Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis, structure, and properties of phosphorus-containing flame retardants Priegert, Andrew Mark 2017

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2017_february_priegert_andrew.pdf [ 21.12MB ]
Metadata
JSON: 24-1.0340871.json
JSON-LD: 24-1.0340871-ld.json
RDF/XML (Pretty): 24-1.0340871-rdf.xml
RDF/JSON: 24-1.0340871-rdf.json
Turtle: 24-1.0340871-turtle.txt
N-Triples: 24-1.0340871-rdf-ntriples.txt
Original Record: 24-1.0340871-source.json
Full Text
24-1.0340871-fulltext.txt
Citation
24-1.0340871.ris

Full Text

Synthesis, Structure, and Properties of Phosphorus-Containing Flame Retardants by  Andrew Mark Priegert  B.Sc. (Hons.), The University of Western Ontario, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  Doctor of Philosophy in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  January 2017  © Andrew Mark Priegert, 2017   ii Abstract  Phosphorus-containing flame retardants were synthesized and a variety of strategies for rendering them non-leachable were investigated. Chapter 1 gives a history of flame retardants, with a focus on the issues associated with their usage. An introduction to the mechanisms of flame retardancy and the effect of flame retardants on the thermal degradation of polymeric materials is also given. An overview of the different methods of incorporating phosphorus-containing flame retardants into polymeric materials is included.  Chapter 2 described the synthesis of a poly(methylene phosphine) and its oxide by the addition polymerization of MesP=CPh2. These polymers are moderately effective non-leachable flame retardants when tested by thermogravimetric analysis, Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00, and Limiting Oxygen Index (LOI). The C-H activated microstructure of two poly(methylene phosphine)s, synthesized by the anionic polymerization of ArP=CPh2 (Ar = 2,4,6-trimethylphenyl, Mes; 2,6-dimethylphenyl, Xyl), is investigated in Chapter 3 using model chemistry and NMR spectroscopic analysis. A mechanism for the anionic polymerization of phosphaalkenes in which a C-H activation occurs is proposed based on kinetic studies, isotopic labelling, and theoretical calculations. The synthesis of the molecular cyclophosphazene-based flame retardant hexakis(2-aminoethyl)aminophosphazene is reported in Chapter 4. This phosphazene is an effective yet leachable flame retardant for paper when tested by thermogravimetric analysis, TAPPI Standard Method T461 cm-00, and LOI.   iii An attempt to covalently link hexakis(2-aminoethyl)aminophosphazene to carboxylates in pulp via carbodiimide coupling is described in Chapter 5. While unsuccessful, carbodiimide coupling can be employed in the synthesis of several simpler phosphazene-amide derivatives.  In Chapter 6, a non-leachable flame retardant treatment for paper using hexakis(2-aminoethyl)aminophosphazene and sodium carboxymethyl cellulose is described. The efficacy of the treatment is evaluated by TAPPI Standard Method T461 cm-00, LOI, and SEM-EDS. The solid precipitate formed in the reaction between hexakis(2-aminoethyl)aminophosphazene and carboxylmethyl cellulose is studied using solid-state CP/MAS 13C NMR and IR spectroscopy, and the interactions between hexakis(2-aminoethyl)aminophosphazene and carboxylmethyl cellulose is modelled using an ammonium-containing phosphazene and carboxylate salt. Chapter 7 summarizes the thesis and gives proposed future work on this project.   iv Preface  This thesis contains work which has been previously published. Chapter 1 contains portions of a review written by myself in collaboration with Benjamin W. Rawe, Dr. Spencer C. Serin, and Prof. Derek P. Gates: Chem. Soc. Rev. 2016, 45, 922. Chapter 2 has been previously published in its entirety: Fire Mater. 2015, 39, 647. Manuscripts based on Chapters 3, 4, 5, and 6 are currently being prepared for submission to peer-reviewed journals. The work described in Chapters 2, 4, 5, and 6 was performed in collaboration with Dr. Thomas Q. Hu at FPInnovations (Vancouver, BC), including flame testing which was performed by myself in the laboratory at FPInnovations. All crystal structures were obtained by Dr. Spencer C. Serin.  In Chapter 2, synthetic work was performed by Dr. Paul W. Siu and myself. Preliminary flame testing was performed by Dr. Paul W. Siu, however the reported results was performed by myself. The pH measurements were taken by Dr. Paul W. Siu.   Chapter 3 was a highly collaborative effort involving several members of the laboratory. In order to give proper context, the entire story is presented here. Compound 3.3a was initially synthesized and characterized by Carl Schiller, with further studies performed by myself. Full characterization of polymer 3.4a was performed by myself and Benjamin W. Rawe. Compounds 3.1b, 3.3b, and 3.4b were synthesized and characterized by Shuai Wang. Kinetic analysis of polymer 3.4b was performed by Shuai Wang, with the data processing and analysis completed by myself. Phosphaalkene d9-3.1a was initially synthesized by Sonja Gerke, and the synthesis, characterization, and kinetic analysis of polymer d9-3.4a was performed by myself. The DFT calculations were performed by Ben Rawe, with guidance from Prof. Pierre Kennepohl. The collection of the 2H spectrum of polymer d9-3.4 was performed by Dr. Paul Xia.   v  In Chapter 4, the synthetic work was entirely performed by myself. Flame testing was performed by myself and Emily Dyck. In Chapter 5, the synthetic work was entirely performed by myself apart from compound 5.8, which was synthesized by Jimmy Tian. In Chapter 6, all synthetic work and flame testing was performed by myself. Dr. Paul Xia assisted with the collection of solid-state 13C NMR spectra. James Drummond at FPInnovations performed the SEM/EDS analyses of coated paper samples.      vi Table of Contents  Abstract .......................................................................................................................................... ii	Preface ........................................................................................................................................... iv	Table of Contents ......................................................................................................................... vi	List of Tables ............................................................................................................................... xii	List of Figures ............................................................................................................................. xiii	List of Schemes .......................................................................................................................... xvii	List of Symbols and Abbreviations .......................................................................................... xix	Acknowledgements .................................................................................................................. xxvi	Dedication ................................................................................................................................ xxvii	Chapter 1: Introduction: Flame Retardants - A Perspective ....................................................1	1.1	 Introduction ..................................................................................................................... 1	1.2	 Historical Background .................................................................................................... 2	1.2.1	 Flammability of Early Plastics ................................................................................ 2	1.2.2	 Halogenated Flame Retardants ............................................................................... 3	1.3	 Mechanisms of Flame Retardancy .................................................................................. 6	1.4	 Phosphorus-Containing Flame Retardants for Plastics ................................................... 9	1.4.1	 Blended Additives ................................................................................................. 10	1.4.2	 Copolymerization of Phosphorus-Containing Monomers .................................... 15	1.4.3	 Post-Polymerization Modification with Phosphorus ............................................ 21	1.5	 Flame Retardants for Cellulosic Materials ................................................................... 22	1.5.1	 Non-Leachable Phosphorus Flame Retardants for Cellulosic Materials .............. 23	  vii 1.6	 Goal of the Project ........................................................................................................ 26	1.7	 Outline of Thesis ........................................................................................................... 26	Chapter 2: Flammability Properties of Paper Coated with Poly(methylene phosphine), an Organophosphorus Polymer .......................................................................................................28	2.1	 Introduction ................................................................................................................... 28	2.2	 Results and Discussion ................................................................................................. 30	2.2.1	 Synthesis of Phosphaalkene Polymers 2.2 and 2.3 ............................................... 30	2.2.2	 Thermogravimetric Analysis ................................................................................ 31	2.2.3	 Flame Testing of Coated Paper Samples .............................................................. 35	2.3	 Summary ....................................................................................................................... 39	2.4	 Experimental Section .................................................................................................... 40	2.4.1	 General Procedures ............................................................................................... 40	2.4.2	 Synthesis of 2.2 ..................................................................................................... 41	2.4.3	 Synthesis of 2.3 ..................................................................................................... 41	2.4.4	 Preparation of Coated Paper Samples ................................................................... 42	2.4.5	 Thermogravimetric Analysis ................................................................................ 42	2.4.6	 Flame Testing by TAPPI T461 cm-00 .................................................................. 43	2.4.7	 Leaching Testing by TAPPI T461 cm-00 ............................................................. 43	2.4.8	 Limiting Oxygen Index ......................................................................................... 44	2.4.9	 Analysis of Char Residue ...................................................................................... 44	Chapter 3: An Addition-Isomerization Mechanism for the Anionic Polymerization of Phosphaalkenes ............................................................................................................................45	3.1	 Introduction ................................................................................................................... 45	  viii 3.2	 Results and Discussion ................................................................................................. 47	3.2.1	 Model Reactivity Studies with a Large Electrophile ............................................ 47	3.2.2	 Synthesis of Polymers 3.4a-b ............................................................................... 52	3.2.3	 Microstructure Analysis by NMR Spectroscopy .................................................. 54	3.2.4	 DFT Calculations of Polymerization Mechanism ................................................. 55	3.2.5	 Kinetic Studies and Isotopic Labelling ................................................................. 57	3.3	 Summary ....................................................................................................................... 61	3.4	 Experimental ................................................................................................................. 61	3.4.1	 X-ray Crystallography .......................................................................................... 61	3.4.2	 General Procedures ............................................................................................... 63	3.4.3	 Computational details ........................................................................................... 65	3.4.4	 Preparation of 3.1b ................................................................................................ 65	3.4.5	 Preparation of 3.3a ................................................................................................ 66	3.4.6	 Preparation of 3.3b ................................................................................................ 67	3.4.7	 Preparation of 3.4a ................................................................................................ 68	3.4.8	 Preparation of 3.4b ................................................................................................ 68	3.4.9	 Preparation of d9-3.4a ........................................................................................... 69	Chapter 4: Synthesis and Flame Retardant Properties of Hexakis(2-aminoethyl)aminophosphazene ..................................................................................................70	4.1	 Introduction ................................................................................................................... 70	4.2	 Results and Discussion ................................................................................................. 74	4.2.1	 Synthesis of Hexakis(2-aminoethyl)aminophosphazene (4.3) ............................. 74	4.2.2	 Thermogravimetric Analysis ................................................................................ 77	  ix 4.2.3	 Flame Testing of Treated Paper Samples ............................................................. 78	4.3	 Conclusion .................................................................................................................... 83	4.4	 Experimental ................................................................................................................. 84	4.4.1	 General Procedures ............................................................................................... 84	4.4.2	 Paper Treatment .................................................................................................... 85	4.4.3	 Preparation of hexakis(2-aminoethyl)aminophosphazene (4.3) ........................... 85	Chapter 5: Carbodiimide-Mediated Amide Couplings of Carboxylic Acids to Phosphazene Derivatives ....................................................................................................................................87	5.1	 Introduction ................................................................................................................... 87	5.2	 Results and Discussion ................................................................................................. 91	5.2.1	 Synthesis of Monofunctional Phosphazene Derivatives ....................................... 91	5.2.2	 X-ray Crystallography .......................................................................................... 93	5.2.3	 Synthesis of Methyl 4-O-Methyl-α-D-glucopyranosiduronic Acid ...................... 95	5.2.4	 EDC Couplings of N3P3(OPh)5NHCH2CH2NH2 (5.7) with Carboxylic Acids .... 96	5.2.5	 Formation of Phosphazene-Functionalized BTMP Handsheets ........................... 98	5.3	 Summary ..................................................................................................................... 100	5.4	 Experimental ............................................................................................................... 101	5.4.1	 X-ray Crystallography ........................................................................................ 101	5.4.2	 General Procedures ............................................................................................. 102	5.4.3	 Preparation of pentaphenoxychlorophosphazene; N3P3(OPh)5Cl (5.6) .............. 103	5.4.4	 Preparation of pentaphenoxy(2-aminoethyl)aminophosphazene; N3P3(OPh)5NHCH2CH2NH2 (5.7) ....................................................................................... 104	  x 5.4.5	 Preparation of pentaphenoxy(2-hydroxyethyl)aminophosphazene N3P3(OPh)5NHCH2CH2OH (5.8) ........................................................................................ 105	5.4.6	 General Carbodiimide Coupling Procedure ........................................................ 106	5.4.7	 Preparation of Pentaphenoxy((N-benzoyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOPh (5.13) ............................................................................... 106	5.4.8	 Preparation of Pentaphenoxy((N-cyclohexylcarbonyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOCy (5.14) .......................... 107	5.4.9	 Preparation of Pentaphenoxy((N-isobutyryl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOCH(CH3)2 (5.15) ................................................................... 107	5.4.10	 Preparation of pentaphenoxy((N-cyclopentylcarbonyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOC5H9 (5.16) ....................... 108	5.4.11	 Preparation of Pentaphenoxy((N-(methyl 4-O-methyl-α-D-glucopyranosiduronyl))2-aminoethyl)aminophosphazene (5.17) ....................................... 108	5.4.12	 Preparation of phosphazene-modified BTMP .................................................... 109	Chapter 6: Carboxymethyl Cellulose and Hexakis(2-aminoethyl)aminophosphazene as a Two-Part Flame Retardant Treatment for Paper ..................................................................110	6.1	 Introduction ................................................................................................................. 110	6.2	 Results and Discussion ............................................................................................... 113	6.2.1	 Flame Testing by TAPPI T461 cm-00 ................................................................ 113	6.2.2	 Flame Testing by Limiting Oxygen Index .......................................................... 115	6.2.3	 Testing of Leaching Under Extreme Conditions ................................................ 117	6.2.4	 Comparison to Related Systems ......................................................................... 118	6.2.5	 Synthesis of 6.1:6.3 Composite .......................................................................... 119	  xi 6.2.6	 Synthesis of Phosphazene-Containing Ammonium-Carboxylate Model (6.5) ... 121	6.2.7	 SEM-EDS Analysis of Treated Paper Samples .................................................. 124	6.3	 Summary ..................................................................................................................... 126	6.4	 Experimental ............................................................................................................... 126	6.4.1	 X-ray crystallography ......................................................................................... 126	6.4.2	 General Procedures ............................................................................................. 128	6.4.3	 Treatment of Paper Samples ............................................................................... 129	6.4.4	 Preparation of 6.1:6.3 Composite ....................................................................... 130	6.4.5	 Leaching of Paper Samples ................................................................................. 130	6.4.6	 Preparation of [N3P3(OPh)5NH(CH2)2NH3][CyCOO] (6.5) ............................... 131	Chapter 7: Summary and Future Work ..................................................................................133	7.1	 Phosphorus-Containing Polymers as Flame Retardants ............................................. 133	7.2	 Phosphazene-Based Non-Leachable Flame Retardants .............................................. 134	References ...................................................................................................................................139	   xii List of Tables  Table 2.1: Summary of weight loss stages for uncoated paper or paper samples treated with either polymers 2.2 or 2.3, or MAP as determined by thermogravimetric analysis. .................... 33	Table 2.2: Summary of results obtained from the flame testing of coated paper samples by TAPPI standard method T461 cm-00. .......................................................................................... 36	Table 3.1: Summary of results obtained from the anionic polymerizations of phosphaalkenes. . 54	Table 3.2: Determination of kp values for the anionic polymerization of phosphaalkenes 3.1a-b and d9-3.1a. ................................................................................................................................... 58	Table 3.3: X-ray crystallographic data of 3.3a and 3.3b .............................................................. 63	Table 4.1: Summary of data obtained from thermogravimetric analysis of paper samples treated with varying quantities of 4.3. ...................................................................................................... 78	Table 4.2: Summary of data obtained from flame testing of paper samples treated with 4.3 by TAPPI Standard Method T461 cm-00. ......................................................................................... 79	Table 4.3: Summary of LOIs obtained for paper samples treated with 4.3. ................................. 81	Table 5.1: Summary of flame testing of BTMP handsheets ....................................................... 100	Table 5.2: X-ray crystallographic data of 5.7 and 5.8 ................................................................ 102	Table 6.1: Summary of results from flame testing by TAPPI Standard Method T461 cm-00 ... 114	Table 6.2: LOI values for treated paper samples. ....................................................................... 116	Table 6.3: LOI values of paper samples treated with 6.1 and 6.3 (15:15 wt%) leached under various conditions for different lengths of time. ......................................................................... 118	Table 6.4: Summary of X-ray collection data for 6.5 ................................................................. 127	   xiii List of Figures  Figure 1.1: The traditional “Fire Triangle”, showing the three necessary conditions for fire at the vertices, and the mechanisms of transfer at the edges. ................................................................... 7	Figure 1.2: The polymer combustion cycle, showing the major steps involved during the burning of a polymeric material ................................................................................................................... 8	Figure 1.3: Strategies of incorporating phosphorus-containing moieties to render a polymeric material flame retardant: a) Physical blending of a phosphorus-containing small molecule or polymer b) Polymerization of a phosphorus-containing monomer c) Modification of an existing polymer by covalently attaching a phosphorus-containing species. ............................................. 10	Figure 1.4: The thermal degradation pathway for cellulose in the absence and in the presence of phosphorus-containing flame retardants. ...................................................................................... 23	Figure 1.5: “Layer-by-layer” assembly of oppositely-charged polyelectrolytes on a cellulosic material. ........................................................................................................................................ 25	Figure 2.1: TGA thermograms of polymers 2.2 and 2.3 under atmospheres of a) air and b) nitrogen. ........................................................................................................................................ 32	Figure 2.2: TGA thermograms of uncoated paper and paper samples treated either with polymers 2.2 or 2.3 or MAP under atmospheres of a) air and b) nitrogen. .................................................. 33	Figure 2.3: Photographs of charred remains of uncoated paper and paper samples coated with either polymers 2.2 or 2.3, or MAP after flame testing by TAPPI T461 cm-00. For coated paper samples, the top group shows samples that had not been leached, while the bottom group shows samples that had been leached prior to flame testing. .................................................................. 37	  xiv Figure 2.4: 31P{1H} NMR spectrum of the aqueous extract of charred residue of paper coated with polymer 2.2 after testing by TAPPI T461 cm-00. ................................................................ 39	Figure 3.1: Common hydrogen transfer processes during radical polymerization (top); proposed C-H activation step during the radical polymerization of phosphaalkene 3.1a. ........................... 46	Figure 3.2: 31P{1H} NMR (162 MHz, toluene, 298 K) spectra of the conversion of 3.2a to 3.3a: a) 3.2a, before addition of Trt-Cl. b) Immediately after addition of Trt-Cl at -78 °C followed by warming to room temperature. c) 72h after addition of Trt-Cl. .................................................... 50	Figure 3.3: Molecular structure of 3.3a (top) and 3.3b (bottom). Thermal ellipsoids are shown at a 50% probability level, all hydrogen atoms except H29 are omitted for clarity. ........................ 51	Figure 3.4 1H-13C HSQC NMR spectrum (600 MHz for 1H, CDCl3, 298 K) of polymer 3.4a. The ordinate shows the 13C APT NMR spectrum and the abscissa shows the 1H NMR spectrum. .... 55	Figure 3.5: Computed intermediates postulated in the mechanisms for the formation of III or IV from I. Energies are electronic energies/free energies in kcal mol-1. ............................................ 56	Figure 3.6: Graph showing linear portion of ln [M]0/[M] vs time (h) plots for the anionic polymerization of 3.1a-b and d9-3.1a with 2% n-BuLi at ~322 K in glyme. The plots for 3.1b and d9-3.1a were used to determine the rate constants given in Table 3.2. .................................. 57	Figure 3.7: a) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer 3.4a. b) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer 3.4b. c) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer d9-3.4a. d) 2H{1H} NMR spectrum (92 MHz, CHCl3, 298 K) of polymer d9-3.4a................................................................................................................................................ 60	Figure 4.1: Common examples of phosphazene motifs. ............................................................... 71	Figure 4.2: 1H and 31P{1H} (inset) NMR spectra of hexakis(2-aminoethyl)aminophosphazene (4.3) ............................................................................................................................................... 75	  xv Figure 4.3: Thermogravimetric analyses of paper samples treated with varying amounts of 4.3. 78	Figure 4.4: Photographs of charred remains of uncoated paper and paper samples coated with phosphazene 4.3 (15 wt%) after flame testing by TAPPI Standard Method T461 cm-00. For coated paper samples, the top sample was not leached prior to flame testing, while the bottom sample was leached. ...................................................................................................................... 80	Figure 4.5: Plot of LOI of treated paper samples vs amount of 4.3 used in treatment. The dashed line marks the concentration of O2 in air (21%). .......................................................................... 82	Figure 5.1: Solid-state molecular structure of 5.7. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms are omitted for clarity. ........................................................... 94	Figure 5.2: Solid-state molecular structure of 5.8. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms are omitted for clarity. ........................................................... 95	Figure 6.1: Cartoon representation of proposed interaction between 6.1, 6.3, and cellulose in which 6.3 acts as a “glue” by binding to both 6.1 and cellulose via non-covalent interactions. 112	Figure 6.2: Photographs of charred remains of uncoated paper and paper samples coated with MAP (15 wt%), 6.1 (15 wt%), and 6.1:6.3 (15:15 wt%) after flame testing by TAPPI Standard Method T461 cm-00. For coated paper samples, the top sample was not leached prior to flame testing, while the bottom sample was leached. ........................................................................... 115	Figure 6.3: 13C{1H} solid-state CP/MAS NMR spectra of the 6.1:6.3 composite (top), 6.1 (middle), and 6.3 (bottom). Asterisks denote spinning sidebands. ............................................. 120	Figure 6.4: IR spectra of 6.3 (blue), 6.1 (red), and the 6.1:6.3 composite (green). The inset is an expanded region of the same spectra. ......................................................................................... 121	Figure 6.5: Molecular structure of (N3P3(OPh)5(NHCH2CH2NH3)]+[O2CCy]– (6.5) (molecule 1 of 2 in the asymmetric unit) as a model for the 6.1:6.3 composite. The filled-in structure   xvi represents one of the two unique ion pairs within the unit cell. Each ammonium moiety forms three H-bonds depicted by one solid structure and two faded structures. .................................. 123	Figure 6.6: Molecular structure of [N3P3(OPh)5(NHCH2CH2NH3)]+[O2CCy]– (6.5), showing approximately three unit cells along the x-axis. Thermal ellipsoids shown at the 50% probability level. All hydrogen atoms and aryl carbon atoms are omitted for clarity. Dashed bonds show H-bonding interactions. Black = carbon, blue = nitrogen, red = phosphorus, cyan = oxygen. ...... 124	Figure 6.7: SEM images at 100x magnification of paper samples without any treatment (top left), unleached samples treated with 6.1 (15 wt%) (top right), unleached samples treated with 6.1:6.3 (15:15 wt%) (bottom left), and leached samples treated with 6.1:6.3 (15:15 wt%) (bottom right). Insets show representative EDS elemental composition analyses for each image. .................... 125	   xvii List of Schemes  Scheme 1.1: Thermal crosslinking of trispirocyclic cyclophosphazene derivative 1.30. ............. 18	Scheme 1.2: Synthesis of a hyperbranched polyphosphonate acrylate (1.37) by UV curing. ...... 20	Scheme 2.1: Isolobal analogy between P=C and C=C bonds, and its relationship to addition polymerization. ............................................................................................................................. 29	Scheme 2.2: Synthesis of poly(methylene phosphine)s 2.2 and 2.3 by the anionic polymerization of phosphaalkene 2.1. ................................................................................................................... 31	Scheme 3.1: Synthesis of 3.3a-b from phosphaalkenes 3.1a-b. ................................................... 48	Scheme 3.2: Synthesis of polymers 3.4a-b from phosphalkenes 3.1a-b. ..................................... 53	Scheme 3.3: Anionic polymerization of the deuterated phosphaalkene d9-3.1a to yield polymer d9-3.4a. .......................................................................................................................................... 59	Scheme 4.1: Reactions of hexachlorophosphazene (4.1) with a bifunctional reagent to yield a variety of products. ....................................................................................................................... 72	Scheme 4.2: Synthesis of hexakis(4-aminophenoxy)phosphazene (4.2) from hexachlorophosphazene (4.1) ....................................................................................................... 73	Scheme 4.3: Synthesis of hexakis(2-aminoethyl)aminophosphazene (4.3) from the reaction of hexachlorophosphazene (4.1) with ethylenediamine. ................................................................... 74	Scheme 4.4 Proposed reaction of a singly-protected ethylenediamine derivative with 4.1 to yield a protected hexa-amino phosphazene derivative (4.5). ................................................................. 76	Scheme 5.1: Mixture of regio- and stereoisomers formed by nucleophilic substitution of hexachlorophosphazene (5.5). ...................................................................................................... 90	  xviii Scheme 5.2: Preparation of monoamino (5.7) and monohydroxy (5.8) phosphazene derivatives from hexachlorophosphazene (5.5) ............................................................................................... 92	Scheme 5.3: Synthesis of methyl 4-O-methyl-α-D-glucopyranosiduronic acid (5.12) from methyl α-D-glucopyranoside (5.9) ............................................................................................................ 96	Scheme 5.4: Synthesis of phosphazene-containing amides (5.13-5.17) using a carbodiimide-mediated coupling ......................................................................................................................... 98	Scheme 6.1: Synthesis of sodium carboxymethyl cellulose (6.3) via etherification .................. 112	Scheme 6.2: Synthesis of an ammonium-carboxylate salt (6.5) between an aminophosphazene (6.4) and cyclohexanecarboxylic acid. ........................................................................................ 122	Scheme 7.1: Synthesis of polyphosphazenes (7.2-7.3) by ring-opening polymerization of hexachlorophosphazene (7.1) followed by nucleophilic substitution. ........................................ 133	Scheme 7.2: Synthesis of a hexa(guanidino)phosphazene (7.7) derivative using S-methylisothiourea hemisulfate salt. ............................................................................................ 136	Scheme 7.3: Synthesis of a tris(amino)phosphoramide (7.14) from phosphorus(V) oxychloride (7.13) ........................................................................................................................................... 138	    xix List of Symbols and Abbreviations  Å     angstrom (1 x 10-10 meters)
  α     alpha (configuational)
  Ac     acetyl anal.     elemental analysis (combustion analysis) ASTM    American Society for Testing and Materials 
  APT     attached proton test (NMR spectroscopy)  Ar     aryl
  ATR     attenuated total resonance Boc     t-butoxycarbonyl br     broad or broadened (NMR spectroscopy)  BTMP    bleached thermomechanical pulp  Bu     butyl
  Bz     benzoyl c     centi (10-2)
  °C     degrees Celsius
  ca.     circa, about cal     calorie
  calcd     calculated
  cat.     catalytic
  cf.     compare COSY    correlation spectroscopy
     xx CP     cross-polarized CTH     controlled temperature and humidity Cy     cyclohexyl δ     chemical shift d     deuterated d     doublet (NMR spectroscopy) D     dextrorotary DCC     N,N’-dicyclohexylcarbodiimide DCM     dichloromethane deg or °   degree ΔE°     change in molar standard enthalpy ΔG‡     change in free energy at transition state DLOI     change in LOI DFT    density functional theory DMF     dimethylformamide DNA    deoxyribonucleic acid dn/dc     refractive index increment DOPO    9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide DS     degree of substitution DSF     dynamic sheet former E     electrophile ε     solvent dielectric constant Ea    activation energy   xxi EDC     N-ethyl-N’-3-(dimethylamino)propylcarbodiimide EDS     energy dispersive X-ray spectroscopy e.g.     for example EI     electron impact en     ethylenediamine equiv.    equivalents ESI    electrospray ionization Et     ethyl etc.    et cetera eV    electron volt g     gram FG     generic functional group FR    flame retardant FT     fourier transform GOF    goodness of fit GPC    gel permeation chromatography {1H}    proton decoupled h     hour HBCD    hexabromocyclododecane HOBt     hydroxybenzotriazole HSQC    heteronuclear single quantum correlation HRMS    high resolution mass spectrometry Hz    Hertz (s-1)   xxii i     iso in     inch in situ    in place or in the reaction in vacuo   under vacuum init.    initiator int     internal (X-ray) IR     infrared J     coupling constant (NMR Spectroscopy) k     kilo (103) K     kelvin kp     rate constant of polymerization KIE     kinetic isotope effect kWh     kilowatt hour L     liter λ     wavelength valence number LOI     limiting oxygen index LRMS    low resolution mass spectrometry M     molarity, mol L-1     mega (106) µ     micro (10-6) m     meta m     multiplet (NMR Spectroscopy)   xxiii meter m/z     mass to charge ratio [M]:[I]    monomer to initator ratio MALS    multi-angle light scattering MAP     monobasic ammonium phosphate MAS     magic angle spinning Me     methyl Mes    2,4,6-trimethylphenyl, mesityl min    minute Mn    number-average molecular weight Mp    peak molecular weight Mw    weight-average molecular weight mol    mole mmol P g-1   millimoles of phosphorus per gram (of paper)  n     normal number of repeat units (polymer) n     nano NMR    nuclear magnetic resonance Nu     nucleophile o     ortho
  ORTEP    oak ridge thermal ellipsoid plot
  %     percent (parts per hundred)
  p    para    xxiv {31P}     phosphorus decoupled (NMR Spectroscopy) PAPTAC    Pulp and Paper Technical Association of Canada PBB    polybrominated biphenyl PBDE    polybrominated diphenyl ether PDI     polydispersity index
  Ph     phenyl PFR    phosphorus-containing flame retardant PMP     poly(methylene phosphine)  POP    persistent organic pollutant ppm     parts per million
  Pr     propyl
   R     generic substituent  residual factor (X-ray) refln     reflection
 (X-ray) Rf     retention factor  RT     room temperature
  s     singlet (NMR spectroscopy) second σ     background (X-ray) coordination number  SEM    scanning electron microscopy θ
     angle t     tertiary   xxv t     time ton triplet (NMR Spectroscopy) T     temperature Tonset    onset temperature (TGA) TAPPI    Technical Association of Pulp and Paper Industry TEMPO   2,2,6,6-tetramethylpiperidyl-1-oxyl TGA     thermogravimetric analysis THF    tetrahydrofuran TLC     thin layer chromatography TMP     thermomechanical pulp  TOF     time of flight Tris    tris(2,3-dibromopropyl)phosphate Trt     triphenylmethyl, trityl UV    ultraviolet ν     vibration V     volume wt%     weight percent x     number of repeat units (copolymer) Xyl     2,6-dimethylphenyl, xylyl y     number of repeat units (copolymer) Z     number of units in a cell (X-ray)     xxvi Acknowledgements  I thank my supervisor Dr. Derek Gates for going out of his way to give me this opportunity.  In the lab, I thank those who initially trained me (Josh, Julien, Ivo, Paul, Eamonn), those who spoke at me (Tom, Spencer), those who spoke with me (Ben, Khatera, Shuai, Han, other Tom), and all others I shared the lab with.  At FPInnovations, I thank Dr. Thomas Hu for his persistent and consistent helpfulness and insight, as well as Michelle Zhao and Dr. Joanne Moszynski for putting up with sudden appearances and long absences.  At UBC, I thank all of the staff of the analytical services and engineering shops for endless patience.  I thank Trion Co. and its endless pursuit of science for science’s sake.  Most importantly I thank my family and friends, especially Morgan who never wanted to date a PhD student.     xxvii Dedication    “Look what a lot of things there are to learn.”  - T.H. White, The Once and Future King   1 Chapter 1: Introduction: Flame Retardants - A Perspective   1.1 Introduction  Perhaps the greatest impact synthetic chemistry has had on the planet during the 20th century is through polymeric materials. Beginning in the late 19th century, the development of methods to cheaply synthesize plastics on enormous scales has led to the permeation of polymeric materials into all facets of our society.1, 2 From the clothes that we wear, to the furniture we sit on, to the containers and packaging in which our food is stored, to even the most mundane of objects such as utensils or shopping bags, in nearly every moment of every day we are in close contact with plastics. While the average consumer may primarily be concerned with the recyclability or durability of these plastics, a much more immediate issue is their flammability. Current safety standards require significant quantities of flame retardants to be incorporated into plastics destined for products that are deemed to pose the greatest risk to consumers, particularly in the case of furniture and electronics. However, due to growing awareness of their long-term health and environmental impact, existing flame retardant additives are no longer suitable for wide use. The challenge for chemists therefore lies not in the synthesis of new flame retardant compounds, but in the development of more effective flame retardant treatments.    2 1.2 Historical Background  1.2.1 Flammability of Early Plastics The inherent flammability of most plastics has been an issue from the earliest days of polymer science. In the 1840’s, Christian Friedrich Schönbein, a chemistry professor at the University of Basel, found that the highly oxidizing mixture of nitric and sulfuric acid reacted with cotton to yield a product that, although appearing to be unchanged, would immediately and vigorously burn upon exposure to flame. The products of cellulose nitration, known then pyroxyline and now as nitrocellulose (1.1), were initially little more than laboratory curiosities. Later, a highly nitrated form would be widely used as an explosive under the name gun cotton.   Soon after, Alexander Parkes found that by mixing only moderately nitrated nitrocellulose with a mixture of oils and organic solvents, a material with some degree of plasticity was formed, which he named parkesine.1, 3, 4 Interestingly, Parkes noted in an early patent from 1855 that the well known flammability of nitrocellulose could be reduced by the addition of compounds “of a less inflammable nature”.5 During the 1860’s John Wesley Hyatt built upon the work of Parkes and others by mixing nitrocellulose with camphor and molding this mixture under heat and pressure to yield a more uniformly plastic material. Thus, Hyatt is usually credited with synthesis of the first commercially useful plastic. This material would be become known as celluloid.1, 3, 4 OOOOOOOOOO n1.1NO2NO2 O2NO2NNO2NO2  3 The high flammability of celluloid combined with the high heat and pressure required to mold the material into desired shapes led to countless industrial accidents in celluloid manufacturing.3, 6 Even after manufacture, celluloid still posed a risk to the public. Early applications of celluloid took advantage of its ability to form a thin film, known as nitrate film. The use of nitrate film to project early motion pictures was responsible for numerous fires as the film could be ignited by sources including the hot bulbs in film projectors, friction from the spinning reel, or an electrical short circuit. Additionally, the dark and crowded environments of cinemas combined with poor or non-existent fire codes in the early 20th century meant that often the majority of deaths occurred not due to the fire itself, but during the ensuing panic. A notorious example from Canada is the Laurier Palace Cinema fire in Montreal in January of 1927, in which 77 children died in a crush after a small fire broke out on the balcony. Not limited to cinemas, perhaps the worst disaster involving nitrate film was the Cleveland Clinic fire of 1929, in which the toxic fumes evolving from burning celluloid X-ray plates in a hospital killed 123 people.7   1.2.2 Halogenated Flame Retardants As newly developed plastics came to replace celluloid, their inherent flammability required the use of flame retardant additives to make them safe for every day use. Halogenated flame retardants such as polybrominated diphenyl ethers (PBDEs, 1.2), hexabromocyclododecane (HBCD, 1.3), or polybrominated biphenyls (PBBs, 1.4) formed the backbone of flame retardant usage in the plastics industry for much of the 20th century. While most often bromine-containing, halogenated flame retardants can also be fluorine- or chlorine-containing. The hydrophobicity of organohalogens makes them ideally suited for use in synthetic   4 polymers (generally petroleum-derived hydrocarbons), as the compatibility of flame retardant additives with the material they are added to is an important concern.8, 9  Beginning in the 1960’s, however, the public perception of flame retardants began to change. In 1965 a Boeing 727 crashed in Salt Lake City after landing short of the runway. Although the impact itself was not serious enough to cause fatalities, the resulting fire killed 43 of the 91 passengers and crew aboard. The plastics used to furnish the interior of the plane were blamed for the intense heat and thick, toxic smoke of the fire. After this and other incidents, the efficacy of flame retardants as well as the utility of the standards used to measure their performance were called into question.10   Specifically, in the case of flame retardants, an infamous case occurred in Michigan in 1973 when between 500 and 1000 pounds of PBBs (1.4) were accidently used in the manufacture of livestock feed.11 This feed was distributed across the state and fed to animals including cattle, pigs, sheep, and chicken. The feed not only poisoned the animals, but also contaminated all of the equipment and machinery used to transport and distribute the feed as well as the products derived from the livestock including meat, eggs, milk, and cheese. Thousands of families are thought to have been exposed to PBBs through ingestion of these contaminated products. Although a solid link between exposure and negative health consequences has yet to be established, studies continue to be conducted generations after the original event.12  OBrn BrmBrBrBrBrBrBr1.2 1.3 1.4Brn Brm  5  Around the same time as the Michigan PBB incident, new flammability standards required manufactures to apply flame retardant additives to children’s sleepwear. Just a few years after this legislation came into effect, the most commonly used of these flame retardants (tris(2,3-dibromopropyl)phosphate (Tris, 1.5) and tris(1,3-dichloro-2-propyl)phosphate (1.6)) as well as their metabolites were found to be mutagenic.13-15 Despite 1.5 being quickly phased out after the publication of these results, 1.6 continues to be widely used in polyurethane foams.16-19  Negative attention towards halogenated flame retardants culminated towards the end of the 20th century when the environmental persistence, bioaccumulation, and toxicity of halogenated flame retardants became apparent. Their presence in the environment stems from how most commonly used brominated flame retardants are small molecules physically blended into plastics. Thus, over time, the flame retardant will leach out of the material and into the surrounding environment. Studies have linked exposure to PBDEs (1.2) to endocrine system disruption, leading to neurological or reproductive deficiencies.20-25 In light of this growing wealth of research and the increased scrutiny stemming from it, various national and international organizations have either banned or limited the production and use of various halogenated flame retardants. The most notable international treaties governing their use include the Rotterdam Convention in 1998 banning of the use of PBBs (1.4) and Tris 1.5POO OOBrBrBrBrBrBrPOO OOClClClClClCl1.6  6 (1.5), while the Stockholm Convention on Persistent Organic Pollutants in 2001 included HBCD (1.3) as well as some PBDEs (1.2) on a list of Persistent Organic Pollutants (POPs). While limiting the usage of halogenated flame retardants is certainly a positive step, the issue of flammability remains. The lack of alternative options can sometimes prevent or limit legislation, leading to the continued use of compounds known to be toxic. For example, the list of POPs resulting from the Stockholm Convention has been amended after its initial ratification to include specific exemptions for the use of HBCD (1.3) in expanded or extruded polystyrene, used as a building material. The development of non-halogenated, non-leachable flame retardants is therefore not only an issue of public safety, but also one of critical commercial importance. However, before a discussion of alternative flame retardants can take place, it is important to understand the different mechanisms by which flame retardants act to better distinguish them not only by their chemical differences, but also their modes of action.  1.3 Mechanisms of Flame Retardancy When describing the concept of fire, the so-called “Fire Triangle” (Figure 1.1) is a simple representation of the relationship between the three conditions necessary for sustained combustion: oxygen, heat, and fuel. All three conditions must be present in balanced proportions for a substance to burn, and the removal of any one vertex of the triangle results in the extinction of the fire: without fuel there is nothing to burn, without oxygen the flame is quenched, and without heat the fire cannot sustain itself.26    7  Figure 1.1: The traditional “Fire Triangle”, showing the three necessary conditions for fire at the vertices, and the mechanisms of transfer at the edges.  While the Fire Triangle is useful in a qualitative sense, a more detailed model is needed to understand the action of flame retardants. Figure 1.2 shows a representation of the polymer combustion cycle. Once ignited, the pyrolysis of a polymer will release both flammable and non-flammable gaseous products. The flammable gases will react with oxygen to radiate heat to the environment and to the surface of the material. The cycle then continues as the radiated heat continues to cause the polymer to undergo pyrolysis. Depending on the exact nature of the polymer, a carbonaceous char can also be formed.26-28 In a manner analogous to the Fire Triangle (Figure 1.1), any interruption in the polymer combustion cycle will stop the fire. Thus, flame retardants are species that interfere with the normal combustion of a material by either chemical or physical mechanisms, resulting in the inhibition or extinction of the flame.29    8  Figure 1.2: The polymer combustion cycle, showing the major steps involved during the burning of a polymeric material  Physical mechanisms for flame retardancy include processes that do not involve a direct reaction between the flame retardant and the materials. This can include cooling, dilution, or the formation of a physical barrier. Aluminum hydroxide [Al(OH)3] and magnesium hydroxide [Mg(OH)2] are examples of flame retardants that perform all three simultaneously. Both salts decompose endothermically, cooling the overall system and thereby preventing pyrolysis. Their decomposition evolves water vapor, diluting the oxygen present in the gaseous phase thereby limiting gaseous reactions. After the decomposition is complete, the final product (either Al2O3 or MgO) acts as a solid, non-flammable layer on top of the substrate, thus blocking further combustion. Intumescence is another example of a physical flame retardant process, in which the flame retardant acts as a blowing agent, swelling the substrate surface thereby insulating the rest of the material.29 Flame retardants that act via chemical mechanisms are divided into two categories, depending if they operate on the material in the gaseous or the condensed phase. Halogenated flame retardants such as 1.2-1.4 act in the gas phase by thermally decomposing to give gaseous   9 free radicals, which in turn react with the free radical species present in flames. By reducing the concentration of gaseous free radicals, the flame is either inhibited or quenched. In the condensed phase, flame retardants can react with a substrate to promote char formation over pyrolysis, thus forming a physical barrier to protect the substrate. This is similar to the physical method of barrier forming, but in this case the char comes from the substrate itself.29 An excellent example of this is how flame retardants for cellulosic materials act, a topic that will be discussed in a later section.    1.4 Phosphorus-Containing Flame Retardants for Plastics  As the need for alternatives to halogenated flame retardants reaches a state of urgency, phosphorus-containing flame retardants (PFRs) are receiving greater attention. PFRs are a large class of flame retardant thought to act primarily through a condensed phase charring mechanism, but in some cases gaseous processes have been proposed.30 While not novel, PFRs are attractive alternatives to halogenated flame retardants as they are generally believed to be less toxic. Additionally, owing to the diversity of oxidation states and bonding environments phosphorus can adopt, phosphorus-containing species can be incorporated into a polymeric material using a variety of strategies (Figure 1.3). Due to the explosive growth within the field of PFRs over the last quarter century, the examples listed in the following sections are intended be representative of techniques rather than an exhaustive list, with emphasis on recent work.31, 32   10  Figure 1.3: Strategies of incorporating phosphorus-containing moieties to render a polymeric material flame retardant: a) Physical blending of a phosphorus-containing small molecule or polymer b) Polymerization of a phosphorus-containing monomer c) Modification of an existing polymer by covalently attaching a phosphorus-containing species.  1.4.1 Blended Additives The easiest strategy from a commercial perspective involves the physical blending of phosphorus-containing flame retardant additives (Figure 1.3a), analogous to the use of halogenated additives 1.2-1.6. Historically, the most commonly used phosphorus-containing flame retardant additives include inorganic salts like ammonium phosphates (e.g. monobasic ammonium phosphate, 1.7), organophosphate esters (e.g. tricresyl phosphate, 1.8), or even elemental red phosphorus.   POO OO1.8POO1.7OHOHNH4  11 Due to the well-known synergistic effect between phosphorus and nitrogen in flame retardants,33-37 recent reports of molecular flame retardant additives attempt to maximize the phosphorus and nitrogen content. Examples include the use of cyclophosphazenes (1.9-1.12),38-40 triazine rings functionalized with 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO) (1.13),41, 42 or melamine-phosphoramidate conjugates (1.14).43   A major drawback of blended flame retardants is their tendency to leach out of a material. This can occur if the molecular additive is not fully compatible with the polymer, causing it to migrate to the surface of the materials where it can either volatilize or be removed by physical action. Additionally, if the additive is water soluble (e.g. 1.7), environmental moisture can also cause leaching. As a consequence of flame retardant leaching, the material will gradually lose its flame retardancy over time. Furthermore, as in the case of halogenated flame retardants 1.2-1.4, the leached additives may be considered environmental pollutants. Despite this, the use NP N PNPRO ORORORRORO1.9R = MeO1.10O1.11NO21.12N NNO OOPPPOOOOOOOHHOOH1.13OPO OPONHONNNHNONNNH2NNH2 H2NNH21.14P OPhP OEtOOEtOOPh  12 molecular non-halogenated PFRs is still preferred because the leached flame retardant additives are presumed to be relatively harmless when compared to halogenated flame retardants. This isn’t always the case, however, particularly with organophosphate esters such as 1.5, 1.6, and 1.8.  In order to fully address the leachability issue, the use of phosphorus-containing polymers as additive flame retardants has been investigated.44 Polyphosphazenes, being the earliest and therefore one of the best studied classes of phosphorus-containing polymers, are particularly well-studied.45, 46 Specifically in the case of additive flame retardants, more recent reports include the use of linear polymer 1.1547 and cross-linked polymer 1.1648 as flame retardants in lithium-ion batteries, and the blending of linear poly(bis(phenoxy)phosphazene) (1.17)49 or cross-linked cyclolinear polyphosphazene 1.1850  with poly(lactic acid).    13   While technically better described as linear or cyclic oligomers, another example of well-studied phosphorus-containing polymeric additives are the polyphosphates. Traditionally, ammonium polyphosphate (1.19) is combined with charring agents (e.g. pentaerythritol51-53 or cyclodextrin54-56) and blowing agents such as melamine. The use of a charring agent is necessary as 1.19 contains no carbon, while the blowing agent creates an intumescent effect.  Polymeric melamine derivatives are also used to act as both charring and blowing agent.57, 58 Melamine P NOOnOOOOP NOOnOOONN NOMeOMeMeONP N PNPOOOOOOOOOOOO1.15 1.161.18P NOPhOPhn1.17  14 polyphosphates (1.20) are also widely used as components of flame retardant mixtures with clay nanoparticles,59, 60 metal phosphinates,61, 62 metal ions,63 or combinations thereof.64-67   Another common structural motif in the synthesis of flame retardant phosphorus-containing polymers is the phosphonate moiety. The popularity of polyphosphonates stems from their ease of synthesis, often formed in a one step condensation polymerization from phenylphosphonic dichloride (PhPOCl2) and a diol. PhPOCl2 is an attractive starting material not only due to its low cost, but its bifunctional nature means linear polymers are easily accessible. In one example, a DOPO-functionalized diol was used to increase the phosphorus content in the resulting polymer (1.21).68 Cyclic oligomers are also accessible, demonstrated by the reaction of PhPOCl2 with bisphenol A (1.22) or phenolphthalein.69 70 Poly(vinylphosphonate)s (1.23) are relatively unexplored for use as flame retardants, despite being a large class of phosphorus-containing polymers. In a recent report, a series of poly(vinylphosphonate)s were synthesized by rare earth metal-mediated group transfer polymerization for use as flame retardant additives in polycarbonates.71  P OOOnN NNNH3H2N NH21.20P OOOnNH41.19  15  1.4.2 Copolymerization of Phosphorus-Containing Monomers  While the use of phosphorus-containing polymers as flame retardant additives can prevent leaching, as with molecular additives poor compatibility between the flame retardant and polymeric material can alter the physical properties of the material in undesirable ways. Additionally, although polymeric flame retardants are inherently non-volatile, phase separation can result in flame retardant leaching. In contrast, direct covalent attachment of phosphorus-containing functionalities to the polymer chain eliminates the problem of leachability and phase separation by making the flame retardant moiety a part of the material itself (Figure 1.3b). The most popular method of covalent attachment involves the incorporation of phosphorus-containing moieties during the synthesis of the polymer by either polymerization of a phosphorus-containing monomer or copolymerization of a phosphorus-containing monomer with a comonomer.44 Due to its high degree of chemical functionality as well as its phosphorus and nitrogen content, the cyclophosphazene motif is a common choice for incorporation into a polymeric structure. Cyclophosphazene-containing epoxides in either monomeric (1.24)72 or polymeric (1.2573, 1.2674) forms have been used to synthesize flame retardant cross-linked resins after curing with amine hardeners. Using the same approach, more complex cross-linked polymeric POPOPhOOnO1.21POPhO O n1.22PRO ORO1.23nR = Et, n-Pr, i-Pr,       p-tolyl  16 structures incorporating cyclophosphazene rings have been synthesized from the hexa(epoxide) cyclophosphazene 1.2775 and the deca(epoxide) bis(cyclophosphazene) bisphenol A derivative 1.28.76   17  PN P NPNPhO OPhOPhRRPhO O OO OR PN P NPNOPhPhOPhO OPhRnR =O OH O1.251.26NP N PNPN NOPhPhOPhO OPhOOOHOHO OOO1.24NP N PNPOOO OOOOO OOOOOOOOOOOOOOOOO ONP N PNPO OOO ONP N PNPO OO OOOOOO OOO OOO1.271.28  18 Other cross-linked cyclophosphazene-containing thermosets include the use of a tris(cyanate) cyclophosphazene derivative (1.29) that, upon heating, forms a flame retardant crosslinked network polymer.77 The spirocyclic cyclophosphazene 1.30 can undergo thermal ring-opening/ring-expanding reactions to form a cross-linked flame retardant polymer (1.31).78 The resulting material can be cross-linked even further through the use of an epoxide-funtionalized bisphenol A derivative.79    Scheme 1.1: Thermal crosslinking of trispirocyclic cyclophosphazene derivative 1.30. NP N PNPOPhOPhO OOPhOOCNNCOOCN1.29NP N PNPHN NHHNNHHNHN1.30130-170 °C1.31NP N PNPNHNHNHNHNHNHNHNHPNPN P NNHNHNH NHNHNH PNPNPNNHNH NHNHNHNHPN P NPN NHNHNH NH  19  Along with their use as flame retardant additives, linear polyphosphazenes have also been widely investigated as flame retardant materials, primarily in the patent literature.45, 46 More recent examples include the blending of polyphosphazene 1.32 with diisocyanate and diol prepolymers to generate flame retardant polyurethane foam,80 and the use of polyphosphazene 1.33 to synthesize flame retardant thermoset epoxy resins.81 Both examples involve the synthesis of poly(dichlorophosphazene) before subsequent substitution.   Polyphosphonates are well represented as flame retardant materials due to their convenient synthesis. For example, 4,4’-dihydrodeoxybenzoin has been used to access polyphosphonates such as polymer 1.3482 as well as copolymers.83, 84 In other cases, the phosphonate moiety is already present in the monomer rather than formed as part of the polymerization step. The bis(isocyanate) 1.35, in which the phosphonate is now part of a side chain, was used to synthesize flame retardant polyurethanes,85 while the bis(acrylate) phosphonate 1.36 was used to synthesize a hyperbranched UV-cured flame retardant polymer (1.37).86  P NOOnCOOHCOOHP NOOnP NOOmOOHOx1.32 1.33  20   Scheme 1.2: Synthesis of a hyperbranched polyphosphonate acrylate (1.37) by UV curing.   As discussed earlier, PhPOCl2 is an attractive starting material for the synthesis of phosphorus-containing polymers as it tends to yield linear polymers. Yet from a material science perspective, branched or cross-linked polymers can be more suitable for particular applications. To this end, phosphorus oxychloride (POCl3) has also been used to synthesize phosphorus-POPhO OOn1.34HN O N OHNPO ONCOOOCN1.35EtO OEtO O P OOOPhOO1.36H2NNNHhvN NNOO O P OOPhOOPOOOPOOOPhOO Ph1.37  21 containing polymers. The organophosphate ester 1.38, a logical extention of phosphonate 1.36, has been used to prepare UV-cured flame retardant coatings,87, 88 while the phosphoramide-containing cross-linked network polymer 1.39 is prepared in a single step from POCl3 and piperazine.89   1.4.3 Post-Polymerization Modification with Phosphorus  Direct covalent attachment of flame retardant phosphorus moieties can also be achieved by post-polymerization modification of existing polymers (Figure 1.3c). Despite its apparent simplicity, however, this strategy is not as widely used. Although the chemical inertness of plastics enables them to be used in a wide variety of applications, this inertness results from the fact that most synthetic polymers lack functional groups. Without any functional groups, chemical modification cannot be conveniently achieved.   An exception to this is poly(vinyl alcohol), which can be modified by simple direct phosphorylation using a chlorophosphine such as diethyl chlorophosphate,90 or cross-linked using dichlorophosphines as bifunctional reagents.91 Other examples of functionalizing poly(vinyl alcohol) include the use of an acyl chloride-containing phosphate ester,92 and the reaction of a DOPO-containing silanol.93  POO OOOOOOOO1.38PONN NNNNP NNOPPNNON NO1.39  22 1.5 Flame Retardants for Cellulosic Materials As part of a growing movement of environmental awareness in which non-renewable resources are being eschewed in favour of “greener” alternatives, demand has shifted away from petroleum-derived materials towards those generated from renewable sources such as plant biomass. Cellulose, derived naturally from cotton or wood pulp, is particularly attractive due to its abundance, renewability, and biodegradability. Despite all of these advantages, the flammability of cellulosic materials still necessitates the use of flame retardants. Although halogenated flame retardants have never been widely used with cellulosics, traditionally used formulations such as boric acid are generally non-durable. Thus, as in the case of petroleum-derived polymeric materials, there is a need for non-leachable flame retardants for cellulosic materials. Phosphorus-containing compounds are extremely effective as flame retardants for cellulose-derived materials due to their ability to influence the mechanism of cellulose degradation during pyrolysis (Figure 1.4).94, 95 During combustion, cellulose will normally depolymerize at about 300 °C to form levoglucosan, which at that temperature will volatilize and combust. Phosphorus compounds generally thermally degrade in air to form phosphorus oxyacids such as phosphoric acid. In the presence of these acids, cellulose will instead crosslink and dehydrate below 300 °C to form a carbonaceous char. This char acts as a physical barrier, shielding the material from further combustion. Therefore, in the specific case of cellulose, phosphorus-based flame retardants primarily act by a condensed phase mechanism (see Section 1.3).   23  Figure 1.4: The thermal degradation pathway for cellulose in the absence and in the presence of phosphorus-containing flame retardants.  1.5.1 Non-Leachable Phosphorus Flame Retardants for Cellulosic Materials Despite the efficacy of PFRs for cellulosic materials, rendering them non-leachable can be challenging since the strategies used for synthetic polymeric materials (Figure 1.3) are not necessarily compatible with cellulose. Physical blending is convenient for melt-processed plastics, but cellulose decomposes before melting. Copolymerization is inherently impossible because cellulose is a naturally occurring polymer. Despite the high degree of chemical functionality in cellulose, covalent attachment is difficult due to the complex macroscopic structure of cellulose fibres. For example, the direct phosphorylation of cellulose is well studied, and yet due to the low efficiency of the process it remains impractical on a large scale.96 Overall, there is surprisingly little research on the use of non-leachable PFRs specifically for cellulose, especially when compared to the vast body of literature on PFRs for synthetic polymeric materials.   24 A recent effort to address the leachability of commonly used flame retardants is the coating of cotton textiles with allyl-functionalized polyphosphazene 1.40, which could then be cross-linked by UV irradiation.97 A unique approach involves the ball-milling of graphite followed by treatment with red phosphorus and exposure to ambient moisture to yield phosphonic acid-containing graphene nanoplatelets (1.41).98 Paper samples treated with these nanoplatelets were found to have increased flame retardancy.    Among recent reports in the area of phosphorus-containing flame retardants for cellulosic materials, the use of the “layer-by-layer” method,99, 100 in which multiple thin layers of oppositely-charged electrolytes are deposited on a substrate, has become popular (Figure 1.5). Despite resembling the simple additive approach, the ionic interactions between the layers as well as possible interactions with the substrate itself makes it somewhat similar to post-polymerization modification. Thus, this intriguing method is something of a hybrid approach.  P NOOn1.40PPCOOHPHOPOHHOOOHHOOOHOHOHOHOO1.41  25  Figure 1.5: “Layer-by-layer” assembly of oppositely-charged polyelectrolytes on a cellulosic material.   A number of systems have been explored in which phosphorus-containing anionic polyelectrolytes including polyphosphates,101-107 poly(vinylphosphonic acid),108 poly(phosphoric acid),109 or phosphonated oligoallylamines110 have been combined with cationic polyelectrolytes (e.g. chitosan, branched polyethylenimine) on cellulosic materials. Bio-sourced phosphorus species such as DNA111 or phytic acid112 have also been investigated for use in flame retardant layer-by-layer assemblies. The durability of layer-by-layer coatings has been improved by the use of a UV-curable species in conjunction with an ammonium polyphosphate/chitosan system.113 Generally the treatment of cellulosic materials by the layer-by-layer method requires the application of as many as 40 individual layers, but recently a one-pot method has been developed.114-116    26 1.6 Goal of the Project  The continued scrutiny of halogenated flame retardants combined with legislation progressively banning their usage has created a dire need for alternatives. Despite phosphorus-based flame retardants having been used for decades, treatment methods to ensure their non-leachability have not advanced beyond a most basic stage. Recent research has focused on the incorporation of phosphorus-containing species into synthetic polymeric materials. However, much less investigation has been done specifically for the case of cellulosic materials. Thus, the goal of this project was to not only synthesize novel phosphorus-based flame retardants, but to develop durable methods of application to render them non-leachable when used to flame-retard cellulosic materials. This thesis therefore lies at the intersection of two large areas of chemical research: the development of non-leachable, non-halogenated flame retardant treatments and the broad field of the utilization of renewable materials.  1.7 Outline of Thesis  Chapter 2 describes the use of phosphorus-containing polymers derived from the polymerization of compounds containing P=C bonds (phosphaalkenes) as additive polymeric flame retardants for paper. These polymers were found to be reasonably effective flame retardants as well as non-leachable with respect to water. Chapter 3 is a mechanistic study of the anionic polymerization of phosphaalkenes by spectroscopic and synthetic methods. The microstructure of these polymers was found to match previously reported results for radical initiation, and a kinetic investigation of a deuterium-labelled monomer gave insight into the mechanism. Chapter 4 describes the synthesis of a phosphazene-derived molecular flame retardant. While susceptible to leaching, this compound was found to be an extremely effective   27 flame retardant for paper. Chapter 5 is a report on an attempt to chemically attach this flame retardant to paper by carbodiimide coupling. Although the synthesis of model compounds was largely successful, only modest flame retardancy was imparted to paper. Chapter 6 details the use of the same phosphazene-derived flame retardant along with carboxymethyl cellulose as a flame retardant coating for paper. The carboxymethyl cellulose was found to bind to both the flame retardant molecule as well as the cellulose in the paper by a combination of non-covalent interactions. These forces are strong enough to yield a robust, non-leachable flame retardant coating. Lastly, Chapter 7 gives an outline of future work possible for this project.      28 Chapter 2: Flammability Properties of Paper Coated with Poly(methylene phosphine), an Organophosphorus Polymer*  2.1 Introduction  Because of their inherent flammability, many commercial products must contain high loadings of flame retardant additives to permit their safe usage. For example, flame retardant additives comprise approximately 20 wt% of cellulose insulation materials made from recycled paper fibers.117, 118 Particularly attractive are phosphorus-containing flame retardants because they tend to offer lower toxicities than their halogen-containing counterparts.31, 32, 119, 120 Molecular and ionic phosphorus-based flame retardants (e.g. ammonium phosphates, phosphonium salts, phosphonates, phosphines, and red phosphorus) have been added to both natural and synthetic polymer-based products for many years. A key disadvantage of using molecular compounds or ionic salts as flame retardants, the majority of which are water soluble, is their tendency to leach out of the product over time and/or when in contact with water or moisture.121, 122  As a consequence of the environmental and toxicological concerns about leachable molecular flame retardants, the development of phosphorus-containing polymers for flame retardant applications is increasingly of interest.44, 123 The use of polymeric phosphorus-based flame retardant additives offers significantly reduced leachability when compared with their molecular counterparts. Despite this advantage, employing phosphorus polymers as flame                                                 *	This chapter has previously been published: A. M. Priegert, P. W. Siu, T. Q. Hu, and D. P. Gates Fire Mater. 2015, 39, 647.   29 retardants remains limited because of the synthetic difficulty of incorporating phosphorus moieties into macromolecules. Perhaps the most well-studied polymeric P-based flame retardants are the polyphosphazenes, [R2P=N]n, which have found commercial use.45, 46  As part of our program to develop new functional polymers containing phosphorus atoms in the main chain, we have been investigating the addition polymerization of the P=C bond in phosphaalkenes (Scheme 2.1).124-129 Our work has shown that P=C bonds, analogous to the C=C bonds of olefins,130, 131 may be polymerized using radical or anionic methods of initiation and that random and block copolymers are accessible containing phosphine functionalities. The resultant polymers, poly(methylene phosphine)s (PMPs), are rare examples of well-defined hydrophobic polymers with a tertiary phosphine moiety in the main chain.132   Scheme 2.1: Isolobal analogy between P=C and C=C bonds, and its relationship to addition polymerization.  Herein, we report our studies of the efficacy of PMPs as non-leachable flame retardant additives for paper made from thermomechanical pulp (TMP). TMP, produced from wood chips in yields of >90%133 and retaining all the major wood components (cellulose, hemicelluloses, and lignin), was chosen as an excellent mimic, in terms of flammability and chemical reactivity, for cellulose insulation materials and for many wood and wood-plastic composite products used in residential construction and the automotive industries.   P C C C C CnP CnPhosphaalkene Olefin Poly(olefin)Poly(methylene phosphine)  30 2.2 Results and Discussion 2.2.1 Synthesis of Phosphaalkene Polymers 2.2 and 2.3 The anionic polymerization of phosphaalkene 2.1 using n-BuLi (1.5 mol%) as initiator afforded polymer 2.2 in high yield (91%) (Scheme 2.2). Compared to the small scales (approximately 1 g of 2.1) used in previous polymerization experiments,125-129 the present work illustrates that the anionic polymerization of 2.1 is amenable to moderate scale-up (16.3 g of monomer) with n-BuLi (1.5 mol%), which is the largest scale reported for the polymerization of 2.1. Analysis of the isolated pale yellow polymer by triple detection GPC-MALS revealed a modest molecular weight (Mn = 27 000 g mol-1) and polydispersity index (PDI = 1.47). The 31P and 1H NMR spectra of 2.2 are consistent with previous data for anionically polymerized monomer. In light of recent data for the radical-initiated polymerization of 2.1,125 the microstructure of 2.2 is shown in Scheme 2.2 with two possible modes of repeating units and connectivity. The phosphine moieties of 2.2 were oxidized to 2.3 by treating a dichloromethane solution of the polymer with an aqueous solution of H2O2. The 31P and 1H NMR spectra of 2.3 were consistent with those of previously characterized samples. Analysis by GPC revealed a slight increase in molecular weight (Mn = 30 000 g mol-1) for 2.3 when compared with 2.2, as anticipated. The slight decrease in polydispersity index (PDI = 1.24) when compared with its parent polymer presumably arises from the fractionation of the polymer sample during the precipitation process.   31  Scheme 2.2: Synthesis of poly(methylene phosphine)s 2.2 and 2.3 by the anionic polymerization of phosphaalkene 2.1.  2.2.2 Thermogravimetric Analysis Thermogravimetric anlysis (TGA) was employed as a means to obtain a preliminary assessment of the flame retardant properties of paper coated with either 2.2 or 2.3. Initially, the polymers were analyzed in the absence of paper under air and nitrogen atmospheres. The TGA thermograms are shown in Figure 2.1(a) and (b), respectively. In both environments, 2.2 exhibited an initial small weight loss (Tonset 180 °C, wt.% = 2.5%) followed by a larger weight loss [Tonset 290 °C, wt.% = 72% (air) or 84% (nitrogen)]. In air, there followed a third, slower stage (Tonset = 686 °C, wt.% = 10%), which was not observed under nitrogen. Interestingly, 2.3 did not exhibit the same initial small weight loss as 2.2 but did show a large weight loss stage [Tonset 335 °C, wt.% = 70% (air), 76% (nitrogen)]. Although the total mass lost during this stage was similar for both 2.2 and 2.3, the phosphine oxide polymer 2.3 appears to have a higher P CMes PhPh n-BuLi (1.5 mol%)P CHPh2CH2 P CMesPhPhxynTHF2.1 2.2P CHPh2CH2 P CMesPhPhxyn2.3OOH2O2  32 stability to weight loss because its Tonset was approximately 45 °C higher than that measured for 2.2. Analogous to 2.2, a third stage of weight loss was observed for 2.3 under air (Tonset = 692 °C, wt.% = 17%) that was not observed under nitrogen.   Figure 2.1: TGA thermograms of polymers 2.2 and 2.3 under atmospheres of a) air and b) nitrogen.  The TGA thermograms of coated paper samples performed in air and nitrogen are shown in Figure 2.2(a) and (b), respectively. Monobasic ammonium phosphate (MAP), a long established phosphorus-based flame retardant for paper,134, 135 was used as a reference, while uncoated paper was used as a control. The major weight loss features are summarized in Table 2.1. All samples showed an initial drying stage where up to ~10% of the initial mass was lost that may be attributed to evaporation of residual moisture/bound water in the paper. After the drying stage, the first stage of weight loss begins just above 300 °C for uncoated paper as well as paper samples coated with 2.2 or 2.3. For uncoated paper, this stage is attributed to the depolymerization of cellulose to primarily afford levoglucosan, which is volatile at that temperature.94, 95, 136-139 For 2.2 and 2.3-coated samples, this depolymerization stage appears to overlap with the onset of weight loss for pure 2.2 and 2.3 described earlier. Noteworthy is the   33 fact that the overall weight lost during this stage is significantly greater for uncoated paper (65%) than for 2.2-coated or 2.3-coated samples (50% and 51%, respectively).   Table 2.1: Summary of weight loss stages for uncoated paper or paper samples treated with either polymers 2.2 or 2.3, or MAP as determined by thermogravimetric analysis.  Drying stage  First stage  Second stage  Weight loss (%)  Tonset (°C)  Weight loss (%)  Tonset (°C)  Weight loss (%) Sample Air N2  Air N2  Air N2  Air N2  Air N2 Uncoated 8 8  315 336  65 67  457 -  27 14 MAP 9 7  251 231  45 37  476 -  42 26 2.2 7 6  305 301  50 57  502 -  39 17 2.3 8 6  315 323  51 62  505 -  39 12   Figure 2.2: TGA thermograms of uncoated paper and paper samples treated either with polymers 2.2 or 2.3 or MAP under atmospheres of a) air and b) nitrogen.  Particularly striking is that the weight loss profiles of paper samples coated with either 2.2, 2.3, or MAP are virtually identical at the completion of the first stage, even though the onset   34 of weight loss for MAP-coated paper (Tonset = 251 °C) is lower than that for 2.2-coated or 2.3-coated paper. (Tonset = 305 and 315 °C, respectively). The flame retardant mechanism for ammonium phosphate-coated cellulosic materials is postulated to involve the thermal-induced formation of phosphoric acid, which promotes cellulose dehydration to afford carbonaceous char, H2O, and gases such as CO2 and CO.94, 135, 137, 140, 141 Therefore, we speculate that by the end of the first stage, a similar char-formation step, the exact mechanism of which is not yet known, may be taking place for the polymer-coated paper samples resulting in flame retardant behavior. It is important to note that both pure 2.2 and 2.3 show no weight loss until 290 and 335 °C, respectively (Figure 2.1). Thus, for the polymer-coated paper, the charring process is delayed until the phosphorus polymer itself starts to degrade, and consequently, both polymer-coated paper samples are stable to higher temperatures than MAP-coated paper. Given that 2.2 and 2.3 both degrade at similar temperatures as cellulose, a gas phase mechanism of flame retardancy is also possible. Such a radical-based mechanism has been proposed for some organophosphorus flame retardants.141, 142 In all cases, the Tonset and percentage by weight lost during the first stage are nearly identical for each sample whether under air or nitrogen, suggesting that the depolymerization stage is largely a thermally driven rather than thermo-oxidatively driven process.  The second stage of weight loss for uncoated paper is attributed to slow degradation of the lignin within the TMP paper.143, 144 For coated paper, the second stage thus becomes a combination of slow oxidative degradation of both lignin and the char formed in stage one. In contrast to the first stage, the second stage is largely thermo-oxidative and, as a result, is slowed considerably when the analysis is performed under nitrogen (the presence of trace oxygen is inevitable because the TGA instrument is not perfectly sealed). The Tonset values for the second   35 stage are significantly higher for air analysis of 2.2-coated and 2.3-coated paper (Tonset = 502 and 505 °C, respectively) when compared with that of MAP-coated or uncoated paper (Tonset = 476 and 457 °C, respectively). Under nitrogen, the TGA thermograms showed no distinct second stage, which reflects the fact that this is primarily a thermo-oxidative process. The char yields at 850 °C were highest for the MAP-coated paper sample (30%) with the 2.2 and 2.3 samples being the same (20%), whereas uncoated paper showed the lowest char yields (11%).   2.2.3  Flame Testing of Coated Paper Samples The flame retardant properties of paper samples coated with 2.2 or 2.3 at a loading of 0.8 mmol P/g paper were also investigated by Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00. The results are summarized in Table 2.2. Figure 2.3 shows the charred remains of samples analyzed before and after leaching treatment. In all cases, the paper samples coated with 2.2, 2.3, or MAP, as well as uncoated paper burned upon contact with the Bunsen flame. Importantly, paper samples coated with MAP, 2.2, and 2.3 were not completely consumed by the flame, and each showed a significant amount of charred residue. The 2.2-coated and 2.3-coated paper samples continued to burn (machine and cross direction: 8 and 9 s, respectively) after the Bunsen flame was removed but extinguished themselves and showed no glowing (Table 2.2). In contrast, uncoated paper samples sustained a flame for slightly longer than coated samples (machine direction: 12 s; cross direction: 11 s) and continued to glow for over 1 minute, stopping only after the paper had been completely consumed. These results suggest that PMP-based polymers have good efficacy as flame retardants. However, they did not perform quite at the level of paper coated with the commercial flame retardant MAP, which ceased to burn even before the Bunsen flame was removed.    36  Table 2.2: Summary of results obtained from the flame testing of coated paper samples by TAPPI standard method T461 cm-00.  TAPPI T461 cm-00    Before leachinga  After leachinga   Sample Flame time (s) Glow time (s)  Flame time (s) Glow time (s) LOI (%) pH of charb Uncoated 12 (11) 102 (63)  - - 19.6 10.5 2.2 8 (8) 0 (0)  9 (7) 0 (0) 23.1 4.9 2.3 9 (9) 0 (0)  17 (9) 0 (0) 25.9 5.2 MAP 0 (0) 0 (0)  13 (12) 133 (125) 34.7 4.0 aValues outside of parentheses are in the machine direction, values inside are for the cross direction. bChar residue was washed with 50 mL of water, and the pH measured.    37  Figure 2.3: Photographs of charred remains of uncoated paper and paper samples coated with either polymers 2.2 or 2.3, or MAP after flame testing by TAPPI T461 cm-00. For coated paper samples, the top group shows samples that had not been leached, while the bottom group shows samples that had been leached prior to flame testing.  After being immersed for 4 h in a bath of distilled water, paper sheets coated with 2.2 show similar flame times (machine direction: 9 s; cross direction: 7 s) compared with the unleached samples (machine and cross directions: 8 s) (Table 2.2). Results for paper sheets coated with 2.3 when cut in the cross direction also show this trend (unleached and leached: 9 s); however, in the case of the 2.3-coated machine direction paper, the flame time increased by a   38 factor of nearly two (unleached: 9 s; leached: 17 s). In all cases, the leached polymer-coated paper sheets had zero glow time and showed charring comparable with when no leaching had been performed (Figure 2.3). In stark contrast to this, leached samples of MAP-coated sheets show flame times (machine direction: 13 s; cross direction: 12 s) and glow times (machine direction: 133 s; cross direction: 125 s) comparable with uncoated samples (flame time, machine direction: 12 s; cross direction: 12 s; glow time, machine direction: 102 s; cross direction: 63 s), indicating that all or nearly all of the MAP had been leached out. Although 2.2 and 2.3 are not quite as effective as MAP as flame retardants, these results show that they are superior when it comes to non-leachability when in contact with water. It is possible that the hydrophobicity of PMPs is responsible for this non-leachability. It is also possible that PMPs may be rendered non-leachable through the chemical attachment of the P moieties to the functional groups of lignin analogous to reactions observed previously for P(CH2OH)3.145, 146   The limiting oxygen index (LOI; ASTM D2863) results for coated paper samples compared with uncoated paper are also shown in Table 2.2. Of course, uncoated paper is flammable even at oxygen levels below the oxygen concentration in air (21%), as is shown by the LOI of 19.6. All coated paper samples showed an increase in LOI compared with that of uncoated paper. Paper samples coated with MAP showed the greatest LOI of 34.7, whereas paper coated with 2.2 or 2.3 had LOIs of 23.1 and 25.9, respectively. These results are consistent with those from TAPPI Standard Method T461 cm-00 in that while 2.2 and 2.3 are good flame retardants, they are not as effective as MAP. The LOIs found for 2.2-coated and 2.3-coated paper are similar to those described previously for molecular and polymeric P-based flame retardants for cellulosic materials (e.g. cotton and rayon) when employed at similar loadings.97, 111, 113, 147, 148    39  Whether a condensed or gas phase mechanism is involved, a key step in the operation of phosphorus-based flame retardants involves the formation of phosphorus oxides and acids (e.g. H3PO4).120, 135, 141 To gain insight into the operation of 2.2 and 2.3 as flame retardants for paper, the char residues after burning were extracted into water, and 31P NMR spectroscopy of the aqueous extract was performed. The spectrum is shown in Figure 2.4, and the observation of a singlet resonance at approximately 0 ppm is consistent with the resonance of phosphoric acid (0 ppm when 85% in H2O). Additionally, the pH of an aqueous extract of the char of uncoated paper was basic, whereas the pH of extracts of the char for all coated paper samples were acidic (Table 2.2). Both of these observations are consistent with the formation of phosphorus oxides and acids during the combustion process.   Figure 2.4: 31P{1H} NMR spectrum of the aqueous extract of charred residue of paper coated with polymer 2.2 after testing by TAPPI T461 cm-00.  2.3 Summary Polymers 2.2 and 2.3 are effective as polymeric phosphorus-based flame retardants for paper made from TMP. The results clearly illustrate that 2.2 and 2.3 can promote the formation of char and inhibit burning. Additionally, 2.2 and 2.3 were demonstrated to be non-leachable   40 with respect to water. LOI measurements showed that 2.2 and 2.3 decrease the flammability of paper. 31P NMR spectroscopic data and pH measurements are consistent with the hypothesis that phosphoric acid may be generated during combustion of polymer-coated paper. TGA showed that coated paper samples have a higher thermal stability than uncoated paper samples.   2.4 Experimental Section 2.4.1 General Procedures  All manipulations of air-sensitive and/or water-sensitive compounds were performed using standard Schlenk or glovebox techniques under nitrogen atmosphere. Monobasic ammonium phosphate (NH4H2PO4; MAP) was purchased from Aldrich (Oakville, Ontario) and used as received. For the preparation of 2.2, THF was dried over sodium/benzophenone ketyl and distilled prior to use. n-BuLi (1.6 M in hexanes) was purchased from Aldrich and titrated prior to use. MesP=CPh2 (2.1) was synthesized in accordance with the literature procedures.149, 150 NMR spectra were recorded on Bruker Avance 300 MHz or 400 MHz spectrometers (Milton, ON) at room temperature. Chemical shifts for 31P NMR spectra are reported relative to H3PO4 as an external standard (85% in H2O, δ = 0).  Molecular weights were determined by gel permeation chromatography with multiple angle light scattering (GPC-MALS) using a Waters liquid chromatograph (Milford, MA) equipped with a Waters 515 high-performance liquid chromatography pump, Waters 717 plus autosampler, Waters Styragel columns (4.6 × 300 mm; HR5E, HR4, and HR2), Waters 2410 differential refractometer, Wyatt tristar miniDAWN (Santa Barbara, CA) (laser light scattering detector-690 nm), and a Wyatt ViscoStar viscometer. A flow rate of 0.5 mL min-1 was used, and samples were dissolved in THF (approximately 2 mg mL-1).    41  2.4.2 Synthesis of 2.2  Polymer 2.2 was prepared using a modification of the original procedure.127 In a glovebox, a hexanes solution of n-BuLi (0.48 mL, 1.6 M, 0.77 mmol) was added to a stirred pale yellow solution of 2.1 (16.3 g, 51.5 mmol) in THF (50 mL), resulting in an immediate color change to deep red. The reaction mixture was stirred at room temperature. After several days, an aliquot was removed from reaction mixture and analyzed by 31P NMR spectroscopy, which showed that the signal assigned to 2.1 (233 ppm) was completely consumed and replaced by a broad signal at -10 ppm. The reaction mixture was removed from the glovebox, and 2.2 was isolated as a pale yellow solid by precipitation into hexanes (500 mL) in air. The polymer was subsequently washed with additional hexanes (2 × 250 mL) and dried in vacuo. Yield = 14.9 g (91%).  GPC-MALS: Mn = 27 000 g mol-1, PDI = 1.47; 31P{1H} NMR (162 MHz, THF): δ = 10 (br s). 1H and 13C{1H} NMR data are identical to those reported previously.129   2.4.3 Synthesis of 2.3  An aqueous solution of H2O2 (10 mL, 30 wt.%) was added slowly to a vigorously stirred solution of PMP (7.0 g, 22 mmol) in dichloromethane (120 mL). The organic layer of the reaction mixture was separated from the aqueous layer and dried with MgSO4. The product was isolated as a pale yellow solid by precipitation into hexanes (500 mL). Yield = 4.5 g (61%).  GPC-MALS: Mn = 30 000 g mol-1, PDI = 1.24; 31P{1H} NMR (121 MHz, THF): δ = 45 (br s). 1H NMR data are available from previous work by Tsang et al.129    42 2.4.4 Preparation of Coated Paper Samples  The paper used for all experiments was laboratory rectangle dynamic sheet former sheets (basis weight = 200 g m-2) made from TMP produced in a pilot plant from black spruce chips with a total energy input of 2468 kWh t-1.133 Sheet samples [70 mm × 210 mm for TAPPI (Technical Association of Pulp and Paper Industry) T461 cm-00, 50 mm × 140 mm for limiting oxygen index (LOI) measurements] were cut (with the machine-direction and cross-direction noted) and conditioned at 23 °C and 50% humidity for 24 h. Each paper sample was weighed, and the mass of flame retardant needed to achieve a loading of 0.8 mmol P g-1 of paper was measured and dissolved in dichloromethane (2.2 or 2.3) or in H2O (for MAP). For 70 mm × 210 mm samples, 10 mL of solvent was used, and for 50 mm × 140 mm samples, 5 mL of solvent was used. The flame retardant solution was then added dropwise using a Pasteur pipet to evenly coat both sides of the paper sample. The samples were conditioned at 23 °C and 50% humidity for 48 h prior to testing. The paper samples were weighed to confirm that the desired loading of flame retardant was achieved. In a typical preparation, 70 mm × 210 mm paper (2.94 g) was coated with a solution of 2.3 (0.68 g, 2.1 mmol P) in dichloromethane (10 mL). After solvent evaporation and conditioning, the actual mass of the coated paper was 3.61 g (expected = 3.62 g).   2.4.5 Thermogravimetric Analysis  Thermogravimetric analyses (TGAs) under air were performed on a Perkin Elmer (Waltham, MA) Pyris 6 Thermogravimetric Analyzer, while TGAs under nitrogen were performed on a Perkin Elmer STA 6000 Simultaneous Thermal Analyzer. Discs 1/8 in. (~3 mm) in diameter were cut with a hole punch from samples treated as outlined in Section 2.4.4. In each   43 experiment, five discs (approximately 12 mg) were heated from 30 to 870 °C at a rate of 10 °C min-1.   2.4.6 Flame Testing by TAPPI T461 cm-00 Testing followed the procedure outlined in TAPPI Standard Method T461 cm-00, using an apparatus fabricated according to the specifications given in the standard method. The paper sample to be tested (prepared as outlined in Section 2.4.4) was suspended such that the lower edge would be 19 mm above the top of a Bunsen burner inside the testing chamber. The burner was lit, and the flame adjusted to a height of 40 mm. With the door of the testing chamber closed, the burner was positioned using an external handle such that the flame was directly in the middle of the sample. The flame was held in contact with the sample for 12 s and then withdrawn. After the burner was withdrawn, the flame time (time the sample continued to sustain a flame) and glow time (time the sample continued to have glowing embers) were measured. The sample was removed from the chamber and photographed. The sample was then tapped to remove brittle or loose char then photographed again.   2.4.7 Leaching Testing by TAPPI T461 cm-00 The leaching of paper samples followed the procedure outlined in the TAPPI Standard Method T461 cm-00. Coated paper samples were placed in a 2000 mL beaker covered with metal mesh. Distilled water was delivered through a glass tube passing through a small hole in the mesh to the bottom of the beaker until the beaker was filled, thereby completely submerging the paper samples. The water flow was maintained for 4 h, at which time the paper samples were removed, blotted dry, and conditioned at 23 °C and 50% humidity for 48 h before testing.    44  2.4.8 Limiting Oxygen Index  Testing was performed using a Govmark OI-1 Oxygen Index Module following American Society for Testing and Materials (ASTM) Standard Test Method D2863. Zero-grade oxygen and nitrogen purchased from Praxair (Mississauga, ON) were used. The top of the sample (prepared as outlined in Section 2.4.4) was lit evenly using a natural gas burner. If the flame burned for longer than 180 s or if the flame traveled more than 80 mm before extinguishing, it was marked as an ‘X’ response. If neither of these criteria were met, it was marked as an ‘O’ response. Using these criteria, identically prepared samples were tested under different oxygen atmospheres until two values separated by 1% oxygen content were found to have opposite responses. Next, the oxygen concentration was increased by 0.2% starting at the value of the opposite pair that gave the ‘O’ response, measuring the response at each increment. The oxygen concentration at which the response changes to an ‘X’ response was taken as the preliminary oxygen index. This value was fine-tuned by measuring the reproducibility of five additional samples at oxygen concentrations close to the preliminary oxygen index.   2.4.9 Analysis of Char Residue  For pH measurements, 50 mL of water was stirred for 30 min with the char remaining from the TAPPI T461 cm-00 test, and the pH recorded. For NMR analysis, a mixture of char and 10 mL of water was boiled for 2 h. Once cool, an aliquot of the aqueous phase was removed for analysis.    45 Chapter 3: An Addition-Isomerization Mechanism for the Anionic Polymerization of Phosphaalkenes  3.1 Introduction  The addition polymerization of C=C bonds, one of the most important methods for the synthesis of polymeric materials, typically involves the regio-regular head-to-tail addition of monomers to a propagating radical, anion or cation. For radical or cationic polymerization, less common side reactions such as hydrogen transfer are important as they lead to termination, chain transfer, or chain branching (Figure 3.1).151 In comparison, regular hydrogen transfers during anionic polymerization are exceedingly rare. A classic example is acrylamide, which polymerizes through the amide functionality, after proton transfer, to afford a polypeptide.152 Vinyl-substituted organosilanes have also been reported to undergo intramolecular proton transfer during anionic polymerization,153 however the regularity of the isomerization appears to be highly dependent on the monomer and reaction conditions.154      46  Figure 3.1: Common hydrogen transfer processes during radical polymerization (top); proposed C-H activation step during the radical polymerization of phosphaalkene 3.1a.  As an entrance into the broad field of inorganic-organic hybrid polymers with unique properties,155 particularly those containing phosphorus,156 we have been interested in the extension of addition polymerization to heteroatom-containing multiple bonds. In particular, the close analogy between P=C and C=C bonds,130 has led us to explore the addition polymerization of phosphaalkenes.125-129, 157-160 Recently, we have reported that polymers resulting from the radical polymerization of phosphaalkenes possess unexpected C-H activated microstructures.125, 159, 160 Unlike other examples of hydrogen transfer polymerizations which isomerize in order to stabilize the propagating species, the radical formed after addition to the P=C bond (A) appears to be forced by steric constraints into a less stable isomer (B) in order for propagation to occur. The repeat unit resulting from simple addition has not been detected, suggesting that the isomerization is highly regular. Evidence for this backbone structure has also been observed in anion-initiated phosphaalkene polymers,158 however given the unprecedented nature of this microstructure within the context of anionic polymerization as well as our interest in the living P CH3CCH2H3CPh PhHP CHPh2H3CCH2H3CHIntramolecularBackbitingH +IntermolecularChain Transfer DisproportionationH +A B  47 anionic polymerization of phosphaalkenes,126, 127 a more thorough investigation was needed. Interestingly, a proton transfer of this type has been proposed to explain observed irregularities during the living anionic polymerization of MesC=CH2, however no structural evidence has been reported.161  Herein we report the regio-regular anionic addition-isomerization polymerization of a P=C bond, for which there is no parallel in olefin polymerization. Specifically, we show through molecular modelling, spectroscopic analysis, and theoretical calculations that the anionic polymerization of phosphaalkenes ArP=CPh2 (Ar = 2,4,6-trimethylphenyl, Mes; 2,6-dimethylphenyl, Xyl) proceeds via an isomerization step before monomer addition, yielding a C-H activated polymeric microstructure.  3.2 Results and Discussion 3.2.1 Model Reactivity Studies with a Large Electrophile We have previously studied the reactions of electrophiles (E+) with the carbanion 3.2a, derived from the addition of MeLi to 3.1a (Scheme 3.1).162 In all cases, the product was MesP(Me)–CPh2E [E = H, Me, P(NEt2)2, SiMe3, SiMe2H]. Despite this, we hypothesized that bulkier electrophiles such as E = Trt (Trt = triphenylmethyl) might favour the C-H activated product, thereby mimicking the microstructure observed in the radical-initiated polymerization. In addition to phosphaalkene 3.1a used in previous studies, we also chose to investigate phosphaalkene 3.1b163, 164 in order to rule out possible intermolecular reactions involving the p-CH3 of the mesityl group.    48  Scheme 3.1: Synthesis of 3.3a-b from phosphaalkenes 3.1a-b.  The addition of MeLi (1 equiv) in diethyl ether to pale yellow solutions of phosphaalkenes 3.1a-b in toluene at -78 °C afforded dark red solutions of carbanions 3.2a-b (Scheme 3.1).162 Upon addition of a solution of Trt-Cl in toluene, the reaction mixture immediately changed colour to yield an orange solution with a white precipitate. The reaction progress could be monitored by 31P NMR spectroscopy. Using the mesityl-substituted species as an example, immediately after the addition of Trt-Cl the resonance corresponding to carbanion 3.2a (-47 ppm) could no longer be observed, with two new resonances (-18 and -38 ppm) in its place (Figure 3.2). Over three days these two intermediates were gradually consumed to form a single product (-26 ppm), assigned to 3.3a. The 1H NMR spectrum of crude 3.3a contained a sharp doublet (4.73 ppm), which was assigned to the P-CHPh2 proton after comparison to the previously synthesized Mes(Me)P-CHPh2.162 Additionally, two new signals (5.11 and 3.72 ppm) were observed and assigned as diastereotopic CH2 protons with the aid of two-dimensional NMR experiments.  P CPhPhTrt-Cl, -78 °C to RTMeMeRP CHPh2MeRTrtMetoluene3.1a (R = CH3)3.1b (R = H)3.3a (R = CH3)3.3b (R = H)XP CMeMeRMeTrtPhPhP CMeMeRMe PhPhMeLi, -78 °C to RTtoluene3.2a (R = CH3)3.2b (R = H)  49 In the analogous reaction with 3.2b, the addition of Trt-Cl yielded one intermediate (-36 ppm), which was gradually consumed to form 3.3b (-24 ppm). This difference in the number of intermediates implies that the intermediate only observed in the reaction with 3.1a (-18 ppm) somehow involves the p-Me group. One possibility is that this intermediate is a species with a p-CH2- carbanion, the product of an intermolecular proton transfer between two equivalents of 3.2a. However, given that neither the second product formed from an intermolecular proton transfer [Mes(Me)P-CHPh2] nor the product of a reaction between p-CH2- carbanion and Trt-Cl are observed, this intermediate is most likely not a p-CH2- carbanion.    50  Figure 3.2: 31P{1H} NMR (162 MHz, toluene, 298 K) spectra of the conversion of 3.2a to 3.3a: a) 3.2a, before addition of Trt-Cl. b) Immediately after addition of Trt-Cl at -78 °C followed by warming to room temperature. c) 72h after addition of Trt-Cl.   51  Figure 3.3: Molecular structure of 3.3a (top) and 3.3b (bottom). Thermal ellipsoids are shown at a 50% probability level, all hydrogen atoms except H29 are omitted for clarity.  Crystals of 3.3a and 3.3b suitable for X-ray diffraction were obtained by the slow evaporation of toluene solutions (Figure 3.3). In both cases, a new C–C bond was formed between a trityl and an o-CH2 substituent. Although the two products crystallized in different space groups (3.3a: P-1; 3.3b: P21/n), the metrical parameters are in close agreement with one another. The new C-C bond (C8-C9) lengths are slightly longer (3.3a: 1.576(6) Å; 3.3b: 1.587(2) Å) than typical carbon-carbon single bonds (~1.54 Å),165 indicative of the strain induced by the   52 bulky trityl group. For comparison, the structure of Mes(Me)P–CPh2Trt was calculated by density functional theory. The CPh2–Trt bond length was found to be 1.707 Å, comparable to the range of calculated bond lengths for the unobtainable compound hexaphenylethane (1.702-1.791 Å).166 The strain induced by the bulky trityl group is also apparent from the C6-C8-C9 bond angle [3.3a: 117.2(3)°; 3.3b: 118.2(1)°], significantly deviated from the typical bond angle for an sp3 carbon (109.5°).   3.2.2 Synthesis of Polymers 3.4a-b Phosphaalkene 3.1a is known to undergo living anionic polymerization,127 however the microstructure of the polymer was initially assigned as resulting from a simple addition mechanism. Given the ability of anion 3.2a to undergo C-H activation with large electrophiles, the microstructure of this polymer, as well as that for the novel polymer resulting from phosphaalkene 3.1b, bears re-examination. Polymers 3.4a-b were therefore synthesized by the addition of n-BuLi (2 mol%, 1.6 M in hexanes) to a THF solution of the corresponding phosphaalkene (Scheme 3.2). The progress of the reaction was monitored by 31P NMR spectroscopy. Upon completion of the reaction, methanol was added to terminate the polymerization. Both polymers were precipitated into hexanes and isolated as pale yellow powders.   53  Scheme 3.2: Synthesis of polymers 3.4a-b from phosphalkenes 3.1a-b.  The molecular weights of polymers 3.4a-b were determined by gel permeation chromatography (GPC). In contrast to the close agreement between the calculated and measured values of Mn for 3.4a, the measured molecular weights of 3.4b were consistently lower than expected (Table 3.1). This is most likely due to trace amounts of oligomeric species leading to tailing in the GPC trace, as the values of Mp match calculated Mn values much better. The PDI values for different samples of 3.4b, while low (1.13-1.20), were generally higher than is typical for anion-initiated phosphaalkene polymerizations.   P CPhPhn-BuLi (2 mol%)P CHPh2CH2 PMeRCPhPhxynMeMeRMe MeR3.1a (R = CH3)3.1b (R = H)THF3.4a (R = CH3)3.4b (R = H)  54  Table 3.1: Summary of results obtained from the anionic polymerizations of phosphaalkenes. Entry Monomer [M]:[I]a Mn calcb (g mol-1) Mn measc (g mol-1) Mp measc (g mol-1) PDI 1 3.1a 50:1 15900 13800 16700 1.10 2 3.1b 25:1 7600 5700 6900 1.18 3 3.1b 30:1 9100 6200 7500 1.17 4 3.1b 35:1 10600 7000 9400 1.20 5 3.1b 43:1 13100 11400 15000 1.13 6 3.1b 50:1 15200 10700 14300 1.17 7 d9-3.1a 50:1 16400 10000 9000 1.08 a[3.1a-b]:[n-BuLi]; bCalculated using the monomer-to-initiator ratio; cAbsolute molecular weights were determined using triple detection GPC. (dn/dc: 3.4a = 0.239; 3.4b = 0.254, determined by direct measurement).   3.2.3 Microstructure Analysis by NMR Spectroscopy The microstructure of polymers 3.4a-b was probed by undertaking detailed solution NMR spectroscopic studies. In the 1H-13C HSQC NMR spectrum of polymers 3.4a-b, in addition to the expected correlations assigned to aromatic C-H groups and CH3 groups, two additional correlations assigned to –CHPh2 and –CH2– moieties were observed (3.4a: Figure 3.4). These assignments were confirmed through a 13C APT NMR experiment. Due to the similarity of the NMR spectra of 3.4a and 3.4b to those of polymers made by radical initiation it is concluded that these polymers are best represented by the microstructure shown in Scheme 3.2 (where x >> y). The microstructure represented by y has never been directly observed, yet it would be presumptuous to assume y = 0 due to the challenge of spectroscopically detecting small structural deviations in polymers.    55  Figure 3.4 1H-13C HSQC NMR spectrum (600 MHz for 1H, CDCl3, 298 K) of polymer 3.4a. The ordinate shows the 13C APT NMR spectrum and the abscissa shows the 1H NMR spectrum.  3.2.4 DFT Calculations of Polymerization Mechanism In light of these results, an intramolecular isomerization mechanism is postulated in which the addition of n-BuLi to 3.1a forms anion I, which then undergoes C-H activation to generate anion II (Figure 3.5). Propagation occurs by the addition of anion II to another molecule of 3.1a, yielding dimer IV. In contrast, a simple addition mechanism would involve the addition of anion I to another molecule of 3.1a to yield dimer III.  DFT was therefore used to gain further understanding of the feasibility of the proposed reaction pathways. The geometries of I-IV and 3.1a were optimized using the B3LYP functional and 6-31G*+ basis set (Figure 3.5). Single point energy calculations were performed using the   56 B97D3 functional and 6-311G**++ basis set, in addition to a solvation correction (in THF, ε = 7.52).   Figure 3.5: Computed intermediates postulated in the mechanisms for the formation of III or IV from I. Energies are electronic energies/free energies in kcal mol-1.  aElectronic energies are only provided due to computational expense of frequency calculation on a large system.  As calculated by DFT, dimer IV was found to be significantly lower in energy than dimer III (ΔE° = -15.8 kcal mol-1 vs +4.4 kcal mol-1). The high energy of dimer III is most likely due to steric strain induced by the hindered Ph2C moiety, which is reflected experimentally by the inability of anions of type I to react with relatively bulky electrophiles such as Trt-Cl. Thus, despite anion II being higher energy than anion I by +12.1 kcal mol-1, overall the formation of IV is exergonic by 15.8 kcal mol-1, supporting the experimentally determined microstructure.   57 3.2.5 Kinetic Studies and Isotopic Labelling The kinetics of the anionic polymerization of 3.1b were studied by monitoring the progress of the reaction by 31P NMR spectroscopy in a manner identical to previously reported experiments for 3.1a (Figure 3.6).157 The polymerization rate constants for phosphaalkene 3.1b were comparable to 3.1a (Entries 1-6, Table 3.2). The activation energy (Ea) for the anionic polymerization of 3.1b was estimated using this data, and was also found to reasonably match that of 3.1a (3.1a: 14.0 ± 0.9 kcal mol-1;157 3.1b: 15.6 ± 2.8 kcal mol-1). These activation barriers are also supported computationally, as a transition state for the C-H activation (TSI-II) was calculated using DFT (ΔG‡ = 17.3 kcal mol-1).  Figure 3.6: Graph showing linear portion of ln [M]0/[M] vs time (h) plots for the anionic polymerization of 3.1a-b and d9-3.1a with 2% n-BuLi at ~322 K in glyme. The plots for 3.1b and d9-3.1a were used to determine the rate constants given in Table 3.2.     58  Table 3.2: Determination of kp values for the anionic polymerization of phosphaalkenes 3.1a-b and d9-3.1a. Entry Ma T (K) kpb (L mol-1 h-1) M T (K) kpc (L mol-1 h-1) 1 3.1b 298.6 22.2 ± 2.6 3.1a 296.3 21.0 ± 2.5 2 3.1b 301.0 40.1 ± 4.7 3.1a 301.8 32.7 ± 3.9 3 3.1b 306.2 69.0 ± 8.2 3.1a 307.4 41.8 ± 4.9 4 3.1b 311.3 127 ± 15 3.1a 313.0 70.7 ± 8.9 5 3.1b 316.6 145 ± 18 3.1a 318.6 125 ± 15 6 3.1b 321.9 152 ± 23 3.1a 324.2 150 ± 17 7 d9-3.1a 321.9 6.6 ± 1.2      a[M]:[I] = 50:1 in all cases. bRate constant of propagation. cReproduced from literature.157  Given the high activation energy observed for the C-H activation, this step is postulated to be rate-limiting. Therefore, a significant kinetic isotope effect (KIE) would be expected for the anionic polymerization of d9-3.1a (Scheme 3.3). To this end, phosphaalkene d9-3.1a was synthesized from d9-mesitylene167 and its polymerization attempted. In contrast to 3.1a, the reaction of a THF solution of d9-3.1a with n-BuLi (2 mol%, 1.6 M in hexanes) at room temperature did not yield any polymeric material. Despite this, polymer d9-3.4a could eventually be synthesized by conducting the polymerization in glyme at 50 °C, and stirring for 24 hours (Scheme 3.3). Although the isolated yield was low (10%), the conversion as measured by 31P NMR was approximately 90%. The molecular weight was determined by GPC, and was found to closely match the calculated value with a low PDI (Entry 7, Table 3.1).    59  Scheme 3.3: Anionic polymerization of the deuterated phosphaalkene d9-3.1a to yield polymer d9-3.4a.  While the 31P and 13C NMR spectra of d9-3.4a resembled those for 3.4a, the 1H NMR spectrum (Figure 3.7c) differed from those for polymers 3.4a-b (Figure 3.7a and b, respectively) due to the lack of aliphatic protons. Instead, only resonances corresponding to the aromatic protons are observed. These aliphatic signals are instead seen in the 2H{1H} NMR spectrum (Figure 3.7d), though with the poorer resolution typical for 2H NMR spectroscopy.  P Cd9-Mes PhPh n-BuLi (2 mol%)P CDPh2CD2 PD3CCD3Cd9-MesPhPhxynd9-3.1a d9-3.4aglyme50 °C, 24 h  60  Figure 3.7: a) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer 3.4a. b) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer 3.4b. c) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of polymer d9-3.4a. d) 2H{1H} NMR spectrum (92 MHz, CHCl3, 298 K) of polymer d9-3.4a.  Kinetic experiments were also performed for d9-3.1a (Figure 3.6; Entry 7, Table 3.2), leading to an apparent kinetic isotope effect of 23. Although this is clearly a significant KIE, it is important to note that the sensitivity of the anion of the growing polymer chain at 50 °C combined with the long reaction time may have led to partial quenching during the   61 polymerization. This would lead to a lower concentration of growing chain in solution, lowering the overall rate. Thus, the apparent kp is likely an under-estimation. Additionally, quantum tunnelling is known to sometimes lead to higher than expected KIE values, however, it is not clear if this is the case for d9-3.1a. Nevertheless, it is apparent that the kinetic isotope effect is significant, indicating that the C-H activation is the rate-limiting step in the anionic polymerization of 3.1a.   3.3 Summary Carbanions generated from phosphaalkenes 3.1a and 3.1b were reacted with Trt-Cl to yield the C-H activated products 3.3a and 3.3b, respectively. Phosphaalkenes 3.1a and 3.1b were also polymerized by anionic initiation, and the isolated polymers (3.4a-b) were shown to have microstructures resulting from an isomerization mechanism. Theoretical calculations using DFT supported this mechanism over a simple addition mechanism, with the C-H activation as the rate-limiting step. Isotopic labelling studies in which the deuterated phosphaalkene d9-3.1a was anionically polymerized to form polymer d9-3.4a revealed a kinetic isotope effect of 23, also suggesting that C-H activation is the rate-limiting step.   3.4 Experimental 3.4.1 X-ray Crystallography The single crystals of 3.3a-b were immersed in oil and mounted on a glass fiber. Data was collected on a Bruker X8 APEX II diffractometer with graphite-monochromated Mo Kα radiation. The structure was solved by direct methods and subsequent Fourier difference techniques. All non-hydrogen atoms were refined anisotropically with hydrogen atoms being   62 included in calculated positions but not refined. The data set was corrected for absorption effects (SADABS), Lorentz, and polarization effects. All calculations were performed using SHELXL-2014 crystallographic software package from Bruker AXS. Additional crystal data and details of data collection and structure refinement are listed in Table 3.3.      63  Table 3.3: X-ray crystallographic data of 3.3a and 3.3b Crystal 3.3a 3.3b Formula C42H39P C41H37P Formula Weight 574.73 560.71 Crystal System Triclinic Monoclinic Space Group P -1 P 21/n Colour Colourless Colourless a (Å) 9.879(1) 12.623(7) b (Å) 12.342(1) 15.37(1) c (Å) 15.455(2) 16.56(1) α (°) 104.703(5) 90 β (°) 101.120(5) 102.68(1) γ (°) 112.177(4) 90 V (Å3) 1597.8(3) 3133(3) Z 2 4 T (K) 90(2) 90(2) µ (Mo Κα) (mm-1) 0.115 0.115 Crystal Size (mm) 0.21 x 0.15 x 0.19 0.42 x 0.26 x 0.15 Dcalcd. (g cm-3) 1.192 1.189 2Θ (max) (°) 60.2 59.5 No. of Reflections 24398 36584 No. of Unique Data 24848 8385 Rint 0.0766 0.0381 Reflections/parameters ratio 18.1 23.4 R1, wR2[I > 2σ(I)]a 0.1077 0.0459 R1, wR2 (all data)b 0.2368 0.1269 GOF 1.376 1.025 aR1 = Σ ||Fo| - |Fc|| / Σ|Fo|. bwR2 = [Σ (w(Fo2 – Fc2)2)/Σ w(Fo2)2]1/2  3.4.2 General Procedures All manipulations of air- and/or water-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk or glovebox techniques. 1H, 31P and 13C NMR   64 spectra were recorded on Bruker Avance 300 MHz, 400 MHz or 600 MHz spectrometers at room temperature unless otherwise specified. Chemical shifts are reported relative to: residual CHCl3 (δ = 7.26 for 1H), C6D5H (δ = 7.16 for 1H); 85% H3PO4 as an external standard (δ = 0.0 for 31P); CDCl3 (δ = 77.0 for 13C). Mass Spectra were acquired using Kratos MS 50 instrument in EI mode (70 eV). Elemental analyses were performed in the University of British Columbia Chemistry Microanalysis Facility. Polymer molecular weights were determined by triple detection gel permeation chromatography (GPC–MALS) using an Agilent liquid chromatograph equipped with an Agilent 1200 series isocratic pump, Agilent 1200 series standard autosampler, Phenomenex Phenogel 5µm narrow bore columns (4.6 x 300 mm) 104 Å (5000-500000), 500 Å (1000-15000), and 103 Å (1000-75000), Wyatt Optilab T-rEx (refractive index detector, λ = 658 nm, 40 oC), Wyatt miniDAWN (laser light scattering detector, λ = 690 nm) and a Wyatt ViscoStar viscometer. A flow rate of 0.5 mL min-1 was used and samples were dissolved in THF (ca. 1.5 mg ml-1). The dn/dc of 3.4b was determined to be 0.2538 using a Wyatt Optilab T-rEx refractive index detector (λ = 658 nm).  Hexanes and toluene were deoxygenated with nitrogen and dried by passing the solvents through a column containing activated basic alumina. THF and glyme were freshly distilled from sodium/benzophenone ketyl before use. Methanol was degassed prior to use. Benzophenone (Aldrich) was sublimed prior to use. MeLi (1.6 M in diethyl ether), n-BuLi (1.6 M in hexanes), trityl chloride, and chlorotrimethylsilane were purchased from Aldrich and used as received.  3.1a149, 150 and XylP(SiMe3)2168 were prepared according to adapted literature procedures. d9-3.1a was prepared in the same manner, however, starting from d9-mesitylene which was prepared according to a literature procedure.167 Kinetics studies performed following a literature procedure.157   65  3.4.3 Computational details Density functional theory calculations were performed in Gaussian 09 (Revision D.01).169 Initial geometry optimizations were performed using the 6-31G*+ basis set and the B3LYP functional (denoted as BS1).170, 171 Frequency calculations were used to confirm the identity of minima (no negative eigenvalues) and transition states (1 negative eigenvalue) using BS1. Single point energy calculations were used to calculate enthalpy for the geometries using the 6-311G**++ basis set on all atoms and the B97D3 functional. Additionally, solvation corrections were implemented using the Polarizable Continuum Model172 using tetrahydrofuran (ε=7.43) as solvent. Frequency calculations were used to use free energies (G°) where possible (systems less than 70 atoms) and are noted in text. Single point energies calculations were rerun with the wB97XD and CAM-B3LYP173 functionals but showed a marginally higher transition state energy for TSI-II. Optimizing the ground state geometries using the 6-31G*+ basis set had little to no change in substrate conformation.   3.4.4 Preparation of 3.1b  To a stirred solution of XylP(SiMe3)2 (6.35 g, 22.5 mmol) in THF (30 mL), MeLi (14.8 mL, 23.6 mmol, 1.6 M in Et2O) was added. The resulting yellow solution was heated to 55 °C for 1.5 h. Upon cooling to -78 °C, a solution of benzophenone (4.10 g, 22.5 mmol) in THF (20 mL) was added, giving a dark red solution. After stirring for 30 min the reaction mixture was warmed to room temperature. After stirring for an additional 30 min, the reaction was again cooled to -78 °C and chlorotrimethylsilane (3.7 mL, 29 mmol) was added, followed by stirring for 30 min. Upon warming to room temperature and stirring for an additional 30 min, the color   66 of the solution turned from dark red to yellow. All volatiles were removed in vacuo, and hexanes (3 × 30 mL) was added to the resulting oil followed by filtration and solvent removal in vacuo. The crude product was distilled under vacuum and subsequently recrystallized from hexanes to yield a crystalline yellow solid. Yield: 2.86 g (42%).  31P{1H} NMR (162 MHz, CDCl3): δ = 231 (s); 1H NMR (400 MHz, CDCl3): δ = 7.62-6.91 (m, 13H, Ar-H), 2.36 (s, 6H, Ar-CH3); 13C{1H} NMR (151 MHz, CDCl3): δ = 193.8 (d, 1JPC = 43 Hz, P=C), 144.8 (d, 2JPC = 24 Hz), 143.2 (d, 2JPC = 14 Hz), 140.6 (d, 2JPC = 7 Hz), 140.0 (d, 1JPC = 44 Hz), 129.1 (s), 129.0 (s), 128.7 (m), 128.6 (s), 128.4 (s), 127.9 (s), 127.7 (s), 127.6 (s), 127.5 (s), 127.4 (s), 22.5 (d, 3JPC = 9 Hz). Anal. Calcd. for C21H19P: C, 83.44; H, 6.29. Found: C, 83.65; H, 6.36.   3.4.5 Preparation of 3.3a To a stirred solution of 3.1a (0.790 mmol) in toluene (3 mL) at -78°C, MeLi (1.6 M in diethyl ether, 0.50 mL, 0.800 mmol) was added. Upon warming, the solution became a deep red colour. 31P NMR analysis of this solution showed complete consumption of the phosphaalkene starting material indicated by the absence of a resonance corresponding to the starting material (234 ppm). Instead a broad singlet (-46 ppm) could be observed, assigned to 3.2a. The reaction mixture was once again cooled to -78°C, at which time a solution of Trt-Cl (0.176 g, 0.806 mmol) in toluene (1 mL) was added. Upon warming the reaction became cloudy and orange. 31P NMR analysis of this solution showed none of the resonance assigned to 3.2a but instead two broad singlet resonances (-18 and -38 ppm) in approximately a 1:3 ratio. After three days of stirring at room temperature, 31P NMR analysis showed almost complete conversion to a new broad singlet resonance (-26 ppm). The reaction mixture was filtered through glass microfiber   67 paper and all volatiles removed under vacuum to give the crude product as an orange-red solid. Crude yield: 0.320 g (87% when taking into consideration amounts removed for NMR analysis).  In a separate reaction done on a larger scale (5 mmol of 3.1a), immediately after the addition of Trt-Cl and warming the reaction was filtered and an aliquot removed. From this aliquot, after slow evaporation of the toluene, crystals suitable for X-ray analysis were isolated. These crystals were also used for characterization. 31P{1H} NMR (121.5 MHz, CD2Cl2): δ = -26 (s); 1H NMR (400 MHz, CD2Cl2): δ = 7.51 (d, 2H, J = 7 Hz, Ar-H), 7.34-7.11 (m, 23H, Ar-H), 6.76 (s, 1H, Ar-H), 6.43 (m, 1H, Ar-H), 5.10 (dd, 1H, 2JHH = 15 Hz, 4JPH = 10 Hz, -CHH-), 4.73 (d, 1H, 2JPH = 5 Hz, P-CHPh2), 3.72 (d, 1H, 2JHH = 15 Hz, -CHH-), 2.56 (s, 3H, o-CH3), 1.90 (s, 3H, p-CH3), 0.89 (d, 3H, 2JPH = 6 Hz, P-CH3); 13C{1H} NMR (101 MHz, CD2Cl2): δ = 147.7 (s), 146.7(s), 143.7(s), 143.6 (d, J = 13 Hz), 143.0 (d, J = 12 Hz), 138.1 (s), 132.7 (d, J = 24 Hz), 130.9 (s), 130.7(s), 130.0 (s), 129.7 (d, J = 5 Hz), 129.5 (s), 129.4 (s), 129.3 (s), 129.2 (s), 129.2 (s), 129.0 (s), 128.9 (s), 128.9 (s), 128.6 (s), 128.0 (s), 127.0 (d, J = 2 Hz), 126.7 (d, J = 2 Hz), 126.4 (s), 59.0 (s, CPh3), 52.2 (d, 3JPH = 16 Hz, -CH2-), 44.3 (d, 1JPC = 34 Hz, P-CHPh2), 23.5 (d, 3JPC = 5 Hz, o-CH3), 21.1 (s, p-CH3), 9.7 (d, 1JPC = 10 Hz, P-CH3); LRMS (ESI-TOF) m/z: 575.5 [M+H]+; Anal. Calcd for C42H39P: C, 87.77; H, 6.84 Found: C, 87.52; H, 6.80.  3.4.6 Preparation of 3.3b Procedure followed the previous procedure but using 3.1b. Yield: 2.21 g (79%).  31P{1H} NMR (162 MHz, CDCl3): δ = -24 (s); 1H NMR (400 MHz, CDCl3): δ = 7.49 (d, J = 8 Hz, 2H, Ar-H), 7.34-7.05 (m, 23H, Ar-H), 6.90 (d, J = 8 Hz, 1H, Ar-H), 6.75 (t, J = 8 Hz, 1H, Ar-H), 6.61 (m, 1H, Ar-H), 5.01 (dd, 2JHH = 15 Hz, 4JPH = 10 Hz, 1H, -CHH-), 4.67 (d, 2JPH   68 = 3 Hz, 1H, P-CHPh2), 3.66 (d, 2JHH = 15 Hz, 1H, -CHH-), 2.56 (s, 3H, Ar-CH3), 0.90 (d, 2JPH = 6 Hz, 3H, P-CH3); 13C{1H} NMR (101 MHz, CDCl3): δ = 147.1 (s), 142.1 (d, J = 26 Hz), 143.3 (s), 142.8 (s), 142.7 (s), 142.1 (d, J = 12 Hz), 136.1 (s), 135.9 (s), 130.7 (s), 130.3 (s), 129.6 (s) 129.5 (s), 129.2 (s), 129.1 (s), 128.7 (s), 128.5 (s), 128.5 (s), 128.3 (d, J = 6 Hz), 128.1 (s), 127.5 (s), 126.6 (s), 126.5 (s), 126.2 (s), 125.9 (s), 58.6 (s, -CPh3), 51.8 (d, 1JPC = 34 Hz, P-CHPh2), 44.0 (d, 3JPC = 17 Hz, -CH2-), 23.4 (d, 3JPC = 5 Hz, Ar-CH3), 9.7 (d, 1JPC = 21 Hz, P-CH3); LRMS (EI) m/z: 560 [M+]; Anal. Calcd. for C41H37P: C, 87.86; H, 6.61 Found: C, 87.94; H, 6.75.  3.4.7 Preparation of 3.4a  Procedure followed an adapted literature procedure.127 To a stirred solution of 3.1a (5 g, 15.8 mmol) in THF (20 mL) at room temperature, n-BuLi (0.21 mL, 0.316 mmol, 1.5 M in hexanes) was added. The conversion of 3.1a (234 ppm) to 3.4a (-10 ppm) was monitored by 31P NMR. Upon complete conversion, the reaction mixture was quenched with 3-4 drops of degassed methanol, and added dropwise to hexanes (3 x 100 mL) to precipitate the product. Yield: 3.85 g (77%).  GPC-LLS (THF): Mn = 17 000 g mol-1, PDI = 1.05. 31P{1H} NMR (162 MHz, CDCl3): δ = -10 (br); 1H NMR (600 MHz, CDCl3) δ = 7.2 (br, Ar-H), 4.8 (br, P-CHPh2), 3.6 (br, -CH2-), 2.3 (br, Ar-CH3); 13C{1H} NMR (151 MHz, CDCl3) δ = 147 (br), 143 (br), 138 (br), 128 (br), 126 (br), 52 (br, P-CHPh2), 32 (br, -CH2-), 23 (br, Ar-CH3), 21 (br, Ar-CH3).  3.4.8 Preparation of 3.4b To a stirred solution of 3.1b (0.488 g, 1.62 mmol) in THF (1.5 mL), n-BuLi (45.0 µL, 64.6 µmol, 1.44 M in hexanes) was added. After 22 h the signal corresponding to the monomer   69 (δ = 231) was consumed and replaced by a broad signal (δ = -11). The reaction was quenched by addition of five drops of degassed methanol, turning the dark red solution reaction mixture yellow. The product was purified by repeated precipitation with dry hexanes (3 x 50 mL) and was isolated as a pale yellow powder. Yield: 0.351 g (72%).  GPC-LLS (THF): Mn = 6700 g mol-1, PDI = 1.14. 31P{1H} NMR (162 MHz, CDCl3): δ = -11 (br); 1H NMR (600 MHz, CDCl3): δ = 7.2 (br, Ar-H), 5.0 (br, P-CHPh2), 3.5 (br, -CH2-), 2.7 (br, Ar-CH3); 13C{1H} NMR (600 MHz, CDCl3): δ = 147 (br), 142 (br), 129 (br), 128 (br), 126 (br), 52 (br, P-CHPh2), 32 (br, -CH2-), 25 (br, Ar-CH3), 23 (br, Ar-CH3).  3.4.9 Preparation of d9-3.4a  To a stirred solution of d9-3.1a (0.25 g, 0.790 mmol) in glyme (2 mL) at 50 °C, n-BuLi (10 µL, 0.0158 mmol, 1.6 M in hexanes) was added. The reaction was monitored over the course of seven days by 31P NMR, at which point the reaction appeared to reach maximum conversion (~90%). The reaction was quenched with degassed MeOH (5 drops), and precipitated with hexanes (30 mL). After filtration, the crude product was dissolved in DCM (5 mL), and the precipitation/filtration procedure repeated twice more to yield a pale yellow powder. Yield: 25 mg (10%).  GPC-LLS (THF): Mn = 11 000 g mol-1, PDI = 1.09. 31P{1H} NMR (162 MHz, CDCl3): δ = -11 (br); 1H NMR (600 MHz, CDCl3): δ = 7.1 (br, Ar-H); 2H{1H} NMR (92 MHz, CHCl3): δ = 4.4 (br, P-CDPh2), 2.1 (br, Ar-CD3); 13C{1H} NMR (151 MHz, CDCl3): δ = 147 (br), 143 (br), 138 (br), 128 (br), 126 (br), 52 (br, P-CDPh2), 30 (br, -CD2-), 22 (br, Ar-CD3), 20 (br, Ar-CD3).   70 Chapter 4: Synthesis and Flame Retardant Properties of Hexakis(2-aminoethyl)aminophosphazene  4.1 Introduction  The toxicity, persistence, and bioaccumulation of halogenated flame retardant additives have led to their use being severely restricted.20-25 Due to this increased scrutiny, phosphorus-based flame retardants have emerged as an alternative.31, 32, 44, 119, 120, 123 Chapter 2 described the use of phosphorus-containing polymers as non-leachable flame retardants for paper.174 While these polymers were found to be somewhat effective, their lengthy synthesis made large quantities needed for flame retardant testing inaccessible. Broadly, despite great academic interest,132 the bulk-scale synthesis of phosphorus-containing macromolecules still poses a challenge. With polymeric additives no longer an option, our attention therefore turned to the covalent attachment of molecular phosphorus-containing species as an alternative strategy used to incorporate phosphorus moieties into polymeric materials.  As seen in many examples in Chapter 1, phosphazenes (Figure 4.1) are a common motif in flame retardant chemistry as phosphorus and nitrogen are known to act synergistically when combined in flame retardant formulations.33-37 Phosphazenes (also known as phosphoranimines, phosphonitriles, or iminophosphoranes) are compounds in which a tetracoordinate, pentavalent (σ4λ5) phosphorus atom is doubly bound to a dicoordinate, trivalent (σ2λ3) nitrogen atom (Figure 4.1).46 Phosphazenes can exist in a variety of structural motifs including monomers (A), cyclic trimers (B) or tetramers (C), and linear (D) or cyclolinear (E) polymers. Among the molecular forms, cyclic trimers (B) are the most common as their higher stability over the monomer (A)   71 and cyclic tetramer (C) leads to preferential formation during phosphazene synthesis. Thus, hexachlorocyclotriphosphazene (4.1, more simply known as hexachlorophosphazene), is a convenient commercially available starting material for phosphazene substitution chemistry.175  Figure 4.1: Common examples of phosphazene motifs.   With the goal of covalently attaching a phosphorus-containing moiety to a polymeric substrate in mind, the synthesis of a chemically functional phosphazene derivative is necessary. The simplest method to achieve this involves the use of a bifunctional reagent (Scheme 4.1). However, in the reaction of 4.1 with simple bifunctional reagents, spiro- (G) and ansacyclic (H) products, or even oligo- and polymeric (I) products can be formed in addition to the desired product (F). For example, over several decades a number of studies on the reaction of 4.1 with aliphatic diamines have been reported in which only the synthesis of compounds of the type G and H is described.72, 78, 176-179  NPNPNPP NnB DNPNPNPEP NAnNP N PNPNPC  72  Scheme 4.1: Reactions of hexachlorophosphazene (4.1) with a bifunctional reagent to yield a variety of products.   Alternative synthetic approaches to hexafunctional phosphazene derivatives exist, however, they generally involve multi-step procedures. For example, hexakis(4-aminophenoxy)phosphazene (4.2) has been synthesized from 4.1 by substitution with sodium 4-nitrophenoxide followed by catalytic reduction with hydrogen,180 or by substitution with 4-acetamidophenol followed by base hydrolysis (Scheme 4.2).181 An uncommon example of a single-step procedure is the use of alumina impregnated with KOH to mediate the reaction between 4.1 and hexamethylenediamine to synthesize hexakis(6-aminohexyl)aminophosphazene.182  NPNPNPFG FGClClClClNPNPNPCl FGClClClClNPNPNPCl FGFGClClClNPNPNPCl ClClClClClFG FGFG4.1FG HNPNPNPCl FGClClFGClNPNPNPFG ClFGClClClI  73  Scheme 4.2: Synthesis of hexakis(4-aminophenoxy)phosphazene (4.2) from hexachlorophosphazene (4.1)   It was therefore of critical importance to develop a facile, one-step procedure for the synthesis of a hexafunctional phosphazene derivative, not only to enable the production of large quantities of flame retardant required for extensive flame testing, but also to make the flame retardant commercially relevant. For this reason, hexakis(2-aminoethyl)aminophosphazene (4.3), the non-cyclized product of a reaction between 4.1 and ethylenediamine, was chosen as a synthetic target. Ethylenediamine is a convenient reagent due to its low cost and more convenient physical properties compared to other simple diamines. Specifically, 1,4-diaminobutane and 1,5-diaminopentane have such offensive odours they are commonly known as putrescine and cadaverine, respectively. Phosphazene 4.3 has been reported previously, however, characterization was limited.183  NPNPNPO OOOOOH2N NH2NH2NH2H2NH2NNPNPNPCl ClClClClCl1. 2. H2, PtO2 (cat.)NaONO2HONHCOCH31.     K2CO32. NaOH4.14.2  74 4.2 Results and Discussion 4.2.1 Synthesis of Hexakis(2-aminoethyl)aminophosphazene (4.3) While schematically straightforward, careful control of stoichiometry and addition rate were necessary to prevent unwanted cyclization or cross-linking reactions during the synthesis of phosphazene 4.3. To this end, a solution of N3P3Cl6 (4.1) was slowly added dropwise to a concentrated solution of ethylenediamine, followed by reflux overnight (Scheme 4.3). After precipitation by adding a methanol solution of the crude product to diethyl ether, the final product (4.3) was isolated as a hygroscopic white powder.   Scheme 4.3: Synthesis of hexakis(2-aminoethyl)aminophosphazene (4.3) from the reaction of hexachlorophosphazene (4.1) with ethylenediamine.   The identity of the isolated product was confirmed through spectroscopic analysis. 31P NMR spectroscopy of a solution of 4.3 revealed a sharp singlet (δ = 20.3, Figure 4.2 inset) rather than the series of multiplets that would arise from a mixture of substitution patterns, or a broad signal if a cross-linked polymer were formed. When analyzed by 1H NMR spectroscopy, the predicted splitting is obscured by the  slightly broadened signals. Despite this, the resonances are distinct enough that their identities can still be assigned (Figure 4.2). The exact assignment of the CH2 groups was made by comparison between the 1H and 1H{31P} NMR spectra, as the protons of the CH2 group closest to the phosphazene ring exhibit 3JPH coupling while those of the other NPNPNPCl ClClClClClH2N NH24.1 4.3NPNPNPHNHNNHHNNHNHNH2H2NH2N NH2NH2H2NTHF  75 CH2 group do not. All NMR spectroscopy was performed using D2O as the solvent, thus exchange between the protons of the primary amines and the deuterium atoms of the solvent completely suppressed any resonance for the NH2 protons. In the 13C{1H} NMR spectrum, the two resonances corresponding to each CH2 group exhibited similar peak broadening as observed in the 1H NMR spectrum.   Figure 4.2: 1H and 31P{1H} (inset) NMR spectra of hexakis(2-aminoethyl)aminophosphazene (4.3)   Batches of 4.3 were always contaminated with the monohydrochloride salt of ethylenediamine, en•HCl, as indicated by a sharp singlet (δ = 2.99 ppm) in the 1H NMR spectrum. This impurity results from the necessity of using a large excess of ethylenediamine in the reaction partly to prevent cyclization and cross-linking, but also as an auxiliary base. Although the majority of the unreacted ethylenediamine was removed in the purification, the similar solubility and chemical functionalities of 4.3 and en•HCl meant that they could not be completely separated. In addition, some of the HCl liberated by the substitution reaction reacts with the amines of phosphazene 4.3 and is therefore retained in the final product. By combining an estimate of the en•HCl content made from the 1H NMR spectrum with the known Cl content   76 from either elemental analysis or AgCl precipitation, the nature of the final product could be estimated to be 4.3•2en•8HCl in a typical batch.  In an attempt to reduce the amount of impurities in the final product, the synthesis of 4.3 was performed using a lower stoichiometry of ethylenediamine (6 equiv vs 12 equiv) in the presence of an excess of a non-nucleophilic base (e.g. triethylamine or 4-(dimethylamino)pyridine). These efforts were unsuccessful, however, as the reaction always yielded a mixture of products judging by the 31P NMR spectrum in which multiple resonances with complex splitting patterns were observed. Another strategy involved using a singly-protected ethylenediamine derivative184 (4.4) in the reaction with phosphazene 4.1 (Scheme 4.4). In principle this would eliminate the need to use an excess of ethylenediamine by completely preventing any cyclization or cross-linking reactions. The expected product of this reaction (4.5) could then be deprotected to yield phosphazene 4.3. However, based on the mixture of resonances in the 31P NMR spectrum of the crude product, side reactions still occurred. This indicates that despite the presence of excess triethylamine, the HCl being liberated by the substitution reaction was able to remove the acid-labile t-butoxycarbonyl (Boc) protecting groups of either 4.4 or partially-substituted intermediates.  Scheme 4.4 Proposed reaction of a singly-protected ethylenediamine derivative with 4.1 to yield a protected hexa-amino phosphazene derivative (4.5).  H2N NH24.1 (1/6 equiv)NEt34.5NPNPNPHNHNNHHNNHNHNHNHHNHNNHNHTHFH2N NHBocBoc2Odioxane0 °CBocBocBocBocBocBoc4.4  77 4.2.2 Thermogravimetric Analysis The high water solubility of 4.3 limits its utility as the sole component of a flame retardant treatment. Despite this, before any strategies for rendering 4.3 non-leachable were attempted it was necessary to fully evaluate the flame retardant properties of paper treated with 4.3. Using the same paper made from thermomechanical pulp used in the previous chapter, treated paper samples were prepared by drip-coating using solutions of 4.3 in water. The samples were allowed to fully dry before testing.  The thermal stability of paper samples coated with 0-15 wt% of 4.3 was tested by thermogravimetric analysis (TGA), as seen in Figure 4.3 and Table 4.1. As discussed in Chapters 1 and 2, phosphorus-based flame retardants for cellulosic materials are thought to function primarily by first decomposing to phosphoric acid, followed by the acid-catalyzed dehydration and cross-linking of cellulose to form a non-combustible char.94, 135, 137, 140, 141 As greater amounts of 4.3 are used to treat paper samples, the Tonset of the first primary mass loss after the initial drying stage decreases from a maximum of 309 °C for samples treated with 1 wt% of 4.3 to 271-280 °C for samples treated with 5-15 wt% of 4.3. This is characteristic of cellulose decomposition by crosslinking and charring rather than the higher temperature process of depolymerization.94, 95, 136-139 This is supported by the change in percent mass lost for the same stage, starting at 60% for untreated paper and decreasing to 30% for samples treated with 15% 4.3. The second stage of decomposition can be attributed to the slow thermal oxidation of both the char formed in the previous step as well as the lignin in the paper.143, 144 No relationship is observed between the amount of 4.3 used and either the Tonset or percent mass lost for this stage.    78 Table 4.1: Summary of data obtained from thermogravimetric analysis of paper samples treated with varying quantities of 4.3.  Drying stage  First stage  Second stage 4.3 (wt%) Weight loss (%)  Tonset (°C) Weight loss (%)  Tonset (°C) Weight loss (%) 0 6  303 60  451 15 1 8  309 52  462 21 5 9  271 37  558 14 10 8  273 33  540 16 15 7  280 30  475 17   Figure 4.3: Thermogravimetric analyses of paper samples treated with varying amounts of 4.3.  4.2.3 Flame Testing of Treated Paper Samples  The flame retardancy of paper samples treated with 0-20 wt% of 4.3 was evaluated by the Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00 (Table 4.2, Figure 4.4). As can be expected, untreated paper was found be flammable, with lengthy flame (machine direction: 14 s; cross direction: 11 s) and glow times (machine direction: 149 s;   79 cross direction: 68 s) resulting in complete combustion of the sample. Despite having comparable flame times (machine direction: 16 s; cross direction: 13 s), the glow times of paper treated with just 1 wt% of 4.3 (machine direction: 24 s; cross direction: 16 s) were much lower compared to untreated paper. When 5 wt% of 4.3 was used, flame times were again relatively constant (machine direction: 12 s; cross direction: 7 s), however, no glow time was observed in any trial. At 10 wt% of 4.3 used, not only was there no glow time in any trial but in some cases the flame time was also zero. Once the amount of 4.3 used reached 15 wt%, appreciable levels of flame retardancy were consistently observed as both the glow and flame times were zero. The same results were obtained with samples treated with 20 wt% of 4.3.   Table 4.2: Summary of data obtained from flame testing of paper samples treated with 4.3 by TAPPI Standard Method T461 cm-00.  Unleacheda  Leacheda Loading of 4.3 (wt%) Flame time (s) Glow time (s)  Flame time (s) Glow time (s) 0 14(11) 149(68)  - - 1 16(13) 24(16)  - - 5 12(7) 0(0)  - - 10 4(0) 0(0)  - - 15 0(0) 0(0)  10(8) 0(0) 20 0(0) 0(0)  - - aValues outside of parentheses are in the machine direction, values inside are for the cross direction.     80  Figure 4.4: Photographs of charred remains of uncoated paper and paper samples coated with phosphazene 4.3 (15 wt%) after flame testing by TAPPI Standard Method T461 cm-00. For coated paper samples, the top sample was not leached prior to flame testing, while the bottom sample was leached.  The limiting oxygen indexes (LOIs) of paper samples treated with 0-25% by weight of 4.3 were also measured, following ASTM Standard Method D2863 (Table 4.3, Figure 4.5). A roughly linear relationship exists between the amount of 4.3 used to treat paper samples and their LOI, from 19.6% for untreated paper to the remarkably large LOI of 43.3% for samples treated with 25 wt% of 4.3. The contribution of residual en•HCl to the overall flame retardancy of 4.3   81 was estimated by measuring the LOI of paper samples coated with 4.4 wt% of en•2HCl. The loading of en•2HCl was chosen as it is equimolar to the amount of en•HCl in samples of paper coated with 15 wt% of 4.3. The LOI was found to be just 22.0%, thus the residual en•HCl in 4.3 has only a modest effect on the flame retardancy of paper samples.   Table 4.3: Summary of LOIs obtained for paper samples treated with 4.3.  LOI (%) Loading of 4.3 (wt%) Unleached Leached 0 19.6 - 1 20.4 - 5 25.1 - 10 27.1 - 12.5 31.0 - 15 35.4 22.7 20 36.5 - 25 43.3 - N.B. LOI of paper sample coated with 4.4 wt% of en•HCl = 22.0%.    82  Figure 4.5: Plot of LOI of treated paper samples vs amount of 4.3 used in treatment. The dashed line marks the concentration of O2 in air (21%).   The leachability of 4.3 when used alone on paper was demonstrated using the leaching component of TAPPI Standard Method T461 cm-00 (Table 4.2, Figure 4.4), in which treated paper samples are submerged in a water bath for four hours followed by drying. Leached samples of paper treated with 15 wt% of 4.3 had increased flame times (Table 4.2) and lower LOIs (Table 4.3) compared to otherwise identical unleached samples, indicative of flame retardant leaching. Interestingly, their glow times remained zero and the LOI was still slightly higher than untreated paper. This suggests that although most of the flame retardant had been washed away, a small amount remains. The observation that some charring still occurs (Figure 4.4) is also evidence that a small amount of 4.3 continues to be present in the paper. Using the LOI data (Table 4.3), the quantity of 4.3 remaining in the leached paper can be estimated to be in the range of 1-5 wt%.    83  In terms of flame retardant performance, phosphazene 4.3 represents a significant improvement over the polymers discussed in Chapter 2. Using TAPPI Standard Method T461 cm-00 for comparison, paper samples treated with 4.3 had flame and glow times of zero and were self-extinguishing before the flame reached the top of the sheet, while polymer-coated paper samples had non-zero flame times and the flame travelled the entire length of the sample. When compared using LOI, polymer-coated paper samples had a maximum LOI of 25.9% while samples treated with 4.3 at a comparable loading (15 wt%) had an LOI of 35.4%.  The significant disadvantage of phosphazene 4.3 compared to the polymers of Chapter 2 is its susceptibility to aqueous leaching. Despite the efficacy of 4.3 as a flame retardant, methods for affixing it to a cellulosic substrate will be needed before it is useful as a flame retardant. Therefore, a comparison to other systems is somewhat irrelevant until a solution to this is found.   4.3 Conclusion  In summary, hexakis(2-aminoethyl)aminophosphazene (4.3) was synthesized and characterized. Though not isolated in a fully purified form, the flame retardant properties of paper treated with 4.3 were evaluated. Thermogravimetric analysis showed that higher treatment levels of 4.3 lead to increased charring at elevated temperatures. When tested by TAPPI T461 cm-00, increasing flame retardant levels led to better performance indicated by decreasing flame and glow times, eventually leading to both values reaching zero. Measurement of the LOIs showed the same trend, as higher LOIs were recorded for samples treated with higher amounts of 4.3. As 4.3 is water soluble, treated paper samples were found to be highly susceptible to aqueous leaching.     84 4.4 Experimental 4.4.1 General Procedures  All manipulations of air and/or water sensitive compounds were performed under inert atmosphere using standard Schlenk techniques. THF was freshly distilled from sodium/benzophenone ketyl prior to use. Ethylenediamine was purchased from Fisher Scientific and purified by distillation prior to use. N3P3Cl6 (4.1) was purchased from Strem, and was sublimed under vacuum and recrystallized from hexanes prior to use. 1H, 13C, and 31P NMR spectra were recorded on Bruker 300 or 400 Avance spectrometers. Chemical shifts for 1H and 13C spectra are reported relative to TMS (δ = 0) and for 31P are reported relative to 85% H3PO4 in H2O (δ = 0). Mass spectra were recorded using a Waters/Micromass LCT. Thermogravimetric Analyses were performed in air on a Perkin Elmer STA 6000 Simultaneous Thermal Analyzer. Discs 1/8” (~3 mm) in diameter were cut with a hole punch from treated paper samples. In each experiment, five discs (ca. 12 mg) were heated from 30 °C to 870 °C at a rate of 10 °C min-1. Flame testing was performed in accordance with Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00, using an apparatus fabricated according to the specifications outlined therein. Limiting Oxygen Index (LOI) testing was performed using a Govmark OI-1 Oxygen Index Module following American Society for Testing and Materials (ASTM) Standard Test Method D2863.     85 4.4.2 Paper Treatment The paper used for all experiments were sheets (basis weight = 200 g m-2) made using a laboratory rectangle dynamic sheet former from thermomechanical pulp produced in a pilot plant from black spruce chips with a total energy input of 2468 kWh t-1. Sheet samples (70 mm × 210 mm for TAPPI T461 cm-00, 50 mm × 140 mm for LOI measurements) were cut (with the machine-direction and cross-direction noted) and conditioned at 23°C and 50% humidity for 24 h. Each paper sample was weighed, and the mass of flame retardant needed to achieve the desired loading was measured and the flame retardant dissolved in H2O. For 70 mm × 210 mm samples, 10 mL of H2O was used, while for 50 mm × 140 mm samples, 5 mL was used. The flame retardant solution was then added dropwise using a Pasteur pipet to evenly coat both sides of the paper sample. The samples were conditioned at 23 °C and 50% humidity for 48 h prior to testing.   4.4.3 Preparation of hexakis(2-aminoethyl)aminophosphazene (4.3) To a stirred solution of ethylenediamine (23 mL, 346 mmol) in THF (30 mL) was added dropwise a solution of hexachlorophosphazene (10 g, 28.8 mmol) in THF (30 mL), immediately forming a white precipitate. Once the addition was complete, the reaction was refluxed overnight. After refluxing, the reaction formed two layers: a clear, colourless and viscous liquid layer of crude product on the bottom and a clear and colourless THF layer on top. The reaction was cooled to room temperature and the top layer decanted off. The bottom layer was dissolved in MeOH (400 mL) with the aid of sonication and heat, followed by addition to a rapidly stirring flask of diethyl ether (1.5 L). The white, hygroscopic precipitate was filtered, quickly transferred to a Schlenk flask, and heated to 60 °C under vacuum for several hours to dry. Upon drying, the   86 white solid could be ground in the absence of moisture to a fine powder. Yield: 18.6 g. The final product is typically contaminated with ethylenediamine hydrochloride, identified by a sharp singlet in the 1H NMR (δ = 2.99 ppm) or 13C{1H} NMR (δ = 39.4 ppm) in D2O.  1H NMR (300 MHz, D2O): δ = 3.11 (br s, 12H, CH2NH2), 3.04 (br m, 12H, PNHCH2), 2.93 (br s, 6H, PNH); 31P{1H} NMR (121 MHz, D2O): δ = 20.3 (s); 13C{1H} NMR (101 MHz, D2O): 40.6 (br s), 38.5 (br s). FT-IR (ATR) νmax (cm-1): 1033, 1087 (C-N), 1180 (P=N), 1434 (CH2), 1505 (N-H), 1603 (N-H), 2065, 2895 (C-H), 3265 (N-H); HRMS (ESI-TOF) m/z: [M+H]+ calcd for C12H43N15P3, 490.3039; found, 490.3044. Anal. found: C, 22.12; N, 29.91; H, 6.51.    87 Chapter 5: Carbodiimide-Mediated Amide Couplings of Carboxylic Acids to Phosphazene Derivatives  5.1 Introduction In Chapter 4 the synthesis of 5.1, a new hexaamino-functionalized phosphazene derivative, was described. Phosphazene 5.1 was found to be an excellent flame retardant for paper but was susceptible to leaching due to its high water solubility. As discussed in Chapter 1, one strategy for rendering flame retardants non-leachable is to have the flame retardant moiety bound to the material through covalent bonds. The high degree of chemical functionality of phosphazene 5.1, specifically the six primary amino groups, offers great potential for direct covalent linkage. Although cyclic or polymeric phosphazenes have been bound to mono-185-190 or disaccharides,191 reactions with more complex macromolecules such as cellulose or cellulose derivatives have not been explored.  Of the numerous linkages that can be formed using amines, amides are attractive due to the strength of the amide bond and the ease with which they can be synthesized. Although many methods for the synthesis of amides have been developed, perhaps the most widely employed involves the use of carbodiimides.192, 193 Beginning in 1955, carbodiimides such as N,N’-dicyclohexylcarbodiimide (DCC, 5.2)194 or N-ethyl-N’-(3-dimethylaminopropyl)carbodiimide (EDC, 5.3)195 have been used for the direct coupling of carboxylic acids with amines under mild NPNPNPHNHNNHHNNHNHNH2H2NH2N NH2NH2H2N5.1  88 conditions. EDC, commonly used as its hydrochloride salt, is particularly notable as its water solubility allows for greater substrate scope and ease of purification of the product. Although linear phosphazene monomers have been used to synthesize carbodiimides,196, 197 the use of carbodiimide reagents in couplings of phosphazene derivatives is unexplored.  Thermomechanical pulp (TMP), made from wood chips in high yield (>90%), retains all the major components of wood including cellulose, hemicelluloses, and lignin.198 Although TMP naturally contains carboxylate groups in the form of glucuronic acid residues in xylan (5.4),199 overall the carboxylate content is low. Bleached thermomechanical pulp (BTMP) has a higher carboxylate content due to the oxidation of lignin during the alkaline peroxide bleaching process.198, 200 This offers the potential for facile chemical modification of pulp fibres via carbodiimide coupling, as has been demonstrated by the synthesis of fluorescent pulp via the EDC coupling of 5-aminofluorescein to BTMP.201 Thus in principle, any amino-functionalized molecule can be coupled to BTMP to impart desired chemical or physical properties.  NMe2NCNNCN5.2 5.3OO OO O O OOO O O OOOOOOHO OHOH3COHOHOHO OHOHOHHOH3COHOHOOCHO HO HOOH OHn5.4  89 Considering the high functionality of 5.1 and the macromolecular structure of pulp, any potential chemistry involving the two could be difficult to study. In order to first assess the potential linking of 5.1 to pulp, it was therefore necessary to use a model system. Relatively simple carboxylic acids can be used as models for the carboxylate-bearing moieties in BTMP, however when designing a model for 5.1, the synthesis of a monoamino phosphazene derivative can pose its own challenges. The selective synthesis of partially substituted derivatives of cyclic phosphazenes is challenging due to the mixtures of products formed in the reaction of N3P3Cl6 (5.5) with nucleophiles (Scheme 5.1).175, 202, 203 Specifically in the case of sodium phenoxide, despite initial efforts yielding only complex mixtures of products after reaction with 5.5,204, 205 it was later found that if the reaction conditions and stoichiometry were carefully controlled the pentasubstituted product could be isolated in more reasonable yields.206 This derivative is attractive as the five phenoxy groups are non-reactive, and can thus be considered to be a type of permanent protecting group. With only one site remaining for substitution on the ring, further chemistry is vastly simplified.   90  Scheme 5.1: Mixture of regio- and stereoisomers formed by nucleophilic substitution of hexachlorophosphazene (5.5).  This chapter describes the synthesis of a monoamino phosphazene derivative and its EDC coupling to a series of carboxylic acids. This includes a glucuronic acid derivative, used as a model for a carboxylic acid moiety present in thermomechanical pulp. Also described is the synthesis of pulp functionalized with phosphazene via EDC coupling, the generation of paper handsheets from this pulp, and an assessment of the flame retardancy of these handsheets.  Nu-NPNPNPCl ClClClClClNPNPNPNu ClClClClClNPNPNPNu NuClClClClNPNPNPNu ClNuClClClNPNPNPNu ClClNuClClNPNPNPNu NuNuClClClNPNPNPNu ClNuClClNuNPNPNPNu ClNuClNuClNPNPNPNu NuNuNuClClNPNPNPNu NuNuClClNuNPNPNPNu NuNuClNuClNPNPNPNu ClNuNuNuNuNPNPNPNu NuNuNuNuNuMonosubstituted: Disubstituted:Trisubstituted:Tetrasubstituted:Pentasubstituted: Hexasubstituted:5.5  91 5.2 Results and Discussion 5.2.1 Synthesis of Monofunctional Phosphazene Derivatives  The reaction of N3P3Cl6 (5.5) with NaOPh can yield a maximum of 12 compounds with different substitutions and/or regio- or stereochemistries (Scheme 5.1). Thus, in order to maximize the yield of the desired pentaphenoxy product (5.6), the literature procedure206 was modified by adding a solution of NaOPh in THF by extremely slow dropwise addition to a chilled solution of 5.5 in THF (Scheme 5.2). After the addition was complete and the reaction warmed to room temperature, the progress of the reaction was monitored by 31P NMR spectroscopy. When five molar equivalents of NaOPh are used, resonances corresponding to the tetraphenoxy (δ = 20.2, d, P(OPh)Cl; 4.9, t, P(OPh)2), pentaphenoxy (5.6) (δ = 22.0, t, P(OPh)Cl; 6.8, d, P(OPh)2) and hexaphenoxy (δ = 8.7, s) products could be observed. By adding additional NaOPh in the same manner as before but in small portions (0.1 equiv), the remaining tetrasubstituted product could be converted to 5.6 and hexaphenoxyphosphazene. After filtration and solvent removal, crude 5.6 was isolated as a viscous, colourless oil.    92  Scheme 5.2: Preparation of monoamino (5.7) and monohydroxy (5.8) phosphazene derivatives from hexachlorophosphazene (5.5)   Although phosphazene 5.6 can be separated from N3P3(OPh)6,204 it was found to be more convenient to simply use the crude product as is, and remove the unwanted hexaphenoxy product after the following step. Crude 5.6 was therefore dissolved in THF and added to a stirred solution of excess ethylenediamine (20 equiv) in THF (Scheme 5.2). After stirring overnight at room temperature and workup, the desired N3P3(OPh)5NHCH2CH2NH2 (5.7) could be separated from N3P3(OPh)6 by repeatedly allowing the product to oil out of a CH2Cl2/hexanes solution and removing the supernatant. After drying under high vacuum, the final product was isolated as a white solid. The 31P{1H} NMR spectrum of 5.7 showed a spectrum similar to that of 5.6, with a doublet of doublets (17.7 ppm, P(OPh)NH) and doublet (8.3 ppm, P(OPh)2).   In an attempt to synthesize an alternative monoamino model compound, phosphazene 5.6 was also reacted with ethanolamine in the presence of triethylamine (Scheme 5.2). Rather than undergoing nucleophilic attack through the oxygen atom, however, ethanolamine instead NPNPNPCl ClClClClClH2N NH2NPNPNPPhO ClOPhOPhPhOPhONPNPNPPhOHNOPhOPhPhOPhONH2NaOPhTHF, 0°     RT5.5 5.65.7THFH2N OH NPNPNPPhOHNOPhOPhPhOPhOOH5.8THF  93 attacked through the amino group to yield the monohydroxy derivative (5.8). While it is still possible to react alcohols with carboxylic acids via carbodiimide coupling to generate esters, given the amino substitution of 5.1 the coupling chemistry of 5.8 was not investigated.  5.2.2 X-ray Crystallography Crystals of 5.7 and 5.8 suitable for X-ray diffraction were obtained by slow evaporation of a CH2Cl2 solution. The molecular structure of 5.7 is shown in Figure 5.1, the molecular structure of 5.8 in Figure 5.2, and the metrical parameters of both structures in Table 5.2. The P-N bond lengths within the phosphazene ring of 5.7 vary within the range 1.582(2)-1.605(2) Å, the average length being 1.584(5) Å. The two longest bonds (P(1)-N(1) 1.596(2), P(1)-N(2) 1.605(2) Å) are the P-N bonds involving the P(OPh)(NHR) moiety. Although many examples of penta(aryloxyl)(amino)phosphazenes exist, only a small number have been characterized crystallographically.207-210 The P-N bond lengths of 5.7 fall within the same range of those few examples (1.50-1.68 Å). The N3P3 ring deviates slightly from planarity, with the sum of the internal angles being 713.6(3)°. The atom with the largest distance from an ideal plane through the N3P3 ring is P(3), with a deviation of 0.181 Å. The exocyclic bond P(1)-N(4) is slightly longer than the endocyclic bonds, at 1.630(2) Å.    94  Figure 5.1: Solid-state molecular structure of 5.7. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms are omitted for clarity.  The P-N bond lengths within the phosphazene ring of 5.8 vary within a range very close to 5.7 (1.577(2)-1.605(2) Å), with a slightly longer average length of 1.589(5) Å.  As with 5.7, the two longest bonds (P(1)-N(1) 1.600(2), P(1)-N(3) 1.605(2) Å) are the P-N bonds involving the P(OPh)(NHR) moiety. The N3P3 ring also deviates slightly from planarity, with the sum of the internal angles being 711.2(3)°. The atom with the largest distance from an ideal plane through the N3P3 ring is N(1), with a deviation of 0.184 Å. Thus, the N3P3 ring in 5.8 is slightly   95 less planar than that of 5.7. The exocyclic bond P(1)-N(4) is slightly longer than the endocyclic bonds, at 1.632(2) Å.  Figure 5.2: Solid-state molecular structure of 5.8. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms are omitted for clarity.  5.2.3 Synthesis of Methyl 4-O-Methyl-α-D-glucopyranosiduronic Acid  While any simple carboxylic acid could be used to mimic the carboxylic acid groups in bleached pulp, it was also desirable to have as accurate a model as possible. For this study, methyl 4-O-methyl-α-D-glucopyranosiduronic acid (5.12) was chosen due to its close   96 resemblance to the glucuronic acid residues in xylan (5.4), a hemicellulose found in thermomechanical pulp.199 Carbohydrate 5.12 has also been used to synthesize compounds modeling naturally occuring lignin-carbohydrate complexes.211 The glucuronic acid derivative 5.12 was synthesized according to a literature procedure (Scheme 5.3).212  Beginning from methyl α-D-glucopyranoside (5.9), the first step involved the selective protection of the 2-, 3-, and 6-hydroxyl groups using bis(tributyltin) oxide and benzoyl chloride to yield the benzoyl ester protected product 5.10. Methylation of the 4-hydroxyl group with iodomethane and silver(I) oxide, followed by deprotection of the benzoyl esters using sodium methoxide, yielded methyl 4-O-methylglucopyranoside (5.11). Lastly, 5.12 was obtained by selective oxidation of the primary hydroxyl group in the C6 position to a carboxylic acid by a TEMPO-catalyzed oxidation with sodium hypochlorite and sodium bromide. This final step was the most difficult to accomplish, as in several attempts 5.12 could not be isolated in higher than 10% yield. Nevertheless, enough was obtained to carry on to the next step.  Scheme 5.3: Synthesis of methyl 4-O-methyl-α-D-glucopyranosiduronic acid (5.12) from methyl α-D-glucopyranoside (5.9)  5.2.4 EDC Couplings of N3P3(OPh)5NHCH2CH2NH2 (5.7) with Carboxylic Acids  All coupling reactions followed a common procedure (Scheme 5.4): to a solution of the acid in CH2Cl2, EDC, hydroxybenzotriazole hydrate (HOBt•xH2O), and excess triethylamine OOMeHOHOHOCH2OH 1. (Bu3Sn)2O2. Bz-Cl OOMeOBzBzOHOCH2OBz1. MeI, Ag2O,     DMF2. NaOMe,     MeOH OOMeHOHOMeOCH2OHTEMPONaOClNaBr OOMeHOHOMeOCOOH5.9 5.10 5.11 5.12toluenerefluxH2O0°C  97 were added and the solution stirred at room temperature. The reaction was monitored by TLC to observe the conversion of the carboxylic acid to the –OBt active ester. Once the activation step was complete, amine 5.7 was added to the solution, and the reaction was stirred at room temperature. After 24 hours, the reaction mixture was diluted with CH2Cl2 and washed five times with distilled water followed by concentration under vacuum and purification by column chromatography. CH2Cl2 was used for all coupling reactions due to the insolubility of 5.7 in water. Other than glucuronic acid derivative 5.12 discussed earlier, the carboxylic acids used in this study (benzoic acid, cyclohexanecarboxylic acid, isobutyric acid, cyclopentanecarboxylic acid) were chosen for their availability. In all cases, the amide-coupling products (5.13-5.17) were isolated and characterized. Importantly, the 1H NMR spectrum of each product showed the complete disappearance of the NH2 signal from the starting material (1.47 ppm) and the appearance of signals corresponding to the NH protons of the amide functionalities in the products (6.1-7.1 ppm). The chemical shifts of these signals fell within the typical range for amides.  Amides 5.13-5.16 were isolated in acceptable yields (30-44%), however the yield of 5.17 was lower (10%). This was partly due to the presence of residual acetic acid in samples of 5.12, left over from the mobile phase used in the purification column. The acetic acid would have also reacted with the EDC and HOBt, and then competed for reactivity with 5.7. This was confirmed by the isolation of the acetyl amide from the column used to purify 5.17. Another factor affecting the yield of 5.17 was the poor solubility of 5.12 in CH2Cl2.    98  Scheme 5.4: Synthesis of phosphazene-containing amides (5.13-5.17) using a carbodiimide-mediated coupling  5.2.5 Formation of Phosphazene-Functionalized BTMP Handsheets  The EDC coupling of 5.1 to BTMP was performed using an adapted literature procedure.201 BTMP, EDC, and 5.1 were mixed together in water at a consistency of 2%, and allowed to stand for three hours before washing and handsheet preparation. Consistency is defined as the weight in grams of oven-dry pulp in 100 grams of pulp-water mixture. The mass of 5.1 was varied between 0-20% of the oven-dried weight of the BTMP, and the EDC was always 50% by mass of 5.1. As a control, handsheets were also prepared by mixing BTMP and 5.1 without any EDC, as well as from BTMP alone.   In order to assess the extent of incorporation of 5.1 into the handsheets, their flame retardancy was tested according to the Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00. The results are shown in Table 5.1. Unsurprisingly, handsheets made from only BTMP (Entry 1, Table 5.1) were found to be very flammable with lengthy flame and glow times (7 s and 77 s, respectively. For the handsheets made from BTMP NPNPNPPhOHNOPhOPhPhOPhONHR-COOHEDCHOBtNEt3ROR =OOMeOHHOMeONPNPNPPhOHNOPhOPhPhOPhONH25.7 5.135.145.155.16PhCyi-PrC5H95.17CH2Cl2  99 treated with 5.1 and EDC, in all cases charring could be observed with zero glow time after the flame self-extinguished. Despite this, only two samples (Entries 4 and 6, Table 5.1) showed a decrease in flame time. Generally the flame times of these handsheets did not change, and in one case actually increased (Entry 7, Table 5.1). The lack of relationship between the mass of 5.1 used and flame retardancy, and the overall poor performance of the prepared handsheets indicates that although a small amount of 5.1 appears to have reacted with BTMP, it is not enough to impart appreciable flame retardancy to the resulting handsheets.  The efficiency of the coupling procedure can be qualitatively evaluated by using the known flame retardant properties of 5.1 from Chapter 4. In that chapter, paper made from TMP treated with 5 wt% of 5.1 had no glow time and a small decrease in flame time, while paper treated with 1 wt% had a small, but non-zero glow time and a slightly increased flame time. Thus, it can be estimated that the amount of 5.1 bound to the BTMP handsheets is in the range of 2-4 wt%. Theoretically, given that BTMP generally has a total carboxyl content of ~270 mmol kg-1,7 of which ~25% (68 mmol kg-1) is likely located on the surface of the fibres,213 one would expect a maximum of ~6 wt% of 5.1 could be attached to the surface of the BTMP fibres, assuming no crosslinking. Although the resulting estimated yield (33-67%) is quite a broad range, it indicates that the coupling reaction was most likely of moderate efficiency. The presence or absence of EDC had no consistent effect on the flame retardancy of the resulting handsheets. This is surprising, since during handsheet preparation the pulp is washed several times with a large amount of water. This should wash away any unreacted 5.1, thus in the cases where no EDC was used one would expect the entire amount of 5.1 to be extracted. Since this is not the case, there may be a different type of non-covalent interaction between the 5.1 and BTMP. As the reaction is carried out at a pH of 4, the amine groups of 5.1 and the carboxylates   100 of BTMP will both be protonated. Although no ionic interaction is possible, however hydrogen bonding may be occurring.  Table 5.1: Summary of flame testing of BTMP handsheets  % Mass   Entry 5.1 EDC Flame Time (s) Glow Time (s) 1 0 0 7 77 2 2 0 7 0 3 2 1 7 0 4 5 0 6 0 5 5 2.5 7 0 6 10 0 5 0 7 10 5 10 0 8 20 0 7 0 9 20 10 7 0  5.3 Summary  In summary, a monoamino model compound (5.7) for hexakis(2-aminoethyl)aminophosphazene (5.1) was prepared and used to synthesize a series of amides (5.12-5.16) via EDC coupling with carboxylic acids. This work represents the first example of using carbodiimide coupling in phosphazene chemistry. Additionally, 5.1 was coupled to BTMP using EDC, followed by the formation of handsheets from this pulp. The flame retardancy of these handsheets was evaluated using TAPPI Standard Method T461 cm-00. They were found to be only modestly flame retardant, indicating low incorporation of 5.1 into the pulp.    101 5.4 Experimental 5.4.1 X-ray Crystallography The single crystals of 5.7 and 5.8 were immersed in oil and mounted on a glass fiber. Data was collected on a Bruker X8 APEX II diffractometer with graphite-monochromated Mo Kα radiation. The structure was solved by direct methods and subsequent Fourier difference techniques. All non-hydrogen atoms were refined anisotropically with hydrogen atoms being included in calculated positions but not refined. The data set was corrected for absorption effects (SADABS), Lorentz, and polarization effects. All calculations were performed using SHELXL-2014 crystallographic software package from Bruker AXS. Additional crystal data and details of data collection and structure refinement are listed in Table 5.2.        102 Table 5.2: X-ray crystallographic data of 5.7 and 5.8 Crystal 5.7 5.8 Formula C32H32N5O5P3 C32H31N4O6P3 Formula Weight 659.55 660.52 Crystal System Triclinic Orthorhombic Space Group P -1 Pca21 Colour Colourless Colourless a (Å) 10.054(4) 11.131(1) b (Å) 12.545(4) 15.449(2) c (Å) 13.341(7) 17.909(2) α (°) 87.166(9) 90 β (°) 72.256(8) 90 γ (°) 76.973(5) 90 V (Å3) 1561.0(11) 3079.5(5) Z 4 4 T (K) 90(2) 183(2) µ (Mo Κα) (mm-1) 0.241 0.246 Crystal Size (mm) 0.35 x 0.24 x 0.12 0.50 x 0.30 x 0.10 Dcalcd. (g cm-3) 1.403 1.425 2Θ (max) (°) 55.04 60.04 No. of Reflections 28032 61872 No. of Unique Data 7934 9672 Rint 0.0348 0.0416 Reflections/parameters ratio 17.1 21.8 R1, wR2[I > 2σ(I)]a 0.0437 0.0378 R1, wR2 (all data)b 0.1198 0.0991 GOF 1.008 1.090 aR1 = Σ ||Fo| - |Fc|| / Σ|Fo|. bwR2 = [Σ (w(Fo2 – Fc2)2)/Σ w(Fo2)2]1/2   5.4.2 General Procedures  All manipulations of air and/or water sensitive compounds were performed under inert atmosphere using standard Schlenk or glovebox techniques. THF was freshly distilled from sodium/benzophenone ketyl prior to use. Hexa(2-aminoethyl)aminophosphazene (5.1) was   103 prepared according to the procedure described in Chapter 4. Methyl 4-O-methyl-α-D-glucopyranosiduronic acid (5.12) was prepared according to literature procedure.212 Dichloromethane, hexanes, ethyl acetate, methanol, ethylenediamine, triethylamine, and benzoic acid were purchased from Fisher Scientific and used as received. Cyclohexanecarboxylic acid, cyclopentanecarboxylic acid, isobutyric acid, and EDC were purchased from Sigma Aldrich and used as received. N3P3Cl6 (5.5) was purchased from Strem, and was sublimed under vacuum and recrystallized from hexanes prior to use. HOBt hydrate was purchased from Oakwood and used as received.  1H, 13C, and 31P NMR spectra were recorded on Bruker Avance 300 or 400 MHz spectrometers. Chemical shifts for 1H and 13C spectra are reported relative to TMS (δ = 0) and for 31P are reported relative to 85% H3PO4 in H2O (δ = 0). Mass spectrometry was performed at the University of British Columbia. BTMP was supplied by FPInnovations. Handsheets were made using a British Sheet Machine in accordance to Pulp and Paper Technical Association of Canada (PAPTAC) Standard Method C.5. Flame testing was performed in accordance with Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00, using an apparatus fabricated according to the specifications outlined therein.   5.4.3 Preparation of pentaphenoxychlorophosphazene; N3P3(OPh)5Cl (5.6) Adapted from a literature procedure.206 To a vigorously stirred solution of 5.5 (6.0 g, 17.3 mmol) in THF (60 mL) chilled in an ice bath, a solution of sodium phenoxide (10 g, 86.5 mmol) in THF (240 mL) was added dropwise as slowly as possible such that the addition took approximately 10 hours. After the addition was complete, an aliquot of the reaction was analyzed   104 by 31P NMR spectroscopy. The two largest signals (δ = 22.0, t; 6.8, d) were assigned to the product, the singlet (δ = 8.7) was assigned to hexaphenoxyphosphazene, and the small doublet (δ = 20.2) and triplet (δ = 4.9) were assigned to tetraphenoxydichlorophosphazene. Another addition of sodium phenoxide (0.2 g, 1.72 mmol) in THF (5 mL) was performed dropwise as before, and the reaction mixture once again analyzed by 31P NMR spectroscopy. This showed just 5.6 and hexaphenoxyphosphazene, so all volatiles were removed in vacuo and the residue dissolved in CH2Cl2 (100 mL). The solution was filtered through Celite and the solvent removed in vacuo to yield the crude product as clear, colourless oil. The crude product was used without further purification.  31P{1H} NMR (121 MHz, THF): δ = 20.9 (t, 2JPP = 84 Hz, PCl(OPh)), 5.72 (d, 2JPP = 84 Hz, P(OPh)2).  5.4.4 Preparation of pentaphenoxy(2-aminoethyl)aminophosphazene; N3P3(OPh)5NHCH2CH2NH2 (5.7) To a stirred solution of ethylenediamine (24 mL, 0.35 mol) in THF (50 mL), a solution of crude 5.6 (17.3 mmol based on previous step) in THF (50 mL) was added. The reaction was stirred at room temperature overnight. After completion of the reaction, the volatiles were removed in vacuo and the residue dissolved in CH2Cl2 (50 mL), filtered through Celite, and washed three times with 1M HCl. The organic phase was then concentrated and added to a flask containing vigorously stirred hexanes to yield a fine precipitate. Upon standing the precipitate oiled out of solution, and the mother liquor could be decanted off. The precipitating/oiling out procedure was repeated until 31P NMR spectroscopy of the oil showed the   105 hexaphenoxyphosphazene (δ = 8.3 ppm, s) was completely removed. The oil was then dried under vacuum to yield the product as a white hygroscopic solid. Yield = 61% (two steps)  1H NMR (300 MHz, CDCl3): δ = 7.33-6.95 (m, 25H, Ar-H), 2.73 (m, 2H, PNHCH2-), 2.70 (m, 1H, NH), 2.55 (m, 2H, -CH2NH2), 1.47 (br s, 2H, NH2); 31P{1H} NMR (121 MHz, CDCl3): δ = 17.7 (dd, 2JPP = 71 Hz, 2JPP = 77 Hz, P-NH), 8.3 (d, 2JPP = 76 Hz, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 150.9 (s), 129.4 (s), 129.3 (s), 124.9 (s), 124.7 (s), 124.5 (s), 121.3 (s), 121.2 (s), 121.0 (s), 43.4 (s), 42.6 (d, 2JPC = 8 Hz). HRMS (ESI-TOF) m/z: [M+H]+ Calcd for C32H33N5O5P3 660.1695; Found 660.1702.  5.4.5 Preparation of pentaphenoxy(2-hydroxyethyl)aminophosphazene N3P3(OPh)5NHCH2CH2OH (5.8) NH2CH2CH2OH (0.10 mL, 1.7 mmol) was added drop-wise to a solution of crude 5.6 (1.00 g, ca 1.57 mmol) and Et3N (7 mL, 50 mmol) in THF (30 mL). The reaction mixture was stirred at room temperature for 24 hours and an aliquot was removed for 31P NMR analysis. If the spectrum revealed signals corresponding to N3P3(OPh)5Cl, then the reaction mixture was heated to 50 ˚C for 15 hours. Upon completion, the volatiles were removed in vacuo to afford a pale yellow, sticky residue. Subsequently, toluene (ca. 45 mL) was added to the residue and filtered. The solvent was removed using a rotary evaporator, leaving a pale yellow oil. N3P3(OPh)5NHCH2CH2OH and N3P3(OPh)6  were separated using silica gel chromatography in hexanes/ethyl acetate (60:40) eluent. Complete removal of the solvent afforded a pale yellow solid. Yield: 0.20 g (20 %)  1H{31P} NMR (300 MHz, C6D6): δ = 6.81-7.29 (m, 25H, Ar-H), 3.13 (q, 2H, 3JHH = 6 Hz, -CH2OH), 2.66 (q, 2H, 3JHH = 6 Hz, PNHCH2), 2.46 (t, 1H, 3JHH = 6 Hz, NH), 1.79 (t, 1H, 3JHH =   106 6 Hz, OH); 31P{1H} NMR (121 MHz, CDCl3): δ = 17.7 (t, 2JPP = 75 Hz, P(OPh)NH),  8.3 (d, 2JPP = 75 Hz, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 150.9 (s) 129.30 (s), 124.86 (s), 121.18 (s), 62.82 (s), 43.36 (s); Anal. Calcd. for C32H31N4O6P3: C, 58.19; H, 4.73; N, 8.48. Found: C, 58.49; H, 4.73; N, 8.50.   5.4.6 General Carbodiimide Coupling Procedure To a stirred solution of carboxylic acid (0.228 mmol) in dichloromethane (5 mL), triethylamine (0.21 mL, 1.52 mmol), EDC-HCl (87 mg, 0.456 mmol), and HOBt hydrate (90 mg) were added. The reaction was monitored by TLC (100% EtOAc) until the complete conversion of the carboxylic acid to –OBt ester could be observed, at which time 5.7 (100 mg, 0.152 mmol) was added. After stirring at room temperature for 24h, the reaction mixture was washed five times with distilled water, concentrated by evaporation, and purified by column chromatography.  5.4.7 Preparation of Pentaphenoxy((N-benzoyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOPh (5.13) Purified by column chromatography (10% MeOH in CH2Cl2) Rf = 0.66. Off-white oil. Yield: 40%.  1H NMR (300 MHz, CDCl3): δ = 7.82 (d, 2H, J = 7 Hz), 7.48-6.94 (m, 28H), 7.10 (m, 1H, NH-COPh) 3.34 (m, 2H, CH2-NHCO), 2.92 (m, 2H, PNH-CH2), 2.73 (m, 1H, P-NH). 31P{1H} NMR (121 MHz, CDCl3): δ = 17.2 (dd, 2JPP = 73 Hz, 2JPP = 78 Hz, P-NH), 8.4 (m, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 167.6 (s), 150.7 (m), 134.2 (s), 131.3 (s), 129.5 (s), 129.4 (s), 128.4 (s), 127.2 (s), 125.0 (s), 124.8 (s), 124.7 (s), 121.1 (s), 121.0 (s), 41.4 (d, 2JPC   107 = 4 Hz), 40.3 (s). HRMS (ESI-TOF) m/z: [M+H]+ Calcd for C39H37N5O6P3 764.1957; Found 764.1953.  5.4.8 Preparation of Pentaphenoxy((N-cyclohexylcarbonyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOCy (5.14) Purified by column chromatography (10% MeOH in CH2Cl2) Rf = 0.48. Off-white oil. Yield: 30%.  1H NMR (300 MHz, CDCl3): δ = 7.31-6.94 (m, 25H), 6.12 (br s, 1H, NH-CO), 3.13 (m, 2H, CH2-NHCO), 2.81 (m, 2H, PNH-CH2), 2.65 (m, 1H, P-NH), 1.98-1.17 (m, 11H). 31P{1H} NMR (121 MHz, CDCl3): δ = 17.2 (dd, 2JPP = 71 Hz, 2JPP = 78 Hz, P-NH), 8.3 (m, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 176.5 (s), 150.9 (m), 129.5 (s), 129.4 (s), 125.0 (s), 124.8 (s), 124.7 (s), 121.2 (s), 121.1 (s), 121.0 (s), 45.4 (s), 40.5 (d, 2JPC = 5 Hz), 40.4 (s), 29.6 (s), 25.7 (s). HRMS (ESI-TOF) m/z: [M+H]+ Calcd for C39H43N5O6P3 770.2426; Found 770.2429.  5.4.9 Preparation of Pentaphenoxy((N-isobutyryl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOCH(CH3)2 (5.15) Purified by column chromatography (5% MeOH in CH2Cl2) Rf = 0.24. Off-white oil. Yield: 42%.  1H NMR (300 MHz, CDCl3): δ = 7.32-6.94 (m, 25H), 6.17 (m, 1H, NH-CO), 3.12 (m, 2H, CH2-NHCO), 2.81 (m, 2H, PNH-CH2), 2.69 (m, 1H, P-NH), 2.21 (sept, 1H, J = 7 Hz, CH(CH3)2), 1.09 (d, 6H, J = 7 Hz, CH(CH3)2). 31P{1H} NMR (121 MHz, CDCl3): δ = 17.2 (dd, 2JPP = 73 Hz, 2JPP = 77 Hz, P-NH), 8.4 (d, 2JPP = 74 Hz, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 177.4 (s), 150.8 (m), 129.5 (s), 129.4 (s), 125.0 (s), 124.8 (s), 124.7 (s), 121.1 (s),   108 121.0 (s), 40.6 (d, 2JPC = 4 Hz), 40.4 (s), 35.4 (s), 19.6 (s). HRMS (ESI-TOF) m/z: [M+Na]+ Calcd for C36H38N5O6NaP3 752.1933; Found 752.1923.  5.4.10 Preparation of pentaphenoxy((N-cyclopentylcarbonyl)2-aminoethyl)aminophosphazene; N3P3(OPh)5NH(CH2)2NHCOC5H9 (5.16) Purified by column chromatography (5% MeOH in CH2Cl2). Off-white oil. Yield: 44%.  1H NMR (300 MHz, CDCl3): δ = 7.32-6.94 (m, 25H), 6.16 (m, 1H, NH-CO), 3.13 (m, 2H, CH2-NHCO), 2.82 (m, 2H, PNH-CH2), 2.72 (m, 1H, P-NH), 2.37 (m, 1H, CH-CONH), 1.72 (m, 4H), 1.50 (m, 2H), 1.31 (m, 2H). 31P{1H} NMR (121 MHz, CDCl3): δ = 17.3 (dd, 2JPP = 73 Hz, 2JPP = 77 Hz, P-NH), 8.3 (d, 2JPP = 73 Hz, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 176.7 (s), 150.8 (m), 129.5 (s), 129.4 (s), 129.3 (s), 125.0 (s), 124.8 (s), 124.6 (s), 121.2 (s), 121.0 (s), 45.7 (s), 40.7 (d, 2JPC = 5 Hz), 40.4 (s), 30.4 (s), 25.9 (s). HRMS (ESI-TOF) m/z: [M+Na]+ Calcd for C38H40N5O6NaP3 778.2089; Found 778.2092.  5.4.11 Preparation of Pentaphenoxy((N-(methyl 4-O-methyl-α-D-glucopyranosiduronyl))2-aminoethyl)aminophosphazene (5.17) Purified by column chromatography (2-10% MeOH in CH2Cl2) Rf = 0.48. White crystalline solid. Yield: 10%.  1H NMR (300 MHz, CDCl3): δ = 7.32-6.91 (m, 25H), 6.83 (m, 1H, NH-CO), 4.71 (d, 1H, J = 4 Hz, C(1)-H), 3.95 (d, 1H, J = 10 Hz, C(5)-H), 3.77 (m, 1H, C(3)-H), 3.52 (s, 3H, C(4)-OCH3), 3.50 (m, 1H, C(2)-H), 3.37 (s, 3H, C(1)-OCH3), 3.28 (m, 1H, C(4)-H), 3.22 (m, 2H, CH2-NHCO), 2.86 (m, 2H, PNH-CH2), 2.66 (m, 1H, P-NH). 31P{1H} (121 MHz, CDCl3): δ = 17.4 (m, P-NH), 8.3 (m, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 169.3 (s), 150.7 (m),   109 129.5 (s), 129.4 (s), 125.0 (s), 124.8 (s), 121.2 (s), 120.9 (s), 99.2 (s), 81.8 (s), 74.2 (s), 71.9 (s), 70.6 (s), 60.5 (s), 55.9 (s), 40.6 (d, 2JPC = 5 Hz), 40.1 (s). HRMS (ESI-TOF) m/z: [M+H]+ Calcd for C40H45N5O11P3 864.2328; Found 864.2335.  5.4.12 Preparation of phosphazene-modified BTMP  BTMP (4 g oven-dried weight), 5.1 (variable mass based on oven-dried weight of pulp), and EDC (50% by mass of 5.1) were mixed and enough distilled water added such that the total weight of the reaction was 200 g (2% consistency). The pH of the reaction was adjusted using H2SO4 such that the pH was 4. The mixture was thoroughly mixed using an overhead stirrer, and then allowed to sit at room temperature without stirring for three hours. The pulp was then filtered and thoroughly washed three times with distilled water (400 mL), followed by handsheet formation according to PAPTAC Standard Method C.5.    110 Chapter 6: Carboxymethyl Cellulose and Hexakis(2-aminoethyl)aminophosphazene as a Two-Part Flame Retardant Treatment for Paper  6.1 Introduction As outlined in Chapter 1, the three major methods of incorporating a phosphorus-containing flame retardant into a polymeric material are a) physical blending, b) copolymerization, or c) post-polymerization modification (Figure 1.3). Chapter 2 described the use of phosphorus-containing polymers as flame retardant additives for paper. Despite being moderately effective as non-leachable flame retardants, the laborious synthesis of the polymers made them unsuitable for wide application. Since copolymerization can be ruled out for cellulosic materials due to cellulose being a naturally sourced polymer, this left covalent attachment of phosphorus-containing moieties as the remaining strategy. Chapter 4 reported the synthesis and flame retardant properties of a hexaamino-functionalized phosphazene derivative (6.1), while in Chapter 5 attempts to couple 6.1 to bleached thermomechanical pulp via carbodiimide coupling were described. Although these efforts were unsuccessful, phosphazene 6.1 remained an attractive flame retardant due to its ease of synthesis and excellent flame retardant properties.   6.1NPNPNPHNHNNHHNNHNHNH2H2NH2N NH2NH2H2N  111 Despite covalent attachment being unproductive, non-covalent attachment is an option. Phosphazene 6.1 is typically isolated with all six primary amine groups protonated to form ammonium groups, thus ionic attraction between the positively charged ammonium groups of 6.1 and negatively charged carboxylates in thermomechanical pulp (TMP) is possible. Hydrogen bonding could also occur between the donating ammonium groups of 6.1 and various potential acceptors within TMP. In Chapter 4, phosphazene 6.1 was found to be highly leachable when coated on paper, thus it can be concluded that the possible non-covalent interactions between 6.1 and pulp are either too weak or not numerous enough to prevent leaching. An additional compound is therefore needed that could bind to both 6.1 and to cellulose.  Sodium carboxymethyl cellulose (6.3) is a water-soluble cellulose (6.2) derivative used as a thickening agent, emulsifier, and solution stabilizer.214-216 In the technical grade its applications include use in the textile, mining, and pulp industries while its finer grade (cellulose gum) is found in detergents, cosmetics, pharmaceuticals, and even food. The most common bulk-scale synthesis of 6.3 is a non-selective etherification of the 2-, 3-, and 6-hydroxyl groups of the anhydroglucose repeat unit of cellulose with sodium monochloroacetate (Scheme 6.1), however more specialized methods have also been developed.217 Samples of 6.3 are typically characterized by two parameters: molecular weight and degree of substitution (DS). As with any polymer, higher molecular weight samples of 6.3 are less soluble and form more viscous solutions. The DS describes the average number of carboxymethyl groups per anhydroglucose unit. Although the theoretical range is 0.0-3.0, commercially available samples are typically in the range of 0.4-1.5. More substituted samples of 6.3 have higher solubility and result in solutions with superior flow properties.    112  Scheme 6.1: Synthesis of sodium carboxymethyl cellulose (6.3) via etherification  Due to 6.3 being negatively charged and the H-bonding potential of its hydroxy and carboxylate groups, it was hypothesized that it may be able to bind to 6.1 via both ionic and H-bonding interactions. The ability of 6.3 to adsorb irreversibly onto cellulose fibres via H-bonding is well known,218-223 thus 6.3 could act as a “glue” by affixing 6.1 to a cellulosic substrate such as paper (Figure 6.1).   Figure 6.1: Cartoon representation of proposed interaction between 6.1, 6.3, and cellulose in which 6.3 acts as a “glue” by binding to both 6.1 and cellulose via non-covalent interactions.  OROOROROROOOROOR nR = H orOONa6.3OHOOHOHOHOOOHOOH n6.2Cl ONaO1. NaOH2.  113 6.2 Results and Discussion 6.2.1 Flame Testing by TAPPI T461 cm-00 Paper samples were coated with 6.1 (15 wt% of paper) followed by 6.3 (15 wt% of paper), each in aqueous solution (final P loading: 1.2 wt%). Between each coating, the sample was air-dried. The flame retardancy was first measured using the Technical Association of Pulp and Paper Industry (TAPPI) Standard Method T461 cm-00 (Table 6.1, Figure 6.2). Paper coated with 6.3 alone (15 wt%) was used as a control. For comparison, the results for uncoated paper as well as paper treated with monobasic ammonium phosphate (MAP) or 6.1 alone from Chapters 2 and 4 are reproduced.  Paper treated solely with 6.3 had slightly smaller flame times but much longer glow times before leaching when compared to untreated paper. Although this is seemingly contradictory, the entire sample is consumed by the flame therefore 6.3 is not considered to be flame retardant. In Chapter 4, 6.1 was shown to be a highly effective flame retardant, yet was susceptible to aqueous leaching. Importantly, not only did samples coated with 6.1 and 6.3 (15:15 wt%) show flame retardancy comparable to MAP-treated paper before leaching, but the flame and glow times remained zero after leaching in the majority of trials. Interestingly, the flame travelled a shorter distance for the leached sample than the unleached sample (10 cm vs. 12 cm).      114  Table 6.1: Summary of results from flame testing by TAPPI Standard Method T461 cm-00  Unleacheda  Leacheda Treatment Flame time (s) Glow time (s)  Flame time (s) Glow time (s) Uncoated 14(11) 149(68)  - - 15 wt% 6.1 0(0) 0(0)  10(8) 0(0) 15 wt% 6.3 10(11) 228(318)  6(7) 144(127) 15 wt% MAP 0(0) 0(0)  13(12) 133(125) 15:15 wt% 6.1:6.3 0(0) 0(0)  2(0) 0(0) aValues outside of parentheses are in the machine direction, values inside are for the cross direction.    115  Figure 6.2: Photographs of charred remains of uncoated paper and paper samples coated with MAP (15 wt%), 6.1 (15 wt%), and 6.1:6.3 (15:15 wt%) after flame testing by TAPPI Standard Method T461 cm-00. For coated paper samples, the top sample was not leached prior to flame testing, while the bottom sample was leached.  6.2.2 Flame Testing by Limiting Oxygen Index The flammability of treated paper samples was also evaluated by measuring their limiting oxygen index (LOI) values by ASTM D2863 (Table 6.2). Paper treated with 6.3 was found to have a slightly higher LOI than uncoated paper, indicating that contrary to the results obtained from the TAPPI Standard Method T461 cm-00, samples treated with 6.3 are mildly flame retardant. Samples treated with both 6.1 and 6.3 were found to have a very high LOI (39.2, 15:15   116 wt%). Interestingly, the increase in LOIs for samples treated in this way (DLOI = 39.2 – 19.6 = 19.6) were significantly higher than the sum of the increase in LOI for paper treated separately with either 15 wt% 6.1 or 15 wt% 6.3 [DLOI  = (35.4 – 19.6) + (24.7 – 19.6) = 10.9], suggesting a synergistic effect between 6.1 and 6.3.  Most importantly, not only did paper treated with 6.1 and 6.3 retain a high LOI after leaching, but the value unexpectedly increased (before: 39.2, after: 41.1). Although the exact reason for this is not known, it could be due to the interactions between 6.1 and 6.3 not fully reaching completion. Since the treatment involves a solution of 6.3 being coated onto a dry sheet of paper on which 6.1 has already been deposited, the fully optimized reaction between the two may not occur until the entire sample is immersed in water. It is also possible that the NaCl formed in the reaction between the ammonium chloride groups of 6.1 and sodium carboxylates of 6.3 somehow suppresses flame retardancy, thus the system shows better performance after leaching when this NaCl has been washed away.  Table 6.2: LOI values for treated paper samples.  LOI (%) Treatment Unleached Leached Uncoated 19.6 - 15 wt% 6.1 35.4 22.7 15 wt% 6.3 24.7 - 15:15 wt% 6.1:6.3 39.2 41.1    117 6.2.3 Testing of Leaching Under Extreme Conditions All of the previously mentioned leaching experiments were conducted according to TAPPI Standard Method T461 cm-00 which involves just one set of conditions: the submersion of treated paper samples in room temperature, neutral pH water for four hours. In order to test the susceptibility of the combined 6.1:6.3 treatment method in different environments, the time spent in the leaching solution as well as the temperature and pH of the solution were varied. As an additional variable, in some cases the volume of the leaching solution was changed in order to test the effect of stoichiometry vs concentration. After being leached under these conditions, the LOIs of the samples were measured (Table 6.3).  It was initially hypothesized that by varying the leaching time it would be possible to move beyond the qualitatively binary “leachable/not leachable” when describing leachability, and instead quantify the susceptibility of a system over time. However, in all cases there was little variability based on how long a sample was leached. For systems in which the coated paper samples are relatively robust (i.e. neutral water), lengthier leaching times may be needed before this is possible. Remarkably, samples heated in boiling water or in mildly acidic or basic solutions (HCl at pH = 5 or NaOH at pH = 9) retained nearly all their flame retardancy with excellent LOIs being retained (LOI: 37.8-40.8; cf. unleached 39.2). Under highly acidic or basic conditions (pH = 1 or 13), however, there was partial leaching of the flame retardant as evidenced by a drop in the LOI, however a flame retardant effect remained (LOI:  24.8-26.2; cf. untreated paper: 19.6). For these extreme pH conditions, there was no significant difference between using 1 L or 8 L of leaching solution.    118 Table 6.3: LOI values of paper samples treated with 6.1 and 6.3 (15:15 wt%) leached under various conditions for different lengths of time.    LOI (%)    Leaching time Leaching Conditions Volume (L) Temperature (°C) 1 h 2 h 4 h 0.1 M HCl 1 RT 26.2 25.6 24.8 0.1 M HCl 8 RT 25.0 25.4 25.0 0.00001 M HCl 1 RT 40.8 38.8 39.0 H2O 1 RT 40.4 39.8 41.1 H2O 1 100 39.8 38.6 37.8 0.00001 M NaOH 1 RT 39.6 38.4 38.8 0.1 M NaOH 1 RT 25.2 25.4 25.4 0.1 M NaOH 8 RT 25.6 25.2 25.2  6.2.4 Comparison to Related Systems Due to the large number of standard methods for evaluating flame retardancy, direct comparisons between this work and published literature can be difficult. Of the two main methods of measuring flame retardancy used in this work, LOIs are more convenient for comparison due to their being more quantitative than the results from TAPPI Standard Method T461 cm-00. While it is important to remember that flame retardancy should not be evaluated based on one test alone, the comparison is still useful.  When compared to other examples of phosphorus-based flame retardants being used to treat cellulosic materials, paper treated with 6.1 and 6.3 possesses superior flame retardancy. In two reports in which “layer-by-layer” assembly (See Section 1.5.1 in Chapter 1) was used to deposit phosphorus-containing species onto cotton, the maximum LOI achieved was just 25.0.111, 113 In another example, cotton treated with a polyphosphazene derivative had an even lower LOI   119 of 23.5.97 Viscose fibers with high LOIs (up to 35.0) have been synthesized by blending a polyphosphazene into the viscose solution prior to spinning, however the wet spinning method is unique to viscose and isn’t adaptable to other systems.224  Another important advantage of this system is the simplicity of the two-step treatment. This is most apparently when compared to “layer-by-layer” methods, which also rely upon non-covalent interactions to affix flame retardant compounds to cellulosics. In a representative example, 20 bilayers (thus 40 layers in total) of chitosan and poly(vinylphosphonic acid) were necessary to impart appreciable flame retardancy to paper handsheets, as tested by a horizontal flame test.108 The system involving 6.1 and 6.3 also compares favourably to very recent work in which cotton fabric is first soaked in a flame retardant solution followed by pH-curing.115, 116   6.2.5 Synthesis of 6.1:6.3 Composite To gain additional insight into the nature of the interaction between 6.1 and 6.3, aqueous solutions of each were mixed to afford an immediate precipitate. The resultant 6.1:6.3 composite is completely insoluble in water and, therefore, was analysed by solid-state 13C{1H} CP/MAS NMR spectroscopy (Figure 6.3). The spectrum of the composite appears nearly identical to that of 6.3, however with a new resonance at 37 ppm. This was assigned to the methylene units of 6.1 based upon comparison to the solid-state 13C{1H} CP/MAS NMR spectrum of 6.1 alone. The 6.1:6.3 composite was also analysed by solid-state IR spectroscopy (Figure 6.4). When compared to 6.3 separately, the most distinctive features are new shoulders at 1187 cm-1 and 1090 cm-1, assigned to the P=N and C-N vibrations of 6.1. Lastly, the formulation of the precipitate as a mixture of the two components was supported by elemental analysis, which revealed a nitrogen content (11.16%) between that of 6.1 (29.91%) and 6.3 (0%). The nitrogen content is not exactly   120 at the midpoint between 6.1 and 6.3 due to the elimination of NaCl as well as possibly ethylenediamine.  Figure 6.3: 13C{1H} solid-state CP/MAS NMR spectra of the 6.1:6.3 composite (top), 6.1 (middle), and 6.3 (bottom). Asterisks denote spinning sidebands.    121  Figure 6.4: IR spectra of 6.3 (blue), 6.1 (red), and the 6.1:6.3 composite (green). The inset is an expanded region of the same spectra.  6.2.6 Synthesis of Phosphazene-Containing Ammonium-Carboxylate Model (6.5) In an effort to obtain more specific structural information regarding the possible ammonium-carboxylate interactions between 6.1 and 6.3, the model salt [N3P3(OPh)5(NHCH2CH2NH3)]+[O2CCy]– (6.5) was prepared from N3P3(OPh)5(NHCH2CH2NH2) (6.4) and cyclohexanecarboxylic acid (Scheme 6.2). Crystals suitable for X-ray analysis were obtained by slow evaporation of a DCM-hexanes solution of 6.5. The molecular structure of the model salt reveals a significant degree of H-bonding with each ammonium group being H-bonded to three different carboxylates whilst each secondary P–NHR is hydrogen bonded to one carboxylate (Figure 6.5). Additionally, each carboxylate is H-bonded to four different nitrogen atoms via the two oxygens. The donor-acceptor N–O distances for the ammonium-carboxylate H-bonds are in the range of 2.728(4)-2.787(3) Å, suggestive of moderate, mostly electrostatic character.225 The N–O distances for the secondary P-NHR-  122 carboxylate hydrogen bonds are slightly longer [2.998(6)-3.012(5) Å], but can still be considered to be of similar electrostatic character. As a consequence of the complexity of these interactions, the model salt crystallizes not as a discrete ion pair, but as a one-dimensional hydrogen-bonded network of phosphazene and carboxylate units along the x-axis (Figure 6.6). Give how these interactions occur even in a relatively simple ammonium-carboxylate salt points towards even more complex bonding in the 6.1:6.3 composite.  Scheme 6.2: Synthesis of an ammonium-carboxylate salt (6.5) between an aminophosphazene (6.4) and cyclohexanecarboxylic acid.  NPNPNPPhOHNOPhOPhPhOPhONH2OOHNPNPNPPhOHNOPhOPhPhOPhONH3OO6.4 6.5  123  Figure 6.5: Molecular structure of (N3P3(OPh)5(NHCH2CH2NH3)]+[O2CCy]– (6.5) (molecule 1 of 2 in the asymmetric unit) as a model for the 6.1:6.3 composite. The filled-in structure represents one of the two unique ion pairs within the unit cell. Each ammonium moiety forms three H-bonds depicted by one solid structure and two faded structures.    124  Figure 6.6: Molecular structure of [N3P3(OPh)5(NHCH2CH2NH3)]+[O2CCy]– (6.5), showing approximately three unit cells along the x-axis. Thermal ellipsoids shown at the 50% probability level. All hydrogen atoms and aryl carbon atoms are omitted for clarity. Dashed bonds show H-bonding interactions. Black = carbon, blue = nitrogen, red = phosphorus, cyan = oxygen.  6.2.7 SEM-EDS Analysis of Treated Paper Samples Treated and untreated paper samples were also analysed by scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDS) (Figure 6.7). As expected, untreated paper contains only carbon and oxygen whereas paper treated with 6.1 additionally shows phosphorus, chlorine, and nitrogen. For paper treated with 6.1 and 6.3 (15:15 wt%), smooth bright regions are observed that have a significantly higher P and Cl content as well as higher N and Na content when compared to darker regions. These bright regions are therefore due to the presence of the 6.1:6.3 composite. Most importantly, after leaching, samples containing this   125 composite retain the majority of their N and P content. However, the Na and Cl contents are considerably lower due to the ready leaching of the NaCl formed when 6.1 and 6.3 are mixed.   Figure 6.7: SEM images at 100x magnification of paper samples without any treatment (top left), unleached samples treated with 6.1 (15 wt%) (top right), unleached samples treated with 6.1:6.3 (15:15 wt%) (bottom left), and leached samples treated with 6.1:6.3 (15:15 wt%) (bottom right). Insets show representative EDS elemental composition analyses for each image.      126 6.3 Summary In conclusion, the polyfunctional molecular phosphazene derivative 6.1 can be rendered virtually non-leachable from paper by using carboxymethyl cellulose (6.3) as a “glue”. This exploitation of H-bonding represents a simple and unique method to easily immobilize a flame retardant molecule within a cellulosic material using traditional water-processing methods. This development opens the door to cost-effective, safe and environmentally friendly uses of renewable cellulosic materials such as paper, wood, cotton or other cellulose fibre-based products where flame retardancy and non-leachability are required.  6.4 Experimental 6.4.1 X-ray crystallography The single crystal of 6.5 was immersed in oil and mounted on a glass fiber. Data was collected on a Bruker X8 APEX II diffractometer with graphite-monochromated Mo Kα radiation. The structure was solved by direct methods and subsequent Fourier difference techniques. All non-hydrogen atoms were refined anisotropically with hydrogen atoms being included in calculated positions but not refined. The data set was corrected for absorption effects (SADABS), Lorentz, and polarization effects. All calculations were performed using SHELXL-2014 crystallographic software package from Bruker AXS. Additional crystal data and details of data collection and structure refinement are listed in Table 6.4.      127 Table 6.4: Summary of X-ray collection data for 6.5 Crystal 6.5 Formula C39H44N5O7P3 Formula Weight 787.70 Crystal System Triclinic Space Group P -1 Colour Colourless a (Å) 10.106(2) b (Å) 18.807(4) c (Å) 20.322(4) α (°) 90.631(4) β (°) 97.484(4) γ (°) 90.641(4) V (Å3) 3828.8(13) Z 4 T (K) 90(2) µ (Mo Κα) (mm-1) 0.212 Crystal Size (mm) 0.70 x 0.05 x 0.05 Dcalcd. (g cm-3) 1.366 2Θ (max) (°) 56.88 No. of Reflections 66884 No. of Unique Data 9827 Rint 0.0603 Reflections/parameters ratio 19.6 R1, wR2[I > 2σ(I)]a 0.0490 R1, wR2 (all data)b 0.1148 GOF 0.993 aR1 = Σ ||Fo| - |Fc|| / Σ|Fo|. bwR2 = [Σ (w(Fo2 – Fc2)2)/Σ w(Fo2)2]1/2    128 6.4.2 General Procedures  All manipulations of air-sensitive and/or water-sensitive compounds were performed using standard Schlenk or glovebox techniques under nitrogen atmosphere. Hexakis(2-aminoethyl)aminophosphazene (6.1) was synthesized according to the procedure in Chapter 4. N3P3(OPh)5NHCH2CH2NH2 (6.4) was synthesized according to the procedure in Chapter 5. Sodium carboxymethyl cellulose (Mw = 90 000 g mol-1, DS = 0.7) (6.3) and cyclohexanecarboxylic acid were purchased from Sigma-Aldrich and used as received. The paper used for all experiments was laboratory rectangle Dynamic Sheet Former (DSF) sheets (basis weight = 200 g m-2) made from thermomechanical pulp (TMP) produced in a pilot plant from black spruce chips with a total energy input of 2468 kWh t-1.  Solution NMR spectra were recorded on Bruker 300 MHz Avance or 400 MHz Avance spectrometers (Milton, ON) at room temperature. Chemical shifts for 31P spectra are reported relative to H3PO4 as an external standard (85% in H2O). The solid-state 13C{1H} NMR data were acquired on a Bruker 400 MHz Avance spectrometer running with xwinnmr 2.6. The MAS speed was set at 5 or 8 kHz. The contact time to establish cross polarization was set to be 2 ms. Total number of scans was around 18 000 with a recycle delay of 5 seconds.  The NMR data were apodized with a 10 Hz Lorentzian broadening function and zero filled once prior to Fourier transformation. Mass spectra were recorded using a Waters/Micromass LCT. IR spectra were recorded using a Perkin Elmer Frontier FT-IR with attenuated total reflectance (ATR). Scanning Electron Microscopy (SEM) was performed using a Hitachi S3000N variable-pressure SEM equipped with a Quartz X-One energy dispersive spectrometry (EDS) system. A 10 mm diameter sub-sample was cut out of treated paper samples and mounted on a 12 mm SEM stub covered   129 with spectroscopically-pure, double-sided carbon tape. Imaging was carried out in back-scattered electron mode.  Flame testing followed the procedure outlined in TAPPI Standard Method T461 cm-00, using an apparatus fabricated according to the specifications given in the standard method. A paper sample was suspended such that the lower edge would be 19 mm above the top of a Bunsen burner inside the testing chamber. The burner was lit, and the flame adjusted to a height of 40 mm. The flame was held in contact with the sample for 12 s and then withdrawn. After the burner was withdrawn, the flame time (time the sample continued to sustain a flame) and glow time (time the sample continued to have glowing embers) were measured.  The limiting oxygen index of paper samples were measured using a Govmark OI-1 Oxygen Index Module following American Society for Testing and Materials (ASTM) Standard Test Method D2863. Within the testing apparatus, the top of the sample was lit evenly using a natural gas burner. If the flame burned for longer than 180 s or if the flame traveled more than 80 mm before extinguishing, it was considered to be able to sustain a flame. Samples were repeatedly tested in this manner until a final LOI value was determined.   6.4.3 Treatment of Paper Samples  All paper samples were pre-conditioned at 23 °C and 50% humidity in a controlled temperature and humidity (CTH) room for 24 h before treatment. Representative example for analysis of paper treated with 6.1 and 6.3 (15:15 wt%) by TAPPI Standard Method T461 cm-00: A sheet of DSF paper (70 mm x 210 mm, 2.95 g) was coated on both sides with a solution of 6.1 (0.44 g) in water (10 mL), and allowed to dry overnight. The sample was then coated on one side with a solution of 6.3 (0.22 g) in water (5 mL). After four hours the sample was dry enough to   130 flip over, and an identical solution of 6.3 (0.22 g) in water (5 mL) was used to coat that side. The sample was dried for 48 hours in a CTH room before further testing. The final mass of the sample was slightly higher than the sum of the original sheet plus the masses of 6.1 and 6.3 used, presumably due to the retention of water within the H-bonded network of 6.1:6.3. For the preparation of samples for LOI analysis, samples were smaller (50 mm x 140 mm) but otherwise treated in the same manner.  6.4.4 Preparation of 6.1:6.3 Composite To a stirred solution of sodium carboxymethyl cellulose (6.3, 1.0 g, Mw = 90 000, DS = 0.7) in water (200 mL), a solution of 6.1 (1.0 g) in water (20 mL) was added, forming a white precipitate. The solid was filtered through a coarse frit, yielding a colourless gummy solid. This solid was sonicated in methanol followed by filtration and grinding with a mortar and pestle to yield a fine white powder. Yield: 0.9 g.  CP/MAS 13C{1H} NMR (101 MHz): δ = 177 (COO-), 104 (C1), 74 (C2-6, C2’), 37 (NH-CH2CH2-NH3). FT-IR (ATR) νmax (cm-1): 1031 (C-O), 1090 (C-N), 1187 (P=N), 1324 (CH2), 1417 (C=O), 1583 (C=O), 2900 (C-H), 3250 (O-H). Anal. found: C, 33.71%; N, 11.16%; H, 6.84%  6.4.5 Leaching of Paper Samples  The leaching of paper samples with room temperature, neutral pH water followed the procedure given in TAPPI Standard Method T461 cm-00. Coated paper samples (70 mm x 210 mm for TAPPI T461cm-00; 50 mm x 140 mm for LOI) were placed in a 2000 mL beaker covered with metal mesh. Deionized water was delivered through a glass tube passing through a   131 small hole in the mesh to the bottom of the beaker until the beaker was filled, thereby completely submerging the paper samples. The water flow was maintained for the desired length of time, at which time the paper samples were removed, blotted dry, and conditioned in a CTH room for 48 h before testing.  For samples leached using boiling water, LOI-sized (50 mm x 140 mm) paper samples were placed in a 1000 mL beaker filled with boiling distilled water and boiled for the desired length of time before drying. Samples were also leached in the same manner using room temperature solutions of strong acid or base at concentrations of 0.1 M HCl (pH = 1) or NaOH (pH = 13), representing an approximately 100x excess of H3O+ or OH- compared to the theoretical number ammonium-carboxylate pairs in each sample. Additionally, concentrations of 0.00001 M HCl (pH = 5) or NaOH (pH = 9) were used in the same manner, representing an approximately 100x excess of the ammonium-carboxylate pairs. For leaching tests done with 8 L of 0.1 M acid or base, large plastic buckets were used.  6.4.6 Preparation of [N3P3(OPh)5NH(CH2)2NH3][CyCOO] (6.5) To a stirred solution of pentaphenoxy(2-aminoethyl)aminophosphazene (6.4) (0.25 g, 0.379 mmol) in dichloromethane (1 mL), cyclohexanecarboxylic acid (49 mg, 0.379 mmol) was added. Removal of solvent in vacuo yielded the product as an off-white solid. Yield: 0.30 g, >99%.  1H NMR (400 MHz, CDCl3): δ = 8.04 (br s, 3H, NH3), 7.30-6.93 (m, 25H, Ar-H), 4.57 (br s, 1H, NH), 2.87 (m, 2H, PNHCH2), 2.70 (m, 2H, CH2NH3), 2.20 (tt, 1H, CHCOO), 1.94-1.17 (m, 10H, (CH2)5); 31P NMR (162 MHz, CDCl3): δ = 18.5 (tt, 2JPP = 74 Hz, 3JPH = 13 Hz, P(OPh)(NH)), 9.45 (d, 2JPP = 74 Hz, P(OPh)2); 13C{1H} NMR (101 MHz, CDCl3): δ = 183 (s,   132 COO), 151 (m), 129 (s), 125 (s), 121 (s), 45 (s, C-COO), 41 (d, 2JCP = 5 Hz, P-NH-CH2), 39 (s, CH2NH3), 30 (s, Cy-CH2), 26 (s, Cy-CH2), 25.8 (s, Cy-CH2); HRMS (ESI-TOF) m/z: [M]+ Calcd for C32H33N5O5P3 (cation) 660.1695; Found 660.1711; Anal. Calcd for C39H44N5O7P3: C, 59.47; H, 5.63; N, 8.89 Found: C, 59.41; H, 5.54; N, 8.59; FT-IR (ATR) νmax (cm-1): 3042, 2922, 2850, 1590, 1487, 1455, 1408, 1252, 1223, 1152, 1069, 1023, 1007.    133 Chapter 7: Summary and Future Work  7.1 Phosphorus-Containing Polymers as Flame Retardants While the use of phosphorus-containing polymers as non-leachable flame retardants remains popular, the reality of inorganic polymer synthesis can limit their application. In Chapter 2 the use of poly(methylenephosphine)s, polymers derived from phosphaalkenes, as non-leachable flame retardants for paper was reported. Although they were found to be moderately effective, the complex, multistep synthesis of these polymers makes them inconvenient. For this reason, attention should turn towards known phosphorus-containing polymers that can be synthesized in few steps. As discussed in Chapter 1, polyphosphazenes are already well established as flame retardants.45, 46 Polyphosphazenes are easily synthesized by thermal ring-opening polymerization of hexachlorophosphazene (7.1) to form poly(dichlorophosphazene) (7.2, Scheme 7.1). The properties of the resulting polyphosphazene can be tuned by nucleophilic substitution of 7.2 to yield substituted polyphosphazenes (7.3).   Scheme 7.1: Synthesis of polyphosphazenes (7.2-7.3) by ring-opening polymerization of hexachlorophosphazene (7.1) followed by nucleophilic substitution.  In Chapter 3, the microstructure of polymers resulting from the anionic polymerization of phosphaalkenes was confirmed to match that of polymers from radical initiation. Using this information along with data from model chemistry and kinetic experiments, an addition-NPNPNPCl ClClClClCl7.1P NClCl nP NNuNu n7.2 7.3250 °C Nu-  134 isomerization mechanism was postulated. Some questions remain, however, as to the possibility of alternative mechanisms during the polymerization. The presence of such mechanisms is hinted at by the observation that at high levels of conversion the reaction rate speeds up at lower temperatures, but slows down at higher temperatures.157 Whether this is indicative of simple addition or another mechanism is unknown, thus further experiments are needed. Lastly, linear phosphaalkene dimers have never been isolated, and are of great interest whether they represent the simple addition mechanism (7.4) or an addition-isomerization mechanism (7.5).   7.2 Phosphazene-Based Non-Leachable Flame Retardants Chapter 4 presented the synthesis of hexakis(2-aminoethyl)aminophosphazene (7.6), a highly functional phosphazene-based flame retardant for cellulosic materials. Setting aside the issue of its leachability which was addressed in later chapters, phosphazene 7.6 was found to be a highly effective flame retardant, thus in that respect it is a success. Future work should therefore not focus on trying to improve the flame retardancy of 7.6 by tweaking its structure, but instead on improving its synthesis and purification. While the current procedure is simple, the end product is contaminated with large amounts of ethylenediamine hydrochloride. Although this impurity does not appear to greatly hinder the flame retardancy of 7.6, conceivably its performance would only increase if it could be isolated in greater purity. P CHPh2BuPMesCHPh2MesP CBuPPhPh MesCHPh27.4 7.5  135   In Chapter 5, the carbodiimide coupling of 7.6 to carboxylate-functionalized wood pulp was attempted in order to render it non-leachable by direct covalent linkage. Although model chemistry indicated that this coupling was feasible, handsheets made from phosphazene-modified pulp was found to be only modestly flame retardant. In light of these results as well as those from the next chapter, this approach is perhaps not the most promising for future investigation. Despite this, in principle the flame retardant performance could be improved by increasing the carboxylate content of the BTMP such that more 7.6 can be coupled to the pulp. Additionally, as has already been discussed, if higher purity 7.6 can be obtained and used in the coupling, this too should improve the flame retardant properties of the resulting handsheets.  Lastly, Chapter 6 reported the use of 7.6 in combination with carboxymethyl cellulose as a two-part flame retardant coating for cellulosic materials which relies on both ionic and H-bonding interactions to render the flame retardant non-leachable. This system was found to be highly effective as a flame retardant, and its performance unexpectedly increased after leaching with water. Other than the previously discussed benefit of using more pure 7.6, the leaching performance of this system could be improved by modifying 7.6 to create a greater degree of H-bonding interactions between the two components. For example, the primary amines of 7.6 could be converted to guanidines in a simple one step procedure to yield a hexa(guanidino)phosphazene derivative (7.7, Scheme 7.2), effectively increasing the number of H-bond donors.  NPNPNPHNHNNHHNNHNHNH2H2NH2N NH2NH2H2N7.6  136  Scheme 7.2: Synthesis of a hexa(guanidino)phosphazene (7.7) derivative using S-methylisothiourea hemisulfate salt.  Altering the H-bond acceptor, for example by using more highly substituted carboxymethyl cellulose, can also influence the degree of H-bonding. From a logistical perspective, using carboxymethyl cellulose of a lower molecular weight would be more convenient due to higher solubility. Additionally, although the ability of carboxymethyl cellulose to bind to cellulose is thought to aid in the prevention of leaching, other potential polymers have not been explored. Common anionic polyelectrolytes include sodium polystyrene sulfonate (7.8) and sodium polyacrylate (7.9). Particularly intriguing is the potential use of phosphorus-containing anionic polyelectrolytes such as ammonium (7.10) or sodium (7.11) polyphosphate, or poly(vinylphosphonic acid) (7.12). With phosphorus in both components of the flame retardant system, in principle less material can be used to achieve the same level of flame retardancy.  NPNPNPHNHNNHNHNH7.7NH2NHHNNHNH2NHHNH2N NH HNH2NNHNPNPNPHNHNNHHNNHNHNH2H2NH2N NH2NH2H2N7.6H2N SMeNH• 1/2 H2SO4HNNHH2NHNNH2NHHN  137   Lastly, this thesis has given a purely academic analysis of the efficacy of several strategies for rendering phosphorus-based flame retardants non-leachable, and from that perspective this research has largely been successful. However, given the strongly application-driven goals of this project, the results of current or future research could be completely irrelevant if not commercially feasible. Moving forward, the continued success of this project therefore depends upon the amenability of these techniques to commercialization. This will largely depend on factors such as the cost of the syntheses involved. The lengthy syntheses of the phosphaalkene-derived polymers in Chapter 2 have already been alluded to, but the synthesis of 7.6 could also be an issue. Although phosphazene 7.6 is synthesized in a single step from hexachlorophosphazene (7.1), both the starting material (7.1) and the reaction solvent (THF) are moderately expensive. Future work in this project should focus on developing new flame retardant compounds structurally similar to 7.6, but synthesized from cheaper starting materials. For example, phosphoramide 7.14 retains the multiple amino groups of 7.6 but instead is synthesized from phosphorus(V) oxychloride (7.13), a much less expensive starting material than phosphazene 7.1 (Scheme 7.3). While the synthesis of 7.14 has been reported previously, its use as a flame retardant has only been investigated as a cross-linking agent for epoxides.226-228  SO3nO OnnOPOONH4nOPOONaPnHO OONaNaNa7.8 7.9 7.10 7.11 7.12  138  Scheme 7.3: Synthesis of a tris(amino)phosphoramide (7.14) from phosphorus(V) oxychloride (7.13)  PONHH2N NH NH2HNNH27.14POCl ClClH2N NH27.13  139 References   1 J. L. Meikle, 'American Plastic: A Cultural History', Rutgers University Press: New Brunswick, New Jersey, 1995. 2 S. Freinkel, 'Plastic: A Toxic Love Story', Houghton Mifflin Harcourt: New York, 2011. 3 R. Friedel, 'Pioneer Plastic: The Making and Selling of Celluloid', University of Wisconsin Press: Madison, Wisconsin, 1983. 4 R. B. Seymour and G. B. Kauffman, J. Chem. Educ., 1992, 69, 311. 5 A. Parkes, 1855, 'Manufacture of Elastic and Adhesive Compounds', British Patent 2359 6 F. Aftalion, 'A History of the International Chemical Industry', University of Pennsylvania Press: Philadelphia, 1991. 7 'This Film is Dangerous: A Celebration of Nitrate Film', Fédération Internationale des Archives du Film: Bruxelles, Belgium, 2002. 8 P. Guerra, M. Alaee, E. Eljarrat, and D. Barceló, 'Introduction to Brominated Flame Retardants: Commercially Products, Applications, and Physicochemical Properties', in 'Brominated Flame Retardants', ed. E. Eljarrat and D. Barceló, Springer-Verlag, Berlin, 2011. 9 M. Alaee, P. Arias, A. Sjodin, and A. Bergman, Environ. Int., 2003, 29, 683. 10 S. Wood, 'Facing the Problems of Fire', in Modern Plastics 45 (June 1968) 82 11 L. J. Carter, Science, 1976, 192, 240. 12 M. L. Terrell, K. P. Hartnett, H. Lim, J. Wirth, and M. Marcus, Chemosphere, 2015, 118, 178. 13 M. J. Prival, E. C. Mccoy, B. Gutter, and H. S. Rosenkranz, Science, 1977, 195, 76. 14 M. D. Gold, A. Blum, and B. N. Ames, Science, 1978, 200, 785. 15 A. Blum, M. D. Gold, B. N. Ames, C. Kenyon, F. R. Jones, E. A. Hett, R. C. Dougherty, E. C. Horning, I. Dzidic, D. I. Carroll, R. N. Stillwell, and J. P. Thenot, Science, 1978, 201, 1020. 16 H. M. Stapleton, S. Klosterhaus, S. Eagle, J. Fuh, J. D. Meeker, A. Blum, and T. F. Webster, Environ. Sci. Technol., 2009, 43, 7490. 17 H. M. Stapleton, S. Klosterhaus, A. Keller, P. L. Ferguson, S. van Bergen, E. Cooper, T. F. Webster, and A. Blum, Environ. Sci. Technol., 2011, 45, 5323. 18 H. M. Stapleton, S. Sharma, G. Getzinger, P. L. Ferguson, M. Gabriel, T. F. Webster, and A. Blum, Environ. Sci. Technol., 2012, 46, 13432. 19 C. C. Carignan, M. D. McClean, E. M. Cooper, D. J. Watkins, A. J. Fraser, W. Heiger-Bernays, H. M. Stapleton, and T. F. Webster, Environ. Int., 2013, 55, 56. 20 C. A. de Wit, Chemosphere, 2002, 46, 583. 21 C. A. de Wit, M. Alaee, and D. C. G. Muir, Chemosphere, 2006, 64, 209. 22 R. J. Law, C. R. Allchin, J. de Boer, A. Covaci, D. Herzke, P. Lepom, S. Morris, J. Tronczynski, and C. A. de Wit, Chemosphere, 2006, 64, 187. 23 M. Frederiksen, K. Vorkamp, M. Thomsen, and L. E. Knudsen, Int. J. Hyg. Environ. Health, 2009, 212, 109.   140 24 A. Covaci, S. Harrad, M. A. E. Abdallah, N. Ali, R. J. Law, D. Herzke, and C. A. de Wit, Environ. Int., 2011, 37, 532. 25 G. Stieger, M. Scheringer, C. A. Ng, and K. Hungerbuhler, Chemosphere, 2014, 116, 118. 26 T. R. Hull and A. A. Stec, 'Polymers and Fire', in 'Fire Retardancy of Polymers: New Strategies and Mechanisms', ed. T. R. Hull and B. K. Kandola, Royal Society of Chemistry, Cambridge, UK, 2009. 27 A. P. Mouritz and A. G. Gibson, 'Fire Properties of Polymer Composite Materials', Springer: Dordrecht, The Netherlands, 2006. 28 D. Price, G. Anthony, and P. Carty, 'Polymer Combustion, Condensed Phase Pyrolysis and Smoke Formation', in 'Fire Retardant Materials', ed. A. R. Horrocks and D. Price, CRC Press, Boca Raton, Florida, 2001. 29 M. Lewin and E. D. Weil, 'Mechanisms and Modes of Action in Flame Retardancy of Polymers', in 'Fire Retardant Materials', ed. A. R. Horrocks and D. Price, CRC Press, Boca Raton, Florida, 2001. 30 B. Schartel, Materials, 2010, 3, 4710. 31 S. V. Levchik and E. D. Weil, J. Fire Sci., 2006, 24, 345. 32 I. van der Veen and J. de Boer, Chemosphere, 2012, 88, 1119. 33 G. C. Tesoro, S. B. Sello, and J. J. Willard, Text. Res. J., 1968, 38, 245. 34 G. C. Tesoro, S. B. Sello, and J. J. Willard, Text. Res. J., 1969, 39, 180. 35 H. B. Pandya and M. M. Bhagwat, Text. Res. J., 1981, 51, 5. 36 D. Bakos, M. Kosik, K. Antos, M. Karolyova, and I. Vyskocil, Fire Mater., 1982, 6, 10. 37 C. Sivriev and L. Zabski, Eur. Polym. J., 1994, 30, 509. 38 H. R. Allcock and J. P. Taylor, Polym. Eng. Sci., 2000, 40, 1177. 39 X. Zhang, Y. Zhong, and Z. P. Mao, Polym. Degrad. Stabil., 2012, 97, 1504. 40 B. Kabisch, U. Fehrenbacher, and E. Kroke, Fire Mater., 2014, 38, 462. 41 Y. Qiu, L. J. Qian, W. Xi, and X. X. Liu, J. Appl. Polym. Sci., 2016, 133. 42 W. Xi, L. J. Qian, Y. Qiu, and Y. J. Chen, Polymer. Adv. Tech., 2016, 27, 781. 43 J. Zhan, L. Song, S. B. Nie, and Y. A. Hu, Polym. Degrad. Stabil., 2009, 94, 291. 44 R. Sonnier, L. Ferry, and J.-M. Lopez-Cuesta, 'Flame Retardancy of Phosphorus-Containing Polymers', in 'Phosphorus-Based Polymers: From Synthesis to Applications', ed. S. Monge and G. David, Royal Society of Chemistry, Cambridge, UK, 2014. 45 C. W. Allen, J. Fire Sci., 1993, 11, 320. 46 H. R. Allcock, 'Chemistry and applications of polyphosphazenes', Wiley-Interscience: Hoboken, 2003. 47 S. T. Fei and H. R. Allcock, J. Power Sources, 2010, 195, 2082. 48 C. H. Tsao, M. Ueda, and P. L. Kuo, J. Polym. Sci. A Polym. Chem., 2016, 54, 352. 49 X. W. Mu, B. H. Yuan, W. Z. Hu, S. L. Qiu, L. Song, and Y. Hu, RSC Adv., 2015, 5, 76068. 50 K. Tao, J. Li, L. Xu, X. L. Zhao, L. X. Xue, X. Y. Fan, and Q. Yan, Polym. Degrad. Stabil., 2011, 96, 1248. 51 G. Camino, L. Costa, and L. Trossarelli, Polym. Degrad. Stabil., 1984, 6, 243. 52 G. Camino, L. Costa, and L. Trossarelli, Polym. Degrad. Stabil., 1984, 7, 25. 53 S. Bourbigot, M. Lebras, and R. Delobel, Carbon, 1993, 31, 1219. 54 J. Alongi, M. Poskovic, A. Frache, and F. Trotta, Polym. Degrad. Stabil., 2010, 95, 2093.   141 55 H. F. Wang and B. Li, Polymer. Adv. Tech., 2010, 21, 691. 56 J. X. Feng, S. P. Su, and J. Zhu, Polymer. Adv. Tech., 2011, 22, 1115. 57 C. H. Ke, J. Li, K. Y. Fang, Q. L. Zhu, J. Zhu, Q. Yan, and Y. Z. Wang, Polym. Degrad. Stabil., 2010, 95, 763. 58 C. M. Feng, M. Y. Liang, J. L. Jiang, J. G. Huang, and H. B. Liu, Polymer. Adv. Tech., 2016, 27, 693. 59 R. C. Zhang, S. M. Hong, and C. M. Koo, J. Appl. Polym. Sci., 2014, 131. 60 J. Sun, X. Y. Gu, M. Coquelle, S. Bourbigot, S. Duquesne, M. Casetta, and S. Zhang, Polymer. Adv. Tech., 2014, 25, 1552. 61 U. Braun, B. Schartel, M. A. Fichera, and C. Jager, Polym. Degrad. Stabil., 2007, 92, 1528. 62 T. Orhan, N. A. Isitman, J. Hacaloglu, and C. Kaynak, Polym. Degrad. Stabil., 2011, 96, 1780. 63 P. Muller and B. Schartel, J. Appl. Polym. Sci., 2016, 133. 64 F. Samyn and S. Bourbigot, Polym. Degrad. Stabil., 2012, 97, 2217. 65 A. D. Naik, G. Fontaine, F. Samyn, X. Delva, Y. Bourgeois, and S. Bourbigot, Polym. Degrad. Stabil., 2013, 98, 2653. 66 A. Ramani and A. E. Dahoe, Polym. Degrad. Stabil., 2014, 105, 1. 67 A. D. Naik, G. Fontaine, F. Samyn, X. Delva, J. Louisy, S. Bellayer, Y. Bourgeois, and S. Bourbigot, RSC Adv., 2014, 4, 18406. 68 Y. L. Chang, Y. Z. Wang, D. M. Ban, B. Yang, and G. M. Zhao, Macromol. Mater. Eng., 2004, 289, 703. 69 N. N. Li, G. W. Jiang, and G. Y. Zhou, Polym. Degrad. Stabil., 2015, 122, 161. 70 N. N. Li, G. W. Jiang, and G. Y. Zhou, RSC Adv., 2016, 6, 2512. 71 D. Lanzinger, S. Salzinger, B. S. Soller, and B. Rieger, Ind. Eng. Chem. Res., 2015, 54, 1703. 72 J. Sun, X. D. Wang, and D. Z. Wu, ACS Appl. Mater. Interfaces, 2012, 4, 4047. 73 J. Liu, J. Y. Tang, X. D. Wang, and D. Z. Wu, RSC Adv., 2012, 2, 5789. 74 Y. W. Bai, X. D. Wang, and D. Z. Wu, Ind. Eng. Chem. Res., 2012, 51, 15064. 75 G. R. Xu, M. J. Xu, and B. Li, Polym. Degrad. Stabil., 2014, 109, 240. 76 H. Liu, X. D. Wang, and D. Z. Wu, Polym. Degrad. Stabil., 2014, 103, 96. 77 D. Mathew, C. P. R. Nair, and K. N. Ninan, Polym. Int., 2000, 49, 48. 78 J. Sun, Z. Y. Yu, X. D. Wang, and D. Z. Wu, ACS Sustain. Chem. Eng., 2014, 2, 231. 79 H. Liu, X. D. Wang, and D. Z. Wu, Thermochim. Acta, 2015, 607, 60. 80 C. S. Reed, J. P. Taylor, K. S. Guigley, M. M. Coleman, and H. R. Allcock, Polym. Eng. Sci., 2000, 40, 465. 81 H. Liu, X. D. Wang, and D. Z. Wu, Polym. Degrad. Stabil., 2015, 118, 45. 82 T. Ranganathan, J. Zilberman, R. J. Farris, E. B. Coughlin, and T. Emrick, Macromolecules, 2006, 39, 5974. 83 T. Ranganathan, B. C. Ku, J. Zilberman, M. Beaulieu, R. J. Farris, E. B. Coughlin, and T. Emrick, J. Polym. Sci. A Polym. Chem., 2007, 45, 4573. 84 T. Ranganathan, M. Beaulieu, J. Zilberman, K. D. Smith, P. R. Westmoreland, R. J. Farris, E. B. Coughlin, and T. Emrick, Polym. Degrad. Stabil., 2008, 93, 1059. 85 H. Y. Ding, C. L. Xia, J. F. Wang, C. P. Wang, and F. X. Chu, J Mater Sci, 2016, 51, 5008.   142 86 H. L. Wang, S. P. Xu, and W. F. Shi, Prog. Org. Coat., 2009, 65, 417. 87 X. F. Wang, J. Zhan, W. Y. Xing, X. Wang, L. Song, X. D. Qian, B. Yu, and Y. Hu, Ind. Eng. Chem. Res., 2013, 52, 5548. 88 X. F. Wang, B. B. Wang, W. Y. Xing, G. Tang, J. Zhan, W. Yang, L. Song, and Y. Hu, Prog. Org. Coat., 2014, 77, 94. 89 L. P. Dong, C. Deng, R. M. Li, Z. J. Cao, L. Lin, L. Chen, and Y. Z. Wang, RSC Adv., 2016, 6, 30436. 90 J. S. Lin, L. Chen, Y. Liu, and Y. Z. Wang, J. Appl. Polym. Sci., 2012, 125, 3517. 91 S. Karpagam and S. Guhanathan, J. Appl. Polym. Sci., 2013, 129, 2046. 92 S. Sauca, M. Giamberini, and J. A. Reina, Polym. Degrad. Stabil., 2013, 98, 453. 93 L. Liu, Y. S. Liu, Y. Liu, and Q. Wang, RSC Adv., 2016, 6, 35051. 94 K. Kishore and K. Mohandas, Fire Mater., 1982, 6, 54. 95 B. K. Kandola, A. R. Horrocks, D. Price, and G. V. Coleman, J. Macromol. Sci., Rev. Macromol. Chem. Phys., 1996, C36, 721. 96 N. Illy, M. Fache, R. Menard, C. Negrell, S. Caillol, and G. David, Polym. Chem., 2015, 6, 6257. 97 T. Mayer-Gall, D. Knittel, J. S. Gutmann, and K. Opwis, ACS Appl. Mater. Interfaces, 2015, 7, 9349. 98 M. J. Kim, I. Y. Jean, J. M. Seo, L. M. Dai, and J. B. Baek, ACS Nano, 2014, 8, 2820. 99 R. K. Iler, J. Colloid Interface Sci., 1966, 21, 569. 100 G. Decher, Science, 1997, 277, 1232. 101 Y. C. Li, S. Mannen, A. B. Morgan, S. C. Chang, Y. H. Yang, B. Condon, and J. C. Grunlan, Adv. Mater., 2011, 23, 3926. 102 F. Carosio, J. Alongi, and G. Malucelli, Carbohyd. Polym., 2012, 88, 1460. 103 J. Alongi, F. Carosio, and G. Malucelli, Polym. Degrad. Stabil., 2012, 97, 1644. 104 J. Alongi, F. Carosio, and G. Malucelli, Cellulose, 2012, 19, 1041. 105 F. Carosio, J. Alongi, and G. Malucelli, Polym. Degrad. Stabil., 2013, 98, 1626. 106 M. Leistner, M. Haile, S. Rohmer, A. Abu-Odeh, and J. C. Grunlan, Polym. Degrad. Stabil., 2015, 122, 1. 107 M. Leistner, A. A. Abu-Odeh, S. C. Rohmer, and J. C. Grunlan, Carbohyd. Polym., 2015, 130, 227. 108 O. Koklukaya, F. Carosio, J. C. Grunlan, and L. Wagberg, ACS Appl. Mater. Interfaces, 2015, 7, 23750. 109 F. Carosio, G. Fontaine, J. Alongi, and S. Bourbigot, ACS Appl. Mater. Interfaces, 2015, 7, 12158. 110 F. Carosio, C. Negrell-Guirao, A. Di Blasio, J. Alongi, G. David, and G. Camino, Carbohyd. Polym., 2015, 115, 752. 111 F. Carosio, A. Di Blasio, J. Alongi, and G. Malucelli, Polymer, 2013, 54, 5148. 112 G. Laufer, C. Kirkland, A. B. Morgan, and J. C. Grunlan, Biomacromolecules, 2012, 13, 2843. 113 F. Carosio and J. Alongi, RSC Adv., 2015, 5, 71482. 114 A. A. Cain, S. Murray, K. M. Holder, C. R. Nolen, and J. C. Grunlan, Macromol. Mater. Eng., 2014, 299, 1180. 115 M. Haile, C. Fincher, S. Fomete, and J. C. Grunlan, Polym. Degrad. Stabil., 2015, 114, 60.   143 116 M. Haile, M. Leistner, O. Sarwar, C. M. Toler, R. Henderson, and J. C. Grunlan, RSC Adv., 2016, 6, 33998. 117 D. Lea, in 'Cellulose:  Building Insulation with High Recovered Content, Low Embodied Energy', Gainesville, FL, 1996. 118 J. E. Stephenson, Paper Technology and Industry, 1985, 27. 119 J. Green, J. Fire Sci., 1996, 14, 353. 120 F. Laoutid, L. Bonnaud, M. Alexandre, J. M. Lopez-Cuesta, and P. Dubois, Mater. Sci. Eng. R-Rep., 2009, 63, 100. 121 M. Lewin, Polym. Degrad. Stabil., 2005, 88, 13. 122 A. R. Horrocks, Polym. Degrad. Stabil., 2011, 96, 377. 123 S. Y. Lu and I. Hamerton, Prog. Polym. Sci., 2002, 27, 1661. 124 J. I. Bates, J. Dugal-Tessier, and D. P. Gates, Dalton Trans., 2010, 39, 3151. 125 P. W. Siu, S. C. Serin, I. Krummenacher, T. W. Hey, and D. P. Gates, Angew. Chem. Int. Edit., 2013, 52, 6967. 126 K. J. T. Noonan, B. H. Gillon, V. Cappello, and D. P. Gates, J. Am. Chem. Soc., 2008, 130, 12876. 127 K. J. T. Noonan and D. P. Gates, Angew. Chem. Int. Edit., 2006, 45, 7271. 128 C. W. Tsang, B. Baharloo, D. Riendl, M. Yam, and D. P. Gates, Angew. Chem. Int. Edit., 2004, 43, 5682. 129 C. W. Tsang, M. Yam, and D. P. Gates, J. Am. Chem. Soc., 2003, 125, 1480. 130 K. B. Dillon, F. Mathey, and J. F. Nixon, 'Phosphorus: the carbon copy', Wiley: New York, 1998. 131 F. Mathey, Angew. Chem. Int. Edit., 2003, 42, 1578. 132 A. M. Priegert, B. W. Rawe, S. C. Serin, and D. P. Gates, Chem. Soc. Rev., 2016, 45, 922. 133 R. Amiri, G. Desilveira, and J. R. Wood, J. Pulp Pap. Sci., 1993, 19, J26. 134 L. A. Lowden and T. R. Hull, Fire Sci. Rev., 2013, 2, 1. 135 J. J. Pitts, 'Inorganic flame retardants and their mode of action', in 'Flame retardancy of polymeric materials', ed. W. C. Kuryla and A. J. Papa, Marcel Dekker, Inc., New York, 1973. 136 F. Shafizadeh, R. H. Furneaux, T. G. Cochran, J. P. Scholl, and Y. Sakai, J. Appl. Polym. Sci., 1979, 23, 3525. 137 Y. Sekiguchi and F. Shafizadeh, J. Appl. Polym. Sci., 1984, 29, 1267. 138 S. Soares, G. Camino, and S. Levchik, Polym. Degrad. Stabil., 1995, 49, 275. 139 J. B. Dahiya and S. Rana, Polym. Int., 2004, 53, 995. 140 B. K. Kandola and A. R. Horrocks, Polym. Degrad. Stabil., 1996, 54, 289. 141 J. Green, J. Fire Sci., 1996, 14, 426. 142 K. H. Pawlowski and B. Schartel, Polym. Int., 2007, 56, 1404. 143 M. V. Ramiah, J. Appl. Polym. Sci., 1970, 14, 1323. 144 M. Brebu and C. Vasile, Cell. Chem. Technol., 2010, 44, 353. 145 R. Chandra, T. Q. Hu, B. R. James, M. B. Ezhova, and D. V. Moiseev, J. Pulp Pap. Sci., 2007, 33, 15. 146 T. Q. Hu, E. Yu, B. R. James, and P. Marcazzan, Holzforschung, 2008, 62, 389. 147 Q. L. Li, X. L. Wang, D. Y. Wang, W. C. Xiong, G. H. Zhong, and Y. Z. Wang, J. Appl. Polym. Sci., 2010, 117, 3066.   144 148 Z. Y. Yang, B. Fei, X. W. Wang, and J. H. Xin, Fire Mater., 2012, 36, 31. 149 G. Becker, W. Uhl, and H. J. Wessely, Z. Anorg. Allg. Chem., 1981, 479, 41. 150 M. Yam, J. H. Chong, C. W. Tsang, B. O. Patrick, A. E. Lam, and D. P. Gates, Inorg. Chem., 2006, 45, 5225. 151 G. G. Odian, 'Principles of Polymerization', Wiley-Interscience: Hoboken, NJ, 2004. 152 For leading references, see: a) D. S. Breslow, G. E. Hulse, and A. S. Matlack, J. Am. Chem. Soc., 1957, 79, 3760. b) K. Yokota, M. Shimizu, Y. Yamashita, and Y. Ishii, Makromolekulare Chem., 1964, 77, 1. c) H. Wexler, Makromolekulare Chem., 1968, 115, 262. d) M. Guaita, G. Camino, and L. Trossarelli, Makromolekulare Chem., 1970, 131, 309. e) J. P. Kennedy and T. Otsu, J. Macromol. Sci.-Revs. Macromol. Chem., 1972, C6, 237. f) Y. Murakami, T. Suzuki, and Y. Takegami, Polym. J., 1985, 17, 855. g) T. Iwamura, I. Tomita, M. Suzuki, and T. Endo, J. Polym. Sci. A Polym. Chem., 1998, 36, 1491. h) T. Iwamura, I. Tomita, M. Suzuki, and T. Endo, J. Polym. Sci. A Polym. Chem., 1999, 37, 465. i) T. Iwamura, I. Tomita, M. Suzuki, and T. Endo, React. Funct. Polym., 1999, 40, 115. j) T. Iwamura, I. Tomita, M. Suzuki, and T. Endo, J. Polym. Chem. A Polym. Chem. 2000, 38, 430. k) T. Iwamura, K. Adachi, and Y. Chujo, Polymer, 2016, 92, 13. 153 a) R. Asami, J. Oku, M. Takeuchi, K. Nakamura, and M. Takaki, Polym. J., 1988, 20, 699. b) J. Oku, T. Hasegawa, T. Kawakita, Y. Kondo, and M. Takaki, Macromolecules, 1991, 24, 1253. c) J. Oku, T. Hasegawa, K. Nakamura, M. Takeuchi, M. Takaki, and R. Asami, Polym. J., 1991, 23, 195. d) J. Oku, T. Hasegawa, T. Takeuchi, and M. Takaki, Polym. J., 1991, 23, 1377. e) J. Oku, T. Hasegawa, Y. Kubota, M. Takaki, and R. Asami, Polym. Bull., 1992, 28, 505. f) J. Oku, M. Takeuchi, A. Saito, and R. Asami, Polym. J., 1992, 24, 1409. 154 a) Y. D. Gan, S. Prakash, G. A. Olah, W. P. Weber, and T. E. Hogen-Esch, Macromolecules, 1996, 29, 8285. b) T. Ganicz, W. A. Stanczyk, N. K. Gladkova, and I. Sledzinska, Macromolecules, 2000, 33, 289. c) J. X. Yang, S. C. Liu, F. H. Zhu, Y. W. Huang, B. Li, and L. Zhang, J. Polym. Sci. A Polym. Chem., 2011, 49, 381. 155 For recent reviews, see:  a) A. M. Priegert, B. W. Rawe, S. C. Serin, D. P. Gates, Chem. Soc. Rev., 2016, 45, 922. b) H. R. Allcock, Dalton Trans., 2016, 45, 1856. c) J. W. Zhou, G. R. Whittell, I. Manners, Macromolecules, 2014, 47, 3529. d) X. M. He, T. Baumgartner, RSC Adv., 2013, 3, 11334. e) A. A. Jahnke, D. S. Seferos, Macromol. Rapid Commun., 2011, 32, 943. f) F. Jakle, Chem. Rev., 2010, 110, 3985. g) T. Baumgartner, R. Reau, Chem. Rev., 2006, 106, 4681. 156 For recent reports of phosphorus polymer synthesis, see: a) C. T. Womble, G. W. Coates, K. Matyjaszewski, and K. J. T. Noonan, ACS Macro Lett., 2016, 5, 253. b) B. W. Rawe and D. P. Gates, Angew. Chem. Int. Ed., 2015, 54, 11438. c) C. Marquardt, T. Jurca, K. C. Schwan, A. Stauber, A. V. Virovets, G. R. Whittell, I. Manners, and M. Scheer, Angew. Chem. Int. Ed., 2015, 54, 13782. d) A. Schäfer, T. Jurca, J. Turner, J. R. Vance, K. Lee, V. A. Du, M. F. Haddow, G. R. Whittell, and I. Manners, Angew. Chem. Int. Ed., 2015, 54, 4836. e) R. Guterman, A. R. Kenaree, J. B. Gilroy, E. R. Gillies, and P. J. Ragogna, Chem. Mater., 2015, 27, 1412. f) T. Wolf, T. Steinbach, and F. R. Wurm, Macromolecules, 2015, 48, 3853. g) Z. C. Tian, C. Chen, and H. R. Allcock, Macromolecules, 2014, 47, 1065. h) Y. Matano, H. Ohkubo, Y. Honsho, A. Saito, S. Seki, and H. Imahori, Org. Lett., 2013, 15, 932. i) X. Wang, K. Cao, Y. Liu, B. Tsang,   145 and S. Liew, J. Am. Chem. Soc., 2013, 135, 3399. j) X. He, A. Y. Y. Woo, J. Borau-Garcia, and T. Baumgartner, Chem. Eur. J., 2013, 19, 7620. k) S. K. Patra, G. R. Whittell, S. Nagiah, C.-L. Ho, W.-Y. Wong, and I. Manners, Chem. Eur. J., 2010, 16, 3240. l) A. Saito, Y. Matano, and H. Imahori, Org. Lett., 2010, 12, 2675. m) S. Greenberg, G. L. Gibson and D. W. Stephan, Chem. Commun., 2009, 304. n) V. L. de Talance, M. Hissler, L. Z. Zhang, T. Karpati, L. Nyulaszi, D. Caras-Quintero, P. Bauerle, and R. Reau, Chem. Commun., 2008, 2200. o) K. Naka, T. Umeyama, A. Nakahashi, and Y. Chujo, Macromolecules, 2007, 40, 4854. p) L. A. Vanderark, T. J. Clark, E. Rivard, and I. Manners, Chem. Commun., 2006, 3332. q) V. A. Wright, B. O. Patrick, C. Schneider, and D. P. Gates, J. Am. Chem. Soc., 2006, 128, 8836. r) R. C. Smith and J. D. Protasiewicz, J. Am. Chem. Soc., 2004, 126, 2268. 157 K. J. T. Noonan and D. P. Gates, Macromolecules, 2008, 41, 1961. 158 B. W. Rawe, C. P. Chun, and D. P. Gates, Chem. Sci., 2014, 5, 4928. 159 S. C. Serin, G. R. Dake, and D. P. Gates, Dalton Trans., 2016, 45, 5659. 160 S. C. Serin, G. R. Dake, and D. P. Gates, Macromolecules, 2016, 49, 4067. 161 D. N. Bhattacharyya, J. Smid, and M. Szwarc, J. Polym. Sci. A Polym. Chem., 1965, 3, 3099. 162 B. H. Gillon, K. J. T. Noonan, B. Feldscher, J. M. Wissenz, Z. M. Kam, T. Hsieh, J. J. Kingsley, J. I. Bates, and D. P. Gates, Can. J. Chem., 2007, 85, 1045. 163 T. C. Klebach, R. Lourens, and F. Bickelhaupt, J. Am. Chem. Soc., 1978, 100, 4886. 164 T. A. Vanderknaap, T. C. Klebach, F. Visser, F. Bickelhaupt, P. Ros, E. J. Baerends, C. H. Stam, and M. Konijn, Tetrahedron, 1984, 40, 765. 165 A. A. Zavitsas, J. Phys. Chem. A, 2003, 107, 897. 166 P. R. Schreiner, L. V. Chernish, P. A. Gunchenko, E. Y. Tikhonchuk, H. Hausmann, M. Serafin, S. Schlecht, J. E. P. Dahl, R. M. K. Carlson, and A. A. Fokin, Nature, 2011, 477, 308. 167 T. S. Chen, Wolinska.J, and L. C. Leitch, J. Labelled Comp., 1970, 6, 285. 168 G. Becker, O. Mundt, M. Rossler, and E. Schneider, Z. Anorg. Allg. Chem., 1978, 443, 42. 169 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, in 'Gaussian 09, Revision D.01', Wallingford, CT, USA, 2009. 170 A. D. Becke, Phys. Rev. A, 1988, 38, 3098. 171 A. D. Becke, J. Chem. Phys., 1993, 98, 5648. 172 J. Tomasi, B. Mennucci, and R. Cammi, Chem. Rev., 2005, 105, 2999. 173 T. Yanai, D. P. Tew, and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51.   146 174 A. M. Priegert, P. W. Siu, T. Q. Hu, and D. P. Gates, Fire Mater., 2015, 39, 647. 175 C. W. Allen, Chem. Rev., 1991, 91, 119. 176 M. Becke-Goehring and B. Boppel, Z. Anorg. Allg. Chem., 1963, 322, 239. 177 S. S. Krishnamurthy, K. Ramachandran, and A. R. Vasudevamurthy, Inorg. Nucl. Chem. Lett., 1977, 13, 407. 178 S. S. Krishnamurthy, K. Ramachandran, A. R. V. Murthy, R. A. Shaw, and M. Woods, J. Chem. Soc. Dalton, 1980, 840. 179 R. A. Shaw, Phosphorus Sulfur Silicon Relat. Elem., 1989, 45, 103. 180 H. R. Allcock, P. E. Austin, and T. F. Rakowsky, Macromolecules, 1981, 14, 1622. 181 J. Barbera, J. Jimenez, A. Laguna, L. Oriol, S. Perez, and J. L. Serrano, Chem. Mater., 2006, 18, 5437. 182 F. Sournies, L. Labrousse, M. Graffeuil, F. Crasnier, J. P. Faucher, M. C. Labarre, and J. F. Labarre, Phosphorus Sulfur Silicon Relat. Elem., 1994, 89, 47. 183 A. F. Nikolaev and E. T. Wan, J. Gen. Chem. USSR., 1964, 34, 1843. 184 X. A. Ton, B. T. S. Bui, M. Resmini, P. Bonomi, I. Dika, O. Soppera, and K. Haupt, Angew. Chem. Int. Edit., 2013, 52, 8317. 185 H. R. Allcock and A. G. Scopelianos, Macromolecules, 1983, 16, 715. 186 H. R. Allcock and S. R. Pucher, Macromolecules, 1991, 24, 23. 187 F. F. Stewart, M. K. Harrup, R. P. Lash, and M. N. Tsang, Polym. Int., 2000, 49, 57. 188 X. Huang, X. J. Huang, A. G. Yu, C. Wang, Z. W. Dai, and Z. K. Xu, Macromol. Chem. Phys., 2011, 212, 272. 189 N. Ren, X. J. Huang, X. Huang, Y. C. Qian, C. Wang, and Z. K. Xu, J. Polym. Sci. A Polym. Chem., 2012, 50, 3149. 190 Y. C. Qian, X. J. Huang, C. Chen, N. Ren, X. Huang, and Z. K. Xu, J. Polym. Sci. A Polym. Chem., 2012, 50, 5170. 191 L. Abbassi, Y. M. Chabre, N. Kottari, A. A. Arnold, S. Andre, J. Josserand, H. J. Gabius, and R. Roy, Polym. Chem., 2015, 6, 7666. 192 E. Valeur and M. Bradley, Chem. Soc. Rev., 2009, 38, 606. 193 A. El-Faham and F. Albericio, Chem. Rev., 2011, 111, 6557. 194 J. C. Sheehan and G. P. Hess, J. Am. Chem. Soc., 1955, 77, 1067. 195 J. C. Sheehan, G. L. Boshart, and P. A. Cruickshank, J. Org. Chem., 1961, 26, 2525. 196 H. Ulrich and A. A. Sayigh, Angew. Chem., 1962, 74, 900. 197 O. A. Attanasi, S. Bartoccini, G. Favi, P. Filippone, F. R. Perrulli, and S. Santeusanio, J. Org. Chem., 2012, 77, 9338. 198 R. Alen, 'Papermaking Chemistry', Finnish Paper Engineers' Association: Jyvaskyla, Finland, 2010. 199 T. Ishii and K. Shimizu, 'Wood and Cellulosic Chemistry', ed. D. N.-S. Hon and N. Shiraishi, Marcel Dekker, New York, 2001. 200 T. Q. Hu, T. Williams, I. I. Pikulik, and J. A. Schmidt, J. Pulp Pap. Sci., 2005, 31, 109. 201 T. Q. Hu and A. Hayek, J-For, 2013, 3, 34. 202 L. F. Audrieth, R. Steinman, and A. D. F. Toy, Chem. Rev., 1943, 32, 109. 203 R. A. Shaw, B. C. Smith, and B. W. Fitzsimmons, Chem. Rev., 1962, 62, 247. 204 B. W. Fitzsimmons and R. A. Shaw, J. Chem. Soc., 1964, 1735. 205 D. Dell, B. W. Fitzsimmons, and R. A. Shaw, J. Chem. Soc., 1965, 4070. 206 E. T. Mcbee, K. Okuhara, and C. J. Morton, Inorg. Chem., 1966, 5, 450.   147 207 H. R. Allcock, S. E. Kuharcik, K. B. Visscher, and D. C. Ngo, J. Chem. Soc. Dalton, 1995, 2785. 208 U. Diefenbach, P. Adamaszek, M. Bloy, M. Kretschmann, and S. Scholz, Z. Anorg. Allg. Chem., 1998, 624, 1679. 209 M. Bloy, M. Kretschmann, S. Scholz, M. Teichert, and U. Diefenbach, Z. Anorg. Allg. Chem., 2000, 626, 1946. 210 V. Chandrasekhar, G. T. S. Andavan, S. Nagendran, V. Krishnan, R. Azhakar, and R. J. Butcher, Organometallics, 2003, 22, 976. 211 C. d'Errico, J. O. Jorgensen, K. B. R. M. Krogh, N. Spodsberg, R. Madsen, and R. N. Monrad, Biotechnol. Bioeng., 2015, 112, 914. 212 K. C. Li and R. F. Helm, Carbohyd. Res., 1995, 273, 249. 213 Y. J. Fu, M. H. Qin, Y. Z. Guo, Q. H. Xu, Z. Q. Li, N. Liu, Z. W. Yuan, and Y. Gao, Wood. Sci. Technol., 2013, 47, 557. 214 D. Zecher and R. van Coillie, 'Cellulose derivatives', in 'Thickening and Gelling Agents for Food', ed. A. Imeson, Blackie Academic & Professional, Glasgow, UK, 1992. 215 R. L. Feddersen and S. N. Thorp, 'Sodium Carboxymethylcellulose', in 'Industrial Gums: Polysaccharides and Their Derivatives', ed. R. L. Whistler and J. N. BeMiller, Academic Press, Inc., San Diego, California, 1993. 216 T. Wüstenberg, 'Cellulose and Cellulose Derivatives in the Food Industry: Fundamentals and Applications', Wiley-VCH: Weinheim, Germany, 2015. 217 T. Heinze and T. Liebert, Prog. Polym. Sci., 2001, 26, 1689. 218 J. Laine, T. Lindström, G. G. Nordmark, and G. Risinger, Nord. Pulp. Pap. Res. J., 2000, 15, 520. 219 J. Laine, T. Lindström, G. G. Nordmark, and G. Risinger, Nord. Pulp. Pap. Res. J., 2002, 17, 50. 220 M. Blomstedt and T. Vuorinen, J. Wood. Sci., 2007, 53, 223. 221 P. Eronen, K. Junka, J. Laine, and M. Osterberg, Bioresources, 2011, 6, 4200. 222 Z. L. Liu, H. Choi, P. Gatenholm, and A. R. Esker, Langmuir, 2011, 27, 8718. 223 R. Kargl, T. Mohan, M. Bracic, M. Kulterer, A. Doliska, K. Stana-Kleinschek, and V. Ribitsch, Langmuir, 2012, 28, 11440. 224 Y. F. He, Y. Chen, Q. K. Zheng, J. Q. Zheng, and S. Chen, Fiber. Polym., 2015, 16, 1005. 225 G. A. Jeffrey, 'An Introduction to Hydrogen Bonding', Oxford University Press: New York, 1997. 226 A. Toldy, P. Anna, I. Csontos, A. Szabo, and G. Marosi, Polym. Degrad. Stabil., 2007, 92, 2223. 227 A. Toldy, A. Szabo, C. Novak, J. Madarasz, A. Toth, and G. Marosi, Polym. Degrad. Stabil., 2008, 93, 2007. 228 A. Toldy, B. Szolnoki, I. Csontos, and G. Marosi, J. Appl. Polym. Sci., 2014, 131.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0340871/manifest

Comment

Related Items