Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The structure and reactivity of poly(methylenephosphine)s Gillon, Bronwyn Hilary 2008

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_2008_fall_gillon_bronwyn.pdf [ 4.41MB ]
Metadata
JSON: 24-1.0060594.json
JSON-LD: 24-1.0060594-ld.json
RDF/XML (Pretty): 24-1.0060594-rdf.xml
RDF/JSON: 24-1.0060594-rdf.json
Turtle: 24-1.0060594-turtle.txt
N-Triples: 24-1.0060594-rdf-ntriples.txt
Original Record: 24-1.0060594-source.json
Full Text
24-1.0060594-fulltext.txt
Citation
24-1.0060594.ris

Full Text

THE STRUCTURE AND REACTIVITY OF POLY(METHYLENEPHOSPHINE)S by BRONWYN HILARY GILLON B.Sc., The University of British Columbia, 2002 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2008 © Bronwyn H. GilIon, 2008 Abstract The initiation and termination steps of the anionic polymerization of P=C bonds have been modeled. The initiation step was investigated through the stoichiometric reaction of MesPCPh2(1) with RL1 (R Me or n-Bu). In each case, the addition was highly regioselective with the formal attack of R at phosphorus to give the carbanion Li[Mes(R)P—CPh2](R = Me (3a); n-Bu (3b)). To simulate the termination step, carbanions 3a and 3b were quenched in situ with various electrophiles: Mes(Me)P—CPh2H (4a), Mes(n-Bu)P—CPh2H(4b), Mes(Me)P— CPh2Me (6a), Mes(Me)P—CPh2—P(NEt (7a), Mes(Me)P—CPh2—SiMeH(8a) and Mes(Me)P— CPh2—SiMe3(9a). The first use of MALDI-TOF MS in the study of the products of RLi (R = Me, n-Bu) initiated oligomerization of I is reported. The detected linear products R[MesP(=O)-CPh2]Hwith R—P and C—H end-groups are consistent with a chain growth mechanism. The oligomerization was extended to other monomers (MesP=CPhAr, Ar = p-C6H4F or p-C6H4OMe). The results suggest oligomers undergo fragmentation during MALDI-TOF analysis. The thermal reactions of M(CO)6 (M = W, Mo, Cr) with polymer n-Bu[MesP-CPh2]H(10) and its model compound 4a are reported. IR was primarily used to determine the success of metal coordination in the polymer. EDXISEM and GPC-LLS were used to determine metal content of the materials. Most metallation was modest (>10%), however, as much as 20% was attained with Mo. The phosphorus-containing polymer 10 was found to be an effective ligand for gold(l) to afford n-Bu[MesP(AuCl)—CPh2]H11, a new class of macromolecule with high gold content. The prepared model compound MeMesP(AuCI)—CPh2H 12 was characterized by X-ray crystallography and NMR spectroscopy. Block copolymers containing phosphorus atoms in the backbone were prepared and metallated with gold(l). The living polymerization of isoprene (I) and then phosphaalkene I affords block copolymers Plmb[Me5PCPh2In(13a: m = 308, n = 46; 13b: m = 222, n = 77). Treating P1mb[MeSPCPh2Inwith (tht)AuCI results in gold-containing macromolecules [Pl]mb [MesP(AuCl)-CPh2](14a and 14b). Upon dissolution in a polyisoprene selective solvent (n heptane), the metallated block copolymers assembled into micelles (14a: spherical; 14b: worm like). The solution self-assemblies were examined by TEM and DLS. III Table of Contents Abstract.ii Table of Contents iv List of Tables x Listof Figures xi List of Symbols and Abbreviations xiii Acknowledgments xvi Dedication xviii Statement of Co-authorship xix Chapter 1 Introduction 1 1.1 Phosphorus 1 1.2 Polymers 4 1.2.1 Inorganic Polymers 5 1.2.2 Hybrid Phosphorus-Carbon Polymers 7 1.3 History of Poly(methylenephosphine)s 11 1.4 Inorganic Micelles 18 1.5 Outline of Thesis 20 1.6 References 22 Chapter 2 Molecular Studies of the Initiation and Termination Steps of the Anionic Polymerization of PC bonds* 27 2.1 Introduction 27 iv 2.2 Preparation of Model Compounds .29 2.2.1 General Preparation of Carbanions Li[Mes(R)P—CPh2](2.3a and 2.3b) 29 2.2.2 Preparation of Mes(Me)P—CPh2H(2.4a) 30 2.2.3 Preparation of Mes(n-Bu)P—CPh (2.4b) 31 2.2.4 Preparation of Mes(Me)P—CPh2e(2.6a) 32 2.2.5 Preparation of Mes(Me)P—CPh--P(NEt)(2.7a) 32 2.2.6 Preparation of Mes(Me)P—CPh2—S1MeR(2.8a and 2.9a) 34 2.3 Conclusion 36 2.4 Experimental 37 2.4.1 Materials and General Procedures 37 2.4.2 Preparation of Mes(Me)P-CPh2Li(2.3a) 38 2.4.3 Preparation of Mes(n—Bu)P-CPhL1(2.3b) 38 2.4.4 Preparation of Mes(Me)P—CPh2H(2.4a) 38 2.4.5 Preparation of Mes(n-Bu)P.--CPh (2.4b) 39 2.4.6 Synthesis of Mes(Me)P—CPh2e(2.6a) 40 2.4.7 Preparation of CIP(NEt2) 40 2.4.8 Preparation of Mes(Me)P—CPh—P(NEt (2.7a) 41 2.4.9 Preparation of Mes(Me)P—CPh2—SiMeH(2.8a) 42 2.4.10 Preparation of Mes(Me)P—CPh—SiMe3(2.9a) 42 2.4.11 X-ray Crystallography 43 2.5 References 45 Chapter 3 Anionic Oligomerization of Phosphaalkenes: Analysis by MALDI-TOF Mass Spectrometry* 47 3.1 Introduction 47 3.2 Results and Discussions 50 V 3.2.1 Preparation of oxidized oligomers from anionically initiated MesP=CPh2to form Me[(Mes)P(=O)-CPh]H,3.9a 50 3.2.2 Origins of 3.1 0a — The “Second Distribution” 54 3.2.2.1 Decomposition due to Oxidation 54 3.2.2.2 Backbiting 57 3.2.2.2.1 Varying Reaction Conditions 58 3.2.2.2.2 Testing A Different Initiator: Preparing n-Bu[(Mes)P(=O)-CPh2]H,3.9gb 59 3.2.2.2.3 Testing Different Phosphaalkenes: Preparation of Me[(Mes)P(=O)-CPhAr]H, 3.15a and 3.26a 61 3.2.2.3 Fragmentation 67 3.3 Conclusion 67 3.4 Experimental 68 3.4.1 Materials and General Procedures 68 3.4.2 Preparation of Me[MesP-CPh2]H,3.8a 69 3.4.3 Preparation of Me[MesP(=O)-CPh]H,3.9a 69 3.4.4 Preparation of Bu[MesP(=O)-CPh2]H,39gb 70 3.4.5 Preparation of Me[MesP(=O)-CPh(p-C64F)]H,3.15a 70 3.4.6 Preparation of Me[MesP(=O)-CPh(p-C5O e ]H, 3.26a 71 3.5 References 72 Chapter4 The Coordination Chemistry of Poly(methylenephosphine)s with Tungsten, Molybdenum and Chromium 74 4.1 Introduction 74 4.2 Results and Discussion 76 4.2.1 Model Compound Reactions 76 vi 4.2.1.1 Coordination of Model Compound 4.4 to Tungsten 76 4.2.1.2 Coordination of Model Compound 4.4 to Molybdenum 79 4.2.1.3 Coordination of Model Compound 4.4 to Chromium 81 4.2.2 Thermal Stability of Poly(methylenephosphine) 4.2 in Solution 82 4.2.3 Preparation of Group 6 Metal-Containing Poly(methylenephosphine)s 83 4.2.3.1 Coordination of Poly(methylenephosphine) 4.2 to Tungsten 84 4.2.3.2 Coordination of Poly(methylenephosphine) 4.2 to Chromium 90 4.2.3.3 Coordination of Poly(methylenephosphine) 4.2 to Molybdenum 92 4.3 Conclusion 94 4.4 Experimental Section 94 4.4.1 General Methods and Instrumentation 94 4.4.2 General procedure for phosphine complexes 96 4.4.3 Preparation of (HCPh2)(Me)MesPW(CO)5(4.5a) 96 4.4.4 Preparation of (HCPh)(Me)MesPMo(CO) (4.5b) 97 4.4.5 Preparation of (HCPh2)(Me)MesPCr(CO)5(4.5c) 97 4.4.6 Metallation of poly(methylenephosphine) 42 with M(CO)6 98 4.4.7 X-ray Crystallography 98 4.5 References 100 Chapter 5 Macromolecular Complexation of Poly(methylenephosphine) to Gold(l): A Facile Route to Highly Metallated Polymers 102 5.1 Introduction 102 5.2 Results and Discussion 103 5.2.1 Coordination of Au(l)to model compound 5.3 103 vii 5.2.2 Coordination of GoId(l) to poly(methylenephosphine) to yield gold-containing polymer 5.5 106 5.3 Conclusion 108 5.4 Experimental 109 5.4.1 Materials and General Procedures 109 5.4.2 Synthesis of [Mes(Me)(CPh2H)PAu l] (5.4) 109 5.4.3 Synthesis of n-Bu(MesP(AuCl)CPh)H(5.5) 110 5.4.4 X-ray Crystallography 111 5.5 References 113 Chapter 6 Inorganic/Organic Hybrid GoId(l)-containing Micelles 115 6.1 Introduction 115 6.2 Results and Discussion 117 6.2.1 The preparation of PImbPMPn block-copolymers 6.9a and 6.9b 118 6.2.2 Coordination of block-copolymers to gold(l), 6.lOa and 6.lOb 120 6.2.3 Solution self-assembly of gold-containing block-copolymers Plmb(PMPAu)n .121 6.3 Conclusion 125 6.4 Experimental 126 6.4.1 Materials and General Procedures 126 6.4.2 Preparation of 6.9a (m = 308, n = 46) 127 6.4.3 Preparation of 6.9b (m = 222, n = 77) 128 6.4.4 Preparation of gold-containing 6.lOa 128 6.4.5 Preparation of gold-containing 6.lOb 129 VIII 6.5 References .130 Chapter 7 Overall Conclusions and Future Work 132 7.1 Summary of Thesis Work 132 7.2 Future Work 135 7.2.1 Uses for model compounds of poly(methylenephosphine) 135 7.2.2 The preparation of polymer with high tungsten-content 135 7.2.3 Gold-containing micelles 136 7.3 Closing Remarks 136 7.4 References 137 ix List of Tables Table 2.1 X-ray crystallographic data for 2.4a, 2.7a, 2.8a and 2.9a 44 Table 3.1 Selected MALDI-TOF MS data for oligomeric mixture 3.9a 53 Table 3.2 Selected MALDI-TOF MS data for oligomeric mixture 3.8a 56 Table 3.3 MALDI-TOF MS data for oligomeric mixture 3.8a with partially oxidized P atoms 57 Table 3.4 Selected MALDI-TOF MS data for oligomeric mixture 3.9b 61 Table 3.5 Selected MALDI-TOF MS data for oligomeric mixture 3.15a 64 Table 3.6 Selected MALDI-TOF MS data for oligomeric mixture 3.26a 66 Table 4.1 Studies on the Thermal Stability of Poly(methylenephosphine) 4.2 in Solution 83 Table 4.2 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with W(CO)6 to Form 4.7a 85 Table 4.3 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with Cr(CO)6 to Form 4.7c 92 Table 4.4 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with Mo(CO)6to Form 4.7b 93 Table 4.5 Details of Crystal Structure Determinations of 4.5a and 4.5b 99 Table 5.1 Details of Crystal Structure Determinations of 5.4 112 x List of Figures Figure 1.1 Examples of Synthetic Routes to Acyclic Phosphaalkenes 2 Figure 1.2 Examples of Reactions of P=C Bonds that Parallel Reactions with C=C Bonds 3 Figure 1.3 The Three Steps of Anionic Addition Polymerization: Initiation, Propagation and Termination 4 Figure 1.4 Preparation of polysilanes 1.1 polysiloxanes 1.2 and polyphosphazenes 1.3 6 Figure 1.5 Reactivity of poly(methylenephosphine) 1.19 16 Figure 2.1 The solid-state molecular structure of 2.4a 31 Figure 2.2 The solid-state molecular structure of 2.7a 34 Figure 2.3 The solid-state molecular structure of 2.8a 35 Figure 2.4 The solid-state molecular structure of 2.9a 36 Figure 3.1 Mechanism of anionic polymerization (initiation, propagation and termination) for P=C bonds 48 Figure 3.2 MALDI-TOF MS of oligomer mixture, 3.9a 53 Figure 3.3 MALDI-TOF MS of oligomeric mixture 3.8a 56 Figure 3.4 Proposed Mechanism of Backbiting of Propagating Oligomer 3.7a at C 58 Figure 3.5 MALDI-TOF MS of n-BuLi initiated oligomer mixture, 3.9mb 61 Figure 3.6 MALDI-TOF MS of the oligomeric mixture 3.15a 64 xi Figure 3.7 MALDI-TOF MS for oligomeric mixture 3.26a 66 Figure 4.1 31P{1H} NMR spectra (diglyme) of 4.4 and 4.5a 77 Figure 4.2 Molecular structure of the tungsten-model adduct 4.5a 79 Figure 4.3 Molecular structure of the molybdenum-model adduct 4.5b 81 Figure 4.4 31P NMR Spectrum in CD2I of: 4.2 and 4.7a 86 Figure 4.5 Infrared Spectra of Metal carbonyl complexes 87 Figure 4.6 X-ray photoelectron spectrum of tungsten-containing 4.7a 88 Figure 4.7 X-ray photoelectron spectrum of the P 2p binding energy of 4.2 and 4.7a 89 Figure 5.1 31P NMR spectra of 5.3, 5.4, 5.2 and 5.5 104 Figure 5.2 The solid-state molecular structure of 5.4 106 Figure 5.3 TGA trace for gold-containing polymer 5.5 108 Figure 6.1 The solution self-assembly of an AB block-copolymer to form a macromolecular structure (shown: cross-section of a spherical micelle) 116 Figure 6.2 TEM images of 6.lOa 122 Figure 6.3 TEM images of 6.lOb 124 Figure 6.4 Sketches of two basic structural motifs for block-copolymers in dilute solution: (a) spherical micelles and (b) cylindrical micelles 125 xl’ List of Symbols and Abbreviations abs. = absolute AD = Anno Domini amu = atomic mass units Anal. = Analysis Ar Aryl n-Bu n-Butyl t-Bu tert-Butyl °C degree Celsius Calcd Calculated cat. = catalytic cf= confer, compare CPS = counts per second d = day dba = Dibenzylideneacetone, (PhCHCH)2C0 Da = Dalton DLS = Dynamic Light Scattering Diglyme = Bis(2-methoxyethyl) ether,CH3OCH2 dn/dc = refractive index increment EDX = Energy dispersive X-ray spectroscopy E or E’ = Inorganic atom E = electrophile E Energy El = Electron Impact equiv = equivalent(s) Et = Ethyl Formvar = polyvinyl formal, TEM grid support film FTIR = Fourier Transform Infrared XIII Glyme = 1 ,2-Dimethoxyethane,CH3OCH2 GPC = Gel Permeation Chromatography h = hour ICP-MS = Inductively Coupled Plasma Mass Spectrometry IR = Infrared MALDI-TOF= Matrix-Assisted Laser Desorption/Ionization - Time of Flight Me Methyl MeO Methoxy Mes = Mesityl = 2,4,6-trimethyiphenyl;C6H2-2,4,6-Me3 Mes* = Supermesityl 2,4,6-tri-tert-butylphenyl;CH2,4,6tBu3 mm = minute M = Number average molecular weight = Weight average molecular weight mol mole MS = Mass Spectrometry NMR = Nuclear Magnetic Resonance PDI = Polydispersity Index Ph = Phenyl ‘Pr isopropyl PS = Polystyrene ref. = reference ret. = relative rt = Room temperature, 25 °C R, R’, or R” = Side group s = second SEM = Scanning Electron Microscopy Tor Temp. = Temperature Tonset = Onset temperature of weight loss TEM = Transmission Electron Microscopy TGA = Thermogravimetric Analysis xiv THF = Tetrahydrofuran TMS = chiorotrimethylsilyl, Si(CH3) TMSCI = chiorotrimethylsilane, (CH3)S1CI VAZO 88 = 1,1’-Azobis(cyclohexanecarbonitrile) wt% Weight percentage xv Acknowledgments This seems an impossible task to thank everyone, but I’m going to try... First and foremost, a big thank you to Dr. Derek Gates for accepting me into the lab and providing advice and encouragement. Fellow Gates group members have made the lab a fun and interesting place to work: Dr. Chi Wing Tsang, Dr. Mandy Yam, Dr. Vince Wright, Kevin Noonan, Joshua Bates, Julien Dugal-Tessier, Cindy Chan, Paul Siu, Tom Hsieh, Yuen-Ying Carpenter, David Riendl, and Bastian Feldscher. Thanks also to those who helped with the instruments for the characterization of stuff: Marshall Lapawa (MALDI-TOF), Dr. Yun Ling (MS), Mary Mager (EDX/SEM), the people at Wyatt (GPC), Dr. Johan Janzen (DLS), Garnet Martens and Derrick Home (TEM). Also, for the characterization that I did not perform myself: Ken Chung Wong (XPS), Josh Bates and Dr. Brian 0. Patrick (x-ray crystallography). Lots of people within the department were particularly helpful making my research possible and/or easier: John Ellis, Xin-Hua Huang, Brian Snapkauskas, Ken Love, and Brian Ditchburn. I have to thank my friends from models night for their endless support and enthusiasm as well as affording some fabulous times. Each of my friends are inspirations to me in some form, for example: Ali Lee: incredibly insightful about pretty much everything, but in particular, people which makes for fabulous conversation; Louisa Stanlake: the funnest (that’s a word... found it in a dictionary once) person ever; Jennifer Kozak: so driven someone should give her nobel prize; Jackie Stewart: so much energy she makes the rest of us look like snails; Courtney Turner: even at such a young age, has a fabulous handle on who she is; Shiva Shoai: future diplomat for the world. xvi Other people in the chemistry department whose friendship over the years has been important and deserve a shout out: Tamara Kunz, Agostino Pietrangelo, Kristin Matkovich, Howie Jong, Paul Hurley, Bryan Shaw, Jason Thomas, Andrew Tait (and Natalie Ruhr), Jay Reid, Pavel Glaze, Timin Hadi, Mark Wood (and Kathleen Wood). During grad school, I have been really lucky with very understanding friends: Roxan Maurcot (Hon’s anyone?), Karmin lp (heart heart everything), Dr. Rebecca Rivera, Scot Ramsay, Neil Anderson, Mike Shapcotte, Stuart Dean, and especially Dale Henshaw. Most importantly, I have to thank my family for being so supportive: my parents, Lynn and Neil Gillon (some day you’ll get to retire and move in to the apartment); my favourite sister Carrie, despite the fact that she hates the polymer she was my best friend (and roommate for so long); my favourite brother Tristan, teacher of the language on the internets (roflmao, who knew...) and conspiracy theories. xvii Dedication This thesis is dedicated to my parents who taught me the joys of reading and discovery. It was for them that I wrote the following paragraph; the only one in entire thesis that they would probably understand (and my favourite). Phosphorus is an element. It has 15 protons, 15 electrons and 16 neutrons in its most abundant form. It exists in 3 allotropes that strangely come in 5 colours: red, white (or yellow) and black (or violet). Sometimes, as the white allotrope, it has no colour at all. In air, it will glow. Sometimes it’s pyrophoric (white), sometimes it’s not (red). It can be found in Group 15, though its chemistry is like that from group 14, sometimes. White phosphorus is always toxic. Though phosphorus is required by every cell in the human body for normal function; it is an essential mineral.1 Naturally, it was first isolated from urine. An alchemist by the name of Hennig Brandt from Hamburg of the Holy Roman Empire failed to make gold. Nevertheless, he landed in the history books for his 1669 AD discovery of the element with the atomic number 15 and the symbol P. It is a truth universally acknowledged, that a phosphorus atom in possession of a lone pair must be in want of a metal. Pride and Preiudice by Jane Austen, if she were a phosphorus chemist. xviii Statement of Co-authorship The material reported in Chapter 2 was published in 2007. Bronwyn H. Gillon, Kevin J. T. Noonan, Bastian Feldscher, Jennifer M. Wissenz, Zhi Ming Kam, Tom Hsieh, Justin J. Kingsley, Joshua I. Bates, and Derek P. Gates, “Molecular studies of the initiation and termination steps of the anionic polymerization of P=C bonds,” Can. J. Chem. 2007, 85(12), 1045-1 052. The chapter was entirely written by me. Undergraduate researchers synthesized all the compounds; I directly supervised Tom Hsieh who synthesized Me[MesP-CPh2]H. I was involved in the characterization of compounds n-Bu[MesP-CPh2}H,Me[MesP-CPh]P(NEt) and Me[MesP-CPh2]SiMeHand Me[MesP-CPh2]SiMe3. A portion of the material reported in Chapter 3 was published in 2004. Bronwyn H. Gillon, and Derek P. Gates, “Analysis of the products of the anionic oligomerisation of a phosphaalkene using MALDI-TOF mass spectrometry,” Chem. Commun. 2004, 1868—1 869. I wrote the chapter and performed all the syntheses and characterizations. The results presented in Chapter 4 will be submitted for publication shortly. I performed all the syntheses and characterizations. Except for Me(Mes)(CPh2H)PW(CO)5which was prepared by Tom Hsieh, an undergraduate whom I supervised. The material reported in Chapter 5 was published in 2008. Bronwyn H. Gillon, Brian 0. Patrick and Derek P. Gates, “Macromolecular Complexation of Poly(methylenephosphine) to Gold(l): A Facile Route to Highly Metallated Polymers,” Chem. Commun., 2008, 2161-2163. I wrote the chapter and performed all the syntheses and characterizations. xix A version of Chapter 6 will be submitted for publication in due course. I wrote the chapter. Kevin J. T. Noonan synthesized the copolymers. I performed the coordination reactions with gold. Together, we prepared the nanostructures and performed their characterization. xx 1Chapter 1 Introduction 1.1 Phosphorus The question I find most difficult to answer; the one which always crops up sooner or later when the subject is mentioned, is, approximately: ‘But how on earth did you come to get yourself mixed up in a crazy affair like this, anyway?’ Web by John Wyndham Since the discovery of phosphorus, phosphorus-containing chemicals are countless in number and boundlessly useful: from organic synthesis to materials science to DNA.1 Common household materials containing phosphorus include: matches, toothpaste, fertilizers, pesticides and detergents. Phosphorus is sourced from the earth in the form of phosphate minerals, in the case of Nauru from its only natural resource: ancient guano deposits.2Elemental phosphorus (P4), being fairly reactive to water and oxygen, must be extracted from minerals (rock phosphate or phosphorite) by heating with coke and sand (Scheme 1.1). From P4, other useful chemicals, like phosphorus trichloride, can be prepared (Scheme 1 .2). Due to its varied reactivity, acting either as an electrophile or a nucleophile depending on reaction conditions, PCI3 is a useful commodity chemical and precursor to widely used chemicals like PCI5, PPh3, BINAP and countless others. A class of significant, albeit lesser-known, phosphorus-containing molecules known as phosphaalkenes, important to this thesis work, can also be prepared from PCI3. 1500°C 2 Ca3(P04)+6 Si02 + 10 C 6 CaSiO3 + 10 CO + p4 Scheme 1.1 References start on page 21 Chapter 1 2 A P4+6Cl2 4PCI3 Scheme 1.2 Phosphaalkenes, compounds with genuine (2p—3p)rr-bonds, are an intriguing class of low-coordinate phosphorus species due in part to their challenging synthesis. The first stable acyclic derivative was reported in the mid-I 970’s from a condensation reaction followed by a 1 ,3-silatropic tautomerization (see Figure 1.1, Route A).5 Since then, other methods of preparation have been developed, including condensation reactions (Routes B — D), phospha Wittig reactions (Route F), base-catalyzed elimination reactions (Route G), and base-catalyzed keto-enol type tautomerization reactions (Route H). Phosphaalkene synthesis via Route E requires the formation of a terminal metal-phosphinidene complex (see Figure 1.1). Li R SiMe3 R R—P’ ÷ O=C H R’ / / \ / / R—P + 0C SiMe. R” R—P + 0C\ \ \ SiMe3 Cl H R” ClSiM\ 1-LiOSiMe3Q R’ Base (cat) ,R’ D P=C R—P + CR’R”X2Base \ H’ CH2 R R’ -2HX H Bas/ F -O=PMe3 NçMO R’ R PML +D_C / ‘SR” Me R’ R R”X H t-Me + R Me R” Figure 1.1 Examples of Synthetic Routes to Acyclic Phosphaalkenes. References start on page 22 Chapter 1 3 Interest in phosphaalkenes continues to grow due to their parallels to the chemistry of C=C bonds and their potential applications.68Phosphaalkenes can undergo reactions similar to alkenes, some examples are shown in Figure 1.2. Like alkenes, phosphaalkenes undergo Diels Alder-type reactivity (reaction A) with butadiene resulting in the formation of a phosphorus heterocycle. Small molecules, like HX or H2, add across the PC bond (reactions B and C, respectively). Reaction D demonstrates the phosphorus analogue of an epoxide formation. Finally, P=C bonds can coordinate in an 2 fashion to metal centers like C=C bonds as shown in reaction E. Incidentally, compounds with P=C bonds are being explored for their effectiveness as ligands in transition-metal catalyzed reactions,913 as well as in polymer science.1421 [M] P=C [O] _______ PC o / \ A P—C- H\ [Rh]caV /X X\ /H PC PC_ or Figure 1.2 Examples of Reactions of P=C Bonds that Parallel Reactions with C=C Bonds. References start on page 22 Chapter 1 4 1.2 Polymers Plastics, an array of polymerization products, are ubiquitous in a North American’s life. In fact, the plastics trade in Canada was worth over 15 billion dollars in 2007.22 Methods of preparing organic polymers are limited to just two methods: (1) step-wise or (2) chain growth polymerization. Step-wise polymerization requires that the monomers contain functional endgroups (for example: hydroxyl groups, carboxylic acids). The polymer is made by a series of condensation reactions eliminating a small molecule (commonly water). For addition polymerization, an unsaturated monomer (containing a C=C bond, for example) is required. The polymer is made by the addition of ions or radicals across the multiple bond of the monomer. Addition polymerization occurs in three stages: (1) initiation: the activation of a C=C 7t bond to generate an ion (heterolytic cleavage) or a free radical (homolytic cleavage); (2) propagation: the continual addition of monomer units which increase the length of the growing chain; and (3) termination, addition of a species such that it quenches the active site of the growing polymer. Figure 1.3 illustrates the mechanism of an anionic addition polymerization. • ____ I IInitiation: R:e c=c R—c—c:e /\ II —A ,--/r 1/ CC • . II II( / ‘ Ii iPropagation. R_C_Cj_.e R-Fc—c :e L n+1 Ii ii El ii Termination: R4C—C—I--:e R-l-C—C-l-E Ii ii m>n Ii iiL Jm L Jm Figure 1.3 The Three Steps of Anionic Addition Polymerization: Initiation, Propagation and Termination. References start on page 22 Chapter 1 5 1.2.1 Inorganic Polymers Most research in polymer science is focused on organic polymers which possess carbon atoms in the backbone. In contrast, the study of inorganic polymers is in a more primitive stage of development. Inorganic polymers contain at least one element other than C, 0 or N in the backbone.23The preparation of such polymers is challenging and continues to be an active area of research due to their unique properties and potential applications. Until recently, inorganic polymers were not synthesized using addition polymerization, a common method of preparing organic macromolecules. Addition polymerization requires a site of unsaturation (E=E’ or EE’ bonds), but unsaturated inorganic monomers that are suitable (stable and readily available) are scarce.23’4Therefore, inorganic polymers have historically been synthesized only by step-wise or ring-opening methods. Some important inorganic polymers include polysilanes (1.1), polysiloxanes (1.2) and polyphosphazenes (1.3). rRl rR 1 rR II I II I II —f--Si--I-- -+-Si—O-+- —I-PN II I II I II[R] [R J, [R 1.1 1.2 1.3 There are multiple methods of synthesizing (each of) polymer 1.1, 1.2 and 13 (see Figure 1.4 for a few examples) most of which are based on polycondensation or ring-opening methods. Polysilanes 1.1 can be made through thermally induced Wurtz coupling of organodichlorosilanes in the presence of sodium metal. Organodichlorosilanes also act as precursors to polysiloxanes 1.2 when introduced to water. Polyphosphazenes 1.3 can be made by thermally induced ring-opening of hexachlorocyclotriphosphazene.2327 References start on page 22 Chapter 1 6 H-Si R Cl—Si—Cl 1.1 R —- Si—C [F! 1.2 Cl Cl rci N’N heat I ii i - —j—P=NCl-..p p Lcl Cl Cl I .3a Figure 1.4 Preparation of polysilanes 1.1 polysiloxanes 1.2 and polyphosphazenes 1.3a Each of these types of polymers has diverse properties and an array of applications.28 Their properties are due to the different functionalities in the backbone and/or the side chains. Polysilanes 1.1 have unique optical and electronic properties due to the delocaliztion of a electrons. Polysiloxanes 1.2 are commonly found in most, if not all, Canadian households. Their applications include sealants, lubricants, contact lenses and medical implants. Their unique properties, such as high flexibility (even at low temperatures) and high thermo-oxidative stability are a consequence of the strong Si—C bonds in the backbone with long Si—C bonds (1.64 A vs 1.54 A for C—C bonds) and large Si—C—Si bond angles (—143° vs —109° for C—C—C bond angles). The strong P—N bonds in the backbone (alternating P—N and P=N bonds) in polyphosphazenes 1.3 result in the material’s high temperature resistance and fire-retardancy. The side chains in polyphosphazenes 1.3 are easily modified by way of simple nucleophilic substitution reactions of the chloride with other organic functionalities. As a result, the properties of 1.3 are tunable and allow for access to new materials with variable properties. Polyphosphazenes have found applications as insulators, optical switches, lenses and Na H20 References start on page 22 Chapter 1 7 elastomers29and belong to an important class of inorganic polymers with phosphorus atoms in the main chain. We are interested mainly in phosphorus-containing organic polymers. 1.2.2 Hybrid Phosphorus-Carbon Polymers The incorporation of phosphorus into the polymer backbone is an emerging area of research which has often led to materials with novel properties. Importantly, phosphorus polymers are of interest for their specialty applications as electroluminescent devices, sensors, biomaterials, and catalyst supports.14’2327, 30 The difficulty in finding suitable methods to incorporate phosphorus atoms into long chains is a major challenge for researchers in this area 31-42 Some recent examples of phosphorus-containing polymers possessing interesting electronic properties are found in analogues of conjugated organic polymers. Some recent examples of such materials include a simple system developed by Lucht and coworkers. Through transition-metal mediated condensation polymerization of diiodobenzene derivatives with primary alkylphosphines, Lucht and coworkers were able to obtain conjugated polymers (j•4)•43’44 Hybrid polyaniline-poly-p-phenylene phosphine polymers (1.5) are prepared in a similar fashion using Pd-catalyzed coupling reactions of primary phosphines and dihaloaromatic diamines.45Gates et a!. use the condensation polymerization method to obtain the first phosphorus analogue of poly(p-phenylenevinylene)s (PPV’s which are used in organic light emitting diodes) with moderate molecular weights (1.6 and 1.7)1617 Protasiewicz and coworkers has shown that similar polymers (1.8) showed evidence of conjugation through analysis by UVNis spectroscopy.46’7A key challenge to the preparation of these types of macromolecules involves their low solubilities which often results in low molecular weight polymers. So far, these polymers have not been used in devices. References start on page 22 Chapter 1 8 PPv ±°if A!1 R 1.4 1.5 OSiMe3 OSiMe3 ] n\ 1.6 In contrast to the above examples, polymers containing phospholes in the backbone have been recently incorporated into electroluminescent devices. Réau and coworkers have integrated phospholes into poly(thiophene)s (1.9). The material, prepared by electropolymerization, has been demonstrated as a detector of Group 16 elements.4849 Also, fused-ring phosphole-containing poly(thiophene)s (1.10) have been prepared which showed 1.7 n n-Hex n 1.8 References start on page 22 Chapter 1 9 photoluminescence properties both in solution and in the solid-state.5051 Recently, by extending the conjugation of similar frameworks (1.11), the emission colours of the material can be tuned from green to yellow.52 Finally, materials incorporating phospholes into conjugated organic backbones have been prepared (1.12) with modest molecular weights.33’5 1.9 1.10 Macromolecules with phosphines in the backbone are also known. For example, the first poly(vinylenephosphine) has been recently reported. The synthesis of 1.13 involves the radical copolymerization of cyclooligophosphines with phenylacetylene resulting in a yellow solid that fluoresces blue-green (see Scheme 1 3)32 With polymers containing chiral P atoms in the backbone, Chujo and coworkers set out to form optically active polymers like 1.14 and 1.11 F 1.12 Spacer = H13C6 F 0C6H13 - -* H13C60 References start on page 22 Chapter 1 56 Using an interesting strategy of anionic ring-opening polymerization of a phosphirene 1.16, Vanderark et al. demonstrates a facile route to poly(vinylenephosphine)s 1.17 (see Scheme j•4)34 10 \ / P—P I \ H radical r I / + Ph —f—C=C initiator I I P L M Scheme 1.3 1.13 BH3 Ph AMe Me 1.16 _f Ph 1 rph Me11/n n-BuLi Pn-Bu )ç4f ?==••-%%[ H[Me M[ _jn m+n=1 n = 0.35 1.17 Scheme 1.4 Incorporation of phosphorus atoms into the polymer backbone is obviously an active area of research. The recent examples above were meant to highlight the strategies used by other researchers (condensation, ring-opening, and electro- polymerizations). Previously, Cl. 1.14 BH3 1.15 n References start on page 22 Chapter 1 11 addition polymerization was often dismissed as unsuitable for inorganic systems despite the numerous advantages of addition polymerization over step-wise methods. For example, condensation polymerization requires very accurate stoichiometry and high monomer purity in order to gain high molecular weight polymers. Also, condensation reactions by their nature eliminate wasteful byproducts that are often toxic (e.g. HCI). The disadvantage in using ring- opening polymerizations can be the lack of control over the resulting polymer’s molecular weight. The Gates group is tackling these problems by developing the addition polymerization of phosphaalkenes (PA) to afford phosphorus-carbon hybrid homo- or co-polymers.18 19,21,57 1.3 History of PoIy(methylenephosphine)s The Gates group set out to prepare inorganic macromolecules through addition polymerization. Several multiple-bond systems were considered as potential monomers; however, P=C bonds were targeted due to their analogous reactivity to CC bonds.6758 For the first time, an inorganic polymer was synthesized through addition polymerization methods. The polymerization of phosphaalkene 1.18 affords poly(methylenephosphine) 1.19; a new functional phosphine polymer with alternating phosphorus and carbon atoms in the main chain.19 lniflal studies suggested that the polymerization of phosphaalkene 1.18 requires high temperatures (> 150 °C), long reaction times, and high concentrations, even in the presence of radical or anionic initiators. Estimated Ma’s of up to 11 500 g mo11 (vs polystyrene) have been reported. Functionalization of the polymer 1.18 is achieved through reactions with H20 and elemental sulfur resulting in oxidized polymer 1.20 and sulfurized 1.21, respectively. References start on page 22 Chapter 1 12 F o Phi I II I t-P-C [Mes Ph Ph Radical or Anionic J..l-J_ 1.20 / \ Initiator I I I IMes Ph >150°C LMeS Ph] S Phi 1.18 1.19 I Mes Ph IL .Jfl 1.21 Notably, characterization of the phosphine polymer 1.19 by GPC (gel permeation chromatography) proved confusing. Our laboratory, at the time, was equipped with a traditional GPC system that required column calibration with polystyrene standards. Consequently, the estimated Me’s of the polymers were artificially diminished as determined by GPC. With a triple detection GPC (equipped with a laser light scattering detector), the absolute Me’s of polymers were determined and were greater than the estimated Me’s. For example, sulfurized polymer 1.21, had an estimated M of 11 900 g mor1 and an absolute M of 32 000 g mo11. This nearly threefold difference in the determined molecular weights can be rationalized due to the molecular weight difference between styrene and MesP=CPh2,as described in the next paragraph. The principle behind GPC is that macromolecules are separated based on their volume (or size).59 Typically, polystyrene (1.22) standards of known Me’s are used to create a calibration curve (M vs elution time). However, there is an inherent problem when analyzing a sample that is chemically different than the standards used to calibrate the column. To illustrate this problem, consider a sample of polystyrene 1.22 and a sample of poly(methylenephosphine) 1.19 each with 100 repeat units. Both samples have approximately the same length and accordingly, approximately the same volume. As a result, both samples have the same elution time. Consequently, a GPC that has been calibrated with polystyrene samples would estimate that the phosphorus-containing 1.19 (n = 100) sample has similar molecular weight to that of the polystyrene. However, with 100 repeat units, the molecular weight for organic 1.22 is 10 400 g References start on page 22 Chapter 1 13 mol1 but for inorganic 1.19 is three times greater at 31 600 g mol1. Therefore, polystyrene makes a poor standard for poly(methylenephosphine)s. Since the Gates group was only equipped with a traditional GPC, we looked for other methods to characterize the absolute Me’s of the phosphorus-containing polymer 1.19. Fortunately, the Chemistry Department at the University of British Columbia possesses an accessible MALDI-TOF MS. We are interested in analyzing poly(methylenephosphine)s by MALDI-TOF MS because it gives absolute masses rather than relative ones (eg. GPC) and usually only molecular ions are observed without fragmentation. f-c-c [ H Ph 1.22 The emergence of soft ionization (i.e. MALDI) mass spectrometric techniques to investigate high molecular weight (MW) macromolecules is revolutionizing the field of synthetic polymers,6062 however their use is still relatively uncommon in the analysis of inorganic systems.6367 A notable advantage in using MALDI-TOF MS to characterize polymers is the detection of the endgroups. Our group had not been able to spectroscopically identify the endgroups of poly(methylenephosphine)s. Numerous attempts were made to detect phosphorus-containing 1.19 by MALDI-TOF MS, but to no avail. Often, smaller chains are easier to ionize. In fact, Tsang et a!. had previously shown that phosphaalkene 1.23 was oligomerizable with a substoichiometric amount of HOTf (see Scheme 1.5). Importantly, the ions of the resulting oligomers 1.24 were detected up to 6 repeat units using electrospray mass spectrometry.68 References start on page 22 Chapter 1 14 H 1/3HOTf F ‘j 1 P=C OTf-l---P—C—IH Mes*= / \ IllMes* H I Mes’ H L fl 1.23 1.24 Scheme 1.5 Another question that must be addressed is whether the initiation, propagation, and termination steps for the addition polymerization of PC bonds are analogous to those known for olefins (recall Figure 1 .3). Analysis of more easily characterized small molecules that are model compounds to homopolymer 1.19 provides invaluable insight into the mechanism of polymerization. Initially, polymerization studies of P=C bonds were focused on radical initiation methods. In fact, with a free-radical source (VAZO), the copolymerization of styrene with phosphaalkene 1.18 was investigated primarily by Dr. Mandy Yam and Dr. Chi Wing Tsang (see Scheme 1.6). The resulting random copolymers 1.25 represent new hybrid inorganic-organic macromolecules that can be used in polymer-supported catalysis. Specifically, copolymer 1.25 has been used in Suzuki cross-coupling reactions that form a C-C bond between aryl halides and aryl-boronic acids (refer to Scheme 1.7). The coupling occurs in the presence of a palladium catalyst (commonly Pd(PPh3)4and a base. Phosphine ligands are used because they are stable and they readily complexed to the Pd in its active zero-valent form. It has been found that copolymer 1.25 is a better ligand than homopolymer 1.19 since the product yield is much greater (90% cf 23%). Without the presence of phosphorus-containing 1.25 (or any other phosphine), the product yield is poor. These results suggest that in fact, the copolymer 1.25 is acting like a molecular phosphine, though there has been no conclusive evidence that it coordinates to palladium. References start on page 22 Chapter 1 15 Ph H Ph H H Mel Ph 2 VAZO Sh]\ E]m] 1.25 Scheme 1.6 + 1.5 B(0H)2 THE Scheme 1.7 The use of the copolymer 1.25 as a ligand in catalytic reactions showed that the polymer can behave chemically like a phosphine. Investigations into the chemical functionality of poly(methylenphosphine)s 1.19 began with simple reactions like oxidation and sulfurization in order to stabilize the polymer to air (see Figure 1.5).19 Since that initial report, polymer 1.19 has been shown to form adducts with borane (resulting in boronated I Additionally, treatment of 1.19 with MeOTf affords methylphosphonium ionomers 1.27.69 Interestingly, only —50% of the phosphines in the backbone can be methylated. Interestingly, when I started this thesis work, homopolymer 1.19 had never been coordinated to a metal system. References start on page 22 Chapter 1 16 Fo Phi FH3B Phi II I I : S8 [Ms hj - 1.19 OTfF S Phi *.. Me Ph Ph III II 1+1 .. +-P—c—+ P—c ‘ P—c liii II II [Mes Ph] Mes Ph \ Mes Ph n x y 1.21 “ x = 0.5, y = 0.5 1.27 Figure 1.5 Reactivity of poly(methylenephosphine) 1.19. After radical methods were developed, the next logical step in the investigation of the polymerization of P=C bonds is the extension to anionic living methods. Living polymerization enables the preparation of macromolecules with controlled architectures (complex structures such as block copolymers, star polymers, and comb polymers).7073 Swarc first demonstrated in 1956 that adding styrene to a mixture of an alkali metal and naphthalene in THE results in an increase in the viscosity of the reaction mixture. However, this increase in viscosity ceases eventually, but upon addition of more monomer, the viscosity increases again. This first example of living polymerization illustrates that the living chain can rest indefinitely. Only upon controlled addition of an electrophile is the polymer terminated. The result is that the prepared polymers have a narrow range of lengths with low polydispersity indices (1.00 < PDI <1 .10) because the polymer chains grow without termination (versus traditional addition polymerization). Over the past 60 years, living polymerization has evolved to include other types of systems: anionic, cationic, ring-opening and free radical.7377 References start on page 22 Chapter 1 17 Block polymers possess two or more chemically different polymer segments, or blocks, connected by a covalent linkage. Notably, it has been recently found that living polymerization methods have been used to obtain block copolymers containing the three main classes of inorganic polymers (polysilanes 1.1, polysiloxanes 1.2, and polyphosphazenes 1.3). Saam et al. were the first to successfully polymerize inorganic monomers in a living fashion.78 By using anionic ring-opening polymerization methods, they have provided routes to block copolymers, polystyrene-b-polydimethylsiloxane 1.28. More recently, the first copolymer containing polysilane blocks (1.29) have been prepared by Sakurai and coworkers.79 Poly(ethyleneoxide) b-polyphosphazene 1.30 represent the expansion of the range of functional inorganic blocks. H H Me 1 _____H Me n-Bu-I—C—C Si—C—I-H n-Bu--j--Si—Si C—C H Ill I I III II[ H Ph m Me ] [ Me Me m H CO2Me 1.28 1.29 CCHCF QCH2CF3 CF3H2O1—P--N PN H [ CHCHOCH m OCH2CF3 1.30 Recently, living polymerization was achieved for MesP=CPh2using substoichiometric amounts of n-BuLi.188°Studying the polymerization at various temperatures, Kevin Noonan of the Gates group has been able to estimate the activation energy for the propagation of P=C bonds in glyme (Ea = 14.0 ± 0.9 kcal mol-1).8°Importantly, homopolymers with controlled molecular weights are now readily prepared as well as block copolymers 1.31 of styrene and phosphaalkene. More complex architectures containing poly(methylenephosphine)s are within reach. References start on page 22 Chapter 1 18 __ Phi n-Bu-f-—C--C P—C-+H III I II[ H Ph m Mes PhJ 1.31 1.4 Inorganic Micelles Complex architectures can result from block copolymers which can self-assemble into mesophases with periodic order with typical repeating distances in the 10-200 nm range both in solutions or melts.81 Vanzo was the first to observe morphological structures in block polymers. He showed that when solutions of diblock polymers (polystyrene-b-polybutadiene 1.32) are evaporated, lamellar structures are formed.82 Using an electron microscope, Vanzo observed that the spacings corresponded to the dimensions of the randomly coiled blocks. Tuning external conditions such as pH or temperature or solvent can control the solution self-assembly of the different phases. The ability to control the order of the phases makes block copolymers compelling candidates for the building blocks of novel materials such as micelles, rods, and vesicles, for example.83Applications for these materials include growth of nanowires, nanoscale architectures for ceramics, and drug delivery. EHH 1 n-Bu--j---C—C PB—j-H Lhm in 1.32 H H HH CH PB= ‘/ C. H C C ..s I I ?CF H H C.HHH H’HH H References start on page 22 Chapter 1 19 Since Vanzo’s first discovery of the solution self-assembly of organic block copolymers, work has been extended to inorganic systems which could result in new structures with novel properties.85Examples of organic-inorganic diblock copolymers that form micellar structures in solution are rare.86 The first example of the solution self-assembly of polymers with an inorganic block has been reported by Saam et al. who observed that poly(styrene-b-dimethylsiloxane) 1.33 “copolymers resemble surfactants due to the extreme differences in solubility between the two blocks.” In cyclohexane, a good solvent for polysiloxanes, the formation of micelles occurs. In fact, at high concentrations of the colourless polymer in the aliphatic solvent, the solution turns blue. Winnik, Manners, and co-workers have reported that poly(ferrocenyldimethylsilane b-isoprene) 1.34 can form rigid cylindrical micelles.8789 The length of these cylindrical micelles can be increased through the addition of more block copolymer that is dissolved in a good solvent for both blocks. The result is the controlled growth of the micelles.9°Cylindrical micelles are formed from aqueous solutions of poly(ferrocenyldimethylsilane-b-[2-(N,N- dimethylamino)ethylmethacrylate]) 1.35, however, only moderate molecular weights have been achieved.91 Redox-active micelles of poly(ferrocenyldimethylsilane-b-2-vinylpyridine) 1.36 have also been formed in alcohols.92’3Simply by using different alcoholic solvents the morphology of the self-assembled structures can be formed as either spherical or cylindrical micelles. rHH1 EMe 1 riIi I I I n-Bu1-_b—C I Si—O-l-H sec-Bu I P1 II I I [HPhjm L Jm‘Me I I I n-Bu P1 = —_:j.’ -‘60% —30% —10% 1.33 1.34 1.35 1.36 n References start on page 22 Chapter 1 20 Micelles (or other complex architectures) are formed in solution when the solubilities of the two blocks are different. In the case of block copolymer 1.31 from styrene and phosphaalkene, both blocks are soluble in polar aprotic solvents and insoluble in non-polar aliphatic solvents. Therefore, in order to induce self-assembly of phosphalkene-derived copolymers, an organic block that is soluble in non-polar aliphatic solvents could be selected. 1.5 Outline of Thesis When I joined the Gates Group in 2002, P=C bonds had been successfully polymerized for the first time through both anionic and radical methods.94The resulting poly(methylenephosphine)s were characterized mainly by GPC and NMR spectroscopy. Some questions remained about these novel phosphorus-containing polymers. We wanted to gain insight into the structure of the polymer, particularly when anionically initiated. More specifically, we were interested in the site of initiation, the nature of the endgroups, and the mechanism of polymerization. With the polymer in hand, we also wanted to functionalize poly(methylenephosphine) at the P atom with transition metals. Chapters 2 and 3 highlight our endeavors to get a handle on the mechanism of polymerization. In Chapter 2, the site of addition of the anion (R) at the P=C bond and the subsequent addition of an electrophile (e.g. H, Me) will be experimentally determined. These model compounds give invaluable information on the structure of the novel polymer 1.19. Chapter 3 deals with the preparation of oligo(methylenephosphine)s. These short-chained oligomers were mainly characterized using MALDI-TOF MS. In Chapter 4, I switch gears somewhat and describe the metallation of poly(methylenephosphine) using classical coordination chemistry with group 6 metal carbonyl complexes. Once metal incorporation was References start on page 22 Chapter 1 21 shown to be possible as a proof of principle, a more valuable metal system was investigated. Complexation of the polymer and its model compound to Au(l) is described in Chapter 5. Chapter 6 deals with utilizing the straightforward functionalization of poly(methylenephosphine)s with Au(l). Therein, we report the preparation of block-copolymers, polyisoprene-b poly(methylenephosphine)s, their coordination to gold and their solution self-assembled macromolecular structures. Finally, Chapter 7 concludes the findings of this thesis and proposes some future directions for this research. References start on page 22 Chapter 1 22 1.6 References 1. J. Emsley, The shocking histoiy of phosphorus a biography of the devThs element. Pan: London, 2001. 2. Republic of Nauru: Permanent Mission to the United Nations. http://www. un. int/nauru/overview. html 3. F. A. Cotton, G. Wilkinson, Advanced Inorganic Chemistiy. 5th ed.; John Wiley & Sons: Toronto, 1988; p 386. 4. J. Emsley, D. Hall, The Chemistiy of Phosphorus: EnvironmentaI Organic, Inorganic, BiochemicaI and Spectroscopic Aspects. John Wiley & Sons: New York, 1976. 5. G. Becker. Z. Anorg. AlIg. Chem. 1976, 423, 242. 6. K. B. Dillon, F. Mathey, J. F. Nixon, Phosphorus: The Carbon Copy. Wiley: West Sussex, 1998. 7. F. Mathey. Angew. Chem. mt. Ed. 2003, 42, 1578. 8. R. Appel, Multiple Bonds and Low Coordination in Phosphorus Chemistiy. Thieme: Stuttgart, 1990; p 157. 9. P. Le Floch. Coord. Chem. Rev. 2006, 250, 627. 10. M. Freytag, S. Ito, M. Yoshifuji. Chem. Asian J. 2006, 1, 693. 11. B. Deschamps, X. Le Goff, L. Ricard, P. Le Floch. Heteroat. Chem. 2007, 18, 363. 12. L. Weber. Angew. Chem. mt. Ed. 2002, 41, 563. 13. J. Dugal-Tessier, G. R. Dake, D. P. Gates. Organometaiics 2007, 26, 6481. 14. T. Baumgartner, R. Réau. Chem. Rev. 2006, 106, 4681. 15. T. Baumgartner, R. Réau. Chem. Rev. 2007, 107, 303. 16. V. A. Wright, B. 0. Patrick, C. Schneider, D. P. Gates. J. Am. Chem. Soc. 2006, 128, 8836. 17. V. A. Wright, D. P. Gates. Angew. Chem. mt. Ed. 2002, 41, 2389. 18. K. J. T. Noonan, D. P. Gates. Angew. Chem. Int. Ed. 2006, 45, 7271. 19. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. References start on page 22 Chapter 1 23 20. C.-W. Tsang, B. Baharloo, D. Riendl, M. Yam, D. P. Gates. Angew. Chem. mt. Ed. 2004, 43, 5682. 21. K. J. T. Noonan, B. 0. Patrick, D. P. Gates. Chem. Commun. 2007, 3658. 22. Canadian Plastic Products lndustiy. Industy Canada. www.ic.qc.ca/epic/site/plastics plastigues. nsf/en/plO 1 406e. html (accessed Feb 2008). 23. J. E. Mark, H. R. Ailcock, R. West, Inorganic Polymers. 2nd Edition ed.; Oxford University Press: 2005. 24. I. Manners. Angew. Chem. Int. Ed. EngI. 1996, 35, 1603. 25. R. D. Archer, Inorganic and Organometallic Polymers. Wiley-VCH: New York, 2001. 26. V. Chandrasekhar, Inorganic and Organometallic Polymers. 1st ed.; Springer: New York, 2005. 27. K. J. T. Noonan, D. P. Gates. Annu. Rep. Prog. Chem., Sect. A: lnorg. Chem. 2007, 103, 407. 28. J. E. Mark, H. R. Alicock, R. West, Inorganic Polymers. 2nd ed.; Oxford University Press: 2005; p 272. 29. H. R. Allcock. Curr. Opin. Solid State Mater. Sci. 2006, 10, 231. 30. H. R. Allcock, Chemistry and Applications of Polyphosphazenes. Wiley: New York, 2003. 31. K. Huynh, A. J. Lough, I. Manners. J. Am. Chem. Soc. 2006, 128, 14002. 32. K. Naka, T. Umeyama, A. Nakahashi, Y. Chujo. Macromolecules 2007, 40, 4854. 33. Y. Morisaki, Y. Ouchi, K. Tsurui, Y. Chujo. Polym. Bull. 2007, 58, 665. 34. L. A. Vanderark, T. J. Clark, E. Rivard, I. Manners, J. C. Slootweg, K. Lammertsma. Chem. Commun. 2006, 3332. 35. M. Raab, G. Schick, R. Fondermann, M. DoIg, W. Henze, U. Weynand, R. M. Gschwind, K. Fischer, M. Schmidt, E. Niecke. Angew. Chem. Int. Ed. 2006, 45, 3083. 36. H.-C. Su, 0. Fadhei, C.-J. Yang, T.-Y. Cho, C. Fave, M. Hissler, C.-C. Wu, R. Réau. J. Am. Chem. Soc. 2006, 128, 983. 37. T. Baumgartner, W. Wilk. Org. Lett. 2006, 8, 503. References start on page 22 Chapter 1 24 38. T. Baumgartner, W. Bergmans, T. Karpati, T. Neumann, M. Nieger, L. Nyulaszi. Chem.-Eur. J. 2005, 11, 4687. 39. T. J. Clark, J. M. Rodenzo, S. B. Clendenning, S. Aouba, P. M. Brodersen, A. J. Lough, H. E. Ruda, I. Manners. Chem. Eur. J. 2005, 4526, 4526. 40. L. B. Han, Z. X. Huang, S. Matsuyama, Y. Ono, C. Q. Zhao. J. Polym. Sd., Part A: Polym. Chem. 2005, 43, 5328. 41. S. Kawasaki, J. Ujita, K. Toyota, M. Yoshifuji. Chem. Lett. 2005, 34, 724. 42. R. C. Smith, J. D. Protasiewicz. J. Am. Chem. Soc. 2004, 126, 2268. 43. Z. Jin, B. L. Lucht. J. Organomet. Chem. 2002, 653, 167. 44. B. L. Lucht, N. 0. St. Onge. Chem. Commun. 2000, 2097. 45. Z. Jin, B. L. Lucht. J. Am. Chem. Soc. 2005, 127, 5586. 46. R. C. Smith, J. D. Protasiewicz. Eur. J. lnorg. Chem. 2004, 998. 47. R. C. Smith, X. F. Chen, J. D. Protasiewicz. lnorg. Chem. 2003, 42, 5468. 48. J. Casado, R. Réau, J. T. L. Navarrete. Chem. Eur. J. 2006, 12, 3759. 49. H. C. Su, 0. Fadhel, C. J. Yang, T. Y. Cho, C. Fave, M. Hissler, C. C. Wu, R. Réau. J. Am. Chem. Soc. 2006, 128, 983. 50. S. Durben, Y. Dienes, T. Baumgartner. Org. Lett. 2006, 8, 5893. 51. Y. Dienes, M. Eggenstein, T. Neumann, U. Englert, T. Baumgartner. Dalton Trans. 2006, 1424. 52. Y. Dienes, S. Durben, T. s. Krpti, T. Neumann, U. Englert, L. s. NyuIszi, T. Baumgartner. Chem. Eur. J. 2007, 13, 7487. 53. H.-S. Na, Y. Morisaki, Y. Aiki, Y. Chujo. Polym. Bull. 2007, 58, 645. 54. Y. Ouchi, Y. Morisaki, T. Ogoshi, Y. Chujo. Chem. Asian J. 2007, 2, 397. 55. Y. Morisaki, Y. Ouchi, K. Naka, Y. Chujo. Chem. Asian J. 2007, 2, 1166. 56. Y. Morisaki, Y. Ouchi, K. Tsurui, Y. Chujo. J. Polym. Sd., Part A: Polym. Chem. 2007, 45, 866. 57. B. H. Gillon, D. P. Gates. Chem. Commun. 2004, 1868. References start on page 22 Chapter 1 25 58. M. Yoshifuji. J. Chem. Soc., Dalton. Trans. 1998, 3343. 59. T. Provder, Size Exclusion Chromatography: Methodology and Characterization of Polymers and Related Materials. American Chemical Society: Washington, D.C., 1984. 60. S. F. Macha, P. A. Limbach. Cu,r Opin. Solid State Mater. Sci. 2002,6, 213. 61. M. J. Stump, R. C. Fleming, W. H. Gong, A. J. Jaber, J. J. Jones, C. W. Surber, C. L. Wilkins. AppI. Spectrosc. Rev. 2002, 37, 275. 62. S. D. Hanton. Chem. Rev. 2001, 101, 527. 63. H. R. Allcock, C. R. de Denus, R. Prange, W. R. Laredo. Macromolecules 2001, 34, 2757. 64. B. M. White, W. P. Watson, E. E. Barthelme, H. W. Beckham. Macromolecules 2002, 35, 5345. 65. F. H. KOhler, A. Schell, B. Weber. Chem. Eur. J. 2002, 8, 5219. 66. J. C. Blais, C. 0. Turrin, A. M. Caminade, J. P. Majoral. Anal. Chem. 2000, 72, 5097. 67. V. Marvaud, D. Astruc, E. Leize, A. Van Dorsselaer, J. Guittard, J. C. Blais. NewJ. Chem. 1997, 21, 1309. 68. C.-W. Tsang, C. A. Rohrick, T. S. Saini, B. 0. Patrick, D. P. Gates. Organometallics 2004, 23, 5913. 69. K. J. T. Noonan, B. Feldscher, J. J. Kingsley, M. Yam, J. I. Bates, D. P. Gates. Dalton Trans. 2008, Accepted. 70. G. Odian, Principles of Polymerization. Wiley & Sons: New York, 1991. 71. 0. W. Webster. Science 1991, 251, 887. 72. N. Hadjichristidis, M. Pitsikalis, S. Pispas, H. latrou. Chem. Rev. 2001, 101, 3747. 73. K. Matyjaszewski, J. Xia. Chem. Rev. 2001, 101, 2921. 74. J. P. Kennedy. J. Polym. ScL, Part A: Polym. Chem. 1999, 37, 2285. 75. 0. W. Webster. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2855. 76. M. Kamigaito, T. Ando, M. Sawamoto. Chem. Rev. 2001, 101, 3689. 77. C. J. Hawker, A. W. Bosman, E. Harth. Chem. Rev. 2001, 101, 3661. 78. J. C. Saam, D. J. Gordon, S. Lindsey. Macromolecules 1970, 3, 1. References start on page 22 Chapter 1 26 79. K. Sakamoto, K. Obata, H. Hirata, M. Nakajima, H. Sakurai. J. Am. Chem. Soc. 1989, 111, 7641. 80. K. J. T. Noonan, D. P. Gates. Macromolecules 2008, 41, 1961. 81. M. Levy. Polymers For Advanced Technologies 2007, 18, 681. 82. E. Vanzo. J. Polym. Sd., Part A: Polym. Chem. 1966, 1727. 83. I. W. Hamley. Nanotechnology 2003, 14, R39. 84. I. W. Ham ley, Block Copolymers in Solution: Fundamentals and Applications. Wiley: 2005. 85. I. Manners. J. Polym. Sd. Part A: Polym. Chem. 2002, 40, 179. 86. D. A. Rider, I. Manners. Polymer Reviews 2007, 47, 165. 87. J. Massey, K. N. Power, I. Manners, M. A. Winnik. J. Am. Chem. Soc. 1998, 120, 9533. 88. L. Cao, I. Manners, M. A. Winnik. Macromolecules 2001, 34, 3353. 89. L. Cao, I. Manners, M. A. Winnik. Macromolecules 2002, 35, 8258. 90. X. Wang, G. Guerin, H. Wang, Y. Wang, I. Manners, M. A. Winnik. Science 2007, 317, 644. 91. X. Wang, M. A. Winnik, I. Manners. Macromol. Rapid Commun. 2002, 23, 210. 92. H. Wang, M. A. Winnik, I. Manners. Macromolecules 2007, 40, 3784. 93. X. Wang, M. A. Winnik, I. Manners. Macromolecules 2005, 38, 1928. 94. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. References start on page 22 27 Chapter 2 Molecular Studies of the Initiation and Termination Steps of the Anionic Polymerization of PC bonds* 2.1 Introduction Five years ago, our group reported the polymerization of phosphaalkene 2.1 to afford poly(methylenephosphine) 2.2; a new functional phosphine polymer with alternating phosphorus and carbon atoms in the main chain (see Scheme 2.1).1 Our initial studies suggested that the polymerization of 2.1 required high temperatures (> 150 °C), long reaction times, and high concentrations, even in the presence of radical or anionic initiators. It was subsequently shown that short-chain oligomers could be obtained when 2.1 was treated with anionic initiators (e.g. MeLi) in THF solution at ambient temperature.2The oligomerization of phosphaalkenes is further discussed in Chapter 3. This solution oligomerization has now been extended to the living anionic polymerization of monomer 2.1 to afford homopolymer 2.2 and block copolymers with controllable architectures.3In order to further develop the living anionic polymerization of P=C bonds and to prepare end-functionalized polymers it is necessary to understand fully the mechanism of anionic chain growth for 2.1. Ph F Phi / Radical or Anionic I .. P=C 4—P—c / \ Initiator I I IMes Ph [Mes Ph 2.1 2.2 Scheme 2.1 * A version of this chapter has been published. Bronwyn H. Gillon, Kevin J.T. Noonan, Bastian Feldscher, Jennifer M. Wissenz, Zhi Ming Kam, Tom Hsieh, Justin J. Kingsley, Joshua I. Bates, and Derek P. Gates, “Molecular studies of the initiation and termination steps of the anionic polymerization of P=C bonds,” Can. J. Chem. 2007, 85(12), 1045-1052. Copyright 2007 NRC Canada. Chapter 2 28 To our knowledge, there is little previous work on the reaction of P=C bonds with typical anionic initiators. Of particular relevance to our studies is the synthesis of phosphine oxide 25a from phosphaalkene 2.1. Specifically, treating 2.1 with MeLi followed by MeOH and air oxidation afforded 2.5a. Both the carbanion (2.3a) and the phosphine (2.4a) were the presumed intermediates in this reaction, however, no characterization data for these species were reported. In a separate study, the addition of MeLi to RP=C(SiMe3)2(R = Mes, t-Bu) afforded Li[MeRPC(SiMe3)2]which was characterized spectroscopically.57In several other studies, the addition of alkyllithium reagents across the P=C bond in phosphinines and triphosphafulvenes similarly afforded only the P-alkyl products.812 These regioselective additions are consistent with the predicted polarity of the P=C bond based on the electronegativities of phosphorus (ô+) and carbon (ô—). Interestingly, alkyllithium reagents can also react with phosphaalkenes with preservation of the P=C bond.1316 In this chapter, the molecular model studies of the initiation and termination steps of the anionic polymerization of 2.1 are discussed. Two initiators were chosen for the present studies: MeLi and n-BuLi. Although n-BuLi is a better initiator for the anionic polymerization of 2.1 , MeLi is optimal in molecular model studies since the small Me-substituent is more likely to afford crystalline products. The new compounds obtained may be utilized as molecular models for the development of the coordination chemistry of polymer 2.2. Moreover, the development of methods to functionalize the endgroups in poly(methylenephosphine)s will afford telechelic polymers which may function as precursors to novel block copolymers. References start on page 45 Chapter 2 2.2 Preparation of Model Compounds 29 Ph / P=C / \ Mes Ph 2.lN H20 or,/ Mel Ph L/ L/ R—P—C—H R—P—C—Me I I I I Mes Ph Mes Ph R = Me 2.4a R = Me 2.6a R = n-Bu 2.4b HO or ‘air F Phj Rj-P-C4H I Mes Ph I L In 2.2 Ph A(n-1)1 RPC:eeLi 2. MeOH I -LiOMe Mes Ph R = Me 2.3a R = n-Bu 2.3b \clP(NEt)>’..qeR’SiCl \LiCl -Lid Ph PhMe I.. .. II R—P—C—P—NEt2 R—P—C—Si—R I I I I I Mes Ph NEt2 Mes Ph Me R = Me 2.7a R = Me; R = H 2.8a R=R=Me 2.9a O Ph II I R—P—C—H I I Mes Ph R = Me 2.5a Scheme 2.2 2.2.1 General Preparation of Carbanions Li[Mes(R)P—CPh2](2.3a, R = Me; 2.3b, R = Bu) A pale yellow solution of phosphaalkene 2.1 (prepared according to literature methods)17 18 in THF was treated with a slight excess of alkyllithium at low temperatures, resulting in an immediate colour change to deep red. Analysis of the reaction mixture using 31P{1H} NMR spectroscopy confirmed the complete conversion of 2.1 to the carbanion 2.3a (ö References start on page 45 Chapter 2 30 -44.3) orto 2.3b (ô = -31.1). These intermediates are believed to be related to the active species in the anionic polymerization of 2.1 (see Scheme 2.2) 2.2.2 Preparation of Mes(Me)P—CPh2H(2Aa) Carbanion 2.3a was quenched by addition of water or methanol to afford model compound 2.4a (ô31p = —24.0). Additionally, 13C{H}, 1H NMR spectroscopy and low resolution El MS (M, m/z = 332) were consistent with the proposed structure (see Scheme 2.2). The 1H NMR spectrum of 2.4a shows a notably downfield doublet resonance at ô = 4.87 (2JHp = 5 Hz) which is assigned to the benzylic proton. Importantly, a doublet resonance is observed in the alkyl region (ó = 1 .31, 2JHP = 6 Hz) which is attributed to the P-CH3 protons. Clearly, the addition of the methyl anion occurred regioselectively at the phosphorus atom of the P=C bond. Also found in the alkyl region are sharp singlets at 2.52 and 2.21 ppm that are assigned to the ortho and para-methyl protons of the mesityl group, respectively. Clear, colourless crystals suitable for X-ray diffraction were obtained from the slow evaporation of a concentrated solution of 2.4a in hexanes. The solid-state molecular structure and the crystallographic data appear in Figure 2.1 and Table 2.1, respectively. The solid-state molecular structure provides insight into the steric nature of the substituents around the P atom. Importantly, the cone angle of model compound 2.4a was estimated to be roughly 140° which is comparable to that of triphenyiphosphine (1450).19 The ligand cone angle 9 was estimated according to Tolman’s method for an unsymmetrical phosphine (PR1R23) ligand. The cone angle is the apex angle of a cylindrical cone centrally positioned 2.28 A away from the P. The edges of the cylindrical cone encompassed the van der Waals radii of the outermost atoms on each substituent on P. Using X-ray crystallographic data for 2.4a, 9 was estimated to be ca. 140°. The P—C bond lengths (A) fall into the typical range for P—C single bonds [P(1)—C(1) = 1.852(1); P(1)—C(10) 1.840(2); P(1)—C(11) = 1.882(1)1.20 Presumably, steric repulsion is References start on page 45 Chapter 2 31 responsible for the slight elongation of P(1)—C(11). For comparison, related P—C bonds where the carbon atom has two aryl substituents (i.e. RArP—CHPh2)are ca. 1 .90 A in length.21 22 Figure 2.1 The solid-state molecular structure of 2.4a (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1 )—C(1) = 1.852(1); P(1)—C(1 0) = 1.840(2); P(1)—C(11) = 1.882(1). Selected bond angles (°): C(10)—P(1)—C(1) = 107.34(7); C(1 0)—P(1)—C(1 1) = 100.51(7): C(1)—P(1)--C(1 1) = 100.18(6). 2.2.3 Preparation of Mes(n-Bu)P—CPh2H(2.4b) Using n-BuLi as the initiator for 2.1, the related compound 2.4b, was prepared by terminating carbanion 2.3b with water, Scheme 2.2. Again, the reaction was quantitative as suggested by 31P{1H} NMR spectroscopy (ô = —12.4). Upon workup, analytically pure 2.4b was isolated in good yield (85 %) as a yellow oil. Akin to the Me-initiated 2.4a, the 1H NMR spectrum of n-Bu-initiated 2.4b showed a characteristic doublet resonance at 4.91 ppm(2JHp = 5 Hz) which is assigned to the benzylic proton. A sharp singlet at 2.20 ppm is assigned to the para methyl mesityl protons. In contrast to the sharp signals observed for Z4a, the ortho-methyl mesityl protons in 2.4b gives rise to a broad signal centered at 2.56 ppm. The broadening may be a consequence of restricted rotation about the P-C (ipso-mesityl) bond. Additionally, there is considerable broadening of the resonance assigned to the ortho-methyl mesityl carbon (ô = 23.8) in the 13C{’H} NMR spectrum. Another interesting feature in the 1H NMR spectrum of 2.4b involves the diastereotopic protons adjacent to the stereogenic phosphorus atom. In particular, C21 Cl 0 References start on page 45 Chapter 2 32 they give rise to two sets of multiplets (ö = 2.02 and 1.55). Assignments were made with the help of1H- COSY, HMQC, and HMBC experiments. 2.2.4 Preparation of Mes(Me)P—CPh2e(2.6a) With the purpose of modelling the functionalization of the endgroups of polymer 2.2, the termination of the methyl-initiated species 2.3a with different electrophiles is investigated. These end-functionalized products are envisaged as model systems for the heretofore-unknown telechelic polymers. In order to obtain Me-terminated 2.6a, carbanion 2.3a was quenched with neat methyl iodide, affording an orange solution (ö31p = — 13.5) (see Scheme 2.2). Model compound 2.6a was characterized by 1H, 13C{H}, 31P NMR spectroscopy and low resolution El MS (M, m/z = 346). Although analytically pure material could not be isolated, analysis by NMR spectroscopy suggests 95% or higher purity from other phosphorus-containing compounds. As with butyl initiated 2.4b, the signal assigned to the ortho-methyl mesityl protons in the 1H NMR spectrum = 1.87) is broad. Two doublets are observed in the alkyl region (1.70 ppm, 2JpH = 13Hz; 1.39 ppm, 3JPH = 7 Hz). The resonance with the larger P-H coupling constant is assigned to the P CH3 group whilst the other signal is attributed to the quenching methyl group. 2.2.5 Preparation of Mes(Me)P—CPh2—P(NEt2) (2.7a) Phosphine end-functionalized 2.7a was prepared by quenching 2.3a with a solution of CIP(NEt2)(1.1 equiv) in THF (see Scheme 2.2). The crude diphosphine was purified by washing the solid with cold hexanes (— 78 °C). As expected, the 31P{1H} NMR spectrum of the product showed that the phosphorus atoms are inequivalent and give rise to two doublets (P2: 8 = 105.3; P1: 8 = —9.4; 2J, = 205 Hz). The observed coupling constant is considerably larger than in a related asymmetric diphosphine[Ph2PCH(NC4H,2Jpp = 147 Hz]23 and suggests References start on page 45 Chapter 2 33 some direct P-P interaction (vide infra). In contrast to the broadened signals observed in the spectra for 2.4b and 2.6a, the resonances due to the ortho-methyl mesityl groups in 2.7a are inequivalent in both the 1H and 13C{H} NMR spectra. Presumably, this results from restricted rotation due to the increased steric bulk in 2.7a. X-ray quality crystals of 2.7a were grown from toluene. Metrical parameters and crystallographic data are shown in Figure 2.2 and Table 2.1, respectively. Interestingly, the P(1 )—C(1 1 )—P(2) bond angle (99.17(9) 0) is more acute than a typical sp3 hybridized C atom. This small bond angle combined with the short PP distance [P(1)—P(2) = 2.966(1) A; cf. rVdW = 3.7 Al may account for the large coupling constant(2Jpp) observed in solution. To confirm that the structure observed in solution is the same as that in the solid state, a solid state 31P NMR spectrum was recorded which revealed a similarly large coupling constant (2Jpp = 255 Hz). Surprisingly, both nitrogen atoms are planar with the sum of the bond angles at nitrogen totaling 3600. Likely, the planarity is due to donation of nitrogen electron density into P(2) and this notion is reinforced by the short P—N distances; P(2)—N(1) = 1.678(2) and P(2)—N(2) = 1.689(2) A.24 In related bis(amino)phosphines, the P—N bonds are often shortened to 1.6 — 1.7 A which is accompanied by planarity at nitrogen.2528 Also, the addition of the functional end-group results in the lengthening of the backbone P—C bonds to beyond typical single bonds [P(1 )—C(1 1) = 1.942(2) and P(2)—C(1 1) = 1.951(2) A].24 References start on page 45 Chapter 2 34 08 G3 C7 C29 - 02804 05 ci N2 C25P1 P2 -., c3-t 09 C24 017 019 Ni 030 010 020 012 026 016 015,- cia 5)C27 023 :014 -C21 022 Figure 2.2 The solid-state molecular structure of 2.7a (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1)—C(1) = 1.851 (2); P(1)—C(10) = 1.843(2); P(1)—C(11) = 1.942(2); P(2)—C(11) = 1.951(2); P(2)—N(1) = 1.678(2); P(2)—N(2) = 1.689(2); P(1)P(2) = 2.966(1). Selected bond angles (°): C(1)—P(1)—C(10) = 105.1(1); C(1)— P(1)—C(11) = 110.05(9); C(10)—P(1)—C(11) = 101.5(1); P(1)—C(11)—P(2) = 99.17(9); C(11)— P(2)—N(1) = 103.87(8); C(11)—P(2)—N(2) = 105.97(9); N(1)—P(2)—N(2) 107.80(9); angles about N(1) atom = 359.7(3)°; angles about N(2) atom = 359.4(3)°. 2.2.6 Preparation of Mes(Me)P—CPh2—SiMeH(2.8a) and Mes(Me)P—CPh2--SiMe (2.9a) Addition of Me2HSiCI or Me3SiCI to carbanion 2.3a produced silane end-functionalized 2.8a and 2.9a, respectively (see Scheme 2.2). These compounds formed quantitatively from 2.3a according to the 31 NMR spectra of the reaction mixtures (2.8a ô = —23.9, 2.9a ó = —25.4), however their high solubility resulted in low isolated yields of pure product after recrystallization (2.8a 50 %, 2.9a 14%). Analogous to that mentioned previously for 2.7a, the 1H NMR signals for the ortho-methyl protons of the mesityl group were inequivalent and two broad signals were observed (2.8a ô = 3.1 and 1.1; 2.9a ó = 2.8 and 0.9). Importantly, for 2.8a a singlet resonance was observed at 4.58 ppm which is assigned to the Si—H proton. Although the methyl protons of the SiMe3 moiety are equivalent in 2.9a, the methyl protons of the SiMe2H moiety are inequivalent in 2.Ba. In particular, the SiHMe2 group in 2.8a shows two sets of doublets at 0.10 ppm (3JHH = 3 Hz) and 0.05 ppm (3JHH = 3 Hz). Both 2.8a and 2.9a were also characterized by low resolution El MS (M: m/z = 390 2.8a; 404 2.9a), elemental analysis (2.8a only) and X-ray crystallography. References start on page 45 Chapter 2 35 Crystals of 2.8a and 2.9a were obtained from hexanes. The solid-state molecular structures and important bond lengths and angles are given in Figure 2.3 and Figure 2.4, and additional data is provided in Table 2.1. As with the other model compounds, 2.4a and 2.7a, the P—CPh2H bond lengths [28a: P(1)—C(11) = 1.902(1); 2.9a: P(1)—C(1) = 1.920(1) A] are significantly longer than the two other P—C bonds [avg. for 2.8a and 2.9a = 1.835(5) A]. Similar to that discussed previously, this elongation may be attributed to steric congestion. The small size of the S1Me2H moiety compared to the SiMe3 moiety is reflected by the smaller P—C—Si bond angles in 2.8a [2.8a: P(1)—C(11)—Si(1) = 106.57(6)°; 2.9a: P(1)—C(1)—Si(1) = 111.52(6)°]. Remarkably, these angles are much larger than the analogous angle in diphosphine 2.7a [P(1)— C(11)—P(2) = 99.17(9)°] in which there is likely a weak PP interaction. 07 C257) H24 CSc4c2 023 021 04 Cli 016 015 _ Figure 2.3 The solid-state molecular structure of 2.8a (50% probability ellipsoids). Hydrogen atoms are omitted for clarity except for H(1). Selected bond lengths (A): P(1)—C(1) = 1.834(2); P(1 )—C(1 0) = 1.824(2); P(1 )—C(1 1) = 1.902(1); Si(1 )—C(1 1) = 1.910(2); Si(1 )—C(24) = 1 .855(2); Si(1)—C(25) = 1.849(2); Si(1)—H(1) = 1.36(2). Selected bond angles (°): C(1)—P(1)—C(10) = 107.39(7); C(1)—P(1)—C(11) = 106.61 (6); C(10)—P(1)—C(11) = 102.33(7); P(1)—C(11)—Si(1) = 106.57(6). References start on page 45 Chapter 2 36 Figure 2.4 The solid-state molecular structure of 2.9a (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1)—C(1) = 1.920(1); P(l)—C(2) = 1.839(2); P(1)—C(3) = 1.844(1); C(1)—Si(1) = 1.944(1). Selected bond angles (°): C(1)—P(1)--C(2) = 101 .38(6); C(2)—P(1)—C(3) 106.97(7); C(1)—P(1)—C(3) = 107.53(5); P(1)—C(1)—Si(1) = 111.52(6). 2.3 Conclusion In closing, the initiation and termination steps in the anionic polymerization of P=C bonds have been modeled. The initiation step was investigated through the stoichiometric reaction of MesP=CPh2(2.1) with RLi (R = Me or n-Bu). In all cases, the addition was highly regioselective with the formal attack of R at phosphorus to give the carbanion Li[Mes(R)P—CPh2}(2.3a/2.3b). To simulate the termination step in the anionic polymerization of 2.1, carbanions 2.3a and 2.3b were quenched in situ with electrophiles such as: a proton (from H20 or MeOH), a methyl cation (from Mel), a phosphenium moiety (from CIP(NEt2)or a silylium moiety (from Me2HSiCI or Me3SiCI). In this way, five new tertiary phosphines and one diphosphine were accessed and will function as models for poly(methylenephosphine)s. Four of these molecules were characterized crystallographically. In future, this chemistry will be extended to the polymer 2.2 with the goal of preparing end-functional telechelic polymers that can be employed in the synthesis of novel block copolymers. In addition, these new compounds will be used to prepare molecular complexes References stan on page 45 Chapter 2 37 that will serve as models for macromolecular coordination complexes of poly(methylenephosphine)s. 2.4 Experimental 2.4.1 Materials and General Procedures All manipulations of oxygen- and/or moisture-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk and/or glovebox techniques. Hexanes, diethyl ether and dichloromethane were deoxygenated with nitrogen and dried by passing through a column containing activated alumina. THF was freshly distilled from sodium/benzophenone ketyl. Distilled water and methanol were degassed prior to use. CDCI3 was distilled from P205 and degassed. CD2I and C6D were purchased from Cambridge Isotope Laboratories and were dried over molecular sieves (3 A). PCI3, HNEt2, Mel, MeLi (1.6 M in Et20), n-BuLi (1.6 M in hexanes) were purchased from Aldrich and used as received. Alkyllithium reagents were titrated prior to use against a THE solution of N-benzylbenzamide. Me2SiHCI and Me3SiCI were distilled over CaH2 and degassed prior to use. MesP=CPh2,2.1, was prepared following literature procedures.17’18 31P, 1H and 13C{H} NMR spectra were recorded on BrukerAV300 orAV400 spectrometers at room temperature. Solid-state NMR experiments were performed on a Bruker Avance DSX400 spectrometer operating at frequencies of 400.13 MHz and 161.98 MHz for 1H and 31P respectively, using a Bruker triple resonance 4mm MAS probe. Chemical shifts for 31P NMR spectra are reported relative to H3P04as an external standard (85% in H20, 3 = 0). Chemical shifts for 13C{H} NMR spectra are reported relative to CD2I (ó = 54.0) or CDCI3 (ó = 77.2). Chemical shifts for 1H NMR spectra are reported relative to residual CHDCI2 (8 = 5.32) or CHCI3 (8 = 7.27) or C6HD5 (8 = 7.16). Mass spectra were recorded on a Kratos MS 50 References start on page 45 Chapter 2 38 instrument and performed by the UBC Mass Spectrometry Centre staff. Elemental analyses were performed by Mr. Minaz Lakha in the Departmental Microanalysis Facility. 2.4.2 Preparation of Mes(Me)P-CPh2Li(2.3a) Typical Procedure: To a cooled (—78°C) solution of phosphaalkene 2.1 (2.06 g, 6.5 mmol) in THF or Et20 (Ca. 150 mL) was slowly added MeLi in Et20 (5.2 mL, 1.5 M, 7.8 mmol). Upon addition, there was an immediate colour change from yellow to dark red. The cooled solution was stirred for 1 h and then warmed to room temperature whereupon it was stirred for an additional 1 h. The concentration for use in subsequent steps was 0.043 M (see preparation of 2.4a). 31P{’H} NMR spectroscopy of an aliquot removed from the reaction mixture showed a single signal (ö = —42.0 in THF, —44.3 in Et20). 2.4.3 Preparation of Mes(n—Bu)P-CPh2Li(2.3b) To a stirred yellow solution of 2.1 (0.64 g, 2.0 mmol) in Et20 (ca. 30 mL) at —78 °C, a solution of n-BuLi in hexanes (3.0 mL, 1.4 M, 4.2 mmol) was added rapidly. The mixture turned bright orange upon addition, and was stirred at —78 °C for 30 mins. Subsequently, the solution was warmed to room temperature whereupon it was stirred for an additional 1 h. Over this time, the solution turned from bright orange to red. The reaction was monitored by 31P{1H} NMR spectroscopy, affording a single signal for 2.3b (ó = —31.1). This solution was used in the preparation of 2.4b. 2.4.4 Preparation of Mes(Me)P—CPh2H(2.4a) To a stirred red solution of 2.3a in THE (150 mL, 0.043 M, 6.4 mmol), was added degassed MeOH (0.4 mL). From the resulting colourless solution, THE was removed in vacuo affording a white solid residue. The soluble fraction was extracted into hexanes (3 x 50 mL) and the resulting solution was evaporated to dryness affording a yellow oil. The oil was dissolved in References start on page 45 Chapter 2 39 minimal hexanes at reflux (Ca. 5 mL). The solution was slowly cooled to room temperature to afford colourless crystals suitable for X-ray diffraction. Yield: 0.50 g (24 %). 31P{1H} NMR (CD2I, 121.5 MHz): ô =—24.0 (s); 1H NMR (CD2I,300 MHz): 7.58— 7.10 (m, 1OH, Ph-I-i), 6.77 (s, 2H, m-Mes-I-I), 4.87 (d, 2JHP= 5 Hz, 1H, CPh2I-i), 2.52 (s, 6H, o Mes-CH3), 2.21 (s, 3H, p-Mes-CH), 1.31 (d, 2JHP= 6Hz, 3H, P—CH3);13C{H} NMR (CD2I, 75.5 MHz) (unassigned): ö = 146.0 (d, Jcp= 15 Hz), 144.3, 144.1, 144.0, 143.8(4 x s or 2 x d), 140.4 (s), 131 .4—1 27.4 (m), 52.8 (d, 1Jcp= 16 Hz), 24.6, 24.4(2 x s or 1 x d), 10.8 (d, 1Jcp 18 Hz, P—CH3); MS (El, 70 eV): m/z (%) = 332 (28) [Mt], 167 (100) [CHPh2], 165 (44) [M—CHPh2], 119 (12) [Mes}. 2.4.5 Preparation of Mes(n-Bu)P—CPh2H(2.4b) To a stirred solution of carbanion 2.3b in Et20 (2.0 mmol; preparation described above) was added dropwise degassed H20 until the reaction mixture turned pale yellow (several drops). The mixture was then dried with MgSO4,filtered, and the soluble fraction was extracted with Et20 (2 x 10 mL). The solvent was removed in vacuo at 60 °C to afford pure 2.4b as a pale yellow oil. Yield: 0.65 g (85 %). 31p NMR (CD2I, 121.5 MHz): ô = —12.4 (s); 1H NMR (CD2I,400 MHz) (assignments made with the aid of a COSY experiment): o = 7.57 (d, JHH = 8 Hz, 2H, Ph-I-i), 7.36 (m, 4H, Ph H), 7.24 (m, 1H, Ph-H), 7.15 (m, 2H, Ph-H), 7.06 (m, 1H, Ph-H), 6.77 (brs, 2H, m-Mes-H), 4.91 (d, 2JPH = 5 Hz, 1 H, CPh2H), 2.56 (br s, 6H, o-Mes-CH3),2.20 (s, 3H, p-Mes-CH3),2.02 (m, 1 H, P-CHI-i), 1.55 (m, 1H, P-CHH), 1.22 (m, 2H, CH3),1.08 (m, 2H, PCH2C), 0.74 (t, JHH = 7 Hz, 3H, CH23);13C{H} NMR (CD2I,100.6 MHz) (assignments made with the aid of HMQC/HMBC experiments): .3 143.6 (d, 2Jcp = 12 Hz, i-Ph-C), 143.4 (d, 2Jcp = 13 Hz, iPh-C), 139.5 (s, p-Mes-C), 130.2 (br, m-Mes-C), 129.7 (d, 1Jc’ = 26 Hz, i-Mes-C), 129.5 (d, 3Jcp = 9 References start on page 45 Chapter 2 40 Hz, o-Ph-C), 129.1 (s, o-Mes-C), 128.9 (d, 3Jcp = 9 Hz, o-Ph-C), 128.6 (s, p-Ph-C), 127.1 (d, 4Jcp = 2 Hz, rn-Ph-C), 126.5 (d, 4Jcp 2 Hz, rn-Ph-C), 51.6 (d, 1Jcp 16 Hz, CHPh2), 30.2 (d, = 20 Hz, P—CH2CH),25.5 (d, 1Jcp 16 Hz, P—CH2), 24.7 (d, 3JCp 13 Hz, CH23), 23.8 (v br, o-Mes-CH3),21.2 (s, p-Mes-CH3), 14.1 (s, CH23).MS (El, 70eV): mlz (%) = 374 (29) [M], 167 (100) [CPh2H], 151 (24) [PMes]. Anal.: calculated = C, 83.39%; H, 8.34%. Found = C, 83.81%; H, 8.34%. 2.4.6 Synthesis of Mes(Me)P—CPh2e(2.6a) Mel (1.6 mL, 26 mmol) was added dropwiseto a solution of 2.3a in Et20 (160 mL, 0.16 M, 26 mmol). The solution was stirred for 30 mm over which time its colour changed from red to orange. The solvent was removed in vacuo, the soluble fraction was extracted with hexanes, and evaporated to dryness. The product was purified by sublimation (140 °C, 0.1 mmHg). Yield = 5.7 g (64 %). 31P NMR (CDCI3, 121.5 MHz): o =—13.5 (s); 1H NMR (CDCI3,300 MHz): ó = 7.47 (m, 2H, Ph-H), 7.21 (m, 6H, Ph-H), 7.01 (m, 2H, Ph-H), 6.74 (s, 2H, Mes-H), 2.23 (s, 3H, p-Mes-CH), 1.87 (br, 6H, o-Mes-CH3), 1.70 (d, 2JPH = 13Hz, 3H, P—CH3), 1.39 (d, 3JpH = 7 Hz, 3H, CPh2H3);13C{H} NMR (CDCI3,75.5 MHz,) (unassigned): ô = 148.3 (d, Jcp= 14Hz), 146.3 (d, Jc = 16 Hz), 144.7 (s), 138.8 (s), 129.9 (s), 128.7—127.4 (m), 125.8 (s), 125.4 (s), 49.1 (d, Jc = 24 Hz), 28.8 (d, Jcp = 15 Hz), 23.5 (d, Jc= 18 Hz), 20.8 (s), 9.0 (d, 1Jcp= 21 Hz, P—CH3); MS (El, 70 eV): m/z (%): 346 (7) [Mt], 181 (100) [M— P(Mes)(CH3)], 165 (16) [M— CPh2(CH3)], 119 (5) [Mes]. 2.4.7 Preparation of CIP(NEt2) CIP(NEt2)was prepared from an adaptation of literature method.29 A solution of HNEt2 (190 mL, 1.84 mol) and CH2I (300 mL) was added slowly to a cooled (-78°C) solution of PCI3 References start on page 45 Chapter 2 41 (40 mL, 0.46 mol) in CH2I (250 mL). The reaction mixture was slowly warmed to room temperature and stirred for 1 h. Solvent was removed in vacuo. The resulting solid was extracted with hexanes (3 x 150 mL) and filtered. Hexanes were removed in vacuo. The resulting oil was distilled under vacuum yielding a colourless liquid (89 g, 92%). 31P{’H} NMR (C6D, 121.5 MHz): ô = 154.8. 1H NMR (C5D6, 300 MHz): ó 2.9—3.1 (m, 8H, CH2); 0.94 (t, 12H, 3JHH = 7 Hz, CH3). 2.4.8 Preparation of Mes(Me)P—CPh2—P(NEt2) (2.7a) To a cooled (0 °C) solution of 2.3a in THF (20 mL, 0.32 M, 6.4 mmol) was slowly added a solution of CIP(NEt2)(1.50 g, 7.1 mmol) in THE (20 mL). An aliquot was removed from the yellow reaction mixture for 31P NMR spectroscopic analysis. After evaporation of the volatiles in vacuo, the soluble fraction was extracted into toluene (30 mL). Removal of the toluene in vacuo afforded the crude product which was washed with hexanes at —78 °C two times. Crystals suitable for X-ray diffraction were obtained by slow evaporation of a saturated toluene solution. Yield: 1.73 g (54 %). 31P NMR (CDCI3, 121.5 MHz): 6 = 105.3 (d, 2Jpp= 205 Hz, PN2), —9.4 (d, 2Jpp= 205 Hz, PMes); 31P MAS NMR (8.5 kHz, recycle delay of 20s, 162.0 MHz): 6 = 108.3 (d, 2Jpp 255 Hz, PN2), -7.9 (d, 2Jpp = 255 Hz, PMes); 1H NMR (CDCI3,400 MHz) 6 = 7.79 (br, 2H, Ph-H), 7.43 (br, 2H, Ph-H), 7.28—7.14 (m, 6H, Ph-I-i), 6.90 (br, IH, m-Mes-I-i), 6.55 (br, 1H m-Mes-H), 3.01 — 2.52 (m, 11 H, NCH2 and o-Mes-CH3),2.25 (s, 3H, p-Mes-CH), 1.38 (d, 2JpH = 9 Hz, 3H, P CH3), 0.97 (m, 9H, NCH2CH3and o-Mes-CH3),0.84 (t, 6H, JHH= 7Hz, NCH2CH3);13C{H} NMR (CDCI3, 100.6 MHz) (assignments made with the aid of HMQC/HMBC experiments): 6 = 147.7 (s, o-Mes-C), 141.3 (br, Ar-C), 141.0 (t, Jpc= 5Hz, Ar-C), 138.6 (s, p-Mes-C), 131.9—127.6 (m, Ar-C), 126.9 (s, Ar-C), 126.4 (s, Ar-C), 124.8 (d, Jc= 4 Hz, Ar-C), 63.9 (dd, 1Jpc 46 and 34 References start on page 45 Chapter 2 42 Hz, P-CPh2-P), 44.5 (d, 2Jpc = 21 Hz, N-CH2), 44.0 (d, 2Jc = 20 Hz, N-CH2), 26.2 (br 5, O Mes-CH3), 21.6 (brs, o-Mes-CH3),21.0 (s, p-Mes-CH3), 14.5 (d, 3Jpc= 3Hz, NCH2CH3), 14.0 (d, 3Jpc = 5 Hz, NCH2CH3),9.7 (t, 1Jpc = 22 Hz, P—CH3). 2.4.9 Preparation of Mes(Me)P—CPh2—S1MeH(2.8a) To the red solution of 2.3a (40 mL, 0.079 M, 3.2 mmol) in THE (40 mL) neat Me2HSiCI (0.5 mL, 4.6 mmol) was added. An immediate colour change to yellow was observed. The solvent was removed in vacuo, and the residue was dissolved in hexanes (3 x 5 mL). Following solvent evaporation, the crude product was recrystallized from a minimal amount of hot hexanes in an inert atmosphere to afford large colourless crystals. Yield: 0.62 g (50%). 31p NMR (C6D, 121.5 MHz): ô =—23.9 (s); 1H NMR(C6D,300 MHz): ô = 7.70 (brs, 2H, Ph), 7.3—6.5 (m, IOH, Ph and m-Mes-H), 4.58 (brs, 1H, Si-H), 3.1 (brs, 3H, o-Mes-CH3),2.18 (s, 3H, p-Mes-CH), 1.45 (d, JPH = 8Hz, 3H, P—CH3), 1.1 (brs, 3H, o-Mes-CH3),0.10 (d, 3JHH = 3 Hz, 3H, Si—CH3), 0.05 (d, 3JHH = 3 Hz, 3H, Si—CH3);13C{H} NMR (CD2I,75.5 MHz) (assignments made with the aid of HMQC/HMBC experiments): ô = 147.6 (s, Mes-C), 147.3 (s, Mes-C), 142.5 (s, Mes-C), 142.1 (d, 1Jpc = 7Hz, i-Mes-C), 139.5 (s, Mes-C), 131 - 128 (m, Ph- C), 125.7 (s, Ph-C), 45.7 (d, 1Jpc = 45 Hz, PCPh2), 21.2 (s, p-Mes-CH3),7.81 (d, 1Jpc = 23 Hz, PCH3), -3.96 (d, 3Jpc = 5 Hz, SiCH3), —5.66 (d, 3Jpc = 6 Hz, SiCH3). MS (El, 70 eV): m/z (%) = 390 (38) [Mt], 375 (17) [Mt— CH3}, 332 (45) [M — SiMe2], 223 (69) [M — PMesMe], 167 (100) [CPh2]. Anal.: calculated = C, 76.88%; H, 8.00%. Found = C, 77.20 %; H, 7.64 %. 2.4.10 Preparation of Mes(Me)P—CPh2—SiMe3(2.9a) To the red solution of 2.3a in Et20 (22 mL, 0.14 M, 3.2 mmol) neat Me3SiCI (0.4 mL, 3.2 mmol) was added. An immediate colour change to yellow was observed accompanied by the formation of a white precipitate. The solution was filtered, the solvent was removed in vacuo, References start on page 45 Chapter 2 43 and the soluble fraction was extracted into Et20 (3 x 15 mL). The Et20 was removed in vacuo. The crude product was dissolved in a minimal amount of hexanes and slow evaporation afforded colourless crystals suitable for X-ray crystallography. Yield = 0.18 g (14 %) 31P NMR (CDCI3, 121.5 MHz): ô = —25.4 (s); 1H NMR (CDCI3,300 MHz): ó = 7.6 (br s, 2H, Ph), 7.3 — 7.1 (m, 8H, Ph). 6.8 (br s, 2H, m-Mes-H), 2.8 (br s, o-Mes-CH3),2.22 (s, 3H, p Mes-CH3), 1.33 (d, 2JPH = 8 Hz, 3H, P—CH3), 0.9 (br s, 3H, o-Mes-CH3),—0.02 (s, 9H, Si(CH3). MS (El, 70 eV): m/z (%) = 404 (100) [M], 389 (26) [M — CH3], 331 (44) [M — SiMe3], 167 (39) [M — CPh2SiMe3],73 (34) [M — MesMePCPh2]. 2.4.11 X-ray Crystallography Crystal data and refinement parameters are listed in Table 2.1.1 All single crystals were immersed in oil and mounted on a glass fiber. Data for 2.4a were collected by Dr. Brian 0. Patrick on a Bruker X8 APEX diffractometer with graphite-monochromated Mo Ka radiation. Data for 2.7a, 2.8a and 2.9a were collected by Joshua Bates on a Bruker X8 APEX II diffractometer. Data was collected and integrated using the Bruker SAINT software package.3° All structures were solved by direct methods and subsequent Fourier difference techniques and were refined anisotropically for all non-hydrogen atoms. Hydrogen atoms were included in idealized positions and refined isotropically except in 2.4a where they were not refined. All data sets were corrected for Lorentz and polarization effects. All calculations were performed using the SHELXTL crystallographic software package from Bruker-AXS.31 Compounds 2.4a, 2.8a and 2.9a did not show any crystallographic complexity. Compound 2.7a was a split crystal and cell_now (within SAINT) was used to determine that the second domain was rotated from the first domain by 4•40• Only the first domain was integrated ‘.cif files have been deposited with the Cambridge Crystallographic Database. Depository numbers: CCDC 655085 (2.4a), CCDC 655086 (2.7a), CCDC 655087 (2.8a), CCDC 655088 (2.9a). References start on page 45 Chapter 2 44 because there was little or no overlap between the two lattices. One of the ethyl groups was disordered at C(28) and C(29). The disorder at carbon was modeled by including two additional carbon atoms [C(28b) and C(29b)] and refining their respective populations [occupancy 0.78(2) and 0.22(2)]. Table 2.1 X-ray crystallographic data for 2.4a, 2.7a, 2.8a and 2.9a. 2.4a 2.7a 2.8a 2.9a formula C23H5P C31H44NP C25H31PS1 C26H33PS1 formula weight (g moF1) 332.40 506.62 390.56 404.58 crystal system monoclinic monoclinic monoclinic triclinic space group P21/c P21/n C2/c p1 colour colourless colourless colourless colourless a (A) 18.867(2) 9.528(5) 20.677(5) 9.165(5) b (A) 6.1290(5) 15.614(5) 8.955(5) 10.757(5) c(A) 16.618(2) 18.935(5) 23.987(5) 12.689(5) a(deg) 90 90 90 79.511(5) f3 (deg) 92.401(3) 94.828(5) 101.555(5) 71.398(5) y(deg) 90 90 90 75.169(5) V(A3) 1919.9(3) 2807.0(19) 4351(3) 1143.3(9) Z 4 4 8 2 T(K) 173(2) 173(2) 173(2) 173(2) 1i(Mo Ka) (cm1 1.44 1.77 1.89 1.82 crystal size (mm) 0.2 x 0.2 x 0.1 1.4 x 0.6x 0.4 1.4 x 1.2x 1.0 lOx lOx 0.5 dcalcd (g cm3) 1.150 1.199 1.192 1.175 20(max) (deg) 56.02 56.04 56.36 55.66 no. of rflns 26221 27995 54740 27579 no. of unique data 4580 6560 5281 5333 R(int) 0.040 0.050 0.033 0.024 rfln/param ratio 20.82 12.84 14.35 13.85 Ri, wR2 [I > 2.000(l)] a 0.0388, 0.0925 0.0418, 0.0944 0.0350, 0.0883 0.0306, 0.0820 Ri, wR2 (all data)a 0.0662, 0.1015 0.0787, 0.1149 0.0446, 0.0950 0.0354, 0.0862 GOF 1.053 1.017 1.029 1.054 a R1 = I F — I FI 1/ZI F01 ; wR2 = [Z(F0’— Fcz)z/Zw(Foz)9z. References start on page 45 Chapter 2 45 25 References 1. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. 2. B. H. Gillon, D. P. Gates. Chem. Commun. 2004, 1868. 3. K. J. T. Noonan, D. P. Gates. Angew. Chem. mt. Ed. 2006, 45, 7271. 4. T. A. van der Knaap, F. Bickeihaupt. Tetrahedron Lett. 1982, 23, 2037. 5. B.-L. Li, R. H. Neilson. lnorg. Chem. 1984, 23, 3665. 6. Z.-M. Xie, P. Wisian-Neilson, R. H. Neilson. Organometallics 1985, 4, 339. 7. R. Appel, C. Casser, F. Knoch. Chem. Ber. 1986, 119, 2609. 8. S. Ito, H. Miyake, M. Yoshifuji, T. HoItzI, T. Veszpremi. Chem.-Eur. J. 2005, 11, 5960. 9. S. Ito, H. Miyake, H. Sugiyama, M. Yoshifuji. Tetrahedron Lett. 2004, 45, 7019. 10. A. Moores, L. Ricard, P. Le Floch. Angew. Chem. mt. Ed. 2003, 42, 4940. 11. G. Markl, A. Merz. Tetrahedron Lett. 1968, 3611. 12. G. Marki, F. Lieb, A. Merz. Angew. Chem. mt. Ed. 1967, 6, 87. 13. S. Ito, M. Freytag, M. Yoshifuji. Dalton Trans. 2006, 710. 14. S. Ito, K. Toyota, M. Yoshifuji. Chem. Commun. 1997, 1637. 15. M. Yoshifuji, H. Kawanami, Y. Kawai, K. Toyota, M. Yasunami, T. Niitsu, N. Inamoto. Chem. Lett. 1992, 1053. 16. F. B. Simon J. Goede. Chemische Berichte 1991, 124, 2677. 17. G. Becker, W. Uhi, H.-J. Wessely. Z. Anorg. AlIg. Chem. 1981, 479, 41. 18. M. Yam, J. H. Chong, C.-W. Tsang, B. 0. Patrick, A. E. Lam, D. P. Gates. lnorg. Chem. 2006, 45, 5225. 19. C. A. Tolman. Chem. Rev. 1977, 77, 313. 20. “Characteristic Bond Lengths in Free Molecules” in CRC Handbook of Chemistry and Physics, 84th Edition. CRC Press: Boca Raton, FL, 2003. References start on page 45 Chapter 2 46 21. P. Maire, S. Deblon, F. Breher, J. Geier, C. Böhler, H. Ruegger, H. SchOnberg, H. Grützmacher. Chem. Eur. J. 2004, 10, 4198. 22. B. Manz, G. Maas. J. Chem. Soc., Chem. Commun. 1995, 25. 23. A. D. Burrows, M. F. Mahon, S. P. Nolan, M. Varrone. lnorg. Chem. 2003, 42, 7227. 24. L. Pauling, The Nature of the Chemical Bond and the Structure of Molecules and Crystals; An Introduction to Modern Structural Chemistry. 3rd ed.; Cornell University Press: New York, 1960; p 664. 25. P. W. Dyer, J. Fawcett, M. J. Hanton, R. D. W. Kemmitt, R. Padda, N. Singh. Dalton Trans. 2003, 104. 26. F. Dahan, P. W. Dyer, M. J. Hanton, M. Jones, D. M. P. Mingos, A. J. P. White, D. J. Williams, A.-M. Williamson. Eur. J. lnorg. Chem. 2002, 732. 27. I. Shevchenko, V. Andrushko, E. Lork, G.-V. Röschenthaler. Eur. J. lnorg. Chem. 2002, 2985. 28. S. A. Reiter, S. D. Nogai, H. Schmidbaur. Dalton Trans. 2005, 247. 29. P. G. Chantrell, C. A. Pearce, C. R. Torer, R. J. Twaits. J. Appi. Chem. 1964, 14, 563. 30. Computer program: SAINT V. 7.03A ed. Bruker AXS Inc., Madison, WI. (1997-2003). 31. Computer program: SHELXTLV. 5.1 ed. BrukerAXS Inc., Madison, WI. (1997). References start on page 45 47 Chapter 3 Anionic Oligomerization of Phosphaalkenes: Analysis by MALDI-TOF Mass S pectrometry* 3.1 Introduction The addition polymerization of olefins is one of the most important methods of polymer synthesis and forms the basis for the production of many commodity organic materials. There has been considerable interest in the synthesis and reactions of phosphaalkenes, and the remarkable parallels that exist between P=C and C=C bonds in molecular chemistry.13Our group has been interested in expanding the so-called “phosphorus—carbon analogy” widely established in molecular chemistry to polymer science, and has recently reported the high temperature polymerization of phosphaalkene 3.1 to give poly(methylenephosphine) 3.2. Ph F Ph] / Radical or Anionic I .. P=C .. -I-—P—C / \ Initiator I I IMes Ph LMeS Ph n 3.1 3.2 Previously, addition polymerization was often dismissed as unsuitable for inorganic systems. Now that 3.1 has been polymerized, one question that must be addressed is whether the initiation, propagation and termination steps are analogous to those known for olefins. If so, by analogy with olefins, living polymers and novel block or end-functionalized polymers may be accessible. Knowledge of the structure of polymer 3.2, including end-groups, will provide insight into the mechanism of polymerization. * A portion of this chapter has been published as a preliminary communication. Bronwyn H. Gillon, and Derek P. Gates, “Analysis of the products of the anionic oligomerisation of a phosphaalkene using MALDI-TOF mass spectrometry,” Chem. Commun. 2004, 1868—1 869. Copyright 2004 Royal Society of Chemistry. Chapter 3 48 The proposed mechanism of propagation occurs in an analogous fashion to that for olefins. A propagating anion will act as a nucleophile and add to the electrophilic phosphorus atom of the P=C bond, increasing the chain length by one repeat unit. This propagating anion can go on to add to another monomer (see Figure 3.1). Finally, when the monomers are completely consumed, addition of an electrophile terminates the polymerization. The polymers are proposed to be linear chains made up of P-C repeat units with terminal groups from the alkyl lithium initiator (R) and the electrophile (E). ___ IInitiation: R:e 15=c R—i5—c:e / \ II T if PC r I.. Il / \ I.. IPropagation. R+P_C+:e R-I-P—C :e L’ I] F ii r F ii Termination: R-’-P—C-I—:e R-1-P—C—ELI I] m>n LI I]m m Figure 3.1 Mechanism of anionic polymerization (initiation, propagation and termination) for P=C bonds. In Chapter 2, the syntheses of model compounds (3.4a-f) were described which provided invaluable insight into the regioselectivity of the addition of alkyllithium initiators (RLi) to the P=C bonds. The addition of the anionic initiator (R) occurs at the phosphorus atom (see Scheme 3.1) of the phosphaalkene 3.1. Subsequent addition of an electrophile (E) occurs at the benzylic carbon of the phosphorus-containing anion (either 3.3a or 3.3b). With the addition of various electrophiles, a range of possible end-groups (from a simple proton to silylium and phosphenium moieties) was tested in the hopes of preparing telechelic polymers. Missing from these model compounds is information on the interactions connecting the monomers in the References start on page 72 Chapter 3 49 polymer 3.2. In an effort to understand the propagation step of the polymerization of poly(methylenephoshine) 3.2, the characterization of oligomers could prove useful. Ph Ph Ph RLi R_P_C:eeLi EX R—P—C—E / ‘ II IIMes Ph Mes Ph - Mes Ph 3.1 3.3a: R = Me; 3.4a: R = Me, E=H; 3.3b: R = n-Bu 3.4b: R = Me, E Me; 3.4c: R = n-Bu, E = H; 3.4d: R = Me, E = P(NEt2) 3.4e: R = Me, E SiMe3; 3.4f: R = Me, E = SIHMe2 Scheme 3.1 The anionically initiated dimerization of a phosphaalkene 3.5 has been previously reported to form 3.6. Like our system, the alkyl anion added across the double bond at the phosphorus atom.6 However, phosphaalkene 3.5 was serendipitously prepared in the reaction mixture and was the proposed intermediate en route to dimer 3.6. The subsequent dimerization was suggested to have occurred through an analogous mechanism to that proposed in Figure 3.1. Upon addition of Me3SiCl, the propagating anion was quenched resulting in the dimeric 3.6 that was isolated as a colourless liquid in high yield and characterized by NMR spectroscopy. SiMe3 MeLi r SiMe3 2 Me3S1CI MeI-P—C SiMe3 / ‘ -Lid II IMe H LMeH 2 3.5 3.6 In this chapter, the ambient temperature oligomerization of 3.1 using anionic initiators and the characterization of the resulting oligomers using MALDI-TOF MS is reported. Also, the anionic oligomerization is extended to other phosphaalkene monomers. To my knowledge, oligomeric phosphaalkenes larger than diphosphetaines (cyclic dimers) have not previously been characterized by mass spectrometry.78Some examples of phosphaalkyne References start on page 72 Chapter 3 50 oligomerizations are known as well, 9-13 though the synthesis of linear species is contentious.’4 15 3.2 Results and Discussions The emergence of soft ionization (i.e. MALDI) mass spectrometric techniques to investigate high molecular weight (MW) macromolecules is revolutionizing the characterization of synthetic polymers,1621 however the use of such methods is still relatively uncommon in the analysis of inorganic systems.2226 Importantly, MALDI-TOF MS gives absolute masses rather than relative (i.e. GPC) and usually molecular ions are observed without fragmentation. The analysis of oligo(methylenephosphine)s 3.2 by MALDI-TOF MS is of particular interest. One objective is to obtain accurate MWs for 3.2 since laser light-scattering has shown that the MW is underestimated by GPC (vs polystyrene standards).4Moreover, MALDI-TOF MS may provide mechanistic insight into this new polymerization reaction by identifying end-groups that have remained undetectable by NMR spectroscopy. To date, analysis of polymeric samples of 3.2 by MALDI-TOF MS has not been successful; therefore, small oligomers have been synthesized that are analyzable using MALDI techniques. 3.2.1 Preparation of oxidized oligomers from anionically initiated MesP=CPh2to form Me[(Mes)P(=O)-CPh]H,3.9a The preparation of the oligomers (3.9a) was planned similarly to the preparation of model compound 3.4a. Instead of quenching the anion 3.3a with an electrophile, the anion was treated with additional equivalents of phosphaalkene 3.1 to favour oligomerization. Phosphaalkene 3.1 was dissolved in THF and cooled to —80 °C, to which was added a slight excess of MeLi in diethyl ether. The reaction mixture was stirred while slowly being warmed to room temperature. An aliquot of the dark red solution was removed and analyzed by References start on page 72 Chapter 3 51 31P NMR spectroscopy. A singlet resonance at —45 ppm was observed and assigned to Mes(Me)P—C(Ph)2Li(3.3a).5A solution of phosphaalkene 3.1 (3 equivs) in THF was added to the reaction mixture which was stirred overnight. Another aliquot was removed from the reaction mixture and was analyzed by 31P NMR spectroscopy. The spectrum revealed a broad resonance centered at —7 ppm assigned to 3.8a which is similar for that of the polymer 3.2 (ó31p = —1 O). Addition of degassed H20 to the reaction mixture afforded a pale yellow solution. Removal of the solvents in vacuo resulted in a yellow oil that was dissolved in diethyl ether and subsequently treated with excess H20 to oxidize the phosphine atoms. After isolation, the oxidized oligomers 3.9a were dissolved in minimal CH2I and were precipitated by adding hexanes rapidly to the solution. Analysis of the precipitated product by 31P NMR spectroscopy revealed that, analogous to the oxidized polymer, only a broad signal (ô = 45 ppm) was observed. The 1H NMR spectrum showed two very broad resonances in the alkyl and aryl regions in the range observed for phosphine polymer 3.2, but offered no other useful information (for example, the presence and nature of endgroups or coupling). Ph Ph r Phi MeLi Me—P—C—Li 2or3equiv3.1 Me-—P—C—Li / EtO -78 °C I I Et20 I I IMes Ph Mes Ph LMeS PhJ 3.1 3.3a 3.7a MeOH or H20 ro Phi r Phi I II I I H20 I .. I I Me-I—P—C--I-H 4 Me+-P—C—I-H II II II II LMes PhJ LMes PhJ 3.9a 3.8a In order to further characterize the products, MALDI-TOF mass spectrometry was employed. A sample for MALDI-TOF MS analysis was prepared using the layer method through the deposition of a THF solution of the matrix 2,5-dihydroxybenzoic acid (DHB) followed by a CH2I solution of oligomers 3.9a.17 A typical mass spectrum is shown in Figure 3.2 and References start on page 72 Chapter 3 52 selective data are collected in Table 3.1. Remarkably, a series of ions spaced by the mass of the monomer unit (332 Da) were observed that were assigned to linear oligomers 3.9a (n = 3 — 11). The masses were within experimental accuracy (± 0.1%) of the calculated masses for protonated species (3.9a +H) (ca ± 2 Da). For example, inset A in Figure 3.2 clearly shows an isotope pattern with lowest mass 1676 Da which is 1 less than (3.95a+H)(C111H0P50;calcd. 1677 g mor1). The isotope pattern is consistent with that expected with the most intense ion due to the presence of I x 13C (observed at 1677). Remarkably, oligomers of up to 11 repeat units [(3.911a+H), 3669 Da; calcd., 3669 g mor1jwere observed and for this ion with 243 C atoms, the most intense ion was due to the presence of 2 x 13C in the oligomer which is consistent with the expected isotope pattern. The observation of ions consistent with the presence of CH3 and H end-groups in 3.9a suggests that linear products are obtained and that the mechanism of chain growth is analogous to that for olefins. References start on page 72 Chapter 3 a. 39a 2000- 1500- 1000- 500- Table 3.1 ____________________________ One puzzling feature of the MALDI-TOF mass spectrum in Figure 3.2, is the unexpected presence of a second series of oligomers spaced by the monomer unit (332 Da). This second series can clearly be seen in Figure 3.2. Remarkably, each peak in this series is situated exactly 3.94a 53 3.95a 3105a InsetA Inset B (3.105a+O)* 3.125a 1830 1850 1870 mlz (3.95a+O)* 3115a 395a __n- -r 1670 1690 1710 m/z 396a 397a 39a I 3Sa LJL..LL L3*?07aL1O8aL 3’V’ 3.9a 3.10a 3104a LJL Figure 3.2 1400 1900 2400 2900 3400 m/z MALDI-TOF MS of oligomer mixture, 3.9a. * tentative assignment. Selected MALDI-TOF MS data for oligomeric mixture 3.9a Calculated Found n MW (g moF1) MW (g moF1) 1 348.2 N/A 2 680.3 679 3 1012.4 1011 4 1344.6 1343 5 1676.7 1675 6 2008.9 2007 7 2341.0 2339 8 2673.1 2671 9 3005.3 3004 10 3337.4 3337 11 3669.5 3669 References start on page 72 Chapter 3 54 halfway (i.e. ±166 Da) between the ions assigned to the linear oligomers (3.9a). This observation might possibly be attributed to multiply charged species. However, on close inspection of the mass spectra this hypothesis can be ruled out since the ions were spaced by integer mass units. If a complex were doubly charged, the difference between peaks would be 0.5 Da. If the P—C bond in a repeat unit were to be broken, each moiety (MeP=O and CPh2) has a mass of 166 Da. Given this fact, these signals are assigned to linear oligomers with an extra phosphorus group at the end of the chain (3.1Oa). Figure 3.2, inset B, shows signals consistent with 3.105awith calculated mass of 1842 Da. Structures with an extra CR2 and P—Me end-group are unlikely since a CPh2—CPh backbone linkage would be necessary to account for the observed masses. The reason for the presence of this second distribution of ions assigned to 3.1Oa is not obvious, although there are a number of possibilities: e.g. backbiting during the oligomerization, decomposition during oxidation, or fragmention during ionization. RLP-C-]-P_H -L-01 J-X-1- LMs Ph] es 3.1Oa R = Me 3.11 3.12 3.1Ob R = n-Bu 3.2.2 Origins of 3.1Oa — The “Second Distribution” Various experiments were performed in order to determine the source of the unexpected “second distribution” of ions. The potential sources were decomposition during oxidation, backbiting during oligomerization or fragmentation during mass spectrometry. 3.2.2.1 Decomposition due to Oxidation Oxidation is a potential source of the “second distribution” of ions that needs to be considered. Perhaps, the conditions for oxidation of the oligomers 3.8a to oxidized oligomers 3.9a with H20 were too severe resulting in decomposition of the backbone. Therefore, another sample of unoxidized oligomers 3.8a was prepared in an analogous fashion to that reported in References start on page 72 Chapter 3 55 section 3.2.1; the exception being that after quenching the anionic oligomers 3.7a with MeOH, the terminated oligomers 3.8a were purified by precipitation under inert atmospheric conditions. A sample for MALDI-TOF analysis was prepared using the layering method described above. From the mass spectrum, it was apparent that a mixture of oligomers was obtained with observable chain lengths of up to n = 7. The masses of the oligomers were within experimental accuracy (± 0.1%) of the calculated masses for protonated species (3.8a+H). Besides signals for linear species 3.8a and their isotopes, additional ions were observed in the spectrum. It is evident from Figure 3.3 that more than the unoxidized 3.8a ions are detected. Some of the additional peaks in the mass spectrum are assigned to partially oxidized oligomers. Unfortunately, the preparation of MALDI-TOF samples necessitates briefly exposure of the samples to air and therefore, partial oxidation of 3-coordinate phosphorus atoms was unavoidable. Consequently, ions that can be assigned to partially oxidized oligomers of 3.8a are observed in the spectrum in Figure 3.3 (see Table 3.2 and Table 3.3). Since the ions of the type 3.1Oa are indeed present in the spectrum, despite the brief exposure to air, oxidation seems an unlikely source of the “second distribution”. References start on page 72 3.84aiL 11idi. . Figure 3.3 Table 3.2 MALDI-TOF MS of oligomeric mixture 3.8a 56Chapter 3 a.i. - 500 - 400 - 300 200 - 1001. 3.82a 11i 38a 1.] IL I I 3.85a 3.86ailil,, .i. I I I I I • I —I I I 600 800 1000 1200 1400 1600 1800 2000 2200 m/z 3.87a Selected MALDI-TOF MS data for oligomeric mixture 3.8a Calculated Found n MW (g moF1) MW (g moF1) 1 332.2 N/A 2 648.3 649 3 964.4 967 4 1280.6 1283 5 1596.7 1598 6 1912.9 1914 7 2229.0 2231 References start on page 72 Chapter 3 57 Table 3.3 MALDI-TOF MS data for oligomeric mixture 3.8a with partially oxidized P atoms Calculated Found n MW (g moF1) MW (g mor1) 2+0 664.3 666 2+20 680.3 682 3+0 980.4 981 3+20 996.4 997 3+30 1012.4 1015 4+0 1296.6 1297 4+20 1312.6 1314 4+30 1328.6 N/A 4+40 1344.6 N/A 5+0 1612.7 1615 5+20 1628.7 1631 5+30 1644.7 N/A 5+40 1660.7 N/A 5+50 1676.7 N/A 6+0 1928.9 1932 6+20 1944.9 1947 6+30 1960.9 N/A 6+40 1976.9 N/A 6+50 1992.9 N/A 6+60 2008.9 N/A 3.2.2.2 Backbiting At the beginning of these studies, only relatively high PDI’s were obtained for the isolated polymers 3.2 (M = 10 300 g mo11 vs polystyrene; PDI = 1.55). If backbiting was a competing pathway in the anionic polymerization mechanism, then the broad range of molecular weights and low yields (Ca 30 — 40%) could be explained.4Backbiting reactions of the type proposed below (Figure. 3.4) are uncommon in olefin polymerization. However, similar backbiting processes are very common in the anionic ring-opening polymerization of cyclic trisiloxanes [(R2SiO)3].7’8A backbiting mechanism necessitates the formation of cyclic oligomers (like 3.11 and 3.12). Close inspection of Figure 3.2 (inset A and B) show that ions are detected exactly 16 Da below 3.9a and 3.1Oa, respectively. A possible mechanism for their formation would involve nucleophilic attack of the backbone of a growing chain with its active end (carbanion). Attack of the chain can potentially occur at either a P or C atom. Attack on C would result in the formation of the cyclic 3.12 and the linear 3.1Oa afterH20/H workup (see Figure 3.4). Conversely, the more likely attack at a P atom would result in the References start on page 72 Chapter 3 58 formation of the cyclic 3.11 and the desired, though shorter, linear 3.9a. Ions that can be assigned to these cyclic species 3.11 and 3.12 are only observed up to n = 5 for the MeLi initiated oligomerization of MesP=CPh2(3.1). Mes Ph ‘ , H20/H + Me{__c_j__H 3.7a 3.12 3.1Oa Figure 3.4 Proposed Mechanism of Backbiting of Propagating Oligomer 3.7a at a C Atom. In order to test whether or not backbiting during oligomerization was indeed the cause of the “second distribution,” a number of experiments was devised: from varying the concentration of the oligomerizations to using different initiators and monomers. In the following subsections, these studies will be discussed along with their impact on the postulated mechanism of anionic oligomerization of P=C bonds. 3.2.2.2.1 Varying Reaction Conditions The first means to test for backbiting was through varying the reaction conditions in order to evaluate the effects on the relative intensity of signals resulting from the postulated backbiting. For example, by increasing the monomer concentration for the oligomerization, the rate of backbiting should decrease because the probability of the reactive chain-end finding a monomer is greater. However, despite varying many reaction conditions (varying the concentration, reaction time, temperature, solvent and addition of coordinating ether 12-crown- 4), the “second distribution” attributed to 3.1Oa was continually observable in the MALDI-TOF MS. At very high concentrations of monomer (overall a —30 fold increase in concentration at 300 mg I 5 mL compared to 150 mg I 40 mL used originally to prepare 3.9a) in which little backbiting is expected to occur, there was evidence for the unwanted species 3.1Oa, 3.11, and 3.l2. In another trial, aliquots were removed at various times during the reaction that were References start on page 72 Chapter 3 59 analyzed by MALDI-TOF MS. After only two hours of reaction time, linear oligomers 3.92a and 3.9a were observable in the MALDI-TOF MS as was 3.102a. It seems unlikely that short chains of two or three repeat units could undergo backbiting due to the steric bulk of the chain. The downside for using MeLi as the initiator is that the difference in MWs between the cyclic species (3.11) and linear species (39a) is 16 g moI1 which corresponds to both the combined masses of the endgroups (Me and H) and the atomic mass of an 0 atom. It is possible that the ions assigned to the cyclic species are in fact under-oxidized linear oligomers. Using n-BuLi as the initiating species instead would separate the linear ions from the cyclic ions by 58 g mo11. 3.2.2.2.2 Testing A Different Initiator: Preparing n-Bu[(Mes)P(=O)-CPh2]H,3.9gb A solution of n-BuLl was added to a stirred solution of phosphaalkene 3.1 in cold diethyl ether. After warming to room temperature, an aliquot was removed for analysis by 31P NMR spectroscopy. A singlet resonance at —31.1 ppm was assigned to anionic 3.3b. To the reaction mixture was added a solution of 2 equivalents of phosphaalkene 3.2 in ether and stirred for 18 h. A second aliquot was removed and was analyzed by 31P NMR spectroscopy. The spectrum showed a broad resonance at —5 ppm and no remaining phosphaalkene (ó = 233) confirming that the monomer 3.1 was consumed. To quench the reaction, a drop of water was added to the reaction mixture and resulted in a colour change from dark red to yellow. The product was oxidized by treating the reaction mixture in diethyl ether with H20 in water. The ethereals were removed in vacuo resulting in a pale yellow oil which was then dissolved in a minimal amount of CH2I.The oligomeric mixture 3.9mb was isolated as a pale yellow powder from precipitation of the CH2I solution into hexanes. The 31P NMR spectrum of the powder dissolved in CD2I showed a broad resonance at 48 ppm that is consistent with oxidized polymer 3.1 and Me initiated oligomers 3.9a. The MALDI-T0F MS sample was prepared in an analogous fashion as the samples for Me-initiated oligomeric mixtures 3.8a and 3.9a. References start on page 72 Chapter 3 60 Ph Ph 10 Phi n-BuLi n-Bu—P—C—Li 1. 2 equiv. 3.1, Et20 n-Bu——P—C—f--H / “ Et0 -78 °C I I 2. H20 or MeOH I I IMes Ph Mes Ph 3. H0 LMeS PhJn 3.1 3.3b 3.9b The MALDI-TOF MS shows that the anionic initiation of P=C bonds occurs with n-BuLi and the oligomerization proceeds in an analogous fashion to the Me-initiated oligomerization. Linear oligomers 3.9mb of up to 6 repeat units are detected (see Table 3.4 and Figure 3.5). Additionally, Na adducts of these linear oligomers are observed and are likely the consequence of washing 3.9b with brine during workup. Again, more ions are present in the mass spectrum than the predicted linear 3.9b. Potential backbiting products like the linear species 3.1 Ob with additional MesP=O moieties can be assigned to ions (n = 2, MW = 888.4 Da; n = 3, MW = 1220.5 Da; n = 4, MW = 1552.6 Da; n = 5, MW = 1884.8 Da). Most notable, however, is the absence of ions that can be assigned to the cyclic species 3.12g. If backbiting were indeed occurring during oligomerization, and the backbiting attack occurred at the C atom in the backbone, the result would be both the presence of ions due to linear 3.1Ob as well as the cyclic 3.i2. Since cyclic species 3.12 are not observable, backbiting is likely not occurring to cause the presence of the “second distribution” assigned to 3.1 Ob. Additionally, when living anionic polymerization was later achieved by Kevin Noonan, the oligomerization was performed under analogous conditions (n-BuLi as the initiator and glyme as the solvent) and found the presence of 3.1Ob in the MALDI-TOF MS.29 30 The mounting evidence suggested that backbiting is not the major contributor of ions assigned to the “second distributions” of 3.1Oa and 3.1Ob. To be certain, the range of the oligomerization was expanded to other similar phosphaalkenes. By tuning the electronic factors of these phosphaalkenes, the degree of backbiting if indeed any was occurring during oligomerization was hoped to be affected. References start on page 72 Chapter 3 61 Selected MALDI-TOF MS data for oligomeric mixture 3.9gb Calculated Found n MW (g moF1) MW (g mor1) 1 390.2 N/A 2 722.4 722.8 (not shown) 3 1054.5 1056 4 1386.6 1388 5 1718.8 1721 6 2050.9 2053 3.2.2.2.3 Testing Different Phosphaalkenes: Preparation of Me[(Mes)P(=O)-CPhAr]H, 3.15a and 3.26a The investigation of the oligomerization of other phosphaalkenes was explored, with the hope that by tuning the electronic factors of the monomers, backbiting could be either eliminated or enhanced. By adding an electron-withdrawing group in the para position of the aryl group, for example, it is expected to minimize backbiting due to the decreased nucleophilicity of the propagating chain. A weaker nucleophile is less likely to attack the bulky backbone resulting in little to no backbiting products. Alternatively, an electron-donating group would result in a more 3.9b 300 200 3.94b 3.10b 100 3.94b+Na 3.104b 3.95b 3.95b÷Na Table 3.4 3,96b 1100 1300 1500 1700 1900 rnlz Figure 3.5 MALDI-TOF MS of n-BuLi initiated oligomer mixture, 3.9mb References start on page 72 Chapter 3 62 nucleophihc chain-end which would be more likely to attack the backbone thereby resulting in more backbiting products. With a fluoro-containing phosphaalkene 3.13, it was hoped that only the linear oligomers Me[(Mes)P(=O)-CPhPhf]H 3.15a would be obtained. Phosphaalkene 3.13 was dissolved in diethyl ether and cooled to —80 °C. The solution was treated with MeLi and immediately darkened from yellow to orange. The reaction mixture was stirred while warming to room, whereupon, two additional equivalents of monomer 3.13 in diethyl ether were added at room temperature and the reaction mixture was stirred for 48 hours. Thereupon, water was added to the red reaction mixture whose colour immediately turned yellow. H20 was used to oxidize the oligomers. A yellow powder was isolated from the precipitation of the crude product mixture in a small amount of CH2I added to hexanes. A broad resonance at 35 ppm was observed in the 31P NMR spectrum and assigned to the oxidized oligomeric mixture 3.15a. MeL1 Me—p—C—Li 1.2 equiv 3.13, Et20 Me—H Mel Ph Et20, -78 °C Mes Ph HO LMs Ph 3.13 3.14 2 2 3.15a Analysis of the precipitated product by MALDI-TOF MS showed evidence of up to only 3 linear repeat units of fluorinated oligomers 3.15a (n = 1 —3) (see Figure 3.6 and Table 3.5). Electron-withdrawing groups on the phenyl substituent appeared to slow the oligomerization. Consumption of the monomer was not complete after two days. In contrast, the oligomerization of MesP=CPh23.1 requires only 18 h. By 31P NMR spectroscopy, oligomerization of fluoro containing 3.13 appeared to stop despite the presence of monomer in solution. Incidentally, inclusion of a second fluorine substituent in a similar phosphaalkene monomer 3.16 resulted in no evidence of oligomerization with substoichiometric amounts of MeLi (see Scheme 3.2). Contrary to predictions, F-containing monomers did not enhance oligomerization and, in fact, References start on page 72 63Chapter 3 resulted in more ions assigned to the “second distribution” than the desired linear oligomers 3.15a. With these confusing results of using an electron-donating phosphaalkene, the oligomerization of an electron-withdrawing containing phosphaalkene was investigated. F(S iD=c’03 equiv MeLi / \ THF,-78°C ?\Mes Scheme 3.2 ro An QI ii I I II Me-I—P—cj_P_H LMes PhJ Mes 3.18a Ar = p-C6H4F 3.21 a Ar = p-COMe r0 Ar1 lMeh 3.19 Ar = p-c6H4F 3.22 Ar = p-COMe o An Ar II H Ip —c--i-—c I I II Mes PhJPh 3.2O Ar = p-c6H4F 3.23 Ar = p-cOMe F 3.16 3.17 References start on page 72 Chapter 3 a.i. - 20000- 15000- Table 3.5 __________________________ The preparation of oligomers with electron-donating groups (OMe) was performed similarly to the oligomerizations reported above. One equivalent of the initiator MeLi in ether was added to a solution of phosphaalkene 3.24 in diethyl ether at —80 °C and the mixture was warmed to room temperature while being stirred. 31P NMR spectroscopy was used to analyze an aliquot taken from the reaction mixture (ö = —43). The reaction mixture was treated with a solution of 2 equivalents of methoxy-containing 3.24 in diethyl ether and was stirred overnight. The reaction mixture was quenched with water and oxidized with H20 afforded oligomers 3.25a that were purified by precipitation from a concentrated solution of 3.26a in CH2I into hexanes. The precipitated product was a yellow powder. 3.152a 10000- 5000- 318a 315a 64 3.182a 315a 2O2JL_ 1000 mlz . LILIILi 3.192a L I • I • I I • 400 600 800 Figure 3.6 MALDI-TOF MS of the oligomeric mixture 3.15a Selected MALDI-TOF MS data for oligomeric mixture 3.15a Calculated Found n MW (g mor1) MW (g mor1) 1 366.2 367.0 2 716.3 717.0 3 1066.4 1067 References start on page 72 Chapter 3 65 OMeOMe OMe MeLi Me—Li 1 2 equiv 3.24, Et20 MeH / “ EtO, -78 °C I I 2. H20 I I I IMes Ph Mes Ph . LMes Ph J, 3.24 3.25 3.26a 31P NMR spectroscopy of a solution of oligomeric mixture 3.26a in CH2I showed a characteristic broad resonance at 40 ppm akin to other oligomeric mixtures. MALDI-TOF MS revealed that linear oligomers with up to 5 monomer units were obtained (see Figure 3.7 and Table 3.6). Notably, the presence of the “second distribution” of linear oligomers 3.21a exists in the MS albeit at smaller relative concentrations to the desired 3.26a. In this case, the nucleophilicity of the propagating chain was greater than for that of MesP=CPh23.1. Therefore, relatively higher concentrations of backbiting ions (3.21a, 3.22a and 3.23a) were expected in the spectrum. Regardless of the changes I made to the reaction, the “second distribution” always appeared in the spectra. References start on page 72 -,.‘.--1 rF Wj” “‘I’! 700 900 1100 1300 1500 Figure 3.7 MALDI-TOF MS for oligomeric mixture 3.26a Table 3.6 ___________________________ The mounting evidence that backbiting could not be the cause of the second distribution is obvious. Though initially disappointing, having backbiting not occur at ambient temperature during propagation means that strict control over polymer structure is possible. In fact, it was later shown that phosphaalkene 3.1 can be anionically polymerized in a living fashion.29 The ultimate explanation for the presence of these “second distribution” of ions is possibly fragmentation during ionization. 66Chapter 3 ai 200 150 100 50 II] Ii 3.21a 1700 m!z Selected MALDI-TOF MS data for oligomeric mixture 3.26a Calculated Found n MW (g moF1) MW (g mor1) 1 378.2 N/A 2 740.3 741.3 3 1102.5 1104 4 1464.6 1466 5 1826.8 1828 References start on page 72 Chapter 3 67 3.2.2.3 Fragmentation Fragmentation of oligomers like 3.9a is possible. However, it is relatively uncommon in polymer MALDI-TOF mass spectrometry.31 If the P—C chains are easily fragmented during ionization, ions like 3.1Oa could be accounted for. However, if fragmentation during ionization is facile, it is expected that oligomeric fragments in attempted MALDI-TOF MS analysis of polymer 3.2 would be observed. Under analogous MALDI conditions, species in mass spectra for the low MW polymer 3.2 (GPC M = 5 x to 1 x io g mor1)were not observed at all. In addition, oxidized 3.2 is thermally stable up to 320 °C as determined by TGA which seems to suggest a stable backbone that is not likely to fragment easily. Since oxidation and backbiting were eventually ruled out as being sources of the “second distribution”, fragmentation was investigated. Some rudimentary MALDI-TOF experiments were performed on Me-initiated oligomers 3.9a. The best spectra (i.e. best resolution and intensities) were acquired while using 2,5- dihydroxybenzoic acid as the matrix, other common matrices were tested (cinnaminic acid, dithranol and anthracenetriol). In every case, the “second distribution” of ions was observed. Based on the observations, the source of the “second distribution” may possibly be, albeit unexpected, the fragmentation of ions of 3.9a to give 3.1Oa. 3.3 Conclusion Our previous work on the polymerization of phosphaalkene 3.1 with MeLi (5%) in a minimum of solvent required a temperature of 150 0C4 This work demonstrated that chain growth proceeds in solution without heating and the observation of endgroups supports that the proposed mechanism of anionic polymerization for P=C bonds is analogous to that for olefins. Anionic oligomerization of phosphaalkenes can be extended to the use of another initiator (n BuLi) and to other monomers with functional groups (F and OMe). The oligomers were mainly References start on page 72 Chapter 3 68 characterized by MALDI-TOF MS and were found to have the proposed linear structure with up to 11 repeat units. However, based on the information at present, a possible source of the “second distribution” is the fragmentation of linear oligomers. 3.4 Experimental 3.4.1 Materials and General Procedures All manipulations of oxygen- and/or moisture-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk and/or glovebox techniques. Hexanes, diethyl ether and dichloromethane were deoxygenated with nitrogen and dried by passing through a column containing activated alumina. THF was freshly distilled from sodium/benzophenone ketyl. Distilled water and methanol were degassed prior to use. CDCI3 was distilled from P205 and degassed. CD2I and C6D were purchased from Cambridge Isotope Laboratories and were dried over molecular sieves (3 A). MeLi (1.6 M in Et20), n-BuLi (1.6 M in hexanes) were purchased from Aldrich and used as received. Alkyllithium reagents were titrated prior to use. MesP=CPh2,MesP=CPh(p-C6H4F),and MesP=CPh(p-C6H4OMe) were prepared following literature procedures.32 31P, 1H and 13C{H} NMR spectra were recorded on BrukerAV300 orAV400 spectrometers at room temperature. Chemical shifts for 31P NMR spectra are reported relative to H3P04as an external standard (85% in H20, ö = 0). Chemical shifts for 13C{H} NMR spectra are reported relative to CD2I (ó = 54.0) or CDCI3 (ô = 77.2). Chemical shifts for 1H NMR spectra are reported relative to residual CHDCI2 (8 = 5.32) or CHCI3 (8 = 7.27) or C6HD5 (8 = 7.16). The MALDI-TOF spectra were acquired using a Bruker Biflex IV spectrometer. The samples were dissolved in CH2I and deposited on the sample target with a layer of matrix References start on page 72 Chapter 3 69 previously deposited. The matrix, 2,5-dihydroxybenzoic acid, was dissolved in THF. The measurements were performed using the following conditions: positive polarity, reflection flight path, 18kV acceleration voltage, 20 shots per sample. Insulin Chain B, bovine insulin (oxidized) and Angiotensin II, human synthetic were obtained from Sigma-Aldrich and used as standards. I would like to thank Dr. Yun Ling and Marshall Lapawa for assistance in troubleshooting while using the MALDI-TOF MS. 3.4.2 Preparation of Me[MesP-CPh2]H,3.8a To a stirred solution of mesityl(diphenylmethylene)phosphine 3.1 (0.050 g, 0.16 mmcl) dissolved in diethyl ether (20 mL) at -78°C was added 1.5 M methyllithium in diethylether (0.13 mL, 0.19 mmol). After warming to room temperature, an aliquot was removed and analyzed by 31P NMR spectroscopy (Mes(Me)P-CPh2Li,ó = -45 ppm). A solution of 1(0.10 g, 0.32 mmol) dissolved in diethyl ether (20 mL) was added slowly to the reaction mixture. After being stirred for 24 hours, the reaction was quenched with degassed water. Analysis of an aliquot by 31P NMR spectroscopy showed a broad resonance centered at —5 ppm like the polymer 3.2. The solvent was removed in vacuo leaving a dark yellow oil. The oil was dissolved in a small amount of dichloromethane. The resulting solution was added to hexanes whereupon the oligomers precipitated as a yellow solid. Yield = 0.09 g (60%). 31p NMR (CD2I, 121.5 MHz): ô = —7 (br); 1H NMR (CD2I,300 MHz): ô = 7.7 — 6.7 (br, aryl), 2.5—1.8 (br, alkyl); 13C{H} NMR (CDCI3,75.5 MHz): 146.7 (br, Mes), 143.2 (br, Mes), 138.2 (br, Mes), 132-1 25 (br mult, Ph), 52.0 (br, P-C-P), 22.9 (br, o-CH3), 20.9 (br, p-CH3). 3.4.3 Preparation of Me[MesP(=O)-CPh2]H,3.9a The preparation was analogous to that of 3.8a except that 3 equivalents of phosphaalkene 3.1 was added to the Mes(Me)P—CPh2Lisolution. Once the unoxidized oligomeric mixture 6a was isolated, To the oil was added dichloromethane (10 mL) and the References start on page 72 Chapter 3 70 suspension was filtered then oxidized with 30% H20 for 30 mm. The organic layer was washed twice with water and dried over magnesium sulphate. The solvent was removed in vacuo leaving a pale yellow oil. The oil was dissolved in a small amount of dichloromethane and precipitated from hexanes leaving a pale yellow solid. Yield = 0.11 g (54%). 31p NMR (CD2I, 121.5 MHz): ó = 46 (br); 1H NMR (CD2I,300 MHz): = 7.6—6.6 (br, aryl), 3.7 — 1.5 (b, alkyl). 3.4.4 Preparation of Bu[MesP(=O)-CPh2]H,3.9mb Prepared analogously to 7a, but with an overall ratio of phosphaalkene:n-BuLi of 2:1 rather than 4:1 above. The product is a yellow solid. Yield = 0.11 g (73%). 31p NMR (CD2I, 121.5 MHz): ô = 48 (br); 1H NMR (CD2I,300 MHz): ó = 7.6—6.6 (b, aryl), 3.7—1.5 (b, alkyl); 13C{H} NMR (CD2I, 100.6 MHz): ô = 143.6 (s), 143.4 (s), 130.2 (s), 129.5 (5), 128.7 (s), 127.1 (d, 4Jc’ = 2 Hz, rn-Ph-C), 126.5 (d, 4Jcp = 2 Hz, rn-Ph-C), 50.6 (bs), 30.2 (bs), 24.7 (bs), 21.2 (s), 15.1 (s). 3.4.5 Preparation of Me[MesP(=O)-CPh(p-C6H4F)]H,3.15a To a cooled (—78°C) solution of fluoro-substituted phosphaalkene 14(0.10 g, 0.30 mmcl) in diethyl ether (10 mL) was added MeLi in Et20 (0.24 mL, 1.5 M, 0.36 mmol). Upon addition, there was an immediate colour change from yellow to dark red. The cooled solution was stirred for 1 h and then warmed to room temperature. To the solution was added 2 equivalents of phosphaalkene 14 (0.20 g, 0.60 mmol) in 30 mL of diethyl ether. The reaction mixture was stirred for 48 h whereupon degassed MeOH (0.4 mL) was added. From the resulting colourless solution, THF was removed in vacuo affording a yellow solid residue. The soluble fraction was extracted into diethyl ether (3 x 10 mL) and the resulting solution was stirred with H20 (30% in H20). The organic layer was isolated and dried over MgSO4. The solvent was evaporated to References start on page 72 Chapter 3 71 dryness affording a yellow solid. The mixture was purified by precipitation from a concentrated solution of CH2I added to hexanes. Yield: 0.07 g (24 %). 31P{1H} NMR (CD2I, 121.5 MHz): = 35 (br); 1H NMR (CD2I,300 MHz): ó = 7.0 (b, aryl), 2.5—1.3 (br, alkyl); 13C{H} NMR (CDCI3,75.5 MHz): 163.4 (br, p-Ar), 159.9 (br, p-Ar), 142-1 25 (br mult, Ar), 115.1 (br, rn-Ar), 55.0 (br, P—C—P), 24.7 (br, o-CH3), 20.9 (br, p-CH3). 346 Preparation of Me[MesP(O)-CPh(p-C6H4O e ]H, 3.26a To a cooled (—78°C) solution of methoxy-substituted phosphaalkene 18 (0.05 g, 0.14 mmol) in diethyl ether (5 mL) was added MeLi in Et20 (0.15 mL, 1.5 M, 0.17 mmol). Upon addition, there was an immediate colour change from yellow to dark red. The cooled solution was stirred for 1 h and then warmed to room temperature. To the solution was added 2 equivalents of phosphaalkene 18(0.10 g, 0.28 mmol) in 15 mL of THE. The reaction mixture was stirred for 18 h whereupon degassed MeOH (0.4 mL) was added. From the resultant colourless solution, THF was removed in vacuo affording a yellow solid residue. The soluble fraction was extracted into diethyl ether (3 x 10 mL) and the resulting solution was stirred with H20 (30% in H20). The organic layer was isolated and dried over MgSO4.The solvent was evaporated to dryness affording a yellow solid. The mixture was purified by precipitation from a concentrated solution of CH2I added to hexanes 3 times. Yield: 0.09 g (70 %). 31P{1H} NMR (CD2I, 121.5 MHz): ó = 40 (br); 1H NMR (CD2I,300 MHz): ó 7.2—6.7 (br, aryl), 3.7 (br, -OCH3), 2.7 — 1 .3 (br, alkyl). References start on page 72 Chapter 3 72 3.5 References 1. F. Mathey. Angew. Chem. mt. Ed. 2003, 42, 1578. 2. K. B. Dillon, F. Mathey, J. F. Nixon, Phosphorus: The Carbon Copy. Wiley: West Sussex, 1998. 3. M. Yoshifuji. J. Chem. Soc., Dalton. Trans. 1998, 3343. 4. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. 5. B. H. Gillon, K. J. T. Noonan, B. Feldscher, J. M. Wissenz, Z. M. Kam, T. Hsieh, J. J. Kingsley, J. I. Bates, D. P. Gates. Can. J. Chem. 2007, 85, 1045. 6. B.-L. Li, R. H. Neilson. inorg. Chem. 1984, 23, 3665. 7. A. Mack, U. BergstraBer, G. J. ReiB, M. Regitz. Eur. J. Org. Chem. 1999, 587. 8. M. Schmitz, S. Leininger, U. Bergstrasser, M. Regitz. Heteroat. Chem. 1998, 9, 453. 9. T. Wettling, J. Schneider, C. G. Kreiter, M. Regitz. Angew. Chem. mt. Ed. EngI. 1989, 28, 1013. 10. R. Bartsch, P. B. Hitchcock, J. F. Nixon. J. Chem. Soc., Chem. Commun. 1989, 1046. 11. V. Caliman, P. B. Hitchcock, J. F. Nixon, M. Hofmann, P. v. R. Schleyer. Angew. Chem. mt. Ed. EngI. 1994, 33, 2202. 12. F. Tabellion, A. Nachbauer, S. Leininger, C. Peters, F. Preuss, M. Regitz. Angew. Chem. mt. Ed. 1998, 37, 1233. 13. D. A. Loy, G. M. Jamison, M. D. McClain, T. M. Alam. J. Polym. Sd., Part A: Polym. Chem. 1999, 37, 129. 14. A. M. Arif, A. R. Barron, A. H. Cowley, S. W. Hall. J. Chem. Soc., Chem. Commun. 1988, 171. 15. M. Yoshifuji, H. Sugiyama, S. Ito. J. Organomet. Chem. 2005, 690, 2515. 16. M. Y. He, H. Chen. Curr. Org. Chem. 2007, 11, 909. 17. G. Montaudo, F. Samperi, M. S. Montaudo. Prog. Polym. Sd. 2006, 31, 277. 18. C. Barner-Kowollik, T. P. Davis, M. H. Stenzel. Polymer 2004, 45, 7791. References start on page 72 Chapter 3 73 19. S. F. Macha, P. A. Limbach. Curr. Opin. Solid State Mater. Sci. 2002, 6, 213. 20. M. J. Stump, R. C. Fleming, W. H. Gong, A. J. Jaber, J. J. Jones, C. W. Surber, C. L. Wilkins. App!. Spectrosc. Rev. 2002, 37, 275. 21. S. D. Hanton. Chem. Rev. 2001, 101, 527. 22. H. R. Allcock, C. R. de Denus, R. Prange, W. R. Laredo. Macromolecules 2001, 34, 2757. 23. B. M. White, W. P. Watson, E. E. Barthelme, H. W. Beckham. Macromolecules 2002, 35, 5345. 24. F. H. KOhIer, A. Schell, B. Weber. Chem. Eur. J. 2002, 8, 5219. 25. J. C. Blais, C. 0. Turrin, A. M. Caminade, J. P. Majoral. Anal. Chem. 2000, 72, 5097. 26. V. Marvaud, D. Astruc, E. Leize, A. Van Dorsselaer, J. Guittard, J. C. Blais. NewJ. Chem. 1997, 21, 1309. 27. M. Barrère, C. Maitre, M. A. Dourges, P. Hémery. Macromolecules 2001, 34, 7276. 28. J. Libiszowski, A. Kowaiski, R. Szymanski, A. Duda, J. M. Raquez, P. Degee, P. Dubois. Macromolecules 2004, 37, 52. 29. K. J. T. Noonan, D. P. Gates. Angew. Chem. lnt. Ed. 2006, 45, 7271. 30. K. J. T. Noonan, B. 0. Patrick, D. P. Gates. Chem. Commun. 2007, 3658. 31. H. J. Räder, W. Schrepp. Acta Polym. 1998, 49, 272. 32. M. Yam, J. H. Chong, C.-W. Tsang, B. 0. Patrick, A. E. Lam, D. P. Gates. lnorg. Chem. 2006, 45, 5225. References start on page 72 74 Chapter4 The Coordination Chemistry of PoIy(methylenephosphine)s with Tungsten, Molybdenum, and Chromium* 4.1 Introduction Metal-containing polymers have garnered interest from scientists due to their interesting properties.18For example, macromolecules with metal species dangling from a polymer side chain have found uses as supported catalysts because they simplify the separation of the metal from the desired organic products. One route to these types of macromolecules is through functionalization of existing polymers. For organic polymers, often functional groups (like phosphines) are attached in multiple steps allowing for a potential coordination site for the metal atoms.9A typical approach (see Scheme 4.1) involves polymerization of 4-bromostyrene followed by a lithiation, then a reaction with chlorodiphenylphosphine.°Metal coordination to the supported phosphine can occur through a phosphine substitution reaction.11 Derivatizations of polymers, though sometimes simple, can be complicated due to the fact that polymers are notoriously difficult to characterize compared to molecular species. If the polymer backbone were to contain a phosphine, then metallation could be attained with a one-step, classic coordination chemistry reaction. * A portion of this chapter will be published shortly. Bronwyn H. Gillon, Tom Hsieh and D. P. Gates, “Metallation of Poly(methylenephosphine)s with Group 6 Metal Carbonyls.” References start on page 99 Chapter 4 75 Br CIPPh2 PPh2 Some time ago, we reported the polymerization of P=C bonds to yield phosphorus- containing polymers 4.2. With the three-coordinate phosphorus built into the backbone, poly(methylenephosphine)s 4.2 is an obvious candidate for coordination chemistry, with potential uses for catalysis. Recently, Tsang et a!. showed that Suzuki cross-coupling reactions between bromobenzene and phenylboronic acid were possible in the presence of a phosphine containing copolymer 4.3, indicative of an interaction between the active catalyst, Pd(0), and the polymer (see Scheme 4.2).12 The interactions between metals and homopolymer 4.2 were explored. Ph E Phi —c initiator Me Ph LMsl’h]n 4.1 4.2 5-10 mol % /rHH Ph Il I I .. I —H-C-C \ P-C II I I I[ H Ph ‘ MesPh + 1.5 5 mcI % Pd2dba3 3.3 equivsCsF THE Scheme 4.2 Since characterizing and functionalizing macromolecules can be challenging, a couple of strategies were devised to overcome these difficulties while showing that P-M bonds can be References start on page 100 Scheme 4.1 Chapter 4 76 formed with poly(methylenephosphine)s. First, Group 6 metal carbonyls, M(CO)6were selected since they are well known to undergo substitution reactions with phosphine ligands.13’ 14 Additionally, the carbonyl ligands provide a distinct IR handle which should simplify characterization. Secondly, a small molecule 4•41517 was prepared in order to model the homopolymer 4.2 to gain insight into the reactivity of the backbone phosphines towards the Group 6 metals. If high metal content is achieved, particularly with tungsten and molybdenum, we could have access to interesting metal phosphide materials useful for hydroprocessing, for example.18’19 4.2 Results and Discussion 4.2.1 Model Compound Reactions (CO)5M Ph M(CO)6 f Ph Me-P—C—H Me-P—C—H I I I Mes Ph Mes Ph 4.4 M=W 4.5a Mo 4.5b Cr 4.5c 4.2.1.1 Coordination of Model Compound 4.4 to Tungsten Initial attempts to synthesize 4.5a photochemically using UV radiation proved difficult with multiple side reactions being observed.20’1The thermal preparation of tungsten phosphine complexes was reported by Magee et al. and proved more fruitful.13 A mixture of 4.4 and W(CO)6 in diglyme was stirred at 160 °C for 4 hours. The reaction was monitored by 31P{1H} NMR spectroscopy noting the disappearance of 4.4 (ó = -23 ppm) and the emergence of a resonance with 183W satellites (6 = 7.9 ppm, 1Jpw = 236.3 Hz) assigned to 4.5a. X-ray quality crystals were grown from an ethanol solution (yield = 20%) to afford colourless, moderately air- References start on page 100 Chapter 4 77 stable crystals. The molecular structure was confirmed by 31P, 13C, 1H NMR spectroscopy; FT-IR spectroscopy; EA; low resolution El MS (M, m/z = 656); and x-ray crystallography. _______ - 20 10 -10 -20 -30 ppm Figure 4.1 31P{1H} NMR spectra (diglyme) of: (a) model compound 4.4 and (b) tungsten adduct 4.5a. The 1H NMR spectrum for tungsten adduct 4.5a, is similar to that of the model compound 4.4, though the resonances are generally shifted slightly downfield. For example, in CD2I, the benzylic proton (CHPh2)in the model compound 4.4 results in a resonance at 4.87 ppm, but the analogous proton in the tungsten adduct 4.5a resonates at 5.22 ppm. The most notable difference is that upon coordination, the two ortho-methyl groups on the mesityl become magnetically inequivalent and show two separate resonances (broad signals at = 2.4 and 2.8 ppm). If rotation about the P(1)-C(1) bond were restricted, the methyl groups would be in different environments since the P atom is a stereogenic center. The most notable features in the 13C NMR spectrum are resonances due to the carbonyl ligands. The carbonyl trans to the phosphine shows a doublet at 202.2 ppm with a relatively large coupling constant (2Jcp = 22.1 Hz). The cis carbonyls show a doublet at 199.8 ppm (2Jcp = References start on page 100 Chapter 4 78 6.4 Hz) with W satellites (1Jcw = 127.1 Hz). The resonances due to the phosphine ligand are similar to those of the free phosphine 4.4. The solid-state molecular structure of 4.5a shows that two of the P-C bonds (P(l)-C(l) and P(1)-C(1O))areslightlyshortened; 1.848Aand 1.831 A(cf 1.852Aand 1.840A)7, respectively. Notably, the other P-C bond (P(1)-C(1 1)) has lengthened from 1.882 (1) A to 1 .915 (2) A upon coordination. The lengthening of this bond is, presumably, reflected in the mass spectrum with 2 of the 3 most intense fragment signals corresponding to fragments formed from the breaking of the P(1)-C(11) bond (M = 167 and 489 m/z). The lengthening of the P(1)-C(11) bond could become a problem for the coordination of the polymer to tungsten. That bond corresponds to the P-C bonds in the polymer backbone. If the backbone bonds are weakened, it is possible that coordination could lead to polymer degradation. This important insight from the solid-state molecular structure will be discussed later for the coordinated polymers. References start on page 100 Chapter 4 79 Figure 4.2 Solid-state molecular structure of the tungsten-model adduct 4.5a (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1)-C(1) = 1.848(2); P(1)-C(10) 1.831(2); P(1)-C(11) = 1.915(2); P(1 )-W(1) = 2.5621(5); C(24)-W(1): 2.030(2); C(25)-W(1): 2.060(2); C(26)-W(1): 1.996(2); C(27)-W(1): 2.040(2); C(28)-W(1): 2.034(2). Selected bond angles (°):C(1 0)-P(1)-C(1)= 110.82(9); C(1 0)-P(1)-C(1 1) = 99.75(9); C(1)-P(1)-C(1 1) = 101 .49(8); C(10)-P(1)-W(1) = 109.02(6); C(1)-P(1)-W(1) = 11.43(6); C(11)-P(1)- W(1) = 123.49(6). Geometry about W atom is nearly octahedral. 4.2.1.2 Coordination of Model Compound 4.4 to Molybdenum The synthesis of 4.5b is analogous to that of the tungsten adduct 4.5a except that the reaction proceeds in THE at reflux for 20 hours. The reaction was monitored by 31P{’H} NMR spectroscopy with the observance of the appearance of a resonance at 28 ppm assigned to the complex 4.5b. X-ray quality crystals are grown from a concentrated ethanol solution to afford colourless, slightly air-stable crystals (yield = 63%). The identity of the compound was confirmed by 31P, 13C, ‘H NMR spectroscopy; FT-IR spectroscopy; EA; low resolution El MS (M, mlz = 572); and x-ray crystallography. Coordination to molybdenum has very little effect on the ‘H NMR spectrum of the phosphine. Like the tungsten-adduct, resonances for the protons of the phosphine are shifted C3 i9 022 C20 C21 References start on page 100 Chapter 4 80 downfield slightly when compared to the free phosphine; the doublet due to the benzylic proton (CHPh2)is found at 5.06 ppm where 2JPH = 6.9 Hz. Also, the two ortho-methyl groups on the mesityl are not magnetically equivalent resulting in two broad signals at ó = 2.66 and 2.15 ppm. The molecular structure of molybdenum adduct 4.5b shows that two of the P-C bonds (P(1)-C(1) at 1.847(2) and P(1)-C(10) at 1.830(2)) are statistically equivalent to the analogous W-adduct 4.5a bonds; 1.848(2) and 1.831 (2) A, respectively. The P(1)-C(11) bond, however, is slightly shorter at 1.907(2) A. It is possible, that coordination of Mo(CO)5moiety could have less impact on the stability of the polymer than the tungsten analogue making Mo a better candidate for metallation. The ‘backbone’ P-C bond in Mo-adduct 4.5b is still longer than the free phosphine 4.4 (1.882 A). However, the most intense signal in the El mass spectrum is the CPh2W fragment (M = 167 mlz). References start on page 100 Chapter 4 81 Figure 4.3 Solid-state molecular structure of the molybdenum-model adduct 4.5b (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1)-C(1) = 1.847(2); P(1)-C(10) = 1.830(2); P(1)-C(11) = 1.907(2); P(1)-Mo(1) = 2.562(1); C(24)-Mo(1) = 1.981 (3); C(25)-Mo(1) = 2.061 (2); C(26)- Mo(1) = 2.029(3); C(27)-Mo(1) = 2.043(3); C(28)-Mo(1) = 2.027(2). Selected bond angles (°): C(1)-P(1)-C(10) = 110.6(1); C(1)-P(1)-C(11) = 101.34(9); C(10)- P(1)-C(11) = 99.6(1); C(1)-P(1)-Mo(1) = 111.46(8); C(10)-P(1)-Mo(1) = 109.06(8); C(11)-P(1)-Mo(1) = 123.82(6). Geometry about Mo atom is nearly octahedral. 4.2.1.3 Coordination of Model Compound 4.4 to Chromium A mixture of 4.4 and Cr(CO)6 in glyme was stirred at 120 °C for 5 days. The reaction was monitored with 31P{1H} NMR spectroscopy by the disappearance of 4.4 (ó = -23 ppm) and the emergence of a resonance ( = 49.6 ppm) assigned to 4.5c. Unfortunately, X-ray quality crystals were never isolated. The molecular structure was confirmed by 31P, 13C, 1H NMR spectroscopy; FT-IR spectroscopy; and low resolution El MS (Mt, mlz = 524). The 1H NMR spectrum for 4.5c is similar to those of 4.5a and 4.5b. The ortho-methyl groups of the mesityl are inequivalent and broadened (ô = 2.4 and 2.8 ppm). Having successfully prepared the model compounds, we were determined to coordinate the polymer to M(CO)5 moieties utilizing the thermal method. C14 013 022 021 References start on page 100 Chapter 4 82 4.2.2 Thermal Stability of Poly(methylenephosphine) 4.2 in Solution Since the preparation of adduct 4.5a required somewhat high temperatures (-160 °C), we were concerned that the polymer 4.2 may decompose under similarly harsh conditions. Therefore, we wanted to test the resiliency of the homopolymer 4.2 to high temperatures. We were hopeful that no significant decomposition would occur because TGA analysis demonstrated that, as a solid, polymer 4.2 does not decompose significantly until approximately 265 OCiG In order to evaluate the solution stability of poly(methylenephosphine)s 4.2 to heat, samples of 4.2 were refluxed in a number of solvents for varying amounts of time. The polymer was then be precipitated once from a concentrated solution of CH2I into hexanes, dried in vacuo and then characterized by triple detection GPC. The results are summarized in Table 4.1. Homopolymer 4.2 can withstand temperatures of up to 110°C in ethereal solvents for approximately one week without an appreciable decrease in M and is recoverable in high yields (see samples 1 and 3). After a few weeks, a small amount of polymer decomposition occurs in refluxing glyme (sample 4). However, in THE, the polymer remains intact after 24 days. Therefore, since refluxing in THF or glyme does not appear to break down the structure of the polymer 4.2, heating is a viable route to coordinating the chromium and molybdenum carbonyls to the polymer. However, upon heating in diglyme at 160°C, the polymer degrades into a brown solid after just 18 hours (sample 6). If the reaction mixture is removed from the heat source after just 4 hours, the polymer maintains its structural integrity, as its M does not significantly decrease. As a result, to coordinate the tungsten carbonyl moiety, a short reaction time is imperative. References stan’ on page 100 Chapter 4 Table 4.1 M(CO)5 r Phi Mej-P-C-fH LMesPh]n M=W 4.6a Mo 4.6b Cr 4.6c M(CO)5 [ Ph fPhl Mel—P-C \ P—C4H LMsTh MesPhjy x+y = 1 M=W 4.7a Mo 4.7b Cr 4.7c 83 Studies on the Thermal Stability of Poly(methylenephosphine) 4.2 in Solution. M(qmoi1)/PDl % Temp. Reco Sample Solvent (°C) Time Before heating After heating vered Colour 1 THF 70 6 days 28 000 I 1.14 28 600 I 1.11 92 yellow 2 THF 70 24 days 54300/ 1.10 54350? 1.19 87 yellow 3 glyme 100 6 days 54 300 / 1 10 50 200 I 1.32 69 yellow 4 glyme 100 24 days 54 300 / 1.10 45 500 / 1.19 87 yellow 5 diglyme 160 4h 21 800/1.3 20 200/1.26 65 yellow 6 diglyme 160 18h 21 800/1.3 bimodal 70 brown 4.2.3 Preparation of Group 6 Metal-Containing PoIy(methylenephosphine)s We chose Group 6 metals carbonyls, M(CO)6 to coordinate to the phosphine-containing polymer 4.2 as a proof of concept; the polymer will behave like a phosphine and form complexes with metals. Moreover, we expected that characterization of a metallated polymer might be difficult. Metal carbonyls have the exploitable IR handle. Carbonyls have strong C—O stretches in the 1900 — 2000 cm1 region. The frequency of the stretch of a carbonyl is also very sensitive to the nature of the ligand in the trans position. Therefore, FTIR will give us conclusive evidence of a phosphorus-metal bond. However, the greatest disadvantage of metallating with M(CO)5 fragments is their size and the close proximity of P atoms in the polymer backbone. Complete functionalization of polymer 4.2 into metallated polymers 4.6a-c will likely be impossible due to steric effects. The most probable metal-containing products will be random copolymers of the type 4.7a-c. F Phi Mej-P-C-f H LMesPh]n 4.2 M(CO)6 A References start on page 100 Chapter 4 84 4.2.3.1 Coordination of Poly(methylenephosphine) 4.2 to Tungsten Given that the coordination of the model compound 4.4 to tungsten ran smoothly under thermal conditions, we set out to emulate the complexation using poly(methylenephosphine) 4.2. The thermal stability studies (section 4.2.2) showed that polymer 4.2 decomposes in solution at high temperature (160°C) after more than 4 hours. Therefore, the polymer should not be exposed to such conditions in diglyme for more than a few hours. A flask equipped with a Teflon valve was loaded with poly(methylenephosphine) 4.2, an excess of W(CO)6,and diglyme. After three freeze-pump-thaw cycles, the reaction mixture was heated to 160 °C while protected from light. After 4 hours, the reaction mixture turned from a light yellow solution to a brown suspension with black flakes. The solution was added to a large amount of hexanes from which a very small amount of brown precipitate formed. After isolation, the brown residue was dried in vacuo. The entire amount was dissolved in a small amount of CD2I in an attempt to characterize the product by 31P NMR spectroscopy. Unfortunately, no discernible resonance was detected. Broadening of the resonance was expected, due to incomplete coordination (not every P atom is coordinated to a tungsten moiety). In addition, partial decomposition of the product was suspected given the colour of the product (brown) and very low yield. The NMR sample was dried in vacuo and analyzed by triple detection GPC. Despite the short reaction time, addition of excess W(CO)6 to a solution of polymer 4.2 in diglyme at 160 °C leads to major decomposition of the polymer. The Mof the polymer sample were determined using GPC-LLS and found to decrease from a modest 10 000 g mor1 to a paltry 600 g mo11. We suspected that the excess W(CO)6was causing decomposition of the polymer, so the experiment with smaller amounts of W(CO)6were repeated. In an analogous fashion to the reaction above, thermal reactions were performed of poly(methylenephosphine) 4.2 with 20 and 50 mol % of W(CO)6 in diglyme at 160 °C for 4 h. The results are summarized in Table 4.2. Decreasing the amount of tungsten in the reaction References start on page 100 Chapter 4 85 decreased the amount of decomposition. With 50 mol % of W(CO)6,the polymer started as a yellow solid with a M of 16 000 g moM and after heating became a brown solid with a M of 6000 g mo11. By adding a mere 20 mol % of W(CO)6,the integrity of the polymer appears to remain intact. Not only is the product an orange solid (indicating some chemical change) but the M of the product is similar to that of the starting material (4.2: 21 800 g mo11; 4.7a: 18 600 g moM). Table 4.2 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with W(CO)6to Form 4.7a. M (g moM) I PDI Metallated Amount of Temp. Polymer polymer Yield %metal Trial Solvent W(CO)6 (°C) Time 4.2 4.7a (%) Colour lated* I diglyme xs 160 4 h 9 600 / 600 / n/a brown n/a 1.06 3.7 2 diglyme 50 mol % 160 4 h 16 000 / 5 700 / 22 brown n/a 1.1 1.1 3 diglyme 20 mol % 160 4 h 21 800 / 18 600 / 48 green n/a 1.06 1.3 4 glyme 50 mol % 100 18 h 54 300 / 5 400 / n/a brown 2% 1.10 2.2 5 THE 50 mol % 80 6 28 000 / 27 000 / 90 orange 5% days 1.14 1.08 6 THE 50 mol % 80 24 54 300 / 57 500 / 97 yellow 6% days 1.10 1.16 * Determined by EDX/SEM 31P NMR spectroscopy was used to further analyze the tungsten-containing polymer 4.7a. The resonance of the product 4.7a in CD2I is broadened significantly with respect to the naked polymer 4.2. The resonance, centered at —10 ppm, has a range of almost 100 ppm (see Figure 4.5). From the model compound studies, I expected a downfield shift for the resonances where P atoms are coordinated to tungsten (see Figure 4.1), and there are some resonances in the expected range (0 to 50 ppm). The incredible broadening is likely due to incomplete complexation: not every P atom is coordinated to a tungsten moiety. The result is that the number of magnetically inequivalent P atoms is greater than in the uncoordinated polymer 4.2. Additionally, given the relatively low molecular weight, decomposition of the polymer likely resulted in further broadening of the resonance. Though coordination is likely to have occurred, References start on page 100 Chapter 4 86 albeit slight, the relative number of P atoms coordinated to W atoms is not determinable from integration in the 31P NMR spectrum. 11p, w”,,’w,1’ 150 100 50 0 -50 -100 ppm Figure 4.4 31P NMR Spectrum in CD2I of: (a) poly(methylenephosphine) 4.2; (b) tungsten- containing poly(methylenephosphine) 4. 7a. With evidence that chemical changes at some P atoms had occurred, i.e. 31P NMR spectroscopy showed a large broadening of the resonance after the reaction with W(CO)6,we needed evidence that W(CC)5moieties were coordinated. The IR spectrum of W(CO)6exhibits a characteristic band at 1976 cm1 in THF assigned to the C-C stretch. Functionalized polymer 4.7a showed a very broad stretch centered at 1890 cm1 (see Figure 4.5). This shift of the C-C stretch to lower frequencies is expected since carbonyls are better t-acceptors than phosphines. After substitution of a carbonyl with a phosphine, backbonding to the remaining carbonyls increases, weakening the C-C bonds. This trend is reflected in the IR spectrum of the model compound 4.5a that has a large signal at 1928 cm1 (assigned to the cis carbonyls). The weaker stretch at 2068 cm1 is assigned to the trans carbonyls (see Figure 4.5). In the case of tungsten-containing polymer 4.7a, we can conclude, therefore, that there are tungsten moieties with polymeric phosphines as ligands. From the lR spectrum, we could not determine the amount of coordinated phosphines in the polymer 4.7a. (a) (b) References start on page 100 Chapter 4 87 0 U) C)) E Cl) as 1. I— > G) 2200 2100 2000 1900 1800 1700 Wavenumbers (cm-i) Figure 4.5 Infrared Spectra of metal-carbonyl complexes with model compound and polymer in CH2I solutions (— 10 mg I mL). To further support the presence of W-P bonds in the polymer, a sample of W-containing 4.7a and fts precursor polymer 4.2 were analyzed by XPS (X-ray photoelectron spectroscopy). This surface technique can provide information on both the composition of a sample and the chemical state of the elements within the material. XPS confirmed the presence of W atoms in the sample by the diagnostic peak from the binding energy of a 4f electron centered around 36 eV (see Figure 4.6). Additionally, the binding energy of a P 2p electron (132 eV) was shifted References start on page 100 CrCO6 4.5c 4.7c M0CO6 &5b 4.7b Wc06 4.5a 4.7a Chapter 4 88 from that of the uncoordinated polymer 4.2 (130 eV) (see Figure 4.7). Such a shift is indicative of a change in the bonding environment at P. Now, convinced that some of the polymer chain is coordinated to W, we set out to determine the percentage of the P atoms that are coordinated to W. 90000 80000 70000 60000 50000 U, 40000 30000 20000 10000 0 1000 800 600 400 200 0 Binding energy (eV) Figure 4.6 X-ray photoelectron spectrum of tungsten-containing 4.7a. References start on page 100 Chapter 4 89 P 2p 1500 1000 500 0 To determine the percent composition of tungsten in the functionalized polymer 4.7a, one would think that traditional elemental analysis techniques would be appropriate. However, polymer 4.7a did not completely combust, and the results were found to be unreliable. A sample was also analyzed for tungsten content using ICP-MS. The procedure required that the sample dissolve in refluxing aqua regia. However, polymer 4.7a did not dissolve completely causing unreliable results. Also, the molecular weight data from GPC-LLS cannot be used to determine percent metallation in this case as upon coordination, the molecular weight of the polymer decreases. In order to get an estimate of tungsten content, EDX (Energy Dispersive X-ray) spectroscopy was employed. Unfortunately, EDX spectroscopy cannot detect the presence of atoms smaller than boron and since poly(methylenephosphine)s 4.2 are approximately 7% hydrogen atoms by mass, EDX spectroscopy will only give an approximation. Samples from [ Phi — Me4—P--C-f-H LMJfl 2 W(CO) Ph t Ph] — Me}-P-C 9H MesPh x\MesPhjy x+y = 1 5000 3500 3000 2500 Cl) 135 133 131 129 127 125 Binding Energy (eV) Figure 4.7 X-ray photoelectron spectrum of the P 2p binding energy of poly(methylenephosphine) 4.2 and tungsten-containing 4.7a References start on page 100 Chapter 4 90 trials 4 to 6 (see Table 4.2) were analyzed by EDX. The results suggested that as much as 7% of the backbone was metallated with W fragments. Low temperature (refluxing THF) and long reaction times (> 3 weeks) result in the highest (though still modest) W content. Coordination of a phosphine to tungsten may cause a weakening in the polymer backbone that results in the ready decomposition of the polymer. Polymer degradation is most likely due to steric repulsion of the large W(CO)5 moiety (relative to the model compound 4.4) and the very congested polymer 4.2. In fact, in the tungsten-containing model compound 4.5a, there is a slight lengthening in the P(1 )-C(l 1) bond (1 .915(2) A) relative to that of the free phosphine 4.4 (1 .882(1) A). These P-C bonds in the model compounds may be compared to the P-C bond within the polymer backbone. Interestingly, the other P-C bonds shorten slightly upon coordination (P(1)—C(1): 4.4 = 1.852(1) A vs 4.5a = 1.848(2) A; P(1)-C(10): 4.4 = 1.840(2) A vs 4.5a = 1.831(2) A). Under high temperatures (>100 °C), the polymer must degrade somewhat, resulting in low M’s as determined by GPC-LLS. The analogous substitution reactions with Mo or Cr require lower temperatures than with W. Therefore, metallation with either Mo or Cr may proceed with less polymer degradation. Complete metallation is, however, still unlikely due to the steric congestion about the metals and polymer 4.2. Based solely on the size of the group 6 metals, coordination to Cr is predicted to be less efficient than for Mo. 4.2.3.2 Coordination of Poly(methylenephosphine) 4.2 to Chromium The reaction of homopolymer 4.2 with Cr(CO)6was carried out in an analogous fashion to that with W(CO)6.Both reagents were loaded into a flask equipped with a Teflon valve and were dissolved in THF. The reaction mixture was degassed by 3 successive freeze-pump-thaw cycles and then heated to 80 °C while being stirred and protected from ambient light. After a period of time, the reaction mixture was cooled to room temperature and added to a relatively References start on page 100 Chapter 4 91 large volume of hexanes in order to precipitate the metallated polymer 4.7c out of solution. Upon isolation, the precipitated solid was heated in vacuo to 40 °C to remove excess Cr(CO)6 and solvent. The purified product was characterized by 31P NMR spectroscopy, GPC-LLS, FT IR and EDX. Just as for the tungsten-containing polymer 4.7a, characterization of chromium- containing 47c proved difficult. The 31P NMR spectrum of 4.7c showed a broadening of the resonance from that of homopolymer 4.2 that was barely discernible from the baseline. FT-IR analysis of the polymer 4.7c showed a shift of the stretching due to the CO groups (broad stretches centered at 1892 and 1858 cm1) relative to Cr(CO)6 (1980 cm1) (see Figure 4.5). Similarly, the chromium-model compound adduct 4.5c shows a strong stretch at 1928 cm1 with a smaller signal for the trans carbonyls at 2068 cm1. Convinced that at least some of the phosphorus atoms in the polymer are coordinated to Cr(CO)5fragments, we set out to maximize the metal content. A number of temperature studies were performed (see Table 4.3 for a summary of the results). Refluxing homopolymer 4.2 and Cr(CO)6 in THF for 4 hours, resulted in a yellow solid whose IR spectrum showed no indication that coordination had occurred. In fact, evidence of coordination could not be obtained until after at least 6 days of reaction time, at which time GPC-LLS analysis of the orange product revealed a small increase in M from that of homopolymer 4.2. This increase of 1100 g mor1 corresponds to approximately 6 of the 88 repeat units in the polymer coordinated to Cr(CO)5moieties. EDX analysis of the yellow solid was consistent with the modest amount of coordination at 5 % Cr by weight. Lengthening the reaction time to 24 days did not significantly change the degree of metallation. A small increase in reaction temperature (to 100 °C), however, leads to some polymer decomposition; the M decreased from 54 000 to 37 300 g mor1. Although modest, thermal metallation of homopolymer 4.2 is possible with Cr(CO)5moieties. References start on page 100 Chapter 4 92 4.2.3.3 Coordination of Poly(methylenephosphine) 4.2 to Molybdenum The reaction of homopolymer 4.2 with Mo(CO)6was carried out in an analogous fashion to that of the other group 6 M(CO) complexes. The reagents were loaded into a flask equipped with a Teflon valve and dissolved in glyme (or THF, or diglyme). After having been degassed, the reaction mixture was shielded from ambient light and was heated to 100 °C (or 80 °C in THF, 160 °C in diglyme) while being stirred. After a period of time, the reaction mixture was cooled to room temperature and precipitated from hexanes (see Table 4.4). The molybdenum- containing polymer 4.7b was heated in vacuo to 40 °C to remove unreacted Mo(CO)6and residual solvent. The purified product was mainly characterized by GPC-LLS, FT-lR and EDX. Table 4.3 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with Cr(CO)6 to Form 4.7c. Amount of Temp. Trial Solvent Cr(CO)6 (°C) Time 1 THF xs 80 4h 2 THF xs 80 6 days 3 THF xs 80 24 days 4 glyme xs 100 18h a determined GPC-LLS; D determined by EDX M(ci moF1)/PDI Metallated Polymer polymer 4.2 4.7c 28000/ 28000/ 1.14 1.1 28000/ 29100/ 1.14 1.12 54300/ 56300/ 1.10 1.21 54300/ 37300/ 1.10 1.10 Yield %metall (%) Colour ated 99 yellow 0% 94 orange 7%a 5%b 78 dark yellow 4%b n/a orange 3%b References start on page 100 Chapter 4 93 Table 4.4 Summary of Conditions for the Thermal Reaction of Poly(methylenephosphine) 4.2 with Mo(CO)6to Form 4.7b. M(p moF1)/PDI Metallated Amount of Temp. Polymer polymer Yield %metall Trial Solvent Mo(CO)6 (°C) Time 4.2 4.7b (%) Colour ated 1 glyme xs 100 7 days 54 300/ 51100/ 56 orange 11%b 1.10 1.46 2 THF xs 80 6 days 28 000 / 27 900 / 79 salmon 0% 1.14 1.12 3 THF xs 80 24 days 54 300 I 54 400 / 60 brown >1 %a 1.10 1.18 4 diglyme 20 mol % 160 4 h 21 800 / 25 700 / 97 yellow 23%a 1.3 1.14 5 diglyme 100 mol % 160 4 h 21 800 / 25 100 / >5 brown 20%a 1.3 1.42 a determined GPC-LLS; b determined by EDX Evidence of coordination of homopolymer 4.2 to molybdenum was confirmed through FTIR. Akin to the observations for metallated 4.7a and 4.7c, the stretching frequency for the cis carbonyls was assigned to the broad absorbance centered at 1920 cm1 in CH2I. In a CH2I solution, Mo(CO)6has an intense carbonyl signal at 1980 cm1, whereas the coordination compound 4b shows a large signal at 1940 cm1 (assigned to the cis carbonyls) with a smaller signal for the trans carbonyls at 2070 cm* Again, from the IR data, the degree of metallation was difficult to discern. The degree of metallation for 4.7b, mainly determined using GPC-LLS and EDX analyses, depended strongly on the reaction conditions. Metallation of around 10% was achieved by heating the homopolymer 4.2 and Mo(CO)5 in glyme to 100 °C, akin to the preparation of model compound 4b. This relatively high metal content, though exciting, was accompanied by a relatively low yield and an overall decrease in M most likely due to decomposition. These metallated polymers appear to be relatively sensitive to heat. Therefore, the reaction was repeated in THF at a slightly cooler 80°C. Although the M’s of the resulting polymers were not less than the M’s of the starting materials, they were also not much larger. Despite a long reaction time, there appeared to be very little coordination of Mo. In fact, 31P References start on page 100 Chapter 4 94 NMR analysis revealed no change in the resonance of homopolymer 4.2 (ó = -7, broad) to that of the isolate product (ö = -7, broad). Metallation of only 10% was too modest. So, despite the polymers instabilities to high temperatures, I heated polymer 4.2 in the presence of 20 mol % of Mo(CO)6to 160 °C for a short time. From analysis by GPC-LLS, the resulting metallated polymer 4.7b, had an M 4 000 g mo11 greater than the homopolymer which corresponds to -20% metallation. Increasing the relative amount of Mo(CO)6under similar conditions does not significantly affect the percent metallation, but can affect the yield greatly. Additionally, the PDI of the resulting polymer is broadened relative to that the starting material indicating a certain amount of decomposition. 4.3 Conclusion Finally, we have shown that poly(methylenephosphine) 4.2 can be partially metallated with Group 6 metal carbonyls through the thermally induced substitution of a CO group. Model compounds for the metallated polymers, phosphine-metal complexes 4.5a, 4.5b and 4.5c were successfully prepared and characterized. The objective was to achieve high metal loading of the polymer 4.2 with the hopes of preparing useful metal-containing materials or ceramic precursors. We now are certain that phosphorus-metal bonds can be formed. However, for total metallation of polymer 4.2, a less sterically hindered, larger metal center should be selected. 4.4 Experimental Section 4.4.1 General Methods and Instrumentation Al! manipulations of air- and/or water-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk or glovebox techniques. Hexanes and dichloromethane were deoxygenated by sparging with nitrogen and dried by passing through a column containing activated alumina. THF, glyme and diglyme were freshly distilled from References start on page 100 Chapter 4 95 sodium/benzophenone ketyl. CDCI3was distilled from P205. Bottled CD2I was purchased from Cambridge Isotope Laboratories and used as received. MesP=CPh24.122 and model compound 44,17 and poly(methylenephosphine) 4.223 were prepared following literature procedures. The metal carbonyls were purified by sublimation before use. 31P, 13C and 1H NMR spectra were recorded on BrukerAV300 MHz and AV400 MHz spectrometers. Chemical shifts for 31P NMR spectra are reported relative to H3P04as an external standard (85% in H20, ó = 0 ppm). Chemical shifts for 13C NMR spectra are reported relative to CD2I (ô = 54.0 ppm). Chemical shifts for 1H NMR spectra are reported relative to residual CHDCI2 (ô = 5.32 ppm). Mass spectra were recorded on a Kratos MS 50 instrument. Elemental analysis was performed by Mr. Minaz Lakha in the Departmental Microanalysis Facility. Molecular weights were estimated by triple detection gel permeation chromatography (GPC-LLS) using a Waters liquid chromatograph equipped with a Waters 515 HPLC pump, Waters 717 plus autosampler, Waters Styragel columns (4.6 x 300 mm) HR5E (2000-4 000 000), HR4 (5000-500 000), and HR2 (500-20 000), Waters 2410 differential refractometer ( = 940 nm, 40 °C), Wyatt tristar miniDAWN (laser light scattering detector operating at = 690 nm), and a Wyatt ViscoStar viscometer. A flow rate of 0.5 mL min1 was used, and samples were dissolved in THF (ca. 2 mg mL1). All IR samples were prepared as solutions in CH2I, and their spectra were recorded on a Thermo Nicolet 4700 FT-IR spectrometer by the author. All XPS spectra were recorded by Ken Chung Wong at the Advanced Materials and Process Engineering Laboratory at the University of British Columbia with a Leybold MAX 200 X-ray Photoelectron Spectrometer (XPS/ESCA) with a Dual anode (Mg Ka and Al Ka) achromatic X-ray source. EDX experiments References start on page 100 Chapter 4 96 were performed by the author at UBC with a Hitachi S-3000N SEM with light element EDX. No conductive coating was used for imaging and analysis of samples. 4.4.2 General procedure for phosphine complexes. This preparation is a modification of the literature procedure according to Magee et a!.13 A bomb was charged with model compound 3 and excess M(CC)6 and an ethereal solvent. The solution was degassed by 3 successive freeze-pump-thaw cycles and heated to reflux with a strict exclusion of ambient light. The reaction was monitored by 31P NMR spectroscopy. The solvent was removed in vacuo. Excess M(CC)6was removed by sublimation from the resulting yellow residue. The compound was purified by recrystallization. 4.4.3 Preparation of (HCPh2)(Me)MesPW(CO)5(4.5a) The reaction flask was charged with W(CC)6 (393 mg, 1.10 mmol), model compound 3 (250 mg, 0.75 mmol) and diglyme (11 mL). Total reaction time was 4 hrs. The product was purified by recrystallization from ethanol giving light sensitive, slightly yellow crystals. Yield= 106mg (20 %). 31P NMR (CD2I, 121.5 MHz): ó 7.9 (s,1Jpw=236.3Hz); 1H NMR (CD2I,300 MHz): ó = 7.44-6.89 (m, 12H, aromatic), 5.22 (d, 1H, CH, 2JHp= 7.48 Hz), 2.80 (brs, 3H, o-Mes-CH3),2.40 (br s, 3H, o-Mes-CH3),2.29 (s, 3H, p-Mes-CH), 2.14 (d, 3H, P-CH32JHP = 5.24 Hz); 13C NMR (CD2I,75 MHz): ó = 202.2 (d, trans-CC, 2Jcp = 22.1 Hz), 199.8 (d, cis-CC, 1Jcw = 127.1 Hz, 2Jcp = 6.4 Hz), 142.4-140.8 (m, Mes-C), 134.1- 129.0 (m, Ph-C), 55.9 (s, P-CH), 27.1 (s, p-Mes CH3), 25.1 (d, P-CH31Jcp 23.9 Hz), 22.7 (s, o-Mes-C). MS (El, 70eV): m/z = 656. Anal. Calcd. forC28H505PW: C, 51.20; H, 3.84. Found: C, 51.40; H, 4.20. lR (CH2I solution): 2068 cm1 (trans-CC), 1928 cm1 (cis-CO). References start on page 100 Chapter 4 97 4.4.4 Preparation of (HCPh2)(Me)MesPMo(CO)5(4.5b). The reaction flask was charged with Mo(CO)6 (260 mg, 1.0 mmol), model compound 3 (250 mg, 0.75 mmol) and glyme (10 mL). Total reaction time was 20 hrs. The product was purified by recrystallization from ethanol giving light sensitive, slightly colourless crystals. Yieldt 301 mg (71 %). 31P NMR (CD2I, 121.5 MHz): 828.5(s); 1H (CD6I2,300 MHz): 8 = 7.31-6.56 (m, 12H, aromatic), 5.06 (d, 2JHp= 6.9 Hz, 1H, CHPh2), 2.66 (brs, 3H, o-Mes-CH3),2.15 (brs, 3H, o-Mes CH3), 1.94 (s, 3H, p-Mes-CH), 1.72 (d, 2JHP = 4.2 Hz, 3H, P-CH3);13CNMR (CD2I,75 MHz): 8 = 202.2 (d, trans-CC, 2Jcp = 22.1 Hz), 199.8 (d, cis-CO, 2Jcp = 6.2 Hz), 141.8-140.2 (m, Mes-C), 134.2- 129.1 (m, Ph-C), 54.7 (s, P-CH), 27.1 (5, p-Mes-CH3),24.3 (d, P-CH31Jcp= 24.0 Hz), 22.5 (s, o-Mes-C). MS (El, 70 eV): m/z = 570. Anal. Calcd. forC28H5O5PMo: C, 59.16; H, 4.43. Found: C, 60.10; H, 4.12. IR (CH2I solution): 2070 cm1 (trans-CC), 1940 cm1 (cis-CO). 4.4.5 Preparation of (HCPh2)(Me)MesPCr(CO)5(4.5c) The reaction flask was charged with Cr(CO)6 (230 mg, 1.0 mmol), model compound 3 (252 mg, 0.75 mmol) and THF (5 mL). Total reaction time was 20 hrs. The product was a yellow solid. Yield= 150 mg (38 %). 31P NMR (CD2I, 121.5 MHz): 839.2(s); 1H (CD6I2,300 MHz): 8 = 7.51-6.47 (m, 12H, aromatic), 5.09 (d, 2JHP = 6.7 Hz, 1 H, CHPh2), 2.69 (br s, 3H, o-Mes-CH3),2.18 (br s, 3H, o-Mes CH3), 2.03 (s, 3H, p-Mes-CH), 1.88 (d, 2JHP = 4.2 Hz, 3H, P-CH3);13CNMR (CD2I,75 MHz): 8 = 203.2 (d, trans-CC, 2Jcp = 22.3 Hz), 200.0 (d, cis-CO, 2Jcp = 5.9 Hz), 142.6-140.4 (m, Mes-C), 134.3- 129.2 (m, Ph-C), 53.7 (s, P-CH), 27.2 (s, p-Mes-CH3),26.2 (d, P-CH31Jcp= 22.7 Hz), 22.6 (s, o-Mes-C). MS (El, 70 eV): m/z = 524. References start on page 100 Chapter 4 98 4.4.6 Metallation of poly(methylenephosphine) 4.2 with M(CO)6 The experimental procedures for 4.7a, 4.7b, and 4.7c are given below as an example: To a Kontes bomb was added poly(methylenephosphine) 4.2 (105 mg, M = 21 800 g mor1, PDI = 1.06) and W(CO)6 (20 mg) and diglyme (2 mL). The reaction mixture was degassed through three cycles of freeze-pump-thaw and then heated to 160 °C in the dark. After heating for 4 hours, the reaction mixture was poured into a Schienk tube containing 50 mL of hexanes. The metallated polymer precipitated immediately from solution and was isolated by filtration. The resulting brown powder was dried in vacuo for at least 24 h. Yield = 50 mg (48 %). See Tables 4.2, 4.3 and 4.4 for the results of all trials. 4.4.7 X-ray Crystallography Crystal data and refinement parameters are listed in Table 4.4. All single crystals were immersed in oil and mounted on a glass fiber. Data for 4.5a were collected on a Bruker X8 APEX diffractometer with graphite-monochromated Mo Ka radiation by Dr.Brian 0. Patrick. Data for 4.5b were collected on a Bruker X8 APEX II diffractometer by Joshua Bates. Data were collected and integrated using the Bruker SAINT software package.24 All structures were solved by direct methods and subsequent Fourier difference techniques and were refined anisotropically for all non-hydrogen atoms. Hydrogen atoms were included in idealized positions and refined isotropically. All data sets were corrected for Lorentz and polarization effects. All calculations were performed using the SHELXTL crystallographic software package from Bruker-AXS 25 References start on page 100 Chapter 4 99 Table 4.5 Details of Crystal Structure Determinations of 4.5a and 4.5b 4.5a 4.5b formula C28H505PW C28H5MoO5P formula weight (g moF1) 656.30 568.39 crystal system monoclinic monoclinic space group P 21/c P 21c colour colourless colourless a (A) 10.5368(4) 10.535(5) b (A) 22.3632(9) 22.334(5) c(A) 11.4187(5) 11.423(5) a(deg) 90 90 f3 (deg) 107.044(2) 107.105(5) y(deg) 90 90 V(A3) 2572.5(2) 2568.8(1 8) Z 4 4 T(K) 173(1) 173(2) i(Mo Ku) (cm1 45.89 0.609 crystal size (mm) 0.35 x 0.25 x 0.12 0.1 x 0.1 x 0.1 dcaIcd (g cm3) 1.695 1.470 20(max) (deg) 55.8 63.04 no. of rflns 44780 43676 no. of uniquedata 6124 8388 R(int) 0.028 0.0564 rfln/param ratio 19.14 20.16 Ri [I > 2.OOa(l) 0.016 0.0364 wR2 (all data) 0.038 0.0826 GOF 1.07 1.036 a R1 = I Fc - FcI 1/ZI F01; b wR2 = [Z(F02— F)2IZw(F02)]1”. References start on page 100 Chapter 4 100 4.5 References 1. A. S. Abd-EI-Aziz. Coord. Chem. Rev. 2002, 233-234, 177. 2. A. S. Abd-EI-Aziz. Macromol. Rapid Commun. 2002, 23, 995. 3. V. Chandrasekhar, Inorganic and Organometallic Polymers. 1st ed.; Springer: New York, 2005. 4. I. Manners. J. Polym. Sd. Part A: Polym. Chem. 2002, 40, 179. 5. I. Manners, Synthetic Metal-Containing Polymers. WiIey-VCH: Weinheim, 2004. 6. K. J. T. Noonan, D. P. Gates. Annu. Rep. Prog. Chem., Sect. A: lnorg. Chem. 2007, 103, 407. 7. I. Manners. Angew. Chem. Int. Ed. EngI. 1996, 35, 1603. 8. J. E. Mark, H. R. Alicock, R. West, Inorganic Polymers. 2nd ed.; Oxford University Press: Toronto, 2005. 9. N. E. Leadbeater, M. Marco. Chem. Rev. 2002, 102, 3217. 10. M. J. Farrall, J. M. J. Fréchet. J. Org. Chem. 1976, 41, 3877. 11. B. M. Trost, E. Keinan. J. Am. Chem. Soc. 1978, 100, 7779. 12. C.-W. Tsang, B. Baharloo, D. Riendl, M. Yam, D. P. Gates. Angew. Chem. mt. Ed. 2004, 43, 5682. 13. C. N. Matthews, T. A. Magee, J. H. Wotiz. J. Am. Chem. Soc. 1959, 81, 2273. 14. T. A. Magee, C. N. Matthews, T. S. Wang, J. H. Wotiz. J. Am. Chem. Soc. 1961, 83, 3200. 15. T. A. van der Knaap, F. Bickeihaupt. Tetrahedron Lett. 1982, 23, 2037. 16. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. 17. B. H. Gillon, K. J. T. Noonan, B. Feldscher, J. M. Wissenz, Z. M. Kam, T. Hsieh, J. J. Kingsley, J. I. Bates, D. P. Gates. Can. J. Chem. 2007, 85, 1045. 18. S. T. Oyama. Journal of Catalysis 2003, 216, 343. 19. S. Rundqvist. Nature 1966, 211, 847. 20. W. Strohmeier, H. Hellmann. Ber. Bunsen Ges. 1964, 5,481. References start on page 100 Chapter4 101 21. G. Schwenzer, M. Y. Darensbourg, D. J. Darensbourg. Inorg. Chem. 1972, 11, 1967. 22. M. Yam, J. H. Chong, C.-W. Tsang, B. 0. Patrick, A. E. Lam, D. P. Gates. Inorg. Chem. 2006, 45, 5225. 23. K. J. T. Noonan, D. P. Gates. Angew. Chem. mt. Ed. 2006, 45, 7271. 24. Computer program: SAINT V. 7.03A ed. BrukerAXS Inc., Madison, WI. (1997-2003). 25. Computer program: SHELXTLV. 5.1 ed. BrukerAXS Inc., Madison, WI. (1997). References start on page 100 102 Chapter 5 Macromolecular Corn plexation of PoIy(methylenephosphine) to Gold(l): A Facile Route to Highly Metallated Polymers* 5.1 Introduction Poly(methylenephosphine)s (PMP), are an unprecedented class of macromolecules which possess alternating phosphorus and carbon atoms (see Scheme 5.1).1 In particular, the phosphaalkene MesP=CPh2(5.1) can be polymerized under ambient temperatures through the addition of substoichiometric amounts of n-BuLi to form PMP 5.2 (Scheme 5.2).2 An appealing feature of PMP’s is their potential functionalization through the phosphine moieties. I am particularly interested in exploiting PMP’s as precursors to well-defined metal-containing macromolecules PMP-M (Scheme 5.1). Previously, our group had shown that PMP-containing copolymers are effective supports in Pd-catalyzed Suzuki cross-coupling.3However, well- defined Pd-complexes have not been isolated. Chapter 4 outlined the difficulties in preparing and characterizing metal coordination to classical organometallic complexes. I wanted to select a valuable metal for which phosphines are good ligands. i rMj Initiator . / \ H J [In PA PMP PMPM Scheme 5.1 * A version of this chapter has been published. Bronwyn H. Gillon, Brian 0. Patrick and Derek P. Gates, “Macromolecular Complexation of PoIy(methylenephosphine) to Gold(l): A Facile Route to Highly Metallated Polymers,” Chem. Commun., 2008, 2161-2163. Copyright 2008 Royal Society of Chemistry. Chapter 5 103 E PhiPh / 1.n-BuLi PzC n-Bu P—C H Me Ph 2. MeOH I I -LiOMe Mes Ph , 5.1 5.2 Scheme 5.2 When choosing target metals for coordination studies, gold(l) has attracted our attention. Gold(l) has a high affinity for phosphine ligands and exhibits a tendency to form linear P-Au-X complexes which would permit high Au loadings in the densely substituted PMP-M. Moreover, gold-containing polymers are attracting considerable interest due to their exciting potential applications in nanochemistry, catalysis, and chemosensors. Much of this work has been in the area of solid state gold coordination polymers that, with the exception of gold(l) polyynes, tend to lose their polymeric structure in solution.4°Gold(l) containing polymers with covalent backbones are less common.11’12 Herein, the preparation of soluble gold-containing macromolecules with high gold loadings through the macromolecular complexation of PMP to Au(l) is reported. First, the potential for PMP 5.2 to act as a ligand for transition metals was evaluated by studying the reactivity of the molecular model 5.3 with Au(l). 5.2 Results and Discussion 5.2.1 Coordination of Au(I) to model compound 5.3 Model compound 5.3, which was prepared from 5.1 13 was treated with (tht)AuCl (tht = tetrahydrothiophene) in CH2I solution. Monitoring the reaction progress by 31P NMR spectroscopy showed that the signal for 5.3 at —23.0 ppm disappeared and was replaced by a new signal at 16.0 ppm assigned to the coordination compound 5.4 (see Figure 5.1 (a) and (b)). Slow diffusion of EtCH into a concentrated CH2I solution of the product afforded 5.4 which was characterized by elemental analysis, El MS (Mt, mlz = 564) and X-ray crystallography. The References start on page 113 Chapter 5 104 Au PhPh / 1. MeLI .. I (tht)AuCI f P=C Me—P—C—H Me—P—C—H “ 2. MeOH CHCI, RTMes Ph - LIOMe Mes Ph - tht Mes Ph r - 5.3 (d) 200 150 100 50 0 -50 -100 ppm Figure 5.1 31P NMR (CD2I,300 K) spectra of (a) model compound 5.3, (b) gold complex 5.4, (c) polymer 5.2 and (d) coordinated polymer 5.5 (* tentatively assigned to phosphine oxide moieties (of oxidized polymer &lp = 471) 1H and 13C NMR spectra of the crystals in C6D solution are consistent with the formulated structure of 5.4. CI 5.1 (a) 5.4 (b) (C) References start on page 113 Chapter 5 105 The solid-state molecular structure of 5.4 (Figure 5.2) displays similar metrical parameters to model compound 5•313 The molecular structure shows that, despite coordination to gold, the solid-state structure of 5.4 does not significantly differ from model compound 5.3. For example, the length of the P-C bonds in 5.3 average 1.858(2) A and after coordination average 1.853(7) A. One interesting feature of 5.4 is that electron-group geometry about the P atom is close to tetrahedral as evidenced by a widening of the C-P-C bond angles with respect to 5.3 [5.4: avg. = 107.0(3)° vs. 5.3: avg. = 102.7(1)°]. The P-Au and Au-Cl bonds [P(1)-Au(1) 2.257(1) A; Au(1)-Cl(1) = 2.305(1) A) are in the long end of the range observed for similar gold(l)-phosphine complexes [Au-Cl: 2.21 — 2.26 A; Au-Cl: 2.23 — 2.31 A].147 Interestingly, the P-Au-Cl bond angle deviates slightly from linearity [P(1)-Au(1)-Cl(1) = 175.69(4)°] with the Cl bending towards the less hindered P-Me moiety. The closest AuAu contact of 9.47 A in 5.4 is much longer than expected if aurophilic interactions were present (ca. 3 A).18 These observations are not surprising since bulky gold(I) phosphines show bent Cl-Au-P angles [e.g. (m-tolyl)3PAuCI: Cl-Au-P = 175.1(1)°J and often do not show aurophilic interactions.19 References start on page 113 Chapter 5 106 cli 020 021 07 Aul 019 022 03 02 018023 013P1 014 01 C4 012 do 08 06 017 05 016 09 Figure 5.2 The solid-state molecular structure of 5.4 (50% probability ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (A): P(1)-C(1) = 1.852(4); P(1)-C(10) = 1.833(4); P(1)-C(11) 1.875(4); P(1)-Au(1) = 2.257(1); Au(1)-Cl(1) = 2.305(1). Selected bond angles (°): C(1)-P(1)-C(10) = 109.0(2); C(1)-P(1)-C(11) = 105.7(2); C(10)-P(1)-C(11) = 106.3(2); C(1 )—P(1 )—Au(1) = 11 5.2(1): C(1 0)—P(1 )—Au(1) = 107.1(1); C(1 I )—P(1 )—Au(1) = 113.2(1); P(1 )— Au(1)-Cl(1) = 175.69(4). 5.2.2 Coordination of Gold(l) to poly(methylenephosphine) to yield gold-containing polymer 5.5 With model complex 5.4 in hand, our next objective was to study the metallation of the polymer. The anionic polymerization of 5.1 in glyme using n-BuLi as the initiator afforded PMP 5.2 (M = 38 900 g mor1, PDI = 1 •34)•2 In an attempt to prepare complexed polymer 5.5, the phosphine polymer 5.2 was treated with (tht)AuCl (1 equiv per P) at room temperature in CH2I solution. After 1 h, an aliquot was removed from the reaction mixture and analyzed by 31P NMR spectroscopy (Figure 5.1 (c) and (d)). Remarkably, the broad resonance for the uncomplexed polymer 5.2 (ó = —7) had entirely disappeared and was replaced by a new broad resonance assigned to 5.5 (ó = 25). For comparison, a similar downfield shift is observed for the model compound (5.3 vs. 5.4: Áö = 39; 5.2 vs. 5.5: Aó = 32). The gold-containing polymer 5.5 was References start on page 113 Chapter 5 107 purified by precipitation of a concentrated CH2I solution into hexanes. The resulting white powder was further characterized by 1H and 13C NMR spectroscopy. CI Au I PhI r Phi I •. I I (tht)AuCI I •‘. I I n-Bu-I--P—C—f-H • n-Buf-P—C--l--H I I I I CH2I,RT[Mes PhJn -tht [Mes Phin 5.2 5.5 Scheme 5.3 The novel gold-containing polymer 5.5 was also characterized by triple detection GPC in THF solution. Importantly, the absolute number average molecular weight for 5.5 (M = 71 600 g moM) increased substantially upon coordination to gold (cf. 5.2: M = 38 900 g moM). The molecular weight of 5.5 is the highest yet observed for a PMP-based polymer and is consistent with a high degree of P-metallation. The polydispersity indices for 5.2 and 5.5 are both narrow (5.2: PDI = 1.34; 5.5: PDI = 1.29) which suggests that the integrity of the backbone is retained upon coordination. Thermogravimetric analysis of 5.5 (Figure 5.3) revealed an onset temperature of polymer degradation (Tonset = 325 °C) which is considerably higher than that for uncomplexed 5.2 (Tonset 265 oc)l Remarkably, after heating to 1000 °C a shiny, lustrous residue was obtained. The residue was 40% of the initial mass of 5.5 which is consistent with the expected gold content (36%). This residue is speculated to be composed primarily of elemental gold. Further studies are underway to confirm this hypothesis and to investigate the potential use of these polymers as polymer-supported catalysts or precursors to well-defined gold nanostructures. References start on page 113 Chapter 5 108 100.0 90.0 80.0 70.0 60.0 C.) 0. 50.0 40.0 30.0 20.0 10.0 0.0 0 100 200 300 400 500 600 700 800 900 1000 Temperature (“C) Figure 5.3 TGA trace for gold-containing polymer 5.5. Collected under a nitrogen atmosphere at a heating rate of 10 °Clmin. 5.3 Conclusion A novel gold-phosphine complex 5.4 was prepared as a model for the more difficult polymer system. For the first time, poly(methylene)phosphine 5.5 was successfully and largely metallated with gold(l) moieties. This achievement paves the way for the possibility of preparing gold-containing materials or macrostructures, micelles for example, as well as using the materials for gold(I) catalytic transformations. References start on page 113 Chapter 5 109 5.4 Experimental 5.4.1 Materials and General Procedures All manipulations of oxygen- and/or water-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk and/or glovebox techniques. Hexanes, toluene and dichloromethane were deoxygenated with nitrogen and dried by passing through a column containing activated alumina. EtCH was dried by distilliing over CaH2 and deoxygenated with nitrogen. CD2I,C6D and CDCI3were purchased from Cambridge Isotope Laboratories and used as received. MesP=CPh2,°Mes(Me)PCPh2H5•3,13 polymer 5.2,2 and (tht)AuCI21 were prepared following literature procedures. 31P, 1H and 13C{H} NMR spectra were recorded on BrukerAV300 orAV400 spectrometers at room temperature. Chemical shifts for 31P NMR spectra are reported relative to H3P04as an external standard (85% in H20, o = 0). Chemical shifts for 13C{H} NMR spectra are reported relative to CDCI3 (8 = 77.2). Chemical shifts for 1H NMR spectra are reported relative to residual CHCI3 (8 = 7.27) or C6HD5 (8 = 7.16). Mass spectra were recorded on a Kratos MS 50 instrument. Elemental analyses were performed by Mr. Minaz Lakha in the Departmental Microanalysis Facility. 5.42 Synthesis of [Mes(Me)(CPh2H)P•Au I] (5.4) To a stirred solution of (tht)AuCI (20 mg, 0.06 mmol) dissolved in CH2I (1 mL) was added model compound 5•313 (21 mg, 0.06 mmol) in CH2I (4 mL). The reaction was monitored by 31P NMR spectroscopy (8 = 16.0) and after 1 h the solvent was removed in vacuo leaving a white residue. The product was extracted with toluene (3 x 2 mL), and the solvent removed in vacuo to afford 5.4 as a colourless solid (33 mg, 94 %). X-ray quality crystals were obtained from slow diffusion of EtCH into a concentrated solution of 5.4 in CH2I. References start on page 113 Chapter5 110 31P NMR (CD2I, 162 MHz): 8 (ppm): 16.0 (s). 1H NMR (C6D,400.13 MHz): 8 (ppm): 7.65 (d, JHH= 8 Hz, 2H, Ph-I-i), 7.17 (m, 2H, Ph-H), 7.04 (m, 3H, Ph-H), 6.84 (m, 3H, Ph-H), 6.45 (brs, 2H, m-Mes-H), 4.52 (d, 2JPH= 16Hz, 1H, P-CPh2H), 2.23 (s, 6H, o-CH3-Mes), 1.89 (s, 3H, p-CH-Mes), 1.15 (d, 2JPH= 9 Hz, 3H, P-Cl-I3). 13C NMR (CDCI3,75.5 MHz) (assignments made with the aid of HMQC experiments) ô(ppm): 143.3 (d, 1Jcp= 10Hz, i-Mes-C), 142.1 (brs, Mes-C), 137.7 (s, Mes-C), 136.7 (s, Mes-C), 132.1 (d, J 9 Hz, Ph-C), 130.1-1 27.8 (m, Ph-C), 52.8 (d, 1Jcp= 32 Hz, P-CH), 25.7 (d, 3Jcp= 9Hz, o-Mes-CH3),21.2 (s, p-Mes-CH3), 16.8 (d, 1Jcp = 35 Hz, P-CH3). MS (El, 70eV): mlz (%) = 564 (0.4) [Mt], 332 (50) [M-AuCl], 167 (100) [CHPh2], 165 (50) [PMeMes], 119 (15) [Mes]. Anal. Caic. forC22H1PAuCI: C, 48.91; H, 4.46. Found: C, 50.38; H, 4.75. 5.4.3 Synthesis of n-Bu(MesP(AuCI)CPh2)H (5.5) To a stirred solution of (tht)AuCI (45 mg, 0.14 mmol) in CH2I (5 mL) was added polymer 5.22 (45 mg, M = 38 900 ± 800 g mor1, PDI = 1.34 ± 0.03) in CH2I (5 mL). The reaction was monitored by 31P NMR spectroscopy (8 = 24) and after 1.5 h the solvent was removed in vacuo leaving a yellow solid. The polymer was precipitated from a solution of CH2I (1 mL) into cold hexanes (15 mL). After filtration, the solvent was removed in vacuo affording 5.5 as a pale yellow solid (58 mg, 74 %). 31P NMR (CD2I, 162.0 MHz): 8 (ppm): 24 (br). 1H NMR (CDCI3, 300 MHz): 8 (ppm): 9.0 - 6.5 (Ar-I-I), 2.4 — 1.7 (Me-H). 13C NMR (CDCI3, 75.5 MHz): 8 (ppm): 146.5 (br, Mes-C), 142.2 (br, Mes-C), 138.0 (br, Mes-C), 132-1 24 (br mult, Ph-C), 52.1 (br, P-C-P), 23.1 (br, o-CH3), 28.2 (br, p-CH3). GPC-LLS (g/mol): M = 71 600 ± 600 g moM, PDI = 1.29 ± 0.01), dn/dc = 0.16 (as determined by 100% mass recovery methods). Satisfactory elemental analyses could not be obtained presumably due to ceramic formation. TGA (N2 atmosphere): 5 % weight loss 25 — References start on page 113 Chapter 5 111 270 °C, 45 % weight loss 270 - 450 °C, 15 % weight loss 450 - 770 °C, plateau 770— 1000 °C with 40 % residue remaining. 5.4.4 X-ray Crystallography Crystal data and refinement parameters are listed in Table 5.1. The single crystal was immersed in oil and mounted on a glass fiber. Data for 5.4 were collected on a Bruker X8 APEX diffractometer with graphite-monochromated Mo Ka radiation. Data was collected and integrated using the Bruker SAINT software package.22The structure was solved by direct methods and subsequent Fourier difference techniques and was refined anisotropically for all non-hydrogen atoms. Hydrogen atoms were included in idealized positions and refined isotropically. The data sets were corrected for Lorentz and polarization effects. All calculations were performed using the SHELXTL crystallographic software package from Bruker-AXS.23 Depository numbers for the .cif file deposited with the Cambridge Crystallographic Database: CCDC 665759. References start on page 113 Chapter 5 112 Table 5.1 Details of Crystal Structure Determinations of 5.4 5.4 formula C23HPAuCI•1/2CHI fw (g moF1) 607.28 cryst syst monoclinic space group 02/c (#1 5) colour colourless a(A) 17.031(1) b (A) 7.7689(5) c(A) 35.310(3) a(deg) 90 fi (deg) 93.431(4) y(de9) 90 V(A) 4663.5(6) Z 8 T(K) 173(2) i(Mo Ku) (cm1 66.13 cryst size (mm) 0.35 x 0.35 x 0.10 dcalcd (g cm3) 1.730 20(max) (deg) 55.7 no. of rflns 30781 no. of unique data 5534 0.0524 rfln/param ratio 21.9 R1; wR2 [I > 2.000(l)]a 0.0304; 0.0696 R1; wR2 (all data)a 0.0369; 0.0722 GOF 1.062 a R1 = I F - I F / F0I ; wR2 = [(F02— )L/w(Fo]h/ References start on page 113 Chapter5 113 5.5 References 1. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. 2. K. J. T. Noonan, D. P. Gates. Angew. Chem. tnt. Ed. 2006, 45, 7271. 3. C. W. Tsang, B. Baharloo, D. Riendi, M. Yam, D. P. Gates. Anyew. Chem. mt. Ed. 2004, 43, 5682. 4. R. J. Puddephatt. Coord. Chem. Rev. 2001, 216-217, 313. 5. M. J. Katz, H. Kaluarachchi, R. J. Batchelor, A. A. Bokov, Z.-G. Ye, D. B. Leznoff. Angew. Chem. mt. Ed. 2007, 46, 8804. 6. C. A. Wheaton, R. J. Puddephatt. Angew. Chem. mt. Ed. 2007, 46, 4461. 7. E. C. Constable, C. E. Housecroft, M. Neuburger, S. Schaffner, E. J. Shardlow. Dalton Trans. 2007, 2631. 8. C. A. Wheaton, M. C. Jennings, R. J. Puddephatt. J. Am. Chem. Soc. 2006, 128, 15370. 9. U. Siemeling, D. Rother, C. Bruhn, H. Fink, T. Weidner, F. Trager, A. Rothenberger, D. Fenske, A. Priebe, J. Maurer, R. Winter. J. Am. Chem. Soc. 2005, 127, 1102. 10. V. J. Catalano, M. A. Malwitz, A. 0. Etogo. lnorg. Chem. 2004, 43, 5714. 11. M. Sebastian, M. Hissler, C. Fave, J. Rault-Berthelot, C. Odin, R. Reau. Angew. Chem. mt. Ed. 2006, 45, 6152. 12. C. Diaz, M. L. Valenzuela, G. G. A. Carriedo, F. J. Garcia-Alonso, A. Presa. Polym. Bull. 2006, 57, 913. 13. B. H. Gillon, K. J. T. Noonan, B. Feldscher, J. M. Wissenz, Z. M. Kam, T. Hsieh, J. J. Kingsley, J. I. Bates, D. P. Gates. Can. J. Chem. 2007, 85, 1045. 14. R. C. Bott, P. C. Healy, G. Smith. Polyhedron 2007, 26, 2803. 15. E. C. Alyea, G. Ferguson, S. Kannan. Polyhedron 2000, 19, 2211. 16. R. C. Bott, C. A. Bowmaker, R. W. Buckley, P. C. Healy, M. C. S. Perera. Aust. J. Chem. 2000, 53, 175. References start on page 113 Chapter5 114 17. E. C. Alyea, G. Ferguson, J. F. Gallagher, J. Malito. Acta Crystallogr., Sect. C: Ciyst. Struct. Commun. 1993, 49, 1473. 18. H. Schmidbaur. Gold Bull. 1990, 23, 11. 19. C. S. W. Harker, E. R. T. Tiekink. Acta Ciystallogr., Sect. C: Cryst. Struct. Commun. 1991, 47, 878. 20. M. Yam, J. H. Chong, C.-W. Tsang, B. 0. Patrick, A. E. Lam, D. P. Gates. lnorg. Chem. 2006, 45, 5225. 21. R. Usón, A. Laguna, M. Laguna. lnorg. Synth. 1989, 26, 85. 22. Computer program: SAINT V. 7.03A ed. BrukerAXS Inc., Madison, WI. (1 997-2003). 23. Computer program: SHELXTLV. 5.1 ed. BrukerAXS Inc., Madison, WI. (1997). References start on page 113 115 Chapter 6 Inorganic/Organic Hybrid GoIdQ)-containing Micelles* 6.1 Introduction Living polymerization is an invaluable tool to form controlled polymer structures.1’2For example, living polymerization provides a facile method of preparing AB block copolymers starting with the polymerization of monomer A. Once the feedstock for polymer A is consumed, the monomer for polymer B is introduced to the reaction mixture. The living chain of A initiates the polymerization of the newly introduced monomer, subsequently forming the AB block copolymer. When polymers A and B have different physical properties (eg. solubility, glass transition temperature (Tg), intermolecular interactions), the resulting block copolymer has hybrid properties that are unattainable from homopolymers.34Our group is particularly interested in the solution self-assembly of AB block copolymers, where the blocks have different solubilities. Dissolving the copolymers in a solvent in which one block would dissolve but the other would not (known as a block selective solvent) can lead to macrostructure formations such as micelles, rods or vesicles.5Micelles, for example, consist of AB block copolymers that have assembled into a sphere consisting of two domains: the core comprised of the insoluble blocks and the corona formed from the soluble blocks (see Figure 6.1).6 Applications for polymers that form ordered assemblies in solution include: surfactants, drug-delivery systems, and nano reactors. 7-12 * A portion of this chapter has been published as a preliminary communication and is reproduced with permission. Kevin J. T. Noonan, Bronwyn H. Gillon, Vittorlo Cappello, and D. P. Gates, “Solution Self- Assembly of Isoprene-Phosphaalkene Block Copolymers: Macromolecular Templates for Tailor Made Gold(I) Nanostructures,”Journal of American Chemical Society, 130 (39), 12876—1 2877, 2008. Copyright 2008 American Chemical Society. Chapter 6 116 Corona (soluble Core (insoluble A B Dissolve in a solvent in which •••••• blockAissolubleand B is insoluble AB block-copolymer Cross-section of a micelle Figure 6.1 The solution self-assembly of an AB block copolymer to form a macromolecular structure (shown: cross-section of a spherical micelle). So far, most work on the preparation of block copolymers through living polymerization is limited to organic systems. Living polymerization of inorganic monomers is a comparatively new area of interest and opens the door to novel inorganic-organic hybrids.1324 Recently, examples of these types of hybrid polymers exhibiting solution self-assembly have been reported.2329 Recently, the living polymerization of phosphaalkenes was discovered by Kevin Noonan in our group, thereby establishing an innovative route to phosphorus-containing block copolymers.2° To date, only block copolymers (6.3) prepared from styrene (6.1) and phosphaalkene 6.2 have been prepared (see Scheme 6.1). Unfortunately, polystyrene (PS) and poly(methylenephosphine) (PMP) have similar solubility properties since they are both soluble in polar organic solvents such as CH2I and THF and are both insoluble in non-polar solvents (eg. n-hexanes) and protic solvents such as H20 and MeOH. Therefore, PS—b—PMP polymers are not suitable candidates for solution self-assembly studies. The search for a different A-block polymer with reasonable solubility in non-polar solvents led to the preparation of polyisoprene block-poly(methylenephosphine) (PlmbPMPn). Polyisoprene (P1) is soluble in linear aliphatics whereas PMP is not. Herein, the preparation of two block copolymers, PImbPMPn (m =308, n References start on page 130 I I I I I I I Chapter 6 117 = 46; and m = 222, n = 77) and the self-assembly of these novel macromolecules into micelles in P1 block selective solvent are reported. E 1 1. n MesP=CPh2(6.2) I Ph1 n-BuLi 25 °C THF 25°C, THF nBu4LPSj_Li 2. MeOH n-Bu-PS m PS-b-PMP Scheme 6.1 6.2 Results and Discussion Our objective was to determine the possibility of forming micelles in a block selective solvent with the block copolymers PlmbPMPn. To ascertain whether the formation of self- assembled structures occurred, transmission electron microscopy (TEM) and dynamic light scattering (DLS) were employed. Good contrast for TEM requires the presence of large atoms (for example, transition metals) in specific domains. The phosphorus atoms in the PMP block can be exploited through metallation with gold(l). Moreover, the block copolymer PlmbPMPn would provide a template to organize well-defined gold nanostructures. In chapter 5, I showed that homopolymer 6.4a and its model compound 6.4b were easily complexed to gold(l) through a simple ligand substitution reaction with (tht)AuCl (tht tetrahydrothiophene) resulting in gold- containing polymer 6.5a and phosphine 6.5b, respectively.30The larger scattering power of gold-containing domains (PMPAu) compared to the domains of the organic P1 provide sufficient contrast for TEM images. As a result, the domains containing gold can be located because they will appear darker than the organic domains. In addition to simply using gold as a contrast agent for TEM imaging, the possibilities to utilize the metal atoms are endless. For example, the gold(l) can be reduced to gold(O) to form PImbPMPn stabilized gold nanoparticles or to prepare gold nanowires. References start on page 130 Chapter6 118 CI r Phi rAu Phi I n (tht)AuCI .1. R-1--P—C---j-H R-{--P—C---j-H I I CH2I,RT I LMes Phjn -n tht [Mes Ph]n 6.4a: n>1, R=n-Bu 6.5a: n> 1, Rn-Bu 6.4b:n=1, R=Me 6.5b:n=1, R=Me 6.2.1 The preparation of PImbPMPn block copolymers 6.9a (m = 308, n = 46) and 6.9b (m = 222, n = 77) The block copolymers PlmbPMPn were prepared through the successive anionic polymerization of isoprene 6.6 and phosphaalkene 6.2. To prepare 6.9a (m = 308, n = 46), a substoichiometric amount of n-BuLi (0.3%) was added quickly to a THE solution of isoprene. The reaction mixture was stirred for 45 minutes. Subsequently, a small amount of the reaction mixture (containing lithiated P1 6.7) was removed and dropped into MeOH (50 mL). The resulting organic polymer 6.8 precipitated with MeOH as a white solid. After isolation, the polyisoprene sample 6.8 was analyzed by triple detection GPC (M = 21 000 g mo11, PDI = 1.02). A polyisoprene chain with a molecular weight of 21 000 g mo11 has 308 monomer repeat units. Since synthetic polymers are mixtures with a distribution of chain lengths, 308 repeat units is just an approximation. Also, the uncertainty in the GPC analysis for the sample is ±2000 g mo11 or 30 repeat units. For illustrative purposes, the P1 block will be treated as if it were exactly 308 repeat units in order to more easily discuss the differences in the polymer samples 6.9a and 6.9b. A solution of MesP=CPh26.3 in THE was added to the remaining reaction mixture containing living poly(isoprenyl) macroinitiators 6.7. The reaction mixture was stirred for 12 hours and then treated with MeOH. 31P NMR spectroscopy was employed to confirm that the phosphaalkene monomer 6.3 (&lp = 233) was completely consumed. The block copolymer 6.9a showed a broad resonance at —5 ppm. The polymer was precipitated from the crude reaction References start on page 130 Chapter6 119 mixture by adding MeOH (100 mL). After a second precipitation of a THF solution with MeOH, the product was isolated as a yellow powder in high yield (87%). After drying in vacuo, 6.9a was analyzed by GPC-LLS (M = 35600 g moF1, PDI = 1.03) and determined to have approximately 46 repeat units in the PMP block, and therefore, is approximately 10 mol % PMP. E 1 1. n MesP=CPh2(6.2) Ph1 n-BuLi I I 25 °C THF I _____ Im 25°C, THF 2. MeOH n..Bu_t[PI m mMesPhn 6.6 6.7 PImbPMPn 6.9a (m = 308, n = 46) MeOH 6.9b (m = 222, n = 77) n-Bu P1 H Urn 6.8 The block copolymer with higher phosphorus content 6.9b (m = 222, n = 77) was prepared in analogous fashion to 6.9a. Before the addition of the phosphaalkene monomer 6.2, an aliquot was removed from the reaction mixture (6.7 in THE) and was quenched with and precipitated with MeOH. After drying, triple detection GPC data suggested P1 6.8 had approximately 222 repeat units (M = 15 100 g moM, PDI = 1.06). To the remaining reaction mixture, a solution of phosphaalkene 6.2 in THE was added. After being stirred for 12 h and subsequent workup of the reaction, the final block copolymer 6.9b was isolated and analyzed by GPC-LLS (M = 39 400 g mor1; PDI = 1.06). Therefore, 6.9b has ca 77 repeat units in the PMP block corresponding to 26 mol % PMP. References start on page 130 Chapter 6 120 6.2.2 Coordination of block copolymers to gold(I), 6.lOa and 6.lOb The PMP blocks are repeating phosphine units that can be readily coordinated to gold(I) moieties. A solution of 6.9a (m 308, n = 46) in CH2I was treated with a solution of (tht)AuCI in CH2I and stirred for 1 h. Afterward, an aliquot was removed for analysis by 31P NMR spectroscopy. Importantly, the resonance for the free block copolymer (631p = — 5 ppm) was not observed and a new resonance at 25 ppm appeared and is assigned to 6.lOa. In comparison to the homopolymer &4a and its model compound 6.4b, similar downfield shifts were observed (homopolymer 6.4a vs. 6.5a: A6 = 32; model compound 6.4b vs. 6.5b: M = 39; block copolymers 6.9a vs. 6.lOa: A5 30). The volatiles were removed in vacuo to afford the metallated polymer 6.lOa as a yellow solid. Further characterization of 6.lOa was gathered using triple detection GPC-LLS (M = 45000 g mo11, PDI = 1.14). The 10600 g mor’ increase in molecular weight after coordination (from 35 600 to 45 000 g mor1) is consistent with that expected for the coordination of AuCI moieties at phosphorus — corresponding well with the approximately 50 P atoms in the backbone. Cl F Phi r Au Phi I ______.. I I n (tht)AuCI ______ n-Bu-f-Pl P—C--f-H n-Bu-f-Pl P—C—f-H I I I 25°C, DCM I I I L m Mes Ph] , -n tht L m Mes Ph] PImbPMPn PImb(PMPAU)n 6.9a (m = 308, n = 46) 6.lOa (m = 308, n = 46) 6.9b (m = 222, n = 77) GlOb (m = 222, n = 77) Analogous to the preparation of 6.lOa, the preparation of gold-containing block copolymer 6.lOb involved the addition of (tht)AuCI to 6.9b (m = 222, n = 77) in CH2I.The resulting gold-containing polymer was analyzed by GPC-LLS and determined to have an M 47 100 g mor1 (PDI = 1.15) tallying to an increase of 7 700 g moF1 or the addition of approximately 30 AuCI units. References start on page 130 Chapter 6 121 6.2.3 Solution self-assembly of gold-containing block copolymers Plmb (PMPAu) With the two metallated block copolymers in hand, solvent conditions in which the polymers would assemble into macromolecular structures were sought. Additionally, the differences in the solubilities of the two block copolymers will be exploited in order to explore the possibility of controlling the structures of the self-assemblies. Both PMP and P1 are soluble in polar solvents like THE and CH2I but only P1 is soluble in aliphatic solvents. Gold-containing polymer 6.lOa (m = 308, n 46) has a relatively shorter PMP block at approximately 10 mol % of the polymer. This block copolymer is expected to be more soluble in non-polar solvents than 6.lOb (m = 222, n 77) with —30 mol % PMP. The nature and size of the self-assembled structures of 6.lOa and 6.lOb were assessed using TEM and confirmed by DLS. Initially, dilute solutions of 6.lOa were prepared in a variety of aliphatic solvents. Promising results were achieved by employing dilute solutions of n-heptane. With the relatively long polyisoprene block, 6.lOa was completely dissolved in n-heptane (3 mg /4 mL). A 3 jL aliquot of the dilute solution was dropped onto a copper TEM grid coated with Formvar (a grid support film) and allowed to evaporate to dryness. Clearly visible in the TEM images (see Figure 6.2 (a) and (b)), are oblong-shaped micelles with lengths ranging from 15 nm to 150 nm (mean = 53 nm; a = 25 nm). In a sample of 30 micelles taken from overall 1/3 of the image, the widths have a relatively narrow range of 15 to 30 nm with an average of 21 nm (a = 4). Analysis of the solution by DLS spectroscopy showed that in solution the species had hydrodynamic radii (Rh) of 94 nm at 25 °C confirming the presence of macromolecular structures. The sizes are larger than those measured by the TEM images. In the TEM images, the size of the micelles where gold is present, i.e. the core containing PMP, can only be measured. The DLS measures the entire micelle, containing both the PMP and P1 blocks. References start on page 130 Chapter 6 122 c - • • ‘•-, •,-- •• •, .1 ;\;:- ;: •100 nm • •‘‘• Figure 6.2 TEM images of 6.lOa (m = 308, n = 46): a) n-heptane (3 mg /4 mL), magnification 1x105; b) n-heptane (3 mg/4 mL), magnification 2x105; C) n-heptane (2mg /4 mL), magnification 5x1 o; d) n-heptane (2 mg / 4 mL), magnification 2x1 o; e) n-heptane:THF (7:1)(3 mg/4 mL). References start on page 130 Chapter 6 123 To ensure that these oblong structures (with average dimensions of 55 nm x 21 nm) were not the result of aggregation during the evaporation process, another more dilute solution in n-heptane (2 mg /4 mL) has been prepared. The TEM images (Figure 6.2 (c) and (d)) demonstrate that the oblong shapes are still present though possess somewhat smaller in dimensions. The widths are comparable to the more concentrated sample, ranging from 10 to 30 nm (mean = 19 nm; y= 6 nm). However, the lengths of the macromolecular structures are noticeably shorter with an average of 34 nm (a = 19 nm). Likely, the larger structures in Figure 6.2 (a) and (b) are due to smaller micelles aggregating into larger masses. Finally, to provide conclusive evidence that the structures were a result of solution self- assembly, 3 mg of block copolymer polymer 6.lOa were dissolved in 4 mL of a mixture of n heptane:THF (7:1). The small amount of THE should solubilize the PMP block thereby destroying the micelles. Figure 6.2 (e) unmistakably shows the lack of large structures as the gold atoms are fairly evenly distributed throughout the image. The block copolymers 6.lOa have been shown to form well-defined nanostructures from self-assemblies in a P1 block selective solvent. In particular, oblong-shaped micelles (with lengths averaging 50 nm and widths of 20 nm) have been observed. Now, the investigation of the possibility of controlling the shape of the micelles, by controlling the proportion of the two blocks while maintaining a similar molecular weight distribution, will be explored. The block copolymer 6.lOb (m = 222, n = 77) that has a shorter P1 block and a longer PMP block than 6.lOa (m 308, n 46) was prepared and therefore expected to have a different assembly in solution. However, with a higher PMP content, block copolymer 6.lOb was not soluble in n heptane alone at the desired concentration (3 mg / 4 mL). A small amount of THF was required to completely solubilize the sample (0.5 mL for every 3.5 mL of n-heptane). See Figure 6.3 (a) and (b) for the TEM images of the evaporated samples. In the solvent mixture, the block copolymer 6.lOb also demonstrated the ability to self-assemble into micelles, albeit from slightly References start on page 130 Chapter 6 124 different solvent conditions, with a different appearance to those from the P1 rich block copolymer 61Oa. Micelles formed from 6.lOb in the solvent mixture are generally oblong in shape with a width of 18 nm (c = 3 nm). These cylindrical micelles have lengths of up to 170 nm (mean = 97 nm; c = 43 nm). The species in solution were found to have an average hydrodynamic radius of 75 nm at 25°C by DLS thereby demonstrating that the micellar structures are retained in solution. To confirm the reproducibility of this remarkable self- assembly, a second sample of 61Ob was analyzed in an analogous fashion (TEM: Figure 6.3 (c) and (d)). a).-. •o.. 1 . •••‘% ( p 4 F — .1 /• ( - —b.. . .-.. - _%• __., ‘ .( pllOOnm. c) f :: Trial References start on page 130 b) .1 ‘F -I I. • 100 nrn lOOnm Figure 6.3 TEM images of 6.lOb (m = 222, n = 77) n-heptane:THF (7:1) (3mg /4 1: a) magnification 1x105; b) magnification 2x105;Trial 2: c) magnification 1x105; d) magnification 2x1 5• Chapter 6 125 The shapes of the micelles from 6.lOb can be interpreted in a couple of ways from the TEM images. Either, the micelles are spherical in shape which have self-organized into necklace-like linear aggregates. Another possibility is that under the solvent conditions used to obtain the TEM images in Figure 6.3, the boundary between spherical and cylindrical micelles has been unexpectedly discovered and therefore both types are observed (see Figure 6.4 (b)). Further experiments to determine the types of micelles are ongoing. Regardless, the observed assemblies are believed to be due primarily to the polymer structure and not the solvent conditions. Recall that a sample of 6.lOa (m = 308, n = 46) in a mixture of n-heptane and THF (7:1) was analyzed under analogous conditions and found no evidence of macromolecular structures from the TEM images. Gold-containing spherical micelles that self-organize into necklace-like aggregates are known though with organic block copolymers and reduced gold.31 (a) Spherical Micelles * Figure 6.4 Sketches of two basic structural motifs for spherical micelles and (b) cylindrical micelles. 6.3 Conclusion New hybrid inorganic/organic block copolymers (Plm—PMPn) with varying m:n ratios were prepared. The functionality of the PMP block was exploited by coordinating gold(l) moieties to the backbone P atoms resulting in new gold-containing polymers. Furthermore, the solution self assembly of the functionalized block copolymers was investigated and the resulting novel (b) Cylindrical Micelles in dilute solution: (a) References start on page 130 Chapter 6 126 nanostructured assemblies with interesting shapes were characterized by TEM and DLS. In P1- specific solvents, P1m (PMPAu) co-polymers will assemble into micelles with a PMPAu core and P1 corona. The micelles varied in shapes (from spherical to oblong) sizes according to the block ratio (m:n). These materials are excellent candidates for applications in nanotechnology, such as the preparation of gold nanowires after reduction of the Au(l) atoms. 6.4 Experimental 6.4.1 Materials and General Procedures All manipulations of oxygen- and/or moisture-sensitive compounds were performed under a nitrogen atmosphere using standard Schlenk and/or glovebox techniques. Hexanes and dichloromethane were deoxygenated with nitrogen and dried by passing through a column containing activated alumina. THE was freshly distilled from sodium/benzophenone ketyl. Distilled methanol was degassed prior to use. CD2I was purchased from Cambridge Isotope Laboratories and were dried over molecular sieves (3 A). n-BuLi (1.6 M in hexanes) and n heptane were purchased from Aldrich and used as received. Alkyllithium reagents were titrated prior to use. Isoprene was purchased from Aldrich and distilled prior to use. MesPCPh2(6.3)32 and (tht)AuCI33were prepared following literature procedures. 31P NMR spectra were recorded on BrukerAV300 orAV400 spectrometers at room temperature. Chemical shifts for 31P NMR spectra are reported relative to H3P04as an external standard (85% in H20, ö 0). Molecular weights were estimated by triple detection gel permeation chromatography (GPC-LLS) using a Waters liquid chromatograph equipped with a Waters 515 HPLC pump, Waters 717 plus autosampler, Waters Styragel columns (4.6 x 300 mm) HR5E (2000-4 000 000), HR4 (5000-500 000), and HR2 (500-20 000), Waters 2410 differential refractometer ( = References stan on page 130 Chapter 6 127 940 nm, 40 °C), Wyatt tristar miniDAWN (laser light scattering detector operating at = 690 nm), and a Wyatt ViscoStar viscometer. A flow rate of 0.5 mL mm-I was used, and samples were dissolved in THE (Ca. 2 mg mL-1). TEM images were collected with the UBC Biolmaging Eacility Hitachi H7600 TEM (120kV tungsten filament) equipped with a side mount AMT Advantage (1 mega-pixel) CCD camera (Hamamatsu ORCA). The sample solutions were deposited on a Eormvar-coated TEM grid and allowed to evaporate to dryness. DLS experiments were performed with a Coulter N4 Plus DLS instrument equipped with a 10-mW helium-neon laser (A = 632.8 nm) and thermoelectric temperature controller. Measurements were taken at a 90° scattering angle in a 3 x 3-mm quartz cuvette on solutions that had been equilibrated at 25 °C. Particle sizes were estimated by SDP weight and intensity analyses. 6.4.2 Preparation of 6.9a (m = 308, n = 46) To a stirred solution of isoprene (0.68 g, 0.010 mol) in THE (0.5 mL) was added 20 iL of n-BuLi (1.36 M, 0.027 mmol) in hexanes via 100 p.L syringe. The reaction mixture turned yellow immediately and was stirred for approximately 45 mm. A drop of the reaction mixture was removed and added to 50 mL of methanol. The organic polymer 6.8 precipitated out of the solution as a white solid. After isolation, polyisoprene 68 was dried in vacuo and analyzed by using GPC-LLS (M = 21 000 (± 2000) g mor1, PDI = 1.02, dn/dc = 0.129). To the reaction mixture containing the active poly(isoprenyl lithium) species, a solution of MesP=CPh2(0.285 g, 0.901 mmol) in a minimal amount of THE was added. The solution immediately turned red and was allowed to stir for 24 h at which point the reaction mixture was almost black in colour. The solution was added dropwise into methanol (100 mL) and filtered. A second precipitation was performed from a concentrated solution of 6.9a in THE into MeOH (100 mL). Volatiles were References start on page 130 Chapter 6 128 removed in vacuo at 60 °C for several hours. The product was isolated as a yellow powder. Yield = 0.844 g (87%). 31P NMR (THE, 162 MHz): = —5 ppm. GPC-LLS (THE): M = 35 600 (± 300) g mor1, PDI 1.03 (±0.01), dn/dc = 0.134. 6.4.3 Preparation of 6.9b (m = 222, n = 77) To a stirred solution of isoprene (0.68 g, 0.010 mol) in THE (0.5 mL) was added 20 jiL of n-BuLi (1.36 M, 0.027 mmol) in hexanes via 100 jtL syringe. The reaction mixture turned yellow immediately and was stirred for approximately 45 mm. A drop of the reaction mixture was removed and added to 50 mL of methanol. The polymer 6.8 precipitated out of the solution as a white solid. After isolation, polyisoprene 6.8 was dried in vacuo and analyzed by using GPC-LLS (M 15 100 (± 1 300), PDI = 1.06, dn/dc = 0.129). To the reaction mixture containing the active poly(isoprenyl lithium) species, a solution of MesP=CPh2(0.520 g, 0.0016 mol) in a minimal amount of THE was added. The solution immediately turned red and was allowed to stir for 24 h. The solution was added dropwise into methanol (2 x 100 mL) and filtered. Volatiles were removed in vacuo at 60 °C for several hours. The product was isolated as a yellow powder. Yield = 0.86 g (72%). 31P NMR (THE, 162 MHz): — 5 ppm. GPC-LLS (THE): M 39 400 (± 300) g mo11, PDI = 1.06 (±0.07), dn/dc = 0.157. 6.4.4 Preparation of gold-containing 6.lOa A solution of 6.9a (98.9 mg, M = 35 600, PDI = 1.03) in CH2I (5 mL) was treated with a solution of (tht)AuCl (33 mg, 103 mmol) in CH2I (5 mL). After being stirred for 1 h, an aliquot was removed and analyzed by 31P NMR spectroscopy ( = 25 br). The solvent was removed in References start on page 130 Chapter 6 129 vacuo resulting in a light yellow solid. The product was dissolved in toluene (3 mL x 2) followed by hexanes (5 mL x 2) and dried in vacuo to ensure the removal of tht. The product was dried overnight in vacuo. Yield = 119 mg (97%). TEM and DLS sample concentrations: 2.9 mg in 4 mL n-heptane. 31P NMR (THF, 162 MHz): = 28 - 21 (broad). GPC-LLS (THF): M = 42 000 (±1 000), PDI = 1.10 (±0.07), dn/dc = 0.137. DLS: Rh = 95 nm. 6.4.5 Preparation of gold-containing 6.lOb To a stirred solution of block copolymer 6.9b (kn491) (50 mg, M = 37 300, PDI = 1.06) in CH2I (3 mL) was added (tht)AuCI (25 mg, 78 mmol) in CH2I (2 mL). The solvent was removed in vacuo resulting in a yellow solid. The product was dissolved in a minimal amount of CH2I and precipitated with hexanes. Toluene (3 mL x 2) followed by hexanes (5 mL x 2) were added to the product and subsequently removed in vacuo to ensure the removal of tht. The product was dried overnight in vacuo. Yield = 54 mg (79 %). TEM sample: 3.1 mg in 0.5 mL THE and 3.5 mL n-heptane. 31P NMR (121.5 Hz, THE): = 25 (broad). GPC-LLS (THF): M = 47 100(1 000) g mo11, PDI = 1.15(2), dn/dc = 0.155. DLS: Rh = 75 nm. References start on page 130 Chapter 6 130 6.5 References 1. G. Odian, Principles of Polymerization. Wiley & Sons: New York, 1991. 2. 0. W. Webster. Science 1991, 251, 887. 3. G. Riess. Prog. Polym. Sci. 2003, 28, 1107. 4. S. B. Darling. Prog. Polym. Sci. 2007, 32, 1152. 5. I. W. Hamley. Soft Matter2005, 1,36. 6. I. W. Ham ley, Block Copolymers in Solution: Fundamentals and Applications. Wiley: 2005; p 300. 7. K. Kita-Tokarczyk, J. Grumelard, T. Haefele, W. Meier. Polymer 2005, 46, 3540. 8. L. F. Lindoy, I. M. Atkinson, Self-Assembly in Supramolecular Systems. 1st ecL; Royal Society of Chemisty: 2001; p 223. 9. R. E. Cohen. Curr. Opin. Solid State Mater. Sci. 1999, 4, 587. 10. I. W. Hamley. Nanotechnology 2003, 14, R39. 11. M. Lazzari, M. A. Lopez-Quintela. Adv. Mater 2003, 15, 1583. 12. C. Park, J. Yoon, E. L. Thomas. Polymer 2003, 44, 6725. 13. C. Kioninger, M. Rehahn. Macromolecules 2004, 37, 1720. 14. M. Tanabe, I. Manners. J. Am. Chem. Soc. 2004, 126, 11434. 15. J. Bauer, N. Husing, G. Kickelbick. J. Polym. Sc, Part A: Polym. Chem. 2004, 42, 3975. 16. S. Ndoni, M. E. Vigild, R. H. Berg. J. Am. Chem. Soc. 2003, 125, 13366. 17. D. A. Rider, K. A. Cavicchi, K. N. Power-Billard, T. P. Russell, I. Manners. Macromolecules 2005, 38, 6931. 18. T. J. Peckham, J. A. Massey, C. H. Honeyman, I. Manners. Macromolecules 1999, 32, 2830. 19. L. Cao, I. Manners, M. A. Winnik. Macromolecules 2001, 34, 3353. 20. K. J. T. Noonan, D. P. Gates. Angew. Chem. Int. Ed. 2006, 45, 7271. References start on page 130 Chapter6 131 21. M. Roerdink, T. S. van Zanten, M. A. Hempenius, Z. Zhong, J. Feijen, G. J. Vancso. Macromol. Rapid Commun. 2007, 28, 2125. 22. Z. Wang, G. Masson, F. C. Peiris, G. A. Ozin, I. Manners. Chem.-Eur. J. 2007, 13, 9372. 23. X.-S. Wang, M. A. Winnik, I. Manners. Macromol. Rapid Commun. 2002, 23, 210. 24. T. Sanji, K. Takase, H. Sakurai. J. Am Chem. Soc. 2001, 123, 12690. 25. J. F. Gohy, B. G. G. Lohmeijer, A. Alexeev, X. S. Wang, I. Manners, M. A. Winnik, U. S. Schubert. Chem. Eur. J. 2004, 10, 4315. 26. H. R. Ailcock, E. S. Powell, Y. Y. Chang, C. Kim. Macromolecules 2004, 37, 7163. 27. A. H. Gabor, E. A. Lehner, G. P. Mao, L. A. Schneggenburger, C. K. Ober. Chem. Mater. 1994, 6, 927. 28. D. A. Rider, I. Manners. Polymer Reviews 2007, 47, 165. 29. H. Wang, M. A. Winnik, I. Manners. Macromolecules 2007, 40, 3784. 30. B. H. Gillon, B. 0. Patrick, D. P. Gates. Chem. Commun. 2008, accepted. 31. 5. Abraham, C. S. Ha, C. A. Batt, I. Kim. J. Polym. ScL, Part A: Polym. Chem. 2007, 45, 3570. 32. M. Yam, J. H. Chong, C.-W. Tsang, B. 0. Patrick, A. E. Lam, D. P. Gates. lnorg. Chem. 2006, 45, 5225. 33. R. UsOn, A. Laguna, M. Laguna. lnorg. Synth. 1989, 26, 85. 34. C. Jackson, Y. J. Chen, J. W. Mays. J. AppI. Polym. Sc!. 1996, 61, 865. References start on page 130 132 Chapter7 Overall Conclusions and Future Work 7.1 Summary of Thesis Work Shortly before I had joined, the Gates group had successfully polymerized PtC bonds for the first time through both anionic and radical methods.1The resulting poly(methylenephosphine)s 7.1 were characterized mainly by convential GPC and NMR spectroscopy. Some questions remained about these novel phosphorus-containing polymers. In particular, we wanted to know if the mechanism of anionic polymerization for phosphaalkenes was analogous to that for alkenes. We were interested in the site of initiation and the nature of the endgroups. Additionaly, we wanted to functionalize poly(methylenephosphine)s at the P atom with transition metals. F Phi I I I L Mes Ph 7.1 The initiation and termination steps in the anionic polymerization of P=C bonds have been modeled and are reported in Chapter 2.2 The initiation step was investigated through the stoichiometric reaction of MesP=CPh2(7.2) with RLi (R = Me or n-Bu). The addition was highly regioselective with the formal attack of R at phosphorus. The termination step in the anionic polymerization of 7.2 was simulated by the subsequent, in situ addition of electrophiles (H from H20 or MeOH; Me from Mel; (NEt2)Pfrom CIP(NEt2);or Me2RSi from Me2HSiCI or Me3SiCI). As a result, five novel tertiary phosphines (7.3a-e) and one diphosphine (7.3f) were prepared and function as models for poly(methylenephosphine)s. References start on page 135 Chapter 7 133 Ph / P=c / \Mes Ph 7.2 Ph Ph RLi R_P_C:eeLi EX R—P—C—E I I -LiX I IMes Ph Mes Ph 7.3a: R = Me, E=H; 7.3b: R = Me, E = Me; 7.3c: R = n-Bu, E = H; 7.3d: R = Me, E = P(NEt2) 7.3e: R = Me, E = SiMe3; 7.31: R = Me, E = S1HMe2 In Chapter 3, I showed that the chain growth step in the anionic polymerization mechanism for P=C bonds proceeds in analogous fashion to olefins.3The oligomeric products were characterized by MALDI-TOF MS from which endgroups were observed. However, analysis by MALDI-TOF MS resulted in the observation of fragments of the oligomers. Addtionally, I showed that high temperatures were not necessary; previous work on the polymerization of phosphaalkene 7.2 with MeLi (5%) in a minimum of solvent required a temperature of 150 oCi Anionic oligomerization of phosphaalkenes can be extended to the use of another initiator (n-BuLi) and to other monomers with functional groups (F and OMe). ro An 11111 H20 R+-P—C-+H I I I I LMeS PhJn 7.4a: Ar = Ph; R = Me 7.4a: Ar = Ph; R = n-BuLl 7.4b: Ar = p-C6H4OMe; R = Me 7.4c: Ar = p-CF; R = Me Chapter 4 describes the coordination of poly(methylenephosphine) to group 6 metal carbonyl complexes. Phosphine 7.3 has been used as a model for 7.1 and was used to develop Ar / RL1P=C / \ Mes Ph Et20, -78 °C 7.2:Ar Ph 7.2a: Ar = p-C6H4OMe 7.2b: Ar = p-CF Ar r Ar1 2or3equiv3.1 I I Et20 R-f_—_1_LiR—P—C—Li Mes Ph LMeS Phjn MeOH or H20 F Ar R-{-P-C*H LMs Ph]n References start on page 137 Chapter 7 134 a route to metal-containing polymers. As a result, phosphine-metal complexes 7.5a, 7.5b and 7.5c were prepared through the thermally induced substitution of a CO group. Partial metallation of polymer 7.1 with W, Mo and Cr was achieved. The objective was to achieve high metal loading of the polymer 7.1 with the hopes of preparing useful metal-containing materials or ceramic precursors. We now are certain that phosphorus-metal bonds can be formed. However, for total metallation of polymer 7.1, a less sterically hindered, larger metal center should be selected. (CO)5M F Phi Me-I—P-C--i-H LMesPhin 7.5a: n=1; M =W 7.5b: n= 1; M= Mo 7.5c: n=1; M =Cr 7.5d: n> 1; M =W 7.5e: n> 1; M = Mo 7.5f: n>1; M =Cr Complexation of the polymer 7.1 and its model compound 7.3a to Au(l) is described in Chapter 54 The complexation was easily extended to polymer 7.1. I showed that most of the P atoms in the polymer backbone were coordinated to gold(l). Cl Au Et Phi n-Bu P—C H [Mes Phjn 7.6a: n I 7.6b: n > 1 We successfully prepared new hybrid inorganic/organic amphiphilic block copolymers (PImbPMPn). Utilizing the straightforward functionalization of poly(methylenephosphine)s 7.1 with Au(l), we easily coordinated the novel block copolymers to gold(l). Furthermore, the solution self-assembly of the functionalized block copolymers has been investigated resulting in References start on page 137 Chapter 7 135 assemblies with interesting shapes. The micelles varied in shapes (from spherical to cylindrical) sizes according to the block ratio (m:n). r ___AUPh1 n-ButPl P—C—j-H L m MsPhj Plm&(PMPAU)n 7.2 Future Work 7.2.1 Uses for model compounds of poly(methylenephosphine) The end-group functionalization of the model compounds can be extended to the polymer. Using functional electrophiles in the termination step of the polymerization would access other types of block copolymers: telechelic polymers. 7.2.2 The preparation of polymer with high tungsten-content A different route for tungsten-containing polymers may result in higher W content could be used as a precursor to WP ceramic materials. Some preliminary investigations have suggested that the anionic initiation of complexes of the type 7.7 is possible with MeLi. Sterically unencumbered phosphaalkenes have been prepared as complexes to tungsten carbonyl fragments.56 This could also be a route to poly(methylenephosphine)s with smaller substituents (7.9). References start on page 137 Chapter 7 136 (CO)5W (CO)5W R MeLi Ph P=C Me—P—C—Li R’ R” R=Mes I IR’=R’=Ph MesPh (CO)W El. i n-Bu---f-P—C-j-H [F R”j 7.9 7.2.3 Gold-containing micelles • Extend the m:n ratio of the block copolymer, PImbPMPn, to further investigate the effect of the ratio on the shape of the self-assembled products. • Reducing the gold(I) cores in the micelles to gold(O) may form poly(methylenephosphine)-stabilized gold nanoparticles useful in nanotechnology and catalysis.7 7.3 Closing Remarks The mechanism of anionic polymerization of P=C bonds was more deeply investigated through the synthesis and characterization of model compounds and short-chain oligomers. This thesis also contains the first reports of metal-containing poly(methylenephosphine)s. Although only moderate amounts of group 6 metals were incorporated into the homopolymer, evidence indicates that M-P bonds were formed. The homopolymers were shown to be much better ligands for gold(l). In fact, gold incorporation was extended to amphiphilic block copolymers which were subsequently shown to form micelles of varying sizes. References start on page 137 Chapter 7 137 7.4 References 1. C.-W. Tsang, M. Yam, D. P. Gates. J. Am. Chem. Soc. 2003, 125, 1480. 2. B. H. Gillon, K. J. T. Noonan, B. Feldscher, J. M. Wissenz, Z. M. Kam, T. Hsieh, J. J. Kingsley, J. I. Bates, D. P. Gates. Can. J. Chem. 2007, 85, 1045. 3. B. H. Gillon, D. P. Gates. Chem. Commun. 2004, 1868. 4. B. H. Gillon, B. 0. Patrick, D. P. Gates. Chem. Commun. 2008, 2161. 5. A. Marinetti, S. Bauer, L. Ricard, F. Mathey. Organometallics 1990, 9, 793. 6. L. Weber, M. Meyer, H. G. Stammler, B. Neumann. Organometallics 2003, 22, 5063. 7. M. C. Daniel, D. Astruc. Chem. Rev. 2004, 104, 293. References start on page 137

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.24.1-0060594/manifest

Comment

Related Items