UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Metallated, beta-substituted phosphine-oligothiophenes Moorlag, Carolyn Patricia 2006

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
831-ubc_2006-130229.pdf [ 9.64MB ]
Metadata
JSON: 831-1.0059340.json
JSON-LD: 831-1.0059340-ld.json
RDF/XML (Pretty): 831-1.0059340-rdf.xml
RDF/JSON: 831-1.0059340-rdf.json
Turtle: 831-1.0059340-turtle.txt
N-Triples: 831-1.0059340-rdf-ntriples.txt
Original Record: 831-1.0059340-source.json
Full Text
831-1.0059340-fulltext.txt
Citation
831-1.0059340.ris

Full Text

METALLATED, BETA-SUBSTITUTED PHOSPHINE-OLIGOTHIOPHENES by C A R O L Y N PATRICIA M O O R L A G B.Sc. (Honours, Co-op), Chemistry, University of Victoria, Canada, 1997^ M.Sc, Metals and Materials Engineering, University of British Columbia, Canada, 2000 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 January 2006 © Carolyn Patricia Moorlag, 2006 Abstract The synthesis and characterization of transition metal complexes combined with rc-conjugated chains to yield hybrid metal-organic materials with new structural, chemical, and electronic properties are reported. P-Phosphinothiophene ligands (9, 11, 13-15, 21-22, 27) were prepared by metal-catalyzed coupling reactions, thienyl backbone conformations were determined by X-ray crystallography, and electronic properties were characterized by absorption and emission spectroscopy and electrochemical methods. Transition metal groups were attached in close proximity to rc-conjugated thienyl chains, at pendent positions or directly to the conjugated backbone, and metal-thienyl interactions were investigated. Pd(II) (29-30) and Au(I) (33-34) complexes were prepared by reaction of the metal groups with bis(phosphino)thiophene ligands. The conformational and electronic properties of the complexes were characterized, and the conjugation of the oligothiophene chains was modified by metal attachment. Ru(II) complexes (41-46, 47-52) are formed from bis(bipyridine) Ru(II) groups and mono(phosphino)thiophenes. Metal-mediated acid-base reactions change the mode of direct attachment of Ru(II) with the oligothiophene chains, between P,S and P,C bonding modes, resulting in different structural and electronic properties, as observed by absorption and emission spectroscopy, electrochemical oxidation and reduction, X-ray crystallography, spectroelectrochemistry, and electron paramagnetic resonance spectroscopy. These two modes of metal bonding to the rc-conjugated backbone constitute a reversible molecular switch. The incorporated bis(bipyridine) Ru(II) groups undergo charge-transfer transitions, and light absorption of 51 appears to remove charge from the pentathiophene chain, suggesting possible light-harvesting applications. Preliminary polymerization studies towards the preparation of polynuclear, metallated polythiophene poly-46 and a p-polysubstituted polythiophene derivative are reported. Long oligomers were observed by ' H N M R spectroscopy and mass spectrometry, but further studies are required. PPh2 C 6 H 1 3 Ph2P R Ph2P R H R X X H n H 1 3 C 6 Ph2P P 2 T 2 (9) n = 0 P 2 hex 2 T 4 ( l l ) n = l PT 3(13) R = R'=H PMeT 3(14) R = CH3, R' = H PMe 2T 3(15) R = R' = CH3 Phex2T5(21) R = C 6H 1 3 ,X = H PDo 2T 5(23) R = C 1 2 H 2 5 ,X = H PBr 2 Do 2 T 5 (27) R = C 1 2 H 2 5 , X = Br n r H 1 3 C 6 Q H 1 3 H 1 3 C 6 C 6 H 1 3 Pd(P 2 T 2 )Cl 2 (29) n = 0 (AuCl) 2 P 2 T 2 (33) n = 0 Pd(P 2hex 2T 4)Cl 2 (30) n = l (AuCl) 2P 2hex 2T 4 (34) n = l RuPT 3 - /»S (41) R = R'=H RuPMeT3-P,S (42) R = CH3, R' = H RuPMezTj-P^ (43) R = R' = CH3 s-N,. , K ^Ru\ | ^PPh2 RuPT3-P,C(47) R = R' = H RuPMeT 3-/» C (48) R = CH3, R' = H RuPMe2T3-/> C (49) R = R' = CH3 RuPhex2T5-P,5'(44) R = C 6H 1 3,X=H RuPhex2T5-P,C(50) R = C 6H 1 3,X = H RuPDo2T5-7>S (45) R = C 1 2 H 2 5 ,X = H RuPDo2T5-J>C(51) R = Ci 2H 2 5,X = H RuPBr 2Do 2T s-i',5 (46) R = C 1 2 H 2 5 , X = Br RuPBr2Do2T5-JP,C(52) R = C 1 2 H 2 5 , X = Br iii Table of Contents Abstract i i Table of Contents : iv List of Tables ; vii List of Figures viii List of Symbols xi List of Abbreviations xii List of Charts xvii List of Schemes xviii Acknowledgements -. xix CHAPTER 1 Introduction 1 1.1 Scope 1 1.2 Organic re-Conjugated Materials 2 1.2.1 Structures of it-Conjugated Materials 2 1.2.2 Synthesis of Oligo- and Polythiophenes 3 1.2.3 Properties and Applications of it-Conjugated Materials 6 1.3 Metal-Organic Hybrid Materials 9 1.3.1 Metallation of it-Conjugated Organic Materials 9 1.3.2 Properties and Applications of Metal-Organic Hybrid Materials 12 1.4 Focus of Present Study 15 1.5 Goals 16 1.6 References 17 CHAPTER 2 Synthesis and Characterization of p-Phosphinothiophene Ligands and Derivatives 20 2.1 Introduction 20 2.2 Experimental 25 2.2.1 General Experimental 25 2.2.2 Synthesis 25 2.3 Results 32 2.3.1 Synthesis 32 2.3.2 Cyclic Voltammetry 38 2.3.3 Electronic Spectroscopy 40 2.4 Discussion 42 2.5 Conclusions 44 2.6 References 45 iv CHAPTER 3 Synthesis and Characterization of Au(I) and Pd(II) Complexes 47 3.1 Introduction ; 47 3.2 Experimental 49 3.2.1 General Experimental 49 3.2.2 X-ray Crystallographic Analysis 50 3.3 Results 50 3.3.1 Synthesis and Crystallographic Studies 50 3.3.2 Cyclic Voltammetry 54 3.3.3 Electronic Spectroscopy 55 3.4 Discussion 57 3.5 Conclusions 59 3.6 References..... < 60 CHAPTER 4 Synthesis and Characterization of P.S-Bound Ru(II) Complexes 62 4.1 Introduction 62 4.2 Experimental 67 4.2.1 General Experimental 67 4.2.2 Synthesis 68 4.2.3 X-ray Crystallographic Analysis 71 4.2.4 Density Functional Theory Calculations 72 4.3 Results 72 4.3.1 Synthesis and Crystallographic Studies 72 4.3.2 Density Functional Theory (DFT) Calculations 77 4.3.3 Cyclic Voltammetry 79 4.3.4 Electronic Spectroscopy and Charge-Transfer Correlations 81 4.3.5 Reversible Molecular Switching 86 4.4 Discussion 88 4.5 Conclusions 92 4.6 References 92 CHAPTER 5 Synthesis and Characterization of P, C-Bound Ru(II) Complexes 96 5.1 Introduction 96 5.2 Experimental 98 5.2.1 General Experimental • 98 5.2.2 Synthesis 99 5.2.3 X-ray Crystallographic Analysis 102 5.2.4 Density Functional Theory Calculations '. 103 5.3 Results 103 5.3.1 Synthesis and Crystallographic Studies 103 5.3.2 Density Functional Theory (DFT) Calculations 106 5.3.3 Cyclic Voltammetry 108 5.3.4 Electron Paramagnetic Resonance 110 5.3.5 Optical Spectroscopy and Charge-Transfer Correlations 113 5.3.6 Spectroelectrochemistry 118 5.4 Discussion 122 5.5 Conclusions 127 v 5.6 References 128 CHAPTER 6 Preliminary Polymerization Studies 131 6.1 Introduction 131 6.2 Experimental 133 6.2.1 General Experimental 133 6.2.2 Polymerization Reactions 133 6.3 Results 135 6.3.1 Coupling of P-Halogenated Oligothiophenes 135 6.3.2 Coupling of Ru(II)-Complexes 138 6.4 Discussion 140 6.5 Conclusions 141 6.6 References 141 CHAPTER 7 Conclusions and Future Directions 142 7.1 Conclusions 142 7.2 Suggestions for Future Work 144 7.3 References 147 Appendix 1 Crystal Structure Data 148 v i List of Tables Table 2-1 Cyclic voltammetry data for (3-phosphinothiophene ligands and derivatives, oligothiophenes, and triphenylphosphine 38 Table 2-2 Electronic spectroscopy data for P-phosphinothiophene ligands and derivatives, and oligothiophenes 40 Table 3-1 Selected interatomic distances (A) and angles (°) for Pd(P2hex2T4)2Ci2 (30) 52 Table 3-2 Cyclic voltammetry data of Pd(II)- and Au(I)-bis(phosphino)thiophene complexes.. 54 Table 3-3 Electronic spectroscopy data of Pd(II)- and Au(I)-bis(phosphino)thiophene complexes 56 Table 4-1 Selected interatomic distances (A) and angles (°) for 41 and 43 73 Table 4-2 Selected interatomic distances (A) and angles (°) for RuPDo 2T 5-P,5' (45) 76 Table 4-3 Calculated H O M O and L U M O energies for RuPT 3-P,5 (41) 79 Table 4-4 Cyclic voltammetry data of Ru(II)-phosphinothiophene-i5,5' complexes 81 Table 4-5 Electronic spectroscopy data for Ru(II)-phosphinothiophene-/>,5' complexes 83 Table 5-1 Selected interatomic distances (A) and angles (°) for 47 and 49 105 Table 5-2 Calculated HOMO and L U M O energies for RuPT 3-P, C (47) 106 Table 5-3 Cyclic voltammetry data of Ru(II)-phosphinothiophene-/5,C complexes3 109 Table 5-4 Electronic spectroscopy data for Ru(II)-phosphinothiophene-/>, C complexes 115 Table 5-5 Spectroelectrochemistry data for P, C Ru(II)-phosphinothiophene complexes and related oligothiophenes 120 Table 6-1 Peaks corresponding to coupled oligomers in the mass spectrum of poly-26T 138 vii List of Figures Figure 1-1 Selected examples of rc-conjugated organic polymers 2 Figure 1-2 Selected examples of soluble polythiophenes 3 Figure 1-3 Metal-catalyzed coupling reactions used to prepare oligo- and polythiophenes 4 Figure 1-4 Mechanism of oxidative polymerization of thiophene 5 Figure 1-5 Energy band diagram of polyacetylene, as derived from overlapping rc orbitals 6 Figure 1-6 Changes in the irreversible oxidation potential, E p (O), and the TC—»TC* absorption wavelength, A , m a x ( A ) with extension of thiophene chain length 7 Figure 1-7 Localized distortions on polythiophene chains and the introduction of energy levels into the band gap due to the formation of (a) a polaron, and (b) a bipolaron 8 Figure 1-8 Classes of transition metal-organic hybrid materials 9 Figure 1-9 Metal-oligothiophene and metal-polythiophene hybrid materials, classed as Type I, Type II, and Type III 11 Figure 1-10 Modification of the band gap energy with metallation of rc-conjugated materials, due to inductive effects or chain twisting to increase the band gap (left), and electron donation or chain planarization (right) 13 Figure 1-11 Examples of planar and twisted conformations of an oligothiophene 13 Figure 1-12 The introduction of metal-based HOMO or L U M O levels into the band gap of a rc-conjugated chain, resulting in (a) a metal-to-ligand charge transfer (MLCT) transition, or (b) a ligand-to-metal charge transfer (LMCT) transition 14 Figure 1-13 General strategies for making (a) Type I and (b) Type II metal-oligothiophene hybrid materials using P-phosphinothiophenes 15 Figure 2-1 Bidentate and monodentate modes of bonding that may arise with metallation of bis(phosphino)thiophene chains 22 Figure 2-2 (a) ORTEP view of (PO) 2T 2 (12). (b) ORTEP view down the C 4 - C 5 axis illustrating the interannular torsion angle 34 Figure 2-3 Cyclic voltammetry of 22 and 24 in CH2CI2 39 Figure 2-4 Correlation between the oxidation potentials, Ei/2,ox, and the number of thienyl rings of the P-phosphinothiophene ligands (+) and the corresponding oligothiophenes (O) 39 Figure 2-5 Correlation between A , ! m a x and the number of thienyl rings of representative phosphinothiophene ligands (+) and the corresponding oligothiophenes (O) 41 Figure 2-6 The 7C<-TC* fluorescence emission of (phosphino)terthiophene ligands 13 and 16, and T 3 42 Figure 3-1 ORTEP view of Pd(P2hex2T4)Cl2 (30) 51 Figure 3-2 (a) ORTEP view of 33 and (b) ORTEP view of 33-tol 53 v i i i Figure 3-3 Cyclic voltammograms of (a) Pd(II) complex 30 and (b) Au(I) complex 34 in CH 2 C1 2 55 Figure 3-4 Absorption spectra of Pd(P2hex2T4)Cl2 (30) in solvents of different polarity 56 Figure 4-1 Schematic representation of a Gratzel cell 64 Figure 4-2 Two possible bidentate bonding modes of a P-phosphinothiophene chain attached to a Ru(bpy)22+ group, P,S and P,C 65 Figure 4-3 (a) ORTEP view of RuPT 3-P,S (41) (conformation A) and (b) ORTEP view of RuPMe2T3-P,S(43) . 73 Figure 4-4 A portion of the unit cell of RuPMe2T3-P,5' (43) viewed normal to the 010 plane... 75 Figure 4-5 X-ray crystal structure of RuPDo2Ts-.P,S'(45) (conformation B) 76 Figure 4-6 Ordering of the calculated HOMO and L U M O energies of RuPT 3-i 5,5' (41), and depictions of frontier orbitals 78 Figure 4-7 Cyclic voltammograms of (a) 41-43, and (b) 45 in C H 3 C N 80 Figure 4-8 Solution absorption spectra of Ru(II)-.P)S complexes 41-43 and 45 in CH 2 Ci2 82 Figure 4-9 Crystals of (a) RuPT 3-P,S (41) and (b) RuPDo 2T 5-P,5 (45) 82 Figure 4-10 Linear fit of Eop versus AE(redox) for R u n D - P S complexes 41-43 and 45 84 Figure 4-11 Solid state absorption spectra of Ru(II)-phosphinothiophene-i3,5' complexes 41-43 and 45 84 Figure 4-12 (a) Emission and excitation spectra of R u P D o 2 T 5 - P , 5 ' (45) in deaerated C H 3 C N solution, (b) Emission spectra of RuPDo2T5-P,5 (45) obtained at 12 ns, 37 ns, 90 ns and 143 ns average times after the incidence of the laser pulse 85 Figure 4-13 Emission spectra of RuPDo2T5-JP,S' (45) (a) at time = 0 h, and (b) at time = 0, 2.5, and 22 h 86 Figure 4-14 3 1 P N M R spectra of RuPDo 2T 5-P,5 (45) and RuPDo 2 T 5 -P,C (51). Addition of HPF6 (cone.) to a solution of 51 results in reversion to 45 88 Figure 5-1 (a) ORTEP view of RuPT 3-P,C (47) and (b) ORTEP view of RuPMe 2T 3-P,C(49) 105 Figure 5-2 (a) Ordering of the calculated HOMO and L U M O energies of RuPT 3-P,C (47), and depictions of the frontier orbitals 107 Figure 5-3 Cyclic voltammograms of (a) 47-49, and (b) 51 in C H 3 C N 109 Figure 5-4 EPR spectra at room temperature (RT = 298 K) and low temperature (110 K) of the first oxidized species of (a) RuPT 3-P, C (47), (b) RuPMeT 3 -P,C (48), and RuMe 2PT 3-P,C(49) 112 Figure 5-5 EPR spectra of the first oxidized species of (a) D02T5 (24) at room temperature (298 K), and (b) RuDo 2 PT 5 -P,C (51) at room temperature (298 K) and low temperature (110K) 113 Figure 5-6 Solution absorption spectra of Ru(II)-.P, C complexes 47-49 and 51 in CH2CI?. solution 114 Figure 5-7 (a) RuPT 3-P, C (47) crystals and (b) RuPDo 2T 5-P,C (51) powder 114 ix Figure 5-8 (a) Plot of the primary CT optical transition (E o p) versus the difference in potential between the first oxidation potentials and first reduction potential (AE) for all Ru(II)-/5, C complexes, and (b) the linear fit for the Ru(II)-(phosphino)terthiophene-P,C complexes 116 Figure 5-9 Solid state absorption spectra of Ru(II)-P, C complexes 47-49 and 51 drop-cast from solution in acetone 117 Figure 5-10 (a) Emission and excitation spectra of 47 (A.ex = 456 nm, A, e m = 748 nm), 48 (A.ex = 459 nm, Xem = 751 nm), and 49 (A,ex = 460 nm, Xsm - 761 nm) in deaerated C H 3 C N . (b) Emission spectra of 47 obtained at 14 ns, 62 ns, 113 ns and 137 ns average times after the incidence of the laser pulse 118 Figure 5-11 Difference spectra of 49 in deaerated C H 3 C N solution with 0.1 M [(«-Bu)4N]PF6 supporting electrolyte 119 Figure 5-12 Difference spectra of (a) complex 51 and (b) pentathiophene 24 in deaerated C H 3 C N solution with 0.1 M [(«-Bu) 4N]PF 6 supporting electrolyte 122 Figure 5-13 Proposed energy-level diagram and transitions, Ti and T 2 , for T n + 125 Figure 5-14 Proposed molecular orbital diagram for 51 + and transitions T ' i , T 2 and T'3, with energy levels O1-O4 126 Figure 5-15 Representation of Ru(II)-phosphino(pentathiophene)-P,C complex 51 as a light-harvesting molecular wire 127 Figure 6-1 ' H N M R spectra of the aromatic region of (a) heptathiophene 28 and (b) the coupled products of the Suzuki cross-coupling reaction of 26 and thiophene (T), to give poly-26T 136 Figure 6-2 MALDI-TOF mass spectrum of poly-26T, 700-4000 m/z region 137 Figure 6-3 3 1 P N M R (162.0 MHz) spectra in CO(CD 3 ) 2 of products of Ni(0) coupling reaction of 46 at 50°C (a) 2 h, (b) 16 h 139 List of Symbols Symbol Description 5 A s X A-Ern A-Ex Arnax V 9 P T o 1 I chemical shift (ppm) difference molar absorptivity (M _ 1 cm _ 1 ) quantum yield hapticity wavelength (nm) emission wavelength (nm) excitation wavelength (nm) wavelength at band maximum (nm) micro energy of X-rays used for crystallographic determination (mm - 1) frequency (s - 1) angle of diffraction density (g cm ) lifetime standard deviation degrees parallel perpendicular the sum of xi List of Abbreviations Abbreviation Description A Angstrom A amperes uA microamperes Ac acetate Anal. analysis aq aqueous a.u. arbitrary units bpy bipyridine Bu butyl cm centimeter cone. concentrated °C degrees Celsius Cp cyclopentadienyl CT charge-transfer d doublet dd doublet of doublets (NMR) dd doublet of doublets of doublets (NMR) DFT density functional theory Do dodecyl D02T5 3,3 ""-didodecyl-2,2': 5 ',2": 5 ",2"': 5 "',2""-pentathiophi dppf 1,1 '-bis(diphenylphosphino)ferrocene dppp 1,3 -bis(diphenylphosphine)propane DSSC dye-sensitized solar cell e~ electron E energy AE(redox) energy difference between redox couples (eV) E1/2 half wave redox potential (V) E c aic redox potential predicted by EL values (V) EL ligand electrochemical parameters (V) E o p transition energy (eV) xii E o x peak potential, oxidation process (V) E p peak potential, irreversible wave (V) E r e d peak potential, reduction process (V) EDOT 3,4-ethylenedioxythiophene EI electron ionization EPR electron paramagnetic resonance eq equivalents Et ethyl eV electron volts Fc ferrocene F c calculated intensity of a reflection from the crystal model F 0 measured intensity of a reflection in a diffraction pattern g * gram gx g-value GOF goodness of fit indicator h Planck's constant hv light energy hex hexyl H H head-to-head HOMO highest occupied molecular orbital HT head-to-tail Hz Hertz I symmetry-related reflection (crystallography) J magnetic coupling constant, N M R coupling constant K Kelvin L ligand L M C T ligand-to-metal charge transfer L U M O lowest unoccupied molecular orbital m multiplet M metal, molarity (mol L" 1), mass MALDI-TOF matrix-assisted laser desorption ionization time of flight M C metal-centered Me methyl mg milligram xiii MHz Megahertz mL milliliter M L C T metal-to-ligand charge transfer mm millimeter mmol millimole M O molecular orbital mol mole umol micromole MS mass spectra mV millivolts M W molecular weight m/z mass-to-charge ratio n number of units in an oligo- or polymeric chain n- normal NBO natural bonding orbital NBS A^-bromosuccinimide near-IR near-infrared NHE normal hydrogen electrode NIS /V-iodosuccinimide nm nanometer N M R nuclear magnetic resonance ns nanosecond OLED organic light emitting diode ORTEP Oak Ridge Thermal Ellipsoid Plot OPO optical parametric oscillator PAT polyalkylthiophene PBr2Do2Ts 5,5 ""-dibromo-3,3 ""-didodecyl-3 "-diphenylphosphino-2,2':5 ',2": 5 ",2 5 "',2 ""-pentathiophene P,C phosphine, thienyl carbon coordination PDo 2T 5 3,3""-didodecyl-3"-diphenylphosphino-2,2':5',2":5",2"':5"',2""-pentathiophene PEDOT pory(3,4-ethylenedioxythiophene) Ph phenyl Phex2T5 3,3""-dihexyl-3"-diphenylphosphino-2,2':5',2":5",2"':5"',2""-pentathiophene xiv PHT polyhexylthiophene PMeT3 3'-(diphenylphosphino)-5-methyl-2,2':5'2"-tertW PMe 2T 3 5,5Mimethyl-3'-(diphe^ (PO)2T2 3,3 '-bis(diphenylphosphoryl)-2,2'-bithiophene ppm parts per million PPV poly(p-phenylenevinylene) P,S phosphine, thienyl sulfur coordination PT3 3'-(diphenylphosphino)-2,2':5'2"-terthiophene ocPT3 2-diphenylphosphino-5,2':5',2"-terthiophene P2hex2T4 3,3"'-dihexyl-3',3"'-bis(diphenylphosphino)-2,5':2',2":5",2"'-quaterthiophene P 2 T 2 3,3'-bis(diphenylphosphino)-2,2'-bithiophene q quartet R residual (in crystallography, a measure of agreement between the crystallographic model and the X-ray diffraction data) RT room temperature (298 K) Ru bis(bipyridyl) Ru(II) group (for complex labels) s singlet (NMR data), second S Siemens sat. aq saturated aqueous SD standard deviation sep septet sh shoulder SCE saturated calomel electrode SOMO singly occupied molecular orbital t triplet T temperature, tesla T x transition T thiophene T 2 bithiophene T 3 terthiophene T 4 quaterthiophene T5 pentathiophene T 7 heptathiophene T„ oligothiophene of indeterminate chain length xv td triplet of doublets THF tetrahydrofuran tht tetrahydrothiophene tol toluene TT tail-to-tail UBC University of British Columbia V Volts V volume w least squares weights X halogen Z number of molecules in a crystallographic unit cell xvi List of Charts Chart 1-1 3 Chart 2-1 ; 20 Chart 2-2 : 21 Chart 2-3 21 Chart 2-4 22 Chart 2-5 23 Chart 2-6 ..... 24 Chart 2-7 24 Chart 3-1 48 Chart 3-2 48 Chart 3-3 '. 49 Chart 4-1 62 Chart 4-2 63 Chart 4-3 66 Chart 4-4 66 Chart 4-5 87 Chart 5-1 96 Chart 5-2 108 Chart 5-3 120 Chart 6-1 132 Chart 6-2 132 Chart 6-3 137 Chart 7-1 146 xvii List of Schemes Scheme 2-1 2 3 Scheme 2-2 3 3 Scheme 2-3 3 4 Scheme 2-4 3 5 Scheme 2-5 3 6 Scheme 2-6 3 7 Scheme 2-7 3 7 Scheme 3-1 5 0 Scheme 3-2 5 3 Scheme 4-1 6 3 Scheme 4-2 6 3 Scheme 4-3 • 6 5 Scheme 4-4 • .• • 7 2 Scheme 4-5 7 5 Scheme 4-6 • • 8 7 Scheme 4-7 • •' 9 1 Scheme 5-1 9 7 Scheme 5-2 1 0 4 Scheme 5-3 1 0 4 Scheme 5-4 1 2 1 Scheme 6-1 1 3 5 Scheme 6-2 1 3 9 Scheme 7-1 1 4 6 Scheme 7-2 1 4 6 xviii Acknowledgements There are many people whom I would like to thank for helping me to carry out and complete this thesis. I thank my supervisor, Mike Wolf, for his encouragement, for many thoughtful discussions, and for supporting me in the many directions that I wanted to take (including overseas) for this work. I appreciated his confidence in me. I would like all my group members, past and present. I was very lucky to have met such extraordinary people, from whom I gained insights about science and life. I especially thank Tracey Stott, who worked on a related project and was always willing to share helpful ideas, and was also a real friend. I thank all the undergraduate students who helped me with this work and contributed towards its final results: Shirley Lam, Mike Bridges, Eagrainie Yuh, Maureen Sam, and Michael Lam. The collaborations carried out were also very important toward the development of this thesis and I would like to thank all the people involved, they were: Daniel Leznoff, Cornelia Bohne, Y u Zhang, Alex Wang, Bipro Sarkar, Wolfgang Kaim, and Peter Bauerle. I am very grateful to Prof. Bauerle for not only inviting me to his group to carry out experiments and sharing his perspectives on thiophene chemistry with me, but for also supporting me while I was for three months in Germany. I also thank NSERC and U B C for scholarships. Finally, I thank my husband, Oliver, who was with me when I first decided that after a Master's, I still wanted to stay years more at university pursue a Ph.D. During the time since, his love, support, and understanding was a positive influence for me that would counteract even the most frustrating times. Without him, this thesis would not have been the same. xix CHAPTER 1 Introduction 1.1 Scope "Materials chemistry" is a rapidly emerging field of chemistry that as yet does not have a formal definition,1 though the Journal of Materials Chemistry describes its content as "the fabrication, properties and applications of materials, including synthesis, structural characterization and modeling".2 By this description, materials chemistry consequently encompasses many areas of scientific study. In accordance with its designation as a materials chemistry topic, this thesis brings together ideas and experimental methods from the areas of organic, inorganic, and physical chemistry, and electronic engineering, for the purpose of preparing and characterizing new metal-organic hybrid materials. It was desired that the materials targeted by this study would find applications in the area of molecular electronics, an expanding area of chemical research conducted with the aim of producing molecular equivalents for conventional electronic components such as wires, switches 3 5 and transistors. " The very small dimensions of molecular components would result in nanoscale devices that would potentially be more energy efficient and transport charge faster than current electronic systems. rc-Conjugated organic materials can conduct current via a delocalized rc-system along an organic molecular chain, and function as "molecular wires".6"8 Extensive research has been carried out on the synthesis and characterization of a variety of conductive rc-conjugated materials, with common examples being poly(p-phenylenevinylene) (PPV), polyaniline and polythiophene.9 Oligo- and polythiophenes were selected for use in this study due to the stability and ease of chemical modification of thiophene chains and their spectroscopic properties such as the absorption and emission of light in the visible region. The conductivity, electrochromism, and electroluminescence displayed for oxidized oligo- and polythiophenes also make these materials very interesting for investigation.10"15 Merging transition metal groups with rc-conjugated chains produces hybrid metal-organic materials with new structural, chemical, and electronic 16-22 properties. 1 1.2 Organic tr-Conjugated Materials 1.2.1 Structures of ^ -Conjugated Materials Organic rc-conjugated materials generally contain alternating double bonds, or linked aromatic rings that form extended, delocalized rc-systems along oligomer or polymer chains. Selected examples of rc-conjugated organic polymers are shown in Figure 1-1, where polythiophene and polypyrrole are examples of polymerized aromatic rings, PPV incorporates both an aromatic ring and a vinyl group within the monomer unit, and polyaniline forms a conjugated pathway along the phenyl ring and the lone pair on the nitrogen atom of the amine linker group.9 H polythiophene polypyrrole poly(p-phenylenevinylene) polyaniline Figure 1-1 Selected examples of rc-conjugated organic polymers. Oligo- and polythiophenes are formed of chains of thiophene rings linked at the a positions (Chart 1-1). For chains of n > 7, these materials are insoluble, and a general strategy to increase the solubility in organic solvents is the substitution of organic groups at P positions along the thienyl backbone. Significant research has also been devoted towards the preparation of regioregular or symmetric polythiophenes, since ordered structures exhibit superior properties such as enhanced conjugation, " improved alignment in the solid state, ' and higher conductivity.2 6'2 7 Polythiophenes have been functionalized with solubilizing alkyl substituents at P positions to yield polyalkylthiophenes (PATs). A regioregular structure can be prepared by linking alkythiophenes head-to-tail (HT) to form HT-PATs, such as polyhexylthiophene (HT-PHT) depicted in Figure 1-2. Polyalkylenedioxythiophenes have also been prepared from alkylenedioxythiophene monomers to yield a symmetric polymer that is soluble in organic 29 30 solvents, and these materials have been investigated for their interesting electronic properties. ' An example of a polyalkylenedioxythiophene, poly(3,4-ethylenedioxythiophene) (PEDOT), is also shown in Figure 1-2. 2 n HT-PHT PEDOT Figure 1-2 Selected examples of soluble polythiophenes. Chart 1-1 1.2.2 Synthesis of Oligo- and Polythiophenes The preparation of oligo- and polythiophenes can be carried out by a variety of synthetic pathways. Since many synthetic pathways are available, and more routes are continually developed, only some of the most commonly used techniques are described, and especially those used for this study. The Handbook of Oligo- and Polythiophenes provides a comprehensive synopsis of thienyl coupling techniques.31 Oligo- and polythiophene coupling reactions can be generally grouped into the categories of metal-catalyzed homo- and cross-coupling reactions, and oxidative polymerization. Metal-catalyzed homocoupling is the coupling of aryl halides, and is a straightforward method involving the coupling of a single aryl halide starting material. The method is convenient for the coupling of symmetric monomers, or if a regioregular material is not required. An example is the Ni(II)-catalyzed coupling of 2-bromothiophene in the presence of zinc to give bithiophene, as depicted in Figure l-3a. 3 2 Kumada coupling is a metal-catalyzed cross-coupling reaction of aryl halides and Grignard reagents,33'34 and is an efficient technique that can be used to couple thiophene rings at a positions in the presence of reactive P-substituents. ' Kumada coupling is frequently used for oligothiophene synthesis, commonly employing [PdCbCdppf)] (dppf = l,l'-bis(diphenyl-phosphino)ferrocene) or [NiC^Cdppp)] (dppp = l,3-bis(diphenylphosphine)propane) as catalysts. 3 The example given in Figure l-3b depicts selective coupling of tribromothiophene at the a positions, leaving the (3-bromo position unsubstituted. Suzuki coupling is the metal-catalyzed cross-coupling of halide and boronic ester groups, and is also often used for oligothiophene coupling reactions. Suzuki coupling reactions are high yielding and display a much higher activity for the iodo group versus other halides, so that the method can be used to prepare regiospecific oligomers via cross-coupling reactions.37 ,38 Figure l-3c shows the coupling of an a,a-chloroiodoalkylthiophene with a thienyl boronic ester, where the iodo group is reacted, and the remaining chloro group can be converted to a boronic ester for further reaction and chain 37 38 extension. ' Ni(II)-HomocoupIing (a) 2 ^ - B r [ N * * P " b M ^y^] Zn, Et 4NI Kumada Coupling , S V S x [PdCl2(dppf)] Br Br Suzuki Coupling M , ^ S ^ _ , - , / S X P ~ \ [ P D < P P H 3 ) 4 ] „ , <c) '^_^CI + Q^\J —~ i x ^ c a R R R R McCullough Method (d) B r ^ C S y - M g B r ^ ( d p p p ) ] - ^ V ^ R R Figure 1-3 Metal-catalyzed coupling reactions used to prepare oligo- and polythiophenes. 4 The McCullough method for preparing regioregular polyalkylthiophenes makes use of Grignard-halide coupling similarly to that of Kumada coupling with a Ni(II) catalyst, using a single 2-bromo-5-bromomagnesio-3-alkylthiophene monomer. Cross-coupling of the monomer results in 98-100% HT-HT coupling, yielding highly regioregular polymers such as HT-PHT. 2 8 ' 3 9 ' 4 0 Figure l-3d displays the general procedure for preparing HT-PAT polymers. An alternative synthetic thienyl coupling method frequently used is Stille coupling, which is a palladium-catalyzed cross-coupling reaction using organotin reagents,41'42 although this method has the disadvantage of handling toxic tin reagents. Oxidative polymerization occurs by loss of an electron from thiophene or an oligothiophene chain that creates a localized radical at an a position, and polymerization proceeds via a radical mechanism. Shown in Figure 1-4 is the oxidation of thiophene to produce the radical cation, followed by coupling of two thiophene radicals. Subsequent loss of two protons yields bithiophene, and chain extension then continues by repetition of the radical mechanism.10 Coupling preferentially occurs at the a positions due to the greater relative reactivity of these positions, though P-couplings can also occur via this technique in a lower ratio than a-coupling. Oxidative polymerization can occur by a chemical redox reaction with the 5 addition of an oxidant such as FeC^, 4 3 or by electrochemical polymerization at an electrode surface 4 4 Thiophenes substituted with electron donating groups or extended oligothiophene chains are more easily oxidized than thiophene, and are therefore better candidates for polymerization via this approach. The electron-rich EDOT group is an example of a substituted thiophene that is polymerized by oxidative polymerization to give PEDOT. 4 4 1.2.3 Properties and Applications of 71-Conjugated Materials A LUMO Energy HOMO Conduction Band Band Gap Valence Band n = l n = 2 n = 3 n = oo Figure 1-5 Energy band diagram of polyacetylene, as derived from overlapping n orbitals. Organic ^-conjugated materials display interesting properties due to the delocalized ^-electrons of these systems. Overlap of the n-orbitals of the double bonds, adjacent aromatic rings, or lone pairs of extended chains results in multiple bonding and antibonding orbitals of similar energy that form a filled valence band and an empty conduction band, separated by a band gap. This band structure can be likened to that of a semiconductor, and n-conjugated materials exhibit similar properties to semiconductors. The molecular orbital diagram displayed 6 in Figure 1-5 shows the overlap of n and n* orbitals of alternating double bonds, resulting in the formation of multiple n-bonding and antibonding orbitals. The valence and conduction bands that are formed for polyacetylene (n = oo) are separated by a band gap that is lower in energy than the difference between the initial HOMO and L U M O levels. The TC orbitals of polythiophene overlap in a similar fashion to yield a band structure. Structure-property relationships are established between the composition and conformation of a material and specific chemical or electronic properties. The changes in electronic properties that arise due to chain extension and the overlap of n orbitals include red shifts of the absorption and emission wavelengths, and lower oxidation potentials. Red shifted absorption and emission wavelengths are due to a decreasing energy gap between the HOMO and the L U M O levels, while a higher HOMO energy level results in easier oxidation of long thiophene chains. Tt-Conjugated chains are also more easily reduced compared with the monomer species, but as the energies of the L U M O levels are generally still high in energy, reduction of the organic chain is more difficult to carry out than oxidation.1 0 Figure 1-6 displays trends observed for the Xmax of the T C - > T C * transition, and oxidation potentials for oligothiophenes of n = 2-6. Number of Thienyl Rings Figure 1-6 Changes in the irreversible oxidation potential, E p (O), and the n—>n* absorption 31 wavelength, Xmax ( A ) with extension of thiophene chain length. Data from Ref. 7 Substantial interest has been generated for the study of n-conjugated materials that is largely stimulated by their ability to conduct current in the doped form.7 The conductivity of regioregular polyalkylthiophenes is increased significantly to maximum conductivities of >1000 Scm" with the introduction of holes (p-doping); ' ' to the same order of magnitude as that of Cu and Ag. The removal of an electron from the rc-orbitals of polythiophene results in the formation of a polaron, a localized distortion of the backbone that changes the alternation of double bonds over five thiophene units, separating a radical-cation pair (Figure l-7a). Polarons are characterized by the insertion of localized levels within the band gap for which the HOMO level is half-filled, as depicted to the right of Figure l-7a. Two polarons can combine to form a lower energy, doubly-charged bipolaron. The structure of a bipolaron is shown in Figure 1 -7b, and the two empty, localized levels that are inserted between the valence and conduction bands are shown to the right of the structure. A bipolaron is thought to be the more stable state that acts as the primary charge carrier for polythiophene conduction.4 5'4 6 Figure 1-7 Localized distortions on polythiophene chains and the introduction of energy levels into the band gap due to the formation of (a) a polaron, and (b) a bipolaron. Due to the introduction of energy levels into the band gap via the creation of polarons and bipolarons with oxidation, the spectroscopic properties of polythiophenes are altered. The colour of polythiophenes is modified with doping and the application of an electrical potential, and this phenomenon is termed electrochromism. While polythiophene displays an absorption band between 400-500 nm to yield a dark-red colour, red shifting of the absorption band with 8 electrochemical oxidation results in a green or black colour of the doped material. Electroluminescence occurs when the application of an electric field results in electrons and holes moving in opposite directions that can combine to form an exciton (electron-hole pair), and emit radiation. This property can be exploited to make organic light emitting diode (OLED) devices where the conjugated material is the emissive layer. Polythiophene usually emits orange to red light, 4 7 but by substitution of the polythiophene chains and modification of the dihedral angles between rings, electroluminescent emission spanning the visible and near-IR regions has been achieved.13 1.3 Metal-Organic Hybrid Materials 1.3.1 Metallation of rc-Conjugated Organic Materials Transition metal groups incorporated into rc-conjugated organic materials can modify the structural, chemical, and electronic properties of the organic component, to generate new classes of metal-organic hybrid materials. Metal groups can be combined with rc-conjugated chains via different attachment modes and with varying proximity to the backbone, and the structural composition and extent of metal-chain interaction strongly affects the resulting properties. Figure 1-8 Classes of transition metal-organic hybrid materials. Figure adapted from Ref. 9 Metal-organic hybrid materials can be classified into three categories, as shown in Figure 48 ' 1-8. For Type I materials, the transition metal center is spaced from the conjugated backbone via a typically saturated tether group. The metal group of Type I materials generally exerts little influence on the electronic properties of the organic chain unless the tether is very short. Polymer l 4 9 is an example of a Type I metal-oligothiophene material that was electropolymerized from the terthiophene monomer, and 2 is a methyl-capped terthiophene complex used to model a polythiophene chain (Figure 1-9). Both 1 and 2 contain long tether groups separating the metal centers from the thienyl backbones, and metal-chain interactions are not observed for these configurations. Complex 3 5 0 is a metal-bis(salicylidene)terthiophene complex where the metal is held in much closer proximity to the conjugated chain. This monomer can be electropolymerized, and the colour of the resulting metallated polythiophene material is dependant on the transition metal center [Ni(II) or Cu(II)]. Binding interactions with cations, occurring near the metal center, also modify the properties of the conjugated chain. The transition metal groups of Type II materials are directly bound to the conjugated backbone without interrupting the conjugation pathway. Compared to Type I materials, the bound metals of Type II materials are more likely to interact electronically with the conjugated chain, dependent on the overlap of orbital energies. Type II metal-oligothiophene complexes include 4 1 9 , which contains a bis(bipyridine) Ru(II) [Ru(bpy)22+] group coordinated via an inserted bipyridine segment that modifies the chain properties (Figure 1-9). Several other Ru(II)-oligothiophene and Ru(II)-polythiophene hybrid materials have been reported with 20 22 51 52 similar backbone structures incorporating an inserted metal-coordinating group. ' ' ' There are fewer examples of metals that are bound directly to oligo- or polythiophenes, and those reported generally contain a tethered metal group that encourages interaction of a metal with the thienyl chain. For instance, when the metal group of Type I complex 2 reacts with A g + to abstract a chloride ligand, the metal then coordinates to a thienyl sulfur to form the Type II complex 5. 1 8 This research group has demonstrated direct binding of a metal via two modes: metal-sulfur coordination, and metal-carbon bonding. Using a p-diphenylphosphine linker to promote metal-chain reactions, the Pd(II) complex 6 5 3 , 5 4 was prepared, and could be electropolymerized to yield metal-crosslinked polythiophenes. 10 Type I 1 2 3 (M = Ni,Cu) Type II 5 6 Type I I I Figure 1-9 Metal-oligothiophene and metal-polythiophene hybrid materials, classed as Type I, Type II, and Type III. Type III materials can also display strong interactions between metal groups and a conjugated chain, since the transition metal is closely situated to the organic chains and positioned directly in the pathway of conjugation. Type III metal-oligothiophene materials include Ru(II) complexes 755 and 856 (Figure 1-9). Complex 7 consists of a hexathiophene chain that is inserted with conjugated acetylene and metal groups, and displays a low-energy thienyl absorption, and ligand to metal charge transfer (LMCT) absorption bands when oxidized. The 11 binuclear cyclometallated complex 8 is also an example of a Type III material; although the metal is not inserted between conjugated chains, the metal groups are attached via a positions and are in the direction of conjugation. Complexes of this type can also be used as models for extended Type III materials.17 Interaction of the thienyl chain with the metal groups of 8 is indicated by an increase in the oxidation potential, though metal-metal interactions via the conjugated chain were not observed. 1.3.2 Properties and Applications of Metal-Organic Hybrid Materials There are a variety of properties that can be altered by the attachment of metals to ^-conjugated materials, and some have already been mentioned in the previous section. In this section, a summary of possible properties modifications that can occur with metallation of conjugated chains is given, along with specific applications that can result. The properties of ^-conjugated materials are generally altered either chemically, structurally, or electronically. Several factors modify electronic properties, including inductive effects, conformation of the conjugated backbone, and the introduction of metal-based orbitals that interact with organic n-orbitals in the hybrid materials. The properties and applications of metal-oligothiophene and metal-polythiophene materials are discussed using the examples given from the previous section. Oligo- and polythiophenes are generally chemically stable; however, the introduction of metal groups can result in new reaction or binding sites that can be exploited for sensing applications. For example, polymers of the bis(salicylidene)terthiophene monomer 3 appear green or orange, depending on whether the complexed metal is Cu(II) or Ni(II), respectively. Additionally, the materials produce characteristic shifts of the thienyl-based redox potentials when the crown-ether ring of the complex selectively chelates B a 2 + and M g 2 + cations, and the resulting device is a chemosensitive electroactive polymer.50 Inductive effects occur when there is shifting of the charge density localized at the thienyl chain towards a comparatively electron poor metal center, and can be most often observed as an increase in the oxidation potential of the conjugated chain. Proximity of the metal to the conjugated chain mediates the strength of the inductive effect, and hence the effects on the electronic properties. Comparison between 2 and 5 demonstrates the importance of metal proximity for the inductive effect.18 The Type I complex 2 displays A, m a x (353 nm) and E\a (510 mV vs. Fc/Fc +) values that are near those of the ligand, while the Type II complex 5 shows a similar X m a x [352 nm, 420 nm (sh)] but an E\a value that is shifted to a higher potential by 12 225 mV. When significant electron density is removed from a rc-conjugated chain, derealization is less effective, and the H O M O / L U M O gap is widened for the conjugated system. This effect is observed as increased absorption and emission energies in the electronic spectra. Conversely, electron donation into the rc-system can occur due to metallation, to increase the conjugation and reduce the energy gap (Figure 1-10). Inductive Effect i or Chain Twisting Conduction Band Band Gap Valence Band Electron Donation • or Chain Planarization Figure 1-10 Modification of the band gap energy with metallation of rc-conjugated materials, due to inductive effects or chain twisting to increase the band gap (left), and electron donation or chain planarization (right). Planar Figure 1-11 Examples of planar and twisted conformations of an oligothiophene. The effective overlap of rc-orbitals largely determines the extent of conjugation, therefore the conformation of the conjugated backbone affects the electronic properties. In the case of a chain composed of aromatic rings, whether the rings are in planar or twisted conformations (Figure 1-11), or some conformation in between, will mediate the conjugation. The H O M O / L U M O gap is decreased by coplanar rings, while chain twisting widens the gap (Figure 1-10). Since Type I and Type II metal-thienyl materials may incorporate metal groups in a 13 bidentate mode along the backbone, changes in the conformation of the chain compared to the free ligand or the corresponding oligo- or polythiophene can occur. Complex 5 is a Type II material that deviates from a terthiophene planar conformation in the solid state and displays a dihedral angle between substituted rings that is 21° from planarity.18 The attachment of metal groups cannot only modify the properties of a rc-conjugated chain, but also introduce new electronic properties into hybrid metal-organic materials. An interesting characteristic of metals is the occurrence of charge transfer processes. Charge transfers with the ligand set and/or with a thiophene chain (Figure 1-12) could occur and would be observed in the electronic spectra. Type II complex 4 displays several changes in properties 2+ due to Ru(bpy)2 complexation, including a red shift of the TC—»TC* absorption band of the conjugated chain due to planarization of the bipyridyl segment of the chain upon metallation.19 Donation of electron density from the Ru(II) group could also contribute toward the red shift. Metal-to-ligand charge transfer (MLCT) transitions, both to the conjugated chain and to the bpy ligands, are introduced and result in a much altered absorption spectrum of the material. M L C T transitions from the metal group to the thienyl chain would result in the introduction of electrons into the conduction band (n-doping, Figure l-12a). Ligand-to-metal charge transfer (LMCT) transitions involving a transition from the conjugated chain to the metal group would alternatively create holes in the valence band (p-doping, Figure l-12b). Both doping mechanisms create a conductive pathway along a rc-conjugated chain with light absorption, and materials of this type constitute light-harvesting materials, having applications in organic-based solar cells. MLCT Metal H O M O Conduction Band A Band Gap Valence Band Metal L U M O LMCT (a) (b) Figure 1-12 The introduction of metal-based HOMO or L U M O levels into the band gap of a rc-conjugated chain, resulting in (a) a metal-to-ligand charge transfer (MLCT) transition, or (b) a ligand-to-metal charge transfer (LMCT) transition. 14 1.4 Focus of Present Study This study is focused on the interactions between transition metals and oligothiophenes of the Type I and Type II classes. These types of structures were targeted so that the conjugation of the oligothiophene chains is not interrupted; as well, metal-oligothiophene interactions could be studied without the complication of an inserted conjugated group. Diphenylphosphine linker groups, substituted at P positions on oligothiophene chains, are used to anchor metal groups. Phosphines can coordinate many metals via a lone pair, modify the electronic properties of metals, and participate in charge transfer transitions.57 These phosphine-oligothiophene ligands are referred to as P-phosphinothiophenes in this thesis. The coordination of a P-phosphinothiophene to a metal group creates a Type I metal-oligothiophene material, where the short phosphine linker allows the metal to lie in close proximity with the conjugated backbone (Figure 1-13a). Type II Figure 1-13 General strategies for making (a) Type I and (b) Type II metal-oligothiophene hybrid materials using P-phosphinothiophenes. Type II metal-oligothiophene hybrid materials are also prepared with P-phosphinothiophene ligands, and are formed when a metal group positioned near the conjugated backbone bonds directly with the adjacent thiophene ring. Metal-thiophene bonding can occur by this strategy either at the sulfur atom or at a P-carbon position (Figure 1-13b). I was especially interested in studying the changes in electronic properties that occur due to close 15 proximity of the attached metal groups, and orbital overlap with the n-conjugated backbone. Both types of metallation can also alter the thienyl backbone structure, with further ramifications on the electronic properties. Once metal groups are attached to the oligothiophene chain, they can also be involved in chemical reactions and the reactivity of the hybrid materials is investigated. These studies, focusing on direct metal-oligothiophene interactions, differ from similar work that has been carried out involving Type II metal-oligothiophene hybrid materials. Most studies have made use of other conjugated groups, inserted into an oligothiophene backbone to directly bond a metal to the conjugated chain, and few instances of direct metal-thienyl interactions along thienyl chains have been reported. The lack of examples in the literature is likely because the thienyl sulfur is a poor coordinating atom, metal-(3-thienyl bonds often require the presence of heat or strong base to form, and both modes usually require chelation for stability.58"60 P-Phosphinothiophenes are used in this study to promote metal interactions with the thienyl backbone via different modes of binding. Oligothiophene chains are used to model polythiophene chains since the metallated materials can be more easily purified and isolated, to prepare well-defined materials and better establish structure-property relationships. 1.5 Goals The goals of this thesis were: (1) to design and synthesize new, metal-oligothiophene and metal-polythiophene hybrid materials, (2) to investigate the properties of these new materials, and (3) to interpret the results from a perspective of materials applications. The initial goal was to design and synthesize hybrid materials where the attached metal group is in close proximity to the thienyl backbone, and can interact structurally, chemically, or electronically with the conjugated chains. The metals were to be attached at either pendent positions or directly to the backbone, to maintain an uninterrupted ^-conjugated thienyl chain. The second goal is comprised of the following objectives: to structurally characterize solid state materials, determine chemical reactivity of the hybrid materials due to the incorporation of metal groups, and to characterize the electronic properties, including the HOMO and L U M O levels, the electronic excitation and emission energies, and oxidation potentials. The properties of the metal-organic hybrid materials in the oxidized state were also to be characterized. The final goal was to identify the structure/property relationships established by this work, and evaluate the metal-organic hybrid materials for molecular electronics or materials applications. 16 References Day, P. "Towards defining materials chemistry," IUPAC, Inorganic Chemistry Division (II) and Polymer Division (IV), 2005 (www.iupac.org/projects/2005/2005-001-l-200.html). Journal of Materials Chemistry Homepage, RSC Publishing, 2005 (www.rsc.org/Publishing/Journals/jm/About.asp). Forrest, S. R. Nature 2004, 428, 911-918. Gurunathan, K. ; Murugan, A . V. ; Marimuthu, R.; Mulik, U . P.; Amalnerkar, D. P. Mater. Chem. Phys. 1999, 61, 173-191. Joachim, C ; Gimzewski, J. K. ; Aviram, A. Nature 2000, 408, 541-548. Shirakawa, H. Angew. Chem. Int. Ed. 2001, 40, 2575-2580. MacDiarmid, A. G. Angew. Chem. Int. Ed. 2001, 40, 2581-2590. Heeger, A. J. Angew. Chem. Int. Ed. 2001, 40. Skotheim, T. A. ; Elsenbaumer, R. L.; Reynolds, J. R. Handbook of Conducting Polymers; 2nd ed.; Marcel Dekker: New York, 1998. Roncali, J. Chem. Rev. 1992, 92, 711-738. Kraft, A. ; Grimsdale, A . C ; Holmes, A. B. Angew. Chem. Int. Ed. 1998, 37, 402-428. Tourillon, G.; Gamier, F. J. Phys. Chem. 1983, 87, 2289-2292. Andersson, M . R.; Berggren, M . ; Inganas, O.; Gustafsson, G.; Gustafsson-Carlberg, J. C ; Seise, D.; Hjertberg, T.; Wennerstrom, O. Macromolecules 1995, 28, 7525-7529. Guay, J.; Kasai, P.; Diaz, A. ; Wu, R.; Tour, J. M . ; Dao, L. H. Chem. Mater. 1992, 4, 1097-1105. Burroughes, J. H. ; Bradley, D. D. C ; Brown, A. R.; Marks, R. N . ; Mackay, K.; Friend, R. H. ; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539-541. Hirao, T. Coord. Chem. Rev. 2002, 226, 81-91. Stott, T. L.; Wolf, M . O. Coord. Chem. Rev. 2003, 246, 89-101. Weinberger, D. A. ; Higgins, T. B.; Mirkin, C. A . ; Stern, C. L . ; Liable-Sands, L. M . ; Rheingold, A. L. J. Am. Chem. Soc. 2001,123, 2503-2516. Zhu, S. S.; Kingsborough, R. P.; Swager, T. M . J. Mater. Chem. 1999, 9, 2123-2131. Wolf, M . O.; Wrighton, M . S. Chem. Mater. 1994, 6, 1526-1533. Walters, K. A. ; Ley, K. D.; Cavalaheiro, C. S. P.; Miller, S. C ; Gosztola, D.; Wasielewski, M . R.; Bussandri, A. P.; van Willigen, H. ; Schanze, K. S. J. Am. Chem. Soc. 2001,123, 8329-8342. Walters, K. A. ; Touillet, L. ; Guillerez, S.; Schanze, K. S. Inorg. Chem. 2000, 39, 5496-5509. Zagorska, M . ; Kulszewicz-Bajer, I.; Pron, A. ; Firlej, L.; Bernier, P.; Galtier, M . Synth. Met. 1991, 45, 385-3S»3. 17 McCullough, R. D.; Tristram-Nagle, S.; Williams, S. P.; Lowe, R. D.; Jayaraman, M . J. Am. Chem. Soc. 1993, 115, 4910-4911. Mena-Osteritz, E.; Meyer, A. ; Langeveld-Voss, B. M . W.; Janssen, R. A. J.; Meijer, E. W.; Bauerle, P. Angew. Chem. Int. Ed. 2000, 39, 2680-2684. Chen, T.-A.; Rieke, R. D. Synth. Met. 1993, 60, 175-177. Chen, T.-A.; Wu, X . ; Rieke, R. D. J. Am. Chem. Soc. 1995,117, 233-244. Loewe, R. S.; Ewbank, P. C ; Liu, J.; Zhai, L. ; McCullough, R. D. Macromolecules 2001, 34, 4324-4333. Akoudad, S.; Roncali, J. Electrochem. Commun. 2000, 2, 72-76. Caras-Quintero, D.; Baeuerle, P. Chem. Commun. 2004, 926-927. Fichou, D. Handbook of Oligo- and Polythiophenes; Wiley-VCH: Weinheim, 1999. Iyoda, M . ; Otsuka, FL; Sato, K.; Nisato, N . ; Oda, M . Bull. Chem. Soc. Jap. 1990, 63, 80-87. Kumada, M . Pure Appl. Chem. 1980, 52, 669-679. Tamao, K. ; Kodama, S.; Nakajima, I.; Kumada, M ; ; Minato, A. ; Suzuki, K. Tetrahedron 1982, 38, 3347-3354. Nakanishi, H. ; Sumi, N . ; Aso, Y. ; Otsubo, T.J. Org. Chem. 1998, 63, 8632-8633. Carpita, A. ; Rossi, R. Gazz. Chim. Ital. 1985. Kirschbaum, T.; Azumi, R.; Mena-Osteritz, E.; Bauerle, P. New J. Chem. 1999, 23, 241-250. Bidan, G.; De Nicola, A. ; Enee, V. ; Guillerez, S. Chem. Mater. 1998,10, 1052-1058. McCullough, R. D.; Williams, S. P.; Tristram-Nagle, S.; Jayaraman, M . ; Ewbank, P. C ; Miller, L. Synth. Met. 1995, 69, 279-282. Loewe, R. S.; Khersonsky, S. M . ; McCullough, R. D. Adv. Mater. 1999,11, 250-253. Pereira, R.; Iglesias, B.; de Lera, A. R. Tetrahedron 2001, 57, 7871-7881. Stille, J. K. Angew. Chem. Int. Ed. 1986, 98, 504-519. Levesque, I.; Leclerc, M . Chem. Commun. 1995, 2293-2294. Roncali, J. J. Mater. Chem. 1999, 9, 1875-1893. Heeger, A . J.; Kivelson, S.; Schrieffer, J. R.; Su, W. P. Rev. Mod. Phys. 1988, 60, 781-850. Bertho, D.; Laghdir, A. ; Jouanin, C. Phys. Rev. B: Condens. Matter 1988, 38, 12531-12539. Yamamoto, T.; Morita, A. ; Miyazaki, Y . ; Maruyama, T.; Wakayama, FL; Zhou, Z.; Nakamura, Y . ; Kanbara, T. Macromolecules 1992, 25, 1214-1223. Wolf, M . O. Adv. Mater. 2001, 13, 545-553. Higgins, T. B.; Mirkin, C. A . Chem. Mater. 1998,10, 1589-1595. Reddinger, J. L. ; Reynolds, J. R. Chem. Mater. 1998, 10, 3-5. 18 Zhu, S. S.; Carroll, P. J.; Swager, T. M . J. Am. Chem. Soc. 1996, 118, 8713-8714. Trouillet, L. ; De Nicola, A. ; Guillerez, S. Chem. Mater. 2000, 12, 1611-1621. Clot, O.; Wolf, M . O.; Patrick, B. O. J. Am. Chem. Soc. 2000,122, 10456-10457. Clot, O.; Wolf, M . O.; Patrick, B. O. J. Am. Chem. Soc. 2001,123, 9963-9973. Zhu, Y. ; Millet, D. B.; Wolf, M . O.; Rettig, S. J. Organometallics 1999, 75, 1930-1938. Pappenfus, T. M . ; Mann, K. R. Inorg. Chem. 2001, 40, 6301-6307. Vogler, A. ; Kunkely, H. Coord. Chem. Rev. 2002, 230, 243-251. Angelici, R. J. Organometallics 2001, 20, 1259-1275. Angelici, R. J. NATO ASISeries 1998, Ser. 3, 89-127. Constable, E. C. Polyhedron 1984, 3, 1037-1057. 19 CHAPTER 2 Synthesis and Characterization of (3-Phosphinothiophene Ligands and Derivatives 2.1 Introduction A strategy to bring metals into close proximity with thiophene chains is the use of coordinating "linker" groups. Examples of linker groups that have been incorporated into oligo-or polythiophenes are imines, pyridines, and thiols. Phosphines are useful in coordinating a variety of transition metals, and my group and others have made use of phosphines to coordinate gold,4 palladium,5 , 6 and ruthenium7'8 centers to ^-conjugated thienyl backbones. Ligands synthesized for this study contain diphenylphosphine linker groups attached to oligothiophenes at the P, or 3- position of a thiophene ring (Chart 2-1), to form P-phosphinothiophene ligands. The attachment of a linker group at a P position rather than at the a, or 2- position allows coordination of a metal without alteration of the backbone composition. Phosphinothiophene ligands metallated at the P positions could be used as molecular wires with the end positions available as points of "electrical contact". Chart 2-1 P h 2 P ^ p, 3-S' a , 2 - ^ ^ P-Phosphinothiophene ligands and derivatives prepared for this study contain 2-7 thiophene units, based on bithiophene (T2), terthiophene (T3), quaterthiophene (T4), pentathiophene (T 5) or heptathiophene (T7) chains (Chart 2-2). P 2 T 2 (9) and P 2 hex 2 T 4 (11) (Chart 2-3) were first synthesized by Dr. O. Clot in this research group.9 These P-bis(phosphino)thiophene ligands contain two adjacent diphenylphosphine groups that permit bidentate coordination of one metal center, or monodentate coordination of two metal centers (Figure 2-1). P-Hexyl substituents are attached to the outer thiophene rings of 11 and the 20 brominated precursor 10 to promote solubility in organic solvents. The (3-phosphine oxide (PO)2T2 (12) was prepared by the oxidation of 9, and conformational information is extracted from the crystal structure. Chart 2-2 O (PO) 2 T 2 (12) P 2 hex 2 T 4 ( l l ) 21 P h 7 P - M L PPh, • p - f t -Ph 2P x pph2 M L n 2ML„ • p - p -Ph2P PPh 2 i I L n M M L n Figure 2-1 Bidentate and monodentate modes of bonding that may arise with metallation of bis(phosphino)thiophene chains. Monophosphinated ligands, such as previously reported (3-(phosphino)terthiophene ligands 13-1510 and the a-(phosphino)terthiophene ligand 16, 1 1 coordinate one metal center (Chart 2-4). While the metal group coordinated by an a-phosphinothiophene is directed away from the thiophene chain due to the structural design of the ligand, P positioning of a phosphine allows an anchored metal group to bind to the adjacent thiophene ring via a thienyl sulfur or P position, provided a vacant site is available at the metal (Scheme 2-1). These bidentate bonding modes employ direct interaction of a metal to a thiophene chain, and have been demonstrated in my group with different metals.5'6'12 Due to the close interaction of the metals with the thiophene chains, the steric and electronic properties of thiophene chains are strongly affected by direct bonding modes. Longer p-phosphinothiophene ligands were prepared for this study by first synthesizing the p-substituted bromodihexyl- and bromodidodecyl-pentathiophenes 17 and 18, and subsequently the iodo analogs 19 and 20. From the P-substituted iodo derivatives, the new P-(phosphino)pentathiophene ligands 21 and 22 were obtained (Chart 2-5). Chart 2-4 aPT 3 (16) 22 Scheme 2-1 Chart 2-5 Phex 2T 5 (21) PDo 2 T 5 (22) Electron-rich hexa- and pentathiophene derivatives 23-24 (Chart 2-6) were prepared and characterized. A tridodecylhexathiophene (23) was obtained as a side product via a non-selective coupling reaction. Didodecylpentathiophene (24) was prepared for comparison of its physical properties to those of the (3-(phosphino)pentathiophene ligand (22). 23 Chart 2-6 The effect that P-substitution and metallation imparts on the properties of polythiophenes is of interest due to the processibility of bulk organic polymers, and for modeling molecular wires. Towards this goal, reactions were carried out as models for two possible routes of polymerization: the polymerization of metal complexes, or the polymerization of functionalized oligomers. Precursor pentathiophenes 25 and 26 that are functionalized at both a positions and at one central p position were prepared (Chart 2-7). By exploiting the higher reactivity of the iodo substituents, the a,a-dibrominated ligand 27 was prepared from 25. Likewise, the P-brominated heptathiophene 28 was prepared from 26. Chart 2-7 24 2.2 Experimental 2.2.1 General Experimental A l l reactions were performed using standard Schlenk techniques with dry solvents under nitrogen. Ligands 9 9 and 14-156 were obtained from Dr. O. Clot, 16 1 1 was obtained from Dr. T. Stott, and ligand 11 was previously synthesized by Dr. O. Clot by another procedure.10 3,5,3',5'-Tetrabromo-2,2'-bithiophene,13 3'-bromoterthiophene,12 2-bromo-3-hexylthiophene,14 2-bromo-3-dodecylthiophene,14 and 2-thienyl-l,3,2-dioxaborinane15 and 136 were prepared by published procedures. A l l other reagents were purchased from Aldrich or Strem Chemicals. A^^Z-dimethylethylenediamine, bis-2-methoxyethylether, and xylenes were distilled before use, and all other reagents were used as received. ' H and 3 1 P{'H} N M R experiments were performed on either a Bruker AC-200E,. Bruker AV-300 or Bruker AV-400 spectrometer, and spectra were 1 31 referenced to residual solvent ( H) or external 85% H3PO4 ( P). Electronic absorption spectra were obtained on an HP8452A diode-array spectrophotometer or on a Cary 5000 in HPLC grade CH2CI2. Microanalyses (C, H, N) were performed at U B C by M . Lakha. Electrochemical measurements were conducted on a Pine AFCBP1 bipotentiostat using a platinum disc working electrode, platinum coil wire counter electrode and a silver wire reference electrode. An internal reference, either decamethylferrocene (-0.12 V vs. SCE) or ferrocene (0.41 V vs. SCE), was added to calibrate the measured potentials with respect to saturated calomel electrode (SCE). 1 6 The supporting electrolyte was [(n-Bu)4N]PF6 that was purified by triple recrystallization from ethanol and dried at 90°C under vacuum for three days. 2.2.2 Synthesis 3,3'"-Dihexyl-3 '3 "-dibromo-2,5':2 ',2 ":5 ",2 '"-quaterthiophene (10) A solution of 2-bromo-3-hexylthiophene (7.34 g, 30 mmol) in THF (10 mL) was added dropwise to a mixture of magnesium (1.41 g, 58 mmol) and trace iodine in THF (30 mL) at reflux and stirred for 2 h. The resulting solution was transferred dropwise at 25°C via cannula to a mixture of 3,5,3',5'-tetrabromo-2,2'-bithiophene (5.72 g, 12 mmol) and [PdCl2(dppf)]-CH2Cl2 (200 mg, 0.2 mmol) in Et^O (30 mL) and toluene (20 mL), and heated at reflux for 3 h. The reaction was then quenched with sat. aq N H 4 C I (50 mL), and filtered through Celite to remove insoluble material. CH2CI2 (50 mL) was added, the organic phase was separated and the aqueous phase was extracted with CH2CI2. The combined organic phases were washed with sat. aq 25 NaHCCh solution and H2O, dried with anhydrous MgSCU, and the solvent was removed to give a thick orange oil. The crude product was purified by column chromatography on silica gel with hexanes as eluent. The first three bands contained small amounts of unreacted 3,5,3',5'-tetrabromo-2,2'-bithiophene, 3-hexylthiophene, and monosubstituted terthiophene side product, respectively. The fourth band contained the desired product 10, and removal of solvent left a bright-yellow, waxy solid. Penta- and hexathiophene derivatives were also isolated in subsequent bands. Yield: 4.02 g (52%). ! H N M R (200.1 MHz, CDCI3): 5 7.23 (d, Jm = 5.2 Hz), 2H), 7.09 (s, 2H), 6.96 (d, JHH = 5.2 Hz, 2H), 2.79 (t, JHH = 7.7 Hz, 4H), 1.64 (m, 4H), 1.34 (m, 12H), 0.90 (t, 6.3 Hz, 6H). Anal. C28H32Br2S4 requires C, 51.22; H, 4.91%. found: C, 51.34; H, 4.72%. 3,3"'-Dihexyl-3 '3 "-bis(diphenylphosphino)-2,5':2 '2 ":5 "2 '"-quaterthiophene (P2hex2T4) (11) To a suspension of 10 (1.0 g, 1.52 mmol) in E t 2 0 (30 mL) at -78°C was added a solution of «-BuLi (1.6 M , 2.10 mL, 3.35 mmol) in dry THF. The mixture was slowly warmed until the suspended solids dissolved at -30°C to give an orange solution, and PPh 2 Cl (1.34 g, 6.09 mmol) was subsequently added. After 5 min at -30°C, a yellow precipitate formed and the mixture was slowly allowed to warm to room temperature, then stirred for 1 h. The reaction was quenched with H 2 0 , dried with anhydrous MgSCM and the solvent was removed to leave a dark-yellow solid/liquid mixture. The crude product was purified by chromatography on silica gel using hexanes-CH2Cl2 (4:1) as eluent. The first two bands contained starting materials, and the third band contained the monosubstituted side product. Removal of solvent from the fourth band yielded 11 as a dark-yellow liquid. Yield: 0.91 g (61%). ' H N M R (200.1 MHz, CDC13): 5 7.31 (m, 20H), 7.10 (d, JHH = 5.1 Hz, 2H), 6.85 (d, J H H = 5.1 Hz, 2H), 6.65 (s, 2H), 2.55 (m, 4H), 1.56-1.07 (m, 16H), 0.88 (m, 6H). 3 1 P{'H} N M R (81.0 MHz, CDC1 3): 8 -25.1 (s). Elemental analysis results were obtained by Dr. O. Clot and were reported.9 3.3 -Bis(diphenylphosphoryl)-2,2-bithiophene (PO)2T2 (12) To a solution of 9 (1.0 g) in CHC1 3 (50 mL) and acetone (50 mL) was added 30% H 2 0 2 (0.4 mL). A white solid precipitated immediately and the mixture was stirred for 1 h. The solid was collected by filtration, redissolved in CHCI3, filtered and precipitated with acetone to give a white solid. Yield: 0.54 g (51%). ' H N M R (200.1 MHz, CDCI3): 5 7.71 (ddd, J= 1.6, 8.2, 14.0 Hz, 8H), 7.53-7.36 (m, 12H), 7.18 (dd, JHH = 5.4 Hz, J H p = 2.2 Hz, 2H), 6.65 (dd, JHH = 5.4 26 Hz, J H p = 4.4 Hz, 2H). 3 1 P{'H} N M R (81.0 MHz, C D C I 3 ) : 5 20.5 (s). Anal. C32H24O2P2S2 requires C, 67.83; H, 4.27. found: C, 67.59 H, 4.32%. 3 "-Bromo-3,3""-dihexyl-2,2 ':5 '2 ":5 "2 "':5 "',2 ""-pentathiophene (17) Magnesium (0.20 g, 7.5 mmol) and a trace amount of iodine were brought to reflux in THF (15 mL), and a solution of 2-bromo-3-hexylthiophene (0.92 g, 3.7 mmol) in THF (2 mL) was added dropwise by syringe. The mixture was heated at reflux for 2 h and allowed to cool. The Grignard solution was then added dropwise by cannula to a condenser-fitted flask containing a solution of 5,3',5"-tribromo-2,2':5',2"-terthiophene (0.91 g, 1.9 mmol) and [PdCl2(dppf)]-CH 2Cl 2 (40 mg, 0.049 mmol) in Et20/toluene (22 mL/17 mL). The orange solution was heated at reflux for 16 h, and then quenched with sat. aq NH 4C1 and stirred for 1 h. The crude product was extracted with CH2CI2, washed with sat. aq NaHC03 once, then three times with H 2 0 to give a bright orange solution that was removed of solvent. The crude product was run through a short silica gel plug with hexanes to remove 3-dodecylthiophene side product and catalyst, and then purified by column chromatography on silica gel using hexanes as eluent. The third band was collected and the solvent was removed to give 17 as a viscous orange liquid. Yield: 316mg(26%). ] H N M R (400.1 MHz, CDC13): 5 7.372 (d, J= 4.0 Hz, 1H), 7.182 (d, J = 5.2 Hz, 1H), 7.176 (d ,J=4.0Hz, 1H), 7.111 (d,J=4.0 Hz, 1H), 7.075 (s, 1H), 7.065 (d,J=4.0 Hz, 1H), 7.015 (d,J= 4.0 Hz, 1H), 6.935 (d, J= 5.2 Hz, 1H), 6.932 (d, J= 5.2 Hz, 1H), 2.774 (t, J=8.4 Hz, 4H), 1.649 (m, 4H), 1.380 (m, 4H), 1.311 (m, 8H), 0.881 (m, 6H). 3 "-Bromo-3,3 ""-didodecyl-2,2 ':5 '2 ":5 ",2 "':5 "',2 ""-pentathiophene (18) Magnesium (1.36 g, 56 mmol) and iodine (5 mg, 0.02 mmol) were brought to reflux in THF (60 mL), and 2-bromo-3-dodecylthiophene (9.29 g, 28 mmol) dissolved in THF (10 mL) was added dropwise by syringe. The green-brown mixture was heated at reflux for 2 h and allowed to cool. The Grignard solution was then added dropwise by cannula to a condenser-fitted flask containing a solution of 5,3',5"-tribromo-2,2':5',2"-terthiophene (6.79 g, 14 mmol) and [PdCl 2(dppf)]-CH 2Cl 2 (300 mg, 0.367 mmol) in Et20/toluene (80 mL/60 mL). The yellow-brown solution was stirred at reflux for 16 h, quenched with sat. aq NH4CI, and stirred for 1 h. The crude product was extracted with CH2CI2, washed with sat. aq NaHC03 once, then three times with H 2 Q to give a bright orange solution that was condensed to give an orange-red liquid. The crude product was run through a short silica gel plug with hexanes to remove 3-dodecylthiophene side product and catalyst, and then purified by column chromatography on 27 silica gel using hexanes as eluent. The initial yellow band was a mixture of tetrathiophene side products. The second orange band was collected and the solvent removed to give 18 as a soft, waxy, bright orange solid. Yield: 7.61 g (66%). ' H N M R (200.1 MHz, CDC1 3): 8 7.37 (d, J = 5.4 Hz, 1H), 7.18 (d, J = 7 . 7 H z , 2H), 7.11 (d, .7=5.7 Hz, 1H), 7.07 (s, 1H), 7.06 ( d , J = 5 . 7 H z , 1H), 7.01 (d, J= 5.4 Hz, 1H). 6.93 (d, J= 7.8 Hz, 2H), 2.77 (m, 4 H ) , 1.65 (m, 4 H ) , 1.25 (m, 36H), 0.87 (t, J= 6.8 Hz, 6 H ) . Anal. C 44H 59S 5Br requires C, 63.81; H , 7.18. found: C, 63.52; H, 7.48%. 3,3 ""-Dihexyl-3 "-iodo-2,2 ':5 ',2 ":5 ",2 '":5 '"2 ""-pentathiophene (19) To a solution of 17 (316 mg, 0.478 mmol), Nal (71.6 mg, 0.956 mmol), and Cul (4.6 mg, 0.024 mmol) in xylenes/bis-2-methoxyethylether (10 mL/4 mL), 7V,A^-dimethylethylenediamine (0.005 mL, 4.2 mg, 0.048 mmol) was added. A white precipitate formed and the resulting mixture was heated to 165°C for 5 h. After cooling, a bright yellow organic layer and a green aqueous layer were formed with the addition of CH2CI2 and H 2 O . The organic layer was separated, washed three times with H 2 0 , dried with anhydrous M g S 0 4 and filtered. The solvent was removed and the xylenes and bis-2-methoxyethylether were distilled off to leave an oily, crude product. Purification by column chromatography on silica gel using hexanes as eluent gave 19 as a soft, waxy, bright orange solid after removal of solvent. Yield: 0.325 g (96%). *H N M R (200.1 MHz, CDC1 3): 8 7.383 (d, J = 4.0 Hz, 1H), 7.184 (d, J = 5.2 Hz, 2H), 7.163 (s, 1H), 7.094 (dd, 7 = 5.2 Hz, J= 4.0 Hz, 2H), 7.012 (d, J= 4.0 Hz, 1H), 6.934 (d, J= 5.2 Hz, 2H), 2.799 (m, 4 H ) , 1.664 (m, 4 H ) , 1.338 (m, 12H), 0.895 (m, 6 H ) . 3,3""-Didodecyl-3 "-iodo-2,2 ' J ',2 ":5 "2 "':5 "',2 ""-pentathiophene (20) To a solution of 18 (7.61 g, 9.18 mmol) , Nal (2.75 g, 18.4 mmol), and Cul (87.5 mg, 0.459 mmol) in xylenes/bis-2-methoxyethylether (160 mL/40 mL), ^ iV-dimethylethylene-diamine (0.098 mL, 81 mg, 0.92 mmol) was added. A white precipitate formed and the resulting mixture was heated to 165°C for 16 h. After cooling, a dark yellow organic layer and a green aqueous layer were formed with the addition of CH2CI2 and H 2 O . The organic layer was separated, washed three times with H 2 O , dried with M g S 0 4 and filtered. The solvent was removed and the xylenes and bis-2-methoxyethylether were distilled away to leave an oily crude product. Purification by column chromatography on silica gel using hexanes as eluent gave 20 as a soft, waxy bright orange solid after removal of solvent. Yield: 7.58 g ( 9 4 % ) . ! H N M R (200.1 M H z , C D C l 3 ) : 8 7.38 (d, J = 4.0 Hz, 1H), 7.19-7.16 (m, 2H), 7.16 (s, 1H), 7.11 ( d , J = 3 . 6 H z , 28 1H), 7.08 (d, J= 4.0 Hz, 1H), 7.01 (d, J = 3.8 Hz, 1H), 6.93 (d, J = 5.2 Hz, 2H), 2.77 (m, 4H), 1.65 (m, 4H), 1.25 (m, 36H), 0.87 (t, J = 6.6 Hz, 6H). Anal. C 44H 59S 5I requires C, 60.38; H, 6.79. found: C, 60.78; H, 6.88%. 3,3""-Dihexyl-3 "-diphenylphosphino-2,2 ':5 '2 ":5 "2 "':5 '"2 ""-pentathiophene (Phex2T5) (21) To a solution of 19 (325 mg, 0.458 mmol) and [Pd(OAc)2] (1.0 mg, 4.6 umol) in C H 3 C N (25 mL), distilled NEt 3 (0.13 mL, 0.093 g, 0.92 mmol) and Ph 2 PH (0.080 mL, 85 mg, 0.46 mmol) were added. The orange suspension darkened to greenish-brown with heating at reflux for 72 h and a dark oil was formed on the flask bottom. Solvent and volatiles were removed under reduced pressure and the crude product was extracted with CH 2 C1 2 , washed with 1 M aq K O H , 2 M aq HC1, then three times with H 2 0 . Removal of solvent left a dark, blackish oil. Purification by column chromatography on silica gel using acetone-hexanes (5:95) as eluent resulted in elution of 21 as the third band. The isolated product was a bright orange viscous liquid after removal of solvent. Yield: 143 mg (41%). 3 1 P{'H} N M R (162.0 MHz, CDC13): 8-23.0 (s). 3.3 ""-Didodecyl-3 "-diphenylphosphino-2,2 ':5 '2 ":5 "2 "':5 "'2 ""-pentathiophene (PD02T5) (22) To a solution of 20 (4.33 g, 4.95 mmol) and [Pd(OAc)2] (5 mg, 0.022 mmol) in C H 3 C N (250 mL), distilled NEt 3 (1.4 mL, 1.0 g, 9.9 mmol) and Ph 2 PH (0.86 mL, 0.92 g, 4.95 mmol) were added. The orange suspension darkened to a greenish-brown with heating at reflux for 16 h and a black, gelatinous layer formed on the flask bottom. Solvent and volatiles were removed under reduced pressure and the crude product was extracted with CH 2 C1 2 , washed with 1 M aq K O H , 2 M aq HC1, then three times with H 2 0 . Removal of solvent left a brown oil. Purification by column chromatography on silica gel using acetone-hexanes (5:95) as eluent resulted in elution of starting material 20, followed by 22 as an orange band, and an orange band containing the oxidized ligand eluted much later. The isolated product was a bright orange viscous liquid after removal of solvent. Yield: 4.29 g (93%). 5 H N M R (200.1 MHz, CDC1 3): 5 7.37 (m, 10H, phenyl), 7.15 (d, J= 5.2 Hz, 2H), 7.03 (d, .7=3.6 Hz, 1H), 7.01 (d, 7=4.0 Hz, 1H), 6.99 (d,J = 4.4 Hz, 1H), 6.96 (d, J= 3.8 Hz, 1H), 6.91 (d, J = 5.0 Hz, 2H), 6.63 (s, 1H), 2.73 (m, 4H), 1.61 (m, 4H), 1.25 (m, 36H), 0.87 (m, 6H). 3 1 P{'H} N M R (81.0 MHz, CDC1 3): 5 -23.5 (s). Anal. C56H69S5P requires C, 72.05; H , 7.45. found: C, 71.65; H, 7.49%. 29 3,3 ,3 -lridodecyt-2,2 :5,2 :5 ,2 :5 ,2 :3 ,2 -hexathiophene (23) Magnesium (0.37 g, 15 mmol) and iodine (3 mg, 0.01 mmol) were brought to reflux in THF (20 mL), and 2-bromo-3-dodecylthiophene (2.53 g, 7.64 mmol) dissolved in THF (5 mL) was added dropwise by syringe. The dark brown mixture was heated at reflux for 2 h and allowed to cool. The Grignard solution was then added dropwise by cannula to a condenser-fitted flask containing a solution of 5,3',5"-tribromo-2,2':5',2"-terthiophene (1.10 g, 2.27 mmol) and [Ni(dppp)Cl2] (40 mg, 0.074 mmol) in Et20/toluene (30 mL/20 mL) and the mixture darkened from orange to red. After heating at reflux for 20 h, there was no change of the product mixture by TLC, and the reaction was quenched with sat. aq NH4CI, and stirred for 1 h. The crude product was extracted with CH 2 C1 2 , washed with sat. aq NaHC03 once, then three times with H 2 0 to give a dark red solution that was condensed to give a dark red liquid. Purification by column chromatography on silica gel using hexanes as eluent resulted in the elution of six bands as follows: 3-dodecylthiophene (0.60 g, 31%), 5,3',5"-tribromo-2,2':5',2"-terthiophene (0.26 g, 24%), mixed (3-substituted tetrathiophenes (0.22 g, 15%), mixed a,[3-substituted pentathiophenes (23 mg, 1.2%), 18 (21 mg, 1.1%), and 23 as orange-red solids. Yield: 0.51 g (23%). ' H N M R (300.1 MHz, CDC1 3): 5 7.32 (d, J= 4.8 Hz, 1H), 7.16 (d, J= 5.1 Hz, 1H), 7.14 (d ,J=3.9Hz, 1H), 7.1 l(d, J=5.1 Hz, 1H), 7.05 (s, 1H), 7.02 (d,J=3.9 Hz, 1H), 6.98-6.92 (m, 4 H), 6.86 (d, J= 5.1 Hz, 1H), 2.77 (t, J= 1.1 Hz, 2H), 2.63 (t, J= 1.1 Hz, 2H), 2.40 (t, J= 7.5 Hz, 2H), 1.64 (m, 2H), 1.54 (m, 2H), 1.45 (m, 2H), [1.26 (m), 1.24 (m), 1.20 (m), 1.16 (m); 54 H], 0.87 (m, 9H). Anal. C 6 oH 8 6 S 6 requires C, 72.09; H, 8.67. found: C, 72.10; H, 8.97%. 3,3""-Didodecyl-2,2 ':5 ',2 ":5 "2 "':5 "',2 ""-pentathiophene (D02T5, 24) A solution of 20 (300 mg, 0.343 mmol) in E t 2 0 (100 mL) was cooled to -20°C and tt-BuLi (0.26 mL, 1.6 M in hexanes, 0.41 mmol) was added. The yellow-orange solution immediately changed color to dark orange. H 2 0 (0.10 mL, 5.55 mmol) was injected into the solution and after 0.5 h and with slow warming, the solution turned yellow. The organic phase was washed three times with H 2 0 , dried with MgS04, filtered, and the solvent was removed to leave a yellow residue. Purification by column chromatography on silica gel using hexanes as eluent gave 24 as a dark yellow powder after removal of solvent. Yield: 0.257 mg (100%). ' H N M R (400.1 MHz, CDCI3): 5 7.17 (d, J= 5.2 Hz, 2H), 7.11 (d,J= 3.6 Hz, 2H), 7.08 (s, 2H), 7.01 (d, J = 3.6 Hz, 2H), 6.93 (d, J = 5.2 Hz, 2H), 2.77 (t, J = 7.8 Hz, 4H), 1.64 (m, 4H), 30 1.37-1.25 (m, 36 H), 0.87 (t, J= 6.6 Hz, 6H). Anal. C 44H 6oS 5 requires C, 70.53; H , 8.07. found: C, 70.93; H , 8.27%. 5,5""-Dibromo-3,3""-didodecyl-3 "-iodo-2,2 ':5 '2 ":5 "2 "':5 "'2 ""-pentathiophene (25) A solution of 20 (1.00 g, 1.14 mmol) and NBS (0.407 g, 2.28 mmol) were stirred in CHCI3/CH3COOH (100 mL/100 mL) at room temperature overnight. The solution was washed with H 2 O , 1 M aq NaOH, then three times with H 2 O , and the aqueous layer was extracted with CH2CI2. The organic phases were dried with MgS04 and the solvent was removed. The crude product was purified by column chromatography on silica gel using hexanes as eluent. The desired product eluted as the first band and was closely followed by a second band containing the monobrominated by-product. Removal of solvent gave 25 as bright orange solids. Yield: 0.982 mg (83%). *H N M R (400.1 MHz, CDCI3): 5 7.353 (d, J= 4.0 Hz, 1H), 7.150 (s, 1H), 7.081 (d, J= 4.0 Hz, 1H), 7.014 (d, J= 4.0 Hz, 1H), 6.946 (d, J= 4.0 Hz, 1H), 6.888 (s, 1H), 6.884 (s, 1H), 2.697 (t, J = 7.2 Hz, 4H), 1.598 (m, 4H), 1.333 (m, 4H), 1.236 (m, 32H), 0.855 (t, J = 6.8 Hz, 6H). Anal. 0 4 4 ^ 7 8 5 ^ requires C, 51.16; H , 5.56. found: C, 51.56; H , 5.59%. 3 "-Bromo-3,3 ""-didodecyl-5,5 ""-diiodo-2,2 ':5 ',2 ":5 ",2 '":5 '"2 ""-pentathiophene (26) A solution of 18 (353 mg, 0.426 mmol) and NIS (0.239 g, 1.06 mmol) in CHCI3/CH3COOH (100 mL/100 mL) was stirred at room temperature for 16 h. A precipitate formed that was extracted with CH2CI2 and the solution was washed with H 2 O , 1 M NaOH, then three times with H 2 O . After drying with MgS04 and filtering, the solvent was removed. A portion of the crude material was lost when crystallization attempts in CH 2 Cl 2 /E tOH solution resulted in oxidation of some product. Purification by column chromatography on silica gel using acetone-hexanes (5:95) as eluent was successful, and the first band eluted contained the product as bright orange solids. Yield: 310mg(67%). ' H N M R (400.1 MHz, CDC13): 5 7.346 (d, J= 4.0 Hz, 1H), 7.093 (d, J= 3.6 Hz, 1H), 7.065 (s, 3H), 7.000 (d, J= 4.0 Hz, 1H), 6.952 (d, J= 3.6 Hz, 1H), 2.708 (q, J = 8.0 Hz, 4H), 1.591 (m, 4H), 1.316 (m, 4H), 1.231 (m, 32 H), 0.852 (t, J= 6.8 Hz, 6H). Anal. C44H57S5I2Br requires C, 48.93; H , 5.32. found: C, 49.21; H , 5.35%. 5,5 ""-Dihromo-3,3 ""-didodecyl-3 "diphenylphosphino-2,2 ':5 ',2 ":5 "2 '":5 '"2 ""-pentathiophene (PBr 2 Do 2 T 5 ) (27) A suspension of 25 (800 mg, 0.826 mmol) and [Pd(OAc)2] (1.0 mg, 4.4 umol) was stirred in C H 3 C N (60 mL), and distilled NEt 3 (0.23 mL, 0.17 g, 1.7 mmol) and Ph 2 PH (0.14 mL, 0.15 g, 31 0.826 mmol) were then injected into the flask. The orange suspension dissolved under reflux conditions and stirred for 20 h. Solvent and volatiles were removed under reduced pressure and the crude product was extracted with CH2CI2, washed with 1 M aq K O H , 2 M aq HC1, then three times with H2O. Removal of solvent left an orange-brown oil. Purification by column chromatography on silica gel using acetone-hexanes (5:95) as eluent resulted in elution of starting material 25, followed by a side-product, then 27 as an orange band, and an orange band containing the phosphine oxide following. The isolated product was a thick, yellow-orange liquid after removal of solvent. Yield: 380 mg (42%). ' H N M R (400.1 MHz, CDCI3): 5 7.345 (m, 10H, phenyl), 7.105 (d, .7=3.6 Hz, 1H), 6.989 (d, .7=4.0 Hz, 1H), 6.923 (d, J=3 .6Hz, 1H), 6.889 (d, J = 3.6 Hz, 1H), 6.855 (s, 2H), 6.577 (s, 1H), 2.632 (m, 4H), 1.548 (m, 4H), 1.270 (m, 4H), 1.231 (s,32H), 0.851 (t, J= 6.6 Hz, 6H). 3 1 P{'H} N M R (162.0 MHz, CDCI3): 5 -23.1 (s). -Bromo-3,3 -aidodecyl-2,5 :2,2 :5 ,2 :5 ,2 :5 ,2 :5 ,2 -neptatniopnene (28) A solution of 26 (50 mg, 0.046 mmol), CsF (63 mg, 0.46 mmol) and [Pd(PPh3)4] (2.3 umol, 5.4 mg) was stirred in THF (20 mL) at reflux, and a solution of 2-thienyl-l,3,2-dioxaborinane (0.093 mmol, 15.6 mg) in THF (10 mL) was added to the mixture. After 2 h, the solution darkened from dark yellow to orange, and the solvent was removed. The product was purified by column chromatography on silica gel using acetone-hexanes (10:90) as eluent. The first band contained 26, the second band was a hexathiophene side-product, and the third band contained 28 as dark orange solids. Yield: 40.9 mg (89%). ' H N M R (400.1 MHz, CDC13): 8 7.381 (d,J= 3.9 Hz, 1H), 7.208 (d, J= 5.1 Hz, 2H), 7.158 (d, J= 3.6 Hz, 2H), 7.123 (d, J = 3.6 Hz, 1H), 7.083 (d, J= 3.9 Hz, 1H), 7.084 (s, 1H), 7.034 (d, J = 3.6 Hz, 1H), 7.009 (s, 2H), 7.007 (dd, J= 5.1 Hz, J= 3.6 Hz, 2H), 2.756 (m, 4H), 1.665 (m, 4H), 1.363 (m, 4H), 1.239 (m, 32H), 0.853 (t, J= 6.9 Hz, 6H). 2.3 Results 2.3.1 Synthesis The preparative routes used to obtain bis(phosphino)thiophene ligands P2T2 (9) and P2hex2T4 (11) are shown in Scheme 2-2.9 The quaterthiophene ligand 11 incorporates two hexyl substituents to enhance solubility in organic solvents. Compound 10 is prepared by coupling 3-hexylthiophene to the a-brominated positions of 3,5,3',5'-tetrabromo-2,2'-bithiophene via palladium-catalyzed Kumada coupling.1 7 Formation of the dianion at the (3-brominated positions 32 by addition of «-butyl lithium to 10, followed by quenching with chlorodiphenylphosphine, yields 11 as a bright yellow solid. Scheme 2-2 P 2hex 2T 4 (11) As attempts to obtain crystals of 9 were not successful, (PO)2T 2 (12) was prepared by oxidation of 9 with hydrogen peroxide, and the X-ray crystal structure of this molecule was determined from a crystal grown from methylene chloride by Dr. O. Clot (Figure 2-2).9 It was observed that the sterically demanding phosphine oxide substituents force the thiophene rings to adopt close to a S-anti conformation in the solid state, with an interannular torsion angle (S-C-C-S) of 124.1(8)°. 33 (A) (B) Figure 2-2 (a) ORTEP view of (PO)2T2 (12). The hydrogen atoms are omitted for clarity and thermal ellipsoids are drawn at 50% probability, (b) ORTEP view down the C4-C5 axis illustrating the interannular torsion angle. Phenyl groups are omitted for clarity. Figure adapted from Ref. (Phosphino)terthiophene ligands 13-15 were also prepared via lithiation with n-butyl lithium, followed by quenching with chlorodiphenylphosphine. The synthesis of PT3 (13) using this procedure (Scheme 2-3) yielded starting material, product, and a side-product. During purification by column chromatography, the three spots eluted at similar retention times; therefore, some product was lost due to discarded mixed fractions. Some oxidation of the product on the silica gel column may also have occurred, and the final yield was moderate at 28%. Scheme 2-3 Br Ph 2P, l ) « - B u L i , - 1 5 ° C 28% 2) Ph 2 PCl , 2 5 ° C 3) HCl(aq) PT 3 (13) 34 Scheme 2-4 C12H25Ph2P // W s v - | W C,7H 1 2 n 2 5 2eq iyMgEv 2.5% [PdCl2(dppf)] /7:8:6 THF-Et20-toluene 75°C C . 7 H 9 66% 2 eq Nal 5% Cul 0.1 eq (NHMeCH2)2 /4:1 xylenes-diglyme 165°C 94% PDo 2 T 5 (22) 1) 2eqNEt2 10% [Pd(OAc)2] 2) Ph2PH /CH3CN 85°C 94% P-Brominated pentathiophenes 17 and 18 were synthesized via selective palladium-catalyzed Kumada coupling 1 7 of hexyl- or dodecylthiophene, respectively, to the a positions of 5,3',5"-tribromo-2,2':5',2"-terthiophene. The corresponding (phosphino)pentathiophene ligands could not be prepared by bromo-lithium exchange followed by addition of chlorodiphenylphosphine, due to low reactivity of the lithio-anion intermediate. An alternative • 1 8 high temperature copper-catalyzed halogen exchange reaction was used to form iodo-substituted pentathiophenes 19 and 20 in high yield, and subsequent palladium-catalyzed cross-coupling with diphenylphosphine19 yielded Phex2Ts (21) and PD02T5 (22). Shown in Scheme 2-4 is the three-step preparation of 22. Selective coupling at the a positions of 5,3',5"-tribromo-2,2':5',2"-terthiophene with the dodecylthiophene Grignard reagent gave 18 in good yield (66%). The use of [PdCl2(dppf)] as a catalyst during the coupling step resulted in preferential reaction at 35 the a positions. Alternatively, the use of [NiCbCdppp)] catalyst resulted in non-selective dodecylthiophene substitution at all three brominated positions and yielded 23 as the major product (Scheme 2-4, bottom). In the case of the palladium-catalyzed coupling, 23 also formed if two equivalents of the Grignard reagent were exceeded, and careful stoichiometry must be maintained. The overall yield of 22 (87%) for the two-step process from 18 is an improvement over the yield for 13 (28%), and other reported yields for phosphinothiophenes prepared via lithiation and addition of chlorodiphenylphosphine.6'12 The compound Phex2Ts (21) and its precursor compounds (17, 19) were prepared in small amounts and consumed for subsequent reactions; consequently, neither elemental analyses nor mass spectrometry were performed. However, the N M R data is analogous to that of the corresponding docecyl-pentathiophenes (18, 20, 22) and free of impurity peaks, so that one can be reasonably confident of the structures and purity of these compounds. Scheme 2-5 20 1) 1.2 eq «-BuLi/Et 2 0 • 2) 16 eq H 2 0 P 12^ 25 Q 2 H 2 5 Do2T5(24) The soluble bis(dodecyl)pentathiophene D02T5 (24) was prepared in quantitative yield by exchange of 20 with n-butyl lithium followed by quenching with water ( Scheme 2-5). It was observed that reaction of the anionic intermediate with water was slow. The resulting dark yellow powder is also easily oxidized, and is susceptible to reaction with oxygen in air. The a,a-dibrominated, P-iodinated pentathiophene 25 was prepared by a selective, room temperature reaction using iV-bromosuccinimide in chloroform/acetic acid solution (Scheme 2-6). Due to the higher reactivity of the iodo substituent relative to the bromo substituents towards palladium-catalyzed cross-coupling with diphenylphosphine, selective phosphine addition yielded the a-brominated P-(phosphino)pentathiophene ligand PBr 2 Do 2 T 5 (27). 36 Scheme 2-6 C,,H 1 2 n 2 5 C,,H 1 2 n 2 5 20 2 eq NBS - 1 2n 2 5 C„H 1 2 n 2 5 /1:1 C H C I 3 - C H 3 C O O H 16 h, RT 1) 2eqNEt3 10% [Pd(OAc)2] 2) 1 eq Ph2PH /CH3CN 85°C C 1 2 H 2 5 Ph2P W J x s x ~\JI PBr 2 Do 2 T 5 (27) 83% 42% The p-brominated pentathiophene 18 was iodinated at the a positions with AModosuccinimide to yield 26 (Scheme 2-7). The higher reactivity of the a-iodo substituents allowed subsequent selective reaction with 2-thienyl-l,3,2-dioxaborinane via Suzuki coupling under efficient, non-aqueous conditions20 to produce the heptathiophene 28 in high yield. Compounds 27 and 28 were both found to be light- and oxygen-sensitive, and some decomposition may account for elemental analysis results that were not within the acceptable range (±0.4%) and are not reported here. Identification relies on the ' H N M R spectra that displayed peaks corresponding to the given structures (see Chapter 6, Figure 6-1 also) and are free from impurity peaks. Scheme 2-7 37 2.3.2 Cyclic Voltammetry Oxidation potentials for ligands 9, 11, and 13-15 have been previously reported to be in the range of 0.99-1.1 V vs. SCE (Table 2-1).6'9 Decreasing oxidation potential with extension of thiophene chain length is generally observed, and incorporation of electron-donating methyl substituents also lowers oxidation potentials, as is observed for ligands 13-15. The first, irreversible oxidation of triphenylphosphine (PPI13) was observed at a substantially higher potential of 1.66 V vs. SCE by our experimental setup. Therefore, first oxidation potentials of the previously reported ligands are most likely thienyl-based. Excluding 1, oxidations of the ligands are higher in potential than those of the corresponding oligothiophenes, as shown in Table 2-1, and are irreversible. Thienyl reductions are too negative in potential to be observed experimentally. Table 2-1 Cyclic voltammetry data for P-phosphinothiophene ligands and derivatives, oligothiophenes, and triphenylphosphine Compound ±0.01 V vs. SCE Compound £ l / 2 , o x ±0.01 V v s . SCE P 2 T 2 (9) +1.10b'c Do 2T 5 (24) +0.82,+1.07a P2hex2T4(11) + 1.02b'c T 2 + 1.19e (PO)2T2 (12) >2 T 3 +0.986 PT 3 (13) 1.30M T 4 +0.92e PMeT 3 (14) 1.18W T 5 +0.656 PMe 2T 3 (15) 1.05M PPh3 + 1.66b,f PDo 2T 5 (22) +0.99b,+1.37b Measurements carried out in CH 2 C1 2 solution containing 0.1 M [(«-Bu)4N]PF6 supporting electrolyte. Veversible wave, E p . cRef. 9 dRef. 6 eRef. 2 1 f C H 3 C N solution. Two irreversible oxidation waves, assigned as thienyl-based, were observed for the (phosphino)pentathiophene ligand 22. The first oxidation potential of 22 is lower than that of 13-15 due to an extended chain length. Compared to the corresponding pentathiophene 24, the thienyl-based oxidations of 22 are anodically shifted and irreversible due to diphenylphosphine substitution (Figure 2-3). Shown in Figure 2-4 is the correlation between the number of the thienyl rings and the oxidation potentials of all the P-phosphinothiophene ligands used and corresponding oligothiophenes. A general trend of decreasing oxidation potential with extension in chain length is observed. P-Phosphination generally increased oxidation potential, although diphosphination of bithiophene to give 9 resulted in a lower oxidation potential than that of T 2 . 38 50 r 40 -30 -< 20 -c fc 10 -V 0 --10 -P D o 2 T 5 (22) 0.6 0.8 1.0 1.2 Volts rv vs. SCE 1.4 1.6 Figure 2-3 Cyclic voltammetry of 22 and 24 in CH2CI2 at 4*10" 3 M concentrations, containing 1 M [(«-Bu)4N]PF6 supporting electrolyte, scan rate = 100 mV/s. w u GO > 1.2 1.1 1.0 0.9 h 0.8 0.7 0.6 - +PT3(13) " OT 2 +PMeT3(14) - +P2T2(9) -+PMe2T3(15) OT 3 +PhexT(ll) 2 2 4 +PDo2T5(22) - OT 4 - ODo2T5(24) 1 1 OT 5 1 1 2 3 4 5 Number of Thienyl Rings Figure 2-4 Correlation between the oxidation potentials, Ei/2,ox, and the number of thienyl rings of the (3-phosphinothiophene ligands (+) and the corresponding oligothiophenes (O). A l l values are from the irreversible oxidation waves, Ep,except that of 24. 39 2.3.3 Electronic Spectroscopy Table 2-2 Electronic spectroscopy data for P-phosphinothiophene ligands and derivatives, and oligothiophenes Compound Absorption3 X m a x /nm [s /IVT'crn"1] Emissionb X, m a x /nm P2T2 (9) 252 (sh) (1.65 x 104)c -P2hex2T4(11) 254 (6.20 x 104), 3 40 (2.83 x 104)c -(PO)2T2 (12) 254 (1.65 x 104), 2 86 (sh) (6.29 x 103) -PT3(13) 254 (3.42 x 104),.354 (1.77 x 104)d 420(sh), 442 (<|> = 0.0057)g PMeT3 (14) 254 (2.33 x 104), 362 (2.15 x 10 4) d ' . 454 PMe2T3 (15) 2 5 2 (3.42 x 104), 282 (sh) (2.57 x 104), 364 (3.63 x 104)d 457 aPT 3 (16) 2 5 0 (1.1 x 104), 374 (2.0 x 104)e 433, 450 (<|> = 0.055)8 PDo2T5 (22) 251 (2.13 x 104), 340 (sh) (1.47 x l O 4 ) , 406 (3.36 x lO 4 ) 499, 528 (<j) = 0.093)g Do2T5 (24) 252 (1.27 x 104),407 (3.27 x 104) 487, 516 (4> = 0.16)g T2 3 02 (4.10 x 104) f -T 3 3 5 5 (4.40 x 104) f 409, 429 O = 0.056)h T 4 3 90 (4.66 x 104) f -T 5 416 (4.74 x lf j 4 ) f 482, 514(<j) = 0.54)h "Measurements carried out in CH2CI2 solution. bDegassed C H 3 C N solution. cRef. 9 dRef. 6 eRef." fThienyl 71—>n* transition only, ref.21 Approximate quantum yields obtained by comparison to T 3 . hRef. 2 2 Absorption spectroscopy has been reported for ligands 9, 11, and 13-15 (Table 2-2).6'9 The bithiophene n—>n* transition of 12 shifts little compared to that of 9, despite the presence of electron-withdrawing phosphine oxide groups and an increase in the oxidation potential. The pentathiophene n—>n* transitions observed for 22 and 24 are red shifted compared with the terthiophene n—»7T.* transitions of ligands 13-15, as anticipated with extension in chain length. Compared with T 5 , the n—>n* transitions of both 22 and 24 are blue shifted by 10 nm. Shown in Figure 2-5 is the correlation between the number of the thienyl rings and the absorption maxima. Generally, the n — t r a n s i t i o n energy decreases as the chain length increases. It is also observed that bis(phosphine) substitution substantially increases the n—>n* transition energy, while a lesser effect is observed with substitution of one phosphine. Phenyl n—»7t* transitions 40 are also expected and a shoulder at 340 nm is observed for 21 that could be the phenyl n—»rc* transition. 450 425 400 375 s 350 j 325 -+ aPT3(16) PT3(13) OT 4 + P 2 hex 2 T 4 ( l l) T O 5 Q Do 2 T 5 (24) PDo2T5(22) - O T 2 - +P2T2(9) i i 250 3 4 5 Number of Thienyl Rings Figure 2-5 Correlation between ^ > m a x and the number of thienyl rings of representative phosphinothiophene ligands (+) and the corresponding oligothiophenes (O). The TX<—7i* fluorescence emission spectrum of PT3 (13) is compared to that of T3 and to the emission of the a-(phosphino)terthiophene ligand aPT3 (16) (Figure 2-6). It is observed that attachment of the phosphine group at the a position for 16 red shifts the emission 21 nm, and has no effect on the intensity of the emission compared to T 3 . Ligand 13 displays n<— n* fluorescence emission that is red shifted by 13 nm from that of T 3 . However, the intensity is greatly reduced to approximately one tenth that of T3. Ligand P D 0 2 T 5 (22) and pentathiophene D02T5 (24) display TC<—n* fluorescence emission that is red shifted by 5-17 nm from that of T 5 , and is also greatly reduced in emission intensity (Table 2-2). 41 0.4 o 0.3 o o o o c 3 O <-> 0.1 0.0 350 400 450 500 550 600 650 700 X/nm Figure 2-6 The T C < - K * fluorescence emission of (phosphino)terthiophene ligands 13 and 16, and T 3 . Absorbance at the excitation wavelength is set to 0.1. 2.4 Discussion P-Phosphinothiophene ligands with thienyl chain lengths of n = 2-4 (9,11,13-15) could be prepared in moderate yields by reaction with «-butyl lithium followed by quenching with chlorodiphenylphosphine. However, this procedure was not feasible for substitution of pentathiophenes due to the formation of an unreactive intermediate anion, and a new route was attempted. Conversion of P-brominated pentathiophene derivatives 17-18 to the corresponding P-iodo derivatives 19-20 and efficient cross-coupling of the reactive iodo substituents with diphenylphosphine successfully yielded the P-(phosphino)pentathiophene ligands Phex 2T 5 (21) and P D 0 2 T 5 (22). The merits of this method are that it is efficient and is also not limited by the number of anionic charges that the conjugated backbone can carry, so that the procedure could be applied to polythiophenes containing several P-iodo functionalized positions to yield conjugated polymers with multiple P-phosphine substituents. The reported X-ray crystal structure of the ligand oxide (PO) 2 T 2 (12)9 was used to estimate the steric effect of adjacent diphenylphosphine groups attached to oligothiophenes (Figure 2-2). It is observed that there is a large interannular torsion angle (124.1(8)°) between the two thienyl rings in the crystal structure of 12, deviating substantially from the near-planar (-180°) torsion angles of solid state oligothiophenes ' and from the interannular torsion angle - 42 calculated for the most stable conformation of T3 (147.6°). A similar angle is likely present for 9 and for the central rings of 3 due to an analogous steric repulsion. The higher oxidation potential of P2hex2T4 (11) compared to T 4 is consistent with expected twisting of the thienyl chains. The lower oxidation potential of P 2 T 2 (9) compared to T 2 suggests less twisting or donation of electron density from the lone pair of the phosphine group. It has been calculated that an oligothiophene HOMO is stabilized by interaction between the n system and an a phosphine, and the degree of lone pair character diminishes with an increase in oligothiophene length.11 The irreversibility of the oxidation potentials of phosphinothiophene ligands could be due to interaction of the oxidized thiophene chains with the phosphine group. The thienyl n—>n* transitions of bis(phosphine) ligands 9 and 11 are both blue-shifted by approximately 50 nm 26 relative to T 2 and T 4 , respectively. The blue shifts are not likely to be due to an electronic withdrawing effect since there is a negligible shift of n—±n* transition observed for P T 3 (13) (^max = 354 nm), 1 2 which has one pendent phosphine attached, relative to T3 (355 nm). 2 6 Therefore, the steric effect of the bulky, adjacent phosphines twisting the thiophene chains out of conjugation is most likely shifting the n—>rt* transition. The transition of 9 is expected to be decreased in energy by electron donation from the phosphine lone pairs, but this contribution appears to be counteracted by chain twisting. This twisting of the thienyl backbone is interesting because the presence of metals could enhance the effect, and alter the conjugation pathway. The (3-(phosphino)terthiophene ligands P T 3 (13), P M e T 3 (14), and P M e 2 T 3 (15) display anodic shifts of the irreversible oxidation potentials compared with T3, as does the P-(phosphino)pentathiophene ligand PDo2T5 (22) compared with D02T5 (24). The thienyl 71—>7i* transition energies of 13 and 22 compared to T 3 and 24 are largely unaffected by phosphine substitution, while the transitions of 14-15 are lower in energy due to methyl substitution. A combination of thienyl ring overlap and donation of electron density from the substituents is likely affecting the oxidation potentials and transition energies. Overall, results suggest that twisting of thiophene chains with addition of one diphenylphosphine group is not substantial. The 21 nm red shift observed for the 71 <— n* fluorescence emission of C1PT3 (16) compared to T3 is consistent with a red shift in the absorption transition due to donation of a phosphine lone pair to the n conjugation." The 7t<— n* fluorescence emission of 13 is 13 nm red shifted from the emission of T3, though the absorption energy is equivalent. Therefore, 13 displays a relatively large Stokes shift. This could indicate the transition has some n—>n* character due to a lone pair lone pair contribution, and change in geometry from a more 43 pyramidal ground state to a planar excited state would result in the Stokes shift. It is also seen that while 01PT3 (16) does not show any difference in intensity compared with T3, emission from the P-substituted ligand PT3 (13) is much weaker in intensity. The emission results of D02T5 (24) indicate that the introduction of dodecyl chains decreases the intensity of emission compared to T 5 , and p-(phosphino)pentathiophene ligand P D 0 2 T 5 (22) displays a further reduction in emission intensity compared to 24. It is probable that extra substituents placed at p positions reduce the thienyl-based emission intensity by the introduction of more degrees of freedom and the creation of more non-radiative relaxation pathways. In the case of constitutional isomers 13 and 16, simply the placement of the phosphine group at the p position rather than the a position reduces the intensity, and it is also possible that torsional oscillations facilitate more efficient non-radiative decay. Exploitation of the difference in reactivity between the bromo- and iodo-substituents towards phosphine cross-coupling and Suzuki reactions was demonstrated. a,a,P-Halogenated pentathiophene derivatives 25 and 26 were prepared for the purpose of developing a strategy to prepare long-chained multi-metallated oligothiophenes with D02T5 as the repeat unit of the thienyl backbone. The selective phosphine cross-coupling reaction of the central P-iodo position of 25 gave PBr2Do2T5 (27), which possesses two remaining bromo substituents at the terminal a positions. These brominated substituents could facilitate coupling reactions between subsequent metal complexes. Alternatively, 26 contains a bromo-substituent at a central P position and highly reactive iodo substituents at the a positions. The preparation of the P-brominated heptathiophene 28 by Suzuki reaction in high yield demonstrated selective cross-coupling at the a positions. Syntheses carried out to prepare compounds 27 and 28 suggest possible reaction pathways that could be taken to prepare P-metallated polymers. 2.5 Conclusions The addition of two adjacent P-phosphine substituents alters thienyl structural and electronic properties via steric and electronic interaction of the phosphine groups, while the addition of one P-phosphine substituent results in little conformational or electronic effect on the thiophene chains. P-Phosphine substitution is shown to diminish thienyl n<-n* fluorescence emission intensity compared with a-phosphine substitution. The phosphine groups of the P-bis(phosphino)thiophene ligands present sites where oligothiophene properties can be altered 44 via conformational change. The configuration of the p-phosphinothiophene ligands provides an opportunity for direct interaction of a phosphine-coordinated metal with the thiophene chain. Towards the preparation of ligands of this class, a new synthetic route from P-(bromo)oligothiophenes to P-phosphinothiophene ligands is demonstrated that is high-yielding and effective for preparation involving longer oligothiophenes. Additionally, the reactivity differences of the bromo- and iodo-substituents can be exploited to yield a-functionalized ligands or p-functionalized oligothiophenes. The P-phosphinothiophene ligands and derivatives thus prepared and characterized provide the foundation for a variety of metallated P-phosphinothiophenes to be studied. 2.6 References (1) Reddinger, J. L. ; Reynolds, J. R. Chem. Mater. 1998, 10, 3-5. (2) Zhu, S. S.; Carroll, P. J.; Swager, T. M . J. Am. Chem. Soc. 1996, 118, 8713-8714. (3) Yam, V . W.-W.; L i , C.-K.; Chan, C.-L. Angew. Chem. Int. Ed. 1998, 37, 2857-2859. (4) Stott, T. L. ; Wolf, M . O.; Patrick, B. O. Inorg. Chem. 2005, 44, 620-627. (5) Clot, O.; Wolf, M . O.; Patrick, B. O. J. Am. Chem. Soc. 2000, 122, 10456-10457. (6) Clot, O.; Wolf, M . O.; Patrick, B. O. J. Am. Chem. Soc. 2001, 123, 9963-9973. (7) Weinberger, D. A. ; Higgins, T. B.; Mirkin, C. A. ; Stern, C. L. ; Liable-Sands, L. ML; Rheingold, A. L. J. Am. Chem. Soc. 2001, 123, 2503-2516. (8) Rogers, C. W.; Wolf, M . O. Chem. Commun. 1999, 2297-2298. (9) Clot, O.; Akahori, Y . ; Moorlag, C ; Leznoff, D., B.; Wolf, M . , O.; Batchelor, R., J.; Patrick, B. , O.; Ishii, M . Inorg. Chem. 2003, 42, 2704-2713. (10) Clot, O. n-Conjugated Materials Containing Transition Metals, Ph.D. Thesis; Department of Chemistry, University of British Columbia, Vancouver, Canada, 2001, 185 pp. (11) Stott, T. L. ; Wolf, M . O.J . Phys. Chem. B 2004, 108, 18815-18819. (12) Clot, O.; Wolf, M . O.; Yap, G. P. A. ; Patrick, B. O. J. Chem. Soc, Dalton Trans. 2000, 2729-2737. (13) Khor, E.; Ng, S. C ; L i , H . C ; Chai, S. Heterocycles 1991, 32, 1805-1812. (14) Kellogg, R. M . ; Schaap, A . P.; Harper, E. T.; Wynbert, H. J. Org. Chem. 1968, 33, 2902-2909. (15) Kobayashi, Y . ; Mizojiri, R.; Ikeda, E. J. Org. Chem. 1996, 61, 5391-5399. (16) Robbins, J. L. ; Edelstein, N . ; Spencer, B.; Smart, J. C. J. Am. Chem. Soc. 1982, 104, 1882-1893. (17) Nakanishi, H. ; Sumi, N . ; Aso, Y . ; Otsubo, T. J. Org. Chem. 1998, 63, 8632-8633. 45 Klapars, A. ; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 14844-14845. Herd, O.; Hessler, A. ; Hingst, M . ; Tepper, M . ; Stelzer, O. J. Organomet. Chem. 1996, 522,69-76. Kirschbaum, T.; Azumi, R.; Mena-Osteritz, E.; Bauerle, P. New J. Chem. 1999, 23, 241-250. Fichou, D. Handbook of Oligo- and Polythiophenes; Wiley-VCH: Weinheim, 1999. Becker, R. S.; de Melo, J. S.; Macanita, A . L. ; Elisei, F. J. Phys. Chem. 1996, 100, 18683-18695. Antolini, L. ; Horowitz, G.; Kouki, F.; Gamier, F. Adv. Mater. 1998, 10, 382-385. Pelletier, M . ; Brisse, F. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1994, C50, 1942-1945. DiCesare, N . ; Belletete, M . ; Marrano, C ; Leclerc, M . ; Durocher, G. J. Phys. Chem. A 1998, 102, 5142-5149. Van Pham, C ; Burkhardt, A. ; Shabana, R.; Cunningham, D. D.; Mark, H. B. J.; Zimmer, H. Phosphorus, Sulfur Silicon Relat. Elem. 1989, 46, 153-168. 46 CHAPTER 3 Synthesis and Characterization of Au(l) and Pd(ll) Complexes 3.1 Introduction One approach to alter the properties of conjugated oligo- or polythiophenes is to attach pendent groups to the conjugated backbone. The attachment of metal groups to ^-conjugated backbones allows tuning of the electronic properties such as the colour, the electrochemical oxidation potential, or the conductivity. Pendent metal groups may electronically couple to an oligo- or polythiophene chain depending on proximity of the metal and the overlap of orbitals. Alternatively, metal coordination can change the conformation of the conjugated chain to modify the properties. P-Bis(phosphino)thiophenes that have two adjacent P-phosphine substituents can coordinate one metal group in a bidentate fashion to form a seven membered ring, or coordinate two metal centers in a monodentate fashion. Complexes formed with these ligands have the potential to lock the conformation of a thiophene chain, or twist the chain due to the steric repulsion of two metal groups. As conjugation along a thienyl backbone is determined by the overlap between the n-orbitals of adjacent thiophene rings,1 metallation that alters the conjugation is also expected to modify electronic properties such as oxidation potential, absorption and emission wavelengths, and conductivity. In order to probe the effect of metallation on the structural and electronic properties, a collaborative project was carried out where model complexes were prepared with P-bis(phosphino)thiophene ligands P2T2 (9) and-P 2 hex 2 T 4 (11). Chelated Pd(II) complexes 29 and 30 (Chart 3-1) were first prepared by Dr. O. Clot in this group.2 The square planar Pd(II) center of these complexes is coordinated in a bidentate fashion by both phosphine groups. In order to complete the characterization of these compounds, complex 30 was prepared and a crystal structure was obtained. Cyclic voltammetry and absorption spectroscopy experiments for 30 were carried out in order to establish consistent trends for the model complexes. 47 Chart 3-1 ci c i \ / Pd / \ Ph2P PPh 2 Pd(P 2 T 2 )Cl 2 (29) Q H 1 3 H 1 3 Q Pd(P 2hex 2T 4)Cl 2 (30) Gold displays the property of aurophilicity,3 where closed-shell, d 1 0 metals interact with one another to form weak bonds. One example of a complex displaying aurophilicity is 31 (Chart 3-2). This complex can be used to sense potassium ions via emission resulting from Au-Au interactions.4 In our group, cc-phosphinothiophenes have been used to prepare Au(I)-thiophene metallocycles, such as 32.5 Emissive states form due to gold-gold interactions in the solid state, and the metallocycles fluoresce. Chart 3-2 Au(I) complexes 33 and 34 were first prepared by collaborator Y . Akahori at Simon Fraser University (Chart 3-3).6 These complexes contain two gold atoms per 48 (3-bis(phosphino)thiophene ligand, and display Au-Au interactions in the solid state. For the purpose of further characterization of these compounds, complexes 33 and 34 were re-prepared, and cyclic voltammetry experiments were carried out. A l l of the results obtained for the Pd(II) and Au(I) complexes are reported, and their properties are compared to one another, to ligands P 2 T 2 (9) and P2hex2T4 (11), to the ligand oxide (PO)2T2 (12), and to T 2 and T 4 (Chapter 2) in order to establish structure/property relationships for these metal-oligothiophene hybrid materials. Chart 3-3 ^ C l ^ C l P h 2 P " A ' U H 1 3 C 6 P h 2 P ^ U ,>PPh 2 A / > P P h 2 C 6 H 1 3 (AuCl)2P2T2 (33) (AuCl)2P2hex2T4 (34) 3.2 Experimental 3.2.1 General Experimental Complex 29 was prepared by Y . Akahori. 6 Complexes 30, 33 and 34 were prepared from P2T2 (9) and P2hex2T4 (11) according to previously reported procedures.6 Electronic absorption spectra were obtained in CH 2 C1 2 on an HP8452A diode-array spectrophotometer. Electrochemical measurements were conducted on a Pine AFCBP1 bipotentiostat using a platinum disc working electrode, platinum coil wire counter electrode and a silver wire reference electrode. An internal reference, either decamethylferrocene (-0.12 V vs. SCE) or ferrocene (0.41 V vs. SCE) was added to correct the measured potentials with respect to saturated calomel electrode (SCE). 7 The supporting electrolyte was 0.1 M [(n-Bu)4N]PF6 that was purified by triple recrystallization from ethanol and dried at 90°C under vacuum for three days. CH 2 C1 2 used for cyclic voltammetry experiments was dried with CaH 2 and distilled 49 3.2.2 X-ray Crystallographic Analysis Data for the X-ray crystallographic analysis of 30 was collected on a Rigaku/ADSC CCD diffractometer from crystals obtained by slow diffusion of hexanes over CH2CI2 solution. Crystal structure acquisition of 30 was completed by Dr. R. Batchelor at Simon Fraser University. Both hexyl groups are disordered, and the structure was modeled in two orientations, with the major disordered fragment refined anisotropically and the minor fragments refined isotropically. Hydrogen atoms were placed in calculated positions. Appendix 1 contains crystallographic data for the crystal structure of 30. The cif file is available online.6 3.3 Results 3.3.1 Synthesis and Crystallographic Studies Scheme 3-1 PPh, [PdCl2] • EtOH/MeCN H20/HC1 50°C CI - £ 1 ' Pd / \ Ph2P PPh2 // w Pd(P 2T 2)Cl 2 (29) [PdCl2] »-EtOH/MeCN H20/HC1 50°C P 2hex 2T 4 (11) C 6 H 1 3 H 1 3 C 6 Pd(P2hex2T 4)Cl 2 (30) Ligands P 2 T 2 (9) and P 2 hex 2 T 4 (11) were chelated to a Pd(II) center to form complexes 29 and 30 that contain a metallocycle at the central rings of the T 2 and T 4 chains. The reaction of [PdCl2] with 9 to yield the Pd(II) complex 29 as a bright yellow solid is shown in Scheme 50 3-1.6 Complex 30 was prepared by the same method and purified by recrystallization from CFbCb/hexanes solution. The orange, air-stable powder is soluble in polar solvents such as methylene chloride and chloroform. Figure 3-1 ORTEP view of Pd(P2hex2T4)Cl2 (30). The hydrogen atoms are omitted for clarity and thermal ellipsoids are drawn at 50% probability. Crystallographic analysis was carried out on a crystal of 30 obtained from hexanes/CH2Cl2 solution. It is seen from the crystal structure data (Figure 3-1) that the p-bis(phosphino)quaterthiophene ligand is chelated to one Pd(II) center via the two phosphine groups, forming a seven-membered ring with a P-Pd-P bite angle of 93.96(3)° (Table 3-1). The Pd(II) center in the solid state structure lies in a distorted square planar geometry with the cis chlorine atoms slightly raised above the P-Pd-P plane. Coordination of Pd(II) to 11 forces the thienyl sulfur atoms of the central bithiophene group to close to a syn conformation with an interannular torsion angle of 56.6(3)°. This angle is slightly higher than the interannular torsion angle that was observed for 29 at 51.1(2)°.6 Both angles are approximately the same magnitude from planar as for the solid state structure of (PO)2T2 (12) at 124.1(8)°. A l l thienyl rings of 30 are oriented in an approximately syn fashion, with the sulfur atoms alternating above and below a 51 plane along the quaterthiophene chain. T4 is nearly planar in the solid state, while structural deviations are sometimes observed due to steric interactions between organic substituents.8 The C8-C9 bond in 30 (1.452(4) A ) is shorter than either of the outer two interannular C-C bonds and the interannular bonds lengths of T 4 , but is consistent with the interannular bond length in 29 (1.456(3) A ) . 6 Table 3-1 Selected interatomic distances (A) and angles (°) for PdtF^hexiT^Ch; (30) Bond length /A Pd,-P, 2.2490(9) Pd,-P 2 2.2510(9) Pd,-Cl, Pd,-Cl 2 2.3426(9) 2.3342(9) C 4 - C 5 C8-C9 • 1.470(5) 1.452(4) C ,2 -C , 3 1.466(5) Bond Angle 1° P,-Pd,-P 2 P,-Pd,-Cl, Pi-Pd,-Cl 2 Cl,-Pd,-Cl 2 93.96(3) 86.04(3) 170.38(4) 91.30(4) P 2-Pd,-Cli P 2 -Pd,-Cl 2 Pd,-P 2 -C 1 0 Pd,-P,-C 7 169.06(4) 90.40(3) 113.77(10) 118.23(10) Torsion Angle 1° Si"C4-C5-S2 S2-C8-C9-S3 58.8(3) 56.6(3) S3"Cl2-Cl3-•s4 69.0(3) The Au(I)-bithiophene complex 33 was synthesized via the reaction of P2T2 (9) with [AuCl(tht)] (tht = 2,3,4,5-tetrahydrothiophene) (Scheme 3-2).6 Two solid state structures were obtained by Y. Akahori; 6 in both cases the Au(I) center binds to one phosphine and one chloride in a nearly-linear geometry (Figure 3-2). The structure determined from crystals grown from CH 2 C1 2 /Et 2 0 solution suggests a Au-Au interaction9,10 due to a distance of 3.3221(4) A between metal atoms. The structure determined from crystals grown from CFbCVtoluene solution appears as toluene adduct 33-tol and displays a greater Au-Au distance that does not suggest an interaction. In solution, only an averaged 3 1 P{ 1 H} N M R signal was observed for 33, and there was no significant shift or decoalescence of the singlet in CD2CI2 or in CD2CI2/C7D8 from 183-298 K. The average Au-P (-2.23 A ) and Au-Cl (-2.28 A) bond lengths are comparable with those observed in substituted gold(I) triphenylphosphine chloride complexes (-2.23 and -2.29 A respectively).11 The crystal structure of 33 displays the two thiophene rings oriented S-anti to one other with an interannular torsion angle of 110.8(5)°, while for 33-tol, the torsion angle is 52 slightly larger, at 115.3(7)°. The ligand conformation despite the lack of a formal Au-Au bond (> 3.6 A) . of 33 -tol is very similar to that of 33, Scheme 3-2 P 2hex 2T 4 (11) (AuCI) 2P 2hex 2T 4 (34) Reaction of quaterthiophene ligand P2hex2T4 (11) with two equivalents of [AuCl(tht)] yielded red solids of 34 by an analogous procedure to that of 33 (Scheme 3-2). The solid-state structure of 34 is similar to that of 33,6 but with a quaterthiophene chain in the place of bithiophene. The Au-Au bond length of 3.0879(7) A in 34 is shorter than that of 33, which may promote the observed decrease in the linearity of the Cl-Au-P angles of 34 (174.28(10)°, 176.87(12)°) compared to those of 33 (178.95(5)°) or 33-tol (179.14(6)°). The interannular torsion angles of the central, internal pair of rings of 34 are more twisted at 100.8(9)°, compared with both 33 (110.8(5)°) and (PO)2T2 (12) (124.1(8)°) respectively, a consequence of the shorter Au-Au bond. The central rings are approaching a 90° torsion angle; hence, minimal 7i-orbital overlap is expected. The sterically less hindered outer rings are only twisted by 28.5(12)° and 12.5(11)°, and are oriented close to an S-syn fashion with respect to one another. 3.3.2 Cyclic Voltammetry The cyclic voltammograms of 29 and 30 contain irreversible first oxidation waves at 1.67 V and 1.40 V , respectively, that are assigned as thienyl-based processes (Table 3-2). These oxidations occur significantly positive of ligands 9 and 11 (1.10, 1.02 V), and of T 2 and T4 (1.19, 1.00 V), respectively. The oxidation potential of 30 is lower than that of 29 due to the longer oligothiophene chain length. Stabilization of the first thienyl oxidation of 30 also supports a second oxidation at 1.51 V (Figure 3-3a). When the working electrode was scanned repeatedly past the first oxidation waves in solutions of the 29 and 30, there was no evidence of film formation on the electrode surface. Table 3-2 Cyclic voltammetry data of Pd(II)- and Au(I)-bis(phosphino)thiophene complexes Compound Ei/2,0x ±0.01 V a vs. SCE Pd(P2T2)Cl2 (29) Pd(P2hex2T4)Cl2 (30) (AuCl)2P2T2 (33) (AuCl)2P2hex2T4 (34) Measurements carried out in C H 3 C N solution containing 0.1 M [(«-Bu)4N]PF6 supporting electrolyte, irreversible wave, E p . +1.67 b(P 2T 2 + / 0) +1.40b (P2hex2T4+ / 0), +1.59b (P 2hex 2T 4 2 + / +) >+2b (P 2 T 2 + / 0 ) +1.49" (P2hex2T4+/0) 54 14 10 12 Pd(P2hex2T4)CI2 (30) < (b) -2 -1 0.0 0.5 1.0 1.5 2.0 2.5 0.75 1.00 1.25 1.50 1.75 2.00 Volts /V vs SCE Volts FV vs SCE Figure 3-3 Cyclic voltammograms of (a) Pd(II) complex 30 and (b) Au(I) complex 34 in CH2CI2 solution containing 1 M [(rc-Bu)4N]PF6 supporting electrolyte, scan rate = 100 mV/s. Plot (a) also shows the Fc/Fc + redox couple at 0.41 V vs. SCE. 7 Cyclic voltammetry of 33 indicates that the bithiophene oxidation potential is >2 V (Table 3-2) and is substantially shifted from the oxidation potentials of ligand 9 and T 2 . Complex 34 also displays an increase in oxidation potential compared to 11 and T 4 . Au(I) complexes 33 and 34 display more anodic oxidation potentials than Pd(II) complexes 29 and 30, which is likely due to an inductive, electron-withdrawing effect of the two complexed Au(I) groups. Repeat scans past the first oxidation wave for 34 showed no evidence of film formation on the electrode surface, though the oxidation wave is irreversible (Figure 3-3b). This result indicates that electropolymerization does not occur. 3.3.3 Electronic Spectroscopy Absorption spectra of 29 and 30 display multiple bands in the region of 258-451 nm. Complex 30 displays broad bands between 275-400 nm (Figure 3-4), which is the region of the Tt—>7i* transition of P2hex 2T 4 (11) (340 nm). A Cl—»Pd L M C T band is also expected at around 340 nm, 1 2" 1 4 and shoulders observed for complexes 29 and 30 at 354 nm and 340 nm, respectively, could be attributed to this transition. Absorption spectra of 30 in solvents of different polarity (Figure 3-4) display a small shift for the weak, lowest energy absorption; however, the shift is within experimental error of the instrument (±1 nm), and the transition 55 cannot be confidently assigned as a charge transfer. The new transitions in the visible regions of 29 and 30 result in yellow and orange coloured complexes, respectively. Table 3-3 Electronic spectroscopy data of Pd(II)- and Au(I)-bis(phosphino)thiophene complexes Compound Absorption3 A, m a x /nm [e /IVr'cm"1] Pd(P 2 T 2 )C l 2 (29) 258 (2.51 x 105), 280 (2.51 x 414 (sh) (9.22 x 103)b 105), 354 (sh) (4.19x 104), Pd(P 2 hex 2 T 4 )Ch (30) 267 (2.06 x 104), 3 02 (1.68 x 451 (2.52 x 103) 104), 340 (sh)(1.42x 104), (AuC l ) 2 P 2 T 2 (33) 239 (sh) (3.52 x 104)b (AuCl)2P2hex2T4 (34) 252 (2.88 x 104), 3 44 (1.66 x 104)b aMeasurements carried out in CH 2 C1 2 solution. bRef. 6 250 300 350 400 450 500 550 600 X/nm Figure 3-4 Absorption spectra of Pd(P 2 hex 2 T 4 )Cl 2 (30) in solvents of different polarity, displaying a shift in the low-energy band. The bithiophene n—>TT* transition of 33 (Table 3-2) is blue shifted from the transition of P2T2 (9) (252 nm). While blue shifting suggests unfavorable thienyl 71-orbital overlap, it is unlikely that any Au-Au interactions persist in solution. In contrast, the quaterthiophene n—*n* transition of 34 is very close in energy to that observed for P2hex2T4 (11). 56 3.4 Discussion The crystal structures of Pd(II) complexes 296 and 30 suggest an overall loss of Tt-orbital overlap across the oligothiophene chains with metallation. The formation of the central Pd(II) metallocycles results in shorter interannular bond distances between the two central rings compared to T 4 . However, interannular torsion angles of 51.1-56.6° for the locked conformations are unfavorable for 71-orbital overlap compared with the near-planar oligothiophene conformations in the solid-state. The unfavorable torsion angles likely counteract any enhanced conjugation due to shorter central bonds and overall conjugation across the central rings is likely to be poor. There are two factors that can contribute to the anodic shifts of 29 and 30 compared to the ligands and oligothiophenes: (a) the inductive effect of the metal and (b) a change in the interannular conjugation that results upon coordination. The interannular torsion angles of the complexes are unfavorable compared to oligothiophenes, but expected to be approximately the same distance from planarity compared to the corresponding ligands. Therefore, the electron-withdrawing effect of the PdCh group15 is likely to be the more significant effect, resulting in ~0.5 V anodic shifts of the oxidation potentials relative to both of the corresponding ligands. Compared to 29, increased n-orbital overlap across the quaterthiophene chain is expected for complex 30, in addition to electron donation from the P-hexyl chains, which likely supports the second, thiophene-based oxidation observed at 1.51 V . There are two possible reasons why complexes 29 and 30 do not electropolymerize. First, low conductivity of metal complexes that couple to form solids would deter film growth on the electrode surface, and it is likely that conductivity across even an extended 7T-conjugated backbone would be poor due to unfavorable rr-orbital overlap. Second, the bulky diphenylphosphine groups attached to the backbone may discourage interactions between chains to result in a non-conductive film. In contrast, uncomplexed oligothiophene chains display near 180° coplanarity of the rings and rc-stacking solid lattices.8 The broad bands that appear to be multiple transitions in the TC—>-TC* transition region of the absorption spectra of 29 and 30 may be a consequence of the "locked" structure of the palladium complexes. Complexation may limit free rotation of the oligothiophene rings in solution and produce different conformations that do not easily interconvert. Bands observed are blue shifted overall from the expected n—>n* transition of T2 and T 4 chains, as expected due to less favorable n-orbital overlap in the locked conformation. Shoulders at 354 nm and 340 nm for 57 29 and 30, respectively, are in the expected region of the Cl—>Pd L M C T band reported for cis and trans PdCl2L2 (L=phosphine) complexes at -340 nm 1 2" 1 4 and may correspond to this charge transfer. Low-energy charge transfer bands are also observed that are weak and substantially red shifted from the reported region of the Cl—>Pd L M C T , and the origin of this transition is unknown. The extension of thiophene chain length appears to lower the energy of this band by 35 nm from 29 to 30. Due to multiple transitions occurring with Pd(II) metallation, yellow and orange coloured complexes are observed. The crystal structure of Au(I)-P-bis(phosphino)bithiophene complex 33 displays a weak Au-Au interaction at a separation of 3.3221(4) A , while the toluene adduct 33-tol demonstrates that an interruption of the interaction occurs easily via a weakly coordinating solvent.6 The 31 1 single P{ H} N M R peak observed in solution also suggests that the Au-Au interaction observed for 33 can be easily disrupted by intermolecular forces. Fixing the conformation of the thiophene rings is a similar result of metal complexation as for the Pd(II) complexes discussed; however, in the case of the Au(I) complexes, the presence of two metals denies the chain the ability to align in an anti or syn planar arrangement and very unfavorable rc-orbital overlap is the result. The eight-membered ring formed aligns the thienyl rings of 33 into a highly twisted configuration with an interannular torsion angle of 110.8(5)°. Similar results are observed for the crystal structure of 34,6 although the Au-Au separation is shorter (3.0879(7) A) , resulting in a less planar torsion angle (100.8(9)°). Crystal packing may be a factor promoting closer proximity of the metal atoms, or the Au-Au interaction appears to be better stabilized when attached to an extended thiophene chain. Since the central interannular torsion angle is near 90°, the quaterthiophene chain can be considered as a pair of adjacent bithiophene units. The lengthened central interannular bond also suggests weak conjugation between the two central rings of 34. The conformations that arise with Au(I) metallation contrast those of Pd(II) metallation, where the C-C bond lengths between the central rings are decreased and rc-orbital overlap is less disrupted. In solution, the very anodic oxidation potential of 33 compared to 9 is most likely due to the loss of extended conjugation of the phosphine lone pairs combined with an electron-withdrawing inductive effect due to complexation of two Au(I) ions. Donation of phosphine lone pairs has been observed to more strongly affect the properties of short-chained phosphinothiophenes.16 More unfavorable steric interaction of the bulkier Au(I) groups is also possible. There is an anodic shift in the oxidation potential of 34 compared with 11, likely for 58 similar reasons though the shift is not as strong, reflecting that the phosphine lone pairs likely do not substantially affect conjugation of the quaterthiophene chain prior to complexation. Anodic shifts are not likely to be due to Au-Au interactions since these do not commonly occur in solution and are not indicated experimentally. Electropolymerization does not occur with repeat scans of the first oxidation wave of 34; it is likely that interannular twisting could occur in the solid state due to Au-Au interactions, and prevent any coupled materials from being sufficiently conductive to allow film growth. The bulky substituents may also decrease conductivity between chains. The bithiophene rc—>rc* transition of 33 is substantially blue-shifted from those of 9 and T 2 . It is expected that twisting of the central rings is similar to that of 9, and the blue shift most likely reflects removal of the phosphine lone-pairs from bithiophene Tt-conjugation. Complex 34 does not display a blue shift of the quaterthiophene rc—>-7x* transition compared to 11. This result again suggests that twisting is comparable to the ligand, and the phosphine lone pairs did not contribute significantly to the conjugation of 11. 3.5 Conclusions Binding metals to oligothiophenes via P-bis(phosphino)thiophene ligands results in substantial effects on the electronic properties of the ^-conjugated system. Changes in the electronic properties that occur with Pd(II) or Au(I) complexation result from a combination of factors including changes in the interannular torsion angles, inductive electron-withdrawing effects of the metals, and charge transfers involving the complexed metal. The interannular torsion angles in the crystal structures of the complexes examined here are large relative to those typically observed in solid-state structures of oligothiophenes. The Pd(II) complexes display central interannular torsion angles between 51-57° that are similar in planarity to that observed for (PCO2T2 (12). The Au(I) complexes display torsion angles in the solid state that are very unfavorable for rc-orbital overlap, between 100-115°, that are most likely a consequence of Au-Au interactions. The choice of metal is therefore controlling the extent of n-orbital overlap. In solution, structures may be different from the solid state crystal structures, especially for the Au(I) complexes that contain weak Au-Au bonds, and electronic spectroscopy and electrochemistry measurements are valid only for complexes in solution. Pd(II) complexation blue-shifts the thienyl rc—>-rc* transitions, likely producing multiple conformations, and also results in the appearance of charge-transfer bands. While Au(I) complexation of P 2 T 2 (9) blue 59 shifts the thiophene TC—>-TC* transition, likely due to the removal of the phosphine lone-pairs from conjugation, Au(I) complexation of P2hex2T4 (11) negligibly shifts the transition. The inductive effect due to the metals of all complexes is evident, as all first oxidation potentials are higher than those of the ligands. Oxidation potentials of the metal complexes are also mediated by chain-length. Overall in solution, Au(I) complexation exerts less influence on the physical properties than Pd(II) complexation that results in the formation of fixed metallocyclic structures. These complexes cannot be electrochemically coupled to produce polymers, likely due to lack of conductivity of solid films that are produced electrochemically. If extended chains of these model complexes are prepared by another method, it would be interesting to study the conductivity of the thiophene chains in fixed conformations resulting from metal complexation. The Au(I) complexes may especially be interesting as a potential conductive switch due the weak Au-Au interactions that may be formed and interrupted. 3.6 References (1) Bredas, J. L. ; Street, G. B.; Themans, B.; Andre, J. M . J. Chem. Phys. 1985, 83, 1323-1329. (2) Clot, O. K-Conjugated Materials Containing Transition Metals, Ph.D. Thesis; Department of Chemistry, University of British Columbia, Vancouver, Canada, 2001, 185 pp. (3) Bardaji, M . ; Laguna, A. J. Chem. Educ. 1999, 76, 201-203. (4) Yam, V . W.-W.; L i , C.-K.; Chan, C.-L. Angew. Chem. Int. Ed. 1998, 37, 2857-2859. (5) Stott, T. L. ; Wolf, M . O.; Patrick, B. O. Inorg. Chem. 2005, 44, 620-627. (6) Clot, O.; Akahori, Y . ; Moorlag, C ; Leznoff, D., B. ; Wolf, M . , O.; Batchelor, R., J.; Patrick, B. , O.; Ishii, M . Inorg. Chem. 2003, 42, 2704-2713. (7) Robbins, J. L. ; Edelstein, N . ; Spencer, B.; Smart, J. C. J. Am. Chem. Soc. 1982, 104, 1882-1893. (8) Lukevics, E.; Barbarella, G.; Arsenyan, P.; Belyakov, S.; Pudova, O. Chem. Heterocycl. Compd. 2000, 36, 630-662. (9) Pathaneni, S. S.; Desiraju, G. R. Inorg. Chem. 1993, 319-322. (10) Leznoff, D. B.; Xue, B. -Y. ; Batchelor, R. J.; Einstein, F. W. B.; Patrick, B. O. Inorg. Chem. 2001, 40, 6026-6034. (11) Leznoff, D. B. ; Rancurel, C ; Sutter, J.-P.; Rettig, S. J.; Pink, M . ; Paulsen, C ; Kahn, O. Inorg. Chem. 1999, 3593-3599. (12) Leung, K . H. ; Szulbinski, W.; Phillips, D. L. Mol. Phys. 2000, 98, 1323-1330. 60 Trzeciak, A . M . ; Bartosz-Bechowski, H. ; Ciunik, Z.; Niesyty, K. ; Ziolkowski, J. J. Can. J. Chem. 2001, 79, 752-759. Verstuyft, A . W.; Redfield, D. A. ; Cary, L. W.; Nelson, J. H . Inorg. Chem. 1976, 75, 1128-1133. Corain, B.; Longato, B.; Favero, G.; Ajo, D.; Pilloni, G.; Russo, U . ; Kreissl, F. R. Inorg. Chim. Acta 1989, 157, 259-266. Stott, T. L. ; Wolf, M . O. J. Phys. Chem. B 2004,108, 18815-18819. 61 CHAPTER 4 Synthesis and Characterization of P,S-Bound Ru(ll) Complexes 4.1 Introduction Direct bonding of metals to the backbone of thiophene chains can be employed to alter the structural, chemical, and electronic properties of the conjugated material. Reported methods of direct bonding to a thiophene backbone include metallation via thiophene,' or via 2 5 6 incorporation of bipyridyl " or bithiazoyl groups. Thiophenes can bond to transition metals by various bonding modes, including ^ '(S), r\2-, n 4 - or r)5-coordination,7 and metal-carbon bonding that is most frequently at an a position (Chart 4-1). C h a r t 4-1 V(S) M-Cbond Ru(II) has been reported to coordinate in an r\l(S) fashion to thiophenes; however, the bond is often weak. Chelation of a metal to a thienyl ligand containing another group can stabilize an r]'(S) bond. Examples of Ru(II) complexes coordinated by chelating thienyl ligands 8 9 are 35 and 36 that contain bis(thienyl)tetrazine and thienylbipyridine ligands, respectively (Chart 4-2). Bonding of metals to a thiophene via a M - C bond may occur by the reaction of a thiophene ring with metal complexes possessing vacant coordination sites,10"12 or by conversion from V(S) coordinated complexes, as has been demonstrated for Re(II)- and Ru(II)-thienyl 62 complexes.1 3 , 1 4 Scheme 4-1 shows an example of conversion from a Re(II) sulfur-bound complex (37) to the carbon-bound complex (38).14 Reversible switching between a M - C bonding mode (39) and an V(S) coordination mode (40) via acid-base mediation has also been reported for the Ru(II) complex depicted in Scheme 4-2. 1 5 Chart 4-2 37 63 The bis(bipyridine) Ru(II) group [Ru(bpy)2 ] exhibits electron and energy transfer processes.16 Ru(bpy)3 2 + and Ru(bpy)2Ln2+ complexes commonly display a Ru(d7i)->bpy(7i*) metal-to-ligand charge transfer (MLCT) to generate a charge-separated state, which allows complexes of this type to be used as sensitizing dyes. A dye molecule adsorbed onto a large band gap, n-type semiconductor that is in contact with a reducing agent dissolved in an electrolyte solution comprises a dye-sensitized solar cell (DSSC), also called a Gratzel Ce l l . 1 7 ' 1 8 An example of a Gratzel cell is the system composed of the trimeric ruthenium dye complex RuL2(u-(CN)Ru(CN)L'2)2 (L = 2,2'-bipyridine-4,4'-dicarboxylic acid, L ' = 2,2'-bipyridine) deposited onto high surface-area titanium oxide and in contact with iodide/triiodide redox electrolyte (Figure 4-1).1 9 The dye is excited with light to generate the charge-separated state, and due to well-matched energies of the excited state and the Ti02 semiconductor, the excited electron is transferred to Ti02. Electrons are thus transferred to the negative electrode, and the dye is replenished of electrons by the redox couple to complete the circuit. The maximum conversion efficiency obtained to date for this system is 12%. semiconductor dye electrolyte counterelectrode — • e Figure 4-1 Schematic representation of a Gratzel cell displaying the energy levels of the cell components and the electronic circuit with cell voltage A V that is created with light absorption of the ruthenium dye complex. Adapted from Ref. 1 9 The approach that is used here to promote direct coordination of a metal to a conjugated backbone is to covalently anchor the metal group via P-phosphinothiophene ligands. An open site available on the metal is provided the opportunity of interacting with the thiophene chain. 64 Two open coordination sites are available for the Ru(bpy)2 group, allowing the possibility of bonding to a P-phosphinothiophene ligand via phosphine-sulfur (P,S) or phosphine-carbon (P,C) bidentate bonding modes (Figure 4-2). Reversible switching between sulfur and carbon bonding modes as demonstrated in Scheme 4-2 could be used to alter polythiophene chain properties. By directly bonding the Ru(II) center to a polythiophene chain, favorable metal-thienyl orbital overlap could result in a charge-separated state due to the M L C T transition, and positive charges could be induced onto the conjugated backbone. The result can be thought of as a model for a molecular wire that conducts upon light absorption, or a light harvesting device. The proposed mechanism is similar to the Gratzel cell shown in that a dye molecule is attached to a substrate where conduction is induced. In the case of polythiophenes, conduction normally occurs with p-doping; therefore, electrons must be removed from the conjugated backbone. A simple schematic (Scheme 4-3) suggests the electron transfer process that could create positive charges, or holes, along an oligothiophene (Tn) chain via a Ru(bpy)2 sensitizing group. Figure 4-2 Two possible bidentate bonding modes of a P-phosphinothiophene chain attached to a Ru(bpy) 2 2 + group, P,S and P, C. P,S bonding mode P, C bonding mode Scheme 4-3 bpy TC* —j bpy TC* - — I Ru dn T, n TC TC 65 Chart 4-3 RuPT3-P,5'(41) RuPMeT3-P,5(42) RuPMe2T3-P,5'(43) RuPhex2T5-.P)S,(44) Chart 4-4 Cl2 H 25 RuPDo2T5-P,5'(45) ^ f l 1(PF 6) C 1 2 H ; RuPBr2Do2T5-,P)S(46) Ru(bpy)2 groups were coordinated via the phosphine and thienyl sulfur to p-(phosphino)terthiophene ligands PT 3 (13), PMeT3 (14), and PMe 2 T 3 (15) to yield Ru(II)-P(lS complexes 41-43 (Chart 4-3). p-(Phosphino)pentathiophene ligands Phex2T5 (21), PDo 2T 5 (22), and PBr 2Do 2T 5 (27) were used to prepare Ru(II)-.P,S complexes 44-46 (Chart 4-4). The effects of the P,S bonding mode on the structural conformation of the thiophene chains, the thienyl redox potentials, and the thienyl rc—>rc* transitions were investigated. Density functional theory calculations were carried out for 41 to estimate the energies and localization of the HOMO and L U M O orbitals. Methyl substituents attached to the P-(phosphino)terthiophene ligands allow observation of the effect of electron-donating substituents on the conformational and electronic properties of the complexes. The electrochemical and spectroscopic properties of the Ru(bpy) 2 2 + group bound to the p-phosphinothiophene ligands in the P,S bonding mode were also studied. Complexes 41-46 react with base, resulting in Ru(II) complexes where a thienyl carbon is bonded to the metal (P,C bonding). The characterization of these Ru(II)-P,C complexes is 66 discussed in Chapter 6. The Ru(II)-P,C complexes revert back to the P,S bonding mode by reaction with acid, and demonstrate an example of metal-thienyl bonding modes that are reversibly switchable. 4.2 Experimental 4.2.1 General Experimental A l l reactions were performed using standard Schlenk techniques with dry solvents under nitrogen. The synthesis and purification of ligands PT3 (13), PMeT3 (14), and PMe2T3 (15) 90 have been reported. The preparation of complex RuPTs-.P.S (41) was previously reported by Dr. O. Clot. 2 1 A l l other reagents were purchased from Aldrich or Strem Chemicals and were 1 *j 1 used as received. H and P N M R experiments were performed on either a Bruker AV-300 or a Bruker AV-400 Spectrometer, and spectra were referenced to residual solvent ('H) or external 85% H3PO4 ( 3 1P). Absorption spectra were obtained with a Cary 5000 in HPLC grade CH 2 C1 2 . Emission spectra were obtained with a Cary Eclipse in HPLC grade CH 2 C1 2 or C H 3 C N , and emission slits were set to 20 nm. Cyclic voltammetry experiments were carried out with a Pine AFCBP1 bipotentiostat using a platinum disk working electrode, platinum coil wire counter electrode, and a silver wire reference electrode.. Either decamethylferrocene (-0.12 vs. SCE) or ferrocene (0.41 vs. SCE) was used as an internal reference to calibrate the measured potentials with respect to saturated calomel electrode (SCE). The supporting electrolyte was 0.1 M [(«-Bu)4N]PF6 that was purified by recrystallizing three times from ethanol and drying for three days at 90°C under vacuum. The lifetime measurement was carried out by Prof. C. Bohne at the University of Victoria by exciting the sample at 480 nm in 7 x 7 mm Suprasil cells at 20 ± 2°C with a Coherent Infinity OPO tunable laser, using the previously described laser flash photolysis system.22 Deoxygenated samples were dissolved in C H 3 C N to achieve absorbences between 0.3 and 0.5 (1 = 7 mm) at 480 nm. Emission decays were measured at fixed wavelengths, averaging at least 5 kinetic traces, and emission spectra were obtained by collecting data at fixed wavelengths and averaging the values between set time windows after the laser pulse. The voltage of the photomultiplier used to detect the emission signal was kept constant throughout the collection of a spectrum. 67 4.2.2 Synthesis [Bis(2,2'-bipyridyl)(3 ~diphenylphosphino-5-methyl-2,2 ':5 '2 "-terthiophene-P,S)ruthenium(II)] [bis(hexafluorophosphate)] (RuPMeT 3-P,S) (42) AgBF4 (230 mg, 1.15 mmol) was added to a deaerated solution of Ru(bpy)2Cl2-2H20 (300 mg, 0.576 mmol) in acetone (30 mL), stirred for 6 h, and filtered under nitrogen. To the red filtrate, PMeT3 (14) (272 mg, 0.610 mmol) was added and the mixture was heated to reflux for 18 h. The resulting solution was concentrated to 10 mL and precipitated by addition to a solution of N H 4 P F 6 (1.89 g, 11.6 mmol) in H 2 0 (100 mL). Recrystallization in ethanol-acetone (9:1) gave 42 as bright yellow crystals. Yield: 0.340 g (51%). *H N M R (300.1 MHz, CO(CD 3 ) 2 : 5 9.13 (d, J= 5.7 Hz, 1H), 8.89 (d, J= 5.2 Hz, 1H), 8.78 (d, J= 8.2 Hz, 1H), 8.69 (d, J= 7.5 Hz, 1H), 8.68 (d, J = 8.1 Hz, 1H), 8.60 (d, J = 8.2 Hz, 1H), 8.28-8.22 (m, 3H), 8.09 (t, J = 8.0 Hz, 1H), 7.96 (d, J= 5.7 Hz, 1H), 7.83-7.81 (m, 1H), 7.69 (t, J = 6.9 Hz, 1H), 7.63-7.55 (m, 3H), 7.50 (d, J= 5.0 Hz, 2H), 7.45-7.32 (m, 6H), 7.22-7.16 (m, 3H), 708 (dd, J= 4.8 Hz, J= 3.9 Hz, 1H), 6.95-6.88 (m, 3H), 6.78-6.77 (m, 1H), 1.48 (s, 3H). 3 1 P{'H} N M R (121.5 MHz, CO(CD 3) 2): 5 28.5 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Anal. C 4 5H 3 5F 1 2 N4S 3 P 3 Ru requires C, 47.00; H, 3.07. found: C, 46.60; H, 3.05%. [Bis(2,2 -bipyridyl)(5,5 "-dimethyl-3 -diphenylphosphino-2,2 ':5 ',2 "-terthiophene-P,S) ruthenium(II)][bis(hexafluorophosphate)] (RuPM^T^-.P.S') (43) AgBF4 (406 mg, 2.38 mmol) was added to a deaerated solution of Ru(bpy)2Cl2-2H20 (530 mg, 1.02 mmol) in acetone (50 mL), stirred for 18 h, and filtered under nitrogen. To the red filtrate, PMe2T3 (15) (497 mg, 1.08 mmol) was added and the mixture was heated to reflux for 24 h. The resulting solution was condensed to 10 mL and precipitated by addition to a solution of N H 4 P F 6 (1.89 g, 11.6 mmol) in H 2 0 (100 mL). Recrystallization in ethanol-acetone (9:1) gave 43 as bright, yellow crystals. Yield: 0.624 g (52%). ! H N M R (300.1 MHz, CO(CD 3) 2): 5 9.14 (d, J = 5.7 Hz, 1H), 8.84 (d, J = 5.0 Hz, 1H), 8.79 (d, J = 8.7 Hz, 1H), 8.69 (q, 2H), 8.61 (d, J = 8.4 Hz, 1H), 8.30-8.23 (m, 3H), 8.10 (t, J = 8.0 Hz, 1H), 7.98 (d, J = 5.4 Hz, 1H), 7.84-7.82 (m, 1H), 7.70 (t, J = 6.8 Hz, 1H), 7.63-7.49 (m, 4H), 7.46-7.35 (m, 5H), 7.23-7.17 (m, 3H), 7.12 (d, J = 3.3 Hz, 1H), 6.94-6.88 (m, 2H), 6.79-6.77 (m, 2H), 6.78 (s, 1H), 2.44 (s, 3H), 1.48 (s, 3H); 3 1 P N M R (121.5 MHz, CO(CD 3) 2): 5 29.0 (s), -143.0 (sep, JP¥ = 708 Hz, PF 6). Anal. C46H 37F 1 2N4S 3P 3Ru requires C, 47.47; H, 3.20; N , 4.81. found: C, 47.54; H, 3.38; N , 4.47%. 68 [Bis (2,2 -bipyridyl)(3,3 ""-dihexyl-3 "-diphenylphosphino-2,2': 5'2": 5 "2 "': 5 '"2 ""-pentathiophene-P,S)ruthenium(II)][bis(hexafluorophosphate)] (RuPhex2T5-7:,,,S) (44) AgBF4 (73 mg, 0.375 mmol) was added to a deaerated solution of R u ^ p y ^ C h ^ ^ O (97 g, 0.186 mmol) in acetone (25 mL), stirred for 3 h and filtered under nitrogen. The resulting red filtrate was added to a suspension of Phex 2 T 5 (21) (143 mg, 0.187 mmol) in deaerated acetone (25 mL) and the mixture was heated at reflux for 20 h. The resulting dark red solution was condensed to 10 mL, added dropwise by pipette into a solution of N H 4 P F 6 (15 g, 92 mmol) in H 2 0 (300 mL) and the resulting orange precipitate was recovered. The precipitate was dissolved in acetone, remaining solids were filtered off, and the solvent was removed. Recrystallization in ethanol-acetone (9:1) gave 44 as bright orange crystals. Yield: 150 mg (55%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 8 9.25 (d, J= 5.6 Hz, 1H), 9.10 (d, J = 5.2 Hz, 1H), 8.71 (d,J = 8.0 Hz, 1H), 8.63 (d ,J= 8.0 Hz, 1H), 8.59 (d, J= 8.4 Hz, 1H), 8.56 (d, J= 8.8 Hz, 1H), 8.31 (dt, J= 1.2 Hz, J= 7.8 Hz, 1H), 8.25 (dt, J= 1.2 Hz, J= 7.8 Hz, 1H), 8.05 (dt, J = 1.2 Hz, .7=7.8 Hz, 1H), 7.98 (dt, J= 1.2 Hz, J= 8.0 Hz, 1H), 7.81 (d, J= 5.6 Hz, 1H), 7.75 (dt, J= 1.2 Hz, J = 6.8 Hz, 1H), 7.63 (d, J = 7.6 Hz, 1H), 7.61 (d, J = 8.8 Hz, 1H), 7.54 (t, J = 6.6 Hz, 1H), 7.49 (dd, J = 1.6 Hz, J = 7.2 Hz, 1H), 7.47 (d, J= 3.6 Hz, 1H), 7.42-7.38 (m, 5H, phenyl), 7.30 (dt, J = 1.2 Hz, J= 6.6 Hz, 1H), 7.27-7.22 (m, 3H), 7.18 (d, J= 5.2 Hz, 1H), 7.14 (d, J= 3.6 Hz, 1H), 7.07 (d, J= 3.2 Hz, 1H), 7.04 (d, J= 5.2 Hz, 1H), 7.02-6.95 (m, 4H), 6.74 (d, JPH= 5.2 Hz, 1H), 2.75 (t, J= 7.8 Hz, 2H), 2.22-2.06 (m, 2H), 1.61 (q, J = 7.2 Hz, 2H), 1.45 (m, 2H), 1.35-1.25 (m, 12 H), 0.86 (t, J= 6.8 Hz, 3H), 0.81 (t, J = 6.8 Hz, 3H). 3 1 P{'H} N M R (162.0 MHz, CO(CD 3 ) 2 ): 8 27.2 (s), -143.0 (sep, J P F = 708 Hz, PF 6). [Bis(2,2 -bipyridyl)(3,3 ""-didodecyl-3"-diphenylphosphino-2,2':5',2":5"2"':5"',2""-pentathiophene-P,S)ruthenium(II)][bis(hexafluorophosphate)] (RuPDo2T5-/>,5) (45) A g B F 4 (1.79 g, 9.18 mmol) was added to a deaerated solution of Ru(bpy) 2Cl 2-2H 20 (2.39 g, 4.59 mmol) in acetone (150 mL), stirred for 3 h and filtered under nitrogen. The resulting red filtrate was added to a suspension of P D 0 2 T 5 (22) (4.29 g, 4.60 mmol) in deaerated acetone (25 mL) and the mixture was heated at reflux for 20 h. The solution was condensed to 10 mL, added dropwise by pipette into a solution of N H 4 P F 6 (15 g, 92 mmol) in H 2 O (300 mL) and the resulting orange precipitate was recovered. The precipitate was dissolved in acetone, remaining solids were filtered off, and the solvent was removed. Recrystallization in ethanol-acetone (9:1) gave 45 as bright orange crystals. Yield: 3.44 g (46%). *H N M R (400.1 MHz, CO(CD 3) 2): 8 9.25 (d ,J=6.0Hz, 1H), 9.09 (d, 7= 5.6 Hz, 1H), 8.71 (d ,J=7.6Hz, 1H), 8.63 (d,J=7.6Hz, 1H), 8.58 (d, J = 8.0 Hz, 1H), 8.55 (d, J = 8.4 Hz, 1H), 8.31 (t, J = 8.0 Hz, 1H), 8.25 (t, J = 8.0 69 Hz, 1H), 8.05 (t,J= 7.8 Hz, 1H), 7.80 (t, J= 8.0 Hz, 1H), 7.81 (d, J= 5.6 Hz, 1H), 7.75 (t, J = 6.6 Hz, 1H), 7.64-7.60 (m, 2H), 7.55-7.48 (m, 2H), 7.46 (d, J = 4.0 Hz, 1H), 7.42-7.38 (m, 5H, phenyl), 7.30 (t, J= 6.8 Hz, 1H), 7.25-7.22 (m, 3H), 7.18 (d, J= 5.2 Hz, 1H), 7.14 (d, J= 4.0 Hz, 1H)„ 7.08 (d, J= 2.8 Hz, 1H), 7.04 (d, J= 4.8 Hz, 1H), 6.98-6.95 (m, 4H), 6.74 (d, JPH = 4.4 Hz, 1H), 2.75 (t, J = 7.8 Hz, 2H), 2.22-2.06 (m, 2H), 1.61 (q, J = 7.4 Hz, 2H), 1.45 (m, 2H), 1.28-1.25 (m, 36 H), 0.89-0.83 (m, 6H). 3 1 P{'H} N M R (162.0 MHz, CO(CD 3) 2): 27.7 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Anal. C 7 6H g 5F 1 2 N4S 5 P 3 Ru requires C, 55.77; H, 5.23; N , 3.42. found: C, 55.38; H , 5.40; N , 3.31%. [Bis(2,2'-bipyridyl)(5,5""-dibromo-3,3""-didodecyl-3 "-diphenylphosphino-2,2 ':5 ',2 ":5 "2 '":5 '"2 ""-pentathiophene-P,S)ruthenium(II)] [bis(hexafluorophosphate)] (RuPBr 2 Do 2 T 5 -P,S) (46) A g B F 4 (99.9 mg, 0.513 mmol) was added to a deaerated solution of Ru(bpy)2Ci2-2H20 (134 mg, 0.256 mmol) in acetone (150 mL), stirred for 3 h and filtered under nitrogen. The resulting red filtrate was added to a suspension of P B r D o 2 T 5 (27) (280 mg, 0.257 mmol) in deaerated acetone (25 mL) and the mixture was heated at reflux for 20 h. The solution was condensed to 10 mL, added dropwise by pipette into a.solution of NH4PF6 (15 g, 92 mmol) in H 2 0 (300 mL) and the resulting orange precipitate was recovered. The precipitate was dissolved in acetone, remaining solids were filtered off, and the solvent was removed. Recrystallization in ethanol-acetone (9:1) gave 46 as pale orange crystals. Yield: 110 mg (24%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 5 9.18 (d, J = 5.6 Hz, 1H), 9.12 (d, J = 5."6 Hz, 1H), 8.75 (d, J = 7.6 Hz, 1H), 8.64 (d, J = 8.0 Hz, 2H), 8.58 (d, J = 7.6 Hz, 1H), 8.33 (t, J = 7.6 Hz, 1H), 8.27 (t, J= 8.0 Hz, 1H), 8.09 (t, J = 7.6 Hz, 2H), 7.85 (d, J = 6.0 Hz, 1H), 7.74 (t, J = 6.6 Hz, 1H), 7.64-7.58 (m, 3H), 7.54 (d,J= 3.6 Hz, 1H), 7.51 (dd, J= 2.0 Hz, J = 7.6 Hz 1H), 7.43-7.38 (m, 5H, phenyl), 7.34 (t, J = 6.8 Hz, 1H), 7.22 (dt, J = 2.8 Hz, J= 7.6 Hz, 2H), 7.17 (d, J = 4.0 Hz, 2H), 7.14-7.12 (m, 1H), 7.13 (s, 1H), 7.04 (d, J = 4.0 Hz, 1H), 6.99 (d, 7=7.2 Hz, 1H), 6.96 (d, J= 7.2 Hz, 1H), 6.80 (s 1H), 2.72 (t, J = 7.8 Hz, 2H), 2.23-2.16 (m, 2H), 1.61 (q, J = 7.6 Hz, 2H), 1.48-1.38 (m, 2H), 1.28-1.26 (m, 36 H), 0.89-0.84 (m, 6H). 3 1 P{ 1 H} N M R (162.0 MHz, CO(CD 3 ) 2 ): 27.1 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Anal. CveFfoF^N^PsB^Ru requires C, 50.86; H, 4.66; N , 3.12. found: C, 51.18; H, 4.64; N , 3.02%. 70 4.2.3 X-ray Crystallographic Analysis Suitable crystals of 41, 43, and 45 were obtained by slow diffusion of hexanes into acetone solution, and were mounted on a glass fiber. The data were collected at -100.0 ± 0.1 °C and the X-ray source in all cases was graphite monochromated M o - K a radiation. The structures were solved by direct methods23 and expanded using Fourier techniques.24 A l l calculations were performed using the teXsan 2 5 crystallographic software package of Molecular Structure Corporation and SHELXL-97 . 2 6 Hydrogen atoms were placed in calculated positions. The crystal structures were obtained and analysis carried out by Dr. B. Patrick at UBC. Data for 41 were collected to a maximum 29 of 55.7° on a Rigaku/ADSC CCD area detector in a series of two scans using 0.50° oscillations with 27.0 second exposures. Data were collected and processed using the d*TREK program.27 The data were corrected for Lorentz and polarization effects. The molecule crystallizes with one molecule of acetone in the asymmetric unit. One thiophene ring is disordered and was modeled in two orientations. The atoms of the minor fragment, as well as the disordered carbons of the major fragment were refined with isotropic thermal parameters. Hydrogen atoms were included in calculated positions. Data for 43 were collected to a maximum 29 of 55.8° on a Bruker X8 diffractometer in a series of two scans using 0.50° oscillations with 12.0 second exposures. Data were collected using the Bruker SAINT 2 8 software package and corrected for absorption effects using the multi-scan technique (SADABS 2 9 ) . The data were corrected for Lorentz and polarization effects. The molecule crystallizes with two molecules of acetone in the asymmetric unit. One PF6~ anion is disordered and was modeled in two orientations with relative populations of 0.88 and 0.12. The atoms of the minor fragment were refined isotropically, all other non-hydrogen atoms were refined anisotropically, and all hydrogen atoms were included in calculated positions. Data for 45 were collected to a maximum 29 of 53.5° on a Rigaku/ADSC CCD area detector in a series of two scans using 0.50° oscillations with 53.0 second exposures. Data were collected and processed using the d*TREK program.27 The data were corrected for Lorentz and polarization effects. The crystal used for analysis contained one [PFg]- and one [BF4]~ counterion each. Both alkyl chains were disordered and were modeled in two orientations. In one case the disorder originates with the orientation of the thiophene rings, with 0.58 and 0.42 populations of the major and minor fragments. The second disordered alkyl chain was modeled with isotropic thermal parameters, to relative populations of 0.77 and 0.23 for the major and minor fragments. Hydrogen atoms were included in calculated positions. Appendix 1 contains 71 the crystallographic data for the crystal structures of 41, 43, and 45. Ci f files are available online. 3 0 ' 3 1 4.2.4 Density Functional Theory Calculations Density functional theory calculations were carried out using the Gaussian 03 Package,32 and the B 3 L Y P 3 3 ' 3 4 method was used to optimize the geometry. The 6-31 G* basis set was used to model non-metallic atoms. A L A N L 2 D Z pseudopotential is used for inner shell calculations of the metal center, with the corresponding L A N L 4 basis set for the pseudopotential. To analyze the chemical bonding, natural bonding orbital (NBO) analysis was used. A l l calculations were carried out by Y . Zhang from the research group of Prof. A . Wang at UBC. Gaussian View software was used to generate depictions of the orbitals. 4.3 Results 4.3.1 Synthesis and Crystallographic Studies Ru(II)-(phosphino)terthiophene-.P,,S' complexes 41-43 were prepared by reaction of Ru(bpy)2Cl2-2H20 with AgBF4 and complexation with the appropriate ligand. The products were metathesized to the [PFe]~ salts and recrystallized in ethanol-acetone to give the air-stable complexes in good yield (Scheme 4-4). Scheme 4-4 3) 20 eq NH4PF6 RuPT3-/',5'(41) R = R' = H (79%) RuPMeT3-/J,5(42) R = Me, R' = H (51%) RuPMe2T3-/5,5(43) R = R' = Me (52%) 72 (a) (b) Figure 4-3 (a) ORTEP view of RuPTa-P.S (41) (conformation A) and (b) ORTEP view of RuPMe2T3-P,£' (43). Hydrogen atoms, counterions, and occluded solvents are omitted for clarity, and thermal ellipsoids are drawn at 50% probability. Table 4-1 Selected interatomic distances (A) and angles (°) for 41 and 43 RuPT3-JP,5(41) RuPMe 2T 3 -^5(43) Bond length /A Bond length /A Rui-Si 2.3640(8) C33-C34 1.340(5) Rui-Si 2.3621(6) C 3 3 - C 3 4 1.347(3) Rui-Pi 2.3215(7) C34-C35 1.426(5) Rui-Pi 2.3397(6) C34-C35 1.433(3) S,-C 33 1.744(3) C35"C36 1.359(5) S,-C 3 3 1.764(2) C 3 5 - C 3 6 1.3 5 6(3) S,-C 3 6 1.751(3) C36"C37 1.452(4) S,-C 3 6 1.750(2) C36-C37 1.449(3) Torsion Angle 1° Torsion Angle 1° S1-C36-C37-S2 -147.02(19) S1-C36-C37-S2 -150.71(12) S2-C40-C41-S3 146.0(2) S2-C40-C41-S3 165.48(12) S2"C40-C4i-S3b -41.3(5) The solid-state structures of 41 and 43 were established by the X-ray crystallography of crystals grown from slow diffusion of hexanes into a solution of the complex in acetone. The structures of 41 and 43 are shown in Figure 4-3, and selected bond lengths and torsion angles are collected in Table 4-1. The (3-(phosphino)terthiophene ligands bind to Ru(II) in a P,S bonding mode, with the metal coordinated to the Si thiophene ring in an rj^S) fashion, resulting in the 73 formation in each case of a six-membered ring. We have observed this bonding mode previously in other ruthenium and palladium complexes.20'35 The Rui-Si bond lengths of 41 and 43 are 2.3640(8) A and 2.3621(6) A , which are shorter than the bond lengths observed for other hemilabile (2.390-2.397 A ) 1 ' 3 6 and inert (2.380 A ) 9 S-bound Ru(II)-thienyl complexes. The plane of the bound thiophene is tilted from the Ru-S bond at angles of 58.3° (41) and 53.6° (43), which minimizes an unfavorable interaction between a sulfur lone pair occupying a n-orbital and the filled metal d y z orbital. The sulfur atom is subsequently hybridized to an sp -type hybridization, removing the lone pair with conjugation from the ring. This is indicated by S1-C33 and S i - C 3 6 bond lengths for 41 and 43 (> 1.744(3) A) that approach C-S single bond lengths38 and are elongated compared with calculated C-S bond lengths for T3 (1.7206 A and 1.7351 A respectively).39 The S1-C36-C37-S2 and S2-C40-C41-S3 interannular torsion angles of 41 for conformation A (147.02(19)°, 146.0(2)°) are close to those calculated for T 3 in solution (147.6°), 3 9 while the S2-C4o-C4i-S3b angle of conformation B (-41.3(5)°) is flipped into a syn arrangement, but is twisted approximately the same amount from planarity. The corresponding angles of 43 (150.71(12)°, 165.48(12)°) indicate increased coplanarity and therefore greater rc-orbital overlap of adjacent thienyl rings compared to 41. In solid-state molecular structures containing aromatic rings, inter-ring separations of 3.3-3.8 A are indicative of molecular interactions due to 7i-stacking.40'41 There is evidence for weak 7i-stacking interactions between the tilted Si thienyl and adjacent N i or N2 pyridyl rings of 41 (plane-to-plane distance between centroids = 3.708 A) and 43 (3.675 A) , and the N4 pyridyl and C21 phenyl rings of both 41 (3.777 A) and 43 (3.580 A) . These intramolecular 71-stacking interactions could promote the preferential crystallization of the diastereomers observed in the crystal structures of 41 and 43, in which the Si thienyl ring is tilted towards the Ni or N2 pyridyl ring rather than towards the edge of the N 4 pyridyl ring, and likely enhances the stability of the complexes. Single peaks are observed in the 3 1 P N M R spectra, also suggesting no evidence for different complex diastereomers in solution. Intermolecular 7x-stacking between rings is observed for 43 but not for 41. The crystal structure of 43 (Figure 4-4) shows intermolecular Si and S3' thienyl rings that exhibit a plane-to-plane distance between centroids of 3.947 A , where closest distances between planes are within range of 71-stacking interactions (>3.583 A). Sulfur atoms of stacked rings are aligned anti to one another. 74 Figure 4-4 A portion of the unit cell of RuPMe 2T 3-P, 4S' (43) viewed normal to the 010 plane. Lines are drawn between thienyl groups that are stacked relative to one other. Hydrogen atoms, counterions, and occluded solvent have been removed for clarity, and thermal ellipsoids are drawn at 50% probability. Ru(II)-(phosphino)pentathiophene-P,5' complexes 44-46 were prepared by a similar procedure as for 41-43 via reactions with the appropriate ligands (Scheme 4-5). Repeated crystallizations of 45 in ethanol containing minimal acetone provided brightly coloured, orange, needle-shaped crystals suitable for X-ray analysis. The crystal used to obtain the X-ray crystal structure of 45 (Figure 4-5) contained one [PF6]~ counterion and one [BF4]~ counterion per Ru(II) complex. Compared to 41 and 43, the Ru-S bond of 45 is shorter (2.3578(14) A, Table 4-2), while the tilt angle between the plane of the thiophene ring and the Ru-S bond (58.4°) is similar. The S 2 -C 5 (1.762(6) A) bond is elongated by 0.028 A compared to the corresponding bonds of the inner thiophene ring calculated for T 3 (1.7342 A).39 The interannular torsion angle between the two bound thiophene rings (147.9(4)°) is very close to the corresponding torsion angle for 41 and expected torsion angles for T 5 . 4 2 ' 4 3 The disorder of the S 4 and S 5 rings of 45 is similar to the disorder observed in the crystal structure of T 5 , 4 4 but may also be due to the long dodecyl chains. Intramolecular TC-stacking for 45 is observed between the N 4 pyridyl and C 4 5 phenyl rings (3.520 A) and between the S 2 thienyl and N i pyridyl rings (3.671 A), and similarly to 41 and 43, there is a preference for the diastereomeric arrangement of the S 2 thienyl ring as depicted. Intermolecular re-stacking is not seen in the crystal structure, and is likely inhibited by the alignment of the dodecyl chains between the molecules that is observed. 75 Scheme 4-5 Ru(bpy)2Cl2-2H20 + 2 AgBF 4 RuPhex2T5-P,1S'(44) RuPDo2T s-P,5 (45) RuPBr2Do2T5-/>,S (46) Figure 4-5 X-ray crystal structure of RuPDo 2Ts-/ J ,,5' (45) (conformation B). Only the first carbon atoms of the dodecyl chains are shown, and the remaining carbon atoms of the chains, all hydrogen atoms and counterions are omitted for clarity. Thermal ellipsoids are shown at 50% probability. 76 Table 4-2 Selected interatomic distances (A) and angles (°) for RuPDo2T5-/ ,,S (45) Bond length /A Ru,-S 2 2.3578(14) S 2 -C 8 1.745(6) C 6 - C 7 1.433(8) Cg-C9 1.469(8) Rui-Pj 2.3404(17) C 5 - C 6 1.358(8) C 7 - C 8 1.353(8) S2-C5 1.762(6) Torsion Angle 1° S1-C4-C5-S2 32.0(6) S3-Cl2"Cl3-S4b 177.1(4) S2-C8-C9-S3 -147.9(4) S4a-Cl6a-Cl7a-S5a -24(3) S3-C]2-Ci3-S4a 127.9(5) S4b-Ci6b-Ci7b"S5b 50.2(19) 4.3.2 Density Functional Theory (DFT) Calculations DFT calculations were carried out for the simplest Ru(II)-(phosphino)terthiophene-.P)>S' complex, RuPTs-P.iS' (41), in order to predict the frontier orbital energies of the complex in the gas phase. The DFT calculations predict that the HOMO is localized on the metal and the terthiophene rings, with a large contribution from the thiophene rings to create a mixed state (Figure 4-6). The H O M O is separated by ~1 eV from the HOMO-1, and the HOMO-1 shows only a contribution from a non-coordinated thiophene ring. The HOMO-2 and the near-degenerate HOMO-3 and HOMO-4 display a greater degree of metal character. The near degenerate L U M O and LUMO+1 are both centered primarily on the bipyridyl rings with a small contribution from the metal center. Bipyridine groups are expected to have a low-lying L U M O level that enable bipyridyl ligands to act as electron acceptor groups.16 The LUMO+2 shows localization of the orbital on the terthiophene rings and the metal center. According to the calculations, the lowest-energy transition from the HOMO to the L U M O is a R11/T3—»Ru/bpy charge-transfer transition with an energy gap of 2.3783 eV. The calculations predict a surprisingly small extent of metal contribution to the HOMO, given that Ru(bpy)2Ln complexes generally exhibit a high-energy metal-based level, giving rise to a Ru—>bpy M L C T as the lowest energy transition. It is noted that the Ru-S bond length is predicted by DFT calculations to be 2.45 A , longer that in the crystal structure of 41 (2.3640(8) A ) and thus the calculations may be predicting less stabilization to the Ru(II) center from the thiophene rings via the Ru-S bond than is occurring. Electronic spectra are also normally measured in solution, and the electronic energy levels are altered by solvent effects that occur in solution. 77 -7 -8 >» cn I -9 LU -10 -11 v t> LUMO+4 LUMO+3 LUMO+2 LUMO+1 L U M O HOMO HOMO-1 HOMO-2 .3-3 HOMO-3/ HOMO-4 Figure 4-6 Ordering of the calculated HOMO and L U M O energies of RuPT 3 -P,5 (41), and depictions of frontier orbitals. General orientation of 41 is shown at bottom. 78 Table 4-3 Calculated HOMO and L U M O energies for RuPT3-P,.S (41) Orbital Energy /eV Orbital Energy /eV HOMO -9.6060 L U M O -7.2277 HOMO-1 -10.740 LUMO+1 -7.1760 HOMO-2 -11.018 LUMO+2 -6.7893 HOMO-3 -11.081 LUMO+3 -6.3583 HOMO-4 -11.114 LUMO+4 -6.1507 4.3.3 Cyclic Voltammetry The cyclic voltammograms of 41-43 are displayed in Figure 4-7a. Oxidation of the complexes results in irreversible peaks at high potentials. Addition of electron-donating methyl substituents progressively lowers the first oxidation wave from 1.48 V (41) to 1.41 V (43) (Table 4-4). First oxidation peaks are assigned as oxidation of the Ru(II) center based on comparisons to calculations using ligand electrochemical parameters to predict the Ru(III/II) redox couple.45 By this approach, it is assumed that all ligand contributions are additive. The expected redox potential (ECA\C) versus the normal hydrogen electrode (NHE) for Ru(II) complexes can be determined by: £ c a l c ( V ) = I £ L 4.1 where EL is the ligand electrochemical parameter for each ligand attached to the Ru(II) center, each representing one sixth of a contribution. Based on the EL values for 2,2'-bipyridine (0.259), triphenylphosphine (0.39) and tetrahydrothiophene (0.30), i i C aic is 1.77 V vs. NHE, or 1.48 V vs. SCE. The calculated value is very close to the potentials of the first oxidation waves observed for 41-43, and are assigned as ruthenium-based. The only prominent thiophene-based oxidation wave observed for 43-45 is an irreversible shoulder at 1.69 V for 43. Oxidation of the terthiophene chains is expected to be significantly anodically shifted due to direct coordination to an electron deficient Ru(II) center, as is observed. New return waves observed on the first and subsequent scans of 41 and 42, observed as minor peaks at 1.14 V and 1.12 V, could be due to oxidative electropolymerization or dimerization at the terthienyl a positions, since a new return wave is not observed for methyl-capped complex 43. The first bipyridyl reduction is irreversible for 41, which could indicate 79 interaction of the reduced bipyridyl group with the relatively electron-poor terthiophene group. Two reversible bipyridine reductions were observed at -1.24 and -1.46 V for 42 and 43; the reversibility is possibly due to stabilization of the reductions by donation of electron density or loss of reactivity at the a positions via methyl substitution. Thienyl-based reduction waves were observed at -1.79 and -1.81 V for 42 and 43, respectively, with a negative shift in potential upon addition of a second methyl substituent. 60 40 20 0 -20 < -40 c QJ fc 60 o 40 20 0 -20 -40 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Volts /V vs. SCE Figure 4-7 Cyclic voltammograms of (a) 41-43, and (b) 45 in C H 3 C N solution at 4.0 x 10"3 M concentrations, containing 0.1 M [(«-Bu)4N]PF6 supporting electrolyte, scan rate = 100 mV/s. Two irreversible oxidation waves were observed by cyclic voltammetry for 45. The first oxidation potential at 1.21 V is significantly lower than the ruthenium oxidation potentials in 41-43 and than the potential predicted using ligand electrochemical parameters. It is likely that the coordinating pentathiophene can no longer be approximated to tetrahydrothiophene. The first oxidation is still likely to be Ru(II)-based and the pentathiophene chain may be contributing 80 1 ' I ; (a) 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' RuPT3-P,5 (41) RuPMeT3-P,5 (42) ftS/ RuPMe2T3-P,5 (43) /jii 1 1 1 f ^ 1 t 1 i 1 i 1 i 1 i 1 i 1 1 1 '. (b) I > | i | i I i | i I i RuPDo2T5-P S (45) i . i 1 . ' 1 to stabilization of the charge, cathodically shifting the oxidation potential. There is a significant anodic shift of the thienyl-based oxidation wave of 45 (1.71 V) from the first potentials of P D 0 2 T 5 (22) (0.99 V), and D o 2 T 5 (24) (0.82 V), but is at a similar oxidation potential to that of 43, which is reasonable since the corresponding ligands values are also alike (Chapter 2). Reduction of 45 shows a bipyridyl reduction wave at -1.22 V that is slightly higher than for 42 and 43. The second quasi-reversible reduction of 45 (-1.36 V) is at a higher potential than would be expected for the second bipyridyl reduction compared to 42 and 43 and may be thienyl-based, as a pentathiophene group has a greater ability to accept an electron due to increased conjugation length. A fast process, often characteristic of desorption from the electrode surface, may explain the sharp return peak of the reduction wave. Table 4-4 Cyclic voltammetry data of Ru(II)-phosphinothiophene-P,iS' complexes3 Compound £i / 2 ,ox±0.01 V v s . SCE £l/2,red±0.01 V V S . SCE RuPT 3 -P ) 1S'(41) + 1.48b (Ru I H / I 1) -1.28 b (bpy0/") RuPMeT 3 - / J , 5 (42) + 1.44b (Ru I I I / H) -1.24 (bpy0/-) -1.46 (bpy-/2~) -1.79 (PMeT 3 0 / _ ) R u P M e 2 T 3 - P , S ( 4 3 ) + 1.41(sh)b (Ru I I I / H) -1.24 (bpy0/-) + 1.69(sh)b (PMe 2 T 3 + / 0 ) -1.46 (bpy" / 2_) -1.81 (PMe 2T 3 0 /~) RuPDo 2 T 5 -P , 5 (45) + 1.21 b(Ru I I I / n) -1.22 (bpy0/-) + 1.71(sh)b (PDo 2 T 5 + / 0 ) -1.36(bpy- / 2-) Measurements carried out in C H 3 C N solution containing 0.1 M [(«-Bu)4N]PF6 supporting electrolyte, irreversible wave, E p . 4.3.4 Electronic Spectroscopy and Charge-Transfer Correlations The absorption spectra of the Ru(II)-.P,lS' complexes 41-43 and 45 are shown in Figure 4-8. Bipyridyl n—»7T,* transitions occur between 280-282 nm (Table 4-5). Terthiophene n—>n* transitions are observed as shoulders at -320 nm for 41-43. In comparison, the pentathiophene 7X—>7t* transition is significantly red-shifted to 371 nm for 45. Phenyl 7 t - > 7 i * transitions are also expected and are likely obscured by other transitions for 41-43. The 323 nm shoulder observed for 45 is in a similar position to the. 340 nm transition observed for P D o 2 T 5 (22) and may 81 for 45 is in a similar position to the 340 nm transition observed for PDo 2T 5 (22) and may correspond to the phenyl transition. The Ru(d7t)—>bpy(7T*) M L C T transition shifts dramatically with extension of the thiophene chain, from -400 nm for 41-43, resulting in the yellow colour of the complex solids, to 465 nm for 45, yielding orange solids (Figure 4-9). 5 4 ° 3 ~2 2 2 w 1 h 300 400 i- RuPT3-P,5(41) RuPMeT3-P,5(42) • \ /1 R„PMe2T3-PJ5(43) y 1 A >J\ / \ RuPDo2T5-P,5 (45) W / \ 1 / \\\ 1 , 1 , i . i . i , 500 A,/nm 600 700 800 Figure 4-8 Solution absorption spectra of Ru(II)-P,S' complexes 41-43 and 45 in CH2CI2 solution. Figure 4-9 Crystals of (a) R u P T ^ S (41) and (b) RuPDo2T5-F,1S' (45), displaying the dramatic colour change with extension of the oligothiophene chain that is primarily due to shifting of the M L C T transition. 82 Table 4-5 Electronic spectroscopy data for Ru(II)-phosphinothiophene-.P,5' complexes Compound Solution Absorption3 Solid-State Absorption Emission11 A,m ax ^max /nm [e /IVT'cm"1] Xmax /nm /nm RuPT 3-P,S 2 80 (3.79 x 104), 3 20 (sh) 304,326,412 — (41) (1.83 x 104), 393 (1.80 x 104) RuPMeT3-P,5 2 82 (3.91 x 104),3 20(sh) 308. 329, 422 — (42) (1.97 x 104), 396 (1.89 x 104) RuPMe2T3-JP,1S 2 82 (4.01 x 104), 3 20 (sh) 305, 331,430 — (43) (2.11 x io4), 404 (1.95 x 104) RuPDozTs-P,^ 2 80 (4.27 x 104), 323 (sh) 281,320 (sh), 377,464 602 (45) (2.24 x i o 4 ) , 371 (2.07 x lO 4 ) , (x< 10ns)c 465 (2.68 x lO 4 ) Measurements carried out in CH2CI2 solution. bDegassed C H 3 C N solution. cLifetime determined from emission at 600 nm. Correlations can be made between electrochemical potentials and optical charge transfer energies for ruthenium bipyridine complexes.46 The observed Ru—>bpy M L C T transition, Eop, is related to the positive energy difference between the Ru(III/II) and bpy(0/-l) redox couples, AE(redox), by: Eop - a A£"(redox) + const. 4.2 where a is usually close to unity, and the constant term collects all the solvent and reorganizational energies of the CT excited state. Figure 4-10 shows the linear correlation for all Ru(II)-i>,5 complexes, which gives: Eop = 1.76 A£(redox) - 1.60 4.3 where sources of uncertainty are the irreversibility of the oxidations and that the oxidation value for 43 that is determined from a shoulder peak. The correlation suggests that M L C T transitions occur and that the first oxidation waves for 41-43 and 45 are ruthenium-based. 83 3.3 t*3 2.40 2.45 2.50 2.55 2.60 2.65 2.70 2.75 A£(redox) /eV Figure 4-10 Linear fit of Eop versus Ais(redox) for RufJI)-/3,!? complexes 41-43 and 45. Solid-state absorption spectra of 41-43 and 45 are also measured (Figure 4-11, Table 4-5). M L C T transitions red-shift 19-26 nm for 41-43 and 12 nm for 45 compared with solution spectra. Thienyl 7c->rc* transitions red shift 6-10 nm, while the bipyridyl T C - » T C * transition remains in the same position for 45, but red shifts ~20 nm for 41-43. Red shifts in the absorption spectra may be due to increased rc-orbital overlap due to stacking interactions in the crystal structures. Overall, smaller red shifts are observed for 8, and it is likely that the presence of the long dodecyl groups results in a disordered solid state upon drop casting. O , C 03 N O 2 R u P T ^ S (41) RuPMeT3-P,5 (42) RuPMe2T 3-P,5 (43) RuPDo2T5-JP,5'(45) 300 400 500 600 K /nm 700 800 Figure 4-11 Solid state absorption spectra of Ru(II)-phosphinothiophene-P,5' complexes 41-43 and 45 drop-cast from solution in acetone. The primary M L C T transition is normalized to the same absorbance intensities for all spectra. 84 Complexation of Ru(bpy)2 groups to P-phosphinothiophenes quenches thienyl 7x<—rc* fluorescence emission. The Ru(d7i)<—bpy(7t*) luminescence that is generally observed with M L C T excitation for complexes containing the Ru(bpy)22+ fragment is quenched for 41-43, possibly due to thermal population to a low-lying non-emissive Ru(II)-based energy level. Very weak luminescence from the M L C T state of 45, at 602 nm (approximate quantum yield of 0.01%), is shown in Figure 4-12. The luminescence lifetime of 45 was measured at 600 nm; however, the lifetime was shorter than the lower limit measurable with the instrumentation used. Thus, the lifetime of 45 is less than 10 ns. The short emission lifetime compared with [Ru(bpy)3] (T = 870 ns) and other [Ru(bpy)2(LL)J complexes, and the low quantum yield support the conclusion that either significant thermal population of a low-lying, non-emissive metal-centered (MC) state is occurring, or vibrational relaxation pathways are competitive with emission from the M L C T state. Figure 4-12 (a) Emission and excitation spectra of RuPDo2T5-.P,>S' (45) in deaerated C H 3 C N solution. Solution abs = 0.1 at the excitation wavelength and the solvent spectra have been subtracted for clarity, (b) Emission spectra of RuPDo2T5-P,5' (45) obtained at 12 ns, 37 ns, 90 ns and 143 ns average times after the incidence of the laser pulse. Very small amounts of a secondary product appear when the complexes are in solution for long periods of time, and more intense emission at -450 nm is observed at similar wavelengths and intensities to the emission of p-phosphinothiophenes. These products form more rapidly in chlorinated solvents such as C H C I 3 or CH2CI2 than in C H 3 C N or acetone. Since the M L C T states are expected to be the lowest energy states of the metal complexes,5 the appearance of higher-energy emission suggests that the ligand detaches from the metal. For 85 example, complex 45 displays a new luminescence band at 514 nm shortly after dissolution in C H 3 C N (Figure 4-13), and the excitation, emission spectra, and the higher intensity of this new emission matches that of PD02T5 (22) (Chapter 3). These secondary products appear to form in very small concentrations and are not easily observed by techniques other than emission measurements. A new product does not appear to form for 45 at higher concentrations (~25 31 mM) since after 11 days in CO(CD3)2, new peaks were not observed in the P N M R spectrum. Similar results were seen for 41-43, though the secondary products form less rapidly. 3 1 P N M R experiments showed that new products formed in solution disappeared completely with the addition of concentrated acid, to reform only the original Ru(II)-/J,.S complexes. These experiments suggest that there may be an equilibrium between the complexes and the unknown products. The observation of emission due to a very minor amount of new product that is forming illustrates how weakly 45 emits. Figure 4-13 Emission spectra of RUPD02T5 -PS (45) (a) at time = 0 h, and (b) at time = 0, 2.5, and 22 h, showing the formation of new emission in C H 3 C N . 4.3.5 Reversible Molecular Switching Ru(II)-phosphinothiophene-P,S' complexes 41-46 dissolved in methanol react with NaOH base under reflux conditions. A color change from yellow (41-43) or orange (44-46) to dark brown or dark red, respectively, indicates the onset of a cyclometallation reaction. The bonding mode of the thiophene ring is converted, or switched, from coordination via a phosphine and a 86 thienyl sulfur (P,S) to bonding via a phosphine and a thienyl carbon to give Ru(II)-phosphinothiophene-P, C complexes 47-52 (Chart 4-5). Switching of the bonding mode, essentially a deprotonation and cyclometallation reaction, does not proceed without heating. Complete reversion of Ru(II)-P,C complexes 47-52 to the Ru(II)-P,5 complexes occurs rapidly with the addition of HPF6 or HC1 at room temperature, concomitant with a color change from 31 deep brown-red to bright yellow-orange. Analysis of 47-52 by P N M R spectroscopy after the addition of acid indicated that this conversion occurred quantitatively with no side products. It was observed, as shown in Figure 4-14, that addition of HPF6 to Ru(II)-P,C complex 51 directly yields the pure Ru(II)-P,S complex 45. Shown in Scheme 4-6 is the general procedure for preparing the Ru(II)-P,C complexes and reversibly switching back to the Ru(II)-P,5' complexes. Detailed procedures and yields for the preparation of Ru(II)-P,C complexes 47-52, followed by characterization studies, are provided in Chapter 5. Chart 4-5 Br RuPT3-.P,C(47) R = R' = H RuPMeT3-.P,C (48) R = Me, R' = H RuPMe2T3-P,C(49) R = R'= Me RuPhex2T5-/5,C(50) R = C 6H 1 3 RuPDo2T5-F,C(51) R = C, 2H 2 5 RuPBr2Do2T5-P,C(52) Scheme 4-6 1 M NaOH /MeOH 65°C 41-46 (P,S) 47-52 (P,C) HPF6 or HC1 87 27.7 ppm RuPDo2T5-P,S(45) 1 1 . 47.8 ppm RuPDojT -P,C (51) HPF6 , 27.7 ppm Product 50 -50 -100 -150 ppm Figure 4-14 3 1 P N M R spectra of RuPDo 2 T 5 -P,S (45) and R U P D 0 2 T 5 - P , C (51). Addition of HPF6 (cone.) to a solution of 51 results in reversion to 45 (bottom) as observed by the reappearance of the peak at 5 27.7 and a dramatic color change from deep red to bright orange. 4.4 Discussion When the structural data of Ru(II)-7:>)5' complexes 41-43 and 45 are examined and compared, the 0.011-0.043 A elongation of the S-C bond lengths of the bound thiophene rings relative to T 3 3 9 indicates that the sulfur atom is partially or fully removed from conjugation, as is expected due to sp3 hybridization of a thienyl sulfur bound to a metal. Bond length changes of the C-C bonds in the bound ring are <0.01 A ; therefore, complexation of a thienyl sulfur is not shown to interrupt conjugation across all of the rings. Despite structural constraints imposed by metal complexation, such as large tilt angles (53.7-58.4°) between Ru(II) and the bound thiophene, and intramolecular rc-stacking between the phenyl and thienyl rings, the torsion angles between the adjacent rings of 41 are close in value to that of T 3 in solution (147.6°). Complex 43 has a more coplanar conformation, which is expected as a result of electron donation by the methyl substituents. The torsion angle between the bound rings of 45 is equivalent to that of terthiophene, while the exterior rings align syn and anti and with multiple 88 conformations that are similar to T 5 . These results suggest that oligothiophene chains are not conformational^ affected by metallation with Ru(bpy)2 2 + groups in the P,S bonding mode, and changes in properties observed with complexation are mainly electronic in nature. A decrease in electron density across the thiophene rings is indicated by a -35 nm blue shift of the thienyl n—>n* transitions of 41-43 and 45 compared to the terthiophene ligands 13-1520 and pentathiophene ligand 22 (Chapter 2), and is likely due to a combination of decreased electron density at the ring coordinated to a Ru(II) center and possibly near the complexed phosphine. Thienyl-based oxidation waves (>1.69 V) that are substantially higher in potential compared with ligands 13-15 (1.05-1.30 V ) 2 0 or 22 (0.99 V) indicate removal of electron density from the thiophene rings, which would contribute to larger thienyl n—>n* energy gaps, as observed by blue shifts in the absorption spectra. Complexation to Ru(II) also removes sufficient electron density to anodically shift the terthiophene-based reduction potentials of 42 and 43, indicated by the reversible third reduction waves observed at -1.79 V and -1.81 V , respectively (Table 4-4). The assignment of the third reduction potential as thiophene-based is in agreement with the DFT calculations of 41 predicting a thiophene-based LUMO+2. Cathodic shifts of the oxidation and reduction potentials reflect the change in electron density of the oligothiophene rings. P-Phosphinothiophenes directly bound to the Ru(II) center also interact with the orbitals of the attached Ru(bpy) 2 2 + group. Compared to the oxidation potential of [Ru(bpy)3] 2 + (1.01 V vs. SCE), 1 6 oxidation potentials are >0.40 V higher for the Ru(II)-P,5' complexes, as expected from calculations with ligand electrochemical parameters,45 and reflecting poorer electron donation from a thienyl sulfur and phosphine compared to bipyridine rings. The M L C T absorptions of 41-43 at 393-404 nm are -50 nm blue-shifted compared to [Ru(bpy)3] 2 + (452 nm). 1 6 Despite blue-shifting of the transitions, observation of the expected correlation between Eop and A£(redox) for 41-43 suggest assignment of Ru—>bpy M L C T transitions. DFT calculations for 41 predict a H O M O / L U M O gap of 2.38 eV that is lower in energy than the band edge of the M L C T transition observed (480 nm = 2.58 eV) and the difference between the first oxidation and reduction (2.76 eV). The electrochemical data is in agreement with the two, near-degenerate bipyridyl-based L U M O and LUMO+1 levels predicted by DFT calculations, while suggesting that the first oxidized level is primarily Ru(II)-based. DFT calculations predict a terthiophene-based HOMO, and the discrepancy may be due to a predicted Ru-S bond length that is longer than observed and also that calculations are carried out for the gas phase. DFT calculations do predict ruthenium/oligothiophene orbital mixing, and complex 45 displays 89 unexpected properties that may be due to orbital mixing, such as a stabilized (cathodically shifted) Ru(III/II) oxidation and an M L C T (465 nm) that is red shifted from that of [Ru(bpy)3]2+. Due to irreversibility of the first oxidation, it is difficult to definitively assign the oxidations via other experiments. 94- 94-It is generally accepted that in Ru(bpy)2L2 , Os(bpy)2L2 and related complexes, absorption is primarily due to a singlet-based M L C T state and emission is from a triplet-based M L C T state, with spin-orbit coupling mixing the singlet and triplet states.5,47"49 It is observed that Ru(II)-/J>,5 complexes 41-43 do not emit; therefore, the MLCT-based luminescence is quenched. Quenching of the M L C T emission is observed in systems containing phosphines due to destabilization of the M L C T state without destabilization of a low- lying, M C state.47,50 This results in a low barrier for energy transfer to the non-emissive M C state and is major deactivation route at room temperature. Weak emission is observed at 602 nm for the Ru(II)-pentathiophene-P.S complex 45 that is expected to have a stabilized M L C T * state according to the red-shifted transition energy; therefore, there may be a higher barrier to the M C state compared with 41-43. The observation of a substantially shorter lifetime (< 10 ns) for 45 than is normally observed for Ru(bpy) 2L n 2 + complexes is in accordance with the presence of an accessible non-emissive M C state; though, due to the possibility of mixing of the orbitals of Ru(II) and the pentathiophene chain, a modified M L C T excited state could also result in the shorter lifetime of the emission. It is evident that the C-H bond cleavage that occurs during the cyclometallation reaction involves the Ru(II) center. Formally, the cyclometallation reaction to a Ru(II)-/J>, C complex is deprotonation, and reversion to a Ru(II)-P,5 complex is protonation; however, oligothiophenes are not normally deprotonated by NaOH. Mo/Co catalysts are known to promote dehydrosulfurization of thienyl rings, and activation of the C-H bonds of thienyl rings has been reported for Ru, Re, and Rh complexes. 1 3 , 1 4 ' 5 1 , 5 2 In these studies, migration of a metal from sulfur to carbon on the ring is observed, and the suggested mechanism of C-H activation via the formation of a r|2-coordinated intermediate is supported. In the presence of base, the metal may migrate to the inner C-C double bond in the Ru(II)-.P,S complexes, and the removal of electron density from the ring is expected to promote deprotonation, providing a site on the thienyl ring for carbon-metal bonding. A diagram of the proposed mechanism for reversible switching of the Ru(II)-(phosphino)terthiophene complexes is presented in Scheme 4-7. 90 Scheme 4-7 From the point of view of incorporating materials of this type into real devices, one can think of a molecular wire composed of polythiophene with one or more Ru(bpy)2 2 + groups attached via the P,S bonding mode. Crystalline and drop-cast solid state samples of the Ru(II)-.P,S complexes display stability in air, and under U V and visible light, and intra- and intermolecular n-stacking of the bpy, phenyl, and thiophene rings may contribute towards complex stability. This characteristic is desirable for device fabrication. For use as a light-harvester, doping of polythiophene chains would have to be initiated by the Ru—>bpy M L C T , dependent on effective overlap of the Ru(II) and thienyl oxidation potentials. The cyclic voltammetry results, combined with the correlation studies, suggest that the Ru(II) oxidation potentials are lower than the thienyl oxidation potentials; therefore, an M L C T transition that effectively removes an electron from Ru(II) is not expected to p-dope the oligothiophene backbones. However, extension of the thienyl chains or addition of electron-rich groups could be used to overlap these potentials and promote doping. Switching of the complexes to a different bonding mode with different properties (Chapter 5) is interesting because this mechanism could be incorporated into a molecular device. If the conductivity of a molecular wire, doped via Ru(bpy) 2 2 + excitation or other means, is altered via a change in metal coordination, P,S to P, C molecular switching would provide a means to control current through a conjugated, organic system. 91 4.5 Conclusions Binding Ru(II) bis(bipyridyl) groups to P-phosphinothiophenes via a direct, metal-thiophene P,S bonding mode affects the electrochemical and spectroscopic properties of oligothiophenes. In solution, the Ru(II)-/J,5' complexes exhibit blue shifting of the thienyl T C — » T C * transitions and anodic shifts in the thienyl oxidation potentials that are electronic inductive effects of the bound Ru(II) metal. Direct metal-thiophene bonding to give the Ru(II)-7:>,5' complexes does not result in unfavorable conformations of thienyl backbones and rc-orbital overlap appears unhindered from the solid-state structures obtained. The properties of the Ru(bpy) 2 2 + groups are affected by coordination of the p-phosphinothiophene ligands and by the length of the thiophene chain. Coordination to the P-(phosphino)terthiophene ligands blue shifts the Ru drc->bpy rc* M L C T transition of 41-43 while the P-(phosphino)pentathiophene ligand of 45 stabilizes the Ru(II) center and red shifts the M L C T transition compared to [Ru(bpy)3J2+. MLCT-based emission is observed for 45; however, the observed emission is too weak for applications. Several factors encourage further investigation of these hybrid materials for molecular device applications. DFT calculations predict a mixed-orbital HOMO, though the lower Ru(II) oxidation potential compared to the thienyl oxidation potential suggests that these oligothiophenes would not be oxidatively doped via the M L C T transition. Hole creation along the conjugated backbone via the M L C T transition could occur for complexes with longer thiophene chains, to create a light-harvesting material. In addition, switching to the P,C bonding mode can be carried out with the addition of base, and all of the Ru(II)-.P,C complexes can be reverted to the P^S-bound form by the addition of acid. Therefore, the two bonding modes are reversibly switchable. The stability of the solid state complexes to air, the conductive properties of oligo- or polythiophenes, and the reversibly switchable bonding modes indicate that P-phosphinothiophene chains incorporating Ru(bpy)2 groups could also have potential as organic conductors, with direct coordination of Ru(II) acting as a handle to alter the electronic properties. 4.6 References (1) Weinberger, D. A . ; Higgins, T. B.; Mirkin, C. A. ; Stern, C. L. ; Liable-Sands, L. M . ; Rheingold, A . L . J . Am. Chem. Soc. 2001,123, 2503-2516. (2) Zhu, S. S.; Swager, T. M . Adv. Mater. 1996, 8, 497-500. 92 Zhu, S. S.; Kingsborough, R. P.; Swager, T. M . J. Mater. Chem. 1999, 9, 2123-2131. Trouillet, L. ; De Nicola, A. ; Guillerez, S. Chem. Mater. 2000,12, 1611-1621. Walters, K. A. ; Touillet, L.; Guillerez, S.; Schanze, K . S. Inorg. Chem. 2000, 39, 5496-5509. Wolf, M . O.; Wrighton, M . S. Chem. Mater. 1994, 6, 1526-1533. Angelici, R. J. NATO ASI Series 1998, Ser. 3, 89-127. Sarkar, B.; Kaim, W.; Klein, A. ; Schwederski, B.; Fiedler, J.; Duboc-Toia, C.; Lahiri, G. K. Inorg. Chem. 2003, 42, 6172-6174. Constable, E. C ; Henney, R. P. G.; Tocher, D. A . J. Chem. Soc, Dalton Trans. 1991, 2335-2347. Wang, R.; Groux, L. F.; Zargarian, D. J. Organomet. Chem. 2002, 660, 98-107. Morikita, T.; Hirano, M . ; Sasaki, A. ; Komiya, S. Inorg. Chim. Acta. 1999, 291, 341-354. Bianchini, C ; Jimenez, M . V. ; Meli, A. ; Moneti, S.; Vizza, F. J. Organomet. Chem. 1995,504,27-31. Benson, J. W.; Angelici, R. J. Inorg. Chem. 1993, 32, 1871-1874. Robertson, M . J.; White, C. J.; Angelici, R. J. J. Am. Chem. Soc. 1994,116, 5190-5195. Constable, E. C ; Dunne, S. J.; Rees, D. G. F.; Schmitt, C. X . Chem. Commun. 1996, 1169-1170. Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin Complexes; Academic Press: London, 1992. Gratzel, M . Energy Resources through Photochemistry and Catalysis; Academic Press: New York, 1983. Kalyanasundaram, K. ; Gratzel, M . Photosensitization and Photocatalysis Using Inorganic and Organometallic Compounds; Kluwer Academic Publishers: Dordrecht, 1993. O'Regan, B.; Graetzel, M . Nature 1991, 353, 737-740. Clot, O.; Wolf, M . O.; Patrick, B. O. J. Am. Chem. Soc. 2001,123, 9963-9973. Clot, O. n-Conjugated Materials Containing Transition Metals, Ph.D. Thesis; Department of Chemistry, University of British Columbia, Vancouver, Canada, 2001, 185 pp. Liao, Y . ; Bohne, C. J. Phys. Chem. 1996,100,134-143. Altomare, A. ; Burla, M . C ; Camalli, M . ; Cascarano, G. L . ; Giacovazzo, C ; Guagliardi, A. ; Moliterni, A . G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115-119. Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de Gelder, R.; Isreal, R.; Smits, J. M . M . "The DIRDIF-94 program system, Technical Report of the Crystallography Laboratory," University of Nijmegen, The Netherlands, 1994. teXsan Crystal Structure Analysis package: Molecular Structure Corporation, 1985 & 1992. SHELXL-97: Sheldrick, G. M . , University of Gottingen, Germany, 1997. 93 d*TREK Area Detector Software, Version 7.11: Molecular Structure Corporation, 2001. SAINT, Version 6.02: Bruker A X S Inc., Madison, Wisconsin, USA, 1999. SADABS Bruker Nonius area detector scaling and absorption correction, Version 2.05: Bruker A X S Inc., Madison, Wisconsin, USA. Moorlag, C ; Clot, O.; Wolf, M . O.; Patrick, B. O. Chem. Commun. 2002, 3028-3029. Moorlag, C ; Wolf, M . O.; Bohne, C ; Patrick, B. O. J. Am. Chem. Soc. 2005, 127, 6382-6393. Gaussian 03: M . J. Frisch, G. W. T., H . B. Schlegel, G. E. Scuseria,; M . A. Robb, J. R. C , J. A . Montgomery, Jr., T. Vreven,; K. N . Kudin, J. C. B., J. M . Millam, S. S. Iyengar, J. Tomasi,; V . Barone, B. M . , M . Cossi, G. Scalmani, N . Rega,; G. A . Petersson, H. N . , M . Hada, M . Ehara, K. Toyota,; R. Fukuda, J. PL, M . Ishida, T. Nakajima, Y . Honda, O. Kitao,; H . Nakai, M . K. , X . L i , J. E. Knox, H. P. Hratchian, J. B. Cross,; C. Adamo, J. J., R. Gomperts, R. E. Stratmann, O. Yazyev,; A . J. Austin, R. C , C. Pomelli, J. W. Ochterski, P. Y . Ayala,; K. Morokuma, G. A . V. , P. Salvador, J. J. Dannenberg,; V . G. Zakrzewski, S. D., A . D. Daniels, M . C. Strain,; O. Farkas, D. K . M . , A . D. Rabuck, K. Raghavachari,; J. B. Foresman, J. V . O., Q. Cui, A . G. Baboul, S. Clifford,; J. Cioslowski, B. B. S., G. Liu, A . Liashenko, P. Piskorz,; I. Komaromi, R. L. M . , D. J. Fox, T. Keith, M . A . Al-Laham,; C. Y. Peng, A . N . , M . Challacombe, P. M . W. Gil l , ; B. Johnson, W. C , M . W. Wong, C. Gonzalez, and J. A . Pople, Gaussian, Inc., Pittsburgh PA, 2003. Becke, A . D. J. Chem. Phys. 1993, 98, 5648-5652. Lee, C ; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter 1988, 37, 785-789. Clot, O.; Wolf, M . O.; Yap, G. P. A. ; Patrick, B. O. J. Chem. Soc, Dalton Trans. 2000, 2729-2737. Weinberger, D. A. ; Higgins, T. B.; Mirkin, C. A. ; Liable-Sands, L. M . ; Rheingold, A . L. Angew. Chem. Int. Ed. 1999, 38, 2565-2568. Harris, S. Polyhedron 1997,16, 3219-3233. Rozsondai, B.; Schultz, G.; Hargittai, I. J. Mol. Struct. 1981, 70, 309-310. DiCesare, N . ; Belletete, M . ; Marrano, C ; Leclerc, M . ; Durocher, G. J. Phys. Chem. A 1998,702,5142-5149. Curtis, M . D.; Cao, J.; Kampf, J. W. J. Am. Chem. Soc. 2004, 726, 4318-4328. Yamamoto, T.; Komarudin, D.; Maruyama, T.; Arai, M . ; Lee, B.-L. ; Suganuma, H.; Asakawa, N . ; Inoue, Y . ; Kubota, K.; Sasaki, S.; Fukuda, T.; Matsuda, H. J. Am. Chem. Soc. 1998,120, 2047-2058. Becker, R. S.; de Melo, J. S.; Macanita, A. L. ; Elisei, F. J. Phys. Chem. 1996, 100, 18683-18695. DiCesare, N . ; Belletete, M . ; Donat-Bouillud, A. ; Leclerc, M . ; Durocher, G. J. Lumin. 1999, 81, 111-125. Azumi, R.; Goto, M . ; Honda, K.; Matsumoto, M . J. Chem. Soc. Jpn. 2003, 76, 1561-1567. Lever, A. B. P. Inorg. Chem. 1990, 29, 1271-1285. Dodsworth, E. S.; Lever, A . B. P. Chem. Phys. Lett. 1986, 124, 152-158. 94 Caspar, J. V. ; Meyer, T. J. Inorg. Chem. 1983, 22, 2444-2453. Mamo, A. ; Juris, A. ; Calogero, G.; Campagna, S. Chem. Commun. 1996, 1225-1226. Kober, E. M . ; Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1984, 23, 2098-2104. Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, / 7, 3334-3341. Dong, L.; Duckett, S. B.; Ohman, K. F.; Jones, W. D. J. Am. Chem. Soc. 1992, 114, 151-160. Angelici, R. J. Organometallics 2001, 20, 1259-1275. 95 CHAPTER 5 Synthesis and Characterization of P,C-Bound Ru(ll) Complexes 5.1 Introduction Different modes of direct bonding of a metal to a thienyl backbone can be used to alter the properties of oligo- or polythiophenes. In the previous chapter, the direct bonding of the Ru(II) bis(bipyridine) group to P-phosphinothiophene chains via a phosphine and a thienyl sulfur atom, to yield P,S-bound Ru(II)-phosphinothiophene complexes, was discussed. These Ru(II)-/J,,5' complexes could be converted to a P,C bonding mode where a thienyl carbon is bonded to Ru(II). The P,C-bound Ru(II)-phosphinothiophene complexes 47-52 (Chart 5-1) that are described in this chapter are formed from the corresponding Ru(II)-/J,5 complexes 41-46, and can be switched back to the .PS-bound complexes with the addition of acid, as was described in Chapter 4. RuPT3-P,C(47) R=R'=H RuPMeT3-P,C(48) R=Me, R-H RuPMe2T3-P,C(49) R=R'=Me Chart 5-1 R RuPhex2T5-/>,C(50) R = C 6H, 3 RuPDo2T5-P,C(51) R = C 1 2 H 2 5 C 1 2 H 2 5 RuPBr2Do2T5-P,C(52) Compared to the Ru(II)-.P,S complexes, the Ru(II)-P,C complexes lack a proton and a [PF6]~ counterion, and carry a +1 rather than a +2 charge on the complex. The P-phosphinothiophene ligands are formally -1 charged anions, and are therefore electron rich ligands. Greater electron density of oligothiophene chains contributes to higher energies of the n orbitals and could result in orbital overlap with the d7i orbital of the Ru(II) center. Scheme 5-1 96 shows the expected shifts of the T n(7i) and Ru(d7i) orbital levels that would occur with conversion from a complex comprised of a less electron-rich oligothiophene to a complex with a more electron-rich oligothiophene. Increased electron donation to Ru(II) via the Ru-C bond would also be expected to result in a higher energy Ru(d7t) orbital, as depicted. The ideal system for light harvesting applications (Scheme 5-1, right side), would result in the Tn(7x) level shifting higher in energy than the Ru(d7i) level, so that electron transfer could occur between these levels due to the Ru(d7i)—>bpy(n*) M L C T transition. A change in bonding mode from the P,S to the P,C bonding mode could result in shifts of the orbital levels as depicted due to binding an electron rich (3-phosphinothiophene ligand, and this possibility is investigated by spectroscopic studies of the model complexes. Scheme 5-1 bpy n* — L Switch Bonding Mode bpy 7 i * — _ Ru dn T „ 71 P,S P,C The characterization of Ru(II)-/J, C complexes 47-52 is similar to that completed for the Ru(II)-/J>)5' complexes, and the physical properties of the complexes in the two different bonding modes can be compared. Density functional theory calculations were carried out for complex 47, predicting the energies and localization of the frontier orbitals. Changes in alignment of the thienyl rings in the solid state structures, thienyl redox potentials, and thienyl n—>n* transitions are observed. Also, the effect that the change in bonding mode from P,S to P,C has on the M L C T absorption and emission is studied. In contrast to the Ru(II)-/J,5' complexes, the Ru(II)-P, C complexes exhibit cathodically shifted and reversible oxidation potentials that enable a more 97 complete study of the electronic states of the complexes via electron paramagnetic resonance spectroscopy and spectroelectrochemistry techniques. 5.2 Experimental 5.2.1 General Experimental A l l reactions were performed using standard Schlenk techniques with dry solvents under nitrogen. The preparation of Ru(II)-P,5 complexes 41-46 was described in Chapter 4. A l l other reagents were purchased from Aldrich or Strem Chemicals and were used as received. ' H and 3 1 P N M R experiments were performed on either a Bruker AV-300 or a Bruker AV-400 Spectrometer, and spectra were referenced to residual solvent ('H) or external 85% H3PO4 ( 3 1P). Absorption spectra were obtained with a Cary 5000 in HPLC grade CH2CI2. Emission spectra were obtained with a Cary Eclipse in HPLC grade CH2CI2 or C H 3 C N , and emission slits were set to 20 nm for measurements on metal complexes. Cyclic voltammetry experiments were carried out on a Pine AFCBP1 bipotentiostat using a platinum disk working electrode, platinum coil wire counter electrode, and a silver wire reference electrode. Either decamethylferrocene (-0.12 vs. SCE) or ferrocene (0.41 vs. SCE) was used as an internal reference to calibrate the measured potentials with respect to saturated calomel electrode (SCE). The supporting electrolyte was 0.1 M [(«-Bu)4N]PF6 that was purified by recrystallizing three times from ethanol and drying for three days at 90°C under vacuum. The lifetime measurement was carried out by Prof. C. Bohne at the University of Victoria by exciting the sample at 480 nm in 7 x 7 mm Suprasil cells at 20 ± 2 °C with a Coherent Infinity OPO tunable laser, using the previously described laser flash photolysis system.1 Deoxygenated samples were dissolved in acetonitrile to achieve absorbances between 0.3 and 0.5 (1 = 7 mm) at 480 nm. Emission decays were measured at fixed wavelengths, averaging at least 5 kinetic traces, and emission spectra were obtained by collecting data at fixed wavelengths and averaging the values between set time windows after the laser pulse. The voltage on the photomultiplier used to detect the emission signal was kept constant throughout the collection of a spectrum. Electron paramagnetic resonance experiments were carried out in deaerated acetonitrile solution on a Bruker System ESP 300 equipped with a Bruker ER035M gaussmeter and an HP 5350B microwave counter, and experiments were conducted by Dr. B. Sarkar in the research group of Prof. W. Kaim at the University of Stuttgart, Germany. The magnetic field was provided by a superconducting magnet (Cryogenics Consultant), which generates fields up to 12 T. Spectroelectrochemical experiments were 98 performed in deaerated acetonitrile that was distilled prior to use, with 0.1 M [(«-Bu)4N]PF6 supporting electrolyte that was recrystallized twice from methanol/ethanol solution and dried at 100°C under vacuum for several days. Optical measurements were conducted in a thin-layer spectrochemical cell under semi-infinite diffusion conditions2 using a reflective platinum electrode and an optical fiber detector, a platinum sheet as the counter electrode, and a Ag/AgCl wire as the reference electrode with a film thickness of 0.5 mm to give a path length of 1.0 mm. Spectroelectrochemical measurements were conducted by the author in the research group of Prof. P. Bauerle at the University of Ulm, Germany. 5.2.2 Synthesis [Bis(2,2 -bipyridyl)(3 -diphenylphosphino- 2,2':5',2''-terthiopheno-P,C)ruthenium(II)] [hexafluorophosphate] (RuPT3-P,C) (47) NaOH (1.2 g, 30 mmol) was dissolved in deaerated MeOH (30 mL) to give a 1.0 M solution. RuPT^-P.S (41) (0.10 g, 0.088 mmol) was dissolved into the solution and stirred at reflux for 18 h. A color change from yellow to dark brown was observed. The solution was then cooled to room temperature, concentrated to 10 mL, and added dropwise by pipette into a solution of NH4PF6 (0.28 g, 1.76 mmol) in H 2 0 (15 mL) to form a brownish-black precipitate. The precipitate was isolated by filtration, washed with water (10 mL), then ether (15 mL), and recrystallized in EtOH-acetone (9:1) to give 47 as black, shiny crystals. Yield: 43.3 g (50%). ' H N M R (300.1 MHz, CO(CD 3 ) 2 ): 8 8.87 (d, J= 5.7 Hz, 1H), 8.57 (d, J= 8.1 Hz, 1H), 8.51 (d, J= 8.1 Hz, 1H), 8.47-8.42 (m, 3H), 8.07 (dd, J = 8.0 Hz, J = 1.9 Hz, 1H), 7.98-7.85 (m, 3H), 7.77-7.64 (m, 4H), 7.52-7.39 (m, 5H), 7.36 (dd, 7 = 4.8 Hz, J= 1.3 Hz, 1H), 7.26 (td, J = 6.8 Hz, J= 1.4 Hz, 1H), 7.18 (dd, J= 3.9 Hz, J= 1.3 Hz, 1H), 7.12 (td, J= 7.2 Hz, J= 1.4 Hz, 1H), 7.06-7.02 (m, 2H); 6.93-6.89 (m, 3H), 6.65 (d, J= 2.7 Hz, 1H), 6.52-6.47 (m, 2H), 6.38 (d, J = 8.1 Hz, 1H). 3 1 P{ ] H} N M R (121.5 MHz, CO(CD 3) 2): 8 46.7 (s), -143.0 (sep, XJ?? = 708 Hz, PF 6). Anal C44H 3 2 F 6 N 4 S 3 P 2 Ru-C 3 H60 requires C 53.86; H, 3.65; N , 5.35. found: C, 53.89; H, 3.41; N , 5.34%. [Bis(2,2 -bipyridyl)(3 -diphenylphosphino-5-methyl-2,2':5 '2 "-terthiopheno-P, C)ruthenium(II)] [hexafluorophosphate] (RuPMeT3-P, Q (48) NaOH (0.72 g, 18 mmol) was dissolved in deaerated MeOH (18 mL) to give a 1.0 M solution. RuPMeTs-PS (42) (0.600 g, 0.521 mmol) was dissolved into the solution and stirred at reflux for 18 h. A color change from yellow to dark brown was observed. The solution was 99 then cooled to room temperature, concentrated to 10 mL, and added dropwise by pipette into a solution of N H 4 P F 6 (1.70 g, 10.4 mmol) in H 2 O (90 mL) to form a brownish-black precipitate. The precipitate was isolated by filtration, washed with water (10 mL), then ether (15 mL), and recrystallized in EtOH-acetone (9:1) to give 48 as black, shiny crystals. Yield: 0.376 g (72%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 8 8.85 (m, 1H), 8.56-8.49 (m, 2H), 8.42-8.38 (m, 3H), 8.03 (m, 1H), 7.92 (m, 2H), 7.83 (m, 2H), 7.6-7.61 (m, 3H), 7.41 (m, 5H), 7.32 (m, 1H), 7.24 (m, 1H), 7.12 (m, 2H), 7.02 (m, 1H), 6.89 (m, 3H), 6.61 (m, 1H), 6.49 (m, 2H), 6.06 (m, 1H), 2.16 (s, 3H). 3 1 P{'H} N M R (162.0 MHz, CO(CD 3) 2): 5 44.9 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Anal. C 49H 34F 6N4S 3P 2Ru requires C, 53.83; H, 3.41. found: C, 54.04; H , 3.49%. [Bis(2,2 -bipyridyl)(5,5 "-dimethyl-3 -diphenylphosphino-2,2 ':5 '2 "-terthiophene-P,C) ruthenium (II)J [hexafluorophosphate] (RuPMe2T3-.P, Q (49) NaOH (0.716 g, 0.0179 mmol) was dissolved in deaerated MeOH (18 mL) to give a 1.0 M solution. RuPMe2T3-P,5' (43) (0.55 g, 0.47 mmol) was dissolved into the solution and stirred at reflux for 18 h. A color change from yellow to dark brown was observed. The solution was then cooled to room temperature, concentrated to 10 mL, and added dropwise by pipette into a solution of N H 4 P F 6 (1.54 g, 9.4 mmol) in H 2 0 (80 mL) to form a brownish-black precipitate. The precipitate was isolated by filtration, washed with water (10 mL), then ether (15 mL), and recrystallized in EtOH-acetone (9:1) to give 49 as black, shiny crystals. Yield: 0.468 g (85%). ' H N M R (300.1 MHz, CO(CD 3 ) 2 ): 5 8.85 (d,J= 5.7 Hz, 1H), 8.57 (d, J= 7.8 Hz, 1H), 8.52 (d, J= 8.1 Hz, 1H), 8.44-8.37 9m, 3H), 8.05 (dd, J = 8.0 Hz, J= 1.5 Hz, 1H), 7.96-7.84 (m, 4H), 7.70-7.62 (m, 3H), 7.47-7.38 (m, 4H), 7.27-7.10 (m, 4H), 6.93-6.88 (m, 3H), 6.70 (dd, J = 3.3 Hz, J= 1.3 Hz, 1H), 6.51-6.46 (m, 3H), 6.07 (d,J= 1.3 Hz, 1H), 6.52-6.47 (m, 2H), 6.38 (d, J = 8.1 Hz, 1H), 2.42 (s, 3H), 2.15 (s, 3H). 3 1 P{'H} N M R (121.5 MHz, CO(CD 3 ) 2 ): 8 46.5 (s), -143.0 (sep, ' j p F = 708 Hz, PF 6). Anal. C 4 6 H 3 6 F 6 N 4 S 3 P 2 R u - C 3 H 6 0 requires C, 54.69; H, 3.93; N , 5.21. found: C, 54.96; H, 3.60; N , 5.22%. [Bis(2,2 -bipyridyl)(3,3 ""-dihexyl-3 "-diphenylphosphino-2,2 ':5 '2 ":5 "2 "': 5 '"2 ""-pentathiopheno-P,C)ruthenium(II)J [hexafluorophosphate] (RuPhex2Ts-.P,C) (50) To a deaerated solution of NaOH (1.2 g, 0.030 mol) dissolved in MeOH (30 mL), RuPhex2T5-P,5' (44) (150 mg, 0.103 mmol) was added and the solution was heated to reflux. After 1 h, the solution turned from orange to a deep red. After 16 h stirring at reflux, the burgundy-red solution was condensed to 150 mL and added dropwise by pipette into a solution 100 of N H 4 P F 6 (600 mg, 3.46 mmol) in H 2 O (50 mL) and stirred 1 h to give a red-black precipitate. Recrystallization from EtOH gave 50 as a very dark, red powder. Yield: 61.2 mg (45%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 5 8.90 (d, J= 5.6 Hz, 1H), 8.63 (d, J= 8.0 Hz, 1H), 8.57 (d, J = 8.0 Hz, 1H), 8.47 ( t , / = 1.6 Hz, 3H), 8.11 (t, .7= 7.6 Hz, 1H), 7.98 (t, .7=6.8 Hz, 2H), 7.94-7.90 (m, 2H), 7.74 ( t , J = 8.4 Hz, 2H), 7.67 (m, 1H), 7.50-7.42 (m, 5H), 7.38 (d, J = 5.2 Hz, 1H), 7.32 (t, .7=5.2 Hz, 1H), 7.20 (d, .7= 4.8 Hz, 2H), 7.14 (m, 1H), 7.10 (d, .7= 4.0 Hz, 1H), 7.03 (d, J = 5.6 Hz, 1H), 6.98-6.92 (m, 3H), 6.87 (d, J = 5.2 Hz, 1H), 6.72 (d, J = 2.4 Hz, 1H), 6.53 (t, J = 8.0 Hz, 2H), 6.44 (s, 1H), 2.77 (m, 2H), 2.45 (m, 2H), 1.63 (q, J = 7.2 Hz, 4H), 1.42-1.20 (m, 12 H), 0.86 (t, J = 6.6 Hz, 6H). 3 1 P{'H} N M R (162.0 MHz, CO(CD 3 ) 2 ): d 45.6 (s), -143.0 (sep, J P F = 708 Hz, PF 6). [Bis(2,2 -bipyridyl)(3,3 ""-didodecyl-3 "-diphenylphosphino-2,2 ':5 ',2 ":5 "2 "': 5 '"2 ""-pentathiopheno-P,C)ruthenium(II)] [hexafluorophosphate] (RuPDo2T5-P,C) (51) To a deaerated solution of NaOH (12 g, 0.30 mol) dissolved in MeOH (300 mL), RuPDo2T5-7,,5'"(45) (1.00 g, 0.611 mmol) was added and the solution was heated to reflux. After 1 h, the solution turned from orange to a deep red. After 16 h stirring at reflux, the burgundy-red solution was condensed to 150 mL and added dropwise by pipette into a solution of N H 4 P F 6 (3.00 g, 17.3 mmol) in H 2 0 (200 mL) and stirred 1 h to give a red-black precipitate. Recrystallization from EtOH gave 51 as a very dark, red powder. Yield: 410 mg (45%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 5 8.89 (d, J= 5.2 Hz, 1H), 8.63 (d, J= 8.0 Hz, 1H), 8.56 (d, J = 1.6 Hz, 1H), 8.48-8.44 (m, 3H), 8.12-8.08 (m, 1H), 8.00-7.89 (m, 4H), 7.77-7.72 (m, 2H), 7.66 (m, 1H), 7.50-7.43 (m, 5H), 7.37 (d, .7= 5.2 Hz, 1H), 7.31 (t, .7= 6.0 Hz, 1H), 7.19-7.13 (m, 3H), 7.06 (d, J = 4.0 Hz, 1H), 7.03 (d, J= 4.8 Hz, 1H), 6.98-6.92 (m, 3H), 6.86 (d, J = 5.2 Hz, 1H), 6.72 (d, J= 2.4 Hz, 1H), 6.54 (t, J= 8.0 Hz, 2H), 6.44 (s, 1H), 2.77 (m, 2H), 2.45 (m, 2H), 1.63 (m, 2H), 1.44 (m, 2H), 1.26 (m, 36 H), 0.84 (m, 6H). 3 1 P{'H} N M R (162.0 MHz, CO(CD 3) 2): 5 44.8 (s), -143.0 (sep, JPF = 708 Hz, PF 6). Anal. C 7 6 H 8 4 F 6 N 4 S 5 P 2 R u requires C, 60.98; H, 5.66; N , 3.74. found: C, 61.29; H, 5.78; N , 4.00%. [Bis(2,2 -bipyridyl)(5,5""-dibromo-3,3 ""-didodecyl-3 "-diphenylphosphino-2,2': 5 ',2 ":5 "2 '":5 '"2 ""-pentathiopheno-P,C)ruthenium(II)J [hexafluorophosphate] (RuPBr2Do2T5-P,Q(52) To a deaerated solution of NaOH (1.2 g, 0.030 mol) dissolved in MeOH (30 mL), RuPBr2Do2T5-P,5' (46) (50 mg, 0.029 mmol) was added and the solution was heated to reflux. After 16 h stirring at reflux, the dark-red solution was condensed to 15 mL and added dropwise 101 by pipette into a solution of N H 4 P F 6 (6.00 mg, 3.46 mmol) in H 2 O (50 mL) and stirred 1 h to give a red-black precipitate. Recrystallization from EtOH-acetone (99:1) gave 52 as a dark, red-purple crystals. Yield: 35.7 mg (75%). ' H N M R (400.1 MHz, CO(CD 3 ) 2 ): 8 8.88 (d, J = 5.2 Hz, 1H), 8.64 (d, J= 8.4 Hz, 1H), 8.58 (d, J= 7.2 Hz, 1H), 8.49-8.45 (m, 3H), 8.12 (t, J = 6.8 Hz, 1H), 7.98 (m, 2H), 7.93-7.89 (m, 2H), 7.74 (t, J = 8.4 Hz, 3H), 7.52-7.43 (m, 5H), 7.32 (d, J= 6.8 Hz, 1H), 7.20 (d,J= 4.0 Hz, 1H), 7.15-7.09 (m, 2H), 7.10 (s, 1H), 6.99-6.92 (m, 3H), 6.94 (s, 1H), 6.75 (m, 1H), 6.55-6.51 (m, 2H), 6.40 (s, 1H), 2.74 (m, 2H), 2.45 (m, 2H), 1.62 (m, 2H), 1.45 (m, 2H), 1.26 (m, 36 H), 0.84 (m, 6H). 3 I P{ 'H} N M R (162.0 MHz, CO(CD 3) 2): 8 46.6 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Anal. CyeH^FeNjSsPaB^Ru requires C, 55.37; H, 5.01; N , 3.40. found: C, 56.78; H, 4.99; N , 3.28%. 5.2.3 X-ray Crystallographic Analysis Suitable crystals of 47 and 49 were obtained by slow diffusion of hexanes into acetone solution, and were mounted on a glass fiber. The data were collected at -100.0 ± 0.1 °C and the X-ray source in all cases was graphite monochromated M o - K a radiation. The structures were solved by direct methods3 and expanded using Fourier techniques.4 Hydrogen atoms were placed in calculated positions. The crystal structures were obtained and analysis carried out by Dr. B. Patrick at UBC. Data for 47 were collected to a maximum 29 of 55.7° on a Rigaku/ADSC CCD area detector in a series of two scans using 0.50° oscillations with 47.0 second exposures. Data were collected and processed using the d*TREK program,5 and corrected for Lorentz and polarization effects. The molecule crystallized with one molecule of acetone in the asymmetric unit. Hydrogen atoms were included in calculated positions. A l l calculations were performed using the teXsan6 crystallographic software package of Molecular Structure Corporation. Data for 49 were collected to a maximum 29 of 58.1° on a Bruker X8 diffractometer in a series of two scans using 0.50° oscillations with 12.0 second exposures using the Bruker SAINT 7 software package. Data were corrected for absorption effects using the multi-scan technique (SADABS 8 ) and corrected for Lorentz and polarization effects. One [PF6]~ anion is disordered and was modeled in two orientations with relative populations of 0.85 and 0.15. The atoms of the minor fragment were refined isotropically, all other non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in calculated positions. A l l refinements were 102 performed using the SHELXL-97 9 crystallographic software package. Appendix 1 contains the crystallographic data for the crystal structures of 47 and 49. C i f files are available online. 1 0 ' 1 1 5.2.4 Density Functional Theory Calculations Density functional theory calculations were carried out using the Gaussian 03 Package,12 and the B 3 L Y P 1 3 ' 1 4 method is used to optimize the geometry. The 6-31 G* basis set was used to model non-metallic atoms. A L A N L 2 D Z pseudopotential is used for inner shell calculations of the metal center, with the corresponding L A N L 4 basis set for the pseudopotential. To analyze the chemical bonding, N B O analysis was used. A l l calculations were carried out by Y. Zhang of the Prof. A . Wang research group at UBC. Gaussian View software was used to generate depictions of the orbitals. 5.3 Results 5.3.1 Synthesis and Crystallographic Studies The Ru(II)-(phosphino)terthiophene-P,C complexes 47-49 were prepared by reaction of the corresponding Ru(II)-P,5' complexes 41-43 with NaOH dissolved in methanol, and heating at reflux for 18 hours (Scheme 5-2). Yields (50-85%) were higher in methanol solution than for reactions in C H 3 C N / H 2 O solution (10-50%). A color change from yellow to dark brown indicated the onset of the reaction, which occurred within one hour of stirring reflux conditions in basic solution. Analysis of crude samples of the Ru(II)-P,C complexes indicated the presence of some oxidized ligand that was removed by recrystallization of the products. The solid-state structures of 47 and 49 (Figure 5-1, Table 5-1) were established by X-ray crystallography from crystals grown by slow diffusion of hexanes into a solution of the complex in acetone. Ru(II) is directly bonded to a thienyl carbon to yield cyclometallated complexes in the P, C bonding mode, with the bound thiophene ring tilted very little from plane the Ru-C bond at angles of 7.3° for 47 and 8.6° for 49. This more planar conformation allows back donation from a Ru(drt) orbital to the terthiophene T C * orbital,15 and the R U 1 - C 3 5 bond lengths (2.076(3) A and 2.095(2) A) are longer than the calculated double bond length in Ru=CH 2 + (1.88 A ) , 1 6 but shorter than Ru-C single bonds reported for ruthenium bound to alkyl ligands (~2.22 A ) , 1 7 ' 1 8 indicating that the bond possesses some double bond character. An antibonding interaction is also indicated by elongated thienyl C-C double bonds, primarily the C35-C36 bond lengths of 47 and 49 that are elongated by 0.044 A and 0.040 A , respectively, compared to the corresponding 103 bond lengths of RuPT 3-P,S (41) and RuPMe 2T 3-P,S (43) (Chapter 4). The S i -C 3 6 -C 3 7 -S 2 torsion angles of 47 (-10.7(7)°) and 49 (-19.7(3)°) show that the bound thienyl rings of the Ru(II)-P,C complexes are more coplanar compared to those of the P.iS-bound complexes. Intramolecular Ti-stacking 1 9 ' 2 0 is observed between the adjacent N 4 pyridyl and C-21 phenyl rings of 47 based on inter-ring separations of 3.510 A , while the corresponding distance for 49 is 3.901 A . Intermolecular thienyl rings in the crystal structures of 47 and 49 are separated by > 4 A . Scheme 5-2 R u P T 3 - P , S (41) R = R' = H RuPMeT-j-.P.S (42) R = Me, R' = H R u P M e 2 T 3 - . P , S (43) R = R' = Me RuPT 3 - / 5 ,C(47 ) R = R' = H (50%) R u P M e T 3 - P , C ( 4 8 ) R = Me, R' = H (72%) R u P M e 2 T 3 - P , C ( 4 9 ) R = R' = Me (85%) Scheme 5-3 RuPhex 2 T 5 -P,5(44) M e Q U RuPhex 2 T 5 - .P, C (50) (45%) RuPDo 2 T s -P ,5 (45) • R u P D o 2 T 5 - P , C (51) (45%) R u P B r 2 D o 2 T 5 - P , S (46) 65°C R u P B r 2 D o 2 T 5 - P , C (52) (75%) 18 h Ru(II)-(phosphino)pentathiophene-P,C complexes 50-52 were synthesized from Ru(II)-P,5 complexes 44-46 by , the' same procedure as for the Ru(II)-(phosphino)terthiophene-P,C complexes (Scheme 5-3). Crystals of RuPDo 2T 5-P,C (51) suitable for X-ray analysis could not be obtained, possibly due to disorder of the pentathiophene chain or the dodecyl substituents combined with a reduced positive charge and number of counterions compared to RuPDo2Ts-P,5 (45). RuPhex2Ts-P,C (50), containing shorter hexyl substituents, was expected to give crystals that exhibit less disorder and greater crystallinity compared to 51, 104 and was prepared for the purpose of obtaining a crystal structure. Recrystallizations carried out using ethanol, methanol, acetone, hexanes, and mixtures thereof did not yield crystals of 50 suitable for X-ray analysis, and often an oily precipitate was obtained. Figure 5-1 (a) ORTEP view of RuPT3-P, C (47) and (b) ORTEP view of RuPMe 2T 3-P,C (49). Hydrogen atoms, counterions, and occluded solvents are omitted for clarity, and thermal ellipsoids are drawn at 50% probability. Table 5-1 Selected interatomic distances (A) and angles (°) for 47 and 49 RuPT 3 -P, C (47) RuPMe 2T 3-PC(49) Bond length /A Bond length /A Ru, -C 3 5 2.076(3) C 3 3 - C 3 4 1.350(4) Ru, -C 3 5 2.095(2) C 3 3 - C 3 4 1.370(4) Rui-P, 2.2737(7) C 3 4 - C 3 5 1.431(4) Rui-P, 2.2954(7) C 3 4 -C 3 5 1.450(3) S,-C 3 3 1.719(3) C 3 5"C 3 6 1.394(4) S i -C 3 3 1.734(3) C 3 5"C 3 6 1.398(3) S,-C 3 6 1.747(3) C 3 6"C 3 7 1.445(4) Sl"C 3 6 1.758(2) C 3 6"C 3 7 1.448(4) Torsion Angle 1° Torsion Angle 1° Si-C36"C37-S2 -10.7(3) Sl"C 36-C 3 7-S2 -19.7(3) S2-C40-C4l-S 3 161.6(2) S 2-C40"C4i -s 3 16.0(3) 105 5.3.2 Density Functional Theory (DFT) Calculations DFT calculations were carried out on RuPT 3-.P, C (47) for comparison to the calculations done for R u P T ^ ^ S (41). Calculations of the localization and energies of the frontier orbitals of the gas phase 1+ charged complex (Table 5-2, Figure 5-2) predict that the HOMO is localized on both the Ru(II) metal and the terthiophene rings, with a larger contribution from the metal than was predicted for 41 (Figure 5-2). Similarly to 41, the HOMO is separated by ~1 eV from the HOMO-1, though for the .P,C-bound complex the HOMO-1 is also composed of mixed metal-terthiophene orbitals. The nearly degenerate L U M O and LUMO+1 are centered on the bipyridine rings and the metal center. Bipyridine ligands are expected to possess a low-lying L U M O level that enable them to act as charge transfer acceptors.21 The LUMO+2 also primarily displays localization on the bipyridine rings. As the (phosphino)terthiopheno ligand is formally -1 charged and electron rich, the terthiophene n* acceptor levels are higher compared with 41. The calculations predict a lowest-energy terthiophene/Ru—»Ru/bpy charge-transfer transition from the HOMO to the L U M O of 2.1062 eV in energy. A Ru->bpy M L C T is expected for Ru(bpy)2Ln2+ complexes, which corresponds to the predicted HOMO—>LUMO transition, with additional contribution from the n orbitals of the terthiophene group. Table 5-2 Calculated HOMO and L U M O energies for R u P T 3 - P , C (47) Orbital Energy /eV Orbital Energy /eV HOMO -6.6935 L U M O -4.5873 HOMO-1 -7.5493 LUMO+1 -4.5272 HOMO-2 -7.7488 LUMO+2 -3.7835 HOMO-3 -8.0484 LUMO+3 -3.7835 HOMO-4 -8.1107 LUMO+4 -3.4913 106 Figure 5-2 (a) Ordering of the calculated HOMO and L U M O energies of RuPT 3-/ > ,C(47), and depictions of the frontier orbitals. General orientation of 47 is shown at bottom. 107 5.3.3 Cyclic Voltammetry The cyclic voltammograms of 47-49 (Figure 5-3) display R u I I I / n oxidations (0.49-0.57 V), terthienyl oxidations (0.96-1.11 V) and two bipyridyl reductions (-1.53 to -1.55 V , -1.78 V). Assignments of the ruthenium oxidation potentials can be made based on ligand electrochemical parameters (EC).22 Due to the absence of a ligand parameter for a carbon-bound thiophene or I I /TTT TO similar aromatic ring, the known Ru oxidation potential of 53 (Chart 5-2) was used to calculate the expected oxidation potential ( i^aic) , using EL values to correct for the measured oxidation potential. Complex 53 possesses similar bonds to the Ru(II)-P )C complexes, with the metal center bound via a carbon atom to a phenyl ligand and to two bipyridyl ligands, while the sixth coordination is to a pyridyl ring rather than to a phosphine. Using the known oxidation potential of 53 (0.464 V vs. SCE), the estimated oxidation potential is of 47 is: where one EL value for 2,2'-bipyridine (0.259) is subtracted and the EL value for triphenylphosphine (0.39) is added to give 0.60 V vs. SCE. This value is very close to the first oxidation potential of the P,C Ru(II) complex 47 (0.57 V) that is assigned to be Ru(II)-based. The oxidation potentials of 48 and 49 are lower in value due to the attachment of the electron donating methyl substituents. £ c a i c (V) = 0.464 - 0.259 + 0.39 = 0.60 5.1 Chart 5-2 53 108 I , I . I . I . I . I . I . I -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Volts / V vs. S C E Figure 5-3 Cyclic voltammograms of (a) 47-49, and (b) 51 in C H 3 C N solution at 4.0 x 10"3 M concentrations, containing 0.1 M [(«-Bu)4N]PF6 supporting electrolyte, scan rate = 100 mV/s. Table 5-3 Cyclic voltammetry data of Ru(II)-phosphinothiophene-P,C complexes3 Compound £i/2,ox ±0.01 V vs. SCE £l/2,red±0.01 V V S . SCE RuPT 3-/>,C(47) +0.57 (Rum /") + 1.11 (PT3 0 /~) -1.53 (bpy07") -1.78 (bpy72") R u P M e T 3 - / J ) C ( 4 8 ) +0.51 (Ru I I I / n) + 1.04(PMeT3 0 /") -1.54 (bpy0'") -1.78(bpy- /2-) R u P M e 2 T 3 j P , C ( 4 9 ) +0.49 (Ru I I I / H) +0.96 (PMe 2T 3 0 /") -1.55(bpy0 /-) -1.78 (bpy72") RuPDo 2 T 5 -P ,C (51) +0.49 (PDo 2T 5 0 /~) +0.80 (Ru I I I / H) + 1.46 (PDo 2 T 5 + / 0 ) -1.52(bpy0 /-) -1.83 (bpy72")" Measurements carried out in C H 3 C N solution containing 0.1 M [(/?-Bu)4N]PF6 supporting electrolyte. 109 Assignments of all oxidation and reduction potentials (Table 5-1) are substantiated by the addition of two methyl substituents to the terthiophene rings having a greater effect on the thienyl oxidation potentials (0.15 V decrease) than on the ruthenium oxidation potentials (0.08 V decrease), and almost no effect on bipyridyl reduction potentials. Conversion to the P,C bonding mode from the Ru^I)-/5,.!) complexes cathodically shifts the ruthenium-based oxidations and terthienyl oxidations ~(-0.9 V) and <(-0.6 V), respectively, while the bipyridyl reductions are less affected. The thienyl oxidations of the Ru(ll)-P,C complexes are also cathodically shifted from those of the corresponding ligands 13-15 (1.05-1.30 V), reflecting how the terthiophene chains are electron rich. It is noted that formally, oxidation of the P-phosphinothiopheno ligands are from a -1 to neutral state. DFT calculations are in agreement with the assignment of both reductions as bipyridyl-based, while predicting a mixed Ru/terthiophene HOMO rather than a Ru(II)-based level. Conversion from Ru(II)-JD,iS' complex 45 to Ru(II)-P,C complex 51 significantly lowers both oxidation potentials from 1.21 and 1.46 V; to 0.49 and 0.80 V . Compared to the Ru(II)-terthiophene-P, C complexes, the first and second oxidations of 45 are close in value, which could indicate nearly degenerate oxidation levels. The assignments of the oxidations of 45 are reversed compared with the terthiophene complexes (Table 5-3), with the first oxidation assigned as thienyl-based, and the second oxidation assigned as Ru(II)-based. The reasoning for these assignments is provided in sections 6.3.4 and 6.3.5 and in the Discussion. The first, thienyl-based oxidation potential of 45 is significantly lower than those of P D 0 2 T 5 (22) (0.99 V) and D02T5 (24) (0.82 V), which can be attributed to the formal negative charge on the bound ligand. 5.3.4 Electron Paramagnetic Resonance Electron paramagnetic resonance (EPR) experiments were carried out in order to definitively assign whether the metal or the ligand is oxidized with the first oxidation couple. The reversibility and stability of the oxidized species of the Ru(II)-.P,C complexes facilitate measurement of EPR spectra of the first oxidation products. Due to the short spin-lattice relaxation times often observed with transition metal radicals, a metal-based signal usually cannot be observed at 298 K ; 2 4 therefore, measurements were also taken of samples frozen at 110 K following oxidation. Due to spin-orbit coupling of transition metals within the crystal field, 110 three signals at gi, g 2 , and g 3 are observed along the different axes for samples in frozen matrices. The anisotropy of the g factor is characterized by: Ag = gi - g 3 5.1 where gi and g 3 are replaced by g i and g\\ values for samples that only generate two peaks. Generally, a larger metal contribution results in a broader signal and a larger Ag value. Hyperfine coupling constants are rarely distinguished for ruthenium complexes of this type except at very low temperature (<5 K). It was observed for complex 47 at 298 K (Figure 5-4a) that the very weak, broad signal expected for a metal-based radical was seen only when the sample was oxidized to the extent that a second oxidation species was also observed at g = 1.9880. This species is identified as a ligand-based radical due to its narrow peak width, although the peak is still broad compared to that expected for a free organic ligand. The ligand is therefore thought to be still complexed to the metal, since the relaxation behavior of slower-relaxing paramagnetic centers is influenced by proximity to a fast-relaxing metal center.24 When the temperature is lowered to 110 K, the ruthenium-based signal of 47 can be more easily observed, though the second oxidation species is still seen. The Ag value observed for the metal-based signal of 47 (0.2505) indicates metal character, but is smaller than for completely metal-based R u I I I / n oxidation. For example, the g tensor reported for polycrystalline Ru(bpy) 3 3 + diluted in diamagnetic powder at 77 K is (gi, g||) = (2.64, 1.14), giving a Ag value of 1.40.25 The methyl substituted complex 48 displays a broad, metal-based signal at 298 K, and three peaks at 110 K, with a Ag value of 0.1892. Both a signal at room temperature and a decreasing Ag value indicate that the radical is more ligand-based than for 47. The dimethylated complex 49 also exhibits a signal at room temperature, and only two peaks, g i and g||, can be distinguished at 110 K with an even lower Ag value of 0.0918. It is evident that the oxidized species exhibit a loss of metal character from 47 to 49; however, the anisotropy observed at low temperature indicate that all oxidations are essentially metal-based and are in agreement with the electrochemical assignments for the Ru(II)-(phosphino)terthiophene-P,C complexes. I l l 3000 3100 3200 3300 3400 3500 3600 3700 3800 B/mT 3000 3100 3200 3300 3400 3500 3600 3700 3800 B/mT 3000 3100 3200 3300 3400 3500 3600 3700 3800 B/mT Figure 5-4 E P R spectra at room temperature (RT = 298 K) and low temperature (110 K) of the first oxidized species of (a) R U P T 3 J P , C (47), (b) RuPMeT 3 -P, C (48), and R u M e 2 P T 3 - P , C (49). 1.12 The EPR result for 51 contrasts those for 47-49. The spectra of 51 at 298 K and 110 K are very similar; single peaks are observed that do not display the anisotropy at low temperature that is characteristic for metal complexes (Figure 5-5). These spectra are shown with the EPR spectrum of the corresponding pentathiophene D02T5 (16) that displays a very narrow signal characteristic of a hydrocarbon radical. Since a metal-based radical is not observed, it is most likely that oxidation of 51 is pentathiophene-based and the broad signal is due to metal attachment decreasing the relaxation time. Detachment of a pentathiophene species from the metal is not indicated, as a very narrow signal would be observed similar to that of 16. Based on the EPR spectra of 51, the assignment of a thienyl-based first oxidation species by cyclic voltammetry is supported. Figure 5-5 EPR spectra of the first oxidized species of (a) D02T5 (24) (Chapter 2) at room temperature (298 K) , and (b) RuDo2PTs-.P, C (51) at room temperature (298 K) and low temperature (110 K). 5.3.5 Optical Spectroscopy and Charge-Transfer Correlations Conversion from Ru(II)-(phosphino)terthiophene-/J>,S' complexes 41-43 to the Ru(lT)-P,C complexes 47-49 alters the absorption spectra with an accompanying color change of the solution from bright yellow to deep brown (Figure 5-6), and the complexes appear black in the solid state (Figure 5-7a). The terthiophene n—>n* transitions red shift ~30 nm and the bipyridyl (a) J g = 1.9653 ' 3100 3200 3300 3400 3500 3600 3700 3800 B/mT 113 71—>7T* transitions red shift -15 nm compared to the Ru(II)-.P,S complexes. Phenyl K—>K* transitions are likely obscured by more intense transitions. The Ru(d7t)—»bpy(7T*) M L C T transitions red shift 56-63 nm and weak transitions assigned as "spin-forbidden" M L C T transitions26"30 are observed as shoulders at 617-633 nm, accounting for the very dark colour of the complexes. Compared to the corresponding terthiophene ligands (Table 3.2, Chapter 3), terthiophene 71—transitions are only -10 nm blue-shifted. 300 400 500 600 700 800 X /nm Figure 5-6 Solution absorption spectra of Ru(H)-P,C complexes 47-49 and 51 in CH 2 C1 2 solution. (a) Figure 5-7 (a) RuPT 3 -P ,C (47) crystals and (b) RuPDo 2 T 5 -P, C (51) powder, displaying the very dark colours of the complexes due to their broad absorption range. 114 Table 5-4 Electronic spectroscopy data for Ru(II)-phosphinothiophene-.P, C complexes Compound Solution Absorption3 A.max /nm [e /IVr'cm"1] Solid-State Absorption ^ m a x /nm Emission0 A, m a x /nm R u P T 3 - P , C (47) 295 (4.63 x 104), 347 (2.00 x 104), 456 (1.84 x 617 (sh)(2.25 x 103) 104), 315,375 (sh), 482, 635 (sh) 754 (x = 22 ± 2 ns)c R u P M e T 3 - P , C (48) 295 (4.72 x 104), 3 5 0 (1.91 x if)4), 459 (2.03 x 628 (sh) (2.20 x 103) 104), 304, 368 (sh), 475, 635 (sh) 763 R u P M e 2 T 3 - P , C (49) 295 (4.64 x 104), 351 (1.97 x 104), 460 (2.07 x 633 (sh) (2.10 x 103) 104), 301,364 (sh), 473, 630 (sh) 772 R U P D 0 2 T 5 - P C (51) 295 (5.24 x 104), 3 60 (2.63 x 104), 380 (sh) (2.59 x l O 4 ) , 485 (3.15 x 615 (sh)(2.62 x 103) 104), 306, 400 (sh), 507, 627 (sh) Measurements carried out in CH2CI2 solution. bDegassed C H 3 C N solution. cLifetime determined from emission at 750 nm. Switching the bonding mode from Ru(II)-(phosphino)pentathiophene-P,5' complex 45 to the P,C-bound complex 51 does not affect the optical transitions as strongly as for the terthiophene complexes. Still, there is a dramatic colour change in solution from orange to dark red, and complex 51 is a very dark red powder in the solid state (Figure 5-7b). While the bipyridyl rc—>Tt* transition and the charge transfer band red shift 15 nm and 20 nm, respectively, the pentathiophene n—>n* transition shifts from 371 nm for 45 to a broad peak at 360 nm with a shoulder at 380 nm for 51. The two transitions observed could correspond to conformations of the pentathiophene chain that are unable to interconvert rapidly in solution due to the constrained nature of the bonding to ruthenium. This is supported by the observation that the polarity of the solvent affects the relative intensities of these peaks; the peak at 360 nm is more intense in CH2CI2, while the peak at 380 nm is prominent in C H 3 C N . Calculations have been reported predicting that the ground state of pentathiophene is twisted while the excited state is planar (quinoidal); therefore, multiple ground state conformations would result in different rc—nr.* transition energies.31 The weak transition observed as a shoulder at 615 nm is assigned as a spin-forbidden M L C T transition and the charge transfer band at 485 nm is significantly more intense than the M L C T bands of the terthiophene complexes. 115 > PJ 2.76 2.72 2.68 2.64 2.60 2.56 (a) P,CRuPMeT3 (48) /J,CRuPMe2T3(49)" P,CRuPT3 (47) iiP,CRuPDo2T5 (51) 1.98 2.00 2.02 2.04 2.06 2.08 2.10 AE /eV > PJ 2.720 2.715 2.710 •2.705 2.700 2.695 - (b) 41 R = 0.99639 SD = 0.00148 . 48 » X -/+ 49 l 1 , 1 , 1 l . l . i 2.04 2.05 2.06 2.07 2.08 2.09 2.10 AE /eV Figure 5-8 (a) Plot of the primary CT optical transition (E o p ) versus the difference in potential between the first oxidation potentials and first reduction potential (AE) for all Ru(II)-P,C complexes, and (b) the linear fit for the Ru(II)-(phosphino)terthiophene-/J,C complexes. For bis(bipyridine) ruthenium(II) complexes, a correlation can be made between the observed Ru->bpy M L C T transition, Eop, and the difference between the Ru(III/II) and bpy(0/-l) redox couples, Afi'(redox).32 Figure 5-8 shows a linear correlation for the Ru(II)-(phosphino)terfhiophene-P, C complexes; however, pentathiophene complex 51 deviates significantly from this linear correlation. The linear fit for 47-49 gives: Eop = 0.38 zl£(redox) + 1.92 5.2 where the slope (a) is 0.38 and the constant term (1.92) collects all of the solvent and reorganizational energies of the excited state. While the fit is good, a deviates from unity, which is the value usually observed for [Ru(bpy)2Ln]2+ complexes. Correlations of this type are generally valid when charge transfer occurs by a very similar process for complexes in a series, and if the oxidation is metal-based and the reduction is bipyridyl-based. Since 51 does not correlate well with the Ru(II)-phosphinoterthiophene complexes, it is likely that the charge transfer occurs by a different process. In addition, it is not expected that 51 should correlate, since the EPR results indicated that the first oxidation of the complex is not metal-based, but thienyl-based. 116 The solid-state absorption spectra of Ru(II)-P, C complexes 47-49 and 51 (Figure 5-9) display red shifts of the bipyridyl n—>n* (9-20 nm), M L C T (13-26 nm), and thienyl rc—»rc* transitions (13-28 nm) compared to solution spectra that could be due to intramolecular stacking interactions such as are observed in the crystal structures of 47 and 49. Broader bands are observed that may be due to different conformers of the oligothiophene chains in the solid state, and/or vibrational fine structure. Figure 5-9 Solid state absorption spectra of Ru(II)-P) C complexes 47-49 and 51 drop-cast from solution in acetone. The primary CT transition is normalized to the same absorbance intensities for all spectra. The Ru(II)-P, C complexes exhibit changes in the luminescence spectra compared to the Ru(II)-/,,5' complexes that are likely related to shifting energies of the emitting states. Ru(II)-(phosphino)terthiophene-P, C complexes 47-49 show emission bands at 754-772 nm that are just red shifted of the spin-forbidden transitions, and are likely from the same low-lying M L C T level. Approximate quantum yields of the emission are -0.001%, and the intensity of the emission decreases with the addition of methyl substituents (Figure 5-10a). The emission lifetime of 47 was determined to be 22 ± 2 ns, though a very short-lived (< 10 ns) species was also observed below 620 nm that is likely due to the formation of a more emissive decomposition product (Figure 5-10b). The Ru(II)-(phosphino)pentathiophene-P,C complex 51 does not display any observable emission, which could indicate increased deactivation of the excited state via vibrational pathways. Both the decrease in the emission intensity of 48-49 and the lack of 117 emission for 51 could be due to more vibrational relaxation modes that are introduced with methyl and dodecyl substitution of the oligothiophene chains. As was observed for the ^(II)-^ ,^ complexes, small amounts of secondary products formed in solutions of the Ru(II)-P,C complexes. These products formed more rapidly in chlorinated solvents and at low concentration and exhibited more intense emission at wavelengths similar to those of the ligands, suggesting that the ligand detaches from the metal. 31 The P N M R spectrum of complex 51 displayed a small peak at 5 41.2 after 1 hour in CO(CD3)2, and multiple minor peaks were formed after 24 hours. Similar results were observed for 47-49, though the secondary products formed less rapidly. The addition of concentrated acid to all Ru(II)-P,C complexes resulted in conversion of all species to the corresponding, pure Ru(II)-P,5' complexes, as observed by peaks in the 3 1 P N M R spectra. 6| 5 A 4J £ 3 J 2| c 1 01 400 RuPT3-P,C(47) RuPMeT3-P,C(48) RuPMe 2T 3-P,C(49) 600 700 X /nm (a) > CO C w 1 1 1 • i • i • 14 ns (b) • * 62 ns - • 113 ns ° 137 ns • • • • • • • • • • • A A A ^ A A A A A • A A A £ A A A " o 5 0 f j 0 Q O i . i i . i 0 0 0 B 1 600 640 680 720 X /nm 760 Figure 5-10 (a) Emission and excitation spectra of 47 (A,ex = 456 nm, A, e m = 748 nm), 48 (A,ex = 459 nm, A, e r a = 751 nm), and 49 (A,ex = 460 nm, A. e m - 761 nm) in deaerated C H 3 C N . Solution abs = 0.1 at the excitation wavelength and the solvent spectra have been subtracted for clarity, (b) Emission spectra of 47 obtained at 14 ns, 62 ns, 113 ns and 137 ns average times after the incidence of the laser pulse. 5.3.6 Spectroelectrochemistry Spectroelectrochemistry experiments were carried out to probe the origin of the charge transfer (CT) bands of the Ru(lT)-P,C complexes, and to further characterize the oxidized species. A requirement for spectroelectrochemistry experiments is that the oxidized species be stable, as was demonstrated by cyclic voltammetry experiments. The complexes were oxidized 118 at progressively higher oxidation potentials to generate the first oxidation species, and the resulting absorption spectra in the visible/NIR region were measured. It was observed that the oxidized complexes were stable and displayed reversibility to oxidation over the duration of an experiment, generally six hours. The effect of oxidation on the CT band of each complex, as well as the formation of new bands, was observed. Figure 5-11 displays the spectra of complex 49 that was oxidized in the region of the first oxidation potential (0.49 V), where the spectrum of neutral 49 was subtracted from each spectrum to display the difference spectra. Oxidation between 300-600 mV resulted in bleaching of the M L C T band at 466 nm, and the formation of a strong absorption band at 577 nm and two broad bands of lesser intensity at 663 nm and 1329 nm. I I I I I I I I I I 1 L 400 600 800 1000 1200 1400 X /nm Figure 5-11 Difference spectra of 49 in deaerated C H 3 C N solution with 0.1 M [(rc-Bu)4N]PF6 supporting electrolyte. The solution was oxidized at progressively higher oxidation potentials every 25 mV and the spectrum of neutral 49 was subtracted. The Ru(II)-(phosphino)terthiophene-.P,C complexes 47 and 48 displayed very similar results to 49, with bleaching of the M L C T band and simultaneous formation of new bands (Figure 5-11). The EPR results allow assignment of the oxidations as metal-based; therefore, the spectra are compared to the spectroelectrochemistry data of [Ru(bpy)3]3+ and related complexes. With oxidation, [Ru(bpy)3] displays a weak band at 675 nm and bleaching of the M L C T band (453 nm), while for [Ru(X2-bpy)3J3+ complexes and mixed-bpy complexes, new bands can form 119 in the range between 640-800 nm. ' These bands are assigned as L M C T transitions from a bipyridine group to Ru(III) that increases in intensity i f the bipyridyl group is electron rich (such as for substituted bipyridine rings) or with decreasing value of Ru(III/II) redox potential.33 L M C T transitions if this type could account for the bands observed at 632-663 nm, and the bands are fairly intense compared to those reported, which can be attributed to the low Ru(III/II) redox values. Table 5-5 Spectroelectrochemistry data for P,C Ru(II)-phosphinothiophene complexes and related oligothiophenes Oxidized Species Absorption3 A . m a x /nm Oxidized Species Absorption A , m a x /nm P , C R u P T 3 + (47+) 558, 632, 1326 54+ 572,880 b P , C R u P M e T 3 + (48+) 562,646,1332 55 + 620, 939c P , C R u P M e 2 T 3 + (49+) 577, 663, 1329 D o 2 T 5 + (24+) (673, 1291),d 728a P C R u P D o 2 T 5 + (51+) 679, 940, 1405 56 + 725, 1240c Measurements carried out in deaerated C H 3 C N solution with 0.1 M [(rc-Bu)4N]PF6 supporting electrolyte. bRef 3 5. cRef 3 6. dPeaks due to formation of dimer. Chart 5-3 56 120 The intense, high-energy transitions (558-577 nm) and the low-energy transitions (1326-1332 nm) of 47-49 are not observed for [Ru(bpy)3] or for related complexes, indicating a role of the attached terthiophenes in the oxidation of these complexes. Spectroelectrochemical studies of the oxidized species of terthiophenes 54 and 55 (Chart 5-3) show the formation of intense bands at -600 nm and weaker bands at -900 nm. The intense transitions are similar in position and intensity to the transitions observed for 47-49. There may be a general equilibrium for the Ru(II)-(phosphino)terthiophene-/J, C complexes between charge localization on the Ru(II) center, and on the terthiophene chain (Scheme 5-4), resulting in the presence of both oxidation features in the spectra. Note that oxidation of a phosphinothiophene group formally results in a neutral ligand. If the terthiophene 7i-orbitals and the Ru(II) drt-orbitals (HOMO) are close in energy, charge transfer from the terthiophene chain to Ru(III) could occur for the oxidized species. The band edge of the low-energy transitions (1600 nm = 0.775 eV) is similar to the energy difference between the Ru I 1 I / n and PMe nT3 0 /~ redox couples (-0.5 eV) seen by cyclic voltammetry, and it is expected that an optical transition would be greater in energy to overcome solvent and reorganizational energies. Scheme 5-4 Ru 3 + -PMe n T 3 - - *~ Ru 2 + -PMe n T 3 n = 0-2 The spectroelectrochemical spectra for the Ru(II)-(phosphino)pentathiophene-P,C complex 51 also show bleaching of the charge-transfer band at 497 nm and the formation of intense and weak bands at 679 and 1405 nm, respectively, that are red shifted from the transitions observed for 47-49 (Figure 5-12). The band at 940 nm is substantially shifted from the 632-663 nm bands of 47-49, and does not indicate a Ru(III)—»bpy L M C T . The spectrum of the oxidized species of D02T5 (16) displays bands that shift with increasing oxidation, suggesting that the 728 nm transition is the oxidized species, while the 673 and 1291 nm bands that form are due to a 2+ charged dimer that is expected to form for oxidized oligothiophenes. Pentamer 56 3 6 (Chart 5-3) displays bands at 725 and 1240 nm when oxidized. The first transitions of 16 + and 56 + are similar to those of 51 + , and it can be speculated from the expected ordering of the energy levels that the transitions of 51 + at 940 nm and 1405 nm could be due to bpy—>PDo2Ts 121 and Ru(II)-»PDo2T5 charge transfers, respectively. If the band edge of the low-energy transition of 5 1 + is taken as 1800 nm by extrapolation of the spectrum, the energy (0.689 eV) is similar to the energy difference between the Ru I I 1 / n and PMe nT3 0 /" redox couples (0.31 eV), as will be further discussed. Figure 5-12 Difference spectra of (a) complex 51 and (b) pentathiophene 24 (Chapter 2) in deaerated C H 3 C N solution with 0.1 M [(«-Bu)4N]PF6 supporting electrolyte. The solution was oxidized at progressively higher oxidation potentials every 25 mV and the spectrum of neutral 51 was subtracted. 5.4 Discussion The P,C bonding mode of complexes 47-52 resulted in a planar conformation of the bound ring that would allow back-donation from the Ru(II) d y z orbital into the unoccupied T C * antibonding orbital of the bound thiophene ring. 1 5 Since sp2 hybridization at the bound carbon does not allow significant deviation from a trigonal planar arrangement, tilt angles of the bound rings of 47 and 49 are only 9-10°, and loss of conjugation around the bound thiophene rings can be inferred by 0.033-0.042 A elongation of the C35-C36 bonds compared to Ru(II)-P,5' complexes 41 and 43 (Chapter 4). While some disruption of conjugation is a consequence of the P,C bonding mode, marked increases in coplanarity of the terthiophene rings compared to the Ru(II)-P,5' complexes are also observed. The interannular torsion angles between the bound rings of 47 (10.7(3)°) and 49 (19.7(3)°) are closer to planar compared with those calculated for T3 (147.6°). 3 7 These results indicate that the conformations of the oligothiophene chains are twisted by metallation with Ru(bpy) 2 + groups in the P C bonding mode. 122 Compared to the Ru(II)-(phosphino)terthiophene-P,lS' complexes, P,C-bound complexes 47-49 are electron-rich, as reflected by thienyl oxidation potentials that are decreased by > 0.7 V , to yield oxidations at 1.11-0.96 V that are quasi-reversible. The terthiophene rings oxidize at higher potentials than the Ru(II) center, which is a different result than the mixed metal/terfhiophene HOMO state predicted by the DFT calculations for RuPT3-P,C (47), but is in agreement with the metal-based EPR spectra observed for the oxidized species of 47-49. The P,C bonding mode evidently stabilizes the ruthenium center to approximately the same extent as for the complexed terthiophene ligands. In contrast, the Ru(II)-(phosphino)pentathiophene-7J,C complex 51 displays a reversible first oxidation at 0.49 V that is thienyl-based, as confirmed by the EPR data, and close in potential to the R u I I / i n oxidation. The cathodic shift of the thienyl oxidation is presumed to be due to extension of the thiophene chain and possibly the electron-donating alkyl chains, to result in pentathiophene- and metal-based orbitals that are close in energy and approximate the relative orbital energies shown in Scheme 5-1. The increased electron density and coplanarity of the terthiophene rings of 47-49 compared to the Ru(II)-/')5' complexes results in -30 nm red shifts of the terthiophene n—>n* transitions and are at similar wavelengths as those of the corresponding ligands (Chapter 3). The two pentathiophene n—MI* transitions at 360 and 380 nm that are observed for 51 are not significantly removed from the transition for RuPDo2T5-JP,S (45) (371 nm). Pentathiophenes do not normally show conformational structure in solution, but it has been observed at low temperature,31 and the P,C bonding mode may result in a barrier to interconversion between two conformers. Interconversion between two conformers would still be relatively fast since only single peaks are observed by 3 1 P N M R spectroscopy and cyclic voltammetry at room temperature. Switching the bonding mode from P,S to P,C modifies the properties of the bound Ru(bpy)2 + group. The ruthenium-based oxidation potentials of terthiophene complexes 47-49 decrease 0.72-0.92 V , reflecting an increase in electron donation from a thienyl carbon to the Ru(II) center compared to a thienyl sulfur. Oxidation potentials of this range have been observed for other cyclometallated Ru complexes.38'39 The effect of switching the bonding mode on the bipyridyl reduction potentials is not substantial. The difference in energy (AE) between first ruthenium-based oxidations and the first bipyridyl reduction is known to correlate to the M L C T transitions (E o p) of [Ru(bpy)3J2+ and related complexes.3 2'4 0 For the terthiophene complexes 47-49, conversion to the P,C bonding mode contracts the AE values and red shifts the M L C T transitions 56-63 nm to also result in lower values for E o p . Since the energies obtained from the 123 electrochemical and optical measurements correlate, this is further corroboration that Ru—»bpy M L C T transitions occur. The AE and E o p values of 51 do not correlate with 47-49 and the charge transfer is thought to be by a different process which is not surprising considering that the first oxidation is not assigned as the R u i n / " redox couple. Mixing of nearly degenerate pentathiophene- and Ru(II)-based orbitals could also result in a M L C T transition that cannot be compared to pure metal to ligand based transitions. The low energy bands observed for all P,C bound complexes are assigned as transitions to a low-lying state that is mostly triplet in character. Transition to a low-lying "forbidden" level is observed as a relatively strong transition in osmium polypyridine complexes 2 9 , 3 0 ' 4 1 due to spin ")f> A") A"X 94- 9A 9ft orbit coupling, ' ' and as a much weaker transition for the [Ru(bpy)3] complex " and some 94- 9Q "30 [Ru(bpy)2(LL)] complexes. ' This transition, due to its very weak intensity, is often covered by superimposed, allowed M L C T transitions that are much stronger in intensity. In the P,C bonding mode, these transitions occur at long wavelengths to form a broad shoulder extending to 700 nm so that the combined spectra cover the visible range, similarly to Ru(bpy)2L2 complexes that have been synthesized as black absorbers for possible applications in photovoltaic cells. 4 4 ' 4 5 The anionic character of the cyclometallated (phosphino)oligothiophene ligands stabilizes the excited states by electron donation to Ru 1 1 1 to shift the bands, a trend that has been observed in Ru/Os polypyridine complexes.30'41 The overall effect of stabilizing the Ru 1" excited state, a reduction in the M L C T absorption energy, is equivalent to incorporating a bpy-type acceptor ligand.4 6 9+ In Ru(bpy)2L2 -type complexes, while absorption is primarily to a singlet-based M L C T state, emission is from a triplet-based M L C T state, with spin-orbit coupling mixing the singlet and triplet states.2 9'4 1'4 7'4 8 Ru(II)-P,C complexes 47-49 display very weak luminescence that approaches the near infrared region (754-772 nm). These singlet-based M L C T states should be low enough in energy to create a barrier to a low-lying metal centered (MC) level, and the formation of a more electron-rich complex has been observed to promote luminescence in another Ru(bpy)2L22+ switchable system.49 However, very low intensity emission was observed, and could be due to competing non-radiative relaxation that increases with increasing vibrational modes introduced by the addition of alkyl substituents to the ligands. The short 22 ± 2 ns lifetime observed for the emission of 47 is in agreement with the presence of competing deactivation pathways, while the decreasing intensity from 47 to 49 is also in accordance with the energy gap law. 5 0 ' 5 1 An alternative explanation for the very weak, or lack of, luminescence observed for all of the Ru(II)-P, C complexes is that the M L C T excited state is being quenched 124 by electron transfer from the oligothiophene to the Ru(III) center that is formed by light absorption, or is partially quenched by derealization of positive charge onto the thienyl chains i f orbital overlap is favorable. When the emission spectra are compared to those of the corresponding Ru(II)-/J,5' complexes, the (phosphino)terthiophene complexes exhibit P,S-OF¥, P . C - O N emission, while the (phosphino)pentathiophene complex exhibits P,S-ON, P ,C-OFF emission. #3 7K 0>2 ~f. 7K Figure 5-13 Proposed energy-level diagram and transitions, Ti and T2, for T n , where energy levels O1-O3 represent the highest fully occupied, singly-occupied and unoccupied antibonding MO's, respectively. Adapted from reference3 . The results of the spectroelectrochemistry studies can be used to further characterize the oxidized states of the Ru(II)-P,C complexes, and suggest the ordering of the energy levels for these metal-thiophene complexes. Bleaching of the M L C T transitions is observed with the initial oxidation of all complexes, so it can be inferred in each case that M L C T transitions involve the species that is oxidized first. Interestingly, all spectroelectrochemical spectra suggest oxidation of the oligothiophene chains, including the (phosphino)terthiophene complexes that display metal-based radicals by EPR spectroscopy. Excluding dimer formation, which is not expected to occur for the Ru(II) complexes, oligothiophene chains are expected to undergo two possible transitions: the first transition (Ti) is from a singly occupied molecular orbital (SOMO) to an empty antibonding M O , and a lower-energy transition (T2) is expected from the highest fully occupied M O to the SOMO (Figure 5-13).36 The higher-energy Ti transitions of Tj+ chains match those at 557-577 nm for 4 7 + - 4 9 + , while the Ti transition observed for Ts + chains is similar 125 in energy to the transition at 649 nm 51+. The lower-energy T 2 transitions could be obscured in the measured spectra, or weak in intensity. The oxidized Ru-(phosphino)terthiophene-/>, C complexes 47+-49+ also display transitions (632-663 nm) characteristic of metal-based bpy—»Ru(III) L M C T transitions,33'34 and low-energy bands that may be PMenT3—^Ru(III) L M C T transitions. The oxidized (phosphino)pentathiophene complex 51+ does not display a transition in this region, consistent with the pentathiophene chain as the first species oxidized. If the highest occupied level of 51+ is expected to be the singly occupied molecular orbital (SOMO) of the PD02T5 ligand, then the proposed M O diagram for the complex, including orbital overlap with the Ru(II) centre, is as depicted in Figure 5-14. The highest-energy transition (T'i = 1.55 eV, as estimated from the band edge) would correspond to a transition from the SOMO-based *D3 to the empty n antibonding M O O 4 . T' 2 (1.13 eV) is tentatively assigned as a transition from the occupied n M O <Di to <I>3, which is higher in energy from that observed for 56+.36 T'3 (0.69 eV) could correspond to a transition from the Ru(II)-based orbital 0 2 to the closely spaced 0 3 . PDO 2T 5(TC*) PDO 2T 5(TC) A SOMO 1 PDo 2 T 5 ( :-) + + 7FT <]>4 T', T' 3 7K <E>3 T' 2 0>i Ru(dTt) Figure 5-14 Proposed molecular orbital diagram for 51+ and transitions T ' i , T' 2 and T'3, with energy levels O1-O4. 126 When the data is taken as a whole, it is interesting to consider that the ordering of the Ru(II) and PD02T5 energy levels can be controlled by the type of coordination mode, and extension of the oligothiophene chain. Due to evidence that with a Ru(bpy)2 -based M L C T transition, positive charges could be introduced into the thienyl backbone complex 51, use of extended complexes of this type as a photoconducting molecular wire as depicted in Figure 5-15 can be visualized. Further investigation into the formation of a charge-separated state with light absorption could be carried out by observing the excited state lifetime via transient absorption spectroscopy. The stability in air and under U V and visible light of crystalline and drop-cast solid-state samples of the Ru(II) P,C complexes encourages further studies of this class of complexes for real applications. Complexes containing longer oligothiophene chains or polythiophene would likely improve the processibility and flexibility of materials of this type. / Figure 5-15 Representation of Ru(II)-phosphino(pentathiophene)-P,C complex 51 as a light-harvesting molecular wire. The M L C T transition results in charge separation and p-doping of the oligothiophene chain, so that the chain can conduct current. 5.5 Conclusions Binding Ru(II) bis(bipyridyl) groups to P-phosphinothiophenes via the P,C bonding mode results in very different electrochemical and spectroscopic properties compared to the Ru(II)-P,5' complexes. Due to a combination of a more electron rich system and increased planarity of the oligothiophene chains with P,C bonding, red shifts are observed for the thienyl TC—>u* transitions, resulting in transition energies very close to those of the ligands, and the thienyl oxidations are cathodically. shifted to potentials lower in value to those of the 127 corresponding ligands. Hence, direct metal-thiophene bonding in the P,C bonding mode does not hinder 71-orbital overlap, as is supported by the solid-state structures obtained. Interaction of the Ru(bpy)2 2 + groups with the P-(phosphino)oligothiophene ligands in the P,C bonding mode results in low-energy M L C T transitions and a broad absorption region, a desirable characteristic of potential light-harvesting materials. Extension of the oligothiophene chain length is shown to result in a reordering of the energy levels, as is inferred from the results of cyclic voltammetry, EPR, optical spectroscopy, and spectroelectrochemistry experiments. MLCT-based emission observed is very weak and is considered to be essentially quenched. Complex 51 exhibits the most promising characteristics for molecular device applications. In contrast to the Ru(II)-P,5' complexes, the P.C-bound complex 51 displays a thienyl-based first oxidation, which could allow oxidative doping of a pentathiophene chain concurrent with Ru—>bpy M L C T transition. The solid state stability of the complexes encourage their investigation as molecular conducting materials. Reversible switching between the P,S and P,C bonding modes, combined with the different physical properties compared with the P,S complexes, result in potential application of these complexes as conductive switches. 5.6 References (1) Liao, Y . ; Bohne, C. J. Phys. Chem. 1996, 100,134-143. (2) Salbeck, J. J. Electroanal. Chem. 1992, 340, 169-195. (3) Altomare, A. ; Burla, M . C ; Camalli, M . ; Cascarano, G. L . ; Giacovazzo, C ; Guagliardi, A. ; Moliterni, A . G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115-119. (4) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de Gelder, R.; Isreal, R.; Smits, J. M . M . "The DIRDIF-94 program system, Technical Report of the Crystallography Laboratory," University of Nijmegen, The Netherlands, 1994. (5) d*TREK Area Detector Software, Version 7.11: Molecular Structure Corporation, 2001. (6) teXsan Crystal Structure Analysis package: Molecular Structure Corporation, 1985 & 1992. (7) SAINT, Version 6.02: Bruker A X S Inc., Madison, Wisconsin, USA, 1999. (8) SADABS Bruker Nonius area detector scaling and absorption correction, Version 2.05: Bruker A X S Inc., Madison, Wisconsin, USA. (9) SHELXL-97: Sheldrick, G. M . , University of Gottingen, Germany, 1997. (10) Moorlag, C ; Clot, O.; Wolf, M . O.; Patrick, B. O. Chem. Commun. 2002, 3028-3029. (11) Moorlag, C ; Wolf, M . O.; Bohne, C ; Patrick, B. O. J. Am. Chem. Soc. 2005, 727, 6382-6393. 128 Gaussian 03: M . J. Frisch, G. W. T., H . B. Schlegel, G. E. Scuseria,; M . A. Robb, J. R. C , J. A . Montgomery, Jr., T. Vreven,; K. N . Kudin, J. C. B. , J. M . Millam, S. S. Iyengar, J. Tomasi,; V . Barone, B. M . , M . Cossi, G. Scalmani, N . Rega,; G. A . Petersson, H. N . , M . Hada, M . Ehara, K . Toyota,; R. Fukuda, J. H. , M . Ishida, T. Nakajima, Y . Honda, O. Kitao,; H . Nakai, M . K. , X . L i , J. E. Knox, H. P. Hratchian, J. B. Cross,; C. Adamo, J. J., R. Gomperts, R. E. Stratmann, O. Yazyev,; A . J. Austin, R. C , C. Pomelli, J. W. Ochterski, P. Y . Ayala,; K . Morokuma, G. A . V. , P. Salvador, J. J. Dannenberg,; V . G. Zakrzewski, S. D., A . D. Daniels, M . C. Strain,; O. Farkas, D. K . M . , A . D. Rabuck, K. Raghavachari,; J. B. Foresman, J. V. O., Q. Cui, A . G. Baboul, S. Clifford,; J. Cioslowski, B. B. S., G. Liu, A . Liashenko, P. Piskorz,; I. Komaromi, R. L. M . , D. J. Fox, T. Keith, M . A . Al-Laham,; C. Y . Peng, A . N . , M . Challacombe, P. M . W. Gil l , ; B. Johnson, W. C , M . W. Wong, C. Gonzalez, and J. A . Pople, Gaussian, Inc., Pittsburgh PA, 2003. Becke, A . D. J. Chem. Phys. 1993, 98, 5648-5652. Lee, C ; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter 1988, 37, 785-789. Harris, S. Polyhedron 1997, 16, 3219-3233. Carter, E. A. ; Goddard, W. A. , Ill y. Am. Chem. Soc. 1986, 108, 2180-2191. Watanabe, M . ; Murata, K.; Ikariya, T. J. Am. Chem. Soc. 2003, 125, 7508-7509. Bibal, C ; Pink, M . ; Smurnyy, Y . D.; Tomaszewski, J.; Caulton, K. G. J. Am. Chem. Soc. 2004, 126, 2312-2313. Curtis, M . D.; Cao, J.; Kampf, J. W. J. Am. Chem. Soc. 2004, 126, 4318-4328. Yamamoto, T.; Komarudin, D.; Maruyama, T.; Arai, M . ; Lee, B.-L. ; Suganuma, H. ; Asakawa, N . ; Inoue, Y . ; Kubota, K. ; Sasaki, S.; Fukuda, T.; Matsuda, H. J. Am. Chem. Soc. 1998, 120, 2047-2058. Kalyanasundaram, K . Photochemistry of Polypyridine and Porphyrin Complexes; Academic Press: London, 1992. Lever, A . B. P. Inorg. Chem. 1990, 29, 1271-1285. Coudret, C ; Fraysse, S.; Launey, J. Chem. Commun. 1998, 663. Pilbrow, J. R. Transition Ion Electron Paramagnetic Resonance; Oxford Science Publications: Oxford, 1990; Vol . I. DeSimone, R. E.; Drago, R. S. J. Am. Chem. Soc 1970, 92, 2343-2352. Lytle, F. E.; Hercules, D. M . J. Am. Chem. Soc. 1969, 91, 253-257. Felix, F.; Ferguson, J.; Guedel, H. U . ; Ludi, A. Chem. Phys. Lett. 1979, 62, 153-157. Felix, F.; Ferguson, J.; Guedel, H. U . ; Ludi, A. J. Am. Chem. Soc. 1980, 102, 4096-4102. Caspar, J. V. ; Meyer, T. J. Inorg. Chem. 1983, 22, 2444-2453. Mamo, A . ; Stefio, I.; Poggi, A. ; Tringali, C ; Di Pietro, C ; Campagna, S. New J. Chem. 1997,27, 1173-1185. Becker, R. S.; de Melo, J. S.; Macanita, A . L. ; Elisei, F. J. Phys. Chem. 1996, 100, 18683-18695. Dodsworth, E. S.; Lever, A. B. P. Chem. Phys. Lett. 1986, 124, 152-158. 129 Nazeeruddin, M . K. ; Zakeeruddin, S. M . ; Kalyanasundaram, K . J. Phys. Chem. 1993, 97, 9607-9612. Scott, S. M . ; Gordon, K. C. Inorg. Chim. Acta 1997, 254, 267-272. Hi l l , M . G.; Penneau, J. F.; Zinger, B.; Mann, K. R.; Miller, L. L. Chem. Mater. 1992, 4, 1106-1113. Bauerle, P.; Segelbacher, U . ; Maier, A. ; Mehring, M . J. Am. Chem. Soc. 1993, 115, 10217-10223. DiCesare, N . ; Belletete, M . ; Marrano, C.; Leclerc, M . ; Durocher, G. J. Phys. Chem. A 1998, 102, 5142-5149. Constable, E. C ; Dunne, S. J.; Rees, D. G. F.; Schmitt, C. X . Chem. Commun. 1996, 1169-1170. Beley, M . ; Collin, J. P.; Louis, R.; Metz, B.; Sauvage, J. P. J. Am. Chem. Soc. 1991, 113, 8521-8522. Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 77, 3334-3341. Mamo, A. ; Juris, A. ; Calogero, G.; Campagna, S. Chem. Commun. 1996, 1225-1226. Kober, E. M . ; Meyer, T. J. Inorg. Chem. 1984, 23, 3877-3886. Kober, E. M . ; Meyer, T. J. Inorg. Chem. 1983, 22, 1614-1616. Anderson, P. A. ; Strouse, G. F.; Treadway, J. A. ; Keene, F. R.; Meyer, T. J. Inorg. Chem. 1994, 33, 3863-3864. Anderson, P. A. ; Keene, F. R.; Meyer, T. J.; Moss, J. A . ; Strouse, G. F.; Treadway, J. A . J. Chem. Soc. Dalton Trans. 2002, 3820-3831. Rillema, D. P.; Mack, K. B. Inorg. Chem. 1982, 27, 3849-3854. Walters, K . A. ; Touillet, L. ; Guillerez, S.; Schanze, K . S. Inorg. Chem. 2000, 39, 5496-5509. Kober, E. M . ; Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1984, 23, 2098-2104. Wang, K. -Z. ; Gao, L . -H. ; Bai, G.-Y.; Jin, L.-P. Inorg. Chem. Commun. 2002, 5, 841-843. Caspar, J. V. ; Meyer, T. J. Am. Chem. Soc. 1983, 105, 5583-5590. Caspar, J. V. ; Meyer, T. J. / . Phys. Chem. 1983, 87, 952-957. 130 CHAPTER 6 Preliminary Polymerization Studies 6.1 Introduction Polynuclear metallated polythiophenes are expected to exhibit properties different from those of mononuclear metallated oligothiophenes, since the interaction between metal centers and the conjugated backbone could be enhanced by multiple points of contact with metal centers. Also of interest to study are the interactions between metal centers that could occur across a rc-conjugated thiophene chain acting as a molecular wire. Electronic interactions between metal centers can be used to probe the distance the electronic modification can be "sensed" along a conjugated molecule.1"4 Since the Ru(II)-phosphinothiophene complexes that have been described displayed properties very different from those of oligothiophenes, it is of interest to observe the properties of the polynuclear Ru(II)-phosphinothiophenes. Therefore, multinuclear metallated counterparts of the Ru(II)-phosphinothiophene-P,5' or Ru(II)-phosphinothiophene-.P, C complexes that incorporate Ru(bpy) 2 2 + groups were targeted for study. These systems should also be good candidates for studying metal-metal interactions across rc-conjugated chains, due to strong interactions between the bound metal group and the conjugated backbones. There are two general methods for preparing polynuclear metallated oligo- or polythiophenes. Most commonly, ligands are prepared with multiple points for metallation along the thiophene chains,5'6 and this approach allows the coordination of different metals. An alternative route is the coupling of mononuclear complexes via functionalities on a thienyl ligand, which has the advantage of better controlling the composition of the polymeric product, though successful coupling is highly dependant on the reactivity of the metal center during the polymerization reaction. Reported methods of polymerization of Ru(II) complexes via functionalized ligands include Ni(0)-catalyzed homocoupling, ' Stille coupling, and Suzuki coupling.9 Both routes towards metallated polymers are investigated for this project and preliminary polymerization reactions have been performed. In Chapter 2, compound 26 (Chart 6-1) was described that is functionalized with iodo substituents at both a positions and with a bromo substituent at one central (3 position. These iodo substituents exhibit higher reactivity than the bromo substituent, as was demonstrated by a Suzuki cross-coupling reaction of 26 to yield the 131 (3-bromo functionalized heptamer 28 (Chapter 2). The preparation of P-brominated polythiophene chains via Suzuki cross-coupling reactions of 26 with a thiophene diboronic ester to give poly-26T (Chart 6-2) was attempted. In this case, the thiophene ring of the diboronic ester functions as a bridging group. The desired polymer poly-26T incorporates alkyl chains and unsubstituted p positions that could bind to an open site of a complexed metal. Chart 6-1 The second route investigated uses Ru(II) complexes as building blocks. Polymerization of the P,5-bound complex was first attempted via Ni(0)-catalyzed homocoupling of the a,a-dibrominated Ru(II) complex 46. This polymerization procedure is the most direct of those reported and does not require cross-coupling with a bridging group; successful coupling of 46 would yield poly-46 (Chart 6-2). As both monomers, 26 and 46, are not symmetric, polymers 132 would not be regioregular as depicted, but all possible coupling orientations would occur for both polymerization routes: head-to-head (HH), head-to-tail (HT) and tail-to-tail (TT). 6.2 Experimental 6.2.1 General Experimental Reactions were performed using standard Schlenk techniques with dry solvents under nitrogen unless otherwise specified. Procedures to prepare 26 and RuPBr2Do2T5-P,S' (46) have been described in previous chapters. [NiBr2(PPh3)2] catalyst,10 activated zinc dust,10 and 2-thienyl-l,3,2-dioxaborinane11 were prepared by published procedures. A l l other reagents were purchased from Aldrich or Strem Chemicals. ' H and 3 1 P{ 1 H} N M R experiments were performed on either a Bruker AV-300 or Bruker AV-400 spectrometer, and spectra were referenced to residual solvent ('H) or external 85% H3PO4 ( 3 1P). Microanalyses (C, H, N) and mass spectrometry experiments were performed at U B C by M . Lakha and M . Lapowa, respectively. 6.2.2 Polymerization Reactions Suzuki coupling, 1:1 ratio: Poly-3 "-bromo-3,3 ""-didodecyl-2,2':5'2":5 "2 "':5 '"2 "":5 ""2 ""'-hexathiophene (poly-26T) A solution of 26 (1.00 g, 0.926 mmol), CsF (647 mg, 4.26 mmol) and [Pd(PPh3)4] (53.2 mg, 0.046 mmol) was stirred in THF (30 mL) at reflux, and a solution of 2,5-bis(l,3,2-dioxaborolan-2-yl)-thiophene (233 mg, 0.926 mmol) in THF (10 mL) was added to the mixture. After 16 h, the solution had darkened from dark yellow to a red-purple suspension. CH2CI2 was added and purple-green solids were filtered off. The solvent was removed to give an orange-red residue. The crude product was run through a chromatographic column using silica gel and CH2Cl2-hexanes as eluent. Starting material was recovered (245 mg, 25%) followed by multiple fractions that varied in colour between orange and dark red but were not isolated in significant amounts. Suzuki coupling, excess diboronic ester: Poly-3 "-bromo-3,3 ""-didodecyl-2,2':5 '2 ":5 "2"': 5 '"2 "":5 ""2 '""-hexathiophene (poly-26T) A solution of 26 (1.00 g, 0.926 mmol), CsF (1.21 g, 7.95 mmol) and [Pd(PPh3)4] (1.00 mg, 0.087 mmol) was stirred in THF (30 mL) at reflux, and a solution of 2,5-bis(l,3,2-133 dioxaborolan-2-yl)-thiophene (436 mg, 1.74 mmol) in THF (10 mL) was added to the mixture. After 24 h, the solution had darkened from dark yellow to a dark red mixture. The solvent was removed to give red-purple solids that were transferred to a Soxhlet finger. Soxhlet extraction was carried out by successive extractions.with EtOH, hexanes, CH2CI2, and C H C I 3 . Dark orange solids (0.236 g) were collected from the hexanes fraction, dark red solids (0.197 g) were collected from the CH2CI2 fraction, and shiny greenish crystals that had a strong odour and are likely catalyst by-product were then collected from the CHCI3 fraction, leaving no remaining solids in the Soxhlet finger. The crude product of the CH2CI2 fraction displayed evidence of containing coupled oligothiophenes by ! H N M R and mass spectroscopy. Estimated yield (monomer M W = 908): 0.197 g (23%). ' H N M R (400.1 MHz, CDCI3): 5 (broad peaks) 7.358 (m, 1H), 7.199 (m, 1H), 7.145 (m, 1H), 7.095 (m, 1H), 7.066 (m, 2H), 7.034 (s, 1H), 6.996 (m, 2H), 2.750 (m, 4H), 1.664 (m, 4H), 1.390 (m, 4H), 1.240 (m, 32H), 0.853 (m, 6H). Nickel-catalyzed cross-coupling, 2 h reaction: Poly-[Bis(2,2 -bipyridyl)(3,3 ""-didodecyl-3 "-diphenylphosphino-2,2 ':5 '2 ":5 "2 '":5 '"2'"'' -pentathiophene-P\S)ruthenium(II)] [bis(hexafluorophosphate)] (poly-46) [NiBr 2(PPh 3) 2] (4.3 mg, 8.9 umol), zinc dust (5.6 mg, 0.086 mmol) and NMe 4 I (11.6 mg, 0.0576 mmol) was stirred for 1 h under nitrogen in THF (10 mL) at 50°C. 46 (50.0 mg, 0.0288 mmol) dissolved in THF (10 mL) was added and the mixture was stirred for 2 h at 50°C. The solution changed colour from orange to orange-brown. The solution was cooled and condensed to 10 mL, and added dropwise by pipette into a solution of N H 4 P F 6 (3.00 g, 18.0 mmol) in H 2 O (60 mL) to form a dark orange precipitate. The suspension was filtered and washed with H 2 O , then dissolved in acetone and the green catalyst and zinc were filtered off. The red filtrate was collected and removed of solvent. The solids were recrystallized from EtOH to give a dark orange precipitate. 3 1 P{'H} N M R (162.0 MHz, CO(CD 3 ) 2 ): 5 27.1 (s), 28.1 (s), -143.0 (sep, J P F = 708 Hz, PF 6). Nickel-catalyzed cross-coupling, 16 h reaction: Poly-[Bis(2,2 -bipyridyl)(3,3 ""-didodecyl-3 "-diphenylphosphino-2,2 ':5 ',2 ":5 "2 "':5 '",2 ""-pentathiophene-P,S)ruthenium(II)] [bis(hexafluorophosphate)] (poly-46) [NiBr 2(PPh 3) 2] (21.5 mg, 44.7 umol), zinc dust (28.0 mg, 0.428 mmol) and NBU4I (53.2 mg, 0.144 mmol) was stirred for 2 h under nitrogen in THF (10 mL) at 50°C, and the mixture changed from light amber to dark brown in colour. 46 (50 mg, 0.144 mmol) dissolved in THF (10 mL) was added and the mixture was stirred for 16 h at 50°C. The solution was cooled 134 and condensed to 10 mL, and added dropwise by pipette into a solution of NH4PF6 (3.00 g, 18.0 mmol) in H 2 O (60 mL) to form a dark red precipitate. The suspension was filtered, and the dark red precipitate (86 mg) was recovered and dried. 3 1 P{'H} N M R (162.0 MHz, CO(CD 3) 2): 8 45.42 (s), 44.65 (s, minor peak), 44.32 (s, minor peak), -143.0 (sep, J P F = 708 Hz, PF 6). The dark red precipitate was dissolved in acetone, catalyst and zinc solids were filtered off, and the solvent was removed and the resulting orange-red solids (53 mg) were dried. Multiple peaks were observed by 3 1 P{'H} N M R (162.0 MHz) in CO(CD 3 ) 2 solution between 45.4-20.0 ppm and a signal at -143.0 (sep, JPF = 708 Hz) due to [PFe]~ (see Results section). 6.3 Results 6.3.1 Coupling of (3-Halogenated Oligothiophenes Palladium catalyzed Suzuki cross-coupling of 26 with the diboronic ester 2,5-bis(l,3,2-dioxaborolan-2-yl)-thiophene was carried out via non-aqueous conditions using CsF as the activating base, as shown in Scheme 6-1. Reaction in a 1:1 ratio of 26 to the diboronic ester resulted in the recovery of significant starting material (25%). Therefore, the reaction was repeated with approximately a 1:2 ratio of 26 to the diboronic ester, and a colour change to a dark red colour indicated the formation of longer oligo- or polythiophene chains. Due to the possibility of chains of varying length, a Soxhlet extraction was carried out to separate the fractions. Soluble salts were extracted with MeOH, and any remaining starting materials, short oligomers and silicon grease were subsequently extracted with hexanes. Dark red solids were then extracted with CH 2 C1 2 , and the amount recovered was estimated to correspond to a yield of 23% if all ends of the starting materials have reacted to form a polymer chain. Scheme 6-1 1) 10 eq CsF 0.05 eq Pd (PPh3)4 THF 26 poly-26T Reflux 2 h 135 7.40 7.30 7.20 7.10 — | i i . 7.00 ppm Figure 6-1 ' H N M R spectra of the aromatic region of (a) heptathiophene 28 and (b) the coupled products of the Suzuki cross-coupling reaction of 26 and thiophene (T), to give poly-26T. The ' r i N M R spectrum of the dark red solids recovered displayed broad peaks that are characteristic of oligomeric or polymeric thiophene chains (Figure 6-lb), at very different positions from those of the spectrum of 26 that does not display any peaks between S 7.346-7.093. In fact, the coupled product displays peaks that are very similar in chemical shift and relative integration to those of the p-brominated heptamer 28 (Figure 6-la), suggesting that thienyl coupling has occurred at the a-iodinated positions. 136 Chart 6-3 n = 1 M = 906/908 n = 2 2M = 1812/1814/1816 n = 3 3M = 2718/2720/2722/2724 n = 4 4M = 3624/3626/3628/3630/3632 0 0 T = 82 —I 127 B - BE = 85 0 .10 J 1771.1 1089. 1253 1818.1 2082.1 1999.2 1900.1 | 1982.1 I 2163. 2727.0 2910.0 2644.0 [ | 2807.9/2807.9 II 2990.9 3739.7 3656.6 1 3820.5 m /z Figure 6-2 MALDI-TOF mass spectrum of poly-26T, 700-4000 m/z region. The results of mass spectroscopy indicate that long oligothiophene chains are present in the cross-coupled product. Chart 6-3 displays the expected mass corresponding to 1-4 monomer units of coupled materials, as well as the mass of end groups that may be present. The spectrum displayed in Figure 6-2 shows four clusters of peaks that appear to correspond generally to the masses of 1 -4 pentamer units. The main peak at 1171 is not the expected mass for the starting material, and may be a fragmentation product. Peaks listed in Table 6-1 can be assigned to 137 specific oligothiophenes, though as thiophene (T) and the boronic ester (BE) have similar masses, the boronic esters listed could also be represented by boronic ester-terminated oligothiophenes. The peak at m/z = 2727 is most likely due to a boronic ester-terminated chain of seventeen thiophene rings. While all peaks cannot be assigned, they are generally spaced by m/z = 82, indicating the fragmentation of thiophene units. Table 6-1 Peaks corresponding to coupled oligomers in the mass spectrum of poly-26T. Peak m/z Oligothiophene 2M(1816) + 2 1818.1 2M(1816) + T + 2 1900.1 3M (2724) - T + 2 2644.0 3M (2723) - T + 2727.0 BE + 1 3M (2724) + T + 2 2807.9 6.3.2 Coupling of Ru(ll)-Complexes Ni(0)-catalyzed homocoupling was attempted for the T^S-bound Ru(II) complex 46 that incorporates two bromo substituents at the a positions of the pentathiophene ligand (Scheme 12 6-2). The reaction also requires the presence of zinc dust and an alkylammonium iodide. Initially, the coupling reaction was carried out for only 2 h at 50°C, as per the reported procedure, and the solution darkened upon addition of the Ru(II) complex. However, mainly starting material was recovered from the reaction products. By 3 1 P N M R spectroscopy, the starting material, with a peak at 5 27.1, was shown to be the major product recovered, but a minor peak at 8 28.1 ppm was also observed (Figure 6-3a). 138 Scheme 6-2 (a) 27.1 28.1 28.5 28.0 27.5 27.0 — 1 — ., , 1 — J. 1 1 50 ppm 1 i i 0 i i 1 , -50 f -100 1 -150 (b) 28.0 45.4 33.8 ,32.3 y 20.0 I27'4 I ' i " " i 1 1 1 1 1 " " i " " r " 1 1 " 45 40 35 30 25 20 ' I r 50 ppm -i i , p 0 Figure 6-3 3 1 P N M R (162.0 MHz) spectra in CO(CD 3 ) 2 of products of Ni(0) coupling reaction of 46 at50°C (a) 2 h, (b) 16 h. ~1~ -50 "i 1 r -100 •150 The coupling reaction was then attempted for a 16 h reaction time, and using greater equivalents of catalyst, zinc, and alkylammonium iodide than shown in Scheme 6-2 (~5 times 139 excess). Orange-red solids were recovered that did not contain the starting material 46 by 3 1 P N M R spectroscopy; however, many peaks were observed in the spectrum, and most were removed from the range of chemical shifts expected for a polymerized product (Figure 6-3b). The peak at 5 20.0 suggests the presence of oxidized ligand, and the peak at 8 45.4 could indicate a cyclometallated product. While the major peak at 8 28.0 is near the chemical shift of 46, the peak is not broadened, as would be expected for a polymer. A small, broad peak is present upfield of the peak at 5 27.4; however, this product would comprise a very small amount of the crude product. 6.4 Discussion As the palladium-catalyzed Suzuki cross-coupling of 26 to give 28 had been successful (Chapter 3), the same non-aqueous conditions were attempted to yield poIy-26T. A 1:1 ratio of 26 to the diboronic ester was not sufficient for all of the starting material to react. When a ratio close to 1:2 was used, all of the starting material was reacted, and broad peaks in the ] H N M R spectrum signaled that long oligomers may have formed. Mass spectrometry also suggests the formation of coupled pentamer chains, linked by thiophene bridging units. One uncertainty is that the central P-brominated position may also couple, and 1 3 C experiments could be carried out to determine substitution along the thienyl backbone. Ni(0)-catalyzed homocoupling of ruthenium complexes was first attempted, as it is a simple procedure requiring only a dihalogenated substrate, and has been reported to be effective for the coupling of brominated aromatic ligands of Ru(II) complexes to yield dinuclear complexes.7'13 Suzuki coupling reactions were not attempted for the .PS-bound complex 46 due to the possibility of conversion to the P,C bonding more during reaction in the presence of base. The multiple peaks observed by 3 1 P N M R spectroscopy after coupling for 16 hours indicate that ligand oxide and a variety of metallated products have formed, and only a small broad peak is present that could indicate polymerized product. The intensities of the peaks observed are very small compared to those of the PF6~ anion, suggesting further that many products are formed in small amounts. The main disadvantage of a polymerization route from 46 to make poly-46 is that the metal center can react and this is most likely occurring for the Ni(0)-catalyzed coupling reactions. The formation of a peak at 45.4 ppm by 3 1 P N M R spectroscopy suggests that a Ru-C bound complex is formed. Neither reaction, for a shorter and longer reaction duration, indicated the majority formation of a polymerized product, since a strong, broadened peak was not 140 observed. Other coupling reactions of metal complexes should be attempted. The P,C-bound Ru(II) complex 52 could provide a better substrate for Suzuki cross-coupling that has proven to be effective for oligothiophenes, since this complex should not be sensitive to reaction with base. 6.5 Conclusions The preliminary results of the polymerization reactions performed to date indicate that Suzuki cross-coupling of p-brominated pentathiophene chains that are iodinated at the a positions is a viable route towards the preparation of a multi-functionalized ligand, and subsequently, polynuclear metallated polymers. While the polymerization of Ru(II) complexes has been attempted without success, other coupling reactions such as Suzuki cross-coupling should be attempted, possibly using the P, C-bound complexes. Both polymerization routes have the inherent disadvantage that the monomers are non-symmetrical and the structure of a resulting metallated polymer cannot be controlled. 6.6 References (1) Stott, T. L. ; Wolf, M . O. Coord Chem. Rev. 2003, 246, 89-101. (2) Pappenfus, T. M . ; Mann, K. R. Inorg. Chem. 2001, 40, 6301-6307. (3) Laye, R. H. ; Couchman, S. M . ; Ward, M . D. Inorg. Chem. 2001, 40, 4089-4092. (4) Cameron, C. G.; Pickup, P. G. J. Am. Chem. Soc. 1999, 121, 11773-11779. (5) Constable, E. C ; Rees, D. G. F. Polyhedron 1998, 77, 3281-3289. (6) Collin, J. P.; Laine, P.; Launay, J. P.; Sauvage, J. P.; Sour, A . Chem. Commun. 1993, 434-435. (7) Griffiths, P. M . ; Loiseau, F.; Puntoriero, F.; Serroni, S.; Campagna, S. Chem. Commun. 2000, 2297-2298. (8) Trouillet, L. ; De Nicola, A. ; Guillerez, S. Chem. Mater. 2000, 72, 1611-1621. (9) Chodorowski-Kimmes, S.; Beley, M . ; Collin, J.-P.; Sauvage, J.-P. Tetrahedron Lett. 1996, 37, 2963-2966. (10) Iyoda, M . ; Otsuka, H. ; Sato, K.; Nisato, N . ; Oda, M . Bull. Chem. Soc. Jap. 1990, 63, 80-87. (11) Kobayashi, Y . ; Mizojiri, R.; Ikeda, E. J. Org. Chem. 1996, 61, 5391-5399. (12) Johansson, K. O.; Lotoski, J. A. ; Tong, C. C ; Hanan, G. S. Chem. Commun. 2000, 819-820. (13) Fanni, S.; Di Pietro, C ; Serroni, S.; Campagna, S.; Vos, J. G. Inorg. Chem. Commun. 2000, 3, 42-44. 141 CHAPTER 7 Conclusions and Future Directions 7.1 Conclusions This thesis describes the synthesis and study of P-phosphinothiophene ligands and derivatives, and Au(I)-, Pd(II)-, and Ru(II)-phosphinothiophene complexes. Modifications of the structural, chemical, and electronic properties of the ligands and the new metal-oligothiophene hybrid materials prepared were observed. This section summarizes the results of this study, and gives conclusions based on the overall data obtained, with reference to the initial goals. Mono- and bisubstituted P-phosphinothiophenes of n = 2-5 thiophene rings were used as coordinating groups to anchor metals to oligothiophene chains. Two adjacent P-phosphine substituents alter the structural and electronic properties of the oligothiophene backbones, while one P-phosphine substituent exerts little influence. The mediation of electronic properties by extension of thienyl chain-length is also observed. P-Bis(phosphino)thiophene ligands present binding sites where oligothiophene properties can be altered via metallation, which is seen with Pd(II) or Au(I) complexation. Strong conformational effects are observed in the solid state complexes as unfavorable interannular torsion angles, where the choice of metal controls the extent of Ti-orbital overlap. Au-Au bonding does not likely persist in solution, but Au(I) complexes could be used to "switch" conductivity using the weak Au-Au interactions. The H O M O - L U M O gaps are increased for both complexation types via inductive electron-withdrawing effects, and additionally by the fixed metallocyclic conformations of the Pd(II) complexes. Pd(II) complexation also introduces L M C T charge transfers and metal-based transitions. Overall, metal complexation with P-bis(phosphino)thiophene generally discourages 7x-orbital overlap across the organic chain, but the introduction of "switchable" bonds such as Au-Au interactions may prove useful to mediate conjugation. Mono-substituted P-phosphinothiophene ligands provide anchored metals an opportunity to directly interact with oligothiophene chains. New P-(phosphino)pentathiophene ligands were prepared by selective substitutions of a,P-functionalized derivatives that display a narrow H O M O / L U M O gap. Binding bis(bipyridine) Ru(II) groups to P-phosphinothiophenes via direct, metal-sulfur (P,S) and metal-P-carbon (P,C) thienyl bonding modes affect the structural and 142 electronic properties of oligothiophenes, and reversible switching between the P,S and P,C bonding modes is mediated by Ru(II)-promoted acid-base chemistry. Thienyl backbone conformations appear favorable from the solid-state structures of both bonding modes, and comparatively, the P,C bonding mode displays increased planarity. The Ru(II)-/J,5' complexes exhibit decreased conjugation via electronic inductive effects of the bound Ru(II) metal, while the electron rich oligothiophene moieties of the Ru(II)-.P,C complexes are equivalent electronically to the corresponding (3-phosphinothiophenes. The bound bis(bipyridine) Ru(II) group exhibits M L C T absorption and emission transitions that are affected by the bonding mode and thienyl chain length. Destabilization of the Ru(II) center in the P,S bonding mode increases the energy of M L C T transitions compared to the P,C bonding mode that displays low-energy M L C T transitions and a broad absorption region. MLCT-based emission is largely quenched due to |3-phosphinothiophene complexation in both binding modes. Investigation of the oxidized states of Ru(II)-P,C complexes suggested that for the Ru(II)-(phosphino)/je«tathiophene-.P,C complex, the charge mainly resides on the thienyl chain, in contrast to the shorter-oligothiophene Ru(II)-P,C complexes and the Ru(II)-P,5 complexes where the charge primarily resides on the metal. These results demonstrate that oligothiophene chain length and bonding mode result in a reordering of the energy levels. DFT calculations of Ru(II)-/>,5' and the Ru(II)-P,C complexes also predicted a mixed metal-thienyl HOMO. The MLCT-based absorption transition of this pentathiophene complex could induce oxidative doping of the rc-conjugated chain, to create a light-harvesting material. The reversibly switchable P,S and P,C bonding modes that mediate conjugation along oligothiophene chains could also find applications as molecular electronic switches. Polymerization reactions towards the preparation of polynuclear polymers have also been attempted, but pure products have not yet been-isolated. Other coupling reactions should be attempted, possibly using the Ru(II)-P, C complexes or new, symmetrical monomers. In conclusion, this thesis has addressed many of the goals initially set for the study. Metal-oligothiophene-hybrid materials have been prepared with attached metal groups that interact with the conjugated chain. Direct metal-oligothiophene (Type II) bonding is found to more strongly alter the properties and display more favorable rc-orbital overlap compared to metals attached in pendent positions to oligothiophene chains (Type I). Structure-property relationships have been established relating chain-length, metal-type and the mode of bonding to the electronic properties. It was intended that this project would be extended to include the synthesis and characterization of metal-polythiophene complexes, but this goal has not yet been attained. Finally, possible applications as molecular electronic wires or switches or as light 143 harvesting materials have been suggested by the electronic properties observed for the metal-oligothiophene materials, though the practicality of these applications require further evaluation. 7.2 Suggestions for Future Work Several projects found within this thesis could be developed further, and new studies could also stem from this work. The p-bis(phosphino)pentafhiophene ligands that were prepared demonstrated a high-yielding new synthetic route suitable for the synthesis of extended P-phosphinothiophenes and multi-substituted P-phosphinothiophene polymers. The preparation of these new ligands could result in materials with improved conductive properties. The Au(I)-bis(phosphino)thiophene complexes displayed aurophilicity1 that could be potentially useful to mediate conjugation, and these materials could be further evaluated in the solid state by absorption and emission spectroscopy. Decreased Ti-orbital overlap is expected when Au-Au bonds are present, and these interactions can also be identified by Au(I)-based luminescence. Electropolymerized Au(I)-oligothiophene materials would be of interest, and modification of the oligothiophene backbone structure via chain extension or the variation of substituents to promote coupled product is worth pursuing. Studies of the P,S and P,C-bound bis(bipyridine) Ru(II)-phosphinothiophene complexes have provided the most interesting results in this thesis, and these compounds are rich with possibilities for future work. P,C Ru(II) complexes would be interesting to study further, due to their reversible oxidation potentials that enable good characterization of the oxidized species. It is desirable to determine i f a long-lived charge-separated state is formed with excitation of the complexes, which would enable light-harvesting applications. Transients created by excitation could be characterized by their lifetimes and absorption spectra using laser flash photolysis techniques, and these studies are already underway. Localization of charge on the oligothiophene chain was observed only for the Ru(II)-(phosphino)pentathiophene-P,C complex, suggesting that a further extended oligothiophene chain would be of interest. Complexes with oligothiophenes of n > 5 chain length could be prepared that may show an improvement in metal-thienyl orbital overlap. Potential materials applications could be evaluated by preparing devices where a Ru(II)-oligothiophene material is drop-cast or spin-coated onto an electrode surface, with contacts deposited on the surface of the film. Photocurrent would be monitored, corresponding to the creation of charge carriers in the material. 144 The investigation of these hybrid materials could also be extended to incorporate Os, as the bis(bipyridine) Os(II) group displays similar charge transfer properties to the bis(bipyridine) Ru(II) group. " Differences in electronic properties for Os(II) complexes include higher energy metal-centered d-orbitals that are less likely to quench MLCT-based emission, a lower Os I I I / n oxidation potential, and increased spin-orbit coupling that can result in more intense transitions to forbidden states.6 Since the Os"1711 oxidation potential is lower than that of Ru 1 1 I / n , it is unlikely that the H O M O would overlap with that of oligo- or polythiophene chains. However, hybrid materials could be potentially used as emitters in organic light emitting diode (OLED) devices if excitons formed along thiophene chains match the M L C T excited state energy. There would be less likelihood in Os(II)-phosphinothiophene systems of emission quenching by low-lying d-orbitals, as was observed for the Ru(II)-phosphinothiophene complexes. Towards the preparation of polynuclear metallated oligo- or polythiophenes, a new route using a symmetric monomer unit is suggested that would result in regioregular materials. Given the results of the coupling reactions that were carried out here, the synthesis of functionalized polythiophene chains followed by metallation is suggested as the most viable route. The challenge for designing a symmetric monomer unit is to ensure that there is an unsubstituted thiophene ring next to each phosphine-substituted ring, so that metal-P-carbon bonding can take place. Alkyl groups should also be incorporated to ensure solubility of the resulting oligomers or polymers. It is suggested that a P,P-dibromo-ct,a-diiodobithiophene could be cross-coupled via Suzuki coupling7 with a thiophene boronic ester to give a bifunctionalized quaterthiophene (Scheme 7-1), or coupled with the bithiophene diboronic ester shown to yield a functionalized polythiophene chain with symmetric quaterthiophene repeating units (Scheme 7-2). The chains would be made soluble in organic solvents by adding two alkyl groups to the central bithiophene units that should arrange anti to one another so rc-orbital overlap is not disrupted. The alkyl groups also block the 3-positions of the starting material so that halogen addition reactions at the desired positions can be more easily performed. P-Brominated polythiophenes thus prepared could be subsequently phosphinated and metallated via procedures that have been described to yield complexes such as the Ru-P,C complexes depicted in Chart 7-1. The binuclear complex could be used to model interactions between metals across a rc-conjugated chain, and the properties could then be compared to the polynuclear metal-polythiophene hybrid material. 145 Scheme 7-1 3) Reflux Scheme 7-2 1) 10 eq CsF 0.05 eq Pd (PPh3)4 THF Br, R R Br 3) Reflux Chart 7-1 Binuclear Metal-Oligothiophene Hybrid Material Polynuclear Metal-Polythioph Hybrid Material 146 7.3 References (1) Bardaji, M . ; Laguna, A. J. Chem. Educ. 1999,76, 201-203. (2) Yam, V. W.-W.; L i , C.-K.; Chan, C.-L. Angew. Chem. Int. Ed. 1998, 37, 2857-2859. (3) Mamo, A. ; Juris, A. ; Calogero, G.; Campagna, S. Chem. Commun. 1996, 1225-1226. (4) Mamo, A. ; Stefio, I.; Poggi, A. ; Tringali, C ; Di Pietro, C ; Campagna, S. New J. Chem. 1997, 21, 1173-1185. (5) Caspar, J. V. ; Meyer, T. J. Inorg. Chem. 1983, 22, 2444-2453. (6) Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin Complexes; Academic Press: London, 1992. (7) Kirschbaum, T.; Azumi, R.; Mena-Osteritz, E.; Bauerle, P. New J. Chem. 1999, 23, 241-250. (8) Stott, T. L. ; Wolf, M . O. Coord. Chem. Rev. 2003, 246, 89-101. 147 Appendix 1 Crystal Structure Data Table Al-1 Selected crystal structure data for Pd(P2hex2T4)Cl2 (30) Formula C52H52P2S4Cl2Pd V/A3 2445.6(3) MW 1044.48 pcalc/g C m 3 1.418 Crystal System triclinic Z 2 Space Group P\ H (Mo-Ka)/mm_1 76 a/A 13.6973(8) Pcaic/g cm"3 1.418 blA 13.7726(9) Rx\ wR2b [/>3o(/)] 0.041,0.082 clA 14.2939(8) i?,a, w^2b (all data) 0.073, 0.089 a/° 73.154(8) v/A3 2445.6(3) p / ° 76.805(9) Reflections Collected /Unique (Rj„t) 22238/9992 (0.059) y/O 73.740(9) GOF 0.93 T/K 173(1) R = S I |F 0 | - |FC| | /E|F 0 | . b Rw = {S[w(F0 2 - F^^/StwfFo 2) 2]} 1" 148 Table A l - 2 Selected Crystal Structure Data for 41, 43, and 45 RuPT 3 -P ) 1 S (41) R u P M e 2 T 3 - / J , 5 ( 4 3 ) R u P D o ^ s - P . S (45) Formula C47H39F, 2N4P3S3ORU C46H37F, 2N4P3S3RU C76H85Fi2N4S.5P3Ru M W 1194.00 1280.11 1578.65 T/K 173(1) 173(1) 173(1) Crystal System triclinic monoclinic triclinic Space Group P\ P2,/a P\ a /A 10.9127(4) 16.122(2) 12.858(1) blk 11.2475(4) 15.480(2) 14.6776(9) elk 20.5749(4) 2.026(3) 20.115(1) a/° 74.704(6) 90.0 77.793(7) p7° 76.606(7) 95.67 89.283(7) y /O 91.102(8) 90.0 87.576(8) v/k3 2361.0(2) 5470(1) 3706.9(4) Pcalc/g C m " 3 1.679 1.554 1.41 z 2 2 2 p. ( M o - K c O / m m " 1 65.7 57.4 47 Reflections Collected/ 22282/9749 192904/12954 32178/14467 Unique (R i n t ) (0.046) (0.057) (0.053) Rx\ wR2b [/>2CT(7)J 0.041,0.100 0.032, 0.085 0.074, 0.165 Ria, wR2b (all data) 0.059,0.106 0.048, 0.091 0.134,0.216 G O F 0.99 1.05 1.02 'R, = Z | |F 0 | - |FC| 1 /S|F 0 | . b wR 2 = {S[w(F0 2 -F c 2 ) 2 ]/Z[w(F 0 2 ) 2 ]}' / 2 149 Table A l - 3 Selected Crystal Structure Data for 47 and 49 . RuPT3-/>,C(47) RuPMe 2 T 3 - / J ,C(49) Formula C47H38F6N40P2RuS3 C 4 6 H 3 6 N 4 F 6 P 2 S 3 R u M W 1064.00 1017.98 T / K 173(1) 173(1) Crystal System triclinic triclinic Space Group P\ P\ a/A 10.1989(3) 9.939(1) b/A 14.9169(5) 11.999(2) clA 15.2939(3) 19.528(3) a/° 77.207(5) 73.361(5) B/° 79.517(5) 77.896(5) y/O 88.763(6) 88.836(5) v/A3 2230.7(1) 2179.8(5) Pcalc/g " l i " 3 1.560 1.551 Z 2 2 /<Mo-Ka)/mm _ 1 63 64.2 Reflections Collected/ 20959/9173 (0.045) 60575/11290 (0.053) Unique (R i n,) Ri\ wR2b 0.034, 0.040 [/>3a(/)] 0.037, 0.088 [/>2a(7)] / f , a ,w/? 2 b (a l l data) 0.063, 0.090 0.063, 0.107 G O F 1.00 1.06 lR, = S I |F 0 | - |FC| | /Z|F 0 | . b wR 2 = {Z[w(F 0 2 - F c 2) 2]/Z[w(F 0 2) 2]} ! ' 150 

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.831.1-0059340/manifest

Comment

Related Items