 21 and 6.1 pg/g dw, respectively. The BDE 99 concentrations were 20 and 14 pg/g dw in the top 0.05 m layer and 0.05-0.25 m layer, respectively. BDE 209 was only found in the top 0.05 m layer at 260 pg/g dw. The concentration of the eight principal congeners ranged between non-detectable (nd) and 260 pg/g dw. This is consistent with previous studies on European reference and background soils where BDE209 had the highest congener concentration (Hassanin et al., 2004; Sellstrom et al., 2005). However, BDE concentration in reference soils from our study are several orders of magnitude lower than for a Swedish agricultural reference soil (Sellstrom et al., 2005).  102  Figure 5.2 PBDE congener concentrations in Totem Field reference soils at different depths. Where there is no bar, the measurement level was non-detectable.  5.3.4 PBDE Concentrations in Biosolids Concentrations of individual BDE congeners in the biosolids applied are in Table 5.5. The ΣPBDE levels in the biosolids are 2.8 to 3.2 times higher than reported in European biosolids (Christensen et al., 2002; de Boer et al., 2003; Eljarrat et al., 2008), but mainly lower than previously measured in North American biosolids (Hale et al., 2001a; Hale et al., 2002; Hale et al., 2003; Reick, 2004; Ciparis and Hale, 2005; Anderson and MacRae, 2006) by up to nearly three orders of magnitude. This is due to a higher North American demand which was estimated to be 31,000 tonnes of PBDE commercial mixtures as opposed to a European demand of 25,980 tonnes in 1999 (BSEF 2004). The major contributors to the ΣPBDE in the biosolids were BDE47, 99 and 209, which are major congeners incorporated in PeBDE, OcBDe and DeBDE 103  commercial mixtures and abundant in domestic wastewater treated by WWTPs (North, 2004; Rayne and Ikonomou, 2005). Previous studies (de Boer and Cofino, 2002; Oberg et al., 2002; de Boer et al., 2003; European Commission, 2001; European Commission; 2004) have reported that these congeners are released to wastewater through the disposal of wash-water containing PBDEs in dust from washing of PBDE-containing textiles, such as upholstery fabrics, curtains and tent materials. Table 5.5 PBDE concentrations in (pg/g dry weight) in biosolids (derived from sewage sludge). Studies Current Study Hale et al. 2002 Hale et al., 2003 North 2004 Reick 2004 Ciparis & Hale 2005 Anderson & MacRae, 2006  47 347×103 544×103 637×103 757×103 1.0×106 400×103  99 425×103 725×103 1.0×103 944×103 920×103 600×103  100 85×103 266×103 165×103 194×103 100×103  153 40×103 88×103 121×103 60×103  154 35×103 -  183 13×103 -  209 509×106 1.5×106 2.7×107 1.2×106 48×103 350×103  Total 510×106 1.5×106 2.1×106 -  457×103  2.1×106  319×103  102×103  -  -  -  -  deBoer et al. 2003 Christensen et al. 2004 Eljarrat et al., 2008  20×103  21×103  -  5.0×103  -  -  99×103  486×103  97×103  86×103  19×103  8.0×103  -  -  248×103  -  27×103  32×103  7.0×103  4.0×103  3.0×103  9.0×103  484×103  570×103  “-” denotes concentration not presented in the study.  5.3.5 PBDEs in Biosolids-Amended Soils Individual PBDE congeners in the biosolids-amended soil changed with time for the 0.00-0.045 m interval. The following changes of PBDE concentrations occurred 4, 8 and 12 months after the biosolids application event: •  In the 0.00-0.05 m layer (Figure 5.3a), the selected PBDE concentrations were 5×103162×103 pg/g dw, 3×103-130×103 pg/g dw and 230-10×103 pg/g dw at 4, 8 and 12 months, respectively, decreasing with time and with precipitation (620, 608 and 193 mm, respectively);  104  •  In the 0.05-0.25 m layer (Figure 5.3b) the selected PBDEs concentrations were n.d.-50 pg/g dw, 6-150 pg/g dw and 22-875 pg/g dw, respectively, thus increasing with time but are inversely related to precipitation; and  •  In the 0.25-0.45 m layer (Figure 5.3c) the selected PBDEs concentrations were almost nd, nd-1×103 pg/g dw and nd-600 pg/g, respectively, thus fluctuating.  Figure 5.4 plots the ratio S/R where S is PBDEs in the biosolids-amended soil and R is PBDEs in the reference soils. The S/R ratios greatly exceeded 1 (500-50,000) in the 0-0.05 m layer, due to inter-mixing of biosolids and soil. Due to non-detectable concentrations of PBDEs in most reference soil layers (0.25-0.85 m), many S/R ratios could not be calculated. The ratios were lower at 8 months and 1 year after application of biosolids than after 4 months. This could be due to PBDEs migrating downwards with the infiltrating water, escaping to the atmosphere and/or degrading. A combination of these possibilities is likely, as the S/R ratios in the underlying 0.05-0.25 m and 0.25-0.45 m exceeded 1 for most BDE congeners. The decrease in S/R of BDE47, 85 and 99 in the 0-0.05 m depth and their increase in the lower depths over time suggests the photodegradation and/or volatilization at the 0-0.05 m depth. Lower brominated congeners such as BDE 47, 85 and 99 are more likely to undergo volatilization than higher brominated congeners such as BDE209 (Gouin et al., 2005). The lower brominated congeners have lower octanol-air partitioning coefficient (KOA) than the higher brominated ones, an indication that they are more susceptible to volatilization (Gouin et al., 2005). Photodegradation primarily degrades BDE 209, resulting in nona- and octa- brominated congeners (Eriksson et al., 2004). The effect of photodegradation can be seen in the last four months of the experiment, where, sunlight exposure periods during the summer months are longer, resulting in a greater decrease in PBDE concentration than in the earlier 8 months (Figure 5.3a and Figure 5.4). 105  (b) 0.05-0.25 m  Concentration (pg/g dw)  (a) 0.00-0.05 m  Concentration (pg/g dw)  (c) 0.25-0.45 m  Figure 5.3 PBDE concentration in biosolids-amended soil from Totem Field at depths of (a) 0.00-0.05 m, (b) 0.05-0.25 m and (c) 0.25-0.45 m. Cumulative concentrations for 0.45-0.65 and 0.65-0.85 m depths intervals were negligible. Error bars indicate 95% confidence limits obtained from repeated measurements (n=3). Note that the concentration axes do not have the same scales.  The cumulative PBDE concentration across a depth of 0.85 m in the biosolids-amended soil is plotted in Figure 5.5. ΣPBDE levels in the 0.00-0.05 m layer increased from approximately 320 to 600×103 pg/g dw after application of biosolids. In the 0.00-0.05 m layers, ΣPBDE decreased from 600×103 to 30×103 pg/g dw over one year after applying the biosolids (Figure 5.5a). For both the 0.05-0.25 m and 0.25-0.45 m intervals, ΣPBDE increased exponentially from 70 to 2.5 ×103 pg/g dw and 8 to 1×103 pg/g dw, respectively, over the year following the application of 80 t/ha biosolids (Figure 5.5b and 5.5c). ΣPBDE was negligible in the 0.45-0.65 and 0.65-0.85 m layers.  106  Figure 5.4 Ratio of PBDE concentration biosolids-amended soil (S) and reference soil (R) in Totem Field biosolids-amended soil between 0.05 m and 0.85 m depth.  The concentrations of tetra, penta, hexa, hepta, octa and decaBDE in each layer are plotted vs. time in Figure 5.6 whereas ΣPBDE is plotted in Figure 5.7. Results are only provided for the 0.00-0.05 m, 0.05-0.25 m and 0.25-0.45 m layers, since the remaining layers had negligible PBDE levels. The exponential functions are described in Table 5.6 and have R2 values between 0.45 and 0.99.  107  (a) 0.00-0.05 m  (b) 0.05-0.25 m  (c) 0.25-0.45 m  Figure 5.5 Cumulative PBDE concentration (pg/g dw) in Totem Field biosolids-amended Soils between August 2006 and July 2007 (8/07) in (a) 0.00-0.05 m, (b) 0.05-0.25 m and (c) 0.25-0.45 m. Note that cumulative concentrations in 0.45-0.65 and 0.65-0.85 m were negligible; error bars cannot be seen due to logarithmic scale used for the vertical axis.  108  Concentration (pg/g dw)  (a) TetraBDE  (b) PentaBDE  0.0091x  y = 0.66 2 R = 0.998  0.0407x  y = 2E-05 2 R = 0.92  Concentration (pg/g dw)  (c) HexaBDE  (d) HeptaBDE  0.0418x  y = 2E-05 2 R = 0.85 0.0375x  y = 2E-05 2 R = 0.75  Concentration (pg/g dw)  (e) OctaBDE  (f) DecaBDE  0.0452x  y = 7E-07 2 R = 0.75  0.0568x  y = 1E-05 2 R = 0.74  Time (Days)  Time (Days)  Figure 5.6 Change of (a) tetra, (b)penta, (c) hexa, (d) hepta, (e) octa and (f) decaBDE concentration (pg/g dw) in each soil layer with time (days). See Table 5.6 for exponential functions.  109  Figures 5.6 and 5.7 show that PBDEs in the surface soil decay exponentially, agreeing with Andrade et al. (2010) who also investigated PBDEs in the surface layer of biosolids-amended soils. The figures show that the rate of decrease of PBDEs in the 0.00-0.05 m interval after applying biosolids to the Totem Field was approximately 500-600 times greater than the rates of accumulation in the 0.05-0.25 m and 0.025-0.45 m layer. In order to account for the difference of volumes due to the varying depth of each layer, the rates were multiplied by 1 m2 area, depth (0.05 m for the top layer, and 0.20 m for each of the two underlying layers), and density of soil (assumed to be 1.6 g/cm3), resulting in an average accumulation rate. The average mass rates of change of ΣPBDE in the 0.00-0.05 m, 0.05-0.25 m and 0.25-0.45 m layers was 2.9×108, 3.9×106 and 4.0 ×106 pg/m2/day, respectively between the period of 120 and 360 days after the biosolids application. The magnitude of the average accumulation rate of change in the 0.00-0.5 m layer is almost 2 orders of magnitude greater than for the 0.05-0.25 m and 0.025-0.45 m layer, while the latter two layers had similar average accumulation rate. This suggests that PBDEs are subjected to similar mechanisms in the 0.050.25 and 0.25-0.45 m layers relative to 0.0-0.05 m surface layer. In addition to the infiltrating water, PBDEs can be subject to biodegradation, photodegradation, volatilization and downward transport by the infiltrating water, to microbes, sunlight and temperature variation (Andrade et al., 2010). The 0.00-0.05 m layer soil is also exposed to atmospheric factors such as sunlight, wind, rain and temperature variations, which influence the fate PBDEs in the soil. These factors have less influence on PBDEs in the soil beneath the 0.000.05 m layer.  The high mass rate of change in the surface soil layer indicates PBDE  volatilization and photodegradation. Gouin et al. (2005) showed that PBDEs, especially the lower brominated ones, could volatilize, given their KOA coefficients. Eriksson et al. (2004) reported that PBDEs in surface soils are susceptible to photodegradation. 110  Compared to the preliminary study on Kamloops biosolids-amended soil (Chapter 3), average PBDE concentrations in the biosolids-amended soil in Totem field were 1-3 orders of magnitude lower. In addition, PBDEs were found at shallower depths in Totem Field (0.45 m vs. 1.05 m in Kamloops). Several factors have led to the different degrees of PBDE contamination. First, the soil properties, OM content and permeability differ. These two properties can directly affect the rate of transport of the contaminants. OM can retard the transport of contaminants, especially species with high KOW, such as PBDEs, since they will have greater affinity to be sorbed on the soil rather than desorbing and partitioning into the infiltrating water. Totem soil had a higher OM content than Kamloops soil, with average OM values of 4.4 and 3.25 respectively. Soil permeability affects the rate at which water infiltrates through the soil, and hence the rate of mobilization of PBDEs. Higher permeability leads to higher rates of infiltration. Totem soils are less permeable, with an average k of 7.4×10-3 m/s about half of that of the Kamloops soils, which have an average k of 1.4×10-2 m/s over the 0.85 m depth. It is also notable that the final sampling in Kamloops occurred 14-18 months after biosolids application, whereas the final sampling event at Totem field was 12 months after the application of biosolids. Hence, there was also more time for PBDEs to migrate downwards in the Kamloops biosolids-amended soil. Although, the results have shown that PBDEs can vary with time in an approximately exponential decay manner, there is considerable spatial variability of PBDE concentration in the biosolids-amended soils. This variability may occur because the biosolids consist of many particles of variable composition (Andrade et al., 2010). It is likely that the biosolids remain bound together for some time before becoming fully available for exchange with soil particles. This assimilation time from one media to another may be a critical factor controlling the fate and availability of PBDEs for mobilization and degradation processes. The diffusion of the chemical species from the interior of biosolids chunks to the surface may also be a contributing factor for availability for photodegradation and volatilization (Andrade et al., 2010). Microbial 111  biodegradation can be another factor, however, further investigations in their microbial degradation in soil environments are required since most investigations have only focused on controlled and isolated cultures (Gerecke et al., 2001; He et al., 2006; Robrock et al., 2008; Robrock et al., 2009). Uptake by plants and biota can also affect the PBDE distribution and fate  Total PBDE Concentration (pg/g dw)  in agricultural soils (Sellstrom et al., 2005; Mueller et al., 2006).  Time (Days)  Figure 5.7 Change of total PBDE concentration (pg/g dw) in each soil layer with time (days).  Table 5.6 Exponential functions describing tetra, penta, hexa, hepta, octa, decaBDE and total PBDE concentration as a function of time. Homologue Group 0.0-0.05 m TetraBDE PentaBDE HexaBDE HeptaBDE OctaBDE DecaBDE Total  Exponential Function y = 933779e-0.0128x y = 1.0×106 e-0.0129x y = 2.0×105 e-0.0127x y = 1.3×105 e-0.013x y = 1.6×105 e-0.0132x y = 1.6×105 e-0.0132x y = 4×106 e-0.0124x  Coefficient of Correlation (R2) 0.85 0.86 0.86 0.87 0.85 0.85 0.84 112  Table 5.6 (continued) Exponential functions describing tetra, penta, hexa, hepta, octa, decaBDE and total PBDE concentration as a function of time. Homologue Group Exponential Function Coefficient of Correlation (R2) 0.05-0.25 m TetraBDE y = 1.6e0.0161x 0.99 0.0171x PentaBDE y = 1.5e 0.98 HexaBDE y = 0.62e0.0138x 0.998 HeptaBDE y = 2×10-5 e0.0418x 0.84 OctaBDE y = 1×105 e0.05x 0.81 5 0.05x DecaBDE y = 1×10 e 0.81 0.0147x Total y = 14.9e 0.97 0.25-0.45 m TetraBDE y = 0.66e0.0091x 0.998 PentaBDE y = 2.0×10-5 e0.0407x 0.92 -0.0002x HexaBDE y = 1.08e 0.45 HeptaBDE y = 2.0×10-6 e0.0375x 0.75 -7 0.0452x OctaBDE y = 7.0×10 e 0.75 -5 0.0568x DecaBDE y = 1.0×10 e 0.74 Total y = 0.44e0.0245x 0.76 y: PBDE concentration in the soil (pg/g dw) x: Time in days  5.4 Conclusions Biosolids applied in a controlled manner on agricultural soil in Vancouver were found to contain PBDEs ranging from 35×103 to 509×106 pg/g dw. BDE 47, 99 and 209 were the predominant congeners likely reflecting the commercial PeBDE and DeBDE products. The presence of PBDEs in the biosolids led to their downward migration and increasing concentrations below the surface of the agricultural soil. Soil-to-reference concentration ratios of the different BDE congeners ranged from 500 to 50,000 in the top 0.00-0.05 m depth intervals, with an increase of total PBDEs from approximately 320 to 600×103 pg/g dw. Over a one-year period, the concentration the surface layer fell to 30×103 pg/g dw at a non-uniform rate. For the 0.05-0.25 and 0.25-0.45 m depths layers, total PBDEs  113  increased to 2.5 ×103 pg/g dw and 1×103 pg/g dw, respectively, over the year indicating downward migration, whereas PBDEs were negligible in the 0.45-0.65 and 0.65-0.85 m layers. Local PBDE concentrations increased or decreased approximately exponentially with time in the soil. The average mass rate of decrease of PBDEs in the top 0.00-0.05 m was almost 2 orders of magnitude greater than the mass accumulation rate in the 0.05-0.25 and 0.25-0.45 m layers. This substantial difference suggests that volatilization and photodegradation of PBDEs may be important in the surface soil.  114  6 PBDE Mobility in Biosolids-Amended Soils using Leaching Column Tests3 6.1 Introduction Polybrominated diphenyl ethers (PBDEs) constitute a family of brominated flame retardant chemicals used in various manufactured products, including foam cushions, plastics within televisions and computers, and surface coatings. There are 209 congeners divided into 10 groups (mono- to deca-bromodiphenyl ethers). However, compounds with fewer than four bromine atoms are rarely found in commercial PBDE products (Darnerud et al. 2001). Congeners are numbered according to the International Union of Pure and Applied Chemistry (IUPAC) system originally designed for polychlorinated biphenyls (PCBs). The penta-brominated diphenyl ether commercial mixture (PeBDE) contains mostly BDE47 and 99 plus traces of BDE100, 153 and 154 (Sellstrom et al., 2005; Sjodin et al., 1998). The octa-brominated diphenyl ether commercial mixture (OcBDE) contains primarily BDE183 (>40%) and other octa-BDE congeners (>30%) (La Guardia et al., 2006). The deca-brominated (DeBDE) commercial product consists of BDE209, as well as minute quantities of octa- and nona-BDE congeners (La Guardia et al., 2006). PBDEs are lipophilic, with n-octanol-water partition coefficient, log (KOW), ranging from 4 to 10, resulting in their accumulation in high-organic-content media (Palm et al., 2002; D'Silva et al., 2004; Gouin and Harner, 2003). They have low water solubility, especially for the higherbrominated congeners (hepta- and above), with solubilities from 1.11×10-8 to 0.09 mg/L (Palm et al., 2002). The PBDE chemical and physical properties are generally similar to those for other halogenated hydrophobic compounds such as PCBs and dioxins, which have low solubilities 3  A version of Chapter 6 has been accepted for publication in Water Air and Soil Pollution Journal on March 31st, 2011. 115  ranging from 0.1 to 0.6 mg/L (Agency for Toxic Substances and Disease Registry [ATSDR], 2000) and 4.75×10-3 and 0.4 mg/L (ATSDR, 1998), respectively. PBDEs, especially the most hydrophobic congeners (hepta- and above), adsorb onto sediments at in lakes and rivers, acting as PBDE reservoirs (ATSDR, 2004). Lower-brominated PBDEs, especially tetra- and penta-brominated are known to accumulate in fish at low concentrations (1×10-8 to 1×10-6 grams of PBDE per gram of fish tissue) (ATSDR, 2004). However, higherbrominated congeners, such as deca-BDE, in general, are not found in fish at measurable concentrations (ATSDR, 2004). PBDEs also bioaccumulate in humans and wildlife, and in some ecosystems their concentrations have been increasing exponentially with time over 5 to 10 years (Ikonomou et al., 2002; Hites, 2004). In general, PBDEs break down very slowly in soil and sediments as they sequester on small organic particles where they are known to remain many years (ATSDR, 2004). One of the routes by which PBDEs are discharged to the environment is through biosolids, which result from processing sludge from wastewater treatment plants (WWTP). PBDE levels in European biosolids have been determined to range from 3×103 to 1,185 ×103 pg/g dw basis (Allchin et al., 1999; Christensen et al., 2002; Matscheko et al., 2002; de Boer et al., 2003; Eljarrat et al., 2008), whereas North American sludges have been found to contain 460×103 to 7,000×103 pg/g dw (Hale et al., 2001; Rayne and Ikonomou, 2005). It is common practice in Europe and North America to dispose of biosolids onto agricultural lands. Given the above concentrations, applying biosolids onto agricultural soils is a potential source of PBDE contamination. Swedish and North American studies have reported PBDE levels of 580 to 840×103 (Matscheko et al., 2002; Sellstrom et al., 2003; Sellstrom et al., 2005; Eljarrat et al., 2008) and 140×103 to 7,600×103 pg/g dw (Reick, 2004), respectively, in biosolids-amended agricultural soils. 116  The few field studies which have investigated PBDE levels in biosolids-amended agricultural soils have focused on the degree of contamination in the top 0.30 m of the soil compared to soils which received no bi PBDEs, especially the most hydrophobic congeners (hepta- and above), adsorb onto sediments at the bottom of bodies of water, such as lakes and rivers, acting as PBDE reservoirs (ATSDR, 2004). Lower-brominated PBDEs (tetra- and penta-brominated congeners in particular) accumulate in fish at low concentrations (1×10-8 to 1×10-6 g PBDE /g of fish tissue) (ATSDR, 2004). However, higher-brominated congeners, such as deca-BDE, in general, are not found in fish at measurable concentrations (ATSDR, 2004). PBDEs also bioaccumulate in humans and wildlife. In some ecosystems their concentrations have increased exponentially over 5 to 10 years ( Ikonomou et al., 2002; Hites, 2004). In general, PBDEs break down very slowly in soil and sediments as they sequester to small organic particles where they remain for many years (ATSDR, 2004). One route by which PBDEs are discharged to the environment is by biosolids (i.e. the treated sludge) from wastewater treatment plants (WWTP). PBDE levels in European biosolids have been determined to range from 3×103 to 1,185 ×103 pg/g dry weight (dw) basis (Allchin et al., 1999; Matscheko et al., 2002; Christensen et al., 2002; de Boer et al., 2003; Eljarrat et al., 2008), whereas North American sludges have been found to contain 460×103 to 7,000×103 pg/g dw (Hale et al., 2001; Rayne and Ikonomou, 2005). It is common in Europe and North America to dispose of biosolids onto agricultural lands. Given the above concentrations, applying biosolids onto agricultural soils is a potential source of PBDE contamination.  Swedish and North  American studies have reported PBDE levels of 580 to 840×103 (Eljarrat et al., 2008; Matscheko et al., 2002; Sellstrom et al., 2003; Sellstrom et al., 2005) and 140×103 to 7,600×103 pg/g dw (Reick, 2004), respectively, in biosolids-amended agricultural soils.  117  The few field studies which have investigated PBDEs levels in biosolids-amended agricultural soils have focused on the degree of contamination in the top 0.30 m of the soil compared to soils which received no biosolids (Matscheko et al., 2002; Sellstrom et al., 2005). A study (Eljarrat et al., 2008) on the fate of PBDEs (tetra- to deca-BDE) in surface soils (top 0.15-0.20 m), four years after applying biosolids, concluded that they were persistent. Studies on hydrophobic compounds, similar in physiochemical properties to PBDEs, have shown that they can sorb on particulate matter suspended in the aqueous phase and become mobilized in the soil phase (McCarthy and Zachara, 1989). This can occur in soils containing macropores due to wormholes and passages created by withered roots. It has also been suggested (Jarvis et al., 1999; Kretzschmar et al., 1999) that suspended matter in percolating water can reach groundwater by passing through these macropores in the unsaturated zone. Mobilization of PBDEs in soil can lead to suspension of particles containing PBDEs and passage of dissolved PBDEs into groundwater, potentially making them bio-available to humans and wildlife. Gorgy et al. (2010), by measuring PBDE concentrations in leachate resulting from water percolating through biosolids, demonstrated that PBDEs were mobilized from the biosolids by the percolated water and partitioned between the aqueous layer and the particles suspended in the leachate. This suggests that PBDEs could also be transported in soils. However, there are no known studies that have comprehensively investigated PBDE transport in soils, their concentration distribution at different depths and PBDE levels in percolating water. PBDEs in agricultural soils are greatly affected by external environmental conditions such as precipitation, temperature and sunlight. Matcheko et al. (2002) reported that PBDE levels in agricultural soil sampled in the spring tended to be higher than in the following autumn . Hence, it is important to limit the effect of external factors on soils being investigated to obtain results with respect to PBDE distribution and mobility in agricultural soils due to water percolating 118  through the soil. Consistency can best be achieved by laboratory investigations utilizing lysimeters or leaching cells to study the mobility of PBDEs in biosolids-amended soils. There can be a risk of PBDE exposure by humans or wildlife from biosolids application. For this to occur, there must be a source of PBDE contamination, one or more pathways, and receptors. In this case, the source of PBDEs is the biosolids-amended soil. Humans and animals in direct or indirect contact with the PBDEs are the receptors. A potential exposure pathway is PBDEs contaminating groundwater due to leaching through the biosolids-amended soil. Hence, there can be a risk of human exposure to PBDEs via groundwater, which should be considered when regulating biosolids application on agricultural lands. Other pathways can include: PBDEs leaching from biosolids-amended agricultural soils into streams, rivers and surface water bodies; uptake of PBDEs by plants, which are then consumed by humans and/or animals; and animals feeding and living on agricultural land treated with PBDE-contaminated biosolids, with the animals then being consumed by humans. These pathways need to be investigated to determine the associated risk. The present study investigates the variation in overall concentration and PBDE congener profiles in biosolids-amended soil and their leachability due to water percolation by laboratory experiments utilizing glass column lysimeters. PBDE concentrations were measured at different soil depths, and their spatial and temporal changes were followed when water was passed through the columns over a four-week period. The findings contribute to understanding the fate of PBDEs in subsurface environments and to the formulation of transport models of PBDEs in soils, hence evaluating the risk of PBDE exposure to humans and wildlife. These can then assist in developing management practices and policies regarding the application of biosolids on agricultural soils.  119  6.2 Materials and Methods 6.2.1 Sample Characterization and Preparation for Leaching Column Tests The agricultural soil was obtained from the University of British Columbia (UBC) experimental agricultural field in Vancouver, Canada. The soil bulk density was determined according to the Cylindrical Core Method (Arshad et al., 1996). The soil moisture content was determined by the procedure outlined by Chapman (1965), i.e. the original weight of soil was compare to the ovendried weight determined after holding at 105±5oC for 24 hours. The organic content of the agricultural soil was determined by ashing 100 g of oven-dried agricultural soil samples in a muffle furnace at 550±5oC for 2 hours (American Association of State Highway and Transportation Officials [AASHTO], 1986). Triplicate samples were tested for quality control and quality assurance (QA/QC). To ensure that the agricultural soil used in the LCT was homogenous, it was dried to constant weight in an oven at 85±5oC and then sieved through a 2 mm screen to retain particles > 2 mm in nominal size. The temperature was chosen to assure adequate drying while avoiding temperatures that might affect the organic matter properties (Hussein Malkawi et al., 1999). The sieving provided a more uniform soil and prevented the formation of large pores when the agricultural soil was added to the leaching columns. The soil, which had passed through the 2 mm sieve, was adjusted to the average moisture content determined previously from oven drying at 105±5oC. The dried and sieved soil was then divided into 250 g batches. Water was added to each batch to return to the average original moisture content. The batches were next integrated and mixed to ensure homogeneity, and the moisture content was then measured to confirm equivalency with the previously-determined average moisture content.  120  The biosolids were obtained from an urban WWTP in Canada, which cannot be named at its request. At this plant, the biosolids were dewatered and underwent thermophilic and aerobic sludge digestion. The biosolids moisture and organic matter contents were determined using procedures similar to those for the agricultural soil. Prior to the experiments, the biosolids were mixed with the sieved and moisture-content-adjusted agricultural soil on a 1:2 weight-weight basis, equivalent to a biosolids loading of approximately 80 t/ha (8.0 kg/m2), a typical biosolids loading on agricultural fields. This mixture is referred to hereafter as the biosolids-soil mixture. 6.2.2 Leaching Column Experiments Since PBDEs are ubiquitous and present in laboratory environments, extensive cleaning procedures were implemented to prepare the apparatus for the leaching experiments.  The  biosolids-soil mixture was added to nine cylindrical solvent-washed glass columns, known as glass leaching cells (glc), shown in Figure 6.1. The solvent washing consisted of 3×20 mL each of acetone, toluene, hexane, dichloromethane (DCM), hexane and DCM to ensure that the glcs were PBDE-free at the start of our experiments. Each column was 56.5 mm tall and 100 mm in diameter. In each, the biosolids-soil mixture weighed ~210 g and formed a 14 mm thick layer at the bottom. The rest of the column was filled with agricultural soil, divided into three stages, each 14 mm thick. The total weight of these three layers was ~600 g, approximately maintaining the bulk density of the agricultural soil in the field. The biosolids-soil mixture layer is referred to as layer 1, whereas the agricultural soil formed layers 2, 3 and 4, as shown in Figure 6.1. Glass fibre filter paper of 100 mm diameter and 5 µm average pore size was located at both ends of the glcs. Porous ceramic discs of average diameter 100 mm and pore size 60 µm were attached to each end of the cell to ensure well-distributed water flow, and then capped at both ends by aluminium end-plates. The porous discs and end-plates were sonicated three times in acetone,  121  toluene, hexane, DCM, hexane and DCM, separately, to ensure that they were not contaminated with PBDEs.  Pressure gauge Outflow De-ionized water reservoir  Aluminum end plate  Layer 4, 14mm thick Porous discs with filter paper on inner surface sides Deionized water  Glass leaching cells 100mm in diameter and 56 mm in height  Layer 3, 14mm thick Layer 2, 14mm thick Layer 1, 14mm thick; biosolids-soil  Agricultural soil  Aluminum end plate Inflow  Figure 6.1 Schematic of apparatus used for leaching column tests. For simplicity the diagram shows the connection between one reservoir and three leaching columns. However, the experiment had three reservoirs, each connected to three leaching columns, with each leaching column connected to a 4 L amber glass bottle.  Each glc was connected to a solvent-washed aluminium reservoir, which held the deionized water, by 6.3 mm copper tubing. The aluminium reservoir was washed with 3×20 mL of acetone, toluene, hexane, DCM, hexane and DCM, separately. The copper tubing was sonicated with these same solvents. The water which passed upwards through the glc (referred to as leachate hereafter) flowed through 6.3 mm copper tubing to a 4.0 L solvent-washed amber bottle where it was collected. The solvent washing procedure was similar to that of glc solvent washing. Note that the water flowed through the biosolids-soil layer (layer 1) first and then through layers 2, 3 and 4 in that order.  122  There were three solvent-washed aluminium reservoirs, each filled with deionized water, pressurized to 14 kPa gauge by compressed air controlled by pressure regulators and monitored by pressure gauges. Each reservoir was connected in parallel to three glcs. The deionized water then permeated the glcs. Three glcs were dismantled after 1 week, three after 2 weeks and the final three after 4 weeks. After terminating each experiment, the leachates were stored in a dark room at 4°C. Each cell was then weighed and transferred to a fume hood, which was wiped with hexane and covered with hexane-washed aluminium foil. The sample was then extruded in 14 mm slices using a hexane washed wire saw providing four layers from the sample (Figure 6.1). Layers 4, 3, 2 and then 1 were extruded in that order to prevent cross-contamination of the upper layers from layer 1, which was likely to have the highest PBDE concentration. The wire saw was washed with deionized water, acetone, toluene, hexane and DCM after each slicing to prevent crosscontamination. From each layer, a sample was extracted to measure the water content. In addition, from each layer, 5 soil samples from the inner area of the sliced layer were combined and stored in hexane-washed amber glass jars prior to analysis (Appendix D). 6.2.3 Sample Clean-up and Analysis The biosolids-soil mixture and agricultural soil samples were stored at −30°C and the leachate samples at 4°C until analysis. Prior to extraction, approximately 10 g of the biosolids-soil mixture or agricultural soil samples were spiked with 25 µL of the EO5100 standard, a suite of 13  C-labeled PBDEs, and  Massachusetts).  13  C-labeled BDE209 (Cambridge Isotope Laboratories, Andover,  These samples were homogenized with dried anhydrous sodium sulphate  (Na2SO4), based on 1:1.5 (sample: Na2SO4 m/m). The samples were then extracted using Dionex ASE 2000, an accelerated solvent extraction (ASE) system with 80:20 (v/v) toluene:acetone as the solvent. The leachate samples were spiked with the EO5100 surrogate internal standard, and 123  extracted three times with approximately equal volumes of toluene.  Details of sample  preparation and extraction are provided in Appendix D. All sample extracts underwent clean-up via a three-step sample procedure where each sample was eluted with hexane and dicholoromethane (DCM) through a multilayer silica column packed with successive layers of silica gel (basic, neutral, acidic, neutral), a column filled with copper filings, and a neutral activated-alumina column, capped with anhydrous Na2SO4 and washed with hexane prior to sample loading. Eluants from the alumina column were concentrated to <10 µL and spiked with 5 µL of the EO5101 standard, a suite of 13C-labeled PBDEs, (Cambridge Isotope Laboratories, Andover, Massachusetts) prior to congener-specific PBDE analyses by a highresolution gas chromatography high-resolution mass spectrometer (HRGC/HRMS), a VGAutospec high-resolution mass spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas chromatograph and an A200S autosampler (CTC Analytics, Zurich, Switzerland). Details of the analytical procedures utilized in the determination of PBDEs were provided in Appendix A, including the composition and amounts of the  13  C-labeled PBDEs internal and  recovery standards, sorbents, solvents and conditions during sample extraction and clean-up steps, as well as instrumental analysis conditions and QA/QC protocols to check method performance and data integrity. The biosolids-soil and agricultural soil samples were processed in batches of 12, each batch consisting of a procedural blank, an in-house reference sample, and nine samples, one of which was analyzed in duplicate. The recovery of the labelled internal samples was between 65 and 115%, and the precision for most analytes from the analysis of the in-house reference samples and the duplicate analyses was better than 20%.  124  6.2.4 Data Analysis In all samples, concentrations of 51 of the potential 209 PBDE congeners were measured (Appendix D). Congener concentrations in the samples were considered non-detectable (nd) when the measured values were less than five times the concentration of the procedural blanks. Otherwise, congener concentrations measured in the procedural blanks were subtracted from the measured concentrations.  All corrected measured congener concentrations were added to  provide a total concentration, referred to hereafter as ΣPBDE. Error bars indicate 95% confidence limits, obtained from repeated analysis on the same sample. 6.2.5 Mole balance The total moles of PBDEs lost from the biosolids-soil mixture layer were estimated from the moles of PBDEs in the three agricultural layers. This was calculated based on the molecular weight of the PBDEs, assuming that PBDEs underwent debromination, but not chemical and/or biological decomposition. Table 6.1 lists the molecular weights of all ten homologue groups. Table 6.1 Molecular weight of PBDE congeners in the same homologue groups. Homologue Mono Di group 168 248 MWof congener MW: Molecular weight (g/mole)  Tri  Tetra  Penta  Hexa  Hepta  Octa  Nona  Deca  328  408  488  568  648  728  808  888  Once the moles were calculated, the moles of PBDE in each layer were calculated by multiplying the molar concentrations in each layer by the corresponding mass of each layer. The moles from all three agricultural layers were then added to obtain the total moles in the agricultural soil layers (layers 2, 3 and 4). The moles of PBDE in the leachates after 1, 2 and 4 weeks of leaching were estimated in a similar manner by multiplying the molar volumetric concentrations of the leachates by the total leachate volume collected for each duration.  125  The total moles of PBDEs were calculated for the biosolids-soil mixture, agricultural soil and deionized water prior to leaching, and for biosolids-soil, agricultural soil and liquid after leaching. The moles gained from the biosolids-soil layers (MB), the three agricultural soil layers (MA) or the leachate (ML) were determined by subtracting the moles prior to leaching from those after leaching. If the result was negative, then PBDEs were lost from that layer.  6.3 Results and Discussion 6.3.1 Physical Properties of the Biosolids Mixture and Agricultural Soil The average moisture content of the biosolids-soil mixture was 25%, whereas the organic content was 18% on a dry weight basis prior to LCT. The average bulk density, moisture and organic contents of the agricultural soil prior to LCT were 1,440 kg/m3, 16% and 6% (dw), respectively. The moisture content increased to approximately 32-35% and 22-25% for the biosolids-soil and the agricultural soil layers, respectively, after LCT. The organic content decreased in the biosolids-soil mixture layers to approximately 15-16%, but increased to 7-8% in the agricultural soil layers after LCT, due to the transfer of organic material in the biosolids-soil mixture to the agricultural soil layers as water permeated through the glcs. 6.3.2 PBDEs in Biosolids-soil and Agricultural Soil Layers Concentrations of the monitored PBDE congeners are presented in Appendix D. Of the 51 congeners analyzed, BDE47, 85, 100, 99, 154, 153, 183 and 209 were most abundant and were detected in most samples examined.  In the following sections, our data analysis and  corresponding discussions focus on these eight congeners. Another reason for focusing on these “principal congeners” is that they have also been reported in many previous studies, facilitating comparison.  126  The concentrations of these eight major PBDE congeners in the biosolids-soil layer prior to LCT are reported in Table 6.2. The predominant congeners in the biosolids-soil mixture were BDE47, 99 and 209, accounting for 80-83% of ΣPBDEs. Concentrations of the eight principal congeners remained within almost 1% of their initial concentrations after one week of leaching. As leaching continued, concentrations of these congeners decreased by 3-10% and 24-98% of their initial values after 2 and 4 weeks of leaching. The ΣPBDEs concentration decreased by 1, 14 and 38% from its initial value after 1, 2 and 4 weeks of leaching, respectively. This decrease is attributed to desorption of PBDEs from the biosolids-soil layer, and is consistent with observations from investigations on the mobility of PCBs in sludge-soil (Adeel et al., 1997). PCBs have chemical structures and chemical and physical properties similar to PBDEs (D'Silva et al., 2004; Hooper and McDonald, 2000). Concentrations of PCBs were reduced by almost 60% from their initial values after the biosolids-soil mixture was leached with distilled water for 40 days (Adeel et al., 1997). The percent reduction of PCBs was greater than for PBDEs because the leaching period was longer, and because PCBs are more soluble (0.1 to 0.6 mg/L) and have lower log (KOW) values (4.5 to 8.2) than PBDEs, making them more mobile than PBDEs. Concentrations of the selected PBDEs in the agricultural soil prior to leaching were mostly nondetectable, as shown in Table 6.2. BDE85 was the only congener detected in all three layers and its concentration generally increased with the duration of leaching. As expected, layer 2 had the greatest increase in ΣPBDE, especially after four weeks of leaching. BDEs 47, 99, 100, 153, 154, 183 and 209 were all non-detectable (nd) in layers 3 and 4, regardless of the duration of leaching. The leachate volumes after 1, 2 and 4 weeks were 1.25, 2.25 and 3.75 L, respectively. The resulting ratios of leachate volume to volume of the biosolids-amended soil layer after 1, 2 and 4 weeks of leaching were 11, 20 and 34, respectively, as shown in Table 6.2. This ratio, useful for  127  interpretation and comparing with environmental conditions, is designated “volume ratio” in the rest of this paper. The tests show that for BDE47, 85, 99, 100, 153 and 154 in layer 1 there were losses from the initial concentration of 0, 3-9% and 35-41% after 1 week (volume ratio of 11), after 2 weeks (volume ratio 20) and after 4 weeks (volume ratio 34) of leaching, respectively. For BDE183, there was a sharp loss of 96, 87 and 98% after 1 week (volume ratio of 11), 2 weeks (volume ratio 20) and 4 weeks (volume ratio 34) of leaching, respectively. Finally for BDE209 after 1 week (volume ratio 11) there was no loss, after 2 weeks (volume ratio 20) there was a 9% loss and after 4 weeks (volume ratio 34) there was a 24% loss. These data show that lighter congeners such as BDE47, 85, 99, 100, 153, 154 and BDE183 have a higher rate of loss than BDE209, which is due to the high KOW value of the latter, causing it to be less mobile in the soil. The BDE homologue group concentrations are reported in Figure 6.2 after 1, 2 and 4 weeks of leaching. The biosolids-soil layer (layer 1) and layer 2 are the only layers represented in Figure 2, since layers 3 and 4 only had measurable concentrations of BDE85. It is evident from the data that the tetra-, penta and deca-BDE homologue groups had the highest concentrations in the biosolids-soil mixture layer prior to and after leaching. This observation is consistent with the PBDE profiles reported by Gorgy et al. (2010), where analyses of the same biosolid samples as in the biosolids-soil mixture samples examined here had levels of tetra-, penta- and deca-BDE homologue groups that contributed most of the ΣPBDE measured in the biosolids. These homologue groups have also been found to be the major constituents in domestic wastewater which finds its way to WWTP facilities which generate biosolids (de Boer et al., 2003; Oberg et al., 2002).  128  Table 6.2 Concentration of eight major PBDE congeners in the biosolids-soil layer during the leaching tests in pg/g dry weight basis (dw). BDE 47 85 Layer 1: Biosolids-soil Layer Biosolids137×103 7.1×103 (1) soil 152×103 7.2×103 11/1 week(2) 127×103 6.4×103 20/2 weeks) 88×103 4.2×103 34/4 weeks Layer 2: Agricultural Soil Layer nd(4) 14 11/1 week nd nd 20/2 weeks 3 3 28×10 2.2×10 34/4 weeks Layer 3: Agricultural Soil Layer nd 6 11/1 week nd 8 20/2 weeks nd 20 34/4 weeks Layer 4: Agricultural Soil Layer nd nd 11/1 week nd 8 20/2 weeks nd 20 34/4 weeks (1)  99  100  153  154  183  209  Σ PBDEs(3)  163×103  34×103  15×103  13×103  5×103  204×103  664×103  170×103 150×103 99×103  35×103 31×103 20×103  17×103 15×103 10×103  15×103 12×103 9×103  135 113 71  215×103 186×103 155×103  657×103 570×103 410×103  nd nd  nd nd  nd 2.2 3 11×10  nd nd 5×10  nd 16 0  nd 840 3 68×10  14 1,100 3 234×10  80×10  3  10×10  3  3  nd nd nd  nd nd nd  nd nd nd  nd nd nd  nd nd nd  nd nd nd  6 20 25  nd nd nd  nd nd nd  nd nd nd  nd nd nd  nd nd nd  nd nd nd  nd 8 20  Biosolids-soil prior to leaching  (2)  Ratio of leachate volume to volume of biosolids-amended soil layer /Duration of leaching. Total concentration of the 51 PBDE congeners analyzed. (4) Non-detectable. (3)  The leaching experiments resulted in 8-42% reduction of the tetra to deca homologue groups from their initial concentrations prior to leaching in the biosolids-soil (layer 1) after 4 weeks of leaching, corresponding to a volume ratio of 34 (Figure 6.2a). As a result of this reduction, the relative contributions of the different homologue groups to ΣPBDE varied. Groups with ≤9 bromines slightly decreased by approximately 1-3% owing to leaching, whereas deca-BDE increased by almost 5% (Figure 6.3a). Similarly, the concentration ratios of BDE47 and BDE99 to ΣPBDE decreased from 23 to 21% and 26 to 24% respectively, while the BDE209/ΣPBDE concentration ratio increased from 80 to 83%. Comparison of these results reveals that the leachability behavior of BDE47 and 99 differs from that of BDE209. BDE 47 and 99 have substantially lower log KOW values than BDE209 and thus are significantly more water soluble. This facilitates their mobilization in soil in the water percolation experiments. On the other hand, BDE209 is virtually insoluble in water and tends to sorb strongly on soil surfaces, especially 129  those of high organic content, and thus it is not readily mobilized between soil layers. Although other factors may contribute to the mobilization of PBDEs between soil layers, the solubility hypothesis is consistent with BDE47 and BDE99 leaching more abundantly from layer 1 than BDE209.  (a) Layer 1 Concentration (pg/g dw)  (b) Layer 2  Number of Bromines  Number of Bromines  Contribution to ∑ PBDE Concentration  Figure 6.2 BDE homologue group concentrations (pg/g dw) in (a) biosolids-soil layer (Layer 1; 0-14 mm) and (b) agricultural soil layer (Layer 2; 14-28 mm). Homologues without bars indicate that they were nd. Agricultural Layers 3 (28-42 mm) and 4 (42-56 mm) had a total PBDE concentration between nd to13 pg/g dw.  (b) Layer 2  (a) Layer 1  Number of Bromines  Number of Bromines  Figure 6.3 Percent contribution of BDE homologue groups to ∑PBDE in (a) biosolids-soil layer (Layer 1; 0-14 mm) and (b) layer 2 (14-28 mm) of the agricultural soil.  130  In the subsequent layers, PBDE concentrations increased with leaching time (Figure 6.2b). The homologue BDE concentrations except for the mono-, octa- and nona-BDE groups in layer 2 increased from nd to 45×103 pg/g dw, as indicated in Figure 6.2b. PentaBDE was the only homologue group detected in layers 3 and 4, with concentrations increasing from 3 to 13 pg/g dw in layer 3 and from nd to 10 pg/g in layer 4. Distributions of the homologue groups in soil layers 2 to 4 differed from those in the biosolidssoil layer. In layer 2, homologue groups with six or fewer bromines (di- to hexa-BDEs), except for penta-BDE, increased to 19% of the total PBDE concentration (Figure 6.3b). Percent contribution of homologue groups with seven to nine bromines (hepta- to nona-) decreased after leaching to negligible. As for deca-BDE, the percent contribution increased from negligible to 76% in the first two weeks of leaching, but then decreased to ~29% after 4 weeks. In layers 3 and 4, pentaBDE was the only homologue group whose concentration increased measurably over the 4 weeks of leaching. The change in the relative contribution of different homologue groups with time shows the influence of the degree of PBDE bromination on the PBDE mobility. The decrease of the lowerbrominated groups (di- to hexa-BDEs) in the biosolids-soil layer, which is contaminated with PBDEs, and their increase in agricultural soil layers with leaching shows that the lessbrominated PBDEs are mobilized more abundantly by the infiltrating water. The corresponding increase in the proportion of the heavier homologue groups like deca-BDE in the biosolids layer, and their decrease or absence in the soil layers is an indication of their lower mobility. The mobility is related to the water solubility and the organic carbon normalized partitioning constant, KOC, which is estimated approximately from KOW, due to lack of accurate KOC data in the literature. KOW is a measure of the hydrophobicity of a compound and is inversely related to its water solubility (Frankki et al., 2006). In addition, the mobility of a compound depends on the 131  organic carbon content of the soil (i.e. the higher the organic carbon content of the soil, the more hydrophobic compounds can sorb on the soil, thus becoming less mobile). Therefore the percent composition of ∑PBDE as hepta to deca-BDEs increased in the biosolids-soil layer because these compounds have relatively high log KOW, ~ 9.4-10, and thus are retained in the biosolidssoil layer which has a higher organic carbon content than the agricultural soil, 18 rather than 6%. Di- to hexa-BDEs have log KOW values of ~5.4-8.0 and thus are more mobile than hepta- to deca-BDE, explaining their decreasing proportion in the biosolids-soil layer as leaching proceeded. 6.3.3 Leachate PBDE Concentrations ΣPBDE increased from nd to 86×103, 108×103, and 310×103 pg/L in the deionized water after 1, 2 and 4 weeks of leaching through the soil (Figure 6.4). The largest increase in the concentration is due to deca-BDE. The concentration of individual PBDE congeners in the leachates are presented in Table 6.3. For all congeners except BDE209, the leachate concentrations are less than their estimated solubilities (SW) (Palm et al., 2002) by 2-3 orders of magnitude. The presence of BDE209 in the leachates suggests that the leachates contain substances other than the deionized water which could have promoted BDE209 removal. These substances include, but are not limited to, dissolved organic matter and fine suspended particles that passed through the 5 µm pore filter, after leaching through the biosolids-soil and agricultural soil layers. The suspended solids would adsorb BDE209, due to this congener’s high log KOW value of 10, and hence would carry BDE209 through the soil pores and filters.  The effects of dissolved organic matter and suspended particles on the partitioning of hydrophobic compounds such as BDE209 have been discussed in several published studies examining the transport of hydrophobic contaminants. Several studies (Gschwend and Wu, 1985; Katsoyiannis and Samara, 2005; Voice et al., 1983) have noted the increase in aqueous 132  concentration of hydrophobic organic contaminants with a decrease in distribution coefficient (Kd), the ratio of the sorbed phase concentration to that of the solution phase. An examination of the distribution of desorbed dioxins from soil between the dissolved organic matter and particulate organic matter in the aqueous phase found that 97% of dioxins in the aqueous phase were associated with the particulate organic matter (Frankki et al., 2006). A study on the effect of suspended solids in the leachate on the PBDE partitioning (Gorgy et al., 2010) indicated that PBDE aqueous concentrations exceeded their solubilities due to their adsorption to suspended particles. The leachate in that case was prepared from the same biosolids as in this study. The finding of suspended solids can hence be applied to the PBDEs in this study since the water  Concentration (pg/L)  infiltrating the biosolids-soil layer had similar physical properties.  Number of Bromines  Figure 6.4 BDE homologue group concentrations (pg/L) in leachate after 1, 2 and 4 weeks of leaching deionized water through soil columns, corresponding to volume ratios of leachate to soil of 11, 20 and 34, respectively.  133  Table 6.3 Estimated water solubilities (pg/L) and concentrations (pg/L) of selected PBDE congeners in the leachates after 1, 2 and 4 week/s of leaching. Congener  Sw(1) (pg/L)  Leaching Duration  St Error(2) 9×106 5.8×103 1.4×103 4.6×103 1.3×103 5.0×103 BDE47 1450 4 4×10 BDE85 173 43 81 21 83 24.07 8×105 4.0×103 1.0×103 2.5×103 2.7×103 BDE99 675 783 5 3 5×10 1.1×10 BDE100 275 740 200 700 203 4×103 BDE153 370 93 200 54 290 84.1 4×103 BDE154 260 65 130 35 205 59.45 2×102 1.4×103 BDE183 511 128 530 143 414.7 4 4 4 4 5 6.5×10 1.6×10 8.4×10 23×10 1.8×10 BDE209 Insoluble 52780 (1) Estimated water solubility (Palm et al., 2002); Sw values for BDE100, 153 and 154 were based on mean values established from corresponding homologues (2) Standard Error (n=3) 1 week  St Error(2)  2 weeks  St Error(2)  4 weeks  6.3.4 PBDE Mole Balance The total moles of PBDEs lost from the biosolids-soil layers was compared to the total mass gained collectively in layers 2, 3 and 4 and in the leachates for each of the three leaching durations. Table 6.4 summarizes the proportion of PBDE moles lost from the biosolids-soil (layer 1), denoted as MB, and gained from in the three agricultural soil layers collectively, designated as MA and from the leachates, denoted by ML. Subtraction of MB from (MA + ML) indicates that ∼1,600, 2,300 and 2,500 µmoles, which is equivalent to after 12, 14 and 15% loss from the initial number of moles after 1 week (volume ratio of 11), 2 weeks (volume ratio 20) and 4 weeks (volume ratio 34), respectively. These amounts were not gained either by the three agricultural soil layers or by the leachate after 1 and 2 weeks of leaching. This suggests underestimation in the mass balance, likely due to PBDEs adsorbed on suspended particles in the leachate which are neither separated nor analyzed. In particular, the liquid-liquid extraction procedure used to extract PBDEs from the leachates may not be able to dissolve all PBDEs attached to suspended ultrafine particles. By contrast, after 4 weeks of leaching, the mass balance slightly overestimated the PBDE moles gained in the agricultural soil layers and the leachate. The number of moles after 4 weeks of leaching was overestimated by 24% of the moles 134  lost from the biosolids-soil layer, while the underestimates after 1 and 2 weeks of leaching were ~ 24 and 53%, respectively, of the moles lost from the biosolids-soil layer. The decrease of the difference between the sum of moles in the leachate collected and in the agricultural soil layer and the moles lost from the biosolids-agricultural is a strong indication of the influence of suspended solids on PBDE transport. As the leachate flows through the soil, the retention of suspended solids and the growth of biomass on soil particle surfaces causes a reduction in pore space (VaGulck and Rowe, 2004). The flow of leachate through the soil decreases owing to the reduction in pore spaces, providing more time for PBDEs, especially less brominated ones, such as tetra- and penta-BDEs, to partition into the biomass and suspended solids retained on the soil surfaces. The increase of organic content of the soils from 6 to 8% indicates increased suspended solids retention and biomass due to potential microbial degradation. Table 6.4 Moles (µmoles) of PBDEs lost or gained in the leaching column tests Matrix Moles (µmoles) lost from biosolids-soil layer ( MB) Moles (µmoles) gained in Layers 2,3& 4 (MA) Moles (µmoles) in leachate (ML) Moles (µmoles) underestimated (-)/ overestimated (+) (MA+ML)- MB  Leaching Duration 1 week  2 weeks  4 weeks  1,600  2,300  2,500  160 225  700 520  4,600 1,000  -1,215  -1,080  3,100  The mole balance can also help to determine the fraction of PBDEs in the four layers to PBDEs in the biosolids-soil prior to leaching. Figure 6.5 indicates that after 4 weeks of leaching of deionized water corresponding to a volume ratio of leachate to soil of 34, the biosolids-soil layer contained approximately 81% of the original PBDE prior to leaching. Layer 2, consisting only of agricultural soil, contained ~ 26% of the PBDEs in the biosolids-soil mixture prior to leaching. The remaining soil layers had less than 1% of the PBDEs in the biosolids-soil prior to  135  leaching, and PBDEs in the leachate were almost 1% of the PBDEs in biosolids-soil mix prior to leaching.  The PBDEs in the four layers, collectively, covering a total depth of 56 mm,  accounted for more than 26% of the PBDEs in the biosolids-soil prior to leaching. This indicates that PBDEs leached from the biosolids-soil layer to the agricultural soil layers. The PBDE mobility was greater for our 672-h study than for the 12-h leaching study of Litz (2002).  Percent of PBDEs in Soil Layers and Leachate to PBDE in Biosolids-Soil Prior to LCT  Figure 6.5 Percent of PBDEs in biosolids-soil mix layer (Layer 1), agricultural soil layers (Layers 2,3 and 4) and leachates to that in the biosolids-soil mix prior to leaching. Note that the percentages are plotted on a logarithmic scale.  6.4 Conclusions PBDEs can leach to agricultural soils from biosolids-amended soils due to the passage of water. ∑PBDE concentrations decreased by 38% over a 4-week period after passing approximately 3.75 L through 210 g of biosolids-amended soil and 600 g of agricultural soil. Leaching column tests showed that lower-brominated PBDEs mobilized more than higher-brominated ones. This can be attributed to the smaller KOW values of the lower-brominated congeners, allowing them to  136  partition more readily into the percolating water. Leachates collected from the tests gained in BDE209 concentrations, even exceeding the BDE209 water solubility. This supports earlier work indicating transfer of BDE209 due to ultrafine particles suspended in the leachate. The increased levels of PBDEs with increased leaching duration and volume of water passing through the biosolids-soil mix layer and the agricultural soils confirms that PBDEs from biosolids application can contaminate water resources. PBDEs can potentially leach and mobilize from biosolids-amended soils and ultimately reach groundwater. Hence the application of biosolids on agricultural soils could cause human exposure to PBDEs, which should be considered when setting policies on the application of biosolids to soils.  137  7 PBDEs in Soils Near Northern Canadian Waste Disposal Sites 7.1 Introduction Environmental release of PBDEs may occur during their synthesis and incorporation into polymers for finished products, or as a result of their ultimate disposal. Alcock et al. (2003) estimated that 80% of PBDEs in discarded consumer products are landfilled, with the remainder incinerated. The relatively low volatilities and aqueous solubilities of PBDEs suggest that the bulk of the environmental burden of these chemicals eventually resides in either sediments or soils (Palm et al., 2002; Hale et al., 2003). Studies quantifying PBDE contamination in the environment due to leaching from municipal solid waste facilities and landfill sites are rare. However, Alcock et al. (2003) and deBoer et al. (2003) reported that sediments in surface water surrounding landfills have elevated concentrations of tetraBDE and pentaBDE congeners, ranging from 21×103 to 370×103 pg/g dw and 9.2×103 to 900×103 pg/g dw, respectively. PBDEs have been found in remote areas such as the Arctic. PBDE concentrations in air samples collected in 1994 from Alert and Tagish, two small Northern Canadian communities, were between 700 and 2,000 pg/m3 (Alaee et al., 2003).  Wania et al. (2005) reported PBDE  concentrations of 3×103 to 29×103 pg/m3 from air samples collected in 2000-01 from several Arctic locations. These contaminants have also been detected in Arctic animals, such as ringed seals, beluga whales and polar bears. Concentrations in ringed seals and beluga whales, between 1981 and 2000, were 572-15,500 pg/g lipid weight, increasing exponentially over this period (Ikonomou et al., 2002). The presence of PBDEs in rural and remote areas, such as the Arctic, is partially attributed to their transport from urban regions via atmospheric deposition, i.e. longrange atmospheric transport (LRAT) (Barrie et al., 1992; Wania and Dugani, 2003). Eckhardt et  138  al. (2003) and Duncan and Bey (2004) found that atmospheric transport in the Northern Hemisphere is in the south-north direction. However, PBDE contamination in the Arctic could also be due to local sources such as solid waste dumpsites. Northern communities use commercial products containing PBDE additives, and dispose their solid waste on land. PBDE levels from old dumpsites and operating waste disposal sites in Northern Canada have been found to be in the range of 350 to 51×103 pg/g dw (Danon-Schaffer, 2010). It is vital to examine the soils surrounding waste disposal facilities to determine the nature and extent of PBDE contamination in Northern remote regions. The food chain transfer and bioaccumulation of PBDEs in terrestrial ecological systems also need to be examined because they could provide an important pathway for human exposure. However, investigations in Northern regions are difficult due to inaccessibility and complex interactions between climate, hydrology and drainage (Marsh et al., 2005). In addition, there is a lack of detailed surveys of industrial activities, random waste burning and unregulated waste disposal sources. The objectives of this study were to: (1) Examine PBDEs contamination in the soil upstream and downstream of waste disposal facilities in Iqaluit, Nunavut and Yellowknife, Northwest Territories; (2) Establish the lateral and vertical distribution trend of these PBDEs in upgradient and downgradient soils; (3) Compare the concentrations with those of other Northern soils; and (4) Examine the relationship between the PBDE contamination in these soils and other environmental matrices from the North. The overall goal is to determine the impact of waste disposal facilities in Northern communities on PBDE contamination in remote environments in order to improve the management of waste disposal facilities in these regions.  139  7.2 Methods In July 2006 sampling was carried out in Iqaluit and Yellowknife to investigate PBDEs within Northern Canadian waste disposal facilities and related soils. These two locations were selected based on a previous study (Danon-Schaffer, 2010) and represented population centres from the east and west of Northern Canada. Iqaluit has a population of 6,184 and a population density of 100 people/km2 (Statistics Canada, 2007a). Yellowknife has a population of 18,700 and a population density of 120 people/km2 (Statistics Canada, 2007b). Soil samples were collected upgradient and downgradient of both solid waste facilities. 7.2.1 Site Background Iqaluit Waste Disposal Facility (IQ) The Iqaluit waste disposal facility, located south of the town, has been operating since 1995. Prior to 2002 the facility practiced open burning at its main dumpsite, and the ash was dumped and covered at different locations in the surrounding area. Iqaluit currently compacts its solid waste. The landfill receives ~6,000 tonnes of waste annually, segregating it into metals, tires, garbage waste, shredded wood, organic compost, and sewage sludge.  A drainage system  surrounds the landfill to collect leachate surface runoff, draining into a nearby bay ~500 m southeast of the landfill (Figure 7.1a). Yellowknife Waste Disposal Facility (YELL) The Yellowknife waste disposal facility, located north of the city near Jackfish Lake, has operated since 1970.  It receives 10,000 tonnes of waste annually from the residential,  commercial, industrial and construction sectors. The landfill is divided into designated areas for scrap metals, construction debris, scrap vehicles, wood/carpet, tires, brush, peat from a municipal sewage treatment lagoon, refrigerators, soils contaminated from oil spills, and garbage bags 140  containing mainly domestic waste (Figure 7.1b). The latter are processed, separated, baled and stacked in a designated area.  (a) Koojesse Inlet  Landfill fence  IQ-1  Scale: 0  5m  IQ-2  10 m  IQ-3  IQ-4  (b)  Wood & Carpets  Peat  Refrigerators Contaminated Soil  Scale: 0  5m  10 m  YELL-2 YELL-3 YELL-4  Tires Garbage  Construction Debris  YELL-1  Scrap Metal  Road  Lagoon  Figure 7.1 Site plans and sampling locations for (a) Iqaluit waste disposal facility, and (b) Yellowknife solid waste disposal facility. Sampling locations for Iqaluit waste disposal site are designated IQ-1, IQ-2, IQ-3 and IQ-4 and for Yellowknife solid waste disposal facility YELL-1, YELL-2, YELL-3 and YELL-4.  141  7.2.2 Soil Sampling In both locations soil samples were collected using a stainless steel trowel which had been thoroughly washed with Alcotab®, a cleaning detergent, dissolved in distilled water before each new soil sample was taken. Each sample was collected in hexane-washed amber glass jars. Iqaluit Waste Disposal Facility (IQ) Soil samples were collected from a location upgradient and ~100 m from the disposal facility, referred to as IQ-1, and at three locations downgradient from the landfill, referred to as IQ-2, IQ3 and IQ-4. The latter were 3, 6 and 9 m from the disposal facility, respectively (Table 7.1). These sites were along the drainage system, and towards the receiving bay. They were chosen to determine the effect of the disposal facility on PBDE contamination in the surrounding environment (Figure 7.1a). At each location a 1 m × 1 m pit was excavated to a depth of 1 m using a backhoe. Each side of the 4 pits was divided into five layers, each of thickness 0.20 m. Due to the high water table levels, samples were only taken from the 0-0.20 m and 0.20-0.40 m intervals. Yellowknife Waste Disposal Facility (YELL) Samples from Yellowknife were collected ~100 m from the landfill (YELL-1), and from three areas downgradient and south of the landfill referred to as YELL-2, YELL-3 and YELL-4, located 3, 6 and 9 m from the disposal facility, respectively (Table 7.1). Due to the rocky nature of the region, 0.30 m deep holes (0.50 m × 0.50 m) were excavated using a pick axe and shovel. The samples were collected from 0-0.15 m and 0.15-0.30 m depths. At YELL-4, baled waste was found ~0.30 m below the surface. This area was in a section of the disposal facility closed since 2000, and was covered with 0.30 m of soil from several excavated locations around Yellowknife (Figure 7.1b).  142  Table 7.1 Sampling locations and elevations (m) above sea level at Iqaluit Waste Disposal Facility and Yellowknife Waste Disposal Facility. Location  Elevation (m)  Coordinates  Iqaluit IQ-1  23  IQ-2  18  IQ-3  9  IQ-4  7  N 63o 93.821 W 68o 32.140 N 63o 93.818 W 68o 32.170 N 63o 93.818 W 68o 32.166 N 63o 93.818 W 68o 32.163  Yellowknife YELL-1  186  YELL-2  199  YELL-3  195  YELL-4  194  N 62o 28.355 W 114o 22.396 N 62o 28.553 W 114o 22.440 N 62o 28.552 W 114o 22.441 N 62o 28.550 W 114o 22.442  7.2.3 Storage and PBDE Analysis Soil samples were sent directly to the Institute of ocean sciences (IOS) in Sidney, BC and stored at -30°C until analysis. All samples were first spiked with 25 µL of EO5100, a suite of labeled PBDEs, and  13  13  C-  C-labeled BDE209 (Cambridge Isotope Laboratories, Andover,  Massachusetts). PBDEs from soil samples were extracted with a Dionex ASE (Dionex, Sunnyvale, CA, USA), using a mixture of toluene:acetone (80:20 v/v), at a pressure of 2,000 psi (13,790 kPa) and a temperature of 180°C.  The extracts were then subjected to extensive  cleaning in preparation for analysis using a HRGC/HRMS using a VG-Autospec high-resolution mass spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas chromatograph and a CTC A200S autosampler (CTC Analytics, Zurich, Switzerland). The GC was operated in the splitless injection mode, with the splitless injector purge valve activated 2 min after sample injection. Detailed procedures for the analyses of 143  PBDEs, QA/QC protocols, analyte identification criteria, and quantification were described in detail by (Ikonomou et al., 2001). All samples were processed in batches of 12, consisting of a procedural blank, a certified reference sample, and nine soil samples, one of which was analyzed in duplicate. 7.2.4 Soil Characterization Moisture content, OM, CEC and particle size distribution were determined for each soil sample. The soil moisture content was determined by the procedure outlined by Chapman (1965), i.e. the original weight of soil was compared to the oven-dried weight determined after holding at 105±5°C for 24 hours. The OM of the soil was determined by ashing 100 g of oven-dried agricultural soil samples in a muffle furnace at 550±5 °C for 2 hours (AASHTO 1986). Triplicate samples were tested for QA/QC.  The CEC was determined according to the  procedures outlined by LRRI (1965), which used the sodium acetate replacement method (Appendix A). The concentrations of exchangeable cations were measured using an atomic adsorption spectrophotometer, Thermo Jarell Ash video (Thermo Fisher Scientific Inc.). The particle size distribution was measured using laser diffraction by a Mastersizer 2000 particle size analyzer. 7.2.5 Data Analysis Concentrations of 51 of the 209 BDE congeners were analyzed and compiled (See appendix E). Concentrations from procedural blanks were subtracted from the measured PBDE concentrations of the samples belonging to the same batch. All of these adjusted concentrations were added to provide a total concentration referred to as ΣPBDE. To determine the distribution trends in the soil, the cumulative sums of specified homologue groups were calculated. For example, the cumulative sum of tetraBDE concentrations includes all congeners with four or fewer bromines. 144  Error bars indicate 95% confidence limits on the mean, obtained from repeated measurements (n=4) on the same sample. Concentrations of often-reported principal congeners: BDE47, 99, 154, 153, 183 and 209 and their %-ratios to ΣPBDE, are also reported. These ratios determine the abundance of the different congeners and their distribution trends. The ratio of BDE47 to BDE99 (denoted as BDE47/99) provides an indication of whether the presence of these compounds originates from commercial mixtures or from debromination of higher-brominated congeners and transport away from their sources of contamination.  7.3 Results and Discussion 7.3.1 Soil Properties The OM, CEC, moisture content and D50 (equivalent particle diameter at which 50% by mass of soil particles are smaller), are summarized in Table 7.2. The data indicate heterogeneity of the soils in both locations, with organic matter content varying between 0.3 and 19% dw; CEC ranging from 0.5 to 3.4 meq/100g dw; moisture content between 4.2 and 13% and D50 from 223 to 836 µm in Iqaluit. Yellowknife soil properties were 0.3-3.7% dw for organic content; 1.20.5meq/100g dw for CEC; 1.2-14% for moisture content and D50 between 37 and 561 µm. 7.3.2 Vertical and Lateral Variations of PBDE Concentration The ΣPBDE concentrations from Iqaluit and Yellowknife disposal facilities are summarized in Figures 7.2 and 7.3, respectively. Cumulative distributions of the PBDE homologue groups are presented in Figure 7.4. The highest concentrations from Iqaluit’s disposal site occurred at IQ-1, upgradient from the facility, in the top (0-0.20 m) layer, with a ΣPBDE concentration of 5,000 pg/g dw. The lowest total concentration of 340 pg/g dw was recorded in the bottom layer of IQ4. It was expected that PBDE levels would decrease away from the disposal site. However, IQ-4 (0-0.20 m) had higher concentrations than IQ-2 and IQ-3 at the same depth. This is linked to 145  black residue patches due to ash from previous burning of municipal waste in the vicinity of IQ4, which had a ΣPBDE concentration of 2,900 pg/g dw. Figure 7.2 shows that the total concentrations in the top layer of IQ-1 are higher than downgradient, indicating that there might be other sources of PBDE contamination upgradient from the disposal facility. However, this cannot be verified due to lack of monitoring of PBDE levels in this region. Table 7.2 Chemical and physical characteristics of Iqaluit (IQ) and Yellowknife (YELL) soil samples. Sample  Depth (m)  IQ-1  0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30  IQ-2 IQ-3 IQ-4 YELL-1 YELL-2 YELL-3 YELL-4  Organic Matter (% by wt) 1.3 19.9 0.3 0.4 0.3 0.3 0.3 0.3 2.1 3.7 1.3 0.8 1.4 1.8 0.3 0.7  CEC (meq/100g)* 2.2 3.4 1.4 1.7 2.7 0.5 1.9 1.9 7.2 10.5 7.1 6.9 2.4 2.2 1.3 1.2  Moisture content (% by wt) 12.4 13.4 4.3 11.7 5.0 4.2 8.4 13.4 4.8 6.7 14.0 9.0 1.4 6.2 1.2 2.5  D50 (µm) 547 836 824 247 737 789 789 223 143 57 37 32 100 276 308 561  * Cation exchange capacity expressed in milliequivalents per 100 g. Na concentration presented in Appendix E.  146  Comparison of the concentrations in the top and bottom layers, to establish the vertical depth trend of the PBDEs, shows that the total concentrations are higher in the 0.20-0.40 m layer than for the 0-0.20 m interval at IQ-3 (Figures 7.2 and Figure 7.4a). However, IQ-1, IQ2 and IQ-4 concentrations in the 0-0.20 m slice are higher than for the 0.20-0.40 m interval.  These  inconsistencies make it difficult to establish a general trend for the vertical PBDE distribution across the overall 0.40 m depth analyzed. The cumulative distributions of the homologue groups from the Yellowknife disposal facility show a much wider spread than at Iqaluit. The effect of old covered waste in the closed section on PBDE contamination is clear from YELL-3 and YELL-4. The 0.15-0.30 m layers at both these locations had the highest total PBDE concentrations, 42,300 and 57,400 pg/g dw (Figure 7.3), respectively, mainly from the tetraBDEs (Figure 7.4b). Koojesse Inlet  Landfill fence  IQ-1  IQ-2  IQ-3 IQ-4  Scale: 0  5m  10 m  Figure 7.2 Cumulative PBDE concentrations (pg/g dw) at Iqaluit Waste Disposal Facility. 147  Wood & Carpets Refrigerators  Peat  Contaminated Soil  Tires  Garbage  Construction Debris  YELL-1  Scrap Metal  YELL-2 YELL-3 YELL-4 Road Lagoon  Scale: 0  5m  10 m  Figure 7.3 Cumulative PBDE concentrations (pg/g dw) at Yellowknife Waste Disposal Facility.  (a) Iqaluit Concentration (pg/g dw)  (b) Yellowknife  Number of Bromines  Number of Bromines  Figure 7.4 Cumulative PBDE concentrations (pg/g dw) in soil upgradient and downgradient of: (a) Iqaluit Waste Disposal Facility, and (b) Yellowknife Waste Disposal facility. Errors bars portray 95% confidence intervals.  148  Groundwater table fluctuation at the Iqaluit site may explain the inconsistent distribution of PBDEs in the samples. Fluctuations in water-table elevation affect the spatial distribution of contaminants within the soil, particularly in the vertical direction. When the water-table descends, the contaminants migrate downwards. In addition, a rise in water-table elevation leads to the reverse process, upward migration of the contaminants (Dobson et al., 2007). The Iqaluit Waste disposal site is on the south shore of Koojesse Inlet. Hence, the groundwater table at the site is directly influenced by the tidal fluctuations. Although PBDEs are hydrophobic, Gorgy et al. (2010) have shown that they can be mobilized via suspended particulate matter present in the aqueous phase, and thus will also move with the groundwater on site. The lack of clear lateral and vertical trends of PBDEs at the sites could be due to their association with the soil. The physiochemical properties of these compounds indicate their strong affinity to partition to organic matter. To investigate the relation between ΣPBDE concentrations and the chemical and physical properties of the soil, ΣPBDE is plotted against moisture content, OM, CEC and D50 of the soil samples in Figure 7.5. After excluding outlying data from the relationships between ΣPBDE and D50 and OM (Figures 7.5a and b, respectively), a few trends can be identified. There is an inverse exponential relationship between the sizes of the soil particles from Iqaluit and ΣPBDE. This inverse relationship is characteristic of sorption of hydrophobic compounds by solid particles (Wu and Gschwend, 1988). In addition, ΣPBDE concentration increased with soil OM content from the samples collected from Iqaluit (Figure 7.5b), which is also characteristic of hydrophobic compounds, such as PBDEs in the environment. Sorption of hydrophobic compounds increases with increasing organic matter of the soil (Grathwohl, 1990). On the other hand, ΣPBDE concentration increased with particle size of the soil samples, and there was very poor correlation between ΣPBDE and the soil OM content from Yellowknife. The CEC and moisture content plots did not show any clear 149  relationship between the concentrations and soil properties. The observed correlations in the Iqaluit soil samples are a strong indication that the presence of PBDEs in soils near Iqaluit Waste Disposal Facility is associated with their environmental fate and transport, which led to their deposition in Iqaluit. The poor correlations for the Yellowknife samples indicate that PBDE presence is not only associated with their environmental fate and transport, but also with a local heterogeneous source of contamination, such as disposed waste.  (a)  (c)  (b)  (d)  Figure 7.5 Correlation between ΣPBDE and: (a) median particle diameter (D50), (b) total organic matter (%), (c) cation exchange capacity (CEC) and (d) moisture content of soil samples from Iqaluit and Yellowknife waste disposal facilities.  150  7.3.3 Distribution Trends of Specific Congeners Table 7.3 summarizes the concentrations of BDE47, 99, 100, 154, 153, 183 and 209. Their presence caused large increases in the cumulative distributions between the tetra- and hepta-BDE homologue groups in Figure 7.4. The largest increases often occur due to BDE47 and BDE209. These increases are indicative of the wide usage of PeBDE and DeBDE formulations. As in Table 7.4, BDE209 contributed 50-90% of the total PBDE concentration, whereas BDE47 and BDE99 comprised almost 2-25% of the total concentration with BDE209 included. The presence of BDE209 leads to two sources of PBDE contamination. The first is the extensive use of the DeBDE commercial mixture, which constitutes > 90% BDE209 in many consumer goods disposed in waste disposal facilities (de Wit et al., 2006). The second source is particle bound LRAT. Dust deposition samples collected from the Arctic had high levels of BDE209 and can be attributed to LRAT (de Wit et al., 2006). In many cases the distribution of PBDE congeners can be misinterpreted as a result of including BDE 209 in the total PBDE concentration. Table 7.5 includes the proportion of BDE47, 99, 100, 154, 153 and 183, normalized to their sum, a method used by Leung et al. (2007) to ascertain whether PBDEs in soil samples originate from the PeBDE, OcBDE, or DeBDE commercial products. The percent ratios were compared to those of PeBDE commercial formulations DE-71 and Bromkal 70-5DE and OcBDE commercial formulations DE-79 and Bromkal 79-8DE (La Guardia et al., 2006). BDE47 and BDE99 were the most abundant congeners from Iqaluit comprising 25 to 66% and 23 to 27%, of the sum of the selected BDE congeners concentration, respectively. This increased percent contribution of BDE47 is an indication of LRAT deposition (deWit et al., 2006).  151  Table 7.3 Mean concentrations (pg/g dw) of five PBDE congeners which are major contributors to the total PBDE concentration in soils from Iqaluit (IQ) and Yellowknife (YELL). Sample  Depth (m)  IQ-2 IQ-3 IQ-4 YELL-1 YELL-2 YELL-3 YELL-4  BDE47  BDE100  BDE99  BDE153  Mean  Stdev  Mean  Stdev  Mean  Stdev  6  41  10.5  21  2  4.3  0.5  0-0.20  3  48  4.4  41  4  7.6  0.6  0.20-0.40  4  38  16.1  21  2  9.4  1.4  0-0.20  4  34  15.4  17  2  3.9  0.5  0.20-0.40  4  52  7.6  24  2  3.8  Blank IQ-1  n  Mean  BDE154  BDE183  Stdev  Mean  Stdev  Mean  8  4.7  1  0.3  81  10.5  4  0.7  7  4.4  2  5  0.7  1  0.8  7  0.8  BDE209  Stdev  Mean  Stdev  12  11  1,010  880  14  8.9  3,960  7,400  1.4  8  4.9  584  613  0.7  6  2.1  520  517  2  1  15  6.7  320  72  0-0.20  5  27  2.2  13  1  1  0.2  4  0.4  1  0.1  5  1.1  540  608  0.20-0.40  5  32  2.9  14  1  0.9  0.3  4  0.6  1  0.3  8  2.7  1,110  716  0-0.20  3  80  26.5  24  2  7.6  0.4  8  2.2  1  0.4  7  6.5  505  363  0.20-0.40  2  26  3.5  14  1  1.6  0.1  4  0.1  0.4  0.5  6  2.5  265  157  0-0.15  2  69  21.4  84  10  11.5  4  19  3.4  8  3.2  8  3.8  3,430  476  0.15-0.30  2  66  27.8  72  9  30.8  6  17  7.2  8  4.2  9  2.8  1,200  1.1  0-0.15  3  131  6.7  171  58  74  62.7  43  17.4  23  12.2  233  351  2,000  1,080  0.15-0.30  3  75  20.7  41  4  12.1  1.4  11  3.9  3  1.7  0.1  0  1,240  1,090  0-0.15  3  60  18.2  29  4  10.7  1.1  8.1  3  2  0.3  10  8.9  1,590  1,300  0.15-0.30  3  19,830  17,120  3,710  3,93  3,210  3,410  3,746  750  3,381  2,930  110  94  2,675  1,580  0-0.15  6  605  380  530  38  300  15.5  100  54  30  13  9  14  1,930  1,580  14,872  17,120  27,821  2,954  32,100  3408  5,182  500  2,536  2,135  83  94  1,730  1,540  0.15-0.30 4 n : Number of samples; Stdev: Standard deviation  152  Table 7.4 Percent of BDE-47, -99, -100, -154, -153, -183 and -209 to total PBDE concentration including BDE-209. Sample Blank IQ-1 IQ-2 IQ-3 IQ-4 YELL-1 YELL-2 YELL-3 YELL-4  Depth (m) 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30  BDE-47 3% 1% 5% 6% 11% 4% 3% 12% 8% 2% 4% 4% 5% 3% 46% 18% 26%  BDE-99 2% 1% 3% 3% 5% 2% 1% 3% 4% 2% 5% 6% 3% 2% 9% 16% 48%  BDE-100 1% 2% 1% 1% 2% 1% 0% 1% 1% 1% 1% 1% 1% 0% 9% 3% 9%  BDE-153 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 2% 0% 0% 9% 1% 5%  BDE-154 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 1% 0% 0% 8% 1% 4%  BDE-183 1% 0% 1% 1% 3% 1% 1% 1% 2% 0% 1% 8% 0% 1% 0% 0% 0%  BDE-209 90% 97% 87% 90% 83% 91% 95% 89% 66% 94% 86% 74% 90% 93% 6% 58% 3%  153  Table 7.5 Percent of BDE-47, -99, -100, -154, -153, and -183 normalized to the their sum of concentration. Sample Blank IQ-1 IQ-2 IQ-3 IQ-4 YELL-1 YELL-2 YELL-3 YELL-4  Depth (m) 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.20 0.20-0.40 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30 0-0.15 0.15-0.30  BDE-47 48% 25% 49% 52% 51% 53% 53% 66% 51% 35% 36% 20% 56% 53% 29% 46% 28%  BDE-99 25% 21% 27% 26% 24% 25% 23% 20% 27% 42% 40% 26% 31% 26% 54% 40% 52%  BDE-100 9% 42% 9% 8% 7% 8% 7% 7% 8% 10% 9% 7% 8% 7% 6% 8% 10%  BDE-154 1% 2% 3% 2% 2% 2% 2% 1% 1% 4% 4% 3% 2% 2% 5% 2% 5%  BDE-153 2% 2% 3% 3% 2% 2% 2% 2% 2% 5% 5% 9% 3% 4% 6% 3% 6%  BDE-183 14% 7% 10% 9% 15% 10% 13% 6% 12% 4% 5% 35% 0% 9% 0% 1% 0%  154  BDE47 and BDE99 in the Yellowknife samples contributed 20 to 56% and 26 to 54%, respectively of ΣPBDE. Soil samples from Yellowknife provide a different distribution of PBDE congeners, with BDE47 and BDE 99 contributing more to the sum of the six common congeners (BDE47, 99, 100, 154, 153 and 183 similar to the PeBDE commercial product, which ~ 38-42% and 45-50% for BDE47 and BDE99, respectively (Sjodin et al., 1998; La Guardia et al., 2006). In addition, the percent contribution in the Yellowknife soil samples is between 6 and 10%, which is similar to its contribution of BDE100 in the PeBDE formulation (Sjodin et al., 1998; La Guardia et al., 2006), hence indicating that PBDE contamination is likely due to local sources, such as disposed waste. Figure 7.6 presents the BDE47/99 ratio at each of the sampling locations from the two solid disposal sites. In Iqaluit, this ratio is 2-3.2, except for one location, the surface layer of IQ-1. BDE47/99 and the percent contribution of these congeners in Table 7.5 suggest that this area is within a region which could be a sink for BDE47. This is likely due to atmospheric transport of lower brominated congeners such as BDE47, which are more volatile than heavier congeners (Gouin et al., 2005), and to the photodebromination of higher-brominated congeners, due to sunlight exposure after melting of snow (Eriksson et al., 2004; Soderstrom et al., 2004). In contrast, most BDE47/99 ratios were ~1 except for YELL-2 in the 0.15-0.30 m layer, and YELL3 in the 0-0.15 m layer. A ratio close to unity is typical of the PeBDE commercial product where the BDE47/BDE99 ratio ranges from 0.7 to 1.2 (Leung et al., 2007). Theses values are also similar to ratios determined in earlier studies of PBDEs in Northern Canada and the European Arctic (Jaward et al., 2004a; Shen et al., 2006). Similar findings were reported by Hassanin et al. (2004) who studied the PBDE distribution pattern in UK and Norwegian remote and rural soils at different latitudes and found BDE47/99 <1 and similar to PeBDE mixtures. This pattern is reversed going north, due to atmospheric transport. This confirms that the PBDE contamination in the soils near the Iqaluit waste disposal site is mainly due to LRAT, whereas that in the soils at 155  the Yellowknife site very likely originates from the waste disposal facility given that their BDE47/99 ratios were similar to those of the PeBDE technical mixtures.  (a) Iqaluit  BDE47/99  (b) Yellowknife  BDE47/99  Figure 7.6 BDE47/99 ratio for soil samples from (a) Iqaluit waste disposal facility and (b) Yellowknife waste disposal facility. Error bars show 95% confidence intervals based on 2 to 6 replicates.  The analysis did not lead to a clear trend of how PBDEs are laterally or vertically distributed in the soil outside the waste disposal facilities. This is likely due to several factors: - Possible external sources other than the landfill which could contribute to PBDE contamination, such as LRAT of PBDEs. - Water infiltration and runoff due to melting snow and seasonally fluctuating groundwater levels.  156  - Heterogeneity of the soil could cause the soil porosity to differ at various locations, allowing water to infiltrate at different rates, and thus PBDEs to be transported at different rates. - High octanol-air partition coefficients (KOA) of PBDEs could lead to seasonally variable concentrations in the surface soil and air. In winter they could accumulate in the soil surface, trapped in the snow and ice, and then volatilize when temperatures rise in the spring and summer, escaping from the soil (Gouin et al., 2005). - Photochemical decomposition, especially of BDE-209, could occur at the surface, producing lower brominated congeners, which could escape from the surface layer (de Wit 2002; Eriksson et al., 2004; Soderstrom et al., 2004). However, the presence of PBDEs in the soil surrounding the waste disposal site indicates potential mobility of these contaminants, whether by atmospheric transport from one location to the other (Gouin and Harner 2003; Hassanin et al., 2004) or by advection as water percolates in the soil due to snow melting.  7.3.4 Comparison to Soils from Different Regions ΣPBDE concentrations at Iqaluit and Yellowknife can be compared to their counterparts in surface soils collected from Northern Canadian waste disposal sites. BDE levels in surface soils from Northern waste disposal sites have previously been measured at Rankin Inlet, Apex Flats, Devon Island and Cambridge Bay and Mt. Relly in Nunavut. The latter is a remote location with no identifiable sources of PBDE contamination (Danon-Schaffer , 2010), and hence can be considered to provide background levels. The concentrations were measured using identical methods of extraction and analysis to our present work. ΣPBDE in the Iqaluit samples is within the same range as their counterparts in Rankin Inlet, Apex Flats, Devon Island and Mt. Relly, from 250 to 4,500 pg/g dw, but 1-3 orders of magnitude lower than at Cambridge Bay. Yellowknife ΣPBDE values are 2-3 times higher than for the other Northern Canadian locations, except for Cambridge Bay (Figure 7.7). 157  Location  ΣPBDE Concentration (pg/g dw)  Figure 7.7 Comparison of ΣPBDE concentrations (pg/g dw) in the soils downgradient of Iqaluit and Yellowknife waste disposal facilities and soils from the Canadian Arctic. RI: Rankin Inlet; AF: Apex Flats; CB: Cambridge Bay; MR: Mount Relly; DI: Devon Island (Danon-Schaffer, 2010).  PBDE levels in Iqaluit soil are similar to the background levels of this region. In view of that, any contamination is likely due to deposition of PBDEs from urban southern regions via LRAT. Deposition of PBDEs due to LRAT is further supported by their increased concentration (Figure 7.5) with increasing organic matter content and decreasing particle size as lighter finer particles are more susceptible to atmospheric transport. On the other hand, PBDEs levels in soils from Yellowknife show high contamination relative to other Arctic soil samples, strongly suggesting contamination due to the waste disposal facility. Figure 7.8 compares the mean BDE47/99 ratios of the Arctic biota, air and soil with those for Iqaluit and Yellowknife samples. High-order predator biota have the highest ratios, almost ten times greater than the Iqaluit and Yellowknife ones. Table 7.6 presents the average 158  concentrations of BDE47 and BDE99 from which BDE47/99 ratios are calculated. The mean BDE47/99 ratio from Yellowknife soil samples is similar to the mean ratio of the Arctic soil samples, whereas the Iqaluit mean ratio is almost 3 times greater, reaffirming that PBDE contamination in Iqaluit soil samples is mostly due to external PBDE sources other than the  BDE47/99  disposal facility.  Figure 7.8 Mean BDE47/99 ratios calculated from data published in studies on PBDE levels in different environmental biota and matrices. Average BDE47 and BDE99 concentrations are presented in Table 7.6. (Lindstrom et al., 1999; Herzke et al., 2001; RAIPON/AMAP/GEF 2001; van Bavel et al., 2001; Christensen et al., 2002; Chernyak et al., 2003; Herzke et al., 2003; Law et al., 2003; Malmquist et al., 2003; Sellstrom et al., 2003; AMAP 2004; Fangstrom et al., 2004; Jaward et al., 2004b; Mariussen et al., 2004; Remberger et al., 2004; Savinova et al., 2004; Vives et al., 2004; Vorkamp et al., 2004a; Vorkamp et al., 2004b; Herzke et al., 2005; Muir et al., 2006).  159  Table 7.6 Average BDE47 and BDE99 concentrations (pg/g) in different environmental matrices in the Arctic used to calculate BDE47/99 ratio in Figure 7.8. Environmental Matrix (units) Air (pg/g)  Average BDE47  Average BDE99  BDE47/BDE99  References  26,796  53,467  1.0  de Wit, et al., 2006; Jaward et al, 2004b  Arctic Soil (pg/g 140 193 0.7 de Wit et al, 2006 dw) Vegetation (pg/g) 57 34 2.0 AMAP, 2004 Terrestrial Mariussen et al., 2004; Animals (pg/g 438 335 1.2 Remberger et al, 2004 lipid wt) Terrestrial Birds Herzke et al., 2001; Herzke et al., 160,800 455,600 0.4 (pg/g lipid wt) 2005; Remberger et al, 2004 Fresh Water Fish 70,477 72,640 1.5 Law et al., 2003 (pg/g lipid wt) Marine Christensen et al., 2002; Invertebrates 6,200 2,800 3.3 Malmquist et al, 2003; AMAP, (pg/g) 2004 Marine Fish RAIPON/AMAP/GEF, 2001; 13,286 693 22.3 (pg/g lipid wt) Johansen et al., 2004 Sea Birds (pg/g 403,729 83,300 6.5 Herzke et al., 2003 lipid wt) Ringed Seal Law et al., 2003; Ikonomoou et 7,908 1,243 14.5 (pg/g lipid wt) al, 2002 Beluga (pg/g Lindstrom et al., 1999; van Bavel 25,220 2,140 11.2 lipid wt) et al., 2001 Polar Bear (pg/g 13,788 1,973 8.3 Law et al., 2003; Muir et al, 2006 lipid wt) Iqaluit Soil (pg/g 48 22 2.2 This study dw) Yellowknife Soil 4,463 8,230 1.1 This study (pg/g dw) Air concentrations are reported in pg/g after converting volumetric mass concentrations using an average air density of 1.3×103 g/m3.  7.4 Conclusion PBDE contamination in Iqaluit is likely due to external sources within either the same geographic region or areas to the south, rather than to locally discarded items. On the other hand, the Yellowknife soil samples indicate that PBDE contamination in this area was mainly due to the waste disposal facility itself. Although PBDE contamination was present in the soils from both Iqaluit and Yellowknife, lateral and vertical trends were inconsistent due to such factors as different environmental conditions and the physical and chemical properties of these compounds. Sources of PBDE contamination in Iqaluit can be mainly linked to long range atmospheric 160  transport, given their inverse and increasing relationship with soil particle size and organic matter content of the soil, and the ratios of BDE47 and 99 to total PBDE concentration. PBDE contamination in Yellowknife can be linked to local PBDE sources, such as the disposed waste, due to the high levels of PBDE contamination and BDE47 and 99 ratios to total PBDE concentrations being similar to the PBDE commercial mixtures.  161  8 PBDE Retention in Sand-Bentonite Liner Material 8.1 Introduction PBDEs incorporated as flame retardants in plastics and textiles are either incinerated or deposited in landfills at the end of their life cycles. PBDEs, which are additives and not chemically bound to the plastic products, readily leach into the landfill environment (Odusanya et al., 2009). In addition, landfills receive biosolids (Bruce and Evan, 2002), solid by-products from wastewater treatment of household sewage, industrial wastewater and storm water runoff. The biosolids have been found to be contaminated with PBDEs. European biosolids have been found to contain 3×103 to 1,185×103 pg/g dw PBDEs (Allchin et al., 1999; Christensen et al., 2002; Matscheko et al., 2002; de Boer et al., 2003; Eljarrat et al., 2008). In North American biosolids, PBDE concentrations were 460×103 to 7×106 pg/g dw (Hale et al., 2001a; Rayne and Ikonomou, 2005; Gorgy et al., 2011). It is estimated (Australian Waste Association [AWA], 2008) that ~19% (~2.0×106 tons/year) of biosolids produced in North America enter landfills, while ~ 47% (~3.0×106 tons/year) of the biosolids produced in Europe are sent to landfills. de Wit (2002) and Oberg et al. (2002) have investigated PBDE concentrations in sewage sludge from several wastewater treatment plants in Sweden, including one receiving leached water from a landfill. The concentrations of the sum of BDE47, 99 and 100 were 400-239×103 pg/g while those of BDE209 were <25×103-11.6×106 pg/g dry weight (dw) basis, respectively. Concentrations of BDE47, 99, 100, 153, 154, 207 and 209, in leachates from 5 landfills in Minnesota ranged widely from non-detectable (nd) to thousands of pg per litre (Oliaei et al., 2002). Osako et al. (2004) found that the sum of the concentrations of BDE47, 99 and 100 in leachate samples from seven Japanese landfills ranged from nd to 4,000 pg/L. Leachate is produced when rainwater percolates through the landfill, dissolving and reacting with organic 162  matter in the waste, i.e., with dissolved organic matter (DOM). DOM acts as a co-solvent in the leachates, enhances the desorption of hydrophobic organic compounds (HOCs) present in the waste (Frankki et al., 2006). Samples containing larger proportions of organic matter were observed to have higher PBDE concentrations (Osako et al., 2004). This observation is consistent with PBDEs having high KOW coefficients, ranging from 104 to 1010 (Palm et al., 2002). Hence, PBDEs can desorb from the waste and enter the leachate stream. An investigation on the leachability of PBDEs from biosolids using leaching column tests found that PBDEs adsorb onto solids suspended within the leachate produced from the infiltrating water through the biosolids. This can also occur to PBDEs found in biosolids and products disposed in landfills. The environmental levels of PBDEs in the leachate may partly arise from leaching from first generation, flame-retarded products dumped in landfills more than a decade ago (Oliaei, 2005). Knowledge of these substances is very limited, preventing environmental authorities from carrying out adequate risk assessments. Wastes containing PBDEs are incinerated, deposited in landfills, discharged to wastewater treatment plants, and emitted into the atmosphere (Oliaei, 2005; Odusanya et al., 2009).  Wastes loaded with such contaminants may move into  groundwater, surface water, air, soil and sediment. PBDEs are discharged into the environment through sewage (North et al., 2004; Rayne and Ikonomou, 2005; Clarke et al., 2010). PBDE wastes buried in landfills may also move off-site into ground water and surface water. The toxicological effects of the PBDEs on humans are unknown. However, laboratory studies on animals indicate that PBDEs, like PCBs, are toxic to the brain, reproductive system and liver and that they disrupt thyroid function (Viberg et al., 2002; Viberg et al., 2003; Hites, 2004). Hite (2004) found that PBDE levels doubled in humans and wildlife over a 5 year period. These findings prompted the banning of two of the three commercial PBDE mixtures in Europe and 163  North America in 2006. However, many products containing PBDEs are still in use, and will end up in landfills when they have outlived their usefulness. Current landfill liner systems are designed primarily as low permeability barriers, commonly incorporating compacted clays or other mineral layers with synthetic membranes in composite or multi-barrier arrangements (Thornton et al., 2001). In most cases, the design criteria require that the clay or mineral component is compacted to achieve a permeability no greater than 10-9 m/s (Seymour, 1992; Thornton et al., 2001). In the last three decades bentonite has often been used as a barrier material because of its low hydraulic conductivity (Li and Denham, 2000; Li and Li, 2001), to prevent landfill leachates from migrating to surrounding areas. Mixing silica sand (D50 of 0.35 mm) with bentonite at a mass ratio of 100:8 can reduce the hydraulic conductivity (k) to as low as 4×10-9 m/s (Li and Li 2001). Bentonite is the primary component of geosynthetic clay liners (GCLs), which consist of a thin layer bentonite sandwiched between two geotextiles, or it may be mixed with an adhesive and attached to a geomembrane (Ruhl and Daniel, 1997). A number of studies have investigated the compatibility of clay material to heavy metal permeants of lead, zinc (Cabral, 1992; Li and Li, 2001), iron and nickel (Peirce et al., 1987) or multiple heavy metals (Li and Li, 2001). Others have investigated the efficacy of clay barriers in containing permeants of organic compounds such as pentachlorophenol, toluene, naphthalene, benzene, methanol, DCM, 1,2-dichloroethane, trichloroethylene, ethanol, isopropyl alcohol and acetic acid (Li and Denham, 2000; Lake and Rowe, 2005; Olgun and Yildiz, 2010). It is notable that most of these studies tested simple solutions of distilled water rather than leachates from landfills. PBDEs are considered hydrophobic since they are virtually insoluble in water (1.11×10-2 to 9×104 pg/L; Palm et al., 2002). Studies have shown that organic compounds, especially HOCs, 164  in the permeating fluid affect k of sand-bentonite mixtures. Chang and Anderson (1968), Sivapullaiah and Sridharan (1984), Smith et al. (1990), Li et al. (1996), Anandarajah (2003) and Paumier et al. (2011) found that k of sand-bentonite mixtures increased by as much as 3 or 4 orders of magnitude when these mixtures were permeated with a leachate containing HOCs relative to deionized water. These changes in k are related to changes in the thickness of the diffuse layer of cations which surround negatively charged bentonite particles (Ruhl and Daniel 1997). HOCs in permeants cause bentonite to shrink due to a decrease in the thickness of the double layer, and thus the diffuse layer of the bentonite particles, allowing them to approach each other more closely and thereby leading to shrinkage (Anandarajah 2003). A double layer consisting of two parallel layers of ions appears on the surface of the bentonite surface. The first layer is the surface charge, which is positive, due to the cations at the surface of the clay particles. The other layer, called the diffuse layer, is in the aqueous region, and is formed from free ions in the fluid under the influence of electric attraction and thermal motion (Overbeek 1952). The clay barrier material surfaces are hydrophilic due to strongly hydrated metal ions, mainly sodium and calcium cations. HOCs, such as PCBs, dioxins and PBDEs, are nonpolar molecules which are repelled from the clay surfaces due to the hydrophilic nature of the clay layers (Anandarajah, 2003). Studies have indicated that bentonite particles tend to coagulate and behave like granular materials in the presence of organic fluids with low dielectric constants. The diffuse layer is also affected (Olgun and Yildiz, 2010). Decreasing dielectric constant of the pore fluid compresses the diffuse double layer thickness around the clay particles, and in turn the electrostatic repulsion which can cause coagulation (Mitchell et al., 1965; Mesri and Olson, 1971). With increasing flame-retarded products disposed in landfills, the efficacy of bentonite liners in containing PBDEs has not been investigated. However, studies (Sai and Anderson, 1992; 165  McCaulou and Huling, 1999) have shown that bentonite-sand admix was incompatible with dense non-aqueous phase liquids (DNAPLs), which are immiscible in water and have low dielectric constant. In addition, bentonite submersed in DNAPLs retained their rigid shape, did not swell, and did not provide a hydraulic barrier (McCaulou and Huling, 1999). Landfills could be a source of PBDEs to the environment if not properly contained. It is important to understand the behaviour of PBDEs in landfill leachates in contact with a sandbentonite barrier. The objectives of this study are to provide data on PBDEs leaching from a sand-bentonite barrier and to determine the effect of PBDE-contaminated leachates on k values of that barrier when it adsorbs PBDEs. Our focus is to investigate leachate originating from biosolids. The results provide insight for the landfill leachate, which contains complicated mixtures of organic and inorganic substances originating from the solid wastes disposed in landfills. Results of this investigation are helpful in assessing the viability of disposing biosolids and consumer products in municipal landfills. It can also assist environmental regulators and city planners in establishing policy and regulations for solid waste disposal facilities, as well as geotechnical engineers in designing containment systems for municipal solid wastes.  8.2 Materials and Methods 8.2.1 Leachate Preparation A biosolids leachate was used in the leaching column tests. The biosolids were from an urban wastewater treatment plant (WWTP) in British Columbia, which cannot be named at its request. Biosolids contain high levels of PBDEs, as presented in Chapters 2 and 4. The leachate was prepared as described in Chapter 4. 850 g of biosolids, expanded by adding 10% in volume deionized water, were inserted in a solvent-washed glass leaching cell (glc) and permeated with deionized water. The permeant was collected in a 4.0 L solvent-washed amber bottle. The 166  biosolids leachate was then filtered using glass microfibre filters with a 1.6 µm retention size to minimize the effects of particulate matter on PBDE mobility in the sand-bentonite (Chapter 4). The filtered leachate was then diluted with 4.0 L of deionized water to form the permeant. Once the permeant was prepared, 2.4 mL of pentaBDE technical mix (50 µg/mL) (Cambridge Isotope Laboratories, Andover, Massachusetts) was dissolved in the permeant to ensure that the leachate contained detectable levels of PBDEs. Sub-samples of the leachate were used to determine the PBDE concentration in the leachate prior to its introduction into the sand-bentonite columns. Leachate quality was examined by measuring pH, total organic carbon (TOC) and total solids (TS). TOC of the permeant was determined by Standard Method 5310B (Standard Methods 1995). The sample was homogenized and a micro portion injected into a heated reaction chamber packed with an oxidative catalyst, barium chromate. TOC was determined by the difference between the total and inorganic carbon measured by an IR analyzer (Shimadzu TOC500, Columbia, USA) with a detection limit of 1 ppm. Sub-samples from the permeant were used to determine TS. 100 ml sub-samples were transferred to ceramic bowls and then dried in an oven at 95 °C until constant weights were obtained. 8.2.2 Barrier Material The barrier material consisted of a 92.5% sand and 7.5% sodium bentonite admix on a dry weight basis. These proportions are similar to those selected by Li (1999), Li and Denham (2000) and Li and Li (2001) to achieve a k of 4×10-9 m/s to complete the investigation in a reasonable time span. The sand was Ottawa silica sand with 10% by mass of the sand particles 243 µm or less in diameter, 50% 328 µm or smaller and 90% 441 µm or smaller. This size distribution was measured using laser diffraction (Mastersizer 2000, Malven Instruments, United Kingdom). The bentonite was sodium bentonite from Canadian Clay Products. 167  The specific surface area was determined using the ethylene glycol monoethyl ether (EGME) method (Cerata and Lutenegger, 2002). 1 g of sample was weighed and then saturated with 3 mL EGME. The excess EGME was then removed to a vacuum desiccator, leaving the EGME formed on the clay surface. After the vacuum process, the sample was weighed every 10–12 h using an analytical balance. If the mass difference between the two measurements was more than 0.001 g, then the sample was placed into the desiccator again for additional vacuum. This process was continued until the sample weight did not change more than 0.001 g. The weight increase due to EGME was used to determine the specific surface area of the clay based on the fact that 2.86×104   g of EGME covers 1 m2 of clay surface with a monomolecular layer (Cerata and Lutenegger,  2002). The CEC was determined according to the modified sodium acetate replacement method (LRRI,1965) and described in Appendix A. The concentrations of Na was measured by an atomic adsorption spectrophotometer, Thermo Jarrell Ash Video 22 (Thermo Fisher Scientific Inc.). The pH of the clay was determined in deionized water. To ensure a homogeneous barrier mixture, eight batches of 9,250 g sand and 750 g bentonite were mixed. Each batch of sand was divided into four sub-batches, each mixed with one quarter of the bentonite, and these were then homogenized into a single batch. Once the barrier material had been prepared in this manner, deionized water was added amounting to 13% by weight of the barrier material. This brought the barrier material to an additional 1% of the optimum moisture content to ensure that the sand-bentonite was close to 100% degree saturation, which was approximately 12%. The optimum moisture corresponds to the maximum unit weight of the sand-bentonite mix, which is essential to provide the barrier material with the lowest possible k, helping its engineering performance, such as preventing the settlement or consolidation of the barrier material (National Research Council, 2007). The optimum moisture content and maximum dry density (ρmax) of the barrier material were 168  determined using the American Standard Test Method (ASTM) D689-91 compaction test for sand-bentonite samples of 11, 12, 13, 14 and 15% moisture content (ASTM, 1995). 8.2.3 Free Swell Test Free swell tests for the bentonite admix were conducted according to (American Society for Testing and material [ASTM], 2001), D5890 to determine the extent of sodium bentonite swelling in the solutions used in the leaching tests. The test will provide insight on the swell behaviour of the admix to PBDE-containing leachate. The degree of swelling is an indication of permeability of the bentonite admix and thus containment ability of the admix to PBDEcontaining leachates. Approximately 90 mL of deionized water, 90 mL pentaBDE spiked deionized water (∼15×10-6 mg/L) and 90 mL of prepared spiked biosolids leachate were each poured into a 100 mL graduated cylinder. The pentaBDE-spiked deionized water was prepared by first dissolving 2.4 mL of the pentaBDE technical mix (50 µg/mL) (Cambridge Isotope Laboratories, Andover, Massachusetts) in 10 mL of acetone and then adding to 90 mL of deionized water to enhance the solubility of PBDEs in the deionized water. 2.0 g of dry powdered bentonite was put into each of the three cylinders in 0.1 g increments. After the bentonite powder was added, deionized water, pentaBDE-spiked deionized water or the prepared leachate was used to dislodge any bentonite particles adhering to the sides of the cylinder, and then filled to the 100 mL mark. The degree of bentonite swelling was measured after 24 h of exposure. 8.2.4 Leaching Column Test The LCT used glass columns 56 mm in length and 100 mm in inner diameter. The barrier material was compacted in three layers in the glass columns according to a modified Proctor procedure (ASTM, 1995), a dynamic process to provide a higher degree of remoulding to  169  remove inconsistencies from uneven mixing and to provide consistent density among the glass columns (Mitchell et al., 1965). The barrier material was compacted to ρmax of 2,640 kg/m3 in seven glass columns enclosed in an aluminium sleeve of the same length, which supported the glass during compaction. To prevent sidewall leakage during permeation/leaching, the inner glass column walls were smeared with a thin layer of bentonite (between 0.8 and 1.0 g dry bentonite with 160-180% water content), prior to inserting and compacting the barrier material (Li and Denham, 2000). Porous ceramic discs of average diameter 100 mm and pore size 60 µm were fixed at both ends of the glass column to ensure that the flow through the cell was uniformly distributed. In addition, 100 mm diameter glass fibre filter paper of 5 µm average pore size was inserted on the inner sides of both porous discs. The glass columns were then capped at both ends by aluminium end-plates as shown in Figure 8.1. Aluminum end-plate Porous ceramic disc with glass fibre filter Glass column Porous ceramic disc with glass fibre filter with glass fibre filter Aluminum end-plate  Figure 8.1 Schematic of glass leaching column used to obtain biosolids leachate and in the leaching column tests.  The permeant flowed from three pressurized aluminium reservoirs to six glass columns in parallel containing the same compacted barrier material, with each tank connected to two glass columns. To begin the leaching, the pressure in the reservoir tank was increased to a gauge pressure of 3.4 kPa (0.5 psi). Every 2 hours, the pressure was increased by ~ 6.9 kPa (1 psi), until it reached a gauge pressure of 49 kPa (7 psi), a pressure chosen to reduce the duration of the test. After passing through the cells, the liquid/leachate from each of the six glass columns was 170  collected in a separate solvent-washed amber glass bottle with a hexane-washed cap. The locations where the copper tubings from the cells passed through the caps of the amber bottles were wrapped with hexane-washed aluminium foil to minimize losses due to volatilization from the leachate collected in the bottles. Amber glass bottles, marked in 100 mL increments and filled with deionized water to a total of 4,000 mL, were used to minimize photodegradation. The bottles were next emptied of the deionized water and washed with 3 × 20 mL of acetone, toluene, hexane, DCM, hexane and DCM. A schematic of the system is presented in Figure 8.2 for one of the glass columns. The leaching column test was conducted for 21 days. To monitor progress, two columns were disconnected from the system after 8 days, then two more after 16 days and the final two after 21 days of permeation. The volumes of leachates from each glass column collected in the glass bottles were recorded during operation at regular time intervals. With the hydraulic gradient imposed by air pressurization, values of hydraulic conductivity, k (m/s), were calculated by means of:  k=  V iAT  (Equation 8.1)  where V is the volume having passed through the cell (cm3), i the hydraulic gradient, A the !  cross-section area of the cell (m2) and T the time of permeation (s). The seventh glass column was connected to an aluminium reservoir containing deionized water only, as a control for the leaching column tests. The set-up procedure for this column was similar to that for the other six glass columns. The leaching duration for the control was 21 days. After terminating the permeation from each cell, the collected leachates were stored in a dark room at 4°C. Large volumes (preferably > 250 mL) of aqueous samples are preferred when analyzing for PBDEs, given that PBDEs are present in low concentrations in the leachate collected due to low water solubilities (Palm et al., 2002), which raises the method detection 171  limits. Therefore, the total volume collected from the LCTs was analysed, rather than collecting smaller aliquots in order to measure the PBDE concentrations more accurately.  Pressure gauge Spiked leachate reservoir  Porous discs with filter paper on inner sides  56 cm  Discharge end Glass leaching cell 100 mm in diameter  Deionized water (permeant)  Leachate collected in amber glass bottles 4L Amber bottle  Figure 8.2 Schematic of leaching column test apparatus.  The barrier material in the glass column was transferred to a fumehood immediately after the column was disconnected from the leaching apparatus. The fumehood had been wiped with hexane solvent and covered with hexane-washed aluminium foil. The barrier material from each glass column was then extruded by 14 mm and sliced using a hexane washed wire saw to provide 4 layers (numbered 1 to 4 from bottom to top (upstream to downstream) from the sample as shown in Figure 8.3. This procedure was carried out in a solvent-wiped fumehood, covered with hexane-washed aluminium foil. The layers were extruded in reverse order to prevent crosscontamination. The wire saw was washed with deionized water, acetone, toluene, hexane and DCM after each slicing to prevent cross contamination. Four samples were taken from the inside  172  area of each layer, combined and stored in hexane-washed amber glass jars at -30oC until analysis. In addition, a sample was taken from each layer to measure the water content.  Flow direction Layer 4, 14 mm thick Layer 3, 14 mm thick Layer 2, 14 mm thick Layer 1, 14 mm thick  Figure 8.3 Layer definition of the cell in each glass column.  8.2.5 Sample Clean-up and Analysis The sand-bentonite samples were processed in batches of 12, each batch consisting of a procedural blank, nine samples, one of which was in duplicate, and a certified reference sample. Prior to extraction, all samples were spiked with 25 µL of the EO5100 standard, a suite of 13Clabeled PBDEs, and  13  C-labeled BDE209 (Cambridge Isotope Laboratories, Andover,  Massachusetts). 10 g from the barrier material samples were homogenized by means of a mortar and pestle with anhydrous Na2SO4, based on 1:1.5 ratio (sample: Na2SO4 m/m), and then transferred to 33 mL stainless steel cells of a Dionex ASE 2000 accelerated solvent extraction system where the solvent was 80:20 (v/v) toluene:acetone. The leachate samples were transferred to separatory funnels and extracted three times with approximately equal volumes of toluene.  173  The sample extracts were next cleaned using a three-step procedure. In the first step, extracts were passed through a multilayer silica column packed with successive layers of silica gel (basic, neutral, acidic, neutral) and eluted with DCM-hexane (1:1 v/v). In the second, a column filled with copper filings and Na2SO4 removed sulphur and residual water, with samples again being eluted with 1:1 v/v DCM-hexane. The third step involved a neutral activated alumina column capped with anhydrous Na2SO4 and washed with hexane, followed by addition of DCM-hexane (1:1 v/v) to elute the analytes of interest. Eluants from the alumina column were concentrated to <10 µL and spiked with 5 µL of the EO5101 standard, a suite of 13C-labeled PBDEs, (Cambridge Isotope Laboratories, Andover, Massachusetts), prior to congener-specific PBDE analyses by high-resolution gas chromatography high-resolution mass spectrometry (HRGC/HRMS) in a VG-Autospec high-resolution mass spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas chromatograph and a CTC A200S autosampler (CTC Analytics, Zurich, Switzerland). The compositions of the internal standards, sorbents, solvents and conditions for sample extraction, in all of the cleanup steps and the quality assurance/ quality control protocols were as reported in detail by Ikonomou et al. (2001). 8.2.6 Data Analysis The analysis procedures permitted the concentrations of 51 of the 209 PBDE congeners to be determined. Congener concentrations in the samples were considered non-detectable (nd) if they were less than five times the concentration of the procedural blanks. In accordance with the standard procedure for the laboratory, congener concentrations measured in the procedural blanks were subtracted from the measured concentrations, which exceeded five times the procedural blank levels. All detected and corrected congener concentrations for each sample analysed were added to provide a total concentration, referred to hereafter as ΣPBDE. To determine trends of congener distributions in the biosolids, leachate and particles, the cumulative 174  sums of specified homologue groups were also calculated. Error bars indicate 95% confidence limits from four repeated measurements on the same sample. Concentrations of the measured PBDE congeners are provided in Appendix F. Of the congeners measured, BDE47, 85, 100, 99, 154, 153, 183 and 209 were most abundant, and each of these was also detected in almost all samples. These congeners are therefore referred to as “principal congeners” below. Detection limits and concentrations of the 51 congeners in the leachate are presented in Table F.2 (Appendix F); detection limits of the congeners in sand-bentonite are presented in Table F.2 (Appendix F). The mass of PBDEs in the permeants and leachates from each cell were estimated by multiplying the concentrations of each PBDE congener by the volumes of leachate collected in the amber glass bottle, and then summed to determine the total mass of PBDEs. The masses of PBDEs retained in the barrier material were estimated by multiplying the concentrations of each measured PBDE congener by the mass of each layer in each cell.  8.3 Results and Discussion 8.3.1 Physical and Chemical Properties The surface area, CEC and pH of the Ottawa sand and Na-bentonite are provided in Table 8.1. The properties for Na-bentonite compared well with these determined by Li and Denham (2000), which were 462 g/m2, 8.38 and 59 cmol/kg, respectively. The average moisture content of the sand-bentonite mixture prior to the leaching column test was 13%, and its organic content was negligible. The average moisture content and organic matter content of the sand-bentonite layers after 8, 16 and 21 days of leaching are presented in Table 8.2. The moisture content varied between 14 and 18% in the sand-bentonite layers. These layers contained little organic matter, 0.4-0.6% dw. 175  Table 8.1 Chemical and physical properties of Ottawa sand and Na-Bentonite. Material Na-Bentonite Ottawa Sand  Specific Surface Area (g/m2) 462 Not available  pH  CEC* (cmol/kg)  8.4 4.7  61 0.1  *Na concentrations used to determine CEC are presented in Table F.1, Appendix F.  Table 8.2 Moisture content and organic content of the sand-bentonite layers during leaching column test. For layer locations, see Figure 8.3. Layer After 8 days Layer 1 Layer 2 Layer 3 Layer 4 After 16 days Layer 1 Layer 2 Layer 3 Layer 4 After 21 days Layer 1 Layer 2 Layer 3 Layer 4  Moisture Content (% weight)  Organic Matter (% dry weight)  16 15 14 16  0.5 0.5 0.6 0.6  18 15 15 17  0.4 0.4 0.5 0.4  18 14 15 17  0.4 0.4 0.5 0.5  The pH, TOC and TSS of the permeant were 7.8, 35 mg/L and 18 mg/L, respectively. Except for TS, these values are within the corresponding ranges reported by Osako et al. (2004) for leachates from Japanese landfills, 6.8-12.7, 1.3-50 and 21-560 mg/L, respectively. The pHs of leachates collected after 8, 16 and 21 days leaching tests were 7.5, 7.2 and 6.9, respectively. TOCs of the leachates in the outlet stream were 28.5, 29 and 29.5 mg/L, respectively, whereas the TS values were 15, 16 and 17 mg/L for the leachates from the leaching column tests after 8, 16 and 21 days, respectively.  176  8.3.2 Swelling Test Results Bentonite swelled 2.5, 2, and 1.5 mL/2g after 24 hours in deionized water, spiked deionized water and spiked biosolids leachate, respectively. Swelling is caused by the mineral interlayer expansion of the bentonite due to hydration (Green et al., 1983; Olgun and Yildiz 2010). The swelling was less for bentonite submerged in the PBDE-spiked deionized water and the PBDEspiked biosolids leachate than when immersed in the un-spiked deionized water. Green et al. (1983) and Olgun and Yildiz (2010) have shown that swelling decreases with decreasing liquid dielectric constants of liquids. The dielectric constant of deionozed water is ∼80 (Fernandez and Quigley 1985; Olgun and Yildiz 2010). The PBDE-spiked deionized water is estimated to have a dielectric constant of ∼74, based on a weighted average of the dielectric constants of water and acetone, 20 for the latter (Green et al., 1983). Since swelling was least for bentonite submerged in the PBDE-spiked biosolids leachate, we can assume that its dielectric constant is less than 74. The degree of swelling of bentonite is also an indication how the hydraulic conductivity of the bentonite will react (an increase in swelling will lead to a decrease in k and visa versa) The swelling test indicates that the k of the sand-bentonite infiltrated by the biosolids leachate in the LCT, will have a higher k than that infiltrated with deionized water. 8.3.3 PBDEs Concentrations in Leachates The concentrations of the eight principal PBDE congeners in the permeant and leachates are reported in Table 8.3. The predominant congeners were BDE47 and 99, contributing 30 and 40%, respectively, of the ΣPBDEs concentration in the permeant and leachates. The total concentrations of the eight principal congeners in the leachates decreased to approximately 4, 18 and 37% of their respective concentrations in the permeant after 8, 16, and 21 days of leaching, respectively, indicating that the sand-bentonite partially adsorbed the PBDEs from the leachate.  177  Table 8.3 Concentrations (pg/L) of eight principal PBDE congeners in the permeant and leachate after 8, 16 and 21 days of leaching through sand-bentonite columns. Matrix  BDE47  BDE85  BDE99  BDE100  BDE153  BDE154  BDE183  BDE209  Permeant Leachate 8 days 16 days 21 days  130×103  9×103  200×103  50×103  20×103  21×103  628  2.8×103  5×103 23×103 50×103  350 1.7×103 3.4×103  7.6×103 36×103 74×103  2×103 9×103 19×103  780 3.7×103 7.6×103  800 4×103 8×103  24 113 234  107 504 1×103  Figure 8.4 shows that the predominant homologue groups in the leachates after 8, 16 and 21 days were tetra and pentaBDEs, representing ~30 and 60% of the total, respectively, with these fractions remaining almost unchanged from the PBDE-contaminated permeant. It is likely that the degradation of PBDEs was insignificant during this period. The ΣPBDEs decreased from approximately 45×104 to 10×104 pg/L after passing 1,850 mL of PBDE-contaminated permeant through the column over 21 days. This decrease indicates that the sand-bentonite mixture in the LCT was able to retain significant amounts of PBDEs, verified by the analytical results for the sand-bentonite samples below. 8.3.4 PBDE Concentrations in Sand-Bentonite The concentrations of eight major PBDEs the sum of the 51 congeners in sand-bentonite layers from the LCTs are compiled in Table 8.4, the concentrations of the 51 measurable congeners in the four layers of the sand-bentonite columns after 8, 16, and 21 days are reported in Tables F.3, F.4 and F.5, respectively of Appendix F. The BDE47 and 99 concentrations were highest in the bottom layer (0-14 mm), whereas BDE99 was most abundant in the succeeding layers. In general, the concentrations of all principal PBDEs increased in all four layers after 21 days of leaching. The concentrations of the eight major PBDEs increased almost threefold in the sandbentonite columns from 8 to 21 days of leaching. BDE183 and 209 were non-detectable in all columns, not surprising since they were almost non-detectable in the leachate.  178  Figure 8.4 PBDE cumulative concentration (pg/L) as a function of degree of bromination in the permeant and leachates.  Table 8.4 Concentrations (pg/g dw) of eight principal PBDE congeners in sand-bentonite pre- and post-leaching. Leaching test duration (days) Sand-bentonite (preleaching)  BDE47  BDE85  BDE99  BDE100  BDE153  BDE154  BDE183  BED209  Total PBDE  nd  nd  nd  nd  nd  1.8  25  nd  nd  Layer 1 (0-14 mm) 8  90  2.9  80  18  6.8  6.5  nd  nd  220  16  220  4.5  140  36  8.4  7.5  nd  nd  440  21  270  7.7  220  53  19  16  nd  nd  680  Layer 2 (14-28 mm) 8  70  2.5  54  13  7  4  nd  nd  160  16  nd  nd  32  9.1  nd  2.2  nd  nd  60  21  nd  nd  40  9.5  4.9  3.5  nd  nd  85  8  nd  nd  nd  nd  4.2  2.3  nd  nd  25  16  70  3.5  46  13  6.2  3.6  nd  nd  160  21  nd  nd  29  nd  nd  2.3  nd  nd  50  Layer 3 (28-42 mm)  Layer 4 (42-56 mm) 8  nd  nd  29  nd  nd  2.3  nd  nd  365  16  nd  nd  nd  nd  nd  0  nd  nd  370  21  160  6.2  150  35  14  12  nd  nd  410  179  The cumulative concentration as a function of the number of PBDE bromines changed with depth/layer in the sand-bentonite columns as shown in Figure 8.5. Cumulative PBDE concentrations in layers 1 (0-14 mm) and 4 (42-56 mm) increased with increasing duration of leaching as shown in the migration profile in Figure 8.6. ΣPBDEs increased from 220 to 440 to 670 pg/g dw, and from 360 to 370 to 410 pg/g dw in layers 1 and 4 respectively from 8 to 16 to 21 days. The accumulation in layer 1 was due to adsorption by the sand-bentonite, as shown by findings from studies on the adsorption of HOCs in permeant fluids by bentonite in sandbentonite (Freitas et al., 2009; Wei et al., 2009; Zhu et al., 2009). However, cumulative PBDE concentrations varied inconsistently in layers 2 (14-28 mm) and 3 (28-42 mm). ΣPBDEs changed from 160 to 60 to 80 pg/g dw and from 24 to 160 to 50 pg/g dw in layers 2 and 3, respectively. This could be due to accumulation of PBDEs in layer 1, during the first 8 days of leaching, followed by gradual increase in PBDE accumulation in layers 2 and 3 after 16 days of leaching resulting from diffusion of PBDEs from layer 1. Organic compounds have been shown to be mobile in clay liners due to diffusion (Shackelford, 1991; Gullick and Weber, 2001). Figure 8.7 plots the cumulative PBDE concentrations of the overall sand-bentonite columns (56 mm), obtained by calculating the weighted average concentrations in the four layers at each leachate measurement time. ΣPBDEs increased from 47,000 to 68,000 to 125,000 pg/g dw after 8, 16 and 21 days of leaching. The cumulative PBDE concentrations in sand-bentonite columns clearly demonstrate that PBDE adsorption increased, as more leachate was eluted.  180  (a) Layer 1: 0-14 mm  (c) Layer 3: 28-42 mm  (b) Layer 2: 14-28 mm  (d) Layer 4: 42-56 mm  Figure 8.5 Cumulative concentrations (pg/g dw) of PBDEs in (a) layer 1: 0-14 mm, (b) layer 2: 14-28 mm, (c) layer 3: 28-42 mm, and (d) layer 4:42-56 mm of the sand-bentonite columns after 8, 16 and 21 days of leaching. Error bars are 95% confidence intervals (n=3).  Figure 8.6 Cumulative PBDE migration profile in sand-bentonite columns.  181  Figure 8.7 Cumulative PBDE concentrations pg/g dw in the sand-bentonite columns (concentrations of all four layers combined). Error bars are 95% confidence intervals (n=3).  8.3.5 PBDE Mass Balance The total masses of PBDEs in the permeant, leachates and the sand-bentonite columns are compiled in Figure 8.8. 4, 18 and 40% of the total mass of the PBDEs in the permeant had passed through the sand-bentonite barrier after 8, 16 and 21 days of leaching. Sand-bentonite retained 12 to 17% of the PBDEs in the permeant. These observations confirm that the sandbentonite adsorbs some PBDEs in the permeant as it percolates through column. However, the total mass of PBDEs retained in the sand-bentonite is much less than the difference between the initial mass of PBDEs in the permeant prior to leaching and in the leachates. This is probably attributable to sorption of PBDEs onto colloidal and suspended solids. Chapter 4 showed that colloidal and suspended solids in leachate/aqueous samples should be filtered and PBDEs extracted separately from the aqueous sample to provide an accurate indication of PBDE levels in leachate samples. The colloids and suspended solids in the collected leachate, which were not collected and analyzed separately from the aqueous portion of the collected leachate, may have 182  retained the PBDEs leading to underestimating the actual PBDE levels which passed through the sand-bentonite barrier.  Figure 8.8 Mass of PBDEs in the leachate post leaching, retained in the sand-bentonite to PBDEs in the leachate pre leaching.  8.3.6 Hydraulic Conductivity and Adsorption Capacity of Sand-Bentonite The hydraulic conductivity (k) of the sand-bentonite from the control LCT (Figure 8.9), was of order 10-7 m/s, in good agreement with k for sand-bentonite in previous LCT experiments (Li and Li, 2001). For the sand-bentonite permeated with PBDE-contaminated leachate, k, between 250 and 1,850 mL of leachate collected (corresponding to 3- 21 days of leaching) varied from 1.0×10-7 to 2.0×10-7 m/s. After passing ~250 mL (~3 days of leaching) of the biosolids leachate through the sand-bentonite, k decreased from 8.5×10-7 to 1.4×10-7 m/s. The k of the sandbentonite permeated by the biosolids leachate was approximately 8.5 times greater than for the sand-bentonite permeated by deionized water. Permeability decreased as more leachate passed due to inter-crystalline swelling of bentonite, as shown by the results of the free swell test 183  described above. The inter-crystalline swelling of bentonite is caused by strong interaction between polar clay surfaces and water molecules (Li et al., 1996). Other factors which may have contributed to decreased k are the blockage of flow-paths within the sand-bentonite resulting from accumulation of suspended solids, biological growth and/or gas production (Ruhl and Daniel 1997). Biological growth is certainly possible given that the leachate is derived from biosolids, which provides the nutrients needed for microbial survival and multiplication. Hydraulic conductivity of the barrier material gradually increased after leaching approximately 1,300 ml. The trend of a decrease and subsequent increase in k is similar to that of clay permeated with methanol, which has a dielectric constant of 32, less than that of water (Anderson et al., 1985; Fernandez and Quigley, 1985). Double-layer theories (Verwey and Overbeek, 1948; Mitchell 1993), explain that a decrease in the fluid’s dielectric constant reduces the double-layer thickness, allowing clay particles to approach each other more closely. This leads to shrinkage of clay clusters. If this occurs for constant, or nearly constant, overall volume, then the permeability increases. As more leachate is passed, the clay particles collapse expelling the water molecules causing an increase in k, similar to observations by Fernandez and Quigley (1985). Although the dielectric constant of the leachate is unknown, we were able to deduce that is it less than deionized water, based on the swelling test results, leading to the shrinkage of the sand-bentonite barrier and ultimately causing k to increase. Another factor contributing to the increase in k is likely to be flocculation of bentonite. Li et al. (1996) hypothesized that bentonite particles can pack more closely as a result of the leachate contents not having strong affinity for the external surfaces of the bentonite. The increased TS of the collected leachate is an indication that flocculation did occur.  184  Figure 8.9 Sand-bentonite hydraulic conductivity and PBDEs in leachate and sandbentonite during leaching tests.  The increases in k of the barrier material and PBDE content in the leachate as more permeant passed through the sand-bentonite show that clay barriers can not retain PBDEs. This study did not investigate the effect of other contaminants such as heavy metals and organic compounds in the leachate on the effectiveness of the sand-bentonite in retaining PBDEs. However, we can speculate on the influence of the other contaminants from published investigations. Biosolids contain a wide array organic compounds (Harrison et al., 2006), affecting the adsorption capacity and k of clay-barrier materials. For many hydrophobic organic chemicals, sorption to biosolids is the primary pathway for removal from wastewater (Bright and Healey 2003). Biosolids from an urban WWTP in British Columbia are certain to contain volatile organics, chlorinated pesticides, PCBs, dioxins/furans, extractable petroleum hydrocarbons, PAHs, phenols, and other contaminants (Bright and Healey 2003). It is expected that these organic compounds, especially those that are soluble, will also be present in the biosolids  185  leachate. Hence, they will have effects similar to those described above, such as causing shrinkage of the clay’s inter-crystalline layers, adsorption on the clay surfaces, formation of channels and increased permeability.  8.4 Conclusions PBDEs are retained to some extent by sand-bentonite. As a result, the concentration of PBDEs within leachates is reduced as they permeate through clay barriers. In our tests, sand-bentonite adsorbed 14% of the PBDEs in the permeant over 21 day period in leaching column tests. The hydraulic conductivity decreased and subsequently increased with time, probably due to an increasing number and size of preferential channels in the sand-bentonite columns, shrinkage of clay interlayers and low adsorption capacity of sand-bentonite. The results indicate that sandbentonite barriers for landfills are not very effective for containment of PBDEs.  186  9  Conclusions and Recommendations  9.1 Conclusions Biosolids from wastewater treatment plants are commonly applied to agricultural soil as amendments because they provide the necessary nutrients for plant growth. This recycles the nutrients to assist growth of crops and reduces the amount of biosolids deposited in municipal landfills. However studies have found that biosolids contain high levels of PBDEs, which are brominated flame retardants used in multiple products, like foam mattresses, televisions, computers, plastics and textiles, which are deposited in landfills at the end of their life-cycle. These chemicals have been found, in rising levels, in humans and animals causing concern due to their similarity to PCBs and dioxins. The thesis focused on understanding the fate and transport of PBDEs in biosolids-amended soils and the effectiveness of clay liners, used as landfill barriers, to retain PBDEs. Research was conducted to determine the concentration and mobility of PBDEs in biosolids and in biosolidsamended soil as well as their retention in clay material. As reported in Chapter 2, PBDEs in biosolids have been investigated since 1988 in Europe, North America, Australia, China and the Middle East. In general North American samples have contained one to two orders of magnitude more PBDEs than European samples. PBDE concentrations in these soils widely varied due to several factors: - different biosolids loadings; - different elapsed times of sampling after the biosolids were applied to the soil; - various sampling, extraction, cleanup and analytical methods; and - targeting and reporting different PBDE congeners 187  Chapter 2 also determined that there was a lack of information on PBDE vertical distribution in biosolids-amended soils. Only one previously published study (Xia et al., 2010) has investigated the PBDEs at depths exceeding 0.20 m. These authors investigated the distribution of the sum of PBDEs over a depth of 1.2 m in fields to which biosolids had been applied. PBDEs decreased sharply with increasing depth. These authors did not report on specific congeners, nor did they investigate the variation of concentrations with time. A preliminary field investigation (Chapter 3) determined the concentration of PBDEs in biosolids-amended agricultural soils to which 20 and 80 t/ha biosolids loading had been applied. Concentration of PBDEs in surface soils (top 0.30 m) which had received a single application of 80 t/ha biosolids were 1-2 orders of magnitude greater than for soil which had received a single application of 20 t/ha of biosolids. Total PBDEs decreased from 10,250 to 220 pg/g dw at the 0.05-1.05 m depths in soils which received 80 t/ha biosolids. Total PBDEs, tetraBDEs, pentaBDEs, hexaBDEs and decaBDE decayed exponentially with depth and had coefficients of correlation ranging from 0.47 to 0.58. Exponential functions plotted to describe the distribution to total PBDEs, tetraBDEs, pentaBDEs, hexaBDEs and decaBDE with CEC of the soil had coefficients of correlation ranging from 0.67 to 0.73. Inconsistent spatial variation of PBDEs in the biosolids-amended soil is probably due to non-uniform biosolids application, heterogeneity of the soil and the unlevelled surface of the field. A laboratory investigation (Chapter 4) was carried out to determine PBDE congener profiles in biosolids, leachability by water, and fractionation of PBDEs in water and suspended solids of different sizes. ΣPBDE increased from non-detectable to 48×106 pg/L in the deionized water after leaching through the biosolids, exceeding their aqueous solubilities, except for BDE47. PBDEs were found to sorb on solid particles suspended in the leachate, allowing PBDEs to be present at concentrations exceeding their aqueous solubilities. The filtration process indicated 188  that PBDEs are associated with fine and ultrafine particles, as their concentration decreased to 95 ×104 and 38×104 pg/L after passing through the fine and ultrafine filters, respectively. Filters of different pore sizes were found to provide a better indication of PBDE levels in the leachate compared to not filtering or using a single filter. PBDEs concentrations were much higher on ultrafine (21×106 pg/g dw) than on fine particles (4×106 pg/g dw), due to greater surface area and higher organic content of the former. This showed that PBDE concentrations may well be underestimated if suspended particulate matter is not considered in the analyses. Chapter 5 described a field investigation where 80 t/ha of biosolids were applied to a fenced agricultural areas, and PBDEs were monitored over a 0.85 m depth for a one-year period. The investigation attempted to limit the influence of external factors, such as ensuring uniform biosolids over the entire study area and a levelled upper surface to prevent surface runoff crosscontamination. PBDEs increased from 80-300 pg/g dw to 300×103-600×103 pg/g dw due to biosolids application, and PBDEs then migrated downwards to depths of at least 0.85 m. PBDE concentrations decreased non-uniformly with depth. Temporal changes to PBDE levels included an exponential decrease in PBDE levels in the topmost biosolids-amended soils layer, and an exponential increase in PBDE levels in the next underlying soil layer over the one-year period. The rate of decrease of total PBDEs in the top 0.00-0.05 m layer was almost two orders of magnitude greater than the rates of increase of total PBDEs in the 0.05-0.25 m and 0.25-0.45 m layers below the surface, indicating that external environmental effects such as sunlight, microbial degradation and temperature likely influence PBDEs in the surface layer. Leaching column tests (Chapter 6) were conducted to determine the mobility of polybrominated diphenyl ethers (PBDEs) in biosolids-amended soils. Deionized water passed upward through a layer of biosolids-amended soils under agricultural soil. The latter was divided into three layers to determine the PBDE distribution along the flow path of the infiltrating water. After 4 weeks of 189  leaching, PBDEs were found to leach from the biosolids-amended soil layer and migrate through the soil. The principal congeners BDE47, 85, 99, 100, 153, 154, 183 and 209 decreased to 398% of their initial concentrations in the biosolids-amended soil, whereas the total PBDE concentration decreased by 38%. PBDE concentrations in the first soil layer increased from nondetectable (nd) to up to 234×103 pg/g dw. Concentrations in the second and third layers increased from nd to 20 and 25 pg/g dw. PBDE in the leachate increased from nd to 310×103 pg/L. The results showed that mobilization of PBDEs is associated with dissolved organic matter and colloids in the infiltrating water as confirmed in Chapter 4. Chapter 7 describes a field study in which PBDE concentrations were examined in soils upgradient and downgradient of two solid waste facilities in Northern Canada, Iqaluit and Yellowknife. Results were highly variable from one site to another, ranging from 450 to 57,400 pg/g dw. There was evidence of migration in the soil. Soils from Iqaluit had concentrations similar to Arctic soil samples analyzed in previous work including background levels, but Yellowknife samples had higher concentrations, 79×103 to 76×108 pg/g dw. The dependence of PBDE concentrations on OM and particle size, as well as the ratios of BDE 47 and 99 to the total PBDE concentration suggest that PBDE contamination in Iqaluit is due to long-range atmospheric transport, whereas that in the Yellowknife soils is mainly from the solid waste facility itself. The adsorption and hydraulic conductivity of bentonite mixed with sand (sand-bentonite), used as clay barrier for landfills, were investigated in leaching column tests to determine whether bentonite can effectively contain and/or immobilize PBDEs in landfill leachate (Chapter 8). The sand-bentonite retained only 11-14% of the total PBDEs in the permeant over a 21 day period. The hydraulic conductivity of the sand-bentonite decreased from 8.5×10-7 to 1×10-7 m/s and then gradually increased to 2.0×10-7 m/s, whereas for the control leaching column test it was 1.0×10-7 190  m/s. The initial decrease in hydraulic conductivity was attributed to the bentonite swelling, while the subsequent increase is due to shrinkage of the clay interlayers caused by the hydrophobicity of the permeant.  9.2 Recommendations Future studies should consider: •  Part of the variability of the results is no doubt due to differing methods of extraction, cleanup, analytical measurement techniques and reporting. This made it difficult to establish trends. Hence, it is important to develop uniform methods in order to determine temporal patterns for PBDEs in biosolids-amended soils.  •  Additional investigations are needed on PBDEs partitioning to the suspended and colloidal matter present in aqueous phases. The relationship between the amount and properties of suspended and colloidal matter on PBDE in the aqueous phase should be determined.  •  Analytical methods measuring PBDEs in aqueous samples should include the particulate matter in the aqueous samples. This will improve the understanding of PBDE fractionating between fixed and mobile phases within the soil environment.  •  The risk of human exposures to PBDEs due to biosolids application should be evaluated  •  Future studies are recommended to investigate the effect of photodegradation, biodegradation and volatilization on the fate of PBDEs in biosolids applied to agricultural soils. Field experiments showed that PBDE losses from the topmost biosolids-amended layer exceeded what was gained in underlying layers. This indicates that PBDEs within the biosolids-amended layer undergo different physical and chemical processes, such as photodegradation, biodegradation and volatilization.  191  •  Management of waste disposal facilities should be re-examined to improve the containment of the solid waste and limit the extent of PBDE contamination on the surrounding environment. The subsequent increase of the hydraulic conductivity of sand-bentonite due to permeating PBDE contaminated leachate also shed light on the necessity of further investigations on the compatibility of sand-bentonite liners to the landfill leachates containing PBDEs to assist regulators and others in establishing policy and regulations for solid waste disposal facilities.  •  Landfill leachates are complex, and thus PBDE interaction with bentonite and other barriers will be influenced by other contaminants in the leachate. These interaction should be investigated in future studies.  192  References Adeel, Z., Luthy, R. G., Dzombak, D. A., Roy, S. B., and Smith, J. R. (1997). "Leaching of PCB compounds from untreated and biotreated sludge-soil mixtures." Journal of Contaminant Hydrology, 28(4): 289-309. Agency for Toxic Substances and Disease Registry (ATSDR). 1998. Toxicological profile for Chlorinated Dibenzo-p-dioxins (CDDs). Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service. Agency for Toxic Substances and Disease Registry (ATSDR) (2000). Toxicological profile for polychlorinated biphenyls (PCBs). U.S. Department of Health and Human Services. Agency for Toxic Substances and Disease Registry (ATSDR) (2004). Toxicological profile for Polybrominated Biphenyls and Polybrominated Diphenyl Ethers. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service. Agency for Toxic Substances and Disease Registry (ATSDR) (2008). Toxicological profile for chlorinated dibenzo-p-dioxins (CDDs). U.S. Department of Health and Human Services. Alaee, M., Muir, D., Cannon, C., Helm, P., Harner, T., and Bidleman, T. (2003). New persistent chemicals in air. Indian and Northern Affairs Canada, Ottawa, ON Canada. Alcock, R. E., Sweetman, A. J., Prevedouros, K., and Jones, K. C. (2003). "Understanding levels and trends of BDE-47 in the UK and North America: an assessment of principal reservoirs and source inputs." Environment International, 29(6): 691-698. Allchin, C. R., Law, R. J., and Morris, S. (1999). "Polybrominated diphenyl ethers in sediments and biota downstream of potential sources in the UK." Environmental Pollution, 105(2): 197-207. American Association of State Highway and Transportation Officials (AASHTO). (1986). AASHTO T 267 Standard method of test for determination of organic content in soils by loss on ignition. American Association of State Highway and Transportation Officials. American Society for Testing and Materials (ASTM) (2001). Standard test method for swell index of clay mineral component of geosynthetic clay liners, D 5890. Philadelphia. Anderson, D. C., Brown, K. W., and Thomas, J. C. (1985). "Conductivity of compacted clay soils to water and organic liquids." Waste Management and Research, 3(4): 339-349. Anderson, T. D., and MacRae, J. D. (2006). "Polybrominated diphenyl ethers in fish and wastewater samples from an area of the Penobscot River in Central Maine." Chemosphere, 62(7): 1153-1160. Andrade, N. A., McConnell, L. L., Torrents, A., and Ramirez, M. (2010). "Persistence of polybrominated diphenyl ethers in agricultural soils after biosolids applications." Journal of Agricultural Food and Chemistry, 58(5): 3077-84.  193  Arshad, M. A., Lowery, B., and Grossman, B. (1996). Physical Tests for Monitoring Soil Quality (Methods for assessing soil quality). Madison. Arctic Monitoring and Assessment Programme (AMAP) (2004). Persistent toxic substances, food security and indigenous peoples of the Russian north. Oslo, Norway. Arctic Monitoring and Assessment Programme (AMAP) (2001). Persistent toxic substances (PTS), food security and indigenous peoples of the Russian North. Arctic Monitoring and Assessment Programme, Oslo, Norway. Anandarajah, A. (2003). "Mechanism controlling permeability change in clays due to changes in pore fluid." Journal of Geotechnical and Geoenvironmental Engineering, 129(2): 163172. Arnold, R. G., Teske, S., Tomanek, M., Engstrom, J., Leung, C., Zhang, J., Banihani, Q., Quanrud, D., Ela, W. P., and Saez, A. E. (2008). "Fate of polybrominated diphenyl ethers during wastewater treatment/polishing and sludge stabilization/disposal." Environmental Challenges in the Pacific Basin, D. O. Carpenter, ed., Blackwell Publishing, Oxford, 394411. Australian Water Association (AWA) (2008). "Fact Sheet-Bioslids." Barrie, L. A., Gregor, D., Hargrave, B., Lake, R., Muir, D., Shearer, R., Tracey, B., and Bidleman, T. (1992). "Arctic Contaminants - Sources, Occurrence and Pathways." Science of the Total Environment, 122(1-2): 1-74. Beck, A. J., Johnson, D. L., and Jones, K. C. (1996). "The form and bioavailability of non-ionic organic chemicals in sewage sludge-amended agricultural soils." Science of the Total Environment, 185(1-3): 125-149. Bergman, Å., and Urika, O. (2001). "Polybrominated flame retardants-A threat?" Organohalogen Compounds, 50: 13-20. Birnbaum, L. S., and Staskal, D. F. (2006). "Brominated flame retardants: Cause for concern?" Environmental Health Perspectives 112(1): 9-17. Bright, D. A., and Healey, N. (2003). "Contaminant risks from biosolids land application: Contemporary organic contaminant levels in digested sewage sludge from five treatment plants in Greater Vancouver, British Columbia." Environmental Pollution, 126(1): 39-49. Bruce, A. M., and Evan, T. D. (2002). Sewage sludge disposal: Operational and environmental issues. FR/R0001, Foundation for Water Research, Bucks. Bromine Science and Environment Forum (BSEF). (2004). Fact sheet on the estimated market demand of brominated flame retardants. Cabral, A. R. (1992). A study of clay compatibility to heavy metal transport in permeability testing. McGill University, Montreal. Canadian Waste Water Association (CWWA) (2010). "FAQ-Biosolids." 194  Cerata, A. B., and Lutenegger, A. J. (2002). "Determination of surface area of fine-grained soils by Ethylene Glycol Monoethyl ether (EGME) method." Geotechnical Testing Journal, 25(3): 1-7. Chapman, H. D. (1965). Methods of soil analysis - Chemical and microbiological properties. Agronomy, C. A. Black, ed., American Institute of Agronomy, Madison, Wisconsin, 891901. Chernyak, S. M., Savinova, T. N., Dahle, S., Matishov, G. G., Hickey, J. P., Begnoche, L. J., and Quintal, R. T. (2003). "Levels and distribution of PBDEs in sediments from the Barents Sea and the Sea of Azov." SETAC 24th Annual Meeting, Austin, Texas. Chiou, C. T., Porter, P. E., and Schmedding, D. W. (1983). "Partition equilibria of nonionic organic compounds between soil organic matter and water." Environmental Science and Technology, 17(4): 227-231. Christensen, J. H., Glasius, M., Pecseli, M., Platz, J., and Pritzl, G. (2002). Polybrominated diphenyl ethers (PBDEs) in marine fish and blue mussels from southern Greenland. Chemosphere, 47(6): 631-638. Christensen, J. H., Groth, B. S., Vikelsøe, J., and Vorkamp, K. (2003). "Polybrominated Diphenyl ethers (PBDEs) in Sewage Sludge and Wastewater " No. 481 Ministry of the Environment, Denmark. Ciparis, S., and Hale, R. C. (2005). "Bioavailability of polybrominated diphenyl ether flame retardants in biosolids and spiked sediment to the aquatic Oligochaete, Lumbriculus variegatus." Environmental Toxicology and Chemistry, 24(4): 916-925. Clarke, B., Porter, N., Symons, R., Marriott, P., Ades, P., Stevenson, G., and Blackbeard, J. (2008). "Polybrominated diphenyl ethers and polybrominated biphenyls in Australian sewage sludge." Chemosphere, 73(6): 980-989. Clarke, B. O., Porter, N. A., Symons, R. K., Marriott, P. J., Stevenson, G. J., and Blackbeard, J. R. (2010). "Investigating the distribution of polybrominated diphenyl ethers through an Australian wastewater treatment plant." Science of the Total Environment, 408(7): 16041611. Covaci, A., Voorspoels, S., and de Boer, J. (2003). "Determination of brominated flame retardants, with emphasis on polybrominated diphenyl ethers (PBDEs) in environmental and human samples - a review." Environment International, 29(6): 735-756. Darnerud, P. O., Eriksen, G. S., Johannesson, T., Larsen, P. B., and Viluksela, M. (2001). Polybrominated diphenyl ethers: occurrence, dietary exposure, and toxicology. Environmental Health Perspectives, 109 (Supplement 1): 49-68. Danon-Schaffer, M. N. (2010). Polybrominated diphenyl ethers in landfills from electronic waste. PhD Thesis. University of British Columbia, Vancouver. de Boer, J., and Cofino, W. P. (2002). "First world-wide interlaboratory study on polybrominated diphenyl ethers (PBDEs)." Chemosphere, 46(5), 625-633. 195  de Boer, J., and Wells, D. E. (2006). "Pitfalls in the analysis of brominated flame retardants in environmental, human and food samples - including results of three international interlaboratory studies." Trac-Trends in Analytical Chemistry, 25(4): 364-372. de Boer, J., Wester, P. G., van der Horst, A., and Leonards, P. E. G. (2003). "Polybrominated diphenyl ethers in influents, suspended particulate matter, sediments, sewage treatment plant and effluents and biota from the Netherlands." Environnemental Pollution, 122(1): 63-74. de la Cal, A., Eljarrat, E., and Barcelo, D. (2003). "Determination of 39 polybrominated diphenyl ether congeners in sediment samples using fast selective pressurized liquid extraction and purification." Journal of Chromatography A, 1021(1-2): 165-173. de Wit, C. A. (2002). "An overview of brominated flame retardants in the environment." Chemosphere, 46(5): 583-624. de Wit, C. A., Alaee, M., and Muir, D. C. G. (2006). "Levels and trends of brominated flame retardants in the Arctic." Chemosphere, 64(2): 209-233. D'Silva, K., Fernandes, A., and Rose, M. (2004). Brominated organic micropollutants - Igniting the flame retardant issue. Critical Reviews in Environmental Science and Technology, 34(2): 141-207. Dobson, R., Schroth, M. H., and Zeyer, J. (2007). "Effect of water-table fluctuation on dissolution and biodegradation of a multi-component, light nonaqueous-phase liquid." Journal of Contaminant Hydrology, 94(3-4): 235-248. Eljarrat, E., and Barcelo, D. (2004). "Sample handling and analysis of brominated flame retardants in soil and sludge samples." Trac-Trends in Analytical Chemistry, 23(10-11): 727-736. Eljarrat, E., de la Cal, A., and Barcelo, D. (2003). "Potential chlorinated and brominated interferences on the polybrominated diphenyl ether determinations by gas chromatography-mass spectrometry." Journal of Chromatography A, 1008(2): 181-192. Eljarrat, E., Marsh, G., Labandeira, A., and Barcelo, D. (2008). Effect of sewage sludges contaminated with polybrominated diphenyl ethers on agricultural soils. Chemosphere, 71(6): 1079-1086. Environment Canada (1999). Georgia Basin Biosolids Inventory and Mercury and Dioxins and Furans Characterization. En40-633/2001E-IN. Environment Commission (2001). European Union risk assessment report: Diphenyl ether, pentabromo derivative. European Commission- Joint Research Centre, Luxembourg. Environment Commission (2003). European Union risk assessment report: Diphenyl ether, octabromo derivative. European Commission-Joint Research Centre, Luxembourg. Eriksson, J., Green, N., Marsh, G., and Bergman, A. (2004). "Photochemical decomposition of 15 polybrominated diphenyl ether congeners in methanol/water." Environmental Science and Technology, 38, 3119-3125. 196  Fangstrom, B., Strid, A., Athanassiadis, I., Grandjean, P., and Weihe, P. "A retrospective study of PBDEs in human milk from the Faroe Islands." The Third International Workshop on Brominated Flame Retardants, Toronto, Canada, 33–36. Fernandez, F., and Quigley, R. M. (1985). "Hydraulic Conductivity of Natural Clays Permeated with Simple Liquid Hydrocarbons." Canadian Geotechnical Journal, 22(2): 205-214. Flury, M., and Qiu, H. (2008). "Modeling colloid-facilitated contaminant transport in the vadose zone." Vadose Zone Journal, 7(2): 682-697. Frankki, S., Persson, Y., Tysklind, M., and Skyllberg, U. (2006). "Partitioning of CPs, PCDEs, and PCDD/Fs between particulate and experimentally enhanced dissolved natural organic matter in a contaminated soil." Environmental Science and Technology, 40(21): 66686673. Gerecke, A. C., Hartmann, P. C., Heeb, N. V., Kohler, H. P., Giger, W., Schmid, P., Zennegg, M., Kohler, M. (2005). "Anaerobic degradation of decabromodiphenyl ether." Environmental Science and Technology, 39(4), 1078–1083. Gevao, B., Muzaini, S., and Helaleh, M. (2008). "Occurrence and concentrations of polybrominated diphenyl ethers in sewage sludge from three wastewater treatment plants in Kuwait." Chemosphere, 71(2): 242-247. Gorgy, T., Li, L. Y., Grace, J. R., and Ikonomou, M. G. (2010). "Polybrominated diphenyl ether leachability from biosolids and their partitioning characteristics in the leachate." Water Air and Soil Pollution, 209(1-4): 109-121. Gouin, T., and Harner, T. (2003). "Modeling the environmental fate of the polybrominated diphenyl ethers." Environment International, 29(6): 717-724. Gouin, T., Harner, T., Daly, G. L., Wania, F., Mackay, D., and Jones, K. C. (2005). "Variability of concentrations of polybrominated diphenyl ethers and polychlorinated biphenyls in air: implications for monitoring, modeling and control." Atmospheric Environment, 39(1): 151-166. Grathwohl, P. (1990). "Influence of organic matter from soils and sediments from various origins on the sorption of some chlorinated aliphatic hydrocarbons: implications on KOC correlations." Environmental Science and Technology, 24(11): 1687–1693. Green, W. J., Lee, G. F., Jones, R. A., and Palit, T. (1983). "Interaction of Clay Soils with Water and Organic-Solvents - Implications for the Disposal of Hazardous Wastes." Environmental Science and Technology, 17(5): 278-282. Greitas, A. F., Mendes, M. F., and Coelho, G. L. V. (2009). "Thermodynamic Study on the Adsorption of Some Organic Acids from Aqueous Solutions by Unmodified and Modified Montmorillonite Clays." Chemical Engineering Communications, 196(7): 824840. Gschwend, P. M., and Wu, S. C. (1985). On the constancy of sediment water partitioncoefficients of hydrophobic organic pollutants. [Article]. Environmental Science and Technology, 19(1): 90-96. 197  Gullick, R. W., and Weber, W. J. (2001). "Evaluation of shale and organoclays as sorbent additives for low-permeability soil containment barriers." Environmental Science and Technology, 35(7): 1523-1530. Hagenmaier, H., She, J., Benz, T., Dawidowsky, N., Dusterhoft, L., and Lindig, C. (1992). "Analysis of sewage-sludge for polyhalogenated dibenzo-p-dioxins, dibenzofurans, and diphenyl ethers." Chemosphere, 25(7-10): 1457-1462. Hale, R. C., Alaee, M., Manchester-Neesvig, J. B., Stapleton, H. M., and Ikonomou, M. G. (2003). "Polybrominated diphenyl ether flame retardants in the North American environment." Environment International: 29(6), 771-779. Hale, R. C., La Guardia, M. J., Harvey, E., and Mainor, T. M. (2002). "Potential role of fire retardant-treated polyurethane foam as a source of brominated diphenyl ethers to the US environment." Chemosphere, 46(5): 729-735. Hale, R. C., La Guardia, M. J., Harvey, E. P., Mainor, T. M., Duff, W. H., and Gaylor, M. O. (2001). Polybrominated diphenyl ether flame retardants in Virginia freshwater fishes (USA). Environmental Science and Technology, 35(23): 4585-4591. Harrison, E. Z., Oakes, S. R., Hysell, M., and Hay, A. (2006). "Organic chemicals in sewage sludges." Science of the Total Environment, 367(2-3): 481-497. Hassanin, A., Breivik, K., Meijer, S. N., Steinnes, E., Thomas, G. O., and Jones, K. C. (2004). "PBDEs in European background soils: Levels and factors controlling their distribution." Environmental Science and Technology, 38(3): 738-745. He, J., Robrock, K. R., Alvarez-Cohen, L. (2006)."Microbial reductive debromination of polybrominated diphenyl ethers (PBDEs)." Environmental Science and Technology, 40(14): 4429-4434. Herzke, D., Berger, U., Kallenborn, R., Nygard, T., and Vetter, W. (2005). "Brominated flame retardants and other organobromines in Norwegian predatory bird eggs." Chemosphere, 61(3), 441-449. Herzke, D., Gabrielsen, G. W., Evenset, A., and Burkow, I. C. (2003). "Polychlorinated camphenes (toxaphenes), polybrominated diphenyl ethers and other halogenated organic pollutants in glaucous gull (Larus hyperboreus) from Svalbard and Bjornoya (Bear Island)." Environmental Pollution, 121(2): 293-300. Herzke, D., Kallenborn, R., Nygard, T., and Sandanger, T. (2001). "Species dependent distribution of polybrominated biphenyls and diphenyl ethers in eggs of Norwegian birds of prey." The Second International Workshop on Brominated Flame Retardants, Stockholm, Sweden, 321–324. Hites, R. A. (2004). "Polybrominated diphenyl ethers in the environment and in people: A metaanalysis of concentrations." Environmental Science and Technology, 38(4): 945-956. Hooper, K., and McDonald, T. A. (2000). "The PBDEs: An emerging environmental challenge and another reason for breast-milk monitoring programs." Environmental Health Perspectives, 108(5): 387-392. 198  Hussein Malkawi, A. I., Alawneh, A. S., and Abu-Safaqah, O. T. (1999). "Effects of organic matter on the physical and the physicochemical properties of an illitic soil." Applied Clay Science, 14(5-6): 270- 278. Ikonomou, M. G., Fraser, T. L., Crewe, N. F., Fisher, M. B., Roger, I. H., He, T., Sather, P. J., and Lamb, P. F. (2001). A comprehensive multi-residue ultra-trace analytical method, based on HRGC/HRMS, for the determination of PCDDs, PCDFs, PCBs, PBDEs, PCDEs, and organochlorines pesticides in six different environmental matrices. 2389, Fisheries and Oceans Canada. Ikonomou, M. G., Rayne, S., and Addison, R. F. (2002). "Exponential increases of the brominated flame retardants, polybrominated diphenyl ethers, in the Canadian arctic from 1981 to 2000." Environmental Science and Technology, 36(9): 1886-1892. Jarvis, N. J., Villholth, K. G., and Ulen, B. (1999). "Modelling particle mobilization and leaching in macroporous soil." European Journal of Soil Science, 50(4): 621-632. Jaward, F. M., Farrar, N. J., Harner, T., Sweetman, A. J., and Jones, K. C. (2004a). "Passive air sampling of PCBs, PBDEs, and organochlorine pesticides across Europe." Environmental Science and Technology, 38: 34-41. Jaward, F. M., Meijer, S. N., Steinnes, E., Thomas, G. O., and Jones, K. C. (2004b). "Further studies on the latitudinal and temporal trends of persistent organic pollutants in Norwegian and UK background air." Environmental Science and Technology, 38(9): 2523-2530. Johansen, P., Muir, D., Asmund, G., and Riget, F. (2004). Contaminants in the traditional Greenland diet. 492, National Environmental Research Institute. Katsoyiannis, A., and Samara, C. (2005). "Persistent organic pollutants (POPS) in the conventional activated sludge treatment process: fate and mass balance." [Article]. Environmental Research, 97(3): 245-257. Knoth, W., Mann, W., Meyer, R., and Nebhuth, J. (2007). "Polybrominated diphenyl ether in sewage sludge in Germany." Chemosphere, 67(9): 1831-1837. Kretzschmar, R., Borkovec, M., Grolimund, D., and Elimelech, M. (1999). "Mobile subsurface colloids and their role in contaminant transport." Advances in Agronomy, 66: 121-193. La Guardia, M. J., Hale, R. C., and Harvey, E. (2006). "Detailed polybrominated diphenyl ether (PBDE) congener composition of the widely used penta-, octa-, and deca-PBDE technical flame-retardant mixtures." [Article]. Environmental Science and Technology, 40(20): 6247-6254. Lake, C. B., and Rowe, R. K. (2005). "A comparative assessment of volatile organic compound (VOC) sorption to various types of potential GCL bentonites." Geotextiles and geomembranes, 23(4): 323-347. Laturnus, F., von Arnold, K., and Gron, C. (2007). "Organic contaminants from sewage sludge applied to agricultural soils - False alarm regarding possible problems for food safety?" Environmental Science and Pollution Research, 14(1): 53-60. 199  Law, R. J., Alaee, M., Allchin, C. R., Boon, J. P., Lebeuf, M., Lepom, P., and Stern, G. A. (2003). "Levels and trends of polybrominated diphenylethers and other brominated flame retardants in wildlife." Environment International, 29(6): 757-770. Leung, A. O. W., Luksemburg, W. J., Wong, A. S., and Wong, M. H. (2007). "Spatial distribution of polybrominated diphenyl ethers and polychlorinated dibenzo-p-dioxins and dibenzofurans in soil and combusted residue at Guiyu, an electronic waste recycling site in southeast China." Environmental Science and Technology, 41(8): 2730-2737. Li, F. (1999). Heavy metal sorption and hydraulic conductivity studies using three types of bentonite admixes. MA Sc thesis, University of British Columbia, Vancouver. Li, J., Smith, J. A., and Winquist, A. S. (1996). "Permeability of earthen liners containing organobentonite to water and two organic liquids." Environmental Science and Technology, 30(10): 3089-3093. Li, L. Y., and Denham, W. T. (2000). "The hydraulic conductivity and adsorptivity of organoclay in a sand-bentonite barrier to hydrophobic organic chemicals." Environmental Technology, 21(12), 1429-1443. Li, L. Y., and Li, F. C. (2001). "Heavy metal sorption and hydraulic conductivity studies using three types of bentonite admixes." Journal of Environmental Engineering, 127(5): 420429. Lindstrom, G., Wingfors, H., Dam, M., and von Bavel, B. (1999). "Identification of 19 polybrominated diphenyl ethers (PBDEs) in long-finned pilot whale (Globicephala melas) from the Atlantic." Archives of Environmental Contamination and Toxicology, 36(3): 355-363. MacDonald, E., Gue, L., and Kooper, K. (2009). An Overview of PBDEs and their Regulatory Status in Canada n Overview of PBDEs and their Regulatory Status in Canada. Ecojustice, David Suzuki Foundation and Canadian Environmental Law Association. Malmquist, C., Bindler, R., Renberg, I., Van Bavel, B., Karlsson, E., Anderson, N. J., and Tysklind, M. (2003). "Time trends of selected persistent organic pollutants in lake sediments from Greenland." Environmental Science and Technology, 37(19): 4319-4324. Mariussen, E., Kalas, J. A., Borgen, A., Nygard, T., and Schlabach, M. (2004). Analysis of brominated flame retardants in liver samples of lynx from the Norwegian biota. SETACEurope, Prague, Czech Republic, 18-22. Marsh, P., Russell, M., and Onclin, C. (2005). The hydrology of lakes in the Western Canadian Arctic – implications to proposed natural gas development. 15th International Northern Research Basins Symposium and Workshop, Luleå to Kvikkjokk, Sweden. Matscheko, N., Tysklind, M., de Wit, C., Bergek, S., Andersson, R., and Sellstrom, U. (2002). "Application of sewage sludge to arable land-soil concentrations of polybrominated diphenyl ethers and polychorinated dibenzo-p-dioxins, dibenzofurans, and biphenyls, and their accumulation in earthworms." Environmental Toxicology and Chemistry, 21(12): 2515-2525. 200  McCarthy, J. F., and Zachara, J. M. (1989). "Subsurface Transport of Contaminants - Mobile Colloids in the Subsurface Environment May Alter the Transport of Contaminants." [Article]. Environmental Science and Technology, 23(5): 496-502. McCaulou, D. R., and Huling, S. G. (1999). "Compatibility of bentonite and DNAPLs." Ground Water Monitoring and Remediation, 19(2): 78-86. McGechan, M. B., and Lewis, D. R. (2002). "Transport of particulate and colloid-sorbed contaminants through soil, part 1: General principles." Biosystems Engineering, 83(3): 255-273. Mesri, G., and Olson, R. E. (1971). "Mechanisms Controlling Permeability of Clays." Clays and Clay Minerals, 19(3): 151-165. Mitchell, J. K. (1993). Fundamentals of soil behaviour, John Wiley & Sons, New York. Mitchell, J. K., Hooper, D. R., and Campanella, R. G. (1965). "Permeability of compacted clay." Journal of the Soil Mechanics and Foundations Division, 91(4): 41-65. Muir, D. C. G., Backus, S., Derocher, A. E., Dietz, R., Evans, T. J., Gabrielsen, G. W., Nagy, J., Norstrom, R. J., Sonne, C., Stirling, I., Taylor, M. K., and Letcher, R. J. (2006). "Brominated flame retardants in polar bears (Ursus maritimus) from Alaska, the Canadian Arctic, East Greenland, and Svalbard." Environmental Science and Technology, 40(2): 449-455. National Research Council (2007). Assessment of the performance of engineered waste containment barriers. Committee to Assess the Performance of Engineered Barriers. The National Academies Press, Washington DC. Nylund, K., Asplund, L., Jansson, B., Jonsson, P., Litzen, K., and Sellstrom, U. (1992). "Analysis of some polyhalogenated organic pollutants in sediment and sewage-sludge." Chemosphere, 24(12): 1721-1730. Oberg, K., Warman, K., and Oberg, T. (2002). Distribution and levels of brominated flame retardants in sewage sludge. Chemosphere, 48(8), 805-809. Odusanya, D. O., Okonkwo, J. O., and Botha, B. (2009). "Polybrominated diphenyl ethers (PBDEs) in leachates from selected landfill sites in South Africa." Waste Management, 29(1): 96-102. Olgun, M., and Yildiz, M. (2010). "Effect of organic fluids on the geotechnical behaviour of a highly plastic clayey soil." Applied Clay Science, 48(4): 615-621. Oliaei, F. (2005). Flame Retardants: Polybrominated Diphenyl Ethers (PBDEs)-Background paper. Minnesota Pollution Control Agency. Oliaei, F., King, P., and Phillips, L. (2002). "Occurrence and concentrations of polybrominated diphenyl ethers (PBDEs) in Minnesota environment." Organohalogen Compounds, 58: 185-188.  201  Osako, M., Kim, Y. J., and Sakai, S. I. (2004). "Leaching of brominated flame retardants in leachate from landfills in Japan." Chemosphere, 57(10): 1571-1579. Overbeek, J. T. G. (1952). Electrochemistry of double layer and the interaction of colloidal particles. Colloid Science, H. R. Kruyt, ed., Elsevier, New York. Palm, A., Cousins, I. T., Mackay, D., Tysklind, M., Metcalfe, C., and Alaee, M. (2002). "Assessing the environmental fate of chemicals of emerging concern: A case study of the polybrominated diphenyl ethers." Environmental Pollution, 117(2): 195-213. Paumier, Touze-Foltz, S.N., Mazeas, L. and Guenne A. (2011)." Quantification of volatile organic compounds diffusion for virgin geosynthetic clay liners and for a GCL after contact with a synthetic leachate." Journal of Geotechnical and Geoenvironmental Engineering. Accepted February 2011. Peng, X. Z., Tang, C. M., Yu, Y. Y., Tan, J. H., Huang, Q. X., Wu, J. P., Chen, S. J., and Mai, B. X. (2009). "Concentrations, transport, fate, and releases of polybrominated diphenyl ethers in sewage treatment plants in the Pearl River Delta, South China." Environment International, 35(2):303-309. Pereg, D., Ryan, J. J., Ayotte, P., Muckle, G., Patry, B., and Dewailly, E. (2003). "Temporal and spatial changes of brominated diphenyl ethers (BDEs) and other POPs in human milk from Nunavik (Arctic) and southern Quebec." Organohalogen Compounds, 61: 127–130. Peirce, J. J., Sallfors, G., Peel, T. A., and Witter, K. A. (1987). "Effects of Selected Inorganic Leachates on Clay Permeability." Journal of Geotechnical and Geoenvironmental Engineering, 113(8): 915-919. Persson, Y., Shchukarev, A., Oberg, L., and Tysklind, M. (2008). "Dioxins, chlorophenols and other chlorinated organic pollutants in colloidal and water fractions of groundwater from a contaminated sawmill site." Environmental Science and Pollution Research, 15(6): 463-471. Rahman, F., Langford, K. H., Scrimshaw, M. D., and Lester, J. N. (2001). "Polybrominated diphenyl ether (PBDE) flame retardants." Science of the Total Environment, 275(1-3): 117. Rayne, S., and Ikonomou, M. G. (2005). Polybrominated diphenyl ethers in an advanced wastewater treatment plant. Part 1: Concentrations, patterns, and influence of treatment processes. Journal of Environmental Engineering and Science, 4(5): 353-367. Reick, R. H. (2004). "Polybrominated diphenyl ethers analysis in fish tissue and other matrices by GC-ECD." LCGC North America, 22(9): 914-925. Remberger, M., Sternbeck, J., Palm, A., Kaj, L., Stromberg, K., and Brorstrom-Lunden, E. (2004). "The environmental occurrence of hexabromocyclododecane in Sweden." Chemosphere, 54(1): 9-21. Richard W. Gullick and Walter J. Weber, J. (2001). "Evaluation of Shale and Organoclays as Sorbent Additives for Low-Permeability Soil Containment Barriers." Environmental Science and Technology , 35(7): 1523-1530. 202  Robrock, K.R., Coelhan, M., Sedlak, D.L., Alvarez-Cohen, L. (2009). "Aerobic biotransformation of polybrominated diphenyl ethers (PBDEs) by bacterial isolates." Environmental Science and Technology, 43 (15), 5705–5715. Robrock, K. R., Korytar, P., Alvarez-Cohen, L. (2008). "Pathways for the anaerobic microbial debromination of polybrominated diphenyl ethers." Environmental Science and Technology, 42 (8), 2845–2852. Ruhl, J. L., and Daniel, D. E. (1997). "Geosynthetic clay liners permeated with chemical solutions and leachates." Journal of Geotechnical and Geoenvironmental Engineering, 123(4): 369-381. Sai, J. O., and Anderson, D. C. (1992). "Barrier Wall Materials for Containment of Dense Nonaqueous Phase Liquid (DNAPL)." Hazardous Waste and Hazardous Materials, 9(4): 317-330. Savinova, T., Savinov, V., Muir, D., Konoplev, A., Alexeeva, L., Surning, V., Samsonov, D., Svetochev, V., Svetocheva, O., Proschemikhin, V., Golikov, A., Belikov, S., and Boltuniv, A. (2004). Contaminant levels in ringed seals from the Russian Arctic. APN2045, Akvaplan-niva, Tromsø, Norway. Schwarzenbach, R. P., Gschwend, P. M., and Imboden, D. M. (2003). Environmental Organic Chemistry, John Wiley and Sons, New Jersey. Sellstrom, U., Soderstrom, G., de Wit, C., and Tysklind, M. (1998). "Photolytic debromination of decabromodiphenyl ether." Organohalogen Compounds, 35: 447-450. Sellstrom, U., Bignert, A., Kierkegaard, A., Haggberg, L., De Wit, C. A., Olsson, M., and Jansson, B. (2003). "Temporal trend studies on tetra-and pentabrominated diphenyl ethers and hexabromocyclododecane in guillemot egg from the Baltic Sea." Environmental Science and Technology, 37(24): 5496-5501. Sellstrom, U., De Wit, A. A., Lundgren, N., and Tysklind, M. (2005). "Effect of sewage-sludge application on concentrations of higher-brominated diphenyl ethers in soils and earthworms." Environmental Science and Technology, 39(23): 9064-9070. Seymour, K. J. (1992). "Landfill Lining for Leachate Containment." Journal of the Institution of Water and Environmental Management, 6(4): 389-396. Shackelford, C. D. (1991). "Laboratory diffusion testing for waste disposal -A review." Journal of Contaminant Hydrology, 7(3): 177-217. Shen, L., Wania, F., Lei, Y. D., Teixeira, C., Muir, D. C. G., and Xiao, H. (2006). "Polychlorinated biphenyls and polybrominated diphenyl ethers in the North American atmosphere." Environmental Pollution, 144(2): 434-444. Sjodin, A., Jakobsson, E., Kierkegaard, A., Marsh, G., and Sellstrom, U. (1998). Gas chromatographic identification and quantification of polybrominated diphenyl ethers in a commercial product, Bromkal 70-5DE. [Article]. Journal of Chromatography A, 822(1): 83-89. 203  Soderstrom, G., Sellstrom, U., De Wit, C. A., and Tysklind, M. (2004). "Photolytic debromination of decabromodiphenyl ether (BDE 209)." Environmental Science and Technology, 38(1): 127-132. Statistics Canada. (2007a). Iqaluit, Nunavut (table). 2006 Community Profiles. Catalogue no. 92-591-XWE, Statistics Canada, Ottawa, ON, Canada. Statistics Canada (2007b). Yellowknife, Northwest Territories (table). 2006 Community Profiles. Catalogue no. 92-591-XWE, Statistics Canada, Ottawa, ON, Canada. Sylvis Environmental. Land application guidelines for the organic matter recylcing regulation and the soil amendment code of practice. Best management practices. Prepared for British Columbia Ministry of Environment. Document # 758-08. Talsness, C. E. (2008). "Overview of toxicological aspects of polybrominated diphenyl ethers: A flame-retardant additive in several consumer products." Environmental Research, 108(2): 158-167. Tchobanoglous, G., Burton, F. L., and Stensel, H. D. (2003). Wastewater engineering: Treatment and reuse, McGraw Hill Companies, Inc, New York, NY. Thornton, S. F., Lerner, D. N., and Tellam, J. H. (2001). "Attenuation of landfill leachate by clay liner materials in laboratory columns: 2. Behaviour of inorganic contaminants." Waste Management and Research, 19(1): 70-88. United States Environmental Protection Agency (USEPA) (1993). Part 503- Standards for the use or disposal of sewage sludge. Office of Waste Management. United States Environmental Protection Agency (USEPA) (2007). Polybrominated diphenylethers (PBDEs) significant new use rule (SNUR) questions and answers. Pollution Prevention and Toxics. VaGulck, J. F., and Rowe, R. K. (2004). "Influence of landfill leachate suspended solids on clog (biorock) formation." Waste Management, 24:723-738. van Bavel, B., Dam, M., Tysklind, M., and Lindstrom, G. (2001). "Levels of polybrominated diphenyl ethers in marine mammals." Organohalogen Compounds, 52: 99-103. Vellend, M., Drummond, E., and Muir, J. (2009). "Ecological differentiation among genotypes of dandelions (Taraxacum officinale)." Weed Science, 57: 410-416. Verwey, E. J. W., and Overbeek, J. T. G. (1948). Theory of the stability of lyophobic colloids. Elsevier, Amsterdam, The Netherlands. Viberg, H., Fredriksson, A., and Eriksson, P. (2002). "Neonatal exposure to the brominated flame retardant 2,2 ',4,4 ',5-pentabromodiphenyl ether causes altered susceptibility in the cholinergic transmitter system in the adult mouse." Toxicological Sciences, 67(1): 104107. Viberg, H., Fredriksson, A., and Eriksson, P. (2003). "Neonatal exposure to polybrominated diphenyl ether (PBDE 153) disrupts spontaneous behaviour, impairs learning and 204  memory, and decreases hippocampal cholinergic receptors in adult mice." Toxicology and Applied Pharmacology, 192(2): 95-106. Vives, I., Grimalt, J. O., Lacorte, S., Guillamon, M., Barcelo, D., and Rosseland, B. O. (2004). "Polyhromodiphenyl ether flame retardants in fish from lakes in European high mountains and Greenland." Environmental Science and Technology, 38(8): 2338-2344. Voice, T. C., Rice, C. P., and Weber, W. J. (1983). "Effect of Solids Concentration on the Sorptive Partitioning of Hydrophobic Pollutants in Aquatic Systems." [Article]. Environmental Science and Technology, 17(9): 513-518 Vorkamp, K., Christensen, J. H., Glasius, M., and Riget, F. F. (2004a). "Persistent halogenated compounds in black guillemots (Cepphus grylle) from Greenland - levels, compound patterns and spatial trends." Marine Pollution Bulletin, 48(1-2): 111-121. Vorkamp, K., Christensen, J. H., and Riget, F. (2004b). "Polybrominated diphenyl ethers and organochlorine compounds in biota from the marine environment of East Greenland." Science of the Total Environment, 331(1-3): 143-155. Wang, Y. W., Zhang, Q. H., Lv, J. X., Li, A., Liu, H. X., Li, G. G., and Jiang, G. B. (2007). "Polybrominated diphenyl ethers and organochlorine pesticides in sewage sludge of wastewater treatment plants in China." Chemosphere: 68(9), 1683-1691. Wania, F., and Dugani, C. B. (2003). "Assessing the long-range transport potential of polybrominated diphenyl ethers: A comparison of four multimedia models." Environmental Toxicology and Chemistry, 22(6): 1252-1261. Webster, L., Russell, M., Adefehinti, F., Dalgarno, E. J., and Moffat, C. F. (2008). "Preliminary assessment of polybrominated diphenyl ethers (PBDEs) in the Scottish aquatic environment, including the Firth of Clyde." Journal of Environmental Monitoring, 10(4): 463-473. Wei, J. M., Zhu, R. L., Zhu, J. X., Ge, F., Yuan, P., He, H. P., and Ming, C. (2009). "Simultaneous sorption of crystal violet and 2-naphthol to bentonite with different CECs." Journal of Hazardous Materials, 166(1): 195-199. Wong, A., Lei, Y. D., Alaee, M., and Wania, F. (2001). "Vapor pressures of the polybrominated diphenyl ethers." Journal of Chemical and Engineering Data, 46(2): 239-242. Wu, S. C., and Gschwend, P. M. (1988). "Numerical Modeling of Sorption Kinetics of OrganicCompounds to Soil and Sediment Particles." Water Resources Research, 24(8): 13731383. Xia, K., Hundal, L. S., Kumar, K., Armbrust, K., Cox, A. E., and Granato, T. C. (2010). "Triclocarban, triclosan, polybrominated diphenyl ethers, and 4-nonylphenol in biosolids and in soil receiving 33-year biosolids application." Environmental Toxicology and Chemistry, 29(3), 597-605. Zhu, R. L., Zhu, L. Z., Zhu, J. X., Ge, F., and Wang, T. (2009). "Sorption of naphthalene and phosphate to the CTMAB-Al-13 intercalated bentonites." Journal of Hazardous Materials, 168(2-3): 1590-1594. 205  Zuurbier, M., Leijs, M., Schoeters, G., Tusscher, G. and Koppe, J. G. (2006). "Children's exposure to polybrominated diphenyl ethers." Acta Paediatrica, 95: 65-70.  206  Appendices Appendix A: Additional Details and Data Related to the Preliminary Field Investigation (Chapter 3) This appendix presents additional information on the preliminary investigation of PBDE mobility in biosolids-amended soil at an experimental farm in Kamloops, British Columbia. It covers the sampling procedure, cation exchange capacity determination (CEC), extraction method for PBDE analysis, and the QA/QC protocols.  A.1 Sampling Sampling was performed on two stages. The first stage took place in December 2004 to provide a preliminary indication of the degree of PBDE contamination in the biosolids-amended soil. The soil samples were collected using 0.1 m diameter stainless core samplers, which were 300 mm in length. The core samplers were washed in the laboratory with 3×50 mL acetone and 3×50 mL toluene. In the field, directly prior to the sampling, the samplers were once more washed with 50 mL toluene. After discarding the first 0.05 m of surface soil, a core sample 0.30 m deep was collected. The samplers were then covered from both ends with toluene washed aluminium foil to avoid contamination from the external environment. Three sample locations were selected, to collect soils from the different application loads. One was from an area did not receive biosolids (reference soil), the second was from an area which received 20 t/ha of biosolids, and the third from an area which received 80 t/ha of biosolids. Each core sample was divided into three 0.10 m layers (Figure A.1). Sub-samples from each layer were stored in hexane-washed amber glass jars at -30ºC until analysis. The second sampling stage in August of 2005 collected samples from five locations: one from an area, referred to as reference soil, where no biosolids had been applied to determine the 207  background levels, and the remaining four from areas that had received 80 t/ha biosolids. The objective of the second sampling program was to determine the PBDE distribution in the biosolids-amended soils over a depth of 1 m. In each of these locations, an open pit was excavated to a depth of 1 m and divided into six layers from which samples were taken corresponding to intervals of 0.00-0.05 m; 0.05-0.25 m; 0.25-0.45 m; 0.45-0.65 m; 0.65-0.85 m; and 0.85-1.05 m. The 0.00-0.05 m layer was excluded from sampling since information on surface treatment was not available. The open pit excavations allowed samples to be obtained from different depths without cross-contamination.  Core sampler Soil column ejected from sampler  Layer 1 Layer 2  3×0.10m layers  Layer 3  Figure A.1 Obtaining the soil column from the core sampler and dividing it into three 0.1 m layers.  A.2 Cation Exchange Capacity The cation exchange capacity (CEC) of the soils and bentonite samples was measured using sodium acetate (NAOAc) method adapted LRRI (1961). 2 to 4 g (dry mass) of material was placed into a centrifuge tube. After adding 33 mL of 1.0 NaOAc solution to the material, the suspension was shaken for 5 min. The sample was then centrifuged at 3,500 rpm for 15 min. After decanting the liquid, this procedure was repeated two more times. 30 mL of 2-propanol was then added to wash the samples, and was again shaken for approximately 5 minutes. After decanting the liquid, this washing procedure was repeated two 208  more times. The 2-propanol rinse was for removing the Na in solution, without affecting Na sorbed onto the soil exchange sites. To displace the sorbed Na from the exchange site, 33 mL of ammonium acetate (NH4OAc) was added to the sample, and was shaken for approximately 5 min. The liquid was then decanted into a 100 mL volumetric flask. This procedure was repeated two more times. Each time, the liquid was decanted into the same volumetric flask. Next the 100 ml flask containing the decanted solution was filled to the 100 mL mark with NH4OAc, and the Na concentration of the solution was determined using Absorption Spectrophotometry. CEC in centimoles of positive charges per kilogram (cmol/kg) was calculated from the following equation: CEC = CNa × Vfl/sample weight × 1g/1,000 mg × 1,000 g/1 kg × 1 mol/23 g × 100 cmol/ 1mol where CNa is the concentration of Na (mg/L) in the volumetric flask and Vfl (L) is the volume of the flask.  A.3 Extraction and Cleanup 10 g samples were pulverized with 30-40 g of Na2SO4. A procedural blank of 50 g Na2SO4 was also pulverized to determine background contamination levels. The samples were then loaded in 33 mL stainless steel extraction cells. Each cell was first loaded with cellulose filter at the bottom, followed by the pulverized samples keeping the level in the cell near one centimetre from the top, then spiked with carbon-13 labelled PBDE internal standards. These standards were 13C-209 and EO5100, which is a mixture of several PBDE congeners (BDE28, 47, 99, 154, 183, and 209). 25 µL of each internal standard was added. Na2SO4 was then added to the top of the cell and covered cellulose filter. Finally the caps were fastened tightly.  209  The samples were loaded onto the ASE 200 carousel, and the extraction solvent reservoir was filled with 80:20 (v/v) toluene: acetone. Prior to the extraction, two cleaning cycles were run in the solvent lines using 4 mL aliquots of 80:20 toluene:acetone, and the rinses were discarded. The extraction schedule for the extraction was then programmed to the ASE 200 system. The extraction program consisted of two cycles. Each cycle included filling the cell with 80:20 toluene:acetone, then heating for nine minutes at 180°C, followed by a static period of 5 min at 2000 psi (13,790 kPa) and then it was flushed with a volume of toluene:acetone over two cycles (50% each). Finally, a purge period of 60 seconds was executed during which nitrogen gas was used to force all of the extract from the cell. The sample extract was collected in a 60 mL vial loaded in the ASE carousel with a final extract volume of approximately 55 mL. Cleanup procedures were performed on the sample extracts to remove interferences such as ketones and sulphurous compounds. Cleanup procedures were conducted according to the Regional Dioxin Laboratory (RDL) protocol (Ikonomou et al., 2001). The extracts were quantitatively transferred to individual 500 mL separatory funnels with 80:20 toluene:acetone. The samples were then washed with 20 mL volumes of concentrated sulphuric acid, discarding aqueous portions, until the organic samples were colourless. The samples were next washed with 40 mL of toluene-extracted water, then washed with 40 mL of 1M potassium hydroxide (KOH). The samples were then washed with 80 mL of NaCl saturated water, followed by a wash of 40 mL NaCl saturated water to clear any emulsions. Finally the aqueous portions were discarded. The water washed samples were gravity filtered through Na2SO4 to remove any aqueous portions that may have remained and collected in 500 mL round bottom flasks. The samples were then reduced to near dryness by rotary evaporation and allowed to dry in the fume hood. The samples  210  were then brought up to 5 mL with 1:1 DCM:Hexane in preparation for five times acidic/basic silica chromatography. Glass columns (one for each sample) were each filled with silica, which was activated, overnight at 170 °C and cooled to room temperature. A portion of the silica was acidified with concentrated sulphuric acid and part was alkalized with 1M sodium hydroxide. Columns were initially conditioned with 225 mL of 1:1 DCM:Hexane. The samples and the procedural blanks were loaded quantitatively with 1:1 DCM:Hexane to the columns passing through neutral, acidic, neutral, and basic silica layers in succession. The columns were then eluted with 275 mL of DCM:Hexane which was collected in 500 mL round bottom flasks. Once more the samples were reduced to near dryness by rotary evaporation and allowed to dry in a fumehood. The samples were then brought up in hexane in preparation for copper clean-up. Glass columns were loaded up to 0.15 m from the bottom with copper granules. The columns were then conditioned with 30 mL of 1M hydrochloric acid, followed by 50 mL toluene extracted water, then 75 mL acetone, and finally 50 mL hexane. Approximately 10 mm of sodium sulphate was added to the top of the columns following the acetone rinse. The samples were then added quantitatively to the columns to remove any sulphur and were eluted with 50 mL of hexane and collected into 250 mL flasks. The samples were then reduced to approximately 5 mL by rotary evaporation in preparation for alumina chromatography. Alumina columns were prepared by adding activated alumina (activated overnight at 170 ºC, and left to cool to room temperature) to a height of 0.20 m. 60 mL of hexane were then added to condition the columns. The samples were then added quantitatively to individual columns and collected in 250 mL flask. Then the columns were further eluted with 25 mL of hexane. The hexane elutions were collected into labelled scintillation vials (two per sample). The desired analytes were then eluted from the column with 60 mL 1:1 DCM:Hexane which was collected 211  into 250 mL round bottom flasks. The samples were then rotary evaporated to approximately 3 mL and transferred quantitatively with 1:1 DCM:Hexane to 15 mL centrifuge tubes, and spiked with identical internal standards identical to those for the samples and procedural blanks. The samples were then further evaporated to 150 µL by nitrogen evaporation. Then they were quantitatively transferred to labelled amber microvials using toluene. The microvials were next reduced to approximately 300 µL by nitrogen evaporation and spiked with 10 µL of the external recovery standard EO4151. Finally, the samples were capped with an aluminum crimp seal with a septum and vortexed prior to analysis by HRGC/HRMS.  A.4 Analysis High-resolution gas chromatography high-resolution mass spectrometry (HRGC/HRMS) was employed for PBDE analyses. The instrument used was a VG-Autospec high-resolution mass spectrometer (Micromass, Manchester, U.K.) equipped with a Hewlett-Packard model 5890 series II gas chromatograph and a CTC A200S autosampler (CTC Analytics, Zurich, Switzerland). For all analyses, the mass spectrometer was operated at 10,000 resolution in the positive ion mode at 39 eV energy and data were acquired in the single ion resolving mode. Analytes were identified by retention time comparison relative to authentic calibration standards. Concentrations were calculated by the internal standard isotope dilution method using mean relative response factors determined from calibration standards, run prior to and following sample analyses. Details on the composition of the internal standards, sorbents, solvents and conditions used during sample extraction and in all of the cleanup steps, as well as the instrumental analyses used and the criteria used for analyte identification and quantification were described in detail by Ikonomou et al. (2001).  212  A.5 QA/QC The analytical methods for PBDEs in water and PBDEs in soil were checked. The linear response ranges for each congener were established by testing duplicate analyses of 5-10 calibration standards. For quality control, each standard was within 7.5% of the nominal value. The limit of detection (LOD) or detection limit (DL) was estimated according to the definition by IUPAC (1976). LOD is defined as corresponding to a signal-to-noise ratio (S/N) of 3. The values extrapolated from calibration standards, and spiked samples are presented in Table A.2. Recoveries for tetra-hexaBDEs were determined from an internal standard BDE77 assuming that this congener was representative of the tetra-hexa BDEs. The soil recovery of BDE209 was determined by standard addition. 200 ng of the congener (corresponding to 100 ng/ml in the final extract) was added to the extraction cell prior to extraction. Interferences were investigated by analyzing the samples without addition of internal and external standards. The samples were divided into batches of 10 samples. In each batch, one blank and one laboratory reference material (LRM) were included. Furthermore, in each batch a sample is analysed in duplicate. Analytical results including average concentration, standard deviation and standard error for samples collected in August 2005 are presented in Table A.3.  213  Table A. 1 Cation Exchange Capacity of Kamloops Reference and Biosolids-Amended Soils. Soil  Number of samples  Depth below surface (m) Samples collected December 2004 20 t/ha 1 0.05-0.35 80 t/ha 1 0.05-0.35 Samples collected August 2005 Reference 2 0.05-0.25 Soil 2 0.25-0.45 2 0.45-0.65 2 0.65-0.85 2 0.85-1.05 4 0.05-0.25 4 0.25-0.45 80 t/ha 4 0.45-0.65 4 0.65-0.85 4 0.85-1.05  Na concentration (mg/L)  Average CEC (meq/100g)  115 184  25 40.0  64.4 23 23 27.6 36.8  14 5 5 6 8  92 36.8 32.2 41.4 50.6  20 8 7 9 11  214  Table A.2 Average Detection Limits (DL) (pg/g) Congener 2-BDE-1 3-BDE-2 4-BDE-3 26-BDE-10 24-BDE-7 24'/33'-BDE-8/11 34-BDE-12 34'-BDE-13 44'-BDE-15 246-BDE-30 24'6-BDE-32 22'4-BDE-17 23'4-BDE-25 244'/2'34-BDE-28/33 33'4-BDE-35 344'-BDE-37 244'6-BDE-75 22'45'-BDE-49 23'4'6-BDE-71 22'44'-BDE-47 23'44'-BDE-66 33'44'-BDE-77 22'44'6-BDE-100 23'44'6-BDE-119 22'44'5-BDE-99 23456-BDE-116 22'44'5-BDE-85 233'44'-BDE-105# 33'44'5-BDE-126 22'44'66'-BDE-155 22'44'56'-BDE-154 22'44'55'-BDE-153 22'344'6'-BDE-140 22'344'5'/2344'56-BDE-138/166 22'344'5'6-BDE-183 22'344'56-BDE-181 (33'44')256/22'6-BDE-190/171 Di(1) Tr(1)  DL (pg/g) NA NA NA 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.4 0.4 0.4 0.4 1.6 1.6 0.2 0.2 0.2 0.2 0.2 0.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.3 0.4  215  Table A.2 (continued) Average Detection Limits (DL) (pg/g). Congener Tr(2) Pe(1) Pe(2) Pe(3) Pe(4)