- Permanent Link:
- Fate, Transport, and Risk Assessment of Biosolids-Borne Triclosan (TCS)
- Place of Publication:
- [Gainesville, Fla.]
- University of Florida
- Publication Date:
- Physical Description:
- 1 online resource (250 p.)
- Doctorate ( Ph.D.)
- Degree Grantor:
- University of Florida
- Degree Disciplines:
- Soil and Water Science
- Committee Chair:
- O'Connor, George A
- Committee Co-Chair:
- Toor, Gurpal Singh
- Committee Members:
- Thomas, John E.
Wilson, Patrick C
James, Margaret O
- Graduation Date:
- Subjects / Keywords:
- American sign language ( jstor )
Bioaccumulation ( jstor )
Biodegradation ( jstor )
Chemicals ( jstor )
Earthworms ( jstor )
Leaching ( jstor )
Risk analysis ( jstor )
Soil science ( jstor )
Soils ( jstor )
Toxicity ( jstor )
ANTIMICROBIAL -- ASSESSMENT -- BIOACCUMULATION -- BIODEGRADATION -- BIOSOLIDS -- EARTHWORMS -- LEACHING -- MICROBES -- MOBILITY -- RISK -- TOXICITY -- TRICLOSAN
Soil and Water Science -- Dissertations, Academic -- UF
- bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Soil and Water Science thesis, Ph.D.
- Triclosan (TCS) is an antimicrobial compound used in many personal care products, and is a common constituent of domestic wastewater. Removal rates of TCS in wastewater treatment plants are typically >95% and most of the TCS partitions (accumulates) in sludge solids. Sludge processed to produce biosolids is often land applied, and can transfer TCS to agricultural soils. The objectives of this research were to evaluate the environmental fate and ecological effects of biosolids-borne TCS following biosolids land application. As previously published TCS risk assessments were based on the data derived from models or unpublished sources, we used measured TCS concentrations, properties, fate, and transport data to conduct a more realistic risk assessment. The TCS concentrations measured in representative biosolids were 1 to 40 mg/kg (mean = 18 ? 12 mg/kg). Water solubility measured at neutral pH was 9 mg/L, and increased to nearly 800 mg/L at two pH units above the pKa of ~8. The log Kd of TCS in biosolids was 4.15 ? 0.03 and log Koc was 4.68 ? 0.07. A TCS biodegradation study determined a primary degradation half-life of ~100 d and a degradation product (Methyl TCS) was identified. Biosolids TCS concentrations of <105 mg/kg did not affect the earthworm survival, suggesting an estimated lethal TCS concentration (LC50) >105 mg/kg. Bioaccumulation factors (BAF) in earthworms were 4.3 to 12 depending on soil texture. Investigation of TCS effects on microbial processes suggested that biosolids TCS concentrations <500 mg/kg had no significant effect on the microbially-mediated processes. Plant bioaccumulation of TCS was minimal in radish (BAF = 0.004) and bahia grass leaves (BAF = <0.001), but greater in lettuce leaves (BAF = 0.04) and radish roots (BAF = 0.43). No significant leaching of TCS occurred beyond the biosolids incorporation depth of 0 to 2.5 cm in amended soil. A preliminary (multi-pathway, multi-target) risk assessment identified American woodcock as the most sensitive species. A tier-2 assessment, utilizing less conservative estimates, suggested minimal risk of biosolids-borne TCS to human and environmental health, when biosolids are land applied in a sustainable manner. ( en )
- General Note:
- In the series University of Florida Digital Collections.
- General Note:
- Includes vita.
- 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.
- Thesis (Ph.D.)--University of Florida, 2011.
- Adviser: O'Connor, George A.
- Co-adviser: Toor, Gurpal Singh.
- Electronic Access:
- RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-04-30
- Statement of Responsibility:
- by MANMEET WARIA.
- Source Institution:
- Rights Management:
- Copyright WARIA,MANMEET. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
- Embargo Date:
- LD1780 2011 ( lcc )
This item has the following downloads:
1 FATE, TRANSPORT, AND RISK ASSESSMENT OF BIOSOLIDS BORNE TRICLOSAN (TCS) By MANMEET WARIA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Manmeet Waria
3 This dissertation is dedicated to my family
4 ACKNOWLEDGMENTS It is a pleasure to thank all those who made this dissertation possible. I am heartily thankful to my supervisor Dr support from preliminary to the concluding level enabled me to develo p understanding of the subject. I would also like to thank my co advisor Dr Gurpal Toor and committee members Drs. John Thomas, Chris Wilson, Ed Topp and Margaret James for their contribution by providing laboratory space as well as guidance at all times. Special thanks go to Dr Andy Ogram, Abid Al Agely and Hee Sung who provided guidance during the analysis of my mi crobiological work; Dr. George Hochmuth and Dr Maria Silveira for their assistance and guidance in my plant uptake work. Thank you to Bob Querns, Dawn Lucas, and Yu Wang for providing expertise on instrumentation in their laboratories. Words get short to describe the help of some folks at the Gulf Coast Research and Education Center especially Dr Bielinski, Nancy West, Elizabeth Golden, Maninder Chahal and Gitta Shurberg I also want to thank the past members of the Soil Chemistry group, including Liz Sny der, Sampson Agyin Birikorang, Matt Miller, Dani el Moura, Augustine Obour, and Jaya Das for all their help and support. I am indebted to the funding agency Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) for funding the research and th eir personnel (Kuldip Kumar, Lakhwinder Hundal) for helping me collect useful field samples. I like to thank the A lmighty for giving me the strength to achieve my goal; my loving husband Varinder Pannu, my parents Sarbdeep Kaur and Balwant Singh, siblings Harveen Bajwa, Naveen Sidhu and Kanwar Sandeep Singh, for their love and support and for being there and loving me always.
5 Lastly, I offer my regards and blessings to all my friends (Gur reet Brar, Raman Brar, Milap Sandhu, Aman Sandhu, Preeti e Sood Rupesh Bhomia, and Maninder Singh ), and to all of those w ho supported me in any respect during the completion of my project.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 11 LIST OF FIGURES ................................ ................................ ................................ ........ 14 LIST OF ABBREVIATIONS ................................ ................................ ........................... 16 ABSTRA CT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION AND PROJECT OBJECTIVES ................................ ................... 19 Background ................................ ................................ ................................ ............. 19 Objective 1: Quant ify TCS Concentrations in Biosolids ................................ .... 27 Objective 2: Determine/Verify Basic Physico Chemical Properties of TCS ...... 28 Objective 3: Determine the Degradation (Persistence) of Biosolids Borne TCS ................................ ................................ ................................ ............... 29 Objective 4: Determine the Impacts of Biosolids Borne TCS to Soil Organisms ................................ ................................ ................................ ..... 29 Objective 5: Determine the Toxicity of Biosolids Borne TCS on Microbial Reactions ................................ ................................ ................................ ...... 30 Objective 6: Quantify the Phytoavailability of Biosolids Borne TCS ................. 30 Objective 7: Quantify the Leaching Potential of Biosolid s Borne TCS .............. 31 Ultimate Objective: Risk Assessment of Biosolids Borne TCS ............................... 31 2 BIOSOLIDS BORNE TCS CONCENTRATIONS ................................ .................... 34 Background ................................ ................................ ................................ ............. 34 Material and Methods ................................ ................................ ............................. 34 Results and Discussion ................................ ................................ ........................... 36 3 BASIC PHYSICO CHEMICAL PROPERTIES OF TCS ................................ .......... 40 Background ................................ ................................ ................................ ............. 40 Material and Methods ................................ ................................ ............................. 43 Solubility ................................ ................................ ................................ ........... 43 Partitioning Coeff icients (K d and K oc ) ................................ ................................ 45 Results and Discussion ................................ ................................ ........................... 47 4 BIODEGRADATION OF BIOSOLIDS BORNE TCS ................................ ............... 53 Background ................................ ................................ ................................ ............. 53
7 Material and Metho ds ................................ ................................ ............................. 56 Chemicals, Biosolids and Soils ................................ ................................ ......... 56 Biodegradation Study Design ................................ ................................ ........... 56 Base Trap Analysis and Soil Sample Extraction ................................ ............... 58 Sequential Extraction Scheme ................................ ................................ ......... 59 Radiological Thin Layer Chromatography (RAD TLC) for Extract Speciation .. 60 Statistical Analysis ................................ ................................ ............................ 61 Results and Discussion ................................ ................................ ........................... 61 Mass Balance and Mineralization of 14 C TCS ................................ .................. 61 Total Carbon Dioxide (CO 2 ) Analyses ................................ .............................. 64 Metabolite Identification ................................ ................................ .................... 65 Half life (Persistence) Determination ................................ ................................ 67 Comparison of TCS Persistence in Amended, Un Amended, and Field Soils .. 70 5 IMPACTS OF BIOSOLIDS BORNE TCS ON SOIL DWELLING ORGANISMS ...... 82 Background ................................ ................................ ................................ ............. 82 Material and Methods ................................ ................................ ............................. 85 Chemicals, Biosolids and Soils ................................ ................................ ......... 85 Range Finding Toxicity Test Design ................................ ................................ 86 Definitive Toxicity Test Design ................................ ................................ ......... 87 Earthworm Bioaccumulation Test ................................ ................................ ..... 88 Design for the laboratory study ................................ ................................ .. 88 Bioaccumulation in field soils ................................ ................................ ..... 88 Sample extraction and derivatization for earthworm bioaccumulation ....... 89 Instrument analyses and quantification ................................ ...................... 90 Results and Discussion ................................ ................................ ........................... 91 Range Finding Toxicity Test ................................ ................................ ............. 91 Definitive Toxicity Test (IFS Soil) ................................ ................................ ...... 92 Bioaccumulation Laboratory Study ................................ ................................ ... 94 Bioaccumulation Field Test ................................ ................................ .............. 99 6 BIOSOLIDS BORNE TCS EFFECTS ON SOIL MICROBES ................................ 106 Background ................................ ................................ ................................ ........... 106 Material and Methods ................................ ................................ ........................... 110 Chemic als, Biosolids, and Soils ................................ ................................ ...... 110 Microbial Toxicity (Range Finding) Test Design ................................ ............. 112 Microbial Toxicity (Definitive) Test Design ................................ ...................... 113 Sample Preparation and Analyses for Microbial Toxicity Test ........................ 113 Microbial Community Structure Test Design ................................ .................. 115 Extraction for Bacterial Count ................................ ................................ ......... 116 Statistical Analysis ................................ ................................ .......................... 117 Results and Discussion ................................ ................................ ......................... 117 Microbial Toxicity (Range Finding) Test ................................ ......................... 117 Microbial Toxicity (Definitive) Test ................................ ................................ .. 118 Effect on respiration rates ................................ ................................ ........ 118
8 Effect on nitrogen cycle ................................ ................................ ............ 119 Bacterial DNA analysis ................................ ................................ ................... 122 Micr obial Community Structure Analysis ................................ ........................ 123 Bacterial Counts ................................ ................................ ............................. 124 7 PLANT TOXICITY AND BIOACCUMULATION OF BIOSOLIDS BORNE TCS .... 134 Background ................................ ................................ ................................ ........... 134 Material and Met hods ................................ ................................ ........................... 139 Soils and Chemicals ................................ ................................ ....................... 139 Toxicty and Bioaccumulation Study Design ................................ .................... 140 Bioaccumulation Field Study ................................ ................................ .......... 142 Plant Harvesting and Sample Preparation ................................ ...................... 143 Instrument Analysis and Quantitation ................................ ............................. 144 Results and Discussion ................................ ................................ ......................... 145 Plant Biomass Yields ................................ ................................ ...................... 145 Uptake in the Above Ground Biomass (Lettuce, Radish and Bahia Grass Leaves) ................................ ................................ ................................ ....... 147 Uptake in the Below Ground Biomass (Radish Root) ................................ ..... 148 Bioaccumulation Field Study ................................ ................................ .......... 149 Model Predicted TCS Concentrations in Plant Tissue ................................ .... 150 Mechanism of Bioaccumulation and Comparison with Other Studies ............ 151 Degradation in Soils ................................ ................................ ....................... 153 Comparison with Real World TCS Concentrations ................................ ......... 154 8 MOBILITY OF TRICLOSAN (TCS) IN BIOSOLIDS AMENDED SOILS ................ 163 Background ................................ ................................ ................................ ........... 163 Materials and Me thods ................................ ................................ .......................... 167 Experimental Design ................................ ................................ ...................... 167 Soils, Biosolids and Chemicals ................................ ................................ ....... 168 Tagging and Application of Biosolids ................................ .............................. 168 Pore Volume Determination ................................ ................................ ........... 169 Leachate Collection and Analysis ................................ ................................ ... 169 Determination of 14 C Activity in the Soil ................................ .......................... 170 Results and Discussion ................................ ................................ ......................... 170 Leachate Recoveries and Tracer Breakthrough ................................ ............. 170 14 C in Leachates and Recoveries by Combustion ................................ .......... 171 14 C Recoveries by Extraction and Extract Speciation ................................ ..... 174 Comparison with CMLS Model ................................ ................................ ....... 175 Implications of TCS Movement ................................ ................................ ....... 177 9 RISK ASSESSMENT OF BIOSOLIDS BORNE TCS ................................ ............ 184 Background ................................ ................................ ................................ ........... 184 Pathways of Exposure ................................ ................................ .......................... 187 Exclusion of Exposure Pathways ................................ ................................ .......... 187
9 Risk to Humans ................................ ................................ .............................. 187 Risk to Aquatic Organisms ................................ ................................ ............. 189 Reference Dose Calculation ................................ ................................ ................. 192 Parameters for Risk Estimation ................................ ................................ ............ 194 Environmental Fate ................................ ................................ ........................ 194 Effect on Soil Dwelling Organisms ................................ ................................ 194 Toxicity to earthworms ................................ ................................ ............. 194 Bioaccumulation in earthworms ................................ ............................... 195 Toxicity and Bioaccmulutaion in Plant Biomass ................................ ............. 195 Avian and Mammalian Toxicity ................................ ................................ ....... 195 Screening Level Assessment ................................ ................................ ............... 196 Exposure Concentration Calculation ................................ .............................. 196 Screening Level Hazard Index Calculation ................................ ..................... 197 Tier 2 Assessment ................................ ................................ ................................ 197 Consideration of TCS Degradation ................................ ................................ 197 Pathway 1: Biosolids soil plant (direct phytotoxicity) ........................... 199 Pathway 8: Biosolids soil soil organism ................................ .............. 199 Pathway 9: Biosolids soil soil organism predator ................................ .... 200 Sources of Uncertainty in Our Risk Estimation ................................ ............... 200 Calculation of Preliminary Biosolids Borne TCS Pollutant Limits ................... 201 Cumulative pollutant loading rates (CPLRs) ................................ ............ 201 Annual pollutant loading rate (APLR) ................................ ....................... 202 Ceiling concentration limit ................................ ................................ ........ 202 Pollutant concentration limit ................................ ................................ ..... 203 10 SUMMARY AND CONCLUSIONS ................................ ................................ ........ 21 9 Summary of Intermediate Objective Results ................................ ......................... 219 Objective 1: Quantify TCS Concentration in Biosolids ................................ .... 219 Objective 2: Determine/Verify Basic Physico Chemical Properties of TCS .... 219 Objective 3: Determine the Degradation (Persistence) of Biosolids Borne TCS ................................ ................................ ................................ ............. 220 Objective 4: Determine the Impacts of Biosolids Borne TCS to Soil Organisms ................................ ................................ ................................ ... 221 Objective 5: Determine the Toxicity of Biosolids Borne TCS on Microbial Reactions ................................ ................................ ................................ .... 222 Objective 6 and 7: Quantify the Phytoavailability and Leaching Potential of Biosolids Borne TCS ................................ ................................ ................... 223 Ultimate Objective: Risk Assessment of Biosolids Borne TCS ....................... 224 Future Studies ................................ ................................ ................................ ...... 225 APPENDIX A EXPLANATION OF THE SEQUENTIAL EXTRACTION SCHEME ....................... 227 B SUPPLEMENATAL DATA FOR CHAPTER 7 ................................ ....................... 229
10 C LIMITS OF DETECTION, QUANTITATION AND RECOVERIES ......................... 230 WORKS CITED ................................ ................................ ................................ ........... 231 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 250
11 LIST OF TABLES Table page 1 1 Toxicity end points of TCS for some aquatic species ................................ ......... 33 2 1 Triclosan (TCS) concentrations in fifteen biosolids (n = 3) obtained from wastewater treatment plants across the U.S. ................................ ..................... 38 3 1 Physico chemical Properties of TCS reported in the literature. .......................... 49 3 2 Water solubility (mg L 1 ) of TCS at various pH values ................................ ........ 50 3 3 Mean log partition coefficients (K d and K oc ) (n=3) standard error (S.E) for TCS ................................ ................................ ................................ .................... 51 4 1 Selected physico chemical properties of the soils and biosolids used in the study ................................ ................................ ................................ ................... 73 4 2 Biodegradation experiment treatments ................................ ............................... 73 4 3 Percent recoveries sta ndard errors of 14 C TCS in IFS (Immokalee fine sand) biotic soil treatment [week (wk) 0 18]. ................................ ...................... 74 4 4 Percent recoveries standard errors of 14 C TCS in IFS (Immokalee fine sand) inhibited soil treatment [week (wk) 0 18]. ................................ .................. 74 4 5 Percent recoveries standard errors of 14 C TCS in ASL (Ashkum silty clay loam) biotic soil treatment [week (wk) 0 18]. ................................ ...................... 75 4 6 Percent recoveries standard errors of 14 C TCS in ASL (Ashkum silty clay loam) inhibited soil treatment [week (wk) 0 18]. ................................ .................. 75 4 7 Rate constants (k) and regression coefficients (R 2 ) obtained for the biodegradation data ................................ ................................ ............................ 76 5 1 Major physico chemical properties of the soils and biosolids utilized in the present study ................................ ................................ ................................ .... 101 5 2 Measured [average; n = 4 and standard error (SE)] TCS concentrations ( mg kg 1 ) and bioaccumulation factors (BAFs) in earthworms grown in the Immokalee fine sand (IFS) ................................ ................................ .............. 102 5 3 Measured (average; n = 4 and SE) TCS concentrations ( mg kg 1 ) and bioaccumulation factors (BAFs) in earthworms grown in th e Ashkum silty clay loam soil (ASL) ................................ ................................ ................................ 102
12 5 4 Measured TCS concentrations ( mg kg 1 ) and bioaccumulation factors (BAFs) in the earthw orms collected from the field equilibrated biosolids amended landscaping soil ................................ ................................ ................................ 103 6 1 The grouping of the various substrates in the biolog ECO plates (Garland and Mills, 1991). ................................ ................................ ............................... 126 6 2 Average bacterial count (number) in soils with varying biosolids applica tion rates, T CS concentrations and textures ................................ .......................... 126 7 1 Selected physico chemical properties of the soils and biosolids used in the present study. ................................ ................................ ................................ ... 157 7 2 Yield of plant parts of three plant species represented by fresh weights (g) in the control an d biosolids amended treatments ................................ ................. 157 7 3 Measured TCS concentrations (average; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the lettuce leaves ................................ ................................ ................................ .... 158 7 4 Measured TCS concentrations (average; n = 3 or 4 and SD) and bioaccumulation f actors (BAF) and BASL4 model calculated BAFs in the radish leaves ................................ ................................ ................................ .... 158 7 5 Measured TCS concentrations (average; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the Bahia grass ................................ ................................ ................................ ....... 158 7 6 Measured TCS concentrations (average ; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the radish root ................................ ................................ ................................ ........ 159 7 7 Bioaccumulations factors (BAF) [average (n=3) and standard error (SE)] obtained in grains of soybean grown in field soils. ................................ ............ 159 7 8 Bioaccumulations factors (BAF) [average (n=3) and standard error (SE)] obtained in leaves of corn grown in field soils. ................................ ................. 159 7 9 Measured TCS soil concentrations in lettuce treatments (means; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % di sappearance. ................................ ................................ ............................. 160 7 10 Measured TCS soil concentrations in radish treatments (means; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % disappearance. ................................ ................................ ............................. 160
13 7 11 Measured TCS soil concentrations in bahia grass treatments (m eans; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % disappearance. ................................ ................................ ..... 160 8 1 Amount and s chedule of leaching events for the empirical and stochastic irrigation regimes. ................................ ................................ ............................. 179 8 2 Average percent recoveries standard devi ation of 14 C (by combustion procedure) by depth in the control and biosolids treatment .............................. 179 8 3 Average percent extraction recoveries of 14 C standard deviation in the top depth of the control and biosolids treatment. ................................ .................... 180 8 4 Speciation of 14 C extracted standard deviation in the top depth of control and biosolids treatme nts for each irrigation regime ................................ ......... 180 9 1 Human and ecological exposure pathways for land applied biosolids (US EPA, 1995). ................................ ................................ ................................ ...... 204 9 2 Acute aquatic toxicity endpoints of TCS from published studies. ...................... 206 9 3 Chronic aquatic toxicity endpoints of TCS from published studies. .................. 207 9 4 Redefined hu man and ecological exposure pathways for land applied biosolids. ................................ ................................ ................................ .......... 209 9 5 Data utilized for the calculation of RfD for humans and animals. ...................... 210 9 6 Various parameters used for conducting the preliminary risk estimation. ......... 211 9 7 Equations used to calculate screening level hazard indices (HI) (considering no TCS degradation). ................................ ................................ ....................... 216 B 1 Average bioaccumulation factors (BAF) in the radish and lettuce leaves after excluding the highest treatment (Trt 3). ................................ ............................ 229 C 1 Limits of detection, quantitation and percent recoveries for TCS and Me TCS in various matrices. ................................ ................................ ........................... 230
14 LIST OF FIGURES Figure page 1 1 Chemical structure of Triclosan [TCS; 5 chloro 2 (2,4 dichloro phenoxy) phenol, CAS 3380 34 5] ................................ ................................ ..................... 33 2 1 Representative biosolids TCS concentrations collected from various sources.. ................................ ................................ ................................ ............. 39 3 1 Calculated octanol water partitioning coefficient (log K ow ) curve and dissociation diagram for TCS. ................................ ................................ ............. 52 4 1 Schematic of the biodegradation experimental design (Adapted from Snyder, 2009) ................................ ................................ ................................ .................. 76 4 2 Mean percent recoveries (n=4) of 14 C TCS in various fractions from biosolids amended Immokalee fine sand (IFS) ................................ ................... 77 4 3 Mean percent recoveries (n=4) of 14 C TCS in various fractions from biosolids amended Ashkum silty clay loam (ASL). ................................ ............. 78 4 4 Mean cumulative CO 2 production standard error bars from biosolids amended (a) Immokalee fine sand (IFS) (b) Ashkum silty clay loam (ASL) soil. ................................ ................................ ................................ ..................... 79 4 5 Typical RAD (top) for (a) week 0 extracts, (b) week 3 18 extracts. ................................ ......... 80 4 6 C hemical structure of Triclosan and Methyl triclosan ................................ ....... 80 4 7 Primary degradation half life (d) estimated using a zero order model from the proportions of 14 C detected as TCS an d Me TCS for biosolids amended soils ................................ ................................ ................................ ................... 81 5 1 The earthworm survival (%) as affected by the biosolids borne TCS concentration and the duration of earthworm ex posure, range finding test ...... 104 5 2 The mean earthworm survival (%) (n=4) standard deviation as affected by the biosolids borne TCS concentration and the duration of earthworm exposure, definitive toxicity test. ................................ ................................ ....... 105 6 1 Total CO 2 evolution (mg) over various times in the (a) IFS and (b) ASL soils amended with biosolids spiked with a range of TCS concentrations, unre plicated range finding test. ................................ ................................ ........ 127
15 6 2 NH 4 + NH 3 concentrations (mg kg 1 ) over time (Days 0 28) in the (a) IFS and (b) ASL soils amended with biosolids spiked with a range of TCS concentrations, unreplicated range finding test. ................................ ............... 128 6 3 N0 3 N0 2 N concentrat ions (mg kg 1 ) over time (Days 0 28) in the (a) IFS and (b) ASL soils amended with biosolids spiked with a range of TCS concentrations, unreplicated range finding test. ................................ ............... 129 6 4 Mean total CO 2 (mg) as a function of TCS concentrations and time (Days 0 28) in (a) ASL and (b) IFS soils (like letters indicate no significant difference between treatments), definitive test. ................................ ................................ 130 6 5 Mean (a) NH 4 N and (b) N0 3 N0 2 N concentrations (n=3) as a function of biosolids TCS concentration and time (Days 0 28) in ASL soils ....................... 131 6 6 Mean (a) NH 4 N and (b) N0 3 N0 2 N concentrations (n=3) as a function of biosolids TCS concentration and time (Days 0 28) in IFS soils ....................... 132 6 7 Average (n=3) well color development (AWCD) for the substrate types in the various soil sa mples developed 40 days after the incubation of the biolog plates. ................................ ................................ ................................ ............... 133 7 1 Representative photos of lettuce (A and B) and radish (C and D) plants that compare plant growth in the control and treatments. ................................ ...... 161 7 2 Representative photos of bahia grass that compare plant growth in various TCS treatments. ................................ ................................ ............................. 162 8 1 Percent recovery of the amount of water that was collected at bottom of the column at various int ervals during the study period ................................ ......... 181 8 2 Chloride breakthrough curves in control and biosolids amended so ils of the (a) empirical and (b) stochastic irrigation regimes. ................................ ........... 182 8 3 Prediction of chemical movement obtained using the CMLS mo del. The graph represents the depth of DCPA movement ................................ .............. 183 9 1 Log of predicted TCS concentrations (mg kg 1 ) assuming no TCS loss, and expected TCS concentrations (considering degradation) ................................ 218 B 1 Comparison of bioaccumulation factors in radish root (below grou nd) and lettuce leaves (above ground) ................................ ................................ .......... 229
16 LIST OF ABBREVIATIONS BAF QSAR Bioaccumulation Factor Quantitative Structure Activity R elationship BASL4 Biosolids amended soil level four model bw Body weight ECOSAR Ecological Structure Activity Relationships HPLC High Performance Liquid Chromatography LC Lethal concentration LC/MS Liquid Chromatography Mass Spectrometry LD Lethal dose LO A EC Lowest O bserved Adver se Effect Concentration LOD Limit of Detection LOEC Lowest O bserved Effect Concentration LOQ Limit of Quantitation MWRDGC Metropolitan Water Reclamatio n District of Greater Chicago NO A EC No Obs erved Adverse Effect Concentration NOEC No O bserved Effect Concentration OPPTS Office of Prevention, Pesticides and Toxic Substances PBT Persistenc e, Bioaccumulation and Toxicity QSAR Quantity S tructure Activity Relationship QSPR Quantitative S tructure Property Relationship RfD Reference dose TNSSS Targeted National Sewage Sludge Survey USEPA United States Environmental Protection Agency wk Week WWTPs Wastewater treatment plants
17 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of th e Requirements for the Degree of Doctor of Philosophy FATE, TRANSPORT, AND RISK ASSESSMENT OF BIOSOLIDS BORNE TRICLOSAN (TCS) By Manmeet Waria May 2011 Chair: Cochair: Gurpal Toor Major: Soil and Water Science Triclosan (TCS) is an antimicrobial compound used in many personal care products, and is a common constituent of domestic wastewater. Removal rates of TCS in wastewater treatmen t plants are typically >95% and most of the TCS partition s (accu mulate s ) in sludge solids. S ludge processed to produce biosolids is often land applied and can transfer TCS to agricultural soils The objectives of this research were to evaluate the environmental fate and ecological effects of biosolids borne TCS following biosolids land application A s p reviously published TCS risk assessments were based on the data derived from models or unpublished sources, we used measured TCS concentrations, propert ies, fate, and transport data to conduct a more realistic risk assessment The TCS concentrations measured in representative biosolids were 1 to 40 mg kg 1 (mean = 18 12 mg kg 1 ). Water solubility measured at neutral pH was 9 mg L 1 and increased to nearly 800 mg L 1 at two pH units above the pK a of ~8 The log K d of TCS in b iosolids was 4.15 0.03 and log K oc was 4.68 0.07 A TCS biodegradation study determined a prim ary degradation half life of ~100 d and a deg radation pr oduct (Methyl TCS) was i dentified. 105
18 mg kg 1 did not affect the earthworm survival suggest ing an estimated lethal TCS concentration (LC 50 ) >105 mg kg 1 B ioaccumulation factors ( BAF ) in earthworms were 4.3 to 12 depending on soil texture. Investigation of TCS effect s on microbial processes suggested that biosolids TCS conce 500 mg kg 1 had no significant effect on the microbially mediated processes. Plant bioaccumulation of TCS was minimal in radish (BAF = 0.004) and ba hia grass leaves (BAF = <0.001), but greater in lettuce leaves (BAF = 0.04) and radish roots (BAF = 0.43). No signifi cant leaching of TCS occurred beyond the biosolids incorporation d e pth of 0 to 2.5 cm in amended soil. A p relimi nary (multi pathway, multi target) risk assessment identified American woodcock as the most sensitive species. A t ier 2 assessment, utilizing less con servative estimates, suggested minimal risk of biosolids borne TCS to human and environmental health, when biosolids are land applied in a sustainable manner
19 CHAPTER 1 INTRODUCTION AND PRO JECT OBJECTIVES Background Chemicals from domestic, municipal, industrial, or agricultural sources that are present in th e environment, but not commonly monitored are termed emerging contaminant s es not infer that the chemical s are new, but that the i nterest of scientific community in the chemical s is recent (Aga, 2009). Such chem ic als have been recently detected in a variety of matrices in cluding surface water, groundw a ter, sediments, and biosolids as a result of improved analytical capabilities. One important group of ECs includes t he pharmaceuticals and personal care products (PPCPs) The primary route of PPCPs entry into the environment is through human use. With exp anding population and increase d human use of PPCPs, the appropriate treatmen t and disposal of the se chem ic als after use is becoming a cau s e of concern. Wastewater produced by domestic use enters and is treated in wastewater treatment plant s (WWTP s ). The products of the WWTP s are reclaimed liquid (effluent) and solid ( sludge). Th e liquid is typically discharged to surface waters and the solid, after process ing to reduce water content and pathogens, is called biosolids and is often suitable for land applica tion B iosolids can contain a variety of ECs (such as PPCPs), due to stability and sorption of ECs to organic rather than aqueous fraction during the WWTP processing. Options f or biosolids disposal include land application, incinerat ion and landfill disposal. The nutrient rich and organic nature of biosolids makes it a valuabl e resource for land application to improve soil fertility and is currently considered the most suitable way of biosolids disposal (Epstein 2002). Despite the be nefits of land appl i c ation
20 careful mo nitoring is essential to ensure the safe and sustainable reuse of biosolids T he USEPA conducted four surv eys over the years (1982 2009) to identify contaminants in biosolids for possible regulatory action One purpose of the most recent survey (USEPA, 2009 a ) was to obtain information on the presence and levels of certain contaminants of emerging concern (such as PPCPs). Triclosan (TCS) was included among several hundred (n = 145) other chemicals such as poly cyclic aromatic hydroc arbons (PAHs), polybrominated di phenyl ethers (PBDEs), antibiotics, drugs, hormones and steroids. The survey results suggested th at land application of biosolids can transfer several of the chem i c als, including TCS, to the terrestri al environment in sizeable amounts. Triclosan [TCS, Irgasan DP 300 (trade name); 5 chloro 2 (2,4 dichloro phenoxy) phenol, CAS 3380 34 5] is an antimicrobial compou nd active against gram positive, gram negative b acteria fungi and viruses Triclosan is added to a variety of personal care produc ts, including soaps, detergents, and cosmetics for its sanitizing properties (Heidler and Halden, 2007). E ffectiveness of TCS as an antimicrobial agent is attributed to inhibition of the enzyme enoyl acyl carrier protei n reductas e that is involved in bacterial lipid biosynthesis (Levy et al., 1999; Heath and Rock, 2000). Brausch et al. (2010) suggested that algal sensitivity to TCS is due to the disruption of lipid biosynthesis, membrane destab lization (Lyrge et a l., 2003; Franz et al., 2008) and uncoupling of phosphorylation (Newton et al., 2005). The t ypical range of TCS concentrations in personal ca re products is 0.1 to 0.3 % (w/w), which is sufficient to inhibit the activity of bacteria, molds, and yeasts (McAvo y et al., 2002).
21 Effects of TCS human exposure through personal care products are well studied. Acceptance of wide usage of TCS containing personal care products is based on the reports of no adverse effects of TCS on humans, except skin irritation and de rmatitis (Dayan, 2007) However, the use of TCS containing products for h ousehold purposes and skin care was implicated in the occurrence of TCS in human breast milk (<20 300 g kg lipid wt 1 ) (Adolfsson Erici et al., 2002; Ye et al., 2005), human urine (2 .4 3790 g L 1 ) (Calafat et al., 2008), and blood (0.010 to 38 ng g 1 ) (Allmyr et al., 2006). Dayan (2007) conducted a human risk assessment in children exposed to TCS containing milk. E ven conservative estimate s (highest reported TCS concentration in consumed milk) of exposure represented minimal risk to children fro m the TCS present in the milk (Dayan, 2007). Other adverse effects that can impact humans have been suggested. Wang et al. ( 2004 ) reported that TCS c an act as a selective inhibitor of sulfotransferase and gl ucuronosyl transferase enzymes with an inhibitory concentration ( IC50 ) range of 430 to 2152 g L 1 in the human liver. The two enzymes catalyze the conjugation of xenobiotics, increasi ng their solub ility, and promoting excretion from the human bodies. James et al. (2010) suggested that TCS (concentration <0.0 3 g L 1 ) might endanger pregnancy in sheep by reducing the total placental estro gen secretion in target tissues critical for the maintenance of pregnancy A thre shold of toxicological concern of 30 g kg 1 d 1 was estimated to represent the highest dose to which a person could be exposed over a lifetime with no expected adv e rse health effect (EPHC, 2008). Several risk assessments conduct ed based on the toxicological, phramacokine tic and clinical data suggested limited toxic ity of TCS from personal care products ( Rodricks et al.,
22 2010; Barbolt, 2002; Gilbert and McBain, 2002; Moss et al., 2000 ) Thus, h uman effects are deemed inconsequential, and a re not directly studied here Following use, much of the unused T CS becomes a co mponent of domestic waste water. The products of WWTPs are liquid s ( effluents ) and solid s ( sludge ). P rocessed sludge forms biosolids, which may then be land applied. Despite monitoring data demonstrating significant TCS removal (98 1 %, Heidler and Halden, 2007) in activated sludge ( consists of aerobic conditions promoting biological processes) systems, the relative roles of biodegradation an d sorption in TCS removal from influents are not conclusively documented (Heidler and Halden, 2007; McAvoy et al., 2002; Sabaliunas et a l., 2003; Singer et al., 2002). Bester (2003) reported 30 % TCS sorption on sludge, whereas Heidler and Halden (2007) est imated that 50 19% of TCS accumulated in the sludge and 48 19% was biotransformed in a full scale activated sludge treatment unit. The 2002 USGS survey of 139 streams expected to be influenced by WWTPs effluent detected TCS in over 60% of the streams at a median concentration of 0.14 g L 1 1 (Kolpin et al., 2002). Triclosan was one of the five most frequently detected contaminants, behind widely recognized contaminants such as coprostanol, cholesterol, N, N diethyl toluamide, and caffeine R ecently, Brausc h and Rand (2010) reported the detection of TCS in 57 % of the surface water (rivers, streams) samples with a median concentration ~ 0.05 g L 1 Benotti et al. (2009) surveyed 19 drinking water treatment plants where the source water was impacted by wastewater and reported TCS median concentration of 0.00 3 g L 1 (detected in six plants) in source water, and 0.0012 g L 1 (detected in one plant) in finished drinking water.
23 Detection of TCS in aqu eous systems is of concern because TCS can photo degrade to dioxins (Latch et al., 2005), and the binuclear aromatic ring of TCS (Fig ure 1 1 ) is also present in dioxins Halden and Paull (2005) and Heidler and Halden (2007) opined that the structural simil arity suggests that TCS may persist and bioaccumulate in a manner similar to dioxins. Adverse effects of TCS in various aquatic species are illustrated in T able 1 1. A s pecies sensitivity distribution (SSD) diagram is another approach to identify the aquatic risk of TCS Capdevielle et al. (2008) conducted a risk assessment of TCS in freshwater environment s using the SSD approach The toxicity distribution was constructed based on chronic toxicity data for several (n=14) aquat ic species, and represents a more realistic threshold of toxicity values (as it considered multiple species) as compar ed to no effect concentration approach based on a single most sensitive species (e.g ., data i n Table 1 1) Ratio s of the p redicted envir on mental concentration (PEC) to the predicted no effect concentration (PNEC) ratios were <1 for most aquatic populations, suggesting low risks to even the most sensitive aquatic species (Capdevielle et al., 2008) Thus, the r isk to sensitive species is low e ve n at the highest likely exposure, which typically occurs immediately downstream of WWTP s Capdevielle et al. ( 2008 ) opined that, at current TCS usage, TCS is not expected to advers ely impact aquatic species Although the TCS aquatic effects are important based on the toxicity data (Table 1 1) the Capdevielle et al. (2008) study results suggested minimal aquatic TCS effects. Therefore, t he focus of the present study is on the occure nce of TCS in soils impacted by land ap plied biosolids. I n the US, WWTPs g enerate approximately 7 milli on Mg of biosolids each year (USGS, 2008), of which 63 % (National Research Council, 20 02) is land applied.
24 Biosolids TCS concentratio ns have been reported in several recent studies ( Langdon et al., 2011; McClellan and Halden, 2 010; Cha and Cupples, 2009; Xia et al., 2009; Heilder and Halden, 2009; USEPA, 2009a ) and suggest a representative concent ration range of 10 to 20 mg kg 1 in typical biosolids. Reported TCS concentrations in the extensive (78 WWTPs ) 2009 Targeted National Sewage Sludge Survey ( TNSS S) (USEPA, 2009a) were 0.4 to 133 mg kg 1 with an overall mean of 16 65 mg kg 1 (including statistical outliers). The treatment plants selected for the survey received secondary treatment or better, but the final products includ ed some materials that were not processed to meet land application standards with respect to disinfection and chemical removal usually obtained by tertiary treatment (USEPA, 2009a). Assuming a TCS concentration of 16 mg kg 1 (mean from TNSSS) in activated sludge, an estimated 1.5 6.5 10 4 kg year 1 TCS is land applied nationwide The mass estimate would differ depending on the actual concentration of TCS in the biosolids, but the practice of biosolids land application clearly represents a mechanism for intro ducing substantial amounts of TCS into the environment. R eports of TCS presence in earthworms (Kinney et al., 2008), algae, snails (Coogan et al., 2007; 2008) wet wt 1 ; Fair et al., 2009) highlight the ability of TCS to accumulate in a variety of organ isms. Additionally, s ome worry that the TCS present in the environment could contribute to the spread of anti bacterial resistance and could threaten human drug therapy (Biro sova and Mikulasova, 2009; Pycke et al., 2010; Rooklidge, 2004; McMurry et al., 1998). Little is known about the risk of biosolids borne TCS that is land applied. Risk quantification of a chemical requires information on the toxi city of the chemical, as w ell as exposure. The presence of chemical in the environment is not a hazard unless
25 humans or other organisms are exposed to concentrations suffici ent to induce an adverse effect. Further, the adverse effect will vary with exposure time s, and depend on whe ther an exposure is chronic or acute especially if a chemical has endocrine effects. The environmental fate and transport of biosolids borne TCS is parti cularly important to identify the relevant exposed populations and to conduct meaningful environmental a nd human health risk assessment s Various studies characterized components of TCS behavior in non biosolids amended (un amended) soils but few studies focuse d on biosolids amended soils The average half life of TCS estimated from studies (Kwon et al., 2010; Lozano et al., 2010; Higgins et al., 2011; Walters et al., 2011) conducted in amended soils was generally >100 d i.e ., TCS was deemed persistent Further, various s tudies (e.g. Miyazaki et al., 1984; McAvoy et al., 2002; Lindstrom et al., 2002, Coogan et al., 2007; 2008) suggested that TCS was methylate d to Me TCS in effluents, biosolids, fish bodies, algae, and snails However, it is uncertain if Me TCS is formed in biosolids amended soils, and if it persists or is further mineralized to CO 2 Limited data available for Me TCS partitioning suggest that Me TCS is more hydrophobic ( log K ow = 5.2 ) than TCS ( log K ow = 4.8 ) (Boehmer et al., 2004 ), and likely to be less bio available than TCS. E arthworm toxicity da ta in un amended soils suggest minimal toxic ity of TCS to earthworms ( Higgins et al., 2011 ). T he bioaccumulation data were variab le in earthworms grown in TCS spiked un amended soil, and in biosolids amended soil creating a need for a definitive assessment Waller and Kookana (2009) and Butler et al. (2011) suggested inhibitory effects of TCS on some microbially mediated reactions and on microbial community structure in soil (no biosolids). Triclosan may behave
26 dif ferently in biosolids amended soil due to reduced bioava i lability of TCS and needs further investigation. Plant TCS accumulation was reported (Wu et al., 2010) in a single crop (soybean Glycine m ax ) grown in a biosolids amended soil but the study was c onfounded by the u se of biosolids with extremely low solid s content (19 g L 1 vs avg of 300 g L 1 ). type of biosolids applied to the soils (Edwards et al., 2009). Model estimated leaching potential of TCS was minimal in amended soil s (Cha and Cupples, 2010). A nother antimicrobial chemical, [ i.e ., triclocarban (TCC ) ] was m inimal ly phytoaccumulated and leach ed (Snyder et al., 2011 ) in a biosolids amended soil. The s a me study reported formation of TCC bo und (non extractable) residues. We speculated that because TCS has a reported range of low water solubility values ( 1.9 17 mg L 1 ) and because the partitioning coefficient of TCS (log K oc = 4.3 ) is even greater than TCC (log K oc = 3.88 ) (Agyin Birikorang et al., 2010), TCS should tend to partition onto soils and sediment s. Fuhr et al. ( 1998) defined b ound residues compounds in soil, plant, or animal which persist in the matrix in the form of the parent substance or its metabolite(s) after extractions. The extraction method must not substantially change the compounds themselves or the structure of the matrix Due to high partitioning coefficient of TCS (log K d = 2.3), it is expected to form bound residues and have minimal potential of leaching and phytoaccumulation in biosolids amended soils, as with TCC. S everal ecological risk assessments were conducted recently (R eiss et al., 2009; Fuchsman et al., 2010; Langdon et al., 2010) that addressed the risk of TCS from biosolids amended soils. Langdon et al. (2010) quantified the risk to aquatic organisms from surface runoff
27 and leaching of T CS from biosolids amended soil s and found some hazard to the most sensitive aquatic species; but, acknowledged that the risk might have been overestimated. Both Reiss et al. (2009) and Fuschman et al. (2010) suggested minimal risk to terrestrial organisms from bioso lids borne TCS but m ost of the data utilized in the assessment s were derived from models or extracted from unpublished sources. Thus, a characterization of biosolids borne TCS fate, transport and risk asse ssment based on measured data is necessary. Based on the available literature for TCS, a ppropriate hypotheses f or the project were formulated as follows: Biosolids borne TCS and its metabolites are persistent (half life >60d) in the environment. Biosolids borne TCS forms bound residues of limited bio and environmental l ability. Exposure to b ios olids borne TCS poses minimal risk to soil micro and macro organisms. Biosolids borne TCS has minimal mobility (leaching tendency ) and phytoavailability. Biosolids borne TCS poses minimal risk to human and environmental health. The ultimate goal of our study was to perform a human and ecological health risk a ss essment of biosolids borne TCS. The following i ntermediate objectives were designed to test the first four hypothese s and to obtain data required to fulfill the ultimate ob jective. Objective 1: Quantify TCS Concentration s in B iosolids The la rgest data base of biosolids TCS concentrations is the 2009 TNSSS (USEPA, 2009a), representing 78 WWTPs across the U S. Reported TCS concentration s range f rom 0.4 to 133 mg kg 1 (mean = 16 65 mg kg 1 including statistical outliers ) The large variability in the TNSSS data likely represent s inclusion of
28 some unique WWTPs producing sewage sludge with exceptionally hig h (minimally processed) and low (probably processed by tertiary treatment) product concentrations of TCS. Our objective was to determine the TCS concentration in additional biosolids Biosolids analyzed in our study included some from the TNSSS (USEPA, 2009a), and some from WWTPs managed by Metropolitan Water Reclamation Distri ct of Greater Chicago (MWRDGC). The biosolids from MWRDGC were chosen as the D istrict serves a large metropolitan area and because the MWRDGC biosolids TCS concentrations were expected to be modest. The low initial concentrations were particularly useful for our studies as we tested the effect of variable (spiked) TCS concentrations on various microbial phenomena ( microbial respiration, ni trification and ammonification ) (Chapter 2) Objective 2: Determine/V erify Basi c Physico C hemical P ropert ies of TCS C w ater solubility and partitioning coefficients (K ow K d and K oc ) are critical to characterizing the fate and transport of environmental contaminants. S olubility and partitioning behavior relate to a chemical s availability for degradation, leaching, plant uptake and bioaccumulation in organisms. Conflicting TCS solubility data exist in the literature, making accurate predictions of environmental fate and transport problematic. T he water solubility of TCS was measured at various pH values as TCS is a weak acid and is expected to be more soluble at pH values greater than its acid dissociation constant ( pK a ) In addition, the partitioning coefficient of TCS in various biosolids was determined (Chapter 3) Agyin B irikorang et al. (2010) measured the partitioning coefficients (K d and K oc ) of TCS in biosolids and biosolids amended soils. But the coefficients of TCS inherent to biosolids have not been determined. The inherent coefficients are likely to be greater than the coefficients in spiked system s as
29 inherent TCS is expected to be thoroughly incorporated and evenly distributed within the organic component of the biosolids. Objective 3: Determine the Degradation (Persiste nce) of Biosolids Borne TCS tence is fundamental to estimating the chemical fate a nd transport in the environment. Further, chemicals may degrade to form metabolites that may be more hydrophilic or lip ophilic than the par ent and behave differently Persistence of TCS in aerobic soils and sediments is reasonably well studied, but TCS degradation in biosolids amended soils is less studied. A f ew studies reported degradation of TCS in amended soils, but the half lives were ba sed on the loss of extractable TCS with time and may not be accurate as l oss of extractable TCS may simply represent the conversion of extractable TCS to non extractable ( bound residue ) forms. We utilized biosolids spiked with 14 C TCS to determine the deg radation (mineralization/ primary degradation) potential of TCS in two biosolids amended soils, identified the metabolites and assessed the bound residue formation (Chapter 4). Objective 4: Determine the I mpact s of Biosolids Borne TCS to Soil Organisms Accurate estimates of TCS toxicity and bioaccumul ation are critical to conduct ecological and human health risk assessments. A f ew researchers [e.g., Kinney et al. (2008); Higgins et al (2011)] have directly addressed the effect of biosolids borne TCS on terrestrial organisms. However the results were confounded either by the detection of significant amounts of TCS in the control soil or by the lack of sufficient number of replicates. We conducted a laboratory study (USEPA, 1996b) using spiked biosolids t o investigate the toxicity and bioaccumulation of biosolids borne TCS to earthworms. Further, long term bioavailability of TCS, was estimated by utilizing earthworms recently
30 collected from a field soil amended with biosolids two years earlier at a high ap plication rate (Chapter 5). Objective 5: Determine the Toxicity of Biosolids Borne TCS on Microbial Reactions Triclosan can a ffect the rate of microbial reactions like respiration and nitrification in un amended soil system s (Waller and Kookana, 2009; Butler et al., 2011). To our knowledge, however, potential impacts of biosolids borne TCS on soil micro organisms or microbially mediated reactions have not been published. We examined the impacts of biosolids borne TCS on micro bial processes using USEPA ( 1996c) methods in two soi ls with contras ting physico chemical properties. Further, changes in microbial counts and community structure due to biosolids borne TCS are not known and may affect ecosystem processes (such as nutrien t recycl ing) and the effectiveness of microbial invasions (such as growth of pathogens) (Garland, 1997). We utilized several sets of biosolids amended soils to perform bacterial count s and to quantify changes in microbial community structure. A direct bact erial count method (Matsunaga et al., 1995 ) and biolog ECO plates were utilized to quantify the effect s of biosolids borne TCS ( Chapter 6 ). Objective 6 : Quantify the Phytoavailability of Biosolids B orne TCS L iterature provides evidence of plant acc umulation of polar and non polar organic compounds A f ew studies of TCS toxicity and accumulation were conducted in soil less cultures (Herklotz et al., 2010) un amended soils (no biosolids) (Liu et al., 2009 ) and saturated systems (wetlands) (Stevens et al., 2009) that suggested the possibility of TCS plant accumulation from land applied biosolids. Because chemical toxicity and bioaccumulation can vary with the species tested (Duarte Davidson and Jones, 1996)
31 we utilized four plant species representing monocotyledons ( monocots ), dicotyledons (dicots) above ground (leaves) and below ground (roots) biomass as well as grasses. A field equilibrated soil previously amended with a high biosolids application rate was utilized as the growth media for the plant s (Chapter 7). Objective 7 : Quantify the Leaching Potential of Biosolids B orne TCS D espite the occurrence of TCS in biosolids, and frequency of biosolids land application, little is known about the mobility of TCS in biosolids amended soils. The partition coefficient of TCS (log K oc = 4.26 ; Agyin Birikorang et al., 2010) measured in amended soils suggests that TCS has a propensity to partition onto soils and sediment s in the environment. Our study explore d the mobility of TCS in a b iosolids amended soil subjected to repeated irrigation s (Chapter 8 ). Ultimate Objective: Risk Assessment of Biosolids Borne TCS Several ecological risk assessments were conducted recently ( Reiss et al., 2009; Fuchsman et al., 2010; Langdon et al., 2010 ) but most of the data use d in the assessments were derived from models or extracted from unpublished sources.Thus, a characterization of biosolids borne TCS behavior, and a risk assessment based on meas ured data are necessary. Studies mentioned in each of the intermediate objectives are described in Chapters 2 to 8 Each intermediate objective is presented as a separ ate chapter in the dissertation, and each chapter has sp ecific hypothese s and objectives. Data accumulated from our studies and from published sources were used to identify various pathways of TCS exposure from biosolids l and application practices and to perform an integrated human and ecological risk assessment of biosolids borne TC S applied to
32 soils (Chapter 9). S uggest ions for the future direction of biosolid s borne TCS w ork are included.
33 Table 1 1. Toxicity end points of TCS for some aquatic species a no observable adverse effect level, b effective concentration, c lethal concentration, d inhibitory concentration Figure 1 1. Chemical structure of Triclosan [TCS; 5 chloro 2 (2,4 dichloro phenoxy) phenol, CAS 3380 34 5 ] Toxic effect Species Toxicity endpoint Reference Endocrine disruption Activation of the human pregnane X receptor >2870 g L 1 Jacobs et al. ( 2005 ) Toxic effects Rainbow trout ( Oncorhynchus mykiss ) NOAEL a of 71.3 g L 1 Orvos et al. ( 2002 ) Toxic effects Daphnia magna EC50 b of 390 gL 1 Orvos et al. ( 2002 ) Toxic effects Amphibian larvae ( Xenopus laevis ) LC 50 c of 259 gL 1 Palenske et al. ( 2010 ) Gene expression American bullfrog ( Rana catesbiana ) 0.15 1.4 g L 1 Veldhoen et al. ( 2006 ) Growth Freshwater microalga ( Pseudokirchneriella subcapitata ) 72 h IC50 d of 0.53 g L 1 Yang et al. ( 2008 ) Genotoxic and cytotoxic effects Zebra mussel hemocytes 0.692 ngL 1 Binelli et al. ( 2009 )
34 CHAPTER 2 BIOSOLIDS BORNE TCS CONCENTRAT IONS Background Numerous biosolids TCS concentrations are reported in the literature. The m ean TCS concentration of 12.6 3.8 mg kg 1 was reported for the composited biosolids archived from the 2001 Targeted National Sewage Sludge S urvey (TNSSS) (McClellan and Halden, 2010). The largest d atabase of biosolids TCS concentrations is the 2009 TNSS S report (USEPA, 2009a), and reports a TCS c oncentration range of 0.4 to 133 mg kg 1 (mean=16 65 mg kg 1 including statistical outliers ) representing 78 WWTPs sampled in the survey. The WWTPs selected for the survey received secondary treatment or better, but the final products included some mater ials not processed to meet land application standards (USEPA, 2009a). Some of the WWTPs in the survey were likely n ot as efficient in removing TCS. Thus, the large variability in the TNSSS data might represent some unique WWTPs producing sewage sludge with exceptionally high (minimally processed) and low product concentrations (probably processed by tertiary treatment) of TCS. Biosolids analyzed in our study included some from the TNSSS (USEPA, 2009a), and some from WWTPs managed by Metropolitan Water Recla mation Distri ct of Greater Chicago (MWRDGC). Our objective was to determine the TCS concentration in representative biosolids including the biosolids utilized in our experiments. Material and Methods Fifteen biosolids were analyzed to determine the TCS co ncentrations as described in Heidler et al. (2006). Four biosolids samples were collected by the MWRDGC personnel and promptly shipped to the laboratory on ice packs. The samples
35 were frozen ( 20 C) until analysis. We also utilized biosolids collected for a previous triclocarban project (biosolids also utilized in the TNSSS by USEPA, 2009) which were frozen at 20 C until analysis. Briefly, the biosolids samples were thawed, lyophilized, and 1 g (dry wt. equivalent) samples were extracted twice with 20 mL o f m ethanol (MeOH) + acetone ( 50:50 v/v) in triplicate. The samples were shaken for 18 h on a platform shaker, followed by sonication (2 h) (Branson 2210, Danbury, CT ; temp. 40 C, 60 sonication s min 1 ) The extract was centrifuged at 800 x g, dried under N 2 gas and reconstituted in 50:50 MeOH: Milli Q water. The samples were then fortified with TCS d7 internal standard and analyzed by Liquid Chromatography Mass Spectrometry (LC/MS) (TSQ Quantum, Thermo Scientific, Williston, VT, USA). Chromatography was per formed on Phenomenex Luna C18 column (3 m particle size, 2 100 mm; P henomenex, Inc., Torrance, CA) with sample injection volume of 5 L min 1 The m obile phase consisted of water: MeOH with 1mM ammonium acetate at a flow rate of 300 L min 1 The g radient consisted of 25:75 water: MeOH (held for 1 min ), increasing to 0:100 (water : MeOH) held for 5 min. and decreasing back to 25:75 water: Me OH (over 0.5 min ). Mass spectrometry was performed using selective ion monitoring in negative ionization mode. Linear calibration consisted of eight standard levels (0 500 ng g 1 ). Detection was based on a characteristic molecular ion TCS (m/z 287) as well as TCS with the naturally occurring isotope 37 Cl (m/z 289, 291). The TCS concentrations were corrected for the TCS recoveries (recovery = 64 5%) determined by spiking known 14 C TCS activities in the biosolids. The samples ran f or 12 min with an ave rage retention time of 8 min. The limit of detection (LOD) was 0.40 ng g 1 and limi t of quantitation (LOQ) was 1 .3 ng g 1 The LOD and LOQ values were calculated as 3 fold
36 and 10 fold, respectively, the standard deviation in the signal from multiple runs of the lowest calibration standard (S ignal/Noise >10) (USEPA, 1984). The details of detection limits and recoveries are pr ovided in the Appendix C. Results and Discussion The measured TCS concentrations in 15 biosolids were 0.40 to 40 mg kg 1 with an average of 18 12 mg kg 1 and a median concentration of 21 mg kg 1 (Table 2 1). The average and median concentration s are consistent with the mean 2009 TNSSS (USEPA, 2009a) value (16 65 mg kg 1 ), and the mean (12.6 3.8 mg kg 1 ) of composited samples archived from the 2001 TNSSS (McClellan and Halden, 2010). The majority of the biosolids analyzed in our study were anaerob ically digested, but differences in TCS concentrations could have occurred due to differences in d igestion periods (time), inputs, or dewatering methods (air dried vs cake) ( Table 2 1). The measured unpublished and published biosolids TCS concentrat ions ar e summarized in Figure 2 1. The Figure 2 1 illustrates variability in TCS concentrations (0.4 133 mg kg 1 ) in a variety of biosolids including the ones from our study and other studies from U S Canad a, Australia, Spain and Germany. A typical representativ e range of biosolids TCS concentration appears to be ~10 to 20 mg kg 1 (Figure 2 1), consistent with the mean TCS concentration (16 65 mg kg 1 ) reported in the 2009 TNSSS (USEPA, 2009a) Xia et al. (2010) found significantly greater med ian TCS concentrat ions in un composted (9.6 mg kg 1 ) biosolids than in composted (1.2 mg kg 1 ) biosolids obtained from 16 WWTPs located in Georgia, South Carolina, Colorado, Illinois, and Califo rnia. They also observed greater TCS concentrations in biosolids obtained from WWTPs serving residential areas than industr ial areas (Xia et al., 2010) Xia et al. (2010) results suggest
37 that TCS concent rations can vary widely, but the concentrations are consistent w ith the wide range of TCS concentrations reported in the 2009 TNSSS and in our study. Of particular interest were the biosolids ( anaerobically digested) obtained from MWRDGC, as the D istrict serves a large metropolitan area and much of the biosolids is land applied. The TCS c oncentrations averaged 6 mg kg 1 in air dried an d 7 mg kg 1 in centrifuge cake samples of Stickney plant biosolids. Calumet plant samples had average TCS concentrations of 0.4 0.0 mg kg 1 in air dried and 4.7 0.1 mg kg 1 in centrifuge cake samples. Our values for the Stickney plant are consistent with data for the TCS concentration (6.4 0.3 mg kg 1 ) of composited three year sampl es ( dewatered) reported by Higgins et al. ( 2011 ). Triclosan concentrations in all MWRDGC materials were less than the nationwide representative concentration range (~10 t o 20 mg kg 1 ) (Fig ure 2 1). If the samples of biosolids furnished to us or the concentrations reported for composited samples (Higgins et al., 2011) are truly representative of the products produced year round, the TCS concentrations in biosolids produced by major Chicago WWTPs can be regarded as modest to low. The concentrations are much smaller than values of 16 65 mg TCS kg 1 (USEPA, 2009a) and 30 11 mg TCS kg 1 (Heidler and Halden, 2007), suggesting that the initial environmental and human health co ncerns expressed by Heidler and Halden (2007) based on an assumed TCS concentration of 30 mg kg 1 may be over estimated. The lower concentrations of TCS in biosolids from Stickney and Calumet WWTPs however, do not necessarily mean minimal ri sk to human or the environmental health. Concerns about biosolids borne persistence, environmental fate, phytoavailability and mobility require quantification, as outlined in the later chapters.
38 Table 2 1. Triclosan (TCS) concentrations in fifteen biosolids (n = 3) ob tained from wastewater treatment plants across the U.S. Biosolids Identification Treatment Process TCS content mg kg 1 UNKB Anaerobic digestion 33 0.7 UNKC Anaerobic digestion (33 d) 21 0.8 UNKD Anaerobic digestion 40 3 UNKE Anaerobic digestion 22 0.4 UNKF Anaerobic digestion 20 1 UNKG Anaerobic digestion 31 0.6 UNKH Anaerobic digestion 25 1 UNKI Anaerobic digestion 1 0.1 UNKJ Unknown 22 0.4 UNKK Unknown 23 1 UNKL Unknown 11 2 CHBC Stickney water reclamation plant, Chicago Anaerobic digestion (air dried) 6 CLBC Stickney water reclamation plant, Chicago Anaerobic digestion (centrifuged cake) 7 0.3 CHAD Calumet water reclamation plant, Chicago Anaerobic digestion (air dried) 0.4 0.0 CHCC Calumet water reclamation plant, Chicago Anaerobic digestion (centrifuged cake) 4.7 0.1 Overall average 18 12 whose locations are unknown
39 Figure 2 1. Representative biosolids TCS concentrations collected from various sources. The n represents the number of biosolids samples. For McCle llan and Halden (2010) n = 110 representing composite biosolids from 94 wastewater treatment plants. Represents the 95th percentile value for TCS concentration in the 2009 TNSSS (USEPA, 2009a) Represent our qualitative asse ssment of the range of typical representative TCS concentrations in biosolids. 0 20 40 60 80 100 120 140 TCS (mg kg 1 ) 25th Percentile Minimum Mean Maximum 75th Percentile
40 CHAPTER 3 BASIC PHYSICO CHEMICAL PROPERTIES OF TCS Background Chemical solubility and partitioning coefficients (K ow K d and K oc ) are important factors for characterizing the fate and transport of environmental contaminants. Relationships between solubility, K ow and K oc are often (log log) linear, but the specific relationship varies with the class of compounds (Lyman, 1990). The relationships can be used to calculate K oc from the available K ow and solubility data. In general, adsorption of a contaminant is directly related to the partitioning coefficient and inversely related to contaminant solubility (Comfort et al., 1994). Water solubility is an important chemical property that strongly influenc es a 2002). Conflicting TCS solubility data exist in the literature, making accurate predictions of environmental fate and transport of TCS problematic. Solubility values c alculated from the Ecological Structure Activity Relationships (ECOSAR) model and PBT (Persistence, Bioaccumulation and Toxicity) Profiler vary from 1.97 4.6 mg L 1 ( Halden and Paull, 2005) (Table 3 1). A ctual solubility measurements vary from 10 mg L 1 (C iba Speciality Chemicals, 2001a) to 17 mg L 1 (MITI, 1992). The variability in the reported and the modeled values of solubility requires resolution. The pK a value of TCS is reported as 8.14 (Jakel, 1990) and TCS water solubility is expected to vary signif icantly with pH because the weak acid dissociates to a more soluble form as pH approaches and exceeds the pK a of TCS. The octanol water partition coefficient (K ow ) is the ratio of the concentration of a chemical in octanol and in water at equilibrium at a specified temperature. Octanol is an
41 organic solvent regarded as a surrogate for hydrophobic lipids The K ow value is used in many environmental models, e.g., Bioaccumulation Factor Quantitative Structure Activity Relationship (BAF QSAR) and Quantitative Structure Property Relationship (QSPR) to estimate the fate of chemicals in the environment. The K ow also provides an web (Dimitrov et al., 2003). The log octanol/water part ition ( K ow ) coefficient of TCS is calculated to be 4.8 (at 25 C, pH 7) using K ow WIN model ( Halden and Paull, 2005) (Table 3 1). The model estimates log K ow values of organic chemicals using an atom/fragment contribution method develo ped by Syr acuse Research Corporation Experimentally determined values of TCS are similar: log K ow 4.76 (Lei and Snyder, 2007; Wezel and Jager, 2002), and 4 .70 (Ying et al., 2007) (Table 3 1). Although the collective data are in close agreement, the pH values at which th e K ow measurements were made are not known and K ow a values. We calculated TCS dissociation and log K ow values as a function of pH using an equation that estimates the log K ow of chlorophenols with varying pH (Nowosielski and Fein, 1998). Calculations suggest decrease s in log K ow of TCS at pH values above pK a The K ow vs pH curve (Figure 3 1) closely resembles an acid dissociation curve, but is o ffset to higher pH values. L og K ow values are minimally affected at pH pK a but a significant decrease at greater pH values. The predicted log K ow becomes pH a +3, with a value 2.9 log units le ss than normally assessed (Fig ure 3 1). A decrease in log K ow values at higher pH predicts lower concentrations partitioned to biosolids and greater concentration in the aqueous phase during sewage treatment involving lime stabilization
42 (pH = 12). However, such high pH values are not common in most conventional sewa ge treatment or biosolids amended soil systems. The consistency of log K ow literature values (4.8 and 4.76) and the negligible change in log K ow values predicted at pH values (<7) represen tative of most sewage treatment systems (or amended soils) convinced us that it was unnecessary to conduct measurement of K ow at different pH values. However, variation of TCS solubility with pH has not been addressed. Therefore, w e measured the TCS solubility at pH = pK a (8.14) and 2 units above (10.14) and below (6.14) t he pK a Mobility of TCS in biosolids amended soil determines the potential for soil and groundwater contamination. The relatively high log K ow (4.8) of TCS suggests extensive retention in (or on) biosolids and limited transport in soil systems but, mobil ity may vary with the soil pH Recent data obtained for a sandy loam and a silt loam soil (no added biosolids) using spiked TCS concentrations of 0.2, 0.5,1 and 2 mg L 1 suggest relatively low values of log K d of ~2.4 and log K oc of ~2.3 (Karnjanapiboonwo ng et al., 2008). Other estimated and measured values of log K oc reported are much greater: 4.26 (estimated, Ying et al. 2007) and 4.6 (measured, Heim, 1997), and a log K d of 4.3 (measured Heim, 1997). The partitioning coefficients of added TCS in biosolid s (K d and K oc ) or in biosolids amended soils have not been determined. Alternatively, K d and K oc d and K oc WWTPs is present throughout the entire biosolids pr solid fraction from the time TCS enters the waste stream to eventual biosolids land application. As such, biosolids borne TCS is expected to be thoroughly incorporated
43 and evenly distrib uted within the organic component of the biosolids. A spike, alternatively, is added to the exterior of the matrix and allowed to react after the biosolids are produced. A TCS spike is not expected to sorb as completely or uniformly as an inherent componen t of biosolids within the time frame of a typical partitioning experiment (<24 h). A K d value estimated using a spiked matrix could then underestimate the true solid phase partitioning of hydrophobic compounds, such as TCS. Inherent K d values were measured for a similar compound TCC (Triclocarban) in several biosolids and differed from spiked K d values (Snyder, 2009). We measured the inherent K d and K oc values of TCS in various biosolids. Material and Methods Solubility Water solubility was measured accord ing to the EPA standard method guideline of Office of Prevention, Pesticides and Toxic Substances ( OPPTS ) 830.7840 ( USEPA, 1996a ). The OPPTS is now renamed to Office of Chemical Safety and Pollution Prevention (OCSPP) but the guideline remains identified as OPPTS, and is u sed throughout the dissertation. The EPA guidance describes three major methods (shake flask, generator c olumn, column elution) for the solubility measurement depending on the expected solubility of the compound. The sh ake flask method is intended for compoun ds with solubilities > 10 mg L 1 whereas the column elution method is intended f or compounds with solubilities 10 mg L 1 The literature contains variable TCS solubility values, but measured values are reportedly >1 0 mg L 1 Thus, the simple shake flask metho d was employed in our study. Solubility measurements were performed at pH values two units above and below the pK a as well at pH = pK a A preliminary test was performed to estimate the quantity of TCS necessary to saturate
44 the desired volume of water. Approximately 5 the quantity of material determined in the preliminary test was weighed into nine Teflon tubes (3 replicates for each pH value) with stoppers. Water (30 mL) was added to the 40 mL tubes and the pH of the water was adjusted to 6.14, 8.14 and 10.14. The pH adjustment was performed by adding acid (HCl) or base ( KOH) to attain target pH values. The tubes were then agit ated at ro om temperature (20 25 C). Additional monitoring of pH was conducted every 12 h during the initial 24 h agitation and at each 24 h after wards. At sampling times of 1, 2 and 3 d the tubes were removed and 1mL of the aliquot was collected The 1 mL of water r emoved was replaced by adding 1 mL additional water to maintain constant volume of the solution. The removed aliquot was centrifuged (8000 g) for 10 min and the TCS concentration in the clear aqueous sample was measured. Samples were quantified on LC/MS (Thermo scientific discovery max TSQ Quantum) in negative ionization mode. Chromatography was carried out on C18 column (5 m particle size, 2.1 x100 mm; Phenomenex ). Mobile phase consisted of water: MeOH with 1mM ammonium acetate at a flow rate of 300 L min 1 The gradient consisted of 25:75 water: MeOH (held for 1 min), increasing to 0:100 (water : MeOH), held for 5 min. and decreasing back to 25:75 water: MeOH (up to 12 min). Linear calibration consisted of eight standard levels (0 500 ng g 1 ). The TCS quantification was performed by the isotope dilution method using the 13 C 12 internal sta ndard with a run time of 12 min The ion monitoring was performed at m/z values of 287 for TCS and 293 for 13 C 12 TCS. The m/z of 289, 291, 295 and 297 were used as qual ifying ions due to the presence of naturally occurring 37 Cl atoms in the organic molecules. The l imit of detection ( LOD ) was 0.4 ng mL 1 and the l imit of quantitation ( LOQ ) was 1 ng mL 1 The LOD and LOQ values were calculated as 3 fold
45 and 10 fold, respect ively, the standard deviation in the signal from multiple runs of the lowest calibration standard (S ignal/Noise >10) (USEPA, 1984). The details of detection limits and recoveries are provided in the Appendix C. Partitioning Coefficients (K d and K oc ) The K d value of TCS was previously measured ( Agyin Birikorang et al., 2010) according to the EPA standard method guideline for soil and sediment (OPPTS 835.1220) (USEPA, 2007). The measurement was performed in biosolids, soils and biosolids amended soils. Detail s are described by Agyin Birikorang et al. ( 2010) Briefly, 0.8 mL of the stock solution (1.29 10 5 dpm 14 C TCS mL 1 ) was utilized to prepare a working solution (250 mL) in 0.01 M CaCl 2 The batch experi ment involved addition of 2.5 mL of the 14 C TCS work ing solution to triplicate 0.5 g (oven dry equivalent) samples of the biosolids (1:5; g solid: mL solution) + Control samples containing TCS spiked CaCl 2 (no biosolids). Ten biosolids used in TCS concentration determination (Chapter 2) were used in the partitioning study along with 5 additional biosolids received as a part of the USEPA TNSSS (USEPA, 2009a) The samples and controls were analyzed for 14 C TCS throughout the experiment to monitor the stability of the TCS concent ration in the solution phase. No, or minimal, degradati on was expected during the 24 h study given the estimated half life of TCS in biosolids amended soils (100 d) ( C hapter 4 ). Replicated blank samples, containi ng 0.5 g of biosolids and 2.5 mL of 0.01 M T CS free CaCl 2, were also included. Samples were equilibrated on an end over end shaker for 24 h (based on a preliminary kinetic study), and subsequently centrifuged (8000 g ) for 10 min at constant temperature (24 2C). Supernatant a liquots (1mL) were add ed to 10 mL of Ecoscint A scintillation cocktail (National Diagnostics, Georgia ) in 20 mL glass scintillation vials. Radioactivity was determined via liquid scintillation counting with
46 background correction against blanks ( 14 C free 0.01 M CaCl 2 ). Controls (no biosolids) were utilized for determining the initial concentration ( C 0 ). The amount of TCS partitioning onto the biosolids solid phase was calculated as a percentage of 14 C TCS initial activity added and the change in activity in the solution phase ove r time. Partition coefficients ( K d ) were estimated for TCS adsorption on seventeen biosolids as: K d = Activity of TCS adsorbed/kg biosolids Activity of TCS in solution/ L solution S d and K oc values. The concept of inherent K d and K oc has been previously utilized for determining the coefficients in pharmaceuticals and musk fragrances (Carballa et al., 2008). Biosolids g 1 Chapter 2, Table 2 1) were selected to improve the probability of detecting TCS in the aqueous phase by LC/MS (limited due to LOQ of 1 n g m L 1 ). The whole biosolids (cake) was extracted with an organic extractant (methanol, acetone mixture) to determi ne the total TCS concentration (sorbed + aqueous) as d escribed in Chapter 2 (Table 2 1). A separate sample of biosolids (cake) was then centrifuged to obtain a sample of the presumed equilibrium solution and analyzed for TCS. The concentration of TCS sorbe d was obtained by subtracting the mass of TCS in supernatant from the total TCS. E stimation of K d d the supernatant TCS. The mean K d and K oc values were calculated based on actual measurem ent of five biosolids supernatant TCS concentrations and on two biosolids concentrations estimated from the LOQ of the instrument.
47 Results and Discussion Measured TCS water solubilities were 9 mg L 1 (pH = 6.14), 27 mg L 1 (pH = pK a = 8.14) and 800 mg L 1 (pH = 10.14) (Table 3 2). The overall 90 fold increase in the solubility resulted from dissoci ation of the neutral acid to anionic species. Increased solubility at higher pH portends greater concentration in the aqueous phase during sewage treatment proces ses utilizing lime stabilizatio n and in high pH soils. Extremely high pH levels are not common in most sewage treatment systems or in most soils, except some sodic soils. The s olubility of TCS measured at pH 6.14 (i.e., 9 mg L 1 ) appears reasonable for pre dicting the fate and transport of TCS in many soils. Thus, typical biosolids applied to typical, near neutral soils would be expected to demonstrate a TCS water solubility of 9 mg L 1 which is similar to reported measured values of 10 and 17 mg L 1 (Ciba Specialty Chemicals, 2001; MITI, 1992), but greater than calculated values of 1.97 to 4.6 mg L 1 obtained using Quantity Structure Activity Relationship (QSAR) and Estimation Programs Interface (EPI Suite v3.10) (Halden and Paull, 2005, Ying et al., 2007). The models underestimated the solubility of TCS and thus, the log K ow values were expected to be overestimated by the model. However, the measured and model estimated log K ow values (4.8 and 4.76) were relatively consistent The unexpected relationship between TCS solubility and log K ow values highlights the importance of measured data collected by standardized methodology. Mean log K d values standard error (S.E ) were 3.76 0.04 and log K oc values were 4.30 0.03 in TCS spiked biosolids (Table 3 3). The log K oc values were similar to estimated values in soil of 4.26 (Ying et al., 2007), but slightly lower than measured value of 4.6 reported by Heim (1997). The log K d and K o c values suggest relatively strong sorption of TCS to the biosolids and low mobility in acid and circum neutral
48 (pH=6.5 7.5) soils. Also, the same biosolids were utilized to measure K d and K oc directly d and K oc e mean (n=7) inherent log K d was 4.15 0.03, and log K oc was 4.68 0.07. Based on the limited number of biosolids analyzed, the differences were not statistically significant, but the inherent log K d and K oc values tend to be greater than the spiked log K d and K oc values (Table 3 3). The log K d values determined for biosolids ( 4.15 0.03 ) was greater than those determined for soils (2.25 0.26) a nd biosolids amended soils (2.31 0.19) (Agyin Birikorang et al., 2010). The difference in the coefficients was attributed to the difference in organic carbon among the various matrices. Following normalization to organic carbon, the coefficients (K oc ) determined in the soils, biosolids and biosolids amended soils were not significantly different and averaged 4. 26 0.31 (Agyin Birikorang et al., 2010). Thus, a specific or narrow range of TCS partitioning coefficient (K oc ) can serve as a first approximation to describe the behavior of TCS in soils or other matrices. The K oc values measured in our study and report ed in other studies suggest significant adsorption of TCS to the soils and or biosolids. Thus, the mobility of TCS in biosolids amended soils is expected to be restricted and the extent of retardation to be highly dependent on the organic carbon content of the amended soils and pH in case of calcareous soils
49 Table 3 1 Physico chemical Properties of TCS reported in the literature. Property Value References Comments Molecular Weight (g mol 1 ) 289.55 Formula ( C 12 H 9 Cl 3 O 2 ) CAS Registry no (3380 34 5) Solubility (mg L 1 ) 1.97 4.6 10 17 (Halden and Paull,2005) (Ciba Speciality Chemicals, 2001a) (MITI, 2002) Estimated Measured log K ow 4.8 4.70 4.76 4.76 (Halden and Paull, 2005) (Ying et al., 2007) (Lei and Snyder, 2007) (Wezel and Jager, 2002) Estimated Kow WIN Estimated PBT Profiler Measured Measured log K oc 4.26 4.6 (Ying et al., 2007) (Heim, 1997) Soils (1.3 % OC) PBT profiler, Measured pK a 8.14 (Jakel,1990) Vapor Pressure (mm Hg at 20 0 C) 4 10 6 (Ciba Special ty Chemicals, 2001b) Non Volatile m 3 mole 1 at 25 0 C) 4.9910 9 (Meylan and Howard, 1991) Estimated Half life (d) 12 0(Soils) 540(Sediment ) 60 (Soils) 240 (Sediment ) 18 58 persistent 540 15 35 12 15 107 56 107 (Ying et al., 2007) (Ying et al., 2007), (Wu et al., 2009) (Miller et al., 2008) (Ciba Speciality Chemicals, 2001a) (Xu et al., 2009) (Lozano et al., 2010) (Kwon et al, 2010) Estimated PBT Profiler, USEPA Level III fugacity model Measured Aerobic, anaerobic soils Sediments Soils Soils Biosolids amended soils Biosolids amended soils
50 Table 3 2 Water solubility (mg L 1 ) of TCS at various pH values pH Time (h) Average 24 48 72 96 mg L 1 6.14 8.5 0.20 8.9 0.47 9.7 0.67 8.4 0.30 8.9 0.26 8.14 14 0.71 30 1.0 26 1.8 26 1.6 26 0.00 10.14 n/d n/d n/d n/d 791
51 Table 3 3 Mean log partition coefficients (K d and K oc ) ( n=3) standard error (S.E) for TCS on spiked ( Agyin Birikorang et al., 2010) and u nspiked (inherent) biosolids ( our study) Spiked K d K oc on biosolids Inherent K d, K oc (Our study) Biosolids Treatment process log K d Fraction Organic carbon log K oc log K d log K oc UNKB Anaerobic digestion 3.68 0.02 0.38 4.19 0.02 3.99 0.01 4.510.00 UNKC Anaerobic digestion 3.79 0.02 0.37 4.28 0.02 3.88 0.04 4.300.14 UNKD Anaerobic digestion 3.74 0.01 0.31 4.35 0.01 UNKE Anaerobic digestion 3.61 0.03 0.32 4.09 0.07 UNKF Anaerobic digestion 3.62 0.04 0.21 4.35 0.04 3.850.03 4.600.03 UNKG Anaerobic digestion 3.76 0.02 0.42 4.21 0.04 4.170.08 4.550.18 UNKH Anaerobic digestion 3.77 0.01 0.30 4.35 0.01 3.880.06 4.440.06 UNKI Anaerobic digestion 3.63 0.04 0.24 4.30 0.04 UNKK Unknown 3.75 0.01 0.28 4.35 0.01 GRU 3.90 0.02 0.38 4.31 0.02 OSBC Anaerobic digestion 3.93 0.01 0.42 4.37 0.01 ORBC BL Untreated (before lime stabilization) 3.85 0.03 0.41 4.26 0.05 ORBC AL Lime stabilization (following lime addition) 3.68 0.03 0.34 4.22 0.07 RCKF Anaerobic digestion 3.85 0.01 0.39 4.31 0.01 CFBC Anaerobic digestion 3.95 0.01 0.41 4.38 0.03 CALC Anaerobic digestion 3.70 0.03 0.28 4.36 0.03 CHCC Anaerobic digestion 3.67 0.02 0.28 4.35 0.02 Average 3.76 0.04 4.30 0.03 4.15 0.03 4.68 0.07
52 Figure 3 1. Calculated octanol water partitioning coefficient (log K ow ) curve and dissociation diagram for TCS. 1.9 2.4 2.9 3.4 3.9 4.4 4.9 0 20 40 60 80 100 0 2 4 6 8 10 12 14 log K ow Percent dissociation pH Percent dissociation log Kow
53 CHAPTER 4 BIODEGRADATION OF BIOSOLIDS BORNE TCS Background The antimicrobial t riclosan (TCS) is a common constituent of domestic waste water and > 50% of TCS mass entering the WWTPs partitions into biosolids (Langdon et al., 2011; McClellan and Halden, 2010; Cha and Cupples, 2009; Xia et al., 2009; Heilder and Halden, 2009; USEPA, 2009a). Land application of biosolids transfer s TCS to soil s where its f ate and transport depend s on TCS persistence in the terrestrial environment Biodegradation half lives for TCS in aerobic soils have been estimated using fugacity based models li ke Persistence, Bioaccumulation, and Toxicity (PBT) Profiler (USEPA) ( t 1/2 = 60 120 d) and measured [t 1/2 = 18 d (Ying et al., 2007) and 2 to 13 d (Kwon et al., 2010)] (Table 3 1). Miller et al. ( 2008 ) suggested long term persistence of TCS in anaerobic sediments (40 years) compared to the estimated (PBT ) half life of 540 d under anaerobic soil conditions. Persistence of TCS in aerobic soils and sediments is reasonably well studied, but TCS degradation in biosolids amended soils is less studied. Degradation studies in biosolids amended soils are important, as TCS pr eferentially partitions into biosolids potentially reducing TCS bioavai l ability to degrading organisms. Lozano et al. (2010) estimated a TCS half life of 107 d in a biosolids amended soil (unknown soil texture) while Kwon et al. (2010) reported mea sured h alf lives ranging f rom 56 to 107 d in two biosolids amended loam soils [fine loam, pH:7.8, coarse loam, pH:4.7] Recently, a mesocosm study conducted in Maryland resulted in an estimated first order half life of 182 193 d (Walters et al., 2010). Higgins et al. (2011) reported a smaller half life (42 d) in a biosolids amended field soil (silty clay loam).
54 However, Higgins et al. (2011) half life estimate was approximate, as the authors acknowledged that the study did not intend to calculate degradation Kwon et al. (2010), Lozano et al. ( 2010), Walters et al. (2010), and Higgins et al. (2011) reported loss of extractable TCS with time, but did not distinguish between compound loss due to TCS mineralization [conversion to carbon dioxide (CO 2 )] or primary degra da tion (formation of metabolite from the parent compound ). Loss of extractable TCS may represent the conversion of initially extractable TCS to non extractable ( bound ) residues with time Snyder et al (2010 ) reported the formation of triclocarban ( TCC ) b o und (non extractable) residues in biosolids amended soils We speculated that because TCS has a reported low water solubility value (1.9 9 mg L 1 ) and because the partitioning coefficient of TCS (log K oc = 4.3 ) is even greater than TCC (log K oc = 3.88 ) (Ag yin Birikorang et al., 2010) TCS should tend to partition into organic C in soils and sediments and form bound residues as well. An accurate measurement of TCS half life should include the dete rmination of TCS mineralization, or primary degradation identification of metabolites, and demonstration of mass balance. Transformation by biological methylation forms methyl ether derivatives that are usually more lipophilic than the parent compounds (Valo and Salkinoja Salonen, 1986; Neilson et al., 1983) Tulp et al. (1979) re ported that TCS degrades by hydroxylation of benzene ring and cleavage of the ether bond forming dichlorophenols in urine samples. Hundt et al. (2000) showed that TCS microbial degradation can form Methyl TCS (Me TCS), dichlorophenol s, and conjugated metabolites where carbohydrates are connecte d to the hydroxyl group of TCS. Miyazaki et al. (1984) reported Me TCS in fish but it was unclear whether the methylation occurred in the su rface water or in the fish body.
55 However, a later stu dy (Balmer et al., 2004) confirmed the presence of Me TCS in surface water as well as fish bodies. Lindstrom et al. (2002) reported that M e TCS formed in wastewater effluents and surface waters through biological methylation. Evidence of TCS methylation wa s also obtained from the presence of the anisole in semi permeable membrane devices (SPMDs). Lindstrom et al. (2002) opined that the presence of only Me TCS (no TCS) in SPMDs might represent significant bioaccumulation potential of Me TCS. Further, McAvoy et al. (2002) detected Me TCS (0.13 0.45 g g 1 ) in sludge and aerobically digested biosolids obtained from two WWTPs in Ohio. Xia (2010) reported that TCS biomethylated to a small quantity (~9%) of Me TCS in a biosolids amended soil (sandy loam) utilized for a TCS leaching study. Thus, we might expect formation of Me TCS under aerobic conditions when biosolids borne TCS is added to the soils. However, it is uncertain if the formed Me TCS persists or further mineralizes to CO 2 Limited data for Me TCS parti tioning suggest that Me TCS is more hydrophobic ( log K ow = 5.2) th an TCS ( log K ow = 4.8 ) (Boehmer et al., 2004), and therefore, Me TCS is likely to be less bioavailable than TCS and more likely to form bound residues. We hypothesize that biosol id s borne TCS and its metabolites are persistent in the environment. The objective of our study was to determine the degradation (mineralization/primary degradation) potential of TCS in two biosolids amended soils. Specifically, we sought to (i) i dentify the TCS metabolites (ii) determine the mineralization and/or primary degradation ha l f lives, and (iii) assess the extent of bound residue formation We conducted a biodegradation study in soils using 14 C TCS spiked biosolids under biotic and inhibited biotic aerobic conditions to determine the
56 degradation rates of TCS in biosolids amended soils. Aerobic conditions were chosen as biosolids borne TCS is expected to undergo primarily aerobic degradation under the typical conditions of surface applied and incorpor ated biosolids. The biodegradation study was conducted according to the United States Environmental Protection Agency (USEPA ) Office o f Prevention Pesticides and Toxic Substances (OPPTS ) Harmonized Fate, Transport and Transformation Test Guidelines; Soil Biodegradation (835.3300 ) (USEPA, 1998). Material and Methods Chemicals, Biosolids and Soils Radiolabeled 14 C TCS uniformly on the chlorophenol ring ( specific activity 48 mCi mmol 1 and 99% purity) was custom synthesized by Tjaden Biosciences ( Burlington, IA ). Ecoscint A liquid scintillation cocktail was purchased from National Diagnostics (Atlanta, GA) and the Me TCS standard from Wellington laboratories ( Shawnee Mission, KS). All other chemicals and solvents were purchased from Fisher Scientif ic (Atlanta, GA). An anaerobically digested b iosolids ( CHCC, s olids: 320 g kg 1 ) was collected from a domestic WWTP in Illinois The biosolids contained 5 mg kg 1 of TCS and non detectable (<0.7 mg kg 1 ) Me TCS (Chapter 2). Two soils, the Immokalee fine sa nd (IFS) ( sandy, siliceous, hyperthermic Arenic Alaquods ) and the Ashkum silty clay loam (ASL) ( f ine, mixed, superactive, mesic Typic Endoaquoll s ) were collected from sites with no known history of receiving land applied biosolids or sludge Select physico chemical properties of soils and the biosolids are presented in T able 4 1. Biodegradation Study Design The biodegradation study consisted of 146 glass, round bottom, 3 0 mL glass centrifuge tubes (2 soils 1 biosolids 1 rate 4 replicates 8 sa mpling period s 2
57 treatments (biotic/inhibited) + 18 controls). The first set of controls consisted of 1 biosolids no spike of 14 C TCS 2 soil s 2 treatments (biotic/inhibited ) 4 replicates (Table 4 2). The absence of evolved 14 CO 2 in controls would confirm no cross contamination in the system. The second set of controls consisted of 14 C TCS spiked into autoclaved water. Evolved 14 CO 2 will indic ate soil independent abiotic TCS d egradation (e.g., photolysis). Treatments were prepared by first weighing CHCC biosolids (0.10 g dry wt. equivalent) in centrifuge tubes. A sub stock of 14 C (1.3 10 6 dpm 14 C TCS g 1 ) was prepared in 1mL methanol (MeOH) and dissolved in 200 mL of deionized distilled ( DDI) water. One mL of sub stock TCS was spiked onto the CHCC biosolids sample and allowed to equilibrate for 24 h at room temperature. Equilibrium between spiked TCS and the biosolids was assumed to ha ve occurred in 24 h as suggested by Agyin B irikorang et al. (2010) who reported TCS sorption equilibrium within 24 h of incubation. The spiked biosolids was mixed with 10 g (dry wt.) of two soils to simulate a realistic field application rate (22 Mg ha 1 or ~10 tons acre 1 equivalent ). The biosolid s amended soils were vortexed for 30 s to promote uniform mixing of the spiked biosolids and the soil sample. T he radioisotope spiking increased the effective biosolids borne TCS concentration to 40 mg TCS kg 1 and the final nominal TCS concentration in the amended soil was 0.4 mg TCS kg 1 dry weight. Treated samples were aerobically incubated under biotic or inhibited conditions, and soil water contents were maintained at field capacity (i.e. 100 g kg 1 for IFS and 300 g kg 1 for the ASL soils) throughou t the experiment. Inhibited treatments quantified the effects of non microbial reactions on TCS degradation and were facilitated by adding 1200 mg kg 1 of 0.1 % sodiu m azide
58 (Fischer et al., 2005). Biosolids amended soil samples were weighed at each of the 8 sampling times, and DDI was added to the samples when the differenc e in weights between initial (week 0) and subsequent samplings was >5%. The b iodegradation study setup was adapted from Snyder (2009) and is shown in F igure 4 1. Briefly, e ach centrifuge t ube was connected to a series of three glass trapping vials (base traps) containing 5 mL of 0.2 M potassium hydroxide (KOH). The base traps collect ed 14 CO 2 (representing 14 C TCS mineraliz ation) and CO 2 (representing microbial respiration). A pump provided samples with humidified air stripped of CO 2 through a series of contain ers containing, sequentially, 1 M KOH, CO 2 free DDI water (prepared by boiling water for extended time) s oda lime [ Calcium hydroxide ( Ca(OH) 2 ) > 80%, KOH < 3% sodium hydroxide (NaOH)<2%, Ethyl violet <1%], 1 M KOH, and CO 2 free DDI water. The first glass vial remained empty to prevent ba ckflow into the centrifuge tube if the pump failed. The samples were incubated at ~23 C in the hood to avoid possible fugitive radiolabel vapor escape into the room. Base Trap Analysis and Soil Sample E xtraction Once a week, base traps at position 1 were removed, the remaining traps were moved forward, and a fresh trap added to the newly op en position 3. The sampling schedule included soil sample removal at 0, 2, 4, 6, 8, 10, 12, 14 and 18 weeks Four samples of each amended soil treatment were periodically removed (sacrificial sampling) and sequentially extracted by sonicating (Branson 2210 Danbury, CT ; temp. 40 C, 60 Hz min 1 ) for 1h with 20 mL each of DDI water (twice), MeOH (twice), and 1 M NaOH (once) and finally centrifuged (~800g for 60 min) to isolate residual soil A subsample (1g dry wt.) of the residual soil was combusted at 900 C in a Harvey Oxidizer, Model OX 500 (Tappan, NY). Ano ther subsample of residual soil was
59 extracted with a mixture of MeOH+acetone (twice) (50:50 v/v). In addition, a ll the base traps attached to the samples (three for each sample) were removed at each fi nal sampling time. Aliquot s (~500 L) of the supernatants removed from various extracts and base traps were mixed with 15 mL Ecoscint A and stored for 24 h. A liquid scintillation counter (LSC) Model LS 6500 (Be ckman Coulter, Irvine, CA, USA) was utilized to quantify 14 C (5 min counts). E volved 14 CO 2 from the oxidizer was also trapped in 15 mL cocktail ( Harvey, Tappan, NY) and the radioactiv ity quantified by LSC. Removed base traps were also analyzed for total CO 2 ( Anderson, 1982 ) to assess possible effects of added 14 C TCS on overall microbial respiration in biotic treatments. Adverse effects were considered possible because of the relatively high total TCS concentration (40 mg kg 1 ) in biosolids spiked with 14 C TCS Sequential Extrac tion S cheme The sequential extraction scheme was modified from a previous study on triclocarban (TCC) biodegradation (Snyder et al., 2010) and represents the extraction of fractions of TCS of assumed varying availabilities (lability) in the soil. The expla nation of the sequential extrac tion scheme is provided in A ppendix A We termed 14 C in the water and MeOH extracts as labile and 14 C in NaOH MeOH+ acetone extracts and the combustible fraction s as non labile (bound) The non labile fractions include d: humic associated (NaOH), loosely sorbed (MeOH+acetone), and bound (combustible) fractions of TCS. The lability assignments are operationally defined (and somewhat arbitrary), but consistent with extraction schemes used b y others ( Semple et al., 2003; Hi ckman and Reid, 2005; Heidler et al., 2006; Snyder et al., 2010 ).
60 Radiological Thin Layer Chromatography (RAD TLC) for Extract S peciation At each sampling time, aliquots of extracts from the first MeOH and the first MeOH+acetone extraction s were separately air dried using an aquarium pump and reconstituted in ~30 L MeOH. Thin layer chromatography (TLC) plates were used to identify 14 C labeled moieties (parent and metabolites) from the various extracts. The plates were spotted with reconstituted MeOH extrac t and developed in a chamber saturated with solvent vapors of mobile phase. The mobile phase consisted of Chloroform+Formic acid (99:1, v/v). The TLC plates were either glass backed Partisil LK5D, silica gel 150 A, 2020 cm (Piscataway, NJ) for performing RAD TLC, or glass backed 2010 cm silica gel 60F 254 plates (E. Merck, Darmstadt, Germany) for fluorescent TLC (used for metabolite identification) Fluorescent TLC plates were analyzed by hand held U V light (Model UVG 54; San Gabriel, CA), and RAD TLC pl ates were analyzed by RAD TLC instant imager ( PerkinElmer Life and Analytical Sciences Waltham, MA ). We initially intended to perform RAD TLC analysis on water and NaOH extracts as well, but the analyses were not possible due to insufficient amounts of radioactivity in the two fractions for peak quantification. The limit of quantitation (LOQ) of all spec ies (parent and metabolites) using RAD TLC imager was ~1000 dpm, which represented ~8% of the total 14 C added in the system. The relative amounts of the TCS and the metabolite were measured at each sampling time in the two extracts. The relative proportio n of 14 C TCS and metabolite in the water fraction was assumed to be same as in the MeOH fraction and the proportion in NaOH and bound fractions were assumed to be same as in MeOH+acetone fraction. The assumptions were made because the LOQ of the RAD TLC im ager was greater than the quantity of 14 C in the water and NaOH extracts. Based on the aforementioned
61 assumption s concentrations of TCS and metabolite were estimated for all the extracts and the data were utilized to estimate the primary degradation half life of TCS to the metabolite. Statistical Analysis Carbon dioxide data were analyzed using the general linear model (PROC GLM) of the SAS software, version 9.1 (SAS institute, 2002). Values of CO 2 evolved over time were compared among the various treatmen ts using the multiple slope comparison option of the regression procedure (SAS institute, 2002). Results and Discussion Mass Balance and Mineralization of 14 C TCS Percent total recoveries of 14 C in the IFS soil treatments varied from 86 to 103% with an average of 92.8% for biotic (Table 4 3), and 93.4% for inhibited treatments (T able 4 4 ) Total recoveries in the ASL soil treatments were 100 to 109% with an average of 104% for biotic (Table 4 5), and 105% for the inhibited treatments ( Table 4 6 ) As t he average recoveries in our system were >90%, we did not expect that TCS volatilized as a gas and escaped the system at any time during the experiment. The p ercent total recoveries obtained in the biodegradation study were deemed acceptable, but to simpli fy the discussion of 14 C TCS partitioning into various fractions the data we re normalized (Figure s 4 2, 4 3 ) by dividing 14 C in each fraction by the total 14 C recovered from each soil at each sampling time In addition, the F igures 4 2 and 4 3 include the 14 C associated with the MeOH+acetone extraction of the combustible ( bound ) fraction. Mineralization of 14 C TCS to 14 CO 2 in the two soils (IFS and ASL) was minimal (<0.5%) through the study period (18 week ) (Figure s 4 2, 4 3). The different percentage
62 of TCS mineralization in the either soil is within the variable total recoveries (range: 86 103%) in the two soil treatments, and is not considered significant. Al Rajab et al. (2009) also re ported minimal (<1 % ) TCS mineralization in a 42 d degradation study of TCS spiked to dewatered biosolids that was then mixed with soil [ sandy loam; Organic matter ( OM ) : 37 g kg 1 ] Further, the two controls did not detect any 14 CO 2 suggesting no cross contamination or abiotic TCS degradation in the system. In the IFS soil biotic treatment (Figure 4 2a ), >90% of spiked 14 C TCS was water +MeOH extractable (labile) at week 0, but extrac tability decreased to ~35% at 18 week The large recove ry in the labile fraction at week 0, suggests initial reversible sorption of 14 C TCS to the solid phase. The percentages of radioactivity in the combustible (bound) and NaOH extractable fractions at week 0 w ere 2% and 0.1%, respectively, but increased to nearly 40% in the bound and 20% in NaOH extractable fractions ( w eek 18). Radioa ctivity in the water fraction remained fairly constant at 2 4 %, much lower than the radioactivity in other fractions. The MeOH+acetone extractable 14 C increased from 1% at week 0 to nearly 10% by week 18. Thus, over the 18 week study period the IFS bioti c treatment data revealed a decrease in water+MeOH extractability ( TCS lability) with time, and concomitant i ncreases in the non labile (NaOH, MeOH+acetone extractable and bound ) fractions. Similar trend s of greater percent extractability in the labile fra ction and low percentages in non labile fractions occurred in the inhibited treatment of IFS soil (Fig ure 4 2b ) but the radioactivity in the bound fraction did not increase over time as in biotic treatment s. The lack of increasing trend over time in the inhibited treatment suggests microbially mediated TCS partitioning into the non labile fraction.
63 Similarly, the data for the ASL soil biotic treatment revealed a trend of reduced extractability of labile 14 C TCS fraction from week 0 ( 40%) through w eek 18 (7%) (Figure 4 3a). About 60 % of the added 14 C in biotic and inhibited treatment s was immediately associated with the non labile (NaOH, MeOH+acetone, bound ) fraction s, suggesting limited TCS lability. The bound 14 C in the ASL biotic tr eatment increa sed from week 0 week 4 (Figure 4 3a) whereas the conversion to the bound fraction was slower in the inhibited treatment, reaching >40% at week 18 (Figure 4 3b ). Faster partitioning of 14 C TCS into the non labile fraction in the biotic treatment again suggests microbially mediated alteration of TCS that hastened the conversion of labile fraction to the non labile bound fraction. Recoveries of 14 C TCS in the labile fraction were consistentl y greater in the IFS (Figure s 4 2a and b) t han in the ASL soil treatments (Figure 4 3a and b ). T he greater clay and organic carbon ( OC ) content s in the ASL soil (340 g kg 1 clay, 34 g kg 1 OC) than in the IFS soil (<10 g kg 1 clay, 11 g kg 1 OC) likely pr omoted stronger interaction of TCS with the ASL soil and reduced TCS lability. At week 0, 3 0% of the 14 C TCS partitioned into bound fraction (non labile) in the ASL soil, whereas the bound fraction was 2 3% in IFS soil suggesting an interactive effect of soil texture and OC content on the partition ing of 14 C TCS into labile and non labile fractions. Agyin Birikorang et al. ( 2010 ) conducted a sorption/desorption study on the same soils used herein and reported that log K d in the IFS soil (1.87 0.21 ) was significantly smaller than in the ASL soil (2.64 0.19 ), suggesting t exture and OC content effects on TCS partitioning into sorbed and aqueous phases
64 Total Carbon Dioxide (CO 2 ) Analyses The TCS spiked ASL soil biotic treatment had the greatest CO 2 evolution rate (Fig ure 4 4b ), and the CO 2 evolution rate of the unspiked ASL soil biotic treatment was similar to the spiked treatment until week 13. The similarity was not surprising as the soil microbes were l ikely utilizing more abundant C sources (bio solids or soil OC ) rather than the small amount of added TCS. In the IFS soil biotic treatments exhibited the greatest CO 2 evolution rates followed by IFS unspiked biotic, IFS inhibited biotic, IFS unspiked inhibited biotic and th e spiked water control (F igure 4 4 a). The trend is expected, as the micro organisms will be actively growing in biotic, but less so in the inhibited biotic treatments. The smaller rates of CO 2 evolution from the unspiked IFS soil biotic treatment, as compared to the unspiked ASL s oil biotic treatment, may represent differences in microbial community size, or the relative abundance of available C sources in the IFS (11 g kg 1 OC) and ASL soils (34 g kg 1 OC ) (Table 4 1). Despite the addition of sodium azide in the inhibited biotic t reatments at week 0, microbial activity continued, but at a reduced rate compared to the biotic treatments. Cumulative CO 2 evolved from spiked water only control (no soil) was the least, as compared to all other treatments, which confirmed the effectiveness of CO 2 scrubbing system. The IFS soil data revealed significan t differences (p<0.05) among all treatments, except between the spiked inhibited biotic and the unspiked inhibited biotic treatmen t, and between the unspiked inhibited biotic and t h e spiked water control (Figure 4 4 a). The ASL soil data revealed no significant differences betw een the biotic and unspiked biotic treatments, nor between the inhibited biotic and unspiked inhibi ted biotic treatments (Figure 4 4b ). However, there were sig nificant differences among the biotic, inhibited biotic and the spiked water control (soil less) in the ASL soil.
65 Cumulative CO 2 increased through week 18 in both biotic soil treatments, suggesting normal microbial activity throughout the bio degradation ex periment (Figure 4 4 ). Thus, the biosolids amended soil TCS concentration (0.40 mg kg 1 ) used in our experiment did not significantly affect CO 2 evolution at any time. The results are in accordance with a study on a similar antimicrobial (TCC) where the CO 2 evolution rates were unaffected by the addition of TCC at concentration range of 0.24 7 mg kg 1 in biosolids amended Florida Immokalee sand (Snyder et al., 2011 ). Waller and Kookana (2009) spiked a sandy soil (clay: 100 g kg 1 OC: 8.5 g kg 1 ) and a clay ey soil (clay: 480 g kg 1 OC: 18.5 g kg 1 ) with 0, 1, 5, 10, 50, and 100 mg TCS kg 1 soil, and monitored changes in substrate (glucose) induced respiration in soils (no biosolids). There were no TCS concentration effect s on microbial respiration up to 100 mg TCS kg 1 soil. Butler et al. (2011) examined basal and substrate induced respiration in t hree soils (no biosolids) sandy loam (OC 17 g kg 1 ), clay (OC 27 g kg 1 ) and loamy sand (OC 23 g kg 1 ) spiked with a range of TCS concentrations (0 1000 mg kg 1 ). Results suggested some inhibition of respiratio n 2 4 days after spiking at TCS concentration s >10 mg kg 1 but respiration recovered to the control level by day 6 and overall, there was no effect of TCS spiking. Metabolite I dentification Each RAD TLC an alysis (up to week 2) of MeOH and MeOH+acetone extracts of corresponding to a 14 C TCS standard (Figure 4 5a). There were no peaks suggesting TCS metabolites up to week 2 A t week 3, an additional peak (peak 3) appeared in the chromatogram of biotic treatments (Figure 4 5b). The additional peak appeared farther from TCS on the TLC plate and was demonstrated to be Me TCS. Methyl TCS is a
66 reported metabolite of TCS (Figure 4 6) with a log K ow of 5.2 (vs log K ow = 4.8 for TCS) (Boehmer et al., 2004) and is formed by the methylation of hydroxyl group of TCS. The p eak could not be conclusively identified with RAD TLC due of unavailability of a 14 C Me TCS standard. Rather, the identification was performed using fluorescent chromatography and a cold Me TCS standard. Chromatography resulted in distinct bands under UV light of Me TCS that aligned with the cold Me TCS standard. Further confirmation was performed by comparing retenti on factors (R f ) The R f values measured in samples taken at various sampling times were 0.6 to 0.65 for TCS and from 0.72 to 0.75 for Me TCS. The relative R f ( R f for TCS/ R f for Me TCS) were 0.83 to 0.86. Based on co migration with an authentic standard, t he metabolite was identified as Me TCS. Analyse s of extracts through week 18 revealed continued presence of Me TCS in biotic treatments for both soils, but no metabolites in inhibited treatments. The formation of Me TCS in the present study corresponds we ll with the previous studies suggesting Me TCS formation in a variety of matrices ( Coogan et al., 2007, 2008 ; Miyazaki et al., 1984; Lindstrom et al., 2002; McAvoy et al., 2002; Bester, 2005). Methyl TCS bioaccumulate d in algae and snails when exposed to e ffluent containing Me TCS in the co ncentration range of 50 to 400 ng L 1 (Coogan et al., 2007, 2008). Hundt et al. (2000) found Me TCS as one of the microbial degradation metabolites in fungus ( T.Versicolor ) cell cultures incubated with TCS. Poulsen and Bester (2010) detected Me TCS (up to 70 ng g 1 ) in sewage sludge composted under thermophilic conditions. The bacterial species Rhodococcus (strain CG 1, CP 2), Mycobacterium (CG 2) and Actinomycetes methylated several chlorophenols (Haggblom et al., 1988; Neilson et al., 1983). Triclosan belongs to the class of chlorophenols and may be
67 methylate d by a similar mechanism. An unpublished report (Christensen, 1 994) estimated that up to 70 80 % of TCS degraded to Me TCS in a 64 d study conducted on three sludge amended soils (Arkansas silt loam, Kansas loam and Wisconsin sandy loam). Unfortunately, details of the experimental design of this unpublished study were not available. In contrast to the above studies Xia (2010) reported that TCS biomethylated to only a small quantity (~9%) of Me TCS after 101 d of bi osolids amended soil ( pH = 5.7, sandy loam ) incubation in a TCS leaching study. The present study is apparently the first published study that suggests Me TCS as a major biodegradation metabolite of TCS in b iosolids amended soils. The mass balance for 14 C TCS for both soils indicated that the total radioactivity decreased in MeOH extracts and increased in the combustible (bound residue) fraction with time. The increase of 14 C in the bound fraction did not affect the relative peak heights of TCS and Me TCS in MeOH extracts in the RAD TLC analysis. The peak for Me TCS and percent radioactivity associated with the peak was always smaller than the corresponding values for TCS. Ho wever, bound residue characterization (MeOH: acetone extract) suggested an increase in peak height of Me TCS and no change in TCS peak height with time, indicating Me TCS formation and partitioning of both TCS and Me TCS to the bound fraction. The i ncrease in peak height of Me TCS with time suggests preferential partitioning of Me TCS over TCS in the bound fraction. Half life (Persistence) D etermination The inhibited biotic treatments produced no detectable metabolites, and there was minimal TCS min eraliza tion (<1%) during the 18 week incubation study. H alf life determination was not possible, and TCS would be regarded as persistent (half life >>120 d) in inhibited biotic s oils
68 Biotic treatment s in both soils also had minimal mineralization (<0.5 %) of adde d TCS. Our study suggests the appearance of a metabolite in the biotic treatment in both soils. T he relative proportions of TCS and Me TCS (of the total 14 C measured) in biotic treatments were utilized to estimate the primary degradation half life of TCS ( Figure. 4 7 a, b). The ratios of the two compounds were different in the two soils. In the IFS extracts, ~2 3 % of Me TCS appeared at week 3, and increased to 10% at week 6, and stayed at ~10% for the rest of the study (Figure. 4 7 a). In the ASL soil, Me TC S also appeared at week 3 (~15%), and increased to 60 % at week 12. After 18 week 80% of 14 C was detected as Me TCS and 20 % as TCS (Figure. 4 7 b). The p roportion s of TCS and Me TCS and the variations with time were fitted us ing zero and first order models ( Table 4 7 ) The zero order model fitted better (R 2 = 0.95 0.97) to the ASL soil data than the first order model (0.76 0.91), and thus, the zero order model was utilized to estimate the time taken for 50% of TCS to disappear and 50% o f Me TCS to appear in the soils. The estimated half lives were 11 week (77 d) for ASL soil (Figure 4 7 b), and >18 week (>126 d) for the IFS soil (F igure. 4 7 a). Our half life estimations are in contrast to expectation based on the lability estimations, as the greater labile or bioavailab le fraction was detected in the IFS soil. Thus, i t appears that the operationally defined extr a c tion scheme described herein may not accurately predict the degradation potential of TCS. However, o ur results were consistent with another study (Kwon et al., 2010) that reported faster TCS transformation in a fine loam soil than a coarse loam soil Kwon et al. (2010) attributed the difference to the greater microbial population in the fine textured soi l (910 6 CFU g 1 dry soil) than in the to the coarse textured soil
69 (510 6 CFU g 1 dry soil). Half lives determined herein are similar to the TCS disappearance (loss of extractable TCS) half life of 107 d (Lozano et al., 2010) determined for a field soil (u nknown soil texture) amended with a range of biosolids application rates (9 25 Mg ha 1 ) (inherent TCS 15.8 mg kg 1 biosolids). Our e stimates are also consistent with a TCS dis appearance (loss of extractable TCS) half life of 50 to 106 d measured in a labor atory incubation of biosolids amended fine loam and coarse loa m soil (Kwon et al., 2010). In contrast, Higgins et al. (2011) reported a smaller half life (42 d) in a biosolids amended field soil (silty clay loam). However, the half life was approximate, as the study was not adequately designed to quantitatively assess degradation. T he loss of extractable TCS as the criterion for TCS degradation used in the above studies is apparently useful for approximating the half life of TCS, but provides no informatio n on the degradation metabolites. A half life range of ~77 to >126 d, average ~100 d would be a reasonable first approximation of TCS persistence in typical biosolids amended soil s We partially accept our hypothesis that TCS is persistent in biosolids amended soil, as a chemical is termed persistent if the half life is >60 d (USEPA, 1999). The persistence of metabolite (Me TCS) was not quantified in the present study and should be a topic of future research. Our h alf life estimation included assumptions that the proportions of TCS and Me TCS in the water are same as in the MeOH fraction and the proportions in MeOH+acetone are same as in the other n on labile fractions. But, the relative proportions of TCS and Me TCS could vary among the variou s fractions and thus, our estimate of half lives could be in error. However, the consistency of half life values
70 determined in the present study and other studies that used loss of extractable TCS as the criteria of persistence leads us to believe that our assumptions were justifiable. Comparison of TCS Per sistence in Amended, Un A mended, and Field Soils Reported degradation rates of TCS in soils not amended with biosolids are typically much greater than the rates measured in biosolids amended soils. Ying et al. (2007) conducted a biodegradation study in a loam soil (pH 7.4; 13 g kg 1 OC) where TCS (1 mg kg 1 ) was spiked in the soil and incubated under aerobic and anaerobic cond itions for 70 d. The half life, quantified via lo ss of extractable TCS over time wa s 18 d under aerobic conditions, and TCS was reported to be persistent under anaerobic conditions. A similar laboratory study was conducted by Kwon et al. (2010) with two soils incubated for 100 d. The soils included a fine loam (pH 7.8; 18 g kg 1 ) and a coarse loam (pH 4.7; 6.5 g kg 1 ) spiked with TCS (1 mg kg 1 ), either directly or as a part of biosolids. The half life estimated (loss of extractable TCS) in th e unamended soils was 2 to 13 d, but the addition of biosolids and /or soil sterilization sign ificantly retarded degradation (Kwon et al., 2010) The reported half lives were 50 to 108 d in bios olids amended treatments and 51 to 60 d in unamended sterilized soil. Al Rajab et al. (2009) compared TCS mineralization rates in soil (sandy loam; OM: 37 g kg 1 ) alone, and soil amended with either liquid or dewatered biosolids. Spiked TCS mineralized more (~5%) in un amended soil and soil amended with liquid biosolids (~17%) than in soil amended with dewater ed biosolids (~1%). The greater mineralization in the liquid biosolids treatment was attributed to biosolids stimulation of TC S degrading micro organisms due to readily available OC in liquid biosolids The slower degradation in the soil amended with dewatered biosolids was attributed either to the sor ption of TCS to the biosolids in forms that are not readily available to degrading
71 organisms, or to the preference of the soil microbes for biosolids, rather than TCS, as a carbon source. The assumption of sorption reducing bioavailability in the dewatered biosolids was also consistent with the zero order mineralization kinetics determined for the dewatered biosolids treatments, as compared to the app roximately first order kinetics for so il alone and the liquid biosolids (Al Rajab et al., 2009). Wu et al. ( 2009) conducted a study involving TCS spiked (2 mg kg 1 ) in two soils (sandy loam and a silty clay), with and without biosolids, and incubated under aerobic conditions. In contrast to the previously mentioned studies (Ying et al., 2007; Al Rajab et al., 20 09; Kwon et al., 2010), Wu et al. (2009) found no significant affect of biosolids on TCS degradation, despite increased TCS sorption; half lives were 20 to 58 d, irrespective of the presence of biosolids R esults of most of the published studies (Ying et a l., 2007; Al Rajab e t al., 2009; Kwon et al., 2010), except Wu et al. (2009), agree with our assessment that the persistence of biosolids borne TCS (cake or dewatered) is considerably great er than the persistence of TCS in un amended soils. P ersistence o f biosolids borne TCS was estimated in laboratory incubation s here, but persistence is expected to vary under field conditions. Al Rajab et al. (2009) opined that TCS degradation rates can vary with temperature, soil moisture conditions, and the degree of biosolids incorporation in soils. Field amended soils, though dominantly aerobic, can contain anaerobic micro sites (especially inside biosolids clumps) that could hinder microbial degradation process. Field soil temperatures and moisture contents are also expected to be much more variable and extreme than in well controlled laboratory incubations. Field degradati on rates can be expected to lie between the degradation rates determined in aerobic laboratory conditions and those in
72 anaerobic sediments, especi ally when the fields are even periodically anaerobic or under drought for a lo ng time. Cha and Cupples (2009) described a field study where biosolids were applied on ten sites between 2003 and 2007, and Xia et al. ( 2010) investigated a site where biosolids had been applied for 33 years. Both studies estimated the TCS concentrations based on the biosolids loading rates and approximate biosolids TCS concentrations. The measured TCS soil concentrations in the two studies [Cha and Cupples (2009); Xia et al. ( 2010) ] were much smaller than the expected concentrations suggesting significant degradation, but half lives were not estimated in the either study. Lozano et al. (2010) suggested TCS degradation half lives (t 1/2 = 107 d) rates from field data (unknown soil texture), similar to our laboratory determination (t 1/2 = 100d) Lozano et al. (2010) suggested a degradation model based on the field data that could be applicable for laboratory and field conditions. There are a variety of conditions that can affect the TCS degradation under fie ld conditions, but based on our data and the available literature, we conclude that laboratory estimations successfully approximate persistence under field conditions.
73 Table 4 1. Select ed physico chemical properties of the soils and biosolids used in the study Sand Silt Clay Organic carbon Water holding capacity pH (1:1) g kg 1 Immokalee fine sand (IFS) 990 <10 <10 11 100 4.5 Ashkum silty clay loam (ASL) 240 420 340 34 300 6.6 Anaerobically digested biosolids (CHCC) 250 750 8.0 Table 4 2 Biodegradation experiment t reatments Soil Spike Treatment Replicates Samples Immokalee 1.3 x 10 6 dpm Biotic 32 Immokalee 1.3 x 10 6 dpm Inhibited biotic 32 Ashkum 1.3 x 10 6 dpm Biotic 32 Ashkum 1.3 x 10 6 dpm Inhibited biotic 32 Controls Immokalee No spike Biotic 4 Immokalee No spike Inhibited biotic 4 Ashkum No spike Biotic 4 Ashkum No spike Inhibited biotic 4 Soil less 1.3 x 10 6 dpm Autoclaved water 2 Total 146
74 Table 4 3 Percent recoveries standard errors of 14 C TCS in IFS (Immokalee fi ne sand) biotic soil treatment [week (wk) 0 18] Time (weeks) Recovery (%) Fraction wk0 wk2 wk4 wk6 wk9 wk12 wk15 wk18 H 2 0 3.70.1 3.50.5 2.20.1 2.50.3 2.20.2 2.00.2 4.60.3 6.40.4 MeOH 881.7 794.3 633.3 624.5 464.6 454.4 352.3 330.4 NaOH 0.10.1 6.24.9 121.9 110.5 141.3 151.5 181.7 180.9 14 CO 2 0.00.0 0.00.0 0.00.0 0.00.0 0.20.0 0.2 0 .0 0.4 0 .0 0.5 0 .0 Combustion 2 .0 0.2 5.40.4 9.41.6 140.8 293.0 312.2 331.5 455.8 Total 943.2 957.2 861.7 913.5 911.5 965.3 913.0 1025 Overall average total recovery (wk0 18) 92.85.9 Table 4 4 Percent recoveries standard errors of 14 C TCS in IFS (Immokalee fine sand) inhibited soil treatment [week (wk) 0 18] Time (weeks) Recovery (%) Fraction wk0 wk2 wk4 wk6 wk9 wk12 wk15 wk18 H 2 0 3.40.4 4.31.2 4.20.8 4.80.3 4.00.4 5.20.4 80.8 7.51.7 MeOH 873.9 712.8 762.9 670.8 734.0 742.7 635.7 724.4 NaOH 1.10.1 8.41.6 2.30.8 7.20.8 4.70.5 7 .0 0.7 233.4 6.71.5 14 CO 2 0.00.0 0.00.0 0.00.0 0.00.0 0.1 0.1 0.1 0.0 0.4 0.0 0.40.0 Combustion 2.90.4 111.5 3.60.5 111.8 7.71.3 5.60.8 8.50.8 7.40.9 Total 953.2 941.7 861.8 911.9 912.2 921.4 1036.7 946.0 Overall average total recovery (wk0 18) 93.45.6
75 Table 4 5 P ercent recoveries standard errors of 14 C TCS in ASL (Ashkum silty cla y loam) biotic soil treatment [week (wk) 0 18] Time (weeks) Recovery (%) Fraction wk0 wk2 wk4 wk6 wk9 wk12 wk15 wk18 H 2 0 0.50.0 0.20.0 0.1 0.1 0.1 0.0 0.30.0 0.3 0.0 0.3 0.0 0.4 0.0 MeOH 416.4 254.0 192.3 202.8 110.5 101.3 151.1 6.72.9 NaOH 1.21.5 7.53.9 5.50.7 50.7 3.10.3 3.70.4 4.60.4 2.50.8 14 CO 2 0.00.0 0.00.0 0.00.0 0.00.0 0.00.0 0.1 0.0 0.3 0.0 0.30.0 Combustion 655.1 715.1 824.3 826.2 869.5 879.9 855.0 903.7 Total 1089.2 1037.2 1064.7 1066.2 1009.1 1018.4 1056.0 1005.6 Overall average total recovery (wk0 18) 1043.2 Table 4 6 Percent recoveries standard errors of 14 C TCS in ASL (Ashkum si lty clay loam) inhibited soil treatment [week (wk) 0 18] Time (weeks) Recovery (%) Fraction wk0 wk2 wk4 wk6 wk9 wk12 wk15 wk18 H 2 0 0.50.0 0.3 0.0 0.20.0 0.2 0.0 0.30.0 3 0.4 0.0 0.50.1 0.3 0.0 MeOH 40 3.0 267.4 234.0 202.9 252 .5 292.8 211.7 7.31.3 NaOH 2 .30.2 5 .82.8 101.5 6.40.6 100.9 110.9 181.7 6.31 .8 14 CO 2 0.00.0 0.00.0 0.00.0 0.00.0 0.00.0 0.0 0.0 0.1 0.0 0.20.2 Combustion 613.0 757 .0 746 .0 768 .2 695.9 595 .9 653.6 956.2 Total 1045.5 1067.8 1064.9 1037.4 1046.4 996.5 1054.7 109 6 .7 Overall average total recovery (wk0 18) 1043.2
76 Table 4 7. Rate constants (k) and regression coefficients (R 2 ) obtained for the biodegradation data according to zero and first order models. Zero order First order k ( g g 1 wk 1 ) R 2 k (wk 1 ) R 2 TCS IFS soil 0.835 0.77 0.004 0.78 ASL soil 4.452 0.9 7 0.036 0.91 Me TCS IFS soil 0.823 0. 74 0.059 0.70 ASL soil 4.460 0.9 5 0.089 0.7 6 Figure 4 1. Schematic of the biodegradation experimental d esign (Adapted from Snyder, 2009) Multi sample rack Biosolids amended soil sample Air flow Air flow Air pump 1 N KOH Base traps
77 *Combustible (bound) fraction Figure 4 2 Mean percent recoveries (n=4) of 14 C TCS in various fractions from biosolids amended Immokalee fine sand (IFS) (a) biotic and (b) inhibited treatments (wk0 18) normalized for total 14 C TCS detected in each treatment. Error bars represent one standard deviation of the mean
78 *Combustible (b ound) fraction Figure 4 3 Mean percent recove ries (n=4) of 14 C TCS in various fractions from biosolids amended Ashkum silty clay loam (ASL) (a) biotic and ( b ) inhibited treatments [week (wk) 0 18] normalized for total 14 C TCS detected in each treatm ent Error bars represent one standard deviation of the mean.
79 Figure 4 4 Mean cumulative CO 2 production standard error bars from biosolids amended (a) Immokalee fine sand (IFS) (b) Ashkum silty clay loam ( ASL) soil treatments over 18 weeks (same letters represent no statistical difference among treatments). (A) (B)
80 Fig ure 4 5 Typical RAD (top) for (a) week 0 extracts, (b) w eek 3 18 extracts for Im mokalee fine sand (IFS) and Ashkum silty clay loam (ASL) soils Peak 1 represent impurities, p eak 2 represents TCS and peak 3 represents Me TCS Fig ure 4 6. Chemical structure of Triclosan [TCS; 5 chloro 2 (2,4 dichloro phenoxy) phenol ], and Methyl triclosan [ Me TCS ; 2,4 dichloro 1 (4 chloro 2 methoxyphenoxy) benzene] Biomethylation
81 Fig ure 4 7 Primary degradation half life (d) estimated using a zero order model from the proportion s of 14 C detected as TCS and Me TCS for biosolids amended (a) Imm okalee fine sand (IFS) biotic (b) Ashkum silty clay loam (ASL) biotic soil treatments The inte rsection of dotted lines in (b) represents our estimation of degradation half life.
8 2 CHAPTER 5 IMPACTS OF BIOSOLIDS BORNE TCS ON SOIL DWELLING ORGANISMS Background Toxicity of a chemical depends on the i n the environment, rather than total chemical concentration in a soil (Alexander, 1999), as the latter often fails to pre dict the toxicity or bioaccumulation in an organism accurately. Further, a c hemical bioavailability and extractability tends to decrease as contact time between the chemical and the soil increases (Alexander et al., 2000) making assessment of bioavailabi lity via determination of total concentration problematic. Bioavailability assessments are performed using toxicity tests (Singh and Ward, 2004 ). The test s generally involve exposure of test organism to the chemical of interest at varying concentrations, a nd subsequent monitoring of biological end points (e.g. mortality, reproduction, growth, and behavioral changes). Effects on reproduction, growth and development are the most useful data for conducting a ri sk assessment. However, only mortality data [e.g. lethal dose (LD 50) or lethal concentration (LC 50)] are measured for majority of chemicals (Singh and Ward, 2004). These data may not be sufficient to protect ecological health, but are critical in conducting screening level risk assessments in the abse nce of chronic toxici ty or reproduction effect data. Triclosan is a constituent of various personal care products and is commonly detected in biosolids (Chapter 2). Concern for TCS effects arise when biosolids are land applied. A T CS degradation half life of ~ 100 d, and the appearance of a metabolite [Methyl TCS (Me TCS) ] (Chapter 4) suggest that both TCS and Me TCS could exist in soils for extended times and have the potential to adversely affect or bioacc u mu late in soil dwelling o rganisms. B ioaccumulation studies c onducted on aquatic species suggest
83 that algae and snail s are go od candidates for assessing distribution in the environment and conducting aquatic risk assessments. Coogan et al. (2007, 2008) reported bio accumu lation of TCS and the deg radation metabolite ( Me TCS ) in algae and snail s grown in surface water and streams affected by wastewater effluent. The measured bioaccmumulation factors (BAFs) were 1000, which suggest rapid TCS accumulation in the organisms in the aquatic environments. However, TCS bioavailability is likely to be smaller in soil s because TCS has hydrophobic properties and tends to partition to the solid fraction of soils and sediments. E arthworms are appropriate model organisms for estimating ch emic and bioavailability in soils. Earthworms live in close contact with soi l, have thin permeable cuticles, and the majority of the worm diet consists of soil (Suter et al., 2000). Further e arthworms are exposed to chemicals through both ing esti on and dermal contact, and represent a significant portion of diet of many vertebrates (Suter et al., 2000). Thus, earthworms are often used as surrogate s fo r other soil dwelling organisms for conducting ecological risk assessments. Samsoe Petersen et al. (2003) found no negative effect of TCS on earthworm 1 in an artificial soil. Lin et al. (2010) mg kg 1 Kinney et al. (2008) screened 70+ organic compounds in earthworms collected 31 and 156 d post application (18 Mg ha 1 ) of biosolids to soil The data suggested TCS accum ulation in earthworms with BAF s ranging from 10.8 to 27. T he authors hypothesized the mobilization of TCS d irectly from biosolids (solid phase) to the earthworms The study results were confounded by the detection of significant TCS
84 (833 g kg 1 ) in the contro l (no amendments for last 7 years ) as well as in biosol ids amended soil Kinney et al. (2008), however, concluded that anthropogenic exposure to chemicals like TCS is widespread bioaccumulat ion is likely, and urged further quantification Reiss et al. (2009) conducted a terrestrial risk assessment for TCS and addressed the exposure to TCS experienc ed by earthworms (no BAFs reported) terrestrial plants and soil organi sms from biosolids amended soil as well as secondary exposure to birds and mammals. Lin et al. (2010) conducted a toxicity test on earthworms ( Eisenia f etida ) exposed to TCS spiked in s oil (no biosolids). The study examined the activity of enzymes such as catalase, superoxide dismutase to assess TCS toxicity. R esults suggested that TCS (>1 mg kg soil 1 ) i s genotoxic and caused oxidative stress in earthworms Unfortunately, Lin et al. (20 10) reported no bioaccmulat ion data Higgins et al. (2011) examined TCS bioaccumulation in earthworms grown in field soils collected from Illinois previously amended with biosolids. T he bioaccumulation results were variable with no clear relationship between TCS exposure levels and concentration in the earthworms The authors acknowledged that the bioaccumulation was estimated based on a few samples (no replicates), and the internal inconsistency of the data led to inconclusive resu lts Internally consistent earthworm toxicity data for un amended soils (Lin et al., 2010; Samsoe Peterson et al., 2003) generally suggest minimal toxicity to earthworm health. However, bioaccumulation data were variable in earthworms grown in TCS spiked u n amended soil and biosolids amended soil. Accurate estimates of biosolids borne TCS toxicity and bioaccumulation are critical to accurately assessing ecological health risk.
85 We hypothesize that: Biosolids borne TCS is not toxic to earthworms but can acc umulate in their tissues. The e xtent of accumulation in earthworms varies with soil TCS concentration, and condition s in which earthworms are grow n (laboratory vs field). The objective was to investigate the toxicity and bioaccu mulation of biosolids borne TCS to earthworms. The laboratory st udy was conducted following the Office of Prevention, Pesticides and Toxic Substances ( OPPTS ) Earthworm Sub chronic Toxicity Test guideline ( Guideline 850.6200 ) (USEPA, 1996b). T he guideline r equire s a range finding and a definitive test, and prescribe s direct addition of the chemical of interest to a natural soil. The protocol was modified herein to deliver TCS as a component of biosolids in an effort to better mimic the primary mechanism of TCS transfer to the soil through biosolids. T o estimate the long er term bioavailability of TCS we also analyzed earthworms recently collected from a field soil previously amended with biosolids and a representative control site. Material and Methods Chemicals, Biosolids and Soils Triclosan (C AS No. 101 20 2; >99.9% purity) standard was purchased from United States Pharmacopeia ( USP ) (Maryland, USA). I nternal standard ( 13 C 12 TCS ), pyridine, bis (trim ethylsilyl) trifluoroacetamide ( BSTFA ) +1% trimethylc hloro silane ( TMCS ) were obtained from Sigma A ldrich (St. Louis, MO ). Methyl TCS standard was purchased from Wellington L aboratories (Shawnee Mission, KS). P otassium chloride and solvents [ methanol (MeOH), acetone] of HPLC grade or greater were purchased f rom Sigma Aldrich (St. Louis, MO) JT Baker (Phillipsburg, NJ) or Fisher Scientific (Atlanta, GA) Anaerobically digested cake b iosol ids (identification code: CHCC) was
86 collected from a domestic WWTP in Illinois; this sample had an inherent TCS concentrati on of 5 mg kg 1 ( Chapter 2 ). Two soils, an Immokalee fine sand (IFS) (sandy, siliceous, hyperthermic Arenic A laquods), and the Ashkum silty clay loam (ASL) (f ine, mixed, sup eractive, mesic Typic E ndoaquoll s) were collected from sites with no known history of receiving land applied biosolids or sludge. An artificial soil (68% silica sand, 20% kaolin clay, 10% sphagnum peat moss 2% calcium carbonat e; all by w eight ) was prepared by mixing the various ingredients in the laboratory as prescribed by OPPTS guide lines (USEPA, 1996 b ) and utilized in the earthworm toxicity test along with the two natural soils. The earth worms were purchased from Carolina biologicals (NC). P rior to use, the earthworms were grown in moist peat moss (growing medium ) and fed with worm f ood (M agic products WI ) each day for 3 weeks. Rang e F inding Toxicity Test Design A range finding test was conducted to identify the appropriate range of biosolid s borne TCS concentrations for a subsequent definitive toxicity assessment Two gram samples of oven dry CHCC biosolids ( TCS = 5 mg kg 1 ) were spiked with 0, 10, 100, 1000, or 10,000 mg TCS kg 1 biosolids using MeOH as the carrier solvent, and were subsequently dried, re wetted, and equilibrated for 48 h. Treatment spikes were in addition to the b iosolids inherent TCS concentr ation, so the nominal final concentr ations were 5, 15, 105, 1005, and 10,005 mg TCS kg 1 biosolids. The EPA guideline prescribes the use of an artificial soil to avoid synergistic effects with unknown chemicals present in natural soils. We also included two natural soils to estimate TCS toxicity in real world scenarios. Biosolids were amended to 200 g ( dry wt.) of the artificial, IFS and ASL soils at a 22 Mg ha 1 equivalent ra te in 800 mL glas s Mason jars. The major physico chemical
87 properties of the soils and biosolids used in the toxicity and bioaccum ulation study are described in T able 5 1. Earthworms were hand picked from the growing medium and washed with deionized water before use in the toxicity test. T he soils were brought to field capacity (artificial, 35%; IFS, 10%; ASL, 30% by wt.) and ten Eisenia fetida earthworms were added to each sample. A lso included were un amended soil control to quantify potential effects of biosolids addition, and carrier solvent free biosolids amended soil control (no additional TCS spike) to quantify carrier solvent effect. Lids were placed loosely on top of the incu bation jars to reduce moisture loss and to prevent earthworm escape. Dead earthworms at the soil surface were counted and removed as necessary each day, and the number of living earthworms were tallied each week for four weeks (28 d). Replicates are not re quired by the OPPTS range finding guideline, nor are extraneous food sources. The USEPA guideline also prescrib es that the test results become 20 % earthworm die off occurs in the control. Definitive Toxicity Test Design Results of the ran ge finding test suggested a narrower range of TCS concentrations for the definitive test with the IFS soil Thus, the test included biosolids spiked at concentrations of 2.5, 5, 10, 50, and 100 mg additional TCS kg 1 biosolids. The test procedure was same as the range finding test except that the definitive test included four replicates for each treatment Accounting for the inherent TCS concentrattion of 5 mg kg 1 the nominal final concentrations were 5, 7.5, 10, 15, 55 and 105 mg kg 1 biosolid s. Also included were the soil only control and the solvent free biosolids amended soil con trol. Statistical analyses were performed with SAS software, version HSD) to assess the treatment effect.
88 Earthworm Bioaccumulation Test D esign for the laboratory study An earthworm TCS bioaccumulation test was run concurrently with the definitive earthworm toxicity test. The test included both IFS and ASL soils to quantify TCS earthworm accumulation in soils of different textures. The TCS treatments included a low er concentration range (as determined by the range finding test) as l ive earthworms were required for quantifying chemical accumulation Two gram samples of CHCC biosolids were spiked with 0, 2.5, 5, 10, 50, and 100 mg additional TCS kg 1 biosolids and amended to the IFS and ASL soils, in quadruplicate as in the definitive toxic ity test. The nominal final concentrations were 5, 7.5, 10, 15, 55 and 105 mg TCS kg 1 b iosolids. Surviving earthworms at the end of week 4 (28 d) were removed, counted, washed, and weighed. The earthworms were allow ed to depurate for 24h in petri dishes lined with moistened filter paper (Banks et al., 2006), and were subsequently frozen until analyses. We recognize that the bi rds and worm eating animals feed on non depura ted earthworms but depuration is prescribed by the OPPTS procedure (USEPA, 1996b) D epuration allows earthworms to excrete TCS contaminated soil or orga nic matter remaining i n the gut, so that the measured TCS concentration in earthworms reflects accumulation in the tissue, rather th an TCS sorbed to gut contents. Bioaccumulation in field soils L ong er term bioavailability of TCS was estimated by utilizing e arthworms and corresponding soil samples collected at a field site in Illinois The site received a single application of CHCC biosolids ( application rate of 228 Mg ha 1 ) in 2008 and the earthworms were collected in 2010 E arthworms were also collected from a n adjacent c ontrol site ( 18 m apart ; same soil texture) that had no history of biosolids application.
89 T he earthworms were collected utilizing the hot mustard extraction method described in Lawre nce and Bowers (2002). Briefly, a mustard (allyl isothio cyanate) solution is applied to the soil surface which encourages earthworm emerge nce to escape the The method was ef f icient in collecting a consistent number 20 from each location ) of earthworms and did not require digging or hand sorting. The collected worms were kept in wet peat moss in aerated bags and promptly shipped to the laboratory under ice packs. Upon arrival, earthworms were cleaned with DDI water to remove peat moss residue, kept in petridishes with wet filter p aper and allowed to depurate for 24 hours. Sample extraction and derivatization for earthworm bioaccumulation A sample extraction technique was modeled after those d escribed by Higgins et al. (2009 ) and Snyde r et al. (2011 ) with few modifications Laborat ory collected frozen earthworms were thawed, and fresh earthworms from the field were transferred to aluminum weigh boats, dried at 50C to a constant weight, and ground. The dried and ground tissue (0.5 1 g) was loaded into 25 mL glass centrifuge tubes. A solvent mixture (10 mL) of MeOH+acetone (50:50, v/v ) was added to each centrifuge tube. The extraction was performed on a platform shaker for 18 h, followed by 60 min of soni cation (Branson 2210, Danbury, CT ; temp. 40 C, 60 sonication s min 1 ) in a water bath. Suspensions were centrifuged at 800 x g, and the supernatant was transferred to 20 mL glass scintillation vials. The extraction procedure was performed twice and the extracts were combined and dried under a gentle nitrogen stream. The extr acts were reconstituted in 1 mL of MeOH and transferred to 2 mL microcentrifuge tubes. The microcentrifuge tubes were then centrifuged at 18,000 g for 30 min, supernatant removed and transferred to 1 mL GC vials, 13 C 12 TCS was added, and the mixture drie d
90 under a mild N 2 stream. We included solvent controls (containing no TCS) that were subjected to the same extr a c tion proc e dure to confirm that TCS was not introd u c ed during the extraction The derivatization was performed according to the method by Sharee f et al. (2006) with slight modification. Briefly, the dried contents were reconstituted in a mixture of 4:1 of derivatization agent BSTFA +1% TMCS and the solvent (pyridine), vortexed for 10 s, an d heated in a dry bath for 1h The samples were then transf erred to fresh G C vials with glass inserts and Teflon lined caps. Fresh earthworms (untreat ed) were also depurated, frozen, thawed and spiked with known concentrations of TCS, 13 C 12 TCS and Me TCS. The spiked samples were incubated for 2 4 h to allow carrier solvent evaporation. Earthworms c ontaining spiked chem i cals were ex tracted with the same solvent mixture of MeOH+acetone (as described above) to determine percent recoveries which were >90% for spiked TCS and Me TCS. Instrument analyses a nd quantification The samples obtained after the derivatization step (as described above) were analyzed by splitless injection (5 L) on the Varian 4000 Gas Chromatograph (GC) equipped with Restek Rxi 5Sil column coupled with Varian 4000 MS/MS. The GC colu mn was initially held at 100 C for 1 min, and then increased to 310 C at a rate of 10 C min 1 with no final hold time. Helium was used as a carrier gas. The ion trap, manifold and transfer line temperatures were 200, 80 and 270 C, respectively, and the ion ization source was internal/electron ionization. Data acquisition monitored two fragment ions for each compound. The ion masses were m/z of 345/347 for TCS trimethylsilylether, 302/304 for TCS OMe, and internal standard 13 C 12 TCS trimethylsilylether mas s w as monitored at 357/359. The samples ran for 22 minutes with an average retention time of 14.8 min for a ll compounds For quantification, an 8 point
91 internal calibration curve was generated in the TCS concentration range of 1 1000 ng g 1 with an average R 2 >0.99 9 The LOD was 0.22 g g 1 and LOQ was 0.7 g g 1 for both TCS and Me TCS in the earthworm tissue. The LOD and LOQ values were calculated as 3 fold and 10 fold, respectively, the standard deviation in the signal from multiple runs of the lowest cali bration standard (S ignal/Noise >10) (USEPA, 1984). The details of detection limits and recoveries are provided in the Appendix C. Results and Discussion Range Finding Toxicity Test 1 had minimal adverse effect on earthworm survival in the AS L and artificial soils. In the ASL soil, all the earthworms in all the treatments sur vived at the end of week 4, except a 10% die off at a concentration of 1005 mg kg 1 (Figure 5 1b) Similarly, in the artificial soil, TCS conce ntration did not adversely affect the worm survival, except for a 20% die off at a concentration of 105 mg kg 1 and 10% die off at 5 mg kg 1 ( solvent free ) (Figure 5 1c) We attributed the death of 20% of the earthworms (2 earthworms out of 10) to natural causes (such as lack of sufficient food or stress) due to lack of adverse effect at greater (>105 mg kg 1 ) concentration s The data suggest that TCS concentration 10,005 mg kg 1 biosolids (equivalent to an amended soil concentration of 100 mg kg 1 ) have no adverse effect on earthworm survival in the ASL and the artificial soils. Earthworms in the IFS soil were adversely affected by TCS addition By week 4, nearly 100% mortality occured at TCS concentrations 5 mg kg 1 biosolids (Figure 5 1 a ). There was a 25% die off at week 4 even in the soil onl y control (no biosolids). D ue to the >20 % die off in the control the experimental results were unacceptable The die off could be attributed to the lack of sufficient food for the worm survival without
92 biosolids as no extraneous food source was supplied. The earthworm survival differed in the TCS treatments and the solvent free control (biosolids amended) had a 100% die o ff by week 4. We tentatively attrib uted the die off in the solvent free control to the absence of solvent residues acting as a food source to the earthworms. No further investigation for the causes of the die off was performed as it was a range finding test. Definitive conclusions and calculation of LC 50 values can only be performed in the definiti ve study, which includes replicates. However, t he EPA guideline s do not require a definitive test if the LC 50 value is greater than the highest concentration tested ( i.e. 10,005 mg kg 1 biosolids) Thus, due to the lack of adverse effects i n the ASL and ar tificial soils, only the IFS soil was utilized in the defin itive toxicity test Definitive Toxicity Test (IFS Soil) The mean earthworm s urvival was >90% in all the treatments as well as in soil only control except at concentrations of 10 and 15 mg TCS kg 1 biosolids (Fig ure 5 2). Unlike in t he range finding test, the soil only control (no biosolids) appeared to prov ide sufficient nutrients for earthworm survival up to week 4 There were no adverse TCS treatment, biosolids addition, or carrier solvent addi tion effect s on the survival of the earthwo rms. T he earthworm survival was slightly affected in the concentration range of 10 to 105 mg TCS kg 1 biosolids, but there was no statistical difference among the various treatments due t o the large standard devia tions (Fig ure 5 2). Thus, no adverse effect was observed on the earthworm survival up to a TCS biosolids concentration of 105 mg kg 1 A dverse effects were not anticipated due to Me TCS exposure, as a suggested degradation half life of TCS i s at least 77 d (Chapter 4) so significant Me TCS formation was not expected in the short length (28 d) of our present study. A d efinitive earthworm LC 50 value cannot be calculated from our study because there was
93 no significant adverse effect up to the maxim um tested concentration. The estimated L C 50 value in the IFS soil is >105 mg TCS kg 1 biosolids. If the biosolids LC 50 value is ~100 mg TCS kg 1 the application of biosolids at 22 Mg ha 1 followed by incorporation at 15 cm depth (~100 fold dilution of the TCS conc entration), results in LC 50 of ~1 mg TCS kg 1 soil. Similarly t he range finding test cannot be use d to calculate an LC 50 value as no replicates were involved; h owever, a rough approximation yields an LC 50 value of >10,005 mg TCS kg 1 biosolids or >10 0 mg TCS kg 1 soil for t he artificial and the ASL soils. The ecological structure activity relationship ( ECOSAR ) program ( USEPA, 2009b) is a model based on the structure activity relationship concept, and predicts the aquatic toxicity of a chemical. The tox icity is estimated based on the similarity of a has been previously measured. Extrapolation from aquatic to terrestrial toxicity may not be accurate ( Hartnik et al., 2008 ) but for the sake of comparison, ECOSAR was utilized to calculate LC 50 value for te rrestrial species. Most calculations in the ECOSAR program are based on the K ow and do not consider soil/sediment parameters where the organism is growing. The earthworm 14 d LC 50 value predicted by the ECOSAR program is ~22 mg TCS kg 1 soil. Our estimated LC 50 value in IFS soil (~1 mg kg 1 ) was mu ch smaller, and the LC 50 value in the other two soils (>100 mg kg 1 ) was considerably greater than the value calculated by the ECOSAR Thus, t he ECOSAR underestimate s the toxicity in one ( IFS) soil and over estimate s the toxicity in the other two (artificial and ASL ) soils T he model estimated LC 50 values should be used with caution.
94 Our toxicity data in the artificial and the AS L soil are consistent with a short term (2 week ) Danish study (Samsoe Petersen et al., 2003) in which there was no negative effect of TCS on earthworm survival a t a concentration 1,026 mg kg 1 in a similar artificial soil. A 14 d TCS exposure study (Lin et al., 2010) was conducted in a soil similar in texture to ASL soil, but with lower OC content (pH = 8.1, OC = 12.8 g kg 1 ) Data suggested no adverse effec t on the earthworm survival at TCS soil concentration s 100 mg kg 1 Although our resul ts were in agreement with the published studies, t he LC 50 value s determi ned in our short term study are best regarded as estimates due to lack of adverse effects at the highest concentration tested F uture investigations c ould include long term TCS impact studies or studies using greater TCS concentrations although greater concentrations are environmentally unrealistic Further, toxicity effects of major metabolite of TCS (Me TCS) were not assessed in ou r study and should be ad dressed. Bioaccumulation Laboratory Study The measured TCS concentrations accumulated by earthworms incubated in two soils amended with biosolids borne TCS are presented in T ables 5 2 and 5 3 The earthworm tissue concentrations were utilized to calculate BAF values for each soil. De gr adation study results (Chapter 4 ) suggested the formation of Me TCS in the biosolids amended soil s (t 1/2 ~77d) We might expect formation of Me TCS in amended soils and/or biosolids and possible subsequent accumulation in eart hworms However, given the short length of the earthworm bioaccumulation study (28 d) and a suggested half life of at le a st 77 d (Chapter 4 ), we did not expect significant Me TCS formation from the added TCS. In add ition, the Me TCS concentration in the bi osolids used in this study was below the LOQ (0.7 mg kg 1 ) of th e instrument. Thus, as expected, we did not
95 detect Me TCS in earthworms exposed to the incubated amended soils or Me TCS inherent to the biosolids Uptake assessment occured through the calculation of bioaccumulation factors (BAFs) expressed as the ratio of TCS concentration in the earthworm tissue to the TCS concentration in the soil in which the worm was exposed to TCS The average BAF value s in the two soils, irrespective of the spiked TCS concentration were 6.5 0.84 for the IFS soil (Table 5 2 ) and 12 3.08 for the ASL soil (Table 5 3 ). The average values were significantly different (p<0.05) from each other. The difference in BAF s in the tw o soils might reflect differences in phys ico chemical properties of the two soils. The soils differed in native so il organic carbon ( OC ) contents (11 g kg 1 for IFS soil and ~34 g kg 1 for ASL soil Table 5 1 ), suggesting more accumulation in soil with greater OC content. Application of b iosolids with an OC content of 250 g kg 1 at a rate of 22 Mg ha 1 adds an additional 2.7 g kg 1 of OC to both soils, but the total OC in the amended IFS soil (13.7 g kg 1 ) was still smaller than the amended ASL soil (36.7 g kg 1 ) Th e trend is possib ly due to eart he soil with greater OC content and ingesting both OC and TCS that were associated with the biosolids. Our results were consistent with a bioaccumulation study (Snyder et al., 2011) conducted on a similar antimicrobial compound triclocarban (TCC) The measured BAF value was greater in the silty clay loam (20 2.1) than in the sand (18 3.5), but the results were not significantly different. Luo et al. (2008) opined that besides the OC content the bioacc umulation may als o vary with the chemical tested, and perhaps with the composition of organic matter.
96 T he BAF s obtained in our study were generally no t a function of spiked TCS concentration. In the IFS soil, the BAFs were not sig nificantly diffe r ent in va rious TCS treatments. S ome c oncentration dependent accumulation was observed in the ASL soil with significantly greater accumulation at a con centration of 0.10 mg TCS kg 1 soil, however, there was no significant difference in accumulation among other TCS treatments. Kinney et al. (2008) determined earthworm TCS BAF s of 10.8 and 27 for a soil (sandy clay loam; OC 12 g kg 1 ) amended with biosolids either 31 or 156 d previously. The inherent TCS concentration of biosolids applied was 10.5 mg k g 1 Earthworm samples were collected from a soybean field site receiving biosolids as a fertilizer for the first time. The range of BAFs obtained in our study (6.5 12.7) is within the range r eported by Kinney et al. (2008), despite the shorter (28 d) expo sure time used in our study. Further, we used Biosolids Amended Soil L evel IV (BASL4) model (Webster and Mackay, 2007) to estimate earth worm bioaccumulation. The program utilizes the concept of fugacity, and model s chemical distribution in soil by assuming that the medium in which organisms are growing could be either at equilibrium, steady state, or non steady state. In our study, e arthworms were exposed to soil in closed jars, so we can assume a steady state condition for modeling purpose s The model para meters were selected based on the physiochemical properties of TCS and soil type where the organisms were grown. BASL4 assumes that the properties of invertebrate and ma mmals are representative of earthworms and shrews, respectively (Hendrik s et al., 1995; Armitage, 2004). The BASL4 predicted average BAF value for IFS soil (5.0 0.0 ) was reasonably c lose to the measured value (6.5 0.8 4) (Table 5 2) For the ASL
97 soil, the predicted average BAF value (2.2 0.0 ) was much less than the measur ed value (12 3.1 ) ( Table 5 3 ). The partitioning theory for traditional hydrop hobic organic contaminants (HOC ) is commonly used to predict the partitioning of chemicals to invertebrates from sediments ( Higgins et al., 2009 ). The bioaccumulation potential is calculated by using the lipid normalized worm and organic carbon normalized soil chemical concentrations. The HOC theory predicts a biota sediment accumulation factor (BSAF) of approximately 1.6 for nonmetabolized organic compounds if the log K ow of a compound is les s than 6 (Morrison et al., 1996). The BSAF values were estimated in our study as: (5 1) The f lip (fraction of lipid) in the earthworm is 0.032 0.01 (Higgins et al., 2010) and f oc (fraction of organic carbon) is 0.011 for IFS soil and 0.034 for ASL soil (Table 5 1). The estimated BSAF values were 1.0 0.27 for the IFS soil and 9.63 3. 29 for the ASL soil. A lower than expected BSAF for the ASL soil may suggest either the metabolism of TCS or decrease d TCS bioavailabi lity. Alternatively, the difference likely suggests that HOC theory does not accurately predict the TCS accumulation in earthworms. Higgins et al. (2011) suggest ed that bioaccumulation of a similar chemical ( TCC ) was somewhat c onsiste nt with the HOC theory; but some reduction in TCC bioavailability was observed. T he HOC theory assumes equilibrium between the soil solid and soil water phases, which is not always true. Further, the expected BSAF value is ~ 1.6 when the earthworms acqui re the chemic al only from the soil solution. S oil dwelling earthworms
98 can accumulate TCS from the soil solution, direct inges tion of soil or biosolids and from direct partitioning of biosoli ds borne TCS to the worm tissue (Kinney et al., 2 008 ). Higgins et al. (2011) examined the bioaccumulation of TCS in earthworms grown in field soils previously amended with biosolids The ra nge of BSAF values was 0.31 to 1.2, and there was no dependence of bioaccumulation on TCS exposure levels. After adjusting the BSAF values from Higgins et al. (2011) to dry weight of soil and earthworms the estimated BAF values were 2 to 2.3 in the sand and 1.3 in the silty cla y loam. The BAF values estimated using Higgins et al (2011) BSAF data were smaller than the range of B AF values obtained in our study (6.5 12.7). The difference in values may be attributed to the lack of replicates in the Higgins et al. (2011) study that minimizes the significance of their data. The TCS c oncentration in earthworm tissue can also be calculated using a mechanistic approach proposed by Jager (1998) that assumes hydrophobic partitioning between the soil pore water and the earthworm tissue. The bioconcentration factor i n that case can be calculated as : (5 2) Inserting the foc (fraction of soil organic carbon) value of 0.011 (IFS) and 0.034 (ASL), a log K ow of 4.8 and log K oc in biosolids amended soils of 4.26 (Agyin Birikorang et al., 2010), yields a bioconcentratio n factor of 1.85 for IFS soil and 4.30 for ASL soil The values underestimate the potential for bioaccumulation reported in the literature and our measured values. Soil pore water earthworm models have been criticized (Suter et al., 2000) for potentially underestimating the contaminant uptake i n cases where absorption via gut is the primary mechanism (e.g earthworms).
99 Models (BASL4 and partitioning) appear to underestimate the bioaccumulation potential of TCS by earthworms, as the models assume bioaccumulation to occur only from soil pore wate r. Earthworm bioaccumulation of TCS also does not appear to follow HOC theory. Thus, the various estimates should be used with caution. Bioaccumulation Field Test The average TCS concentration measured in the earthworm s collected in a field equilibrated s oil was 4.3 1.9 mg kg 1 corresponding to a BAF value of 4.3 0.7 (Table 5 4) G rab soil sample s collected from the same area where the earthworms were collected average d 0.99 mg TCS kg 1 amended soil. The BAF value in the earthworms collected from the three locations within the amended soil were 2.68 to 5.93. T he TCS concentrations in the earth worm tissues were variable among the three locations ideally representing varying activity p atterns in the soil. Earthworms occupy a range of soil depths dep ending on the season species an d life history stage (Bouche and Gardner, 1984) and may be expo sed to varying amounts of chemical. Triclosan concentrations were non detect in earthworms and soil collected from a representative control site. T he texture of the field soil was similar to the A SL soil used in our lab oratory study, but the average BAF value in the field collected earthworms was significantly smaller (p<0.05) than the average values in laboratory stud y. Even when we compare the average BAF obtained in the field soil with the BAF in the similar ASL laboratory treatment (1 mg TCS kg 1 soil, Table 5 3), the BAF was significan t ly greater in the laboratory study (12 3.1 ) than in the field soil (4.3 1.9 ) The difference may reflect greater TCS availability under laboratory conditions, as TCS was spiked to the laboratory study soils as opposed to inherent TCS in field soils; the inherent log K d
100 values tend to be greater than the spiked log K d value s (Chapt er 3) We also speculated that high rate of biosolids borne TCS applied in the field conditions, though dominantly aerobic, may experience anaerobic micro sites (especially insi de biosolids clumps) that could hinder movement and bio availabil ity and, hence, lower the accumulation in earthworms. The BAF values (2.68 12.7) (laboratory and field) in the present study was smaller than the BAF values (10.8 27) reported by Kinney et al. ( 2008). The implications of earthworm bioaccumulation to TCS ecological risk assessment are unknown, but Snyder (2009) estimated that the earthworm predator pathway was limiting in a biosolids borne TCC risk assessment. The risk estimation of biosolids borne TCS to earthworms and other organisms are critical to iden tify potential pathways of concern. B iosolids borne TCS accumulates in, b ut is not toxic to, earthworms (h ypothesis 1). We partially accept the second hypothesis also, as the earthworm accumulation varied significantly between laboratory and field conditio ns, but the accumulation did not vary with TCS concentration in biosolids. Implications of TCS accumulation in earthworms are assessed via a risk estimation of biosolids borne TCS (Chapter 9).
101 Table 5 1. Major physico chemical properties of the soils and biosolids utilize d in the present study Soil Texture Organic carbon pH (1:1) g kg 1 Immokalee fine sand (IFS) Sand 11 4.5 Ashkum silty clay loam (ASL) Silty clay loam 34 6. 6 Artificial soil Peat moss 10% Ca c arbonate 2 % 48 7.7 B iosolids (CHCC) n/d 250 8.0 Field c ontrol landscaping Clay loam 40 7.8 Field amended landscaping Silty clay loam 80 7.1
102 Table 5 2 Measured [average ; n = 4 and standard error (SE)] TCS concentrations ( m g k g 1 ) and bioa ccumulation factors (BAF s ) in earthworms grown in the Immokalee fine sand (IFS) (same letters represent no statistical difference among treatments). Table 5 3 Measured (average ; n = 4 and SE) TCS concentrations ( m g k g 1 ) and bioa ccumulation factors (BAF s ) in earthworm s grown in the As hkum silty clay loam soil (ASL) (s ame letters represent no statistical difference among treatments ). Nominal spiked soil concentration Measured ear thworm tissue concentration Mean BAF s dry wt. Calculated BAF (BASL4) m g k g 1 0 <14 2. 2 0.025 <4.7 2.2 0.0 5 1.10.1 7.42.3 a 2.2 0. 10 1 1.60.3 12.60.7 181.3 b 120.7 a 2.2 2. 2 Average 123.1 Â§ 2.20.0 The BAF values were calculated by accounting for the total TCS concentration ( 5 m g k g 1 of inherent biosolids and the spiked) 10). The limit of quantitation of TCS was 0.7 ug g 1 for the earthworm tissue. The concentrations < LOQ were estimated as the LOQ of the instrument. Â§ The average BAFs excluded the samples where TCS concentrations in earthworms are
103 Table 5 4 Measured TCS concentrations ( m g k g 1 ) and bioaccumulatio n factors (BAF s ) in the earthworm s collected from the field equilibrated biosolids amended landscaping soil (same letters represent no statistical difference among treatments). Inherent amended soil concentration Measured earthworm tissue concentration BAF s dry wt. m g k g 1 Location 1 0.99 2.60.2 2.70.2 a Location 2 0.99 4.42.5 4.42.5 a Location 3 0.99 5.9 0.9 5.90.9 a Overall a verage 4.31. 9 4.3 0. 7
104 Figure 5 1 The earthworm survival (%) as affe cted by the biosolids borne TCS concentration and the duration of earthworm ex posure (weeks) in the biosolids amended (a) I mmokalee fine sand (IFS), (b) Ashkum silty clay loam (ASL), and (c) Artificial soils range finding te st a b c
105 Fig ure 5 2. The mean earthworm survival (%) (n=4) standard deviation as affected by the biosolids borne TCS concentration and the duration of earthworm e xposure in (weeks) in biosolids amended Immokalee fine sand (IFS), definitive toxicity test
106 CHAPTER 6 BIOSOLIDS BORNE TCS EFFECTS ON SOIL MICROBES Background The n utrient and organic rich solid produc t of wastewater treatment plants (WWTPs) is called biosolids. L and application of biosolids recycles the nutrients to accelerate plant grow th and supply abundant carbon for sustainable agriculture I n the U S WWTPs generate approximately 7 million Mg of biosolids each year ( USGS, 2008), 63% of which is land applied ( NRC 2002) Land application is considered the most suitable means of biosolids disposal /use (Epstein 2002). Besides acting as a nutrient boo st for agriculture, bi o solids addition changes diversity, rich ness, and structure of plant, animal and microbial communities (USEPA, 1991). Microbial community changes may affect ecosystem processes (such as nutrient recycling) and the effectiveness of microbial invasions (such as growth of pathogens) (Garland, 1997). Dennis and Fresquez (1989) suggested that biosolids (22 90 Mg ha 1 ) application improved soil fertility by changing the soil microbial composition. Garcia Gil et al. (2004) found that microbial biomass, respiration, and enzymatic activities increased following biosolids (36 Mg ha 1 ) application Sullivan et al (2006 a ) analyzed the long term (12 y e a rs) impac ts of biosolids application on microbial communities u sing b iolog plate analysis that compare d substrate utilization by microbial communities in control and amended soils. Analysis indicated quicker substrate utilization by communities in amended than in control soil plots, suggesting that biosolids addition changed the microbial community structure (Sullivan et al., 2006 a ). Similarly, Zer z ghi et al. (2010a) observed that long term ( 20 years ) biosolids application enhanced the a ctivity and density of micro bes and changed the microbial community composition.
107 Along with the microbial community changes, biosolids addition may alter soil microbial diversity and microbial processes Soil microbial diversity is critical in maintaining soil processes such as deco mposition of organic matter and nutrient cycling (Garbeva et al., 2004). Zer z ghi et al. (2010 b ) examined the microbial diversity in soil s following biosolids (8 72 Mg ha 1 ) applied each year for 20 years. The results suggested no adverse e ffect of biosolids application on the microbial diversity. The diversity in amended soil either remained the same or increased following biosolids a ddition Rojas Oropeza et al. (2010) amended a saline sodic sand with biosolids (28 115 Mg ha 1 soil), and observe d i ncrease d microbial diversity and rates of microbial processes ( ammonification and nitrification ) Further, b iosolids addition elevated microbial respiration ( CO 2 production ) rates in soils (Barbarick et al., 2004; Sullivan et al., 2006 a ) even 6 to 12 ye ars following the biosolids amendment H olt et al. (2010) observed increase d N mineralization (conversion of organic N to NH 4 N ) due to organic N additio n through biosolids application Dennis and Fresquez (1989) reported that biosolids application increased most of the microbial populations but, fungal diversity decreased initially. Consistently, K ourtev et al. (2003) found that bacteria dominate in fertile (e.g., biosolids applied) soil s and the fungi occur more frequently in infertile ecosystems. Thus, the biosolids application may, or may not, alter m icrobial diversity, but may change individual microbial species and their activities (e.g. N cycling, respiration) (Lawlor a et al., 2000) Besides supplying nutrients and carbon to soil biosolids add a variety of organic contaminants. A recent Targeted National Sewage Sludge Survey (TNSSS, USEPA, 2009a) found various co ntaminants of emerging concern in the biosolids Triclosan
108 (TCS) was i nclu ded among several hundred chemicals such as polycyclic aromatic hydroc arbons (PAHs), polybrominated di phenyl ethers (PBDEs), antibiotics, drugs, hormones and steroids. Triclosan is an antimicrobial compound frequently added in liquid soaps, detergents and household products, which lead to transfer of TCS to WWTPs and then partition to biosolids. The TNSSS results suggested that biosolids land applicat ion can transfer several chemicals, including TCS, to the soil environment. Previous studies suggested changes in microbial community diversity, structure, and microbially mediate d reactions following biosolids (unknown TCS concentrations) addition, but the effects of TCS have not been quantified. Previous studies (Hansen et al., 2001; Ingerslev et al., 2001; Schmitt et al., 2004 ) suggest adverse effect s of antibiotics on the number of soil micro organisms and microbial community structure. S imilar to antibiotics, TCS (an anti microbial) may also affect soil micro organisms. Studies (Liu et al., 200 9 ; Wal ler and Kookana, 2007) conducted in soils (no biosolids) suggest TCS effects on microbial diversity and structure. Li u et al. (2008 ) evaluated the effect s of TCS spiked in soil (no biosolids) on the soil microbial n umber and functional diversity. The resul ts suggested that soil TCS concentrations ( >10 mg kg 1 ) increased carbon source utiliz ation, but had no effect on the microbial population numbers and diversity Waller and Kookana (2009) found increased carbon utilization at spiked TCS concentration <50 mg kg 1 soil and suggested microbial structure changes. Svenningsen et al. (2011) suggested a decrease in cultivable microbial numbers when a sewage drain field soil was exposed to a TCS (4 mg kg 1 soil). Further, the same study suggested increased persis tence of ibuprofen and alkylbenzene sulfonate in the presence of TCS (>0.16 mg kg 1 soil) in the
109 same soil. The only study of TCS spiked in biosolids amended soil i s p reliminary work by Young (2011 unpublished, personal communication ), and suggest ed adver se effect s on th e overall community composition. Thus, definitive effect s of biosolids borne TCS on microbial community structure changes need quantification. Further, changes in microbial count (number) following TCS addition have not been reported. Simi l a r to affects of biosolids addition, biosolids borne TCS may affect microbial ly mediated reactions. Zha o ( 2006) suggested that TCS inhibited nitrification via competitive inhibition of the enzyme ammonia monooxygenase (AMO) in Nitrosomonas europaea, and the inhibit ion occurs i n both sludge and soils (Stasinakis et al., 2007). In a laboratory study, McBai n et al. (2004) reported occurre nce of TCS resistant strains of E. coli however, the study found no TCS resistant ammonia oxidizers even after chronic TC S exposure Waller and Kookana (2009) measured inhibitory effects of TCS concentration (50 500 mg kg 1 soil) on nitrification, respiration and enzyme activity in Australian soils ( un amended, no biosolids) Butler et al. ( 2011) evaluated basal and substrate induced respiration following TCS spiking (0 1000 mg kg 1 ) in un amended soils (sandy loam, loamy sand, and clay). Results suggest that TCS inhibited initial respiration in the unacclimated soils, but following the respiking, TCS acted as a C source and stimulated respiration, at least at TCS concentration <100 mg TCS kg 1 soil Overall, the l iterature suggests that microbes and microbially mediated reactions can be altered by biosolids addition or TCS spiked in to soils. However, biosolids borne TCS may have a reduced bioavailability and its effect may be differen t from TCS spiked
110 directly in the soil We hypothesize that biosolids bo rne TCS has no adverse effect on soil micro organisms or microbial mediated reactions. Objective s o f our study were to: Determine the impacts of biosolids borne TCS on microbial processes. Determine the microbial community structure changes in soils following the addition of biosolids borne TCS Determine the changes in microbial count (number ) in soi ls following the addition of biosolids borne TCS. E ffects on microbial processes were assessed according to the United States Environme ntal Protection Agency (USEPA) Office of Prevention, Pesticides, and Toxic S ubstances (OPPTS) Soil Microbial Community To xicity Test (Guideline 850.5100) (USEPA, 1996c). The guideline focuses on the effect of added chemical on the microbial processes of ammonification, nitrification, and respiration. The guideline requires a range finding and a definitive test, and prescribe s direct addition of the chemical of interest to a natural soil. The protocol was modified herein to deliver TCS as a component of biosolids in an effort to better simulate the primary mechan ism of TCS transfer to the soil in land applied biosolids. Microb ial community structure changes were assessed by the community level physiological profiling (CLPP) t echnique (Schmitt et al., 2004), utilizing biolog plates. The bacterial count (n umber) was assessed by dir ect counting (Matsunaga et al., 1995). Material and Methods Chemicals, Biosolids, and Soils Triclosan (C AS No. 101 20 2; >99.9% purity) standard was purchased from United States Pharmacopeia (USP) (Maryland, USA). M ethanol (MeOH) phenol phthalein indicator, potassium chloride (KCl) potassium hydroxide (KOH) sodium hydroxide (NaOH) barium chloride ( BaCl 2 ) hydrochloric acid (HCl) nitrate (NO 3 ) and
111 ammonium (NH 4 + ) standards were purchased from Sigma Aldrich (St. Louis, MO) or Fishe r Scientific (Atlanta, GA) Anaerobically digested biosolids (solid content = 320 g kg 1 ) (identification code: CHCC) was collected from a domestic WWTP plant in Illinois and had an inherent TCS concentration of 5 mg kg 1 ( Chapter 2 ). Two soils, an Immokalee fine sand (IFS) (sandy, siliceous, hyperthermic Arenic Alaquods) [organic carbon ( OC ) : 11 g kg 1 ] and the Ashkum silty clay loam (ASL) (f ine, mixed, superactive, mesic Typic Endoaquolls ) (OC: 34 g kg 1 ) were collected from sites with no known history of receiving la nd applied biosolids or sludge and utilized in microbi al toxicity test. Bacterial count and microbial community structure analysis included three sets of soil samples One set consisted of the control soils (no biosolids) and the soils ( lowest and highest spi ked TCS concentration treatments) from the microbial toxicity test The second set of samples consisted of the field landscaping control (no biosolids) (clay loam, OC: 40 g kg 1 ), and t he landscaping soil field equilibrated for two years following amendment with 228 Mg ha 1 biosolids (silty clay lo am, OC: 84 g kg 1 TCS concentration: ~1 mg kg 1 amended soil ). The landscaping soils were collected from Illinois fields in 2010, air dried and promptly sent to our laboratory. A th ird set of samples were obtained fr o m Will (WL silty clay loam ) and Kanka kee (KK fine sand ) Counties ( Illinois ) field research plots. The selected WL and KK soils were the control s (no biosolids), and th e soil amended with 118 Mg ha 1 biosolids in 2006 ( WL TRT ), and 155 Mg ha 1 biosolids in 2007 ( KK TRT ); soils were sampled in 2008. The WL and KK samples (both control and treatment) were air dried and stored at room temperature The soils were shipped to our laboratory in 2010 and utilized as such Collectively, the
112 three sets of samples offered a wide ar ray of soil textures, biosolids loading rates TCS soil concentration s and time since last biosolids amendment. Microbial Toxicity ( Range Finding ) Test Design The microbial toxicity test included a range finding and a definitive test. Th e range finding test was conducted to identify the appropriate range of biosolids borne TCS concentrations for a subsequent definitive toxicity asses sment. One gram samples of oven dry CHCC biosolids ( Inherent TCS 5 mg kg 1 ) were spiked with 0, 10, 100, 1000, or 10,000 mg TCS kg 1 biosolids us ing MeOH as the carrier solvent and were subsequently dried, re wetted, and equilibrated for 4 8 h in 300 mL glass Mason jars. Treatment spikes were added in addition to th e inherent TCS concentrations, so the final nominal concentrations were 5, 15, 105, 1 005, and 10,005 mg TCS kg 1 biosolids. Biosolids were amended to 100 g (dry wt.) of IFS and ASL soils at an equivalent rate of 22 Mg ha 1 and amended soils brought to fie ld capacity (10% for IFS and 30 % for ASL by wt.). A soil only control, a carrier solv ent free biosolids amended soil control ( ~5 mg TCS kg 1 biosolids) and an empty jar (to confirm efficacy of the CO 2 scrub ber system) were also included in the unreplicated study. The jars were aerated with CO 2 free humidi fied air at approximately 22C. Incoming air was stripped of CO 2 and humidified by pumping ambient air first through 2 M KOH, followed by CO 2 free water, a column of soda lime chips (Ca (OH) 2 >80%, KOH< 3%, NaOH< 2%, Ethyl violet<1%), additional 2 M KOH, and once more through CO 2 free wat er. Carbon dioxide evolved from each treatment jar was taken as a measure of microbial respiration, and was collected in a series of two base traps each containing 100 mL of 0.15 M KOH.
113 Mi crobial Toxicity (Definitive) Test Design In response to adverse effects ob served in the ASL and IFS soils, a concentration range of 0 to 1000 mg TCS kg 1 biosolids was selected for the definitive toxicity test. Biosolids were spiked with 5, 10, 50, 100 and 500 mg TCS kg 1 biosolids in the ASL soil. An additional treatm ent of 1000 mg TCS kg 1 biosolids was included in the IFS soil Treatment spikes were add ed in addition to the inherent TCS concentration, so the nominal final TCS biosolids concentrations were 5, 10, 15, 55, 105, 505 and 10 05 mg TCS kg 1 Three replicates were prepared for each treatment and the control, and for each sampling period. Sample Preparation and Analyses for Microbial Toxicity Test Samples were removed from the aeration unit and analyzed in the same manner in both the range find ing and definitive test s. B ase traps (containing KOH) connected to the treatment jars were removed and re placed with fresh KOH on days 5, 14, and 28, and analyzed for CO 2 (Anderson, 1982). A solution of BaCl 2 was used to first precipitate the carbonates (r epresenting trapped CO 2 ) in a known volume of base, and subsequently centrifuged at ~2,000g for 10 min One mL of the supernatant was then transferred to a glass 20 mL scintillation vial, treated with phenolphthalein ind icator, and titrated to pH of 8 to 9 with 0.1 M HCl. The unused KOH remaining in the base trap was used to calculate the moles of KOH neutralized by evolved CO 2 Also on days 0, 5 and 28, subsets (10 g dry wt.) of amended soil were sampled and shaken with 100 mL 1 M KCl to extract N0 3 N0 2 N and NH 4 N (Bremner, 1996). Extracts were filtered (0.45 m) and analyzed for NO 3 NO 2 N (USEPA, 1993a, Method 353.2) and NH 4 N (USEPA, 1993b, Method 350.1) to assess TCS impacts on nitrification and ammonification, respectively. Method 353.2 accomplish es NO 3
114 reduction by passing soil extract through a copper cadmium coil, and the resultant total NO 2 is reacted with multiple reagents to form an azo dy e for colorimetric analysis. Results were reported as mg N0 3 N0 2 N per kg s oil or biosolids amended so il. Me thod EPA 350.1 results in conversion of extracted NH 4 + to NH 3 which then reacts with phenol to yield indophenol for colorimetric analysis. Results are reported as mg NH 4 N per kg s oil or biosolids amended soil. A r apid flow analyzer ( Alpkem, VA ) was utilized for NO 3 NO 2 N analysis, and AQ2 discrete analyzer ( SEAL, WI ) was utilized for the NH 4 N analysis. Soil samples from control and highest TCS spiked treatment s (IFS 500 and ASL 500) utilized in the definitive test were subjected to molecul ar (DNA) analysis. The analysis included DNA isolation and amplification. The DNA was extrac ted from the soil s (0.25 g wet weight) using MoBio PowerSoil DNA Isolation Kit (MoBio Laboratories, Solana Beach, CA) according to the manufacturer instruction s and stored at 20 C until use. Polymerase chain reaction ( PCR ) amplification of bacterial amoB gene was performed using primers amoA 1F (5' GGGGTTTCTACTGGTGGT 3') and am oA 2R (5' CCCCTCKGSAAAGCCTTCTTC 3') ( Rotthauwe et al. 1997) and GoTaq Green Master Mix (Promega, Madison, WI ). The reaction mixture contained Master Mix, 0.5 M of each primer, 2 of a ten fold DNA dilute, and distilled water to make a total volum e of 50 l The P CR was conducted with a BioRad iCycler thermal cycler (Hercules, CA) under the following thermo cycling conditions: initial enzyme activation and d enaturation at 95C for 15 min, 35 cycles of 95C for 30 sec, 55C for 45 sec, and 72C for 45 sec with a final e xtension step at 72C for 7 min. From the PCR reaction, a 491bp fragment of AOB gene was amplified
115 The PCR ampl ification of amoA was conducted using primer s A19F (5' ATG GTC TGG CT (AT) AGA CG 3') and 643R (5' TCC CAC TT (AT) GAC CA (AG) GCG GCC ATC CA 3') ( Leininger et al. 2006), which produce a 624bp amplicon The s ame PCR mixture used for amoB was used for amoA, except for the primers The thermo cycle consisted of the denaturation of 15 min at 95C, followed by 30 sec at 95C, 45 sec at 55C, and 45 sec extension at 72C for 30 cycles, and a f inal extension of 72C for 7min. The PCR products were analyzed by electrophoresis through 1.5 % Tris acetate EDTA (TAE) agarose gels and visualized by staining ethidium bromide and exposure under UV light Microbial Community Structure Test Design Microbial community structure changes were assessed with the community level physiological profiling (CLPP) technique (Schmitt et al., 2004), utilizing micr otiter biolog plates Prefilled microtiter plates (ECO microplate; Biolog, Hayward, CA) contain ed triplicates of 31 organic substrates such as sugars, and amino acids as well as a tetrazolium dye. M icrobial grow th on the substrates is assessed from the co lor development in tetrazolium dye contained in the wells (Rutgers et al., 1998). Tetrazolium dye can be reduced to a soluble puple formazan product by live microbial cells. The amount of formazan is m easured spectrophotometically at 590 nm. Quantity of co lor development or formazan product is directly proportional to the number of living and respiring cells (Cory et al., 1991). Subsam ples (10 g dry wt.) of soil were extract ed with 100 mL phosphate buffer, prepared by mixing 61.5 mL of 1 M dipotassium hydrogen phosphate ( K 2 HPO 4 .3H 2 O ) and 38.5 mL of 1 M potassium dihydrogen phosphate ( KH 2 PO 4 ) The s uspension was shaken for 20 min on a reciprocating sha ker, and then centrifuged for 10 min at 8000 g.
116 Supernatants were transferred into micro centrifuge tubes immediately frozen under liquid nitroge n, and stored in the freezer until analysis. A n ine fold dilution (estimated in a preliminary study as being sufficient to contain appropriate number of bacteria ) was performed on the extracts using the phosphate bu ffer, and 100 L of each diluted extract was transferred to biolog plate wells. The plates were covered with alumi num f oil and stored in the dark at 20 to 25 C Temporal analysis of plates included measuring the absorbanc e of the whole plate at the wavelength of 590 nm every day for 10 d and ever y other day for the next 20 d on the Bio TekFL600 micro plate reader (MTX labs, VA). The color development ( absorba n ce ) in the plate wells represented the extent of substr ate utilization by the microbe s, representing growth. The absorbance was separately meas ured for each substrate, but the substrates were later grouped into categories (Table 6 1): acids, amino acids, polymers, amines, carbohydrates, and miscellaneous based on the classificat ion by Gar land and Mills (1991). The substrate absorbance was calculated by subtracting the absorbance of water controls measured at corresponding times. Extraction for Bacterial Count The total bacterial count was estimated by direct counting using a fluorescein is othiocyanate ( FITC ) stain (Matsunaga et al., 1995) Five gram s of soil (d ry wt. ) was shak en vigorously with 45 mL of 0.1 % agar solution. After 30s, a 10 L of aliqu ot was spread evenly onto a pre marked 1 cm 2 area on a s lide. The slide was kept on a warmer to heat fix the bacter ia. One drop of FITC stain was added on the slide and left to dry for 3 min E xcess stain was washed off with 0.5 M sodium carbonate bicarbonate ( Na 2 CO 3 NaHCO 3 ) buffer, and the washed slide mounted with glycerol (pH = 9.6) befo re putting the co ver slide Bacteria were counted within four microscopic fields at 400
117 using an Optiphot Biological UV Micr oscope (Nikon, NY), and the mean counts recorded. The mean count s represent only viable bacteria, as the technique fails to stain the dead bacteria. Statistical Analysis Statistical analysis was performed using SAS software, version 9.1 (SAS institute, 2002). The T differences at each sampling period across treatments. Regre ssion analysis with the ESTIMATE procedure assessed the statistical differences between sampling periods within treatments. Results and Discussion Microbial Toxicity (Range Finding) Test Respiration measures the overall activi ty of a microbial community. The substrate (biosolids) induced respiratio n was quantifie d following OPPTS guideline using CO 2 evolution as a me asure of microbial respiration. Biosolids addition i ncrease d microbial respiration (CO 2 evolution) over time, but some respiration inhibition occurred due to TCS spiking ( Figure 6 1). At 28 d, biosolids TCS spiking caused a reduction of ~27% ( at TCS concentration= 1005 mg kg 1 ), and ~42% ( at TCS concentration= 10,005 mg kg 1 ) in total CO 2 evolution in the IFS treatmen ts (Figure 6 1a). For ASL treatments, biosolids TCS ( concentration= 10,005 mg kg 1 ) caused ~10% reduction in total CO 2 evolution relative to the unspiked biosolids amend ed control (Figure 6 1b). Further, b iosolids addition increased the NH 4 N production in both soil s. U p to day 5, T CS spiking did not inhibit the NH 4 N production (Figure 6 2 ), and a t th e study termination (28d), TCS inhibition of NH 4 N only occurred at a TCS concentration of 10,005 mg kg 1 biosolids in the IFS soil (Figure 6 2a). B iosolids or TCS addition did not affect the production of N0 3
118 N0 2 N in the IFS soil (Figure 6 3a), but the ASL soil behaved d ifferently. Biosolids TCS concentrations 15 mg kg 1 reduced the N0 3 N0 2 N production by 75% a s compared to unspiked biosolids amended control ( Figure 6 3b). Microbial Toxicity ( Definitive ) Test Effect on respiration rates B iosolids TCS concentration s 1005 mg kg 1 in the IFS soil, and 505 mg kg 1 in ASL soil, did not affect cumulative respiration at any sampling time. Addition of biosolids significantly increased the total CO 2 production in both soils (Figure 6 4a, b ), except for one treatment (15 mg TCS kg 1 bio solids) in the IFS soil (Figure 6 4b ). The cumulative CO 2 evolution in the 15 mg TCS kg 1 treatment was signi ficantly smaller than the other TCS treatments, and likely represented a n unexplained analytical error. The o nly e ffect on microbial respiration was due to biosolids additio n likely attributable to increased substrate induced respiration following the addition of labile organic carbon. Our results agree with the results from previous studies (Barbarick et al., 2004; Sullivan et al., 2006a). Barbarick et al. (2004) found increased microb ial respiration rates in shrubland and grassland soils amended with biosolids (30 40 Mg ha 1 ) 6 years prior T he increase d respiration was attributed to the availability of additional carbon substrate s furnished by the biosolids (Barbarick et al., 2004) Sullivan et al. ( 2006 a) p ositive ly correlated biosolids addition and C O 2 evolution i n semi arid rangeland soil (sandy loam, pH=6.2). The C O 2 production was enhanced, and the metabolism of the microbial c ommunity elevated, even 1 2 years following biosolids application O ur CO 2 evolution data su ggest that TCS concentration s 505 mg kg 1 biosolids do not adversely affect respiration rates, similar to studies assessing the effects of TCS spiked in to soils ( Butler et al., 2011; Waller and Kookana, 2009 ). Waller and Kookana
119 (2009) found no negative effect on substrate induced respiration at spiked TCS concentration s <50 mg kg 1 soil ( equivalent to our 5000 mg TCS kg 1 biosolids treatment). Butler et al. (2011) assessed basal and substrate induced respiration changes following TCS spiking in un amended soils. Three soils sandy loam [organic carbon (OC) 17 g kg 1 ], clay (OC 27 g kg 1 ) and loamy sand (OC 23 g kg 1 ) were spiked with a range of TCS concentr ations (0 1000 mg kg 1 ) Results suggest respiration inhibition at TCS concentration >10 mg kg 1 soil ( equivalent to our >1000 mg kg 1 biosolids treatment ) within 6 h from the initial TCS spiking, but the inhibition disappeared after 2 to 4 d. Res piking of TCS to the same soil, caused an initial microbial stimulation of respiration up to day 4, but the res piration rates returned to the baseline by day 6. Authors reported that TC S inhibited initial respiration rates (6h) in the unacclimated soils, but following the respiking, TCS acted as a C source and stimulated respiration at least at the low TCS concentr ations (<100 mg kg 1 ) (Butler et al., 2011) Thus, the stimul ation of microbial respiration observed in our study could be a combin ed effect of TCS and biosolids serving as C sources, especially at TCS concentration 10 mg kg 1 amended soil. S piked TCS concentrations d id not significantly affect total CO 2 evolved, and the results were consistent with our range finding test, and previous two studies conducted in spiked soils (no biosolids) (Waller and Kook ana, 2009; Butler et al., 2011). Our study suggest ed that b iosolids addition, however, stimulated microbial respiration. Effect on nitrogen c ycle Ashkum silty clay loam (ASL) soil A T CS concentration 505 mg kg 1 biosolids did not significan tly affect the NH 4 N production in biosolids amended soil or the control at 5d (Figure 6 5a ). At day 28, NH 4 N production surprisingly declined in all treatments
120 independent of bioso lids or TCS additions so there was no significant effect of biosolids or TCS concentration on NH 4 N produc tion (Figure 6 5a). The abrupt decrease in NH 4 N production from day 5 to 28 cannot be easily explained because of a gap between the sampling times. Consistent with our st udy, a 28 d incubation study ( Snyder et al., 2011 ) revealed insignificant differences in NH 4 N production at triclocarban concentrations up to ~717 mg kg 1 biosolids following biosolids addition to a silty loam soil. Biosolids addition significantly decreased the N0 3 N0 2 N production at day 5 (Fig ure 6 5b ). However, N0 3 N0 2 N production in spiked biosolids treatment s was not significantly different than the un spiked biosolids amended control (Inherent TCS=5 mg kg 1 ). At day 28, N0 3 N0 2 N concentrations in creased (as compared to day 5), resulting in no overall TCS or biosolid s addition effect on N0 3 N0 2 N production at biosolids TCS concentration of 505 mg kg 1 Given reports (Zhao, 2006; Stasinakis et al., 2007) that TCS inhibits the nitrification process in sludge and soils, we compared the relationship between the ammon ification and nitrification processes in our study. A ccumulation of NH 4 N at day 5 coincided with a decrease in N0 3 N0 2 N production that might suggest TCS inhibition of nitr ification. However, between days 5 and 28, unknown changes in the study conditions appeared to disturb the microbial community structure (or conditions), which overcame any TCS inhibition of nitrification and caused a rapid conversion of NH 4 N to N0 3 N0 2 N Immokalee fine sand (IFS ) soil A similar study conducted with the IFS soil included an additional sampling period of 14d to avoid the unexplainable changes observed in ASL soil data interpretation due to long sampling intervals.
121 Ammonium production increased following biosolids addition (Figure 6 6a ). At day 5, a TCS biosolids concentration of 1005 mg kg 1 caused the greatest NH 4 N production, but TCS additio n effect was no t significant All the bioso lids amended treatments increased NH 4 N production at day 5, except the treatment with 505 mg TCS kg 1 biosolids where the effect did not appear until day 14 (Figure 6 6a). At day 5, the biosolids or TCS addition s did not affect N0 3 N0 2 N production. A t day 28, biosolids addition caused a significant reduction in N0 3 N0 2 N concentrations ( Figure 6 6a, b). Holt et al. (2010) investigated the production of NH 4 N from a soil (unknown texture) amended with biosolids (18 Mg ha 1 ) The results suggested that diazotrophs fixed more N followi ng biosolids addition up to 42 d but the stimulation disappeared at 84 d causing no difference in NH 4 N production in control and amended soils. Authors concluded that biosolids amendment added organic N causing rapid NH 4 N production, and that inh erent biosolids borne chemicals ( e.g., PPCPs unknown concentrations ) did not adverse ly affect N cycling Similarly Barbarick et al. (2006) attributed elevated N mineralization rates in soil to addition of organic N through biosolids application. Waller and Kookana (2009) monitored changes in N min eralization in sand (OC: 8.5 g kg 1 ) and clay (OC: 18.5 g kg 1 ) soil s spiked with 0, 1, 5, 10, 50, and 100 mg TCS kg 1 soil. Nitrification process was not adversely affected at TCS concentration s ranging from 5 to 50 mg kg 1 soil (corresponds to equivalent concentration s of 500 to 5000 mg kg 1 biosolids). A s with CO 2 evolution and NH 4 N production the N0 3 N0 2 N production data at study termination (28 d) suggest signif icant biosolids addition effect s but no TCS treatment effect, up to 1005 mg TCS kg 1 biosolids in the IFS soil.
122 Bacterial DNA a nalysis Zhao (2006) reported that TCS inhi bited ammonia monooxygenase (AMO ), the first enzyme in the two step nitrification process. The AMO is encoded by two genes from Arch ea l am oA and bacterial amoA. Archeal a moA dominates in most environments (Leininger et al., 2006). In a preliminary study, we measured the relative concentrations of the two forms of bacterial DNA from IFS control (soil only), 5 (only biosolids inherent TCS), and 505 mg TCS kg 1 biosol ids (spi ked TCS) treatments. There were no significant differences in archeal amoA concentrations among the treatments but high variability made the final results inconclusive The preliminary data suggest minimal effects of biosolids borne TCS addit ion on th e archeal amoA that is likely to be involved in N cycling at least in one soil. T he inconclusive results with bacterial amoA due to TCS spiking requires further investigation perhaps in various soils. Collectively, the soil microbial toxicity test data su ggest no TCS spiking effects on respiration, nitrification or ammonification up to a TCS biosolids concentration of 505 mg kg 1 amended to the ASL soil, and up to 1005 mg kg 1 amended to the IFS soil. The differences in NH 4 N and NO 3 NO 2 N production bet ween the two soils may reflect differences in the availability of TCS, or other C contain ing substrates in the two soils ( Table 5 1 ). Triclosan availability could differ in the two soils due to less TCS retention in the IFS soil (log K d = 1.8 7 0.21), than in the ASL soil (log K d = 2.64 0.19) (Agyin Birikorang et al., 2010) Waller and Kookana (2009) monitored changes in N mineralization in two soils and suggested that the TCS concentration required to affect nitrification was ten times greater in the clay soil (50 mg TCS kg 1 ) than in the sand (5 mg TCS kg 1 ).
123 Thus, TCS concentrations of 25 to 50 fold greater (500 mg kg 1 ) than normally found (~10 to 20 mg kg 1 ) in most land applied bi osolids did not cause microbial toxicity, at least using microbial respiration and nitrogen cycling as indicator s The effects of long term bi osolids applications are known. O ur study and other published st udies (Waller and Kookana, 2009 ; Butler et al., 2011) are short term and may not represent the effect s of long term biosolids borne TCS application. Microbial Community Structure Analysis The average well color development (AWCD ) in various substrates is illustrated in F igure 6 7 D ata obtained for each substrate were combined into the general category of carboxylic acids, amino acids, polymers, amines, carbohydrates, and miscellaneous. At study termination (day 40), the well color deve lopment varied sug gesting different utilization of the six types of carbon subst rates in different soils (Figure 6 7 ). Polymers were uti lized most readily, carbohydrates w ere utilized least readily, and the remaining substrate groups were moderately utilized. A ddition of biosolids or TCS did not appear to increase substrate utilization as compared to the control, except i n IFS soil where the addition of biosolids alone (IFS 0) inhibit ed the utilizat ion of most groups of substrates, except the polymers. T he addition of biosolids spiked with TCS in the IFS soil (IFS 500) resulted in s imilar color development as in the control (IFS control) H igh (landscaping ) and multiple (WL and KK) loads of biosolids and associated TCS did not increase substrate utilization as compared to the respective control s Thus, neither inhibition n or stimulation of carbon substrate utilization occurred du e to TCS or biosolids addition, suggest ing minimal disturbance of the microbial community structure at least with AWCD as an indicator. Results are at odds with the assumption that
124 microbial community structures would vary with soil texture or biosolids and TCS addition. Young (2011 unpublished, personal communication ) suggest ed adverse effect s of TCS spiked in biosolids on the overall community composition u sing biolog plates and PCA (p r inciple component analysis). However, a direct comparison and evaluation of the two studies was not possible due different quantification methods Additional investigation of our data should involve the quantification of the overall well color development for all substrates in each treatment and plot ting the various microbial communities using a PCA method. Various studies (Barbarick et al., 2004; Garcia Gil et al., 2004; Sullivan et al., 2006b) suggested shifts in soil microbial co mmunity structure due to the addition of biosolids alone. Thus, the changes in community structure observed in Young (2011) study may simply be a biosolids effect. In fact, he observed a greater effect of biosolids as compared to TCS concentration effects Bacterial Counts T he average bacteria l counts were greater in the ASL, landscaping and WL soil cont rol than in the IFS and KK control soils but the differences were not significant. The trends may be a texture effect, because ASL and landscaping soils (soil collected from an area utilizing for landscaping purposes in Illinois) are silty clay loam s, whereas, the IFS and KK are sand s A ddition of biosolids increased the bacterial counts in all soils as compared to the control soils (Table 6 2) Biosolids addition likely provide d organic carbon for enhancing the bacterial growth T he ASL soil treated with TCS spiked biosolids ( ASL 500 ) had significantly greater bacterial counts than t he corresponding IFS treatments (IFS 500) (p<0.05).
125 B iosolids application r ates were same in the two soils, but greater bacterial number enhancemen t occured in the ASL soil spiked with TCS concentration (ASL 500) which is likely a TCS addition effect in the ASL soil T he landscaping biosolids amended treatment (landscaping TRT) had significantly greater (p<0.0001) bacterial cou nts than the landscaping control soil or the biosolids amended ASL, IFS, WL and KK soils which may have occurred following the application of high loads (228 Mg ha 1 ) of biosolids Addition of TCS signifi cantly increase d bacterial count s in the ASL 500 treatment as well as in the WL amended treatment (WL TRT) compared to the soil control (ASL control and WL control) and the biosoli ds amended unspiked treatments (ASL 0) There is no evidence of bacterial growth inhibition following high (landscaping TRT) or multiple (WL TRT, KK TRT) biosolids applications or high TCS concentrations ( 990 u g k g 1 ) in amended soil T he bacterial counts were greater in biosolids amended silty clay loam soil (ASL) than in amend ed sand (IFS) suggesting a texture effect. Zer z ghi et al. (2010a) evaluated the effect of 20 years of biosolids land application (8 24 Mg ha 1 year 1 ) on the microbial population and activity in soil ( s outhwestern desert soil ) D irect cou nt s suggested difference s in total microbial counts between the control and a mended soils, but the difference was not significant. The a uthors concluded that even long term (20 years ) biosolids application s resulting in high total biosolids loads ( 160 480 Mg ha 1 ) did not affect microbial numbers of bacteria, actinomycetes and fungi. W e accept our hypothesi s that biosolids borne TCS has no adverse effect on soil micro o rganisms, using microbially mediated processe s, community structure, and bacterial count s as indicato r s
126 Table 6 1. The grouping of the various substrates in the biolog ECO plate s (Garland and Mills, 1991). Acids Galactonic acid Galacturonic acid 2 hydroxy benzoic acid 4 hydroxy benzoic acid Itaconic acid Ketobutryic acid Malic acid Pyruvic acid Glucosaminic acid Amino acids Arginine Asparagine Phenylalanine Serine Threonine Glutamic acid Polymers Tween 40 Tween 80 Glycogen Amines Pheny ethyl amine Glucosamine Putresoine Carbohydrates Cellobiose Lactose Methyl glucosidase D xylose Miscellaneous Erthristol Mannitol Cyclodextrin Glucose 1 phosphate Glycerol phosphate Table 6 2. A verage bacterial count ( number ) in soils with varying biosolids application rates TCS concentration s and textures (same letters represent no significant difference among treatments) Treatment Biosolids rates (Mg ha 1 ) TCS concentration ( g kg 1 ) Texture Average standard deviation of count of bacteria per g of soil Landscaping control na nd Clay loam 2.5*10 8 1.1*10 8 b Landscaping TRT 228 (one time) 990 Silty clay loam 1.0*10 9 2.0*10 8 a ASL control na nd Silty clay loam 3.4*10 8 2.3 *10 8 b ASL 0 22 5 Silty clay loam 4.0*10 8 2.3 *10 8 b ASL 5 00 22 500 Silty clay loam 5.3*10 8 2.0 *10 8 c IFS control na nd Sand 1.7*10 8 6.8*10 7 b IFS 0 22 5 Sand 2.7*10 8 1.6 *10 8 b IFS 50 0 22 500 Sand 3.8 *10 8 1.4*10 8 b WL control na 1 Silty clay loam 2.3*10 8 1.5*10 8 b WL TRT 116 (multiple) 41 Silty clay loam 5.2*10 8 1.6*10 8 c KK control na 0.8 Sand 2 .1*10 8 7.3*10 7 b KK TRT 154 (multiple) 18 Sand 2.8*10 8 1.4*10 8 b nd na
127 Figure 6 1 Total CO 2 evolution (mg) over various time s in the (a) IFS and (b) ASL soil s amended with biosolids spiked with a range of TCS concentrations unreplicated range finding test 0 50 100 150 200 250 300 CO 2 Evolved (mg) Biosolids TCS concentration (mg kg 1 ) Days 0-5 Days 5-28 Days 0-28 0 60 120 180 240 300 360 420 CO 2 Evolved (mg) Biosolids TCS concentration (mg kg 1 ) a. IFS soil b. ASL soil
128 Fig ure 6 2 NH 4 + NH 3 concentrations (mg kg 1 ) over time (Days 0 28) in the (a) IFS and (b) ASL soils amended with biosolids spiked with a range of TCS concentrations unreplicated range finding test 0 20 40 60 80 100 120 140 160 NH 4 + NH 3 in biosolids amended soils (mg kg 1 ) Biosolids TCS concentration (mg kg 1 ) Day 0 Day 5 Day 28 0 5 10 15 20 25 30 NH 4 + NH 3 in biosolids amended soils (mg kg 1 ) Biosolids TCS concentration (mg kg 1 ) Day 0 Day 5 Day 28 a. IFS soil b. ASL soil
129 Fig ure 6 3 N0 3 N0 2 N concentrations (mg kg 1 ) over time (Days 0 28) in the (a) IFS and (b) ASL soils amended with bioso lids spiked with a range of TCS concentrations unreplicated range finding test 0 1 2 3 4 5 6 7 8 NO 3 NO 2 in biosolids amended soils (mg kg 1 ) Biosolids TCS concentration (mg kg 1 ) Day 0 Day 5 Day 28 0 10 20 30 40 50 NO 3 NO 2 in biosolids amended soils (mg kg 1 ) Biosolids TCS concentration (mg kg 1 ) Day 0 Day 5 Day 28 a. IFS soil b. ASL soil
130 Figure 6 4. Mean t otal CO 2 (mg) as a function of TCS concentrations and time (Days 0 28) in (a) ASL and (b) IFS soils (like letters indicate no significant difference between treatments) definitive test.
131 Figure 6 5 Mean (a) NH 4 N and (b) N0 3 N0 2 N concentrations (n=3) as a function of biosolids TCS concentration and time (Days 0 28) in ASL soils (like letters i ndicate no significant difference between treatments) definitive test.
132 Figure 6 6. Mean (a) NH 4 N and (b) N0 3 N0 2 N concentrations (n=3) as a function of biosolids TCS concentration and time (Days 0 28) in IFS soils (like letters indicate no significant difference between treatments) definitive test.
133 Figure 6 7 A verage (n=3) well color development (AW CD) for the substrate types in the va rious soil samples developed 40 days after the incubation of the biolog plates.
134 CHAPTER 7 PLANT TOXICITY AND BIOACCUMULATION OF BIOSOLIDS BORNE TCS Background Food chain bioaccumulation begins with chemical uptake by plants. Chemicals can enter plant s by partitioning from contami nated soil solution s to the roots, and be translocated through x ylem tissue (t he xylem transports water from roots to leaves by transpiration). C hemicals may enter the vegetation directly by gas phase and particle phase deposition on le aves, enter the stomata and be translocated through phloem tissue (t he phloem transp orts photosynthesis pr oducts from leaves to plant parts) (Simonich and Hites, 1995). Chemicals may also directly partition from soil particles to the root epidermis or the cortex and accumulate in the root (Leewen and Vermeire, 2007). The pathway of chemic al up take by plants depends on (a) such as water solubility, lipophilicity, an d vapor pressure; (b) environmental conditions like temperature, organic carbon (OC) content of soil and (c) plant species characteristics like surface are a of the leaf and root mass (Simonich and Hites, 1995). Root u ptake from soil solution is a dominant pathway for hydrophilic compounds characterized by high water solubility low vapor pressure and low log K ow values (< 4). Such compounds move from the outer to the inner root, and are translocated via the xylem to various plant parts (Paterson et al., 1994). However, the translocation to plant parts is limited by the TSCF (transpiration stream concentration factor). The TSCF is a ow and is the ratio of chemical concentration in the xylem sap to the chemical concentration in the solution surrounding the root. L ipophilic organic compounds (log K ow > 4) suc h as organochlorine pesticides and polycyclic aromatic hydrocarbons do not e nt er the root xylem, but preferenti ally partition to the root
135 epide rmis or the soil solids (Paterso n et al., 1994; Schroll et al., 1 994) and accumulate in the root. Peterse n et al. (2003) proposed an alternate mechanism of plant contamination through rainfall induced splashing of soils when biosolids are applied. However, the authors later concluded that the added biosolids did not increase the risk of rain induced sp lashing and the propose d mechanism is insignificant (Peterse n et al., 2003). Thus pla nt uptake of a chemical depends on chemical properties (solubility, K ow ), plant characterstics (lipid content, leaf orientation, TSCF) and soil properties (organic carbon). Hulster et al. (1994) found that three cucurbitaceae species ( z ucchini, pumpkin an d cucumber) grown in soils accumulate d and translocate d high concentrations of highly lipophilic PCDD/Fs ( Polychlorinated Dibenzo p dioxins and d ibenzofurans ). U ptake by cucumber occurred th r ough particle phase deposition. The exact mechanism of fruit (zuc chini and pumpkin) accumulation was not elucidated but Hulster et al. (1994) opined that uptake occurred through roots due to increase d chemical availability in the presence of uniq ue species specific root exudates. Several studies (Farkas et al., 2009; Kumar et al., 2005; Boxall et al ., 2006; Dolliver et al., 2007) reported plant accumulation and phytotoxic effects of antibiotics in un amended and manure amended soils. T riclosan (TCS) is a n antimicrobial, and a common component of waste water At the was tewater treatment, TCS partitions mainly in to biosolids. Biosolid s borne TCS can be land appli ed, and plants exposed to TCS may exhibit phytotoxicities or accumulate TCS in a manner similar to antibiotics In a recent presentation, Kumar ( 2010 ) presented a rule of 3 phar maceuticals and personal care products for plant uptake studies. According to the rule, the chemicals that have a molecular
136 weight less than 450, log K ow less than 3, number of H bond donors less than three, and number of H bond acceptors of less than six should be studied in field uptake studies for conducting risk assessment. Triclosa n obeyed all the rules except that the log K ow of TCS is greater than t hree. Herklotz et al. (2010) studied TCS toxicity in hydroponic system and suggested adverse effects of TCS on the germination of cabbage ( Brassica rapa var. pekinensis ) and Wisconsin fast plants ( Brassica rapa ) with a lethal concentration (LC) of 0.44 mg TCS L 1 water. Stevens et al. (2009) evaluated TCS effect s on the germination and bioaccumulation in three wetland plants ( Sesbania herbacea, Eclipta prostrata and Bidens frondosa ) A TCS concentration of 100 mg L 1 water reduced the r ate of s eed germination in two species ( S. herbacea and B. frondosa ) Triclosan also accumulated in the shoots of S. herbacea with a bioco ncentration factor (BCF) of <10, and in roots of B. frondosa with a bioaccumulation fa ctor (BAF) ranging fro m 53 to 101. A f ew unpublished studies (Hoberg, 1992; Schwab and Heim, 1997 ; cited in Reiss et al., 2009 ), and a p ublished study (Liu et al., 200 9 ) reported TCS plant toxicity in soil system s not amended with biosolids Hoberg (1992) s uggested adverse effects of TCS on c u cumber, with a shoot length No Obs erved E ffect Concentration (NOEC) of 0.096 mg kg 1 soil, and a Lowest Observed Effect Concentration (LOEC) of 0.28 mg kg 1 soil. Schwab and Heim (1997) evaluated TCS effect s on cucumber seedling emergence, shoot length, ro ot and shoot weight and reported a NOEC of 1 mg kg 1 soil. Liu et al. (2008) suggested a shoot height inhibition effect concentration (EC 10 ) of 6 mg TCS kg 1 soil for cucumber, and a LOEC of 10 mg kg 1 soil for rice root length. The studies described above suggested toxicity and bioaccumulation of TCS to plants grown in
137 hydroponics and in un amended soil. Simlar to accumulation in un a mended soil p lants grown in biosolids amended soil may accumulate TCS or experience TCS toxicity. Thus, the consumers eating plants grown in amended soil m ay be unknowingly ingesting TCS. A f ew studies evaluated TCS plant accumulation following land applied biosolids. Xia et al. (2010) found TCS concentration s <6.5 ng g 1 [Limit of quantitation ( LOQ ) of the instrument] in corn ( Zea Mays ) stover collected from long term biosolids amended (various rates) field soil s (calcareous mine spoil; pH 7.8) with TCS soil con centration s ranging from 10 to 50 ng g 1 The results suggest minimal TCS accumulation by corn, a monocotyledon ( monocot ) However, the toxicity and bioaccumulation of TCS might vary with the plant species and characterstics of soil (Duarte Davidson and Jones, 1996; Suter, 2007) Wu et al. (2010) evaluated the TCS uptake by a soybean ( Glycine max ) a dicotyledon (dicot), under green house conditions. The biosolids were spiked with TCS and amended (11 Mg ha 1 ) to soil (sand, pH = 5.1, OC = 16 g kg 1 ) to achieve a final TCS concentration of 7 0 ng g 1 amended soi l Results suggested TCS contamination of root tissue and transl ocation to above ground biomass The p lants, harvested at the full seed stage (110 d) accumulated TCS in the root (76.8 3.1 ng g 1 ), stem (136 66 ng g 1 ), leaf (120 37 ng g 1 ), and beans (12.6 2.3 ng g 1 ). The BCF values were measured at first harvesting (60d) and at full seed stage (110d). The BCF from soils to roots were 3 to 6.5 and 1 to 2 in the roots to leaves (Wu et al., 2010) The Wu et al. (2010) study w as conducted in a spiked system, where TCS bioavailability and accumulation potential may differ from system s with inherent TCS (biosolids borne) Further the study utilized a single crop, and the biosolids had a solid s content ( 19 g L
138 1 ) which is much less than the solid s content of biosolids (~30 0 g kg 1 ) routinely u sed for land appli cation in the U S. Chemical transport potential and bioavailability is like l y influenced by the type of biosolids applied to the soils (Edwards et al., 2009), with a greater chemical availability in biosolids with low solid s co ntent. In addition, a TCS degradation half life of 100 d and the appearance of a metabolite [Methyl TCS (Me TCS)] (Chapter 4) of greater K ow suggest that both TCS and Me TCS could exist in soils for extended time s following biosolids amendment. The metabol ite may also possess phytoaccumulative potential. The present study investigated the TCS toxicity and accumulation in multiple food crops [ lettuce ( Lactuca sativa ), radish ( Raphanus sativus ), cucumber ( Cucumic sativus ) bahia grass ( Paspalum notatum ) ] utilizing cake biosolids ( commonly us ed in the U.S ) and a field equilibrated soil. The soil received a single large application (228 Mg ha 1 ) of biosolids that was incorporated and mixed with 15 to 20 cm of soil in 2008 The amended soil sampled in 2010 contained an inherent TCS concentration of ~ 1 mg kg 1 soil. A chemical similar to TCS [i.e triclocarban, (TCC)] was minimal ly phyto accumulated by bahia grass (Snyder et al., 2011 ) grown in a biosolids amended soil. We speculated that, because the partiti oning coefficient of TCS (log K oc = 4.26 ) is even greater than TCC (log K oc = 3.82 ) (Agyin Birikorang et al., 2010) TCS should partition more extensively to soil OC and be less phytoavailable W e h ypothesize that biosolids borne TCS has minimal phyto toxic ity and phytoaccumulation potential. Our study objective was to evaluate the toxicity and plant uptake of TCS by several crops grown in a soil previo usly am ended with biosolids. Further, the uptake of an expected metabolite
139 was assessed by analyzing the plant s for Me TCS. Plant toxicity and accumulation was investigated in vege table s grown for edible portions in temp erature controlled growth rooms, and bahia grass grown in a greenhouse. In addition, empirical models and equations were us ed for bioaccumulation estimation and comparison with measured values. Material and Methods Soils and Chemicals Triclosan (C AS No. 101 20 2; >99.9% purity) standard was purchased from United States Pharmacopeia (USP) (Maryland, USA). Internal standard ( 13 C 12 TCS) pyridine, BSTFA (bis (trimethylsilyl) trifluoroacetamide) +1% TMCS (trimethylch loro silane) were obtained from Sigma A ldrich (St. Louis, MO). Methyl TCS (Me TCS) standard was purchased from Wellington laboratories ( Shawnee Mission, KS). Potassium chloride (KCl), methanol (MeOH), acetone of HPLC grade or greater were purchased from Sigma Aldrich (St. Louis, MO) JT Baker (Phillipsburg, NJ) or Fisher Scienti fic (Atlanta, GA) The biosolids amended soil was collected in 2010 from a land scaping site in Illinois last amended with biosolids ( ~228 Mg ha 1 ) in 2008. Control soil samples were collected from an adjoining site (18 m away same soil texture ) with no known history of receiving land applied biosolids or sludge. The collected soil s amples were air dried for 1 week and sent to our laboratory in ~18 L buckets. Select physico chemical properties of the control and the biosolids amen ed soil s are presented in T able 7 1 Water soluble fertilizer was supplied as M iracle G row [ 24 8 16 ; nitro gen ( N ) phosphorus ( P ) potassium ( K ) ] obtained from a local store. The vegetable seeds were purchased from Burpee seeds (Warminster, PA). Oasis HLB extraction cartridges
140 (3 cm 3 250 mg) for solid phase extraction (SPE) were purchased from Waters (Mississauga, ON, Canada). Toxicty and Bioaccumulation Study Design The study included 4 crops 3 treatments 4 replicates and a control (four replicates). B iosolids amended and control soils were air dried and sieved t hrough a 2 mm sieve. Organic carbon was determined using the standard Walkley Black method (W alkley and Black, 1934), assuming t hat 77% of the total OC was oxidized (Nelson and Somers, 1996). Soil ammonium concentration was determined by extracting with 2M KCl (U SEPA, 1993b, Method 350.1). A pproximat ely 2 kg (air dried) of control and biosolids a mended soil s were weighed in 15 cm (diameter) pots (bulk density of 1.6 g cm 3 ) lined with a cloth screen to hold the soil. After weighing, the soil s were thinly spread in plastic trays to facilitate uniform application of treatments The treatments included amended soil with inherent TCS (990 ng g 1 ), and the same soil spiked with 500 0 and 10 ,0 00 ng TCS g 1 soil Treatment (Trt) spikes were supplemented to the inherent TCS con centration in the amended soil; thus, t he final nominal concentrations were 99 0 ( Trt 1 ) 5 99 0 ( Trt 2 ), and 10, 9 00 n g TCS g 1 soil ( Trt 3 ) The TCS concentration s utilized are much greater than normally expected in soils amended with typical biosolids applied at agronomic rates, but were useful to obtain a dose response relationship and to maximize potential plant accumulation of TCS. T he Trt 1 was unique as TCS occurred inherent to the biosolids and is e xpected to represent a real world (field equilibrated) scenario. Methanol was utilized as a carrier solvent for spiking in the other treatments and was allowed to evaporate in a hood as the soils equilibrated with the spike s for 48 h. After spik ing and eq uilibration, the soil was packed into pots and brought to pot holding capacity. In a preliminary study, the pot holding capacity was
141 determined by saturating a pot with excess water, and allowing fre e drainage for 24 h. W ater retained by the soil column af ter free drainage stopped was considered the pot holding capacity. Pot holding capacity (water content = 400 g kg 1 soil) is greater than the field capacity (water content = 300 g kg 1 ) due to the textural discont in uity between the soil, the cloth liner, and drainage holes present in the pots. The crops consisted of gourmet blend lettuce ( Lactuca sativa ), crimson giant radish ( Raphanus sativus ), b ahia grass ( Paspalum notatum ), and baby cucumber ( Cucumic sativus ). Crop sp ecies represented monocot s (bahia grass ), dicot s ( lettuce, radish cucumber ), above ground ( lettuce, cucumber, radish leaves and bahia grass ), and below ground ( radish) biomasses All the crops were planted from seeds. For bahia grass, the recommended seed ing rate of 10.9 Mg ha 1 was increased to 22 Mg ha 1 (equivalent to 2 g seeds per pot) to ensure sufficient and rapid soil surface coverage. Lettuce and radish were planted using an excess se eding rate of ~30 seeds per pot, and cucumber was pl anted using four seeds per pot. S eeds were covered with wet paper towels and misted every 3 to 4 h until germination. Growth conditions for vegetab les i ncluded controlled chambers with a 24 h tempera ture maintained between 20 to 25C, and a provision of 40 watt flores cent bulbs for 15 h photoperiod, corresponding to a light intensity of 700 0 to 10 00 0 l ux B ahia grass was grown in a greenhouse, maintained at a suggested temperature range of 26 to 35C (Newman et al., 2010) N anopure water was utilized for irrigation when the water content in the pots was 8 0% of the pot holding capacity or when the soil surface appeared dry. On an average, plants received water once daily. Saucers placed beneath the pot s collected
142 any leachate and the collected water was immediately poured back on to the soils to avoid TCS leaching loss es The f irst vegetable thinning occurred at a plant height of at lea st 2.5 cm, which was 5 d for radish, and 10 d for lettuce and cucumber. No thinning was required for bahia grass. During the first t hinning, excess or weak plants we re removed. The s econd thinning ( 15 d ) reduced the number of plants to 6 to 10 in each pot for radish and lettuce and to two plants for cucumber. A suggested solution for growing pot vegetables consisted of 30g of 20 20 20 ( N P K) analysis water soluble fertilizer mixed with 20 L of water (Stephens, 2009). A commercially available so luble fertilizer Miracle grow (N P K: 24 8 1 6) was deemed sufficiently close to the recommended dosage, and was prepared by mixin g 4 g of solid fertilizer with 4 L of wa ter. Ten days after the germination, control pots were pro vided with liquid fertilizer at week ly interval s until harvesting Biosolids amended soil was deemed to contain sufficient nutrients to permit healthy plant g rowth as illustrated by the physico chemical properties of the amended soil (Table 7 1 ), and were not fertilized. Bioaccumulation Field S tudy We utilized paired soil and plant samples from a site in Fulton County, IL amended at hi gh biosol ids application rates for 16 years r esulting in total loads of 1084 to 1180 Mg biosolids ha 1 The last biosolids application occurred in 2000, and the plant samples were collected in 2001, 2002, 2004 and 2006. Plant samples included three field replicates for each crop and for each year. The soil and plant samples were collected, air dried and stored in a cool and dry place at room temperature until they were sent to us in 2010. The available samples included a monocot (corn leaves) and a dicot
143 ( soybean grains ) The fiel d samples addressed the possibility of long term TCS and Me TCS plant bioa ccumulation in a monoct and a dicot. Plant Harvesting and Sample Preparation Harvesting of radish and lettuce plants occurred 40 d after germination. B ahia grass was harvested 6 0 d after the initial soil preparation, as the grass required reseeding due to poor initial crop cover. H arvested plants were washed with nano pure water fresh weigh t s determined and biomass (leaves and roots) was separated dried separately at 50C to constant weight, and the dried tissue ball milled S oils were collected from each treatment and replicate immediately before and after the study and dried to constant weight at 50C. A sample extraction t echnique for TCS was modeled after that described b y Snyder et al. (2011 ) w ith some modifications. The dried soil and plant tissue s (0.5 1 g) were loaded int o 25 mL glass cen trifuge tubes and extracted with 10 mL of MeOH+acetone (50:50, v/v). The extraction was performed on a platform shaker for 18 h, foll owed by 60 min of soni cation in a water bath (Branson 2210, Danbury, CT ; temp. 40 C, 60 sonication s min 1 ) Suspensions were centrifuged at 800 x g, and the supernatant transferred to 20 mL glass scintillation vials. The extraction procedure was performed twice and the extracts were combined, dried (under N 2 ), reconstituted in MeOH and transferred to microcentrifuge tubes. The microcentrifuge tubes were centrifuged at 18,000 g for 30 min, and supernatant transferred to GC vials, and again dried ( under N 2 ) Dried e xtracts were then r econstituted in MeOH followed by cleanup using Oa sis HLB SPE cartridges (Chu and Metcalfe, 2007). After the cleanup, the extract was transferred to amber GC vials, an internal standard 13 C 12 TCS (50 ng mL 1 ) was added and the extracts dried.
144 D erivatization was performed according to Shareef et al. (2006) with slight modification. Briefly, the dried extracts were reconstituted in a mixture of 4:1 of derivatization agent ( BSTFA +1% TMCS ) and a solvent (pyridine), vortexed for 10 s, and heated in a dry bath for 1h. The samples were then transferred to fresh GC vials with glass inserts and T eflon lined caps. A s et of plant samples from the control soil was spiked with TCS and Me TCS, and subjecte d to the same extraction procedure. The a verage percent recoveries of the spiked TCS a nd Me TCS in the plants were >90 %. Instrument Analysis and Quantitation The samples obtained after the derivatizati on step were analyzed by sp litless injection (5 L) on a Varian 4000 Gas Chromatograph equipped with a Restek Rxi 5Sil column coupled with a Varian 4000 MS/MS. The GC/MS conditions were modeled after Balmer et al. (2004) with some modifications. The GC column temperature was initially held at 100C for 1 min, and then increased to 310C at a rate of 10C min 1 with no final hold time. The carrier gas was helium, t he ion trap, manifold and transfer line temperatures were 200, 80 and 270C, respectively, and the ionization source was internal/electron ionization. Data acquisition monitored two fragment ions for each compound. The ion masses were m/z of 345/347 for TCS trimethy lsilylether, 302/304 for Me TCS. The internal standard 13 C 12 TCS trimethylsilyl ether mass was monitored at 357/359. The sa mples ran for 22 min with an average retention time of 14.8 min for all compounds. For quantification, an 8 point internal calibration curve was generated in t he TCS concentration range of 1 to 1000 ng g 1 average R 2 >0.999 The detection limits were determined at a sign al to noise (S/N) ratio of > 10. The limit of detection ( LOD ) was 0.28 ng g 1 and limit of quantitation ( LOQ ) was 1 ng g 1 for both TCS and Me TCS in the plant tissues. The LOD and LOQ values were calculated as 3 fold and
145 10 fold, respectively, the standard deviation in the signal from multiple runs of the lowest calibration standard (S ignal/Noise >10) (USEPA, 1984). The details of detection limits and recoveries are provided in the A ppendix C Results and Discussion Plant Biomass Yields Assessment of a dverse e ffect s of bi osolids borne TCS on plant growth was by visual examination and by measuring plant biomass yields. Figure s 7 1 and 7 2 are represent ative p ictures of lettuce, radish and bahia grass, and show no obvious difference s in the health of plan ts exposed to various soil TCS concentrations. H eight and vigor appeared to be superior in all the plants grown in treated soils ( biosolids amended) than in the control soil irrespective of the TCS concent ration ; i.e. TCS spiking did not adversely affect the appearance of any plant species. Our observation was consistent with reports of better growing grass ( Poa pratensis ) in fields of the biosolids amended soil than the control soils utilized in our study D ifference s in field growth were not quantified, but were visually obvious ( Kuldip Kumar, personal communication 2011 ). Biosolids addition significantly increase d the let tuce yields (fresh and dry wt.) in Trt 1 relative to the control. Lettuce growth in Trt 2 and Trt 3, however, was inhibited and plant fresh weights were significantly smaller than in Trt 1 (Table 7 2) The observed difference appeared at a concentration of 5990 ng TCS g 1 amended soil. The lettuce dry weights were not adversely aff ected at any concentrat ion used F resh weights of bahia grass were significantl y greater in all of the TCS treatments than in the controls, suggesting enhancement of plant growth by biosolids or TCS addition (Table 7 2) F resh weight s of r adish root in control, Trt 1 and Trt 2 were
146 not significantly different, but the weight in Trt 3 was significantly smaller than all other treatments (Table 7 2) The data suggest some inhibition of radish root growth at an amended soil concentration of 9150 ng TCS g 1 Thus, using fresh biomass yields as the criteria, an amended soil concentration of 9150 ng TCS g 1 (Trt 3) can be regarded as the LOEC for radish root growth For lettuce, Trt 1 (4570 ng g 1 ) is regarded as the LOEC and for bahia grass the NOEC is at least the highest TCS concentration (9 1 50 ng g 1 ) utilized in the study. In contrast, dry biomass yield data suggest essentially no inhibition effect of spiked TCS. Only the radish es grown in Trt 2 had significantly smaller yield s, but the yields were similar to the control, su ggesting no overall growth inhibition (Table 7 2). Greater yield as a result of biosolids treatments was the only significant effect. Thus, based on dry biomass yields, NOEC for all the plants is at least 9150 ng g 1 (Trt 3). Most TCS toxicity studies conducted in soils are either un published or limited to TCS effects on germination rates [Stevens et al. ( 2009 ); Hoberg (1992 ) Schwab and Heim (1997) ]. Only Liu et al. (200 9 ) quantified TCS toxicity by growing cucumber and rice plants exposed to a range o f TCS concentrations (1 300 mg kg 1 ) The a uthors suggested a shoot height inhibit ion effect concentration of 6 mg TCS kg 1 soil for cucumber and a LOEC of 10 mg TCS kg 1 soil for rice root length. Comparison s of our study results with those of Liu et al. (200 9 ) are confounded because the latter study involved no biosolids. Nevertheless, o ur estimated LOEC and NOEC value s are within the range reported in the Liu et al. (200 9 ) study. Data suggest that fresh biomass yields of lettuce and radish are adversely affected at TCS concentration s ranging from 5 990 to 10 990 n g g 1 amen ded soil, but no TCS concentration affect ed the dry biomass yields.
147 Cucumber plants utilized in our study behaved different than the other crops D ue to d elay in fruiting of the plants, w e allowed the cucu mber to grow for a longer time. T he cucumber variety chosen for the study was cross pollinated, but the closed growth rooms had no facility for bee pollination. T hus, we hand pollinated the female flowers when they appeared, but were apparently un successful as few fruit formed and plants started to senesce before fruiting Overall, cucumber growth was poor in all the treatments and is unexplained. However, w e do not believe the severe senescence and poor fruit growth was a biosolids o r TC S treatment effect because s imilar problems occurred with plants grown in the control treatment Due to problems in growing the cucumber plants and poor fruit yield, bioaccumul ation data w ere of limited use. A few fruit were extracted to qua ntify the bi oaccumulation. T he TCS and Me TCS concentrations were below the detection limit (
148 accumulated no detectable TCS (< LOQ of 1 ng g 1 ), resulting in a n estimated BAF of <0.0001 for all the treatments (Table 7 5) On an average, lettuce lea ves experienced 10 fold greater TCS accumulation than the radish leaves and ~40 fold more than bahia grass. Duarte Davidson and Jones ( 1996 ) suggested that physico chemical pro perties of organic chemical s and plant species affect the accumulation of polar and non polar compound s Schroll and Scheu nert ( 1992 ) suggested greater bioaccumulation of hexachlorobenzene in dicots (lettuce and carrots) than in monocots (like maize, oats and barley) but did not suggest a n explanation Our stud y included both dicots (radish, lettuce) and monocots (bahia grass), and TCS accumulation was crop specific. G reater T CS accumulation ( 0.004 0.43) occured in the dicots (radish and lettuce) than in the monocot (bahia grass) (Tables 7 3 through 7 5) T he 10 fold difference in BAF value between the radish and lettuce leaves suggest variable accumulation in plants due to different plant characterstics (e.g., lipid content) Suter (2007) suggested that water and lipid content in plant tissues may affect the con taminant uptake by plants. B 1996). Thus, the data suggest inconsequential accumulation in radish (BAF = 0.004) and b ahia grass leaves (BAF = <0.001) but some accumulation in lettuce leaves (BAF = 0.04) grown in soil spiked with exceptionally high TCS concentrations The bioaccumulation potential increased with TCS concentration Uptake in the Below G round Biomass ( Radish Root ) Radish es grown in soil spiked with Trt 3 (TCS concentration = 9905 702 ng g 1 ) accumulated the greatest amount of TCS (9150 1187 ng g 1 ), corresponding to a BAF value of 0.93 0.14 (Table 7 6) The average BAF in radish root ( 0.43 0.38 ) was 10 fold the BAF (0.04 0.04) in lettuce and 100 fold the BAF (BAF = 0.004) in radish
149 leaves Figure B 1 (Appendix B ) compares the BA F values in the radish root and lettuce leaves at various TCS concentrations. The data suggest significantly (p<0.05) greater TCS accumulation in below ground bioma ss (i.e radish roots ) than in abov e ground biomass (i.e. lettuce leaves). Triclosan plant concentration s a nd BAFs increased with TCS concentration in amended soil ( Trt 1 to 3 ), but t he accumulation in Trt 3 was significantly greater (p <0.001) than in the Trt 1 and 2 ( Table 7 6 ). The above ground and below ground plant tissues analyzed for TCS were simultaneously analyzed for Me TCS, and the Me TCS concent r a tions were below the detection limit (1 ng g 1 ) for all the treatments. Bioaccumulation Field Study Concentrations in the soils collected from soybean fields were relatively constant (51 56 ng TCS g 1 soil) in various years (Table 7 7) whereas soil concentrations in the corn leaf fields were variable (43 100 ng TCS g 1 soil) (Table 7 8) The plant sampl es also displayed variable TCS accum ulation (Table 7 7 and 7 8 ) The monocot (corn leaf) accumulated less TCS than soybean (dicot) grain. Soybean grain BAF values varied widely across years with a greater accumulation (average BAF = 0.16 0.15) in 2002 than 2001 (average BAF = 0.06 0.09) (Table 7 7). Two of the three soybean samples from 2001 had BAF values <0.01, representing minimal accumulation (Table 7 7). The BAF values for all the corn leaf samples collected in 2006 were <0.01 (Table 7 8) representing inconsequential accumulation. Some corn samples collected in 2004 accumulated non negligible amounts of TCS, with an average BAF value of 0.07 0.05 (Table 7 8 ) but variability among the samples was large. Me thyl TCS concentrations were no n detectable in all the field grown plant s. Collectively, the fi e l d samples represented longer term measures of TCS phytoavailability than the growth chamber
150 studies. The field results were highly variable, but, were generally con sistent with the growth ch amber and greenhouse data Thus, TCS accumulation was slightly greater in the dicot ( soybean ) than the monocot ( corn ) plants. The maximum BAF value calculated for the field grown plants wa s 0.16. Model Predicted TCS Concentrations in Plant Tissue The measured BAFs in various plant s can be compared to BAFs predicted by a Biosolids Amended Soil: Level 4 (BASL4) c omputer model (Webster and Mackay, 200 7 ) The m odel parameters were chosen based on the physico chemical properties of TCS and soil s in which th e plants were grown. The m odel predicted different BAFs for above ground and below ground biomass es and grass, but failed to distinguish between the radish and lettuce leaves. Further, t he model predicts no significant difference among BAF s at various soil TCS concentrations. The BASL4 predicted BAF s (dry wt. ) were 23 1.2 for radish root 6.1 0.32 for lettuce and radish leaf, and 10.8 0.45 for bahia grass tissue (Tables 7 3 through 7 6). M odel ed (using BASL4) BAF values over predi ct s bioaccum ulation in all plant s and plant parts, and suggest significant TCS bio accumula tion. In contrast, measurement suggest that plant uptake of biosolids borne TCS is generally minimal and strongly influenced by soil TCS concentrations. The only similarity between the m odeled and measured values is the prediction of greater TCS accumulation in below ground (i.e roots) than the above ground (i.e leaves) biomass. Pla nt accumulation of organic chemicals has been predicted using empirical equations and the accumulation is be lieved to depend heavily on chemical partitioning coeffic i e n t (K ow ) with no consideration of soil characterstics or chemical exposure times. Such equations were previously utilized for risk estimation of various chemicals by USEPA. Travis and Arms (1988) derived an equation (equation 7 1) to describe
151 bioaccumulation of 29 hydrophobic chemicals including pesticides and dioxins in above ground plant biomass (multiple crop species ) The equation was derived for pesticides with a log K ow ranging from 1.75 and 6.15. The log K ow of TCS is 4.8 and thus, TCS bioaccumulation may be expected to follow the prediction. ( 7 1 ) Where Up : U ptake coefficient ( equivalent to BAF ) and the numerical values are regression parameters ; u sing log K ow of TCS as 4.8, C alculated uptake coefficient (Up) for TCS was 0.065 for all the above ground biomass. The estimated BAF value exceeded the measured BAF values for lettuce, radish and bahia grass leaves and underestimate d BAF for radish root Suter (2007) opined that the accumulation of chemicals is less likely to follow the empirical model predictions if the chemical, soil and plant species are different than the ones used to derive the model. A definitive prediction could only be made using the emp irical models if the models are validated to a wide range of chemicals, soils and plants. Thus, the mode l s and empirical equations (Equation 7 1) overpredict the phytoaccumulation of TCS and the es timates should be used with caution. Mechanism of Bioaccum ulation and Comparison with Other Studies The BAF v alues obtained in our study suggest some accumulat ion in the below ground biomass (roots) a nd translocation to the above ground biomass ( leaves ) but accumulation was greater in th e root crop than in above ground biomass M echanistic explanations for the difference are not known. T fo r TCS at 25 C is small (10 9 ), so vapor phase movement is likely minimal. Rather, movement in the aqueous phase and/or direct partitioning from soil to l ipid rich plant tissue is likely criti cal
152 (Trapp and McFarlene, 1995). Trapp and McFarlene ( 1995 ) and Wild and Jones (1992) suggest ed that l ipophilic compounds with log K ow > 4 ( like TCS ) have a l imited transport across t he endodermis membrane through soil solution. Further, Trapp and McFarlene (1995) suggested that non ionized chemical s entering the plant s through root or leaves move in the plant with water flow in the xy lem, but TCS movement with water is limited because of the low water solubility. Thus, TCS is expected to move (partition) from soil solid s to lipid enriched outer root parts and accumulate there The abovementioned concept is partially true in our study as we observed a g reater accumulation in the radish root s than in the leaves suggesting a direct partitioning to roots, but d etection of small levels of TCS in the leaves suggests that at least some translocation occurred from the root to the leaves even when the expectation of movement was minimal Wu et al. (2010) evaluated the TCS uptake by soybean ( Glycine max ) from a soil (sand, pH = 5.1, OC = 16 g kg 1 ) amended with spiked biosolids (11 Mg ha 1 ) to obtain a final TCS concentration of 7 0 ng g 1 amended soil Data suggested plant TCS accumulation, and the r ange of BAFs ( 1 6.5 ) was greater than the range of BAFs obtained in our greenhouse and field samples (<0.0001 to 0.43) D ifference in biosolids type and application rates likely caused different bioaccumulation in the two studies. The biosolids utilized in the Wu et al. (2010) study contained much less solid s content (19 g L 1 ) than biosolids (30 0 g kg 1 ) used in our studies ; lower solid s content would mean a greater relative TCS aqueous phase concentration, greater bioavailability and greater plant accumulation. In addition, the two studies differed in the biosolids loading rates. Wu et al. (2010) applied a biosolids load of 11 Mg ha 1 compare d with load ing rate of 228 Mg ha 1 used herein, which may have affect ed TCS bioavailability. Trapp
153 and McFarlene ( 1995 ) suggested that h igh biosolids application increases the OC content of soils and reduce s TCS uptake by roots and trans l ocati on via xylem ; thus, reducing the extent of plant accumulation. Previous studies also suggested variable accumu lation of antimicrobials and antibiotics (Boxall et al., 2006; Dolliver et al., 2007) in various plant parts Boxall et al. (2006) investigated the accumulatio n of antibiotics (concentration = 1 mg kg 1 soil) in soil (loamy sand), and found that some of th e an tibiotics accumulated more in the body of the carrot whereas some accumulated more in the outer peel of the carrot. Similarly, Dolliver et al. (2007) evaluated the accumulation of antibiotic sulfamethazine in pota toes grown in silt loam soil ( pH: 7.0; OC: 27 g kg 1 ). Sulfamethazine accumulated more in the outer skin of the tuber than the center of the potato. The m olecular weight, pK a and structure of TCS is similar to the antimicrobials (trimethoprim, sulfadiazine) de scribed in Boxall et al. (2006), a nd was expected to accumulate in a similar manner. Thus, similar to our study, the above mentioned studies suggest variable accumulation in various plant parts. Degradation in Soils The percent disappearance of TCS was calculated based on the TCS concentrations measured in soil immediately b efore and after the greenhouse study (Table 7 8 to 7 10) The d ata suggest TCS disappearance in all treatments utilized in our study. T he soil used to grow lettuce and radish was collected after 40 d wherea s soil sample used to grow bahia grass was collected after 6 0 d Previous TCS degradation data (Chapter 4) and values obtained from the literature suggest a primary degradation half li fe ( time for 50 % disappearance) of TCS of ~ 100 d (across soil type s ), an d formation of a metabolite (Me TCS). Thus a fter 40 60 d of plant growth in the soil
154 we expect ed ~ 20 25 % degradation/ disappearance of TCS. D ata (Table 7 9), however, suggest a maximum disappearance of 15 % The estimated percent disappearance was variable (shown by large standard error s associated with the data) and did not increase with TCS concentration (Tables 7 8 through 7 10 ) Further, no metabolite (Me TCS) was detected in the system suggesting that either the metabolite (Me TCS) appeared and became non extractable with time, or the Me TCS did not form due to reduced bioavailability of TCS. Reduction in TCS b ioavailability may occur because high biosolids application (228 Mg ha 1 ) is expected to favor strong interaction of TCS with OC matrix (Alexande r, 1995) Wil d and Jones (1991) observed slower degradation o f diethylhexyl phthalate (DEHP) in soil amended with high biosolids application (90 Mg ha 1 ), than in same soil amended at agronomic rates ( 22 Mg ha 1 ) These a uthors opined that the high biosolids dosages either decreased the DEHP bioavai la bility or created partial a naerobic soil conditions that slowed the degradation O ur study may also possess anaerobic microsites due to high water content (450 g kg 1 soil ) of the soils. Appearance of Me TCS and subsequent conversion to non extractable fraction with time may also occur, similar to previous observations (Chapter 4). Further, Me TCS was not detected in any of the plant parts, consistent with the absence of detectable Me TCS in biosolid s and soils utilized herein Comparison with Real World TCS Concentrations The spiked TCS concentration range utilized in the present study was 990 10 99 0 n g g 1 amended soil. The TCS concentration in Trt 1 (~ 990 n g g 1 amended soil) is equivalent to an agr onomic application rate (22 Mg ha 1 ) of biosolids containing a TCS concentration of 16 000 ng g 1 (mean concentration from TNSSS, 2009) for at least 6 years assuming no TCS losses. The Trt 1 was unique among other treatments as TCS
155 occured inherent to the biosolids (no spike additions) and is expected to represent a real world scenario. Soils with high biosolids applications resulting in high TCS concentrations similar to Trt 1 utilized herein are sometimes used in real world for landscap ing purposes. In fact, a biosolids application of 228 Mg ha 1 occurred on the landscaping field equ ilibrated soil utilized herein. Landscaping soils similar to the one mentioned abov e may be used for growing grass, but likely not for vegetables Thus, the vegetables grown in the landscaping soil represents a worst case estimate. Treatments 2 (5 990 ng g 1 ) and Trt 3 ( 10990 ng g 1 ) were included in the study design to obtai n a dose response relationship B iosolids contain ing a TCS concentration of 133 mg TCS kg 1 (highest biosolids concentration from TNSSS, 2009) applied for 10 years and assuming no TCS loss, r esults in a TCS concentration of 1 33 00 ng TCS g 1 amended soil Thus the TCS concentration in Trt 3 was unrealistic The Trt 1 and 2 are more reasonable, but still repre sent application s of representative biosolids at agron omic rates for up to 6 30 years, and assume no TCS loss over time. T he BAF s were reestimated by excluding Trt 3 The resulting average BAF values we re 0.02 for lettuce leaves and 0.26 for radish roots (Table B 1, Appendix B ), a s compared to the 0.04 and 0.43 with all treatments included Further, plant accumulation may differ when TCS is inherent vs spiked into the biosolids. Langdon (2010, per sonal communication) found different dissipation/ degradat ion rates of TCS when TCS occurs inherently vs when TCS is spiked into biosolids. Difference in dissipation rates may occur as the spiked TCS is likely to sorb to the outer portion of the biosolids and therefore be more available to microbes in the
156 presence of O 2 causing faster TCS degradation. Different dissipation rates between the in herent vs spiked systems might have created variable bioavailabilities in our Trt 1 (inherent TCS), Trt 2 and 3 (spiked TCS ) causing different accumulation. A r elationship between accumulation and chemical occuring in spiked vs inherent form could not be e stablished, as our study was conf ounded by the different concent r a tions in the two treatments; however, accumulation data in plants exposed to Trt 1 are likely more accurate. Thus, w hile the TCS concentrations utilized in our study were unrealistic for typ ical biosolids a pplied at agronomic rates, results may mimic phytoavailability following l ong term biosolids applications or exceptionally high application rates Results suggest that inherent or spiked biosolids borne TCS applied at excessively high conc entrations and applied for multiple years will not result in significant a ccumulation of TCS in above ground plant parts; however some accumu lation is possible in root crops like radish.
157 Table 7 1. Selected physico chemical prop erties of the soils and biosolids used in the present study Soils Organic carbon Mehlich 1 P NH 4 + N pH (1:1) EC Texture g kg 1 mg kg 1 S cm 1 Control 30 0.4 3.2 0.2 8.9 0.2 6.0 1.7 512 25 Clay loam Filed b iosolids a mended 84 0.3 790 12 27 1.4 7.1 0.0 506 2.3 Silty clay loam Table 7 2. Yield of plant parts of three plant species represented by fresh weights (g) in the control and biosolids amended treatments (same letters represent no statistical difference among treatments) Plant type Amended soil TCS concentration (ng g 1 ) (Treatment) Fresh weight (g) Dry weight (g) Lettuce Control 990 (Trt 1 ) 5990 (Trt 2 ) 10 990 (Trt 3 ) 25.7 0.90 c 79.7 6.77 b 55.6 9.23 a 55.9 6.28 a 6.17 1.00 a 13.2 4.94 b 13.7 4.15 b 15.2 3.69 b Bahia grass Control 990 (Trt 1 ) 5990 (Trt 2 ) 10 990 (Trt 3 ) 12.5 1.07 a 42.2 3.70 b 30.2 7.23 b 27.7 9.83 b 0.10 0.02 a 7.82 0.61 b 4.54 1.26 b 6.41 4.78 b Radish Control 990 (Trt 1 ) 5990 (Trt 2 ) 10 990 (Trt 3 ) 22.5 6.31 a 26.9 9.70 a 17.3 2.60 a 11.1 2.10 b 3.70 0.64 a 7.98 0.97 b 4.55 0.75 a 7.56 3.16 b
158 Table 7 3. Measured TCS concentrations ( average ; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the lettuce leaves grown in a biosolids amended silty clay loam soil ( same letters represent no statisti cal difference among treatments) Type of application (Treatment) Measured biosolids amended soil concentration Measured plant tissue concentration Mean BAF dry wt. Calculated BAF (BASL4) ng g 1 Inherent (T rt 1 ) 1015 94.2 10.7 2.1 0.01 0.00 a 5.92 a Spiked (T rt 2 ) 4570 448 119 14 0.03 0.00 b 6.47 a Spiked (T rt 3 ) 10145 1043 897 117 0.09 0.01 c 5.93 a Average 0.04 0.04 6.1 0 0.32 Table 7 4. Measured TCS concentrations ( average ; n = 3 or 4 and SD ) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the radish leave s grown in a biosolids amended silty clay loam soil ( same letters represent no statisti cal difference among treatments) Type of application (Treatment) Measured biosolids amended soil concentration Measured plant tissue concentration Mean BAF dry wt. Calculated BAF (BASL4) ng g 1 Inherent (Trt 1 ) 989 68.4 <1 <0.001 a 5.92 a Spiked (Trt 2 ) 4531 439 16.9 2.20 0.004 0.00 a 6.47 a Spiked (Trt 3 ) 9905 702 62.8 21.3 0.006 0.00 b 5.93 a Average 0.004 0.002 6.1 0 0.32 Table 7 5. Measured TCS concentrations ( average ; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the Bahia grass grown in a biosolids amended silty clay loam soil ( same letters represent no statisti cal difference among treatments) Type of application (Treatment) Measured biosolids amended soil concentration Measured plant tissue concentration Mean BAF dry wt. Calculated BAF (BASL4) ng g 1 Inherent (Trt 1 ) 977 68.4 <1 <0.0001 a 10.9 a Spiked (Trt 2 ) 4492 395 <1 <0.0001 a 11.2 a Spiked (Trt 3 ) 9221 832 <1 <0.0001 a 10.3 a Average <0.0001 10.8 0.45
159 Table 7 6. Measured TCS concentrations ( ave r a ge ; n = 3 or 4 and SD) and bioaccumulation factors (BAF) and BASL4 model calculated BAFs in the radish root grown in a biosolids amended silty clay loam soil ( same letters represent no statisti cal difference among treatments) Type of application (Treatment) Measured biosolids amended soil concentration Measured plant tissue concentration Mean BAF dry wt. Calculated BAF (BASL4) ng g 1 Inherent (Trt 1 ) 989 68.4 101 18.7 0.10 0.02 a 22.4 a Spiked (Trt 2 ) 4531 439 1244 132 0.27 0.04 a 24.5 a Spiked (Trt 3 ) 9905 702 9150 1187 0.93 0.14 b 22.4 a A vearge 0.43 0.38 23 .1 1.2 1 Table 7 7 Bioaccumulations factors (BAF) [average (n=3) and standard error (SE)] obtained in grains of soybean grown in field soils. Soybean grain Y ear collected TCS soil concentration (ng g 1 ) TCS plant concentration (ng g 1 ) BAF (dry wt.) 2001 51 .8 <1 <0.01 5 2.8 <1 <0.01 55. 0 10. 1 0.18 Av erage SE 0.06 0.09 2002 51 .0 <1 <0.01 54.9 17.1 0.31 55.8 10.1 0.18 Av erage SE 0.16 0.15 Table 7 8 Bioaccumulations factors (BAF) [average (n=3) and standard error (SE)] obtained in leaves of corn grown in field soils. Corn leaves Y ear collected TCS soil concentration (ng g 1 ) TCS plant concentration (ng g 1 ) BAF (dry wt.) 2004 55.2 6.62 <0.01 100 <1 <0.10 43.8 5.38 0.12 Av erage SE 0.07 0.05 2006 5 3.1 <1 <0.01 5 2.0 <1 <0.01 5 5.0 <1 <0.01 Av erage SE <0.01 The BAF in non detects were calculated by assuming a TCS soil concentration of LOQ (1 ng g 1 ) of the instrument
160 Table 7 9 Measured TCS soil concentrations in lettuce treatments (means; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % disappearance Typ e of application (Treatment) I nitial TCS soil concentrations F inal TCS soil concentrations % disappearance ng g 1 Inherent (Trt 1 ) 1015 94.2 873 95.2 13 10 Spiked (Trt 2 ) 4570 448.4 4372 438 3.2 17 Spiked (Trt 3 ) 10145 1043 8826 875 12 5.9 Table 7 10 Measured TCS soil concentrations in radish treatments (means; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % disappearance Type of application (Treatment) I nitial TCS soil concentrations F inal TCS soil concentrations % disappearance ng g 1 Inherent (Trt 1 ) 989 68.4 874 93 11 10 Spiked (Trt 2 ) 4531 439 4224 386 7 6 Spiked (Trt 3 ) 9905 702 8422 472 15 9 Table 7 11 Measured TCS soil concentrations in bahia grass treatments (means; n = 3 or 4 and SD) before and after the plant accumulation study and the corresponding % disappearance Type of application (Treatment) I nitial TCS soil concentrations F inal T CS soil concentrations % disappearance ng g 1 Inherent (Trt 1 ) 977 68.4 956 43.4 1.8 8.0 Spiked (Trt 2 ) 4492 395 4080 330 8.8 8.9 Spiked (Trt 3 ) 9221 832 8654 613 5.8 8.0
161 Figure 7 1 Representative photos of lettuce (A and B ) and radish (C and D ) plants that compare plant growth in the control and treatments Control: no biosolids, Trt 1(no spike) = T CS spiked concentration of 990 ng g 1 Trt 2 (5 ppm TCS) = TCS spiked concentration of 5990 ng g 1 A B C D A B C D
162 Figure 7 2. Representative photos of bahia grass that compare plant growth in various TCS treatments Trt 1(no spike) = TCS spiked concentration of 990 ng g 1 Trt 2 (5 ppm TCS) = TCS spiked concentration of 5990 ng g 1 ppm) = TCS spiked concentration of 10,990 ng g 1 A. B.
163 CHAPTER 8 MOBILITY OF TRICLOSAN (TCS) IN BIOSOLIDS AMENDED SOILS Background Triclosan [5 chloro 2 (2,4 dichlorophenoxy) phenol] (TCS) is an antimicrobial agent commonly added to a number of consumer products. Plastic an d textile manufacturing industries use TCS as an additive. The inherent antibacterial property of TCS makes it a n effective agent for use in household articles like sp onges and k itchen chopping boards The incorporation of TCS in a vast array of products results in its discharge to wastewater treatment plants (WWTPs). The products of WWTPs are liquids (effluents) an d solids (sludge). Processed sludge forms biosolids, which may then be land applied The hydrophobic nature of TCS (log K ow 4.6 to 4. 8: Halden and Paull, 2005; Ying et al., 2007) portends sign ificant partitioning of TCS originally in wastewater streams int o biosolids. Reported biosolids TCS concentrations in the extensive (78 WWTPs ) 2009 Targeted National Sewage Sludge Survey (TNSSS) (USEPA, 2009a) were 0.4 to 133 mg kg 1 with an overall mean of 16 65 mg kg 1 ( including statistical outliers) An average WWTP in U.S produces 240 kg dry weig ht of biosolids per million lit r e s of wastewater treated (Ki nney et al., 2006). Nationally, WWTPs produce ab o ut 7 million Mg of biosolids each year ~ 63 % of which is land applied ( NRC, 2002 ). Assuming an average biosolids borne TCS concentration of 16 mg kg 1 (mean value from TNSS S ), land application represents a potentially significa nt source (~70 Mg year 1 ) of TCS to the land. S ome worry that the T CS presence in the environment could contribute to the spread of bacterial resistance and threaten human drug therapy (Birosova and Mikulasova, 2009; Pycke et al., 2 010 ; McMurry et al., 1998). McMurry et al. (1998)
164 recovered TCS resistant E.coli on agar plates with a mutation of fab 1 gene. Besides, TCS affect s aquatic organisms by blocking enzyme carrying proteins, causing concerns of the possible build up of bacterial resistance in these organisms (Nghiem and Coleman, 2008). Numerous aquatic TCS toxicity data are summarized in Table 1 1 (Chapter 1). Triclosan can be toxic to fish with chronic toxicity value of 22 mg L 1 (Lindstro m et al., 2002). Orvos et al. (2002 ) reported TCS toxic ity to Daphnia magna with a 48 h median effective concentration (EC 50 ) of 390 mg L 1 and fathead minnow ( Pimephales promelas ) with a 96 h median lethal concentration (LC 50 ) of 260 mg L 1 Therefo re movement of TCS through and over biosolids amended soils to surface water s could be an environmental concern. Despite the occurrence of TCS in biosolids, and frequency of biosolids land a pplication, little is known about the mobility of TCS in amended soils. Cha and Cupples (2010) assessed the leaching po tential of TCS in biosolids amended soils using biodeg radation and sorption data in a simple leaching model developed for pesticides ( Gustafson, 1989). The model involved calculation of leaching potential using groundwater ubiquity scores (GUS); and chemical s with GUS <2.8 were termed non leachable. This m odel assumed reversible chemical sorption and first order chemical degradation half life. The model predicted that TCS in biosolids amended to sandy and loam soils (concentration = 0.05 2 mg TCS kg 1 amended soil) wo uld experience insignificant leaching (GUS<0.7) Topp et al. (2008 ) used simulated rainfall in fi e l d studies to estimate the runoff potential of TCS in soil amended with surface applied and incorporated liquid biosolids. Triclosan appeared in runoff water from the surface applied liquid biosolids treatment collected immediately and 266 d after the biosolids
165 application The r ea son for TCS appearance at day 266 was not known but was attributed t o the persistence of TCS in bound residues especially during the winter months. Sabo u rin et al (2009) conducted a simil ar TCS leaching study following the application of dewatered biosolids Less than 1 % of the total biosolids TCS mass appeared in the runoff water suggesting minimal mobility. Lapen et al. ( 2008) suggested a potential f or TCS transport via shallow tile drainage systems to surface waters. Triclosan was detected in ti le drains when soils received either liquid or dewatered biosolids (Lapen et al., 2008; Edwards et al., 20 09 ). Greater TCS concentrations ( 3.68 g L 1 ) in tile drain s appeared following liquid biosolids ap plication (Lapen et al., 2008) than appearing follo wing dewatered biosolids application (0.24 g L 1 ) (Edwards et al., 2009). Xia et al. (2010) reported that 49 to 64% of total TCS mass applied in biosolids treatm ents was found at a depth of 30 to 120 cm, suggesting significant leaching in soils amended with biosolids for 33 years. T he studies described above represent somewhat unique conditions Lapen et al. (2008) utilized liquid biosolids (11.9 g solids L 1 biosolids) the soil in Xia et al. ( 2010) was a calcareous mine spoil (pH = 7.8 ) and the biosolids used in Edwards et al. (2009) had high pH (~7.5). Liquid biosolids portends greater amounts of TCS in the aqueous phase as the fluidity of liquid biosolids i s similar to water. The high pH in two systems was close to the pK a of TCS (~8) ; likely making the dissociated TCS more available for transport within and over the soil surface. The log K oc values determined in soils, biosolids, and biosolids amended soils for TCS were large (>4), and reasonably constant for all solid ma trices (Agyin Birikorang et al., 2010). The large values suggest a propensity of TCS to partition onto organic
166 carbon ( OC ) associated with biosolids, soil and sediment in the environment. H ysteresis coefficient s f or TCS determined in soils, biosolids, and biosol ids amended soils matrices were <<1 ( 0.002 to 0.005 for biosolids, 0.25 to 0.30 for biosolids amended soils, and 0.41 to 0.42 for un amended soils) suggesting highly restricted desorption (Agyin Birikorang et al., 2010). The small hysteresis coefficient s particu larly for the biosolids, suggest that reincorporation of TCS to the solution during desorption is practically negligible (Mamy and Barriuso, 2007). Triclocarban (TCC) with soil log K oc (3.82 ) similar to log K oc of TCS (4.26 ) experienced minimal leaching (<1%) in a biosolids amended soil, and formation of TCC bound residues (Snyder 2009 ) Triclosan b ound residue s were also observed in biodegradation study of TCS conducted in biosolids amended soil (Chapter 4). Based on the exp ected behavior of TCS, we hypothesized that the mobility of biosolids borne TCS is strongly retarded in amended soils. Besides TCS, a degradation metabolite of TCS is likely to appear in soil s as the primary degradation half life of TCS in amended soil is 100 d (Chapter 4). The objective of the present study was to determine the mobility of TCS ( and possibly Me TCS ) in bio solids amended soil. L eachability was assessed in a laboratory column study utilizing a sand soil with a pH
167 event, and at study termination, the soil s in the columns were sec tioned and extracted to quantify the depth of TCS and Me TCS movement. Material s and Methods Experimental Design The d esign consisted of incorp oration of TCS spiked biosolids, or soils, into soil columns and application of two irrigation regimes T he irrigation regimes were based on the rainfall received in Gainesville, FL area. The 40 year ( 1970 2009) average annual rainfall in Gainesville is 132.8 cm (oc curring in 126 days). More than half of the rainfall (74.8 cm) occurs in the summer (June September, occu ring in 49 days). I n the summer, an average of ~ 1.53 cm rainfall occurs at a frequency of 2.7 days, yielding a weekly average of 3.97 cm. Irrigation consisted of empirical and stochastic regimes and soil treatments included controls (no biosolids), and biosolids i ncorporated with the top 2.5 cm of soil The treatment combinations were: Empirical control Stochastic control Empirical biosolids treatment Stochastic biosolids treatment The irrigation regime consisted of the following: Empirical irrigation : Qua ntit y of water applied to the columns was based on the long term L 0.5 pore volumes/ week Stochastic irrigation : Random water application s (quantity and frequency) to mimic actual long term rainfall pattern s but the total quantity of water applied over the experimental period was the same as the quantity applied in the empirical irr igation regime Chloride was utilized as a non interactive conservative tracer At each leaching event, i rrigation water was mixed with chloride to assess water movement during the experiment.
168 Soils, Biosolids and Chemicals As the study w as conducted in Gaine sville, we utilized soil (sand; OC: 3 g kg 1 ) collected from the E horizon of a n uncoated Myakka fine sand ( sandy, siliceous hyperthermic Aeric Alaquods). 14 C TCS radiolabeled uniformly on the chlorophenol ring (specific activity 48 mCi mmol 1 and 99% purity) was custom synthesized by Tjaden Biosciences (Burlington, IA). Ecoscint A liquid scintillation cocktail was purchased from Nat ional Diagnostics (Atlanta, GA). P ota ssium chromate, silver nitrate, methanol (MeOH) acetone and othe r solvents were purchased from Fisher Scientific (Atlanta, GA). An anaerobically digested class A cake b iosolids ( s olids content : 320 g kg 1 ) was collected from a domestic WWTP in Illinois The biosolids contained 5 mg kg 1 of TCS and non detectable (<0.7 mg kg 1 ) Me TCS (Chapter 2). Tagging and Application of Biosolids The biosolids (100 g) were spiked with 0.06 mCi of 14 C TCS (specific activity = 48 mCi mmol 1 ), resulting in a total TCS biosolids concentration of ~8 .6 mg TCS kg 1 Although the biosolids TCS concentration was below the mean concentration (16 mg kg 1 ) found in TNS S S it was within the range reported in the TNS S S (USEPA, 2009a) The biosolids (10 g dry weight equivalent/tube) were weighed into 50 mL Teflon tu bes (total of 10 tubes), and spik ed with 5 mL of MeOH 14 C TCS The 14 C TCS spiked biosolids samples were se aled and equilibrated for 24 h, th e MeOH was evaporate d for an additional 24 h in a hood T he biosolids samples from various tubes were composited and thoroughly mixed. The columns utilized in the study were 30 cm in length and 5 cm in diameter. Plain sand filled 20 cm of the column and was supported by porous plate. To encourage uniform water distribution, 10 cm of pea gravel was placed at the top and a suction of 0.1 bar was applied to the column Appropriate
169 quantities of the spiked thoroughly mixed biosolids (~6 g) were incorporated to a depth of 2.5 cm into the soil at equivalent rate of 22 Mg ha 1 (1%, by wt). Control treatments consisted of spraying 14 C directly on the top of the soil (no biosolids) to achieve the same TCS concentration as in the biosolids amended treatments The suction head of 0.1 bar applied to the columns prevent ed saturation at the column bottom, and maintain ed uniform moisture conte nt ( 10% by weight, roughly field capacity ) throughout the column. The s chedule of the l eaching events and the amount of irrigatio n applied i s described in T able 8 1. Pore Volume Determination A p reliminary experiment was conducted to determine a represent ative column pore volume (PV) using the gradual wetting method. Four columns were filled with 6 00 g of dry soil and weighed. W ater was gradually introduced into the columns throug h the column bottom using Ma riotte bottles, until the soil was saturated. A suction head of ~0.1 bar (similar to the experimental conditions) was then applied to the columns until t he excess water drained out. T he columns were then re weighed a nd the difference in weight (~148 g) was assum ed to be 1 PV. Leachate Coll ection and Analysis After each leaching event, leachate was collected from each column, and analyzed for tracer ( chloride ) concentration and 14 C activity. C hloride concentration was determined in an aliquot (5 mL) of leachate using the Argentometric titra tion procedure (Sheen and Kahler, 1938; Kumar, 2006). The method consisted of titration with silver nitrate using potassium chromate as indicator An other aliquot (1 mL ) of the leachate was mixed with 10 mL of scintillation cocktail to determine 14 C activity Measurement was performed for 5 min on a liquid scintillation counter (LS 6500, Beckman Coulter
170 Inc., Fullerton, CA, USA) with background correction against blank samples ( 14 C free cocktail ). Determination of 14 C Activity in the Soil After the s tudy termination pea gravel on the soil surface was carefully removed and the soil in each column was gently pushed up from the bottom of the columns with a plung er. The soils in biosolids amended treatment and controls were sectioned into 0.5 cm incremen ts up to 2.5 cm and at 2.5 cm increments from 5 to 20 cm. A subsample of each soil section was combusted at 900C in a Harvey Oxidizer Model OX 500 (Tappan, NY), and analyzed for total 14 C acti vity. Additional ly, a subsample from the zone of biosolids incorporation (0 2.5 cm ) from each column was extrac ted (twice) with MeOH: acetone (50:50; v/v) solvent as described previously (Chapter 4) The extract was then subjected to RAD TLC analyses ( for 14 C speciation) as described previously ( Chapte r 4 ) Result s and Discussion Leachate Recoveries and Tracer Breakthrough L eachate recoveries were calculated as the ratio of the volume of irrigation water collected beneath the columns to the volume of irrigation water applied to the columns. R ecoveries in the empiri cal and stochastic irrigation treatments at each leaching event were similar and >90% and approximately constant for the entire study period (62 d) (Figure 8 1). R ecoveries in the control and biosolids treatments were also similar Most of the applied water was recovered with no indication of soil saturation and biosolids application did not retard water movement. Consistent with the constant leachate recoveries column masses were nearly constant for the study period and similar t o the masses measured at time ze ro (data not
171 presented). Constant column masses suggest that the study conditio ns remained constant through the study period The tracer (chloride) breakthrou g h curve is represented in Figure 8 2. Breakthrough of Cl did not differ between the control and biosolids am ended treatments and was similar for both irrigation treatments. Similar breakthrough indicates similar water flow despite different frequency of water application. At 1 PV, 50% of Cl appeared in the leachate sugg esting uniform convective transport and lack of preferential flow (Mamo et al., 2005). However, t he breakthrough curve for the empirical treatments (Figure 8 2 a ) was more symmetrical than for the stochastic treatments (Figure 8 2b ) After leaching for about 4 PV, the relative concentration of Cl was almost zero suggesting that all the Cl was reco vered from the columns. 14 C in Leachates and Recoveries by Combustion Leachate collected from columns in all the treatments contained no detectable 14 C throughout the study period, suggesting that all the applied 14 C TCS remained in the soil Addition of up to 4PV of water did not move 14 C to the bottom of the columns. Different irrigation regimes and TCS spiked onto biosolids or directly on to the soil surface affect the TCS leaching behavior T he percent 14 C recoveries from the soils sectioned at various depths are presented in T able 8 2. Average r ecov ery of 14 C from the top (0 2.5 cm ) depth in the empirical treatment was 97.8 0.83%. The soil segment immediately below incorporation (2.5 5 cm) contained 2.08 0.84%, but only 0.01% of 14 C was detected at the depth >5 cm. Similarly, f or the stochastic treatment, the average recovery was 93.7 2.05 % at the 0 2.5 cm depth, 6.17 2.04% at 2.5 to 5 cm de p th, only 0.01% of 14 C beyond the 5 cm depth Data f or biosolids treatments suggest that the majority of the
172 14 C (93 97%) was contained with in the depth of incorporation ( 0 2.5 cm ) irrespective of the irrigation treatment s (Table 8 2) Triclosan association with biosolids was likely greater than in the soils, as the partitioning coefficient of TCS is greater in the biosolids (log K d = 3.76 ) than in the soil s alone (log K d = 2.25 ) (Agyin Birikorang et al., 2010) A ppearance of 14 C below 2.5 cm likely represents contamination from the top depth during the soil sectioning process. Minimal 14 C detected at depth s beyond 5 cm suggests that biosolids borne TCS did not move beyond the depth of biosolids incorporation ( 0 2.5 cm ) except f or a little 14 C in the 2.5 to 5 cm depth. R ecove r ies in the controls were 77 to 82% i n the 0 to 1 cm depth a nd from 17 to 22% in the soils in 1 to 2 cm depth (Table 8 2). As the initial TCS spiking occurred at t he surface of the soil column, significant 14 C detected at 1 to 2 cm depth suggest some TCS movement. Absence of biosolids in the controls likely allowed some move ment with irrigation water, but TCS did not move beyond 2 cm. Triclosan experienced g reater s orption in biosolids amended soils than in the un amended soils (Agyin Birikorang et al., 2009). Lower sorptive capacity of an un amended soil suggests a great er movement with soil solution. Cha and Cupples (2010) predicted the leaching of TCS spiked (0.05 2 mg kg 1 ) in biosolids and amended to sandy and loam soils, using biodegradation and sorption data in a simple leaching model Predictions for both soils are similar to our results suggesting insignificant TCS leaching in the presence of biosolids. In contrast, Lapen et al. (2008) reported leaching of TCS into tile drains (concentration of 3.68 g L 1 ) located 80 cm below the soil surface following a liquid biosolids application. Liquid biosolids (11.9 g solids L 1 biosolids) have greater (relative) amounts of TCS aqueous phase and could be expected to promote greater TCS
17 3 mobility. Triclosan in the aqueous phase may move by gravity based flow through large pores and worm borrows (preferential flow) (Turpin et al., 2007), and may intersect the shallow tile drains. Further, movement by facili tated flow may occur when TCS associates with dissolved organic carbon (DOC). Edwards et al. (2008) conducted a similar leaching study in soils amended with dried biosolids, and found detetable TCS concentrations (0. 24 g L 1 ) in tile drains, but the concentrations were much lower than in the Lapen et al (2008) study. Leaching occuring with dried biosolids application was due to the pH (7.5) of the biosolids that was sufficiently close to the pk a (~8) to favor greater TCS concentrations in aqueous phase. Similarly, Xia et al. (2010) reported TCS leaching to a depth of 30 to 120 cm in soils annually amended with biosolids for 33 years. Soil was calcareous min e spoil (pH = 7.8) and the high pH likely increased TCS solubility making it more sus c eptible to movement. A column study was con ducted with biosolids borne TCC leached with 0.5 PV at each leaching event resulting in ~4PV of leachate at the study termination. The amount of water (PV) was similar to our study a nd a leaching of <1% was observed in a 3.5 month leaching study conducted in 17 cm long columns (Sn yder, 2009). The study utilized biosolids with varying OC contents. The authors opined that partitioning coefficient (K d ) influenced the leaching; as a small er K d portends a greater re a dily leachable (aqueous) fraction and a greater mobility Gibson et al. (2010) measured the leaching of TCS (84 1032 ng L 1 ) applied with irrigation water to 30 cm long soil columns for 90 days. No TCS was detected in the column leachate; and s oil sectioning confirmed minimal leaching beyond the 10 cm soil de p th Greater leaching of TCS in the Gibson et al. (2010) study than our study was likely due to greater TCS concentration in the aqueous phase. The stud y thus reemphasizes that the
174 relative TCS movement would be greater when TCS is present in the aqueous form as opposed to TCS occurring bound to biosolids OC 14 C Recoveries by Extraction and Extract Speciation Extraction was performed on soil samples tak en from the zone of biosolids incorporation ( 0 2.5 cm), as the combustion data (Table 8 2) indicated that the majority of 14 C added was present at this depth. The average recoveries of 14 C in amended treatments for both irrigation treatments (85 86% ) (Table 8 3) were less than the recoveries obtained through combustio n (93 97% ) (Table 8 2). Extraction r ecoveries were expected to be smaller than recoveries through combustion, as the combustion procedure measures total (i.e. TCS associated with labile an d non extractable fractions) concentrations. The recoveries in the controls were 96 to 98 % in the both irrigation treatments (Table 8 3) similar to combustion procedure recoveries. Data suggest that biosolids application reduces the extractable 14 C recove ries There are no published reports of the leaching or run off potential of Me TCS Because >95% of 14 C was recovered in the 0 2.5 cm depth; these samples were subjected to RAD TLC analysis to determine 14 C speciation. The data (Table 8 4) suggest degradation of TCS and appearance of a metabolite (Me TCS). Appearance of Me TCS was similar to our previous observations (Chapter 4). I n t he biosolids amended treatments, the average 14 C identified as TCS was 51 to 62%, and the average 14 C identified as Me TCS was 37 to 4 8%, irrespective of the irrigation treatment. The TCS recoveries were generally greater in the empirical than the stochastic irrigation regime T he corresponding Me TCS recoveries were great er in the stochastic trea t ments. In the control s 58 to 62 % of 14 C was ide n tified as TCS and 38 to 42 % as Me TCS The TCS and Me TCS recoveries were variable among the treatments; but the recoveries were
175 not significantly different suggesting no effect of irrigation or biosolids treatments on TCS speciation/degradation. In biosolids treatments, >95% of 14 C remained in the zone of biosolids incorporation ( 0 2.5 cm ) Extraction and speciation data from that depth suggest that Me TCS was formed but remained in the top depth segment, suggesting minimal movement in amended soils. In controls, appearance of TCS and Me TCS beyond the initial surface spiking of TCS suggest that either TCS leached to the lower dept h and deg r aded to Me TCS or Me TCS appeared in the top depth and subsequently leached t o the low er depth. The first mechanism is more probab le as the reported log K ow (5) of Me TCS i s greater than the log K ow (4.8) of TCS and, therefore, Me TCS movement is likely more retarded than TCS. Thus, Me TCS mobility /leaching would be equal or smaller than to the mobility of TCS, and rate of TCS infiltration is sufficiently slow to allow degradation. The speciation data allowed the estimation of time for 50 % disappearance of TCS. Data su g gest that ~ 40 % of Me TCS (Table 8 4) wa s detected after 60 days which suggests a 50% disappearance time of ~ 75 days. The disappearance time calculated herein i s within the range measured in our incubation study (Chapter 4), and other published studies of TCS half lives (Lozano et al., 2010; Kwo n et al., 2010; Wu et al., 2009). Comparison with CML S M odel The C hemical M ovement in L ayered S oils (CMLS) is a model that can be utilized to estimate the chemical movement in response to movement of water as well as chemical degradation. The CMLS has the ability to provide leaching estimates varying with weather and spatial variability of soils and uncertainties in chemical properties.
176 The model estimates the depth of center of mass of a chemical (non pola r) with time and assumes that a chemical o nly moves in the liquid phase in response to soil water environment. Water that is already in the profile is pushed by the inflowing wat er ( piston flow ) and water is also lost from the root zone by evapotranspiration and deep percolation. C hemical movement is assumed to be retarded due to reversible sorption on the soil solids. The model makes several simplifying and conservative assumptions, including: complete first order degradation (metabolites not considered), uniform soil and chemical proper ties withi n a soil layer and linear and reversible chemical sorption P referential flow i s not considered The model estimates the location of the center of chemical mass rather than chemical concentration profile The simplifying assumptions are not true for TCS m ovement in biosolids ame n ded soil as our study (Chapter 4) suggested that TCS degradation fo llowed zero order kinetics, the formation of a metabolite, and strongly hysteretic desorption ( Agyin Birikorang et al., 2010). Thus, CML S serves a s only a first ap proximation of biosolids borne TCS movement. Figure 8 3 present CML S model simulation of the movement of a model default chemical [Dimethyl Tetrachloroterephthalate (DCPA)] with K d (log K d = 3.7) and half life (100 days) values similar to TCS. Simulations suggest that a chemical with such charac teristics applied to a sand soil (with low organic carbon c ontent ), under the rainfall conditions similar to the average annual rainfall in Gainesville require s many years to move completely through a soil column of 20 cm. A fter 1 year of rainfall, the chemical is predicted to remain in the top 2 cm of the soil profile Our column study results were similar to the CML S model prediction as we observed a TCS movement to 2.5 cm in ~ 2 months (6 0 days). Our data are also c onsistent with the recent modeling
177 re sults by Cha and Cupples (2010). As the CML S modeling approach i s relatively simple, f uture work should include simulation s utilizing more sophisticated models that consider irereversible and non equilibrium chemical so rption and the fate and transport of the metabolite as well. Implications of TCS Movement Leaching of TCS depends on the equilibrium between TCS in the soil solution and sorbed phase. As expected, the leaching was slower in a biosolids amended soil than a soil without biosolids, as TCS associates with the OC of the biosolids causing retardation to movement. Similar to TCS, Me TCS formed in the control and amended soil remained in the top depth (0 2.5 cm) Mobility of contaminants in a soil can be ass essed using the relative mobility factor (RMF) (USEPA, 2008 a ). The RMF is defined as the ratio of leaching distance of test substance (i.e. TCS or Me TCS ) to the leaching distance of the reference substance (i.e. chloride). The chemical is defined as immob ile if the RMF is 0.15 (USEPA, 2008). Triclosan and Me TCS moved to a maximum depth of 2.5 cm and the reference chemical moved up to 20 c m depth. The calculated RMF is 0.12 5 and both TCS and Me TC S would be regarded as immobile in an un amended and a bio solids amended soil in 2 months of leaching period. Our column leachin g study represents a scenario where biosolids borne TCS was amended to a sand soil of low OC content and subjected to extensive (4PV) leaching Mobility of TCS depends on the partitionin g coefficient (K d ). The K d in a sandy soil is 5.88 smaller than the K d in a silty loam soil (Agyin Birikorang et al., 2010) meaning that a greater fraction of TCS would be available for leaching in a sandy soil as compared to a loam or clay soil. Gibson et al. (2010) observed a variable leaching potential of pharmaceuticals due to difference in K d values. Thus, t he leaching of TCS is expected
178 to be even less under comm o n field conditions characterized by soils with greater organic carbon contents a nd vegetative cover. Previous studies found TCS concentrations (0.24 3.68 g L 1 ) in tile drains in s oils amended with liquid and dried biosolids (Edwards et al., 2008; Lapen et al., 2008) It may be noted that the tile drain TCS concentrations would be di luted upon interception with the surface water and further reduce TCS concentrations. Implications of TCS concentrations to aquatic organisms are critical. Langdon et al. (2010) quantified the risk of biosolids borne TCS to aquatic organisms from surface r unoff and leaching and suggested that maximum conce ntration of TCS reported in the leaching and runoff water did not adversely affect the aquatic ecosystem. Mobility data obtained herein and other data available in the literature are utilized in our risk d iscussion (Chapter 9).
179 Table 8 1. Amount and schedule of leaching events for the emp irical and stochastic irrigation regimes Empirical irrigation regimes Stochastic irrigation regimes Day Amount ( mL ) Day Amount ( mL ) 5 79 5 50 12 78 11 30 19 78 15 70 26 78 17 75 33 78 26 70 40 78 31 90 47 78 41 80 54 78 46 65 61 78 55 60 68 78 67 90 75 78 73 80 82 78 79 60 86 65 Total 936 Total 885 Table 8 2 Average p ercent recov eries standard deviation of 14 C (by combustion procedure) by depth in the control and biosolids treatment for each irrigation regime Depth (cm) 14 C Recovery (%) Empirical Stochastic Treatment Treatment 0 2.5 97.8 0.83 93.7 2.05 2.5 5 2.08 0.84 6.17 2.04 5 20 0.0 1 0.00 0.0 1 0.0 0 Control Control 0 1 77.4 3.3 1 82.5 3.7 0 1 2 22.4 3.2 0 17.4 3.7 0 2 2.5 0.06 0.0 0 0.01 0.0 0 2.5 20 0 .0 0 0.00 0 0 0 0.00
180 Table 8 3 Average p ercent extraction recoveries of 14 C standard deviation in the top depth of the control and biosolids tr eatment for each irrigation regime Depth (cm) 14 C Recovery (%) Empirical Stochastic Treatment Treatment 0 2 .5 85.1 10 86.4 10 Control Control 0 2 .5 96.2 10 97 .9 8.1 Table 8 4. Speciation of 14 C extracted standard deviation in the top de p th of control and biosolids treatments for each irrigation regime (same letters represent no significant difference among treatments). Depth (cm) TCS (%) Me TCS (%) Empirical treatment 0 2.5 62.5 0.7 a 37.5 0.7 b Stochastic treatment 0 2.5 51.9 5. 3 a 48.1 5. 3 b Empirical control 0 2 .5 62 2.5 a 37.7 2.5 b Stochastic control 0 2 .5 58 2.6 a 42 2.6 b
181 Figure 8 1. Percent reco very of the amount of water that was collected at bottom of the column at various intervals during the study period in the (a) empirical and (b) stochastic irrigation regimes 0 20 40 60 80 100 0 10 20 30 40 50 60 Leachate recovery (%) Time (days) a. Empirical Control Treatment 0 20 40 60 80 100 0 10 20 30 40 50 60 Leachate recovery (%) Time (days) b. Stochastic Control Treatment
182 Figure 8 2 Chloride b reakthrough curves in control and biosolids amended soils of the (a) empirical and (b) stochastic irrigation regimes 0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 6 Relative concentration (C/C0) Pore volume a. Empirical Control Treatment 0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 6 Relative concentration (C/C0) Pore volume b. Stochastic Control Treatment
183 Fig ure 8 3 Pr ediction of chemical movement obtained using the CMLS model. The graph rep resents the depth of DCPA movement (chem ical properties similar to TCS) in a sandy soil with time.
184 CHAPTER 9 RISK ASSESSMENT OF B IOSOLIDS BORNE TCS Background Risk assessment is a process of assigning magnitudes and probabilities to the adverse effects of human activities. The process involves identifying a hazard, and qua n tifying the relationship between e vent and its effect (Suter, 2000 ). Screening level risk assessments are often conducted initially by assuming worst case scenario s to narrow the scope of subsequent assessments. Specific pathways in which the screenin g level assessment s are above critical limit s are selected for more detailed risk evaluations ( Suter, 2007). Human health risk assessments are sometimes considered as alternative s to ecological assessments. T he notion is that protection of humans (typically, the most sensitive receptors) will automatically protect non human species. However non hum an receptors are possib ly exposed to greater chemical conce ntrations, or are more sensitive than humans. Thus, rather than conducting independent assessment s ecological and human risk (combined) assessment s should comp lement each other (Suter, 2007) and be more protective overall. Several ecological risk assessments were conducted recently ( Reiss et al., 2009; Fuchsman et al., 2010; Langdon et al., 2010 ) to address the risk of biosolids borne triclosan ( TCS ) Fuchsman et al. (2010) estimated the exposure to terrestrial organisms (soil microbes, plants, mammals, birds) using a probabi listic fugacity based model. Triclosan concentrations in various organisms and their e xposures were estimated assuming either equi librium or steady state conditions Toxicity v alues from the literature were utilized to establish effect c oncentrations and to identify sensitive species
185 Results suggested no adverse effect s of biosolids borne TCS ( except for nitrogen cycling disruption ) under even worst case scenarios ( considering t he most sensitive toxicity endpoint), and the effects were transient The study addressed ecological risks utilizing simple fugacity models and a probabilistic approach, but did not consider TCS degradation. Further, corresponding data for validati on of th e probabilistic model are lack ing (MacKay and Barthouse, 2010). A screening level assessment (Langdon et al., 2010) qua n ti fied the risk of surface runoff and leaching of biosolids borne TCS to aquatic organisms using the haza rd quotient (HQ) approach. Run off concentrations were predicted using literature values for biosolids TCS concentrations and modeled TCS physico chemical properties (K ow K oc K d ). The assessment identified some risk (HQ>1) to the most sensitive aquatic species (gre en algae) i n a worst case scenario using the greatest predicted run off concentrations Additional assessment that protected 95 % of the stu dy population resulted in lower risk but HQ values remained >1 Langdon et al. (2010) opined that hydrophobic chemicals like TCS are expected to have low aqueous phase concentration s due to higher parti tio ning coefficient s than used in their study and to hysteret ic sorption desorption behavior Consistently, Agyin Birikorang et al. (2010) reported strong hysteretic desorption behavior o f TCS incomplete desorption, and lower aq ueous phase TCS concentration s than expected. Reiss et al. (2009) followed the European Union System for the Evaluation of Substances (EUSES) model and supplemented his approach with United States regulatory guida nce The m argin of safety (MOS) concept was utilized to assess the risk of biosolids borne TCS (5 and 15 mg TCS kg 1 biosolids ) Soil TCS concentrations were
186 estimated from agronomic biosolids applications (5 10 Mg ha 1 ) to assess risks to earthworms, soil organisms and plants. E stimated TCS concentrations in earthworm tissues were utilized to as sess risk to earthworm and fish eating birds and mammals. The a uthors suggested no significant risk (MOS>>100) to the species fro m the TCS concentrations typically found in soils even after applying the biosolids annually for 10 year s Most of the data (half lives, K d microbial and plant toxicity, earthworm accumulation) utilized in the Reiss et al. (2009) and Fushman et al. (201 0) assessment s were either derived from models or were measurement s extracted from unpublished sources. Reiss et al. (2009) considered a relatively short half life (35d) for TCS in amended soils and Fushman et al (2010) did not consider TCS degradation. Thus, a characterization of biosolids borne TCS behavior, and a risk assessment based on measured data is expected to be more accurate Our approach for the preliminary risk estimation was similar to Reiss et al. (2009) We began by identifying the relevan t exposure pathways from the list of pathways used in the Part 503 Biosolids Rule (USEPA, 1994 ) We then chose the most relevant measured parameters for the assessment (half life, LC 50 NOAEL, K oc etc.), and calculated haza rd indices (HI) for each pathway Hazard indices are the ratio of exposure concentration of a species to the corresponding species endpoint. Pathways in which HI values exceeded one were identified and subsequently subjected to detailed evaluation Our risk estimation utilized measured d ata generated herein to build on the previous assessments and extended the risk evaluation t o humans. Thus, our results provide a more comprehensive evaluation of risk of biosolids borne TCS than previously published evaluations
187 Pathways of Exposure The biosolids borne TCS risk assessment included 14 exposure pathways utilized in the risk assessment supporting the Part 503 Biosolids Rule (Table 9 1 ). Similar to the recent triclocarban ( TCC ) risk assessment (Snyder, 2009), additional pathways were iden tified to incorpor ate terrestrial animal exposure to contaminated fish (pathway 15) and aquatic organism exposure to contaminated surface water (pathway 16) The receptor for each pathway was a h ighly exposed individual (HEI), the indiv idual from a population that has the greatest realistic exposure The HEI differs from a most e xposed individu al (MEI) sometimes used in risk assessment, who has unrealistic exposures and does not actually exist (Epstein, 2003). Exclusion of Exposure Pathways Risk to H uman s P revious assessments examined the human risk to TCS exposure through the use of personal care products containing TCS (NICNAS, 2009) or through drinking water contaminated with TCS (USEPA, 2008 b ). NICNAS (2009) utilized the mar gins of exposure (MOE) concept; t he MOE is similar to the HI approach of USEPA for characterizing o ccupational and domestic risk of chemicals to huma ns The MOE is a measure of the likelihood that a partic ular adverse health effect occur s in an organism after an exposure as compared to an unexposed organism The g reater the MOE, the smaller is the potential of an adverse effect or risk. If MO E > 100, an adverse a ffect is not expected. The exposure path ways in the assessment include d inhalation and dermal exposure through T CS containing household products (NICNAS, 2009). R esults suggested that even repeat dosage s over long period s of time ( ~2 y ear ) posed no risk to human s as all the MOE s were >100. R isk s to children (assumed to b e the most
188 sensitive population) through exposure to TCS in toy paints and breast milk were also calculated The MOE s for all the exposure pathways for children were > 100 and there was minimal expected risk f rom inhalation an d dermal exposure to TCS from the use of personal care products (NICNAS, 2009) The US EPA conducted a screening assessment of TCS human risk through consumption of TCS containing drinkin g water as a part of the Reregistration Eligibility Decision (RED) for TCS ( USEPA, 2008 b ) The assessment utilized TCS concentration s of 56 ng L 1 in drinking water found in southern California ( Loraine and Pettigrove, 2006) Assuming a water intake rate of 2 L the calculated TCS i ntake rate i s 1.4 ng kg 1 d 1 for a 70 kg adult. The USEPA suggested a TCS reference dose (RfD) of 300,000 ng kg 1 d 1 (McMahon, 1998), which was same for the acute and chronic dietary end points. The TCS intake rate ( 1.4 ng kg 1 d 1 ) was many fold smaller than the RfD ( 3 00,000 n g kg 1 d 1 ) and t he exposure concentration never reach ed the RfD. H azard was deemed negligible even after a 100 y ear expo sure Further, TCS degrades in aquatic environments (Canosa et al., 2005) so the actual exposure concentration would be even smaller. As the screening assessment of TCS from drinking water did not suggest a risk, a qua ntitative risk estimation w as not conducted by US EPA. More r ecently, Benotti et al. (2009) reported TCS concentrations in sou rce water and finished drinking water derived from streams impacted by wastewater effluent The maximum (6.4 ng L 1 ) and mean (3 ng L 1 ) TCS concentrations were nearly 10 fold smaller than the California concentration s used for the US EPA risk estimation above Thus, estimated risk s using these concentrations would be even smaller. R ealistic TCS concentration s in drinking water will no t re present a ris k to humans even over long term ( 100 y ear) exposure.
189 Risk to Aquatic Organisms Several a quatic risk assessments for TCS have been conducted (Lyndall et al., 2010 ; Langdo n et al ., 2010; Brausch et al., 2010). Abundant TCS toxicity data are available for aquatic organisms like fish es amphibians algae and vascular plants (T ables 9 2 and 9 3 ), and include values from both acute and chronic toxicity st udies. Algae and invertebrates are generally m ore sensitive to TCS exposure than fish and vas cular pla nts. Brausch et al. (2010) attributed algal sensitivity to the disruption of lipid biosynthesis through fab 1 (fatty acid synthesis) McMurry et al. ( 1998 ) and Lu and Arc her ( 2005 ) identified TCS toxicity to FASII (enoyl acyl carri er protein reduc tase) pathways Lyrge et al. ( 2003 ) and Franz et al. ( 2008 ) identified membrane destabilization and Newton et al. ( 2005 ) identified uncoupling of phosphorylation Langdon et al (2010) focused on the risk assessment of pharmaceuticals and personal care products entering the aquatic environment through land applied biosolids using the HQ approach Land a pplication allows TCS transfer to soils and a potential for TCS mig ration to surface waters through leaching or runoff. E stimated runoff water TCS concentration predicted using an eq uilibrium partitioning (reversible desorption) approach was 4.5 g L 1 P redic ted environmental concentration s were compared with the aquatic toxicity end points to dete rmine possible adverse ef fects. Conservative a ssumptions included using the greatest run off TCS concentrations, and toxicity end points for the most sensitive aquatic species. A p reliminary a ssessment, utilizing conservative parameters, suggested some hazard of TCS (HQ> 1) and warranted detailed assessment. Consistently a preliminary assessment by Braush and Rand (2010) also suggested some risk of TCS to aquatic species ( HQ >1 ) but the authors acknowledged that the estimated risk was for a worst case scenario. Detailed
190 i nvestigation by Langdon et al. (2010) included more realistic aquatic toxicity and runoff concentrations to protect 95 % of the aquatic species. The HQ was still > 1 impl ying that a runoff TCS concentration (0.59 g L 1 ), predicted to protect 95 % of the popu lation pose d some threat to the most sensitive species. However, t he runoff TCS concentration (0.59 g L 1 ) used exceeded the range of TCS concentration s ( 0.025 0.10 g L 1 ) previously reported in runoff water an d tile drainage from field application of cake biosolids ( Topp et al., 2008; Sabourin et al., 2009 ; Edwards et al., 2009 ). T ile drain TCS concentrations of 3.68 g L 1 were measured following liquid biosolids application but may represent unique conditions. A pplication o f l iquid biosolids portends greater TCS aqueous phase concentrations as the fluidity of liquid biosolids is similar to water Langdon et al. (2010) utilized an estimate d log K oc of 3.97 that was smaller than the measured value of 4.26 (Agyin Birikorang et al., 2010) and hysteretic adsorption was not considered (Agyin Birikorang et al., 2010). Additionally, Langdon et al. (2010) ignored degradation or dilution of TCS despite previous evidence of signif icant degradation of TCS in b i osol ids amended soil over time (Lozano et al., 2010; Kwon et al., 2010; Wu et al., 2009). Thus, the risk predicted by Langdon et al. (2010) study was conservative and the actual risk to the aquatic organisms is likely less. Lyndall et al (2010) evaluated TCS risk to aquatic (al g a e, bacteria, and invertebrates ), sediment dwelling organism s and aquatic feeding wildlife from WWTP effluent entering the surface waters Triclosan concentrations in surface water were either obtained from literature or predicted from a fugacity based model assuming steady state condition s Wild life exposures were calculated using standard wildlife dietary exposure equation s (USEPA, 1993 c ). Toxicty to aquatic organisms was
191 evaluated based on a species sensitivity distribution (SSD) approach which Capedevielle et al. ( 2008) used to assess TCS risk in freshwater environments. The SSD consisted of a toxicity distribution based on chronic toxicity data fo r several species, and is regarded as more realistic than NOEC toxicity values bas ed on the most sensitive species. The estimated 95 th percentile TCS concentrations in surface waters, sediments, and biota tissues were well below the 5 th percentile of the respective species sen sitivity distribution s suggesting no a dverse effect of TCS e xposure Triclosan risk assessme nt s in aquatic species (Langdon et al., 2010, Lyndall et al., 2010) consistently conclude that TCS risk to aquatic organisms is unlikely. Thus, we did not perform an independent assessment on aquatic species in our study. Th e risk estimations conducted for humans (USEPA, 2008 b ; NICNAS, 2009) and aquatic organisms (Langdon et al., 2010; Lyndall et al., 2010) suggested limited TCS risk. Thus, the aquatic and human exposure pathways initially included for our risk assessment can be ex cluded. The pathways were excluded when detailed studies were either previously conducted (aquatic organisms) or t here was minimal predicted risk through particular pathways (human) The six pathways that were excluded include B iosolids soil airborne dust human Biosolids soil air human Biosolids soil groundwater human Biosolids soil surface water human Biosolids soil surface water animal Biosolids soil surface water aquatic organism The remaining pathways considered in the risk evaluation are presented in Table 9 4. Further, the ingestion exposure of a non gardener human is considered similar to the home gardener and represents the worst case scenario for the exposure. P athways 1 and 2 (Table 9 1 ) were grouped toget her and considered in pathway 3 (Table 9 4).
192 Reference Dose Calculation The USEPA (1993 c ) defines the human reference dose ( RfD ) as an estimate of t o be without an appreciable risk of deleteri is typically expressed in units of milligrams per kilogram of bodyweight per day (mg kg 1 bw d 1 ). Reference doses can be calculated from non human endpoints as follows: Where : UF: uncertainty factors MF: modifying factor based on professional judg e ment (default = 1) NOAEL: n o o bserved adverse effect leve l Uncertain ty factors for obtaining human RfD for non cancerous chemicals are calculated according to the following cr iteria (USEPA, 1993). U ncertainty factors are applied when extrapolating toxicity data from : Animal study data to humans (10 a ) Prolonged exposure data to average healthy humans (10 b ) U se of a less than chronic study exposure (3 s to 10 s ) Deriving an RfD from a LOAEL rather than a NOAEL (10 l ) Modifying factors a ccoun t ing for scientific uncertain ties typically range from 3 mf to 10 mf As discussed earlier, acute t oxicity of TCS to humans via oral and dermal routes is low. Triclosan also did not cause in vivo genotoxicity, carcinogenicity reproductive or developmental toxicity in rodents. Various animal toxicity studies are available (Table 9 5) that suggest rat and mouse sensitivity to TCS The m ajority of the studies were either derived from unpublished literature or lack suffici ent evidence of the credibility. A 13
193 week repeat TCS t oxicity study conducted in mice suggested a NOAEL of 25 mg kg 1 d 1 (NICNAS, 2009), and the mice was considered more sensitive to TCS exposure However, the adverse effects were perox isome proliferator type, considered ir relevant for human risk assessments. So, this end point (mouse study) was not appropriate for calculating the RfD for humans. A c h ronic study (2 years ) tested the sensitivity of rat to mild clinical chemistry and hematology changes ( hypertrophy and hepatocyte vacuoliza tion ) in male cells (NICNAS, 2009 ). An i ngestion NOAEL of 40 mg kg 1 bw d 1 was identif ied (Table 9 5) in the study, and was deemed appro priate for our risk calculations. The RfD for humans was calculated using uncertainty factors for extrapolation of animal to human study (10 a ) and prolonged exposure data to average healthy human (10 b ) as follows: RfD = (40 mg kg 1 bw d 1 ) / (10 a 10 b ) = 0.40 mg kg 1 bw d 1 Similarly, an RfD value for terrestrial animal s was calculated from the same rat study utilizing uncertain ty factors specific for ecosystem risk assessment. An uncertainty factor of 5 is applied if the test species and endpoint species of interest are in the same class but different order (5 a ). Differences in species sensitivity, laboratory field extrapolation, and intraspecific variability can each be accounted for with uncertainty factor s of two each (2 b 2 c 2 d respectively) (Suter, 2007). The RfD for all terrestrial animals was calculated as follows: RfD = (40 mg kg 1 bw d 1 ) / (5 a 2 a 2 b 2 c ) = 1 mg kg 1 bw d 1 Predators like American woodcock herring gu ll, and short tailed shrew feed on earthworms, and may consume TCS that has accumulated in the worm bodies. Reiss et al. (2009) suggested an acute LD 50 of 862 mg kg 1 d 1 for avian species. The RfD for
194 predators is estimated by applying appropriate uncertainty factors. An uncertain ty factor of 10 is applied if the toxicity end point is derived from less than chronic studies (10 s ) if endpoint is based on LOAEL (or LD 50 ) rather than NOAEL (10 l ), and a modifying factor of 3 to account for scientific uncertainties (3 mf ). RfD = (862 mg kg 1 bw d 1 ) / (10 s 10 l 3 mf ) = 2.87 mg kg 1 bw d 1 Parameters for Risk Estimation Environmental Fate Although our study (Chapter 4 and 8) and other published studies (Lozano et al., 2010; Wu et al., 2009; Xu et al., 2009) suggested TCS degradation in b iosolids amended soils, we assumed no loss of TCS through degra dation (worst case scenario) for the screening level risk estimation. Effect on Soil Dwelling Organisms Toxicity to earthworms The acute toxicity to earthworms was assessed according to OPPTS g uidelines ( USEPA, 1996b ). Mortality was assessed in a 28 d exposure of earthworms to TCS spiked in biosolids, and amended to sand (IFS), silty clay loam (ASL) and an artificial soil (Chapter 5 ). A n unreplicated range finding test suggested TCS effects on the earthworm mortality exposed to TCS concentration ( 5 mg kg 1 biosolids) only in the sand A replicated definitive test conducted on the sand showed no effects of TCS on earthworm mortality, and no abnormal symptoms were observed in the live earthworms up to a TCS concentration of 105 mg kg 1 biosolids (Chapter 5). The estimated 28 d LC 50 value was >105 mg kg 1 biosolids or >1 mg kg 1 amended sand soil. The estimated 28d LC 50 for ASL and artificial soil was >10,000 mg kg 1 biosolids or >100 mg kg 1 amende d
195 soils. For screening level risk assessment, the most sensitive toxicity end point of 1 mg kg 1 amended soil was considered. Bioaccumulation in earthworms The measured earthworm BAF values were 6.5 0.84 in the IFS soil and 12 3.08 in the ASL soil The BASL4 model underestimated the bi oaccumulation potential of TCS, with calculated BAF values of 5 0.0 ( IFS soil) and 2.2 0.0 (ASL soil) The earthworms collected from field soil (similar in texture to ASL soil) had an average BAF of 4.3 0.7 ( Chapte r 5 ). A conservative estimate of 10 as the earthworm BAF was utilized to calculate the predicted environmental concentration s in earthworms ( Table 9 6 ) grown in soils amended with biosolids borne TCS Toxicity and Bioaccmulutaion in Plant Biomass The average BAF values in a greenhouse study were 0.43 0.38 in radish root 0.04 0.04 in lettuce 0.004 0.002 in radish leaves and the BAF value in bahia grass was <0.0001. T he greatest BAF value (0.93) was obtained in the highest TCS treatment (10 mg kg 1 amen ded soil) utilized in our study. The BAF values were also measured in the field collected samples. The average BAF in field grown soybean gra in were 0.06 to 0.16 (Chapter 7). The BASL4 model greatly over estimated the bioaccumulatio n potential o f TCS in plant tissues. The calculated a verage BAF values were 6 to 23. A worst case conservative estimate of 0. 93 was used for the screening level risk estimation. Avian and Mammalian Toxicity The risk posed to birds that feed on earthworms was examined using pre dicted earthworm concentrations. In a manner similar to Reiss et al. (2009), we selected two avian species that consume high quantities of earthworms, the American woodcock and
196 the herring gull. The risk posed to mammalian species w as also examine d in a species with high earthworm consumption, the short tailed shrew. As no other reliable data source s were available, t he TCS toxicity data (Table 9 6) for the avian and mammalian species were obtained from Reiss et al (2009) and appropriate uncertain ty factors were applied to calculate RfD that applies to all predators. Screening Level Assessment Exposure Concentration Calculation According to the US EPA Part 503 risk assessme nt f or the biosolids rule (USEPA, 1994), a n HI ca lculation is performed to determine if a particular pollutant/pathway is a potential problem. The HI is the ratio of the predicted environmental concentration (PEC) of a chemical to the established terrestrial or human health criteria (values obtained from toxicity studies). The terrestrial toxicity values are NOAEL, LOAEL, LC 50 or LD 50 and the human toxicity values are represented by RfD s A chemical/ pathway with a n HI of >1 is considered problematic and requires a tier 2 assessment of hazard rankings ( USEPA, 1995). Reiss et al. (2009 ) conducted a terrestrial risk assessment utilizing Snyder (2009) used the USEPA HI approach and conducted an integrated human and environmental risk assessment of biosolids borne tricl ocarban (TCC). We utilized the equations described in Reiss et al. (2009 ) and Snyder (2009) for our initial risk estimation Triclosan exposures were calculated by multiplying concentrations in relevant environmental media by the corresponding animal or h uman intake rates in that media. Exposure concentrations were calculated using measured or predicted data. S creening level exposures assumed two biosolids application scenarios: a one time application rate of 50 Mg ha 1 an application of 5 Mg ha 1 for 100 years
197 s were similar to the Part 503 Biosolids Ru le risk assessment (NRC, 2002). T ypical one time biosolids application rates are 5 10 Mg ha 1 as opposed to 50 Mg ha 1 considered here, and land application at a single location is unlikely to occur for 100 consecutive years. The TCS concentration used was the 95 th percentile TCS concentration (62 mg kg 1 ) from TNSSS (USEPA, 2009a), using the 95 th percentile contaminant concentration was specified in the Part 503 risk assessment. All other parameters and equations required for the risk estimation are described in Table 9 6. Screening L evel Hazard Index Calculation Hazard i n dices for each exposure pathway were calculated as the ra tio of exposure concentration to the RfD for the corresponding species endpoint ( T able 9 7 ) The screening level assessment indicated an HI value of well below one for the majority of the exposure pathways, under both biosolids application scenarios (worst case and 100 year). The resulting HI values suggest that the HEIs would experience minimal risk from exposure to biosolids borne TCS over their life time even when exposed to the 95 th percentile TCS concentration The critical pathways (pathways 1 8 and 9 ) are the only ones in which th e HI value was greater than one, a nd those were selected for the second tier assessment. Tier 2 Assessment Consideration of TCS Degradation For the tier 2 assessment, we considered TCS d egrad a tion to recalculate the HI for the pathways that were critical (HI>1) in the screening level risk assessment. T he time required for 50% disappearance of biosolids borne TCS estimated from the 14 C TCS biodegradation study was 77 d for the ASL soil (si lty clay loam) and >126 d for the
198 IFS (sand) soil (Chapter 4). The leaching study (Chapter 8) estimated time taken for and bioaccumulation criteria ( http://www .Pbt profiler.n et/criteria.asp ), a chemical is termed persistent if the half life in soil is greater than 60 d. Thus, biosolids borne TCS is considered persistent in both soils (USEPA, 1999). For simplicity, we assumed a half life of 100 d for risk estimation (Table 9 6). Further, our short term (18 week ) laboratory incubation study suggested zero order kinetics for TCS disappearance, but other published studies assumed a first order half life. Th us, for simpli fi cation and conservatism we also assu med a first order rate equation, as t he zero order kinetics result in non detectable TCS after a year Assuming an average half life of 100 d, the average TCS concentrations in soils (CS) over time wer e calculated as follows : (9 2 ) Where: At t=0, (Table 9 6) k = first or der decay rate constant (0.693/ t 1/2 ) = 0.693/100 = 0.007 d 1 t = time after biosolids application P arameters r equired for the calculation are described in Table 9 6. A comparison of TCS concentrations estimated using a half life of 100d, with concentrations assuming no TCS loss (no degradati on) is described in Figure 9 1. The results suggest that assuming degrada tion, ~10% of TCS r emains in the soil after 1 year, and the TCS
199 concentration did not change (reached steady state) even when the biosolids containing TCS are applied for 100 years. Pathway 1: Biosolids soil plant (direct phytotoxicity) P lants grown in amended soils (pathway 1) may experience some risk of TCS following biosolids application. Biosolids with 95 th percentile TCS concentration applied at worst case application rates (50 Mg ha 1 ) did not present a risk (HI = 0.15 9 ); however, lo nger term (100 year) scen ario presented a risk (HI = 1.59). The HI in the 100 year scenar io was 10 fold greater than the risk associated with biosolids applied through the single application (Table 9 7) If the biosolids TCS concentration is adjusted to th e mean concentration from the TNSSS (16 mg TCS kg 1 ), the HI values for the 100 year scenario reduces to 0.38, which is well below the critical value of one. Alternatively, consider ing a TCS degradation hal f life of 100 d, the majority of TCS would disappe ar in 1 year (Figure 9 1), and the HI for both the worst case and 100 year scenario is less than one, even when considering the 95 th percentile TCS concentration. Pathway 8: Biosolids soil soil organism An HI of one was exceeded for the acute toxicity to Eisenia Foetida (earthworms) for the 100 year scenario. The HI values wer e calculated using a conservative toxicity end point obtained from our study in sand soil T he end point s in the silty clay loam and the artificial soils were 10 fold greater, and the HI reduces to 0.16 for the 100 year scenario in these soils If we adjust the HI using the more realistic average TCS concentration from TNSSS (16 mg kg 1 ), the adjusted HI reduces to 0.04 for the worst case and 0.41 for the 100 year scenarios Further, c onsidering TCS degradation, very little TCS would b e detected at the end of 1 year (Figure 9 1) or 100 years and the adjusted HI would be much less than o ne.
200 Pathway 9: Biosolids soil soil organism predator For the worst case scenario, the HI was less than one for herring gull, but greater than one for the American woodcock and short tailed shrew The 100 year scenario resulted in HI of 33 for the Ame rican woodcock, 14 for the short tailed shrew and 3.3 for herring gull (Table 9 7). The different HI values for the predators reflected variable earthworm diets. The most sensitive sp ecies was the American woodcock, whose earthworm diet was assumed to be 77%. Considering a TCS degradation half life of 100d, the majority of TCS would disappear in 1 yea r and the HI for both the worst case and 100 year scenario would be l ess than one, even using the 95 th percentile biosolids TCS concen tration. Thus, the adjusted HI values for plants, earthworms and predators were all less than one when TCS degradation was considered However, the HI value of greater than one for various organisms (estimated in the screening risk assessment) under 100 year scenario suggests that biosolids borne TCS pollutant limits are needed to guide sustainable use of long term biosolids land application practice. Sources of Uncertainty in Our Risk Estimation Absence of reliable data on TCS toxicity to predators Using the estimated toxicity values for earthworms and plants. T he highest concentration values selected in our studies did not cau se any adverse effects, and the toxicity end points were estimated to be greater than the highest concentration tested in each study So, the actual toxicity to organism s may be much lower (i.e greater toxic end points ) Exclusion of toxi city and ac cumulation of major metabolite of TCS i.e. Me TCS in the risk assessment as the fate and transport data f or Me TCS is essentially unknown.
201 Calculation of Preliminary Biosolids Borne TCS Pollutant Limits The USEPA Part 503 Biosolids Rule ( USEPA, 1995 ) incl udes pollutant limits to ensure the sustainable land application of biosolids. T he various pollutant limits include : Cumulative pollutant loading rates (CPLRs), Annual pollutant loading rates, Ceil ing concentration limits, and Po llut an t concentration limit s. The pollutant limits were calculated herein based on the most sensitive pathway tha t had an HI of greater than Cumulative pollutant loading rates (CPLRs) A CPLR is the maximum pollutant load (kg ha 1 ) that can be applied to a soil through biosolids application. When the pollutant load reaches CPLR at a site, biosolids application does not have to cease, but only biosolids that meets the pollutant concentration limits can be applied thereafter The CPLRs for TCS were obtained from the risk assessment results and represent the maximum cumulative TCS l o a d in amen ded soil that remains protective of human and ecological health. The CPLR can be calculated by setting the HI for a relevant exposure pathway to on e and back calculating the TCS concentration in the amended soil. Our screening level assessment suggested that American woodcock under the 100 yea should be protective of the bird. The CPLR is the concentration in soils calculated by setting the HI equation (pathway 9) equal to one as follows: HI = 1= (FI WD* CS BAF worm / BW) / RfD Where: FI = food ingestion rate (g d 1 d.w.) WD = Worm diet (%)
202 CS = Soil concentration (mg TCS amended soil 1 ) BAF worm = Bioaccumulation factor in the worm BW = Body wt. (g) RfD = Reference dose The calculated CPLR for the American woodcock is 5.8 kg ha 1 (equivalent to 2.9 mg TCS kg 1 amended soil) The CPLR is thus the amount of TCS that can be land applied without expectation of adverse effects. The calculated CPLR translates to one time application of biosolids containing TNSSS 95 th percentile TCS concentration (62 mg kg 1 TCS kg 1 ) at a rate of 93 Mg ha 1 Annual pollutant loading rate (APLR) The APLR is the maximum amount of pollutant that can be annually applied from biosolids or given away in a bag or container for land application ( USEPA, 1995 ). Estimation is based on a conservative estimate of a 20 year of site life of lawns and home gardens. The APLR for TCS is calculated by dividing the CPLR by 20 assuming 20 years of land app lication and no TCS loss. The resulting APLR (0. 29 kg TCS ha 1 y 1 ) is protective of HEI present in the amended soils. Ceiling concentration limit Cei ling concentration limits are the maximum pollutant concentration s allowable in land applied biosolids. The limits were set to prevent the use of low quality biosolids containing extremely high pollutant concentrations. A ceiling concentration limit is the less stringent of the 99 th percentile TCS concentration in biosol ids and pollutant limit calcul ated from risk assessment. The 99th percentile TCS concentration fr om the TNSSS (USEPA, 2009a) was 197 mg kg 1 biosolids. Alternatively, t he ceiling concentrati on was calculated from the CPLR value obtained in the amended soils The
203 resu lting ceiling concentration wa s 127 mg TCS kg 1 biosolids f or the worst case scenario and 12.7 mg TCS kg 1 biosolids for the 100 year scenario. Pollutant concentration limit Pol lutant concentration limits are the most stringent concentration limits f or pollutant land applied in biosolids Biosolids that meet the pollutant concentration limits have fe w regulatory requirements for land application. The p ollutant concentration lim i t assumes a total biosolids load of 1000 Mg over 100 years or less and is calculated by dividing the CPLR by 100 years of application at 10 Mg ha 1 y ear 1 The resulting risk based TCS concentration limit based on the most sensitive organism (predator, pathway 9) for is 6.3 mg TCS kg 1 biosolids. The concentration is much lower than the repre sen tative TCS concentration (10 20 mg kg 1 ) in the b iosolids across U.S (Chapter 2), and mean concentration (16 mg k g 1 ) reported in the TNSSS (USEPA, 2009a), but similar to the TCS concentration of CHCC biosolids (5 mg kg 1 ) utilized in our study. We accept our overall hypothesis of minimal risk of biosolids borne TCS to human and environmental health. Thus, b iosolids containing a 95 th percentile TCS concentration (62 mg kg 1 ) applied at agronomic rates (5 Mg ha 1 ) for multiple years (up to 100 years) did not adversely affect the majority of the organisms (according to tier 2 assessment) included in the exposure pathway s described herein.
204 Table 9 1 Human and ecological exposure pathways for land applied biosolids ( US EPA, 1995 ). Pathway Description of HEI 1. Biosolids soil plant human Human (except for home gardener) lifetime ingestion of plants grown in biosolids amended soil 2. Biosolids soil plant human Human (home gardener) lifetime ingestion of plants grown in biosolids amended soil 3. Biosolids soil human Human (child) ingesting biosolids 4. Biosolids soil plant animal human Human lifetime ingestion of animal products (animals raised on forage all of which is grown on biosolids amended soil) 5. Biosolids soil animal human Human lifetime ingestion of animal products (animals ingest biosolids directly) 6. Biosolids soi l plant animal Animal lifetime ingestion of plants grown on biosolids amended soil 7. Biosolids soil animal Animal lifetime ingestion of biosolids 8. Biosolids soil plant Plant toxicity due to uptake of bios olids borne TCS when grown in biosolids amended soils 9. Biosolids soil soil organism Soil organism (e.g earthworms microbes ) ingesting biosolids/soil mixture 10. Biosolids soil soil organism predator Predator of soil organisms that exist in biosolids amended so ils 11. Biosolids soil airborne dust human Adult human lifetime inhalation of particles (dust)
205 Table 9 1 Continued Pathway Description of HEI 12. Biosolids soil surface water human Human lifetime drinking surface water a nd ingesting fish containing TCS from biosolids 13. Biosolids soil air human Adult human lifeti me inhalation of volatilized TCS from biosolids amended soil 14. Biosolids soil groundwater human Hum an lifetime drinking well water containing TCS from biosolids that lea ched from soil to ground water 15. Biosolids soil surface water animal Animal lifetime drinking surface water a nd ingesting fish containing TCS from biosolids 16. Biosolids soil surface water aquatic organism Aquatic organism exposed to runoff water containing TCS from biosolids
206 Table 9 2. Acute aquatic t oxicity endpoints of TCS from published studie s. Species Trophic group End point Result (LC 50 ,mg L 1 ) Source D.magna Invert. 48 h 0.39 Orvos et al. (2002) Ceriodaphnia dubia Invert. 24, 48 h (pH= 7) 0.2 Orvos et al. (2002) Pimephales promelas Fish 24,48, 72, 96 h 0.36, 0.27, 0.27, 0.26 Orvos et al. (2002) Lepomis macrochirus Fish 24, 48, 96 h 0.44, 0.41, 0.37 Orvos et al. (2002) Oryzias latipes Fish 96 h 0.602 (larvae) 0.399 (embryos) Ishibashi et al. (2004) Xenopus laevis Amphibian 96 h 0.259 Palenske et al. (2010) Acris blanchardii Amphibian 96 h 0.367 Palenske et al. (2010) Bufo woodhousii Amphibian 96 h 0.152 Palenske et al. (2010) Rana sphenocephala Amphibian 96 h 0.562 Palenske et al. (2010) Pseudokirch niriella subcapitata Algae 72 h growth 0.53 g L 1 Yang et al. (2008)
207 Table 9 3 Chronic aquatic t oxicity endpoints of TCS from published studies Species Trophic group End point Result (LC 50 ,mg L 1 ) Source D. magna Invert. 21d survival, reproduction 200 (NOEC) 200 (LOEC) Kopperman et al. (1974) C. dubia Invert. 7 d survival, reproduction 50 6 Kopperman et al. (1974) C. dubia Invert. 7 d survival, reproduction IC 25=170 Carlson and Caple (1977) Chironomus riparius Invert. 28 d survival, emergence 440 Schultz and Riggin (1985) Chironomus tentans Invert. 10 d survival, growth LC25= 100 Schultz and Riggin (1985) Hyalella Azteca Fish 10 d survival, growth LC25=60 Schultz and Riggin (1985) O. mykiss Fish 96 d Hatching, survival 71 3 Kopperman et al.(1974) O. latipes Fish 14 d hatchability IC25=290 Carlson and Caple (1977) Gambusia affinis Fish 35 d sperm count 101.3 Orvos et al. (2002) Danio rerio Amphibian 9 d hatchability IC25=160 Carlson and Caple (1977) Xenopus laevis Amphibian 21 d metamorphosis No effect Ishibashi et al.(2004) Rana catesneiana Amphibian 18 d development 300 Palenske et al.(2010) Rana pipiens Amphibian 24 d survival, growth 230 2.3 Yang et al.(2008)
208 Table 9 3. Continued Species Trophic group End point Result (LC 50 ,mg L 1 ) Source S. capricornutum Algae 96 h growth EC50 =4.46 EC25=2.44 Kopperman et al. (1974) S. subspicatus Algae 96 h biomass, growth EC50=1.2 0.5 Kopperman et al. (1974) S. costatum Algae 96 h growth EC50>66 Kopperman et al. (1974) flos aquae Algae 96 h biomass EC50=0.97 Kopperman et al. (1974) P. subcapitata Algae 72 h growth EC25=3.4 Ura et al. (2002) N. pelliculosa Algae 96 h growth EC50=19.1 Kopperman et al.(1974) Natural algal assemblage Algae 96 h biomass 0.12 Marchini et al.(1992) Closterium ehrenbergii Algae 96 h growth NOEC=250 Canton et al. (1985) Dunaliella tetriolecta Algae 96 h growth NOEC=250 Calamari et al. (1982) L. gibba Plant 7 d growth EC50> 62.5 Kopperman et al. (1974) S. herbacea Plant 28 d seed germination, morphology 100 100 Abernathy et al.(1986) E. prostrata Plant 28 d seed germination, morphology No effect 1000 Abernathy et al.(1986) B.frondosa Plant 28 d seed germination, morphology 100 10 Abernathy et al.(1986)
209 Table 9 4. Redefined human and ecological exposure pathways for land applied biosolids Pathway Description of HEI 1.Biosolids soil plant Plant toxicity due to uptake of biosolids borne TCS when grown in biosolids amended soils 2. Biosolids soil human Human and child ingesting biosolids 3. Biosolids soil plant human Human ( home gardener ) lifetime ingestion of plants grown in biosolids amended soil 4. Biosolids soil plant animal Animal lifetime ingestion of plants grown on biosolids amended soil 5. Biosolids soil plant animal human Human lifetime ingestion of animal products (animals raised on forage grown on biosolids amended soil) 6. Biosolids soil animal Animal lifetime ingestion of biosolids 7. Biosolids soil animal human Human lifetime ingestion of animal products (animals ingest biosolids directly) 8. Biosolids soil soil organism Soil organism ingesting biosolids/soil mixture 9. Biosolids soil soi l organism predator Predator of soil organisms that have been exposed to biosolids amended soils
210 Table 9 5. Data utilized for the calculation of RfD for humans and animals Study type Species Endpoint Exposure Result ( mg kg 1 day 1 ) References Subchronic toxicity Rat Hamster NOAEL Dietary feeding exposure for 90 d Dietary feeding for 13 weeks 10 0 5 0 75 SCF ( 2000 ) Ciba ( 2000 ) Ciba ( 2000 ) Chronic/carcinogenicity Rat Hamster NOAEL NOAEL Dietary feeding for 2 years Dietary feeding for 90 95 weeks 52 75 Borzelleca ( 1992 ) Ciba ( 2000 ) Reproductive or developmental Rat Mouse NOAEL NOAEL Exposure by gastric intubations and effects up to 2 generations 150 (reproductive performance) 50 (offspring) 25 (maternal development) Borzelleca ( 1992 ) Ciba ( 2000 ) Borzell eca ( 1992 ) Repeat toxicity Mouse NOAEL 13 week Changes in liver weight 25 NICNAS( 2009 ) Repeat toxicity Rat NOAEL Oral feeding (2 years) Clinic al chemistry and liver changes 40 (male) 56 (female) NICNAS ( 2009 )
211 Table 9 6. Various parameters used for conducting the preliminary risk estimation Abbreviation Parameter definition Value Assumptions/Explanation Pathway Reference BAF Chemical concentration in an organism divided by the concentration in an environmental medium, when the concentrations are near steady state, and multip le uptake routes contribute (Suter, 2007) BAF worm = 10 (worm, d.w.) BAF p = 0.93 (plant, d. w.) Conservative BAF values obtained from our study Conservative BAF value assuming a worst case 9 3, 4, 5 Chapter 5 Chapter 7 BAR Biosolids application rate 50 Mg ha 1 (d.w.) 5 Mg ha 1 y 1 (d.w.) x 1 0 0 y One time application rate Application rate; applied annually for 10 0 years 1, 2, 3, 4, 5, 6, 7, 8, 9 NRC ( 2002 ) BW Body weight (live weight) Adult: 70 kg Child: 16 kg Cow: 590 kg American woodcock: 0.181 kg Short tailed shrew: 0.016 kg Herring gull: 1.09 kg Mean 2, 4, 5, 6, 7, 9 USEPA ( 1997 ) U SEPA ( 1993 c ) USEPA (1993c) USEPA (1993c)
212 Table 9 6. Continued Abbreviation Parameter definition Value Assumptions/Explanation Pathway Reference CA TCS concentration in consumed animal either eating plant or soil Concentrations vary with the animal species CA m (animal meat) C S FS* FI/BW CA* FF 4, 5 6, 7 Snyder (2009) CB T CS concentration in biosolids 62 mg kg 1 95 th percentile concentration in biosolids in TNSSS 1, 2, 3, 4, 5, 6, 7, 8, 9, USEPA ( 2009a ) CS Predicted c oncentration in soils Worst case:1589 g kg 1 100 y ear: 15897 g kg 1 1, 2, 3, 4 5, 6, 7, 8, 9 DT Depth 15 cm Depth of biosolids incorporation 1, 2, 3, 4 5, 6, 7, 8, 9 D S or Rho D ensity 13 00 kg m 3 Bulk density of soils Brady and Weil (2002)
213 Table 9 6. Continued Abbreviation Parameter definition Value Assumptions/ Explanation Pathway Reference C W TCS concentration in earthworm s Worst case: 15897 gkg 1 100 year: 158974 g kg 1 PEC soil BAF worm Earthworms acquire chemical via ingestion of the soil 8, 9 Chapter 5 FF Fat fraction in meat Beef: 0.10 Pork: 0.090 Poultry: 0.060 Mean 5, 7 USEPA, 1997 FI Food ingestion rate Cow: 9100 g d 1 (d.w.) American woodcock: 1 3 9 g d 1 (d.w.) Short tailed shrew : 13 g d 1 Herring gull : 213 g d 1 4, 5, 6, 7, 9 Nagy (1987) in Suter (2007) In Reiss et al. (2009) FS Soil fraction of diet Cow: 0.025 Value used in Part 503 biosolids risk assessment 6 USEPA (1995 ) FVC Combined fruit and vegetable consumption 7.7 g kg 1 d 1 (w.w.) Mean 3 USEPA (1997)
214 Table 9 6. Continued Abbreviation Parameter definition Value Assumptions/Explanation Pathway Reference HFS Hectare furrow slice mass 2.2 x 10 6 kg Soil bulk density = 1.3 g cm 3 1, 2, 3, 4, 5, 6, 7, 8, 9, Brady and Weil, 2002 MC Meat consumption Adult Beef: 90 g d 1 Pork: 27 g d 1 Poultry: 67 g d 1 Child Beef: 1.8 g kg 1 d 1 Pork: 0.84 g kg 1 d 1 Poultry: 1.5 g kg 1 d 1 Mean Mean 5, 7 USEPA, 1997 USEPA, 1997 LD 50 Lethal dose of TCS that kills 50% of the organisms Avian toxicity 862 mg kg 1 d 1 Short tailed shrew: NOAEL: 75 mg kg 1 d 1 The organisms feed on earthworms as their primary diet 9 In Reiss et al. (2009) PC Concentration in the plant tissue Worst case: 1478 g kg 1 100 year: 14784 g kg 1 CS BAF p 3, 4 5 Chapter 7 PEC PEC predators Predator TCS concentrations vary with the species Calculated for two biosolids application scenarios 9 Chapter 5
215 Table 9 6. Continued. Abbreviation Parameter definition Value Assumptions/Explanation Pathway Reference RfD Reference dose : daily dose of chemical that appears to be without appreciable risk during an entire life time. Human: 0.40 mg kg 1 bw d 1 Animal: 1 mg kg 1 bw d 1 Predators: 2.87 mg kg 1 bw d 1 Calculated using toxicity end points obtained from literature 1, 2, 3, 4, 5, 6, 7, 8, 9 Chapter 9 SI Soil ingestion Adult: 0.05 g d 1 Child: 0.2 g d 1 Pica child: 10 g d 1 Mean 2 USEPA (1997) WD Percent of worm diet Hearing gull : 7.7 American woodcock: 78 Short tailed shrew: 31.4 9 Reiss et al. (2009)
216 Table 9 7. Equations used to calculate screening level hazard indices (HI ) (considering no TCS degradation) Pathway Hazard Index Equations Worst case Hazard Index 100 year Hazard Index Comments/ Assumptions 1. Biosolids soil plant C S / RfD CF 0.15 9 1. 59 Direct TCS effects on plant growth 2. Biosolids soil human ( C S SI / BW) / RfD CF Adult: 2.6*10 6 Child: 4.5*10 5 Pica child: 2.2*10 3 Adult: 2.6*10 5 Child: 4.5*10 4 Pica child: 2.2*10 2 3 Biosolids soil plant human (PC FVC) / RfD CF Adult: 0.026 Adult: 0.26 All produce consumed grown in biosolids amended soil 4. Biosolids soil plant animal (PC*FI/BW)/RfD*CF Cow: 0.023 Cow: 0.23 100% of diet consists of plants growing biosolids amended soil 5. Biosolids soil plant animal human (CA m MC / BW) / RfD CF Beef: 6.7*10 6 Pork: 1.8*10 6 Poultry: 2.9*10 6 Beef: 6.7*10 5 Pork: 1.8*10 5 Poultry: 2.9*10 5 Individual HI values calculated f or each meat product consumed 100% of diet consists of plants grown in biosolids amended soil
217 Table 9 7. Continued Pathway Hazard Index Equations Worst case 100 years Comments/ Assumptions 6 Biosolids soil animal ( C S FS FI / BW) / RfD CF Cow: 6.1*10 4 Cow: 6.1*10 3 7 Biosolids soil animal human (CA m MC / BW) / RfD CF Beef: 1.8*10 7 Pork: 4.8 *10 8 Poultry : 8 *10 8 Beef: 1.8*10 6 Pork: 4.8 *10 7 Poultry: 8 *10 7 HI values calculated assuming consumption of each meat product 8 Biosolids soil soil organism SC / RfD CF Eisenia foetida: 0.16 Eisenia foetida: 1.59 Based on the most sensitive LC 50 Lifetime spent in biosolids amended soil 9 Biosolids soil soil organism predator (FI* WD *CS* BAF worm / BW) / (RfD CF) Herring gull : 0.08 American woodcock : 3.3 Short tailed shrew : 1.4 Herring gull: 0. 83 American woodcock: 33 Short tailed shrew: 14 Earthwom diet by the birds is variable among species
218 Figure 9 1. Log of predicted TCS concentrations (mg kg 1 ) assuming no TCS loss and expected TCS concentrations (considering degradation) varying with time (years) of annual biosolids applications. 0 0 1 10 100 0 20 40 60 80 100 120 log of amended soil TCS concentration (mg kg 1 ) Years of biosolids application Expected considering degradation No degradation
219 SUMMARY AND CONCLUSIONS Summary of Intermediate Objective Results Objective 1: Quantify TCS Concentration in B iosolids The measured TCS concentration s in 15 biosolids were 0.40 to 40 mg kg 1 with an average of 18 12 mg kg 1 and a median concentration of 21 mg kg 1 The average and median concentration s were consistent with the mean value (16 65 mg kg 1 ) reported in TNSSS (USEPA, 2009a) and majority of rece ntly published studies. B iosolids utilized in our study were anaerobically digested, but differences in TCS concentrations occu red likely due to differenc es in digestion periods (time), inputs, or dewater ing methods (air dried vs cake). Triclosan c oncentra tions may also differ between WWTPs serving residential and industrial communities. A typical representative range of biosolids TCS concentration appears to be ~10 to 20 mg kg 1 Objective 2: Determine/Verify Basic Physico C hemical P ropert ies of TCS degradation rate in soils. Measured TCS water solubilities were 9 mg L 1 (pH = 6.14), 27 mg L 1 (pH = pK a = 8.14) and 800 mg L 1 (pH = 10.14). The overall 90 fold increas e in solubility when pH exceeded the pK a resulted from the dissociation of the neutral acid to anionic species. Increased solubility at higher pH portends greater concentration in the aqueous phase during sewage treatment processes utilizing lime stabiliza tio n and in high pH soils. H igh pH levels are not common in most sewage treatment systems or in most soils, except in some sodic soils. Solubility of TCS measured at pH 6.14 (i.e., 9 mg L 1 ) appears reasonable for predicting the fate and transport of TCS i n many soils. Partitioning coefficient (K ow ) values determined by using models and published measu rements were in close agreement, and the value (log K ow = 4.8) changed
220 negligibly when pH exceeded the p K a A log K ow of 4.8 was considered representative at pH values typically found in most sewage treatment systems and in amended soils. Mobility of TCS in biosolids amended soil determines the potential for soil and groundwater contamination. The low water solubility (9 mg L 1 ) an d relatively high log K ow (4.8) of TCS suggests extensive retention in (or on) biosolids and limited transport in soil systems. The mean (n=7) inherent log K d values for biosolids was 4.15 0.03 and log K oc was 4.68 0.07. The m ean (n=18) spiked log K d v alue for biosolids was 3.76 0.04 and log K oc value was 4.30 0.03. Based on the limited number of biosolids (n=7) analyzed for inherent K d and K oc the differences between inherent and spiked coefficients were not statistically significant, but the inherent log K d and K oc values tend to be greater than the spiked values. The log K d v alues determined for biosolids, soils a nd biosolids amended soils were variable Following normalization to organic carbon, the coefficients (K oc ) determined in all matrices were not significantly different and averaged 4.26 0.31 (Agyin Birikorang et al., 2010). Thus, a specific or narrow range of TCS partitioning coefficient s (K oc ) can serve as a first approximation to describe the behavior of TCS in soils or o ther matrices. The high log K oc suggest s preferential partition ing of TCS into organic material and thus, the mobility of biosolids borne TCS is expected to be low, with minimal risk of leaching. Objective 3: Determine the Degradation (Persistence) of Biosolids Borne TCS Biodegradation of TCS was studied in biosolids amended sand (IFS) and silty clay loam (ASL) soil s. Mineralization of 14 C TCS to 14 CO 2 was minima l (<0.5%) in both soils through 18 week in both biotic and inhibited biotic treatments B io solids amended soil TCS concentration s ( up to 0.40 mg kg 1 ) did not adversely affect CO 2 evolution at any time Methyl TCS was a major metabolite of TCS in the biotic treatment in both
221 soils. Absence of Me TCS in inhibited treatment suggested that Me TCS i s formed by a biomethylation reaction Biosolids borne TCS is persistent with a time for 50% disappearance of 77 to >126 d depending on soil texture Triclosan form ed bound (non extractable) residues of expected limite d lability with time. D egradation data were inconsistent with the TCS lability expectation s according to our operationally defined extraction scheme Triclosan degraded faster in silty clay loam soil (than the sand soil) but the silty clay loam has a smaller labile pool. So, t he operationally defined scheme imperfectly distinguishes the labile and b ound pools, and /or some of the compound in bound fraction may also be bioavailable/biodegradable. Further investigation is needed of the lability inferred extraction schemes and the relationship to other measures of lability (degrada tion and bioaccumulation data). Triclosan d egradation is typically slower in biosolids amended soils (laboratory study) than in un amended soils but faster than in field soils. W e accept our hypothesis that biosolids bor ne TCS is persistent in the environment and forms bound residues; however, the fate of metabolite (Me TCS) is not known. Objective 4: Determine the I mpact s of Biosolids Borne TCS to Soil Organisms Earthworm survival in the silty clay loam ( ASL ) soil and th e artificial soil was not adversely affected at 1 biosolids (equivalent to a maximum amended soil concentration of 100 mg kg 1 ) suggesting an LC 50 of ~10,005 mg kg 1 biosolids. The estimated LC 50 value in the IFS soil is ~ 105 mg TCS kg 1 biosolids. T he application of biosolids at 22 Mg ha 1 followed by incorporation at 15 cm depth (~100 fold dilution of the TCS concentration), results in LC 50 of ~1 mg TCS kg 1 soil in the IFS soil and 100 mg TCS kg 1 soil in the ASL and art ificial soils.
222 Earthworms grown in the silty clay loam soil (ASL) accum ulated more TCS than in the sand (IFS) soil. The average BAF value s in the two soils, irrespective of t he spiked TCS concentration, were 6.5 0.84 for the IFS soil and 12 3.08 for t he ASL so il. The average values were significantly different (p<0.05) from each other and likely reflected variable physico chemical soil properties. The soils differed in native soil organic carbon (OC) contents (11 g kg 1 for IFS soil and ~34 g kg 1 for ASL soil ), suggesting more accumulation in soil with greater OC content. The average TCS concentration measured in the earthworm s collected in a field equilibrated soil was 4.3 1.9 mg kg 1 corresponding to a BAF value of 4.35 0.7 Greater BAF val ues occurred where TCS was spiked into biosolids (laboratory conditions) as opposed to inherent TCS in field soils. The spiked log K d values are less than the inherent log K d value s suggesting more availability in a spiked system. B iosolids borne TCS accum ulated in, but was not toxic to, earthworms. We accept the hypothesis that biosolids borme TCS is not toxic to earthworms but acknowledge that earthworms may accumulate TCS. E arthworm accumulation of TCS varied significantly between laboratory and field co nditions, but the accumulation did not vary with TCS concentration in biosolids. Earthworm TCS bioaccumulation would be similar irrespective of the TCS concentration in soils where the worm is grown. Objective 5: Determine the Toxicity of Biosolids Borne T CS on Microbial Reactions The TCS concentration of 505 mg kg 1 biosolids ( 5.05 mg kg 1 soils ) was regarded as NOAEL for the microbial commun ity toxicity test in both soils Biolog plate analyses suggested that micro organism exposed to TCS do not differ i n substrate utilization and t hus, community structure effects were not anticipated. B iosolids borne TCS has no
223 adverse effect on soil micro organisms, using microbially mediated processes, community structure, and bacterial counts as indicators and our h ypothesis of no adverse TCS effects was accepted Assuming a representative biosolids TCS concentration of 16 mg kg 1 biosolids applied at 22 Mg ha 1 y ea r 1 would have to be applied for at least 30 years (assuming no loss of compound and no decrease in bi oavailability) to expect possible adverse effects on the processes o f ammonification, nitrification, or microbial respiration. Objective 6 and 7 : Quantify the Phytoavailability and Leaching Potential of Biosolids B orne TCS Plant toxicity and bioaccumulation tests were performed using three plant species including rad ish, lettuce, and bahia grass. U sing fresh biomass yields as the criteria, an amended soil concentration of 9150 ng g 1 was regarded as the LOEC for radish root growth. For lettuce 4570 ng g 1 was the LOEC, and for bahia grass the NOEC is at least 9150 ng g 1 Based on dry biomass yields, NOEC for all the plants is at least 9150 ng g 1 Food chain bioaccumulation begins with chemical uptake by plant s. Pla nt bioaccumulation test s su ggested minimal accumulation in radish (BAF = 0.004) and b ahia grass leaves (BAF = <0.001), but some accumulation in lettuce leaves (BAF = 0.04) and more in radish roots (BAF = 0.43 ). Dicots (radish, lettuce) accumulated more TCS than the monots (bahia grass) The BAF values obtained in our study suggest some accumulation in the below ground biomass (roots), and translocation to the above ground biomass (leaves), but accumulation was gre ater in the root than in the leaves. Th us, a diet of tuber ( root) plants is likely to pose greater risk of TCS to animals and humans than a diet of leafy plants (lettuce).
224 As expected from the measured K d and K oc values, the mobility of TCS was minimal in leached soil columns amended with 14 C TCS spiked biosolids. T he leaching was less in a biosolids amended soi l than a soil without biosolids, as TCS association with the OC of the biosolids retards movement. In amended soil, both TCS and Me TCS remained within the depth of biosolids incorporat ion (0 2.5 cm), but in controls, surface applied TCS ( and Me TCS ) moved to a depth of 2.5 cm. Thus, a relatively long time will be required for both chemical s to move through a soil profile and reach groundwater. Estimated time for 50% TCS degradation in the leaching study was 7 6 d, consistent with 50% disappearance time in the incubation study (77 to >126d). We accept our hypothesis that biosolids borne TCS has minimal mobility and phytoavailability. Ultimate Objective: Risk Assessment of Biosolids Borne TCS A s creening level risk assessment suggested minimal risk to human, terrestrial or aquatic populat ion s (HI<1) in most of pathways, except in pathway s 1, 8 and 9 (HI>1) A t ier 2 assessment considering TCS loss through degradation reduced the HI value for all critical pathway s to less than one suggesting minimal r isk to even sensitive organisms like American woodcock, herring gull, short tailed shrew, earthworms and plants We accept our overall hypothesis of minimal risk of biosolids borne TCS to human and environmental heal th B iosolids containing a 95 th percentile TCS concentration (62 mg kg 1 ) applied at agronomic rates (5 Mg ha 1 ) for multiple years (up to 100 years) did not adverse ly affect the majority of the organisms (according to tier 2 assessment) included in the exposure pathways described herein. American woodcock was identified as the most sensit ive species to TCS exposure and, thus, pathway 9 was utilized for the calculation of preliminary pollutant limits The estimated CPLR was 5.8 kg TCS ha 1 The re sulting pollutant concentration limit was 6.3
225 mg TCS kg 1 biosolids which portends that if biosolids contain a TCS concentration <6.3 mg kg 1 biosolids land application may not have restrictions regarding TCS risk. However, our pollutant limits are preli minary estimates not intended to make changes to the biosolids land application regulations. Rather, we calculate d the limits to guide future TCS work. We utilized th e measured data (collected from our study ) wherever possible for risk estimation, but the assessment f or some organisms (e.g. predators) was limited by t he lack of reliable data sources Snyder (2009) identified the predator pathway (pathway 9) as the most critical pathway in the TCC risk assessment, and sam e pathway was critical for TCS. O ur e stimates are preliminary, as the toxicity end point in the most critical exposure pathway (pathway 9) was de rived from unpublished source s and the risk was assessed based on our best scientific judgement (use of uncertainty factors) The critical exposure pathways (HI>1) identified here in need s re ass essment based on reliable toxicity endpoints Additionally, the TCS pollutant limits were established assuming no loss of TCS through degradation, but TCS disappears with a relativel y sh ort half life (100d) and should be considered when estimating the pollutant limits. As Me TCS was deemed as a major metabolite, risk assessment of biosolids borne TCS should consider the contribution of Me TCS. Future Studies Appearance of metabo lite (Me T CS) raises some concern, as limited information is available about Me TCS. Leaching behavior of Me TCS addressed in our leaching study was similar to TCS However, basic physicochemical properties of Me TCS (solubility, K ow K d ), biodegradation, toxicity, plant accumulation are essentially unknown. A recent study detected Me TCS in soils amended with biosolids borne TCS (Lozano, personal communication, 2011), and f uture studies should address the fate
226 and transport of Me TCS in biosolids amended s oils. Furt her, the behavior of bound fract ion of TCS is still not certain. Our short term (18 week ) incubation study suggested formation of bound resiues but long term studies are needed to explore possible change s in bioavaila bility of TCS bound fraction with time. Data gaps also remain regarding the potential development of antibiotic resistance and endocrine disruption effects of TCS Conflicting data repo rts the ability of TCS to cause endocrine disruption (Veldhoen et al., 2006; Fort et al., 2010) in aquatic environments. Veldho en et al. (2006) suggested endocrine disruption effects at TCS concentrations (0.15 1.4 g L 1 ), as opposed to no adverse effects observed in Fort et al. (2010) study. Improved r isk assessment s also require more reliable toxicity end po ints for terrestrial species. Other research might include TCS leachability studies to explore the effect of soils, biosolids, application methods and grou n d cover on the TCS movement. Aquatic risk assessments only consider TCS concentrations in the aqueou s phase but given the high partitioning coefficients of TCS, it is expected to sorb to dissolve d organic carbon (DOC) and reduce the TCS availability to aquatic organisms. Characterization of bioavailability of DOC associated TCS could improve the risk ass essments of runoff and tile drian age waters entering the surface waters and affecting aquatic environments
227 APPENDIX A EXPLANATION OF THE SEQUENTIAL EXTRACTION SCHEME Labile is defined as the fraction of the chemical that is readily transformed by micro organisms or readily available to plants (https://www.soils.org /publications/soils glossary#). Mild extraction methods are commonly used to mimic the chemical uptake b y soil organisms and plants and, thereby, estimate the labile fraction of the chemical. Common extraction agents used to characterize lability of compounds similar to TCS (chlorophenols) are a mixture of water and MeOH (Yu et al., 2005; Hu et al., 2005). Wa ter extractions alone do not always correlate well with bioavailability (Hickman and Reid, 2005). In particular, the bioavailability of hydrophobic chemicals can be underestimated by water extractions because low water solubility limits the amount of chemi cal extracted (Semple et al., 2003). Combining the amount of chemical extracted by water and MeOH is one approach to circumvent the problem with chemicals characterized by low water solubility. The humic fractions with expected low lability can be successf ully extracted by a strong base like NaOH (Shirshova et al., 2006). The residual activity left in the dried soil after the sequential extraction s is expected to represent t he bound or non labile fraction. However, chemicals considered bound are sometimes s orbed to the OM, Fe and Al oxides by either weak electrostatic forces or covalent bonds. The TCS sorbed by weak electrostatic forces is expected to be ext racted by a mixture of MeOH+ acetone (50:50 v/v), as the two solvents differ in polarity and should ext ract the loosely sorbed TCS. Indeed, t he solvent mixture is utilized to determine total TCS and triclocarban concentrations in biosolids (Snyder et al., 2010 ; Heidler et al., 2006) and is expected to be a more rigorous extractant than MeOH alone. The 14 C b ound by covalent bonds was considered bound to the soil. The
228 sequential extraction scheme is proposed to extract labile (water, MeOH) and the non labile TCS in the soils. The non labile fractions will include: humic associated, (NaOH ) loosely sorbed (MeOH+ acetone), and bound (combustible) fractions of TCS.
229 APPENDIX B SUPPLEMENATAL DATA FOR CHAPTER 7 Table B 1. Average b ioaccumulation f actors (BAF) in the radish and lettuce leaves after excludin g the highest treatment (Trt 3) Plant Average BAF including all treatments BAF excluding Trt 3 Radish roots 0.43 0.26 Lettuce leaves 0.04 0.02 Figure B 1 Comparison of bioa ccumulation factors in radish root (below ground) and lettuce leaves (above ground) grown in a biosolids amended soil with a range of TCS concentrations (same letters represent no statistical difference among treatments) 0 0.2 0.4 0.6 0.8 1 Radish Lettuce BAF (dry wt.) Crops Control 990 ng/g (Trt 1) 5990 ng/g (Trt 2) 10,900 ng/g (Trt 3) a a b a b c
230 APPENDIX C LIMITS OF DETECTION, QUANTITATION AND RECOVERIES Tabl e C 1. Limits of detection, quantitation and percent recoveries for TCS and Me TCS in various matrices. Matrix Chemical % recovery S. E LOD (ng g 1 ) LOQ (ng g 1 ) Instrument Biosolids TCS 64 5.0 0.40 1.3 LC/MS Soil TCS Me TCS 95 7.6 90 8.6 0.28 1.0 GC/MS Earthworm TCS Me TCS 93 5.0 91 4.5 0.22 0.7 GC/MS Plants TCS Me TCS 90 11 89 6.9 0.28 1.0 GC/MS
231 WORKS CITED Abernathy, S., A. M. Bobra, W.Y. Shiu, P.G. Well and D. Mackay. 1986. Acute lethal toxicity of hydrocarbons and chlorinated hydrocarbons to two planktonic crustaceans: the key role of organism water partitioning. Aquat. Toxicol. 8 : 163 174. Adolfsson Erici, M., M. Pettersson, J. Parkkonen, and J. Sturve. 2002. Triclosan, a commonly used bactericide found in human milk and in the aquatic environment in Sweden. Chemosphere 46:1485 1489. Aga, D. 2009. Fate of pharmaceuticals in the environment and in water treatment systems. CRC press. Boca Raton, FL Agyin Birikorang, S., M. Miller., TCC and TCS in soils and sediments. Env iron. Toxicol. Chem. 29: 1925 1933. Alexander, M. 1995. How toxic are toxic chemicals in s oil? Environ. Sci. Technol. 29: 2713 2717. Alexander, M. 1999. Biodegradation and Bioremediation. Academic Press, San Diego, California. Alexander, M. 2000. Aging, bioavailability, and overestimation of risk from environmental pollutants. Env iron. Sci. Technol. 34: 4259 4265. Allmyr, M., M. Adolfsson Erici, M. S. M cLachlan, and G. Sandborgh Englu nd. 2006. Triclosan in plasma and milk from Swedish nursin g mothers and their exposure via personal care products. Sci. Total Environ. 372:87 93. Al Rajab, A. J., L. Sabourin, A. Scott, D. R. Lapen, and E. Topp. 2009. Impact of biosolids on the persistence and dissipation pathways of triclosan and triclocarban in an agricultural soil. Sci. Total Environ. 407:5978 5985. Anderson, J. P. 1982. Soil respiration, in Page, A.L., Miller, and R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis, Part 2 Chemical and Microbiological Properties, Second Edition : 842 842. Armi tage, J.M. 2004. Development and evaluation of a terrestrial food web bioaccumulation model. Masters Thesis, Simon Fraser University, Burnaby, BC. Balmer, M. E., T. Poiger, C. Droz, K. Romanin, P. Bergqvist, M.D. Muller, and H. Buser. 2004. Occurrence of m ethyl triclosan, a transformation product of the bactericide triclosan, in fish from various lakes in Switzerl and. Environ. Sci. Technol. 38: 390 395.
232 Banks, M. K., A. P. Schwab, N. Cofield, J. E. Alleman, M. Switzenbaum, J. Shalabi, and P. Williams. 2006 Biosolids amended soils: Part I. Effect of biosolids application on soil quality and ecotoxicity. Water Environ. Res. 78:2217 2230. Barbarick, K. A., K. G. Doxtader, E. F. Redente, and R. B. Brobst. 2004. Biosolids effects on microbial activity in shrub land and grassland soils. Soil Sci. 169:176 187. Barbolt, T. A. 2002. Chemistry and safety of triclosan, and its use as an antimicrobial coating on VICRYL plus bacterial suture (Coated polyglactan 910 suture with triclosan). Surg. Infect. 3:S45 S53. Beno tti, M. J., R. A. Trenholm, B. J. Vanderford, J. E. Holady, B. D. Stanford, and S. A. Snyder. 2009. Phamaceuticals and endocrine disrupting compounds in US drinking water. Environ. Sci. Technol. 43:597 603. Bester, K. 2003. Triclosan in a sewage treatment process balances and monitoring data. Water Res. 37:3891 3896. Bester, K. 2005. Fate of triclosan and triclosan methyl in sewage treatment plants and surface waters Arch. Enviro n. Contam. Toxicol. 49:9 18 Binelli, A., D. Cogni, M. Parolini, C. Riva, and A. Provini. 2009. Cytotoxic and genotoxic effects of in vitro exposure to triclosan and trimethoprim on zebra mussel (Dreissena polymorpha) hemocytes. Comp. Biochem. Physiol. C. Toxi col. Pharmacol. 50:50 56. Birosova, L., and M. Mikulasova. 2009. Developme nt of triclosan and antibiotic resistance in Salmonella enterica serovar typh imurium. J. Med. Microbiol. 58: 436 441. Boehmer, W., H. Ruedel, A. Wenzel and C. Schroeter Kermani. 2004. Retrospective monitoring of triclosan and methyl triclosan in fish: resul ts from the German environmental specimen bank. Organohalogen Compd. 66: 1516 1521. Bouche, M. B., and R. H. Gardner. 1984. Earthworm functions VIII, Population 63 Boxall, A.B., P. Johnson, E J. Smith, C. J. Sinclair, E. Stutt, and L. S. Levy. 2006. Uptake of veterinary medicines from soils into plants. J. Agric. Food Chem. 54:2288 2297. Brady, N. C., and R. R. Weil. 2002. The nature and properties of soils. Prentice hall, NJ. Brausch, J. M ., and G. M. Rand. 2010. A review of peron al care products in the aquatic environment: Environmental concentrations and toxicity. Chemosphere 82:1518 1532.
233 Bremner, J. M. 1996. Nitrogen total. P. 1238 1255. In C. A. Black et al. (ed.) Methods of soil analy sis. Part 2. Agron. Monogr. 9. ASA and SSSA, Madison, WI. Butler, E., M. J. Whelan, K. Ritz, R. Sakrabani, and R. van Egmond. 2011. Effects of triclosan on soil microbial respiration. Environ. Toxicol. Chem. 30:360 366. Calafat, A. M., X. Ye, L. Wong, J. A. Reidy, and L. L. Needham. 2008. Urinary concentrations of triclosan in the U.S. population: 2003 20 04. Environ. Health Persp. 116: 303 307. Calamari, D., S. Galassi, F. Setti and M. Vighi. 1983. Toxicity of selected chl oro benzenes to aquatic organisms, Chemosphere 12 : 253 262. Canosa, P ., S. Morales, I. Rodriquez, E. Rubi, R. Cela, and M. Gomez. 2005. Aquatic degradation of triclosan and formation of toxic chlorophenols in presence of low concentrations of free chlorine. Anal. Bioanal Chem. 383:1119 26. Canton J.H, W. Sloof, H.J. Kool, J. Struys, T.J.M. Po uw, R.C.C. Wegman and G.J. Piet 1985. Toxicity, biodegradability, and accumulation of a number of chlorine/nitrogen containing compounds for classification and estab lishing water quality criteria. Regul. Toxicol. Ph arm 5 : 123 131. Capdevielle, M ., R. Van Egmond M. Whelan D. Versteeg M. Hofmann Kamensky J. Inauen, V. Cunningham and D Woltering 2008. Consideration of exposure and species sensitivity of triclosan in the freshwater environment. Integr Environ Assess Manag. 4:15 23. Carballa, M., F. Guido, O. Francisco, J. M. Lema, and T. Ternes. 2008 Determination of the solid water distribution coefficient (Kd) for pharmaceuticals, estrogens and musk fragrances in d igested sludge. Water Res. 42: 287 295. Carlson, R.M., and R. Caple. 1977. Chemical/Biological Implications of using Chlorine and Ozone f or Disinfection. EPA 600/3 77 066, Duluth, MN, 88p. Cha, J., and A.M. Cupples. 2009. Detection of the antimicrobials triclocarban and triclosan in agricultural soils following land appl ication of municipal biosolids. Water Res. 43 : 2522 2530. Cha, J., and A M. Cupples. 2010. T riclocarban and triclosan biodegradation at field concentrations and the resulting leaching potentials in three agric ultural soils. Chemosphere 81: 494 499 Christensen, K P. Triclosan Determination of aerobic biodegradation in soils. Springborn Laboratories, Inc., Wareham (MA): 1994; Prepared for Colgate Palmolive Company, Piscataway (NJ). SLI Report 93 5 4770. Available at http://www.fda.gov/ohrms/dockets/dockets/75n0183h/75n 0183h sup0013 11 Attachment 04 01 vol202.pdf A ccessed on Jan 20 2011.
234 Chu, S., and C.D. Metcalfe. 2007. Simultaneous determination of triclocarban and triclosan in municipal biosolids by liqu id chromatog raphy tandem mass spectrometry. J Chromatogr A 1164 : 212 218. Ciba Specialty Chemicals. 2001a. General information on chemical,physical and microbiological properties of Irgasan DP300, Igracare MP and Irgacide LP10. Brochure 2520. Publication Ag B2520e.02.2001. Basel Switzerland. Ciba Specialty Chemicals. 2001b. Irgasan DP 300, Irgacare MP.Toxicological and ecolological data. Official registrations. Technical Brochure#2521, Basel, Switzerland. Comfort, S. D., P. J. Shea, F. W. Roeth. 1994. Underst anding pesticides and water quality in Nebraska. Nebraska Cooperative Extension EC 94 135. University of Nebraska, Lincoln, NE 68583, p 16. Coogan, M. A., R. E. Edziyie, T. W. La Point, and B. J. Venables. 2007. Algal bioaccumulation of triclocarban, tricl osan, and methyl triclosan in a North Texas wastewater treatment plant receiving stream Chemosphere 10 : 1911 1918 Coogan, M. A., and T. W. La Point. 2008. Snail bioaccumulation of triclocarban, triclosan, and methyltriclosan in a North Texas, USA, stream affected by wastewater treatment plant runoff. Environ T oxicol. Chem. 27 :1788 17 93. Cory, A. H ., T. C. Owen, J. A. Barltrop, and J. G. Cory. 1991. Use of an aqueous soluble tetrazolium/formazan assay for cell growth assays in culture. Cancer Commun. 3:207 212. Dayan, A. D. 2007. Risk assessm ent of triclosan [Irgasan] in human breast milk. Food Chem. Toxicol. 45:125 129 Delgado, E.J. 2002. Predicting aqueous solubility of chlorinated hydrocarbons from molecular structure. Fluid Phase Equilibria 199:101 107 Dennis, G. L., and P. R. Fresquez 1989. The soil microbial community in sewage sludge amended semi arid grassland. Biol. Fertil. Soil. 7:310 317. Dimitrov, S. D., N. C. Dimitrova, J. D. Walker, G. D. Veith, and O.G. Mekenyan. 2003. Bioconcentration potential predictions based on molecula r attributes An early warning approach for chemicals found in humans, birds, fish and wildlife. QSAR Comb. Sci. 22:58 68. Dolliver, H., K. Kumar, and S. Gupta. 2007. Sulfamethazine uptake by plants from manure amended soil. J. Environ. Qual. 36:1224 1230. Duarte Davidson, R. and K.C. Jones. 1996. Screening of the environmental fate of organic contaminants in sewage sludge applied to agricultural soils: II. The potential for transfers to plants and grazing an imals. Sci. Total Environ. 185: 59 70.
235 Edwards, M. E. Topp, C. D. Metcalfe, H. Li, N. Gottschall, P. Bolton, W. Curnoe ,M. Payne, A. Beck, S. Kleywegt, and D. R. Lapen. 2009. Pharmaceutical and personal care products in tile drainage following surface spreading and injection of dewatered municipal biosol ids to an agricultural field. Sci Total Environ. 407:4220 4230 Environmental Protection and Heritage Council (EPHC). 2008. Australian Guidelines for Water Recycling: Managing Health and Environmental Risks (Phase 2): Augmentation of Drinking Water Suppli es. Adelaide, South Australia, Australian Health Ministers' Conference, Natural Resource Management Ministerial Council. Epstein, E., 2003. Land Application of Sewage Sludge and Biosolids. CRC Press, Boca Raton, Florida. Fair, P. A., H. B. Lee, J. Adams, C. Darling, G. Pacepavicius, M. Alaee, G. D. Bossart, N. Henry, and D. Muir. 2009. Occurrence of triclosan in plasma of wild Atlantic bottlenose dolphin (Tursiops truncates) and in their environment. Environ. Pollut. 157:2248 2254. Farkas, M. H ., E. R. Mojica, M.Patel, D. S. Aga, J. O. Berry. 2009. Development of a rapid biolistic assay to determine changes in relative levels of intracellular calcium in leaves following tetracycline uptake by pinto bean plants. Analyst 134:1594 1600. Fischer, A., C. Oehm M. Selle, and P. Werner 2005. Biotic and abiotic transformations of Methyl tertiary Butyl Ether (MTBE). Envir on. Sci. Pollut. Res. Int. 12:381 386 Fort, D. J R. L. Rogers, J. W. Gorsuch, L. T. Navarro, R. Peter, and J. R. Plautz. 2010. Triclosan and anuran metamorphosis: no effect on thyroid mediated metamorphosis in Xenopus laevis Toxicol Sci. 113:392 400. Franz, S., R. Altenburger, H. Heilmaeir and M. Schmidtt Jansen. 2008. What contrib utes to the sensitivity of microalgae to triclosan? Aquat. Toxicol. 90 :102 108. Fuchsman, P., J. Lyndall, M. Bock, D. Laur en, T. Barber, K. Leigh, E. Perruchon, and M. Capdevielle. 2010. Terrestrial ecological risk evaluation for tric losan in land applied biosolids. Int. Environ. Assess. Manag. 6:405 418. Garbeva, P., J. A. van Veen, and J. D. van Elsas. 2004. Microbial diversity in soil: selection of microbial populations by plant and soil type and implications for disease suppressiveness Ann. Rev. Phytopathol. 42:243 270. Garci a Gil, J. C., C. Plaza, N. Senesi, and G. Brunetti. 2004. Effects of sewage sludge amendment on humic acids and microbiological properties of a semi arid Mediterranean s oil. Biol. Fertil. Soil. 39:320 328
236 Garland J.L. 1997. Analysis and interpretation of community level physiological profiles in microbial eco logy. FEMS Microbiol. Ecol. 24: 289 300. Garland, J.L., and A. L. Mills. 1991 Classification and characterization of heterotrophic microbial communities on the basis of patterns of community level sole carbon source utilization Appl. Environ. Microbiol. 57: 2351 2359. Gibson, R., J. C. Duran Alvarez, K. L. Estrada, A. Chavez, and B. J. Cisneros. 2010. Accumulation and leaching potential of some pharmaceuticals and potential endocrine disruptors in soils irrigated with wastewater in the Tula Valley, Mexico. Chemosphere 81:1437 1445. Gilbert, P., and A.J. McBain. 2002. Literature bas ed evaluation of the potential risks of impregnation of medical devi ces and implants with triclosan. Sur g Infect. 3 :S55 S63. Gusta fson, D. I. 1989. Groundwater ubiquity score a simple method for as sessing pesticide leachability. Environ. Toxicol. Chem. 8:339 357. Haggblom, M. M., J. H. A. Apajalaht i, and M. S. Salkinoja Salonen. 1988. O Methylation of chlorinated para Hydroquinones by Rhodococcus chlorophenolicus. Applied Environ. Micro. 54: 1818 1824. Halden, R. U., and D. H. Paull. 2005. Co Occurrence of triclocarban and triclosan in U.S. water resources. Environ. Sci. Technol. 39:1420 1426. Hansen, L. H., B. Ferrari, A. H. Srensen, D. Veal, and S. J. Ssen. 2001. Detection of oxytetracyline production by Streptomyces rimosus in soil microcosoms by co mbining whole cell biosensors and flow cytometry. Ap pl. Environ. Microbiol. 67: 239 244. Hartnik, T., L. E. Sverdrup., and J. Jensen. 2008. Toxicity of the pesticide alpha cypermethrin to four soil nontarget invertebrates and implications for risk assessment. Environ. Toxicol. Chem. 27:1408 1415. Heath, R. J., and C. O. Rock. 2000. A triclosa n resistant bacter ial enzyme. Nature (Lond). 406: 145 146. Heidler, J., A. Sapkota, and R. Halden. 2006. Partitioning, persistence, and accumulation in digested sludge of the topical antiseptic Triclocarban during wastewater treatment. Environ. Sci. Technol 40:3634 3639. Heidler, J., and R. U. Halden. 2007. Mass balance of triclosan removal during conventional sewage treatment. Chemosphere 66:362 369. He idler, J., and R. U. Halden. 2009. Fate of organohalogens in US wastewater treatment plants and estimated chemical releases to soils nationwide from biosolid s recycling. J. Environ Monitor. 11: 2207 2215.
237 Heim, LG. 1997. Adsorption of 14 C triclosan to suspended solids. Report 43356. Ciba Specialty Chemical, Basel, Switzerland. Hendriks, A. J., W. C. Ma, J. J. B rouns, E. M. de Ruiter Dijkman, and R. Gast. 1995. Modelling and monitoring organochlorine and heavy metal accumulation in soils, earthworms, and shr ews in Rhine Delta floodplains. Archive s Environ. Contam. Toxicol. 29: 115 127. Herklotz, P. A., P. Gurung, H. Vanden Heuvel, and C. A. Kinney. 2010. Uptake of human ph a rmaceuticals by plants grown under hydroponic conditions. Chemosph e re 78:1416 1421. Hickman, Z. A., and B. J. Reid. 2005. Towards a more appropriate water based extraction for the assessment of organic contaminant availability. Environ. Pollut. 138:299 306. Higgins, C. P., Z. J. Paesani, T. E. Abbott Chalew, R. U. Halden, and L. S. Hundal 2011 Persistence of triclocarban and triclosan in soils after land application of biosolids and bioaccumulation in Eisenia foetida Environ. Toxicol. Chem. 30:556 563. Hoberg, J. R. 1992. Determination of effects on seedling growth of six plant species. Springborn laborat ories, Inc., Wareham (MA). Submitted to The Procter & Gamble Company, Cincinnati (OH). SLI report 90 12 3574. Holt, L M., A. E. Laursen., L.H. McCarthy, I. Vadim Bostan, and A. L. Spongber g 2010. Effects of land application of municipal biosolids on nitro gen fixing bacteria in agricultural s oil. Biol. Fertil. Soil. 46:407 41. Hu, X.Y., B. Wen, X. Q. Shan, and S. Z. Zhang. 2005. Bioavailability of pentachloro phenol to earthworms (Eisenia foetida) in artificially contaminated soils. J. Environ. Sci. Health A. 40:1905 1916. Hulster, A., J. F. Muller, and H. Marschner. 1994. Soil plant transfer of polychlorinated dibenzo p dioxins and dibenzofurans to vegetables of the cucumber family (Cucurbitac ea). Environ. Sci. Technol. 28: 1110 1115. Hundt, K., D. Martin, E. Ham mer, U. Jonas, M. K. Kindermann, and F. Schauer. 2000. Transformation of Triclosan by Trametes versicolor and Pycnoporus cinnabarinus. Applied Environ. Micro. 66:4157 4160. Ingerslev, F., and B. Halling Srensen. 2001. Biodegradability of metronidazo le, olaqiondox, and tylosin, and formation of tylosin degradation products in aerobic soil/man ure slurries. Environ. Saf. 48:
238 Ishibashi, H., N. Matsumura, M. Hirano, M. Matsuoka, H. Shiratsuchi, Y. Ishibashi, Y. Takao, and K. Arizono. 2004. Effects of triclosan on the early life stages and reproduction of medaka Oryzias Latipes and induction of hepatic vitellogenin. Aquat. Toxicol. 67:167 179. Jacobs, M. N., G. T. Nolan, and S. R. Hood. 2005. Lignan, bacteriocides and organochlorine compounds activa te the human pregnane X receptor (PXR). Toxicol. Appl. Pharmacol. 209:123 133. Jager, T. 1998. Mechanistic approach for estimating bioconcentration of organic chemicals in earthworms (Oligochaeta). Environ. Toxicol Chem. 17:2080 2090. Jakel, K. 1990. I rgasan DP 300: Report on dissociation constant in water. Ciba Speciality Chemical, Basel, Switzerland. James, M. O., W. Li, D. P. Summerlot, L. Rowland Faux, and C. E. Wood. 2010 Triclosan is a potent inhibitor of estradiol and estrone sulfonation i n shee p placenta. Environ. Int. 36:942 949. Karnjanapiboonwong, A., C. Qingsong, A.N. Morse, and T.A. Anderson. 2008. Sorption of estrogens, caffeine and triclosan to a sandy loam and a silt loam soil. SETAC North America 29th Annual Meeting. Tampa, FL. Nov. 16 20. Kinney, C. A., E. T. Furlong, D. W. Kolpin, M. R. Burkhardt, S. D. Zaugg, S. L. Werner, J. P. Bossio, and M. J. Benotti. 2008. Bioaccumulation of pharmaceuticals and other anthropogenic waste indicators in earthworms from agricultural soil amended with biosolids or swine manure. Environ. Sci. Technol. 42:1863 1870. Kolpin, D.W., E.T. Furlong, M.T. Meyer, E.M. Thurman, S.D. Zaugg, L.B. Barber, and H.T. Buxton. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 199 9 2000: A national reconnaissance. Environ. Sci. Technol. 36:1202 1211. Kopperman, H. L., R. M. Carlson, and R. Caple. 1974. Aqueous chlorination and ozonation studies. I. Structure toxicity corelations of phenolic compounds to Daphnia magna. Chem Biol. I nteract. 9:245 251. Kourtev, P. S., J. G. Ehrenfeld, and M. Haggblom. 2003. Experimental analysis of the effect of exotic and native plant species on the structure and function of soil microbial communities. Soil Biol. Biochem. 35:895 905. Kumar, K., S. C. Gupta, S. K. Baidoo, Y. Chander, and C. J. Rosen. 2005. Antibiotic uptake by plants from soil fertilized with animal manure. J. Environ. Qual. 34:2082 2085. Kumar, P.R. 2006. An experimental methodology for monitoring contaminant transport through geotechnical centrifuge models. Environ. Moni t. Assess. 117:215 233.
239 Kumar, K. 2010. A framework to predict uptake of pharmaceuticals and personal care products (PPCPs) by plants. SSSA International meetings, Long Beach, CA. Kwon, J. W., K. L. Arm brust, and K. Xia. 2010. Transformation of triclosan and triclocarban in soils and biosolids applied soils. J. Environ. Qual. 39: 1139 1144. Langdon, K. A., M. S. Warne, and R. S. Kookana. 2010 Aquatic hazard assessment for pharmaceuticals, personal care p roducts, and endocrine disrupting compounds from biosolids amended land. Int. Environ. Assess. Manag. 6:663 676. Langdon, K. A., M. S. Warne, R. J. Smernik, A. Shareef, and R. S. Kookana. 2011. Selected personal care products and endocrine disruptors in biosolids: An Australia wide survey Sci. Total Environ. 409:1075 1081. Lapen, D.R., E. Topp, C. D. Metcalfe, H. X. Li, M. J. Edwards, N. Gottschall, P. Bolton, W. E. Cu rnoe, M. Payne, and A. Beck. 2008. Pharmaceutical and personal care products in tile dr ainage following land application of municipal bios olids. Sci. Total Environ. 399: 50 65. Latch, D. E., J. Packer, B. Stender, J. Van Overbeke, W. Arnold, and K. McNeill. 2005. Aqueous photochemistry of triclosan: formation of 2,4 Dichlorophenol, 2,8 Dichlo rodibenzo p Dioxin, and oligomerization p roducts. Environ. Toxicol. Chem. 24:517 525. Lawlora, K ., B. P. Knighta, V. L. Barbosa Jeffersona P. W. Laneb A. K. Lilleyc G. I. Patond, S. P. McGratha S. M. O'Flahertya and P. R. H irscha. 2000. Comparison of methods to investigate microbial populations in soils under different agricultural management. FEMS Microbiol. Ecol 33:129 137. Lawrence, A. P., and M. A. Bowers. 2002. A test of the hot mustard extraction method of sampling earthworms. Soil Biol. Biochem. 34:549 552. Leewen, C. J. van., and T. G. Vermeire. 2007. Risk assessment of chemicals: An introduction. 2 nd edition, Springer, Netherlands. Lei, H., and S. A. Snyder. 2007. 3D QSPR models for the removal of trace organic contaminants by ozone and free chlorine. Water Res. 41:4051 4060. Leininger, S., T. Urich, M. Schloter, L. Schwark, J. Qi, G. W. Nicol, J. I. Prosser, S. C. Schuster and C. Schleper. 2006. Archaea predominates among ammonia oxidizing prokaryotes in soils. Nature 442 :806 809 Levy, C. W., A. Roujeinikova, S. Sedelnikova, P. J. Baker, A. R. Stuitje, A. R. Slabas, D. W. Rice, and J. B. Rafferty. 19 99. Molecular basis of triclos an activity. Nature (Lond) 398: 383 384.
240 Lin, D., Q. Zhou, X. Xie, and Y. Liu. 2010. Potential biochemical and genetic toxicity of triclosan as an emerging pollutant on earthworms (Eisenia fetida). Chemosphere 81:1328 1333. Li ndstrom, A., I. J. Buerge, T. Poiger, P. Bergqvist. M. D. Muller, and H. Buser. 2002. Occurrence and environmental behavior of the bactericide triclosan and its methyl derivative in surface waters and in wastewater. Environ. Sci. Technol. 36: 2322 2329. Liu, F., G. G. Ying L. H. Yang and Q. X. Zhou. 2009. Terrestrial ecotoxicological effects of the antimicrobial agent triclosan. Ecotoxicol. Environ Saf. 72:86 92. Loraine, G.A., and M.E Pettigrove. 2006. Seasonal variations in concentrations of p ha r maceuticals and personal c are products in drinking water and reclaimed wastewater i n So uthern California. Environ. Sci. Technol 40:687 95. Lozano, N., C. P. Rice, M. Ramirez, and A. Torrents. 2010. Fate of triclosan in agricultural soils after biosolid ap plications. Chemosphere 78: 760 766. Lu, S., and M. Archer. 2005. Fatty acid synthesis is a potential target for the chemoprevention of breast cancer. Carcinogenesis 26 :153 157. Lyman, W. 1990. Adsorption coefficient for soils and sediments. Pages 4 1 4 31 in W. J. Lyman, W. F. Reehl, and D. H. Rosenblatt (eds.), Handbook of chemical property estimation methods. American Chemical Society, Washington DC. Lyndall, J., P. Fuchsman, M. Bock, T. Barber, D. Lauren, K. Leigh, E. Perruchon, and M. Capdevielle. 2010. Probabilistic risk evaluation for triclosan in surface water, sediments and aquatic biota tissues. Int Environ. Assess. Manag. 6:419 440. Lyrge, H., G. Moe, R. Skalevik and H. Holmsen. 2003. Interaction of triclosan with eukaryotic membrane lipids. Eur. J. Oral. Sci. 111 :216 222. Mackay, D., and L. Barnthouse. 2010. Integrated risk assessment of household chemicals and consumer products: addressing concerns about triclosan. Int. Environ. Assess. Manag. 6:290 392. Mamo, M., S. C. Gupta, C. J. Rosen, and U. B. Singh. 2005. Phosphorus leaching at cold temperatures as affected by wastewater application and soil phosphorous levels. J. Environ. Qual. 34: 1243 1250. Mamy, L., and E. Barriuso E. 2007. Desorption and time dependent sorption of herbicides in soils Eur. J Soil Sci. 58:174 187. Marchini, S., M. L. Tosato, T. J. Norberg King, D. E. Hammermeister and M. D. Hoglund. 1992. Lethal and sublethal toxicity of benzene derivatives to the fathead m innow, using a short term test. Environ. Toxicol. Chem 11: 187 195.
241 Matsunaga, T., M. Okochi, and N. Satoshi. 1995. Direct count of bacteria using fluorescent dyes: application to assessment of electrochemical d isinfection Anal. Chem 67:4487 4490. McAvoy, D. C., B. Schatowitz, M. Jacob, A. Hauk, and W. S. Eckhoff. 20 02. Measurement of triclosan in wastewater treatment systems. Environ. Toxicol. Chem. 21:1323 1329. McBain, A., R. Ledder, P. Sreenivasan, and P. Gilbert. 2004. Selection for high level resistance by chronic triclosan exposure i s not universal. J. Antimicr Chem. 53: 772 777. McClellan K., and R. U. Halden. 2010. Pharmaceuticals and personal care products in archived U.S. biosolids from the 2001 EPA national sewag e sludge survey. Water Res. 44: 626 636. McMahon, T. 1998. An AD Memo by Tim McMahon to Jess R owland, Executive McMurry, L. M., M. Oethinger and S. B. Levy. 1998. Triclosan targets lipid biosynt hesis. Nature 394: 531 532. Meylan, W.M., a nd P.H. Howard. 1991. HENRYWIN s oftware available from syracuse research Corp., Environmental science center, Syracuse, NY 13210. Environ. Toxicol. Chem. 10:1283 1293. Miller, T. R., J. Heidler, S. N. Chillrud, A. DeLaquil, J. C. Ritchie, J. N. Milalic, R. Boop, and R. U. Halden. 2008. Fate of triclosan and evidence for reducti ve dechlorination of triclocarban in estuarine sediment. Environ. Sci. Technol. 42:4570 4576. MITI. 1992. Biodegradation and bioaccumulation data of existing chemicals based on the CSCL Japan. Japan chemical industry ecology toxicology & information center Miyazaki, T., T Yamagishi, and M Matsumoto 1984. Residues of 4 chloro 1 (2,4 dichlorophenoxy) 2 methoxybenzene (triclosan methyl) in aquatic biota. Bull. E nviron. Contam. Toxicol. 32:227 232. Morales, S., P. Canosa, I. Rodriguez, E. Rubi, and R. Cela. 2005. Microwave assisted extraction followed by gas chromatography with tandem mass spectrometry for the determination of triclosan and two related chlorophenols in sludge and sediments. J Chromatogr. A. 1082:128 135. Morrison, H. A., F. A. P. C. Gobas, R. Lazar, and G. D. Haffner. 1996. Development and verification of a bioaccumulation model for organic contaminan ts in benthic invertebrates. Environ. Sci Technol. 30:3377 3384.
242 Moss, T., D. Howes, and F.Mm. Williams. 2000. Percutaneous penetration and de rmal metabolism of triclosan trichloro hydroxydiphenylether) Food Chem. Toxicol. 38 :361 370. Nagy, K. A. 1987. Field metabolic rate and food requirement scaling in mammals and bi rds. Ecolog. Monographs 57: 111 128. National industrial chemicals notification and assessment scheme (NICNAS). 2009. Priority existing chemicals assessment report no. 30. Australian Government Department of Health and Ageing. National Research Council (NRC) 2002. Biosolids applied to land: advancing standards and practices, National Academy Press. Washington, DC. Neil, O. 2006. The Merck Index, 14th ed. Merck & Co., Inc., USA. Neilson, A. H., A. Allard, P. Hynning, M. Remberger, and L. Landner. 1983. Bacterial methylation of chlorinated phen ols and guaiacols; Formation of veratroles from guaiacols and high molecular weight chlorinated lignin. Applied Environ. Micro. 774 783. Newman, Y., J. Vendramini, and A. Blout. 2010. Bahiagrass (Paspalum notatum): Overview and management. University of Fl orida, IFAS extension publication # SS AGR 332. Newton, P., S. Cadena, M. Rocha, E. Carnieri and M. Oliveira. 2005. Effect of triclosan (TRN) on energy linked functions of rat liver mitochondria. Toxicol. Lett. 160 :49 59. Nghiem, L. D., and P.J. Coleman. 2008. NF/RO filtration of the hydrophobic ionogenic compound triclosan: transport mechanisms and the influence of membrane fouling. Separat. Purif Technol. 62 :709 716. Nowosielski, B. E., and J. B. Fein. 1998. Experimental study of octanol water partition coefficients for 2,4,6 trichlorophenol and pentachlorophenol: Derivation of an empirical model of chlorophenol partitioning behavior. Appl. Geochem. 13:893 904. amended soils and their poten tial for uptake by crop plants. Sci. Total Environ. 185: 71 81. Orvos, D. R., D. J. Versteeg, J. Inauen, M. Capdevielle, A. Rothenstein, and V. Cunningham. 2002. Aquatic toxicity of triclosan. Environ. Toxicol. Chem. 21:1338 49. Palenske, N. M., G. Nallani, and E. M. Dzialowski. 2010. Physiological effects and bioconcentration of triclosan on amphibian larvae. Comp. Biochem. Physiol. C. 152:232 240.
243 Pa terson, S., D. Mackay, and C. McFarlane. 1994. A Model of organic chemical u pta ke by plants from soil and the a tmosph er e. Environ. Sci. Technol. 28: 2259 2266. Petersen, S. O., K. Henriksen, G. K. Mortensen, P. H. Krogh, K. K. Brandt, J. Sorensen, T. Madsen, J. Petersen, and C. Gron. 2003. Recycling of sewage sludge and household compost to arable land: fate and effects of organic contaminants, and impact on soil fertility. Soil Tillage Res. 72:139 152. Poulsen, T. G., and K. Bester. 2010. Organic micropollutant degradation in sewage sludge during compositing under thermophilic conditi ons. Environ. Sci. Technol. 44: 5086 50 91. Pycke, B. F. G., A. Crabbe, W. Verstraete, and N. Leys. 2010. Characterization of triclosan resistant mutants reveals multiple antimicrobial resistance mechanisms Rhodospirillium Rubium S1H. Amer. Soc. Microbiol. 76:3116 3123. Reiss, R., G. Lewis, and J. Griffin. 2009. An ecological risk assessment for triclosan in the terrestrial environment. Environ. Toxicol. Chem. 28:1546 1556. Rodricks, J. V., J. A. Swenberg, J. F. Borzelleca, R. R. Maronpot, and A. M. Shipp 2010. Triclosan: A critical review of the experimental data and development of margins of safety for consumer products Critical Rev. Toxicol. 40:422 484. Rojas Oropeza, M., L. Dendooven, L. Garza Avendao, V. Souza, L. Philippot and N. Cabirol. 20 10. Effects of biosolids application on nitrogen dynamics and microbial structure in a saline sodic soil of the former Lake Texcoco (Mexico). Biores. Technol. 101:2491 2498. Rooklidge, S J. 2004. Environmental antimicrobial contamination from terraccumulation and diffuse pollution pa thways. Sci. Total Environ. 325: 1 13. Rotthauwe, J.H., K.P. Witzel, and W. Liesack. 1997. The ammonia monooxygenase structural gene amoA as a functional marker: molecular fine scale analysis of natural ammonia oxidizing populations. Appl. Environ. Microbiol. 63:4704 4712. Rutgers, M., I. M. van't Verlaat, B. Wind, L. Posthuma, and A. M. Breure. 1998. Rapid method for assessing pollution induced community t olerance in contaminated soil Environ. Toxicol. Chem.17: Sabaliunas, D., S. F. Webb, A. Hauk, M. Jacob, and W.S. Eckhoff. 2003. Environmental fate of triclosan in the river Aire basin, UK. Water Res. 37:3145 3154. Sabourin, L., A. Beck, P. W. Due nk, S. Kleywegt, D. R. Lapen, H. X. Li, C. D. Metcal fe, M. Payne, and E. Topp. 2009. Runoff of pharmaceuticals and personal care products following application of dewatered municipal biosolids to an agricultural field. Sci. Total Environ. 407: 4596 4604.
244 Sa msoe Petersen, L., M. Winther Nielsen, and T. Madsen. 2003. Fate and effects of triclosan. Project 861. Danish Environmental Protection Agency, Copenhagen, Denmark (www2.mst.dk/Udgiv/ publications/2003/87 7972 984 3/pdf/87 7972 985 1.pdf). SAS institute. 2 002. Online doc. Version 9.1.3. SAS Inst., Cary, NC. Schmitt, H., P. van Beelen, J. Tolls and C.L. van Leeuwen. 2004. Pollution induced community tolerance of soil microbial communities caused by the a ntibiotic sulfachloropyridazine. Environ. Sci. Technol. 38 :1148 1153. Schroll, R., B. Bierling, G. Cao, U. Drfler, M. Lahaniati, T. Langenbach, I. Scheunert, and R. Winkler. 1994. Uptake pathways of organic c hemicals from soil by agricultural plants. Chemosphere 28:297 30. Schroll, R., and I. Scheunert. 1992. A laboratory system to determine separately the uptake of organic chemicals from soil by plants and by l eaves after vaporization. Chemosphere 24: 97 108 Schultz, T.W., and G.W. Riggin. 1985. Predictive correlations for the toxicity of alkyl and halogen substituted phenols Toxicol. Lett. 25 : 47 54. Schwab, D., and L. G. Heim. 1997. Seedling growth phytotoxicity test. Prepared for Colgate Palmolive Company. Piscatway, New Jersey, and Ciba Geigy Corporation, Greensboro (NC). Prepared by ABC laborat ories, InC., Columbia (MO). Study 95 005. Semple, K A. Morris, and G. Paton. 2003. Bioavailability of hydrophobic organic contaminants in soils: Fundamental concepts and techniques for analysis. Eur. J. Soil Sci. 54:809 8 18 Shareef, A., M. J. Angove, an d J. D. Wells. 2006. Optimization of silyation of using N methyl N (trimethylsilyl) trifluoroacetamide, N,O bis (trimethylsilyl) trifluoro acetamide and N (tert butyldemethylsilyl) N methyltrifluoroacetamide for the determination of the estrogens, estrone and 17 alpha ethinylestradiol by gas chromatography mass spect rometry. J. Chromatogr. A 1108: 121 128. Sheen, R.T., and H.L. Kahler. 1938. Effects of ions on Mohr method for chloride determina tion. Ind. Eng. Chem. Anal. 10: 628 629. Shirshova L T E. A. Ghabbour, and G. Davies 2006. Spectroscopic characterization of humic acid fractions isolated from soil using different extraction procedures. Geoderma 133:204 16. Simonich, S. L., and R.A. Hites. 1995. Organic pollutant accumulation in v egetation. Enviro n. Sci. Technol. 29:2905 2914.
245 Singer, H., S. Muller, C. Tixier, and L. Pillonel. 2002. Triclosan: Occurrence and fate of a widely used biocide in the aquatic environment: Field measurements in wastewater treatment plants, surface waters, and lake sediment s. Environ. Sci. Technol. 36:4998 5004. Singh, A., and O.P. Ward. 2004. Biod egradation and Bioremediation. Springer, Berlin, Germany. Snyder, E. H. 2009. Fate, transport, and risk assessment of biosolids borne triclocarban (TC C). Ph.D. dissertation. Uni. of Florida, Gainesville, FL. 237 pages. Snyder, Fate of 14 C triclocarban in biosolids amended soils. Sci. Total Environ. 408:2726 27 32. Snyder, and D. McAvoy. 2011 Toxicity and bioaccumulation of biosolids borne triclocarban (TCC) in terrestrial organisms. Chemosphere 82:460 467 Stasinakis, A., A. Petalas, D. Mamais, N. Thomaidis, G. Gatidou, and T. Lekkas. 2007. Investigation of triclosan fate and toxicity in continuous flow ac tivated slu dge systems. Chemosphere 68:375 381. Stephens, J. M. 2009. Mini gardening (Growin g vegetables in containers). Un iversity of Florida, IFAS extension publication. HS708. Available at http://edis.ifas.ufl.edu/pdffiles/VH/VH03200.pdf Accessed on Feb, 7, 2011 Stevens, K.J., S. Y. Kim, S. Adhikari, V. Vadapalli and B. J. Venables. 2009. Effects of triclosan on seed germination and seedling development of three wetland plants: Sesbania herbacea Eclipta prostrate and Bidens frondosa Environ Toxicol Chem 28:2598 2609. Sullivan, T. S., M. E. Stromberger, M. W. Paschke, and J. A. Ippolito. 2006a. Long term impacts of in frequent biosolids applications on chemical and microbial propert ies of a semi arid rangeland soil. Biol. Fertil. Soil. 42:258 266. Sullivan, T. S., M. E. Stromberge, and M. W. Paschk e 2006b. Parallel shifts in plant and soil microbial communities in response to biosolids in a semi arid grass land. Soil Biol. Biochem. 3 8:449 45 9 Suter, G.W. 2007. Ecological Risk Assessment. CRC Press, Boca Raton, Florida. Suter, G.W., R.A. Efroymson, B.E. Sample, and D.S. Jones. 2000. Ecological risk assessment for contaminated sites. CRC Press, Boca Raton, Florida. Svenningsen, H., T. Henriksen, A. Prieme, and A. R. Johnsen. 2011. Triclosan affects the microbial community in simulated sewage drain field soil and slows down xenobiotic degradation. Env iron. Pollut. Available at 10.1016/j.envpol.2011.02.052.
246 Topp, E., S. C. Monteiro, A. B eck, B. Ball Coelho, A. B. Boxall, and P. W. Duenk. 2008. Runoff of pharmaceuticals and personal care products following application of biosolids to an agricultural f ield. Sci. Total Environ. 396:52 5 9. Trapp, S., and J. C. McFarlane. 1995. Plant Contamina tion. Modeling and simulation of organic chemical processes. CRC Press, Inc. Travis, C. C., and A. D. Arms, 1988. Bioconcentration of organics in beef, milk and vegetation. Environ. Sc i. Technol. 22:271 292. Tulp, M. T. M., G. Sundstro m, L. B. J. M. Martron, and O. Hutzinger. 1979. Metabolism of chlorodiphenyl ethers and I rgasan DP 300. Xenobiotica 9:65 77. Turpin, K M., D. R. Lapen M. J. Robi n E. Topp M. Edwards W. E. Curnoe, G. C. Topp, N. B. McLaughlin, B. Ball Coelho, and M. Payne. 2007. Slur ry application implement tim e modification of soil hydraulic properties under different soil water content conditions for silt clay loam soils. Soil Tillage Res 95:120 32. United State Geological Survey (USGS). 2008. Household chemicals and drugs found in biosolids from wastewater treatment plants toxic substances hydrology program. Available from http://toxics.usgs.gov/highlights/biosolids.html Accessed on February 14, 2009 United States Environmental Protection Agenc y (USEPA). 1984. Calculation of precision, bias, and method detection limit for chemical and physical measurements. Office of research and development, USEPA, Washington, D.C. 20460. United States Environmental Protection Agen cy (USEPA). 1991. Intera gency policy on beneficial use of municipal sewage sludge on federal land; notice. Fed Regist 5 6:30448 30450. United States Environmental Protection Agency (USEPA). 1993a. Determination of nitrate nitrite by automated colorime try, EPA Method 353.2 (Revision 2.0). Methods for the determination of inorganic substances in environmental samples. EPA/600/R 93/100. USEPA, Cincinnati, Ohio. United States Environmental Protection Agency (USEPA). 1993b. Determination of ammonia nitrogen by semi automated colorimetry, EPA Method 350.1 (Revision 2.0). Methods for the determination of inorganic substances in environmental samples. EPA/600/R 93/100. USEPA, Cincinnati, Ohio. United States Environmental Protection Agency (USEPA). 1993c. Wildli fe Exposure Factors Handbook. EPA/600/R 93/187. USEPA, Washington, DC. United States Environmental Protection Agency (USEPA). 1995 A Guide to the Biosolids Risk Assessments for the EPA Part 503 Rule. EPA/832 B 93 005. USEPA, Washington, DC.
247 United States Environmental Pro tection Agency (USEPA). 1996a. Product properties test guidelines, OPPTS 830.7840, Water solubility: colum n elution method; shake flask. OPPTS, Washington, D.C. United States Environmental Protection Agency (USEPA). 1996b. Product p roperties test Guidelines, OPPTS 850.6200, earthworm sub chronic toxicity test. OPPTS, Washington, D.C. United States Environmental Pro tection Agency (USEPA). 1996c. Product properties test guidelines, OPPTS 850.5100, Soil microbial community toxicity test OPPTS, Washington, D.C. United States Environmental Protection Agency (USEPA). 1997. Exposure factors h andbook. National Center for Environmental Assessment, Office of Research and Development, Washington, DC. United States Environmental Protection Age ncy (USEPA). 1998. Fate, transport and transformation test guidelines, OPPTS 835.3300. Soil biodeg. OPPTS, Washington, D.C. United States Environmental Protection Agency (USEPA). 1999. Category for persistent, bioaccumulative, and toxic new chemical substa nces. http://www.epa.gov/fedrgstr/EPA TOX/1 999/November/Day 04/t28888.htm ,Accessed on Jan 20, 2010 United States Environmental Protection Agency (USEPA). 2008 a Fate, transport and transformation test guidelines, OPPTS 835.1240. Leaching studies. OPPTS, Washington, D.C. United States Environmental Protection Agency (USEPA). 2008 b Reregistration eligibility decision (RED) for triclosan. Office of Prevention, Pesticides, and Toxic Substances (OPPTS). USEPA, Cincinnati, Ohio. United States Environmental Pr otection Agency (USEPA), 2009a. Targeted National Sewage Sludge Survey Sampling and Analysis Technical Report ("Technical Report", EPA 822 R 08 016). USEPA, Washington, D.C. United States Environmental Protection Agency (USEPA), 2009b. Estimation Program s Protection Agency, Washington, DC, USA. Ura, K., T. Kai, S. Sakata, T. Iguchi, and K. Arizono. 2002. Aquatic acute toxicity testing using the nematode Caenorhabditis elagans J. Health Sci. 48 : 583 586 Valo, R., and M. Salkinoja Salonen. 1986. Microbial transformation of polychlorinated phenoxy phenols. J. Gen. Appl. Microbiol. 32:505 517.
248 Veldhoen, N., R. C. Skirrow, H. Osachoff, H. Wigmore, D.J. Clapson, M.P. Gunderson, G. Van Aggelen, and C.C. Helbing. 2006. The bactericidal agent triclosan modulates thyroid hormone associated gene expression and disrupts postembryonic anura n development. Aquatic Toxicol. 80:217 227. Waller, N. J., and R. S. Kookana. 2009. Effect of triclosan o n microbial activity in Australian soils. Environ. Toxicol. Chem. 28:65 70. Walters, E., K. McClellan and R. U. Halden. 2010. Occurrence and loss over three years of 72 pharmaceuticals and personal care products from biosolids soil mixtures in outdoor mesocosms. Water Res. 44:6011 6020. Wang, L. Q., C.N. Falany, and M.O. James. 2004. Triclosan as a substrate and inhibitor of 3' phosphoadenosine 5' phosphosulfate sulfotransferase and udp glucuronosyl transferase in human liver fractions. Drug Metab. Disp os. 32:1162 1169. Webster, E., and D Mac kay. 2007. Modeling the fate of contaminants introduced into agricultural soils from biosolids: BASL4 model: users' manual report to environment Canada. CEMC Report No. 200702 Trent University, Peterborough, Ontario Wezel, A. P., and T. Jager. 2002. Comparison of two screening level risk assessment approaches for six disinfectants and pharmaceuticals. Chemosphere 47:1113 1128. Wild, S. R. and K. C. Jones. 1991. Organic Contaminants in Wastewaters and Sewage Sludges: Transfer to the Environment Following Disposal. In: K.C. Jones (Ed.), Organic Contaminants in the Environment, Environmental Pathways and Effects, Elsevier, 1991. Wild, S. R., and K. C. Jones. 1992. Organic chemicals entering agricultural soils in sewage sludges: screening for their potential to transfer to crop plants and live stock. Sci. Total Environ. 199: 85 119. Wu, C., A. L. Spongberg, and J. D. Witter. 2009. Adsorption and degradation of triclosan and triclocarban in soils and biosolids amended soils. J. Agric. Food Chem. 57: 4900 4905. Wu, C., A. L. Spongberg, J. D. Witter, M. Fang, and K. P. Czakowski. 2010. Uptake of pharmaceutical and perso nal care products by soybean plants from soils amended with biosolids and irrigated with contaminated water. Environ. Sci. Technol. 44:6157 6161. Xia, K. 2010. Organic compounds of emerging concern in biosolids and biosolids applied soils: occurrence, accumulation, and transformation. W2170 presentation, June 2010.
249 Xia, K., L. S. Hundal, K. Kumar, K. Armbrust, A. E. Cox, and T. C. Granato. 2010. Triclocarban, triclosan, polybrominated diphenyls ethers, and 4 nonylphenol in biosolids and in soil receivin g 33 year biosolids applicati on. Environ. Toxicol. Chem. 29: 597 605. Xu, J., L. S. Wu, and A. C. Chang. 2009. Degradation and adsorption of selected pharmaceuticals and personal care products (PPCPs) in agricultural soils. Chemosphere 77: 1299 1305. Yang, L H., G. G. Ying, H. C. Su, J. L. Stauber, M. S. Adams, and M. T. Binet. 2008. Growth inhibting effects of 12 antibacterial agents and their mixtures on the freshwater microalga Pseudokirchneriella subcapitata Environ.Toxicol. Chem. 27:1201 1208. Yaron, B ., R. Calvet, and R. Pr ost. 1996. Soil pollution, processes and d ynamics. Springer, Heidelberg, Germany. Ye, X., Z. Kuklenyik, L. L. Needham, and A. M. Calafat. 2005. Automated on line column switching HPLC MS/MS method with peak focusing for the determi nation of nine environmental phenols in urine. Anal. Chem. 77:5407 5413. Ying, G. G., and R.S. Kookana. 2007. Triclosan in wastewaters and biosolids from Austral ian wastewater treatment plants. Environ. Int. 33 : 199 205 Ying, G. G., X. Y. Yu, and R. S. Koo kana. 2007. Biological degradation of triclocarban and triclosan in a soil under aerobic and anaerobic conditions and comparison with environmental fate modeling. Environ. Pollut. 150: 300 305. Young, T. 2011. Triclosan, triclocarban (and other trace organi cs) in biosolids. CWEA Biosolids Workshop Jan 2011. Yu, Y. L., X. M. Wu, S. N. Li, H. Fang, Y. J. Tan, J. Q. Yu. 2005. Bioavailability of butachlor and myclobutanil residues in soil to earthworms. Chemosphere 59:961 967. Zerzghi, H., C. P. Gerba, and I. L. Pepper. 2010 a Long term effects of land application of class B biosolids on soil chemical properties. J. Res. Sci. Technol. 7:51 61. Zerzghi, H., J. P. Brooks, C. P. Gergba, and I. L. Pepper. 2010 b Influence of long term land application of class B bios olids on soil bacterial diversity. J. Applied Microbol. 109:698 706. Zhao, F. 2006. Biodegradation of triclosan by a triclosan degrading isolate and an ammonia oxidizing bacterium. Master of Scien ce Thesis, Texas A&M University.
250 BIOGRAPHICAL SKETCH Manmeet Waria was born in t he beautiful state of Punjab (India), and is the youngest of the 3 three sisters with a younger brother She received a Bachelor of Science (Hons.) in a griculture from Punjab Agricultural U niversity in India and moved to U .S in 2005. She obtained a Master of Science in e nvironmental soil s cience from University of Nebraska, Linc oln in 2007. In the spring of 2008, she moved to s unny Florida and joined Ph.D. in the Soil and Water Science D epartment and graduated in spring 20 11 She is a married woman and alon g with her Veterinarian husband, she dreams to travel the world.