Identification and Analysis of Halogenated Contaminants ... · Identification and Analysis of Halogenated Contaminants Formed in Thermal Decomposition of Halogenated Materials Doctor
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Identification and Analysis of Halogenated Contaminants Formed in Thermal
Decomposition of Halogenated Materials
by
Anne Lindsay Myers
A thesis submitted in conformity with the requirements
Identification and Analysis of Halogenated Contaminants Formed in Thermal
Decomposition of Halogenated Materials
Doctor of Philosophy Degree, 2014
Anne Lindsay Myers
Department of Chemistry, University of Toronto
ABSTRACT
Fires release a wide array of contaminants that contribute to detrimental
environmental and health effects. The thermal decomposition of halogenated materials in
fires produces complex mixtures that include unidentified environmentally persistent and
toxic halogenated contaminants. Resulting complex mixtures pose an analytical challenge
in assessing the hazards of a fire site where burn conditions and contents are unknown.
Non-targeted analytical approaches can facilitate identification of novel compounds in
complex mixtures to direct further study. This thesis investigates the thermal
decomposition mechanisms and products of halogenated materials using non-targeted
analytical techniques, and assesses the environmental relevance of these thermal
processes.
Mass defect filtering of high-resolution mass spectra is an effective approach to
visually resolve and interpret complex chemical information. This non-targeted approach
was used to characterize the thermal decomposition products of
polychlorotrifluoroethylene (PCTFE) and polyvinylidene fluoride (PVDF). High-
resolution mass spectra, obtained by Fourier transform ion cyclotron resonance mass
spectrometry (FTICR-MS), were interpreted using mass defect plots based on unique
mass scales. Novel halogenated compounds were identified, including
iii
perchlorinated/fluorinated polycyclic aromatic hydrocarbons (X-PAHs, X=Cl,F), and
polyfluorinated polycyclic aromatic hydrocarbons (F-PAHs). Congener profiles of
thermal decomposition products provided evidence of associated decomposition
mechanisms.
The potential of non-targeted analyses to characterize complex environmental
samples was further exemplified through studies examining the environmental relevance
of combustion-derived contaminants. Using mass defect plots as a screening tool, a
variety of brominated contaminants, including polybrominated/chlorinated dibenzofurans
(PXDFs, X=Br/Cl), were identified in air and particulates collected from simulation fire
studies. Mass defect filtering, in combination with comprehensive two-dimensional gas
chromatography high-resolution time of flight mass spectrometry (GCxGC-HRToF) and
gas chromatography tandem mass spectrometry (GC-MS/MS) was used to identify
bioaccumulative contaminants in aquatic species exposed to soil from an industrial fire
site. Compounds identified as bioaccumulative included polybrominated/chlorinated
anthracenes/phenanthrenes and pyrenes/fluoranthenes (X-PAHs, X=Br,Cl).
The majority of environmental analyses is targeted and thereby only examines a
small subset of compounds present in the environment. This work demonstrates the
potential of non-targeted techniques to screen complex environmental samples for
halogenated contaminants. This is particularly applicable to combustion-derived
mixtures, the complexity of which varies with fire conditions and contents.
iv
ACKNOWLEDGEMENTS During my time at the University of Toronto I have been fortunate to work with
many amazing individuals from whom I have learned so much.
First, thank you to my co-supervisors, Eric and Scott. Eric, thank you for your
support and encouragement, and for sharing your vast mass spectrometry and dioxin
expertise. Scott, thank you for your support and excellent professional advice, I have
learned so much from working with you.
Thanks to Jen for serving on my committee and being a mentor to me. I greatly
appreciate your advice and perspective. Thanks also to Derek for serving on my
committee, it has been an honour working with such an accomplished environmental
chemist.
I am indebted to my Mabury group colleagues, past and present, for their incredible
support and friendship. Thank you Jess for your outstanding kindness and patience, you
are a great friend and mentor. Derek, it’s been so fun working with you and I’ve learned
so much, you are a gifted teacher. Amila, your friendship, support and mentorship has
meant so much. Amy, it was great working with you, I truly admire your spontaneity and
toughness. Holly, you make everything look so easy, such a talent! Lisa, thanks for much
needed coffee breaks and for being my California conference buddy. Thank you Leo for
sharing your exceptional analytical expertise and for so many useful discussions. Keegan,
I strive to be as calm, cool and collected as you, thanks for all the useful discussions and
feedback. Thanks also to Erin, Craig, Cora, Shona, Rob, Barbara, Rui and Angela, it has
been a pleasure working with such talented and kind people.
The Environmental Chemistry department is filled with so many great people, but I
must thank Zamin, Sumi, Dorea, Rachel, Rob, Jeff, Sarah, and Alex for being such
wonderful colleagues. Thanks to Jamie and Frank for helpful feedback and advice, as
well as Anna Liza for administrative support. To my colleagues at the MOE, thank you
Karl, Miren, and Li, for your support and guidance.
Thank you to my parents, Shirley and Dick, and my brother, Ken, for providing
unwavering strength and support through challenging times. Katie, thanks for being the
most understanding and kind friend I could ask for. Finally, thanks to Greg for being so
caring, thoughtful and supportive, you are the best.
v
TABLE OF CONTENTS
CHAPTER ONE - Thermal Decomposition of Halogenated Materials ...................... 1!A. L. Myers, S. A. Mabury, and E. J. Reiner 1.1! Environmental Impact of Fire .................................................................................... 2!1.2! Definitions.................................................................................................................. 2!1.3! Chemistry of Uncontrolled Fires ............................................................................... 3!1.4! Halogenated Polymers and Flame Retardants ........................................................... 6!
1.4.1! Production and Commercial Use ................................................................................. 6!1.4.2! Thermal Decomposition .............................................................................................. 7!1.4.3! Environmental Fate and Toxicity of Thermal Decomposition Products ..................... 9!
1.5! Characterization of Thermal Decomposition Products ............................................ 13!1.5.1! Pyrolysis and Combustion Studies ............................................................................ 13!
1.5.2! Non-targeted Analysis ............................................................................................... 14!1.5.2.1! Mass Defect Filtering of High-Resolution Mass Spectra ................................................ 15!1.5.2.2! GCxGC-HRToF .............................................................................................................. 18!
1.5.3! Applications of Non-targeted Analysis ..................................................................... 19!1.6! Goals and Hypotheses .............................................................................................. 20!1.7! References ................................................................................................................ 22! CHAPTER TWO - Using Mass Defect Plots as a Discovery Tool to Identify Novel Fluoropolymer Thermal Decomposition Products ...................................................... 30!A. L. Myers, K. J. Jobst, S. A. Mabury, E. J. Reiner, J. Mass Spectrom. 49, 291-296 (2014) 2.1! Abstract .................................................................................................................... 31!2.2! Introduction .............................................................................................................. 31!2.3! Materials and Methods ............................................................................................. 33!
CHAPTER THREE - Identification and Environmental Relevance of Fluoropolymer Thermal Decomposition Products ...................................................... 48!A. L. Myers, S. A. Mabury, and E. J. Reiner 3.1! Abstract .................................................................................................................... 49!3.2! Introduction .............................................................................................................. 49!3.3! Materials and Methods ............................................................................................. 52!
3.3.1! Thermal Decomposition Experiment ........................................................................ 52!3.3.2! XAD Adsorbent Tube Extraction .............................................................................. 52!3.3.3! FTICR-MS Analysis .................................................................................................. 52!3.3.4! Mass defect filtering of FTICR-MS data .................................................................. 53!3.3.5! GC-MS Analysis ....................................................................................................... 54!3.3.6! Quality Assurance/Quality Control (QA/QC) ........................................................... 54!
3.4.3! Sources of Fluoropolymer Thermal Decomposition Products to the Environment .. 63!3.4.3.1! Controlled and Regulated Industrial Practices ................................................................ 63!3.4.3.2! Uncontrolled and Unregulated Thermal Decomposition Events ..................................... 64!
3.4.4! Environmental Fate and Toxicity of Fluoropolymer Thermal Decomposition Products ................................................................................................................................. 65!
3.4.4.1! Halogenated Alkanes and Alkenes .................................................................................. 65!3.4.4.2! PFCAs and PXCAs .......................................................................................................... 66!3.4.4.3! Halogenated Aromatics ................................................................................................... 67!
3.4.5! Next Steps in Understanding the Environmental Relevance of Fluoropolymer Thermal Decomposition Products ......................................................................................... 68!
3.5! Acknowledgements .................................................................................................. 69!3.6! References ................................................................................................................ 69! CHAPTER FOUR - Screening for Halogenated Contaminants in Fire Samples Using Mass Defect Plots .................................................................................................. 76!A. L. Myers, K. J. Jobst, K. Organtini, S. Fernando, B. Ross, B. McCarry, F. Dorman, S. A. Mabury, and E. J. Reiner 4.1! Abstract .................................................................................................................... 77!4.2! Introduction .............................................................................................................. 77!4.3! Materials and Methods ............................................................................................. 79!
4.3.1! Fire Simulation Experiments ..................................................................................... 79!4.3.2! Sample Preparation .................................................................................................... 81!
4.3.2.1! Foil Samples .................................................................................................................... 81!4.3.2.2! Air Samples ..................................................................................................................... 81!4.3.2.3! Water Run-off Samples ................................................................................................... 82!
4.3.3! FTICR-MS Analysis and Mass Defect Filtering ....................................................... 82!4.3.4! Quality Assurance/Quality Control (QA/QC) ........................................................... 84!
4.4! Results and Discussion ............................................................................................ 84!
vii
4.4.1! Thermal Decomposition Products ............................................................................. 84!4.4.1.1! Run-off Water .................................................................................................................. 84!4.4.1.2! Household Fire ................................................................................................................ 84!4.4.1.3! Electronics Fire ................................................................................................................ 86!
4.4.2! Sources and Fate of Thermal Decomposition Products ............................................ 88!4.4.2.1! PBDEs ............................................................................................................................. 88!4.4.2.2! PBDFs and PXDFs .......................................................................................................... 89!4.4.2.3! Br-Anthracenes/Phenanthrenes ....................................................................................... 91!
4.4.3! Potential for Rapid Broad Screening Approach ........................................................ 92!4.5! Acknowledgements .................................................................................................. 92!4.6! References ................................................................................................................ 93! CHAPTER FIVE - Analysis of Mixed Halogenated Dibenzo-p-dioxins and Dibenzofurans (PXDD/PXDFs) in Soil by Gas Chromatography Tandem Mass Spectrometry (GC-MS/MS) ........................................................................................... 99!A. L. Myers, S. A. Mabury, E. J. Reiner, Chemosphere 87, 1063-1069 (2012). 5.1! Abstract .................................................................................................................. 100!5.2! Introduction ............................................................................................................ 100!5.3! Materials and Methods ........................................................................................... 102!
5.4! Results and Discussion .......................................................................................... 108!5.4.1! Method Performance ............................................................................................... 108!5.4.2! PXDD/PXDF Isomer Peak Patterns and Ratios ...................................................... 108!5.4.3! Confirmation of PXDD/PXDFs by GC-HRToF ..................................................... 111!5.4.4! PXDD/PXDFs in Soil from the Plastimet Inc. Fire ................................................. 112!
5.5! Conclusions ............................................................................................................ 114!5.6! Acknowledgements ................................................................................................ 114!5.7! References .............................................................................................................. 114! CHAPTER SIX - Complementary Non-targeted and Targeted Mass Spectrometry Techniques to Determine Bioaccumulation of Halogenated Contaminants in Freshwater Species ........................................................................................................ 118!A. L. Myers, T. Watson-Leung, K. J. Jobst, L. Shen, S. Besevic, K. Organtini, F. L. Dorman, S. A. Mabury, and E. J. Reiner (submitted to Environ. Sci. Technol.) 6.1! Abstract .................................................................................................................. 119!6.2! Introduction ............................................................................................................ 119!6.3! Materials and Methods ........................................................................................... 121!
6.3.1! Chemicals ................................................................................................................ 121!6.3.2! Bioaccumulation study ............................................................................................ 122!6.3.3! Lipid and Total Organic Carbon (TOC) Analysis ................................................... 123!
6.5! Discussion .............................................................................................................. 134!6.5.1! Variation in Contaminant Uptake Between Species ............................................... 134!6.5.2! BSAFs of PCNs, Cl-PAHs, X-PAHs, PCDFs, and PXDD/PXDFs ........................ 135!6.5.3! Considerations for Combustion-derived Contaminant Bioaccumulation ................ 137!6.5.4! Potential and Limitations in Non-targeted Analysis of Environmental Contaminants ................................................................................................................................. 138!
6.6! Acknowledgements ................................................................................................ 139!6.7! References .............................................................................................................. 139! CHAPTER SEVEN - Summary, Conclusions, and Future Directions .................... 147!A. L. Myers, S. A. Mabury, and E. J. Reiner 7.1! Summary and Conclusions .................................................................................... 148!7.2! Future Directions ................................................................................................... 151!7.3! References .............................................................................................................. 153!
ix
LIST OF TABLES CHAPTER FIVE Table 5.1. PXDD/PXDF parent-product transitions, corresponding collision energies and peak area ratios used in GC-MS/MS analysis. ................................................................ 105!Table 5.2. GC-MS/MS method performance data. ........................................................ 107! CHAPTER SIX Table 6.1. Mean BSAFs for halogenated compounds identified in biota extracts with associated standard deviations (SD). BSAFs for PCNs, Cl-PAHs, X-PAHs, and PCDFs analyzed by GCxGC-HRToF represent isomers. BSAFs for PXDD/PXDFs analyzed by APGC-MS/MS represent congener groups or multiple isomers. Shaded cells indicate BSAFs were statistically greater than (in bold) or less than 1 (p ! 0.05, Student's t test). ND indicates analyte was not detected. LD indicates analyte was detected, but did not meet peak identification requirements. *BSAFs based on analyte instrumental responses < LOQ. **BSAFs based on extrapolated sediment concentrations. †n=4. ..................... 129 Table 6.2. Of the isomers identified by GCxGC-HRToF in Table 1, six were confirmed through comparison of retention time coordinates with analytical standards. ND indicates analyte was not detected. LD indicates analyte was detected, but did not meet peak identification requirements. Corresponding estimated values for logKOW and water solubility were generated by the United States Environmental Protection Agency’s EPISuiteTM. .................................................................................................................... 130
x
LIST OF FIGURES CHAPTER ONE Figure 1.1. General formation of aromatic compounds in thermal decomposition. .......... 5!Figure 1.2. PBDE thermal decomposition pathway to form PBDD/PBDFs. .................... 8!Figure 1.3. Thermal decompositon of PVDF. .................................................................. 10!Figure 1.4. General structures of some halogenated thermal decomposition products. .. 10!Figure 1.5. Simplified depiction of an HRToF. ............................................................... 16!Figure 1.6. Simplified depiction of an FTICR-MS. ......................................................... 17!Figure 1.7. Separation of three unresolved peaks using GCxGC separation followed by transformation to two-dimensional chromatogram. .......................................................... 19! CHAPTER TWO Figure 2.1. A partial section of the mass defect plot of polar products produced in the thermal decomposition of PCTFE at 400oC. The series of ions (black diamonds) represent elemental compositions corresponding to perchlorofluoroalkanoic aicds (PXCAs, X=Cl,F). ............................................................................................................. 38!Figure 2.2. Mass defect plot of polar products produced in the thermal decomposition of PCTFE at 400oC (X=Cl,F). Number of carbon atoms ranged from two to 18 and number of chlorine atoms ranged from one to eight. ..................................................................... 39!Figure 2.3. Mass defect plot of non-polar products produced in the thermal ecomposition of PCTFE at 800oC (X = Cl,F). Bold elemental compositions correspond to proposed X-PAH structures. Non-bold elemental compositions are proposed X-PAH fragment ions from chlorine loss or X-PAH molecular ions with an additional CF2 group. ................... 42!Figure 2.4. (a) Mass spectrum of a trichloroheptafluoropyrene isomer obtained by GC-FTICR-MS analysis of non-polar extract for PCTFE thermal decomposition at 800oC. Labelled peaks correspond to the molecular ion and two proposed fragment ions. (b) Mass defect plot obtained using a DIP-FTICR-MS experiment, which shows series corresponding to X-pyrene isomers and fragment ions. ................................................... 43!Figure 2.5. (a) The fluorine/chlorine ratios of the most abundant X-PAH molecular ions for each series identified in the thermal decomposition of PCTFE at 800oC. The trend line indicates the increasing F/Cl ratio with increasing unsaturation. (b) The average fluorine/chlorine ratios and associated standard deviations for PXCA molecular ions identified containing zero (n=15), and two (n=6) double bond equivalents (DBE). ........ 44!
xi
CHAPTER THREE Figure 3.1. Mass defect plot of PVDF thermal decomposition products where n = 2-7. Inset demonstrates –H/+F (+17.99058) spacing between polyfluorinated naphthalene congeners. ......................................................................................................................... 55!Figure 3.2. Mass spectra corresponding to C12H4OF4
+·, C13H4F6+·, C12H6F4
+·, and C16H5F5
+·, as well as identifying mass spectral fragments and proposed structures. ........ 57!Figure 3.3. Relative abundance of mass spectral peaks corresponding to specific congeners of PFNs, PFDFs, PFBs, F-fluorenes/phenalenes (F-FLO/PHNs), F-anthracenes/phenanthrenes (F-ANT/PHEs), F-pyrenes/fluoranthenes (F-PYR/FLUs), and F-tetracenes identified on the x-axis by the elemental composition H/F ratio. Congeners with H/F ratios of 1:1 were generally the most prevalent within each congener group, indicating loss of HF during thermal decomposition of PVDF. ....................................... 58!Figure 3.4. Proposed thermal decomposition pathways for PTFE(29-35). ..................... 61!Figure 3.5. Proposed thermal decomposition pathways for PCTFE(13, 25, 27, 28, 31, 36, 37).. ................................................................................................................................... 62!Figure 3.6. Proposed thermal decomposition pathways for PVDF(18, 24, 25, 27, 28). .. 63! CHAPTER FOUR Figure 4.1. Burn cell photos before and after household fire simulation. ........................ 80!Figure 4.2. Burn cell photos before and after electronics fire simulation. ....................... 80!Figure 4.3. Burn cell fire simulation and sample collection set-up. ................................ 81!Figure 4.4. Mass defect plot based on H/Cl mass scale for household fire air sample extracts. Highlighted peaks correspond to PBDFs, PXDFs, and PBDEs, along with corresponding fragment series and unknown series, C14O2H13-nBrn
+!. ............................. 85!Figure 4.5. Selected ion chromatogram (SIC) for household fire air sample extract for 323.878 ± 0.010 demonstrating chromatographic separation of dibromodibenzofuran (C12H6OBr2
+!) (A) and dibromodiphenyl ether (C12H6OBr4+!) (B). Corresponding mass
spectra show fragmentation for each compound. ............................................................. 86!Figure 4.6. Mass defect plots based on H/Cl mass scale for electronics fire air sample extract (A) and foil extract (B) Highlighted peaks correspond to PBDFs, PXDFs, and PBDEs, along with corresponding fragment series and unknown series, C14O2H13-nBrn
+!. ................................................................................................................ 87!Figure 4.7. Relative abundance of PBDF and PXDF monoisotopic peaks with varying degrees of halogenation (X) for electronics fire foil extract (A) and household fire air extract (B). ........................................................................................................................ 91!
xii
CHAPTER FIVE Figure 5.1. Corresponding peak patterns, transitions, and peak ratios for monobromo-dichlorofurans in a 38.4pg/"L Standard and Plastimet Soil A extract. .......................... 110!Figure 5.2. General peak patterns observed for monobromo-dichlorodibenzofurans (A), monobromo-dichlorodibenzo-p-dioxins (B), monobromo-trichlorodibenzofurans (C), dibromo-dichlorodibenzo-p-dioxins (D), monobromo-tetrachlorodibenzo-p-dioxins (E), monobromo-pentachlorodibenzo-p-dioxins (F), monobromo-hexachlorodibenzo-p-dioxins (G), and monobromo-heptachlorodibenzo-p-dioxins (H). Chromatographic transitions shown correspond to M-COBrCl for dibenzofurans and M-(CO)2BrCl for dibenzo-p-dioxins. .......................................................................................................... 111!Figure 5.3. Concentrations of PXDD/PXDFs in soil samples with associated standard deviation (n=2). *For sample D, concentrations and standard deviations correspond to triplicate extractions and duplicate injections (n=6). ...................................................... 112! CHAPTER SIX Figure 6.1. Mass defect plot, based on the H/Cl mass scale, generated from FTICR-MS analysis of a Hexagenia spp. extract. Highlighted peaks represent PCNs, Cl-PAHs, X-PAHs and PCDFs, and proposed chemical structures. ................................................... 127!Figure 6.2. Two-dimensional selected ion chromatograms for mass range 299.8400-299.8800 highlighting bioaccumulative isomers of C10H3Cl5. Circled isomer peaks were detected in all sample replicates for the species indicated and peak intensities were used to determine BSAFs relative to those corresponding peaks in the Plastimet sediment mix (Table 1). Isomers a and c were confirmed with a PCN standard solution as 1,2,3,5,7-pentachloronaphthalene and 1,2,3,4,6-pentachloronaphthalene, respectively. The two remaining unlabeled peaks in the PCN standard chromatogram correspond to 1,2,3,6,7-pentachloronaphthalene and 1,2,3,5,8-pentachloronaphthalene. .................................... 131!Figure 6.3. APGC-MS/MS chromatograms corresponding to MRM transitions of bioaccumulative PXDD/PXDFs: A) C12OH5BrCl2 (313.8 > 206.8), B) C12OH4BrCl3 (347.7 > 240.8), C) C12OH3BrCl4 (381.8 > 274.5), D) C12OH4Br2Cl2 (393.6 > 286.9), E) C12OH3Br2Cl3 (427.6 > 320.9), F) C12O2H5BrCl2 (329.6 > 159.8), and G) C12O2H4BrCl3 (363.7 > 256.7). Top and bottom chromatograms correspond to Plastimet sediment mix and L. variegatus extracts, respectively. Peaks shaded in black correspond to peak areas used in BSAF calculations. ............................................................................................. 133
xiii
LIST OF APPENDICES APPENDIX A - Supporting information for Chapter Two ........................................... 156!APPENDIX B - Supporting information for Chapter Five ............................................ 160!APPENDIX C - Supporting information for Chapter Six ............................................. 163 APPENDIX D - Supporting data files for Chapters Two, Three, Four, Five, and Six...172
xiv
PREFACE
This thesis consists of two published manuscripts, one manuscript that has been
submitted to a peer-reviewed scientific journal, and four unpublished chapters. All
chapters were written by Anne L. Myers with editorial comments provided by Eric J.
Reiner. Contributions of all co-authors are provided in detail below.
CHAPTER ONE - Thermal Decomposition of Halogenated Materials
Author list: Anne L. Myers, Scott A. Mabury and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Eric J.
Reiner.
CHAPTER TWO - Using Mass Defect Plots as a Discovery Tool to Identify Novel
Fluoropolymer Thermal Decomposition Products
Published as: J. Mass Spectrom. 2014, 49, 291-296.
Author list: Anne L. Myers, Karl J. Jobst, Scott A. Mabury, and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Karl J.
Jobst, Scott A. Mabury, and Eric J. Reiner. Anne L. Myers was responsible for designing
and performing thermal decomposition experiments, sample extraction, LC-MS/MS
analysis, and data interpretation. Karl J. Jobst performed sample analysis by FTICR-MS
with assistance from Anne L. Myers.
CHAPTER THREE - Identification and Environmental Relevance of Fluoropolymer
Thermal Decomposition Products
Author list: Anne L. Myers, Scott A. Mabury and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Eric J.
Reiner. Anne L. Myers was responsible for designing and performing thermal
decomposition experiments, sample extraction, as well as data analysis and interpretation.
xv
CHAPTER FOUR - Screening for Halogenated Contaminants in Fire Samples Using
Mass Defect Plots
Author list: Anne L. Myers, Karl J. Jobst, Kari Organtini, Sujan Fernando, Brian Ross,
Brian McCarry, Frank Dorman, Scott A. Mabury and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Eric J.
Reiner. Anne L. Myers led the planning and organization of the simulation fire
experiments. Experimental design was initially conceived by Anne L. Myers and refined
through collaborative discussion and shared expertise of all co-authors. Anne L. Myers,
Karl J. Jobst, Kari Organtini, Sujan Fernando, Brian McCarry, Frank Dorman, and Eric J.
Reiner carried out the experiments in a collaborative effort with fire fighters from FESTI.
Anne L. Myers was responsible for water sample extraction, Kari Organtini was
responsible for aluminum foil extraction, and Sujan Fernando was responsible for air
sample extraction. Karl J. Jobst performed sample analysis by FTICR-MS with assistance
from Anne L. Myers.
CHAPTER FIVE - Analysis of Mixed Halogenated Dibenzo-p-dioxins and
Dibenzofurans (PXDD/PXDFs) in Soil by Gas Chromatography Tandem Mass
Spectrometry (GC-MS/MS)
Published as: Chemosphere 2012, 87, 1063-1069.
Author list: Anne L. Myers, Scott A. Mabury, and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Scott
A. Mabury and Eric J. Reiner. Anne L. Myers was responsible for sample extraction, GC-
MS/MS method development and analysis, GC-HRToF analysis, and data interpretation.
xvi
CHAPTER SIX - Complementary Non-targeted and Targeted Mass Spectrometry
Techniques to Determine Bioaccumulation of Halogenated Contaminants in Freshwater
Author list: Anne L. Myers, Trudy Watson-Leung, Karl J. Jobst, Li Shen, Sladjana
Besevic, Kari Organtini, Frank L. Dorman, Scott A. Mabury, and Eric J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Trudy
Watson-Leung, Karl J. Jobst, Scott A. Mabury, and Eric J. Reiner. Experimental design
was initially conceived by Eric J. Reiner. Trudy Watson-Leung was responsible for
exposure studies and Li Shen and Sladjana Besevic performed sample extractions. Karl J.
Jobst performed sample analysis by FTICR-MS. Anne L. Myers was responsible for
sample analysis by GCxGC-HRToF and APGC-MS/MS, as well as data interpretation.
CHAPTER SEVEN - Summary, Conclusions, and Future Directions
Author list: Anne L. Myers, Scott A. Mabury, and Eric. J. Reiner
Contributions: Prepared by Anne L. Myers with editorial comments provided by Eric J.
Reiner.
xvii
Other Publications During PhD: K. L. Organtini, A. L. Myers, K. J. Jobst, J. W. Cochran, B. Ross, B. E. McCarry, E. J. Reiner, F. L. Dorman, Comprehensive characterization of the halogenated dibenzo-p-dioxin and dibenzofuran contents of residential fire debris using GCxGC-TOFMS. 2014, submitted to J. Chromatogr. A. S. B. Gewurtz, et al., Perfluoroalkyl acids in the Canadian environment: multi-media assessment of current status and trends. Environ. Int. 59, 183-200 (2013). S. A. Styler, A. L. Myers, D. J. Donaldson, Heterogeneous photooxidation of fluorotelomer alcohols: a new source of aerosol-phase perfluorinated carboxylic acids. Environ. Sci. Technol. 47, 6358-6367 (2013). A. L. Myers, et al., Fate, distribution, and contrasting temporal trends of perfluoroalkyl substances (PFASs) in Lake Ontario, Canada. Environ. Int. 44, 92-99 (2012). R. J. Mitchell, A. L. Myers, S. A. Mabury, K. R. Solomon, P. K. Sibley, Toxicity of fluorotelomer carboxylic acids to the algae Pseudokirchneriella subcapitata and Chlorella vulgaris, and the amphipod Hyalella azteca. Ecotox. Environ. Safe. 74, 2260-2267 (2011). P. A. Helm, et al., Lake-wide distribution and depositional history of current- and past-use persistent organic pollutants in Lake Simcoe, Ontario, Canada. J. Great Lakes Res. 37, 132-141 (2011). A. L. Myers, S. A. Mabury, Fate of fluorotelomer acids in a soil-water microcosm. Environ. Toxicol. Chem. 29, 1689-1695 (2010).
1
1 CHAPTER ONE
Thermal Decomposition of Halogenated Materials
Anne L. Myers, Scott A. Mabury, and Eric J. Reiner Contributions: Anne L. Myers prepared this chapter with editorial comments provided by Eric J. Reiner.
2
1.1 Environmental Impact of Fire
Accidental and uncontrolled fires release a wide variety of undesirable
contaminants that contribute to detrimental environmental and health effects. Common
combustion-derived products include carbon dioxide (CO2), carbon monoxide (CO),
detection in air has been associated with point sources, such as municipal waste
incinerators, and there is evidence of long-range transport(55). The hydrophobic
nature of PCDD/PCDFs has led to their detection in soil and sediments, with
concentrations varying with proximity to point sources(55). Aquatic species are
particularly susceptible to PCDD/F bioaccumulation, however only certain isomers are
considered important. Of the 75 different PCDD congeners and 135 PCDF congeners,
only 17 are considered toxic, with 2,3,7,8-tetrachlorodibenzo-p-dioxin being the most
potent(55). The Stockholm convention recognizes PCDD/PCDFs as persistent organic
pollutants(56) and changing waste incineration technology has led to decreases in
PCDD/PCDF environmental detection(55).
The environmental behaviour of PBDD/PBDFs varies slightly from
PCDD/PCDFs. As a result of the weaker carbon-bromine bond, PBDD/PBDFs are more
susceptible to UV degradation and are less persistent(57). Higher molecular weights and
lower vapour pressures also limit atmospheric transport of PBDD/PBDFs relative to
PCDD/PCDFs(57). Typically PBDD/PBDF contamination occurs in close proximity to
point sources, such as incineration of wastes containing BFRs(57). However,
PBDD/PBDFs have also been measured in air, indoor dust, soil, sediment, wastewater
treatment plant sludge, and biota(57). Relative to PCDD/PCDFs and PBDD/PBDFs,
considerably less is known of PXDD/PXDFs. This is likely a result of there being 1550
PXDD congeners and 3050 PXDF congeners(58), and very few corresponding analytical
standards. To date, PXDD/PXDFs have been largely associated with incinerator fly
ash(39, 43-45, 59, 60).
The toxicity of polyhalogenated dibenzo-p-dioxins and dibenzofurans relates to
the number and position of halogen substituents. One isomer of considerable potency in
this contaminant group is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), which binds with
high-affinity to a specific cellular protein called the aryl hydrocarbon receptor (AhR),
interfering with normal gene expression(61). Effects of AhR-mediated activities include
wasting, gastric lesions, chloracne, and hepatotoxicity(62). The 2,3,7,8-
tetrabromodibenzo-p-dioxin (TBDD) has a slightly lower AhR binding affinity relative to
TCDD, however 2,8-dibromo-3,7-dichloro-dibenzo-p-dioxin has demonstrated twice the
binding affinity of TCDD(63). TBDD has a similar elimination half-life to TCDD,
12
however 2,3,7,8-tetrabromodibenzofuran (TBDF) has a longer half-life than
tetrachlorodibenzofuran (TCDF). This may be a result of the steric hindrance of the
bromine atoms blocking metabolism(63).
PCNs consist of two aromatic rings substituted with chlorine atoms. Thermal
processes, such as waste incineration, are a major source of PCNs, however PCNs may
also be present as flame-retardants, industrial technical mixtures (e.g. Halowax), and
impurities in polychlorinated biphenyl (PCB) formulations(64). Like PCDD/PCDFs,
PCNs are hydrophobic and persistent environmental contaminants that have been
detected in air, water, sediment, and biota(64). The bioaccumulative nature of PCNs
causes diet to be an important exposure route for humans(64). With 75 congeners, the
congener profiles observed in environmental samples vary widely. All PCN congeners
are planar, structurally similar to PCDD/PCDFs, and as a result bind to the AhR, inducing
toxic effects(65). Similar to PCDD/PCDFs, the position and number of chlorine
substituents determines the degree of AhR-mediated activity(66).
Polyhalogenated anthracenes and phenanthrenes are two variations on a three
membered aromatic ring structure substituted with bromine and/or chlorine atoms. These
compounds are released to the environment through waste incineration and electronic
waste recycling. With structural similarities to PCDD/PCDFs, Br- and Cl-
anthracenes/phenanthrenes are of similar environmental and toxicological concern. A
recent review by Sun et al. summarizes the identification of Br- and Cl-
anthracenes/phenanthrenes in air, water, snow, soil, sediment, fly ash, and biota(67). As
hydrophobic species, sediment partitioning and bioaccumulation are likely environmental
fates of Br- and Cl-anthracenes/phenanthrenes(67). It has been shown that photochemical
reactions may also alter the toxicity or bioavailability of Cl-anthracenes/phenanthrenes in
the environment(68, 69). Like PCDD/PCDFs, the AhR-mediated effects of PAHs with 1-
3 halogen substituents depend on halogen number and position(70, 71). It has been
observed that Br-anthracenes/phenanthrenes do not induce significant AhR-mediated
responses(72).
13
1.5 Characterization of Thermal Decomposition Products
1.5.1 Pyrolysis and Combustion Studies
Studying the complex nature of thermal decomposition presents numerous
challenges. These processes involve highly variable temperatures and pressures, multiple
phases, unstable species, and transient processes(1). Experiments may be performed with
online or off-line analysis, and designed to address specific research interests. Major
parameters of interest in experimental design include temperature, pressure, flow, particle
size, and species concentrations(1).
1.5.1.1 Online Analysis There are several methods for studying thermal decomposition directly. Two
widely used techniques are thermogravimetric analysis (TGA) and pyrolysis-gas
chromatography mass spectrometry (Py-GC/MS).
In thermal analysis using TGA, a small solid sample is placed on the arm of a
recording microbalance positioned inside a furnace. The temperature of the furnace is
increased while nitrogen, oxygen, or air flows over the sample. The system records
changes in sample weight with temperature, indicating temperatures and rates at which a
material thermal decomposition occurs(1). While providing important mechanistic
information, the relevance of this technique to uncontrolled fires is limited as a result of
the heating rate, amount of material, and lack of heat feedback. In fires, heating rates of
10-100oC/s are common, but TGA heating rates are often around 10oC/min. Fires also
contain large quantities of mixed materials and energy is resupplied at varying rates to
thermally decomposing material. These conditions are not simulated in TGA(73).
In contrast to TGA, Py-GC/MS provides specific information of volatile and
semi-volatile thermal decomposition products. In this technique, a flow-through reactor
(furnace-type pyrolyzer) is directly connected to the analytical GC column via a heated
transfer line. Mass spectrometry provides the required sensitivity to capture thermal
decomposition products produced from a small sample (up to a few milligrams). The
most common ionization method used in Py-GC/MS is positive electron impact
ionization (EI+). Some disadvantages of this technique include its inability to identify
non-volatile thermal decomposition products, large molecules, and highly polar analytes.
14
For online analysis, pyrolysis is usually performed in an inert gas that couples as the
GC/MS carrier gas. Thermal decomposition in other gases may be performed as long as
gas-exchange is performed prior to GC/MS analysis(16).
1.5.1.2 Off-line Analysis Indirect analysis of thermal decomposition products offers more diverse options
in terms of experimental conditions, sampling design and sample analysis. In the interest
of studying uncontrolled fires where incomplete combustion occurs, thermal
decomposition experiments are performed in air. Experiments may range from
laboratory-scale quartz tube furnace studies to large-scale simulation fire or incinerator
studies. A variety of sampling media is used to collect air-borne particulates, gases, and
post-combustion debris and ash. Extraction methods may be targeted to identify specific
analytes of interest and trace-level contaminants. Sample extracts may be analyzed by
multiple analytical methods to provide complementary and confirmatory information. In
some cases, these types of combustion experiments provide a more realistic glimpse of
the thermal decomposition products formed in fires that may enter the environment and to
which firefighters and civilians are exposed. However, it should be noted that these
studies are almost always qualitative with limited reproducibility.
1.5.2 Non-targeted Analysis
In the field of non-targeted environmental sample analysis, a quote by American
politician and businessman, Donald Rumsfeld, is often referenced:
“There are known knowns. These are things we know that we know.
There are known unknowns. That is to say, there are things that we
know we don’t know. But there are also unknown unknowns. There
are things we don’t know we don’t know.”
In terms of characterizing the thermal decomposition products in an uncontrolled fire
event, there are also known knowns, known unknowns, and unknown unknowns. It is
known that a fire containing hydrocarbon fuel will burn to produce a measurable quantity
of carbon dioxide, which has environmental effects as a greenhouse gas. It is also known
that PCDD/PCDFs form in combustion processes, however their formation mechanisms
15
have been debated in the literature(13) and congener and isomer profiles vary widely
between samples. In these known unknown cases, targeted analytical methods, such as
gas chromatography tandem mass spectrometry (GC-MS/MS), are used to measure
concentrations of known analytes in environmental samples alongside commercially
available analytical standards. There are many unknowns associated with the complex
chemical mixtures produced in uncontrolled fires. The chemical composition of the
mixture and the sources and mechanisms of its formation are unknown. The mixture may
contain previously unidentified persistent and toxic environmental contaminants for
which there are no developed targeted analytical methods. In these cases, a non-targeted
analytical approach can provide an overview of the sample mixture contents and direct
further targeted analysis through identification of analytes of interest. Two variations of
non-targeted analyses include mass defect filtering of high-resolution mass spectra and
comprehensive two-dimensional gas chromatography high-resolution mass spectrometry
(GCxGC-HRToF).
1.5.2.1 Mass Defect Filtering of High-Resolution Mass Spectra High-resolution mass spectrometry describes a mass analyzer with a resolving
power greater than 10,000. The resolving power is defined as m/#m, where m is the
accurate mass of a mass spectral peak and #m is the full width at half maximum height
(FWHM) of that peak. This value refers to the ability of the mass analyzer to distinguish
between two ions of different elemental composition(74). The error in mass measurement
accuracy is expressed in parts per million (ppm) and calculated as: [(exact mass –
accurate mass) / (exact mass)] $ 106. Precision is enhanced when more data points are
collected per mass spectral peak width and the mass spectrum is internally calibrated on
multiple ions of known m/z values(74).
Two types of high-resolution mass spectrometry applied in environmental
analyses are time of flight mass spectrometry (HRToF) and Fourier Transform ion
cyclotron resonance mass spectrometry (FTICR-MS).
The basic principle behind HRToF is that ions of different m/z travel at different
velocities in a field-free drift path towards a detector(75). In this technique, ions are
trapped following ionization and then accelerated simultaneously into the drift region,
where they are separated by m/z (Figure 1.5)(76). Equation 1 describes the time (T)
16
required for an ion of m/z, accelerated by a voltage (U), to travel a distance (s), where e is
electron charge. In this relationship, T is proportional to the square route of m/z.
! !!!!!!!!!! ! !!!!"
!!!
Therefore, ions of higher m/z will require more time to reach the detector than
ions of lower m/z(75). The resolving power (m/#m) of HRToF is equivalent to T/2#t,
where T is the total ion flight time and t is the mass spectral peak width. Instrument
resolution can be enhanced by increasing the ion flight time(74) and recent developments
have achieved resolving powers greater than 100,000(77). Advantages of HRToF include
a very large mass range, high sensitivity, relatively low-cost, and accurate mass
information(75, 77).
Figure 1.5. Simplified depiction of an HRToF.
FTICR-MS uses the principles of cyclotron motion and Fourier transform to
create a high performance analytical technique that achieves high sensitivity, resolving
power, and mass accuracy. In this technique, ions are trapped and accumulated following
ionization. Once the ion accumulation period is complete, a shutter opens to allow ions to
travel through an ion guide, where they are focused upon entering the uniform field of a
superconducting magnet. The analyzer cell is positioned at the end of the ion guide, in the
center of the superconducting magnet (Figure 1.6)(78). Within the analyzer cell, ion
cyclotron motion results from the force acting on charged particles moving perpendicular
to a magnetic field and is the basis for FTICR-MS. Excitation of ions in cyclotron motion
is used to accelerate ions to a larger orbital radius, to increase ion kinetic energy in order
Detector
Field-Free Drift Region
Ion Acceleration
Region
Ionization Region
Sample Inlet
17
to perform ion dissociation or ion-molecule reactions, and to remove ions from the
instrument(79). When ions are cycling at a larger orbit, receiver plates on the analyzer
cell detect image current signals. These signals, which are sinusoidal frequencies
corresponding to the number of ions present, undergo Fourier transform analysis(78). In a
uniform magnetic field (B), an ion of mass (m) and charge (z), will have a circular path
with an angular velocity (!) or cyclotron frequency (v) (Equation 2)(80):
(2) ! = zB/m = 2"v
An important feature of Equation 2 is that the mass-to-charge ratio (m/z) is inversely
proportional to the cyclotron frequency, which is also independent of ion velocity(78).
This feature allows FTICR-MS to achieve much higher resolution than other methods
that rely on ion energy and velocity for detection, such as HRToF. While most mass
analyzers have resolving powers ranging 2000 to 20,000, FTICR-MS can achieve a
resolving power greater than 300,000(78).
Figure 1.6. Simplified depiction of an FTICR-MS.
High-resolution mass spectra of complex mixtures obtained by HRToF and
FTICR-MS contain thousands of mass spectral peaks. These complex data sets are
difficult to interpret when presented as traditional mass spectra. Mass defect filtering of
high-resolution mass spectral data is an effective approach to visually resolve and
interpret this complex chemical information(81-83). The mass defect is the difference
between the exact mass and the nominal mass of a chemical compound(83). By plotting
the mass defect of each mass spectral peak versus the corresponding nominal mass,
Ionization Region
Sample Inlet
Superconducting Magnet
Ion Guide
Shutter
Analyzer Cell
Ion Accumulation
Region
18
thousands of mass spectral peaks are resolved to reveal congener and homologue
patterns(81, 82). The mass defect plot can be altered to meet specific research interests by
changing the mass scale on which mass defects are calculated. By converting mass
spectral data from the International Union of Pure and Applied Chemistry (IUPAC) mass
scale (C = 12.0000 Da) to the Kendrick mass scale (CH2 = 14.0000 Da), hydrocarbon
homologous series are revealed(81, 82, 84). More recently, this approach has been used
to identify halogenated contaminants in environmental samples using a mass scale based
on a hydrogen for chlorine substitution (-H/+Cl = 34.0000 Da)(85, 86). The visual
resolution coupled with accurate mass information in a mass defect plot allows for
identification of novel contaminants in complex mixtures.
1.5.2.2 GCxGC-HRToF Comprehensive two-dimensional gas chromatography (GCxGC) is a high-
resolution chromatography technique incorporating two GC columns. The first dimension
column is typically a nonpolar stationary phase and the second dimension column is
shorter in length with a more polar stationary phase. The two columns are connected by a
modulator interface that controls the flow of analytes from the primary column to the
secondary column. The modulator acts to trap and refocus subsets of analytes as they
travel from the primary to the secondary column(87). The advantage of GCxGC over
one-dimensional GC is improved peak separation through high peak capacity, which
helps to reduce matrix interferences in complex samples(88). As an example, consider a
one-dimensional GC peak that contains three different unresolved analyte peaks. The
modulator essentially slices the peak into smaller scale chromatographic separations that
take place on the second dimension column, meanwhile preserving the separation
achieved in the first dimension column. This data can be transformed into a two-
dimensional chromatogram that effectively resolves three different analyte peaks
unresolved by one-dimensional GC (Figure 1.7).
19
Figure 1.7. Separation of three unresolved peaks using GCxGC separation followed by transformation to two-dimensional chromatogram.
When GCxGC is coupled to HRToF, the superior chromatographic separation,
wide mass range and accurate mass information allows for elemental composition
estimation for unknown peaks and isomer separation. One drawback is the relatively slow
acquisition rate of HRToF relative to the narrow modulated peaks in GCxGC. Without
achieving adequate measurements over a peak, some loss in chromatographic integrity
may occur. While recently used as a qualitative tool for identifying halogenated
contaminants in environmental samples(89-91), GCxGC-HRToF shows promise for
quantitative non-targeted analysis of complex environmental mixtures.
1.5.3 Applications of Non-targeted Analysis
While mass defect filtering of high-resolution mass spectra and GCxGC-HRToF
are described here as non-targeted analytical techniques, it should be noted that in some
respects, there is a targeted element to these analyses. For instance, a crude Soxhlet
extraction of ash with toluene will target only water-insoluble compounds and GC
analysis is limited to volatile and semi-volatile species. Despite these considerations,
these techniques remain non-targeted in the fact that they provide a broad scan of a
complex chemical mixture with superior chromatographic and mass spectral resolution.
Modulation
Transformation
1st Dimension
2nd
Dim
ensi
on
GC
GCxGC
20
This allows for elemental composition estimation via accurate mass and identification of
novel compounds. In addition, these analyses produce complex data sets that may be
revisited at any time for further re-evaluation and identification of unknowns.
Non-targeted analytical approaches are valuable to the characterization of thermal
decomposition products in fires for their ability to interpret complex chemical mixtures of
unknown composition. Currently these techniques are primarily qualitative in nature,
however when used as an initial screening tool, they can direct further studies and
development of targeted analytical methods.
1.6 Goals and Hypotheses
With the widespread use of halogenated materials, the complexity of thermal
decomposition mechanisms and products in a modern fire presents numerous unknowns.
There is a good sense of the environmental fate and toxicity of many known combustion-
derived products, such as PCDD/PCDFs, however with new patents and changing
markets, the contents of accidental and uncontrolled fires will continually change. Non-
targeted analytical approaches, such as mass defect filtering and GCxGC-HRToF, offer a
top-down approach to environmental analysis of complex mixtures. Without
preconceived notions of sample composition, novel environmentally persistent and toxic
contaminants may be revealed, directing further studies and targeted method
development. The motivation behind this thesis was to investigate the thermal
decomposition mechanisms and products of halogenated materials using non-targeted
analytical techniques, and to assess the environmental relevance of these thermal
processes.
On a small scale, understanding the thermal decomposition of halogenated
materials through laboratory studies contributes to understanding their behaviour in large-
scale fires. A previous laboratory study identified environmentally persistent
perfluorinated carboxylic acids (PFCAs) as thermal decomposition products of
PTFE(52), suggesting other fluoropolymers may thermally decompose to form
contaminants of concern. Chapters 2 and 3 investigate the thermal decomposition of
fluoropolymers PCTFE and PVDF, respectively, through quartz tube furnace
experiments. Chapter 2 also serves to showcase the potential of mass defect filtering to
21
characterize complex environmental samples. To place these fluoropolymer thermal
decomposition studies in a larger context, Chapter 3 includes a review of fluoropolymer
thermal decomposition mechanisms, products and their environmental relevance.
A variety of halogenated contaminants have been measured in accidental fires(3,
5) and informal electronic waste recycling processes(92). The targeted analytical
approaches used in these studies are limited in their assessment of these complex
contaminant mixtures as a whole. In Chapter 4, the non-targeted analytical approach of
FTICR-MS and mass defect filtering was applied to screen for halogenated contaminants
in large-scale electronics and household fire simulation experiments performed at the Fire
and Emergency Services Training Institute (FESTI) in Toronto, Ontario. This analysis
provided clues to the halogenated materials initially present in the fires and identified
contaminants of concern produced therein.
The 1997 Plastimet Inc. plastics recycling plant fire in Hamilton, Ontario released
PCDD/PCDFs to the surrounding area(3). Previously, PCDD/PCDFs in incinerator fly
ash have been associated with PXDD/PXDFs(45), suggesting possible PXDD/PXDF
formation in the Plastimet fire. Chapter 5 describes development of a targeted GC-
MS/MS method for identifying PXDD/PXDFs, which was applied in the analysis of
archived soil collected from the Plastimet fire.
Following an accidental or uncontrolled fire, contaminants are released to the
surrounding environment. Access to archived Plastimet fire soil provided a unique
opportunity to investigate bioaccumulative contaminants in these complex combustion-
derived mixtures. Chapter 6 describes an aquatic exposure study and the identification of
bioaccumulative halogenated contaminants using FTICR-MS and mass defect filtering, as
well as GCxGC-HRToF. A targeted analysis of bioaccumulative PXDD/PXDFs using
GC/MS/MS was also performed.
The final chapter summarizes the contributions of this research to understanding
thermal decomposition mechanisms and products of halogenated materials, as well as
their environmental relevance. Future research directions to improve and integrate non-
targeted approaches in environmental analysis are discussed.
22
1.7 References
1. M. Lackner, A. B. Palotas, F. Winter, Combustion: From Basics to Applications
(Wiley-VCH, Weinheim, Germany, 2013).
2. K. M. Butler, G. W. Mulholland, Generation and transport of smoke components. Fire Technol. 40, 149–176 (2004).
3. Ontario Ministry of the Environment, Plastimet Inc. fire, Hamilton, Ontario, July 9-12, 1997. https://ia700309.us.archive.org/24/items/plastimetincfire00sochuoft/plastimetincfire00sochuoft.pdf (1997).
4. P. J. Landrigan et al., Health and environmental consequences of the World Trade Center disaster. Environ. Health Persp. 112, 731–739 (2004).
5. P. J. Lioy et al., Characterization of the dust/smoke aerosol that settled east of the World Trade Center (WTC) in lower Manhattan after the collapse of the WTC 11 September 2001. Environ. Health Persp. 110, 703–714 (2002).
6. J. H. Offenberg et al., Persistent organic pollutants in the dusts that settled across Lower Manhattan after September 11, 2001. Environ. Sci. Technol. 37, 502–508 (2003).
7. R. Zeig-Owens et al., Early assessment of cancer outcomes in New York City firefighters after the 9/11 attacks: an observational cohort study. Lancet 378, 898–905 (2011).
8. S. D. Shaw et al., Persistent organic pollutants including polychlorinated and polybrominated dibenzo-p-dioxins and dibenzofurans in firefighters from Northern California. Chemosphere 91, 1386–1394 (2013).
9. American Society for Testing and Materials International, Terminology of Fire Standards (ASTM International, West Conshohocken, PA, 2012).
10. I. Ericsson, R. P. Lattimer, Pyrolysis nomenclature. J. Anal. Appl. Pyrol. 14, 219–221 (1989).
11. S. R. Turns, An Introduction to Combustion: Concepts and Applications J. P. Holman, J. Lloyd, Eds. (McGraw-Hill, Boston, MA, ed. 2, 2000).
12. Canadian Council of Ministers of the Environment, National Guidelines for Hazardous Waste Incineration Facilities (CCME, 1992).
13. B. R. Stanmore, The formation of dioxins in combustion systems. Combust. Flame 136, 398–427 (2004).
14. National Fire Protection Association, All About Fire. http://www.nfpa.org/press-
15. I. Aracil, R. Font, J. A. Conesa, Semivolatile and volatile compounds from the pyrolysis and combustion of polyvinyl chloride. J. Anal. Appl. Pyrol. 74, 465–478 (2005).
16. S. C. Moldoveanu, Pyrolysis of Organic Molecules with Applications to Health and Environmental Issues (Elsevier, Amsterdam, Netherlands, 2010).
17. J. Scheirs, Fluoropolymers (Rapra Technology Limited, Shawbury, UK, 2001).
20. S. Ebnesajjad, Introduction to Fluoropolymers (Elsevier, Waltham, MA 2013).
21. S. G. Patrick, Practical Guide to Polyvinyl Chloride (Smithers Rapra, Shawbury, UK, 2005).
22. M. Alaee, P. Arias, A. Sjodin, A. Bergman, An overview of commercially used brominated flame retardants, their applications, their use patterns in different countries/regions and possible modes of release. Environ. Int. 29, 683–689 (2003).
23. A. B. Morgan, J. W. Gilman, An overview of flame retardancy of polymeric materials: application, technology, and future directions. Fire Mater. 37, 259–279 (2013).
24. E. Eljarrat, D. Barcelo, Eds., Brominated Flame Retardants (Springer-Verlag, Berlin, 2011).
25. World Health Organization (WHO), Brominated diphenyl ethers (International Programme on Chemical Safety, Environmental Health Criteria 162, 1994).
26. United Nations, Adoption of Amendments to Annexes A, B, and C (Stockholm Convention on Persistent Organic Pollutants, Stockholm, Sweden, 2009; http://chm.pops.int/Portals/0/download.aspx?d=UNEP-POPS-TREATY-NOTIF-CN524-2009.En.pdf).
27. U.S. Environmental Protection Agency, Polybrominated Diphenyl Ethers (PBDEs) Action Plan (U.S. Environmental Protection Agency, 2009; http://www.epa.gov/oppt/existingchemicals/pubs/actionplans/pbdes_ap_2009_1230_final.pdf).
28. W. Ma, G. Hoffmann, M. Schirmer, G. Chen, V. S. Rotter, Chlorine
24
characterization and thermal behavior in MSW and RDF. J. Hazard. Mater. 178, 489–498 (2010).
29. Z. Wang et al., HCl Formation from RDF pyrolysis and combustion in a spouting-moving bed reactor. Energy Fuels 16, 608–614 (2002).
30. G. Montaudo, C. Puglisi, Evolution of aromatics in the thermal degradation of poly(vinyl chloride): a mechanistic study. Polym. Degrad. Stab. 33, 229–262 (1991).
31. D. Wang, X. Xu, S. Chu, Q. X. Li, Polychlorinated naphthalenes and other chlorinated tricyclic aromatic hydrocarbons emitted from combustion of polyvinylchloride. J. Hazard. Mater. B138, 273–277 (2006).
32. R. Font, A. Gálvez, J. Moltó, A. Fullana, I. Aracil, Formation of polychlorinated compounds in the combustion of PVC with iron nanoparticles. Chemosphere 78, 152–159 (2010).
33. J. Vehlow, Reduction of dioxin emissions from thermal waste treatment plants: a brief survey. Rev. Environ. Sci. Biotechnol. 11, 393–405 (2012).
34. P. Weber, E. Dinjus, L. Stieglitz, The role of copper(II) chloride in the formation of organic chlorine in fly ash. Chemosphere 42, 579–582 (2001).
35. I. Dalrymple et al., An integrated approach to electronic waste (WEEE) recycling. Circuit World 33, 52–58 (2007).
36. R. Weber, B. Kuch, Relevance of BFRs and thermal conditions on the formation pathways of brominated and brominated-chlorinated dibenzodioxins and dibenzofurans. Environ. Int. 29, 699–710 (2003).
38. M. Altarawneh, B. Z. Dlugogorski, A Mechanistic and kinetic study on the formation of PBDD/Fs from PBDEs. Environ. Sci. Technol. 47, 5118–5127 (2013).
39. W. Schafer, K. Ballschmiter, Monobromo-polychloro-derivatives of benzene, biphenyl, dibenzofurane and dibenzodioxine formed in chemical-waste burning. Chemosphere 15, 755–763 (1986).
40. K. H. Schwind, J. Hosseinpour, H. Thoma, Brominated/chlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 17, 1875–1884 (1988).
41. G. W. Sovocool et al., Bromo- and bromochloro-polynuclear aromatic hydrocarbons, dioxins and dibenzofurans in municipal incinerator fly ash. Biomed.
25
Environ. Mass 15, 669–676 (1988).
42. J. Hosseinpour, K. H. Schwind, H. Thoma, Brominated/chlorinated dibenzo-p-dioxins and furans: synthesis of standards and detection in fly ash from a municipal waste incinerator. Chemosphere 19, 109–114 (1989).
43. R. L. Harless, R. G. Lewis, D. D. McDaniel, A. E. Dupuy Jr, Identification of bromo/chloro dibenzo-p-dioxins and dibenzofurans in ash samples. Chemosphere 18, 201–208 (1989).
44. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
45. H. Y. Tong, S. J. Monson, M. L. Gross, L. Q. Huang, Monobromopolychlorodibenzo-p-dioxins and dibenzofurans in municipal waste incinerator flyash. Anal. Chem. 63, 2697–2705 (1991).
46. G. Söderström, S. Marklund, PBCDD and PBCDF from incineration of waste-containing brominated flame retardants. Environ. Sci. Technol. 36, 1959–1964 (2002).
47. C. Rittmeyer, J. Vehlow, Decomposition of organohalogen compounds in municipal solid waste incineration plants. part I: chlorofluorocarbons. Chemosphere 26, 2129–2138 (1993).
48. R. Weber, H. Hagenmaier, Synthesis and analysis of mixed chlorinated-fluorinated dibenzo-p-dioxins and dibenzofurans and assessment of formation and occurrence of the fluorinated and chlorinated-fluorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 34, 13–28 (1997).
49. S. Sakai, M. Hiraoka, K. Shiozaki, Thermal behaviour of chlorofluorocarbons (CFCs) and formation of PCDD/PCDFs & PFDD/PFDFs. Organohalogen Compd. 23, 347–352 (1995).
50. T. Yamada, P. H. Taylor, R. C. Buck, M. A. Kaiser, R. J. Giraud, Thermal degradation of fluorotelomer treated articles and related materials. Chemosphere 61, 974–984 (2005).
51. P. H. Taylor, T. Yamada, R. C. Striebich, J. L. Graham, R. J. Giraud, Investigation of waste incineration of fluorotelomer-based polymers as a potential source of PFOA in the environment. Chemosphere 110, 17–22 (2014).
52. D. A. Ellis, S. A. Mabury, J. W. Martin, D. C. G. Muir, Thermolysis of fluoropolymers as a potential source of halogenated organic acids in the environment. Nature 412, 321–324 (2001).
53. D. A. Ellis, J. W. Martin, D. C. G. Muir, S. A. Mabury, The use of 19F NMR and
26
mass spectrometry for the elucidation of novel fluorinated acids and atmospheric fluoroacid precursors evolved in the thermolysis of fluoropolymers. Analyst 128, 756–764 (2003).
54. G. Montaudo, C. Puglisi, E. Scamporrino, D. Vitalini, Correlation of thermal degradation mechanisms: polyacetylene and vinyl and vinylidene polymers. J. Polym. Sci. 24, 301–316 (1986).
55. C. Rappe, Sources and environmental concentrations of dioxins and related compounds. Pure & Appl. Chem. 68, 1781–1789 (1996).
56. United Nations, Listing of POPs in the Stockholm Convention. http://chm.pops.int/TheConvention/ThePOPs/ListingofPOPs/tabid/2509/Default.aspx (2008).
57. A. Schecter, Ed., Dioxins and Health: Including Other Persistent Organic Pollutants and Endocrine Disruptors (John Wiley & Sons, ed. 3, 2012).
58. H.-R. Buser, Brominated and brominated/chlorinated dibenzodioxins and dibenzofurans: potential environmental contaminants. Chemosphere 16, 713–732 (1987).
59. W. Chatkittikunwong, C. S. Creaser, Bromo-, bromochloro- and chloro-dibenzo-p-dioxins and dibenzofurans in incinerator fly ash. Chemosphere 29, 559–566 (1994).
60. H. Preud'homme, M. Potin-Gautier, Optimization of accelerated solvent extraction for polyhalogenated dibenzo-p-dioxins and benzo-p-furans in mineral and environmental matrixes using experimental designs. Anal. Chem. 75, 6109–6118 (2003).
61. A. Schecter, L. Birnbaum, J. J. Ryan, J. D. Constable, Dioxins: An overview. Environ. Res. 101, 419–428 (2006).
62. A. Poland, J. C. Knutson, 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related halogenated aromatic hydrocarbons: examination of the mechanism of toxicity. Annu. Rev. Pharmacol. 22, 517–554 (1982).
63. L. S. Birnbaum, D. F. Staskal, J. J. Diliberto, Health effects of polybrominated dibenzo-p-dioxins (PBDDs) and dibenzofurans (PBDFs). Environ. Int. 29, 855–860 (2003).
64. J. L. Domingo, Polychlorinated naphthalenes in animal aquatic species and human exposure through the diet: a review. J. Chromatogr. A 1054, 327–334 (2004).
65. J. Falandysz, Polychlorinated naphthalenes: an environmental update. Environ. Pollut. 101, 77–90 (1998).
27
66. A. L. Blankenship et al., Relative potencies of individual polychlorinated naphthalenes and halowax mixtures to induce Ah receptor-mediated responses. Environ. Sci. Technol. 34, 3153–3158 (2000).
67. J.-L. Sun, H. Zeng, H.-G. Ni, Halogenated polycyclic aromatic hydrocarbons in the environment. Chemosphere 90, 1751–1759 (2013).
68. K. Sankoda, K. Nomiyama, T. Kuribayashi, R. Shinohara, Halogenation of polycyclic aromatic hydrocarbons by photochemical reaction under simulated tidal flat conditions. Polycycl. Aromat. Comp. 33, 236–253 (2013).
69. P. P. Fu, Q. Xia, X. Sun, H. Yu, Phototoxicity and environmental transformation of polycyclic aromatic hydrocarbons (PAHs)—light-induced reactive oxygen species, lipid peroxidation, and DNA damage. J. Environ. Sci. Heal. C 30, 1–41 (2012).
70. P. P. Fu, L. S. Von Tungeln, L.-H. Chiu, Z. Y. Own, Halogenated-polycyclic aromatic hydrocarbons: a class of genotoxic environmental pollutants. J. Environ. Sci. Heal. C 17, 71–109 (1999).
71. T. Ohura, M. Morita, M. Makino, T. Amagai, K. Shimoi, Aryl hydrocarbon receptor-mediated effects of chlorinated polycyclic aromatic hydrocarbons. Chem. Res. Toxicol. 20, 1237–1241 (2007).
72. Y. Horii et al., Relative potencies of individual chlorinated and brominated polycyclic aromatic hydrocarbons for induction of aryl hydrocarbon receptor-mediated responses. Environ. Sci. Technol. 43, 2159–2165 (2009).
73. C. L. Beyler, M. M. Hirschler, in The SFPE Handbook of Fire Protection Engineering, (Society of Fire Protection Engineers, Quincy, MA, 2002), pp. 1–110–1–131.
74. F. Xian, C. L. Hendrickson, A. G. Marshall, High resolution mass spectrometry. Anal. Chem. 84, 708–719 (2012).
75. J. H. Gross, Mass Spectrometry (Springer-Verlag, Berlin, ed. 2, 2011).
76. M. Guilhaus, Principles and instrumentation in time-of-flight mass spectrometry. J. Mass. Spectrom. 30, 1519–1532 (1995).
77. W. R. Plaß, T. Dickel, C. Scheidenberger, Multiple-reflection time-of-flight mass spectrometry. Int. J. Mass Spectrom. 349-350, 134–144 (2013).
78. R. T. McIver, J. R. McIver, Fourier Transform Mass Spectrometry: Principles and Applications (IonSpec Corporation, Lake Forest, CA, 2006).
79. A. G. Marshall, C. L. Hendrickson, G. S. Jackson, Fourier transform ion cyclotron resonance mass spectrometry: a primer. Mass Spectrom. Rev. 17, 1–35 (1998).
28
80. S. Guan, A. G. Marshall, Ion traps for Fourier transform ion cyclotron resonance mass spectrometry: principles and design of geometric and electric configurations. Int. J. Mass Spectrom. 146/147, 261–296 (1995).
81. C. A. Hughey, C. L. Hendrickson, R. P. Rodgers, A. G. Marshall, K. Qian, Kendrick mass defect spectrum: a compact visual analysis for ultrahigh-resolution broadband mass spectra. Anal. Chem. 73, 4676–4681 (2001).
82. Z. Wu, R. P. Rodgers, A. G. Marshall, Two- and three-dimensional van Krevelen diagrams: a graphical analysis complementary to the Kendrick mass plot for sorting elemental compositions of complex organic mixtures based on ultrahigh-resolution broadband Fourier transform ion cyclotron resonance mass measurements. Anal. Chem. 76, 2511–2516 (2004).
83. L. Sleno, The use of mass defect in modern mass spectrometry. J. Mass. Spectrom. 47, 226–236 (2012).
84. A. G. Marshall, R. P. Rodgers, Petroleomics: the next grand challenge for chemical analysis. Acc. Chem. Res. 37, 53–59 (2004).
85. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass Spectrom. 21, 1918–1921 (2010).
86. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
87. L. Mondello, Ed., Comprehensive Chromatography in Combination with Mass Spectrometry (John Wiley & Sons, New Jersey, 2011).
88. E. Skoczy!ska, P. Korytár, J. de Boer, Maximizing chromatographic information from environmental extracts by GCxGC-ToF-MS. Environ. Sci. Technol. 42, 6611–6618 (2008).
89. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
90. S. Hashimoto et al., Global and selective detection of organohalogens in environmental samples by comprehensive two-dimensional gas chromatography–tandem mass spectrometry and high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3799–3810 (2011).
91. S. Hashimoto et al., Selective extraction of halogenated compounds from data measured by comprehensive multidimensional gas chromatography/high resolution time-of-flight mass spectrometry for non-target analysis of environmental and
29
biological samples. J. Chromatogr. A 1282, 183–189 (2013).
92. M. H. Wong et al., Export of toxic chemicals – A review of the case of uncontrolled electronic-waste recycling. Environ. Pollut. 149, 131–140 (2007).
30
2 CHAPTER TWO
Using Mass Defect Plots as a Discovery Tool to Identify Novel Fluoropolymer Thermal Decomposition Products
Anne L. Myers, Karl J. Jobst, Scott A. Mabury, and Eric J. Reiner
Published as: J. Mass Spectrom. 2014, 49, 291-296. Contributions: Anne L. Myers was responsible for designing and performing thermal decomposition experiments, sample extraction, LC-MS/MS analysis, and data interpretation. Karl J. Jobst performed sample analysis by FTICR-MS with assistance from Anne L. Myers. Anne L. Myers prepared this manuscript with editorial comments provided by Karl J. Jobst, Scott A. Mabury, and Eric J. Reiner.
A partial section of the resulting plot (Figure 2.1) displays a series of ions
highlighted as black diamonds. This series was easily recognized amongst numerous
other peaks based on its distinct grid pattern. The top and bottom rows consist of four and
five clusters of peaks, respectively, and each cluster is horizontally separated by 50 Da
(CF2). In the vertical direction, the rows are separated by 16 Da, which corresponds to the
exchange of a fluorine for a chlorine (-F/+Cl). Each cluster corresponds to a
monoisotopic peak and related 37Cl isotopic peaks. Based on this grid pattern,
37
identification of one molecular formula can lead to the assignment of a whole series of
ions. This would be difficult to achieve using a mass spectrum depicted in the traditional
manner (intensity vs. mass).
Accurate mass measurements and isotope ratios were used to determine that the
highlighted peaks of Figure 2.1 corresponded to perchlorofluoroalkanoic acids (PXCAs,
X=Cl,F). Consider for example the first cluster of peaks in the top row series, the
monoisotopic peak of which is located at coordinates (194.9232, 0.9356). The mass
defect of 0.9356, calculated using Equations 1 and 2, corresponds to 194.9356 Da on the
CF2 mass scale and 194.9232 Da on the IUPAC mass scale. The cluster has two (37Cl)
isotopic peaks with relative abundances that correspond to a molecular ion containing
two chlorine atoms. Using this information, the peak was identified as a PXCA with
molecular formula C2F3Cl2CO2- (0.5ppm deviation). In the same vein, all the ions in the
row at mass defect 0.935 correspond to PXCAs varying in chain length from three to six
carbons, each containing two chlorine atoms.
Moving vertically down the plot, a fluorine atom is exchanged for a chlorine
atom. The exact IUPAC masses of fluorine and chlorine are 18.9984 Da and 34.9689 Da,
respectively. Using Equations 1 and 2, these are transformed first into masses on the CF2
mass scale, and then mass defects based on the CF2 mass scale (0.9996 and 0.9711 for
fluorine and chlorine respectively). Therefore, exchanging a fluorine for a chlorine atom
results in an IUPAC mass shift of 34.9689 – 18.9984 = 15.9705 Da and a CF2 mass
defect shift of 0.9711 - 0.9996 = -0.0284. As an example, the coordinates of C3F5Cl2CO2-
in Figure 2.1 are (244.9199, 0.9355) and the corresponding elemental composition
resulting from a fluorine for chlorine exchange is C3F4Cl3CO2- at (260.8904, 0.9071),
representing a relative positioning on the plot of (+15.9705, -0.0284).
38
Figure 2.1. A partial section of the mass defect plot of polar products produced in the thermal decomposition of PCTFE at 400oC. The series of ions (black diamonds) represent elemental compositions corresponding to perchlorofluoroalkanoic aicds (PXCAs, X=Cl,F).
The remaining elemental compositions and series were identified in the mass
spectrum using this approach, as demonstrated in the mass defect plot of Figure 2.2.
Based on the assigned elemental compositions, two series of ions (Cn-1X2n-3CO2- and Cn-
1X2n-5CO2-), were proposed to be PXCAs containing one and two double bond
equivalents (Figure 2.2). Whether this unsaturation corresponds to alkenoic structures,
ringed structures, or a mixture of both cannot be established from this experiment.
However, we are quite confident of the carboxylic acid functional group, which is
consistent with the anion-exchange cleanup procedure and the strong signal response in
ESI-. In addition to the three PXCA series, a variety of sodium and proton-bound cluster
ions containing PXCAs and sodium chloride ion pairs were identified (Figure 2.2).
Aliphatic and alkenoic PXCAs have been proposed previously as PCTFE thermal
decomposition products by Ellis et al.(11, 13, 14), on the basis of carefully performed
GC-MS and 19F NMR experiments. The results presented here show that mass defect
filtering with high-resolution mass spectrometry offers an alternative, straightforward
means of identifying the same class of perhalogenated compounds. Excluding sodium
and proton-bound cluster ions, 29 PXCA congener groups were identified in Figure 2.2.
39
Each identified molecular ion (compound) may well represent a multitude of structural
isomers.
Figure 2.2. Mass defect plot based on CF2 mass scale of polar products produced in the thermal decomposition of PCTFE at 400oC (X=Cl,F). Number of carbon atoms ranged from two to 18 and number of chlorine atoms ranged from one to eight.
40
2.4.2 Identification of PCTFE Thermal Decomposition Products at 800oC
Unlike the aliphatic acids observed at 400oC, aromatic species were the dominant
thermal decomposition products at 800oC. The complex nature of the polar extract posed
a challenge in assigning elemental compositions to all peaks in the spectrum. However,
series of peaks corresponding to Cl/F aromatic carboxylic acids were identified in the
mass defect plot (not shown). Pentafluorobenzoic acid was confirmed as a thermal
decomposition product of PCTFE by LC-MS/MS using the characteristic loss of CO2
(Figure S2).
The non-polar extract from the thermal decomposition of PCTFE at 800oC was
introduced to the FTICR-MS via GC and DIP. The observed products ranged in mass
from m/z 150-450 in the GC experiment and m/z 150-750 for the DIP experiment. This
difference may be attributed to higher mass analytes being retained on the GC injector
liner and column. For this reason, the mass defect plot has been constructed using the DIP
mass spectral data.
The mass defect plot for non-polar PCTFE thermal decomposition products at
800oC is based on the fluorine for chlorine substitution (-F/+Cl = 16.0000 Da) mass scale.
The mass spectral data set was converted to the F/Cl mass scale using Equation 3, and the
mass defect values within the converted data set were calculated using Equation 4.
Plotting the mass defect values based on the F/Cl mass scale versus the original IUPAC
mass scale m/z values yielded the mass defect plot shown in Figure 2.3.
pyrenes/fluoranthenes, and benzo[a]pyrenes. Even higher mass elemental compositions
were identified with proposed structures containing seven and eight Cl/F substituted
aromatic rings.
The remaining 10 series shown in Figure 2.3 are expected to be fragment ions
formed in the EI source during analysis. Two key fragmentation patterns were identified
and attributed to the consecutive losses of chlorine. For example, Figure 2.4 shows the
mass spectrum of ions C16F7Cl3+·, a Cl/F-pyrene isomer, which was obtained from the
analysis of the non-polar extract by GC-FTICR-MS. Aside from a strong molecular ion,
the spectrum also displays fragment ions resulting from chlorine losses. From the mass
defect plot, various series could be attributed to these types of fragment ions. The mass
spectrum of Figure 2.4 also displays a peak at m/z 479.8920, which corresponds to the
elemental composition, C17F9Cl3+·. We suspect that this is another molecular ion with an
additional CF2 group.
The identification of X-PAHs as non-polar thermal decomposition products was
supported by a GC-FTICR-MS experiment with octafluoronaphthalene. Corresponding
chromatographic peaks in the standard and non-polar extract were observed at 12.2
minutes, confirming octafluoronaphthalene as a thermal decomposition product of
PCTFE at 800oC (Figure S3).
42
Figure 2.3. Mass defect plot of non-polar products produced in the thermal decomposition of PCTFE at 800oC (X = Cl,F). Bold elemental compositions correspond to proposed X-PAH structures. Non-bold elemental compositions are proposed X-PAH fragment ions from chlorine loss or X-PAH molecular ions with an additional CF2 group.
43
Figure 2.4. (a) Mass spectrum of a trichloroheptafluoropyrene isomer obtained by GC-FTICR-MS analysis of non-polar extract for PCTFE thermal decomposition at 800oC. Labelled peaks correspond to the molecular ion and two proposed fragment ions. (b) Mass defect plot obtained using a DIP-FTICR-MS experiment, which shows series corresponding to X-pyrene isomers and fragment ions.
2.4.3 Thermal Decomposition Mechanism of PCTFE
An examination of fluorine/chlorine ratios in the thermal decomposition products
of PCTFE gave some indication of the decomposition mechanism. For both the PXCAs
produced at 400oC and the X-PAHs produced at 800oC, the fluorine to chlorine ratio
increased as the unsaturation increased, indicating that chlorine is lost in the formation of
double bonds (Figure 2.5). Considering that the carbon-chlorine bond has the lowest
average bond energy (339 kJ/mol), relative to the carbon-fluorine and carbon-carbon
bonds (453 kJ/mol and 347 kJ/mol, respectively)(21) present in PCTFE, the initial loss of
chlorine in thermal decomposition is expected. In addition, the bond energy of a carbon-
carbon bond increases with fluorine substitution(22), further emphasizing the relative
weakness of the carbon-chlorine bond in PCTFE. Ellis et al. proposed a thermal
decomposition mechanism for polytetrafluoroethylene (PTFE, [-CF2-CF2-]n) in which
perfluorinated carboxylic acids (PFCAs) were formed from CF2 carbene units, and
suggested a similar decomposition pathway for PCTFE(13). However, as a result of the
weak carbon-chlorine bond, there are likely several other thermal decomposition
pathways that lead to the unsaturated products observed in this study.
44
Figure 2.5. (a) The fluorine/chlorine ratios of the most abundant X-PAH molecular ions for each series identified in the thermal decomposition of PCTFE at 800oC. The trend line indicates the increasing F/Cl ratio with increasing unsaturation. (b) The average fluorine/chlorine ratios and associated standard deviations for PXCA molecular ions identified containing zero (n=15), and two (n=6) double bond equivalents (DBE).
2.5 Conclusion
Ultrahigh-resolution mass spectrometry and mass defect plots were used to
identify novel thermal decomposition products of PCTFE in a complex mixture. At
400oC, PCTFE thermal decomposition products consisted of PXCAs and Cl/F
alkanes(13, 14, 16), while at 800oC, X-PAHs were the major non-polar products
identified. This study showcases the potential of using mass defect plots to identify
unknown halogenated compounds in a complex mixture. This approach may be
performed using any high-resolution technique, such as time of flight mass spectrometry.
!
"
#
$
%
&
'
!"#$%&'()*
+*,(%-./01'0(%234-5%+*$67/$'8%9*0,
!
#
%
'
(
"!
!)*+, ")*+, #)*+,
!"#$%&'()*
42#-,
':
.:
45
Numerous studies have investigated PAHs with bromine or chlorine
substituents(23), as well as perfluorinated carboxylic acids(24, 25). However, few studies
have examined either PXCAs(26-28) or mixed halogenated PAHs(29-31), and there are
no known reports of X-PAHs (X=Cl,F) in the environment. These are contaminants of
concern for their potential environmental persistence and toxicity. With the increasing
use of PCTFE in commercial applications, we suspect the formation of the compounds
identified in this study through thermal processing, disposal and fires. Further
investigation into the environmental prevalence of PXCAs and X-PAHs and their
potential environmental risks is therefore warranted.
2.6 Acknowledgements
We thank John Ford, Jack O’Donnell, and Dr. Leo Yeung from the University of
Toronto for assistance in design and assembly of the thermal decomposition apparatus.
We are also grateful to Dr. Vince Taguchi from the Ontario Ministry of the Environment
for access to the FTICR-MS.
2.7 References
1. A. T. Lebedev, Ed., Comprehensive environmental mass spectrometry (ILM
Publications, Lichfield, 2012).
2. P. H. Howard, D. C. G. Muir, Identifying new persistent and bioaccumulative organics among chemicals in commerce. Environ. Sci. Technol. 44, 2277–2285 (2010).
3. C. A. Hughey, C. L. Hendrickson, R. P. Rodgers, A. G. Marshall, K. Qian, Kendrick mass defect spectrum: a compact visual analysis for ultrahigh-resolution broadband mass spectra. Anal. Chem. 73, 4676–4681 (2001).
4. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass Spectrom. 21, 1918–1921 (2010).
5. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
6. Z. Wu, R. P. Rodgers, A. G. Marshall, Two- and three-dimensional van Krevelen
46
diagrams: a graphical analysis complementary to the Kendrick mass plot for sorting elemental compositions of complex organic mixtures based on ultrahigh-resolution broadband Fourier transform ion cyclotron resonance mass measurements. Anal. Chem. 76, 2511–2516 (2004).
7. L. Sleno, The use of mass defect in modern mass spectrometry. J. Mass. Spectrom. 47, 226–236 (2012).
8. E. Kendrick, A mass scale based on CH2=14.0000 for high resolution mass spectrometry of organic compounds. Anal. Chem. 35, 2146–2154 (1963).
9. A. G. Marshall, R. P. Rodgers, Petroleomics: the next grand challenge for chemical analysis. Acc. Chem. Res. 37, 53–59 (2004).
12. J. Scheirs, Fluoropolymers (Rapra Technology Limited, Shawbury, UK, 2001).
13. D. A. Ellis, S. A. Mabury, J. W. Martin, D. C. G. Muir, Thermolysis of fluoropolymers as a potential source of halogenated organic acids in the environment. Nature 412, 321–324 (2001).
14. D. A. Ellis, J. W. Martin, D. C. G. Muir, S. A. Mabury, The use of 19F NMR and mass spectrometry for the elucidation of novel fluorinated acids and atmospheric fluoroacid precursors evolved in the thermolysis of fluoropolymers. Analyst 128, 756–764 (2003).
15. F.-Y. Hshieh, H. D. Beeson, A preliminary study on the toxic combustion products testing of polymers used in high-pressure oxygen systems (NASA Technical Report, 2004; http://naca.larc.nasa.gov/search.jsp?R=20100041342&qs=N%3D4294504844).
16. S. Zulfiqar, M. Zulfiqar, M. Rizvi, A. Munir, I. C. McNeill, Study of the thermal degradation of polychlorotrifluoroethylene, poly(vinylidene fluoride) and copolymers of chlorotrifluoroethylene and vinylidene fluoride. Polym. Degrad. Stabil. 43, 423–430 (1994).
17. D. Wang, X. Xu, S. Chu, Q. X. Li, Polychlorinated naphthalenes and other chlorinated tricyclic aromatic hydrocarbons emitted from combustion of polyvinylchloride. J. Hazard. M. B138, 273–277 (2006).
18. S. Taniyasu et al., Analysis of trifluoroacetic acid and other short-chain perfluorinated acids (C2–C4) in precipitation by liquid chromatography–tandem
47
mass spectrometry: comparison to patterns of long-chain perfluorinated acids (C5–C18). Anal. Chim. Acta 619, 221–230 (2008).
19. M. E. Aydin, S. Ozcan, A. Tor, Ultrasonic solvent extraction of persistent organic pollutants from airborne particles. Clean. 35, 660–668 (2007).
20. C. Bruce, M. A. Shifman, P. Miller, E. E. Gulcicek, Probabilistic enrichment of phosphopeptides by their mass defect. Anal. Chem. 78, 4374–4382 (2006).
21. M. S. Silberberg, Chemistry: The Molecular Nature of Matter and Change (McGraw-Hill, New York, NY, USA, 2012).
22. R. D. Chambers, Fluorine in Organic Chemistry (CRC Press, Boca Raton, FL, USA, 2004).
23. J.-L. Sun, H. Zeng, H.-G. Ni, Halogenated polycyclic aromatic hydrocarbons in the environment. Chemosphere 90, 1751–1759 (2013).
24. M. Houde, A. O. De Silva, D. C. G. Muir, R. J. Letcher, Monitoring of perfluorinated compounds in aquatic biota: an updated review. Environ. Sci. Technol. 45, 7962–7973 (2011).
25. L. Ahrens, Polyfluoroalkyl compounds in the aquatic environment: a review of their occurrence and fate. J. Environ. Monit. 13, 20 (2011).
26. M. L. Hanson, P. K. Sibley, S. A. Mabury, D. C. Muir, K. R. Solomon, Chlorodifluoroacetic acid fate and toxicity to the macrophytes Lemna Gibba, Myriophyllum Spicatum, and Myriophyllum Sibiricum in aquatic microcosms. Environ. Toxicol. Chem. 20, 2758–2767 (2001).
27. M. L. Hanson, K. R. Solomon, Haloacetic acids in the aquatic environment. Part I: macrophyte toxicity. Environ. Pollut. 130, 371–383 (2004).
28. M. L. Hanson, K. R. Solomon, Haloacetic acids in the aquatic environment. Part II: ecological risk assessment. Environ. Pollut. 130, 385–401 (2004).
29. H.-G. Ni, E. Y. Zeng, Environmental and human exposure to soil chlorinated and brominated polycyclic aromatic hydrocarbons in an urbanized region. Environ. Toxicol. Chem. 31, 1494–1500 (2012).
30. G. W. Sovocool et al., Bromo- and bromochloro-polynuclear aromatic hydrocarbons, dioxins and dibenzofurans in municipal incinerator fly ash. Biomed. Environ. Mass. 15, 669–676 (1988).
31. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
48
3 CHAPTER THREE
Identification and Environmental Relevance of Fluoropolymer Thermal Decomposition Products
Anne L. Myers, Scott A. Mabury and Eric J. Reiner Contributions: Anne L. Myers was responsible for designing and performing thermal decomposition experiments, sample extraction, as well as data analysis and interpretation. Anne L. Myers prepared this chapter with editorial comments provided by Eric J. Reiner.
49
3.1 Abstract
Fluoropolymers, such as polytetrafluoroethylene (PTFE), polyvinylidene fluoride
(PVDF), and polychlorotrifluoroethylene (PCTFE), have a wide range of industrial and
commercial applications as a result of their uniquely high thermal and chemical inertness.
Some fluoropolymer applications include electrical wire and cable insulations,
waterproof clothing, architectural coatings, and transparent films. Despite their high heat
resistance, fluoropolymers are susceptible to thermal decomposition to produce
potentially environmentally persistent and toxic compounds, including perhalogenated
carboxylic acids (PXCAs, X=Cl,F) and perhalogenated polycyclic aromatic hydrocarbons
(X-PAHs, X=Cl,F). Such thermal decomposition could occur during processing,
recycling, backyard barrel waste burning or accidental fire events. In the present study,
the thermal decomposition products of PVDF were investigated using a non-targeted
analytical technique incorporating high-resolution mass spectrometry and mass defect
filtering. Products included a wide variety of polyfluorinated polycyclic aromatic
hydrocarbons (F-PAHs), including polyfluorinated naphthalenes (PFNs) and
polyfluorinated dibenzofurans (PFDFs). These findings are included in a review of
thermal decomposition products and mechanisms of PTFE, PVDF, and PCTFE in an
effort to assess their environmental relevance.
3.2 Introduction
Fluoropolymers are widely used in industrial and commercial applications for
their high thermal and chemical inertness, low coefficient of friction, and dielectric
properties(1). In 2010, the global fluoropolymer market revenue was estimated at almost
$6 billion, with dominant consumer markets in North American and Asia-Pacific
regions(2). Polytetrafluoroethylene ([CF2CF2]n, PTFE) is the most widely used
fluoropolymer, followed by polyvinylidene fluoride ([CH2CF2]n, PVDF), and to a lesser
extent, polychlorotrifluoroethylene ([CFClCF2]n, PCTFE). The fluoropolymer industry
first began with the discovery of PCTFE in 1937(3) and PTFE in 1938(4), from which
other fluoropolymers were developed, including PVDF in 1948(2). Since then, the
corresponding monomers have also been incorporated in a significant number of
copolymers(2).
50
Fluoropolymer chemical structure significantly alters the processing and
commercial applications. With a continuous sustainable use temperature of 260oC(2),
PTFE applications range widely to include electrical wire insulation, pipes, liners,
bearings, non-stick coatings, waterproof clothing, and medical devices(1). Despite lower
heat resistance, PVDF and PCTFE have an advantage over PTFE in that they are melt
processible(2). Major PVDF applications include architectural coatings, wire and cable
insulations, and semi-conductor manufacturing(1). The presence of a chlorine atom in
PCTFE allows greater intermolecular attraction and limits close packing of molecular
chains. As a result, PCTFE has greater mechanical strength over PTFE and may be
processed to form transparent films through reduced crystallinity(1). Important
applications of PCTFE involve gas and moisture barrier films, particularly in
pharmaceutical blister packaging and cryogenic seals and gaskets(1, 2). Despite the high
thermal resistance of these fluoropolymers in commercial applications, at higher
temperatures they thermally decompose to produce a wide variety of potentially toxic and
environmentally persistent contaminants.
Over the course of a fluoropolymer product life cycle, it may undergo a variety of
thermal treatments, through manufacturing, disposal, recycling, and/or accidental fire.
While the fluoropolymer industry provides safe handling guidelines for manufacturers(2,
5), with their widespread use, it is important to consider the significance of fluoropolymer
thermal decomposition as a source of halogenated contaminants to the environment. To
date, other halogenated polymers have received considerably more attention in this
regard. For example, waste incineration of polyvinyl chloride (PVC) has been shown to
form polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/PCDFs)(2, 6). For
fluoropolymers, the majority of thermal decomposition studies stem from an interest in
testing polymer performance, ensuring worker safety, and optimizing chemical recycling
parameters(1, 7, 8). However, the release of fluoropolymer thermal decomposition
products may also be relevant to the informal electronic waste recycling practices in
countries such as China, India and Ghana, where scrap is burned in the open air(9, 10).
Open backyard barrel burning and accidental fires may also emit fluoropolymer thermal
decomposition products to the environment.
51
Numerous laboratory studies have investigated the thermal decomposition
products of PTFE and PVDF, with fewer reports on PCTFE. Many of these studies were
performed in inert atmospheres which provide valuable mechanistic and product
information, however thermal decomposition processes are often accelerated in oxidizing
atmospheres (air or oxygen)(11), which is more relevant to fire. The analytical
instrumentation used varies, but largely involves a pyrolyzer chamber coupled to a gas
chromatograph mass spectrometer (Py-GC-MS) with electron impact (EI) ionization. Off-
line methods are also used in which thermal decomposition products are collected via
sorptive materials, cold traps, or solvents, then analyzed by mass spectrometry or other
techniques(12). In comparing results of fluoropolymer thermal decomposition studies it is
important to consider the range of parameters used and their influence on the studies’
findings. Some of these parameters include sample mass, temperature, heating duration,
heating rate, flow of pyrolysis gas, and quality of sample(12). Despite variations in
experimental design, the current literature has repeatedly identified certain fluoropolymer
thermal decomposition products. The fluoropolymer industry lists hydrogen fluoride
The relative abundance of mass spectral peaks corresponding to PVDF thermal
decomposition products showed a pattern indicative of the thermal decomposition
mechanism proposed in previous studies, which generates polyene sequences through
dehydrofluorination(24-28). Figure 3.3 demonstrates the relative abundance of specific
congeners in relation to the elemental composition hydrogen/fluorine ratio (H:F). For the
PFN, PFDF, F-anthracene/phenanthrene, F-pyrene/fluoranthene, and F-tetracene series,
the most prevalent congener group was that containing an equal number of hydrogen and
fluorine atoms (H:F = 1:1). These results correspond to the major PVDF thermal
decomposition products observed by Montaudo et al.(18). For PFB and F-
fluorene/phenalene series, the most prevalent congener group corresponded to an H:F
ratio of 6:4 and 4:6, respectively.
Thermal decomposition products and congener abundance profiles were
confirmed in the additional thermal decomposition experiment performed at a faster
temperature ramp (data not shown). The only exceptions were for PFBs and F-
fluorenes/phenalenes. The PFB congener group with an H:F of 5:5 was approximately
two times more abundant than the congener group with an H:F of 6:4. For the F-
fluorenes/phenalenes, only congener groups with H:F ratios of 6:4 and 5:5 were
observed. Figure 3.3. Relative abundance of mass spectral peaks corresponding to specific congeners of PFNs, PFDFs, PFBs, F-fluorenes/phenalenes (F-FLO/PHNs), F-anthracenes/phenanthrenes (F-ANT/PHEs), F-pyrenes/fluoranthenes (F-PYR/FLUs), and F-tetracenes identified on the x-axis by the elemental composition H/F ratio. Congeners with H/F ratios of 1:1 were generally the most prevalent within each congener group, indicating loss of HF during thermal decomposition of PVDF.
polymer (PFA), chlorotrifluoroethylene vinylidene fluoride copolymer (CTFE-VDF), and
tetrafluoroethylene hexafluoropropylene vinylidene fluoride terpolymer (THV)(2). A
study of the thermal decomposition of CTFE-VDF under inert atmosphere observed
similar products as the corresponding monomers with an increase in hydrogen chloride
(HCl) production(27). Similarly, thermal decomposition products of ECTFE in air
resembled those of PCTFE(36). These studies indicate that there may be thermal
decomposition marker compounds common to fluoropolymers and their copolymers.
As described here, fluoropolymer thermal decomposition and corresponding
products are clearly complex. The contents, temperature, and various conditions under
which a thermal decomposition event occurs, greatly alters its potential environmental
impact. It is unlikely that products from laboratory thermal decomposition of an
individual fluoropolymer will match those of a large-scale combustion event
incorporating a variety of materials. Indeed, full-scale fire studies involving
perfluoropolymers have reported lower smoke toxicity compared to laboratory
studies(87). In addition, the acute effects of gases such as CO and HF may outweigh the
chronic health effects of X-PAHs. Toxicity aside, it is unknown to what extent
environmentally persistent fluoropolymer thermal decomposition products are released
under various combustion conditions. By measuring emissions at potential sources using
non-targeted analytical techniques, it may be possible to identify whether the
fluoropolymer thermal decomposition products discussed here, or other novel products,
present environment or health concerns. This information could help in initiatives to
protect fire fighters and others exposed to fluoropolymer thermal decomposition
products.
69
3.5 Acknowledgements
We thank John Ford and Jack O’Donnell from the University of Toronto for
assistance in design and assembly of the thermal decomposition apparatus. We thank
Derek Jackson for assistance with GC-MS analysis and Karl Jobst for assistance with
FTICR-MS analysis. We are also grateful to Dr. Vince Taguchi from the Ontario
Ministry of the Environment for access to the FTICR-MS.
3.6 References
1. J. Scheirs, Fluoropolymers (Rapra Technology Limited, Shawbury, UK, 2001).
2. S. Ebnesajjad, Introduction to Fluoropolymers (Elsevier, Waltham, MA 2013).
3. I. G. Farbenindustrie Aktiengesellschaft, Manufacture of Polymerisation Products (Patent 465520, London, UK, 1937).
4. R. J. Plunkett, Tetrafluoroethylene polymers (U.S. Patent 2230654, Wilmington, DE, 1941).
5. PlasticsEurope, Guide for the Safe Handling of Fluoropolymer Resins. http://www.plasticseurope.org/cust/documentrequest.aspx?DocID=51888 (2011).
6. T. Shibamoto, A. Yasuhara, T. Katami, in Reviews of Environmental Contamination and Toxicology, (Springer, New York, NY, 2007), vol. 190, pp. 1–41.
7. C. M. Simon, W. Kaminsky, Chemical recycling of polytetrafluoroethylene by pyrolysis. Polym. Degrad. Stabil. 62, 1–7 (1998).
8. E. Meissner, A. Wróblewska, E. Milchert, Technological parameters of pyrolysis of waste polytetrafluoroethylene. Polym. Degrad. Stabil. 83, 163–172 (2004).
9. A. Sepúlveda et al., A review of the environmental fate and effects of hazardous substances released from electrical and electronic equipments during recycling: examples from China and India. Environ. Impact Asses. 30, 28–41 (2010).
10. M. Akormedi, E. Asampong, J. N. Fobil, Working conditions and environmental exposures among electronic waste workers in Ghana. Int. J. Occup. Env. Heal. 19, 278–286 (2013).
11. C. L. Beyler, M. M. Hirschler, in The SFPE Handbook of Fire Protection Engineering, (Society of Fire Protection Engineers, Quincy, Mass., 2002), pp. 1–110–1–131.
70
12. S. C. Moldoveanu, Pyrolysis of Organic Molecules with Applications to Health and Environmental Issues (Elsevier, Amsterdam, The Netherlands, 2010).
13. A. L. Myers, K. J. Jobst, S. A. Mabury, E. J. Reiner, Using mass defect plots as a discovery tool to identify novel fluoropolymer thermal decomposition products. J. Mass. Spectrom. 49, 291–296 (2014).
14. M. E. Aydin, S. Ozcan, A. Tor, Ultrasonic solvent extraction of persistent organic pollutants from airborne particles. Clean. 35, 660–668 (2007).
15. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass Spectrom. 21, 1918–1921 (2010).
16. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
17. M. M. Hirschler, Effect of oxygen on the thermal decomposition of poly(vinylidene fluoride). Eur. Polym. J. 18, 463–467 (1982).
18. G. Montaudo, C. Puglisi, E. Scamporrino, D. Vitalini, Correlation of thermal degradation mechanisms: polyacetylene and vinyl and vinylidene polymers. J. Polym. Sci. 24, 301–316 (1986).
19. S.-S. Choi, Y.-K. Kim, Microstructural analysis of poly(vinylidene fluoride) using benzene derivative pyrolysis products. J. Anal. Appl. Pyrol. 96, 16–23 (2012).
20. R. D. Chambers, Fluorine in Organic Chemistry (CRC Press, Boca Raton, FL, USA, 2004).
21. R. Weber, D. Schrenk, H. J. Schmitz, A. Hagenmaier, H. Hagenmaier, Polyfluorinated dibenzodioxins and dibenzofurans - synthesis, analysis, formation and toxicology. Chemosphere 30, 629–639 (1995).
22. D. Herzke, R. Thiel, W. D. Rotard, D. Neubert, Kinetics and organotropy of some polyfluorinated dibenzo-p-dioxins and dibenzofurans (PFDD/PFDF) in rats. Life Sci. 71, 1475–1486 (2002).
23. M. I. Bruce, M. A. Thomas, Fluorinated tropylium ions in the mass spectra of some polyfluoroaromatic compounds. Org. Mass Spectrom. 1, 417–423 (1968).
24. N. N. Loginova, L. Y. Madorskaya, N. K. Podlesskaya, Relations between the thermal stability of partially fluorinated polymers and their structure. Polymer Science USSR 25, 2995–3000 (1983).
25. J. Lonfei, W. Jingling, X. Shuman, Mechanisms of pyrolysis of fluoropolymers. J. Anal. Appl. Pyrol. 10, 99–106 (1986).
71
26. G. Montaudo, C. Puglisi, Evolution of aromatics in the thermal degradation of poly(vinyl chloride): a mechanistic study. Polym. Degrad. Stabil. 33, 229–262 (1991).
27. S. Zulfiqar, M. Zulfiqar, M. Rizvi, A. Munir, I. C. McNeill, Study of the thermal degradation of polychlorotrifluoroethylene, poly(vinylidene fluoride) and copolymers of chlorotrifluoroethylene and vinylidene fluoride. Polym. Degrad. Stabil. 43, 423–430 (1994).
28. M. Rizvi, A. Munir, S. Zulfiqar, M. Zulfiqar, Pyrolysis gas-chromatographic studies of homopolymers and copolymers of chlorotrifluoroethylene with vinylidene fluoride and methyl methacrylate. J. Thermal Anal. 45, 1597–1604 (1995).
29. R. K. Steunenberg, G. H. Cady, Pyrolysis of fluorocarbons. J. Am. Chem. Soc. 74, 4165–4168 (1952).
30. L. A. Errede, The application of simple equations for calculating bond dissociation energies to thermal degradation of fluorocarbons. J. Org. Chem. 27, 3425–3430 (1962).
31. D. A. Ellis, S. A. Mabury, J. W. Martin, D. C. G. Muir, Thermolysis of fluoropolymers as a potential source of halogenated organic acids in the environment. Nature 412, 321–324 (2001).
32. E. E. Lewis, M. A. Naylor, Pyrolysis of polytetrafluoroethylene. J. Am. Chem. Soc. 69, 1968–1970 (1947).
33. H. Arito, R. Soda, Pyrolysis products of polytetrafluoroethylene and polyfluoroethyleneepropylene with reference to inhalation toxicity. Ann. Occup. Hyg. 20, 247–255 (1977).
34. S. Morisaki, Simultaneous thermogravimetry-mass spectrometry and pyrolysis-gas chromatography of fluorocarbon polymers. Thermochim. Acta 25, 171–183 (1978).
35. A. N. García, N. Viciano, R. Font, Products obtained in the fuel-rich combustion of PTFE at high temperature. J. Anal. Appl. Pyrol. 80, 85–91 (2007).
36. D. A. Ellis, J. W. Martin, D. C. G. Muir, S. A. Mabury, The use of 19F NMR and mass spectrometry for the elucidation of novel fluorinated acids and atmospheric fluoroacid precursors evolved in the thermolysis of fluoropolymers. Analyst 128, 756–764 (2003).
37. A. J. Pidduck, Mass spectrometric analysis of halogenated polymers. J. Anal. Appl. Pyrol. 7, 215–229 (1985).
38. G. Acerboni, N. R. Jensen, B. Rindone, J. Hjorth, Kinetics and products formation of the gas-phase reactions of tetrafluoroethylene with OH and NO3 radicals and
72
ozone. Chem. Phys. Lett. 309, 364–368 (1999).
39. R. E. Kupel, M. Nolan, R. G. Keenan, M. Hite, L. D. Scheel, Mass spectrometric identification of decomposition products of polytetrafluoroethylene and polyfluoroethylenepropylene. Anal. Chem. 36, 386–389 (1964).
40. L. A. Wall, J. D. Michaelsen, Thermal decomposition of polytetrafluoroethylene in various gaseous atmospheres. J. Res. Nat. Bur. Stand. 56, 27–34 (1956).
41. K. L. Paciorek, R. H. Kratzer, J. Kaufman, Oxidative thermal degradation of polytetrafluoroethylene. J. Polym. Sci. 11, 1465–1473 (1973).
42. J. A. Conesa, R. Font, Polytetrafluoroethylene decomposition in air and nitrogen. Polym. Eng. Sci. 41, 2137–2147 (2001).
43. H. A. Watson, H. J. Stark, L. E. Sieffert, L. B. Berger, Decomposition temperatures of polytetrafluoroethylene and polymonochlorotrifluoroethylene as indicated by halogen liberation (United States Department of the Interior, Report of Investigations 4756, Pittsburgh, PA, 1950).
44. M. L. O'Shea, C. Morterra, M. J. D. Low, Spectroscopic studies of carbons. XVII. Pyrolysis of polyvinylidene fluoride. Mater. Chem. Phys. 26, 193–205 (1990).
45. T. Yamada, P. H. Taylor, R. C. Buck, M. A. Kaiser, R. J. Giraud, Thermal degradation of fluorotelomer treated articles and related materials. Chemosphere 61, 974–984 (2005).
46. P. H. Taylor, T. Yamada, R. C. Striebich, J. L. Graham, R. J. Giraud, Investigation of waste incineration of fluorotelomer-based polymers as a potential source of PFOA in the environment. Chemosphere 110, 17–22 (2014).
47. P. M. Lemieux, C. C. Lutes, J. A. Abbott, K. M. Aldous, Emissions of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans from the open burning of household waste in barrels. Environ. Sci. Technol. 34, 377–384 (2000).
48. B. Hedman, M. Naslund, C. Nilsson, S. Marklund, Emissions of polychlorinated dibenzodioxins and dibenzofurans and polychlorinated biphenyls from uncontrolled burning of garden and domestic waste (backyard burning). Environ. Sci. Technol. 39, 8790–8796 (2005).
49. B. K. Gullett et al., PCDD/F, PBDD/F, and PBDE emissions from open burning of a residential waste dump. Environ. Sci. Technol. 44, 394–399 (2010).
50. K. M. Butler, G. W. Mulholland, Generation and transport of smoke components. Fire Technol. 40, 149–176 (2004).
51. B. B. Baker, M. A. Kaiser, Understanding what happens in a fire. Anal. Chem. 63,
73
79A–83A (1991).
52. M. Baasandorj et al., Rate coefficients for the gas-phase reaction of the hydroxyl radical with CH2%CHF and CH2%CF2. J. Phys. Chem. A 114, 4619–4633 (2010).
53. T. J. Wallington et al., The environmental impact of CFC replacements - HFCs and HCFCs. Environ. Sci. Technol. 28, 320A–326A (1994).
54. Ø. Hodnebrog et al., Global warming potentials and radiative efficiencies of halocarbons and related compounds: a comprehensive review. Rev. Geophys. 51, 300–378 (2013).
55. G. L. Kennedy Jr., Toxicology of fluorine-containing monomers. CRC Cr. Rev. Toxicol. 21, 149–170 (1990).
56. Ontario Ministry of Labour, Current occupational exposure limits for Ontario workplaces required under Regulation 833. http://www.labour.gov.on.ca/english/hs/pdf/ontario_oels.pdf (2013).
57. Z. Wang, I. T. Cousins, M. Scheringer, R. C. Buck, K. Hungerbühler, Global emission inventories for C4–C14 perfluoroalkyl carboxylic acid (PFCA) homologues from 1951 to 2030, part II: the remaining pieces of the puzzle. Environ. Int. 69, 166–176 (2014).
58. K. Prevedouros, I. T. Cousins, R. C. Buck, S. H. Korzeniowski, Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 40, 32–44 (2006).
59. D. J. Bowden, S. L. Clegg, P. Brimblecombe, The henry's law constant of trifluoroacetic acid and its partitioning into liquid water in the atmosphere. Chemosphere 32, 405–420 (1996).
60. D. J. Bowden, S. L. Clegg, P. Brimblecombe, The henry's law constants of the haloacetic acids. J. Atmos. Chem. 29, 85–107 (1998).
61. J. C. Boutonnet, et al., Environmental risk assessment of trifluoroacetic acid. Hum. Ecol. Risk Assess. 5, 59–124 (1999).
62. J. W. Martin et al., Detection of chlorodifluoroacetic acid in precipitation: a possible product of fluorocarbon degradation. Environ. Sci. Technol. 34, 274–281 (2000).
63. J. W. Martin et al., Airborne haloacetic acids. Environ. Sci. Technol. 37, 2889–2897 (2003).
64. C. P. Higgins, R. G. Luthy, Sorption of perfluorinated surfactants on sediments. Environ. Sci. Technol. 40, 7251–7256 (2006).
65. C. Lau et al., Perfluoroalkyl acids: A review of monitoring and toxicological
74
findings. Toxicol. Sci. 99, 366–394 (2007).
66. C. M. Butt, U. Berger, R. Bossi, G. T. Tomy, Levels and trends of poly- and perfluorinated compounds in the arctic environment. Sci. Total Environ. 408, 2936–2965 (2010).
67. M. L. Hanson, K. R. Solomon, Haloacetic acids in the aquatic environment. Part I: macrophyte toxicity. Environ. Pollut. 130, 371–383 (2004).
68. M. L. Hanson, K. R. Solomon, Haloacetic acids in the aquatic environment. Part II: ecological risk assessment. Environ. Pollut. 130, 385–401 (2004).
69. M. M. MacDonald Phillips, M. J. A. Dinglasan-Panlilio, S. A. Mabury, K. R. Solomon, P. K. Sibley, Fluorotelomer acids are more toxic than perfluorinated acids. Environ. Sci. Technol. 41, 7159–7163 (2007).
70. R. J. Mitchell, A. L. Myers, S. A. Mabury, K. R. Solomon, P. K. Sibley, Toxicity of fluorotelomer carboxylic acids to the algae Pseudokirchneriella subcapitata and Chlorella vulgaris, and the amphipod Hyalella azteca. Ecotox. Environ. Safe. 74, 2260–2267 (2011).
71. A. A. Rand, J. P. Rooney, C. M. Butt, J. N. Meyer, S. A. Mabury, Cellular toxicity associated with exposure to perfluorinated carboxylates (PFCAs) and their metabolic precursors. Chem. Res. Toxicol. 27, 42–50 (2014).
72. R. Weber, H. Hagenmaier, Synthesis and analysis of mixed chlorinated-fluorinated dibenzo-p-dioxins and dibenzofurans and assessment of formation and occurrence of the fluorinated and chlorinated-fluorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 34, 13–28 (1997).
73. P. Haglund, T. Alsberg, A. Bergman, B. Jansson, Analysis of halogenated polycyclic aromatic hydrocarbons in urban air, snow and automobile exhaust. Chemosphere 16, 2441–2450 (1987).
74. G. W. Sovocool et al., Bromo- and bromochloro-polynuclear aromatic hydrocarbons, dioxins and dibenzofurans in municipal incinerator fly ash. Biomed. Environ. Mass 15, 669–676 (1988).
75. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
76. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
77. J.-L. Sun, H. Zeng, H.-G. Ni, Halogenated polycyclic aromatic hydrocarbons in the
75
environment. Chemosphere 90, 1751–1759 (2013).
78. T. Ohura, Environmental behavior, sources, and effects of chlorinated polycyclic aromatic hydrocarbons. Sci. World J. 7, 372–380 (2007).
79. P. P. Fu, Q. Xia, X. Sun, H. Yu, Phototoxicity and environmental transformation of polycyclic aromatic hydrocarbons (PAHs)—light-induced reactive oxygen species, lipid peroxidation, and DNA damage. J. Environ. Sci. Heal. C 30, 1–41 (2012).
80. K. Sankoda, K. Nomiyama, T. Kuribayashi, R. Shinohara, Halogenation of polycyclic aromatic hydrocarbons by photochemical reaction under simulated tidal flat conditions. Polycycl. Aromat. Comp. 33, 236–253 (2013).
81. S. Sakai, M. Hiraoka, K. Shiozaki, Thermal behaviour of chlorofluorocarbons (CFCs) and formation of PCDD/PCDFs & PFDD/PFDFs. Organohalogen Compd. 23, 347–352 (1995).
82. J. Shi et al., Theoretical study on hydrophilicity and thermodynamic properties of polyfluorinated dibenzofurans. Chemosphere 84, 296–304 (2011).
83. C. Zhang, X. Sun, Gas-phase tropospheric chemistry of 2,3,7,8-tetrafuorinated dibenzo-p-dioxin. Sci. Total Environ. 468-469, 104–110 (2014).
84. T. Ohura, M. Morita, M. Makino, T. Amagai, K. Shimoi, Aryl hydrocarbon receptor-mediated effects of chlorinated polycyclic aromatic hydrocarbons. Chem. Res. Toxicol. 20, 1237–1241 (2007).
85. R. Weber, H. J. Schmitz, D. Schrenk, H. Hagenmaier, Metabolic degradation, inducing potency, and metabolites of fluorinated and chlorinated-fluorinated dibenzodioxins and dibenzofurans. Chemosphere 34, 29–40 (1997).
86. H.-J. Schmitz et al., 2,3,7,8-Tetrafluorodibenzo-p-dioxin: a potent agonist of the murine dioxin receptor. Environ. Toxico. Phar. 3, 105–113 (1997).
87. F. B. Clarke et al., The toxicity of smoke from fires involving perfluoropolymers: full-scale fire studies. J. Fire Sci. 10, 488–527 (1992).
76
4 CHAPTER FOUR
Screening for Halogenated Contaminants in Fire Samples Using Mass Defect Plots
Anne L. Myers, Karl J. Jobst, Kari Organtini, Sujan Fernando, Brian Ross, Brian McCarry, Frank Dorman, Scott A. Mabury and Eric J. Reiner
Contributions: Anne L. Myers led the planning and organization of the simulation fire experiments. Experimental design was initially conceived by Anne L. Myers and refined through collaborative discussion and shared expertise of all co-authors. Anne L. Myers, Karl J. Jobst, Kari Organtini, Sujan Fernando, Brian McCarry, Frank Dorman, and Eric J. Reiner carried out the experiments in a collaborative effort with fire fighters from FESTI. Anne L. Myers was responsible for water sample extraction, Kari Organtini was responsible for aluminum foil extraction, and Sujan Fernando was responsible for air sample extraction. Karl J. Jobst performed sample analysis by FTICR-MS with assistance from Anne L. Myers. Anne L. Myers prepared this chapter with editorial comments provided by Eric J. Reiner.
77
4.1 Abstract
Urban fire events release complex mixtures of contaminants to the surrounding
environment as a result of the wide array of polymers and flame-retardants used in
plastics, electronics and furniture. Heat resistant and flame retarding materials often
incorporate halogens to improve product performance, however at elevated temperatures
these materials thermally decompose to environmentally persistent and toxic halogenated
contaminants. Resulting complex mixtures pose an analytical challenge in assessing the
dangers associated with a fire site where burn conditions and contents are unknown. A
rapid non-targeted analytical approach to qualitatively identify unknown halogenated
contaminants in complex mixtures may provide an important first step to this assessment.
To this aim, recent studies have employed high-resolution mass spectrometry and mass
defect filtering. In this approach, large mass spectral data sets are simplified by
generating a mass defect plot from which series of halogenated contaminants are
revealed. In this study, complex mixtures were produced in simulation household and
electronics fires. Air, particulate, and water run-off samples were collected and analyzed
by Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FTICR-MS) and
mass defect filtering. In both fire simulations, polybrominated diphenyl ethers (PBDEs),
X=Br,Cl), and polybrominated anthracenes/phenanthrenes were identified. This study
demonstrates the merits of a mass defect plot screening approach to provide valuable
health and safety information to firefighters, while directing remediation measures.
4.2 Introduction
Air and particulate samples collected following the September 11, 2001 attack on
the World Trade Center in New York demonstrated the complex nature of contaminants
formed in urban fire events(1). In addition to the immense volume of glass fibers,
asbestos, lead, and cement dust released, halogenated contaminants such as
polybrominated diphenyl ethers (PBDEs), polychlorinated biphenyls (PCBs), and
polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/PCDFs) were identified(1-
3). Halogenated contaminants are released in urban fires because polymers and flame-
retardants incorporating bromine, chlorine, and fluorine are widely used in plastics,
78
electronics, cable and wire insulations, and furniture. While halogenated materials are
recognized for their superior heat resistance and flame retarding capabilities, at elevated
temperatures they may release environmentally persistent and toxic halogenated
contaminants. In the case of PBDE flame retardants, a fire event may lead to release of
PBDEs in their commercial form, or as corresponding thermal decomposition products,
such as polybrominated dibenzo-p-dioxins and dibenzofurans (PBDD/PBDFs)(4). The
complex nature of samples collected at urban fire sites is a result of thermal
decomposition of both natural and anthropogenic materials.
An array of persistent halogenated organic contaminants has been identified in
firefighter blood(5). While firefighters face a number of physical and acute chemical
(e.g., carbon monoxide) exposure risks, the long-term health effects of their exposure to
complex mixtures of halogenated contaminants is difficult to determine. A study of
firefighters involved in the 2001 World Trade Center fire showed increased incidences of
cancer over non-exposed firefighters, however the authors suggested this could be
attributed directly to known carcinogens, such as PCDD/PCDFs, and/or indirectly
through chronic inflammatory disorders(6). Understanding the environmental and health
impacts of halogenated contaminants formed in fires is also challenging due to large
variations in burn content and duration, temperature, and scale.
Previous studies of halogenated contaminants formed in fire events have used
targeted analytical approaches to measure established analytes of environmental concern.
While generating quantitative and specific information, targeted approaches are limited in
understanding complex contaminant mixtures as a whole. A rapid broad scanning
approach to qualitatively identify unknown halogenated contaminants in complex
mixtures may serve as an important first step to understanding the specific dangers of a
fire involving halogenated materials. High-resolution mass spectrometry and mass defect
filtering have been used recently to identify halogenated compounds in complex
mixtures(7-9). In this non-targeted approach, large mass spectral data sets are simplified
by generating a mass defect plot from which series of halogenated contaminants are
revealed(10, 11).
This study uses the high-resolution technique of Fourier Transform Ion Cyclotron
Resonance Mass Spectrometry (FTICR-MS) and mass defect filtering to characterize
79
complex samples obtained from simulation electronics and household fires with a focus
on halogenated contaminants.
4.3 Materials and Methods
4.3.1 Fire Simulation Experiments
Simulation household and electronics fire experiments were performed at the Fire
and Emergency Services Training Institute (FESTI) in Toronto, Ontario. The burn cell
was a converted steel shipping container used for firefighter training.
The household fire simulation contents included a mattress, a sofa chair, a wood
and plastic chair, five pillows, one carpet, and one television. The fire was ignited using a
road flare and allowed to burn for approximately 12 minutes, after which it was partially
extinguished and a smouldering period of approximately 30 minutes was performed,
followed by water extinguishment. Figure 4.1. shows the household fire simulation
contents before and after the burn experiment.
The electronics fire simulation contents included two televisions, one microwave,
two printers, two computer monitors, one computer, and cables. Since polymers
associated with electronics do not burn easily, a small amount of gasoline, a quarter bale
of straw, and a road flare were used to ignite the fire. After approximately 30 minutes of
burning, the fire was extinguished with water. Figure 4.2 shows the electronics fire
simulation contents before and after the burn experiment.
Several types of samples were collected throughout the fire simulations and are
depicted in Figure 4.3. Sheets of aluminum foil were taped to the interior walls of the
burn cell to capture airborne particulates. Air samples were collected by pumping air
from the burn cell through glass tubing (0.64 cm x 30 cm), a Teflon filter (37 mm, 2.0
µm, Concept Controls, Calgary, AB, CAN), and an XAD-2 cartridge (100mg/50mg,
Concept Controls, Calgary, AB, CAN) at approximately 2 L/min. The sampling media
was clipped to the door of the burn cell so the glass tubing extended into the cell, while
the filter and XAD were situated just outside the cell. Due to issues with melting and
80
Figure 4.1. Burn cell photos before and after household fire simulation.
Figure 4.2. Burn cell photos before and after electronics fire simulation.
blocked filter cartridges, filters and XADs were replaced as needed during the burn
experiments with firefighter assistance. During knockdown of the fire, run-off water was
collected in plastic catch basins and transferred to amber glass jars. Following
extinguishment, aluminum foil samples were transferred to plastic bags.
For the household fire simulation, a firefighter (Mike Hutchison, FESTI) stood
inside the burn cell to confirm the contents were smouldering. During this time, he wore
an air sampler, complete with pump, filter, and XAD for approximately 12 minutes.
81
Figure 4.3. Burn cell fire simulation and sample collection set-up.
4.3.2 Sample Preparation
4.3.2.1 Foil Samples Electronics fire simulation aluminum foil samples were extracted for non-polar
species by Kari-Lynn Organtini (Pennsylvania State University) using Soxhlet extraction
in toluene over 22 hours. A blank aluminum foil sample was extracted simultaneously.
Unfortunately aluminum foil samples from the household fire simulation were lost during
fire knockdown. Crude toluene extracts were stored in glass gas chromatography (GC)
vials at 4oC until analysis.
4.3.2.2 Air Samples Filter and XAD samples were extracted for non-polar species by Sujan Fernando
(McMaster University). The XAD cartridges were extracted by ultra sonication for 20
minutes in acetonitrile and then concentrated to 1 mL in toluene. The filters were
extracted on an accelerated solvent extractor (ASE 350, Dionex, Bannockburn, IL, USA)
using dichloromethane and then concentrated to 1 mL in toluene. The XAD and filter
extracts were combined in a 1:1 ratio to provide a representative total gas and particulate
phase air extract. Extracts were stored in glass GC vials at 4oC until analysis.
82
4.3.2.3 Water Run-off Samples Water run-off was extracted using a method based on that of Taniyasu et al. for
water soluble acids(12). Prior to extraction, run-off water samples were shaken, 40 mL
subsamples were transferred to 50 mL polypropylene (PP) falcon tubes, and particulate
was allowed to settle. Weak anion exchange solid phase extraction (WAX SPE)
cartridges (Waters Oasis WAX 6 cc, 150 mg, 30 µm) were conditioned with 4 mL 0.1%
NH4OH in methanol, 4 mL methanol, and 4mL deionized water. Run-off water
subsamples were loaded on SPE cartridges, followed by 4 mL of 25 mM ammonium
acetate buffer (pH4). Cartridges were dried and rinsed with 4 mL methanol before polar
analytes were eluted with 4 mL 0.1% NH4OH in methanol and collected. Extracts were
taken to dryness under a nitrogen stream and reconstituted in 1 mL methanol. Extracts
were transferred to 1.2 mL PP vials and stored at 4oC until analysis.
4.3.3 FTICR-MS Analysis and Mass Defect Filtering
The FTICR-MS methods have been previously published(9) and are reported in
Chapter 2. Briefly, both Varian 901 and 920 FTICR-MS mass spectrometers, positioned
in a Varian 9.4 Tesla superconducting magnet, were used to obtain ultrahigh-resolution
mass spectra using negative ion electrospray ionization (ESI-) and positive ion electron
ionization (EI) respectively.
This study examined five samples, which included water run-off, aluminum foil,
and air extracts from the electronics fire, and water run-off and air extracts from the
household fire. The water run-off extracts, expected to contain polar analytes, were
infused directly at a flow rate of 5 µL/min to a negative electrospray ionization (ESI-)
source. Of the extracts containing non-polar analytes, the foil extract was injected via the
direct insertion probe (DIP) without GC separation, while the air extracts were injected
on a Varian CP-3800 gas chromatograph (GC) with electron impact ionization (EI)
equipped with a DB-5HT GC column (15 m x 0.25 mm ID x 0.10 µm, J&W Scientific,
USA).
For the water run-off samples, the FTICR-MS was operated at a resolution of
80,000 at m/z 400 full width at half maximum (FWHM) and mass spectra were obtained
for a mass range of m/z 150-1,000. For the air and foil samples, the FTICR-MS was
83
operated at a resolution of 80,000 and 150,000, respectively, at m/z 400 FWHM. Mass
spectra for air and foil were obtained for a mass range of m/z 150-1,000 and 150-650,
respectively. The acquisition times were either 265 or 525 ms, and the cycle time was 1.5
s.
The FTICR-MS instrumental detection limit (IDL) for tetrachlorodibenzo-p-
dioxin injected via GC was 1 pg(7) and for perfluorinated carboxylic acids infused
directly to the ESI- source, the IDL was 1 pg. For data acquired for non-polar and polar
analytes, external mass calibration was performed using perfluorotributylamine (PFTBA)
and a mixture of perfluorosulfonic acids, respectively. Mass spectra were internally
calibrated on known background siloxane and phthalate ions. Elemental compositions
were determined from accurate mass measurements using the Varian Elemental
Composition Calculator. For each assignment, the measured masses were within 16 ppm
of theoretical values.
Mass spectra were interpreted using mass defect filtering on a H/Cl mass scale (–
H/+Cl = 34.0000 Da), which has been described previously(7, 8). Briefly, FTICR-MS
analysis of each extract produced a complex mass spectrum, from which the detected m/z
values were converted to the –H/+Cl mass scale, using Equation 1.
(1) H/Cl mass = !!"#$%!!"##!! ! !"!!!!"#$%
From the converted data set, mass defect values were calculated by subtracting the
nominal mass (rounded down) from the accurate mass (–H/+Cl mass scale), using
Equation 2.
(2) mass defect (H/Cl scale)
= accurate mass (H/Cl scale) – nominal mass (rounded down, H/Cl scale)
Resulting mass defect values based on the H/Cl mass scale (y-axis) were plotted against
the corresponding IUPAC mass scale m/z values (x-axis) to yield a mass defect plot.
84
4.3.4 Quality Assurance/Quality Control (QA/QC)
Blank samples were extracted alongside samples collected during fire simulations
to account for contamination. These included a water sample collected directly from the
firefighting hose, a clean aluminum foil sample, and an air sample collected from the
burn cell for 15 minutes prior to the experiments.
4.4 Results and Discussion
4.4.1 Thermal Decomposition Products
4.4.1.1 Run-off Water Mass defect plots corresponding to run-off water extracts did not demonstrate any
halogenated contaminants for either the household or electronics fire simulations. It is
suspected that if halogenated compounds were present, the samples may have been too
dilute for identification via FTICR-MS, as a result of the large amount of water used to
extinguish the fire relative to the 40mL run-off water sample that was extracted.
4.4.1.2 Household Fire The extract of the air sample collected on the firefighter that stood in the burn cell
during the smouldering period contained a variety of brominated compounds. Through
the mass defect plot in Figure 4.4, PBDEs (C12H10-nOBrn+!, n=2-7) and PBDFs (C12H8-
nOBrn+!, n=1-6) were both identified as products released in the household fire
simulation. It was confirmed by their chromatographic separation that PBDFs were not
merely mass fragments formed in the ion source corresponding to loss of Br2 from
PBDEs (Figure 4.5). Fragment mass series for PBDFs corresponding to loss of COBr
(C11H7-nBrn+!, n=1-4) and COBr3 (C11H5-nBrn
+!, n=1-2) were observed. In addition, series
of polyhalogenated dibenzofurans (PXDFs, X=Br,Cl) containing one chlorine (C12H7-
nOClBrn+!, n=1-4) and Br-anthracenes/phenanthrenes (C14H10-nBrn
+!, n=3-4) were
identified. An unknown series corresponding to the elemental composition, C14H13-
nO2Brn+! (n=2-4) may be mass fragments of a higher mass compound series or a product
of cluster ions formed in the ion source.
85
Figure 4.4. Mass defect plot based on H/Cl mass scale for household fire air sample extracts. Highlighted peaks correspond to Br-anthracenes/phenanthrenes (C14H10-nBrn
+!!), PBDFs (C12OH8-
nBrn+!!), PXDFs (C12OH7-nClBrn
+!!), and PBDEs (C12OH10-nBrn+!!), along with corresponding fragment
series (C11H7-nBrn+!! and C11H5-nBrn
+!!) and unknown series, C14O2H13-nBrn+!!.
86
Figure 4.5. Selected ion chromatogram (SIC) for household fire air sample extract for 323.878 ± 0.010 demonstrating chromatographic separation of dibromodibenzofuran (C12H6OBr2
+!!) (A) and dibromodiphenyl ether (C12H6OBr4
+!!) (B). Corresponding mass spectra show fragmentation for each compound.
4.4.1.3 Electronics Fire The extract of an air sample that was collected near the top of the burn cell for
approximately 10 minutes produced a simple mass defect plot for halogenated
compounds with only PBDEs (C12H10-nOBrn+!, n=4-5) and a mass fragment series
corresponding to loss of Br2 (C12H8-nOBrn+!, n=1-6) (Figure 4.6A).
In contrast, the extract of a foil sample that collected particulate on the burn cell
wall demonstrated a contaminant profile similar to that of the household fire air (Figure
87
4.6B). A similar number of PBDF congeners were observed (C12H8-nOBrn+!, n=1-5) and
corresponding mass fragment series corresponding to loss of Br2 (C12H6-nOBrn+!, n=2),
COBr (C11H7-nBrn+!, n=1-4) and COBr3 (C11H5-nBrn
+!, n=2), however only one PBDE
congener (C12H10-nOBrn+!, n=2) was identified. In addition to PXDFs that contained one
chlorine (C12H7-nOClBrn+!, n=1-3), a series of PXDFs with two chlorines (C12H6-
nOCl2Brn+!, n=1-2) was observed. An increased number of Br-anthracenes/phenanthrenes
(C14H10-nBrn+!, n=1-3) were identified relative to the household fire air. The unknown
series corresponding to C14H13-nO2Brn+! (n=1-2) was also observed.
Figure 4.6. Mass defect plots based on H/Cl mass scale for electronics fire air sample extract (A) and foil extract (B) Highlighted peaks correspond to Br-anthracenes/phenanthrenes (C14H10-nBrn
+!!), PBDFs (C12OH8-nBrn
+!!), PXDFs (C12OH7-nClBrn+!! and C12OH6-nCl2Brn
+!!), and PBDEs (C12OH10-
nBrn+!!), along with corresponding fragment series (C12OH6-nBrn
+!!, C11H7-nBrn+!! and C11H5-nBrn
+!!) and unknown series, C14O2H13-nBrn
+!!.
88
4.4.2 Sources and Fate of Thermal Decomposition Products
4.4.2.1 PBDEs All three samples examined contained PBDEs. It is likely PBDEs were present in
the contents of both the electronics and household fire simulations, as their widespread
applications include cables, furniture, coatings, and circuit boards(13). As additive flame
retardants, PBDEs are not chemically bound to the associated polymeric material and
may escape from the finished product(14). This is of environmental concern because
levels of PBDEs incorporated into polymeric materials have been reported to range from
5 to 30%(13), and upon exposure may induce toxic effects(15). Bromine is specifically
incorporated into flame retardant materials because it can capture vapour phase free
radicals produced in combustion, thereby reducing flame propagation(16, 17). While
effective in this regard, PBDEs are also released in thermal events containing brominated
flame retardants (BFRs) such as waste incineration(18, 19). Release of PBDEs to the
surrounding environment has led to their widespread detection in environmental samples
and several reviews addressing their environmental occurrence, persistence, long range
atmospheric transport, human exposure and bioaccumulation(15, 20-24).
The degree of bromination may determine their environmental fate, as PBDE
vapour pressures decrease with increasing bromine substituents(25). Indeed, an
atmospheric study in Kyoto, Japan, reported increased levels of higher brominated
PBDEs in the particulate phase relative to the corresponding lower brominated
species(26). Unfortunately in the present study we were unable to differentiate between
PBDEs in the gas and particulate phase, as the two extracts were combined. Several
studies have noted that the variation in PBDE concentrations and congener profiles
identified around the world may reflect relevant regulations and use(20, 22, 24, 27). The
three commercially important PBDE congener formulations are deca-, octa-, and penta-
BDEs. Due to environmental and health concerns, production and import of items
containing penta- and octa-BDEs have been restricted or banned(28, 29), with similar
regulations in development for deca-BDEs(29).
In the present study, the PBDEs identified had varying degrees of bromination.
The household fire air extract contained di- to hepta-BDEs. In the electronics fire, the air
extract contained tetra- and penta-BDEs, while only di-BDE was observed in the foil
89
extract. It is possible the variation in PBDE congeners observed is a result of differences
in sampling duration, media and location, making it difficult to pinpoint a particular
PBDE source. Penta-BDEs, which have been used in polyurethane foam(23), may have
been present in the mattress, sofa chair and pillows burned in the household fire
simulation, leading to identification of lower brominated PBDEs. Octa-BDEs, which
have been used in hard plastics for computer casings, and deca-BDEs, which have been
used in polystyrene plastics for furniture, television casings, and electronic
equipment(23), may have been present in the television and chairs burned in the
household fire simulation, leading to identification of higher brominated PBDEs.
The range of PBDE congeners that may be released in a fire event varies greatly.
A study examining PBDEs released at an informal electronic waste recycling site where
waste is burned in the open air, identified complex congener profiles ranging from mono-
to deca-BDEs in surrounding air, soil, and sediments(30). In contrast, a study of PBDEs
in flue gases from municipal waste incinerators and a coal-fired power plant, observed
octa-BDE as the predominant PBDE released(19). In the present study, a range of di- to
hepta-BDEs was observed; although it is possible the variation is a result of differences in
sampling duration, media and location. In future studies, knowledge of burn material
PBDE content, burn cell temperatures, and replicate and comparable air and particulate
samples could allow further interpretation of PBDE sources and congener profiles. These
findings do indicate the significant use of BFRs in electronics and household items, and
demonstrate the potential for PBDEs to be released in a fire event.
4.4.2.2 PBDFs and PXDFs The electronics fire foil extract and the household fire air extract contained
PBDFs and PXDFs. The identification of PBDFs and PXDFs is often associated with the
analogous polybrominated and polyhalogenated dibenzo-p-dioxins (PBDDs and PXDDs,
X = Br,Cl), however the detection limits of the FTICR-MS may not have been low
enough to capture these compounds. Unlike PBDEs, these compounds are not produced
commercially, but formed in the thermal decomposition of brominated materials. Initial
pyrolysis studies by Buser identified PBDD/PBDFs as thermal decomposition products of
PBDEs(4). Since then, several laboratory studies have identified PBDD/PBDFs as
thermal decomposition products of various BFRs(31). These compounds may also form
90
at lower temperatures through de novo synthesis(31-33). Both PBDD/PBDFs and
PXDD/PXDFs have been identified previously in waste incinerator fly ash(34, 35) and
flue gas(36, 37), as well as soot from an accidental fire(38). Environmental measurements
have indicated a correlation between PBDE and PBDD/PBDF levels(26, 39) which has
been attributed to both PBDE thermal decomposition and impurities present in PBDE
mixtures(40). With the identification of PBDEs in the present study, it is likely PBDFs
and PXDFs were identified as a result of BFR thermal decomposition. Like the analogous
PCDD/PCDFs, PBDD/PBDFs and PXDD/PXDFs are of environmental and toxicological
concern(41-43).
The mechanisms and conditions under which PBDD/PBDFs form have been
studied previously. Proposed mechanisms of PBDE thermal decomposition to form
PBDFs have included elimination of HBr or Br2(31), however a recent theoretical study
suggests loss of a Br or H atom followed by ring closure is the most likely reaction
pathway to PBDF formation(44). It has been observed that PBDD/PBDF yield decreases
with increasing bromination of PBDE precursors(31), and the presence of polymer
matrix, metal oxide, and water may increase production of PBDFs(31, 45-47).
The PXDFs identified were primarily monochloro-polybrominated dibenzofurans
(Br = 1-4), however in the electronics foil extract, dichloro-polybrominated
dibenzofurans (Br = 1,2) were also observed. Previous studies have identified
PXDD/PXDFs as thermal decomposition products of PBDEs in the presence of a chlorine
source(31, 48). In both samples, monoisotopic peak intensity was higher for PBDFs than
PXDFs, as shown in Figure 4.7. This is consistent with other studies that observed
PXDD/PXDF concentrations to range 1-20% of the corresponding PCDD/PCDF
concentrations(31, 35, 49, 50).
The congener profiles observed in the electronics fire foil extract and the
household fire air extract differed. The household fire air PBDF congener profile (Figure
4.7B) closely resembled that reported for television residue collected from an accidental
fire site in which di-BDF was the most concentrated, followed by tri-, tetra-, and penta-
BDF(51). It is possible the television in the household fire simulation contained similar
BFR materials as the television reported on in that study, however the electronics fire
simulation contained two televisions and shows a very different PBDF congener profile.
91
Without knowing the BFR content of the burn materials, it is difficult to interpret the
different congener profiles of the two fires. It is also important to consider here that
Figure 4.7 PBDF peak intensities were obtained from summed transient mass spectra that
do not distinguish between PBDE mass fragments and PBDFs. Chromatography is
required to distinguish these as shown in Figure 4.5. Figure 4.7. Relative abundance of PBDF and PXDF monoisotopic peaks with varying degrees of halogenation (X) for electronics fire foil extract (A) and household fire air extract (B).
4.4.2.3 Br-Anthracenes/Phenanthrenes The electronics fire foil extract and the household fire air extract contained Br-
anthracenes/phenanthrenes, ranging in degree of bromination from 1-3 and 3-4,
respectively. Monobromo-anthracene/phenanthrene has been identified previously in
municipal waste incinerator fly ash along with PXDD/PXDFs(52, 53). Another study
identified monobromo-anthracene/phenanthrene in fly ash and noted that total Cl-PAH
concentrations were 10 times that of total Br-PAH concentrations(54), however no Cl-
PAHs were identified in the present study. Thermal decomposition experiments with
BFR, decabromodibenzyl (DDB), produced Br-phenanthrenes (Br =1-4), whereas deca-
BDE did not(55). This may indicate that a BFR other than PBDE was present, leading to
Br-anthracene/phenanthrene formation.
92
To date, few studies have examined Br-PAHs in the environment, although they
have been identified in air, soil and sediments(56-59). They are susceptible to photolysis
to produce monobromo- and parent PAHs(60). Toxicity of Br-PAHs varies with halogen
substitution and position(61, 62), but relative to Cl-PAHs, Br-PAHs may be less toxic
due to the steric hindrance of the larger bromine atom(63). It has also been observed that
Br-anthracenes/phenanthrenes do not induce significant aryl hydrocarbon receptor-
mediated responses(64).
4.4.3 Potential for Rapid Broad Screening Approach
As demonstrated in the present study, high-resolution mass spectrometry and
mass defect filtering offer a rapid broad screening approach to assessing the
contamination present at a fire site. Realistically, air and particulate samples can be
collected, extracted and analyzed within 24 hours. Interpretation of mass defect plots is
currently the limiting factor. Creating a mass defect database of established halogenated
contaminants formed in fire (e.g. PCDD/PCDFs, PBDD/PBDFs) could quickly
distinguish known contaminants from unknowns and assist in safe management of a fire
site. This screening approach could provide valuable health and safety information to
firefighters and nearby residents, while directing clean-up measures and further
environmental site investigations.
4.5 Acknowledgements
We thank Lindsay Jobst for assisting with electronics donations and Brian Kellow
from Goodwill for donation of burn materials. We are grateful to Mike Hutchison, Phil
Bott, Sam Marshall, and Pike Krpan for sharing their firefighting expertise and their
assistance in experimental design and execution. We also thank Vince Taguchi from the
Ontario Ministry of the Environment for access to the FTICR-MS.
93
4.6 References
1. P. J. Landrigan et al., Health and environmental consequences of the World Trade
Center disaster. Environ. Health Persp. 112, 731–739 (2004).
2. P. J. Lioy et al., Characterization of the dust/smoke aerosol that settled east of the World Trade Center (WTC) in lower Manhattan after the collapse of the WTC 11 September 2001. Environ. Health Persp. 110, 703–714 (2002).
3. J. H. Offenberg et al., Persistent organic pollutants in the dusts that settled across Lower Manhattan after September 11, 2001. Environ. Sci. Technol. 37, 502–508 (2003).
5. S. D. Shaw et al., Persistent organic pollutants including polychlorinated and polybrominated dibenzo-p-dioxins and dibenzofurans in firefighters from Northern California. Chemosphere 91, 1386–1394 (2013).
6. R. Zeig-Owens et al., Early assessment of cancer outcomes in New York City firefighters after the 9/11 attacks: an observational cohort study. Lancet 378, 898–905 (2011).
7. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass Spectrom. 21, 1918–1921 (2010).
8. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
9. A. L. Myers, K. J. Jobst, S. A. Mabury, E. J. Reiner, Using mass defect plots as a discovery tool to identify novel fluoropolymer thermal decomposition products. J. Mass. Spectrom. 49, 291–296 (2014).
10. C. A. Hughey, C. L. Hendrickson, R. P. Rodgers, A. G. Marshall, K. Qian, Kendrick mass defect spectrum: a compact visual analysis for ultrahigh-resolution broadband mass spectra. Anal. Chem. 73, 4676–4681 (2001).
11. Z. Wu, R. P. Rodgers, A. G. Marshall, Two- and three-dimensional van Krevelen diagrams: a graphical analysis complementary to the Kendrick mass plot for sorting elemental compositions of complex organic mixtures based on ultrahigh-resolution broadband Fourier transform ion cyclotron resonance mass measurements. Anal. Chem. 76, 2511–2516 (2004).
94
12. S. Taniyasu et al., Analysis of fluorotelomer alcohols, fluorotelomer acids, and short- and long-chain perfluorinated acids in water and biota. J. Chromatogr. A 1093, 89–97 (2005).
13. World Health Organization (WHO), Brominated diphenyl ethers (International Programme on Chemical Safety, Environmental Health Criteria 162, 1994).
14. O. Hutzinger, G. Sundstrom, S. Safe, Environmental chemistry of flame retardants part I. introduction and principles. Chemosphere, 3–10 (1976).
15. C. A. de Wit, An overview of brominated flame retardants in the environment. Chemosphere 46, 583–624 (2002).
16. M. Alaee, P. Arias, A. Sjodin, A. Bergman, An overview of commercially used brominated flame retardants, their applications, their use patterns in different countries/regions and possible modes of release. Environ. Int. 29, 683–689 (2003).
17. A. B. Morgan, J. W. Gilman, An overview of flame retardancy of polymeric materials: application, technology, and future directions. Fire Mater. 37, 259–279 (2013).
18. C. Agrell et al., Polybrominated diphenyl ethers (PBDES) at a solid waste incineration plant I: atmospheric concentrations. Atmos. Environ. 38, 5139–5148 (2004).
19. L.-K. Tu, Distribution of polybrominated dibenzo-p-dioxins and dibenzofurans and polybrominated diphenyl ethers in a coal-fired power plant and two municipal solid waste incinerators. Aerosol Air Qual. Res. 11, 596-615 (2011).
20. A. Sjodin, D. G. Patterson Jr, A. Bergman, A review on human exposure to brominated flame retardants - particularly polybrominated diphenyl ethers. Environ. Int. 29, 829–839 (2003).
21. R. A. Hites, Polybrominated diphenyl ethers in the environment and in people: a meta-analysis of concentrations. Environ. Sci. Technol. 38, 945–956 (2004).
22. Y. Wang, G. Jiang, P. K. S. Lam, A. Li, Polybrominated diphenyl ether in the East Asian environment: A critical review. Environ. Int. 33, 963–973 (2007).
23. C. A. de Wit, D. Herzke, K. Vorkamp, Brominated flame retardants in the Arctic environment — trends and new candidates. Sci. Total Environ. 408, 2885–2918 (2010).
24. A. Besis, C. Samara, Polybrominated diphenyl ethers (PBDEs) in the indoor and outdoor environments – A review on occurrence and human exposure. Environ. Pollut. 169, 217–229 (2012).
25. S. A. Tittlemier, G. T. Tomy, Vapor pressures of six brominated diphenyl ether
26. K. Hayakawa, H. Takatsuki, I. Watanabe, S.-I. Sakai, Polybrominated diphenyl ethers (PBDEs), polybrominated dibenzo-p-dioxins/dibenzofurans (PBDD/Fs) and monobromo-polychlorinated dibenzo-p-dioxins/dibenzofurans (MoBPXDD/Fs) in the atmosphere and bulk deposition in Kyoto, Japan. Chemosphere 57, 343–356 (2004).
27. I. Watanabe, S.-I. Sakai, Environmental release and behavior of brominated flame retardants. Environ. Int. 29, 665–682 (2003).
28. United Nations, Adoption of Amendments to Annexes A, B, and C (Stockholm Convention on Persistent Organic Pollutants, Stockholm, Sweden, 2009; http://chm.pops.int/Portals/0/download.aspx?d=UNEP-POPS-TREATY-NOTIF-CN524-2009.En.pdf).
29. U.S. Environmental Protection Agency, Polybrominated Diphenyl Ethers (PBDEs) Action Plan (U.S. Environmental Protection Agency, 2009; http://www.epa.gov/oppt/existingchemicals/pubs/actionplans/pbdes_ap_2009_1230_final.pdf).
30. M. H. Wong et al., Export of toxic chemicals – a review of the case of uncontrolled electronic-waste recycling. Environ. Pollut. 149, 131–140 (2007).
31. R. Weber, B. Kuch, Relevance of BFRs and thermal conditions on the formation pathways of brominated and brominated-chlorinated dibenzodioxins and dibenzofurans. Environ. Int. 29, 699–710 (2003).
32. D. Heinbuch, L. Stieglitz, Formation of brominated compounds on fly ash. Chemosphere 27, 317–324 (1993).
33. R. Weber, B. Kuch, T. Ohno, T. Sakurai, De novo synthesis of mixed brominated-chlorinated PXDD/PXDF. Organohalogen Compd. 56, 181–184 (2002).
34. K. H. Schwind, J. Hosseinpour, H. Thoma, Brominated/chlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 17, 1875–1884 (1988).
35. W. Chatkittikunwong, C. S. Creaser, Bromo-, bromochloro- and chloro-dibenzo-p-dioxins and dibenzofurans in incinerator fly ash. Chemosphere 29, 559–566 (1994).
36. T. Oberg, K. Warman, J. Bergstrom, Brominated aromatics from combustion. Chemosphere 16, 2451–2465 (1987).
37. B. Du et al., Mixed polybrominated/chlorinated dibenzo-p-dioxins and dibenzofurans in stack gas emissions from industrial thermal processes. Environ. Sci. Technol. 44, 5818–5823 (2010).
96
38. H.-R. Buser, Selective detection of brominated aromatic compounds using gas chromatography/negative chemical ionization mass spectrometry. Anal. Chem. 58, 2913–2919 (1986).
39. H. Terauchi, S. Takahashi, P. K. S. Lam, B.-Y. Min, S. Tanabe, Polybrominated, polychlorinated and monobromo-polychlorinated dibenzo-p-dioxins/dibenzofurans and dioxin-like polychlorinated biphenyls in marine surface sediments from Hong Kong and Korea. Environ. Pollut. 157, 724–730 (2009).
40. N. Hanari et al., Occurrence of polybrominated biphenyls, polybrominated dibenzo-p-dioxins, and polybrominated dibenzofurans as impurities in commercial polybrominated diphenyl ether mixtures. Environ. Sci. Technol. 40, 4400–4405 (2006).
41. H. Olsman et al., Relative differences in aryl hydrocarbon receptor-mediated response for 18 polybrominated and mixed halogenated dibenzo-p-dioxins and -furans in cell lines from four different species. Environ. Toxicol. Chem. 26, 2448–2454 (2007).
42. G. Mason, T. Zacharewski, M. A. Denomme, L. Safe, S. Safe, Polybrominated dibenzo-p-dioxins and related compounds: quantitative in vivo and in vitro structure-activity relationships. Toxicology 44, 245–255 (1987).
43. L. S. Birnbaum, D. F. Staskal, J. J. Diliberto, Health effects of polybrominated dibenzo-p-dioxins (PBDDs) and dibenzofurans (PBDFs). Environ. Int. 29, 855–860 (2003).
44. M. Altarawneh, B. Z. Dlugogorski, A mechanistic and kinetic study on the formation of PBDD/Fs from PBDEs. Environ. Sci. Technol. 47, 5118–5127 (2013).
45. R. Dumler, D. Lenoir, H. Thoma, O. Hutzinger, Thermal formation of polybrominated dibenzofurans and dioxins from decabromo-diphenyl ether flame retardant: Influence of antimony(III) oxide and the polymer matrix. Chemosphere 20, 1867–1873 (1990).
46. R. Luijk, H. Wever, K. Olie, H. A. J. Govers, J. J. Boon, The influence of the polymer matrix on the formation of polybrominated dibenzo-p-dioxins (PBDDs) and polybrominated dibenzofurans (PBDFs). Chemosphere 23, 1173–1183 (1991).
47. D. Lenoir, B. Zier, D. Bieniek, A. Kettrup, The influence of water and metals on PBDD/F concentration in incineration of decabrombiphenyl ether in polymeric matrices. Chemosphere 28, 1921–1928 (1994).
48. S. Rupp, J. W. Metzger, Brominated–chlorinated diphenyl ethers formed by thermolysis of polybrominated diphenyl ethers at low temperatures. Chemosphere 60, 1644–1651 (2005).
97
49. R. L. Harless, R. G. Lewis, D. D. McDaniel, A. E. Dupuy Jr, Identification of bromo/chloro dibenzo-p-dioxins and dibenzofurans in ash samples. Chemosphere 18, 201–208 (1989).
50. H. Y. Tong, S. J. Monson, M. L. Gross, L. Q. Huang, Monobromopolychlorodibenzo-p-dioxins and dibenzofurans in municipal waste incinerator flyash. Anal. Chem. 63, 2697–2705 (1991).
51. V. Zelinski, W. Lorenz, M. Bahadir, Brominated flame retardants and resulting PBDD/F in accidental fire residues from private residences. Chemosphere 27, 1519–1528 (1993).
52. G. W. Sovocool et al., Bromo- and bromochloro-polynuclear aromatic hydrocarbons, dioxins and dibenzofurans in municipal incinerator fly ash. Biomed. Environ. Mass 15, 669–676 (1988).
53. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
54. Y. Horii, G. Ok, T. Ohura, K. Kannan, Occurrence and profiles of chlorinated and brominated polycyclic aromatic hydrocarbons in waste incinerators. Environ. Sci. Technol. 42, 1904–1909 (2008).
55. G. Grause et al., TG–MS investigation of brominated products from the degradation of brominated flame retardants in high-impact polystyrene. Chemosphere 85, 368–373 (2011).
56. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
57. H.-G. Ni, E. Y. Zeng, Environmental and human exposure to soil chlorinated and brominated polycyclic aromatic hydrocarbons in an urbanized region. Environ. Toxicol. Chem. 31, 1494–1500 (2012).
58. J.-L. Sun, H.-G. Ni, H. Zeng, Occurrence of chlorinated and brominated polycyclic aromatic hydrocarbons in surface sediments in Shenzhen, South China and its relationship to urbanization. J. Environ. Monit. 13, 2775 (2011).
59. J.-L. Sun, H. Zeng, H.-G. Ni, Halogenated polycyclic aromatic hydrocarbons in the environment. Chemosphere 90, 1751–1759 (2013).
60. T. Ohura, K.-I. Sawada, T. Amagai, M. Shinomiya, Discovery of novel halogenated polycyclic aromatic hydrocarbons in urban particulate matters: occurrence, photostability, and AhR activity. Environ. Sci. Technol. 43, 2269–2275 (2009).
98
61. P. P. Fu, L. S. Von Tungeln, L.-H. Chiu, Z. Y. Own, Halogenated-polycyclic aromatic hydrocarbons: a class of genotoxic environmental pollutants. J. Environ. Sci. Heal. C 17, 71–109 (1999).
62. P. P. Fu, Q. Xia, X. Sun, H. Yu, Phototoxicity and environmental transformation of polycyclic aromatic hydrocarbons (PAHs)—light-induced reactive oxygen species, lipid peroxidation, and DNA damage. J. Environ. Sci. Heal. C 30, 1–41 (2012).
63. E. J. LaVoie, L. Tulley-Freiler, V. Bedenko, D. Hoffmann, Mutagenicity of substituted phenanthrenes in Salmonella typhimurium. Mutat. Res. 116, 91–102 (1983).
64. Y. Horii et al., Relative potencies of individual chlorinated and brominated polycyclic aromatic hydrocarbons for induction of aryl hydrocarbon receptor-mediated responses. Environ. Sci. Technol. 43, 2159–2165 (2009).
99
5 CHAPTER FIVE
Analysis of Mixed Halogenated Dibenzo-p-dioxins and Dibenzofurans (PXDD/PXDFs) in Soil by Gas Chromatography Tandem Mass Spectrometry (GC-MS/MS)
Anne L. Myers, Scott A. Mabury, and Eric J. Reiner Published as: Chemosphere 2012, 87, 1063-1069. Contributions: Anne L. Myers was responsible for sample extraction, GC-MS/MS method development and analysis, GC-HRToF analysis, and data interpretation. Anne L. Myers prepared this manuscript with editorial comments provided by Scott A. Mabury and Eric J. Reiner.
dioxin (1-Br-2,3,6,7,8,9-Cl6DD), and 1-bromo-2,3,4,6,7,8,9-heptachlorodibenzo-p-dioxin (1-Br-
2,3,4,6,7,8,9-Cl7DD) were obtained from Cambridge Isotope Laboratories (Andover, MA, USA).
All other materials were obtained from Caledon Laboratory Chemicals (Georgetown, ON, CAN)
unless otherwise indicated.
5.3.2 Soil Extraction
In July of 1997, a fire at the Plastimet Inc. plastics recycling facility in Hamilton,
Ontario, Canada consumed 400 tonnes of polyvinyl chloride plastic and polyurethane foam,
resulting in the emission of numerous hazardous substances to the surrounding area(27). Nearby
soil samples of high ash content were collected immediately and extracted for dioxin analysis
103
following the Ontario Ministry of the Environment (MOE) method DFPCB-E3418(28). For the
present study, four archived soil samples were selected for PXDD/PXDF analysis based on
elevated levels of PCDD/PCDFs observed in the original 1997 analysis.
Soil extraction procedures are described in greater detail in MOE method DFPCB-
E3418(28). Briefly, 5.0 g of homogenized dry soil was weighed into a glass extraction thimble
containing activated silica (Rose Scientific Ltd., Edmonton, AB, CAN). To each sample, 5.0 "L
of a 100 pg/"L 13C12-2,3-Br2-7,8-Cl2DD solution in nonane was added using a 5 "L capillary
syringe. A glass wool plug was placed in each thimble prior to positioning them in a Soxhlet
extractor. A Soxhlet extraction was performed using 200 mL toluene for approximately 12 hours.
Following extraction, toluene extracts were brought to dryness by rotary evaporation and
reconstituted in 5.0 mL hexane.
Sample clean-up consisted of an 11 mm diameter glass column containing 1.5 g 10%
silver nitrate silica packing, 1.0 g silica, 2.0 g 33% sodium hydroxide silica packing, 1.0 g silica,
4.0 g 44% sulphuric acid silica packing, 2.0 g silica, and 2.0 g anhydrous sodium sulphate. The
column was conditioned with 50 mL hexane prior to transferring the sample to the column, and
eluting the extract with 100 mL hexane. The extract was reduced to 1.0 mL by rotary
evaporation. A 6 mm diameter glass column containing 5.0 g activated basic alumina and 2 g
anhydrous sodium sulphate was conditioned with 40 mL hexane. The sample was transferred to
the column and the first fraction was eluted with 100 mL hexane followed by 20 mL of hexane
solution containing 10% carbon tetrachloride (Sigma Aldrich, Oakville, ON, CAN). The second
fraction, expected to contain PXDD/PXDFs, was eluted with 50 mL dichloromethane.
Sample extracts were reduced to dryness by rotary evaporation and reconstituted in 1.0
mL hexane. Extracts were transferred to amber glass gas chromatography vials (Canada Life
Science, Peterborough, ON, CAN) and brought to dryness via nitrogen evaporation. Sample vials
were stored at 4.0 oC until analysis. Prior to analysis, samples were reconstituted in 25 "L
nonane.
5.3.3 GC-MS/MS Analysis
Analyses were performed on a gas chromatograph (GC) (Agilent HP 6890 GC, Agilent
Technologies, Mississauga, ON, CAN) coupled to a tandem quadrupole mass spectrometer
(Waters Quattro Micro tandem quadrupole mass spectrometer, Waters, Milford, MA, USA)
104
using a non-polar stationary phase GC column (Agilent J&W DB-5 GC column, 40 m x 0.18
mm, 0.18 "m, Chromatographic Specialties, Brockville, ON, CAN). A 1.0 "L injection volume
was used with an injector temperature of 270oC. The temperature program began at 140oC and
increased to 200oC at a rate of 55oC/min, after which the temperature increased at a rate of
2.5oC/min to 300oC, where it was held for two minutes. The GC-MS/MS system was operated in
multiple reaction monitoring mode (MRM) with electron impact ionization. The mass
spectrometer source temperature was 250oC and the electron energy was 70 eV.
Parameter optimization was performed via the solids probe. A small amount of each
PXDD/PXDF standard solution was transferred to a separate glass capillary and the solvent was
allowed to evaporate so the analyte coated the interior of the capillary tube. The capillary was
inserted into the tip of the solids probe, secured with a tungsten strip, and inserted into the inner
source of the instrument. The temperature of the solids probe was raised gradually until a signal
was observed on which optimization could be performed. The temperature at which parameters
were optimized varied from 30oC to 90oC. The unique parent-product transitions and the
corresponding parameters are presented in Table 5.1. For each transition, the dwell time was 0.1
seconds and the delay time was 0.05 seconds.
Quantification was performed by internal calibration for all sample analyses. The linear
calibration range was 2.4 pg/"L to 1920 pg/"L and 13C12-2,3-Br2-7,8-Cl2DD was used as an
internal standard for quantification of all analytes. All PXDD/PXDF standard analytes were
separated chromatographically with the exception of co-eluting 8-Br-2,3,4-Cl3DF and 3-Br-
2,7,8-Cl3DF. Quantification of PXDD/PXDF standard analytes or corresponding congener
groups was performed using the sum of peak areas of corresponding transitions. Isomer
groupings of PXDD/PXDF congener groups were identified by peak patterns and fragment peak
ratios were monitored to ensure sample response was within 40% of standard peak ratios,
indicating few or no interferences.
105
Table 5.1. PXDD/PXDF parent-product transitions, corresponding collision energies and peak area ratios used in GC-MS/MS analysis.
Compound Mass to Charge Ratio (m/z) Reaction Collision Energy (eV)
Standard Peak Area Ratio
Transitions Used in Quantification Parent Ion Product Ion
8-Br-2,3-Cl2DF 315.7 209.0 M - COBr 18 0.82 ü 172.0 M - COBrCl 33 1 ü
7-Br-2,3-Cl2DD 331.9 269.1 M - COCl 18 0.71 225.1 M - COBr 19 0.60 ü 162.0 M - (CO)2BrCl 36 1 ü
8-Br-2,3,4-Cl3DF 349.9 243.0 M - COBr 27 0.42 206.1 M - COBrCl 36 1 ü
3-Br-2,7,8-Cl3DF 349.9 243.0 M - COBr 27 0.42 206.1 M - COBrCl 36 1 ü
2,3-Br2-7,8-Cl2DD 409.8 347.0 M - COCl 23 0.48 303.0 M - COBr 23 1.77 ü 240.1 M - (CO)2BrCl 36 1 ü
1-Br-2,3,7,8-Cl4DD 399.8 337.0 M - COCl 18 0.96 293.1 M - COBr 17 0.65 ü 230.0 M - (CO)2BrCl 36 1 ü
2-Br-1,3,7,8-Cl4DD 399.8 337.0 M - COCl 18 0.96 293.1 M - COBr 17 0.65 ü 230.0 M - (CO)2BrCl 36 1 ü
2-Br-3,6,7,8,9-Cl5DD 435.8 373.0 M - COCl 19 0.44 329.1 M - COBr 19 0.74 ü 266.0 M - (CO)2BrCl 38 1 ü
1-Br-2,3,6,7,8,9-Cl6DD 469.8 406.9 M - COCl 22 0.50 362.9 M - COBr 22 0.50 300.0 M - (CO)2BrCl 36 1 ü
1-Br-2,3,4,6,7,8,9-Cl7DD 503.8 440.8 M - COCl 18 0.58 396.9 M - COBr 17 0.52 ü 334.0 M - (CO)2BrCl 37 1 ü
13C12-2,3-Br2-7,8-Cl2DD 421.7 357.9 M - COCl 20 2.07 ü 314.0 M - COBr 20 1.79 ü 250.0 M - (CO)2BrCl 36 1 ü
106
5.3.4 GC-HRToF
Confirmation of PXDD/PXDFs in soil was performed on a gas chromatograph (Agilent
HP 6890 GC, Agilent Technologies, Mississauga, ON, CAN) coupled to a time of flight mass
spectrometer (Waters GCT Premier time of flight mass spectrometer, Waters, Milford, MA,
USA) (GC-HRToF) using a non-polar stationary phase GC column (Agilent DB-5ms GC
column, 60m x 0.25mm, 0.25µm, Chromatographic Specialties, Brockville, ON, CAN). A 1.0
µL injection volume was used with an injector temperature of 250oC. The mass spectrometer was
operated at a resolution of 7000 (half height full width). The temperature program began at
140oC and increased to 200oC at a rate of 55oC/min, after which the temperature increased at a
rate of 2.5oC/min to 300oC, where it was held for three minutes. The GC-HRToF system scanned
for masses between m/z 50 and 550 at a scan time of 0.3 seconds with electron impact ionization.
The mass spectrometer source temperature was 250oC and the electron energy was 70 eV.
5.3.5 Quality Assurance/Quality Control
Spike and recovery studies were performed to assess the accuracy and precision of the
soil extraction procedure. An extraction thimble containing 5.0 g of Ottawa Sand was spiked
with 100 µL of native PXDD/PXDF hexane solution (1.20-12.5 pg/µL) and 5.0 µL of 100 pg/µL 13C12-2,3-Br2-7,8-Cl2DD nonane solution (Table 5.2). Two additional spike and recovery studies
were performed in which 20 µL of 250 pg/µL native PXDD/PXDF nonane solution and 500 µL
of native PXDD/PXDF hexane solution (1.20-12.5 pg/µL) were spiked (Table S1). The
extraction procedure was the same as the standard MOE Soxhlet extraction method for
PCDD/PCDFs in soils(28). Precision of the extraction procedure was also assessed for the soil
matrix through triplicate extraction of soil D. Instrumental precision was assessed through
duplicate analysis of each sample. All data is presented as the mean of replicate samples with the
The instrumental detection limit (IDL) was defined by a peak greater than or
equal to a signal to noise (S/N) of 2.5 and the instrumental quantification limit (IQL) was
defined by a peak greater than or equal to S/N of 10. Analyte peak responses were
integrated using Savitzky-Golay smoothing and the S/N was determined by peak-to-peak
analysis. The IDL and IQL values are presented in Table 5.2 for each standard analyte.
The method detection limit (MDL, pg/g) (Table 5.2) was determined by multiplying the
standard deviation (pg) of spike and recovery samples (n=7) by the relevant student’s t
value for a 99% confidence interval, and dividing by 5 g.
Good instrument reproducibility was shown among standards run before, during,
and after GC-MS/MS analysis of soil extracts. Relative standard deviations between
standards run in triplicate ranged from 0.55% to 10%. Triplicate extraction of soil D
demonstrated good precision for the soil extraction method with relative standard
deviations between extractions ranging from 0.22% for monobromo-
trichlorodibenzofurans and 9.1% for monobromo-hexachlorodibenzo-p-dioxins. Spike
and recovery studies showed good accuracy with PXDD/PXDF recoveries ranging from
94% to 124% (Table 5.2).
5.4.2 PXDD/PXDF Isomer Peak Patterns and Ratios
The number of PXDD/PXDF isomers varies with each congener group and will
determine the complexity of each peak pattern. The number of possible isomers for each
PXDD/PXDF congener group is reported by Buser(21), and those relevant to this study
are presented in Table 5.2. The simplest example is monobromo-heptachlorodibenzo-p-
dioxins, for which the two possible isomers were chromatographically separated and
observed in all corresponding transitions for all soil samples. The most complex example
is monobromo-trichlorodibenzofurans, for which there are 140 possible isomers. For this
congener group, the corresponding large asymmetrical peak patterns observed in both
transitions suggests some isomers are more prevalent than others and/or instrumental
109
parameters are not optimized for all isomers present (Figure 5.1). Due to the large
number of possible isomers, significant co-elution of isomers was expected.
Although PXDD/PXDF concentrations varied between soil samples, peak patterns
were generally the same for each congener group. This was expected since soil samples
were all collected from the Plastimet Inc. fire site, however PXDD/PXDF peak patterns
may vary between sites based on materials consumed in the fire and other factors
determining PXDD/PXDF formation. The general peak patterns observed for each
PXDD/PXDF congener group in the soil extracts are presented in Figure 5.2. Tong et al.
have shown similar isomer peak patterns for monobromo-pentachlorodibenzo-p-dioxins,
monobromo-hexachlorodibenzo-p-dioxins, and monobromo-heptachlorodibenzo-p-
dioxins in municipal incinerator fly ash using GC-HRMS(4). The similarity between
isomer peak patterns may be a result of common PXDD/PXDF precursors present during
incineration or similar formation mechanisms that favour particular isomers.
In order to properly identify and quantify PXDD/PXDFs in soil extracts, the peak
area ratios of corresponding transitions were compared with those of PXDD/PXDF
standards (Figure 5.1). In comparison, observed peak area ratios in soil extracts varied
greatly from standard peak ratios, with deviations ranging from 1% to 660%. For certain
transitions, large deviations from the standard peak ratio were observed consistently as
interferences and therefore these transitions were excluded from quantification
procedures, but included in visual PXDD/PXDF confirmation. Only transitions with peak
areas within 40% of standard peak ratios were used for quantification and are reported in
Table 5.1.
110
Figure 5.1. Corresponding peak patterns, transitions, and peak ratios for monobromo-dichlorofurans in a 38.4pg/µL Standard and Plastimet Soil A extract.
mass/7000. These values ranged from ±0.045 amu for monobromo-
dichlorodibenzofurans to ±0.072 amu for monobromo-heptachlorodibenzo-p-dioxins. The
GC-HRToF PXDD/PXDF mass values for [M]+, [M+2]+, and [M+4]+ isotopic peaks
were all within ± 0.043 amu of the corresponding theoretical values. In each case,
isotopic peak ratios were also generally within 15% of theoretical values.
5.4.4 PXDD/PXDFs in Soil from the Plastimet Inc. Fire
Analysis of soil extracts, A, B, C, and D, showed varying concentrations of
PXDD/PXDFs within the ng/g range (Figure 5.3). Monobromo-dichlorodibenzofurans
and monobromo-trichlorodibenzofurans were observed at higher levels than PXDDs,
reaching concentrations of 180 ng/g and 77 ng/g in soil D, respectively. Of the PXDDs
examined, monobromo-hexachlorodibenzo-p-dioxin and monobromo-
pentachlorodibenzo-p-dioxin were the most prevalent, reaching concentrations of 47 ng/g
and 41 ng/g in soil D, respectively. Concentrations of dibromo-dichlorodibenzo-p-dioxin
were also highest in soil D at 10 ng/g.
Figure 5.3. Concentrations of PXDD/PXDFs in soil samples with associated standard deviation (n=2). *For sample D, concentrations and standard deviations correspond to triplicate extractions and duplicate injections (n=6).
0
25
50
75
100
125
150
175
A B C D*
Con
cent
ratio
n (n
g/g
dwt)
monobromo-dichlorodibenzofuran
monobromo-dichlorodibenzo-p-dioxin
monobromo-trichlorodibenzofuran
dibromo-dichlorodibenzo-p-dioxin
monobromo-tetrachlorodibenzo-p-dioxin
monobromo-pentachlorodibenzo-p-dioxin
monobromo-hexachlorodibenzo-p-dioxin
monobromo-heptachlorodibenzo-p-dioxin
113
Archived PCDD/PCDF data obtained from the same soil samples was compared
with PXDD/PXDF concentrations from the present study. Concentrations of monobromo-
trichlorodibenzofuran were found to be 10 ± 2.6% (n=4) of tetrachlorodibenzofuran
concentrations. For the dioxins, dibromo-dichlorodibenzo-p-dioxins were 36 ± 5.0%
(n=3) of tetrachlorodibenzo-p-dioxin concentrations, monobromo-tetrachlorodibenzo-p-
dioxins were 8.0 ± 1.5% (n=4) of pentachlorodibenzo-p-dioxin concentrations,
monobromo-pentachlorodibenzo-p-dioxins were 9.5 ± 1.6% (n=4) of hexachlorodibenzo-
p-dioxin concentrations, monobromo-hexachlorodibenzo-p-dioxins were 9.0 ± 2.6%
(n=4) of heptachlorodibenzo-p-dioxin concentrations, and monobromo-
heptachlorodibenzo-p-dioxins were 6.2 ± 1.6% (n=4) of octachlorodibenzo-p-dioxin
concentrations. The relatively consistent fractions of PXDD/PXDFs to PCDD/PCDFs
between soil samples are consistent with those observed in samples obtained from waste
incinerators(2, 13). Congener distribution among PXDD/PXDFs and PCDD/PCDFs
showed similar patterns, suggesting similar formation mechanisms for both, as proposed
by Harless and Lewis(2). The one exception was dibromo-dichlorodibenzo-p-dioxins,
which had a relative percentage approximately four times that of the other PXDD/PXDFs
examined. This may reflect the fact that as the Br/Cl substitution ratio reaches one, the
number of potential isomers increases. In addition, this may indicate large quantities of
brominated products were consumed in the fire, thereby enhancing production of
polybrominated PXDD/PXDFs. Tetrahalogenated dibenzo-p-dioxins are of particular
concern for their high toxicity relative to other halogenated dibenzo-p-dioxins(11, 12),
and with 114 possible isomers, it is difficult to chromatographically separate the three
relatively toxic 2,3,7,8-substituted dibromo-dichlorodibenzo-p-dioxin isomers from other
less harmful tetrahalogenated isomers. As these results demonstrate, polybrominated
tetrahalogenated PXDDs may be formed in higher concentrations than expected and may
contribute significantly to total toxic dioxins formed in fires containing brominated and
chlorinated materials. More polybrominated 2,3,7,8-substituted PXDD standards are
required to properly identify the contribution of these toxic compounds of varying Br/Cl
ratios in dioxin formation.
114
5.5 Conclusions
The commercial application of BFRs and other halogenated materials is
widespread, and through waste incineration, recycling processes, and accidental fires, it is
expected that large amounts of PXDD/PXDFs were formed. Due to limited understanding
of sources and formation mechanisms of PXDD/PXDFs, as well as limited availability of
PXDD/PXDF standards, it is likely that overall dioxin concentrations are significantly
underestimated in industrial practice. With an overwhelming number of congeners,
PXDD/PXDFs may contribute significantly to toxicological issues occurring from fires.
Relative to GC-HRMS techniques, the GC-MS/MS method presented here offers a cost-
effective, simplified analysis for assessing PXDD/PXDF concentrations in environmental
samples.
5.6 Acknowledgements
We thank Wellington Laboratories for PXDD/PXDF standard solutions, Gareth
Cleland and Charles Valentine for GC-MS/MS technical assistance, and Kim Hong for
assistance in soil extraction.
5.7 References
1. W. Schafer, K. Ballschmiter, Monobromo-polychloro-derivatives of benzene,
biphenyl, dibenzofurane and dibenzodioxine formed in chemical-waste burning. Chemosphere 15, 755–763 (1986).
2. R. L. Harless, R. G. Lewis, D. D. McDaniel, A. E. Dupuy Jr, Identification of bromo/chloro dibenzo-p-dioxins and dibenzofurans in ash samples. Chemosphere 18, 201–208 (1989).
3. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
4. H. Y. Tong, S. J. Monson, M. L. Gross, L. Q. Huang, Monobromopolychlorodibenzo-p-dioxins and dibenzofurans in municipal waste incinerator flyash. Anal. Chem. 63, 2697–2705 (1991).
5. W. Chatkittikunwong, C. S. Creaser, Bromo-, bromochloro- and chloro-dibenzo-p-dioxins and dibenzofurans in incinerator fly ash. Chemosphere 29, 559–566
115
(1994).
6. H. Preud'homme, M. Potin-Gautier, Optimization of accelerated solvent extraction for polyhalogenated dibenzo-p-dioxins and benzo-p-furans in mineral and environmental matrixes using experimental designs. Anal. Chem. 75, 6109–6118 (2003).
7. B. Du et al., Mixed polybrominated/chlorinated dibenzo- p-dioxins and dibenzofurans in stack gas emissions from industrial thermal processes. Environ. Sci. Technol. 44, 5818–5823 (2010).
8. K. Hayakawa, H. Takatsuki, I. Watanabe, S.-I. Sakai, Polybrominated diphenyl ethers (PBDEs), polybrominated dibenzo-p-dioxins/dibenzofurans (PBDD/Fs) and monobromo-polychlorinated dibenzo-p-dioxins/dibenzofurans (MoBPXDD/Fs) in the atmosphere and bulk deposition in Kyoto, Japan. Chemosphere 57, 343–356 (2004).
9. K. Kannan, I. Watanabe, J. P. Giesy, Congener profile of polychlorinated/brominated dibenzo-p-dioxins and dibenzofurans in soil and sediments collected at a former chlor-alkali plant. Toxicol. Environ. Chem. 67, 135–146 (1998).
10. M. Unger et al., Polybrominated and mixed brominated/chlorinated dibenzo-p-dioxins in sponge (Ephydatia fluviatilis) from the Baltic Sea. Environ. Sci. Technol. 43, 8245–8250 (2009).
11. G. Mason, T. Zacharewski, M. A. Denomme, L. Safe, S. Safe, Polybrominated dibenzo-p-dioxins and related compounds: quantitative in vivo and in vitro structure-activity relationships. Toxicology 44, 245–255 (1987).
12. H. Olsman et al., Relative differences in aryl hydrocarbon receptor-mediated response for 18 polybrominated and mixed halogenated dibenzo-p-dioxins and -furans in cell lines from four different species. Environ. Toxicol. Chem. 26, 2448–2454 (2007).
13. R. Weber, B. Kuch, Relevance of BFRs and thermal conditions on the formation pathways of brominated and brominated-chlorinated dibenzodioxins and dibenzofurans. Environ. Int. 29, 699–710 (2003).
14. G. Söderström, S. Marklund, PBCDD and PBCDF from incineration of waste-containing brominated flame retardants. Environ. Sci. Technol. 36, 1959–1964 (2002).
15. S. Rupp, J. W. Metzger, Brominated–chlorinated diphenyl ethers formed by thermolysis of polybrominated diphenyl ethers at low temperatures. Chemosphere 60, 1644–1651 (2005).
16. L. Q. Huang, H. Tong, J. R. Donnelly, Characterization of
116
dibromopolychlorodibenzo-p-dioxins and dibromopolychlorodibenzofurans in municipal waste incinerator fly ash using gas chromatography/mass spectrometry. Anal. Chem. 64, 1034–1040 (1992).
17. C. S. Evans, B. Dellinger, Formation of bromochlorodibenzo-p-dioxins and furans from the high-temperature pyrolysis of a 2-Chlorophenol/2-Bromophenol mixture. Environ. Sci. Technol. 39, 7940–7948 (2005).
18. C. S. Evans, B. Dellinger, Formation of bromochlorodibenzo-p-dioxins and dibenzofurans from the high-temperature oxidation of a mixture of 2-Chlorophenol and 2-Bromophenol. Environ. Sci. Technol. 40, 3036–3042 (2006).
19. P. O. Steen, M. Grandbois, K. McNeill, W. A. Arnold, Photochemical formation of halogenated dioxins from hydroxylated polybrominated diphenyl ethers (OH-PBDEs) and chlorinated derivatives (OH-PBCDEs). Environ. Sci. Technol. 43, 4405–4411 (2009).
20. H. Thoma, G. Hauschulz, O. Hutzinger, Pyrolysis of dibenzodioxin, dibenzofuran and 1,2,3,4-tetrabromodibenzodioxin with different chlorine donors and catalysts. Chemosphere 18, 1213–1217 (1989).
21. H.-R. Buser, Brominated and brominated/chlorinated dibenzodioxins and dibenzofurans: potential environmental contaminants. Chemosphere 16, 713–732 (1987).
22. K. Jay, L. Stieglitz, Interferences in the analysis of mixed halogenated dibenzofurans with diphenyl ethers. Chemosphere 35, 1227–1231 (1997).
23. H.-R. Buser, Rapid photolytic decomposition of brominated and brominated/chlorinated dibenzodioxins and dibenzofurans. Chemosphere 17, 889–903 (1988).
24. A. J. Murk et al., Chemical-activated luciferase gene expression (CALUX): a novel in vitro bioassay for Ah receptor active compounds in sediments and pore water. Fund. Appl. Toxicol. 33, 149–160 (1996).
25. E. J. Reiner et al., Application of tandem quadrupole mass spectrometry for the ultra-trace determination of polychlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 20, 1385–1392 (1990).
26. E. J. Reiner, D. H. Schellenberg, V. Y. Taguchi, Environmental applications for the analysis of chlorinated dibenzo-p-dioxins and dibenzofurans using mass spectrometry/mass spectrometry. Environ. Sci. Technol. 25, 110–117 (1991).
27. Ontario Ministry of the Environment, Plastimet Inc. fire, Hamilton, Ontario, July 9-12, 1997. https://ia700309.us.archive.org/24/items/plastimetincfire00sochuoft/plastimetincfire00sochuoft.pdf (1997).
117
28. Ontario Ministry of the Environment, The Determination of Polychlorinated Dibenzo-p-dioxins, Polychlorinated Dibenzofurans and Dioxin-like Polychlorinated Biphenyls (DLPCBs) in Environmental Matrices by Gas Chromatography-High Resolution Mass Spectrometry (GC-HRMS) (Laboratory Services Branch Method DFPCB-E3418, 2010).
118
6 CHAPTER SIX
Complementary Non-targeted and Targeted Mass Spectrometry Techniques to Determine Bioaccumulation of Halogenated Contaminants in Freshwater Species
Anne L. Myers, Trudy Watson-Leung, Karl J. Jobst, Li Shen, Sladjana Besevic, Kari Organtini, Frank L. Dorman, Scott A. Mabury, and Eric J. Reiner
Submitted: Environ. Sci. Technol. (Manuscript ID: es-2014-03090s) Contributions: Experimental design was initially conceived by Eric J. Reiner. Trudy Watson-Leung was responsible for exposure studies and Li Shen and Sladjana Besevic performed sample extractions. Karl J. Jobst performed sample analysis by FTICR-MS. Anne L. Myers was responsible for sample analysis by GCxGC-HRToF and APGC-MS/MS, as well as data interpretation. Anne L. Myers prepared this manuscript with editorial comments provided by Trudy Watson-Leung, Karl J. Jobst, Scott A. Mabury, and Eric J. Reiner.
119
6.1 Abstract
Assessing the toxicological significance of complex environmental mixtures is
challenging due to the large number of unidentified contaminants. Non-targeted
analytical techniques may serve to identify bioaccumulative contaminants within
complex contaminant mixtures without the use of analytical standards. This study
exposed three freshwater organisms (Lumbriculus variegatus, Hexagenia spp., and
Pimephales promelas) to a highly contaminated soil collected from a recycling plant fire
site. Biota extracts were analyzed by Fourier Transform ion cyclotron resonance mass
spectrometry (FTICR-MS) and mass defect filtering to identify bioaccumulative
halogenated contaminants. Specific bioaccumulative isomers were identified by
comprehensive two-dimensional gas chromatography time of flight high-resolution mass
spectrometry (GCxGC-HRToF). Targeted analysis of mixed brominated/chlorinated
dibenzo-p-dioxins and dibenzofurans (PXDD/PXDFs, X = Br and Cl) was performed by
atmospheric pressure gas chromatography tandem mass spectrometry (APGC-MS/MS).
Relative sediment and biota instrument responses were used to estimate biota-sediment
accumulation factors (BSAFs). Bioaccumulating contaminants varied between species
and included polychlorinated naphthalenes (PCNs), polychlorinated dibenzofurans
(PCDFs), chlorinated and mixed brominated/chlorinated anthracenes/phenanthrenes, and
pyrenes/fluoranthenes (Cl-PAHs and X-PAHs, X = Br and Cl), as well as PXDD/PXDFs.
Bioaccumulation potential between isomers also varied. This study demonstrates how
complementary high-resolution mass spectrometry techniques identify persistent and
bioaccumulative contaminants (and specific isomers) of environmental concern.
6.2 Introduction
Non-targeted analytical techniques are unconstrained by preconceived notions of
contaminants of concern and are becoming increasingly important in the analysis of
complex environmental mixtures. Recently, non-targeted techniques, such as
comprehensive two-dimensional gas chromatography time of flight mass spectrometry
(GCxGC-ToF), have been used to identify unknown halogenated contaminants in
environmental samples(1-4). The use of high-resolution mass spectrometry further
facilitates unknown identification by determining elemental compositions of accurate
120
mass measurements. Fourier transform ion cyclotron resonance mass spectrometry
(FTICR-MS), an ultra-high resolution technique, and mass defect filtering have been used
to identify chlorinated polycyclic aromatic hydrocarbons (Cl-PAHs) in vegetation
exposed to an industrial fire(5) and Lake Ontario Lake Trout(6). Comprehensive two-
dimensional gas chromatography high-resolution time of flight mass spectrometry
(GCxGC-HRToF) has been used to identify Cl-PAHs and mixed brominated/chlorinated
polycyclic aromatic hydrocarbons consisting of three or more aromatic rings (X-PAHs, X
= Br and Cl) in several environmental matrices(7-9). These approaches may serve as
complementary techniques in identifying previously unknown halogenated contaminants.
First, the mass defect filtering of high-resolution mass spectral data reveals compound
classes within complex data sets and provides initial screening information to direct
further analysis. Secondly, the high-resolution chromatographic separation and mass
spectral data achieved via GCxGC-HRToF allows examination of specific isomers within
congener groups identified in initial screening. Finally, analytes of interest discovered
through these non-targeted approaches may direct development of targeted analytical
methods, such as gas chromatography tandem mass spectrometry (GC-MS/MS) analysis
of mixed brominated/chlorinated dibenzo-p-dioxins and dibenzofurans (PXDD/PXDFs,
X = Br and Cl)(10).
This complementary analytical approach provides the high-resolution
chromatographic separation and accurate mass information necessary to characterize
complex mixtures of halogenated aromatic contaminants. Recently, a similar approach
was used to characterize soil collected from the 1997 Hamilton, Ontario Plastimet Inc.
recycling plant fire site, in which a variety of halogenated contaminants were
identified(11). Combustion-derived Cl-PAHs, X-PAHs, and PXDD/PXDFs have received
little attention relative to other combustion products, such as polychlorinated
naphthalenes (PCNs) and polychlorinated dibenzo-p-dioxins and dibenzofurans
(PCDD/PCDFs), despite toxicological concerns(12-16). While there has been widespread
environmental detection of Cl-PAHs(17, 18), there are fewer reports of X-PAHs(7, 19-
21) and PXDD/PXDFs(10, 20-34). Although Cl-PAHs have been measured in air and
airborne particulates(15, 19, 35-40), physicochemical properties suggest that, once
deposited to water, they will partition into sediments and may be more bioaccumulative
121
than corresponding parent PAHs(18). Similar environmental fates are expected for X-
PAHs and PXDD/PXDFs. As a result, exposure of freshwater organisms to sediments
contaminated with Cl-PAHs, X-PAHs and PXDD/PXDFs is of concern.
Biota-sediment accumulation factors (BSAFs) estimate the bioaccumulation of
contaminants that cannot be effectively metabolized and eliminated. Bioaccumulation
describes absorption of contaminants by an organism through all routes of exposure in a
natural environment, including dietary uptake(41). Since Cl-PAHs, X-PAHs, and
PXDD/PXDFs are comprised of thousands of congeners of which toxicological
significance is specific to halogen substitution and position(13, 42), BSAFs for specific
isomers may be more informative than the corresponding congener group BSAF. For
example, 2,3,7,8-tetrachlorodibenzo-p-dioxin exhibits increased toxicological effects
over other tetrachlorodibenzo-p-dioxin isomers because it can bind with the aryl
hydrocarbon receptor as a result of its particular stereochemistry(43, 44). In addition,
identifying bioaccumulative isomers can direct future studies to compounds that pose a
greater toxicological risk. However, assessing BSAFs of particular isomers without
corresponding standards presents a complex analytical challenge. To date, the study of
Cl-PAHs, X-PAHs, and PXDD/PXDFs has relied on limited commercially available or
monoisotopic accurate masses to be within 20ppm of theoretical masses, isotopic peak
ratios to be within 20% of theoretical ratios, the presence of corresponding mass spectral
fragments, and the peak to be present in every replicate biota extract.
127
6.4 Results
6.4.1 FTICR-MS Analysis
Initial screening of Hexagenia spp. sample extracts via FTICR-MS analysis and
mass defect filtering identified bioaccumulative analytes of interest. A mass defect plot
identifying PCNs, Cl-PAHs, X-PAHs, and PCDFs in Hexagenia spp. exposed to the
Plastimet sediment mix is presented in Figure 6.1. Congener classes were identified by
their common mass defect and characteristic compound spacing of 33.9610 Da,
representing the exchange of hydrogen for chlorine. For example, PCNs (C10H8-nCln+, n =
4-6) are represented by three clusters of grey squares with an average mass defect value
of 0.2074. Moving from left to right across the plot, PCN chlorine substitution increases
from 4 to 6. Each cluster represents the relevant chlorine isotopic peaks of that particular
PCN congener with characteristic isotope spacing of 1.9970 Da. The congener classes
identified helped direct GCxGC-HRToF analysis. Figure 6.1. Mass defect plot, based on the H/Cl mass scale, generated from FTICR-MS analysis of a Hexagenia spp. extract. Highlighted peaks represent PCNs, Cl-PAHs, X-PAHs and PCDFs, and proposed chemical structures.
Analysis by GCxGC-HRToF demonstrated the complex nature of the Plastimet
sediment mix and variation in isomer bioaccumulation. Estimated BSAFs (and standard
deviations) for PCN, Cl-PAH, X-PAH and PCDF isomers that met peak identification
criteria are presented in Table 6.1. BSAFs ranged 0.34 (± 0.10) to 2.9 (± 0.6) for PCNs,
0.43 (± 0.40) to 21 (± 20) for Cl-anthracenes/phenanthrenes, and 1.1 (± 0.4) to 2.5 (± 2.0)
for Cl-pyrenes/fluoranthenes. BSAFs ranged 0.10 (± 0.04) to 2.2 (± 1.7) for X-
anthracenes/phenanthrenes and the BSAF for the X-pyrene/fluoranthene isomer
(C16H8BrCl) was 3.5 (± 1.7). The BSAFs for PCDFs ranged 0.040 (± 0.030) to 1.2 (±
0.7).
Variation in C10H3Cl5 isomer bioaccumulation (circled peaks) between species
relative to the Plastimet sediment mix to which they were exposed is shown in Figure 6.2.
As a result of mismatched peak retention time coordinates between the Plastimet
sediment mix and biota extracts, other isomers (uncircled) could not be confirmed as
bioaccumulative. Similarly, several compounds identified in the mass defect plot of the
Hexagenia spp. extract (Figure 6.1) do not have corresponding BSAFs in Table 6.1
because positive identification of bioaccumulative isomers was restricted to peaks
observed in every biota extract replicate. Despite these analytical limitations, Figure 6.2
illustrates that the number and type of bioaccumulative isomers varies between species.
From available standard solutions it was possible to confirm the identity of six
isomers presented in Table 6.2. Isomers (a) and (c) of C10H3Cl5 were identified as
1,2,3,5,7-pentachloronaphthalene and 1,2,3,4,6-pentachloronaphthalene, respectively.
Isomer (a) of C10H2Cl6 was identified as 1,2,3,4,6,7-hexachloronaphthalene and/or
1,2,3,5,6,7-hexachloronaphthalene (co-eluting isomers). Isomer (a) of C14H8Cl2
corresponded to 9,10-dichlorophenanthrene. The isomer of C12OH4Cl4 was identified as
2,3,7,8-tetrachlorodibenzofuran and isomer (b) of C12OH3Cl5 corresponded to 1,2,3,7,8-
pentachlorodibenzofuran.
129
Table 6.1. Mean BSAFs for halogenated compounds identified in biota extracts with associated standard deviations (SD). BSAFs for PCNs, Cl-PAHs, X-PAHs, and PCDFs analyzed by GCxGC-HRToF represent isomers. BSAFs for PXDD/PXDFs analyzed by APGC-MS/MS represent congener groups or multiple isomers. Shaded cells indicate BSAFs were statistically greater than (in bold) or less than 1 (p ≤ 0.05, Student's t test). ND indicates analyte was not detected. LD indicates analyte was detected, but did not meet peak identification requirements. *BSAFs based on analyte instrumental responses < LOQ. **BSAFs based on extrapolated sediment concentrations. †n=4.
Compound
Classes
Congener Groups and
Isomers
P. promelas (n=3) Hexagenia spp. (n=4) L. variegatus (n=5)
BSAF SD BSAF SD BSAF SD
PCNs
C10H3Cl5
a 1.0 0.3 LD 1.1 1.1 b 2.1 0.7 LD 2.9 0.6 c ND 1.6 0.2 LD d LD LD 0.39 0.10 e ND LD 1.3 0.6 f ND 0.53 0.15 0.72 0.56
C10H2Cl6
a 0.62 0.11 LD 1.0 0.2 b 0.34 0.10 LD 1.9 0.5 c LD ND 0.48 0.40 d ND LD 0.36 0.16
Cl-Anthracenes/ Phenanthrenes
C14H8Cl2 a LD 21 20 LD b ND LD 4.6 2.3
C14H7Cl3 a LD 9.0 7.2 9.5 4.9 b ND ND 3.2 1.7 c ND ND 0.91 0.37
C14H6Cl4 a ND LD 0.43 0.40
X-Anthracenes/ Phenanthrenes
C14H8BrCl a ND LD 0.18 0.09 b ND LD 0.26 0.15 c ND ND 2.2 1.7
C14H7BrCl2 a ND ND 0.20 0.11 b ND ND 0.10 0.04
Cl-Pyrenes/ Fluoranthenes
C16H8Cl2 a ND LD 2.3 0.6 b LD 1.1 0.4 2.5 2.0
C16H7Cl3 a ND 1.3 0.7 2.1 0.6 X-Pyrenes/
Fluoranthenes C16H8BrCl a ND LD 3.5 1.7
PCDFs
C12OH5Cl3 a ND ND 1.2 0.7 C12OH4Cl4 a LD LD 0.18 0.10
Table 6.2. Of the isomers identified by GCxGC-HRToF in Table 1, six were confirmed through comparison of retention time coordinates with analytical standards. ND indicates analyte was not detected. LD indicates analyte was detected, but did not meet peak identification requirements. Corresponding estimated values for logKOW and water solubility were generated by the United States Environmental Protection Agency’s EPISuiteTM.
Congener Groups and
Isomers
BSAF Confirmed Isomer logKOW
Solubility (25oC, mg/L) P.
promelas Hexagenia
spp. L.
variegatus
C10H3Cl5 a 1.0 LD 1.1 1,2,3,5,7-
pentachloronaphthalene 6.4 0.043
c ND 1.6 LD 1,2,3,4,6-pentachloronaphthalene 7.0 0.013
C10H2Cl6 a 0.62 LD 1.0 1,2,3,4,6,7- or 1,2,3,5,6,7-hexachloronaphthalene
7.7 or 7.0
0.0020 or 0.0075
C14H8Cl2 a LD 21 LD 9,10-dichlorophenanthrene 5.6 0.029
C12OH4Cl4 a LD LD 0.18 2,3,7,8-tetrachlorodibenzofuran 6.6 0.0019
C12OH3Cl5 b ND ND 0.040 1,2,3,7,8-pentachlorodibenzofuran 7.3 0.00034
131
Figure 6.2. Two-dimensional selected ion chromatograms for mass range 299.8400-299.8800 highlighting bioaccumulative isomers of C10H3Cl5. Circled isomer peaks were detected in all sample replicates for the species indicated and peak intensities were used to determine BSAFs relative to those corresponding peaks in the Plastimet sediment mix (Table 1). Isomers a and c were confirmed with a PCN standard solution as 1,2,3,5,7-pentachloronaphthalene and 1,2,3,4,6-pentachloronaphthalene, respectively. The two remaining unlabeled peaks in the PCN standard chromatogram correspond to 1,2,3,6,7-pentachloronaphthalene and 1,2,3,5,8-pentachloronaphthalene.
132
6.4.3 APGC-MS/MS Analysis
Due to low concentrations, PXDD/PXDFs were not observed in FTICR-MS initial
screening and a more sensitive technique was required. Analysis by APGC-MS/MS
provided chromatograms with isomer peak patterns that were unique to the
PXDD/PXDFs examined (Table S1). Peak pattern retention time regions were determined
using PXDD/PXDF standards and the Plastiment sediment mix extract. Relative isomer
peak patterns observed for PXDD/PXDF congener groups in Plastimet sediment mix and
L. variegatus extracts are shown in Figure 6.3. For PXDFs (C12OH5BrCl2, C12OH4BrCl3,
C12OH3BrCl4, and C12OH4Br2Cl2), BSAFs correspond to the congener group rather than
individual isomers because entire isomer peak patterns in the Plastimet sediment mix
matched those in the biota extracts. In contrast, for other PXDD/PXDFs (C12OH3Br2Cl3,
C12O2H5BrCl2, and C12O2H4BrCl3), BSAFs correspond to selected isomer peaks that were
more prominent in biota extracts than other isomer peaks in the Plastimet sediment mix,
suggesting particular isomers may be more bioaccumulative than others. These peaks
were not considered interferences, as they were not observed in solvent blanks, extraction
method blanks, pre-exposed biota extracts, or biota extracts from control sediment test
jars.
Of the PXDD/PXDFs examined, C12OH3BrCl4, C12OH4Br2Cl2, C12OH3Br2Cl3, and
C12O2H4BrCl3 were identified in L. variegatus extracts, while C12OH5BrCl2,
C12OH4BrCl3, and C12O2H5BrCl2 were identified in both L. variegatus and Hexagenia
spp extracts. Estimated BSAFs (and standard deviations) for PXDFs and PXDDs ranged
0.14 (± 0.05) to 1.3 (± 0.2) and 0.89 (± 0.13) to 3.0 (± 0.3), respectively (Table 6.1).
PXDDs with six to eight halogen substituents (C12O2H2BrCl5, C12O2HBrCl6, and
C12O2BrCl7) were observed in the Plastimet sediment mix, but not in biota extracts.
PXDDs, C12O2H3BrCl4 and C12O2H4Br2Cl2, were observed in both the Plastimet sediment
mix and biota extracts, however isomer peak patterns were too weak to meet quantitative
method requirements.
133
Figure 6.3. APGC-MS/MS chromatograms corresponding to MRM transitions of bioaccumulative PXDD/PXDFs: A) C12OH5BrCl2 (313.8 > 206.8), B) C12OH4BrCl3 (347.7 > 240.8), C) C12OH3BrCl4 (381.8 > 274.5), D) C12OH4Br2Cl2 (393.6 > 286.9) , E) C12OH3Br2Cl3 (427.6 > 320.9), F) C12O2H5BrCl2 (329.6 > 159.8), and G) C12O2H4BrCl3 (363.7 > 256.7). Top and bottom chromatograms correspond to Plastimet sediment mix and L. variegatus extracts, respectively. Peaks shaded in black correspond to peak areas used in BSAF calculations.
Time16.00 17.00 18.00 19.00 20.00
%
0
100
16.00 17.00 18.00 19.00 20.00
%
0
100
Time (min)
Intensity
(%)
Time19.00 20.00 21.00 22.00 23.00
%
0
100
19.00 20.00 21.00 22.00 23.00
%
0
100
Time23.00 24.00 25.00 26.00 27.00
%
0
100
23.00 24.00 25.00 26.00 27.00
%
0
100
Time21.00 22.00 23.00 24.00 25.00
%
0
100
21.00 22.00 23.00 24.00 25.00
%
0
100
Time25.00 26.00 27.00 28.00 29.00 30.00
%
0
100
25.00 26.00 27.00 28.00 29.00 30.00
%
0
100
Time17.00 17.50 18.00 18.50 19.00
%
0
100
17.00 17.50 18.00 18.50 19.00
%
0
100
Time20.00 20.50 21.00 21.50 22.00 22.50
%
0
100
20.00 20.50 21.00 21.50 22.00 22.50
%
0
100
A
B
C
D
E
F
G
134
6.4.4 Method Performance
Precision of the GCxGC-HRToF was demonstrated by a triplicate injection of a P.
promelas biota extract which generated BSAFs with percent relative standard deviations
(%RSD) for C10H2Cl6 isomers (a) and (b) of 46% and 40%, respectively. All
experimental masses were within 20 ppm of corresponding theoretical masses. Isotopic
peak ratios were generally within 20% of theoretical values. A three point internal
standard calibration curve constructed from 10, 40, and 200 pg/µL C12-2,3,7,8-Cl4DF
(with 100 pg/µL 13C12-2,3,7,8-Cl4DF) standard solutions gave an R2 value of 0.9943.
Good precision was shown for APGC-MS/MS analysis with %RSDs for multiple
PXDD/PXDF standard injections ranging from 0.55% for 20 pg/µL 8-Br-2,3,4-Cl3DF (n
= 4) to 23% for 1 pg/µL 1,3-Br2-2,7,8-Cl3DF (n = 3). A triplicate injection of an L.
variegatus biota extract generated BSAFs with %RSDs ranging from 1.5% for
C12O2H4BrCl3 to 7% for C12OH3Br2Cl3.
For BSAFs that were statistically greater than or less than 1 (Table 6.1), %RSDs
ranged 8-93% (mean = 37%, n = 31). The range in error associated with BSAFs is likely
a cumulative result of GCxGC-HRToF instrumental precision error and variation
between bioaccumulation study test jars.
6.5 Discussion
6.5.1 Variation in Contaminant Uptake Between Species
Several factors were considered when comparing contaminant uptake by different
organisms. First, species behaviour influences contaminant exposure in aquatic systems.
L. variegatus and Hexagenia spp. are invertebrates that burrow into and ingest sediments
and therefore have direct interactions with contaminants that are sorbed to sediment and
dissolved in pore water(52). P. promelas are exposed to contaminants through ingestion
and re-suspension of sediments, as well as water-soluble contaminants via absorption
through gills and skin(52). Second, lipid content influences uptake of halogenated organic
contaminants. In the present study, approximate lipid contents were 1-2%, 0.5-2%, and
3% for L. variegatus, Hexagenia spp., and P. promelas, respectively. Third, the
135
organism’s ability to metabolize and eliminate the contaminant reduces bioaccumulation.
Finally, as a result of these factors, the time required for test jars to reach steady state
varies between species.
Of the 36 PCN, Cl-PAH, X-PAH and PCDF isomers and PXDD/PXDF congener
groups listed in Table 6.1, 94% were identified in L. variegatus extracts, while 25% and
11% were identified in Hexagenia spp. and P. promelas extracts, respectively. The high
bioaccumulation potential of these contaminants exhibited in L. variegatus may
correspond to its direct interactions with sediment and slow metabolism of PAHs(53-55).
Previous studies have shown similar bioaccumulation trends for polychlorinated aromatic
contaminants in L. variegatus and Hexagenia spp.(46, 56), however in the present study,
fewer contaminants were identified in Hexagenia spp. tissue extracts relative to L.
variegatus. This may be due to variation in contaminant uptake and elimination among
individuals, although mechanisms of metabolism in Hexagenia spp. are unknown.
Despite P. promelas having the highest lipid content among the three species tested, only
four PCN isomers were identified in P. promelas tissue extracts. This may correspond to
previous studies that demonstrated contaminant concentrations in fish do not always
reflect exposure(57) due to metabolism of organic contaminants, such as PAHs(58), or
due to limited direct interactions with sediment. An exposure study with polychlorinated
biphenyls (PCBs) observed higher uptake and elimination rates for Hexagenia spp. and L.
variegatus relative to P. promelas(56). The authors attributed the findings to differences
in exposure routes and suggested slower uptake rates in P. promelas may influence time
required for the system to reach steady state(56).
6.5.2 BSAFs of PCNs, Cl-PAHs, X-PAHs, PCDFs, and PXDD/PXDFs
Based on the equilibrium partitioning model, the theoretical range for organic
contaminant BSAFs is between 1 and 2(59). The BSAFs presented in Table 6.1 represent
a simplified model for examining the relative bioaccumulation potential of non-targeted
contaminant isomers and congener groups in three species. The BSAF model used in the
present study makes several assumptions that may lead to BSAFs outside the theoretical
range of 1-2. It was assumed that at 28 days test jars were at steady state, sediment was
the only contaminant source, no metabolic degradation occurred, and all types of lipid
136
and organic carbon were equal(56, 60). In addition, it was assumed that contaminant
concentration in the organism is a linear uptake function of sediment concentration(61),
and concentrations of neutral organic contaminants are a function of organism lipid and
sediment organic carbon content(62).
Congener groups of PCNs (C10H3Cl5 and C10H2Cl6) were identified in all three
organisms with BSAFs ranging 0.34-2.9. A study of PCNs in a Baltic Sea benthic food
chain identified a similar BSAF range of 0.69-1.4(63), and a Lake Ontario study observed
C10H3Cl5 and C10H2Cl6 to be prevalent PCNs in benthic species(64). PCNs are known
contaminants of concern and their environmental occurrence and toxicity has been
reviewed by Falandysz(65) and Domingo(66). In the present study, GCxGC-HRToF
allowed identification of specific bioaccumulative isomers without the use of analytical
standards. As shown in Table 6.1, most PCN isomers were identified in L. variegatus,
whereas only certain PCN isomers exhibited bioaccumulation potential in Hexagenia spp.
and P. promelas. Of the analytes examined, only PCNs were identified to have
bioaccumulation potential in P. promelas. This may be a result of increased water
solubility and associated bioavailability in the water phase of smaller PCN molecules.
This is evident from the confirmed isomers presented in Table 6.2 substituted with four or
more chlorine atoms for which estimated PCN solubilities (0.0020-0.043) are greater than
those of PCDFs (0.00034-0.0019)(51).
The BSAFs estimated for Cl-PAHs and X-PAHs varied widely between isomers
(Table 1). The Cl-anthracenes/phenanthrenes exhibited the largest range of BSAFs from
0.43 to 21. Isomer (a) of C14H8Cl2 identified as 9,10-dichlorophenanthrene had the
highest BSAF of 21 in Hexagenia spp., however this value was not found to be
statistically greater than 1. A total of five X-anthracene/phenanthrene isomers
corresponding to C14H8BrCl and C14H7BrCl2 were identified in L. variegatus with BSAFs
ranging 0.10-2.2. Among X-anthracenes/phenanthrenes, isomer (c) of C14H8BrCl had the
highest bioaccumulation potential with a BSAF of 2.2, however this value was not found
to be statistically greater than 1. Relative to other contaminants, the Cl-pyrenes
demonstrated higher bioaccumulation potential in both Hexagenia spp. and L. variegatus
extracts with BSAFs ranging 1.1-2.5. The only X-pyrene identified was C16H8BrCl with a
BSAF of 3.5. To date, only two studies have reported Cl-PAHs in environmental biota
137
samples(6, 67), despite increasing evidence of the toxicological risks they pose(13, 15,
16). The present study provides further evidence of the bioaccumulation potential of Cl-
PAHs and new evidence of the bioaccumulation potential of X-PAHs.
The BSAFs estimated for PCDFs were not found to be statistically greater than 1.
Similarly, Van Geest et al. observed BSAFs less than 1 for PCDD/PCDFs in L.
variegatus, Hexagenia spp. and P. promelas exposed to various sediments(46). Yunker
and Cretney identified higher BSAFs for PCDDs relative to PCDFs in dungeness
crabs(68). In the present study, PCDDs were not observed, which was due to inadequate
instrumental detection limits. Yunker and Cretney also identified higher BSAFs for
2,3,7,8-substituted PCDD/PCDFs compared to other isomers(68). Similarly, in the
present study, the only tetrachlorodibenzofuran isomer identified corresponded to the
retention time coordinates of 13C12-2,3,7,8-Cl4DF, confirming this isomer as 2,3,7,8-
tetrachlorodibenzofuran.
While only one or two isomers were identified for each PCDF congener group by
GCxGC-HRToF analysis, isomer peak patterns of PXDFs (C12OH5BrCl2, C12OH4BrCl3,
C12OH3BrCl4, and C12OH4Br2Cl2) observed in APGC-MS/MS analysis indicated
bioaccumulation potential for nearly all isomers in the congener group (Figure 6.3). In
contrast, chromatograms for one PXDF (C12OH3Br2Cl3) and PXDDs (C12O2H5BrCl2 and
C12O2H4BrCl3) indicated selected isomer peaks were more prominent in biota extracts
than others observed in the Plastimet sediment mix (Figure 6.3). Higher bioaccumulation
potential was shown for C12OH5BrCl2 and C12O2H5BrCl2 with BSAFs statistically greater
than 1. To date, only three studies have reported PXDD/PXDFs in environmental biota
samples(32-34), despite known toxicological risks(12, 14). The present study provides
evidence of the bioaccumulation potential of PXDD/PXDFs in freshwater invertebrates.
6.5.3 Considerations for Combustion-derived Contaminant Bioaccumulation
In some instances the bioaccumulation potential of a contaminant may be
predicted by its octanol-water coefficient (KOW) or hydrophobicity. The estimated
logKOW values for confirmed isomers in Table 6.2 range from 5.6 to 7.7(51), suggesting
bioaccumulation will be influenced by both organism uptake rates and contaminant
desorption from sediment(55). For contaminants with logKOW values greater than 6, the
138
time required for a test system to reach equilibrium, as well as measurement and
extraction techniques, may also influence (and possibly underestimate) bioaccumulation
potential(69).
A correlation between bioavailability of planar molecules and black carbon
content of sediment has been reported previously(70-74). It is expected that strong π-π
interactions exist between soot particles and planar contaminants, which decreases their
desorption rate from sediments, thereby affecting their bioavailability and
bioaccumulation(72). In the present study, the extraction method isolated planar
compounds through a carbon column cleanup, and the soil collected from the Plastimet
fire site undoubtedly contained significant amounts of black carbon as a result of
incomplete combustion. Since PCNs, Cl-PAHs, X-PAHs, PCDFs and PXDD/PXDFs are
strongly associated with combustion processes, their bioaccumulation potential may be
influenced by their association with black carbon.
6.5.4 Potential and Limitations in Non-targeted Analysis of Environmental
Contaminants
The non-targeted approaches applied here are accessible to a modern laboratory.
Mass defect filtering may be performed with any high-resolution mass spectral data set
and serves as a valuable initial step in screening sample extracts for analytes of interest. It
has been applied in many fields of study(75), including petroleomics(76-78), but has only
recently been used to identify halogenated contaminants in environmental samples(5, 6,
11).
Analysis by GCxGC-HRToF provides detailed information about isomer specific
bioaccumulation. The advantage of chromatographic peak separation and high-resolution
mass spectral information for distinguishing between a Cl-PAH and X-PAH with a
common unit mass is demonstrated in Figure S1 and has been described previously(11).
While the peaks are clearly chromatographically separated in this example, corresponding
high-resolution mass spectra provide additional confirmation of their identity through
accurate mass, isotopic peak ratios and mass fragments. Chromatographic separation is
most valuable in the present study for its ability to distinguish between isomers in an
unknown mixture that would not be resolved by high-resolution mass spectrometry alone.
139
Further characterization of bioaccumulative isomers would entail comparison of
chromatographic peaks of interest with analytical standards of the suspected isomer.
Isomeric structure may also be deduced using relative retention time coordinates of other
isomers in the congener group.
An important limitation in the present study was the slow acquisition rate of the
HRToF. Typically, GCxGC-MS analysis is performed at an acquisition rate greater than
100Hz with unit mass resolution, however low mass resolution is limiting in the
identification of unknown contaminants in complex mixtures(7). In the present study, the
acquisition rate of 10 Hz produced peak slices with less than 10 points over the peak,
which did not meet typical quantitative requirements. Since the intensity of the most
prominent peak slice was used for analysis, the analytical accuracy and precision relied
on a single mass spectrum. This likely contributed to peak intensity error in replicate
injections and BSAFs, despite accurate masses and isotopic peak ratios meeting other
method requirements. Coupling GCxGC to an HRToF capable of higher acquisition rates,
as several recent studies have done(7, 9, 79), serves as a powerful enhancement in non-
targeted analysis of complex environmental samples.
6.6 Acknowledgements
We thank Wellington Laboratories for PXDD/PXDF solutions, as well as Vince
Taguchi for use of the FTICR-MS. Thanks to Liad Haimovici, Christie Hartley, Leo
Yeung, Miren Pena, Bert van Bavel, Frank Wania, and Derek Muir for useful discussions
and feedback. Thanks also to Kurunthachalam Kannan for Cl-PAH standards and the
MOE Aquatic Toxicology Unit.
6.7 References
1. E. Skoczyńska, P. Korytár, J. de Boer, Maximizing chromatographic information
from environmental extracts by GCxGC-ToF-MS. Environ. Sci. Technol. 42, 6611–6618 (2008).
2. K. Pangallo, R. K. Nelson, E. L. Teuten, B. E. Pedler, C. M. Reddy, Expanding the range of halogenated 1′-methyl-1,2′-bipyrroles (MBPs) using GC/ECNI-MS and GC×GC/TOF-MS. Chemosphere 71, 1557–1565 (2008).
140
3. E. Hoh et al., Capabilities of direct sample introduction−comprehensive two-dimensional gas chromatography−time-of-flight mass spectrometry to analyze organic chemicals of interest in fish oils. Environ. Sci. Technol. 43, 3240–3247 (2009).
4. E. Hoh et al., Nontargeted comprehensive two-dimensional gas chromatography/time-of-flight mass spectrometry method and software for inventorying persistent and bioaccumulative contaminants in marine environments. Environ. Sci. Technol. 46, 8001–8008 (2012).
5. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass. Spectrom. 21, 1918–1921 (2010).
6. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
7. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
8. S. Hashimoto et al., Global and selective detection of organohalogens in environmental samples by comprehensive two-dimensional gas chromatography–tandem mass spectrometry and high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3799–3810 (2011).
9. S. Hashimoto et al., Selective extraction of halogenated compounds from data measured by comprehensive multidimensional gas chromatography/high resolution time-of-flight mass spectrometry for non-target analysis of environmental and biological samples. J. Chromatogr. A 1282, 183–189 (2013).
10. A. L. Myers, S. A. Mabury, E. J. Reiner, Analysis of mixed halogenated dibenzo-p-dioxins and dibenzofurans (PXDD/PXDFs) in soil by gas chromatography tandem mass spectrometry (GC–MS/MS). Chemosphere 87, 1063–1069 (2012).
11. S. Fernando et al., Identification of the halogenated compounds resulting from the 1997 Plastimet Inc. fire in Hamilton, Ontario, using comprehensive two-dimensional gas chromatography and (ultra)high resolution mass spectrometry. Environ. Sci. Technol., 48, 10656-10663 (2014).
12. G. Mason, T. Zacharewski, M. A. Denomme, L. Safe, S. Safe, Polybrominated dibenzo-p-dioxins and related compounds: quantitative in vivo and in vitro structure-activity relationships. Toxicology 44, 245–255 (1987).
13. P. P. Fu, L. S. Von Tungeln, L.-H. Chiu, Z. Y. Own, Halogenated-polycyclic
141
aromatic hydrocarbons: a class of genotoxic environmental pollutants. J. Environ. Sci. Heal. C 17, 71–109 (1999).
14. H. Olsman et al., Relative differences in aryl hydrocarbon receptor-mediated response for 18 polybrominated and mixed halogenated dibenzo-p-dioxins and -furans in cell lines from four different species. Environ. Toxicol. Chem. 26, 2448–2454 (2007).
15. T. Ohura, K.-I. Sawada, T. Amagai, M. Shinomiya, Discovery of novel halogenated polycyclic aromatic hydrocarbons in urban particulate matters: occurrence, photostability, and AhR activity. Environ. Sci. Technol. 43, 2269–2275 (2009).
16. H. Sakakibara et al., Organ-specific distribution of 7-chlorinated benz[a]anthracene and regulation of selected cytochrome P450 genes in rats. J. Toxicol. Sci. 38, 137–143 (2013).
17. T. Ohura, Environmental behavior, sources, and effects of chlorinated polycyclic aromatic hydrocarbons. Sci. World J. 7, 372–380 (2007).
18. J.-L. Sun, H. Zeng, H.-G. Ni, Halogenated polycyclic aromatic hydrocarbons in the environment. Chemosphere 90, 1751–1759 (2013).
19. P. Haglund, T. Alsberg, A. Bergman, B. Jansson, Analysis of halogenated polycyclic aromatic hydrocarbons in urban air, snow and automobile exhaust. Chemosphere 16, 2441–2450 (1987).
20. G. W. Sovocool et al., Bromo- and bromochloro-polynuclear aromatic hydrocarbons, dioxins and dibenzofurans in municipal incinerator fly ash. Biomed. Environ. Mass 15, 669–676 (1988).
21. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
22. W. Schafer, K. Ballschmiter, Monobromo-polychloro-derivatives of benzene, biphenyl, dibenzofurane and dibenzodioxine formed in chemical-waste burning. Chemosphere 15, 755–763 (1986).
23. R. L. Harless, R. G. Lewis, D. D. McDaniel, A. E. Dupuy Jr, Identification of bromo/chloro dibenzo-p-dioxins and dibenzofurans in ash samples. Chemosphere 18, 201–208 (1989).
24. H. Y. Tong, S. J. Monson, M. L. Gross, L. Q. Huang, Monobromopolychlorodibenzo-p-dioxins and dibenzofurans in municipal waste incinerator flyash. Anal. Chem. 63, 2697–2705 (1991).
25. W. Chatkittikunwong, C. S. Creaser, Bromo-, bromochloro- and chloro-dibenzo-p-
142
dioxins and dibenzofurans in incinerator fly ash. Chemosphere 29, 559–566 (1994).
26. H. Preud'homme, M. Potin-Gautier, Optimization of accelerated solvent extraction for polyhalogenated dibenzo-p-dioxins and benzo-p-furans in mineral and environmental matrixes using experimental designs. Anal. Chem. 75, 6109–6118 (2003).
27. T. Oberg, K. Warman, J. Bergstrom, Brominated aromatics from combustion. Chemosphere 16, 2451–2465 (1987).
28. B. Du et al., Mixed polybrominated/chlorinated dibenzo-p-dioxins and dibenzofurans in stack gas emissions from industrial thermal processes. Environ. Sci. Technol. 44, 5818–5823 (2010).
29. K. Hayakawa, H. Takatsuki, I. Watanabe, S.-I. Sakai, Polybrominated diphenyl ethers (PBDEs), polybrominated dibenzo-p-dioxins/dibenzofurans (PBDD/Fs) and monobromo-polychlorinated dibenzo-p-dioxins/dibenzofurans (MoBPXDD/Fs) in the atmosphere and bulk deposition in Kyoto, Japan. Chemosphere 57, 343–356 (2004).
30. K. Kannan, I. Watanabe, J. P. Giesy, Congener profile of polychlorinated/brominated dibenzo-p-dioxins and dibenzofurans in soil and sediments collected at a former chlor-alkali plant. Toxicol. Environ. Chem. 67, 135–146 (1998).
31. H. Terauchi, S. Takahashi, P. K. S. Lam, B.-Y. Min, S. Tanabe, Polybrominated, polychlorinated and monobromo-polychlorinated dibenzo-p-dioxins/dibenzofurans and dioxin-like polychlorinated biphenyls in marine surface sediments from Hong Kong and Korea. Environ. Pollut. 157, 724–730 (2009).
32. M. Unger et al., Polybrominated and mixed brominated/chlorinated dibenzo-p-dioxins in sponge ( Ephydatia fluviatilis) from the Baltic Sea. Environ. Sci. Technol. 43, 8245–8250 (2009).
33. A. R. Fernandes et al., Mixed brominated/chlorinated dibenzo-p-dioxins, dibenzofurans and biphenyls: simultaneous congener-selective determination in food. J. Chromatogr. A 1218, 9279–9287 (2011).
34. D. Zacs, J. Rjabova, V. Bartkevics, Occurrence of brominated persistent organic pollutants (PBDD/DFs, PXDD/DFs, and PBDEs) in Baltic Wild Salmon (Salmo salar) and correlation with PCDD/DFs and PCBs. Environ. Sci. Technol. 47, 9478–9486 (2013).
35. T. Ohura, A. Kitazawa, T. Amagai, Seasonal variability of 1-chloropyrene on atmospheric particles and photostability in toluene. Chemosphere 57, 831–837 (2004).
143
36. T. Ohura, A. Kitazawa, T. Amagai, M. Makino, Occurrence, profiles, and photostabilities of chlorinated polycyclic aromatic hydrocarbons associated with particulates in urban air. Environ. Sci. Technol. 39, 85–91 (2005).
37. A. Kitazawa, T. Amagai, T. Ohura, Temporal trends and relationships of particulate chlorinated polycyclic aromatic hydrocarbons and their parent compounds in urban air. Environ. Sci. Technol. 40, 4592–4598 (2006).
38. T. Ohura, S. Fujima, T. Amagai, M. Shinomiya, Chlorinated polycyclic aromatic hydrocarbons in the atmosphere: seasonal levels, gas-particle partitioning, and origin. Environ. Sci. Technol. 42, 3296–3302 (2008).
39. J. Ma et al., Airborne PM 2.5/PM 10-associated chlorinated polycyclic aromatic hydrocarbons and their parent compounds in a suburban area in Shanghai, China. Environ. Sci. Technol. 47, 7615–7623 (2013).
40. T. Ohura, Y. Horii, M. Kojima, Y. Kamiya, Diurnal variability of chlorinated polycyclic aromatic hydrocarbons in urban air, Japan. Atmos. Environ. 81, 84–91 (2013).
41. J. A. Arnot, F. A. Gobas, A review of bioconcentration factor (BCF) and bioaccumulation factor (BAF) assessments for organic chemicals in aquatic organisms. Environ. Rev. 14, 257–297 (2006).
42. A. Poland, J. C. Knutson, 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related halogenated aromatic hydrocarbons: examination of the mechanism of toxicity. Annu. Rev. Pharmacol. 22, 517–554 (1982).
43. M. Van den Berg et al., Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. Environ. Health Persp. 106, 775–792 (1998).
44. A. B. Okey, D. S. Riddick, P. A. Harper, The Ah receptor: mediator of the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Toxicol. Lett. 70, 1–22 (1994).
45. Ontario Ministry of the Environment, Bioaccumulation of sediment-associated contaminants in freshwater organisms (Laboratory Services Branch Method 3495, 2001.
46. J. L. Van Geest, D. G. Poirier, K. R. Solomon, P. K. Sibley, A comparison of the bioaccumulation potential of three freshwater organisms exposed to sediment-associated contaminants under laboratory conditions. Environ. Toxicol. Chem. 30, 939–949 (2011).
47. Ontario Ministry of Environment, Plastimet Inc. fire, Hamilton, Ontario, July 9-12, 1997. https://ia700309.us.archive.org/24/items/plastimetincfire00sochuoft/plastimetincfire00sochuoft.pdf (1997).
144
48. Ontario Ministry of the Environment, The determination of total carbonate-carbon in soil and sediments by coulometry (Laboratory Services Branch Method 3012, 2010).
49. Ontario Ministry of the Environment, The determination of polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans and dioxin-like polychlorinated biphenyls (DLPCBs) in environmental matrices by gas chromatography-high resolution mass spectrometry (GC-HRMS) (Laboratory Services Branch Method 3418, 2010).
50. Ontario Ministry of the Environment, The determination of polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, dioxin-like polychlorinated biphenyls (DLPCBs), polychlorinated naphthalenes (PCNs) and polybrominated diphenyl ethers (PBDEs) in biota samples by automated sample extraction/cleanup and gas chromatography-high resolution mass spectrometry (GC-HRMS) (Laboratory Services Branch Method 3481, 2012).
51. U.S. Environmental Protection Agency, Exposure Assessment Tools and Models. Estimation Program Interface (EPI) Suite, Version 4.11 (Office of Pollution Prevention and Toxics, 2012) http://www.epa.gov/opptintr/exposure/pubs/episuite.htm.
52. J. L. Van Geest, D. G. Poirier, P. K. Sibley, K. R. Solomon, Measuring bioaccumulation of contaminants from field-collected sediment in freshwater organisms: A critical review of laboratory methods. Environ. Toxicol. Chem. 29, 2391–2401 (2010).
53. M. T. Lappanen, J. V. Kukkonen, Fate of sediment-associated pyrene and benzo[a]pyrene in the freshwater oligochaete Lumbriculus variegatus (Muller). Aquat. Toxicol. 49, 199–212 (2000).
54. L. J. Schuler, M. Wheeler, A. J. Bailer, M. J. Lydy, Toxicokinetics of sediment-sorbed benzo[a]pyrene and hexachlorobiphenyl using the freshwater invertebrates Hyalella azteca, Chironomus tentans, and Lumbriculus variegatus. Environ. Toxicol. Chem. 22, 439–449 (2003).
55. M. Lyytikainen, S. Pehkonen, J. Akkanen, M. Leppanen, J. V. Kukkonen, Bioaccumulation and biotransformation of polycyclic aromatic hydrocarbons during sediment tests with oligochaetes (Lumbriculus variegatus). Environ. Toxicol. Chem. 26, 2660–2666 (2007).
56. J. L. Van Geest, D. MacKay, D. G. Poirier, P. K. Sibley, K. R. Solomon, Accumulation and Depuration of Polychlorinated Biphenyls from Field-Collected Sediment in Three Freshwater Organisms. Environ. Sci. Technol. 45, 7011–7018 (2011).
57. R. van der Oost, J. Beyer, N. P. Vermeulen, Fish bioaccumulation and biomarkers in environmental risk assessment: a review. Environ. Toxicol. Phar. 13, 57–149
145
(2003).
58. U. Varanasi, J. E. Stein, M. Nishimoto, in Metabolism of polycyclic aromatic hydrocarbons in the aquatic environment, (CRC Press, Boca Raton, Florida, 1989).
59. L. P. Burkhard, Estimation of biota sediment accumulation factor (BSAF) from paired observations of chemical concentrations in biota and sediment (EPA/600/R-06/047, U.S. Environmental Protection Agency, 2009).
60. J. L. Van Geest, D. G. Poirier, P. K. Sibley, K. R. Solomon, Validation of Ontario's new laboratory-based bioaccumulation methods with in situ field data. environmental Toxicology and Chemistry 30, 950–958 (2011).
61. Bechtel Jacobs Company LLC, Biota Sediment Accumulation Factors for Invertebrates: Review and Recommendations for the Oak Ridge Reservation (BJC/OR-112, Oak Ridge, TN, 1998).
62. U.S. Environmental Protection Agency, Methods for measuring the toxicity and bioaccumulation of sediment-associated contaminants with freshwater invertebrates (EPA 600/R-99/064, Duluth, MN, 2000).
63. K. Lundgren, M. Tysklind, R. Ishaq, D. Broman, B. Van Bavel, Polychlorinated naphthalene levels, distribution, and biomagnification in a benthic food chain in the Baltic Sea. Environ. Sci. Technol. 36, 5005–5013 (2002).
64. P. A. Helm et al., Occurrence and biomagnification of polychlorinated naphthalenes and non- and mono-orthoPCBs in Lake Ontario sediment and biota. Environ. Sci. Technol. 42, 1024–1031 (2008).
65. J. Falandysz, Polychlorinated naphthalenes: an environmental update. Environmental Pollution 101, 77–90 (1998).
66. J. L. Domingo, Polychlorinated naphthalenes in animal aquatic species and human exposure through the diet: a review. J. Chromatogr. A 1054, 327–334 (2004).
67. Y. Horii, T. Ohura, N. Yamashita, K. Kannan, Chlorinated polycyclic aromatic hydrocarbons in sediments from industrial areas in Japan and the United States. Arch. Environ. Contam. Toxicol. 57, 651–660 (2009).
68. M. B. Yunker, W. J. Cretney, Bioavailability of chlorinated dibenzo-p -dioxins and dibenzofurans to dungeness crab (Cancer magister) at marine pulp mill sites in British Columbia, Canada. Environ. Toxicol. Chem. 19, 2997–3011 (2000).
69. M. T. O. Jonker, S. A. van der Heijden, Bioconcentration factor hydrophobicity cutoff: an artificial phenomenon reconstructed. Environ. Sci. Technol. 41, 7363–7369 (2007).
146
70. G. Cornelissen et al., Extensive sorption of organic compounds to black carbon, coal, and kerogen in sediments and soils: mechanisms and consequences for distribution, bioaccumulation, and biodegradation. Environ. Sci. Technol. 39, 6881–6895 (2005).
71. J. V. Kukkonen et al., The contrasting roles of sedimentary plant-derived carbon and black carbon on sediment-spiked hydrophobic organic contaminant bioavailability to Diporeia species and Lumbriculus variegatus. Environ. Toxicol. Chem. 24, 877–885 (2005).
72. O. Gustafsson, F. Haghseta, C. Chan, J. Macfarlane, P. M. Gschwend, Quantification of the dilute sedimentary soot phase: implications for PAH speciation and bioavailability. Environ. Sci. Technol. 31, 203–209 (1997).
73. B. Sundelin, A.-K. E. Wiklund, G. Lithner, O. Gustafsson, Evaluation of the role of black carbon in attenuating bioaccumulation of polycyclic aromatic hydrocarbons from field-contaminated sediments. Environ. Toxicol. Chem. 23, 2611–2617 (2004).
74. G. Cornelissen, O. Gustafsson, Prediction of large variation in biota to sediment accumulation factors due to concentration-dependent black carbon adsorption of planar hydrophobic organic compounds. Environ. Toxicol. Chem. 24, 495–498 (2005).
75. L. Sleno, The use of mass defect in modern mass spectrometry. J. Mass. Spectrom. 47, 226–236 (2012).
76. C. A. Hughey, C. L. Hendrickson, R. P. Rodgers, A. G. Marshall, K. Qian, Kendrick mass defect spectrum: a compact visual analysis for ultrahigh-resolution broadband mass spectra. Anal. Chem. 73, 4676–4681 (2001).
77. Z. Wu, R. P. Rodgers, A. G. Marshall, Two- and three-dimensional van Krevelen diagrams: a graphical analysis complementary to the Kendrick mass plot for sorting elemental compositions of complex organic mixtures based on ultrahigh-resolution broadband Fourier transform ion cyclotron resonance mass measurements. Anal. Chem. 76, 2511–2516 (2004).
78. A. G. Marshall, R. P. Rodgers, Petroleomics: The next grand challenge for chemical analysis. Acc. Chem. Res. 37, 53–59 (2004).
79. N. Ochiai et al., Stir bar sorptive extraction and comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry for ultra-trace analysis of organochlorine pesticides in river water. J. Chromatogr. A 1218, 6851–6860 (2011).
147
7 CHAPTER SEVEN
Summary, Conclusions, and Future Directions
Anne L. Myers, Scott A. Mabury, and Eric. J. Reiner Contributions: Anne L. Myers prepared this chapter with editorial comments provided by Eric J. Reiner.
148
7.1 Summary and Conclusions
This thesis investigated the thermal decomposition mechanisms and products of
halogenated materials using non-targeted analytical techniques, and assessed the
environmental relevance of these thermal processes.
The quartz tube furnace thermal decomposition studies described in Chapters 2
and 3 uncovered novel fluoropolymer thermal decomposition products of
polychlorotrifluoroethylene (PCTFE) and polyvinylidene fluoride (PVDF), respectively.
Previously, Ellis et al. reported the production of a range of perchlorinated/fluorinated
carboxylic acids (PXCAs, X=Cl,F) through PCTFE thermal decomposition in air up to
500oC(1). In Chapter 2, these findings were confirmed for PCTFE thermal decomposition
at 400oC using Fourier transform ion cyclotron resonance mass spectrometry (FTICR-
MS) and mass defect filtering. Chapter 2 also reported a range of novel
dibenzofurans (PBDFs), polyhalogenated dibenzofurans (PXDFs, X=Br,Cl), and
polybrominated anthracenes/phenanthrenes as the main halogenated contaminants
released. These results suggested brominated flame-retardants (BFRs) were the primary
halogenated materials present in both fires. These contaminants have been identified
previously from thermal processes and are of environmental and toxicological concern(8-
15). This study demonstrated the potential of mass defect plots to serve as a screening
tool for contaminants formed in fires of unknown contents.
In Chapter 6, FTICR-MS with mass defect filtering was also used to screen for
bioaccumulative contaminants in freshwater species exposed to soil collected from the
Plastimet recycling plant fire site. Investigation of specific bioaccumulative isomers was
150
performed by GCxGC-HRToF and biota-sediment accumulation factors (BSAFs) were
determined. A range of combustion-derived contaminants were identified as
bioaccumulative, including polychlorinated naphthalenes (PCNs), polychlorinated
dibenzofurans (PCDFs), chlorinated and mixed brominated/chlorinated
anthracenes/phenanthrenes and pyrenes/fluoranthenes (Cl-PAHs and X-PAHs, X=Br,Cl).
These contaminants have been identified previously in the environment and are of
toxicological concern(12, 16-21). The superior chromatographic separation of GCxGC,
coupled with the accurate mass information obtained by HRToF, allowed identification
of specific bioaccumulative isomers without a corresponding analytical standard. This
type of analysis may direct further studies, synthesis of analytical standards, and
development of targeted analytical methods.
With 4600 possible congeners(22) and few available analytical standards,
environmental analysis of polyhalogenated dibenzo-p-dioxins and dibenzofurans
(PXDD/PXDFs, X=Br,Cl) has been limited, despite toxicological concerns(9-11). To
date, PXDD/PXDF identification has been largely associated with incinerator fly ash(23-
28). A targeted gas chromatography tandem mass spectrometry method (GC-MS/MS)
was developed in Chapter 5 for analysis of PXDD/PXDFs in complex environmental
mixtures. This method was used to measure PXDD/PXDF concentrations in soil collected
from the Plastimet recycling plant fire site. Of the PXDFs examined, monobromo-
dichlorodibenzofurans were the most prevalent, reaching a concentration of 180 ng/g. Of
the PXDDs examined, monobromo-hexachlorodibenzo-p-dioxins were the most
prevalent, reaching a concentration of 47 ng/g. This method was also applied in Chapter 6
to identify bioaccumulative PXDD/PXDFs in freshwater species. PXDD/PXDFs
identified as bioaccumulative included monobromo-dichlorodibenzofurans and
monobromo-dichlorodibenzo-p-dioxins. Targeted analytical methods, such as that
developed in Chapter 5, provide quantitative data to complement and support non-
targeted analyses of complex environmental samples.
151
7.2 Future Directions
This thesis presents novel analytical approaches to identifying halogenated
contaminants in complex environmental mixtures, with particular interest in those formed
through thermal processes. The non-targeted techniques of mass defect filtering and
GCxGC-HRToF are promising candidates to greatly enhance the environmental analysis
of halogenated contaminants through a top-down approach. Going forward, there are
several aspects of these analyses that should be considered to improve accessibility and
data quality.
In this work, FTICR-MS was used primarily for the acquisition of high-resolution
mass spectral data. The superior mass resolving power and mass accuracy of FTICR-MS
is valuable in determining the elemental composition and fragmentation patterns
indicative of chemical structure, particularly for the identification of unknowns. Despite
the advantages, FTICR-MS instruments are costly with a price range of $500,000 to $1.4
million, plus the additional costs of maintaining a high-powered Tesla magnet. As a
result, this type of instrumentation is not accessible to the average laboratory(29). The
alternative option of HRToF comes with increased sensitivity and at a lower price of
approximately $450,000. The drawback of HRToF relative to FTICR-MS is lower
resolution(29). Despite this, HRToF instruments operate at resolving powers greater than
10,000 and resulting mass spectra may still be interpreted using mass defect filtering. It is
anticipated that mass defect filtering of high-resolution mass spectra collected by HRToF
will advance the non-targeted analyses of environmental samples.
Today, the majority of laboratories perform target compound analysis, an
approach that only examines a small subset of compounds present in the environment.
There is a need for an automated system to screen data for non-targeted compounds. The
current limitation in regards to analysis by mass defect filtering is the lengthy time
required to interpret a single mass defect plot. While the plot itself reveals homologous
series of ions, the interpreter must investigate each peak or peak cluster in the series for
accurate masses, isotopic peak patterns, fragment ions and low intensity monoisotopic
peaks, in order to assign a reasonable elemental composition with low error. To reduce
the labour intensive nature of this analysis, development of data interpretation software
and an environmental contaminant mass defect database would greatly enhance the speed
152
of mass defect plot interpretation. In this way, many more contaminants could be
identified and monitored.
Similarly, data interpretation software for GCxGC-HRToF chromatograms could
serve as a powerful tool for identification of unknowns in complex environmental
samples. Visually, trace contaminants in a GCxGC-HRToF chromatogram of a complex
environmental sample cannot be identified unless through a selected ion chromatogram,
which requires some notion of analyte masses of interest. Automated software may allow
identification of unknown trace contaminants in such complex data sets. A current
limitation that may hinder this approach is the slow acquisition rate of HRToF relative to
the narrow modulated peaks in GCxGC. Without achieving adequate measurements over
a peak, chromatographic inconsistences would make automated data interpretation
challenging. The GCxGC-HRToF analysis in Chapter 6 used an acquisition rate of 10 Hz,
however several recent studies have examined halogenated organics using GCxGC-
HRToF with acquisition rates of approximately 25 Hz(30-32). Continued developments
in the coupling of GCxGC and HRToF will be important to non-targeted analysis of
complex environmental samples.
As demonstrated in this thesis, thermal decomposition of halogenated materials
produces a complex mixture of environmentally persistent and toxic halogenated
contaminants. The environmental impact of a fire involving halogenated materials varies
greatly depending on the fire contents, temperatures, and conditions. As discussed in
Chapter 3, understanding the link between laboratory thermal decomposition studies and
large-scale uncontrolled fires is challenging. Non-targeted analytical techniques provide a
screening tool to identify novel combustion-derived contaminants and possible source
materials at fire sites. This approach may facilitate identification of environmental
markers to link halogenated material sources to corresponding thermal decomposition
products in the interest of reducing the environmental impact of uncontrolled fires.
153
7.3 References
1. D. A. Ellis, J. W. Martin, D. C. G. Muir, S. A. Mabury, The use of 19F NMR and
mass spectrometry for the elucidation of novel fluorinated acids and atmospheric fluoroacid precursors evolved in the thermolysis of fluoropolymers. Analyst 128, 756–764 (2003).
2. G. Montaudo, C. Puglisi, E. Scamporrino, D. Vitalini, Correlation of thermal degradation mechanisms: polyacetylene and vinyl and vinylidene polymers. J. Polym. Sci. 24, 301–316 (1986).
3. C. A. Hughey, C. L. Hendrickson, R. P. Rodgers, A. G. Marshall, K. Qian, Kendrick mass defect spectrum: a compact visual analysis for ultrahigh-resolution broadband mass spectra. Anal. Chem. 73, 4676–4681 (2001).
4. Z. Wu, R. P. Rodgers, A. G. Marshall, Two- and three-dimensional van Krevelen diagrams: a graphical analysis complementary to the Kendrick mass plot for sorting elemental compositions of complex organic mixtures based on ultrahigh-resolution broadband Fourier transform ion cyclotron resonance mass measurements. Anal. Chem. 76, 2511–2516 (2004).
5. L. Sleno, The use of mass defect in modern mass spectrometry. J. Mass. Spectrom. 47, 226–236 (2012).
6. V. Y. Taguchi, R. J. Nieckarz, R. E. Clement, S. Krolik, R. Williams, Dioxin analysis by gas chromatography-Fourier transform ion cyclotron resonance mass spectrometry (GC-FTICRMS). J. Am. Soc. Mass Spectrom. 21, 1918–1921 (2010).
7. K. J. Jobst et al., The use of mass defect plots for the identification of (novel) halogenated contaminants in the environment. Anal. Bioanal. Chem. 405, 3289–3297 (2013).
8. C. A. de Wit, An overview of brominated flame retardants in the environment. Chemosphere 46, 583–624 (2002).
9. H. Olsman et al., Relative differences in aryl hydrocarbon receptor-mediated response for 18 polybrominated and mixed halogenated dibenzo-p-dioxins and -furans in cell lines from four different species. Environ. Toxicol. Chem. 26, 2448–2454 (2007).
10. G. Mason, T. Zacharewski, M. A. Denomme, L. Safe, S. Safe, Polybrominated dibenzo-p-dioxins and related compounds: quantitative in vivo and in vitro structure-activity relationships. Toxicology 44, 245–255 (1987).
11. L. S. Birnbaum, D. F. Staskal, J. J. Diliberto, Health effects of polybrominated dibenzo-p-dioxins (PBDDs) and dibenzofurans (PBDFs). Environ. Int. 29, 855–860 (2003).
154
12. P. P. Fu, L. S. Von Tungeln, L.-H. Chiu, Z. Y. Own, Halogenated-polycyclic aromatic hydrocarbons: a class of genotoxic environmental pollutants. J. Environ. Sci. Heal. C 17, 71–109 (1999).
13. P. P. Fu, Q. Xia, X. Sun, H. Yu, Phototoxicity and environmental transformation of polycyclic aromatic hydrocarbons (PAHs)—light-induced reactive oxygen species, lipid peroxidation, and DNA damage. J. Environ. Sci. Heal. C 30, 1–41 (2012).
14. E. J. LaVoie, L. Tulley-Freiler, V. Bedenko, D. Hoffmann, Mutagenicity of substituted phenanthrenes in Salmonella typhimurium. Mutat. Res. 116, 91–102 (1983).
15. Y. Horii et al., Relative potencies of individual chlorinated and brominated polycyclic aromatic hydrocarbons for induction of aryl hydrocarbon receptor-mediated responses. Environ. Sci. Technol. 43, 2159–2165 (2009).
16. T. Ohura, K.-I. Sawada, T. Amagai, M. Shinomiya, Discovery of novel halogenated polycyclic aromatic hydrocarbons in urban particulate matters: occurrence, photostability, and AhR activity. Environ. Sci. Technol. 43, 2269–2275 (2009).
17. H. Sakakibara et al., Organ-specific distribution of 7-chlorinated benz[a]anthracene and regulation of selected cytochrome P450 genes in rats. J. Toxicol. Sci. 38, 137–143 (2013).
18. J. L. Domingo, Polychlorinated naphthalenes in animal aquatic species and human exposure through the diet: a review. J. Chromatogr. A 1054, 327–334 (2004).
19. J. Falandysz, Polychlorinated naphthalenes: an environmental update. Environ. Pollut. 101, 77–90 (1998).
20. C. Rappe, Sources and environmental concentrations of dioxins and related compounds. Pure & Appl. Chem. 68, 1781–1789 (1996).
21. A. Schecter, L. Birnbaum, J. J. Ryan, J. D. Constable, Dioxins: An overview. Environ. Res. 101, 419–428 (2006).
22. H.-R. Buser, Brominated and brominated/chlorinated dibenzodioxins and dibenzofurans: potential environmental contaminants. Chemosphere 16, 713–732 (1987).
23. W. Chatkittikunwong, C. S. Creaser, Bromo-, bromochloro- and chloro-dibenzo-p-dioxins and dibenzofurans in incinerator fly ash. Chemosphere 29, 559–566 (1994).
24. R. L. Harless, R. G. Lewis, D. D. McDaniel, A. E. Dupuy Jr, Identification of bromo/chloro dibenzo-p-dioxins and dibenzofurans in ash samples. Chemosphere
155
18, 201–208 (1989).
25. H. Preud'homme, M. Potin-Gautier, Optimization of accelerated solvent extraction for polyhalogenated dibenzo-p-dioxins and benzo-p-furans in mineral and environmental matrixes using experimental designs. Anal. Chem. 75, 6109–6118 (2003).
26. W. Schafer, K. Ballschmiter, Monobromo-polychloro-derivatives of benzene, biphenyl, dibenzofurane and dibenzodioxine formed in chemical-waste burning. Chemosphere 15, 755–763 (1986).
27. G. W. Sovocool et al., Analysis of municipal incinerator fly ash for bromo- and bromochloro-dioxins, dibenzofurans, and related compounds. Chemosphere 18, 193–200 (1989).
28. H. Y. Tong, S. J. Monson, M. L. Gross, L. Q. Huang, Monobromopolychlorodibenzo-p-dioxins and dibenzofurans in municipal waste incinerator flyash. Anal. Chem. 63, 2697–2705 (1991).
29. M. P. Balogh, Debating resolution and mass accuracy in mass spectrometry. Spectroscopy 19, 34–40 (2004).
30. T. Ieda, N. Ochiai, T. Miyawaki, T. Ohura, Y. Horii, Environmental analysis of chlorinated and brominated polycyclic aromatic hydrocarbons by comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3224–3232 (2011).
31. S. Hashimoto et al., Global and selective detection of organohalogens in environmental samples by comprehensive two-dimensional gas chromatography–tandem mass spectrometry and high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1218, 3799–3810 (2011).
32. N. Ochiai et al., Stir bar sorptive extraction and comprehensive two-dimensional gas chromatography coupled to high-resolution time-of-flight mass spectrometry for ultra-trace analysis of organochlorine pesticides in river water. J. Chromatogr. A 1218, 6851–6860 (2011).
156
APPENDIX A
SUPPORTING INFORMATION FOR CHAPTER TWO
Using Mass Defect Plots as a Discovery Tool to Identify Novel Fluoropolymer Thermal Decomposition Products
157
LIST OF TABLES AND FIGURES
Figure S1. Thermal decomposition apparatus. ............................................................... 158 Table S1. LC-MS/MS gradient method ......................................................................... 158 Figure S2. LC-MS/MS chromatograms confirming pentafluorobenzoic acid as a thermal decomposition product of PCTFE at 800oC, with corresponding peaks at 5.14 minutes. ........................................................................................................................... 159 Figure S3. GC-FTICR-MS chromatograms confirming octafluoronaphthalene as a thermal decomposition product of PCTFE at 800oC with corresponding peaks at 12.2 minutes. ........................................................................................................................... 159
158
Figure S1. Thermal decomposition apparatus.
Sample Preparation for LC-MS/MS Analysis
Buffer extracts in methanol were diluted four times in a 300µL polypropylene
(PP) LC vial with a 50:50 methanol:water mixture for LC-MS/MS analysis.
Figure S2. LC-MS/MS chromatograms confirming pentafluorobenzoic acid as a thermal decomposition product of PCTFE at 800oC, with corresponding peaks at 5.14 minutes.
Figure S3. GC-FTICR-MS chromatograms confirming octafluoronaphthalene as a thermal decomposition product of PCTFE at 800oC with corresponding peaks at 12.2 minutes.
Analysis of Mixed Halogenated Dibenzo-p-dioxins and Dibenzofurans (PXDD/PXDFs) in Soil by Gas Chromatography Tandem Mass Spectrometry (GC-
MS/MS)
161
LIST OF TABLES AND FIGURES Table S1. Spike and recovery results for three different PXDD/PXDF spiking concentrations. ................................................................................................................ 162
162
Table S1. Spike and recovery results for three different PXDD/PXDF spiking concentrations.
Complementary Non-targeted and Targeted Mass Spectrometry Techniques to Determine Bioaccumulation of Halogenated Contaminants in Freshwater Species
164
LIST OF TABLES AND FIGURES Table S1. APGC-MS/MS MRM method parameters. .................................................... 169 Figure S1. Two-dimensional selected ion chromatogram for mass range 313.9000-313.9700 highlighting two chromatographically separated peaks corresponding to C14H6Cl4
and C16H8BrCl in an L. variegatus extract. Corresponding mass spectra identify elemental compositions through accurate mass, isotopic ratios and fragments. ............. 170
165
Chemicals
This study incorporated 16 PXDD/PXDF standards of which 8-bromo-2,3-
Figure S1. Two-dimensional selected ion chromatogram for mass range 313.9000-313.9700 highlighting two chromatographically separated peaks corresponding to C14H6Cl4
and C16H8BrCl in an L. variegatus extract. Corresponding mass spectra identify elemental compositions through accurate mass, isotopic ratios and fragments.
200 250 300150
100
% In
tens
ity
m/z
313.9445
315.9429
317.9403235.0259
200.0582
C16H8BrCl+313.94924
-Br
-BrCl
313.9208
315.9138
317.9102243.9819
245.9803
200 250 300150
100
% In
tens
ity
m/z
C14H6Cl4+
313.92181
-2Cl
A
B
A
B
171
References 1. Ontario Ministry of the Environment, The determination of total carbonate-carbon
in soil and sediments by coulometry (Laboratory Services Branch Method 3012, 2010).
2. Ontario Ministry of the Environment, The determination of polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans and dioxin-like polychlorinated biphenyls (DLPCBs) in environmental matrices by gas chromatography-high resolution mass spectrometry (GC-HRMS) (Laboratory Services Branch Method 3418, 2010).
3. Ontario Ministry of the Environment, The determination of polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, dioxin-like polychlorinated biphenyls (DLPCBs), polychlorinated naphthalenes (PCNs) and polybrominated diphenyl ethers (PBDEs) in biota samples by automated sample extraction/cleanup and gas chromatography-high resolution mass spectrometry (GC-HRMS) (Laboratory Services Branch Method 3481, 2012).
172
APPENDIX D
SUPPORTING DATA FILES FOR CHAPTERS TWO, THREE, FOUR, FIVE, AND SIX
Data used to generate figures found in this thesis are available in a supplemental file.