<%BANNER%>

Magnetic Separation as a Tool for Metal Removal from Soils and Sediments

Permanent Link: http://ufdc.ufl.edu/UFE0021603/00001

Material Information

Title: Magnetic Separation as a Tool for Metal Removal from Soils and Sediments
Physical Description: 1 online resource (222 p.)
Language: english
Creator: Feng, Nan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: chemical, heavy, magnetic, plant, sediment, soil, toxicity
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Heavy metal contamination of soils and sediments is of increasing concern. The main objective of this research was to investigate the effectiveness of a magnetic treatment method for removing heavy metals from contaminated soils and sediments. The treatment approach was based on adsorbing the metal contaminants onto iron filings seed and removing the metal-laden filings by magnetic separation. The heavy metal binding capacity of different types of soils was first assessed using the MetPLATE test which responds specifically to heavy metal toxicity. The binding capacities of Cu, Zn, or Hg in different types of soils followed this trend: clay rich soil > organic rich soils > sandy soils. The magnetic treatment was conducted by using 5% (w/w) iron filings and a 3-h contact time between iron filings and soil. The effectiveness of magnetic separation of Cu, Zn and Cd from spiked soils was evaluated using the MetPLATE assay, the 48-h Ceriodaphnia dubia and the 96-h Selenastrum capricornutum toxicity tests. Results showed that the magnetic treatment method worked best on Cu-spiked soils, followed by the Zn-spiked soils and the Cd-spiked soils. The toxicity removal for Cu varied between 81.4% and 99.9% after a single treatment, where as the toxicity removal for Zn and Cd were, respectively, 81.4%-98.7% and 80.1%- 96.2% after two or three treatments. Chemical analysis indicated that the metals were removed from both the soil matrix and the soil extracts, and the energy dispersive X-ray spectroscopy (EDS) further confirmed the adsorption of Cu, Zn, and Cd onto the added iron filings. No significant reduction in magnetic separation efficiency was observed as regards Cu and Zn removal from aged soils. The retrieved iron filings could be regenerated in 1 N HNO3 for 1 hr and then retreated by 1 M NaOH for 72 hrs prior to reuse. The magnetic separation method also worked effectively in Pb contaminated shooting range soils and artificially spiked sediments. More than 77.8% of the toxicity produced by Pb was removed from five shooting range soils. For Cu-, Zn-, and Hg- spiked sediments, the toxicity removal after a single treatment varied between 83.0% and 99.98%. The type of sediment and metal did not affect the treatment effectiveness, and the metals were removed from both the sediment matrix and the sediment extracts. MetPLATE also showed great potential in predicting heavy metal phytoavailability in different types of soil. Results from a plant study showed that if a soil extract showed approximately 90% inhibition by MetPLATE assay, this soil could probably cause phytotoxicity in lettuce (Lactuca sativa) and Indian mustard (Brassica juncea). The plant study also demonstrated the effectiveness of the proposed magnetic treatment on reducing Cu phytoavailability. After treatment, the growth of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) was significantly enhanced, and the Cu content in plants shoots and roots also significantly decreased after treatment.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nan Feng.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Bitton, Gabriel.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021603:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021603/00001

Material Information

Title: Magnetic Separation as a Tool for Metal Removal from Soils and Sediments
Physical Description: 1 online resource (222 p.)
Language: english
Creator: Feng, Nan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: chemical, heavy, magnetic, plant, sediment, soil, toxicity
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Heavy metal contamination of soils and sediments is of increasing concern. The main objective of this research was to investigate the effectiveness of a magnetic treatment method for removing heavy metals from contaminated soils and sediments. The treatment approach was based on adsorbing the metal contaminants onto iron filings seed and removing the metal-laden filings by magnetic separation. The heavy metal binding capacity of different types of soils was first assessed using the MetPLATE test which responds specifically to heavy metal toxicity. The binding capacities of Cu, Zn, or Hg in different types of soils followed this trend: clay rich soil > organic rich soils > sandy soils. The magnetic treatment was conducted by using 5% (w/w) iron filings and a 3-h contact time between iron filings and soil. The effectiveness of magnetic separation of Cu, Zn and Cd from spiked soils was evaluated using the MetPLATE assay, the 48-h Ceriodaphnia dubia and the 96-h Selenastrum capricornutum toxicity tests. Results showed that the magnetic treatment method worked best on Cu-spiked soils, followed by the Zn-spiked soils and the Cd-spiked soils. The toxicity removal for Cu varied between 81.4% and 99.9% after a single treatment, where as the toxicity removal for Zn and Cd were, respectively, 81.4%-98.7% and 80.1%- 96.2% after two or three treatments. Chemical analysis indicated that the metals were removed from both the soil matrix and the soil extracts, and the energy dispersive X-ray spectroscopy (EDS) further confirmed the adsorption of Cu, Zn, and Cd onto the added iron filings. No significant reduction in magnetic separation efficiency was observed as regards Cu and Zn removal from aged soils. The retrieved iron filings could be regenerated in 1 N HNO3 for 1 hr and then retreated by 1 M NaOH for 72 hrs prior to reuse. The magnetic separation method also worked effectively in Pb contaminated shooting range soils and artificially spiked sediments. More than 77.8% of the toxicity produced by Pb was removed from five shooting range soils. For Cu-, Zn-, and Hg- spiked sediments, the toxicity removal after a single treatment varied between 83.0% and 99.98%. The type of sediment and metal did not affect the treatment effectiveness, and the metals were removed from both the sediment matrix and the sediment extracts. MetPLATE also showed great potential in predicting heavy metal phytoavailability in different types of soil. Results from a plant study showed that if a soil extract showed approximately 90% inhibition by MetPLATE assay, this soil could probably cause phytotoxicity in lettuce (Lactuca sativa) and Indian mustard (Brassica juncea). The plant study also demonstrated the effectiveness of the proposed magnetic treatment on reducing Cu phytoavailability. After treatment, the growth of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) was significantly enhanced, and the Cu content in plants shoots and roots also significantly decreased after treatment.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nan Feng.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Bitton, Gabriel.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021603:00001


This item has the following downloads:


Full Text

PAGE 1

1 MAGNETIC SEPARATION AS A TOOL FO R METAL REMOVAL FROM SOILS AND SEDIMENTS By NAN FENG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Nan Feng

PAGE 3

3 To my family -my parents, Jianguo Feng and Shuying Zhang, for their tremendous love and support in my life; my husband, Shiwei Zha ng, whose love, understand ing, and encouragement will fortify me forever

PAGE 4

4 ACKNOWLEDGMENTS I express my sincerest gratitude to my a dvisor, Dr. Gabriel Bitton, for his support, guidance, inspiration and humor throughout my graduate study in the University of Florida. I also thank the other members of my committee, Dr. Jean Claude Bonzongo, Dr. Willie Harris, and Dr. Angela Lindner for their valuable advice help, and support in the past four years. I thank Dr. Lena Ma and Dr. Uttam Saha for letting me use their greenhouse and providing me shooting range soils. I am also grateful to Li sa Stanley and Dr. Xinde Cao for their assistance with soil characterization. My thanks are extended to Peter Meyers and Craig Watts at Hydrosphere Research. They trai ned me in toxicity tests and provided the starter cultures of Ceriodaphnia dubia and Selenastrum capricornutum I thank Dr. Ali Boularbah for his help. I also extend my gratitude to my fellow students in the Department of Environmental Engineering Sciences for their kindness and help. Special thanks go to my wonderful parents for their tremendous love and support in my life. Finally, my heartfelt a ppreciation goes to my husband, Shiwei Zhang, for his love, help, encouragement, and accompaniment through many l ong nights in the lab. Without them, I could not have achieved this goal.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......12 ABSTRACT....................................................................................................................... ............14 CHAPTER 1 INTRODUCTION..................................................................................................................16 1.1 Problem Statement.......................................................................................................... ..16 1.2 Heavy Metal Behaviors in Soils and Sediments...............................................................17 1.3 Soil and Sediment Re mediation Techniques....................................................................18 1.4 Magnetic Separation........................................................................................................ .21 1.5 Research Objectives and Dissertation Outline.................................................................22 2 TOXICOLOGICAL APPROACH FOR ASSESSING THE HEAVY METAL BINDING CAPACITY OF SOILS........................................................................................31 2.1 Introduction............................................................................................................... ........31 2.2 Material and Methods.......................................................................................................32 2.1.1 Soils Used...............................................................................................................32 2.2.2 Methodology for Assessing Soil Heavy Metal Binding Capacity (SHMBC)........33 2.3 Results and Discussion.....................................................................................................34 2.4 Conclusions................................................................................................................ .......36 3 HEAVY METAL REMOVAL FROM SOILS USING MAGNETIC SEPARATION.........41 3.1 Introduction............................................................................................................... ........41 3.2 Material and Methods.......................................................................................................43 3.2.1 Soils Used...............................................................................................................43 3.2.2 Chemicals Used......................................................................................................43 3.2.3 Recovery of Iron Filings from Soils.......................................................................44 3.2.4 Determination of Iron Filings Concentration.........................................................44 3.2.5 Determination of the Contact Time between Iron Filings and Soil Matrix............45 3.2.6 Magnetic Separation of Heavy Meta ls from Four Metal-Spiked Soils..................45 3.2.7 Toxicity of Soil Extracts.........................................................................................45 3.2.7.1 MetPLATE protocol.....................................................................................46 3.2.7.2 48-hour Ceriodaphnia dubia acute bioassay................................................46 3.2.7.3 96-hour Selenastrum capricornutum chronic toxicity test...........................46 3.2.8 Chemical Analysis..................................................................................................47 3.2.9 Sequential Extraction of Metals from Soils............................................................47

PAGE 6

6 3.2.10 Energy Dispersive X-ray Spectroscopy of Retrieved Iron Filings.......................49 3.2.11 Regeneration of Iron Filings.................................................................................49 3.3 Results and Discussion.....................................................................................................50 3.3.1 Recovery of Iron Filings from Soils.......................................................................50 3.3.2 Effect of Iron Filings Concentrati on and Contact Time on Metal Removal from Soils.....................................................................................................................50 3.3.3 Evaluation of Magnetic Separation of Cu, Zn and Cd from Spiked Soils, Using MetPLATE........................................................................................................51 3.3.4 Evaluation of Magnetic Separation of Cu, Zn and Cd from Spiked Soils, Using Ceriodaphnia dubia Acute Toxicity Test..........................................................55 3.3.5 Evaluation of Magnetic Separation of Cu, Zn and Cd from Spiked Sandy Soils, Using Selenastrum capricornutum Chronic Toxicity Test................................55 3.3.6 Assessment of Metal Removal Effi ciency Using Chemical Analysis....................56 3.3.7 Mass Balance of Metals in Soils............................................................................59 3.3.8 Sequential Extraction of Metals in Soils................................................................59 3.3.9 Energy Dispersive X-ray Spectroscopy of Retrieved Iron Filings.........................61 3.3.10 Regeneration of Iron Filings.................................................................................62 3.4 Conclusions................................................................................................................ .......64 4 EFFECT OF AGING OF METAL-SPIKE D SOILS ON METAL TOXICITY AND REMOVAL USING MAGNE TIC SEPARATION...............................................................88 4.1 Introduction............................................................................................................... ........88 4.2 Material and Methods.......................................................................................................90 4.2.1 Soils Used...............................................................................................................90 4.2.2 Chemicals Used......................................................................................................90 4.2.3 Soil Aging without Wet/Dry Cycle........................................................................90 4.2.4 Soil Aging with Wet/Dry Cycle.............................................................................91 4.2.5 Toxicity of Aged Soil Extracts...............................................................................91 4.2.6 Magnetic Treatment of Aged Soils.........................................................................91 4.3 Results and Discussion.....................................................................................................92 4.3.1 Effect of Aging on Cu and Zn Toxicity..................................................................92 4.3.1.1 Toxicity of Cu and Zn in aged sandy soil....................................................92 4.3.1.2 Toxicity of Cu and Zn in aged organic rich soil..........................................94 4.3.2 Effect of Aging on Magnetic Treatment................................................................95 4.4 Conclusions................................................................................................................ .......97 5 LEAD REMOVAL FROM SHOOTING RANGE SOILS USING MAGNETIC SEPARATION.....................................................................................................................100 5.1 Introduction............................................................................................................... ......100 5.2 Material and Methods.....................................................................................................102 5.2.1 Soils and Chemicals Used....................................................................................102 5.2.2 Toxicity of Soils under Study...............................................................................102 5.2.3 Magnetic Separation of Pb from Shooting Range Soils.......................................102 5.2.4 Chemical Analysis................................................................................................103 5.3 Results and Discussion...................................................................................................103

PAGE 7

7 5.3.1 Toxicity of Soils under Study...............................................................................103 5.3.2 Evaluation of Magnetic Separation of Pb from Shooting Range Soils, Using MetPLATE and Ceriodaphnia dubia Acute Toxicity Tests......................................105 5.3.3 Assessment of Pb Removal Effi ciency Using Chemical Analysis.......................106 5.4 Conclusions................................................................................................................ .....106 6 HEAVY METAL REMOVAL FROM SEDIMENTS USING MAGNETIC SEPARATION.....................................................................................................................111 6.1 Introduction............................................................................................................... ......111 6.2 Material and Methods.....................................................................................................113 6.2.1 Sediments Used....................................................................................................113 6.2.2 Chemicals Used....................................................................................................114 6.2.3 Sediment Heavy Metal Binding Capacity............................................................114 6.2.4 Magnetic Separation of Heavy Meta ls from Metal-Spiked Sediments................115 6.2.5 Toxicity of Sediment Extracts..............................................................................116 6.2.6 Chemical Analysis................................................................................................116 6.3 Results and Discussion...................................................................................................117 6.3.1 Sediment Heavy Metal Binding Capacity............................................................117 6.3.2 Use of MetPLATE to Evaluate the Effectivene ss of Magnetic Separation of Cu, Zn and Hg from Spiked Sediments.....................................................................119 6.3.3 Use of Ceriodaphnia dubia Acute Toxicity Test to Evaluate the Effectiveness of Magnetic Separation of Cu, Zn and Hg from Spiked Sediments..........................120 6.3.4 Assessment of Metal Removal Effi ciency Using Chemical Analysis..................121 6.3.5 Mass Balance of Metals in Sediments..................................................................123 6.4 Conclusions................................................................................................................ .....126 7 PLANT GROWTH STUDY TO DEMONSTRATE METAL REMEDIATION BY MAGNETIC SEPARATION...............................................................................................138 7.1 Introduction............................................................................................................... ......138 7.2 Material and Methods.....................................................................................................141 7.2.1 Assessment of Cu Phyt otoxicity Using MetPLATE............................................141 7.2.1.1 Soils used....................................................................................................141 7.2.1.2 Soil spiking with Cu...................................................................................141 7.2.1.3 Pot experiment............................................................................................141 7.2.1.4 Toxicity of soils, as determined by MetPLATE.........................................142 7.2.2 Use of Plants to Evaluate the Eff ectiveness of Magnetic Separation on CuSpiked Soils...............................................................................................................143 7.2.2.1 Soils preparation.........................................................................................143 7.2.2.2 Treatment of spiked soils...........................................................................143 7.2.2.3 Pot experiment............................................................................................144 7.2.2.4 Toxicity of soils used for growing plants...................................................144 7.3 Results and Discussion...................................................................................................145 7.3.1 Assessment of Cu Phyt otoxicity Using MetPLATE............................................145 7.3.1.1 Cu phytotoxicity.........................................................................................145 7.3.1.2 Toxicity of soils, as determined by MetPLATE.........................................148

PAGE 8

8 7.3.1.3 Cu uptake by lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea )................................................................................................................150 7.3.2 Evaluation of Iron Treatment on Cu -Spiked Soils Using Plant Study.................152 7.3.2.1 Effect of treatments on pl ant growth in a sandy soil..................................152 7.3.2.2 Effect of treatments on plan t growth in organic soil 2...............................154 7.3.3 Toxicity of Iron Treated soils as Determined by MetPLATE.............................155 7.3.4 Copper Uptake by Plants in Treated and Non-treated Soils.................................156 7.3.4.1 Copper uptake by plants grown in sandy soil............................................157 7.3.4.2 Copper uptake by plants grown in organic rich soil 2................................158 7.4 Conclusions................................................................................................................ .....159 8 SUMMARY AND CONCLUSIONS...................................................................................174 8.1 Summary.................................................................................................................... .....174 8.2 Conclusions................................................................................................................ .....175 APPENDIX A DETAILED PROCEDURE FO R TOXICITY TESTS........................................................178 A.1 MetPLATE Procedure....................................................................................................178 A.2 48-h Ceriodaphnia dubia Acute Toxicity Test..............................................................180 A.2.1 Preparation of culture medium and food.............................................................180 A.2.2 Maintenance of Ceriodaphnia dubia cultures.....................................................181 A.2.3 Test procedure.....................................................................................................181 A.3 96-h Selenastrum capricornutum Chronic Toxicity Test..............................................182 A.3.1 Preparation of algal medium...............................................................................182 A.3.2 Maintenance of Selenastrum capricornutum cultures........................................182 A.3.3 Algal assay procedure.........................................................................................183 B DETAILED PROCEDURE FOR TOTAL METAL ANALYSIS.......................................185 B.1 U.S. EPA Method 3010A..............................................................................................185 B.2 U.S. EPA Method 3050B...............................................................................................186 B.3 Total Mercury Determination........................................................................................187 B.4 Plant Digestion for Total Metal Analysis......................................................................187 C ADDITIONAL MATERIALS FOR PLANT STUDY.........................................................189 LIST OF REFERENCES.............................................................................................................196 BIOGRAPHICAL SKETCH.......................................................................................................222

PAGE 9

9 LIST OF TABLES Table page 1-1 Health effects of selected heavy metals on humans...........................................................24 1-2 US and FL regulations on drinking water and soil levels of Cu, Cd, Zn, Pb and Hg........24 1-3 Background level of Cu, Zn, and Cd in natural soils and plants.......................................25 1-4 Chemical species of trace metals with re gard to their bioavailability and potential toxicity to organisms..........................................................................................................25 1-5 Summary of soil re mediation techniques...........................................................................26 1-6 US EPA categories of treatments potentially applicable to sediments..............................29 1-7 Applications of magnetic separation..................................................................................30 2-1 Soils characteristics...................................................................................................... ......38 2-2 EC50s as determined by MetPLATE of water extracts from five soils and Ottawa sand........................................................................................................................... .........38 3-1 Soils characteristics...................................................................................................... ......66 3-2 Effect of iron filings co ncentration and cont act time between iron filings and soil matrix on the removal of heavy metals from a spiked sandy soil, as determined by the MetPLATE toxicity test...............................................................................................67 3-3 Copper, zinc and cadmium toxicity removal from a sa ndy soil by magnetic treatment, as determined by MetPLATE...........................................................................69 3-4 Copper, zinc and cadmium toxicity removal from a red sandy soil by magnetic treatment, as determined by MetPLATE...........................................................................70 3-5 Copper, zinc and cadmium toxicity rem oval from an organic rich soil by magnetic treatment, as determined by MetPLATE...........................................................................71 3-6 Copper, zinc and cadmium toxicity re moval from a Georgia clay rich soil by magnetic treatment, as determined by MetPLATE............................................................72 3-7 Effect of magnetic treatment on the remova l of Cu, Zn, and Cd from four soils as determined by the 48-h acute Ceriodaphnia dubia toxicity test........................................73 3-8 Effect of magnetic treatment on the remova l of Cu, Zn, and Cd from four soils as determined by the 96-h chronic Selenastrum capricornutum toxicity test........................75

PAGE 10

10 3-9 Effect of magnetic trea tment on the removal of Cu2+ from spiked soils, as determined by chemical analysis..........................................................................................................77 3-10 Effect of magnetic trea tment on the removal of Zn2+ from spiked soils, as determined by chemical analysis..........................................................................................................78 3-11 Effect of magnetic trea tment on the removal of Cd2+ from spiked soils, as determined by chemical analysis..........................................................................................................79 3-12 Mass balance of Cu in spiked soil s before and after magnetic treatment..........................80 3-13 Mass balance of Zn in spiked soil s before and after magnetic treatment..........................81 3-14 Mass balance of Cd in spiked soil s before and after magnetic treatment..........................82 3-15 Effect of magnetic treatment on the re moval of Cu, Zn, and Cd from each soil fraction, as determined by sequential extraction................................................................85 3-16 Effect of contact time be tween iron filings and 1 N HNO3 on the recovery of fresh iron filings................................................................................................................... .......87 3-17 Comparison of the toxicity of sandy soil ex tracts treated with fresh iron filings and regenerated iron filings, as determined by MetPLATE.....................................................87 3-18 Comparison of the toxicity of sandy soil extracts treated by fresh iron filings and regenerated iron filings, as determined by the 48-h acute Ceriodaphnia dubia toxicity test.................................................................................................................. .......87 5-1 Soils characteristics...................................................................................................... ....108 5-2 Lead toxicity removal from five shoo ting range soils by magnetic treatment, as determined by MetPLATE...............................................................................................109 5-3 Lead toxicity removal from five shoo ting range soils by magnetic treatment, as determined by the 48-h acute Ceriodaphnia dubia test...................................................109 5-4 Effect of magnetic treatment on the rem oval of Pb from shooting range soils, as determined by chemical analysis.....................................................................................110 6-1 Sediments characteristics.................................................................................................128 6-2 EC50s as determined by MetPLATE of water extracts from four sediments and Ottawa sand.................................................................................................................... ..128 6-3 Copper, zinc and mercury toxicity re moval from a spiked sandy sediment by magnetic treatment, as determined by MetPLATE..........................................................130 6-4 Copper, zinc and mercury toxicity remova l from a spiked organic rich sediment by magnetic treatment, as determined by MetPLATE..........................................................131

PAGE 11

11 6-5 Copper, zinc and mercury toxicity re moval from a spiked sandy sediment by magnetic treatment, as determined by the 48-h acute Ceriodaphnia dubia toxicity test........................................................................................................................... .........132 6-6 Copper, zinc and mercury toxicity remova l from a spiked organic rich sediment by magnetic treatment, as determined by the 48-h acute Ceriodaphnia dubia toxicity test........................................................................................................................... .........133 6-7 Effect of magnetic trea tment on the removal of Cu2+, Zn2+ and Hg2+ from a spiked sandy sediment, as determined by chemical analysis......................................................134 6-8 Effect of magnetic trea tment on the removal of Cu2+, Zn2+ and Hg2+ from a spiked organic rich sediment, as dete rmined by chemical analysis............................................135 6-9 Mass balance of Cu, Zn and Hg in a spiked sandy sediment before and after magnetic treatment...........................................................................................................136 6-10 Mass balance of Cu, Zn and Hg in a spik ed organic rich sediment before and after magnetic treatment...........................................................................................................137 7-1 Soils Characteristics...................................................................................................... ...161 7-2 Copper toxicity in spiked sandy soil, or ganic rich soil, and mixed soil used for growing plants, as determined by MetPLATE................................................................165 7-3 Copper uptake by lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in spiked sandy soil and organic soil, as determined by chemical analysis.........166 7-4 Effect of different treatments on copper to xicity in sandy soil and organic rich soil 2 used for growing plants, as determined by MetPLATE..................................................172 7-5 Copper uptake by lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in treated and non-tre ated sandy soil, as determined by chemical analysis..........173 7-6 Copper uptake by lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in treated and non-treated organic rich soil 2, as determined by chemical analysis....................................................................................................................... ......173 A-1 Chemical parameters of moderately hard water (MHW)................................................181 A-2 Components of preliminary alga l assay procedure (PAAP) medium..............................183

PAGE 12

12 LIST OF FIGURES Figure page 1-1 Interactions between soil c onstituents and heavy metals...................................................25 2-1 Soil HMBC (SHMBC) methodology.................................................................................39 2-2 SHMBC for three metals (Cu, Zn, Hg) and five soils.......................................................40 3-1 Recovery of iron filings from soils under different condition...........................................68 3-2 Distribution of Cu, Zn, and Cd fractions in a spiked sandy soil........................................83 3-3 Distribution of Cu, Zn, and Cd fract ions in a spiked organic rich soil..............................84 3-4 Energy dispersive x-ray spect roscopy (EDS) of iron filings.............................................85 4-1 Toxicity of Cu and Zn in an aged sandy soil and an organic rich soil over a 4-month period and 20 wet-dry cycles.............................................................................................98 4-2 Effect of soil aging on magnetic treatment of spiked sandy soil and organic rich soil over a 4-month period and 20 wet-dry cycles....................................................................99 5-1 Toxicity of shooting ra nge soil extracts, as dete rmined by MetPLATE assay................108 6-1 Sediment HMBC for three metals (Cu, Zn, Hg) and four sediments (Ottawa sand served as the reference)....................................................................................................129 6-2 Sediment HMBC for three metals (Cu, Zn, Hg) and three sediments (Little Hatchet Creek sediment served as the reference)..........................................................................129 7-1 Effect of Cu concentrations on dry biomass of shoots and roots of lettuce ( Lactuca sativa ) grown in spiked sandy soil...................................................................................161 7-2 Effect of Cu concentrations on dry biom ass of shoots and root s of Indian mustard ( Brassica juncea ) grown in spiked sandy soil.................................................................162 7-3 Effect of Cu concentrati ons on shoots length of lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in spiked sandy soil...................................................162 7-4 Effect of Cu concentrations on dry biomass of shoots and roots of lettuce ( Lactuca sativa ) grown in spiked organic rich soil.........................................................................163 7-5 Effect of Cu concentrations on dry biom ass of shoots and root s of Indian mustard ( Brassica juncea ) grown in spiked organic rich soil.......................................................163 7-6 Effect of Cu concentrati ons on shoots length of lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in spiked organic rich soil..........................................164

PAGE 13

13 7-7 Effect of Cu concentrations on dry biomass of shoots and roots of lettuce ( Lactuca sativa ) grown in spiked mixed soil..................................................................................164 7-8 Effect of Cu concentrati ons on shoots length of lettuce ( Lactuca sativa ) grown in spiked mixed soil.............................................................................................................165 7-9 Effect of different treatments on dry biomass of shoots and roots of lettuce ( Lactuca sativa ) grown in a sandy soil...........................................................................................167 7-10 Effect of different treatments on dry biom ass of shoots and roots of Indian mustard ( Brassica juncea ) grown in a sandy soil..........................................................................168 7-11 Effect of different treatme nts on shoots length of lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in sandy soil...............................................................169 7-12 Effect of different treatments on dry biomass of shoots and roots of lettuce ( Lactuca sativa ) grown in organic rich soil 2.................................................................................170 7-13 Effect of different treatments on dry biom ass of shoots and roots of Indian mustard ( Brassica juncea ) grown in organic rich soil 2................................................................171 7-14 Effect of different treatme nts on shoots length of lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) grown in organic rich soil 2..................................................172 A-1 MetPLATE protocol........................................................................................................179 C-1 Phytotoxicity of 100 mg/kg Cu to lettuce ( Lactuca sativa ) in sandy soil after 4 weeks exposure....................................................................................................................... ....189 C-2 Phytotoxicity of 100 mg/ kg Cu to Indian mustard ( Brassica juncea ) in sandy soil after 4 weeks exposure.....................................................................................................189 C-3 Effect of iron treatmen t on the growth of Lettuce ( Lactuca sativa ) in sandy soil...........190 C-4 Effect of iron treatment on the growth of Indian mustard ( Brassica juncea ) in sandy soil........................................................................................................................... .........192 C-5 Effect of iron treatmen t on the growth of lettuce ( Lactuca sativa ) in organic rich soil 2.............................................................................................................................. ..........193 C-6 Effect of iron treatment on the growth of Indian mustard ( Brassica juncea ) in organic rich soil 2.................................................................................................................... ......194 C-7 Effect of iron treatment on plan t roots in organic rich soil 2...........................................195

PAGE 14

14 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MAGNETIC SEPARATION AS A TOOL FO R METAL REMOVAL FROM SOILS AND SEDIMENTS By Nan Feng December 2007 Chair: Gabriel Bitton Major: Environmental Engineering Sciences Heavy metal contamination of soils and sedi ments is of increasing concern. The main objective of this research was to investigate th e effectiveness of a magnetic treatment method for removing heavy metals from contaminated soils and sediments. The treatment approach was based on adsorbing the metal contaminants onto iron filings seed and removing the metal-laden filings by magnetic separation. The heavy metal binding capacity of different types of soils was first assessed using the MetPLATE test which responds specifically to he avy metal toxicity. The binding capacities of Cu, Zn, or Hg in different types of soils followed this trend: clay rich so il > organic rich soils > sandy soils. The magnetic treatment was conducted by using 5% (w/w) iron filings and a 3-h contact time between iron filings and soil. The effectiven ess of magnetic separation of Cu, Zn and Cd from spiked soils was evaluated using the MetPLATE assay, the 48-h Ceriodaphnia dubia and the 96-h Selenastrum capricornutum toxicity tests. Results show ed that the magnetic treatment method worked best on Cu-spiked soils, followed by the Zn-spiked soils and the Cd-spiked soils. The toxicity removal for Cu varied between 81.4% and 99.9% after a single treatment, where as the toxicity removal for Zn and Cd were, respectively, 81.4%-98.7% an d 80.1%96.2% after two

PAGE 15

15 or three treatments. Chemical analysis indicated that the metals were removed from both the soil matrix and the soil extracts, and the energy dispersive X-ray spect roscopy (EDS) further confirmed the adsorption of Cu, Zn, and Cd ont o the added iron filings. No significant reduction in magnetic separation efficiency was observed as regards Cu and Zn removal from aged soils. The retrieved iron filings coul d be regenerated in 1 N HNO3 for 1 hr and then retreated by 1 M NaOH for 72 hrs prior to reuse. The magnetic separation method also worked effectively in Pb contaminated shooting range soils and artificially spiked sediments. More than 77.8% of the toxicity produced by Pb was removed from five shooting range soils. Fo r Cu-, Zn-, and Hgspiked sediments, the toxicity removal after a single treatment va ried between 83.0% and 99.98%. The type of sediment and metal did not affect the treatment e ffectiveness, and the metals were removed from both the sediment matrix and the sediment extracts. MetPLATE also showed great po tential in predicting heavy metal phytoavailability in different types of soil. Results from a plant study showed that if a soil extract showed approximately 90% inhibition by MetPLATE assay, this soil could probably cause phytotoxicity in lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ). The plant study also demonstrated the effectiveness of the proposed magnetic treatment on reducing Cu phytoavailability. After treatmen t, the growth of lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) was significantly enhanced, and the Cu co ntent in plants shoots and roots also significantly decreased after treatment.

PAGE 16

16 CHAPTER 1 INTRODUCTION 1.1 Problem Statement Heavy metal contamination of soils and sedime nts is of increasing concern. Sources of heavy metals in soils and sediments mainly include natural occurr ence and anthropogenic sources (e.g. mining, smelting, energy and fuel production, vehicle emission, waste disposal, urban wastewater effluents, fertilizer applicati on, military operations, etc.) (Cui et al., 2004a; Hong et al., 2002; Nedelkoska a nd Doran, 2000). Since heavy metals are not biodegradable, they tend to persist over long-term peri ods in the soil matrix. They can be transported to groundwater (an important source of drinking water) or taken up by agricultural crops which leads to concerns over human and animal health (Mulligan et al ., 2001a; OConnor, et al ., 2003; Sas-Nowosielska et al., 2004). Heavy metal-contaminated sediments have direct adverse e ffects on aquatic life and ecosystems. Potential problems caused by contam inated sediments include poisoning of food chain and loss of recrea tional enjoyment (US EPA, 1998a). Tabl e 1-1 lists the he alth effects of Cu, Cd, Zn, Pb, and Hg on humans, and Table 12 provides the U.S. EPA national primary or secondary drinking water standards (US EPA, 2003a; 2003b) and Florid a soil cleanup target levels (SCTLs) for these five metals (FDEP, 2005). A number of remediation tec hniques have been developed to decontaminate heavy metal polluted soils and/or sediments. These methods ha ve been successful under specific situations, such as in laboratory experiment s, but may not be implemented in large scale yet (Virkutyte et al., 2002). Factors like variable so il type and texture, decrease of soil productivity, high cost, and safety concerns can also limit the applicability of some of the existing techniques. Therefore, research is now being directed at developing alternative, lowcost and environmentally safe

PAGE 17

17 methods for heavy metal clean-up for soils and sediments (Nedelkoska and Doran, 2000; Reed et al., 1996; Sas-Nowosielska, 2004). 1.2 Heavy Metal Behaviors in Soils and Sediments The partitioning of heavy metals in solids can be divided into five fractions, including exchangeable, bound to carbonate, bound to Fe-Mn oxi des, bound to organic matter, and residual fraction (Tessier et al., 1 979). Factors that affect the speciation of metals in soils are soil texture, pH, organic matter, clay minerals, carbonate, an d redox potential (Mulligan et al., 2001a). Metal ions undergo a series of reactions with different soil constituents, including adsorption/desorption, precipitati on, biological mobilization and immobilization, redox reaction and penetration into the crystal structure of minerals (Letan et al., 2003). Moreover, some microorganisms can convert heavy metals into mo re toxic forms, e.g., methylation of Hg (Zhou et al., 2000). Figure 1-1 displays the interactions between soil constituents and heavy metals. Table 1-3 shows the background levels of Cu, Cd, and Zn in natural soils and plants. Bioavailability of heavy metals is often associ ated with their distri bution among soil fractions (Tu et al., 2001), and toxicity of heavy metals is associated with their bioavailability. Bioavailable metals are most likely found in wate r soluble and exchangeable forms, whereas the residual fraction is tightly bound and unlikely to be bioavailable under natural conditions (Ma and Rao, 1997). Therefore, total metal content us ually cannot be used directly for determining the bioavailability of a metal. Soil pH, redox potential, clay minera ls, organic matter, as well as the presence of hydrous oxides of aluminum, iron, and manganes e can affect the solubility, mobility and thereby the bioavailability of hea vy metals in soils and sediments. Generally speaking, an increase in soil pH clay and organic matter cont ent, and a decrease in redox potential can lower the solubil ity and availability of heavy metals (Morel, 1997; Trivedi and

PAGE 18

18 Axe, 2000). Table 1-4 summarizes the chemical speci es of trace metals with regard to their bioavailability and potential toxicity to organisms. Within the sediments, under anoxic conditions, me tals usually occur as insoluble sulfides. The bioavailability of sediment-borne metals is related to sediment ch aracteristics, including organic matter and clay content, particle si ze distribution, pH, cation exchange capacity, etc. Heavy metals in sediments can be transferred to the water column via a variety of processes, such as diffusion, resuspension, and biot ransfer through organisms (Adriano, 2001a). Sequential extraction is a commonly used met hod to identify metal fractionations in soils and sediments (Li et al., 1995; Ma and Rao, 1997; Nvoa-Muoz et al., 2007; Reddy et al., 2001). This technique is based on us ing a series of reagents with different chemical properties to extract a faction of metals linked to a specific form (Tessier et al., 1979). 1.3 Soil and Sediment Re mediation Techniques Soil and sediment remediation techniques include physical, biological and chemical treatments (Hamby, 1996), and these tec hniques can also be divided into ex situ remediation and in situ remediation. A number of soil remediatio n techniques and thei r applications are summarized in Table 1-5. Among the numerous reme diation strategies discussed, methods such as excavation, containment, bioremediation, vi trification, electrokinetic remediation, soil washing and flushing, chemical immobilizati on, and phytoremediation ar e all commonly used remediation techniques for heavy metal-contaminated soils. The mechanisms of heavy metal removal by mi croorganisms can be categorized as either metabolic or non-metabolic uptakes. The former is due to the fact that some heavy metals are essential micro-nutrients for micr obial growth, while the latter includes organic binding to the cell wall and extracellular biopolymers (Antsuki et al., 2003). Microorg anisms are able to convert metals into more soluble or mobile forms by protonation, chelation, or chemical

PAGE 19

19 transformation (e.g., redox reaction), which can be removed from solid matrices. In contrast, immobilization of heavy metals by microorgani sms can occur by precipitation, crystallization, sorption, uptake and intracellular sequestrati on. These immobilization processes may have applications for in situ remediation (Gadd, 2004). The microbial reduction of Cr6+ to Cr3+ and As5+ to As3+ have been attracting resear chers interests and show promising results for soil remediation (Anderson and Cook, 2004; Chirwa a nd Wang, 2000; Faisal et al., 2005; Luli et al., 1983; Megharaj et al., 2003). Microbial reduction of U6+ and Se6+ have provided possibilities to concentrate uranium and selenium from contamin ated soils (Abdelouas et al., 1998; Ganesh et al., 1997; Ike et al., 2000; Or emland et al., 1999; Spear et al., 2000). The ability of microorganisms to convert Hg2+ to volatile Hg0 has shown great potential for mercury bioremediation (Nakamura et al., 1999; Wiatrows ki et al., 2006; Zer oual et al., 2001). Although successful experiments have been conducted in labor atories, the applicati on of bioremediation in large scales is still limited so far. Phytoremediation is an environmentally friendly alternative that uses plants (hyperaccumulators) to remove metals from so ils (Chaney et al., 1997; Wong et al., 2004). A hyperaccumulator is defined as a plant whose leaves may contain >100mg/kg Cd >1,000mg/kg Ni, or Cu, or >10,000mg/kg Zn or Mn (dry weight ) when grown in metal-rich media. Some of the identified hyperaccumulators include Zea mays, Brassica jun cea, Alyssum bertolonii, Berkheya coddii, Helianthus, and Thalspi caerulescens (Zavoda et al., 2001). Due to the small size and low biomass of most hyperaccumulators, much research is being conducted to enhance the availability of heavy metals in soils by a dding chemical amendments, such as ethylene diaminetetraacetate (EDTA) and nitriloacetate (N TA) (Wong et al., 2004). However, the use of synthetic chelators can enhance the mobility and bioavailability of metals in soils and thus

PAGE 20

20 increase the potential risks of leaching in to groundwater and of posing harm to soil microorganisms (Meers et al., 2005; Wong et al., 2004). Organic acids (e.g., citric acid), which can degrade quickly in soils, may be a pr omising alternative to the persistent aminopolycarboxylic acids such as EDTA for incr easing metal bioavailability (Meers et al., 2005). In addition, the disposal of contaminated cr op material after harvesting is also a concern (Sas-Nowosielska et al., 2004). Amendments utilized in chem ical immobilization of heavy metal-contaminated soils include alkaline-based products, such as lim e and phosphate-based materials (Basta and McGowen, 2004; Eighmy et al., 1997; Hettiarach chi et al., 2000; McGowen et al., 2001; Wang et al., 2001). In addition, organi c matter (biosolids) (Brown et al., 2003, 2004; Farfel et al., 2005), as well as various industria l products, such as zeolites (Boul arbah et al., 1996; Edwards et al., 1999; Friesl et al., 2003; Os te et al., 2002) and paper mill sludge (Calace et al., 2005), can also be used as soil amendments. Chemical immob ilization is a cost-effectiv e alternative and will not present adverse environmenta l or health efffects (Hamby, 1996). Remediation techniques for contaminated sediments can also be grouped into in situ and ex situ treatments. In situ treatments avoid handling of sedi ments and usually cost less than ex situ treatments. However, in situ treatment is almost always less effective than ex situ treatment, and sometimes lacks process control (US EPA, 1993) The categories and techniques used by the U.S. EPA are listed in Table 1-6, among which only a few, such as capping, chemical treatment, ground freezing, washing, and solidification/st abilization, can be used for heavy metalcontaminated sediments. However, each remedi ation technique has its own applicability and limitations. In situ chemical treatment may cause secondary contamination, and is difficult to ensure complete mixing of the treatment reagents w ith contaminated sediments. Ground freezing

PAGE 21

21 has very limited applicat ion due to its high cost. In situ solidification/stabi lization has not yet been proven or accepted by the U.S. EPA for contam inated sediments since little is known about the cost, effectiveness, and possible byproducts for large-scale treatment. Among ex situ treatments, only soil washing and solidification/ stabilization can be used for heavy metalcontaminated sediments. Sediment washing is no t effective for sediments with high contents of fine particles (e.g., silt, clay and humic substa nces) or sediments with low permeability. The success of solidification/stabilization is very dependent on the selection of proper stabilizing agents and requires a low orga nic concentration of less than 20 percent (US EPA, 1993). 1.4 Magnetic Separation Four categories of magnetic separation was reported by Parker (1977), and they are the following: low intensity dry magnetic separati on, low intensity wet magnetic separation, high intensity dry magnetic separation, and high inte nsity wet magnetic separation, which is also known as high gradient magnetic separation (HGMS). Magnetic force acting on particles during thei r passage through the system is defined by Equation 1-1, provided that the particle is su fficiently small for the magnetic field to be considered uniform throughout its volume. Greater magnetic force generate s greater possibility of successful separation (Parker, 1977). Fm= 2 1 0 (k-km) (H2) (1-1) Where, Fm is the magnetic force, 0 is the permeability of free space, is the particle volume, k is the susceptibility per unit volume, km is the susceptibility of the particle-bearing medium that can be neglected in most circumstances, and H is the magnetic field strength (Parker, 1977). In a separation process, tails refers to the non-magnetic product collected after passing through the system, and the magnetic concentrate is called mags. The separation efficiency is

PAGE 22

22 measured in terms of recovery (R) and grade (G). R is the ra tio of magnetic material in the mags to that in the original feed; G is the fraction of the magnetic component in the mags (Parker, 1977). Table 1-7 lists some conventi onal and modern applications of magnetic separation in different fields. The advantages of magnetic sepa ration, such as low cost, simplicity and ability to work at high flow rates, have encouraged th e application of this technique to environmental problems (Karapinar, 2003). A number of studi es have shown that high gradient magnetic separation (HGMS) can be used to remove magn etic and non-magnetic pollutants such as solid particles (DeLatour, 1975), phosphorus (Bitton et al., 1974; Franzreb and Hll, 2000; Karapinar et al., 2004), organic compounds (Sakai et al., 1 997), algae (Bitton et al., 1975), bacterial viruses (Bitton and Mitchell, 1974), co lor and turbidity (Anderson et al., 1983) from water and wastewater effluents. A succe ssful separation of non-magnetic c ontaminants consists of adding a magnetic seeding agent, such as magnetite, Fe2O3, Cr2O3, or MnO2, to which the pollutants will be attached and an additiona l treatment such as precipita tion /flocculation /coagulation (Karapinar, 2003). In addtion, in vestigations have indicated th at the adsorption properties of magnetite can be used in conjunction with a ma gnetic field to remove heavy metals, metal colloids, and nanoparticles from aqueous efflue nts (Anand et al., 1985; Navratil and Tsair, 2002; Terashima et al., 1986). 1.5 Research Objectives and Dissertation Outline The purpose of this doctoral research is to de velop an effective approach to remove heavy metals from soils and sediments. The approach is based on using iron filings as an adsorbent and subsequently recovering the iron filings by magnetic separation. This dissertation is organized into eight chapters. Chapter 1 co vered the background and research objectives. Chapter 2 developed a toxico logical approach for assessing the heavy metal

PAGE 23

23 binding capacity of soils. Chapter 3 investigated the effectiveness of heavy metal removal from artificially contaminated soils using magnetic se paration. Chapter 4 discussed the effect of soil aging on the magnetic separation process. Magne tic separation of heavy metals from Pbcontaminated soils and artificially contaminat ed sediments were covered in Chapter 5 and Chapter 6, respectively. Chapter 7 in vestigated the ability of MetPLATETM, a bacterial toxicity test, in predicting heavy metal phytoavailability, as well as the use of plants to assess the effectiveness of magnetic separation for removing heavy metals from soils. Chapter 8 summaried the findings of all of thes e research experiments.

PAGE 24

24 Table 1-1. Health effects of selected heavy metals on humans Toxicant Exposure routes Signs and symptoms of exposure Carcinogenicity by EPA Copper Inhalation Oral Dermal Nose, mouth and eyes irritation, headaches, dizziness, nausea, vomiting, diarrhea, abdominal pain, immune system damage, liver and kidneys damage Group D (not classable) Zinc Inhalation Oral Dermal Chest pain, cough, dyspnea, reduced lungs volume, nausea, chills, malaise, leukocytosis, vomiting, abdominal cramps, diarrhea, skin irritation, copper deficiency Group D (not classable) Cadmium Inhalation Oral Dermal Extreme fragility of bones, severe pain in bones and joints, severe nausea, vomiting, salivation, abdominal cramps and diarrhea, renal disturbances, lung insufficiency, osteomalacia, anemia and anosmia Group B1 (probable human carcinogen) Lead Inhalation Oral Dermal Fatigue, tremor, vomiting, abdominal pain, weight loss, weakness in fingers, wrists, or ankles, small increases in blood pressure, anemia, nervous system damage, brain and kidneys damage, miscarriage Group B2 (Probable human carcinogen-based on sufficient evidence of carcinogenicity in animals) Mercury Inhalation Oral Dermal Cough, lungs irritation, nausea, vomiting, diarrhea, increase in blood pressure or heart rate, skin rashes, eye irritation, personality changes, tremor, changes in vision, deafness, muscle incoordination, loss of sensation, difficulties with memory, nervous system damage, kidneys damage Elemental mercury is group D (not classable); Methylmercury is group C (Possible human carcinogen) Source: US DHHS, 1999a, 1999b, 2004 2005a, 2005b; Francis, 1994; Table 1-2. US and FL regulations on drinking wa ter and soil levels of Cu, Cd, Zn, Pb and Hg FL SCTL Direct Exposure (mg/kg)c Metal Drinking Water regulations (mg/L) Residential Commercial/Industrial FL SCTL Leachability based on groundwater criteriac (mg/kg) Cu 1.3a 150 89,000 NA Cd 0.005a 82 1,700 7.5 Zn Pb Hg (inorganic) 5b 0.015a 0.002 a 26,000 400 3 630,000 1,400 17 *** *** 2.1 a US EPA national primary drinking water regulations, 2003a; b US EPA national secondary drinking water regulations, 2003b; c FDEP soil cleanup target levels (SCTLs), 2005; *** Leachability values may be derived using the SPLP test to calculate site-specific SCTLs or may be deterimined using TCLP in the event oily wastes are present; NA= not available at time of rule adoption

PAGE 25

25 Table 1-3. Background level of Cu, Zn, and Cd in natural soils and plants Metal Concentration in soils (ppm) Concentration in plants (ppm) Cu 2-100 5-30 Zn 30-150 10-150 Cd <1 0.005-0.02 Source: Mulligan et al., 2001a Figure 1-1. Interactions between soil constitu ents and heavy metals (from Adriano, 2001b). Table 1-4. Chemical species of trace metals wi th regard to their bioavailability and potential toxicity to organisms Dominant chemical speciesa Metal Soil Water Most toxic speciesb Cu Cu2+ Cu2+-fulvate Cu2+ Zn Zn2+ Zn2+ Zn2+ Cd Cd2+ Cd2+ Cd2+ Pb Pb2+ Pb(OH)+ Pb2+ Hg Hg2+; Hg2+-fulvate Hg(OH)2; HgCl2; CH3Hg CH3Hg a Does not account for ion-pairs or complex-ion species; b Considers degree of bioavailability Source: Adapted from Adriano, 2001a.

PAGE 26

26Table 1-5. Summary of soil remediation techniques Technique Description Appli cability Comments Reference Excavation Excavation and burial of the contaminated soils at a hazardous waste site Organics/ inorganics Expensive, only transfers problems from one location to another McGowen et al. (2001); Maenpaa et al. (2002) Containment Use physical barriers to retain, immobilize, or isolate contaminated soil (e.g. slurry walls and landfills) Organics / inorganics Require extensive preparation of the site and/or pretreatment of the waste Mulligan et al. (2001a); US EPA (1997) Soil washing Ex situ, aqueous-based technique to extract and separate contaminants from soils Organics/ inorganics Not very effective for soils containing high silt and clay content or contaminated with high concentrations of mineralized metals or hydrophobic organics Mulligan et al. (2001a); US EPA (1997); Andrade et al. (2007); Isoyama and Wada (2007); Ehsan et al. (2006a); Ehsan et al. (2006b) Soil flushing In situ technique based on injection or infiltration of chemical solutions through soils to extract contaminants, similar to soil washing. Organics / inorganics Not very effective for soils with high clay and organic matter. Large-scale treatment is limited to metals Mulligan et al. (2001a); US EPA (1997); Park and Bielefeldt (2005); Tsang et al. (2007); Zhu et al. (2005) Thermal treatment Use direct or indirect heat exchanges to desorb, vaporize, or separate contaminants from soils Volatile / Semi-volatile organics Can be highly cost intensive due to high moisture content US EPA (1997); Harjanto et al. (2002); Kasai et al. (2000); Merino and Bucal (2007)

PAGE 27

27Table 1-5. Continued Technique Description Appli cability Comments Reference Incineration Ex situ remediation technique, which is also known as controlled-flame combustion or calcinations Organics Very expensive to treat wastes with very high moisture content or low organic content US EPA (1997); Leuser et al. (1990); Anthony and Wang (2006) Vitrification In situ technique based on heating the soil electrically around 1600-2000C, then allow the molten volume to solidify as it cools Organics/ inorganics Expensive technique, high clay and moisture contents can affect the efficiency Mulligan et al. (2001a); Hamby (1996); Spalding (1994) Electrokinetic remediation In situ or Ex situ technique based on by passing electric current in contaminated soils Organics / inorganics Mainly applicable for saturated soils with low groundwater flow rates and fine-grained soils of low hydraulic permeability Mulligan et al. (2001a) ; Reddy et al. (2003) ; Kim et al. (2002); Altin and Degirmenci (2005); Ottosen et al. (1997); Page and Page (2002) Phytoremediation In situ technique using plants to make soil contaminants less harmful or non-toxic Organics / inorganics An environment friendly technique but requires a very long treatment period, disposal of contaminated crop is also a concern Chaney et al. (1997); Banuelos et al. (1997); Barocsi et al. (2003); Begonia et al. (2003); Garbisu and Alkorta (2001); Chemical immobilization A technique by introducing chemicals into the soil to reduce the solubility, mobility or availability of contaminants Inorganics Less expensive than excavation and landfilling and may provide a longterm remediation solution Hamby (1996); Yang et al. (2001); Basta and McGowen (2004); Illera et al. (2004)

PAGE 28

28Table 1-5. Continued Technique Description Appli cability Comments Reference Vapor Extraction In situ or ex situ technique based on using vacuum pump or blower to induce air flow through the waste to extract contaminants Volatile / semi-volatile organics Contaminants must have low water solubility and above the water table, soil moisture content must be low and the soil must be sufficiently permeable Hamby (1996); US EPA (1997); Upreti et al. (2007); Travis and Macinnis (1992); Poulsen et al. (1998); Park et al. (2005) Bioremediation In situ or ex situ technique based on using microbial species to degrade and transform contaminants Organics / inorganics Release of unpleasant odors, added nutrients may be carried to surface water or leach to groundwater US EPA (1997); Wilson and Jones (1993); White et al. (1998); Rojas-Avelizapa et al. (2007); Van Dillewijn et al. 2007)

PAGE 29

29Table 1-6. US EPA categories of treatme nts potentially applicable to sediments Technique Description Applicability In situ treatment Capping Underwater covering of c ontaminated sediments with less contaminated sediments with or without lateral walls Organics/ inorganics Solidification/stabilization Add chemicals or cements to encapsulate sediments and/or make them less soluble, mobile or toxic Organics/ inorganics Biological treatment Add microorganisms and/or chemicals to initiate or enhance bioremediation Organics Chemical treatment Treat sediments by neutralization, precipitation, oxidation, and chemical dechlorination Organics/ inorganics Ground freezing Form a wall of fr ozen sediment by placing refrigeration probes Organics/ inorganics Ex situ treatment Biological treatment Use microor ganisms to breakdown organic contaminants Organics Dechlorination Potentially effective in detoxifying specific types of aromatic organic contaminants Aromatic organic compounds Solvent extraction Volume reduction technique leaches contaminants with organic solvents Organics Soil washing Water-based process to mechanically scrub excavated sediments Organics/ inorganics Thermal desorption Remove contaminants by heating the sediment at a temperature below combustion Volatile/ semi-volatile organics Solidification/ stabilization Add materials such as fly ash to reduce mobility or solubility of waste constituents Most effective on inorganics and metals, not effective on volatile organics Incineration A thermal treatment to destroy organic compounds Organics Source: US EPA 1998a; 1993

PAGE 30

30 Table1-7. Applications of magnetic separation Fields of application Purpose References Chemical and allied industries; food, drink and tobacco manufacturing; coal processing; metals industries Remove tramp metal to protect machine Parker (1977) Industrial raw materials processing Extract ferrous contamination Parker (1977) Mineral dressing industries Extract/enrich magnetic ores Parker (1977) Clinical Separation of red cells from whole blood Melville et al. (1975); Melville at al. (1982); Takayasu et al. (2000) Food industry Remove contaminants from milk Kaminski et al. (2000) Environmental remediation Oil spill remediation Industrial sludge treatment Water purification Wastewater treatment Nuclear waste treatment Chun et al. (2001); Yanagisawa et al. (1981); Petrakis and Ahner (1976); Navratil and Tsair (2002); Chiba et al. (2002); DeLatour (1973); Bitton et al. (1974, 1975); Gokon et al. (2002); Ebner et al. (1999)

PAGE 31

31 CHAPTER 2 TOXICOLOGICAL APPROACH FOR ASSE SSING THE HEAVY METAL BINDING CAPACITY OF SOILS 2.1 Introduction Soils are widely used as sinks for organic a nd inorganic contaminants such as metals. They are useful in mitigating the impact of contaminants on groundwater resources and surface waters. Soils are impacted by several types of toxic wastes, includi ng industrial wastes, biosolids, mining, and construction and demolition wastes. Toxic metals in the applied wastes have attracted the attention of regulatory agencies because they can be transported to groundwater or taken up by agricu ltural crops, leading to con cerns over human and animal health. Toxic metals tend to bind to soils, thus becoming less available to the biota and to roots of agricultural crops (Adriano, 2001a; Salo mons, 1995). Metal phytoavailability (i.e., availability to plants) is c ontrolled by several factors, in cluding metal speciation, soil characteristics (e.g., pH, clay type and content, organic matter content, moisture content) and contact time between soil and metals (Naidu et al., 2003; Weng et al., 2002). Moreover, bioavailability of metals in so ils also depends on the type of clay and organic matter (Lock and Janssens, 2001). Soil amendment with clay minera ls (e.g., bentonite, zeolite), iron oxides (e.g., goethite, hematite), and phosphate fe rtilizers has been reported to be effective in reducing metal availability to wheat (Triticum aestivum) (Usman et al., 2005). Metal binding and immobilization in soils involves several mechanisms, such as adsorption, ion exchange, complexation by humic substances, and precipit ation reactions (Weng et al., 2002). Sequential extraction procedures provide a good indication of metal partitioning in soils. They involve a range of ch emical reagents that extract different meta l fractions (soluble, exchangeable, carbonate-bound, oxide/hydroxid e-bound, organic matter-bound, and residual fractions) in soils (Balasoiu et al., 2001; Smith et al., 1999; Tack and Verloo, 1995; Tessier et al.,

PAGE 32

32 1979; Yong et al., 2001). The sol uble and, potentially, th e exchangeable fractions are generally believed to be available to plants and the biota, and the total concentration of metals in soils does not indicate their availa bility to plants (Adriano, 2001a). Phytoavailability may vary among different soils contaminated with the same tota l metal concentration, suggesting that the soil matrix plays an important role in phytoavailabil ity and, ultimately, phytotoxicity (Naidu et al., 2003). The metal fractionation chemical procedures need, however, to be complemented with toxicity testing to obtain information about the biological activity of me tals in soils. Some investigators have used both chemical and toxico logical approaches to as sess the bioavailability of metals in solid matrices (Kong and Bitton, 2003; Schultz et al., 2004; De Vevey et al., 1993). In this chapter, we developed a relatively rapid test to assess the heavy metal binding capacity (HMBC) of five soils. The test is based on the use of MetPLATETM, a bioassay that responds specifically to heavy meta l toxicity (Bitton et al., 1994). The test compares the relative toxicity of a metal in a given soil to metal t oxicity in a reference soil (Ottawa sand). 2.2 Material and Methods 2.1.1 Soils Used Three soil types were used to assess their cap acity to bind metals, such as copper, zinc and mercury. Two sandy soils were collected from th e top 4 feet at two di fferent sites and were chosen because they are represen tatives of the soils prevailing in North Central Florida. An organic rich soil (organ ic soil 1) was collected from the first few top inches along Hogtown Creek in Gainesville, FL. The s econd organic rich soil (organic soil 2) was a top soil purchased from a local landscaping store. A clay rich soil (Georgia clay soil, top 20 cm) was collected in Atlanta, Georgia. Table 2-1 s hows some characteristics of the so ils under study. The soil pH was measured according to the U.S. EPA method 9045D (US EPA, 2004). Soil redox potential (Eh)

PAGE 33

33 was measured by Fisher Scientidic Accument Model 15 pH/mV meter. Part icle size distribution was determined according to the USDA Soil Survey Lab Method (USDA, 1992). The Walkley & Black Method (Walkey and Black, 1934) was used to measure the soil organic carbon content. Soil effective cation exchange capacity (CEC) wa s determined according to a method developed by Sumner and Miller (1996). Ottawa Sand, due to its low ability to bind metals, was selected as a reference soil. 2.2.2 Methodology for Assessing Soil Heavy Metal Binding Capacity (SHMBC) Briefly, as shown in Figure 2-1, the test cons ists of adding metal-laden solutions to soils under study, allowing the mixtures to reach equilib rium, separating the solid phase from the pore water by centrifugation, and assaying for metal toxicity of the soil extracts. A similar methodology was used for the Ottawa sand, which serves as a reference soil. Soils were first air-dried, screened (sieve # 16; 1.19 mm particle s), and homogenized. Subsequently, serial dilutions of metal-spiked so lutions were prepared in moderately hard water (60 mg/L Ca, 60 mg/L Mg, pH = 7.4-7.8) for soil sp iking. The solutions were labeled A, B, C, D, and E. A sixth solution labeled F, was not spiked with metals and served as the negative control for the soil. Five other metal-spiked solutions were added to Ottawa sand (reference soil) and were labeled A0, B0, C0, D0, and E0. A sixth solution, labeled F0, was prepared without any metal added. Tw enty milliliter of each solution was added to 5 g of soil or Ottawa sand in 50-mL Erlenmeyer flasks The flasks were covered with parafilm and placed on a shaker at 300 rpm for 4 hours. Afte r shaking, the soils were centrifuged at 10,000 rpm for 15 minutes. The metal toxicity of the soil extracts was assayed with MetPLATETM, a microbial test which responds specifically to heavy metal toxicity. The MetPLATETM assay was carried out according to B itton et al. (1994). Followi ng rehydration of the MetPLATETM bacterial reagent, 0.1 mL of bacterial suspensi on was added to 0.9 mL of the soil extracts in

PAGE 34

34 small culture tubes. The tubes were vor texed and placed in an incubator at 35oC for 90 minutes. Following incubation, 0.2 mL of the content fr om each tube was transferred to a 96-well microplate. 0.1 mL of rehydrated MetPLATETM chromogenic substrate was added to each well. The plate was shaken gently and returned to the incubator until a purple color developed (after about 1 hr) in the negative controls (solutions F and Fo). The absorbance was determined with a microplate spectrophotometer (Maxline Microplate Readers, Molecular Devices, Sunnyvale, CA) at 570 nm. All HMBC tests were run in triplicate and three MetPLATETM toxicity tests were run for each HMBC test. Regression analysis was used to determine the EC50 for both the soil under study and Ottawa sand, the reference soil. The soil heavy metal binding capacity (SHMBC) was determined by dividing the EC50 for the soil sample by the EC50 for the metal in the reference soil, Ottawa Sand. The SHM BC was obtained as followed: EC50 of field soil spiked with a given metal SHMBC = (2-1) EC50 of Ottawa sand spiked with the same metal The SHMBC methodology, summarized in Figure 2-1, was used to determine the Cu, Hg and Zn binding capacity of the five soils A detailed procedure for the MetPLATETM assay along with an example of the EC50 calculation was included in Appendix A. 2.3 Results and Discussion Soil heavy metal binding capacity (SHMBC) was determined for five soils sampled in Florida and Georgia. Three metals (Cu, Zn, and Hg) were tested for their binding to the soils. Table 2-2 shows the EC50s (expressed as metal added in mg/kg soil) of the three metals in the different soils under study. The Ottawa sand (refe rence soil) extracts were quite toxic, with EC50s of 1.1 mg/kg for Cu, 0.9 mg/kg for Zn, and 1. 5 mg/kg for Hg, indicating that the Ottawa

PAGE 35

35 sand displayed a relatively low binding capacity fo r metals, thus justifying its selection as a reference soil. It is worth mentioning that the higher the EC50, the lower the toxicity of the soil extracts, indicating that the metal was bound by the soil and, thus, unavailable to the test organisms. The Georgia clay soil displayed the highest EC50s (i.e., lowest toxicity to MetPLATE) for the three metals tested. Among the five soils under study, the EC50s, expressed as metal added to soils in mg/kg, varied between 19.1 and 416 for Cu, between 14.3 and 296.9 for Zn, and between 8.4 and 448.4 for Hg (Table 2-2). The SHMBC for the three metals and five so ils tested is shown in Figure 2-2. As regarding their binding capacity towards the three metals, the so ils were classified in the following order: Georgia clay rich soil > organi c rich soils > sandy soils The organic (e.g., humic substances) and inorga nic (e.g., clay minerals) colloidal particles in soils generally play a significant role in bindi ng metals. Georgia clay rich soil contains 21% clay and thus displayed the highest binding capac ity towards all three metals. Similarly, the organic rich soils (1 & 2) show ed much higher SHMBCs than the sandy soils, due to their much higher organic matter content ( 12.2-18.8%) (Figure 2-2). Thus the higher the SHMBC, the higher is metal binding to the soils. The HMBC concept, as reviewed by Bitton et al. (2005), was previously used to assay metal bioavailability in surf ace waters (Huang et al., 1999) a nd municipal landfill leachates (Ward et al., 2005). The present research shows the first application to date of this bioassay to soils. It is a relatively rapid methodology to as sess metal bioavailabil ity in soils. This methodology is based on toxicity testing of soil extracts with MetPLATE, a test specific for

PAGE 36

36 heavy metal toxicity (Bitton et al., 1994). The ability of this test to predict metal uptake and subsequent toxicity to plants is discussed in Ch apter 7. Culture studies with plants show that metal uptake by plants varies with the type of so il, with the uptake bei ng higher in sandy than in clay soils which generally display a higher me tal binding capacity (Naidu et al., 2003). For example, terrestrial plants shoul d be protected from Cu toxicity at the benchmark concentration of 100 mg/kg (Will and Suter, 1995). Howeve r, no phytotoxicity was found in Australian orchard soils with a range of copper concentrati ons of 11 to 320 mg/kg (Merry et al., 1983). Cu and Cr toxicity to barley was lower in a spiked natural forest soil (82.8 sand, 9.2% silt, 8.0% clay, 3.8% organic matter conten t) than in an artificial sandy soil (100% sand; pH 7.80, 0.27% organic matter) (Ali et al., 2004). Around a Peru vian copper mine, phytoxicity was found to be higher in soils with low organic matter (Bech et al., 1997). These findings c onfirm that total soil metal concentrations do not give an indication of metal bioavailability and phytotoxicity, with the soil matrix playing an important role in metal toxicity. Our findings, using a bacterial toxicity test, confirm that, at least for Cu, Zn and Hg, the SHMBC for clay and organic soils is much higher than for sandy soils. 2.4 Conclusions This new technique, based on a toxicological bi oassay, shows a novel approach to evaluate heavy metal binding to soils and, hence, bioava ilability. We have shown that the soil metal binding capacity (SHMBC) varies with the type of soil, with clay rich and organic rich soils displaying a higher metal binding than sandy soils, which was used in latter experiments to determine the heavy metal concentrations in spiked soils. This relatively ra pid test could be used in a number of applications, mainly ecological ri sk assessment. The SHMBC test could also be used to assess the suitability of soils to receive metallic wastes. The test could simulate more realistic conditions by determining the SHM BC of metal-spiked soils which have been

PAGE 37

37 subsequently aged for a few weeks or a few mont hs. Moreover, it would be valuable to run SHMBC tests in parallel with metal uptake by terrest rial plants to determine their application in assessing phytoavailability a nd potential for phytotoxicity.

PAGE 38

38 Table 2-1. Soils characteristics Characteristic Red sandy soil Sandy soil Organic soil 1 Organic soil 2 Georgia clay soil pH 6.25.75.75.3 5.7 Eh (mV) 485.0422.0403.0337.0 324.0 % Organic carbon 0.10.56.44.1 0.6 % Organic matter 0.31.618.812.2 1.8 % Sand 90.796.9293.192.6 56.4 % Silt 3.20.022.00.8 22.6 % Clay CEC (cmolc/kg) 6.1 14.4 3.06 14.1 4.8 230.1 6.6 107.8 21.0 79.6 Table 2-2. EC50s (metal added in mg/kg soil), as determined by MetPLATETM, of water extracts from five soils and Ottawa sand Soil type Heavy metal EC50 (mg/kg soil) Ottawa sand (reference soil) Red sandy soil Sandy soil Organic soil 1 Organic soil 2 Georgia clay soil Cu Zn Hg Cu Zn Hg Cu Zn Hg Cu Zn Hg Cu Zn Hg Cu Zn Hg 1.1 0.2* 0.9 0.1 1.5 0.3 19.1 1.3 14.3 2.1 8.4 0.2 54.3 3.1 55.0 2.5 88.4 8.0 86.4 2.7 153.3 36.0 268.9 53.1 128.4 3.3 129.3 4.4 300.7 22.5 416.0 16.2 296.9 21.5 448.4 22.3 *mean of 3 replicates one standard deviation

PAGE 39

39 Figure 2-1. Soil HMBC (SHMBC) methodology Screen (no. 16 sieve) and homogenize soil Determine dry weight of soil Prepare 5 metal spike solutions in serial dilutions (A, B, C, D, and E). Solution F has [M] = 0 and serves as negative control. Add 20 mL solutions A, B, C, D, E, and F to 5g (D.W.) of field soil in small flasks (liquid/solid ratio = 4). Add 20 mL solutions A0, B0, C0, D0, E0, and F0 to 5g (D.W.) of Ottawa sand in small flasks (liquid/solid ratio = 4). Cover field soil and Ottawa sand slurries with parafilm and place flasks on shaker for 4 hours at 300 rpm. Centrifuge mixtures to separate aqueous phase from solid p hase ( 10 000 r p m for 15 min ) Place 0.9 mL of supernatant from each solution in small test tube, and add 0.1 mL of MetPLATE bacteria. Mix and incubate at 35C for 90 min. Place 0.2 mL of solutions on a microplate, and add 0.1 mL of CPRG (substrat e). Shake plate gently. Return plate to incubator until color develops. Read absorbance at 570nm. Determine EC50 for soil sample and Ottawa sand. Soil HMBC = EC50 soil / EC50 Ottawa sand Prepare 5 metal spike solutions in serial dilutions (A0, B0, C0, D0, and E0). Solution F0 has [M] = 0 and serves as ne g ative control.

PAGE 40

40 17 49 79 117 378 16 61 144 6 58 201 170 330 179 2990 100 200 300 400 500Red Sandy SoilSandy SoilOrganic Soil 1Organic Soil 2Georgia Clay SoilSoil HMBC Cu Zn Hg Figure 2-2. SHMBC for three metals (Cu, Zn, Hg) and five soils.

PAGE 41

41 CHAPTER 3 HEAVY METAL REMOVAL FROM SOILS USING MAGNETIC SEPARATION 3.1 Introduction Heavy metal contamination of soils and sedime nts is a worldwide problem that has been attracting considerable attenti on over the past decades. Heavy metals in soils frequently accumulate in the upper horizons and can advers ely impact soil microbial activities, crop productivity, and the food chain (OConnor et al., 2003). Moreover, toxi c metals are highly persistent in the environment, with a residence time of hundreds of years or even more (Mulligan and Wang, 2006; Sas-Nowosielska et al., 2004). In re sponse to these adverse effects, a variety of decontamination strategies, including in situ and ex situ techniques, have been developed or are under ongoing development. Soil excavation and bur ial at landfills is the most traditional method, but it is very costly a nd only transfers the heavy meta l problem from one location to another (Maenpaa et al., 2002; McGo wen et al, 2001). Soil washing with acid or chelating agents such as EDTA is another frequently used technique; however, it can decrease soil productivity and change the chemical and physical structure of soils (Reed et al., 1996). The recovery of heavy metal-EDTA complexes is also difficult (Hong et al., 2002). Therefor e, a large number of alternative options have been investigated whic h are considered less intrusive and more cost effective. One of the alternatives that has recei ved a considerable amount of attention is heavy metal immobilization in soils via addition of various of amendments (Gray et al., 2006). Examples of these amendments include lime, phosphate-based materials (Basta and McGowen, 2004), organic matter (biosolids) (Brown et al., 2003, 2004; Farfel et al ., 2005), as well as various industrial products, such as zeolites (Boularbah et al., 1996; Edwards et al., 1999; Friesl et al., 2003; Oste et al., 2002) Immobilization is less expens ive than excavation and may provide a long-term remediation solution (M cGowen et al., 2001). Phytoremediation and

PAGE 42

42 electroremediation are other emerging hea vy metal treatment techniques. The goal of phytoremediation is to extract metals by using livi ng plants and it is incr easingly being regarded as a cost-effective and environmentally friendly alternative (Garbisu a nd Alkorta, 2001; Wu et al., 2004). However, phytoremediation technology is still a new field and holds some drawbacks and limitations that require further investigation. Electroremediation cons ists of passing a low intensity electric current between appropria tely distributed elec trodes imbedded in the contaminated soil (Mulligan et al., 2001a; Page and Page, 2002). The process can be applied in situ or ex situ. This method is more effective for clay soils as compared to soil washing, however, the demonstrations of this technology ar e limited so far (Mulligan et al., 2001a). In this chapter, experiments were performed to evaluate the effectiveness of removal of heavy metals (Cu2+, Zn2+, and Cd2+) from four soils using iron filings followed by a magnetic separation step. Iron (hydr)oxides are known to bind heavy metals by adsorption or coprecipitation to decrease hea vy metal mobility in soils (Crawfor d et al., 1993; Hartley et al., 2004; Mckenzie, 1980; Mench et al., 1994), and have been used to remove heavy metals from wastewater and liquid hazardous wastes (B enjamin et al., 1996; Yeager, 1998). Magnetic separation has been used extensively in industries, such as in the processing of minerals. In recent years, the application of magnetic separa tion technology to environmental problems has received considerable attention (Orbell et al ., 1997). As reported by Kochen and Navratil (1997), magnetic polymer resins were able to remove actinides and heavy metals efficiently from contaminated water. Studies ha ve also proven that high gradie nt magnetic separation (HGMS) can also be used for the removal of non-magne tic water pollutants, such as suspended solid particles (DeLatour, 1975), phosphorus (Bitton et al., 1974; Franzreb and Hll, 2000; Karapinar et al., 2004), organic co mpounds (Sakai et al., 1997), and alg ae (Bitton et al., 1975). Little is

PAGE 43

43 known about the application of ma gnetic separation methods to so il remediation. Macek et al. (2002) demonstrated the possi bility of removal of cesiu m-137, strontium-85, and europium-152 from artificially contamin ated soils by a magnetic sorbent. After the determination of best treatment c onditions, the soil extracts were tested by MetPLATETM assay, 48-h Ceriodaphnia dubia acute toxicity test, and 96-h Selenastrum capricornutum chronic toxicity test to assess the re duction of heavy metal toxicity in soils. Chemical analysis and mass balance studies were also performed to investigate heavy metal distribution in the soil matrix and extracts. In addition, a sequential extraction procedure was employed to assess the fractionations of the adde d heavy metals in a sandy soil and an organic rich soil after magnetic treatment. The regeneration of iron filings was also investigated, and the retrieval of Cu, Cd and Zn by iron filings was further examined, using energy dispersive X-ray spectroscopy (EDS). 3.2 Material and Methods 3.2.1 Soils Used Four soils were used to inve stigate the effectiven ess of the proposed magnetic treatment method. A sandy soil was sampled from the top 4 feet at the McCarty W oods on the University of Florida campus. A red sandy soil was sampled from Perdido Landfill in Cantonment, FL. An organic rich soil (organic soil) was collected from the firs t few top inches along Hogtown Creek in Gainesville, FL. A clay rich soil (Georgia cl ay soil, top 20 cm) was collected in Atlanta, Georgia. Table 3-1 shows the main characteris tics of the soils under study. All soils samples were first air-dried, screened (sieve # 10; 2.0 mm particles), and homogenized prior to use. 3.2.2 Chemicals Used Three heavy metal solutions (Cu2+, Cd2+, and Zn2+) and a heavy metal mixture (a solution consisting of equal concentrations of Cu2+, Cd2+, and Zn2+) were used. Copper solution was

PAGE 44

44 prepared from copper sulfate (CuSO4H2O, Sigma, St. Louis, OM). Cadmium solution was prepared from cadmium chloride (CdCl2.5H2O) purchased from Fisher (Pittsburgh, PA). Zinc solution was prepared from zinc chloride (ZnCl2, Sigma). Iron filings (Fisher, 40 mesh) were placed in 1 M NaOH for 72 hours to increase the adsorption capacity and then washed thoroughly with distilled wate r before use (Yeager, 1998). 3.2.3 Recovery of Iron Filings from Soils The recovery of iron filings from the so ils under study, using a Ferrimag rectangular magnet (Scientifics, 152 mm, 3.4 megagauss oerste ds) was investigated. Iron filings removal was studied under three different soil conditions (air-dried, fi eld capacity, and watersaturated) and three concentrations of iron filings (1%, 2%, and 5%, w/w). Under each soil condition, dry iron filings were added to 50 g of soils and were retr ieved with the magnet following 2-hr incubation. The retrieved iron filings were washed thoroughly with distilled water and dried at 70C overnight and then weighed to determine the percent retrieval. All samples were run in triplicate. 3.2.4 Determination of Iron Filings Concentration Fifty gram of sandy soil, weighed in a 250 ml centrifuge tube, was spiked with 40 mL of a heavy metal mixture containing 50 mg/L Cu2+, 50 mg/L Cd2+, and 50 mg/L Zn2+ (resulting in 40 mg/kg Cu2+, 40 mg/kg Cd2+, and 40 mg/kg Zn2+). The soil slurry was shaken for 1 hour. Then, three concentrations (2.5%, 5%, and 10%, w/w) of iron filings were added and the system was shaken for 6 hours. The mixture was then transf ered into a plastic container (34 cm 20cm 10cm), and the iron filings were magnetically retrieved by moving the magnet back and forth above the surface of the soil slurry The soil slurry was then transf ered back to the centrifuge tube, and 60 mL of distilled wate r were added to the system to bring the total so lution volume to 100 mL. The soil slurry was centrifuged at 10,00 0 rpm for 15 minutes. Th e supernatant (soil

PAGE 45

45 extract) was removed with a pipet. Each soil sample was run in triplicate. Soil without iron treatment served as the control. 3.2.5 Determination of the Contact Time between Iron Filings and Soil Matrix We studied the effect of the contact time betw een the metal-spiked soils and iron filings. Metal removal was determined following cont act times of 1.5 hr, 3 hrs and 6 hrs. 3.2.6 Magnetic Separation of Heavy Met als from Four Metal-Spiked Soils Fifty gram of sandy soil, red sandy soil, organic rich soil, and Georgia clay rich soil were spiked with 40 mL of Cu2+, Cd2+ or Zn2+ solutions. Three concentr ations (250 mg/L, 500 mg/L, and 1000 mg/L) of heavy metal solutions were us ed, resulting in 200 mg-metal/kg soil, 400 mgmetal/kg soil, and 800 mg-metal/kg soil. The so il slurry was shaken for 1 hour at room temperature. Metal separation efficiency was studied under the followi ng conditions derived from the previous experiments: addition of iron filings at 5% (w/w) conc entration; contact time of 3 hrs. Then, the iron filings were magnetically retrieved and the soil sl urry was centrifuged at 10,000 rpm for 15 minutes and the supernatant (so il extract) was removed with a pipet. The retrieved iron filings and the soil matrix afte r centrifugation were dried at 70C overnight. Soil without iron treatment served as the control. Toxicity tests were undertaken for all soil extracts before and after magnetic treatment. Chemical an alysis was performed for all factions, including soil extracts, retrieved iron filings, and soil matrix. Each soil sample was run in triplicate. 3.2.7 Toxicity of Soil Extracts Three toxicity assays, MetPLATETM, the 48-h Ceriodaphnia dubia acute bioassay, and the 96-h Selenastrum capricornutum chronic toxicity test were used to assess the toxicity of the soil extracts before and after iron treatment. To determine the EC50 for the soil extracts, 4 to 5 dilutions of the soil extracts were prepared, and a regression analysis was used to calculate the EC50s (See Appendix A for details). All sa mples were run in triplicate.

PAGE 46

46 Toxicity unit (TU) (Bitton, 1998), defined in the Equation 3-1, was us ed as an expression of metal toxicity. A higher TU value indicates higher toxicity. 100 TU = (3-1) EC50 3.2.7.1 MetPLATETM protocol The MetPLATETM assay (as described in Section 2.2. 2, also see Appendix A for further details), a microbial test which is specific fo r heavy metal toxicity a nd does not respond to organic toxicity (Bitton et al., 1994 ), was used to determine the toxicity of the soil extracts. Moderately hard water (60 mg/L Ca, 60 mg/L Mg, pH = 7.4-7.8) was used as the negative control. 3.2.7.2 48-hour Ceriodaphnia dubia acute bioassay The 48-h acute Ceriodaphnia dubia bioassay was carried out according to the U.S. EPAs standard method (US EPA, 2002a). Ne onate daphnids (first instar Ceriodaphnia less than 24 hours old) were used for testing a nd were fed 2 hours before the test started. Five neonates were exposed in each plastic cup containing 20 mL of the sample. Moderately hard water (60 mg/L Ca, 60 mg/L Mg, pH = 7.4-7.8) was used as th e negative control. The test temperature was 25oC. After 48 hours, the number of motile and dead daphnids was counted (see Appendix A for details). 3.2.7.3 96-hour Selenastrum capricornutum chronic toxicity test. The 96-h chronic Selenastrum capricornutum test was carried out according to the U.S. EPAs standard method (US EPA, 2002a). Th e preliminary algal assay procedure (PAAP) medium was used as the negative control. When te st began, 1 mL of the algal seed, with a cell density of 55 cells/mL, was spiked into each test flask containing 50 mL of the sample. Then all flasks were immediately foam-stoppered and placed under continuous light condition (400

PAGE 47

47 40ft-c). The test temperature was 25oC. The test flasks were shaken and rearranged randomly at least once per day. After 96 hours, growth inhi bition was measured by counting the cell density in each sample and negative control by a hem acytometer (Hausser Scientific, Horsham, PA) under microscope (see Appendix A for details). 3.2.8 Chemical Analysis Chemical analysis was undertaken for all fract ions, including the soil matrix, soil extracts and iron filings, before and after magnetic treatmen t. Soil extracts were digested according to the U.S. EPA method 3010A (US EPA, 1992), and soils a nd iron filings were digested according to the U.S. EPA method 3050B (US EPA, 1996). All di gested samples were analyzed for metals using inductively coupled plasma-atomic emission spectroscopy (ICP-AES). Mass balance studies were also performed for the Cu2+, Zn2+, and Cd2+ spiked soils. All samples were run in triplicate. Detailed digestion procedures are included in Appendix B. 3.2.9 Sequential Extraction of Metals from Soils A sequential extraction procedure adapted fr om Tessier et al. ( 1979) and Ma and Rao (1997) was used to assess the frac tionations of Cu, Zn, and Cd in spiked soils before and after magnetic treatment. The sandy soil and organi c rich soil were spiked with 200 mg/kg Cu2+, 200 mg/kg Zn2+, or 200 mg/kg Cd2+, respectively. The spiked soils were allowed to dry at room temperature, followed by adding distilled water to reached saturation, mixing very well, and letting them dry again at room temperature (3 days for sandy soil, and 5 days for organic rich soil). After four wet-dry cycles, a portion of each spiked soil was trea ted by magnetic separation according to the following procedure: 100 mL of distilled water was added to 50 g of soil, and the soil slurry was shaken for 1 hour at room temperature. Then, 5% (w/w, 2.5 g) of iron filings were added, and the system was shaken for 3 hours followed by magnetic retrieval of the iron

PAGE 48

48 filings. The soil slurry after treatment was dried at 70C overnight. Both treated and non-treated soils were used for sequential extraction acco rding to the procedure described below: Fraction 1: Exchangeable. One gram of soil was extracted at room temperature for 1 hour with 8 mL of 1 M sodium acetate (pH = 8.2) with continuous agitation. Fraction 2: Bound to carbonates. The residue from faction 1 was extracted at room temperature for 5 hours with 8 mL of 1M sodium acetate (pH adjusted to 5.0 with acetate) with continuous agitation. Fraction 3: Bound to Fe-Mn oxides. The residue from faction 2 was extracted with 20 mL of 0.04 M hydroxylamine hydrochloride in 25% (v/v) acetate for 6 ho urs at 96 3C with occasional agitation. Fraction 4: Bound to organic matter. Three milliliter of 0.02 M nitric acid and 5 mL of 30% hydrogen peroxide (pH adjusted to 2 with nitric acid) were adde d to the residue from fraction 3, and the mixture was heated to 85 2 C for 3 hours with intermittent agitation. After cooling, 5 mL of 3.2 M ammonium acetate in 20% (v/v) nitric acid was added and the sample was diluted to 20 mL and agitated continuously for 30 min. Fraction 5: Residual. The residual fraction was a theore tical value that was calculated by subtracting the sum of the previous four factions from the total. To minimize losses of solid material, the extractions were conducted in 50mL polypropylene centrifuge tubes. Between each succe ssive extraction, separation was performed by centrifuging at 10,000 rpm for 30 min, and the supernatant wa s removed with a pipet and filtered through a 0.2 m membrane filter. The residue was washed with 8mL of deionized water by vigorous hand shaking, and af ter centrifugation for 30 min, th is second supernatant was

PAGE 49

49 discarded. Metal concentrations in each fraction were determin ed by ICP-AES. All extractions were conducted in triplicate. 3.2.10 Energy Dispersive X-ray Spectroscopy of Retrieved Iron Filings Energy dispersive X-ray spectroscopy (EDS ), measured by SEM JEOL JSM 6400, was also employed to further examine both unused and used iron filings following magnetic retrieval from 800 mg/kg Cu, Zn and Cd-spiked sandy soil s. Since three magnetic treatments were required for 800 mg/kg Znand Cdspiked soils, the iron filings retrieved from each single treatment were combined and mixed very well for spectroscopy analysis. 3.2.11 Regeneration of Iron Filings Nitric acid (1 N) was used to treat the magneti cally retrieved iron fili ngs. First of all, the effect of the contact time between HNO3 and iron filings on the recovery of iron filings was studied. Two point five gram of unused ir on filings was treated with 10 ml 1 N HNO3 for 24 hrs, 2 hrs, and 1hr, and then washed thoroughly with distilled water and drie d at 70C overnight and then weighed to determine metal recovery. Each sample was run in triplicate. Once the optimal regeneration time (i.e., 1 hour) was determined, the effectiveness of the regenerated iron filings on the adsorption of Cu, Zn, and Cd from a spiked sandy soil was tested. Fifty gram of sandy soil was spiked with 40 mL of Cu, Zn, or Cd solution to reach a final metal concentration of 400 mg/kg. The soil slurry was shaken at room temperature for 1 hour followed by a magnetic treatment using 5% (w/w, 2.5 g) of fresh iron filings acco rding to the procedure described in Section 3.2.5. After treatment, th e retrieved iron filings we re regenerated in 1 N HNO3 for 1 hour and retreated in 1 M NaOH fo r 72 hours, and then washed thoroughly by distilled water and dried at 70C overnight. The effectiveness of the iron filings after the first regeneration were tested by treating the same sp iked sandy soil, and the retrieved iron filings from this treatment were regenerated (i.e., second regeneration) and tested ag ain. The toxicity of

PAGE 50

50 the soil extracts after treatment was determined by both MetPLATETM and the 48-h acute Ceriodaphia dubia assay, and a comparison of the toxicity of the soil extracts treated by fresh iron filings and regenerated iron filings was made All experiments were conducted in triplicate. 3.3 Results and Discussion 3.3.1 Recovery of Iron Filings from Soils The iron filings introduced in soils to trap meta ls must be retrieved efficiently to achieve a good removal of metals. Thus, iron filings were adde d at different concentrations to a sandy soil, red sandy soil, organic rich soil, and Georgia clay rich soil, a nd their recovery from the soil matrix was determined. The recovery of iron fili ngs from these four soils by magnetic separation under three different conditions (dry soil, soil at field capacity, and saturated soil) is shown in Figure 3-1. With regard to the sandy soil, unde r all of the three condi tions, the iron filings recovery was quite high and varied from 92.6% to 98.6%, and both iron f iling concentrations and degree of soil water satu ration did not significantly influence the recovery. In the case of the other three soils, the recovery of iron filings under air-dried and water saturated conditions was very high, varied from 93.4% to 99.2%, and was not significantly affected by neither soil type nor iron filing concentrations. However, magnetic separation from red sandy soil, organic rich soil, and Georgia clay rich soil at field capacity showed somewhat lower iron recoveries (from 62.0% to 86.1%) than under the other two condi tions (air-dried and water saturated soil). Therefore, all following experiments were carried out under water-saturated conditions. 3.3.2 Effect of Iron Filings Concentration and Contact Time on Metal Removal from Soils Heavy metal removal was investigated usi ng increasing iron filing concentrations and different contact time periods be tween iron filings and a sandy soil spiked with a mixture of metals (Cu2+, Cd2+, and Zn2+). The metal removal was assessed, using the MetPLATETM toxicity test. The effect of iron filings concentration is sh own in Table 3-2. The addition of iron filings at

PAGE 51

51 a final concentration of 2.5% (w/w) resulted in an increase of the EC50 of the soil extract from 3% before treatment to 6.8% after treatment, t hus showing a decrease in toxicity of the soil extract (i.e., toxicity decreases as EC50 increases). In the presence of 5% (w/w) and 10% (w/w) iron filings, the EC50s of the soil extracts were 25.5% and 37.5%, respectively, showing a substantial drop in soil toxicity. The EC50s were converted to toxicity units (TU, see Equation 31) to calculate the percent removal of heavy metal toxicity from a sandy soil following the magnetic iron filings (MIF) treatment. At 2.5% iron filing concentration, the toxicity removal was 55.5%, as compared to 87.9% and 92.0% at iron concentrations of 5% and 10%, respectively. In subsequent experiments we us ed 5% iron filings as a best cost-effective treatment concentration. Equa tion 3-2 shows the calculation of the % toxicity removal. TUbefore treatment TUafter treatment % Toxicity removal (TR) = 100% (3-2) TUbefore treatment Regarding the contact time between iron filings and the soil matrix, Table 3-2 shows that, after 1.5-hr contact time, the EC50 of the soil extract increased fr om 3.8% to 8.4% whereas after 3and 6-hr contact time, the EC50s were 44% and 36.4%, respect ively, suggesting a more substantial decrease in soil toxicity. Following conversion of EC50s to TUs, the percent toxicity removal was 54.8% after 1.5-hr contact time, as compared to 92.2% and 92.1% after 3-hr and 6hr contact times, respectively. A 3-hr contact time was chosen in subsequent experiments. 3.3.3 Evaluation of Magnetic Separation of Cu, Zn and Cd from Spiked Soils, Using MetPLATETM The magnetic removal of individual metals (Cu2+, Zn2+, Cd2+) from spiked sandy soil, red sandy soil, organic rich soil, and Ge orgia clay rich soil were inves tigated. Tables 3-3, 3-4, 3-5, and 3-6 show that the toxicity removal efficiency from the spiked soils varied with the type of metal and soil. For the three metals under study, we tested the toxicity of the soil extracts by

PAGE 52

52 determining the EC50s which were then converted into toxic ity units (TUs). The percent toxicity removal was calculated acco rding to Equation 3-2. Table 3-3 shows that the TUs in the sandy soil extracts varied from 21.9 TUs at 200 mg Cu/kg soil to 1661 TUs at 800 mg Cu2+/kg. The percent toxicity removal from the sandy soil was generally higher than 95%. With regard to the Zn-spiked sandy soil, 96.1% of the Zn toxicity was removed at the input concentration of 200 mg Zn2+/kg soil (Table 3-3). However, at 400 mg/kg and 800 mg Zn2+/kg soil, only 70.0% and 49.6% of the toxicity was removed by magnetic treatment. Therefore, additional treatments were necessary to achieve toxicity removal higher than 90%. As shown in Table 3-3, at 400 mg Zn2+/kg soil, 98.7% toxicity removal was achieved after two successive treatments whereas at 800 mg Zn2+/kg soil, 90.1% removal was obtained after three successive treatments. In the case of Cd-spiked sandy soil, a single magnetic treatment removed 51.1% of the Cd toxicity when the sp iked Cd concentration was 200 mg/kg soil. Cd removal increased to 92.6% following a second magnetic treatment (Table 3-3). At 400 mg Cd/kg soil the removal increased from 27.5% after one treatment to 91.9% after a second treatment. At 800 mg Cd/kg soil, metal removal reached 90.4% only after 3 magnetic treatments (Table 3-3). With regard to the Cu-spiked red sandy soil and Georgia clay rich soil, the percent toxicity removal was generally higher than 97% at all of the three Cu concentrations (200 mg/kg, 400 mg/kg, and 800 mg/kg) (Tables 3-4 and 3-6). In the case of Zn-spiked red sandy soil and Georgia clay rich soil, 97.6% and 97.4% of the toxicity was removed respectively from the soils when the spiked Zn concentration was 200 mg/kg (Tables 3-4 and 3-6). However, as the spiked Zn concentration increased to 400 mg/kg and 800 mg/ kg, additional treatments were required to reach a higher toxicity removal. Table 3-4 shows that at 400 mg Zn2+/kg red sandy soil, the

PAGE 53

53 toxicity removal increased from 68.5% to 98.5% after a second magnetic treatment, whereas at 800 mg Zn2+/kg red sandy soil, 87.5% removal was ach ieved after three successive treatments. Similar results were obtained toward Zn-spiked Georgia clay rich soil (Table 3-6), when Zn concentration was 400 mg/kg soil, 91.8% toxicity removal was achie ved after two treatments; at 800 mg Zn2+/kg soil, 90.3% removal was obtained after three successive treatments. After a single treatment of the Cu-spiked organic rich soil, the TUs of the soil extracts varied from 1.1 to 2.4 at Cu input concentratio ns from 200 mg/kg to 800 mg/kg, which result in a toxicity removal from 55.4% to 90.1% (Table 3-5) As discussed in Chapter 2, organic rich soil has a higher metal binding capacity than sandy soil s; therefore, compared to sandy soils, the spiked organic rich soil showed lower TUs be fore magnetic treatment, which could lead to a lower toxicity removal. Besides, as shown in Table 3-5, at 200 mg Zn2+/kg soil, the soil extract was not toxic after one treatment, at 400mg Zn2+/kg soil, 87.85 of the toxi city was removed after two successive treatments whereas at 800mg Zn2+/kg soil, 85.2% removal was obtained after three successive treatments. Cd-spiked soils showed higher toxicity than Znand Cu-spiked soils (Tables 3-3 through 3-6). In the case of Cd-spiked red sandy soil, organic rich soil, and Ge orgia clay rich soil, even at 200 mg Cd2+ /kg soil, two successive treatments were required to obtain a toxicity removal higher than 80.1%, and three magnetic treatment s were necessary for 400 mg/kg and 800 mg/kg Cd-spiked soils to achieve a toxicity removal be tween 80.1% and 95.4% (Tables 3-4, 3-5, and 36). In all, comparing the three heavy metals used, at equal concentration, Cd resulted in the highest toxicity (or highest TU va lues) in the soil extracts than Zn and Cu either before or after treatment, while Cu produced the lowest to xicity. Moreover, for each metal at the same

PAGE 54

54 concentration, the toxicity they produced in different soils increased as follows: sandy soils> Georgia clay rich soil > organic rich soil, which is associated with the soil heavy metal binding capacity (SHMBC; see Chapter 2). Moreover, th e magnetic treatment method worked best on Cu-spiked soils, followed by the Zn-spiked soils a nd the Cd-spiked soils. In a study of adsorption of divalent metals on iron (III) oxide, Tamura et al. (1994) found similar results and reported that the affinity for metals increased in the following order: Zn2+
PAGE 55

55 and potential problems may be en countered when highly water soluble contaminants leach out from the root zone (Cunningham et al., 1995) However, the magnetic treatment method proposed in our study can potentially be applied ex situ even on a soil contaminated with mixed metals. 3.3.4 Evaluation of Magnetic Separation of Cu, Zn and Cd from Spiked Soils, Using Ceriodaphnia dubia Acute Toxicity Test Metal toxicity removal from soils following ma gnetic treatment was also evaluated with the 48-hr Ceriodaphnia dubia acute toxicity test. The soils spiked with the highest metal concentration (800 mg/kg soil) used in the previous experiments were tested in this section. The results are shown in Table 3-7. The 48-h C. dubia test showed higher TU values than the MetPLATETM assay, due to the higher sensitivity of the daphnid test. For Cu, the toxicity removal from the four soils varied from 90.1% to 99.9% by MetPLATETM, and from 81.4% to 99.9% by C. dubia test. For Zn, after three treatments, the toxicity removal from the soils ranged from 85.2% to 90.3% by MetPLATETM, and from 82.7% to 89.2% by C. dubia test. As regards Cd-spiked soils, the toxicity removal percentage s after three treatments varied from 80.1% to 95.4% and from 84.8% to 96.2% as shown by MetPLATETM and C. dubia, respectively. The two tests showed, however, a simila r trend as regards toxicity removal of Cu, Zn or Cd from soils. 3.3.5 Evaluation of Magnetic Separation of Cu Zn and Cd from Spiked Sandy Soils, Using Selenastrum capricornutum Chronic Toxicity Test The soil extracts tested by the 48-h C. dubia assay were also evaluated with the 96-h Selenastrum capricornutum test. As shown in Table 3-8, th e 96-h algae test showed the highest TU values among the three toxicity tests. In othe r words, the 96-h algae te st displayed the highest sensitivity toward the three metals used. The three tests, however, showed a similar trend as regards toxicity removal of Cu, Zn or Cd from the soils. For Cu, the toxicity removal from the

PAGE 56

56 four soils varied from 90.0% to 99.9% as shown by the algae test. As regards Znand Cdspiked soils, after three treatments, the toxicity remova l percentages varied from 90.2% to 98.7% and from 94.9% to 96.0%, respectively. Comparing with the results discussed in Section 3.3.4, the metal toxicity removal shown by algae test was slightly higher th an that indicated by MetPLATETM and C. dubia. 3.3.6 Assessment of Metal Removal Effi ciency Using Chemical Analysis We have used toxicity tests to assess soil toxicity following the proposed magnetic treatment. We then used chemical analysis to study Cu, Zn, and Cd distributions in the soil matrix and extracts and to demonstrate that the metals were indeed adsorbed and concentrated on the iron filings. As shown in Tables 3-9, 3-10, an d 3-11, for each heavy metal, in the same soil, metal removal from the soil extract was always higher than that from the soil matrix. Cu removal. The sandy soil was spiked with copper to reach final concentrations of 200 mg/kg, 400 mg/kg and 800 mg/kg. Cu removal from the soil matrix ranged from 48.7% at 200 mg Cu/kg and 80.7% at 800 mg Cu/kg. Cu remova l from the soil extracts ranged from 96.9% at 200 mg Cu/kg and 99.5% at 800 mg Cu/kg. Table 3-9 also shows that Cu was adsorbed to the iron filings and reached Cu concentrations ranging from 2022 mg/kg filings at soil concentration of 200 mg Cu/kg and 12,186 mg/kg at soil concentration of 800 mg Cu/kg. Chemical analysis for the red sandy soil, organic rich soil and Geor gia clay rich soil spiked with 800mg Cu/kg soil was also performed. As shown in Table 3-9, the retrieved iron f ilings following magnetic treatment reached Cu concentrations of 10, 140 mg/kg, 6,262 mg/kg, and 11,400 mg/kg for red sandy soil, organic rich soil, a nd Georgia clay rich soil, respectively. The metal removal percentages from soil extracts were 98.6% for red sandy soil and 99.9% for Georgia clay rich soil, whereas the metal removal percentages from soil matrices were lower, which were 53.9% for red sandy soil and 85.2% for Georgia clay ri ch soil. In the case of 800 mg/kg Cu spiked

PAGE 57

57 organic rich soil, the metal removal percentages from both soil extract (83.9%) and soil matrix (49.1%) were lower than those from sandy so ils and Georgia clay rich soil. Zn and Cd removal. The soils (sandy, red sandy, orga nic rich, and Geor gia clay rich) spiked with the highest Zn and Cd concentra tion (800 mg/kg soil) were chemically analyzed. Since three successive treatments were required fo r these spiked soils, the chemical analysis was performed on the soil matrices and soil extracts after the final treatment, and a mixture of the iron filings retrieved form each treatment. As regards Zn removal (Table 3-10), no si gnificant difference was observed among the four types of soil extracts (sa ndy soil extract, 83.4%; red sandy soil extract, 72.0%; organic rich soil extract, 78.8%; Georgia clay rich soil extract, 71.1%). However, the removal of Zn from the soil matrices showed somewhat larger difference, varied from 4% in Georgia clay rich soil to 67.3% in sandy soil, with red sandy soil (14.6%) and organic rich soil (18.5%) in between. The concentrations of Zn adsorbed to the retrieved iron filings, varied from 559.2 mg/kg (from Georgia clay rich soil) to 3,371.9 mg/kg (from sa ndy soil), also demonstrated that the metal was indeed adsorbed and concentrated on the iron filings (Table 3-10). In the case of Cd removal, Table 3-11 indicates that the percentage removal from the soil extracts, ranged from 49.9% to 60.6%, was not si gnificantly affected by the type of soil. Whereas the removal of Cd from the soil matrices fo llowed this order: organic rich soil (5.2%) red sandy soil (5.7%) < Georgia clay rich soil (19.7%) < sandy soil (49.5%). The retrieved iron filings reached Cd concentrations form 117.6 mg /kg (from organic rich soil) to 2,386.9 mg/kg (from sandy soil). In all, comparing the three heavy metals (C u, Zn, and Cd) studied, the metal removal percentages from both the soil extr act and soil matrix followed this order in sandy soil, red sandy

PAGE 58

58 soil, and organic rich soil: Cu > Zn > Cd. However, in the Georgia clay rich soil, the same trend was found for metal removal from soil extracts, while a different trend (Cu> Cd> Zn) was shown for metal removal from soil matrix. The chemical analysis discussed above suggest ed that Cu, Zn, and Cd were removed from both the soil extracts and the soil matrix. As regards chemical immobilization, it is a remediation technique that uses chemical amendments to decrease the concentration of dissolved contaminants by sorption or precipitation (Bas ta and McGowen, 2004). Nu merous studies have been carried out to investigate the effectiv eness of lime and phosphate-based materials on immobilizing heavy metals in soils (Eighmy et al ., 1997; Hettiarachchi et al., 2000; Illera et al., 2004; Maenpaa et al., 2002; McGowen et al., 2001; Raicevic et al., 2005; Yoon et al., 2007). Ma et al. (1995) reported that phosphate rocks reduced water-soluble Pb from a contaminated soil by 56.8-100%. Chen et al. (1997) used mineral apatite to stabilize a Pb, Cd, and Zn contaminated soil, and they found that the removal of Cd a nd Zn by the apatite was pH dependent, whereas removal of Pb was not. The removals were 0. 729 mmol of Pb, 0.4891.317 mmol of Cd, and 0.596-2.187 mmol of Zn/g of apatite. Kumpiene et al. (2007) also evaluated the effectiveness of coal fly ash and natural organic matter (peat) in reducing soil Cu and Pb mobility. Their results indicated that the amount of leached Cu and Pb decreased by 74.5% and 61.0% after the addition of 5% organic matter, and by 91.1% and 87.1% afte r the addition of 5% coal fly ash. In another study carried out by Shanableh and Kharabsheh (1996) they studied the stab ilization of Cd, Ni, and Pb in a contaminated soil using a natural zeolite. At 500mg metal/kg soil, Pb leaching was reduced by more than 97% using a minimum of 25% zeolite. Using up to 50% zeolite, Ni and Cd leaching was reduced by a maximum of 50% and 60%, respectively. However, no matter what amendment is used, the aim of chemical imm obilization is to redu ce the metal solubility,

PAGE 59

59 mobility and toxicity, but the metal itself is stil l retained in the soil which may require further monitoring. Compared with the immobiliza tion approach, our magnetic separation method results in both immobization of metals as well as their physical removal from the soil matrix. 3.3.7 Mass Balance of Metals in Soils Tables 3-12, 3-13, and 3-14 display the mass bala nce of Cu, Zn, and Cd in the spiked soils (sandy, red sandy, organic rich, Georgia clay rich) before and after magnetic treatment. As shown in these tables, a large portion of the meta ls was immobilized in the soil matrix before treatment, but the metals associated with the so il matrix and in the soil extracts were both reduced following magnetic treatment. The tota l recovery of metals was not significantly affected by the soil type. Before treatment (i.e ., no iron filings added), the total recovery of added Cu (81.3%-91.3%), Zn (83.2%87.4%) and Cd (80.8%-91.3%) in the four soils were very close. After magnetic treatment, 82.2% to 95% of Cu, 80.8% to 88.3% of Zn, and 79.7% to 88.4% of Cd was recovered from the soils under study. 3.3.8 Sequential Extraction of Metals in Soils A sequential extraction procedure was performe d to show the distri bution of Cu, Zn, and Cd in a sandy soil and an orga nic rich soil before and after magnetic treatment. The sandy soil and organic rich soil were spiked with Cu, Zn, or Cd to reach a final metal concentration of 800 mg/kg. The results are shown in Fi gures 3-2, 3-3 and Table 3-15. Metal fractionations in sandy soil. Many factors, such as so il organic matter, cation exchange capacity, soil pH, and metal properties, can affect th e distribution of heavy metals (Buanuam et al, 2005; Ma and Rao, 1997). The most important sinks for metals are the Fe-Mn oxides, organic matter, sulfides, and carbonate s (Adriano, 2001c). Figure 3-2 displays the distribution of Cu, Zn, and Cd in treated and non -treated sandy soil. Before treatment, a large portion of Cu was bound to the carbonate phase (52.6%), and much lower proportions of Cu

PAGE 60

60 were extracted from the exchangeable (6.7%) Fe-Mn oxide bound (19%), organic bound (4.6%), and residual phases (17.1%). The percentage distribution of Zn in non-treated sandy soil was different from Cu, as 35.6% and 30.9% of Zn were associated with the carbonate and Fe-Mn oxide phase, respectively. Other re searchers have also found Zn to be associated with carbonates and Fe-Mn oxides (Chao et al., 2006; Ma and Rao, 1997; Ramos et al., 1994; urija and Branica, 1995). The exchangeable and residual Zn wa s 18.8% and 14.1%, respectively, whereas insignificant amount of Zn (0.6%) was bound to or ganic matter due to the very low organic content of the sandy soil. The distribution of Cd in non-residual fractions followed this order: exchangeable (54.9%) > car bonate-bound (20.6%) > Fe/Mn oxi de-bound (2.5%) > organicbound (1.2%). Maiz et al., (2000) s howed a similar Cd distribution in a polluted soil, which is exchangeable + carbonate phase (51%) > Fe-Mn oxide phase (29%) > organic phase (5%). As shown in Table 3-1, under a soil pH of 5.7, metals extracted by sodium acetate probably came from the soil fractions that are similar to carbonates. After magnetic treatment, as shown in Table 3-15, metal removal from the spiked sandy soil was the highest in the exchangeable phase (79.9% for Cu, 79.2% for Zn, and 65.8% for Cd), followed by the carbonate phase (46% for Cu, 44.0% for Zn, and 36.1% for Cd). However, Cu was not removed from the Fe-Mn oxide and the organic fractions. The percentage removal of Zn from the Fe-Mn oxide and organic factions were much lower, which were 24.4% and 8.1%, respectively. 29.2% of Cd was removed from the organic bound phase, whereas no Cd removal was observed in the Fe-Mn oxide fraction. Metal fractionations in organic soil. The distribution of Cu, Zn, and Cd factions in spiked organic rich soil was different from that in sandy so il (Figure 3-3). Before magnetic treatment, the larges t portion of Cu was bound to the or ganic matter (49.6%), followed by carbonate-bound (20.7%) and Fe-Mn oxide bound ( 20.3%). Much lower amount of Cu was

PAGE 61

61 retained in the exchangeable phase (1.9%) and residual phase (7.5%). The major association of Cu with the organic fraction is probably due to the high sorption capacity of organic matter for Cu and the stability of organic-Cu complexe s (Adriano, 2001c). Some other fractionation studies have also shown that appreciable am ounts of Cu are bound to organic matter (Kishk et al., 1973; Kuo et al., 1983; Ma and Rao, 1997). As regards Zn-spiked or ganic rich soil, only 2.1% of Zn was bound to the organic matter, whic h is the lowest fractio n among the five phases. Much higher proportions of Zn were distributed in the carbonate phase (33.7%), Fe-Mn oxide phase (34.4%), and residual phase (24.6%), and a small amount of Zn was found in the exchangeable fraction (5.2%). Figur e 3-3 also illustrates the dist ribution of Cd in the spiked organic rich soil, which followed the order: Ca rbonate (44.3%) > Exchangeable (26.9%) > FeMn oxide (21.8%) > residual (5.5%) > organic (1 .5%). After magnetic treatment, unlike in sandy soil, the removal of Cu, Zn, and Cd from the orga nic rich soil was observed in all fractions as shown in Table 3-15. The grea test removal was from the orga nic-bound phase, which was 45.1% for Cu, 48.3% for Zn, and 53.3% for Cd. The second highest removal was from the exchangeable phase for Cu (32.4%) and the Fe-Mn oxide phase for Zn (34.8%) and Cd (45.5%). Cu removal from the carbonate and Fe-Mn oxide factions we re somewhat lower, which were 22.3% and 29.5%, respectively. The removal of Zn and Cd from the exchangeable and carbonate phases was very close, varied from 16.1% to 21.5%. 3.3.9 Energy Dispersive X-ray Spectroscopy of Retrieved Iron Filings Energy dispersive X-ray anal ysis is a nondestructive method for the microanalysis of element composition (Heckmann et al., 2007). Numerous researchers have employed this technique to detect, measure, and determine the location of chemical elements within samples (Choel et al., 2005; Heckmann et al., 2007; Helsen and Van den Bulck, 1998; Lewis et al., 2000; Otulakowska and Nicholson, 2006; Petry et al., 2006). Al -Asheh and Duvnjak (1999) examined

PAGE 62

62 the mechanisms of metal biosorption by moss fr om solutions using scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopies (EDS), and they found that metal ions were sorbed mainly at the cell wa ll of the moss and only a small amount of ions diffused into the cytoplasm. Liu et al. (2007) also used SEM and EDS to investigat e the localization of Cd in the root tissue of Allium cepa L. In our study, the comparison of EDS between unused and used iron filings retrieved from Cu, Zn, and Cd-spiked sa ndy soils confirmed that Cu, Zn, and Cd were indeed adsorbed onto the iron fi lings (Figure 3-4). The aluminum detected on the retrieved iron filings probably came from th e sandy soil matrix. The silica pe aks shown in the EDS of both unused and used iron filings can be consid ered as impurity in the iron filings. 3.3.10 Regeneration of Iron Filings Yeager (1998) had treated the used iron filings with concentrated HNO3 for 10 minutes with shaking and then washed to remove h eavy metals from the surface; however, this regeneration method was ineff ective as displayed by MetPLATETM toxicity test results. Therefore, we investigated longer contact times (i.e. 1-hr 2-hr, and 24-hr) between HNO3 and iron filings. Table 3-16 shows the recove ry of iron filings from 1N HNO3 after three different regeneration periods. The 24-hr contact time rec overed only 43.2% of iron filings. Thus, shorter regeneration periods (2-hr and 1-hr) were used, and the recove ry of iron filings after two successive regenerations was studied for both of the two regeneration times. As regards 2-hr contact time, the recovery of iron filings reached 73.9% after the first regeneration; however, much lower recovery (48.4%) was obtained follow ing the second regeneration. In the case of 1hr regeneration, 86.2% and 85.3% of iron filings was recovered respectively following the first and second regeneration. Therefore, the 1-hr rege neration period was used for later experiments. To compare the effectiveness of magnetic tr eatments using fresh and regenerated iron filings for the removal of Cu, Zn, and Cd fr om a spiked sandy soil, two toxicity tests,

PAGE 63

63 MetPLATETM (Table 3-17) and the 48-h C. dubia (Table 3-18) were used to assess the toxicity of the soil extracts afte r treatment. The sandy soil was spiked with metal solutions to reach a Cu, Zn or Cd concentration of 400 mg/kg soil. As sh own in Table 3-17, for Cu-spiked sandy soil, the TUs of treated soil extracts were 1.1 TUs, 1.1TUs, and less than 1 TU by using fresh iron, iron regenerated once, and iron regene rated twice, respectively. For Zn-spiked sandy soil, 45.9 TUs, 25.8 TUs, and 18.5 TUs were produced respectively by the soil extracts treated with fresh iron, iron regenerated once, and iron regenerated twice. With regard to Cd-spiked sandy soil, the TUs of fresh iron treated soil extr act was 251.5 TUs, which were reduced to 161.2 TUs and 124.7 TUs by using iron filings regenerate d once and twice, respectively. The results obtained by using the 48-h C. dubia assay were quite similar (Table 3-18). After being treated with fresh iron filings, Cu-, Zn -, and Cdspiked soils extracts generated 24.8 TUs, 590.6 TUs, and 719.2 TUs, respectively. However, by using iron filings regenerated once and twice, the TUs were respectively reduced to 15.5 TUs and 6.4 TUs for Cu-spiked soil extracts, 174.6 TUs and 107.8 TUs for Zn-spike d soil extracts, and 389.2 TUs and 291.3 TUs for Cd-spiked soil extracts. Therefore, the adsorp tion capacity of regenerated iron filings was completely restored and even slightly highe r than that of the fresh iron filings. Other regeneration methods for iron (hydr)oxides have also b een evaluated by researchers. Peng et al. (2006) used montmor illonite-Cu(II)/Fe(III)oxides magnetic material as an adsorbent to successfully remove humic acid from solutions and they also found that the magnetic material can be thermally regenerated at 300C for 3 hrs. The regenerated adsorbent was still magnetic and had as good adsorption capacity as the unus ed material. In another study performed by Kornmuller et al. (2002), granulat ed iron hydroxide was used as a sorbent to remove reactive dye in textile wastewater. They reported that the spent iron hydr oxide could be regenerated by

PAGE 64

64 hydrogen peroxide at room temperature, and a regeneration time of 3.5-hr was necessary for decolorization and 6-hr was required to re store the adsorption capacity completely. 3.4 Conclusions The feasibility of decontamination of h eavy metal-contaminated soils by magnetic separation was demonstrated in laboratory experi ments with a sandy soil, red sandy soil, organic rich soil, and Georgia clay rich soil artificially spiked with Cu, Cd, and Zn. The results of this study showed a significant reduction of toxicity generated by Cu, Cd or Zn in soil extracts after one to three magnetic treatments. As regards toxi city reduction in soil extracts, this magnetic treatment method worked best on Cu-spiked so ils, followed by Zn-spiked soils and Cd-spiked soils. The speciation of Cu, Zn, and Cd dete rmined by sequential extractions was found to depend on the soil characteristics as well as the metal type. The comparison of metal fractionations in spiked soils before and after tr eatment suggested that, in the spiked sandy soil, the removal of Cu, Zn, and Cd was the greatest from the exchangeable (65.8%-79.7%) fraction, followed by the carbonate fraction (36.1%-46.0% ). Cu was not removed from both the Fe-Mn oxide and organic fractions, and Cd was not rem oved from the organic phase. In the spiked organic rich soil, the removal of Cu, Zn, and Cd was found in all fractions, in which the greatest removal was from the organic-bound phase (45.1% -53.3%), and the second highest removal was from the exchangeable phase for Cu (32.4%) an d the Fe-Mn oxide phase for Zn (34.8%) and Cd (45.5%). In addition, the ener gy dispersive X-ray spectroscopy (EDS) of magnetically retrieved iron filings confirmed the adsorption of Cu, Zn, and Cd on iron filings. Chemical analysis by ICP-AES s uggested that all the three me tals were removed from both the soil matrix and the soil extracts. However, me tal removal from the soil matrix was lower than removal from the soil extracts.

PAGE 65

65 Finally, the regeneration of used iron filings was also investigated, and the results indicated that the adsorption capacity of the iron filings for Cu, Zn, and Cd was completely restored by regenerating in 1N HNO3 for 1 hr and then in 1M NaOH for 72 hrs. In all, we conclude that this magnetic trea tment method shows great potential as a rapid ex situ remediation technology for heavy metal-contamin ated soils, and the retrieved iron filings could be regenerated and reused.

PAGE 66

66 Table 3-1. Soils characteristics Characteristic Red sandy soil Sandy soil Organic soil Georgia clay soil pH 6.25.75.75.7 Eh (mV) 485.0422.0403.0487.5 % Organic carbon 0.10.56.40.06 % Organic matter 0.31.618.80.1 % Sand 90.796.9293.241.0 % Silt 3.20.022.043.8 % Clay CEC (cmolc/kg) 6.1 14.4 3.06 14.1 4.8 230.1 15.2 27.6

PAGE 67

67Table 3-2. Effect of iron filings concentr ation and contact time between iron filings and soil matrix on the removal of heavy m etals from a spiked sandy soil, as determined by the MetPLATETM toxicity test EC50 c of soil extract (% soil extract) Toxicity unitsd of soil extract Spiked heavy metal mixturea (mg/kg) Study factors No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) 2.5 3.00.2% 6.80.7%* 33.52.1 14.91.5* 55.54.6%* 5 3.00.2% 25.55.0%** 33.52.1 4.00.8** 87.92.4%** Iron filings conc.b (%) 10 3.00.2% 37.43.8%*** 33.52.1 2.70.3*** 92.00.8%*** 1.5 3.80.2% 8.41.4%* 26.91.1 12.21.9* 54.87.2%* 3 3.50.6% 44.01.2%** 29.14.9 2.30.1** 92.20.2%** Cu2++Cd2++Zn2+ (40mg/kg for each) Iron filingse and soil contact time (h) 6 2.90.3% 36.44.7%*** 35.03.9 2.80.4** 92.11.0%** a 50 g of sandy soil was spiked with 40 mL of co mbined heavy metal solution containing 50 mg/L Cu2+, 50 mg/L Cd2+, and 50 mg/L Zn2+; b Iron filings and soil contact time was 6 hours; c Mean of 3 replicates 1 standard deviation; d Toxicity units were expressed as 100/EC50; e 5% (2.5g) iron filings were added to 50 g of spiked sandy soil; f Means followed by the same number of ast erisks within the same column are not significantly different at the 5% level.

PAGE 68

68 0 20 40 60 80 100 120Sandy soil Red sandy soilOrganic soilGeorgia clay soilARecovery of iron filings (%) 1% Fe 2% Fe 5% Fe 0 20 40 60 80 100 120Sandy soilRed sandy soilOrganic soilGeorgia clay soilBRecovery of iron filings (%) 1% Fe 2% Fe 3% Fe 0 20 40 60 80 100 120Sandy soilRed sandy soilOrganic soilGeorgia clay soilCRecovery of iron filings (%) 1% Fe 2% Fe 5% Fe Figure 3-1. Recovery of iron fili ngs from soils under different c ondition. A) Air-dried condition. B) Field capacity condition. C) Water satura ted condition. (Field capacities of sandy, red sandy, organic rich, and Georgia clay rich soil were 0.2 mL/g, 0.4 mL/g, 0.5 mL/g, and 0.7 mL/g, respectively. Error bars represent standard deviation of three replicates).

PAGE 69

69Table 3-3. Copper, zinc and cadmium toxicity re moval from a sandy soil by magnetic tr eatment, as determined by MetPLATETM. EC50 b of soil extract (% soil extract) Toxicity unitsc of soil extract Heavy metal type Heavy metal concentration in spiked soil (mg/kg) Number of treatmentsa No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) 200 1 4.60 0.6%>100% 21.9 2.8 <1g>95.4% 400 1 0.20 0.03%96.2 4.0% 481.0 57.2 1.04 0.04 99.8 0.03% Cud 800 1 0.06 0.0%51.3 4.9% 1661.2 96.1 2.0 0.2 99.9 0.0% 200 1 3.20 0.2%86.8 7.6% 30.9 1.81.2 0.296.1 1.0% 1 0.60 0.03%2.10 0.3% 164.2 8.149.3 7.370.0 4.2% 400 2 0.50 0.06%40.7 5.6% 200.6 23.62.5 0.398.7 0.3% 1 0.25 0.1%0.50 0.01% 411.3 94.3198.7 2.349.6 13.5% 2 0.26 0.02%0.53 0.05% 386.1 29.8188.6 17.651.2 1.3% Zne 800 3 0.20 0.03%2.10 0.2% 499.7 79.448.1 4.790.1 2.3% 1 1.50 0.1%3.40 1.0% 66.5 4.131.6 8.451.9 14.8% 200 2 0.20 0.03%3.30 1.0% 438.9 53.632.6 10.092.6 1.5% 1 0.30 0.01%0.40 0.05% 357.1 11.9250.6 32.327.5 7.8% 400 2 0.09 0.01%1.10 0.1% 1121.8 64.091.1 5.691.9 0.2% 1 0.10 0.00%0.13 0.01% 1013.7 15.8772.3 59.623.8 6.5% 2 0.03 0.01%0.10 0.01% 3962.4 1246.81006.0 95.072.5 10.7% Cdf 800 3 0.01 0.00%0.08 0.01% 12632.3 1591.41205.3 146.690.4 1.0% a 5% (2.5g) iron filings were a dded to the spiked sandy soil and the contact time was 3 hours; b Mean of 3 replicates 1 standard deviation; c Toxicity units = 100/EC50; d 50 g of sandy soil was spiked with 40 mL of Cu solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cu2+respectively; e 50 g of sandy soil was spiked with 40 mL of Zn solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Zn2+ respectively; f 50 g of sandy soil was spiked with 40 mL of Cd solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cd2+respectively; g Non-toxic.

PAGE 70

70Table 3-4. Copper, zinc and cadmium toxicity re moval from a red sandy soil by magnetic tr eatment, as determined by MetPLATETM. EC50 b of soil extract (% soil extract) Toxicity unitsc of soil extract Heavy metal type Heavy metal concentration in spiked soil (mg/kg) Number of treatmentsa No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) 200 1 1.1 0.1%44.9 8.4 % 94.1 7.1 2.3 0.597.6 1.0% 400 1 0.2 0.03%36.1 3.4% 552.2 74.5 2.8 0.3 99.5 0.1% Cud 800 1 0.05 0.01%7.3 2.2% 2055.6 419.6 14.5 3.7 99.3 0.3% 200 1 0.8 0.03%34.8 6.7% 125.2 5.32.9 0.597.6 0.5% 1 0.36 0.06%1.2 0.1% 280.6 48.487.8 10.768.5 1.9% 400 2 0.34 0.04%23.7 5.9% 299.8 35.24.4 1.398.5 0.6% 1 0.12 0.02%0.21 0.02% 867.5 119.1485.7 36.843.4 8.1% 2 0.12 0.01%0.40 0.01% 812.0 37.0252.1 3.768.9 1.9% Zne 800 3 0.12 0.01%0.94 0.01% 858.6 43.7106.8 0.787.5 0.7% 1 0.25 0.05%0.77 0.07% 417.9 94.1131.1 12.067.4 8.4% 200 2 0.20 0.02%2.3 0.4% 495.4 53.244.7 7.290.8 2.3% 1 0.09 0.01%0.17 0.03% 1120.4 125.3589.3 100.147.5 6.6% 400 2 3 0.07 0.01% 0.01 0.00% 0.36 0.04% 0.20 0.04% 1527.8 240.6 11111.1 0.00 279.6 27.0 516.4 77.2 81.2 4.8% 95.4 0.7% 1 0.04 0.01%0.09 0.02% 2777.8 481.11179.9 222.456.5 12.3% 2 0.02 0.01%0.08 0.01% 4444.4 962.31216.9 183.372.0 5.6% Cdf 800 3 0.003 0.00%0.07 0.02% 30555.6 4811.31402.1 278.795.4 0.4% a 5% (2.5g) iron filings were added to the spiked red sandy soil and the contact time was 3 hours; b Mean of 3 replicates 1 standard deviation; c Toxicity units = 100/EC50; d 50 g of red sandy soil was spiked with 40 mL of Cu solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cu2+respectively; e 50 g of red sandy soil was spiked with 40 mL of Zn solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Zn2+ respectively; f 50 g of red sandy soil was spiked with 40 mL of Cd solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cd2+respectively.

PAGE 71

71Table 3-5. Copper, zinc and cadmium toxicity re moval from an organic rich soil by magnetic treatment, as determined by MetPLATETM. EC50 b of soil extract (% soil extract) Toxicity unitsc of soil extract Heavy metal type Heavy metal concentration in spiked soil (mg/kg) Number of treatmentsa No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) 200 1 40.5 2.1%90.8 3.5 % 2.5 0.1 1.1 0.0455.4 0.7% 400 1 13.9 4.8%75.8 6.2% 7.7 2.3 1.3 0.1 81.7 5.7% Cud 800 1 4.0 0.3%41.2 4.0% 24.9 2.1 2.4 0.2 90.1 1.6% 200 1 33.7 7.9%> 100% 3.1 0.8< 1g> 66.3% 1 3.8 0.3%11.6 1.6% 26.3 2.08.7 1.267.0 2.3% 400 2 3.7 0.3%31.0 5.8% 27.4 2.53.3 0.687.8 3.0% 1 1.8 0.1%4.9 0.5% 54.6 1.720.6 2.162.4 2.8% 2 1.8 0.3%6.3 0.4% 56.6 9.616.0 0.971.4 3.3% Zne 800 3 2.2 0.03%15.0 1.0% 45.1 0.66.7 0.485.2 0.8% 1 28.1 1.3%47.2 3.4% 3.6 0.22.1 0.140.5 1.9% 200 2 19.9 1.7%> 100% 5.0 0.4< 1g> 80.1% 1 8.2 0.6%20.0 4.7% 12.2 0.95.2 1.157.0 11.3% 400 2 3 4.7 0.5% 4.1 0.1% 19.7 2.5% 27.1 0.6% 21.7 2.3 24.6 0.8 5.1 0.7 3.7 0.1 76.3 2.3% 84.9 0.8% 1 2.3 0.1%3.4 0.2% 42.9 2.229.7 1.430.7 1.1% 2 1.0 0.03%2.7 0.02% 96.2 2.437.6 0.360.9 0.7% Cdf 800 3 0.8 0.02%3.9 0.1% 129.3 2.525.3 0.680.4 0.3% a 5% (2.5g) iron filings were adde d to the spiked organic rich soil and the contact time was 3 hours; b Mean of 3 replicates 1 standard deviation; c Toxicity units = 100/EC50; d 50 g of organic rich soil was spiked with 40 mL of Cu solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cu2+respectively; e 50 g of organic rich soil was spiked with 40 mL of Zn solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Zn2+ respectively; f 50 g of organic rich soil was spiked with 40 mL of Cd solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cd2+respectively; g Non-toxic.

PAGE 72

72Table 3-6. Copper, zinc and cadmium toxicity rem oval from a Georgia clay rich soil by magnetic treatment, as determined by MetPLATETM. EC50 b of soil extract (% soil extract) Toxicity unitsc of soil extract Heavy metal type Heavy metal concentration in spiked soil (mg/kg) Number of treatmentsa No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) 200 1 2.8 0.4%> 100 % 35.8 4.8 < 1g> 97.2% 400 1 0.4 0.1%85.9 3.3% 234.3 35.2 1.2 0.1 99.5 0.1% Cud 800 1 0.16 0.01%53.2 3.2% 640.5 45.3 1.9 0.1 99.7 0.04% 200 1 2.1 0.2%81.4 13.2%47.9 4.81.2 0.2 97.4 0.2% 1 0.96 0.03%3.9 0.6%104.9 7.025.9 3.775.2 5.2% 400 2 0.43 0.05%5.2 0.6%233.9 2.719.3 2.291.8 0.8% 1 0.23 0.03%0.49 0.02%438.1 53.9206.4 9.052.4 7.9% 2 0.16 0.01%0.52 0.03%627.5 55.5192.6 10.569.3 1.0% Zne 800 3 0.11 0.02%1.2 0.07%903.2 136.987.1 5.490.3 0.9% 1 0.38 0.05%1.2 0.3%264.7 36.889.5 25.566.5 5.0% 200 2 0.44 0.07%3.7 0.7%230.2 37.027.3 5.088.2 0.3% 1 0.11 0.02%0.2 0.04%933.8 142.1485.0 92.448.1 2.0% 400 2 3 0.11 0.02% 0.11 0.01% 0.6 0.1% 0.7 0.1% 972.2 196.4 954.5 64.3 183.3 41.7 139.6 13.7 80.3 8.3% 85.4 0.5% 1 0.07 0.02%0.12 0.01%1594.5 464.7871.2 53.643.4 13.1% 2 0.06 0.001%0.13 0.01%1724.7 42.4801.3 45.353.6 1.5% Cdf 800 3 0.04 0.01%0.18 0.01%2916.7 589.3571.9 23.180.1 3.2% a 5% (2.5g) iron filings were added to the spiked Ge orgia clay rich soil and the contact time was 3 hours; b Mean of 3 replicates 1 standard deviation; c Toxicity units = 100/EC50; d 50 g of Georgia clay rich soil was spiked with 40 mL of Cu solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cu2+respectively; e 50 g of Georgia clay rich soil was spiked with 40 mL of Zn solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Zn2+ respectively; f 50g of Georgia clay soil was spiked with 40mL of Cd solution containing 250 mg/L, 500 mg/L, and 1000 mg/L Cd2+respectively; g Non-toxic

PAGE 73

73Table 3-7. Effect of magnetic treatment on the removal of Cu, Zn, and Cd from f our soils as determined by the 48-h acute Ceriodaphnia dubia toxicity test. EC50 c of soil extract (% soil extract) Toxicity unitsd of soil extract Soila type Metal conc. in spiked soila (mg/kg) Number of treatmentsb No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) Cu2+ (800mg/kg) 1 0.005 0.001%5.3 1.2%22148.6 5142.019.8 5.299.9 0.01% 1 0.016 0.0003%0.03 0.003%6062.7 125.33473.2 380.542.7 6.0% 2 0.017 0.001%0.04 0.005%5859.6 343.92308.8 309.960.7 3.2% Zn2+ (800mg/kg) 3 0.016 0.002%0.1 0.01%6164.4 600.0979.9 113.983.9 3.0% 1 0.014 0.002%0.03 0.01%7333.3 1154.83614.2 1262.151.8 10.6% 2 0.011 0.003%0.03 0.02%9792.0 2102.23426.7 1535.666.3 9.1% Sandy Cd2+ (800mg/kg) 3 0.009 0.0005%0.07 0.001%10759.0 588.81477.3 32.185.8 0.5% Cu2+ (800mg/kg) 1 0.005 0.001%0.6 0.1%20988.6 2669.8173.6 37.699.2 0.3% 1 0.017 0.001%0.02 0.002%5896.0 347.44444.4 481.124.7 4.6% 2 0.024 0.002%0.05 0.007%4181.8 314.91866.7 230.954.9 8.5% Zn2+ (800mg/kg) 3 0.022 0.004%0.1 0.006%4753.1 950.31036.7 59.577.4 6.3% 1 0.015 0.002%0.03 0.002%6869.7 742.33387.3 293.150.2 8.1% 2 0.008 0.001%0.04 0.002%13112.8 1197.62504.2 125.480.8 2.5% Red sandy Cd2+ (800mg/kg) 3 0.008 0.001%0.05 0.004%13461.5 1665.42111.1 157.184.8 1.1% a 50 g of soil was spiked with 40 mL of heavy metal solution containing 1000 mg/L Cu2+, 1000 mg/L Zn2+, and 1000 mg/L Cd2+ respectively; b 5% (2.5 g) iron filings were added to the sp iked soil and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Toxicity units = 100/EC50.

PAGE 74

74Table 3-7. Continued. EC50 c of soil extract (% soil extract) Toxicity unitsd of soil extract Soila type Metal conc. in spiked soila (mg/kg) Number of treatmentsb No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) Cu2+ (800mg/kg) 1 0.9 0.2%4.9 0.6% 112.4 18.620.8 2.481.4 1.3% 1 0.5 0.1%1.3 0.2% 196.7 41.376.3 8.859.4 13.5% 2 0.5 0.1%1.6 0.2% 211.8 42.162.1 6.169.8 7.2% Zn2+ (800mg/kg) 3 0.60.1%5.9 0.4% 158.2 16.017.0 1.289.2 1.7% 1 0.7 0.1%0.9 0.1% 151.0 23.7107.2 12.728.7 3.1% 2 0.4 0.1%0.8 0.05% 247.7 65.0121.4 8.149.4 9.6% Organic rich Cd2+ (800mg/kg) 3 0.3 0.07%2.2 0.5% 386.8 87.5 51.2 5.885.8 3.0% Cu2+ (800mg/kg) 1 0.070.003%4.4 0.1%e 1481.559.522.9 0.898.5 0.01% 1 0.080.004%0.1 0.003% 1180.152.7831.5 18.829.4 4.7% 2 0.080.004%0.14 0.004% 1282.172.5737.2 22.742.5 1.5% Zn2+ (800mg/kg) 3 0.070.007%0.4 0.04% 1547.6168.4265.2 30.182.7 3.8% 1 0.020.001%0.040.004% 4591.6363.12418.4206.051.2 0.6% 2 0.020.001%0.06 0.02% 5558.3174.71669.0412.669.8 8.4% Georgia clay rich Cd2+ (800mg/kg) 3 0.0040.0005%0.1 0.01% 25731.23113.8 983.8 81.896.2 0.1% a 50 g of soil was spiked with 40 mL of heavy metal solution containing 1000 mg/L Cu2+, 1000 mg/L Zn2+, and 1000 mg/L Cd2+ respectively; b 5% (2.5 g) iron filings were added to the sp iked soil and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Toxicity units = 100/EC50.

PAGE 75

75Table 3-8. Effect of magnetic treatment on the removal of Cu, Zn, and Cd from four soils as determined by the 96-h chronic Selenastrum capricornutum toxicity test. EC50 c of soil extract (% soil extract) Toxicity unitsd of soil extract Soila type Metal conc. in spiked soila (mg/kg) Number of treatmentsb No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) Cu2+ (800mg/kg) 1 0.003 0.000%2.3 0.4% 35744.7 1277.445.5 9.399.9 0.02% 1 0.007 0.001%0.03 0.006% 13639.0 1164.13848.3 714.171.4 7.4% 2 0.006 0.001%0.04 0.008% 17725.9 2693.52573.6 565.484.9 5.9% Zn2+ (800mg/kg) 3 0.006 0.002%0.07 0.005% 17474.3 6062.81475.5 106.190.9 2.9% 1 0.003 0.0004%0.008 0.001% 28912.1 3498.112584.6 1067.056.2 5.5% 2 0.003 0.0004%0.02 0.001% 31100.5 4850.85179.5 310.583.2 1.7% Sandy Cd2+ (800mg/kg) 3 0.003 0.0006%0.05 0.001% 41483.1 9259.62105.5 31.395.5 0.1% Cu2+ (800mg/kg) 1 0.003 0.001%0.6 0.1% 31846.2 6832.4181.6 29.599.4 0.2% 1 0.007 0.001%0.03 0.001% 12823.6 1785.02904.3 100.877.1 2.3% 2 0.005 0.001%0.08 0.004% 18856.7 2893.71196.7 54.293.6 1.1% Zn2+ (800mg/kg) 3 0.004 0.001%0.15 0.002% 28170.4 8165.6688.4 10.797.5 0.7% 1 0.001 0.0002%0.014 0.001% 70277.7 8825.37192.6 325.889.7 1.1% 2 0.001 0.0001%0.02 0.004% 76907.7 6313.04703.8 973.893.9 0.8% Red sandy Cd2+ (800mg/kg) 3 0.001 0.0002%0.03 0.006% 69444.0 7750.43516.8 801.994.9 0.9% a 50 g of soil was spiked with 40 mL of heavy metal solution containing 1000 mg/L Cu2+, 1000 mg/L Zn2+, and 1000 mg/L Cd2+ respectively; b 5% (2.5 g) iron filings were added to the sp iked soil and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Toxicity units = 100/EC50.

PAGE 76

76Table 3-8. Continued. EC50 c of soil extract (% soil extract) Toxicity unitsd of soil extract Soila type Metal conc. in spiked soila (mg/kg) Number of treatmentsb No Treatment Magnetic Treatment No Treatment Magnetic Treatment Toxicity removal (%) Cu2+ (800mg/kg) 1 0.4 0.04%4.4 0.4% 231.2 19.323.0 2.290.0 1.4% 1 0.2 0.02%0.5 0.09% 439.9 46.3188.9 30.756.3 11.1% 2 0.2 0.01%0.7 0.04% 510.4 18.6138.2 7.672.9 1.7% Zn2+ (800mg/kg) 3 0.2 0.01%3.2 0.2% 423.3 15.930.2 0.492.7 0.2% 1 0.07 0.001%0.2 0.01% 1433.6 10.5527.0 34.263.4 2.2% 2 0.05 0.004%0.5 0.1% 1888.8 133.6191.0 37.689.8 2.6% Organic Cd2+ (800mg/kg) 3 0.04 0.004%1.1 0.04% 2375.1 222.094.7 4.196.0 0.4% Cu2+ (800mg/kg) 1 0.003 0.0006%2.8 0.1% 30921.3 5341.336.3 0.799.9 0.02% 1 0.004 0.0001%0.09 0.002% 22554.3 719.01148.9 21.094.9 0.1% 2 0.003 0.001%0.12 0.02% 29081.1 4640.5870.0 157.097.0 0.1% Zn2+ (800mg/kg) 3 0.003 0.0002%0.21 0.02% 36377.4 2408.9482.1 39.798.7 0.02% 1 0.002 0.000%0.02 0.003% 52265.7 753.25890.5 996.488.7 1.7% 2 0.002 0.000%0.03 0.0003% 60243.9 3356.73701.3 41.693.8 0.4% Georgia clay Cd2+ (800mg/kg) 3 0.002 0.000%0.04 0.003% 57939.9 7624.52578.0 196.195.5 0.9% a 50 g of soil was spiked with 40 mL of heavy metal solution containing 1000 mg/L Cu2+, 1000 mg/L Zn2+, and 1000 mg/L Cd2+ respectively; b 5% (2.5 g) iron filings were added to the sp iked soil and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Toxicity units = 100/EC50.

PAGE 77

77Table 3-9. Effect of magnetic treatment on the removal of Cu2+ from spiked soils, as determined by chemical analysis. Cu conc.c in soil matrix, extracts, and iron filingsd Cu removal from soil fractions Soil type Initial Cu conc. in spiked soil (mg/kg)a In soil matrix (mg/kg) In soil extract (mg/L) Adsorbed on iron filings (mg/kg) From soil matrix (%) From soil extract (%) No treatment 171.9 3.1e3.8 0.6 No iron filings added 200 Magnetic treatmentb 88.0 9.90.1 0.04 2022.466.9 48.76.4%96.90.3% No treatment 279.8.739.4 1.1 No iron filings added 400 Magnetic treatment 71.3 5.60.4 0.07 5583.4424.3 74.24.2%99.10.2% No treatment 362.0.6144.2 1.0 No iron filings added Sandy 800 Magnetic treatment 69.82.00.8 0.07 12186.7200.7 80.70.8%99.50.05% No treatment 453.1 21.1277.6 7.8 No iron filings added53.9.8%98.6.1% Red sandy 800 Magnetic treatment 209.0 19.13.8 0.4 10140.0402.9 No treatment 705.8 21.12.6 0.3 No iron filings added49.1 3.8%83.9 1.5% Organic rich 800 Magnetic treatment 358.9 20.10.4 0.02 6261.7 124.4 No treatment 591.5 34.4110.3 3.8 No iron filings added85.2 3.6%99.9 0.04% Georgia clay rich 800 Magnetic treatment 87.3 19.00.1 0.04 11400 542.9 a 50 g of soil was spiked with 40 mL of Cu solution containing 250 mg/L, or 500 mg/L, or 1000 mg/L Cu2+; b 5% (2.5 g) iron filings were added to the spiked soil, and the contact time was 3 hours; c Detected by ICP-AES; d Iron filings had a Cu background value of 2503.3 mg/kg. This concentration was subtracted from the Cu concentration in filings after treatment; e Mean of 3 replicates one standard deviation.

PAGE 78

78Table 3-10. Effect of magnetic treatment on the removal of Zn2+ from spiked soils, as determined by chemical analysis. Zn conc.c in soil matrix, extracts, and iron filingsd Zn removal from soil fractions Soil type Initial Zn conc. in spiked soil (mg/kg)a In soil matrix (mg/kg) In soil extract (mg/L) Adsorbed on iron filings (mg/kg) From soil matrix (%) From soil extract (%) No treatment 423.411.3e 242.6 4.4 No iron filings added Sandy 800 Magnetic treatmentb 142.4 19.740.5 8.6 3371.9170.6 67.35.6%83.43.2% No treatment 383.4.3288.2 8.3 No iron filings added14.6.0%72.0.8% Red sandy 800 Magnetic treatment 327.4 5.580.5 4.6 1590.641.9 No treatment 694.4 19.55.8 0.5 No iron filings added18.5 5.2%78.8 3.8% Organic rich 800 Magnetic treatment 565.5 20.41.2 0.3 559.2 20.5 No treatment 477.5 4.1187.9 3.6 No iron filings added4.0 1.7%71.1 0.7% Georgia clay rich 800 Magnetic treatment 458.6 12.154.2 0.4 1291.9 36.8 a 50 g of soil was spiked with 40 mL of Zn solution containing 1000 mg/L Zn2+; b 5% (2.5 g) iron filings were a dded to the spiked soil, and the contact time was 3 hours; c Detected by ICP-AES; d Iron filings had a Zn background value of 13.1 mg/kg. This concentration was subtracted from the Zn concentration in filings after treatment; e Mean of 3 replicates one standard deviation.

PAGE 79

79Table 3-11. Effect of magnetic treatment on the removal of Cd2+ from spiked soils, as determined by chemical analysis. Cd conc.c in soil matrix, extracts, and iron filingsd Cd removal from soil fractions Soil type Initial Cd conc. in spiked soil (mg/kg)a In soil matrix (mg/kg) In soil extract (mg/L) Adsorbed on iron filings (mg/kg) From soil matrix (%) From soil extract (%) No treatment 434.738.5e 235.8 5.9 No iron filings added Sandy 800 Magnetic treatmentb 217.810.295.0 1.3 2386.9102.9 49.56.4%59.70.8% No treatment 384.7.5261.6 4.1 No iron filings added5.7.8%54.0.0% Red sandy 800 Magnetic treatment 362.79.1120.3 0.9 1084.933.3 No treatment 726.2 38.13.9 1.2 No iron filings added5.2 1.8%60.6 7.7% Organic 800 Magnetic treatment 688.2 22.81.5 0.2 117.6 9.7 No treatment 507.6 22.8152.6 2.4 No iron filings added19.7 7.4%49.9 0.6% Georgia clay 800 Magnetic treatment 407.0 19.281.0 0.4 955.9 66.0 a 50 g of soil was spiked with 40 mL of Cd solution containing 1000 mg/L Cd2+; b 5% (2.5 g) iron filings were a dded to the spiked soil, and the contact time was 3 hours; c Detected by ICP-AES; d Iron filings had a Cd background value of 108.4 mg/kg. This concentration was subtracted from the Cd concentration in filings after treatment; e Mean of 3 replicates one standard deviation.

PAGE 80

80Table 3-12. Mass balance of Cu in spiked soils before and after magnetic treatment. Soila type Initial Cu mass (mg) Cu in soil matrix (mg) Cu in soil extract (mg) Cu adsorbed on iron filingsc (mg) Total recovered Cu mass (mg) Total recovery of Cu (%) Before treatmentb 8.60.20.40.1No iron filings added9.00.3d90.03% 10 After treatment 4.40.50.010.0045.10.29.50.795.07% Before treatment 14.01.53.90.1No iron filings added17.91.689.58% 20 After treatment 3.60.30.040.0114.01.117.61.488.07% Before treatment 18.10.214.40.1No iron filings added32.50.381.30.8% Sandy 40 After treatment 3.50.10.10.0130.50.534.10.685.31.5% Before treatment 22.71.113.90.4No iron filings added36.50.991.32.3% Red sandy 40 After treatment 10.51.00.20.0225.41.036.00.690.01.5% Before treatment 35.31.10.10.01No iron filings added35.41.188.52.6% Organic rich 40 After treatment 17.91.00.020.0015.70.333.60.784.11.8% Before treatment 29.61.75.5 0.2No iron filings added35.11.687.73.9% Georgia clay rich 40 After treatment 4.41.00.010.00228.51.432.91.382.23.1% a 50 g of soil was spiked with 40 mL of Cu solution containing 250 mg/L, or 500 mg/L, or 1000 mg/L Cu2+; b 5% (2.5 g) iron f ilings were added to the spiked soil for 3 hours; c The amount of background Cu mass in iron filings was s ubtracted from the Cu mass in filings after treatment; d Mean of 3 replicates one standard deviation.

PAGE 81

81Table 3-13. Mass balance of Zn in spiked soils before and after magnetic treatment. Soila type Initial Zn mass (mg) Zn in soil matrix (mg) Zn in soil extract (mg) Zn adsorbed on iron filingsc (mg) Total recovered Zn mass (mg) Total recovery of Zn (%) Before treatment 21.20.612.10. 2No iron filings added33.30.8d83.31.9% Sandy 40 After treatment 7.11.02.00.425.31.334.41.186.12.8% Before treatment 19.20.514.40.4No iron filings added33.60.884.01.9% Red sandy 40 After treatment 16.40.34.00.211.90.332.30.780.81.7% Before treatment 34.71.00.20.02No iron filings added35.01.087.42.5% Organic rich 40 After treatment 28.31.00.050.014.20.232.50.981.32.3% Before treatment 23.90.29.40.2No iron filings added33.30.483.21.0% Georgia clay rich 40 After treatment 22.90.62.70.029.70.335.30.988.32.2% a 50 g of soil was spiked with 40 mL of Zn solution containing 1000 mg/L Zn2+; b 5% (2.5 g) iron filings were a dded to the spiked soil for 3 hours; c The amount of background Zn mass in iron filings was s ubtracted from the Zn mass in filings after treatment; d Mean of 3 replicates one standard deviation.

PAGE 82

82Table 3-14. Mass balance of Cd in spiked soils before and after magnetic treatment. Soila type Initial Cd mass (mg) Cd in soil matrix (mg) Cd in soil extract (mg) Cd adsorbed on iron filingsc (mg) Total recovered Cd mass (mg) Total recovery of Cd (%) Before treatment 21.71.911.80. 3No iron filings added33.51.7d83.84.2% Sandy 40 After treatment 10.90.54.80.117.90.833.50.583.91.2% Before treatment 19.20.413.10.2No iron filings added32.30.280.80.6% Red sandy 40 After treatment 18.10.46.00.18.10.332.30.680.71.4% Before treatment 36.31.90.20.06No iron filings added36.52.091.34.9% Organic rich 40 After treatment 34.41.10.070.010.880.0735.41.288.43.0% Before treatment 25.41.17.60.1No iron filings added33.01.382.53.2% Georgia clay rich 40 After treatment 20.41.04.10.027.50.531.90.579.71.1% a 50 g of soil was spiked with 40 mL of Cd solution containing 1000 mg/L Cd2+; b 5% (2.5 g) iron filings were added to the spiked soil for 3 hours; c The amount of background Cd mass in iron filings was s ubtracted from the Cd mass in filings after treatment; d Mean of 3 replicates one standard deviation.

PAGE 83

83 a a a aa a b b0 10 20 30 40 50 60 70 ExchangeableCarbonatesbound Fe/Mn oxidesbound Organic matter-boundCuDistribution (%) Before treatment After treatment a a a a a b b b 0 10 20 30 40 50 60 ExchangeableCarbonatesbound Fe/Mn oxidesbound Organic matter-boundZnDistribution (%) Before treatment After treatment a a a a a a b b0 10 20 30 40 50 60 70 ExchangeableCarbonatesbound Fe/Mn oxides-bound Organic matter-boundCdDistribution (%) Before treatment After treatment Figure 3-2. Distribution of Cu, Zn, and Cd fractions in a spiked sandy soil. (Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Students t -test)

PAGE 84

84 a a a a b b a a 0 10 20 30 40 50 60 ExchangeableCarbonatesbound Fe/Mn oxidesbound Organic matter-boundCuDistribution (%) Before treatment After treatment a a a a b b b b 0 10 20 30 40 ExchangeableCarbonatesbound Fe/Mn oxidesbound Organic matter-boundZnDistribution (%) Before treatment After treatment a a a a b b b a 0 10 20 30 40 50 60 ExchangeableCarbonatesbound Fe/Mn oxidesbound Organic matter-boundCdDistribution (%) Before treatment After treatment Figure 3-3. Distribution of Cu, Zn, and Cd fractions in a spiked organic rich soil. (Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Students t -test).

PAGE 85

85 Table 3-15. Effect of magnetic treatment on the removal of Cu, Zn, and Cd from each soil fraction, as determined by sequential extraction. Metal removal from each fraction after treatmentb (%) Soila type Metal type Exchangeable Carbonates-boundFe/Mn oxidesbound Organic matterbound Cu 79.9 3.3%c46.0 12.4%-7.6 4.4 0 0% Zn 79.2 1.7%44.0 2.3%24.4 4.5% 8.1 3.3% Sandy Cd 65.8 6.2%36.1 5.9%0 0% 29.2 6.4% Cu 32.4 8.6%22.3 11.6%29.5 6.1% 45.1 2.1% Zn 16.1 5.1%19.6 3.4%34.8 8.0% 48.3 5.5% Organic Cd 21.5 5.0%17.0 2.1%45.5 2.8% 53.3 11.7% a Soil was spiked with 200 mg/kg of Cu, Zn, and Cd; b 5% (2.5 g) iron filings were added to the spiked soil for 3 hours; c Mean of 3 replicates one standard deviation. Figure 3-4. Energy dispersive x-ray spectroscopy (EDS) of iron filings. A) was unused Fe. B) was retrieved from 800 mg/kg Cu-spiked sandy soil, C) was retrieved from 800 mg/kg Zn-spiked sandy soil, and D) was retrieved from 800 mg/kg Cd-spiked sandy soil. A B

PAGE 86

86 Figure 3-4. Continued. D C

PAGE 87

87Table 3-16. Effect of contact time between ir on filings and 1 N HNO3 on the recovery of fresh iron filings. Contact time between iron filings and 1N HNO3 24-hr 2-hr 1-hr Number of regenerations 1st regeneration 1st regeneration 2nd regeneration 1st regeneration 2nd regeneration Recovery of iron filings (%) 43.2 1.3% 73. 9.6% 48.4 2.5% 86.2 0.9% 85.3 0.4% a Mean of 3 replicates one standard deviation Table 3-17. Comparison of the toxicity of sandy soil extracts treated with fresh ir on filings and regenerated iron filings, as determined by MetPLATETM. Fresh Fe After 1st regeneration After 2nd regeneration Heavy metal in spiked soila EC50 of treated soil extract (% soil extract) TUsb of treated soil extract EC50 of treated soil extract (% soil extract) TUs of treated soil extract EC50 of treated soil extract (% soil extract) TUs of treated soil extract Cu 87.01.6%c1.10.0291.69.0%1.10.1>100%<1d Zn 2.20.1%45.92.34.11.3%25.88.05.40.5%18.51.8 Cd 0.40.02%251.512.20.60.1%161.214.90.80.02%124.73.4 a 50 g of sandy soil was spiked with 40 mL of Cu, Zn, or Cd solution containing 500 mg/L Cu2+, Zn2+, and Cd2+, respectively; b Toxicity unit (TU) = 100/EC50; c Mean of 3 replicates one standard deviation; d Non-toxic. Table 3-18. Comparison of the toxicity of sandy soil extracts treated by fresh iron fi lings and regenerated iron filings, as de termined by the 48-h acute Ceriodaphnia dubia toxicity test. Fresh Fe After 1st regeneration After 2nd regeneration Heavy metal in spiked soila EC50 of treated soil extract (% soil extract) TUsb of treated soil extract EC50 of treated soil extract (% soil extract) TUs of treated soil extract EC50 of treated soil extract (% soil extract) TUs of treated soil extract Cu 4.00.2%c24.81.46.72.0%15.54.515.60.2%6.40.1 Zn 0.20.003%590.610.60.60.3%174.618.01.00.2%107.825.3 Cd 0.140.01%719.241.00.260.04%389.261.20.340.02%291.319.0 a 50 g of sandy soil was spiked with 40 mL of Cu, Zn, or Cd solution containing 500 mg/L Cu2+, Zn2+, and Cd2+, respectively; b Toxicity unit (TU) = 100/EC50 ; c Mean of 3 replicates one standard deviation.

PAGE 88

88 CHAPTER 4 EFFECT OF AGING OF METAL-SPIKED SO ILS ON METAL TOXICITY AND REMOVAL USING MAGNETIC SEPARATION 4.1 Introduction After soluble metals are added to soil, severa l reactions may occur, which can change the partitioning of metals between the aqueous and solid phases in soil and thereby their mobility and bioavailability to organism s (Adriano, 2001b; Ma et al., 2006a ). Sorption is considered the most important process controlling the partitioning of metals in soils, which may represent the combined effects of ion exchange, specific adso rption, precipitation, and complexation (Adriano, 2001b). Factors influencing metal behavior in soil s include soil characteristics, metal species, and the contact time between soil and metals (Adriano, 2001b; Naidu et al., 2003a; Weng et al., 2002). Soil pH is a main factor that controls metal partitioning, followed by organic matter which is a time-sensitive parameter (Daous t et al., 2006) and subject to long-term transformations (e.g. chemical or biological proce sses) (Martnez et al., 20 03). It has been shown that an increase of soil pH and organic matter ge nerally decrease the mobili ty and, thereby, the (phyto)toxicity of metals (Adriano, 2001b; Alva et al., 2000; Sa uv et al., 2000). The term aging refers to the process during which the availability of certain compounds changes as the compounds stay in soil for so me time (Alexander, 1995). Some studies have revealed that as the residence time increases, th e binding of metals to so ils tends to increase, which could result in lower toxicity and availability of metals in aged contaminated soils than in freshly spiked soils (Alexander, 1995; Amorim et al., 2005; Smit and Van Gestel, 1998; Ma et al., 2006a, b; Song et al., 2006; Oorts et al., 2007; Stewart et al., 2003). Therefore, the bioavailability of metals derive d from laboratory experiments usi ng freshly spiked soils may be overestimated (Van Straalen and Denneman 1989), and the addition of metal salts to soils may

PAGE 89

89 result in a different metal speciation than in the field (Pedersen et al., 2000). The decrease of metal mobility with time could be generall y due to micropore diffusion, cavity entrapment, occlusion in soil phase, or surface precipitation (Ma et al., 2006a). Although long-term aging is a very important pr ocess which can change metal availability and toxicity overtime, it is usuall y not considered in the risk asse ssment of metals in field soils, and very limited research on this subject has b een done so far (Ma et al., 2006a). Ma et al. (2006a) have studied the aging of Cu added to 19 European soils based on a two-year period. The results indicated that the to tal Cu in soil pore water from leached soils decreased rapidly initially followed by further decrease at slower rates. They concluded that soil pH was a vital factor affecting the aging rate of Cu. More over, by employing a semi-mechanistic model, they found that when the soil pH was below 5, only sl ow processes (likely diffusion) occurred. Another study conducted by Arias-Estves et al. ( 2007) has investigated th e influence of aging on Cu fractionation in an acid soil. After 500 days of incubation, the change of soil pH was negligible. However, Cu in the exchangeable and organically-bound fac tions decreased with increasing incubation time, whereas Cu concentrati on in the residual faction increased. Ma and Uren (2006) also studied the effect of aging on the phytoavailability of Zn in a slightly calcareous soil. This twoyear green house st udy provided biological evidence that long-term aging decreases the bioavailability of Zn to corn. In addition, an investiga tion of the transport of Cd in an aged sandy soil was carried out by Seunt jens et al. (2001). No conclusive evidence was found that aging affects Cd transport. In our study, we examined the effect of agi ng on the change of Cu and Zn toxicity in spiked soils. Besides, the aging effect of contaminants should also be considered when performing remediation in field soils. In other wo rds, remediation strategies that show promising

PAGE 90

90 results in laboratory experiment s using freshly spiked soils may not work as well in aged soils under field conditions. Thus, we also examined the effectiveness of the proposed magnetic treatment on decontaminating aged Cuand Zn-spiked soils. 4.2 Material and Methods 4.2.1 Soils Used The sandy soil and organic rich soil used in Chap ter 2 were further tested in this section of study. Table 3-1 shows the main characteristics of the soils under study. All soils samples were first air-dried, screened (sieve # 10 ; 2.0 mm particles) and homogenized. 4.2.2 Chemicals Used Two heavy metal solutions (Cu2+ and Zn2+) were used. Copper solution was prepared from copper sulfate (CuSO4H2O, Sigma), and zinc solution was prepared from zinc chloride (ZnCl2, Sigma). Iron filings (Fisher, 40 mesh) were placed in 1M NaOH for 72 hours to increase the adsorption capacity and then washed thoroughly with distilled water be fore use (Yeager, 1998). 4.2.3 Soil Aging without Wet/Dry Cycle Three kilogram of sandy soil was spiked with 1L of 600 mg/L Cu2+ or Zn2+ solution to reach a final soil metal concentr ation of 200 mg/kg. Three kilogram of organic rich soil was spiked with 2 L of 1200 mg/L Cu2+ or Zn2+ solution, resulting in 800 mg-metal /kg soil. These metal concentrations were based on the toxic leve ls determined in Chapter 2. The volume of the heavy metal solutions was above the soil field cap acity to ensure a sufficient contact between the metal ions and the soil particles. The spiked soil slurries were mixed very well and dried at room temperature. A portion of each dried soil was re gularly tested for toxicity and treated by magnetic separation over a 4-month period, with 1 m onth interval. Freshly spiked soils served as controls and all tests were run in triplicate.

PAGE 91

91 4.2.4 Soil Aging with Wet/Dry Cycle After 4-month aging as discussed above, a wetdry procedure (3 days for sandy soil, 5 days for organic rich soil) was performed to the spiked sandy soil and organic rich soil to simulate the field conditions in Florida duri ng the rainy season. The dried so ils were wetted with distilled water to reached saturation, and then the soil sl urries were mixed very well and allowed to dry again at room temperature. After four wet-dry cycles, a portion of dried soil was tested for toxicity and treated by magnetic separation me thod. The entire aging experiment lasted approximately 7 months, including the initial 4-month aging and 20 wet-dry cycles. All tests were run in triplicate. 4.2.5 Toxicity of Aged Soil Extracts One hundred milliliter of distilled water was added to 50 g of spike soil, and the soil mixture was shaken for 4 hours. Then the soil slurry was centrifuge d at 10,000 rpm for 15 minutes and the supernatant (soil extract) was re moved with a pipet and was used for toxicity tests after overnight settling in the refr igerator. Two toxicity assays, MetPLATETM and the 48-h Ceriodaphnia dubia acute bioassay (as discussed in Sect ion 3.2.6, also see Appendix A for detailed procedures) were used to assess the toxi city of the aged soil extracts. To determine the EC50s for the soil extracts, a regre ssion analysis was performed. Th e toxicity units (TU) were then calculated according to Equation 3-1 (see Chap ter 3). All samples were run in triplicate. 4.2.6 Magnetic Treatment of Aged Soils One hundred milliliter of distilled water was added to 50 g of spiked soil, and the soil slurry was shaken for 1 hour at room temperature. Then, 5% (2.5 g) of iron filings were added and the system was shaken for 3 hours followed by magnetic retrieval of th e iron filings. The soil extracts were then separated from the solid phase by centrifuging at 10, 000 rpm for 15 mins, and the supernatant (soil extract) was removed with a pipet. All experi ments were run in triplicate.

PAGE 92

92 4.3 Results and Discussion 4.3.1 Effect of Aging on Cu and Zn Toxicity 4.3.1.1 Toxicity of Cu and Zn in aged sandy soil Figure 4-1 represents the change of Cu and Zn toxicity in spiked sandy soil over a 4-month period and 20 wet-dry cycles. In 200 mg/kg Cuspiked sandy soil, the toxicity of Cu in soil extracts determined by both MetPLATETM and the 48-hr C. dubia test did not show significant ( = 0.05) overall change over the 4-month dr y aging period, with TU values of 43.3 by MetPLATETM and 443.7 TUs by C .dubia at time 0, and 40.9 TUs by MetPLATETM and 461.4 TUs by C.dubia at the 4th month. However, after subjecti ng the sandy soil to 20 wet-dry cycles, Cu toxicity decreased significantly ( = 0.05) and gradually with time. After 20 wet-dry cycles, the toxicity of the sandy so il extracts decreased to 11.8 TUs when using MetPLATETM and to 148.6 TUs when using the C. dubia test. The change of Zn toxicity in aged sandy soil is also displayed in Figure 4-1. After 1-month aging, the toxicity of Zn did not change significantly as shown by neither MetPLATETM nor C. dubia test. However, after aging for 2 months, Zn toxicity in the sandy so il started to decrease significantly ( = 0.05) as shown by both toxicity tests. When using MetPLATETM toxicity test, the TU values of the soil extracts decreased from 268.7 TUs at month 1 to 182.6 TUs at month 4, and then continued decreasing to 88.0 after the 20th wet-dry cycle. The 48-h C .dubia test showed similar trends as MetPLATETM. Over the entire aging period (4 -month dry aging plus 20 wet-dry cycles), the Zn toxicity in soil extracts droppe d from a initial value of 520.7 TUs to 174.4 TUs, during which a very dramatic d ecrease was observed after the 12th wet-dry cycle (Figure 4-1). These findings generally agreed with those repo rted by other researchers (Arias-Estves et al., 2007; Lock and Janssens, 2003; Ma et al., 2006a; Ma and Ur en, 2006; Oorts et al., 2007; Song et al., 2006; Stewart et al., 2003; Tye et al., 2003). Pede rsen and van Gestel (2001)

PAGE 93

93 compared the toxicity of copper to the collembolan Folsomia fimetaria L. in a spiked soil and in a soil from old Cu-contaminated field site. La rge differences in effects were found between spiked soil and field soil when concentrations were expressed on the basis of total soil Cu concentration. The EC10 and EC50 values for reproduction of Folsomia fimetaria in spiked soil were 700 and 1400 mg Cu/kg soil, whereas no effects were found in field soil at Cu concentrations up to 2500mg/kg. Ma et al. (2006a) studied the long-term (2 years) effect of aging on Cu in 19 soil samples. They found that upon the addition of water soluble Cu to the soils, free Cu2+ activity in pore water in the soils with lo w pH (<5.5) and high organic matter content (>12.94%) was not significantly a ffected by incubation time. However, for the other soils, the free Cu2+ activity in soil pore water d ecreased with incubation time. Tye et al. (2003) examined the changes of Cd and Zn concentrations in so il pore water in 23 soils over an 818-day period. They reported that the changes in labile Cd a nd Zn were dependent on both time and pH. At low pH values, only a small decrease in metal concentrations was found over the aging period, whereas at high soil pH levels, greater decrease with time was observed. In addition, both Zn2+ and Cd2+ activities in soil pore water decreased with time. They also pointed out that the changes in metal concentrations and activities were proba bly due to time-dependent fixation of added metals and the readjustment of soil pH. In a nother study performed by Ma and Uren (2006), the effect of aging on the availability of Zn in a slightly calcareous soil was investigated. Their results showed that uptake of Zn by corn decr eased over a 2-year aging period. Meanwhile, they also studied the effect of agi ng on the speciation and extractabili ty of Zn added to the soil by a sequential extraction procedure. They concluded th at when water-soluble Zn was added to a soil, it changed gradually into less available forms, mo stly into the forms asso ciated with Fe (Al) oxides and minerals.

PAGE 94

94 4.3.1.2 Toxicity of Cu and Zn in aged organic rich soil Soil organic matter is a critical component for the retention of heavy metals in soils; however, both the character and stability of the organic materials can influence the partitioning, speciation and environmental fate of metals (Martnez et al., 2003). In our study, as shown in Figure 4-1, in 800 mg /kg Cu-spiked organic ri ch soil, the toxicity of Cu did not significantly ( = 0.05) change during the entire aging period (4-month dry aging plus 20 wet-dry cycles) as determined by C. dubia, and TU values ranged from 18.1 to 25.7. When using MetPLATETM toxicity test, the toxicity of Cu was slightly increased from 4.6 TUs at time 0 to 5.5 TUs at month 4 and after the 20th wet-dry cycle, the TU value further increased to 7.6. Long-term chemical or biological transf ormation of soil organic matter may result in the release of dissolved organic mater and a poten tial increase in heavy metal mobility and bioavailability (Martnez et al ., 2003). Martnez and his colleague s (2003) studied the combined effect of time and temperature on metal release and speciation from a metal contaminated soil (120 mg/kg Pb, 39.3 mg/kg Cd, 420 mg/kg Cu 1480 mg/kg Zn, 100 mg/kg Ni, 800 mg/kg Mn, and 13 mg/kg Mo; 12.3% organic carbon) over a twomonth period. Their results indicated that dissolved organic carbon (DOC) increased both with temperature and incubation time, and in general, total soluble metal re lease measured by inductively c oupled plasma spectroscopy (ICP) paralleled the behavior of DOC. As regards the change of Zn toxicity in spik ed organic rich soil (Figure 4-1), when using the MetPLATETM toxicity test, from month 2 to the 12th wet-dry cycle, Zn toxicity slightly increased from 45.4TUs to 57.2TUs. However, after the 12th cycle, Zn toxicity started dropping, and the TU value after the 20th cycle decreased to 40.1, which was significantly ( = 0.05) lower than the TU value at time 0. When using C.dubia test, from time 0 to the 12th wet-dry cycle, no significant change of Zn toxicity was observed, whereas from the 12th to the 20th cycle, the

PAGE 95

95 toxicity of Zn decr eased significantly ( = 0.05) from 108.6 TUs to 63.7 TUs. Lock and Janssen (2003) studied the effect of aging on Zn toxicity and bioavailability in 20 soils (organic matter content ranged from 1.5% to 12%), and they repo rted that in freshly spiked soils, Zn partitioning is mainly determined by pH and cation exchange capacity, whereas only pH determines metal availability in aged soil. They also found that Zn availability was the highes t in soils with low pH values, and aging has little effect on Zn bioavailability in these soils. 4.3.2 Effect of Aging on Magnetic Treatment The effect of aging on magnetic separation of Cu and Zn from soils was also investigated. A comparison of treating freshly contaminated soils and aged soils is shown in Figure 4-2. After testing the toxicity of the trea ted soil extracts by both MetPLATETM and C. dubia, it is found that soil aging had no adverse effect on magnetic separation of Cu from spiked sandy soil and organic rich soil. Throughout the entire ag ing period, after magnetic treatment, the TU values of all Cuspiked sandy soil extracts were less than 1 as determined by MetPLATETM, and decreased from 6.2 TUs at time 0 to 2.8 TUs after the 20th wet-dry cycle, as determined by the 48-h C .dubia test. In the case of Cu-spiked or ganic rich soil, the toxicity of all treated soil extracts did not change much over the aging period, varyi ng from 1.8 TUs to 2.2 TUs when using MetPLATETM, and decreased slightly from 5.4 TUs at time 0 to 2.8 TUs at the end when using C. dubia test. In the case of Zn removal from the aged sandy soil, after magnetic treatment, the TU values of freshly spiked so il extract were 1.2 by MetPLATETM and 5.0 by C. dubia, and the TU values of aged soil extracts va ried from 1.1 to 1.7 by MetPLATETM, and from 3.5 to 5.3 by C. dubia. With regard to Zn removal from organic rich soil, compared to the TU values of treated soil extract at time 0 ( 30.1 by MetPLATETM and 30.8 by C. dubia), the toxicity of treated soil extracts after different aging periods va ried from 21.0 TUs to 27.9 TUs by MetPLATETM, and

PAGE 96

96 from 24.8 TUs to 32.0 TUs by C. dubia. In conclusion, the aged so ils extracts showed less or very close TU values to the freshly spiked soil extracts after magnetic tr eatment, which indicated that aging had no significant adverse effect on magnetic separation of Zn from sandy soil and organic rich soil. It has been recognized that, in most cases, the chemical state and bioa vailability of freshly spiked metals is very different from that of the same material in aged environmental medium (Sauv, 2003). Therefore, ignoring changes in meta l bioavailability due to the effect of aging may lead to underor overestimation in risk assess ment of contaminated soils (Lock and Janssen, 2003). However, it has also been reported that agi ng has little effect on me tal availability in soils with low pH values; therefore, it should not be assumed that aging will eventually resolve all problems associated with metal contaminated soils (Lock and Janssen, 2003). Moreover, the effects of aging also raise quest ions about the possible effectiv eness of remediation of sites containing compounds that may have aged and the application of existing models to predict the fate of chemicals at real fi eld sites (Hatzinger and Alexande r, 1995). Therefore, caution should be taken when relating results derived from benc h scale laboratory tests using spiked soils to remediation of aged contaminated soils under fi eld conditions. Ottosen et al. (2006) compared the effectiveness of electrodialytic removal of Cu from spiked kaolinite, spiked soil and industrially polluted soils under the same operati onal conditions. The results showed that the removal rate was higher in kaolinite than in bot h spiked soil and industr ially polluted soils. The duration of spiking was also important in simula tion remediation of industr ially polluted soils. It was found that spiking for 2 days was too short; however, after 30 da ys the spiked soil showed a pattern similar to that of i ndustrially polluted soils both rega rding sequential extraction and remediation result, though the remediation still progr essed slightly faster in the spiked soil. In

PAGE 97

97 our study, as regards the effect of aging on magnetic separation of Cu and Zn from sandy soil and organic soil, we did not find significant adverse effect over the 7-month aging period, which indicated the potential of utilizing this magnetic separation method to remediate contaminated soils under field conditions. 4.4 Conclusions In this study, the change of Cu and Zn toxic ity in aged soils and the effect of aging on magnetic separation of Cu and Zn from soils were investigated. During the initial 4-month aging, as determined by either MetPLATETM or the 48-h C. dubia test, the toxicity of Cu in sandy soil did not show significant change. However, afte r 20 wet-dry cycles, Cu toxicity decreased significantly and gradually with ti me. Zn toxicity in aged sandy soil gradually decreased after 2month aging until the 20th wet-day cycle as shown by both toxi city tests. Compared to freshlyspiked organic rich soil, Cu toxicity di d not change significantly as determined by C. dubia test, whereas MetPLATETM showed a slight increase in Cu toxicity Zn toxicity in aged organic rich soil only showed some decrease after the 12th wet-dry cycle as shown by both toxicity tests. As regards the effect of aging on ma gnetic separation of Cu and Zn from aged soils, no significant adverse effect was observed. Further investig ation using longer aging periods would be necessary to evaluate the significance of this study.

PAGE 98

98 Figure 4-1. Toxicity of Cu and Zn in an aged sandy soil (A) dry aging (B) wet-dry aging, and an organic rich soil (A) day aging (B) wet-dry aging over a 4-month period and 20 wet-dry cycles, as determined by MetPLATETM and the 48-h acute Ceriodaphia dubia assay (TU value indicated by one asterisk was significantly lower at the 5% level than the TU value at time 0; TU value indicated by two asteri sks was significantly higher at the 5% level than the TU value at time 0. Error bars represent standard deviation of three replicates). A A B B

PAGE 99

99 Figure 4-2. Effect of soil aging on magnetic treatment of spiked sandy soil (A) dry aging (B) wet-dry aging, and organic rich s oil (A) dry aging (B) wet-dry aging over a 4-month period and 20 wet-dry cycles (T oxicity of treated soil extracts were measured by MetPLATETM and the 48-h acute Ceriodaphnia dubia assay. All Cu-spiked sandy soil extracts showed less than 1 TU after treatment, as determined by MetPLATETM. Error bars represent standard deviation of three replicates). A A B B

PAGE 100

100 CHAPTER 5 LEAD REMOVAL FROM SHOOTING RANGE SO ILS USING MAGNETIC SEPARATION 5.1 Introduction Lead (Pb), as a well known toxic heavy metal, has been attracting much attention due to its widespread distribution and poten tial adverse effect on the envi ronment (Zeng et al., 2007). Pbcontaminated soils can be found in both urban and rural environments (Chen et al., 2003). Humans can be exposed to Pb vi a air, drinking water, food, pa int, and industrial releases (Adriano, 2001d). The Florida Depa rtment of Environmental prot ection (FDEP) has established a soil cleanup target level (SCTL) of 400 mg/kg for Pb in residential and commercial/industrial areas (FEDP, 2005). The interact ion between Pb and soil compone nts mainly includes specific adsorption, precipitation, and the formation of complexes or chelates with organic matter (Adriano, 2001d). Specific adsorption of Pb to Mn (hydr)oxide is pr oven to be greater than that to any other metal (hydr)oxides (H ettiarachchi et al., 2000). Due to the strong affinity of Pb for organic matter and its generally immobile nature lead usually accumulates in the surface layers of soils (Adriano, 2001d). The mobility and bioa vailability of Pb is highly site-specific (Turpeinen et al., 2000) and also depends on th e forms and species of Pb (Adriano, 2001d). To clean up Pb-contaminated soils, many remedi ation strategies have been proposed. Soil washing is one of the commonly used ex-situ technique, with a goa l of transferring the contaminants from the solid to the aqueous pha se (Grasso et al., 1997). Many extraction agents including acids, neutral salts, and chelating ag ents have been tested and showed successful applications under certain condi tions (Cline and Reed, 1995; Is oyama and Wada, 2007; Kos and Letan, 2003; Lin et al ., 2001; Neilson et al., 2003; Reed et al., 1996; Van Benschoten et al., 1997). Electroremediation and phytoremediation are emerging techniques for removing metals from soils. Some laboratory and pilot-scale experi ments have demonstrated the effectiveness of

PAGE 101

101 electroremediation to extract Pb from soils (Altin and Degirm enci, 2005; Amrate et al., 2005; Chung and kang, 1999; Yang and Lin, 1998). To apply phytoremediation successfully, hyperaccumulators are required. By definition (Baker at al., 1994), a hyperaccumulator for Pb should reach more than 1000 mg-P b /kg-shoot dry matter. Cui et al. (2004b) have shown that Indian mustard (Brassica juncea L.) and winter wheat (Triticum aestivum L.) can extract Pb from soils with the addition of elemental sulphur and EDTA; some other studies have shown that the use of vetiver grass (Vetiveria zizanoides), corn (Zea mays L.), or morning glory (Ipomoea lacunose L.) associated with the application of ED TA showed great pote ntial as a remedial strategy (Hovsepyan and Greip sson, 2005; Kambhampati et al., 2003; Wilde et al., 2005). In-situ Pb immobilization has been considered as a cost-effective alternative. Many amendments including coal fly ash, zeolite, na tural organic matter, and gypsumand lime-rich industrial by products have been utilized in chemical immobilizati on of Pb (Illera et al., 2004; Kumpiene at al., 2007; Shanableh and Kharabsheh 1996), and numerous studies have indicated the successful immobilization of Pb by phosphor us-containing material s (Chen et al., 2003; Hettiarachchi et al., 2000; Hettiarachchi and Pier zynski, 2002; Melamed et al., 2003; Tang et al., 2004; Yang et al., 2001; Yoon et al., 2007). In this study, we investigated the effectiven ess of removing Pb from shooting range soils using magnetic separation. The toxicity of the so il extracts before and after magnetic treatment was determined by MetPLATETM assay and the 48-h Ceriodaphnia dubia acute toxicity test. Chemical analysis by ICP-AES was also performe d to investigate Pb distribution in the soil matrix and extracts.

PAGE 102

102 5.2 Material and Methods 5.2.1 Soils and Chemicals Used Five soil samples were collected from differe nt out-door shooting ra nges in Florida. All these shooting facilities were operated for years. Soil 1 was collected from the mid-berm of a rifle shooting range in Ocala, Florida. Soil 2 and soil 3 were collected from the mid-berm of two different rifle areas at the sa me shooting range in Osceola, Florida. Soil 4 and soil 5 were collected respectively from a pistol range and a ri fle range located in Gainesville, Florida, and both of them were sampled from the mid-berm. All soils samples were first air-dried, screened (sieve # 10; 2.0 mm particles) and homogenized. Table 5-1 shows the soil characteristics and the total Pb concentrations in the soil samples. Iron filings (Fisher, 40 mesh) were placed in 1 M NaOH for 72 hours to increase the adsorption capacity and then washed thoroughly with distilled water be fore use (Yeager, 1998). 5.2.2 Toxicity of Soils under Study Twenty milliliter of distilled water was added to 10 g of soil samples. The soil slurry was shaken for 4 hours at room temperature, and then centrifuged at 10,000 rpm for 15 minutes and the supernatant (soil extract) was removed with a pipet. The toxici ty of the soil extracts were determined by measuring the EC50s of the samples using the MetPLATETM toxicity assay (see Appendix A for detailed MetPLATETM protocol), and then the EC50s were converted to toxicity units (TUs) according to Equation 3-1 (see Chapter 3) All soil samples as well as toxicity tests were run in triplicate. 5.2.3 Magnetic Separation of Pb from Shooting Range Soils Twenty milliliter of distilled water was added to 10 g of soil samples. The soil slurry was shaken for 1 hour at room temperature. Then, 5% (0.5 g) of iron fili ngs were added and the system was shaken for 3 hours followed by magnetic retrieval of the iron filings. The supernatant

PAGE 103

103 (soil extract) was separated from the solid pha se by centrifuging at 10,000 rpm for 15 minutes. The retrieved iron filings and the soil matrix following centrifugation were dried at 70C overnight. Soil without iron treatment served as the control. Toxicity tests (i.e. MetPLATETM and the 48-h acute Ceriodaphnia dubia assay) were performed for all soil extracts before and after treatment. Chemical analysis was carried ou t for all factions of samples, including soil extracts, retrieved iron filings, and the soil matr ix. Each soil sample was run in triplicate. Detailed procedures for MetPLATETM assay and the 48-h acute Ceriodaphnia dubia test were included in Appendix A. 5.2.4 Chemical Analysis Chemical analysis was undertaken for all fract ions, including the soil matrix, soil extracts and iron filings, before and after magnetic treatmen t. Soil extracts were digested using the U.S. EPA method 3010A (US EPA, 1992). Soils and iron filings were digested according to the U.S. EPA method 3050B (US EPA, 1996). All digested samples were analyzed for metals using inductively coupled plasma-atomic emission spectroscopy (ICP-AES). See Appendix B for detailed digestion procedures. 5.3 Results and Discussion 5.3.1 Toxicity of Soils under Study The total Pb concentration in the five soil samples varied from 1,538 mg/kg to 70,000 mg/kg, with an order of soil 1> soil 2> soil 5> so il 3> soil 4, which i ndicated Pb contamination of the shooting range soils. These findings suppo rt many recent studies that revealed very high Pb concentrations in the soils of shooting ra nge (Cao et al., 2003; Chen and Daroub, 2002; Darling and Thomas, 2003; Hardison et al., 2004). Figure 5-1 shows the toxicity of the shooting range soil extracts determined by MetPLATETM assay. The toxicity units of these five soil extracts followed this order: soil 2 (36.6 TUs) > soil 3 (16.8 TUs) > soil 5 (14.2 TUs) > soil 1

PAGE 104

104 (5.9 TUs)> soil 4 (4.5 TUs), which did not comple tely comply with the trend of the total Pb concentrations, possibly due to metal speciation a nd bioavailability. Therefore, it confirmed that like most trace metals, total Pb concentration in soils usually is not a good indicator of Pb bioavailability (Adriano, 2001d). Park et al. (2003) examined the bioavailability of metals at eight military shooting ranges in the Kyungkido a nd Kangwondo districts in Korea by measuring the exchangeable and soluble metal fractions as well as the metal content in the tissues of the plants growing at these sites. Although all of the sites were se riously contaminated with heavy metals, especially Pb and Cd, their results showed that only a small fraction of the total metal was soluble and as a result the metal bioavailability was very low. In ecotoxicology, bioavailability can be broa dly defined as the portion of a chemical in the environment which is available for biologic al action (e.g. uptake by an organism) (Adriano, 2001a). Generally speaking, bioavailability is a f unction of the solubility and mobility of metals (Adriano, 2001a), which mainly depends on the me tals chemical behavior, soil properties and the individual characteristics of the receptor (S iebielec et al., 2006). It is recognized that increasing soil pH leads to decrease of Pb content in plant roots. Be sides, Pb has a strong affinity for organic matter and is generally immobile in soil (Adriano, 2001d). Pb can also be immobilized in soil by Fe and Mn oxides through specific ad sorption and/or coprecipitation processes (Hettiarachchi and Pierzynski, 2004). Various analytical methods have been used to assess metal availability, among which the chemical extraction technique is the most commonly used method. The soluble metal content in soil solution plus the exchangeable fraction provide a good measure of the plant-available amount (Adriano, 2001a). However, plant physiol ogy and rhizosphere biochemistry can change the relationship between the extr actant and plant tissu e concentration (Basta et al., 2005). The

PAGE 105

105 use of organisms (e.g. plant or soil microorgani sms) to assess metal bioavailability could be more reliable but there are still concerns a bout the most suitable or ganisms and about the extrapolation of results from the laboratory to the field (Adriano, 2001a). Metal bioavailability and toxicity to soil microorganisms can be esti mated by the impact of metals on the microbial community at either the populati on level (size of the population, community structure, species diversity, etc.) or at the f unctional level (e.g. respirati on, element cycling, C and N mineralization, etc.) (Naidu et al., 2003b). Si ebielec et al. (2006) evaluated the metal bioavailability and toxicity in 40 long-term c ontaminated soils sampled from the Tarnowskie Gory area (a metal mining area) of Poland. Metal availability was measured using chemical extraction, microbial activity and wheat (Triticum aestivum, L.) growth. Their study showed that despite high metal contents in most soils, the bioa vailability of Zn, Pb, and Cd was relatively low and mostly independent on total metal contents. Th ey also reported that soil contamination with metals did not reduce microbial activity, such as nitrification potential in soils, and the most contaminated soils had the highest microbial ac tivity due to their relatively high organic matter and clay contents and neutral pH optimal for bacteria activity. 5.3.2 Evaluation of Magnetic Separation of Pb from Shooting Range Soils, Using MetPLATETM and Ceriodaphnia dubia Acute Toxicity Tests After a single magnetic treatment of the shooting range soils, as determined by MetPLATETM (Table 5-2), all soil extracts became non-t oxic with TU values less than 1, and the percent toxicity removal from the soils was gene rally higher than 77.8%. Pb toxicity removal from shooting range soils following magnetic tr eatment was also evaluated with the 48-hr Ceriodaphnia dubia acute toxicity test (Table 5-3). As discussed in Chapter 3, the 48-h C. dubia test showed higher TU va lues than the MetPLATETM assay, due to the higher sensitivity of the daphnid test. The TU values of soil 1 and so il 2 were respectively 18.2 and 32.7 before

PAGE 106

106 treatment, which then decreased to 1.5 and 1.3 after single magnetic treatment, and 91.6% and 95.9% of toxicity was removed from soil 1 and so il 2, respectively. The percent toxicity removal from Soil 3 and soil 4 was higher than 88.7%, wh ich resulted in non-toxi c soil extracts after magnetic treatment. As regards soil 5, the TU values of the soil extract decreased from 15.3 to 1.4 after treatment, which repres ents 91.0% toxicity removal. 5.3.3 Assessment of Pb Removal Effi ciency Using Chemical Analysis Following toxicity testing to assess Pb toxicity removal by the proposed magnetic treatment, we then used chemical analysis to study Pb distributions in the soil matrix and extracts. As shown in Table 5-4, for each soil, a large portion of Pb was retained in the soil matrix, and little amount of Pb was present in soil extracts, which also indicated the low bioavailability of these aged shooting range soils. Pb removal from soil extracts was very high, ranged from 78.0% to 98.3%, whereas much lowe r Pb removal (2.1% to 12.5%) was found from the soil matrix. Table 5-4 also shows that Pb was adsorbed to the iron filings and reached Pb concentrations ranging from 242.4 mg/kg fo r soil 4 and 3529.5 mg/kg for soil 1. 5.4 Conclusions By comparing toxicity of the soil extracts and th e total Pb in the soils, it is confirmed that total metal concentrations may not be used to pred ict the bioavailability a nd thus the toxicity of this metal in natural soils. As regards the effectiveness of the magnetic separation method on removing Pb from shooting range soils, the results have shown that a great reduction of toxicity generated by Pb was obtained after single magnetic treatment. The chemical analysis suggested that a large portion of Pb was retained in the soil matrix, and although Pb was removed from both of the soil matrix and soil extracts, the remova l from soil matrix was always lower than that from the soil extracts. Moreover, the chemical an alysis also demonstrated that Pb was indeed adsorbed and concentrated on the iron filings Therefore, this magnetic separation method

PAGE 107

107 showed great potential in decontaminating heavy metal-contaminated soils that have been aged for years under field condition.

PAGE 108

108 Table 5-1. Soils characteristics Characteristic Soil 1Soil 2 Soil 3 Soil 4 Soil 5 pH 6.25.15.66.4 6.8 % Organic carbon 0.10.40.40.4 0.5 % Organic matter 0.20.70.70.7 0.9 % Sand 86.678.389.493.8 88.0 % Silt 9.517.27.92.9 7.1 % Clay CEC (cmolc/kg) Total Pb (mg/kg) 3.9 11.1 70,000 4.5 11.0 12,400 2.7 5.8 3,256 3.3 4.9 16.5 24.9 1,538 11,490 5.9 4.5 16.8 36.6 14.2 0 10 20 30 40 50 Soil 1Soil 2Soil 3Soil 4Soil 5TUs of soil extracts Figure 5-1. Toxicity of shooting range soil extracts, as determined by MetPLATETM assay (error bars represent standard devi ation of three replicates)

PAGE 109

109Table 5-2. Lead toxicity removal from five shooting ra nge soils by magnetic treatment, as determined by MetPLATETM EC50 of soil extract (% soil extract) Toxicity unitsc of soil extract Soil ID Pb2+ conc. in soil (mg/kg) Treatmenta times No treatment After Magnetic treatment No treatment After Magnetic treatment Toxicity removal (%) Soil 1 70,000 1 17.1 2.9% b>100%5.9 1.0<1d>82.9% Soil 2 12,400 1 2.8 0.7%>100%36.6 8.6<1>97.2% Soil 3 3,256 1 6.0 0.5%>100%16.8 1.5<1>94.0% Soil 4 1,538 1 22.2 1.2%>100%4.5 0.2<1>77.8% Soil 5 11,490 1 7.2 1.4%>100%14.2 2.7<1>92.8% a 5% (0.5 g) iron filings were added to the shooting range soils, and the contact time was 3 hours; b Mean of 3 replicates one standard deviation; c Toxicity units = 100/EC50; d Non-toxic. Table 5-3. Lead toxicity removal from five shooting range soils by magnetic treatment, as determined by the 48-h acute Ceriodaphnia dubia test EC50 of soil extract (% soil extract) Toxicity unitsb of soil extract Soil ID Pb2+ conc. in soil (mg/kg) Treatmenta times No treatment After Magnetic treatment No treatment After Magnetic treatment Toxicity removal (%) Soil 1 70,000 1 5.5 0.3%c 65.5 0.8 18.2 0.8 1.5 0.02 91.6 0.5% Soil 2 12,400 1 3.1 0.1% 75.4 1.0% 32.7 1.2 1.3 0.02 95.9 0.1% Soil 3 3,256 1 3.0 0.1% >100% 33.1 0.6 <1d >97.0% Soil 4 1,538 1 11.3 0.2% >100% 8.8 0.1 <1 >88.7% Soil 5 11,490 1 6.5 0.1% 72.5 5.3% 15.3 0.3 1.4 0.1 91.0 .5% a 5% (0.5 g) iron filings were added to the shooting range soils and the contact time was 3 hours; b Toxicity units = 100/EC50; c mean of 3 replicates one standard deviation; d Non-toxic.

PAGE 110

110Table 5-4. Effect of magnetic treatment on the removal of Pb from shooting range so ils, as determined by chemical analysis Pb conc.b in soil matrix, extracts, and iron filings Pb removal from soil fractions (%) Soil ID Initial Pb conc. in soil (mg/kg) In soil matrix (mg/kg) In soil extract (mg/L) Adsorbed on iron filings(mg/kg) From soil matrix From soil extract No treatment 6768552.610.7 0.8No iron filings added Soil 1 70,000 Magnetic treatmenta 662701.50.2 0.013529.5 274.1c2.1 0.9%d 98.3 0.01%No treatment 109326.931.4 2.7No iron filings addedSoil 2 12,400 Magnetic treatment 9950.531.22.1 0.3916.1 64.7c9.0 0.7%93.2 1.5%No treatment 2962.0 17.823.5 5.0No iron filings addedSoil 3 3,256 Magnetic treatment 2862.0 49.91.8 0.2575.2 92.8c 3.3 2.0%92.4 0.6%No treatment 1328.5 74.24.8 .01No iron filings addedSoil 4 1,538 Magnetic treatment 1163.1 72.20.5 0.1242.4 6.4c12.5 .5%88.7 1.3% No treatment 9960 149.918.0 0.7No iron filings addedSoil 5 11,490 Magnetic treatment 9252.510.43.9 0.7737.3 90.1c7.1 4.5%78.0 4.9% a 5% (0.5 g) iron filings were added to the cont aminated soils and the contact time was 3 hours; b Detected by ICP-AES; c Iron filings had a Pb background value of 56.8 mg/kg. This concentration was subtracted fr om the Pb concentration in filings after treatment; d Mean of 3 replicates one standard deviation.

PAGE 111

111 CHAPTER 6 HEAVY METAL REMOVAL FROM SEDIME NTS USING MAGNETIC SEPARATION 6.1 Introduction Sediments are considered a mixture of assort ed materials that settle to the bottom of a water body (US EPA, 1993). During the sediment ation process, water will be trapped and entrained in the sediment, forming the pore wate r (Batley and Giles, 1979 ). Once toxicants enter into the aquatic system, they tend to be sorbed onto sediments which act as sinks that accumulate higher levels of chemicals than th e overlying water column (Mendil and Uluzl, 2007; Nowierski et al., 2006; Wang and Chen, 2000). Within the sediment matrix, chemical species tend to reach equilibrium between th e pore water and solid phase; however, the concentrations of chemical spec ies in the pore water, which are useful in determining sediment quality and contamination, are not necessarily the same as those in the overlying water (Bufflap and Allen, 1995). Recently, issues related to heavy metal cont aminated-sediments have been attracting increasing attention of regulatory agencies a nd researchers. Metals tend to accumulate in sediments and a variety of sediment constituents including clay minerals iron oxides, manganese oxides, and organic matter are considered metal adsorbents (Jenne, 1995; Wang and Chen, 2000). The accumulated heavy metals in sedime nts may be remobilized by natural and manmade processes (Lin et al., 2003), and become available for biological uptake and then contaminate the food chain. Metals which are no t directly available for biological uptake may become available later through changes in physic ochemical conditions or erosion of sediment deposits (Tarras-Wahlberg and Lane, 2003). Th e cycling of trace metals in sediments is governed by precipitation and di ssolution of minerals (Ouddane et al., 2004), whereas the mobility and toxicity of metals are generally affected by metal speciation, sediment composition,

PAGE 112

112 occurrence of complexing agents and fundamental physicochemical conditions, such as pH and Eh (Lin et al., 2003; Luther, 1995; Ouddane et al ., 2004; Pardue et al., 1995; Tack et al., 1996). Like heavy metal fractionations in soil, sequential ex traction methods can also be applied to sediments to divide the heavy metals into five fractions (exchangeable carbonate-bound, Fe/Mn oxide-bound, organic matter/sulfid e-bound, and residual) (Lin et al., 2003; Pardo et al., 1990; Salomons and Frstner, 1980; Tessier et al., 1979). Metals in exch angeable, carbonate-bound, and Fe/Mn oxide-bound forms are considered to be more mobile and bioava ilable than organicbound and residual forms. Therefore, the determin ation of the amount of metal in each binding form is more useful than the measurement of total metal content in predicting the potential environmental effects caused by contamin ated sediments (Lin et al., 2003). Although numerous remediation methods have b een tested and successfully applied to heavy metal contaminated-soils, much less is kno wn about sediment treatment (Mulligan et al., 2001b). Particle size and organic ma tter content in sediments signifi cantly affect th e selection of treatment strategy. Fine textured sediments ha ve a much higher affinity for all types of contaminants, which are more difficult to reme diate (US EPA, 1993). The remediation of metalcontaminated sediments can be accomplished either in situ or ex situ by physical or chemical treatments (Mulligan et al., 2001b; US EPA, 19 93). Pretreatment of dredged or excavated sediments such as dewatering is usually re quired to modify the physical and chemical characteristics of the sediments (US EPA, 2005). Currently, contaminated dredged sediments are mainly buried in landfills; however this cannot be considered as a long-term solution (Lser et al., 2005). Other main treatments for metal-contaminated sediments include solidification/stabilization, washing, and bioremediation. The purpose of solidification/stabilization is to reduce the m obility of contaminants by treating them with

PAGE 113

113 reagents (lime, fly ash, cement, etc.); howev er, long-term monitoring is required since this process can be reversible (Mullig an et al., 2001b; US EPA, 1993). Sediment washing is primarily useful for sands and gravels, the washing solution s can be solely water or water in combination of organic solvents, chelating compounds, surfactan ts, acids or bases (Mulligan et al., 2001b; US EPA, 1993). Biological processes have recently been under development and gained increasing attention since they are environmentally fr iendly and economical (Chen and Lin, 2001). However, techniques for metal remediation ar e not as developed as those for organic contaminants (Mulligan et al., 2001b). Some stud ies have investigated the effectiveness and kinetics of bioleaching to decontaminate sedi ments (Chen and Lin, 2001; Lser et al., 2005, 2006, 2007). Metal removal by the bioleaching process is based on dissolving toxic metals in sulfuric acid which is produced by microorga nisms through oxidation or reduction of sulfur compounds (Chen and Lin, 2001). The objective of this research was to evaluate the effectiveness of removal of heavy metals (Cu2+, Zn2+, and Hg2+) from four sediments using the proposed magnetic separation method which had been shown to be effective for trea ting metal-contaminated soils. The sediment extracts were tested by MetPLATETM assay and the 48-h Ceriodaphnia dubia acute toxicity test to assess the reduction of heavy metal toxicity in sediments. In addition, chemical analysis and mass balance studies were also performed to inve stigate heavy metal distribution in the sediment matrix and extracts. 6.2 Material and Methods 6.2.1 Sediments Used Two types of sediments were collected at 4 diffe rent water bodies in Gainesville, Florida. The sandy sediments were sampled from the Little Hatchet Creek and Hogtown Creek, respectively. An organic rich sediment wa s collected from a pond near Home Depot in

PAGE 114

114 Southwest Gainesville. The second organic rich sediment was obtained from Lake Alice on the University of Florida campus. All sediments were collected from the top few inches. Table 6-1 shows the main characteristics of the sediment s under study. All sediments samples were first air-dried and stored in the refrigerator until use. 6.2.2 Chemicals Used Three heavy metal solutions (Cu2+, Zn2+, and Hg2+) were used. Copper solution was prepared from copper sulfate (CuSO4H2O, Sigma). Zinc solution was prepared from zinc chloride (ZnCl2, Sigma). Mercury solution was prepared from mercury chloride (HgCl2) purchased from Mallinckrodt. Iron filings (Fisher, 40 mesh) were placed in 1M NaOH for 72 hours to increase the adsorption cap acity and then washed thoroughly with distilled water before use (Yeager, 1998). 6.2.3 Sediment Heavy Metal Binding Capacity The purpose of determining the heavy metal bi nding capacity (HMBC) of sediments is to get an idea of how much metal would be necessary for spiking sediments to produce metal toxicity. The spiked sediments would subseque ntly be used to assess the effectiveness of magnetic treatment. Briefly, the methodology for testing sediment heavy metal binding capacity is similar to that used for assessing soil heavy metal bi nding capacity (SHMBC) (see Chapter 2). The sediments were first screened (sieve # 16; 1.19 mm particles) and homogenized. Then, serial dilutions of metal-spiked solutions were prepared in moderately hard water 60 mg/L Ca, 60 mg/L Mg, pH = 7.4-7.8) for sediment spiking. Three metal solutions containing Cu2+, Zn2+, or Hg2+ were prepared. Twenty milliliter of each solution were added to 5 g (dry weight) of sediment in 50-mL Erlenmeyer flasks. The flasks were covered with parafilm and placed on a shaker at 300 rpm for 4 hours. After shaking, the solid phase was separa ted from the pore water

PAGE 115

115 by centrifugation at 10,000 rpm for 15 minutes. The metal toxicity of the sediment extracts was assayed with MetPLATETM, a microbial test which responds sp ecifically to heavy metal toxicity (See Appendix A for detailed MetPLATETM procedure). Sediments spiked with moderately hard water served as negative cont rols. Regression analysis was then used to determine the EC50s for both the sediments under study and the reference sedi ment. Two sets of reference material were used including Ottawa sand and one of th e sediments which showed the lowest EC50 (i.e. bound the smallest amount of metals). The sedi ment HMBC was determined by dividing the EC50 for the sediment sample by the EC50 for the metal in the reference sediment as defined in Equation 6-1. EC50 of field sediment spiked with a given metal Sediment HMBC = (6-1) EC50 of reference sediment spiked with the same metal A detailed description of the HMBC proce dure was discussed in Chapter 2 and the corresponding methodology was summarized in Figure 2-1. All HMBC tests and MetPLATETM toxicity tests were run in triplicate. 6.2.4 Magnetic Separation of Heavy Metals from Metal-Spiked Sediments After the determination of the sediments heavy metal binding capacity, a sandy sediment and an organic rich sediment were selected to assess the effectiveness of removing Cu2+, Zn2+, and Hg2+ from spiked sediments by magnetic separation. Fifty gram (dry weight) of sedime nt was spiked with 40 mL of Cu2+, Zn2+ or Hg2+ solutions. Based on the HMBC values, two concen trations (125 mg/L a nd 250 mg/L) of heavy metal solutions were spiked into the Hogt own creek sandy sediment, resulting in 100 mgmetal/kg sediment and 200 mg-metal/kg sediment. Fo r the organic rich sediment collected from the Home Depot pond, much higher concentrati ons, 500 mg/L and 1000 mg/L, were used, which generated metal concentrations of 400 mg/kg se diment, and 800 mg/kg sediment, respectively. The spiked sediment was shaken for 1 hour at room temperature, then 5% (2.5 g iron /50 g

PAGE 116

116 sediment) of iron filings were added and the mixture was shaken for 3 hours at room temperature. Then, the iron filings were magne tically retrieved using a ferrimag rectangular magnet (Scientifics, 152 mm, 3.4 megagauss oersteds), and 60 mL of distilled water was added into the system to bring total solu tion volume to 100 mL. The sediment slurry was centrifuged at 10,000 rpm for 15 minutes. The supe rnatant (sediment extract) was removed with a pipet and was used after overni ght settling in the refrigerator. After centrifugation the retrieved iron filings and the sediment matrix were dr ied at 70C overnight. Toxicity testing was undertaken for all sediment extracts before and after magnetic treatment. Sediment without iron treatment served as the control. Chemical anal ysis by ICP-AES was perf ormed for all factions, including sediment extracts, retr ieved iron filings, and sediment matrix. Each sediment sample was run in triplicate. 6.2.5 Toxicity of Sediment Extracts Two toxicity assays, MetPLATETM and the 48-h Ceriodaphnia dubia acute bioassay were used to assess the toxicity of the sediment extracts before and after iron treatment. The EC50 of the sediment extract was determined by regression analysis, and was then converted to toxicity unit (TU = 100/EC50). Higher TU values indicate higher toxi city. All toxicity tests were run in triplicate and the detailed test proced ures are described in Appendix A. 6.2.6 Chemical Analysis Chemical analysis was performed for all fracti ons, including the sediment matrix, sediment extracts and iron filings, before and after magnetic treatment. For Cu and Zn spiked sediment fractions, the sediment extracts were dige sted according to the U.S EPA method 3010A (US EPA, 1992), and sediments and iron filings were digested according to the U.S. EPA method 3050B (US EPA, 1996). Then, all digested samples were analyzed for total Cu and Zn using inductively coupled plasma-atomic emission spect roscopy (ICP-AES). In the case of Hg-spiked

PAGE 117

117 sediment, different digestion and analysis methods were employed. The U.S. EPA method 1631(US EPA, 2002b) was utilized fo r digesting sediment extracts, whereas the sediment matrix and iron filings were digested wi th a 7:3 (v: v) mixture of HNO3/H2SO4 in 50 mL Teflon containers in a hot block overn ight (adapted from Warner et al., 2003). All digested samples were then analyzed by SnCl2 (tin chloride) reduction technique (US EPA, 2002b) with a cold vapor atomic fluorescent spectroscopy (CV-AFS). Mass balance studies were also undertaken for the Cu2+, Zn2+, and Hg2+ spiked sediments. All samples were run in triplicate. Detailed digestion procedures are included in Appendix B. 6.3 Results and Discussion 6.3.1 Sediment Heavy Metal Binding Capacity The methodology for assessing sediment heavy metal binding capacity (HMBC) is based on toxicity testing of extracts from metal-spiked sediments with MetPLATETM, a test specific for heavy metal toxicity (Bitton et al., 1994). Three me tals (Cu, Zn, and Hg) were tested for their binding capacity to four sediments sampled in Ga inesville, Florida. Table 6-2 shows the EC50s (expressed as added metal in mg/kg sediment) of the three metals in the different sediments under study. As discussed in Chapter 2, the Ot tawa sand displayed a very low binding capacity for metals, with EC50s of 1.1 mg/kg for Cu, 0.9 mg/kg for Zn, and 1.5 mg/kg for Hg, which were lower than those shown by the sediments under study. Therefore, the Ottawa sand was still utilized as one of the reference sediments. Among the four sedime nts studied, the sandy sediment sampled from Little Hatchet Creek displayed the lowest EC50s (highest toxicity to MetPLATETM) for the three metals tested, and its EC50s, expressed as metal added to sediments in mg/kg, were 4.3 for Cu, 14.3 for Zn, and 18.4 for Hg (Table 6-2). Thus, the sediment from Little Hatchet Creek was employed as the second reference due to its lowest binding capacity for metals among the sediments under study. Anothe r sandy sediment from Hogtown Creek showed

PAGE 118

118 a little higher EC50s (25.8 for Cu, 35.6 for Zn, and 31.2 for Hg) than Little Hatchet sediment, whereas, the EC50s of the two organic rich sediments fo r the three metals, varied between 92.3 and 111.2 for Cu, between 122.6 and 139.7 for Zn, and between 162.7 and 644.1 for Hg, were much higher than those of the sandy sediments (Table 6-2). The sediment heavy metal binding capacity for the three metals and four sediments tested is shown in Figure 6-1, using Ottawa sand as th e reference. Figure 6-2 displays the sediments HMBC, using Little Hatchet sediment as the re ference. In both figures, the organic rich sediments showed higher binding capacity toward s the three metals than the sandy sediments due to their higher organic matter content (5.8%-10. 5%). This finding consisted with the results shown in Chapter 2, that the organic rich soils had higher SHMBC than the sandy soils. It is well known that organic matter content is one of the key factors controlling metal mobility and toxicity in soils and sediment s (Adriano, 2001a, 2001b; Jenne, 1995; Ouddane et al., 2004). Through studying sediment cores collected from 20 lakes in the Muskoka region of Ontario, Canada, El Bilali et al. (2002) found that organic ma tter played a key role in the enrichment of Cu, Zn, Hg, Pb, and Cd in surf ace sediments. The HMBC concept was previously used to assay metal bioavailabi lity in surface waters (Huang et al., 1999) and municipal landfill leachates (Ward et al., 2005). Chapter 2 showed the first application to da te of the bioassay to assess the metal bioavailability in soils. Similarly, the sediment HMBC method could be used to predict the metal bioavailability to aquatic orga nisms. Frana et al. (2005) investigated the enrichment of heavy metals in benthic invertebra tes and fish in three salt marsh areas in the Tagus estuary, Portugal. Their results revealed that although Hortas salt marsh (contained more sand and lower organic matter) cont ained lower Cu concentration than the other marshes, the enrichment of Cu in the invertebrates and fish in this area was the highest. Therefore, the total

PAGE 119

119 metal concentrations in sediments may not give an accurate indication of metal bioavailability, since the sediment characteristics play an important role in determining metal solubility and toxicity. Our findings in this se ction, confirm that, at least fo r Cu, Zn and Hg, the Sediment HMBC for organic rich sediments is much higher than for sandy sediments. This research should be expanded to include more sediments and metals. 6.3.2 Use of MetPLATETM to Evaluate the Effectiveness of Magnetic Separation of Cu, Zn and Hg from Spiked Sediments. A sandy sediment and an organic rich sedi ment, collected respectively from Hogtown Creek and Home Depot Pond, were sp iked with individual metal (Cu2+, Zn2+, or Hg2+) solutions based on their HMBC values. For the three metals under study, the toxicity of the sediment extracts before and after magnetic treat ment was determined by measuring the EC50s. The EC50 was then converted to toxicity unit (TU = 100/EC50). It is worth mentioning that higher TU values indicate higher toxicity. The percent toxicity removal was calculated according to Equation 3-2. Table 6-3 shows that the toxi city removal efficiency from the spiked sandy sediment varied with the type of metal. Comparing the three heavy metals used, at equal concentrations, Zn resulted in the highest toxic ity (or highest TU values) in the sediment extracts than Hg and Cu before treatment. Cu toxicity in the sandy se diment extracts before magnetic treatment varied from 10.4 TUs when the sample was spiked with 100 mg Cu2+/kg sediment to 158.6 TUs when the sample was spiked with 200 mg Cu2+/kg. As regards Hg2+, toxicity was 25.1 TUs and 250.7 TUs at spiking levels of 100 mg and 200 mg Hg2+/kg sediment, respectively. Zn toxicity was 63.7 TUs and 514.8 TUs at spiking le vels of at 100 mg and 200 mg Zn2+/kg sediment, respectively (Table 6-3). After a single magnetic treatment, the TU values of the sediment extracts were all less than 1 for all three heavy metals, which indicated significant reductions in

PAGE 120

120 toxicity. The percent toxicity removal from Cu-, Zn-, and Hg-spiked sandy sediment was generally higher than 90.4% (Table 6-3). With regard to metal-spiked organic rich sedi ment, as shown in Table 6-4, the toxicity (or TU values) generated by the three metals in se diment extracts before treatment followed this order: Zn> Cu > Hg. For Cu-spi ked organic rich sediment, before treatment, the TU values of the sediment extracts were 12.3 and 70.1 at spiking levels of 400 mg and 800 mg Cu2+/kg sediment, respectively. After a single magnetic tr eatment of Cu-spiked sediment, the TU values of the sediment extracts d ecreased to 1.7 and 1.9 at Cu2+ input concentrations of 400 mg and 800 mg/kg, respectively. This represen ts a toxicity reduction of 86.4% to 97.2%. In the case of Znspiked organic rich sediment, after a single magnetic treatment, the TUs of the sediment extracts decreased from 53.5 to 3.2 at 400 mg Zn2+/kg sediment, and from 242.0 to 7.3 at 80 mg Zn2+/kg sediment, generating a toxicity removal of 94. 0% and 97.0%, respectively (Table 6-4). As regards Hg-spiked organic rich sediment, no t oxicity (TU<1) was observed in the sediment extract after a single magnetic trea tment at Hg spiking level of 400 mg/kg. At a spiking level of 800 mg Hg2+/kg sediment, toxicity decreased from 49.9 TUs to 1.3 TUs, leading to a toxicity reduction of 97.3% (Table 6-4). Considering the e ffect of sediment characteristics, this magnetic treatment worked effectively in both sandy sedi ment and organic rich sediment. Moreover, the metal type did not exert significant effect on the metal removal efficiency. 6.3.3 Use of Ceriodaphnia dubia Acute Toxicity Test to Evaluate the Effectiveness of Magnetic Separation of Cu, Zn and Hg from Spiked Sediments. Metal toxicity removal from sediments follo wing magnetic treatment was also evaluated with the 48-hr Ceriodaphnia dubia acute toxicity test. The results are shown in Tables 6-5 and 66. Comparing with MetPLATETM assay, the 48-h C. dubia test showed higher TU values due to the higher sensitivity of the daphnid test, which had already been shown in Chapters3, 4, and 5.

PAGE 121

121 However, as regards toxicity removal of Cu Zn, and Hg from sediments following magnetic treatment, the two tests showed a similar trend. For Cu-, Zn-, and Hg-spiked sandy sediment, the toxicity removal from the sediments was generally higher than 90.4% when using MetPLATETM, and higher than 99.0% when using the C. dubia test. The sandy sediment spiked with 100 mg/kg Cu2+, 200 mg/kg Cu2+, or 100 mg/kg Zn2+ all became non-toxic after a single magnetic treatment. For Cu-, Zn-, and Hg-spiked organi c rich sediment, after a single treatment, the toxicity removal from the metal-spiked se diments was higher than 83.0% using MetPLATETM, and ranged from 80.5% to 99.8% when using the C. dubia test. 6.3.4 Assessment of Metal Removal Effi ciency Using Chemical Analysis Besides toxicity tests, we also employed chemi cal analysis to assess the removal efficiency of Cu, Zn, and Hg from metal-spiked sediments, as well as the distributions of these three metals in the sediment matrix and extracts. All fract ions, including the sediment matrix, sediment extracts and iron filings, before and after magnetic treatment from the sediments spiked with the highest Cu, Zn and Hg concentrations (200 mg -metal/kg for sandy sediment and 800 mg-metal /kg for organic rich sediment) were digested an d chemically analyzed using inductively coupled plasma-atomic emission spectrosc opy (ICP-AES) (for Cu and Zn) and cold vapor atomic fluorescent spectroscopy (CV-AFS) (for Hg). For sandy sediments, as shown in Table 6-7, the metal removal from the sediment matrix was the highest for Hg (79.3%), followed by Cu (38.2%) and Zn (25.9%). However, the removal of Cu, Zn and Hg from sediment extracts was mu ch higher, and varied from 95.8% to 99.5%. In addition, very high concentra tions of Cu (1700.2 mg/kg), Zn (1886.5 mg/kg), and Hg (1923.4 mg/kg) were found in the retrieved iron filings, wh ich demonstrated that the metals were indeed adsorbed and concentrated on the iron filings.

PAGE 122

122 Table 6-8 shows the percent removal of Cu, Zn and Hg re from organic rich sediment factions. Among the three metals studied, Hg s howed the highest removal from both sediment matrix (58.5%) and extracts (93.3%). The remova l of Cu and Zn from the sediment matrix were 18.5% and 27.9%, respectively. Cu removal from the sediment extracts was 69.3%, as compared to 35% removal for Zn. The concentrations of Cu, Zn and Hg adsorbed onto the iron filings followed this order: Hg (6,014.2mg/kg) > Cu (2,441.5mg/kg) > Zn (1,789.3mg/kg). Oxides and oxyhydroxides of iron play a signific ant role in sequestering elements due to their large surface area and strong affinity for many elements (Parida et al., 1997). The sorptive properties of iron (hydr)oxides for various metals and metalloid including Cu, Cd, Zn, Pb, Co, Ni, and As have been extensively investigated (Bryce et al., 1994; Es madi and Simm, 1995; Lee and Anderson, 2005; Nayak et al., 2006; Swallo w et al., 1980; Wilkie and Hering, 1996; Yamaguchi and Okazaki, 2002). A study conducte d by Namasivayam and Senthilkumar (1999) used Fe (III)/Cr (III) hydroxide as an absorbent to remove Cu2+ from aqueous solutions. Besides, Namasivayam and Ranganathan (1995) ha d investigated the adsorption of Pb2+, Cd2+, and Ni2+ on Fe (III)/Cr (III) hydroxide. These studies showed that the adsorption of metal ions increased when the adsorbent concentration increased and the particle size decreased, and the adsorption obeyed the Freundlich isotherm model. Howeve r, much fewer studies on mercury adsorption onto iron (hydr)oxides have been carried out. Kim et al. (2004a) employed extended X-ray absorption fine structure (E XAFS) spectroscopy to study Hg2+ sorption to iron (hydr)oxides and their results indicated that Hg2+ formed inner-sphere sorption complexes to goethite (-FeOOH) over a pH range of 4 to 8. Another study carried out by Kim et al. (2004b) re vealed that chloride ligands resulted in reduced Hg2+ sorption to goethite, whereas sulfate could enhance the Hg2+ binding to goethite.

PAGE 123

123 In all, the chemical analysis suggested that Cu, Zn, and Hg were removed from both the sediment extracts and the sediment matrix. However, for each heavy metal, in a given sediment, metal removal from the sediment extracts was always higher than that from the sediment matrix, which agreed with the results in chapte r 3 on metal removal from soil fractions. 6.3.5 Mass Balance of Metals in Sediments Tables 6-9 and 6-10 display the mass balan ce of Cu, Zn, and Hg in the spiked sandy sediment and organic rich sediment before a nd after magnetic treatment. As shown in these tables, a large portion of the metals was immobilized in the sediment matrix before treatment, but the metals associated with the sediment matrix and in the sediment extracts were both reduced following magnetic treatment, these findings are consistent with those derived from the mass balance study of metals in soils (see in chap ter 3). The total recovery of metals was not significantly affected by the sediment type. In the sandy sediment, 87.4% of Cu was accounted for before treatment whereas 81.5% of Cu was acc ounted for after treatment. In the organic rich sediment, the mass balance accounted for 94.1% Cu before treatment and 91.2% after treatment. As regards Zn, the mass balance accounted for 77. 4% and 83.5% Zn before magnetic treatment of the sandy and organic rich sediment, respecti vely. Following magnetic treatment, 81.2% and 70.8% of Zn was accounted for in the sandy and orga nic rich sediment, resp ectively. The total recovery of added Hg from sediments was somewh at lower than that of Cu and Zn. Before treatment, the recovery of Hg was 76.7% and 67. 6% in the sandy sediment and the organic rich sediment, respectively. After magnetic treatmen t, only 60.7% and 65.1% of Hg was recovered from the sandy sediment and organic rich sedime nt, respectively. The lower recovery of Hg probably came from the digestion pro cess of Hg-containing iron filings. Comparing with ex situ treatment of contaminated sediments, fewer techniques can be used for in situ treatment due to a variety of limitations such as saturated conditions, anaerobic

PAGE 124

124 environments and ambient temperatures; however, in situ treatment is generally considered to be less expensive than ex situ treatment or disposal of cont aminated sediments (US EPA, 1998b). Table1-6 (see Chapter 1) lists the commonly used in situ and ex situ techniques for sediment remediation by the U.S. EPA. In 1995, Environm ent Canada carried out a pilot-size (100 m m) demonstration project of capping metal and me talloid (Zn, Cu, Pb, Cr Ni, Cd, As, and Hg) contaminated sediments in Hamilton Harbor, Lake Ontario. A layer of clean medium to coarse sand with the average thickness of 35 cm was pl aced at the site (Azcue et al., 1998). One year later, the mobility of trace elements through the cap material and the physi cal stability of the cap were assessed by Azcue et al. ( 1998). Significant reductions in the vertical fluxes (up to 80%) of all trace elements were observed after capping of the contaminated sediments; however, they also reported that a thin layer (1 to 3 cm thick) of fresh mode rately contaminated sediments had started to develop on the top of the capping layer. Therefore, long-term m onitoring of the cap is essential to ensure that its in tegrity has not been compromise d by water body and other effects (US EPA, 1993). Ex situ treatment of contaminated sediments usua lly involves dredging a nd/or pretreatment of sediments to remove debris and dewater the dredged sediments (M ulligan et al., 2001b). Dredging of sediments causes resuspension of se diment; however, the spread of resuspended sediment can be limited through the use of silt cu rtains (US EPA, 1993). US EPA (1993) divided the dredging methods into three categories including mechanical, hydraulic, and pneumatic dredge. The amount of dewatering depends on the t ype of dredging used and the technique to be used for treatment. Centrifugati on, filtration or gravity thicke ning can be used for dewatering purposes (Mulligan et al., 2001b). Extraction of heavy metals from dredged sediments requires mobilization of metals since metals are often strongly re tained in sediment under natural

PAGE 125

125 conditions (Nystroem et al., 2006). Calmano et al. ( 1993) indicated that acidic and chelating agents were most effective in removing heavy metals from sediments. Extraction studies in laboratory scale have showed promising results. Yu et al. (1996) evaluated the remobilization abilities of EDTA (ethylenedia mine tetraacetic acid), DTPA (diethylenetriamine pentaacetic acid), and EGTA (ethylene glycol tetraacetic acid) on zinc from Ell-Ren river sediment in Taiwan. Their results showed that DPTA had hi gher remobilization effect on Zn than EDTA and EGTA. McCready et al. (2003) compared the e ffectiveness of 1 M HCl and 0.05 M EDTA on extracting Zn, Pb, Cu, and Cd from sixty sedime nts with different textures collected from Sydney Harbor. They reported that 1M HCl extr acted a large percentage (60-100%) of heavy metals in oxic sediments; whereas the extractabil ity of metals with 0.05 M EDTA was generally lower by 20%. Sediment washing is a large scal e application of extract ion method; however, it cannot efficiently treat sediment s with fine particles and high humic content or low permeability (US EPA, 1993). Recently, electrodialytic remediation has also showed good results for remediation of heavy metal contaminated harbor sediments (N ystroem et al., 2005, 2006; Ottosen et al., 2007). Nystroem et al. (2006) tested the use of different desorbing agents (HCl, NaCl, citric acid, lactic acid, ammonium citrate and distilled water) in el ectrodialytic remediatio n of harbor sediment. Their results revealed that the use of desorbing agents did not generally enhance the heavy metal removal. The removal was 48% Cu, 80% Zn, 96% Pb, and 98% Cd when using distilled water. In another study, Ottosen et al (2007) successfully removed Cu and Cd from a sediment sampled from Sisimiut Harbor, Greenla nd by electrodialytic remediati on. The Cu concentration was reduced from 97 to 16 mg/kg and the Cd concen tration was reduced from 0.55 to 0.03 mg/kg

PAGE 126

126 after 28 days with an applied current density of 1.2 mA/cm2, and the major removal of the two heavy metals was obtained during the first 3 days. Ex situ stabilization/solidification is another tech nique suitable for sediments contaminated with inorganic and metals, however, fine particle sizes and the presence of soluble salts of manganese, tin, copper and lead can reduce th e treatment effectiveness (US EPA, 1993). Besides, since immobilization leads to an increas e in volume, larger areas of land are required for disposal. Cementor silicate-based processe s are useful for sediment s and economical if the end product can be used for landfill closure or in other applications (Mulligan et al., 2001b). Mller and Pluquet (1998) i nvestigated the effectiven ess of five iron-bearing materials (red mud, sludge from drinking water treatme nt, bog iron ore, unused steel sh ot and steel shot waste) on immobilization of Cd and Zn in a contaminated sediment dredged from the port of Bremen (Germany). After treatment, the uptake of Cd and Zn by plants was reduced by 20-50%, and the NH4NO3 and DTPA (diethylenetriamine pentaacetic acid) extractable amounts of Cd and Zn were reduced by 50% and 20%, respectively. It was also demonstrated that red mud and Febearing sludge were the most effective material s. Meng et al. (1991) treated a Cd-contaminated lake sediment with aluminum nitrate at pH 9.5. The NH4OAc-extractable Cd in the sediment treated with 1.6 mmol aluminum per gram of se diment was 80% less than the extractable Cd in the untreated sediment. They also found that free ze-thaw treatment of th e Al-treated sediment significantly reduced the volume of the settled solids and did not affect the tendency for the treatment to enhance Cd retention. 6.4 Conclusions The sediment heavy metal binding capacity (H MBC) test showed that sediment HMBC varied with the type of sediment, with organi c rich sediments displa ying a much higher metal binding than sandy sediments. In addition, after demonstrating the use of magnetic separation in

PAGE 127

127 treating heavy metal-contaminated soils in Chapter 3, we also examined the effectiveness of this treatment approach in removing Cu, Zn, and Hg fro m spiked sediments. The results of this study showed a significant reduction of toxicity generate d by Cu, Zn or Hg in sediment extracts after a single magnetic treatment. As regards toxicity reduction in sediment extracts, the type of sediment and metal did not affect the treatmen t effectiveness. Moreover, chemical analysis suggested that all the th ree metals were removed from both the sediment matrix and the sediment extracts. However, metal removal from the sediment matrix was lower than that from the sediment extracts. In conclusion, the results indicated that, in addition to decontaminating metalcontaminated soils, the magnetic treatment could also be used to treat metal contaminated sediments. Comparing with other sediment remedi ation techniques, this me thod is relatively easy to operate, and time-saving. Future studies on decontaminating natural/aged contaminated sediments would be necessary to assess the f easibility of applying this method under field conditions.

PAGE 128

128 Table 6-1. Sediments characteristics Characteristic Little Hatchet Creek Sediment Hogtown Creek Sediment Home Depot Pond Sediment Lake Alice Sediment pH 6.05.75.8 5.9 Eh (mV) 313.4288.6381.1 347.8 % Organic carbon 0.00.46.4 6.1 % Organic matter 0.00.75.8 10.5 % Sand 62.099.487.0 95.7 % Silt 37.80.56.0 0.3 % Clay CEC (cmolc/kg) 0.2 3.6 0.1 8.3 7.0 192.9 4.0 218.5 Table 6-2. EC50s (added metal in mg/kg sediment), as determined by MetPLATETM, of water extracts from four sediments and Ottawa sand Sediment Type Heavy Metal EC50 (mg/kg sediment) Ottawa Sand Little Hatchet Creek sediment Hogtown Creek Sediment Lake Alice Sediment Home Depot Pond Sediment Cu Zn Hg Cu Zn Hg Cu Zn Hg Cu Zn Hg Cu Zn Hg 1.1 0.2* 0.9 0.1 1.5 0.3 4.3 0.9 14.3 2.5 18.4 2.3 25.8 5.3 35.6 1.5 31.2 2.0 111.2 14.2 139.7 15.0 644.1 22.8 92.3 8.6 122.6 2.0 162.7 16.6 *mean of 3 replicates one standard deviation

PAGE 129

129 3.9 23.5 83.9 101.1 15.9 39.6 136.2 12.3 20.8 108.5 155.2 429.4 0 100 200 300 400 500 Little Hatchet CreekHogtown CreekHome Depot PondLake AliceSediment HMB C Cu Zn Hg Figure 6-1. Sediment HMBC for three metals (C u, Zn, Hg) and four sediments (Ottawa sand served as the reference). 6 2.5 8.6 9.8 1.7 8.8 25.9 21.5 35 0 5 10 15 20 25 30 35 40 Hogtown CreekHome Depot PondLake AliceSediment HMB C Cu Zn Hg Figure 6-2. Sediment HMBC for three metals (Cu, Zn, Hg) and three sediments (Little Hatchet Creek sediment served as the reference).

PAGE 130

130Table 6-3. Copper, zinc and mercury toxicity removal from a spiked sandy sediment by magnetic treatment, as determined by MetPLATETM. EC50 of sediment extract (% sedim ent extract) Toxicity unitsb of sediment extract Heavy metal type Heavy metal concentration in spiked sediment (mg/kg) No treatment Magnetic treatmenta No treatment Magnetic treatment Toxicity removal (%) 100 9.6 1.1%f>100% 10.4 1.2 <1g>90.4% Cuc 200 0.6 0.002%>100% 158.6 0.4 <1 >99.4% Znd 100 200 1.6 0.02% 0.2 0.02% >100% >100% 63.7 0.6 514.8 40.3 <1 <1 >98.4% >99.0% 4.0 0.08%>100% 25.1 0.5 <1 >96.0% Hge 100 200 0.4 0.06%>100% 250.7 36.3 <1 >99.6% a 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; b Toxicity units = 100/EC50; c 50 g of sediment was spiked with 40 mL of Cu solution containing 125 mg/L and 250 mg/L Cu2+, respectively; d 50 g of sediment was spiked with 40 mL of Zn solution containing 125 mg/L and 250 mg/L Zn2+, respectively; e 50 g of sediment was spiked with 40 mL of Hg solution containing 125 mg/L and 250 mg/L Hg2+, respectively; f Mean of 3 replicates 1 standard deviation; g Non-toxic.

PAGE 131

131Table 6-4. Copper, zinc and mercury toxicity removal from a spiked or ganic rich sediment by magnetic treatment, as determined by MetPLATETM EC50 of sediment extracts (% sed iment extract) Toxicity unitsb of sediment extracts Heavy metal type Heavy metal concentration in spiked sediment (mg/kg) No Treatment Magnetic treatmenta No treatment Magnetic treatment Toxicity removal (%) 400 8.2 0.8%f60.1 2.7% 12.3 1.1 1.7 0.1 86.4 0.6% Cuc 800 1.4 0.11%51.4 0.2% 70.1 5.4 1.9 0.01 97.2 0.2% Znd 400 800 1.9 0.04% 0.4 0.01% 30.91.1% 13.9 2.6 53.5 1.2 242.0 6.6 3.2 0.1 7.3 1.4 94.0 0.1% 97.0 0.5% 17.0 0.2% >100% 5.9 0.1 <1g >83.0% Hge 400 800 2.0 0.3%78.5 5.7% 49.9 7.2 1.3 0.1 97.3 0.6% a 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; b Toxicity units = 100/EC50; c 50 g of sediment was spiked with 40 mL of Cu solution containing 500 mg/L and 1000 mg/L Cu2+, respectively; d 50 g of sediment was spiked with 40 mL of Zn solution containing 500 mg/L and 1000 mg/L Zn2+, respectively; e 50 g of sediment was spiked with 40 mL of Hg solution containing 500 mg/L and 1000 mg/L Hg2+, respectively; f Mean of 3 replicates 1 standard deviation; g Non-toxic.

PAGE 132

132Table 6-5. Copper, zinc and merc ury toxicity removal from a spiked sandy sedime nt by magnetic treatment, as determined by the 4 8-h acute Ceriodaphnia dubia toxicity test EC50 of sediment extracts (% sedim ent extract) Toxicity unitsb of sediment extracts Heavy metal type Heavy metal concentration in spiked sediment (mg/kg) No treatment Magnetic treatmenta No treatment Magnetic treatment Toxicity removal (%) 100 0.3 0.02%f>100% 342.2 23.8 <1g>99.7% Cuc 200 0.03 0.00%>100% 3611.0 27.2 <1 >99.98% Znd 100 200 0.61 0.12% 0.11 0.01% >100% 11.7 0.1% 168.5 33.6 887.1 51.6 <1 8.5 0.1 >99.4% 99.0 0.0% 0.27 0.04%76.1 1.6%373.0 51.61.3 0.0299.6 0.1% Hge 100 200 0.03 0.01%65.9 4.0%3532.2723.81.5 0.199.96 0.01% a 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; b Toxicity units = 100/EC50; c 50 g of sediment was spiked with 40 mL of Cu solution containing 125 mg/L and 250 mg/L Cu2+, respectively; d 50 g of sediment was spiked with 40 mL of Zn solution containing 125 mg/L and 250 mg/L Zn2+, respectively; e 50 g of sediment was spiked with 40 mL of Hg solution containing 125 mg/L and 250 mg/L Hg2+, respectively; f Mean of 3 replicates 1 standard deviation; g Non-toxic.

PAGE 133

133Table 6-6. Copper, zinc and mercury toxicity removal from a spiked organic rich sedime nt by magnetic treatment, as determined b y the 48-h acute Ceriodaphnia dubia toxicity test EC50 of sediment extracts (% sedim ent extract) Toxicity unitsb of sediment extracts Heavy metal type Heavy metal concentration in spiked sediment (mg/kg) No treatment Magnetic treatmenta No treatment Magnetic treatment Toxicity removal (%) Cuc 400 2.4 0.1%f 68.8 5.3% 41.4 1.9 1.5 0.1 96.5 0.4% 800 0.16 0.02 64.5 0.8% 633.1 87.2 1.6 0.02 99.8 0.04% Znd 400 1.2 0.07% 6.5 0.7% 85.3 4.9 15.6 1.6 81.7 3.0% 800 0.22 0.03% 1.1 0.03% 462.5 68.9 89.1 2.4 80.5 3.4% Hge 400 1.5 0.2% 34.5 0.8% 69.1 10.0 2.9 0.06 95.8 0.5% 800 0.14 0.01% 13.1 0.2% 714.3 28.3 7.5 0.1 98.9 0.1% a 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; b Toxicity units = 100/EC50; c 50 g of sediment was spiked with 40 mL of Cu solution containing 500 mg/L and 1000 mg/L Cu2+, respectively; d 50 g of sediment was spiked with 40 mL of Zn solution containing 500 mg/L and 1000 mg/L Zn2+, respectively; e 50 g of sediment was spiked with 40 mL of Hg solution containing 500 mg/L and 1000 mg/L Hg2+, respectively; f Mean of 3 replicates 1 standard deviation.

PAGE 134

134Table 6-7. Effect of magnetic treatment on the removal of Cu2+, Zn2+ and Hg2+ from a spiked sandy sediment, as determined by chemical analysis. Metal conc. in sediment matrix, extracts, and iron fili ngs Metal removal from sediment fractions Initial metal conc. in spiked sediment (mg/kg)a In sediment matrix (mg/kg) In sediment extracts (mg/L) Adsorbed on iron filings (mg/kg) From sediment matrix (%) From sediment extracts (%) No treatment 128.5 3.6c24.5 2.9No iron filings added Cu2+ (200mg/kg) Magnetic treatmentb 80.4 3.30.17 0.061700.2 152.6d 38.2 3.1%99.5 0.1% No treatment 98.5 2.933.4 0.4No iron filings added Zn2+ (200mg/kg) Magnetic treatment 75.7 5.41.4 0.31886.5 139.5e 25.9 6.3%95.8 1.0% No treatment 118.5 3.117.5 1.0No iron filings added Hg2+ (200mg/kg) Magnetic treatment 24.5 1.00.4 0.11923.4 22.7f 79.3 1.3%98.0 0.6% a 50 g of sediment was spiked with 40 mL of Cu, Zn or Hg solution containing 250 mg/L metal; b 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Iron filings had a Cu background value of 2503.3 mg/kg. This concentration was subtracted fro m the Cu concentration in filings after treatment; e Iron filings had a Zn background value of 13.1 mg/kg. This concentration was subtracted fro m the Zn concentration in filings after treatment; f Iron filings had an Hg background value of 0.02 mg/kg. This concentration was subtracted fro m the Hg concentration in filings after treatment.

PAGE 135

135Table 6-8. Effect of magnetic treatment on the removal of Cu2+, Zn2+ and Hg2+ from a spiked organic rich sediment, as determined by chemical analysis. Metal conc. in sediment matrix, extracts, and iron filings Metal removal from sediment fractions Initial metal conc. in spiked sediment (mg/kg)a In sediment matrix (mg/kg) In sediment extract (mg/L) Adsorbed on iron filings (mg/kg) From sediment matrix (%) From sediment extract (%) No treatment 750 50.7c5.2 0.2No iron filings added Cu2+ (800mg/kg) Magnetic treatmentb 611.7 29.81.6 0.12441.5 365.9e 18.5 2.0%69.3 3.4% No treatment 627.7 15.534.3 1.3No iron filings added Zn2+ (800mg/kg) Magnetic treatment 460.9 23.222.3 0.51789.3 149.3f 27.9 2.0%35.0 3.8% No treatment 528.1 22.86.4 0.5No iron filings added Hg2+ (800mg/kg) Magnetic treatment 219.0 1.90.4 0.026014.2 769.1g 58.5 1.5%93.3 0.8% a 50 g of sediment was spiked with 40 mL of Cu, Zn or Hg solution containing 1000 mg/L metal; b 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d Iron filings had a Cu background value of 2503.3 mg/kg. This concentration was subtracted fro m the Cu concentration in filings after treatment; e Iron filings had a Zn background value of 13.1 mg/kg. This concentration was subtracted fro m the Zn concentration in filings after treatment; f Iron filings had an Hg background value of 0.02 mg/kg. This concentration was subtracted fro m the Hg concentration in filings after treatment.

PAGE 136

136Table 6-9. Mass balance of Cu, Zn and Hg in a spiked sand y sediment before and after magnetic treatment. Initial metal mass (mg)a Metal in sediment matrix (mg) Metal in sediment extract (mg) Metal adsorbed on iron filings (mg) Total recovered metal mass (mg) Total recovery of metal (%) Before treatment 1.3 0.04c0.5 0.06No iron filings added1.7 0.187.4 4.2% Cu (2 mg) After treatmentb 0.8 0.030.003 0.0010.9 0.08d1.6 0.181.5 5.5% Before treatment 0.9 0.030.7 0.01No iron filings added1.5 0.0377.4 1.8% Zn (2 mg) After treatment 0.7 0.050.03 0.010.9 0.1e 1.6 0.181.2 4.6% Before treatment 1.2 0.030.35 0.02No iron filings added1.6 0.0276.7 2.4% Hg (2 mg) After treatment 0.2 0.010.01 0.0021.0 0.01f1.2 0.0160.7 0.6% a 50 g of sediment was spiked with 40 mL of Cu, Zn or Hg solution containing 250 mg/L metal ; b 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d The amount of background Cu mass in iron filings was subtracted from the Cu mass in filings after treatment; e The amount of background Zn mass in iron filings was subtracted from the Zn mass in filings after treatment; f The amount of background Hg mass in iron filings was subt racted from the Hg mass in filings after treatment.

PAGE 137

137Table 6-10. Mass balance of Cu, Zn and Hg in a spiked organic rich sediment before and after magnetic treatment. Initial metal mass (mg)a Metal in sediment matrix (mg) Metal in sediment extract (mg) Metal adsorbed on iron filings (mg) Total recovered metal mass (mg) Total recovery of metal (%) Before treatment 7.4 0.5c0.1 0.004No iron filings added7.5 0.5d94.1 6.4% Cu (8 mg) After treatmentb 6.0 0.30.030.0031.2 0.2d 7.3 0.491.2 5.3% Before treatment 6.0 0.20.7 0.03No iron filings added6.7 0.283.5 2.2% Zn (8 mg) After treatment 4.3 0.20.4 0.010.9 0.1e5.7 0.270.8 1.9% Before treatment 5.3 0.20.1 0.01No iron filings added5.4 0.267.6 2.9% Hg (8 mg) After treatment 2.2 0.020.01 0.00053.0 0.45.2 0.465.1 4.6% a 50 g of sediment was spiked with 40 mL of Cu, Zn or Hg solution containing 1000 mg/L metal; b 5% (2.5 g) of iron filings were added to the spiked sediment, and the contact time was 3 hours; c Mean of 3 replicates one standard deviation; d The amount of background Cu mass in iron filings was subtracted from the Cu mass in filings after treatment; e The amount of background Zn mass in iron filings was subtracted from the Zn mass in filings after treatment; f The amount of background Hg mass in iron filings was s ubtracted from the Hg mass in filings after treatment.

PAGE 138

138 CHAPTER 7 PLANT GROWTH STUDY TO DEMONSTRATE METAL REMEDIATION BY MAGNETIC SEPARATION 7.1 Introduction The Environment may be polluted by heavy metals as a result of indus trial activities, such as mining, smelting, electroplating, energy pr oduction, military operations, or sewage sludge disposal (Lim et al., 2004; Nedelkoska and Doran, 2000; Quartacci et al., 2005). Land application of sewage sludge is more economical than incineration or disposal into landfills (Blais et al., 2004). However, a large portion of sewage sludge from industrial sources often contains toxic metals, such as lead (Pb), cadmium (Cd), nickel (Ni), chromium (Cr), copper (Cu) and zinc (Zn), that can persist in the top cultivated layer of soils (Wani et al., 2007). If these metals are phytoavailable, they may accumulate in plants and pose potential threat to humans and grazing animals (Boularbah et al., 2006). Also, soil phytotoxicity may cause disappearance of natural vegetation cover which could exert risk for the surrounding areas (Boisson et al., 1999). Some metals like Cd and Zn are very mobile in soils and thus read ily available to plants (Madejn et al., 2004). Metal phytoavailability is mainly determined by the metal speciation (Notten et al., 2005), soil characteristics, such as pH, texture, organic matter and clay content, cation exchange capacity and redox potential (Planquart et al., 199 9), and duration of contact with the surface binding these metals (Naidu et al., 2003a). Other factors including the distribution of metals across the soil profile and rooting depth are also ab le to affect the uptake of metals by plants (Notten et al., 2005). Naidu et al. (2003a) have sh own that at any given total Cd concentration, the phytoavailable metal fraction is higher in Oxisol s than in Vertisols un less the pH of Oxisols is higher than 6 to enhance their binding capacity of Cd. Vegetation can be used to indicate or monitor environment contamination by heavy meta ls (Mertens et al., 2005; Pugh et al., 2002).

PAGE 139

139 For example, mosses have been used as bioi ndicators in many countri es in the Northern Hemisphere to estimate the atmospheric deposit ion of metal particles (Denayer et al., 1999). Madejn et al. (2004) have survey ed the content of eight trace elements (As, Cd, Cu, Fe, Mn, Ni, Pb, Zn) in leaves and stems of white poplar (Populus alba) trees, and found a significant and positive correlation between the concentration of trace elements in poplar leaves and the soil available Cd and Zn, which indicated that poplar leaves could be used as biomonitors for soil pollution of these two metals. To quantitatively predict the tr ansfer of metals from soils to plants, both mechanistic and empirical models have been proposed. Since me chanistic models are based on plant and soil parameters which are difficult to determine, ma ny more studies employed empirical models, and linear functions are pref erred in most cases (Krauss et al ., 2002). However, in reality, plant uptake of metals is a non-linear process. Krauss et al. (2002) evaluated the use a curvilinear Freundlich-type function (CPlant = b CSoil a, where b and a are the empirical Freundlich coefficients) to predict Cd, Cu, Pb, and Zn concentrations in wheat (Triticum aestivum L.) grain and leaf (CPlant) from soil concentrations (CSoil). They showed that this Freundlich-type function was suitable to predict Cd and Zn concentratio ns in wheat grain and leaf from the EDTAextractable metal concentrations, whereas the pred iction of Cu and Pb concentrations was poorer. Recently, studies have been carried out to in vestigate the metal tolerance mechanisms in different plant species (Clemens, 2001; Horiguc hi, 1987; Meharg, 2005; Sasaki et al., 1995). The mechanisms involve exclusion, active removal, bi osorption, precipitation or bioaccumulation in external and intracellular spaces which can influen ce the metal solubility and the bioavailability to plants (Carlot et al., 2002). The bioaccumulation of metals in plants depends on both the plant species and the type of metal (Naidu et al., 2003a). Hyperaccumul ation is a sub-class of metal

PAGE 140

140 resistance of plant species, which is less common than exclusion strategi es (Meharg, 2005). The first hyperaccumulators characterized were me mbers of Brassicaceae and Fabaceae families (Salt et al., 1998). Presently, more than 400 plan t species have been identified as metal hyperaccumulators (Tang et al., 2003). The poten tial use of hyperaccumulators to remediate heavy metal-contaminated soils has attracted researchers interests. For instance, Alyssum bertolonii, A. tenium, and A. troodii are recognized hyperaccu mulators of Ni, and Thlaspi caerulescens is a recognized hyperaccumulator of Cd and Zn (Nedelkoska and Doran, 2000). Sahi et al. (2002) reported that Sesbania drummondii possessed the ability of hyperaccumulating Pb. In addition, a hyperaccumulator of As, Pteris vittata L. (brake fern) was discovered by Ma et al. (2001). These studies on metal hyperaccumulati ng plants have revealed the feasibility of utilizing phytoremediation as a mean to clean up metal-contaminated soils. However, although phytoremediation is being regarded as a promis ing, economical, and environmentally friendly remediation alternative, some limitations have been addressed by researchers. The major problem affects plant remediation efficiency is that some of the metals are immobile in soils and thus their bioavailability is limited to the root surfaces (Wu et al., 2004). Moreover, the majority of the hyperaccumulators have very low bioma ss production (Tang et al., 2003), and the disposal of contaminated crop materi al is another concern (Sas -Nowosielska, et al., 2004). In this chapter, we investigated the ability of MetPLATETM, a bacterial toxicity test, in predicting heavy metal phytoavailability in different t ypes of soils, as well as the use of plants to assess the effectiveness of magnetic separa tion for removing heavy metals from soils.

PAGE 141

1417.2 Material and Methods 7.2.1 Assessment of Cu Phyt otoxicity Using MetPLATETM 7.2.1.1 Soils used Three soil types were used to study the suitability of the MetPLATE toxicity test to predict heavy metal phytoavailability. A sandy soil was sa mpled from the top 4 feet at the McCarty Woods on the University of Florida campus, which is representative of the soils prevailing in North Central Florida. An organic rich soil wa s collected from the first few top inches along Hogtown Creek in Gainesville, FL. A mixed so il contained 50% (w/w) of the red sandy soil (sampled from Perdido landfill in Cantonment, FL) and 50% (w/w) of the organic rich soil. Table 7-1 shows the main characte ristics of the soils unde r study. All soils samples were first airdried, screened (sieve # 10; 2.0 mm particles) and homogenized. 7.2.1.2 Soil spiking with Cu According to the soil Cu binding capacity obta ined in Chapter 2, different ranges of Cu solution (prepared from CuSO4H2O, Sigma) were spiked into the three types of soils. The sandy soil was spiked with 25, 50, and 100 mg/kg of Cu; the organic rich soil was spiked with 50, 500, and 700 mg/kg of Cu; while the mixed so il was spiked with 100, 200, and 500 mg/kg of Cu. The spiked soils were air-dried for 10 days to reach equilibrium and promote the adsorption of the added metal, and then screened (sieve # 10; 2.0 mm particles) and homogenized again prior to toxicity test s and pot experiments. 7.2.1.3 Pot experiment The pot experiment was performed in a greenhouse under controlled conditions (temperature 25 3 C, day-night cycle 16/8 h), and two plant species, lettuce (Lactuca sativa) and Indian mustard (Brassica juncea), were used. Five hundred gram of spiked soil was placed in each plant pot (5-inch diameter). NH4NO3 and KH2PO4 were applied as fertilizers at the rates

PAGE 142

142 of 0.43 and 0.33 g/kg, respectively (Wu et al., 2004) Nine seeds of each plant species were sown in each pot and thinned to four seedlings for lett uce and three seedlings for Indian mustard after 14 days. All pots were watered daily with distil led water by adjusting the water content to 70% of the soil water holding capacity, and fertilized once per week to maintain vigorous plant growth (water holding capacity of each type of so il was determined in advance). For each type of soil spiked with one Cu concentration, three rep licates were set up for each plant species. Nonspiked soils served as controls. After eight weeks, all plants were harv ested and the shoots (a boveground parts) were separated from the roots. The length of the shoot s were measured and recorded. Then, the shoots and roots were washed with distilled water, and oven dried at 70C for 24 h to measure their dry biomass. The dried plant tissue was then gr ound through a 20 mesh sieve (0.85 mm opening), and the Cu concentration in plant roots and shoots were measured by inductively coupled plasma-atomic emission spectrosc opy (ICP-AES) after wet acid di gestion of the plant tissue using concentrated nitric acid and 30% hydrogen peroxide (Mills and Jone s, 1996). The detailed plant digestion procedure was included in Appendix B. 7.2.1.4 Toxicity of soils, as determined by MetPLATETM The toxicity of the spiked soils that were used for growing plants was tested by the MetPLATETM assay. Twenty milliliter of distilled wate r were added to 10 g of spiked soil, and the soil mixture was shaken for 4 hours. Then the soil slurry was centr ifuged at 10,000 rpm for 15 minutes. The supernatant (soil extract) was removed with a pipet. The MetPLATETM microbial test was carried out according to Bitt on et al. (1994), and the detailed procedure is described in Appendix A. To determine the EC50 for the soil extracts, 4 to 5 dilutions of the soil extracts were prepared, and a regression analysis was used to calculate the EC50. Moderately

PAGE 143

143 hard water (60 mg/L Ca, 60 mg/L Mg, pH = 7.4 -7.8) was used as the negative control. All experiments were run in triplicate. 7.2.2 Use of Plants to Evaluate th e Effectiveness of Magnetic Se paration on Cu-Spiked Soils 7.2.2.1 Soils preparation The same sandy soil sampled from the McCa rty Woods on the Univ ersity of Florida campus was spiked with Cu2+ solution (prepared from CuSO4H2O, Sigma) to reach a concentration of 100 mg Cu2+/kg soil. Another organic rich soil (organic soil 2), purchased from a local landscaping store, was spiked to reach a concentr ation of 500 mg Cu2+/kg soil. The spiked soils were air-dried for 10 days, and then screened (sieve # 10; 2.0 mm particles) and homogenized prior to iron treatment followed by magnetic separation. The iron filings (Fisher, 40 mesh) were placed in 1M NaOH for 72 hours to increase the adsorpti on capacity and then washed thoroughly with distilled water before use (Yeager, 1998). 7.2.2.2 Treatment of spiked soils Treatment design. To investigate the effectiveness of magnetic separation of Cu from spiked soils, as well as the potential effect of iron on plants, the following five treatments were designed for plant study: (1) control (non-spiked so il); (2) control with 5% (5g iron/ 100 g soil) of iron filings; (3) spiked soil with iron treatm ent, and the iron filings were then magnetically retrieved; (4) spiked soil with iron treatment, but the iron filings were not magnetically retrieved (to investigate the effect of Cu immobilization in soil); (5) spiked soil without iron treatment. For the sandy soil, two concentrations of iron fili ngs (2.5% and 5%) were used and magnetically retrieved, whereas only 5% of ir on filings were used to treat Cu -spiked organic rich soil 2. Due to the large amount of soil required for growing plants, each treatment was performed in several batches.

PAGE 144

144 Magnetic separation of Cu from soils. For Cu-spiked sandy soil and organic rich soil 2, 80 ml of distilled water were added to 100 g of sp iked soil, and the soil slurry was shaken for 1 hour. Then, iron filings (2.5% and 5% for sandy soil, 5% for organic rich soil 2) were added to the soil slurry, followed by shaking for another 3 hours. Then, the iron filings were retrieved by a Ferrimag rectangular magnet (Scientifics, 152 mm, 3.4 megagauss oersteds). Each soil sample treated by the same amount of iron filings from different batches were combined and mixed thoroughly, air-dried, and ho mogenized for growing plants. Iron immobilization of Cu in soils. The procedure of this treatment was the same as that described above except that th e added iron filings (5%) were not removed from the soil. Therefore, after treatment, the soil and the ir on filings were air-dried and homogenized for growing plants. 7.2.2.3 Pot experiment Soils with the various treatments mentioned abov e were used to grow plants in clay pots under controlled condit ions in a greenhouse. The pot experime nt in this section was the same as that described in Section 7.2.1.3. Th e same two plant species, lettuce (Lactuca sativa) and Indian mustard (Brassica juncea), were planted. For each type of treatment including controls, three replicates were set up for each plant species. After eight weeks, all plants were harvested, and the length and bi omass of the shoots and roots were recorded. The Cu concentration in digested plant tissue was also measured by inductively coupled plasma-atomi c emission spectroscopy (ICP-AES). 7.2.2.4 Toxicity of soils used for growing plants The effect of different treatments on the toxici ty of the sandy soil and the organic rich soil 2 used for growing plants was also tested by MetPLATETM assay. One hundred milliliter of distilled water was added to 50 g of soil, and the soil mixture was shaken for 4 hours followed by

PAGE 145

145 centrifugation at 10,000 rpm for 15 minutes. The supernatant (soil extract) was removed with a pipet and was tested for toxicity after overnight settling in th e refrigerator. All experiments including toxicity tests were run in triplicate. Moderately hard water (60 mg/L Ca, 60 mg/L Mg, pH = 7.4 -7.8) served as the negative cont rol. The detailed procedure for MetPLATETM test was included in Appendix A. 7.3 Results and Discussion 7.3.1 Assessment of Cu Phyt otoxicity Using MetPLATETM 7.3.1.1 Cu phytotoxicity The growth of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) in Cu-spiked soils was investigated. Three types of soils (san dy soil, organic rich soil, and mixed soil) were spiked with different Cu concentrations base d on their Cu binding capacity (see Chapter 2). Compared with the control (i.e. non-spiked) so il, no Cu phytotoxicity was observed in lettuce and mustard during germination and growth in 25 mg/kg and 50 mg/kg Cu-spiked sandy soil. However, in 100 mg/kg Cu-spiked sandy soil, after germination, purple spots started to appear on the leaves of lettuce and mustard, and the seedlings of the plants stopped growing and eventually died, which indicated Cu phytotoxicity Figures 7-1 and 7-2 show the dry biomass of lettuce and mustard grown in spiked sandy soil, respectively. The dry biomass varied between 2.5 g/pot to 2.7 g/pot for shoots, and 0.7 g/pot to 0.8 g/pot for roots (Figure 7-1). No significant difference (5% level according to the Tukey's Studentized Range (HSD) Test) was observed regarding the dry biomass of le ttuce shoots and roots grown in the control, 25 mg/kg, and 50 mg/kg Cu-spiked sandy soils. As re gards Indian mustard, when the Cu input concentration was 25 mg/kg, the dry biomass of shoots (3.5 g/pot) a nd roots (1.2 g/pot) did not differ significantly from those grown in the control soil. However, when the Cu concentration was increased to 50 mg/kg, the dry biomass of mustard decreased to 2.9 g/pot for shoots and 0.9 g/pot for roots, and

PAGE 146

146 the reduction of biomass was significant at the 5% level by performing Tukey's Studentized Range (HSD) Test (Figure 7-2). The effect of Cu concentrations in sandy soil on shoots length of the two plant species was displayed in Figure 73. As regards lettuce, shoot length did not significantly differ in 0 mg/kg (control), 25 mg /kg, and 50 mg/kg Cu-spiked sandy soil, and ranged from 20.9 cm to 24.5 cm. For Indian mustar d, shoot length was 24.3 cm and 22.3 cm in 0 mg/kg and 25 mg/kg Cu-spiked soil, respectively. However, the shoot length was 18.6 cm at 50 mg/kg Cu concentration, which was significantly di fferent from the control soil. Therefore, both plants did not grow at 100 mg/ kg Cu while soil spiking at 50 mg/kg Cu caused little inhibition only to Indian mustard, and 25 mg/kg Cu did not inhibit the gr owth of any of the two plant species. With regard to Cu phytotoxicity in organic rich soil, Figures 7-4, 7-5, and 7-6 show the dry weight and height of lettuce and Indian mu stard grown in 0 mg/kg (control), 50 mg/kg, 500 mg/kg, and 700 mg/kg Cu-spiked or ganic rich soil. In 50 mg/kg Cu -spiked organic rich soil, the dry biomass of both plant species was slightly enhanced (6.7 g/pot and 1.7 g/pot for lettuce shoots and roots, respectively; 6.0 g/pot and 1.5 g/ pot for mustard shoots a nd roots, respectively), as compared to the control plants (5.5 g/pot and 1.2 g/pot for lettuce shoots and roots, respectively; 5.4 g/pot and 1.2 g/pot for mustar d shoots and roots, respectively). Besides, the shoots length of lettuce (20.8 cm) and Indian mu stard (28.0 cm) in 50 mg/kg Cu-spiked organic rich soil were very close to those grown in c ontrol soil (22.6 cm for lettuce and 30.5 cm for mustard). When Cu concentration was increased to 500 mg/kg soil, the growth of lettuce was moderately inhibited. Although th e shoots length of le ttuce (18.2 cm) did not significantly differ from that of the control plant (22.6 cm), the dry biomass of shoots (3.6 g/pot) and roots (0.8 g/pot) significantly decreased. As regards Indian mustard, the growth of mustard in 500 mg/kg

PAGE 147

147 Cu-spiked organic rich soil was se verely inhibited, with total dr y biomass (shoots plus roots) of 0.04 g/pot and shoots length of only 2.1 cm. In 700 mg/kg Cu-spiked or ganic rich soil, the growth of both lettuce and Indian mustard were severely inhibited. As shown in Figure 7-4, the dry biomass of lettuce shoots and roots were only 0.2 g/pot and 0.04 g/pot, respectively. The dry biomass of mustard grown in 700 mg/kg Cu-spiked organic rich soil was 0.02 g/pot for shoots and 0.01 g/pot for roots (Figur e 7-5). In addition, the shoot s length of lettuce and mustard decreased to 6.2 cm and 2.1 cm (see Figure 7-6). In the case of Cu-spiked mixed soil, the grow th of lettuce was not affected in 100 mg/kg and 200 mg/kg Cu-spiked mixed soil. Figure 7-7 s hows that, in 0 mg/ kg (control), 100 mg/kg and 200 mg/kg Cu-spiked mixed soil, the dry biom ass of lettuce varied from 2.2 g/pot to 2.4 g/pot for shoots and 0.17 g/pot to 0.21 g/pot for ro ots. The shoots length of lettuce in 100 mg/kg (26.4 cm) and 200 mg/kg (25.5 cm) Cu-spiked mixe d soil was very close to that grown in the control soil (24.9 cm) (Figure 7-8). When the Cu concentration increased to 500 mg/kg soil, the growth of lettuce was moderately inhibited. Th e shoots length of lettu ce decreased to 14.5 cm (Figure 7-7), and the plant yield decreased to 0.6 g/pot for shoots and 0.03 g/pot for roots (Figure 7-8). The data for Indian must ard were unfortunately lost due to heavy pest infestation. It is known that at sufficiently high concentra tion, heavy metals can cause severe damage to physiological and biochemical activities of pl ants (Nicholls and Mal, 2003), such as the interruption of essentia l enzymes activities, photosyntheti c processes, and nutrient uptake (Sayed, 1999). Many researchers ha ve investigated the phytoxicity of heavy metals to different plant species (Davis and Beckett, 1978; Fj allborg and Dave, 2004; Luo and Rimmer, 1995; Mukherji and Gupta, 1972; Nicholls and Mal, 2003; Sonmez et al., 2006 ). Mukherji and Gupta (1972) studied the effect of Cu2+ on the growth of lettuce (Lectuca sativa) seedlings in cupric

PAGE 148

148 sulfate solution, and they found that their r oot growth was completely inhibited at a Cu2+ concentration of 0.05 M, and the seed germination stopped at a Cu2+ concentration of 0.1 M. They also reported that the inhibition of root growth was relatively stronger than that of hypocotyl growth. Davis and Becke tt (1978) investigated the toxic effect of copper, zinc and nickel in young barley (Hordeurn vulgare L.), wheat (Triticum aestivum L.), rape (Brassica napus L.), lettuce (Lactuca sativa L.) and ryegrass (Lolium perenne). Their results indicated that the dry biomass of these young plants was independe nt on the concentration of Cu, Ni, or Zn in their photosynthesizing tissues until reach to a critical concentra tion (upper critical level). Another study conducted by Nicholls and Mal (2003) assessed the effects of Cu and Pb on the growth of an invasive weed, Lythrum salicaria, in a sandy soil which was spiked with single metal solution and a mixture of Cu and Pb to reach 1000 mg/kg and 2000 mg/kg metal concentrations. 55 days after sp iking, the aboveground parts of the plants were completely withered and died. 7.3.1.2 Toxicity of soils, as determined by MetPLATETM It is known that the total metal content in soil s is a poor indicator of its mobility, toxicity and (phyto)availability. Metal sp eciation is an important factor controlling the metal uptake by plants. Generally, water-soluble a nd exchangeable forms of metals are most available to plants (Tokahoglu and Kartal, 2004) A study conducted by Zhang et al. (1998) showed that the Zn and Mn contents in plant were pos itively correlated with their Fe/Mn oxides-bound fraction in the soil; however, Ca uptake was negatively correlat ed with Ca bound to carbonates in the soil. Sequential extraction techniques have been used extensively to id entify phytoavailable forms of metals in soils (Cajuste et al., 2000; Szakova et al., 2005; Tokahoglu a nd Kartal, 2004; Zhang et al., 1998). However, the multi-step extraction methods are usually very time-consuming;

PAGE 149

149 therefore, we investigated the feasibility of using a rapid heavy metal-specific microbial test, MetPLATETM, to predict heavy metal phytotoxicity in different types of soil. The toxicity of the three t ypes of soil (sandy, organic ri ch, and mixed soil) used for growing plants was tested, and the results are sh own in Table 7-2. For each spiked soil, the EC50 (expressed as % soil extract) of the soil extracts generally decreased as the Cu concentration increased, showing a raise in soil toxicity. In 25 mg/kg and 50 mg Cu2+/kg sandy soil, 50 mg Cu2+/kg organic rich soil, and 100 mg Cu2+/kg mixed soil, the EC50s of the soil extracts were all greater than 100%, showing very little toxicity. Similarly, no Cu phytotoxicity was found in the plants grown in the above Cu-spiked soils. In 100 mg Cu2+/kg sandy soil, the EC50 of the soil extract decreased to 15.3% while in Cu-spiked organic rich soil, the EC50s of the soil extracts were 3.7% and 2.4% at Cu concentrations of 500 mg/kg and 700 mg/kg, respectively. With regard to 200 mg/kg and 500 mg/ kg Cu-spiked mixed soil, the EC50s of the soil extracts were 47.7% and 11.5%, respectively. To find a potential relationship between MetPLATETM toxicity and phytotoxicity, the percent inhibition of undiluted soil extract was also tested by MetPLATETM. Among all spiked soils at different Cu concentrations, the follo wing did not cause phytotoxicity: 25 mg/kg and 50 mg Cu2+ /kg sandy soil, 50 mg Cu2+/kg organic rich soil, and 100 mg/kg and 200 mg Cu2+/kg mixed soil. The percent inhibition of theses und iluted soil extracts varied from non toxic to 67.8%, as determined by MetPLATETM. However, for the spiked soils that caused moderate to high phytotoxicity, the percent inhibition of thei r soil extracts ranged fr om 84.8% to 90.1%. Therefore, based on the results, we could prelimin arily conclude that if a soil extract generates around 90% inhibition by MetPLATETM assay, this soil could probably cause phytotoxicity in lettuce (Lactuca sativa) and Indian mustard (Brassica juncea).

PAGE 150

1507.3.1.3 Cu uptake by lettuce ( Lactuca sativa ) and Indian mustard ( Brassica juncea ) Table 7-3 presents the Cu concentrations in plant shoots and roots grown in different types of soils. Generally speaking, in Cu-spiked soils, the plant uptake of Cu increased with the increase of the input Cu concen tration, indicating an increase in phytoavailable Cu. Comparing the two plant species, in the same type of soil, the Cu concentration in Indian mustard shoots was generally higher than that in le ttuce shoots; however, the roots upt ake of Cu in these two plant species was very close. Besides, for both lettu ce and mustard, their roots always accumulated much more Cu than the shoots. A study carried out by Jordao et al. (200 6) found similar results. They reported that in Cu-enriched vermicompost-amended soil, the Cu concentrations in the roots of lettuce (Lactuca sativa) (76.3 to 244.6mg/kg) were much higher than those in lettuce leaves (5.9 to 13.9mg/kg). Fargasova (2001) also showed that the accumulation of Cd, Cu, Zn, Pb, and Fe in mustard (Sinapis alba) seedlings was higher in the roots than in the shoots, no matter whether the metals were added individual ly or in pairs. By studying Cu uptake from solutions by lettuce (Lactuca sativa), Cheng and Allen (2001) repor ted that Cu uptake by plant roots was affected by free Cu ion activity, pH, and the concentration of other competing ions such as Ca2+. Moreover, Cu concentration in lettuce shoots was much lo wer than that in lettuce roots. As shown in Table 7-3, chemical analysis data for plants grown in 100 mg/kg Cu-spiked sandy soil are not available, due to plant death. However, as the spiked Cu concentration increased from 0 mg/kg (control) to 50 mg/kg, the uptake of Cu by lettuce increased from 4.4 mg/kg to 16.6 mg/kg in shoots, and from 5.7 mg/k g to 344.4 mg/kg in roots. Likewise, the Cu concentrations in Indian mustard increased from 5.2 mg/kg to 65.6 mg/kg in shoots, and from 7.4 mg/kg to 359.7 mg/kg in roots. Plant growth was not inhibited at these Cu levels as shown by the pot experiments. Davis and Beck ett (1978) investigated the mini mum concentrations of Cu in

PAGE 151

151 plant shoots necessary to cause toxic reac tions. They reported that the critical Cu2+ concentrations in plant shoots to cause phytot oxicity were 19 mg/kg for spring barley (Hordeurn vulgare L.), 21 mg/kg for ryegrass (Lolium perenne), 21 mg/kg for lettuce (Lactuca sativa L.), 16mg/kg for rape (Brassica napus L.), and 18 mg/kg for wheat (Triticum aestivum L.) Their findings confirm our pot experiments results th at no phytotoxicity was obs erved in 50 mg/kg Cuspiked sandy soil since the lettuce shoots concen tration was only 16.6 mg/kg, less than 21 mg/kg. In Cu-spiked organic rich soil, as the increase of the Cu input concentration from 0 mg/kg (control) to 50 mg/kg soil, the Cu concentrations in the plant s hoots slightly increased, from 3.5 mg/kg to 6.7 mg/kg in lettuce shoots, and from 4.7 mg/kg to 8.4 mg/kg in mustard shoots; whereas the Cu concentrations in the plant root s increased largely from 8.4 mg/kg to 79.9 mg/kg for lettuce and 7.4 mg/kg to 71.4 mg/kg for must ard. With further increase of the Cu input concentration to 500 mg/kg soil, Cu uptake by le ttuce increased to 50.2 mg/kg in shoots and 670.0 mg/kg in roots; however, the dry bioma ss of Indian mustard was too low to perform chemical analysis on the Cu content. At 700mg Cu2+/kg organic rich soil, only lettuce shoots generated enough biomass for chemical analysis, with a Cu concentration of 106.4 mg/kg. In addition, one common Cu input c oncentration (50 mg/ kg soil) was shared by the sandy soil and organic rich soil. Comparing the Cu uptake by plan ts grown in these two soils, we found that for the same plant, the uptake of Cu from the sandy soil was higher than that from the organic rich soil, which confirmed the higher metal binding ca pacity of the organic soil as discussed in Chapter 2. Table 7-3 also shows the Cu concentrations in lettuce harvested from the mixed soil. For 0 mg/kg (control), 100 mg/kg, 200 mg/kg, and 500 mg/kg Cu-spiked mixed soil, the Cu

PAGE 152

152 concentrations in lettuce shoots increased from 7.8 mg/kg to 318.5 mg/kg, whereas the uptake of Cu by lettuce roots was much higher, ranged from 51.7 mg/kg to 1199.0 mg/kg. 7.3.2 Evaluation of Iron Treatment on Cu-Spiked Soils Using Plant Study The effectiveness of different iron treat ments followed by magnetic separation on 100 mg/kg Cu-spiked sandy soil and 500 mg/kg Cu-spike d organic rich soil 2 were assessed, using plant growth experiment in a greenhouse under co ntrolled conditions. Besides, the effect of added iron filings on plant growth was also investig ated. Iron is an essential nutrient for plants; however, it can cause phytotoxicity when hyperaccumu lated or maldistributed within plant cells (Pich et al., 2001). As mentioned in Section 7.2.2.2, the experiment consisted of the following five treatments: (1) control (non-spiked so il); (2) control with 5% iron filings; (3) spiked so il with iron treatment, and the iron filings were then magnetically retrieve d; (4) spiked soil with iron treatment, but the iron filings were NOT magnetically retrieved (iron immobilization); (5) spiked soil without iron treatment. 7.3.2.1 Effect of treatments on plant growth in a sandy soil Figures 7-9 and 7-10 present the effect of th ese treatments on the dry biomass of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in a sandy soil. Figure 7-11 shows the shoots length of these two pl ant species. As shown in Figures 7-9 and 7-11, the growth of lettuce, as shown by determination of dry biom ass and shoots length, did not differ significantly in control soil (no Cu2+ added) with 5% iron (2.1 g/pot fo r shoots dry biomass, 0.43 g/pot for roots dry biomass, and 22.9 cm for shoots length) and in control soil with out iron (2.2 g/pot for shoots dry biomass, 0.67 g/pot for roots dry biom ass, and 25.0 cm for s hoots length), indicating that the added iron filings did not affect the growth of lettuce at 5% concentration. In nontreated sandy soil, lettuce could not survive due to high phytoxicity. However, after 2.5% or 5%

PAGE 153

153 iron treatment, the growth of lettuce was substant ially enhanced. The dry weight (0.56 g/pot for roots; 3.1 g/pot for shoots) and height (24.8 cm ) of lettuce harvested from 2.5% iron treated soil was also close to the plants grown in control soils. Moreover, the growth of lettuce in 5% iron treated sandy soils was also completely rest ored. The plant weight and height did not significantly differ between pots where iron was re trieved (0.56 g/pot for roots, 3.0 g/pot for shoots, and 27.3 cm for shoots length) and pots where the iron was not retrieved from the soil (0.52 g/pot for roots, 3.2 g/pot for shoots, and 27.5 cm for shoots length) (Figures 7-9 and 7-11). Similar results were obtained as regards the growth of Indian mustard in sandy soil (Figures 7-10 and 7-11). No si gnificant effect of iron filings was observed on the growth of mustard. The dry weight and height of Indian mustard grown in control soil (no Cu2+ added) with 5% iron (4.6 g/pot for shoots, 1.6 g/pot for roots, and 23.4 cm for shoots le ngth) were quite close to that grown in control soil without iron filings (4.2 g/pot for shoots, 1.9 g/pot for roots, and 22.3 cm for shoots length). Indian mustard coul d not survive in non-treated sandy soil; however, its growth was significantly impr oved in iron-treated soil. After the 5% iron treatment, the growth of Indian mustard was also completely restored. Moreover, the plant weight and height did not significantly differ between soils where iron was retrieved (1.8 g/pot for roots, 4.2g/pot for shoots, and 21.2cm for shoots length) and soils where the iron was not re trieved (1.6 g/pot for roots, 4.4 g/pot for roots, and 24.8 cm for shoots length). In the 2.5% iron-treated sandy soil, as compared with plants in the cont rol soil, the dry plant biomass (2.4 g/pot for shoots and 0.5 g/pot for roots) was somewhat lower, although the shoots length of Indian mustard (20.0 cm) was not significantly different. This indicated that treatmen t with 2.5% iron filings was not as effective as 5% iron filings for reducing Cu phytot oxicity towards Indian mustard.

PAGE 154

1547.3.2.2 Effect of treatments on plant growth in organic soil 2 As shown in Figures 7-12 and 7-13, and 7-14, the growth of lettuce and Indian mustard in control soil with or without iron filings showed no significan t difference. However, in nontreated 500 mg/kg Cu-spiked organi c rich soil 2, the gr owth of lettuce and Indian mustard was significantly inhibited, with redu ced dry biomass of 1.2 g/pot (s hoots) and 0.2 g/pot (roots) for lettuce, and 0.21 g/pot (shoots) and 0.03 g/pot (roots) for mustard. Moreover, the height of lettuce and Indian mustard grow n in non-treated organic rich so il 2 also decreased to 15.5 cm and 4.6 cm, respectively. After 5% iron treatment followed by magnetic separation, the growth of lettuce and Indian mustard was improved to the similar le vel of plants grown in noncontaminated soil. In 5% iron treated organic ri ch soil 2 where the iron filings were magnetically retrieved, the dry biomass and shoots length of le ttuce were 5.8 g/pot (sh oots), 0.7 g/pot (roots), and 38.9 cm, and the dry biomass and height of Indian mustard were 8.1g/pot for shoots, 2.9g/pot for roots, and 29.0cm. When the iron fi lings were not removed, the dry weight and height for plants were 4.3 g/pot (shoots), 0.6 g/pot (roots) and 32.3 cm for lettuce. For Indian mustard, the dry weight and hei ght for plants were 7.5 g/pot (s hoots), 2.9 g/pot (roots) and 27.3 cm for mustard (see Figure s 7-12, 7-13, and 7-14). Many other researchers have also used plant growth to study the effectiveness of different soil amendments in reducing heavy metal mobility and bioavailability in contaminated soils (Boisson et al., 1999; Castaldi et al., 2005; Chen et al., 2000 ; Smeulders et al., 1983a, b; Smeulders and Vandegeijn, 1983). Boisson et al. (1999) applied hydroxyapatite (HA) to treat a metal (Zn, Pb, Cu, Cd) and As-contaminated soil, and the treatment effectiveness was examined by growing maize (Zea mays L.) and bean (Phaseolus vulgaris) in treated and non-treated soil. Their data indicated that plant growth was partly restored at 0.5% and 1% hydroxyapatite treated soils. However, at 5% hydroxyapatite applicatio n rate, plant growth was inhibited again which

PAGE 155

155 may due to the simultaneous immobilization of essential nutrients. In another study reported by Castaldi et al. (2005), the effect s of three amendments (zeolite, compost, and calcium hydroxide) on the immobilization of Pb, Cd and Zn in a c ontaminated soil were determined and their influence on white lupin (Lupinus albus) growth was also investigat ed. Results showed that the growth of white lupin was enhanced in amended soils. With respect to unamended soil, the plant shoots biomass increased with a factor of 1.8 (so il amended with zeolite), 3.6 (soil amended with compost), and 3.1 (soil amended with calcium hydr oxide), and the roots biomass increases with a factor of 1.4 (soil amended w ith zeolite), 5.6 (soil amended with compost), and 4.8 (soil amended with calcium hydroxide). 7.3.3 Toxicity of Iron Treated soils, as Determined by MetPLATETM The above plant study has shown the effectiven ess of iron treatment on Cu-spiked soils. It was found that following iron treatment, the phytoxicity caused by Cu was substantially decreased. We paralle lly used the MetPLATETM assay to test the toxicity of these iron-treated soils used for growing plants. The results are show n in Table 7-4. With regard to sandy soil, no toxicity was found in the control soil with 5% iron filings. Afte r 2.5% and 5% iron treatment, whether the iron filings were magnetically retrie ved or not, the treated soil extracts hardly showed any toxicity (7.5% to 8.5% inhibition), with EC50s all greater than 100%, which agreed with the plant study showing no phytotoxicity in these treated soils. However, in non-treated 100mg/kg Cu-spiked sandy soil, the EC50 of the soil extract was 15.4%, and the non diluted soil extract generated an inhi bition of 86.8% by MetPLATETM, which caused very high phytotoxicity resulting in plant death. The toxicity of the organic rich soil 2 with or without iron treatments was also displayed in Table 7-4. The plant study showed that th e growth of lettuce a nd Indian mustard was significantly inhibited in the nontreated 500 mg/kg Cu-spiked organi c rich soil 2, whereas in 5%

PAGE 156

156 iron-treated soils, no phytotoxicity wa s observed. After performing MetPLATETM toxicity test to the same soils, we found that the non-treated soil extract produced an inhibition of 78.2%; however, after 5% iron treatment, the EC50s of the soil extracts were all greater than 100%, and much lower percent inhibitions (16.6% in ir on retrieved soil extrac t and 14.9% in iron immobilized soil extract) were generated. The MetPLATETM assay showed consistent data with the plant study. In a study carri ed out by Wundram et al. (1996) a new phytotoxicity test, based on the inhibition of photosynthesis of the green algae Chlamydomonas reinhardtii, was investigated and compared with other phytotoxicity tests using Lemna (growth) and Lepidium (root elongation). The experiment was conducted in solutions with distinct heavy metals and solutions with complex mixture of heavy metals. Their data indicated that Chlamydomonas was most sensitive to Hg and Cd, whereas Lepidium and Lemna had highest sensitiv ity to Cu and Cd. In the solution containing a mixture of various heavy metals including Cd, Cu, Zn, and Pb, Chlamydomonas showed the highest sensitivity. Trapp et al. (2000) developed a short-term acute toxicity assay based on the change of transpira tion of willow tree cutting grown in contaminated solution. The sensitivity of the test was evaluated with 3,5-dichlorophenol and the EC50s were found between 5.8 and 9.6 mg/L after 48-h and 72-h exposure, which were similar to the results from algal (Scenedesmus subspicatus and Raphidocelis subcapita) growth tests. 7.3.4 Copper Uptake by Plants in Treated and Non-treated Soils Chemical analysis by ICP-AES wa s also performed for lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) harvested from treated and non-tr eated sandy soil and organic rich soil 2. For each plant species, in the same soil, the uptake of Cu accumulated mainly in the roots, and this is agreed with results referred by other authors (Castald i et al., 2005; Cheng and Allen, 2001; Fargasova, 2001; Jordao et al., 2006). Moreover, the shoots Cu concentration in Indian

PAGE 157

157 mustard was generally higher than that in lettuce; however, the roots uptake of Cu in these two plant species was similar. These finding agreed with the results shown in Section 7.3.1.3. 7.3.4.1 Copper uptake by plants grown in sandy soil As shown in Table 7-5, Cu uptake by lettuce shoots in sandy soil followed this order: 2.5% iron treated soil iron retrieved (18.9 mg/kg) > 5% iron treated soil iron immobilized (13.1 mg/kg) 5% iron treated soil iron retrieved (1 2.0 mg/kg) > control soil with 5% iron (7.2mg/kg) control soil without iron (6 .4 mg/kg). However, the uptak e of Cu by lettuce roots in control with 5% iron (109.0 mg/ kg) was significantly higher than that in control without iron filings (9.2 mg/kg), which was probably due to the Cu background in iron filings (2503.3 mg/kg). Besides, 125.4 mg/kg and 182.3 mg/kg of Cu were found resp ectively in lettuce roots at 5% and 2.5% iron treated sandy soil where the ir on filings were magnetically retrieved. When the 5% iron filings were not removed from the tr eated sandy soil, a little higher Cu concentration (204.3 mg/kg) was found in lettuce roots, which may also result from the Cu background in the iron filings. Table 7-5 also shows the Cu uptake by Indi an mustard in treated and non-treated sandy soil, which followed the same trend as the Cu upt ake by lettuce. The shoots Cu concentrations of mustard grown in the control with and with iron filings were 7.4 mg/kg and 5.5 mg/kg, respectively. The shoots Cu uptake by mustard fr om 5% and 2.5% iron-treated sandy soil were somewhat enhanced, varied from 30.9 mg/kg to 51.3 mg/kg. As regards roots Cu uptake by Indian mustard, 115.4 mg/kg Cu was found in the roots grown in 5% iron treated control, as compared to 8.3 mg/kg in roots from non-treated control (i.e. no iron filing s added). Besides, the roots Cu concentrations in iron treated soil va ried from 135.2 mg/kg to 215.9 mg/kg, with the highest value found in the soil where the 5% ir on filings were not magnetically retrieved.

PAGE 158

158 Chemical analysis for plants grown in 100 mg/ kg Cu-spiked sandy soil was not available due to plant death. 7.3.4.2 Copper uptake by plants grow n in organic rich soil 2 The Cu contents in lettuce and Indian mustar d harvested from the organic rich soil 2 are displayed in Table 7-6. The shoots Cu contents in control soil with 5% iron (7.4mg/kg for lettuce and 5.4 mg/kg for mustard) were very close to that in control soil without iron (6.6 mg/kg for lettuce and 5.1 mg/kg for Indian mustard). Without iron treatment, in 500mg/kg Cu-spiked organic soil 2, the shoots Cu uptake was 75.3m g/kg for lettuce and 128.9 mg/kg for mustard. However, after iron treatment, s ubstantial decrease of Cu conten t in plant shoots was obtained, indicating a drop of phytoavailable Cu. The shoots Cu uptake in 5% iron-treated organic rich soil 2, where iron filings were magnetically retrieve d, was 18.4 mg/kg for lettuce and 25.4 mg/kg for mustard. When the iron flings we re not removed, a little higher Cu concentrations, 30.0 mg/kg and 31.5 mg/kg, were respectively found in th e shoots of lettuce and Indian mustard. With regard to Cu uptake in plant roots, as shown in Table 7-6, 10.4 mg/kg and 9.4 mg/kg Cu were found in the roots of lettuce and Indian mustard, resp ectively, grown in control soil without the addition of iron filings. In the presence of 5% iron, the Cu content in roots grown in non-spiked organic rich soil 2 (c ontrol) increased to 61.7 mg/kg for lettuce and 55.0 mg/kg for mustard. However, these values were lower th an those found in the sandy soil (109.0 mg/kg for lettuce and 115.4 mg/kg for mustard) due to the higher Cu binding capacit y of organic rich soil as shown in Chapter 2. Without iron treatment, in 500 mg/kg Cu-spiked organic rich soil 2, 1131.2 mg/kg and 1025.2 mg/kg Cu was accumulated in the roots of lettuce and Indian mustard, respectively. However, after treatment with 5% iron filings, the roots Cu content was significantly reduced to 398.6 mg /kg for lettuce and 352.9 mg/kg fo r Indian mustard when the iron filings were magnetically retr ieved. However, when the iron filings were not removed from

PAGE 159

159 the treated soil, 448.9 mg/kg and 460.7 mg/kg Cu was found in the roots of lettuce and Indian mustard, respectively. Lombi et al. (2002a) utilized a Fe-oxide rich material (red mud) to immobilize heavy metals (Cd, Zn, Cu, Ni, and Pb) in contaminated so ils, at an application rate of 2% (w/w). After performing sequential extraction ex periment, they found that this soil amendment shifted metals from the exchangeable forms to the Fe-bound fractions. In a companion paper (Lombi et al., 2002b), Lombi and his colleagues also examined the effects of the above soil amendment by using biological indicators, such as plants growth, metal uptake, and soil microbial activity. The results showed that the application of red mud reduced phytotoxicity of heavy metals, enhanced plant yields and decreased the metal concentra tions in plants, which supports quite well the findings in our study. 7.4 Conclusions The feasibility of using PletPLATETM toxicity test to predic t Cu phytotoxicity was first investigated in this Chapter. The result s of the plant study showed that lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) could not survive in 100 mg/kg Cu-spiked sandy soil, whereas in 25 mg/kg and 50 mg/ kg Cu-spiked sandy soil, no phytot oxicity was observed. In Cuspiked organic rich soil, 50 mg/kg Cu concentrat ion did not cause any pl ant inhibition; however, at 500 mg/kg and 700 mg/kg Cu concentrations, th e growth of the plants was significantly inhibited, indicating the phytot oxicity of Cu. In 100 mg/kg, 200 mg/kg, and 500 mg/kg Cuspiked mixed soil, only the highest Cu concentr ation, 500 mg/kg soil, in hibited the growth of lettuce to some extent. The data for Indian mu stard grown in mixed soil was unfortunately lost due to heavy pest infestation. The toxicity of the above soils was also tested, using MetPLATETM. We concluded that if a soil extrac t showed approximately 90% inhibition by MetPLATETM assay, this soil could probabl y cause phytotoxicity in lettuce (Lactuca sativa) and

PAGE 160

160 Indian mustard (Brassica juncea). Moreover, chemical analysis on Cu content in plant tissue suggested that the plant uptake of Cu increased with the increase of the input Cu concentration, indicating the increase of phytoavailable Cu. Beside s, for each plant in the same type of soil, roots could always accumulate much more Cu th an shoots. Comparing th e two plant species, the shoots Cu concentration in Indian mustard was gene rally higher than that in lettuce; however, the roots uptake of Cu in these tw o plant species was similar. In addition, we employed pot experiments to demonstrate the effectiveness of iron treatment followed by magnetic separation for reduc ing phytoavailable Cu. The results indicated that iron filings at 5% concentr ation did not exert any adverse effect on plant growth. In Cuspiked sandy soil and organic ri ch soil 2, the plant growth in non-treated soil was substantial inhibited. However, after iron treatment, whethe r the iron filings were magnetically retrieved or not, the growth of lettuce and Indian mustard wa s significantly enhanced, and the plant height and weight were similar to thos e observed in control soils that were not spiked with Cu. Moreover, chemical analysis showed a great reduction of the Cu c ontent in plant shoots and roots after iron treatment of the Cu-spi ked organic rich soil 2, which al so indicated the decrease of phytoavailable Cu. In all, we conclude that MetPLATETM toxicity test showed great potential in predicting Cu phytotoxicity. The effectiveness of the iron treatment fo r reducing phytoavailable Cu in soils was also confirmed by the plant study.

PAGE 161

161 Table 7-1 Soils Characteristics Characteristic Sandy soil Organic soil Mixed soil Organic soil 2 pH 5.75.75.55.3 Eh (mV) 422403374.7337 % Organic carbon 0.56.43.34.1 % Organic matter 1.618.89.612.2 % Sand 96.993.292.092.6 % Silt 0.022.02.60.8 % Clay CEC (cmolc/kg) 3.06 14.1 4.8 230.1 5.4 122.3 6.6 107.8 a a a a a a100mg/kg Cu 100mg/kg Cu0 1 2 3 4 5 ShootsRootsDry biomass (g/pot) Control (no Cu added) 25mg/kg Cu 50mg/kg Cu 100mg/kg Cu Figure 7-1. Effect of Cu con centrations on dry biomass of shoots and roots of lettuce (Lactuca sativa) grown in spiked sandy soil (No grow th was observed in lettuce grown 100 mg/kg Cu-spiked sandy soil. Error bars re present standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 162

162 a a a a b b100mg/kg Cu 100mg/kg Cu0 1 2 3 4 5 ShootsRootsDry biomass (g/pot) Control (no Cu added) 25mg/kg Cu 50 mg/kg Cu 100mg/kg Cu Figure 7-2. Effect of Cu concen trations on dry biomass of shoot s and roots of Indian mustard (Brassica juncea) grown in spiked sandy soil (No growth was observed in 100 mg/kg Cu-spiked sandy soil. Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test). a a a a b a100mg/kg Cu 100mg/kg Cu0 10 20 30 40LettuceMustardShoots length (cm) Control (no Cu added) 25mg/kg Cu 50mg/kg Cu 100mg/kg Cu Figure 7-3. Effect of Cu concentr ations on shoots length of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in spiked sandy soil (No growth was observed in lettuce and mustard grown in 100 mg/kg Cu -spiked sandy soil. Error bars represent standard deviation of three replicates. Valu es followed by the same letter within the same group do not differ significantly at th e 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 163

163 a a b bc c d d 0 2 4 6 8 ShootsRootsDry biomass (g/pot) Control (no Cu added) 50mg/kg Cu 500mg/kg Cu 700mg/kg Cu Figure 7-4. Effect of Cu con centrations on dry biomass of shoots and roots of lettuce (Lactuca sativa) grown in spiked organic rich soil (Erro r bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level accordi ng to the Tukey's Stude ntized Range (HSD) Test). a a a a b b b b 0 2 4 6 8 ShootsRootsDry biomass (g/pot) Control (no Cu added) 50mg/kg Cu 500mg/kg Cu 700mg/kg Cu Figure 7-5. Effect of Cu concen trations on dry biomass of shoot s and roots of Indian mustard (Brassica juncea) grown in spiked organic rich soil (Error bars represent standard deviation of three replicates. Values follo wed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 164

164 a a a a b a b b 0 10 20 30 40 LettuceMustardShoots length (cm) Control (no Cu added) 50mg/kg Cu 500mg/kg Cu 700mg/kg Cu Figure 7-6. Effect of Cu concentr ations on shoots length of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in spiked organic rich soil (Error bars represent standard deviation of three replicates. Valu es followed by the same letter within the same group do not differ significantly at th e 5% level according to the Tukey's Studentized Range (HSD) Test). a a a a a a b b 0 1 2 3 4 ShootsRootsDry biomass (g/pot) Control (no Cu added) 100mg/kg Cu 200mg/kg Cu 500mg/kg Cu Figure 7-7. Effect of Cu con centrations on dry biomass of shoots and roots of lettuce (Lactuca sativa) grown in spiked mixed soil (Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 165

165 b a a a 0 10 20 30 40 0100200500Cu concentration in spiked mixed soil (mg/kg)Shoots length (cm) Figure 7-8. Effect of Cu concentr ations on shoots length of lettuce (Lactuca sativa) grown in spiked mixed soil (Error bars represent standa rd deviation of three replicates. Values followed by the same letter do not differ signi ficantly at the 5% level according to the Tukey's Studentized Range (HSD) Test). Table 7-2. Copper toxicity in spiked sandy soil, organic rich soil, and mixed soil used for growing plants, as determined by MetPLATETM. Soil Type Cu Conc. in spiked soil (mg/kg) EC50 of soil extract (% soil extract) % Inhibition of undiluted soil extract 25>100%Not Toxic 50>100%31.4 2.1% Sandy 10015.3 2.0%a89.9 0.4% 50>100%23.6 3.0% 5003.7 0.1%89.9 0.9% Organic rich 7002.4 0.06%90.1 0.8% 100>100%31.6 1.1% 20047.7 1.3%67.8 4.2% Mixed 50011.5 0.6%84.8 2.3% a Mean of 3 replicates 1 standard deviation

PAGE 166

166 Table 7-3. Copper uptake by lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in spiked sandy soil and organic soil, as determined by chemical analysis. Cu conc. in shoots (mg/kg) Cu conc. in roots (mg/kg) Soil Type Initial Cu conc. in spiked soil (mg/kg) Lettuce Mustard Lettuce Mustard 0 4.4 0.6a5.2 0.95.7 1.4 7.4 0.8 25 7.6 1.527.0 3.2123.8 12.3 132.1 15.3 50 16.6 2.165.6 8.5344.4 0.6 359.7 24.6 Sandy 100 N/AbN/Ab N/Ab N/Ab 0 3.5 1.14.7 1.3 7.4 1.1 8.4 0.9 50 6.7 1.28.4 0.871.4 3.4 79.9 3.8 500 50.2 .5N/Ac 670.0 6.2 N/Ac700 106.4 19.4N/Ac N/Ac N/Ac Organic rich Mixed 0 100 200 500 7.8 0.9 36.9 6.1 66.2 11.9 318.5 46.0 N/Ad N/Ad N/Ad N/Ad 51.7 2.5 243.9 26.4 706.3 43.4 1199.0 113.2 N/Ad N/Ad N/Ad N/Ad a Mean of 3 replicates 1 standard deviation; b Not available due to death of the plants; c Not available due to low biomass; d Data were lost due to heavy pest infestation.

PAGE 167

167 a a a a a a a a a aNon-treated Non-treated0 1 2 3 4 5 6 ShootsRootsDry Biomass (g/pot) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Treated w/2.5% iron, iron retrieved Non-treated Figure 7-9. Effect of different treatments on dry biomass of shoots and roots of lettuce (Lactuca sativa) grown in a sandy soil (Cu concentrati on in spiked sandy soil was 100 mg/kg before treatment. No growth was observed for lettuce grown in non-treated sandy soil. Error bars represent standard deviati on of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 168

168 a a a a a a a a b bNon-treated Non-treated0 1 2 3 4 5 6 ShootsRootsDry Biomass (g/pot) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Treated w/2.5% iron, iron retrieved Non-treated Figure 7-10. Effect of different treatments on dr y biomass of shoots and r oots of Indian mustard (Brassica juncea) grown in a sandy soil (Cu concentr ation in spiked sandy soil was 100 mg/kg before treatment. No growth wa s observed for Indian mustard grown in non-treated sandy soil. Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test).

PAGE 169

169 a a a a a a a a a aNon-treated Non-treated0 10 20 30 40 50 LettuceMustardShoots length (cm) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Treated w/2.5%, iron retrieved Non-treated Figure 7-11. Effect of different trea tments on shoots length of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in sandy soil (Cu concen tration in spiked sandy soil was 100 mg/kg before treatment. No growth was observed in both Lettuce and Indian mustard grown in non-treated sandy soil. Erro r bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level accordi ng to the Tukey's Stude ntized Range (HSD) Test).

PAGE 170

170 a a a a a a a a b b 0 2 4 6 8 10 ShootsRootsDry Biomass (g/pot) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Non-treated Figure 7-12. Effect of different treatments on dry biomass of shoots and roots of lettuce (Lactuca sativa) grown in organic rich soil 2 (Cu con centration in spiked organic soil 2 was 500 mg/kg before treatment with iron filings Error bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level accordi ng to the Tukey's Stude ntized Range (HSD) Test).

PAGE 171

171 a a a a a a a a b b 0 2 4 6 8 10 12 ShootsRootsDry Biomass (g/pot) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Non-treated Figure 7-13. Effect of different treatments on dr y biomass of shoots and r oots of Indian mustard (Brassica juncea) grown in organic rich soil 2 (C u concentration in spiked organic soil 2 was 500 mg/kg before treatment. Erro r bars represent standard deviation of three replicates. Values followed by the same letter within the same group do not differ significantly at the 5% level accordi ng to the Tukey's Stude ntized Range (HSD) Test).

PAGE 172

172 a a a a a a a b b c0 10 20 30 40 50 60LettuceMustardShoots length (cm) Control (no Cu added) Control w/5% iron Treated w/5% iron, iron retrieved Treated w/5% iron, iron immobilized Non-treated Figure 7-14. Effect of different trea tments on shoots length of lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in organic rich soil 2 (Cu concentration in spiked organic soil 2 was 500 mg/kg before treat ment. Error bars represent standard deviation of three replicates. Values follo wed by the same letter within the same group do not differ significantly at the 5% level according to the Tukey's Studentized Range (HSD) Test). Table 7-4. Effect of different treatments on copper toxicity in sandy soil and organic rich soil 2 used for growing plants, as determined by MetPLATETM Soil Type Initial Cu conc. in spiked soil (mg/kg) Treatment EC50 of soil extract (% soil extract) % Inhibition of undiluted soil extract 0Control w/ 5% of iron >100% -0.9 0.2%a Treated w/ 5% of iron, iron retrieved >100% 8.2 0.2%bTreated w/ 5% of iron, iron immobilized >100% 7.5 0.9% Treated w/ 2.5% of iron, iron retrieved >100% 8.4 1.5% Sandy 100 100 100 100Non-treated 15.4 3.4% 86.8 0.1% 0Control w/ 5% of iron >100% 11.9 1.8% Treated w/ 5% of iron, iron retrieved >100% 16.6 1.8% Treated w/ 5% of iron, iron immobilized >100% 14.9 2.7% Organic rich 500 500 500Non-treated 15.9 0.4% 78.2 1.6% a Non-toxic; b Mean of 3 replicates 1 standard deviation.

PAGE 173

173 Table 7-5. Copper uptake by lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in treated and non-tre ated sandy soil, as determined by chemical analysis Plant Type Initial Cu conc. in spiked soil (mg/kg) Treatment Cu conc.a in shoots (mg/kg) Cu conc. in roots (mg/kg) Control 6.4 0.5 9.2 0.7 0 0Control w/ 5% of iron 7.2 2.0 109.0 7.2Treated w/ 5% of iron, iron retrieved 12.0 2.0 125.4 11.7Treated w/ 5% of iron, iron immobilized 13.1 1.9 204.3 10.9Treated w/ 2.5% of iron, iron retrieved 18.9 2.7 182.3 8.9 Lettuce 100 100 100 100 Non-treated N/Ab N/Ab Control 5.5 1.9 8.3 2.5 0 0Control w/ 5% of iron 7.4 0.3 115.4 19.0Treated w/ 5% of iron, iron retrieved 35.7 0.3 135.2 0.9Treated w/ 5% of iron, iron immobilized 30.9 0.5 215.9 11.7Treated w/ 2.5% of iron, iron retrieved 51.3 5.6 198.8 6.6 Mustard 100 100 100 100 Non-treated N/Ab N/Ab a Mean of 3 replicates 1 standard deviation; b Not available due to plant death. Table 7-6. Copper uptake by lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) grown in treated and non-treated organic ri ch soil 2, as determined by chemical analysis Plant Type Initial Cu conc. in spiked soil (mg/kg) Treatment Cu conc. in shoots (mg/kg) Cu conc. in roots (mg/kg) Control 6.6 1.7a 10.4 1.9 0 0Control w/ 5% of iron 7.4 1.3 61.7 8.0 Treated w/ 5% of iron, iron retrieved 18.4 2.6 398.6 26.8Treated w/ 5% of iron, iron immobilized 30.0 8.5 448.9 18.4 Lettuce 500 500 500 Non-treated 75.3 3.8 1131.2 79.6 Control 5.1 0.1 9.4 2.1 0 0Control w/ 5% of iron 5.4 1.8 55.0 5.9Treated w/ 5% of iron, iron retrieved 25.4 2.5 352.9 47.9Treated w/ 5% of iron, iron immobilized 31.5 4.9 460.7 43.5 Mustard 500 500 500 Non-treated 128.9 11.1 1025.2 37.2 a Mean of 3 replicates 1 standard deviation

PAGE 174

174 CHAPTER 8 SUMMARY AND CONCLUSIONS 8.1 Summary Heavy metal contamination of soils and sedime nts is a worldwide issue. The main purpose of this research was to inves tigate the effectiv eness of a magnetic treatment method for removing heavy metals from contaminated soils and sediments. The treatment approach was based on adsorbing the metal contaminants onto iron filings seed and removing the metal-laden filings by magnetic separation. This dissertation covered six experiment al studies. The first study dealt with a toxicological approach to assess the heavy metal binding capacity of different types of soils. The toxicity test used was MetPLATETM, a rapid bacterial assay that is specific for heavy metal toxicity. The binding capacity of soils towards Cu2+, Zn2+, and Hg2+ was assessed. In the second study we determined the conditions for adsorptio n of the metals to iron filings followed by magnetic treatment. The parameters studied incl ude the concentrations of added iron filings, level of soil saturation, and th e contact time between the iron filings and the soil matrix. The effectiveness of magnetic treatment on heavy metal (Cu2+, Cd2+, and Zn2+) removal from artificially contaminated soils was also evaluated. Toxicity tests, chemical analysis, mass balance study, and sequential extraction of metals were a ll performed for treated and non-treated soils. In the following study, we investigated the effects of aging on the magnetic separation process, as well as the change of Cu and Zn toxicity in ag ed soils. In the fourth study we investigated the magnetic separation of Pb from shooting range soils, and the fifth study focused on removing Cu2+, Zn2+, and Hg2+ from artificially contaminated sediments by the same magnetic separation method. In the final study, we discus sed the feasibility of using MetPLATETM assay to predict

PAGE 175

175 heavy metal phytotoxicity, as well as the use of plants to assess the effectiveness of magnetic separation for removing Cu from soils. 8.2 Conclusions Based on the above studies, the fo llowing conclusions were drawn: A novel toxicological approach is proposed to determine the heavy metal binding capacity of soils. Cu2+, Zn2+, and Hg2+ binding capacity for different types of soils followed this trend: Georgia clay rich soil > organic rich soils > sandy soils. The recovery of iron filings from a sandy soil was very high under all three soil saturation conditions (dry soil, soil at field capacity, and water-saturated soil). For red sandy soil, organic rich soil, and Georgia cl ay rich soil, the recovery of iron filings from saturated soil was lower than that from soils unde r dry and field capacity conditions. The conditions for magnetic separation of me tals from soils were: An iron filings concentration of 5% (w/w) and a contact time between iron filings and the soil matrix of 3 hours. Cd-spiked soils showed higher toxicity than Cu and Znspiked soils. Besides, at the same concentration, the toxicity of each metal in di fferent soils was shown to follow this trend: sandy soils> Georgia clay rich soil > organic rich soil. The magnetic treatment method proposed worked best on Cu-spiked soils, followed by the Zn-spiked soils and the Cd-spiked soils. In a ddition, the effectiveness of this magnetic treatment was not significantly affected by the type of soil in Cu-spiked soils. However, as regards Znand Cd-spiked soils, this method wo rked better in sandy so ils than in organic rich soil and Georgia clay rich soil. A large portion of the added metals was imm obilized in the soil matrix. After magnetic treatment, Cu, Zn, and Cd were removed from both the soil extracts and the soil matrix. However, for each heavy metal, in the same soil, metal removal from the soil extract was always higher than that from the soil matrix. After magnetic treatment, the removal of Cu, Zn, and Cd was the greatest from the exchangeable fraction, followed by the carbonate fraction in sandy soil. In organic rich soil, the removal of these metals was found in all fractions, in which the highest removal was from the organic-bound phase, and th e second highest removal was from the exchangeable phase for Cu, and the Fe-Mn oxide phase for Zn and Cd. Energy dispersive X-ray spectroscopy (EDS) c onfirmed the adsorption of Cu, Zn, and Cd on magnetically retrieved iron filings.

PAGE 176

176 The retrieved iron filings coul d be regenerated in 1 N HNO3 for 1 hr and then retreated by 1 M NaOH for 72 hrs prior to reuse. The regenerated iron filings worked as well as fresh iron filings even after two regeneration cycles. To simulate field conditions, we studied the eff ect of soil aging on Cu t oxicity in soils. As in the sandy soil, Cu toxicity did not show significant change duri ng the initial 4-month dry aging as determined by both MetPLATETM and the 48-h Ceriodaphnia dubia test. However, after 20 wet-dry cycles, Cu toxicity decreased gradually with time. Zn toxicity in aged sandy soil gradually decreased after 2month aging as shown by both toxicity tests. In organic rich soil, Cu toxicity di d not change significantly when using C.dubia test, however, MetPLATETM showed slightly increase in Cu toxicity. Zn toxicity in aged organic rich soil only showed some decrease after the 12th wet-dry cycle by both toxicity tests. No significant reduction in magnetic separation efficiency wa s observed as regards Cu and Zn removal from aged soils. The magnetic separation method also showed gr eat potential in treating Pb contaminated shooting range soils. Our proposed treatment achieved a great reduction of toxicity generated by Pb in soils. Moreover, Pb was removed from both the soil matrix and the soil extracts, although the removal from the soil matrix was always lower than that from the soil extracts. We also used the toxicological approach to assess the heavy metal binding capacity of aquatic sediments. Organic rich sediments showed higher metal (Cu2+, Zn2+, and Hg2+) binding capacity than sandy sediments. Cu2+, Zn2+, and Hg2+ could also be removed effec tively from sediments by magnetic separation. The type of sediment and metal di d not affect the treatment effectiveness. The metals were removed from both the sediment matrix and the sediment extracts. However, as observed in soils, metal removal from the sediment matrix was lower than that from the sediment extracts. The sensitivity of the toxicity tests were shown in this order: the 96-h Selenastrum capricornutum test > the 48-h Ceriodaphnia dubia test > MetPLATETM assay. However, the three tests showed similar tr ends as regards toxicity rem oval of metals from soils and sediments. To investigate Cu phytotoxicity, we conducte d a plant growth study and the following results were obtained. Lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) could not survive in 100 mg/kg Cu-spiked sandy so il, whereas in 25 mg/kg and 50 mg/kg Cuspiked sandy soil, no phytotoxicity was observe d. In Cu-spiked organic rich soil, 50 mg/kg Cu concentration did not cause any growth i nhibition of both Lettuce and Indian mustard; however, at 500 mg/kg and 700 mg/kg soil Cu concentrations, plant growth was significantly inhibited. In 100 mg/kg, 200 mg/kg, and 500 mg/kg Cu-spiked mixed soil, only the highest Cu concentration, 500 mg/ kg soil, inhibited the growth of lettuce (Lactuca sativa) to some extent. The data for Indian mustar d was lost due to heavy pest infestation.

PAGE 177

177 Cu uptake by Lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) increased as the increase of the Cu input c oncentrations in soils. Plant ro ots always accumulated more Cu than the aerial shoots. Furthermore, Cu c oncentration in the shoots of Indian mustard was generally higher than that in the shoots of lettuce. However, Cu uptake by the roots was similar in both plants. MetPLATETM toxicity test showed great potential in predicting metal phytotoxicity. If a soil extract showed approximately 90% inhibition by MetPLATETM assay, this soil could probably cause phytotoxicity in lettuce (Lactuca sativa) and Indian mustard (Brassica juncea). The plant growth study also demonstrated the effectiveness of the proposed magnetic treatment on reducing Cu phytoavailability. Afte r treatment, whether the iron filings were magnetically retrieved from the so il or not, the growth of Lettuce (Lactuca sativa) and Indian mustard (Brassica juncea) was significantly enhanced, and the plant height and weight were very similar to plants grown in control soils (i.e. no Cu added). The Cu content in plant shoots a nd roots also significantly decreased after treatment. The plant growth study confirmed the results obtained through the use of short-term toxicity tests such as MetPLATETM. This suggests that these short-term toxicity tests could be used to assess the effectiveness of present and future treatment methods for soil and sediment decontamination.

PAGE 178

178 APPENDIX A DETAILED PROCEDURE FOR TOXICITY TESTS A.1 MetPLATETM Procedure MetPLATETM is a rapid, quantitative microbial a ssay developed by Bitton et al. (1994). The particularity of this test is that it is spec ific for heavy metal toxicity and is not affected by relatively high concentrations of organic toxicants. MetPLATETM assay is based on inhibition of the activity of -galactosidase in a mutant strain of Escherichia coli. The MetPLATETM kit includes freeze-dried E. coli (bacterial reagent), moderately hard water (diluent), buffered chromogenic enzyme substrate (chlorophenol-red -galactopyranoside (CPRG)), a positive control, and a 96-well microplate. In the presen ce of the active enzyme, CPRG will change from yellow to red-purple. Otherwise, absence or redu ced color change indicates the inhibition caused by heavy metals. The degree of color change can be quantified by reading the optical density with a microplate spectrophotometer at 570nm. Th e percent inhibition was calculated according to Equation A-1. (Negative control absorbance Sample absorbance) %Inhibition = % (A-1) Negative control absorbance To determine the EC50 (the concentration that causes 50% inhibition of the test organisms in a given sample), we usually fi rst undertook a range-finding toxic ity test to determine the range of concentrations th at could cover the EC50 point, then 4 to 5 dilutions of the sample within this range were prepared prior to th e definitive test. Then, a regr ession analysis was performed by plotting the percent inhibitions pr oduced by these sample dilutions versus their concentrations, and the EC50 was calculated from the Equation A-2. 50Yintercept EC50 = (A-2) Slope

PAGE 179

179 Figure A-1. MetPLATETM protocol Add 5 ml of diluent to the bacterial reagent Mix the bacterial suspension Add 0.1 ml of bacterial suspension to 0.9 ml of sample in test tube, the mixture is vortexed Incubate the mixture at 35C for 90 min Transfer 0.2 ml from each tube to micro p late wells Add 0.1 ml of substrate to each well Incubate at 35C for color development (Yellowred-purple) Measure absorbance with m icro p late reader at 575n m

PAGE 180

180 A.2 48-h Ceriodaphnia dubia Acute Toxicity Test A.2.1 Preparation of culture medium and food Moderately hard water (MHW) was used as th e growth medium to culture the daphnids. The characteristics of MHW are summ arized in Table A-1. The food of Daphnia consisted of concentrated Selenastrum capricornutum suspension (3.5 107 cells/mL) and YCT (a mixture of yeast, cereal leaves and trout chow). The prepar ation of YCT started with a 7-day digestion of trout chow. Five gram of trout chow was adde d into 1 L of nanopure water, and the mixture was blended in a blender at low speed for 5 minutes. The trout chow so lution was then transferred to a digestion apparatus where conti nuous aeration was maintained by using a glass pipette attached to an aquaculture pump. After 7 days, the di gestion was complete and the digestate was transferred to a 1 L graduated cylinder and cove red with Parafilm. Then, the cylinder was placed in the refrigerator to settle for at least 1 hour. Meanwhile, on day 7 of the trout chow digestion, 5 g of cereal leaves was added in to 1 L of nanopure water in a blender, and the mixture was blended at low speed for 5 minutes. Then, the cer eal leaves suspension was transferred to a 1 L graduated cylinder, covered with Parafilm and placed in the refrigerator to settle for at least 1 hour. Simultaneously, on day 7 of the trout chow di gestion, 5 g of dry yeast were also added to 1 L of nanopure water, and the mixture was blen ded at low speed for 5 minutes. However, the yeast suspension was not allowed to settle and was combined immediately with trout chow and cereal leaves in the next step. Six hundred milliliter of each of YCT components (yeast suspension, trout chow and cereal leaves supernatant) were combin ed in a 2 L pitcher by sieving through a fine mesh (55 m). The mixture was mi xed thoroughly and transferred to a half gallon bottle labeled as Pre-YCT. Ten milliliter of th e Pre-YCT was placed in an aluminum weight dish and dried overnight at 70C, and the tota l suspended solids (TSS) content of the pre-YCT

PAGE 181

181 was then adjusted to 1800 mg/L by the additi on of nanopure water as needed. The YCT food was apportioned into 400 mL plastic bottles, labeled, and stored in the freezer (-40C) until use. Table A-1. Chemical parameters of moderately hard water (MHW) Parameters Amount NaHCO3 96 mg/L CaSO4 .2H2O 60 mg/L MgSO4 60 mg/L KCl 4 mg/L Hardness 80-100 (mg/L as CaCO3) Alkalinity 60-70 pH 7.4-7.8 A.2.2 Maintenance of Ceriodaphnia dubia cultures A starter culture of Ceriodaphnia dubia was donated by Hydr osphere Research (Gainesville, FL). The Daphnids were culture d in 1 L glassware containing 500 mL of MHW and kept in a PervicalTM environmental chamber (model # E-30 BX) at 25C with a day-night cycle of 16/8 h. The daphnids were fed every ot her day with equal volume of YCT (6.67 ml) and Selenastrum capricornutum suspension (6.67 ml) per liter of culture. Neonates (less than 24-hour old) were separated from adults daily and used for toxicity test or for starting new cultures. A.2.3 Test procedure The 48-h Ceriodaphnia dubia bioassay was carried out a ccording to the U.S. EPAs standard method (US EPA, 2002). To determine the EC50 for a given sample, five or more sample dilutions were prepared. MHW served as th e diluent as well as the negative control. All sample dilutions and negative control were prepar ed in triplicate. Neon ates (less than 24-hour old) were separated from adults and fed 2 hours prior to starting the te st. Five neonates were exposed in plastic cups containing 20 ml of sample dilution or nega tive control. The test containers were placed in the same environmen tal chamber (model # E-30 BX) at 25C with a

PAGE 182

182 day-night cycle of 16/8 h. After 48 hours, the test containers were placed on a light table and the number of motile and dead daphnids was record ed. The percent inhibition was calculated as followed: (# of motile in negative control # of motile in sample) %Inhibition = 100% (A-3) # of mo tile in negative control A.3 96-h Selenastrum capricornutum Chronic Toxicity Test A.3.1 Preparation of algal medium The preliminary algal assay pr ocedure (PAAP) medium was pr epared from eight nutrient stock solutions, including ma gnesium chloride (MgCl2H2O), calcium chloride (CaCl2H2O), sodium nitrate (NaNO3), magnesium sulfate (MgSO4H2O), potassium bicarbonate (K2HPO4), sodium bicarbonate (NaHCO3), disodium ethylenediaminetetr acetate (EDTA), and the trace salts solution (FDEP, 1997). Tabl e A-2 lists the concentration of each salt component. The PAAP medium was prepared by combining 1ml of each major salt solution with 1mL of the trace salt solution in a 1 L volumetric flask filled w ith nanopure water. The pH of the PAAP medium was adjusted to 7.5 0.1 with either 0.1 N NaOH or 0.1 N HCl. The medium was then filtered through a 0.45 m pre-sterilized filter and stored in the refrigerator before use. PAAP medium without EDTA was used for toxicity testing, while PAAP with EDTA was used for growing algae cultures. A.3.2 Maintenance of Selenastrum capricornutum cultures A pure culture of S. capricornutum was obtained from Hydrosphere Research (Gainesville, FL). The algae culture was grown in PAAP me dium (with EDTA) under controlled conditions (25C, 24 hours light source (400 40ft-c)). To maximize the li ght reflection and minimize the temperature change, black plastic sheeting was used to surround the light unit, and the interior of the sheeting was covered with aluminum foil. Be sides, continuous aeration was also provided by

PAGE 183

183 putting a 1 mL glass pipette in th e culture which was connected to an air pump. The culture flask was covered with Parafilm and shaken at least on ce per day. After 3-5 days, the algae cells were used for toxicity test; after 7 days, a small portio n of the mature algae culture was transferred to 2 L fresh medium to start a new culture. The surp lus 7-day old algae culture was concentrated by centrifugation and then resuspended in a small vol ume of distilled water. The density of this recovered algae solution was maintained at of 3.5107 cells/mL and served as Ceriodaphnia dubia food. Table A-2. Components of preliminary algal assay procedure (PAAP) medium Major salts Concentration (g/L) MgSO4H2O MgCl2H2O CaCl2H2O NaHCO3 NaNO3 K2HPO4 EDTA 14.700 12.164 4.410 7.500 25.500 0.818 0.300 Trace salts Concentration (mg/L) H3BO3 MnCl2 FeCl3H2O 185.520 415.380 159.760 ZnCl2 CoCl2H2O Na2MoO4H2O CuCl2H2O 3.270 1.428 7.260 0.012 Source: FDEP, 1997. A.3.3 Algal assay procedure The 96-h chronic Selenastrum capricornutum test was carried out according to the U.S. EPAs standard method (US EPA, 2002). To determine the EC50 of a given sample, 5 or more samp le dilutions were prepared. PAAP medium (without EDTA) was used as sample diluen t as well as negative control. All sample dilutions and negative control were prepared in triplicate. Fifty milliliter of each sample dilution was added to a 125 ml autoclaved Erlenmeyer flask with styrofoam stopper, and the sample

PAGE 184

184 solution was then inoculated with 1 mL of algae inoculum to reach a cell density of 5105 cells/mL. To prepare the inoculum, 10 ml of 35 days old algae suspension was centrifuged at 4,000 rpm for 15 minutes. The supernatant was dis carded and the algae seed was resuspended in 20 ml of PAAP medium (without EDTA). Then the density of the algae seed was determined with a hemacytometer under a microscope. Accordi ng to the density, the inoculum was prepared by diluting the seed with PAAP solution (without EDTA) to a final density of 5105 cells/mL. All inoculated flasks were then placed under the fluorescent lights and were shaken and rearranged at least once per day. After 96 hours exposure, the cell density in each flask was measured with a hemacytometer under a microsco pe. The percent inhibition was calculated by the following Equation: (Negative control cell density sample cell density) %Inhibition = 100% (A-4) Negative control cell density

PAGE 185

185 APPENDIX B DETAILED PROCEDURE FO R TOTAL METAL ANALYSIS B.1 U.S. EPA Method 3010A The U.S. EPA method 3010A is an acid dige stion procedure for aqueous samples and extracts. After digestion, the total metals in the sample can be analyzed by flame atomic absorption spectroscopy (FLAA) or inductively c oupled argon plasma spectroscopy (ICP). The following elements can be digested by this met hod: Al, As, Ba, Be, Cd, Ca, Cr, Co, Cu, Fe, Pb, Mn, Mg, Mo, Ni, K, Se, Na, TI, V, and Zn (US EPA, 1992). Fifty milliliter aliquot of the well-mixed sample was transferred to a 150 mL Griffin beaker and 3 mL of concentrated nitric acid (HNO3) was added, the beaker was covered with a ribbed watch glass and placed on hotplat e under the hood. The liquid was evaporated to a low volume (5 mL) without boiling. Then, the beaker was allo wed to cool, and another 3 mL of concentrated HNO3 was added. The beaker was covered with a watch glass and refluxed on the hotplate. Additional concentrated HNO3 was added and the system was refluxed continuously until the digestate was light in color or did not change in appearance with conti nued refluxing. The system was evaporated to 3 mL without boiling. The be aker was allowed to cool, and then 10 ml of 1:1(v/v, diluted in distilled wa ter) hydrochloric acid (HCl) was added. The beaker was covered with a watch glass and refluxed fo r an additional 15 minutes. The beaker was allowed to cool. The digestate was slowly filtered through a filter (Whatman 42 ashless) into a 50 mL volumetric flask. The beaker, watch glass, filter, and funnel we re rinsed with distilled water, and the solution was brought to 50 ml with distilled water. The sa mple was now ready for analysis by inductively coupled plasma spectrometry.

PAGE 186

186 B.2 U.S. EPA Method 3050B The U.S. EPA method 3050B is an acid diges tion procedure for solid samples including sediments, sludges, and soils. The following elements can be analyzed by flame atomic absorption spectroscopy (FLAA) or inductively coupled plasma atomic emission spectroscopy (ICP-AES) after digestion: Al, Sb, Ba, Be, Cd, Ca, Cr, Co, Cu, Fe, Pb, V, Mg, Mn, Mo, Ni, K, Ag, Na, TI, V, and Zn (US EPA, 1996). One to two gram of sample was weighed out in a 250 mL Erlenmeyer flask. Ten milliliter of 1:1(v/v, diluted in distil led water) nitric acid (HNO3) was added, and the flask was covered with a ribbed watch glass and placed on the hot plate under the hood. The system was refluxed for 10-15 minutes without boiling. The flask was removed from the hotplate and 5 mL of concentrated HNO3 was added. The system was refluxed again without boiling for 30 minutes. If brown fumes were generated, addi tional 5 mL of concentrated HNO3 was added until no more fumes were formed. The system was evaporated to 5 mL without boiling for a maximum of 2 hours. The flask was allowed to cool and 2 mL of distilled water and 3 mL of 30% Hydrogen Peroxide (H2O2) were added. The system was refl uxed and additional 1 mL of 30% H2O2 was added to a maximum of 10 mL until effervescence subsided. The system was refluxed until the sample volume was 5 mL. Then 10 mL of concentr ated hydrochloric acid (HCl) was added, and the system was refluxed again for 15 minutes. The flask was allowed to cool. The digestate was slowly filtered through a filter (Whatman 42 ashless) into a 100 mL volumetric flask. The Erlenmeyer flask, watch glass, fi lter, and funnel were rinsed with distilled water, and the solution was brought to 100 mL with dist illed water. The sample was now ready for analysis by inductively coupled plasma spectrometry.

PAGE 187

187 B.3 Total Mercury Determination Aqueous sample. The U.S. EPA method 1631 was used for determination of mercury (Hg) in filtered and non-filtered water by cold vapor atomic fluorescence spectroscopy (CVAFS). The method detection limit (MDL) for Hg has been determined to be 0.2 ng/L when no interferences are present (US EPA, 2002). Fi fty milliliter of sample containing 0.5-1% hydrochloric acid was oxidized in a 125 mL Teflon bottle by adding 0.5 mL of bromine monochloride (BrCl). The sample was then allowed to digest ove rnight. After oxidation, 0.2 mL of 30% hydroxylamine hydrochloride (300 g of NH2OHHCl was dissolved in 1000 mL of nanopure water) was added. The mixture was swirle d and allowed to react for at least 5 minutes. The sample was now ready for analysis by CV-AFS. Solid sample. The method to digest solid sample wa s adapted from Warner et al. (2003). Point five to one gram of sample was weighed in 50 mL Teflon hot block digestion tube. Twenty milliliter of a mixture of HNO3/H2SO4 (7:3, v: v) was added to the Teflon tube. The tube was capped loosely and the sample was allowed to dige st in a hot block at 100 C overnight. Then, the sample was allowed to cool, and another 30 mL of distilled water was added. The sample was now ready for analysis by CV-AFS. After digestion, all digestates were analy zed for total Hg by stannous chloride (SnCl2) reduction technique using a CV-AFS. B.4 Plant Digestion for Total Metal Analysis A wet acid digestion procedur e using nitric acid (HNO3) and 30% hydrogen peroxide (H2O2) was utilized for total metal analysis in plant tissue (Mills and Jones, 1996). Point five gram of dried (70C) and gr ound (20 mesh) plant tissue was weighed in a 150 mL Griffin beaker, and 8 mL of concentrated HNO3 was added. The sample was then allowed to digest overnight. On the next day, the beaker was cove red with a watch glass and heated for one hour

PAGE 188

188 on a hot plate at 120C. Then, the beaker was re moved from the hot plate and was allowed to cool. Four milliliter of 30% H2O2 was added, and the addition of 30% H2O2 was repeated until the digest was colorless. After the digestion wa s complete, the watch glass was removed and the residue was taken to dryness at 80 C. Once all the acid had been evaporated, the beaker was immediately taken off the hot plate and allowed to cool. The residue was then dissolved in 10 mL of 1:10 HNO3 (v/v). The sample was now ready for analysis by inductively coupled plasma spectrometry.

PAGE 189

189 APPENDIX C ADDITIONAL MATERIALS FOR PLANT STUDY Figure C-1. Phytotoxicity of 100 mg/kg Cu to lettuce (Lactuca sativa) in sandy soil after 4 weeks exposure Figure C-2. Phytotoxicity of 100 mg/kg Cu to Indian mustard (Brassica juncea) in sandy soil after 4 weeks exposure

PAGE 190

190 Figure C-3. Effect of iron trea tment on the growth of Lettuce (Lactuca sativa) in sandy soil. A1 and A2) front view. B) Top view. (The sandy soil was spiked with 100 mg/kg Cu before iron treatment. Lettuce could not survive in non-treated soil). Control W/O Iron Treated W/5% Iron IronRetrieved W/O Treatment W/O Treatment Control W/5% Iron Treated W/5% Iron IronImmobilized Treated W/2.5% Iron Iron Retrieved A 2 A1

PAGE 191

191 Figure C-3. Continued. Control W/5% Iron Treated W/5% Iron IronImmobilized Treated W/2.5% Iron Iron Retrieved W/O TreatmentB

PAGE 192

192 Figure C-4. Effect of iron treatment on the growth of Indian mustard (Brassica juncea) in sandy soil. A) Front view. B) Top view. (The sandy soil was spiked with 100 mg/kg Cu before iron treatment, plants grown in non iron-treated so il did not grow). Control W/5% Iron Control W/O Iron W/O Treatment Treated W/5% Iron IronImmobilized Treated W/5% Iron IronRetrieved B Control W/O Iron Treated W/ 5% Iron IronRetrieved W/O Treatment Treated W/ 2.5% Iron IronRetrieved A

PAGE 193

193 Figure C-5. Effect of iron trea tment on the growth of lettuce (Lactuca sativa) in organic rich soil 2. A) Front view. B) Top view. (The orga nic rich soil 2 was spiked with 500 mg/kg Cu before iron treatment). W/O Treatment Treated W/5% Iron IronImmobilized Treated W/5% Iron IronRetrieved Control W/5% Iron Control W/O IronA W/O Treatment Control W/O Iron Control W/5% Iron Treated W/5% Iron IronImmobilized Treated W/5% Iron IronRetriev ed B

PAGE 194

194 Figure C-6. Effect of iron treatment on the growth of Indian mustard (Brassica juncea) in organic rich soil 2. A) Front view. B) Top view. (The or ganic rich soil 2 was spiked with 500 mg/kg Cu before iron treatment). Treated W/5% Iron IronImmobilized Control W/5% Iron Control W/O Iron Treated W/5% Iron IronRetrieved W/O Treatment A Control W/OIron Treated W/5% Iron IronRetrieved W/O Treatment B Treated W/5% Iron IronImmobilized Control W/5% Iron

PAGE 195

195 Figure C-7. Effect of iron treatment on plant roots in organic rich soil 2. A) Lettuce (Lactuca sativa) roots. B) Indian Mustard (Brassica juncea) roots. (The organic rich soil 2 was spiked with 500 mg/kg Cu before iron treatment). Control W/5% Iron W/O Treatment Control W/O Iron Treated W/5% Iron IronRetrieved Treated W/5% Iron IronImmobilized A Control W/O Iron Control W/5% Iron Treated W/5% Iron IronRetrieved Treated W/5% Iron IronImmobilized W/O TreatmentB

PAGE 196

196 LIST OF REFERENCES Abdelouas, A., Lu, Y.M., Lutze, W., Nuttall, H.E., 1998. Reduction of U(VI) to U(IV) by indigenous bacteria in contam inated ground water. Journal of Contaminant Hydrology 35, 217-233. Abramovitch, R.A., Lu, C.Q., Hicks, E., Sina rd, J., 2003. In situ remediation of soils contaminated with toxic metal ions using microwave energy. Chemosphere 53, 1077-1085. Adriano, D.C., 2001a. Bioavailability of trace metals. In: Trace elements in terrestrial environments: biogeochemistry, bioava ilability, and risks of metals, 2nd ed. Springer, New York, pp. 62-89. Adriano, D.C., 2001b. Biogeochemical processes regulating metal behavior. In: Trace elements in terrestrial environments: biogeochemistry bioavailability, and risks of metals, 2nd ed. Springer, New York, pp. 30-59. Adriano, D.C., 2001c. Copper. In: Trace elements in terrestrial environments: biogeochemistry, bioavailability, and risks of metals, 2nd ed. Springer, New York, pp. 500-546. Adriano, D.C., 2001d. Lead. In: Trace elements in terrestrial environments: biogeochemistry, bioavailability, and risks of metals, 2nd ed. Springer, New York, pp. 350-410. Al-Asheh, S., Duvnjak, Z., 1999. Sorption of heavy meta ls from synthetic metal solutions and industrial wastewater using plan t materials. Water Quality Research Journal of Canada 34, 481-503. Alexander, M., 1995. How toxic are toxic ch emicals in soil. Environmental Science & Technology 29, 2713-2717. Ali, N.A., Ater, M., Sunahara, G.I, Robidoux, P.Y., 2004. Phytotoxicity and bioaccumulation of copper and chromium using barley (Hordeum vulgare L.) in spiked artificial and natural forest soils. Ecotoxicolo gy and Environmental Safety 57, 363-374. Altin, A., Degirmenci, M., 2005. Lead (II) removal from natural soils by enhanced electrokinetic remediation. Science of the Total Environment 337, 1-10. Alva, A.K., Huang, B., Paramasivam, S., 2000. Soil pH affects copper fractionation and phytotoxicity. Soil Science Society of America Journal 64, 955-962. Amorim, M.J.D., Rmbke, J., Schallnass, H.J., Mo rtagua, A., Soares, V.M., 2005. Effect of soil properties and aging on the toxicity of copper for Enchytraeus albidus, Enchytraeus luxuriosus, and Folsomia candida. Environmental Toxicology and Chemistry 24, 18751885. Amrate, S., Akretche, D.E., Innocent, C., Seta, P., 2005. Removal of Pb from a calcareous soil during EDTA-enhanced electrokinetic extracti on. Science of the Total Environment 349, 56-66.

PAGE 197

197 Anand, P., Etzel, J.E., Friedlaender, F.J., 1985. Heavy metals removal by high gradient magnetic separation. IEEE Transacti ons on Magnetics 21, 2062-2064. Anderson, C.R., Cook, G.M., 2004. Isolation and ch aracterization of arsena te-reducing bacteria from arsenic-contaminated sites in Ne w Zealand. Current Microbiology 48, 341-347. Anderson, N.J., Priestley, A.J., 1983. Color a nd turbidity removal with reusable magnetite particles. 5. process-developmen t. Water Research 17, 1227-1233. Andrade, M.D., Prasher, S.O., Hendershot, W. H., 2007. Optimizing the molarity of a EDTA washing solution for saturated-soil remedi ation of trace metal contaminated soils. Environmental Pollution 147, 781-790. Anthony, E.J., Wang, J.S., 2006. Pilot plant inve stigations of thermal remediation of tarcontaminated soil and oil-contaminated gravel. Fuel 85, 443-450. Antsuki, T., Sano, D., Omura, T., 2003. Functiona l metal-binding proteins by metal-stimulated bacteria for the development of an innova tive metal removal technology. Water Science and Technology 47, 109-115. Arias-Estvez, M., Nvoa-Muoz, J.C., Pateiro, M ., Lpez-Periago, E., 2007. Influence of aging on copper fractionation in an aci d soil. Soil Science 172, 225-232. Azcue, J.M., Zeman, A.J., Mudroch, A., Rosa, F., Patterson, T., 1998. Assessment of sediment and porewater after one year of subaqueous capping of contaminated sediments in Hamilton Harbour, Canada. Water Sc ience and Technology 37, 323-329. Baker, A.J.M., McGrath, S.P., Sidoli, C.M.D., Reeves, R.D., 1994. The possibility of in-situ heavy-metal decontamination of polluted soil s using crops of metal-accumulating plants. Resources Conservation and Recycling 11, 41-49. Balasoiu, C.F., Zagury, G.J., Deschnes, L., 2001. Partitioning and speciation of chromium, copper, and arsenic in CCA-contaminated soils : influence of soil composition. Science of the Total Environment 280, 239-255. Banuelos, G.S., Ajwa, H.A., Terry, N., Zayed, A., 1997. Phytoremediation of selenium laden soils: a new technology. Journal of So il and Water Conservation 52, 426-430. Barocsi, A., Csintalan, Z., Kocsanyi, L., Dushenkov, S., Kuperberg, J.M., Kucharski, R., Richter, P.I., 2003. Optimizing phytoremediation of hea vy metal-contaminated soil by exploiting plants stress adaptation. International Journal of Phytoremediation 5, 13-23. Basta, N.T., McGowen, S.L., 2004. Evaluation of chemical immobilization treatments for reducing heavy metal transport in a smelter-c ontaminated soil. Environmentsl Pollution 127, 73-82. Basta, N.T., Ryan, J.A., Chaney, R.L., 2005. Tr ace element chemistry in residual-treated soil: key concepts and metal bioavailability. J ournal of Environmental Quality 34, 49-63.

PAGE 198

198 Batley, G.E., Giles, M.S., 1979. Solvent displace ment of sediment interstitial waters before trace-metal analysis. Wa ter Research 13, 879-886. Bech, J., Poschenrieder, C., Lluga ny, M., Barcel, J., Tume, P., Tobias, F.J., Barranzuela, J.L., Vsquez, E.R., 1997. Arsenic and heavy meta l contamination of soil and vegetation around a copper mine in Northern Peru. Scien ce of the Total Envi ronment 203, 83-91. Begonia, M.T., Begonia, G.B., Butler, A.D., Griffin, U., Yong, C., 2003. Chemically-enhanced phytoextraction of cadmium-contaminated soils using wheat (Triticum aestivum L.). Bulletin of Environmental Contam ination and Toxicology 71, 648-654. Benjamin, M.M., Sletten, R.S., Bailey, R.P., Bene tt, T., 1996. Sorption and filtration of metals using iron-oxide-coated sand. Water Research 30, 2609-2620. Bitton, G., 1998. Formula Handbook for Environmen tal Engineers and Scientists. John Wiley and Sons, New York, pp. 256. Bitton, G., Fox, J.L., Strickland, H.G., 1975. Re moval of algae from Florida lakes by magnetic filtration. Applied Microbiology 30, 905-908. Bitton G., Jung, K., Koopman, B., 1994. Evalua tion of a microplate assay specific for heavy metal toxicity. Archives of Environmental Contamination and Toxicology 27, 25-28. Bitton, G., Mitchell, R., 1974. Removal of Escherichia-coli bacteriophage T7 by magnetic filtration. Water Reseach 8, 549-551. Bitton, G., Mitchell, R., DeLatour, C., Maxw ell, E., 1974. Phosphate removal by magnetic filtration. Water Research 8, 107-109. Bitton, G., Ward, M., Dagan, R., 2005. Determ ination of the heavy metal binding capacity (HMBC) of environmental samples. In: Blaise, C., Frard, J.F. (Eds), Small-scale freshwater environment toxic ity test Methods, Volume 2. Kluwer Academic Publishers, New York, pp. 215-231. Blais, J.F., Meunier, N., Merc ier, G., Drogui, P., Tyagi, R.D., 2004. Pilot plant study of simultaneous sewage sludge digestion and metal leaching. Journal of Environmental Engineering-ASCE 130, 516-525. Boisson, J., Ruttens, A., Mench, M, Vangronsve ld, J., 1999. Evaluation of hydroxyapatite as a metal immobilizing soil additive for the remediat ion of polluted soil. Part 1. influence of hydroxyapatite on metal exchangeability in soil, plant growth and plant metal accumulation. Environmental Pollution 104, 225-233. Boularbah, A., Morel, J.L., Bitton, G., Mench, M., 1996. A direct solid phase assay for heavy metal toxicity. 2. Assessment of heavy metal immobilization in soils and bioavailability to plants. Journal of Soil Contamination 5, 395-404.

PAGE 199

199 Boularbah, A., Schwartz, C., B itton, G., Morel, J.L., 2006. Heavy metal contamination from mining sites in south Morocco: 1. use of a biot est to assess metal toxicity of tailing and soils. Chemosphere 63, 802-810. Brown, S., Chaney, R. L., Hallfrisch, J.G., Xue, Q., 2003. Effect of bios olids processing on lead bioavailability in an urban soil. Journa l of Environmental Quality 32, 100-108. Brown, S., Chaney, R., Hallfrisch, J., Ryan, J.A ., Berti, W.R., 2004. In situ soil treatments to reduce the phytoand bioavailab ility of lead, zinc, and cadmi um. Journal of Environmental Quality 33, 522-531. Bryce, A.L., Kornlcker, W.A., Elzerman, A.W ., 1994. Nickel adsorption to hydrous ferric oxide in the presence of EDTA: effects of compone nt addition sequence. Environmental Science & Technology 28, 2353-2359. Buanuam, J., Shiowatana, J., Pongsakul, P., 2005. Fractionation and elemental association of Zn, Cd, and Pb in soils contaminated by Zn minings using a continuous-flow sequential extraction. Journal of Envir onmental Monitoring 7, 778-784. Bufflap, S.E., Allen, H.E., 1995. Sediment pore water collection methods for trace metal analysis: a review. Water Research 29, 165-177. Cajuste, L.J., Cruz-Diaz, J., Garcia-Osorio, C., 2000. Extraction of heavy metals from contaminated soils: I. sequentia l extraction in surface soils and their relationships to DTPA extractable metals and metal plant uptake. Journal of Environmental Science and Health Part AToxic/Hazardous substances & E nvironmental Engineering 35, 1141-1152. Calace, N., Campisi, T., Iacondini, A., Leoni, M ., Petronio, B.M., Pietroletti, M., 2005. Metalcontaminated soil remediation by means of paper mill sludges addition: chemical and ecotoxicological evaluation. Envi ronmental Pollution 136, 485-492. Calmano, W., Hong, J., Forstner, U., 1993. Bi nding and mobilization of heavy-metals in contaminated sediments affected by pH and redox potential. Water Science and Technology 28, 8-9. Cao, X.D., Ma, L.Q., Chen, M., Hardison, D.W., Harris, W.G., 2003. Weathe ring of lead bullets and their environmental effect s at outdoor shooting ranges. Journal of Environmental Quality 32, 526-534. Carlot, M., Giacomini, A., Casella, S., 2002. Asp ects of plant-microbe in teractions in heavy metal polluted soil. Acta Biotechnologica 22, 13-20. Castaldi, P., Santona, L., Melis, P., 2005. Heavy metal immobilization by chemical amendments in a polluted soil and influence on white lupin growth. Chemosphere 60, 365-371. Chaney, R.L., Malik, M., Li, Y.M., Brown, S.L., Brewer, E.P., Angle, J.S., Baker, A.J.M., 1997. Phytoremediation of soil metals. Curre nt Opinion in Biotechnology 8, 279-284.

PAGE 200

200 Chao, L., Zhou, Q., Chen, S., Cui, S., 2006. Speci ation distribution of l ead and zinc in soil profiles of the Shenyang smeltery in northeast China. Bulletin of Environmental Contamination and Toxicology 77, 874-881. Chen, H.M., Zheng, C.R., Tu, C., Shen, Z.G., 2000. Chemical methods and phytoremediation of soil contaminated with heavy metals. Chemosphere 41, 229-234. Chen, M., Daroub, S.H., 2002. Characterization of lead in soils of a rifle/pi stol shooting range in central Florida, USA. Soil and Se diment Contamination 11, 1-17. Chen, M., Ma, L.Q., Singh, S.P., Cao, R.X., Mela med, R., 2003. Field demonstration of in situ immobilization of soil Pb using P amendments. Advances in Environmental Research 8, 93-102. Chen, S.Y., Lin, J.G., 2001. Effect of subs trate concentration on bioleaching of metalcontaminated sediment. Journal of Hazardous Materials 82, 77-89. Chen, X.B., Wright, J.V., Conca, J.L., Peurr ung, L.M., 1997. Effects of pH on heavy metal sorption on mineral apatite. Environm ental Science & Technology 31, 624-631. Cheng, T., Allen, H.E., 2001. Prediction of uptake of copper from solution by lettuce (Lactuca sativa Romance). Environmental Toxicology and Chemistry 20, 2544-2551. Chiba, A., Okada, H., Tada, T., Kudo, H., Nak azawa, H., Mitsuhashi, K., Ohara, T., Wada, H., 2002. Removal of arsenic from geothermal water by high gradient magnetic separation. IEEE Transactions on Applie d Superconductivity 12, 952-954. Chirwa, E.N., Wang, Y.T., 2000. Simultaneous chromium(VI) reduction a nd phenol degradation in an anaerobic consortium of b acteria. Water Research 34, 2376-2384. Choel, M., Deboudt, K., Osan, J., Flament, P., Van Grieken, R., 2005. Quantitative determination of low-Z elements in single atmospheric particles on boron substrates by automated scanning electron microscopy Energy-dispersive X-ray spectrometry. Aanalytical chemistry 77, 5686-5692. Chun, C.L., Park, J.W., 2001. Oil spill remediat ion using magnetic separation. Journal of Environmental Engineering-ASCE 127, 443-449. Chung, H.I., Kang, B.H., 1999. Lead removal from contaminated marine clay by electrokinetic soil decontamination. Engineering Geology 53, 139-150. Clemens, S., 2001. Molecular mechanisms of plan t metal tolerance and homeostasis. Planta 212, 475-486. Cline, S.R., Reed, B.E., 1995. Lead removal from soils via bench-scale soil washing techniques. Journal of Environmental Engineering-ASCE 121, 700-705.

PAGE 201

201 Crawford, R.J., Harding, I.H., Mainwaring, D. E., 1993. Adsorption and c oprecipitation of single heavy-metal ions onto the hydrated oxides of iron and chromium. Langmuir 9, 3050-3056. Cui, Y.J., Zhu, Y.G., Zhai, R.H., Chen, D.Y., Hu ang, Y.Z., Qiu, Y., Liang, J.Z., 2004a. Transfer of metals from soil to vegetables in an area near a smelter in Nanning, China. Environmental International 30, 785-791. Cui, Y.S., Wang, Q.R., Dong, Y.T., Li, H.F., Chri stie, P., 2004b. Enhanced uptake of soil Pb and Zn by Indian mustard and winter wheat follo wing combined soil application of elemental sulphur and EDTA. Plant and Soil 261, 181-188. Cunningham, S.D., Berti, W.R., Huang, J.W., 1 995. Phytoremediation of contaminated soils. Trends in Biotechnology 13, 393-397. Daoust, C.M., Bastien, C., Deschnes, L., 2006. Influence of soil properties and aging on the toxicity of copper on compost worm and barl ey. Journal of Environmental Quality 35, 558567. Darling, C.T.R., Thomas, V.G., 2003. The distri bution of outdoor shooting ranges in Ontario and the potential for lead pollution of soil and wa ter. Science of the Total Environment 313, 235-243. Davis, R.D., Beckett, P.H.T., 1978. Upper critical levels of toxic el ements in plants. 2. critical levels of copper in young barley, wheat, rape, lettuce and ryegrass, a nd of nickel and zinc in young barley and ryegrass. New Phytologist 80, 23-32. DeLatour, C., 1973. Magnetic separation in wa ter-pollution control. IEEE Transactions on Magnetics 9, 314-316. DeLatour, C., Kolm, H., 1975.Magnetic separa tion in water pollution control.2. IEEE Transactions on Magnetics 11, 1570-1572. Denayer, F.O., Van Haluwyn, C., de Foucault, B., Schumacker, R., Colein, P., 1999. Use of bryological communities as a diagnostic tool of heavy metal soil contamination (Cd, Pb, Zn) in northern France. Plant Ecology 140, 191-201. De Vevey, E., Bitton, G., Rossel, D., Ramos, L.D., Munguia-Guerrero, L., Tarradellas, J., 1993. Concentration and bioavailability of heavy meta ls in sediments in Lake Yojoa (Honduras). Bulletin of Environmental Contamin ation and Toxicology 50, 253-259. Ebner, A.D., Ritter, J.A., Nuez, L., 1999. Highgradient magnetic separation for the treatment of high-level radioactive wastes. Sepa ration Science and Technology 34, 1333-1350. Edwards, R., Rebedea, I., Lepp, N.W., Lovell, A.J., 1999. An investigat ion into the mechanism by which synthetic zeolites reduce labile metal concentrations in soils. Environmental Geochemistry and Health 21, 157-173.

PAGE 202

202 Ehsan, S., Prasher, S.O., Marshall, W.D., 2006a. A washing procedure to mobilize mixed contaminants from soil: 1. Polychlorinated biphenyl compounds. Journal of Environmental Quality 35, 2146-2153. Ehsan, S., Prasher, S.O., Marshall, W.D., 2006b. A washing procedure to mobilize mixed contaminants from soil: 2. Heavy metals. Jo urnal of Environmenta l Quality 35, 2084-2091. El Bilali, L., Rasmussen, P.E., Hall, G.E.M., Fo rtin, D., 2002. Role of sediment composition in trace metal distribution in lake sedime nts. Applied Geochemistry 17, 1171-1181. Eighmy, T.T., Crannell, B.S., Butler, L.G., Cartle dge, F.K., Emery, E.F., Oblas, D., Krzanowski, J.E., Eusden, J.D., Shaw, E.L., Francis, C.A., 1997. Heavy metal stabilization in municipal solid waste combustion dry scrubber residue using soluble phosphate. Environmental Science & Technology 31, 3330-3338. Esmadi, F., Simm, J., 1995. Sorption of cobalt (I I) by amorphous ferric hy droxide. Colloids and Surfaces A-Physicochemical and Engineering Aspects 104, 265-270. Faisal, M., Hameed, A., Hasnain, S., 2005. Ch romium-resistant bacter ia and cyanobacteria: impact on Cr(VI) reduction potential and plant growth. Journal of I ndustrial Microbiology & Biotechnology 32, 615-621. Farfel, M.R., Orlova, A.O., Chaney, R.L., Lees, P.S.J., Rohde, C., Ashley, P.J., 2005. Biosolids compost amendment for reducing soil lead hazards: a pilot study of Orgro amendment and grass seeding in urban yards. Scien ce of the Total Envi ronment 340, 81-95. Fargasova, A., 2001. Effect of Cd in combinati on with Cu, Zn, Pb and Fe on root prolongation and metal accumulation in the roots and cotyledones of mustard (Sinapis alba) seedlings. Rostlinna Vyroba 47, 97-103. Fjallborg, B., Dave, G., 2004. Toxicity of Sb a nd Cu in sewage sludge to terrestrial plants (lettuce, oat, radish), and of sludge elutriate to aquatic organisms (Daphnia and Lemna) and its interaction. Water Air and Soil Pollution 155, 3-20. FDEP, 1997. Selenastrum capricomutum chronic toxicity testing me thods. Florida Department of Environmental Protection, Tallahassee, Florida. FDEP, 2005. Soil Cleanup Target Levels. Division of Waste Management, Florida Department of Environmental Protection, Tallahassee, Florida. Frana, S., Vinagre, C., Caador, I., Cabral, H.N ., 2005. Heavy metal concen trations in sediment, benthic invertebrates and fish in three salt marsh areas subjected to different pollution loads in the Tagus Estuary (Portugal) Marine Pollution Bulletin 50, 993-1018. Francis, B.M., 1994. Hazardous waste disposal. In : Toxic Substances in the Environment. John Wiley & Sons, New York, pp.140.

PAGE 203

203 Franzreb, M., Hll, W.H., 2000. Phosphate remo val by high-gradient ma gnetic filtration using permanent magnets. IEEE Transactions on Applied Superconductivity 10, 923-926. Friesl, W., Lombi, E., Horak, O., Wenzel, W., 2003. Immobilization of heavy metals in soils using inorganic amendments in a greenhouse st udy. Journal of Plant Nutrition and Soil Science 166, 191-196. Gadd, G.M., 2004.Microbial influence on metal m obility and application for bioremediation. Geoderma 122, 109-119. Ganesh, R., Robinson, K.G., Reed, G.D., Sayler G.S., 1997. Reduction of hexavalent uranium from organic complexes by sulfatea nd iron-reducing bacteria. Applied and Environmental Microbiology 63, 4385-4391. Garbisu, C., Alkorta, I., 2001. Phytoextraction: a cost-effective plant-based technology for the removal of metals from the environm ent. Bioresource Technology 77, 229-236. Gokon, N., Shimada, A., Hasegawa, N., Kaneko, H., Kitamura, M., Tamaura, Y., 2002. Ferrimagnetic coagulation process for phosphate ion removal using high-gradient magnetic separation. Separation Scie nce and Technology 37, 3781-3791. Grasso, D., Butkus, M.A., OSullivan, D., Nikolaidis, N.P., 1997. Soil-washing design methodology for a lead-contaminated sandysoil. Water Research 31, 3045-3056. Gray, C.W., Dunham, S.J., Dennis, P.G., Zhao, F.J., McGrath, S.P., 2006. Field evaluation of in situ remediation of a heavy metal contam inated soil using lime and red-mud. Environmental Pollution 142, 530-539 Hamby, D.M., 1996. Site remediation techniques supporting environmental restoration activitiesa review. Science of the To tal Environment 191, 203-224. Hardison, D.W., Ma, L.Q., Luongo, T., Harris, W.G., 2004. Lead contamination in shooting range soils from abrasion of lead bullets and subsequent weathering. Science of the Total Environment 328, 175-183. Harjanto, S., Kasai, E., Terui, T., Nakamura, T ., 2002. Formation and transport of PCDD/Fs in the packed bed of soil containing organic chlo ride during a thermal remediation process. Chemosphere 49, 217-224. Hartley, W., Edwards, R., Lepp, N.W., 2004. Ar senic and heavy metal mobility in iron oxideamended contaminated soils as evaluated by shortand long-term leaching tests. Environmental Pollution 131, 495504. Hatzinger, P.B., Alexander, M., 1995. Effect of aging of chemicals in soil on their biodegradability and extractability. Environmental Sc ience & Technology 29, 537-545.

PAGE 204

204 Heckmann, R.A., Amin, O.M., Standing, M.D., 2007. Chemical analysis of metals in acanthocephalans using energy dispersive x-ray analysis (EDXA) in conjunction with a scanning electron microscope (SEM). Comparative Parasitology 74, 388-391. Helsen, L., Van den Bulck, E., 1998. The microdistribution of copper, chromium and arsenic in CCA treated wood and its pyrolysis residue us ing energy dispersive X-ray analysis in scanning electron microscopy. Holzforschung 52, 607-614. Hettiarachchi, G.M., Pierzynski, G.M., 2002. In si tu stabilization of soil lead using phosphorous and manganese oxide: influence of plant grow th. Journal of Environmental Quality 31, 564-572. Hettiarachchi, G.M., Pierzynski, G.M., 2004. Soil lead bioavail ability and in situ remediation of lead-contaminated soils: a review. Environmental Progress 23, 78-93. Hettiarachchi, G.M., Pierzynski, G.M., Ransom, M. D., 2000. In situ stabil ization of soil lead using phosphorous and manganese oxide. E nvironmental Science & Technology 34, 46144619. Hong, K.J., Tokunaga, S., Kajiuchi, T., 2002. Ev aluation of remediation process with plantderived biosurfactant for recovery of heavy me tals from contaminated soils. Chemosphere 49, 379-387. Horiguchi, T., 1987. Mechanism of manganese toxici ty and tolerance of plants. 2. deposition of oxidized manganese in plant-tissues. Soil Science and Plant Nutrition 33, 595-606. Hovsepyan, A., Greipsson, S., 2005. EDTA-enha nced phytoremediation of lead-contaminated soil by corn. Journal of Plant Nutrition 28, 2037-2048. Huang, F., Bitton, G., Kong, I.C., 1999. Determ ination of the heavy metal binding capacity of aquatic samples using MetPLATE: a preliminar y study. Science of th e Total Environment 234,139-145. Ike, M., Takahashi, K., Fujita, T., Kashiwa, M., Fujita, M., 2000. Selenate reduction by bacteria isolated from aquatic environment free from selenium contamination. Water Research 34, 3019-3025. Illera, V., Garrido, F., Serrano, S., Garca-Gonz lez, M.T., 2004.Immobilization of the heavy metals Cd, Cu and Pb in an acid soil am ended with gypsumand lime-rich industrial byproducts. European Journal of Soil Science 55, 135-145. Isoyama, M., Wada, S.I., 2007. Remediation of Pb-contaminated soils by washing with hydrochloric acid and subsequent immobilizatio n with calcite and allophonic soil. Journal of Hazardous Material 143, 636-642. Jenne, E.A., 1995. Metal adsorption onto and desorpti on from sediments: 1.rates. In: Allen, H.E. (Ed), Metal contaminated aquatic sediment s. Ann Arbor Press, Chelsea, Michigan.

PAGE 205

205 Jordao, C.P., Fialho, L.L., Cecon, P.R., Matos, A.T., Neves, J.C.L., Mendonca, E.S., Fontes, R.L.F., 2006. Effects of Cu, Ni and Zn on le ttuce grown in metalenriched vermicompost amended soil. Water Air and Soil Pollution 172, 21-38. Kambhampati, M.S., Begonia, G.B., Begonia, M. F.T., Bufford, Y., 2003. Phytoremediation of a lead-contaminated soil using morning glory (Ipomoea lacunose L.): effects of a synthetic chelate. Bulletin of Environmental C ontamination and Toxicology 71, 379-386. Kaminski, M.D., Nuez, L., Pourfarzaneh, M ., Negri, C., 2000. Cesium separation from contaminated milk using magnetic particle s containing crystallin e silicotitanates. Separation and Purification Technology 21, 1-8. Karapinar, N., 2003. Magnetic sepa ration: an alternative method to the treatment of wastewater. The European Journal of Mineral Processi ng and Environmental Protection 3, 215-223. Karapinar, N., Hoffmann, E., Hahn, H.H., 2004. Magnetic seeded precipitation of phosphate. Water Research 38, 3059-3066. Kasai, E., Harjanto, S., Terui, T., Nakamura T., Waseda, Y., 2000. Thermal remediation of PCDD/Fs contaminated soil by zone com bustion process. Chemosphere 41, 857-864. Kim, C.S., Rytuba, J.J., Brown, G.E., 2004a. E XAFS study of mercury(II) sorption to Feand Al-(hydr)oxides.1. effects of pH. Journal of Colloid and Interface Science 271, 1-15. Kim, C.S., Rytuba, J.J., Brown, G.E., 2004b. EX AFS study of mercury(II) sorption to Feand Al-(hydr)oxides.2. effects of chloride and sulf ate. Journal of Colloid and Interface Science 270, 9-20. Kim, S.O., Kim, K.W., Stben, D., 2002. Evaluati on of electrokinetic removal of heavy metals from tailing soils. Journal of Environm ental Engineering-ASCE 128, 705-715. Kishk, F.M., Hassan, M.N., Ghanem, I., Elsissy, L., 1973. Status of copper in some calcareous and non-calcareous soils of E gypt. Plant and Soil 39, 487-496. Kochen, R.L., Navratil, J.D., 1997. Removal of radioactive materials an d heavy metals from water using magnetic resin. United States Patent # 5595666. Kong, I.C., Bitton, G., 2003. Co rrelation between heavy metal toxi city and metal fractions of contaminated soils in Korea. Bulletin of Environmental Contamination and Toxicology 70, 557-565. Kornmuller, A., Karcher, S., Jekel, M., 2002. Adsorption of r eactive dyes to granulated iron hydroxide and its oxidative regeneration. Water Science and Technology 46, 43-50. Kos, B., Letan, D., 2003. Induced phytoextrac tion/soil washing of lead using biodegradable chelate and permeable barriers. Envir onmental Science & Technology 37, 647-629.

PAGE 206

206 Krauss, M., Wilcke, W., Kobza, J., Zech, W., 2002. Predicting heavy metal transfer from soil to plant: potential use of Freundlich-type functions. Journal of Plant Nutrition and Soil Science 165, 3-8. Kumpiene, J., Ore, S., Lagerkvist, A., Maur ice, C., 2007. Stabilization of Pband Cucontaminated soil using fly ash and peat Environmental Pollution 145, 365-373. Kuo, S., Heilman, P.E., Baker, A.S., 1983. Dist ribution and forms of copper, zinc, cadmium, iron, and manganese in soils near a copper smelter. Soil Science 135, 101-109. Lee, S., Anderson, P.R., 2005.EXAFS study of Zn sorption mechanisms on hydrous ferric oxide over extended reaction time. Journal of Colloid and Interfac e Science 286, 82-89. Letan, D., Grcman, H., Zupan, M., Bacac, N ., 2003. Relationship of soil properties to fractionation of Pb and Zn in soil and their uptake into plantago lanceolata. Soil and Sediment contamination 12, 507-522. Leuser, R.M., Velazquez, L.A., Cohen, A., Janssen, J., 1990. Remediation of PCB soil contamination by on-site incineration. J ournal of Hazardous Materials 25, 375-385. Lewis, D.M., Yan, G., Julia, R.M., Coderch, L., Erra, P., 2000. Chromium distribution in wool by electron microscopy and x-ray energy dispersive analysis. Te xtile Research Journal 70, 315-320. Li, X.D., Coles, B.J., Ramsey, M.H., Thornton, I., 1995. Sequential ex traction of soils for multielement analysis by ICP-AES. Chemical Geology 124, 109-123. Lim, J.M., Salido, A.L., Butcher, D.J., 2004. P hytoremediation of lead using Indian mustard (Brassica juncea) with EDTA and electrodics. Microchemical Journal 76, 3-9. Lin, H.K., Man, X.D., Walsh, D.E., 2001. Lead removal via soil washing and leaching. JOMJournal of the Minerals Metals & Materials Society 53, 22-25. Lin, J.G., Chen, S.Y., Su, C.R., 2003. Assessment of sediment toxicity by metal speciation in different particle-size fractions of river sediment. Water Science and Technology 47, 233241. Lock, K., Janssens, C.R., 2001. Effect of clay and organic matter type on th e ecotoxicity of zinc and cadmium to the potworm Enchytraeus albidus. Chemosphere 44, 1669-1672. Lock, K., Janssens, C.R., 2003. Influence of ageing on zinc bioavailability in soils. Environmental Pollution 126, 371-374. Lombi, E., Zhao, F.J., Zhang, G.Y., Sun, B., F itz, W., Zhang, H., McGrath, S.P., 2002a. In situ fixation of metals in soils using bauxite re sidue: chemical assessment. Environmental Pollution 118, 435-443.

PAGE 207

207 Lombi, E., Zhao, F.J., Wieshammer, G., Zhang, G.Y., McGrath, S.P., 2002b. In situ fixation of metals in soils using bauxite residue: biol ogical effects. Enviro nmental Pollution 118, 445452. Lser, C., Zehnsdorf, A., Grsch, K., Seidel, H., 2005. Bioleaching of heavy metal polluted sediment: kinetics of leaching a nd microbial sulfur oxi dation. Engineering in Life Sciences 5, 535-549. Lser, C., Zehnsdorf, A., Grsch, K., Seidel, H., 2006. Remediation of heavy metal polluted sediment in the solid bed: comparison of ab iotic and microbial leaching. Chemosphere 65, 9-16. Lser, C., Zehnsdorf, A., Hoffmann, P., Seidel H., 2007. Remediation of heavy metal polluted sediment by suspension and solid-bed leaching: estimate of metal removal efficiency. Chemosphere 66, 1699-1705. Luli, G.W., Talnagi, J.W., Strohl, W.R., Pfis ter, R.M., 1983. Hexavalent chromium-resistant bacteria isolated from river sediments. Applied and Environmental Microbiology 46, 846854. Luo, Y.M., Rimmer, D.L., 1995. Zinc-copper inte raction affecting plant growth on a metalcontaminated soil. Environmental Pollution 88, 79-83. Luther, G.W., 1995. Trace metal chemistry in porewaters. In: Allen, H.E. (Ed), Metal contaminated aquatic sediments. An n Arbor Press, Chelsea, Michigan. Ma, L.Q., Rao, G.N., 1997. Chemical fractiona tion of cadmium, copper, nickel, and zinc in contaminated soils. Journal of Environmental Quality 26, 259-264. Ma, L.Q., Komar, K.M., Tu, C., Zhang, W.H., Cai, Y., Kennelley, E.D., 2001. A fern that hyperaccumulates arsenic. Nature 409, 579. Ma, Q.Y., Logan, T.J., Traina, S.J., 1995. L ead immobilization from aqueous solutions and contaminated soils using phosphate rocks. Environmental Science & Technology 29, 11181126. Ma, Y.B., Lombi, E., Oliver, I.W., Nolan, A. L., McLaughlin, M.J., 2006a. Long-term aging of copper added to soils. Environmen tal Science & Technology 40, 6310-6317. Ma, Y.B., Lombi, E., Nolan, A.L., McLaughlin, M.J., 2006b. Short-term natural attenuation of copper in soils: effects of time, temperatur e, and soil characteri stics. Environmental Toxicology and Chemistry 25, 652-658. Ma, Y.B., Uren, N.C., 2006. Effect of aging on the av ailability of zinc adde d to a calcareous clay soil. Nutrient Cycling in Agroecosystems 76, 11-18. Macek, F., Navratil, J.D., Dulanska, S., 2002. Magnetic sorbent for soil remediation-a waste for waste treatment. Separation Science and Technology 37, 3673-3692.

PAGE 208

208 Madejn, P., Maran, T., Murillo, J.M., Robinson, B., 2004. White poplar (Populus alba) as a biomonitor of trace elements in contaminated riparian forests. Environmental Pollution 132, 145-155. Maenpaa, K.A., Kukkonen, J.V.K., Lydy, M.J., 2002. Remediation of heavy metal-contaminated soils using phosphorous: evalua tion of bioavailability usi ng an earthworm bioassay. Archives of Environmental Contam ination and Toxicology 43, 389-398. Martnez, C.E., Jacobson, A.R., McBride, M.B ., 2003. Aging and temperature effects on DOC and elemental release from a metal contam inated soil. Environmental Pollution 122, 135143. McCready, S., Birch, G.F., Taylor, S.E., 2003. Extraction of heavy metals in Sydney Harbour sediments using 1M HCl and 0.05M EDTA and implications for sediment-quality guidelines. Australian Journal of Earth Sciences 50, 249-255. McGowen, S.L., Basta, N.T., Brown, G.O., 2001. Use of diammonium phosphate to reduce heavy metal solubility and transport in smelter-contaminated soil. Journal of Environmental Quality 30, 493-500. Mckenzie, R.M., 1980. The adsorption of lead and other heavy metals on oxides of manganese and iron. Australian Journal of Soil Research 18, 61-73. Meers, E., Lesage,E., Lamsal, S., Hopgood, M., Vervaeke, P., Tack, F.M.G., Verloo, M.G., 2005. Enhanced phytoextraction: 1. Effect of EDTA and citric acid on heavy metal mobility in a calcareous soil. Internationa l Journal of Phytoremediation 7, 129-142. Megharaj, M., Avudainayagam, S., Naidu, R., 2003. Toxicity of hexavalent chromium and its reduction by bacteria isolated from soil co ntaminated with tannery waste. Current Microbiology 47, 51-51. Meharg, A.A., 2005. Mechanism of plant resistan ce to metal and metalloid ions and potential biotechnological applications Plant and Soil 274, 163-174. Melamed, R., Cao, X.D., Chen, M., Ma, L.Q., 2003. Field assessment of lead immobilization in a contaminated soil after phospha te application. The Science of the Total Environment 305, 117-127. Melville, D., Paul, F., Roath, S., 1975. High gr adient magnetic separation of red cells from whole blood. IEEE Transacti ons on Magnetics 11, 1701-1704. Melville, D., Paul, F., Roath, S., 1982. Fracti onation of blood components using high gradient magnetic separation. IEEE Transactions on Magnetics 18, 1680-1685. Mench, M.J., Didier, V.L., Lffler, M., Gomez, A., Masson, P., 1994. A mimicked in-situ remediation study of metal-contaminated soils with emphasis on cadmium and lead, Journal of Environmental Quality 23, 58-63.

PAGE 209

209 Mendil, D., Uluzl, .D., 2007. Determination of trace metal levels in sediment and five fish species from lakes in Tokat, Turkey. Food Chemistry 101, 739-745. Meng, X.G., Letterman, R.D., 1991. Immobilization of cadmium in sediment by treatment with an aluminum salt and freeze-thaw dehydrat ion. Colloids and Surfaces 54, 219-233. Merino, J., Bucal, V., 2007. Effect of temper ature on the release of hexadecane from soil by thermal treatment. Journal of Hazardous Materials 143, 455-461. Merry, R.H., Tiller, K.G., Alston, A.M., 1983. Accu mulation of copper, lead and arsenic in some Australian orchard soils. Australian Journal of Soil Research 21, 549-561. Mertens, J., Luyssaert, S., Verheyen, K., 2005. Use and abuse of trace metal concentrations in plant tissue for biomonitoring and phytoextraction. Envir onmental Pollution 138, 1-4. Mills, H.A., Jones, J.B., 1996. Preparation and analysis. In: Plant analysis handbook II: a practical sampling, preparation, analysis and interpretation guide. Micro-Macro Publishing, Athens, Georgia, pp.122. Morel, J.L., 1997. Bioavailability of trace elements to terrestrial plants. In: Tarradellas, J., Bitton, G., Rossel, D. (Eds), Soil Ecotoxicology. Le wis Publishers, CRC Press, Boca Raton, FL, pp. 145-147. Mukherji, S., Gupta, B.D., 1972. Characterizat ion of copper toxicity in lettuce seedlings. Physiologia Plantarum 27, 126-129. Mller, I., Pluquet, E., 1998. Immobilization of heavy me tals in sediment dredged from a seaport by iron bearing materials. Water Science and Technology 37, 379-386. Mulligan, C.N., Wang, S., 2006. Remediation of a heavy metal-contaminated soil by a rhamnolipid foam. Engineering Geology 85, 75-81. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001a. Remediation technologies for metalcontaminated soils and groundwater: an ev aluation. Engineering Geology 60, 193-207. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001b. An evaluation of technologies for the heavy metal remediation of dredged sediments. Journal of Hazardous Materials 85, 145-163. Naidu, R., Oliver, D., McConnell, S., 2003. Hea vy metal phytotoxicity in soils. In: Langley, A., Gilbey, M., Kennedy, B. (Eds), Proceedi ngs of the fifth national workshop on the assessment of site contamination. Nationa l Environment Protection Council, Adelaide, Australia, pp.235-241. Naidu, R., Rogers, S., Gupta, V.V.S.R., Kooka na, R.S., Bolan, N.S., Adriano, D.C., 2003b. Bioavailability of metals in the soil plant environment and its potential role in risk assessment. In: Naidu, R., Gupta, V.V.S.R., Rogers, S., Kookana, R.S., Bolan, N.S., Adriano, D.C. (Eds), Bioavailability, toxicity and risk relationships in ecosystems. Science Publishers, Enfield, NH, pp. 21-57.

PAGE 210

210 Nakamura, K., Hagimine, M., Sakai, M., Fur ukawa, K., 1999. Removal of mercury from mercury-contaminated sediments using a co mbined method of chemical leaching and volatilization of mercury by bacter ia. Biodegradation 10, 443-447. Namasivayam, C., Ranganathan, K., 1995. Remo val of Pb(II), Cd(II), Ni(II) and mixture of metal-ions by adsorption onto waste Fe(II I)/Cr(III) hydroxide and fixed-bed studies. Environmental Technology 16, 851-860. Namasivayam, C., Senthilkumar, S., 1999. Adsorp tion of copper (II) by waste Fe(III)/Cr(III) hydroxide from aqueous soluti on and radiator manufactur ing industry wastewater. Separation Science and Technology 34, 201-217. Navratil, J.D., Tsair, M.T.S., 2002. Magnetic se paration of iron and hea vy metals from water. Water Science and Technology 47, 29-32. Nayak, B., Hossain, M.A., Sengupta, M.K., Ahamed S., Das, B., Pal, A., Mukherjee, A., 2006. Adsorption studies with arsenic onto ferric hydroxide gel in a nonoxidizing environment: the effect of co-occurring solutes and speci ation. Water Quality Research Journal of Canada 41, 333-340. Nedelkoska, T.V., Doran, P.M., 2000. Character istics of heavy metal uptake by plant species with potential for phytoremediation and phytom ining. Minerals Engineering13, 549-561. Neilson, J.W., Artiola, J.F., Maier, R.M., 2003. Characterization of lead removal from contaminated soils by nontoxic soil-washing ag ents. Journal of Envi ronmental Quality 32, 899-908. Nicholls, A.M., Mal, T.K., 2003. Effects of lead and copper exposure on growth of an invasive weed, Lythrum salicaria L. (purple loosestrife). Ohio Journal of Science 103, 129-133. Notten, M.J.M., Oosthoek, A.J.P., Rozema, J., Ae rts, R.R., 2005. Heavy metal concentrations in a soil-plant-snail food chain along a terrestr ial soil pollution gradient. Environmental Pollution 138, 178-190. Nvoa-Muoz, J.C., Queijeiro, J. M.G., Blanco-Ward, D., lvare z-Olleros, C., Garca-Rodeja, E., Martnez-Cortizas, A., 2007. Arsenic fractionation in agricu ltural acid soils from NW Spain using a sequential extraction procedure. Science of the Tota l Environment 378, 1822. Nowierski, M., Dixon, D.G., Borgmann, U., 2 006. Lac Dufault sediment core trace metal distribution, bioavailab ility and toxicity to Hyalella azteca. Environmental Pollution 139, 532-540. Nystroem, G.M., Ottosen, L.M., Villumsen, A., 2005. Acidification of harbor sediment and removal of heavy metals induced by water sp litting in electrodialytic remediation. Separation Science and Technology 40, 2245-2264.

PAGE 211

211 Nystroem, G.M., Pedersen, A.J., Ottosen, L.M., Villumsen, A., 2006. The use of desorbing agents in electrodialytic remediation of harbour sediment. Science of the Total Environment 357, 25-37. OConnor, C.S., Lepp, N.W., Edwards, R., S underland, G., 2003. The combined use of electrokinetic remediation and phytoremediation to decontaminate metal-polluted soils: a laboratory-scale feasibility st udy. Environmental Monitoring and Assessment 84, 141-158. Oorts, K., Ghesquiere, U., Smolders, E., 2007. L eaching and aging decrease nickel toxicity to soil microbial processes in soils freshly sp iked with nickel chloride. Environmental Toxicology and Chemistry 26, 1130-1138. Oremland, R.S., Blum, J.S., Bindi, A.B., Do wdle, P.R., Herbel, M., 1999. Simultaneous reduction of nitrate and selenate by cell su spensions of selenium -respiring bacteria. Applied and Environmental Microbiology 65, 4385-4392. Orbell, J.D., Godhino, L., Bigger, S.W., Nguyen, T.M., Ngeh, L.N., 1997. Oil spill remediation using magnetic particles. Jour nal of Chemical Education 74, 1446-1448. Oste, L.A., Lexmond, T.M., Riemsdijk, W.H.V., 2002. Metal immobiliza tion in soils using synthetic zeolites. Journal of E nvironmental Quality 31, 813-821. Ottosen, L.M., Hansen, H.K., Laursen, S., Villu msen, A., 1997. Electrodialytic remediation of soil polluted with copper from wood preserva tion industry. Environmental Science & Technology 31, 1711-1715. Ottosen, L.M., Lepkova, K., Kubal, M. 2006. Comparison of elect rodialytic removal of Cu from spiked kaolinite, spiked soil and industrially polluted soil. Journal of Hazardous Materials 137, 113-120. Ottosen, L.M., Nystrom, G.M., Jensen, P.E., Villumsen, A., 2007. Electrodialytic extraction of Cd and Cu from sediment from Sisimiut Harbour, Greenland. Journal of Hazardous Materials 140, 271-279. Otulakowska, J., Nicholson, J.W., 2006. Scanning electr on microscopy and energy dispersive Xray study of a recovered dental implant. Journal of Mate rials Science-Materials in Medicine 17, 277-279. Ouddane, B., Abbasse, G., Halwani, J., Fischer, J.C., 2004. Determination of metal partitioning in porewater extracted from the Seine rive r estuary sediment (France). Journal of Environmental Monitoring 6, 243-253. Page, M.M., Page, C.L., 2002. Electroremedia tion of contaminated soils. Journal of Environmental Engineering-ASCE 128, 208-219. Pardo, R., Barrado, E., Perez, L., Vega, M., 1990. Determination and speciation of heavy-metals in sediments of the Pisuerga ri ver. Water Research 24, 373-379.

PAGE 212

212 Pardue, J.H., Patrick, W.H., 1995. Changes in me tal speciation following alternation of sediment redox status. In: Allen, H.E. (Ed), Metal contam inated aquatic sediments. Ann Arbor Press, Chelsea, Michigan. Parida, K.M., Gorai, B., Das, N.N., Rao, S. B., 1997. Studies on ferri c oxide hydroxides. 3. adsorption of selenite (SeO3 2-) on different forms of iron oxyhydroxides. Journal of Colloid and Interface Science 185, 355-362. Park, G., Shin, H.S., Ko, S.O., 2005. A laborator y and pilot study of thermally enhanced soil vapor extraction method for the removal of semi -volatile organic contaminants. Journal of Environmental Science and Health 40, 881-897. Park, J.S., Chang, Y.Y., Bae, B., Kim, O.K., Cho, K.S., Lee, I.S., 2003. Low heavy metal bioavailability in soil at c ontaminated Korean shooting sites. Journal of Environmental Science and Health Part A-T oxic/Hazardous Substances & E nvironmental Engineering 38, 1285-1297. Park, S.K., Bielefeldt, A.R., 2005. Non-ionic surfactant flushi ng of pentachlorophenol from NAPL-contaminated soil. Water Research 39, 1388-1396. Parker, M.R., 1977. The physics of magnetic se paration. Contemporary Physics 18, 279-306. Pedersen, M.B., Elmegaard, C.K.N., 2000. Toxicity and bioaccumulation of copper to black bindweed (Fallopia convolvulus) in relation to bioavailability and the age of soil contamination. Archives of Environmenta l Contamination and Toxicology 39, 431-439. Pedersen, M.B., van Gestel, C.A.M., 2001. Toxicity of copper to the collembolan Folsomia fimetaria in relation to the age of soil contamination. Ecotoxicology and Environmental Safety 49, 54-59. Peng, X.J., Luan, Z.K., Zhang, H.M., 2006. Montmorillonite -Cu(II)/Fe(III) oxides magnetic material as adsorbent for removal of humic acid and its thermal regeneration. Chemosphere 63, 300-306. Petrakis, L., Ahner, P.F., 1976.Use of high gr adient magnetic separa tion techniques for the removal of oil and solids from water efflue nts. IEEE Transactions on Magnetics 12, 486488. Petry, R., Mastalerz, R., Zahn, S., Mayerhofer, T.G., Volksch, G., Viereck-Gotte, L., KreherHartmann, B., Holz, L., Lankers, M., Popp, J., 2006. Asbestos mineral analysis by UV Raman and energy-dispersive X-ray sp ectroscopy. Chemphyschem 7, 414-420. Pich, A., Manteuffel, R., Hillmer, S., Scholz, G ., Schmidt, W., 2001. Fe homeostasis in plant cells: does nicotianamine play multiple roles in the regulation of cytoplasmic Fe concentration. Planta 213, 967-976.

PAGE 213

213 Planquart, P., Bonin, G., Prone, A., Massiani, C., 1999. Distribution, movement and plant availability of trace metals in soils amended with sewage sludge composts: application to low metal loadings. Science of th e Total Environment 241, 161-179. Poulsen, T.G., Moldrup, P., Yamaguchi, T., Massmann, J.W., Hansen, J.A., 1998. VOC vapor sorption in soil: soil type dependent model and implications for vapor extraction. Journal of Environmental Engineering-ASCE 124, 146-155. Pugh, R.E., Dick, D.G., Fredeen, A.L., 2002. Heavy metal (Pb, Zn, Cd, Fe, and Cu) contents of plant foliage near the anvil range lead/zin c mine, Faro, Yukon terr itory. Ecotoxicology and Environmental Safety 52, 273-279. Quartacci, M.F., Argilla, A., Baker, A.J.M., Navari-Izzo, F., 2006. Phytoextraction of metals from a multiply contaminated soil by I ndian mustard. Chemosphere 63, 918-925. Raicevic, S., Kaludjerovic-Radoi cic, T., Zouboulis, A.I., 2005. In situ stabilization of toxic metals in polluted soils using phosphates: theoretical prediction and experimental verification. Journal of H azardous Materials 117, 41-53. Ramos, L., Hernandez, L.M., Gonzalez, M.J., 1994. Sequential fractionation of copper, lead, cadmium and zinc in soils from or near D onana-National-Park. Journal of Environmental Quality 23, 50-57. Rao, G.R., Laitinen, H.A., 1974. Studies of heavy metal adsorption by hydrous iron and manganese oxides. Analytical Chemistry 46, 2022-2026. Raskin, I., Smith, R.D., Salt, D.E., 1997. Phytorem ediation of metals: using plants to remove pollutants from the environment. Current Opinion in Biotechnology 8, 221-226. Reddy, K.R., Chaparro, C., Saichek, R.E., 2003. I odide-enhanced electrokin etic remediation of mercury-contaminated soils. Journal of Environmental Engineering-ASCE 129, 11371148. Reddy, K.R., Xu, C.Y., Chinthamreddy, S., 2001. A ssessment of electrokinetic removal of heavy metals from soils by sequential extraction an alysis. Journal of Hazardous Materials B84, 279-196. Reed, B.E., Carriere, P.C., Moore, R.J., 1996. Flus hing of a Pb (II) contaminated soil using HCl, EDTA and CaCl2. Journal of Environmental Engineering-ASCE 122, 48-50. Rojas-Avelizapa, N.G., Roldn-Carrillo, T ., Zegarra-Martnez, H., Muoz-Colunga, A.M., Fernndez-Linares, L.C., 2007. A field trial for an ex-situ bioremediation of a drilling mud-polluted site. Chemosphere 66, 1595-1600. Sahi, S.V., Bryant, N.L., Sharma, N.C., Si ngh, S.R., 2002. Characterization of a lead hyperaccumulator shrub, Sesbania drummondii. Environmental Science & Technology 36, 4676-4680.

PAGE 214

214 Sakai, Y., Miama, T., Takahashi, F., 1997. Si multaneous removal of organic and nitrogen compounds in intermittently aerated activated sludge process using magnetic separation. Water Research 31, 2113-2116. Salomons, W., 1995. Environmental impact of meta ls derived from mining activities: Processes, predictions, prevention. Journal of Geochemical Exploration 52, 5-23. Salomons, W., Frstner, U., 1980. Trace-metal anal ysis on polluted sediments. 2. evaluation of environmental-impact. Environmenta l Technology Letters 1, 506-517. Salt, D.E., Smith, R.D., Raskin, I., 1998. Phytor emediation. Annual Review of Plant Physiology and Plant Molecular Biology 49, 643-668. Sasaki, M., Kasai, M., Yamamoto, Y., Matsumot o, H., 1995. Involvement of plasma-membrane potential in the tolerance mech anism of plant-roots to alum inum toxicity. Plant and Soil 171, 119-124. Sas-Nowosielska, A., Kucharski, R., Malkowsk i, E., Pogrzeba, M., Kuperberg, J.M., Kry ski, K., 2004. Phytoextraction crop disposal-an un solved problem. Environmental Pollution 128, 373-379. Sauv, S., 2003. The role of chemical speciati on in bioavailability. In: Naidu, R., Gupta, V.V.S.R., Rogers, S., Kookana, R.S., Bolan, N. S., Adriano, D.C. (Eds), Bioavailability, toxicity and risk relationships in ecosystems Science Publishers, Enfield, NH, pp. 59-82. Sauv, S., Hendershot, W., Allen, H.E., 2000. Solid-solution partitioning of metals in contaminated soils: dependence of pH, to tal metal burden, and organic matter. Environmental Science & Technology 34, 1125-1131. Sayed, S.A., 1999. Effects of lead and kine tin on the growth, and some physiological components of safflower. Plant Growth Regulation 29, 167-174. Schultz, E., Joutti, A., Risnen, M.L., Lin tinen, P., Martikainen, E., Lehto, O., 2004. Extractability of metals and ecotoxicity of soils from two old wood impregnation sites in Finland. Science of the To tal Environment 326, 71-84. Seuntjens, P., Tirez, K., Simunek, J., Van Ge nuchten, M.T., Cornelis, C., Geuzens, P., 2001. Aging effects on cadmium transport in undi sturbed contaminated sandy soil columns. Journal of Environmental Quality 30, 1040-1050. Shanableh, A., Kharabsheh, 1996. Stabilization of Cd, Ni and Pb in soil using natural zeolite. Journal of Hazardous Materials 45, 207-217. Siebielec, G., Stuczy ski, T., Korzeniowska-Puculek, R., 2006. Metal bioavailability in longterm contaminated Tarnowskie Gory soils. Po lish Journal of Envi ronmental Studies 15, 121-129.

PAGE 215

215 Smeulders, F., Maes, A., Sinnaeve, J., Cremer s, A., 1983a. Insitu im mobilization of heavymetals with tetraethylenepentamine (TETREN) in natural soils and its effect on toxicity and plant-growth. 1. ion-exchange equilibria of metal-TETREN complexes in natural soils. Plant and Soil 70, 34-47. Smeulders, F., Sinnaeve, J., Cremers, A., 1983b. Insitu immobilization of heavy-metals with tetraethylenepentamine (TETREN) in natural so ils and its effect on toxicity and plantgrowth. 2. effect of complex-formation with TETREN on copper and zinc uptake in corn from nutrient solutions. Pl ant and Soil 70, 49-57. Smeulders, F., Vandegeijn, S.C., 1983. Insitu immobilization of heavy-metals with tetraethylenepentamine (TETREN) in natural so ils and its effect on toxicity and plantgrowth. 3. uptake and mobility of copper and its TETREN-complex in corn plants. Plant and soil 70, 59-68. Smit, C.E., Van Gestel, C.A.M., 1998. Effect s of soil type, prepercolation, and ageing on bioaccumulation and toxicity of zinc for the springtail Folsomia candida. Environmental Toxicology and Chemistry 17, 1132-1141. Smith, D.C., Sacks, J., Senior, E., 1999. Irrig ation of soil with synthetic landfill leachate: speciation and distribution of selected pollu tants. Environmental Pollution 106, 429-441. Song, J., Zhao, F.J., McGrath, S.P., Luo, Y.M ., 2006. Influence of soil properties and aging on arsenic phytotoxicity. Environmental Toxicology and Chemistry 25, 1663-1670. Spalding, B.P., 1994. Volatilization of cesium-13 7 from soil with chloride amendments during heating and vitrificati on. Environmental Scien ce & Technology 28, 1116-1123. Spear, J.R., Figueroa, L.A., Honeyman, B.D., 2000. Modeling reduction of uranium U(VI) under variable sulfate concentrations by sulfate-re ducing bacteria. Applied and Environmental Microbiology 66, 3711-3721. Stewart, M.A., Jardine, P.M., Brandt, C.C., Barnett, M.Q., Fendorf, S.E., McKay, L.D., Mehlhorn, T.L., Paul, K., 2003. Effects of c ontaminant concentrati on, aging, and soil properties on the bioaccessibility of Cr(II I) and Cr(VI) in soil. Soil and Sediment Contamination 12, 1-21. Sumner, M. E., Miller, W. P., 1996. Cation exch ange capacity and exchange coefficients. In: Sparks, D.L. (Ed.), Methods of Soil Analys is. Part 3: Chemical Methods. American Society of Agronomy, Soil Science Societ y of America, Madison, WI, pp. 1201-1229. urija, B., Branica, M., 1995. Distribution of Cd Pb, Cu and Zn in carbonate sediments from the Krka river estuary obtained by sequential ex traction. Science of the Total Environment 170, 101-118. Swallow, K.C., Hume, D.N., Morel, F.M.M., 1980. Sorption of copper and lead by hydrous ferric oxide. Environmental Sc ience & Technology 14, 1326-1331.

PAGE 216

216 Szakova, J., Tlustos, P., Pavlikova, D., Balik, J ., 2005. Modification of standardized European soil sequential extraction method for evaluati on of plant-available amounts of selected elements. Chemicke Listy 99, 502-508. Tack F.M., Callewaert, O.W.J.J., Verloo, M.G., 1996. Metal solubility as a function of pH in a contaminated, dredged sediment affected by oxidation. Environmental Pollution 91, 199208. Tack F.M., Verloo, M.G., 1995. Chemical speciation and fractionation in soil and sediment heavy metal analysis: a revi ew. International Journal of Environmental Analytical Chemistry 59, 225-238. Takayasu, M., Kelland, D.R., Minervini, J.V., 2000. Continuous magnetic separation of blood components from whole blood. IEEE Transac tions on Applied Superconductivity 10, 927930. Tamura, H., Okita, K., Katayama, N., Furuichi, R., 1994. Modeling of diva lent heavymetal ion adsorption on the surface of iron (III) oxi de particles. Bunseki Kagaku 43, 831-836. Tang, S., Xi, L., Zheng, J., Li, H., 2003. Response to elevated CO2 of Indian mustard and sunflower growing on copper cont aminated soil. Bulletin of Environmental Contamination and Toxicology 71, 988-997. Tang, X.Y., Zhu, Y.G., Chen, S.B., Tang, L.L., Chen, X.P., 2004. Assessment of the effectiveness of different phosphorus fertiliz ers to remediate Pb-contaminated soil using in vitro test. Environment International 30, 531-537. Tarras-Wahlberg, N.H., Lane, S.N., 2003. Suspende d sediment yield and metal contamination in a river catchment affected by El Nio events and gold mining activities: the Puyango river basin, southern Ecuador. Hydr ological Processes 17, 3101-3123. Terashima, Y., Ozaki, H., Sekine, M., 1986. Re moval of dissolved heavy-metals by chemical coagulation, magnetic seeding and high-gradient magnetic filtration. Water Research 20, 537-545. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the speciation of particulate trace meta ls. Analytical Chemistry 51, 844-851. Tokahoglu, S., Kartal, S., 2004. Bi oavailability of soil-extractable metals to tea plant by BCR sequential extraction procedure. Instru mentation Science & Technology 32, 387-400. Trapp, S., Zambrano, K.C., Kusk, K.O., Karlson, U., 2000. A phytotoxicity test using transpiration of willows. Archives of Envi ronmental Contamination and Toxicology 39, 154-160. Travis, C.C., Macinnis, J.M., 1992. Vapor extrac tion of organics from subsurface soils-is it effective. Environmental Science & Technology 26, 1885-1887.

PAGE 217

217 Trivedi, P., Axe, L., 2000. Modeling Cd and Zn sorption to hydrous metal oxides. Environmental Science & Technology 34, 2215-2223. Tsang, D.C.W., Zhang, W.H., Lo, I.M.C., 2007. Copper extraction effectiveness and soil dissolution issues of EDTA-flushing of artificially contaminated soil. Chemosphere 68, 234-243. Tu, C., Zheng, C.R., Chen, H.M., 2001. Distributi on of copper and zinc fractions in red soil as influenced by fertilizer application. Comm unications in Soil Science and Plant Analysis 32, 661-673. Turpeinen, R., Salminen, J., Kairesalo, T., 2000 Mobility and bioavailability of lead in contaminated boreal forest soil. Envi ronmental Science & Technology 34, 5152-5156. Tye, A.M., Young, S.D., Crout, N.M.J., Zhang, H., Preston, S., Barbosa-Jefferson, V.L., Davison, W., McGrath, S.P., Paton, G.I., Kilham, K., Resende, L., 2003. Predicting the activity of Cd2+ and Zn2+ in soil pore water from the radio-labile metal fraction. Geochimica et Co smochimica Acta 67, 375-385. Upreti, S.R., Lohi, A., Kapadia, R.A., El-Haj R., 2007. Vapor extraction of heavy oil and bitumen: A review. Energy & Fuels 21, 1562-1574. U.S. Department of Agriculture. 1992. Soil Survey Laboratoty Methods Manual. pp. 14. Washington, D.C. U.S. Department of Health & Human Servi ces, 1999a. Toxicological profile for cadmium. Agency for toxic substances and disease registry. Atlanta, Georgia. U.S. Department of Health & Human Servi ces, 1999b. Toxicological profile for mercury. Agency for toxic substances and disease registry. Atlanta, Georgia. U.S. Department of Health & Human Services, 2004. Toxicological profile for copper. Agency for toxic substances and diseas e registry. Atlanta, Georgia. U.S. Department of Health & Human Services, 2005a. Toxicological profile for zinc. Agency for toxic substances and disease registry. Atlanta, Georgia. U.S. Department of Health & Human Services, 2005b. Toxicological profile for lead (draft for public comment). Agency for toxic substances and disease registry. Atlanta, Georgia. U.S. Environmental Protection Agency, 1992. Me thod 3010A: acid digestion of aqueous samples and extracts for total metals for analysis by FLAA or ICP spectroscopy. Office of Solid Waste. Washington, D.C. U.S. Environmental Protection Agency, 1993. Selecting remediation techniques for contaminated sediment. Office of Water. Washington, D.C.

PAGE 218

218 U.S. Environmental Protection Agency, 1996. Method 3050B: acid digestion of sediments, sludges, and soils. Office of Solid Waste. Washington, D.C. U.S. Environmental Protection Agency, 1997. Best management practices (BMPs) for soils treatment technologies. Office of Solid Waste. Washington, D.C. U.S. Environmental Protection Agency, 1998a. EPAs contaminated sediment management strategy. Office of Water. Washington, D.C. U.S. Environmental Protection Agency, 1998b. In situ treatment of contaminated sediments. Office of Solid Waste and Emergency Response. Technology Innovation Office. Washington, D.C. U.S. Environmental Protection Agency, 2002a. S hort-term methods for estimating the chronic toxicity of effluents and rece iving waters to freshwater or ganisms. Office of Water. Washington, D.C. U.S. Environmental Protection Agency, 2002b. Me thod 1631, Revision E: mercury in water by oxidation, purge and trap, and cold vapor atomic fluorescence spectrometry. Office of Water. Washington, D.C. U.S. Environmental Protection Aagency, 2003a. National primary drinking water standards. Office of Water. Washington, D.C. U.S. Environmental Protection Aagency, 2003b. National secondary drinking water standards. Office of Water. Washington, D.C. U.S. Environmental Protection Aagency, 2004. Method 9045D: soil and waste pH. Office of Soild Waste. Washington, D.C. U.S. Environmental Protection Aagency, 2005.Cont aminated sediment remediation guidance for hazardous waste sites. Office of Solid Waste and Emergency Response. Washington, D.C. Usman, A.R.A., Kuzyakov, Y., Stahr, K., 2005. Effect of immobilizing substances and salinity on heavy metals availability to wheat grown on sewage sludge -contaminated soil. Soil & Sediment Contamination 14,329-344. Van Benschoten, J.E., Matsumoto, M.R., Young, W.H., 1997. Evaluation and analysis of soil washing for seven lead-contaminated soils. Jo urnal of Environmental Engineering-ASCE 123, 217-224. Van Dillewijn, P., Caballero, A., Paz, J.A., G onzalez-Perez, M.M., Oliva, J.M., Ramos, J.L., 2007. Bioremediation of 2,4,6trinitrotoluene under field conditions. Environmental Science & Technology 41, 1378-1383. Van Straalen, N.M., Denneman, C.A.J., 1989. Ecotoxi cological evaluation of soil quality criteria. Ecotoxicology and Environmental Safety 18, 241-251.

PAGE 219

219 Virkutyte, J., Sillanp, M., Latostenmaa, P ., 2002. Electrokinetic soil remediation-critical overview. Science of the Total Environment 289, 97-121. Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid ti tration method. Soil Science 37, 29-38. Wang, F.Y., Chen, J.S., 2000. Relation of sediment characteristics to trace metal concentrations: a statistical study. Water Research 34, 694-698. Wang, Y.M., Chen, T.C., Yeh, K.J., Shue, M.F., 2001. Stabilization of an elevated heavy metal contaminated site. Journal of Hazardous Materials 88, 63-74. Wani, P.A., Khan, M.S., Zaidi, A., 2007. Impact of heavy metal toxicity on plant growth, symbiosis, seed yield and nitr ogen and metal uptak e in chickpea. Australian Journal of Experimental Agriculture 47, 712-720. Ward, M., Bitton, G., Townsend, T., 2005. Heavy metal binding capacity (HMBC) of municipal solid waste landfill leachates. Chemosphere 60, 206-215. Warner, K.A., Roden, E.E., Bonzongo, J.C., 2003. Microbial mercury transformation in anoxic freshwater sediments under iron-reduci ng and other electro-accepting conditions. Environmental Science & Technology 37, 2159-2165. Weng, L., Temminghoff, E.J., Lofts, S., Tipping, E., Van Riemsdijk, W.H., 2002. Complexation with dissolved organi c matter of heavy metals in a sa ndy soil. Environm ental Science & Technology 36, 4804-4810. White, C., Sharman, A.K., Gadd, G.M., 1998. An integrated microbial process for the bioremediation of soil contaminated with toxic metals. Nature Biotechnology 16, 572-575. Wiatrowski, H.A., Ward, P.M., Barkay, T., 2006. Novel reduction of mercury (II) by mercurysensitive of dissimilatory metal reducing bacteria. Envi ronmental Science & Technology 40, 6690-6696. Wilde, E.W., Brigmon, R.L., Dunn, D.L., Heitk amp, M.A., Dagnan, D.C., 2005. Phytoextraction of lead from firing range soil by Vetiver grass. Chemosphere 61, 1451-1457. Wilkie, J.A., Hering, J.G., 1996. Adsorption of arsenic onto hydrous fe rric oxide: effects of adsorbate/adsorbent ratios and co-occu rring solutes. Colloids and Surfaces APhysicochemical and Engineering Aspects 107, 97-110. Will, M.E., Suter II, G.W., 1995. Toxicological benchmarks for screening potential contaminants of concern for effects on terr estrial plants: 1995 Revision. En vironmental Restoration Risk Assessment, Oak Ridge National Laboratory, Tennessee. Wilson, S.C., Jones, K.C., 1993. Bioremediation of soil contaminated with polynuclear aromatichydrocarbons (PAHS)-a review. Envi ronmental Pollution 81, 229-249.

PAGE 220

220 Wong, J.W.C., Wong, W.W.Y, Wei, Z.G., Jagad eesan, H., 2004. Alkaline biosolids and EDTA for phytoremediation of an acidic loamy soil sp iked with cadmium. Science of the Total Environment 324, 235-246. Wu, L.H., Luo, Y.M., Xing, X. R., Christie, P., 2004. EDTA-e nhanced phytoremediation of heavy metal contaminated soil with Indian mu stard and associated potential leaching risk. Agriculture Ecosystems & Environment 102, 307-318. Wundram, M., Selmar, D., Bahadir, M., 1996. The Chlamydomonas test: A new phytotoxicity test based on the inhibition of algal photosynthesis enables the assessment of hazardous leachates from waste disposals in salt mines. Chemosphere 32, 1623-1631. Yamaguchi, N.U., Okazaki, M., 2002. Micro volume changes due to Pb(II) and Cu(II) sorption on amorphous Fe(III) hydroxide. Journal of Colloid and Interface Science 249, 489-491. Yanagisawa, Y., Hasuda, T., Iwasaki, J., 1981. Fine treatment of industrial sludges by high gradient magnetic separation. IEEE Transactions on Magnetics 17, 3317-3319. Yang, G.C.C., Lin, S.L., 1998. Removal of l ead from a silt loam soil by electrokinetic remediation. Journal of H azardous Materials 58, 285-299. Yang, J., Mosby, D.E., Casteel, S.W., Blanch ar, R.W., 2001. Lead immobilization using phosphoric acid in a smelter-contaminated urban soil. Environmental Science & Technology 35, 3553-3559. Yeager, R.P., 1998. Remediation of metal-cont aminated soils and sediments using magnetic adsorbents. MS Thesis. University of Florida, Gainesville, Florida. Yong, R.N., Yaacob, W.Z.W., Bently, S.P., Harri s, C., Tan, B.K., 2001. Partitioning of heavy metals on soil samples from column tests. Engineering Geology 60, 307-322. Yoon, J.K., Cao, X.D., Ma, L.Q., 2007. Applic ation methods affect phosphorous-induced lead immobilization form a contaminated soil. J ournal of Environmental Quality 36, 373-378. Yu, K.C., Ho, S.T., Tsai, L.J., Chang, J.S., Lee, S.Z., 1996. Remobilization of zinc from ell-ren river sediment fractions affected by EDTA, DTPA a nd EGTA. Water Science and Technology 34, 125-132. Zavoda, J., Cutright, T., Szpak, J., Fallon, E ., 2001. Uptake, selectiv ity, and inhibition of hydroponic treatment of contaminants. Journal of Environmental Engineering-ASCE 127, 502-508. Zeng, L.S., Liao, M., Chen, C.L., Huang, C.Y., 2007. Effects of lead contamination on soil enzymatic activities, microbial biomass, and rice physiological indi ces in soil-lead-rice (Oryza sativa L.) system. Ecotoxicology and Environmental Safety 67, 67-74.

PAGE 221

221 Zeroual, Y., Mutaouakkil, A., Blaghen, M., 2001. Volatilization of mercury by immobilized bacteria (Klebsiella pneumoniae) in different support by usin g fluidized bed bioreactor. Current Microbiology 43, 322-327. Zhang, T.H., Shan, X.Q., Li, F.L., 1998. Comparis on of two sequential extraction procedures for speciation analysis of metals in soils a nd plant availability. Communications in Soil Science and Plant Analysis 29, 1023-1034. Zhou, Z.Y., Fan, Y.P., Wang, M.J., 2000. Heavy metal contamination in vegetables and their control in China. Food Reviews International 16, 239-255. Zhu, K., Hart, W., Yang, J.T., 2005. Remedia tion of petroleum-contaminated loess soil by surfactant-enhanced flushing technique. Journa l of Environmental Sc ience and Health 40, 1877-1893.

PAGE 222

222 BIOGRAPHICAL SKETCH Nan Feng, born in Liaoning province, Chin a, is the only child of Jianguo Feng and Shuying Zhang. Nan Feng received her bachelors degree in environmental science from Nankai University in 2003. Upon completion of her underg raduate study, she was admitted as a graduate student with an Alumni Fellowship award in th e Department of Envir onmental Engineering and Sciences at the University of Florida, where she started her doctoral resear ch in soil remediation under the direction of Dr. Gabriel Bitton.