Citation
Look into the Light

Material Information

Title:
Look into the Light Tuning Platinum Acetylide Complexes for Enhanced Photophysical Properties
Creator:
Winkel, Russell W
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (245 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
SCHANZE,KIRK S
Committee Co-Chair:
WAGENER,KENNETH B
Committee Members:
MILLER,STEPHEN ALBERT
VEIGE,ADAM S
CARROLL,BRUCE F
Graduation Date:
8/8/2015

Subjects

Subjects / Keywords:
Absorption spectra ( jstor )
Chromophores ( jstor )
Fluorescence ( jstor )
Lasers ( jstor )
Ligands ( jstor )
Molecules ( jstor )
Orbitals ( jstor )
Phosphorescence ( jstor )
Platinum ( jstor )
Wavelengths ( jstor )
Chemistry -- Dissertations, Academic -- UF
nhc -- photophysics -- platinum-acetylide
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Chemistry thesis, Ph.D.

Notes

Abstract:
My research focuses on enhancing the photophysical properties of trans-platinum(II) acetylide compounds by altering the properties of the ligand environment around the metal center. Platinum acetylide compounds have been studied for decades due to their rich photophysical properties and potential material applications. The synthetic work described herein has three main facets. The first involves replacing the typical tribuylphosphine (PBu3) ancillary ligands on the platinum atom with N-heterocyclic carbenes (NHCs) to reduce the overall spin-orbit coupling of the system and extend the excited state lifetime. Second, a series of platinum acetylide dimers, tethered by one, two, or three flexibilizing units was used to study excited state dynamics. Specific properties include inter- versus intramolecular excimer formation and triplet-triplet annihilation. Finally, a series of complexes featuring one aryl chromophore and one aryl-acetylide chromophore, with the connectivity being trans-(Ar-Pt-CCAr), was studied, as a method to monosubstitute an aryl ligand to platinum was recently developed by our Group. These mixed chromophore compounds show dual emission in both the singlet and triplet manifolds, giving rise to emission across most of the visible region. Additionally, the ligands can be tuned such that the emission intensity of the fluorescence and phosphorescence is similar in intensity, so the emitted light appears white. Finally, two manuals are presented. One is for use of our Groups femtosecond laser, the other for use of Gaussian on the UF HPC. The laser manual covers basic operation, trouble shooting, two-photon induced fluorescence measurements, and two-photon absorption cross section measurements. The Gaussian/HPC manual lays out instructions for creation, execution, and interpretation of computational results. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2015.
Local:
Adviser: SCHANZE,KIRK S.
Local:
Co-adviser: WAGENER,KENNETH B.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-08-31
Statement of Responsibility:
by Russell W Winkel.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
8/31/2016
Classification:
LD1780 2015 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

LOOK IN TO THE LIGHT: TUNING PLATINUM ACETYLIDE COMPLEXES FOR ENHANCED PHOTOPHYSICAL PROPERTIES By RUSSELL W. WINKEL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF TH E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2015

PAGE 2

© 2015 Russell W. Winkel

PAGE 3

To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS of. Kirk Schanze, for giving me the freedom to work in an unrestricted setting, having the patience to allow me to pursue side projects, and always being available to provide guidance when it was reinforced my understanding of the basics of photophysics, while pushing the limits of advanced materials. Next I would like to thank my Committee members: Prof. Ken Wagener, Prof. Stephen Miller, Prof. Adam Veige, and Prof. Bruce Carroll, not only for the ir guidance when needed, but for making the final defense scheduling completely painless as well. The entire Schanze Group has been phenomenal in my time here. The dynamic of the group , and the willingness of its members to work with an d help each other , I believe , is unrivaled in our department. Starting with the former members, thank Dr. Galyna Dubinina for showing me the finer points of experimental details and crystallization. Dr. Randi Price was instrumental in teaching me how to use all of the laser systems housed in our lab s. Dr. Ali Gundogan, who pioneered the methodology of the Stille type coupling to generate platinum aryl complexes, which was the basis of my last project. For pointing me in the right direction computationally, I would like to thank Dr. M. Erkan Kose. Finally, I am grateful to Dr. Abby Hobbs Shelton for her work on platinum acetylide chromophores which led directly into my first project. For current members, I would like to start by thanking Subhadip Goswami for his insi ghts on organometallic synthesis and general conversations. Seda Cekli has also been very helpful in helping me operate the lesser used Old TA laser and general instrumental troubleshooting. Samantha Phan has put in a ton of work fighting with and

PAGE 5

5 optimiz i ng protocol for the Tsunami laser. An exchange student, Louise Fouquat, was also very helpful in working with and test ing the li mitations of the Tsunami laser. While not a member of the Schanze Group, I also need to thank Zach Morseth of the Papanikolas Gr oup at UNC for running the MD simulations found in Chapter 3. These members and all of the others not mentioned here made my time in the Schanze Group great. My time spent in grad school has been made relatively painless thanks to the support of all of my family and friends. I need to first thank my parents, Bob and Mary, for always pushing me to be better and never letting me quit on something that I had started. Those two values have helped propel me to where I am today. I also need to thank my brother, C country life. As for my friends, I need to thank my roommates Mike Costanzo and Andrew Mowson for games, sporting events, or j ust , living with them always kept things light and entertaining. Last but certainly not least, I need to thank my girlfriend Adeline for always being there. Regardless of how things were going , or how much work I needed to do at home, h er love and support for me never wavered .

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 23 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 25 Overview ................................ ................................ ................................ ................. 25 Plati num Carbene Complexes ................................ ................................ ................ 25 In Non chromophoric Systems ................................ ................................ ......... 25 As Chromophoric Ligands ................................ ................................ ................ 26 As Ancillary Ligands to Acetylene Chromophores ................................ ............ 34 Excimers and Dimeric Complexes Tethered by Flexible Linking Units ................... 38 Overview ................................ ................................ ................................ .......... 38 Applications ................................ ................................ ................................ ...... 39 Excimeric Complexes Containing Platinum ................................ ...................... 40 Monosubstituted Platinum Aryl Compounds as Precursors to Mixed Ligand Platinum Aryl/Acetylene Systems ................................ ................................ ........ 42 2 PHOTOPHYSICAL PROPERTIES OF TRANS PLATINUM ACETYLIDE COMPLEXES FEATU RING N HETEROCYCLIC CARBENE LIGANDS ................ 47 Background ................................ ................................ ................................ ............. 47 Synthesis and Structure ................................ ................................ .......................... 48 Photophysical Properties ................................ ................................ ........................ 52 Electronic Structure Calculations ................................ ................................ ............ 56 Summary and Conclusions ................................ ................................ ..................... 57 Experimental ................................ ................................ ................................ ........... 59 General Remarks ................................ ................................ ............................. 59 Fourier Transform Infrared Spectroscopy (FTIR) ................................ ............. 60 X ray Structure Determinations ................................ ................................ ........ 60 Absorption and Emission Spectroscopy ................................ ........................... 61 Nanosecond Transient Absorption (TA) Spectroscopy ................................ ..... 62 Open Aperture Z Scan ................................ ................................ ..................... 63 Computational Details ................................ ................................ ...................... 63 Synthesis of trans (ICy) 2 PtCl 2 (1) ................................ ................................ ..... 64 General Procedure for the Hagihara Coupling Reaction (2a c) ........................ 65 trans (ICy) 2 Pt(PE2) 2 (2a) ................................ ................................ .................. 65 trans (ICy) 2 Pt(BTF) 2 (2b) ................................ ................................ .................. 66

PAGE 7

7 trans (ICy) 2 Pt(DPAF) 2 (2c) ................................ ................................ ............... 66 3 PLATINUM DIMERS LINKED BY ONE, TWO, OR THREE FLEXIBILE GROUPS FOR STUDY OF EXCITED STATE DYNAMICS ................................ .... 68 Background ................................ ................................ ................................ ............. 68 Synthesis ................................ ................................ ................................ ................ 70 Photophysical Properties ................................ ................................ ........................ 72 Molecular Dynamics Simulations ................................ ................................ ............ 80 Summary and Conclusions ................................ ................................ ..................... 82 Experimental ................................ ................................ ................................ ........... 85 General Remarks ................................ ................................ ............................. 85 Absorption and Emission Spectroscopy ................................ ........................... 86 Nanosecond Transient Absorption (TA) Spectroscopy ................................ ..... 86 Transient Absorption Power Dependence ................................ ........................ 87 Computational Details ................................ ................................ ...................... 87 Structure optimization ................................ ................................ ................ 87 Molecular dynamics calculations ................................ ................................ 87 Synthesis of bis(4 bromobenzyl) ether (COC Br) ................................ ............. 88 Genera l Procedure for the Sonogashira Coupling Reaction ............................. 88 1,2 bis(4 (trimethylsilylethynyl)phenyl)ethane (CC TMS) ........................... 89 Bis(4 (trimethyl silylethynyl)benzyl) ether (COC TMS) ................................ 89 General Procedure for the TMS Deprotection Reaction ................................ ... 89 1,2 bis(4 ethynylphenyl)ethane (C C H) ................................ ..................... 90 Bis(4 ethynylbenzyl) ether (COC H) ................................ .......................... 90 Synthesis of monosubstituted platinum precursor trans (PBu 3 ) 2 Pt(PE2)Cl (PE2 Pt Cl) ................................ ................................ ................................ .... 90 Synthesis of model complex trans (PBu 3 ) 2 Pt(PE2)(CC p tolyl) (tol PtPE2) ...... 90 General Procedure for the Hagihara Coupling React ion for Platinum Dimers. ................................ ................................ ................................ .......... 91 trans [(PE2)Pt(PBu 3 ) 2 (CC p Ph)] 2 CH 2 (C PtPE2) ................................ ...... 92 trans [(PE2)Pt(PBu 3 ) 2 (CC p Ph)] 2 1,2 ethane (CC PtP E2) ........................ 92 trans [(PE2)Pt(PBu 3 ) 2 (CC p Benzyl)] 2 O (COC PtPE2) .............................. 92 4 PHOTOPHYSICAL PROPERTIES OF COMPLEXES FEATURING A MIXED ARYL PLATI NUM ACETYLIDE CHROMOPHORE MOTIF ................................ .... 94 Background ................................ ................................ ................................ ............. 94 Synthesis ................................ ................................ ................................ ................ 95 Ph otophysical Properties ................................ ................................ ........................ 98 Absorption ................................ ................................ ................................ ........ 98 Fluorescence ................................ ................................ ................................ .. 101 Phospho rescence ................................ ................................ ........................... 103 Transient Absorption ................................ ................................ ...................... 105 Electronic Structure Calculations ................................ ................................ .......... 107 Summary and Conclusions ................................ ................................ ................... 114 Experimental ................................ ................................ ................................ ......... 117

PAGE 8

8 General Remarks ................................ ................................ ........................... 117 Absorption and Emission Spectroscopy ................................ ......................... 118 Nanosecond Transient Absorption (TA) Spectroscopy ................................ ... 118 Computational Detail s ................................ ................................ .................... 119 General Procedure for the Hagihara Coupling Reaction for Mixed Platinum Aryl Acetylide Compounds. ................................ ................................ ......... 119 trans (PBu 3 ) 2 Pt(BTF) (CCBTF) (BTF Pt CCBTF) ................................ ..... 120 trans (PBu 3 ) 2 Pt(DPAF)(CCDPAF) (DPAF Pt CCDPAF) .......................... 120 trans (PBu 3 ) 2 Pt(BTF)(CCDPAF) (BTF Pt CCDPAF) ................................ 121 trans (PBu 3 ) 2 Pt(DPAF)(CCBTF) (DPAF Pt CCBTF) ................................ 121 APPENDIX A TSUNAMI FEMTOSECOND LASER MANUAL ................................ .................... 123 Overview ................................ ................................ ................................ ............... 123 Laser Start up ................................ ................................ ................................ ....... 124 Laser Wavelength Selection, Mode Locking, and Output Power Optim ization ..... 125 Wavelength Selection and Mode Locking ................................ ...................... 125 Output Power Optimization ................................ ................................ ............. 127 Two Photon Excited Fluorescence (2PEF) ................................ ........................... 127 Two Photon Cross Section Measurements ................................ ........................... 128 Data Conversion ................................ ................................ ................................ ... 130 Instrument Shut Down ................................ ................................ .......................... 130 Laser Powers and Bandwidths at Given Wavelengths ................................ ......... 130 Data Workup ................................ ................................ ................................ ......... 131 Troubleshooting ................................ ................................ ................................ .... 132 Pump Laser is not Coming Up to the Set Power ................................ ............ 132 Laser Can Not Be Tuned Below 780 nm. ................................ ....................... 133 Advanced Troubleshooting ................................ ................................ ................... 133 B COMPUTATIONAL METHODS ................................ ................................ ............ 136 Overview ................................ ................................ ................................ ............... 136 Accessing the HPC ................................ ................................ ............................... 137 General Considerations ................................ ................................ ........................ 138 Using Gaussian 09 and GaussView 5 via the HPC ................................ .............. 139 Creating Input Files and Submitting Jobs ................................ ............................. 140 Geometry Optimization/Standard Input Parameters ................................ ....... 141 Frequency ................................ ................................ ................................ ...... 145 Time Dependent Density Function al Theory (TDDFT) ................................ ... 147 Triplet Computations ................................ ................................ ...................... 148 General considerations ................................ ................................ ............ 148 Estimation of phosphorescence energy ................................ ................... 149 Stability ................................ ................................ ................................ ........... 149 Population ................................ ................................ ................................ ...... 149 Volume ................................ ................................ ................................ ........... 151

PAGE 9

9 PBS Script ................................ ................................ ................................ ...... 151 Reading and Interpreting Output Files ................................ ................................ .. 155 Geometry Optimization ................................ ................................ ................... 15 6 Frequency ................................ ................................ ................................ ...... 157 TDDFT ................................ ................................ ................................ ............ 160 Stability ................................ ................................ ................................ ........... 161 Population ................................ ................................ ................................ ...... 161 Volume ................................ ................................ ................................ ........... 165 Visualizing Result s ................................ ................................ ................................ 166 GaussView ................................ ................................ ................................ ..... 167 Molecular geometry ................................ ................................ ................. 167 Molecular orbitals ................................ ................................ ..................... 167 Chemcraft ................................ ................................ ................................ ....... 169 Generating cube files ................................ ................................ ............... 169 Molecular geometry/mol ecular orbitals ................................ .................... 170 Charge difference density ................................ ................................ ........ 171 Considerations for Triplet Graphics Generation ................................ ............. 174 C SUPPORTING INFORMATION FOR CHAPTER 2 ................................ .............. 176 X ray Experimental and Data ................................ ................................ ................ 176 NMR Spectra ................................ ................................ ................................ ........ 181 IR Spectra ................................ ................................ ................................ ............. 185 Emission Lifetime Decays ................................ ................................ ..................... 187 Transient Absorption Decays ................................ ................................ ................ 189 Triplet Triplet Absorption Power Dependence ................................ ...................... 191 Computational Studies ................................ ................................ .......................... 192 D SUPPORTING INFORMATION FOR CHAPTER 3 ................................ .............. 200 NMR Spectra ................................ ................................ ................................ ........ 200 Emission Lifetime Decays ................................ ................................ ..................... 211 Transient Absorption Decays ................................ ................................ ................ 213 E SUPPORTING INFORMATION FOR CHAPTER 4 ................................ .............. 217 NMR Spe ctra ................................ ................................ ................................ ........ 217 Emission Lifetime Decays ................................ ................................ ..................... 226 Transient Absorption Decays ................................ ................................ ................ 228 Computational Results ................................ ................................ .......................... 230 Monosubstituted Platinum Model Complexes ................................ ................. 230 Mixed Ligand Platinum Aryl/Acetylide Complexes ................................ .......... 232 LIST OF REFERENCES ................................ ................................ ............................. 239 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 245

PAGE 10

10 LIST OF TABLES Table page 2 1 Key interatomic distances (Ã…) and angles (deg) for platinum acetylide complexes. ................................ ................................ ................................ ......... 50 2 2 Summary of photophysical data for platinum acet ylides 2a c and 3a c . ............. 53 2 3 Summary of TD DFT computations for the S 0 1 transition of . ............ 57 3 1 Summary of photophysica l data for R PtPE2 . ................................ .................... 73 4 1 Summary of photophysical data for mixed aryl/acetylide ligated Pt(II) complexes. ................................ ................................ ................................ ....... 100 4 2 Summary of T D DFT computations for the S 0 1 transition of monosubstituted platinum aryl and platinum acetylide compounds. ................. 108 4 3 Summary of TD DFT computations for transitions of oscillator strength > 0.1 and for the m ixed platinum aryl/acetylide complexes. .................... 111 A 1 Laser powers and bandwidths at given wavelengths. ................................ ....... 130 C 1 Summary of crystallogr aphic data ................................ ................................ .... 177 C 2 Summary of TDDFT computations for vertical excitations of oscillator strength greater than 0.1 for compounds . ................................ ............... 193

PAGE 11

11 LIST OF FIGURES Fi gure page 1 1 Structure of a generic NHC showing resonance from nitrogen lone pairs. ......... 27 1 2 Core structure of the pla tinum NHCs explored by the Strassner Group. ............ 27 1 3 Structures of 1 aryl and 4 aryl 1,2,4 triazole 5 ylidenes studied in Reference 27. Triazole rings are numbered. ................................ ................................ ........ 27 1 4 Nitrile containing Pt NHC, featuring the mesacac ligand, with an 85% phosphorescence quantum yield in a 2 wt% PMMA film. ................................ ... 28 1 5 Dibenzofuran containin g Pt NHC with a 90% quantum yield in a 2 wt% PMMA film. ................................ ................................ ................................ ......... 29 1 6 Representative crystal packing diagrams of Pt NHCs with A) low steric bulk of the ancillary ligand allowing close packing and dim eric interactions, and B) high steric bulk of the ancillary ligand ................................ ................................ . 31 1 7 Nitro functionalized Pt NHCs which are modestly emissive despite bearing a ketoamine ancillary ligand. ................................ ................................ ................. 32 1 8 Pt NHCs featuring tetradentate bonding motifs. ................................ ................. 33 1 9 Platinum carbene and platinum phosphine complexes with phenylacetylene chromophore s. The carbene complex shows room temperature phosphorescence in deaerated solution ................................ ............................. 34 1 10 Qualitative d orbital splitting diagram for a square planar platinum(II) phosphine complex versus a square planar platinum(II) carbene complex.. ...... 35 1 11 A Pt NHC complex featuring monodentate carbene ligands and trans stereochemistry. ................................ ................................ ................................ . 36 1 12 Structure of the Pt NHC featuring monodentate carbene ligands demonstrating an 80% emission quantum yield in a PMMA film. ....................... 37 1 13 Structure of the Pt NHC featuring monodentate c arbene ligands demonstrating an 80% emission quantum yield in a PMMA film. ....................... 37 1 14 A Jablonski diagram illustrating the energy levels of monomeric and excimeric excited states. ................................ ................................ .................... 38 1 15 The structure of 1,3 (di 2 pyrenyl)propane. ................................ ........................ 39 1 17 The structure of PE2 octyl. ................................ ................................ ................. 41

PAGE 12

12 1 18 Transient absorption spectrum of 57 mM PE2 octyl in deoxygenated benzene after 476 nm laser excitation. Decay traces start at t=0 and were taken in increments of 150 ns. ................................ ................................ ............ 42 1 13 The four mixed ligand platinum aryl/acetylide complexes synthesized by the Low Group. ................................ ................................ ................................ ......... 45 2 1 One pot synthesis of trans (ICy) 2 PtCl 2 ( 1 ). ................................ ......................... 48 2 2 ORTEP diagram of trans (ICy) 2 PtCl 2 ( 1 ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. ................................ ................................ ................................ ................. 48 2 3 Hagihara reaction to generate the platinum acetylide compounds 2a c . ............ 49 2 4 ORTEP diagrams of the molecular structures of 2a c , with thermal ellipsoids given at the 50% probability level. The hydrogen atoms are omitted for clarity. C: Black, N: Blue, S: Yellow, Pt: Pink, Chlorine: Green ........................... 50 2 5 2a c . .............. 51 2 6 The A) ground state absorption, B) normalized fluorescence, and C) normalized phosphorescence spectra of 2a c in THF. E xcitation was at the ground state absorption maxima, and the plots were normalized. ...................... 53 2 7 Principal component transient absorption spectra of 2a c , with excitation at 355 nm, 10 ns pulse widt absorptivity value of 0.62 at 355 nm after four freeze pump thaw cycles. .......... 55 2 8 Open aperture z scan transmittance using 606 nm light. Solut ions were 1 3c was used as the reference. ................................ ................................ ................................ ..... 56 2 9 DFT optimized structures (top row), and charge difference densities (bottom row) for t he calculated transitions corresponding to the experimental ground state absorption maxima. ................................ ................................ ................... 57 3 1 Structure of the target platinum acetylide dimers and mono platinum model. .... 68 3 2 Synthesis of diarylacetylides tethered by 1 3 flexibilizing units. .......................... 71 3 3 Synthesis of platinum diarylacetylides and a mono platinum model complex. ... 72 3 4 The A) ground state absorption, B) normalized fluorescence, and C) normalized phosphorescence spectra of R PtPE2 in THF. Excitation was at the ground state absorption maxima, a nd all plots were normalized. ................. 73

PAGE 13

13 3 5 Principal component transient absorption spectra of R PtPE2 , with excitation an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles. ...... 75 3 6 First five time slices of the transient absorption spectra of COC PtPE2 with a camera delay interval of 400 ns. Excitation was at 355 nm, with a 10 ns pulse ................................ ........................ 76 3 7 Transient absorption power dependence of A) R PtPE2 monitored at 550 nm, with B) an expansion of the low fluence region to show initial slopes. ......... 76 3 8 Transient absorption lifetime decays mea sured at 550 nm with variable fluence. A) C PtPE2 . B) CC PtPE2 . C) COC PtPE2 . D) tol PtPE2 . .................. 78 3 10 Normalized transient absorption lifetime decays measured at 550 nm for R PtPE2 on a logarith mic scale at A) low fluence (1.7 ± 0.3 mJ/(cm 2 pulse)), and B) high fluence (61 ± 7 mJ/(cm 2 pulse)). ................................ .............................. 79 3 11 Histograms showing the probability of the internuclear Pt Pt distance for A) C PtPE2 , B) CC PtPE2 , and C) COC PtPE2 . ................................ .................... 81 3 12 DFT optimized gas phase structures of A) C PtPE2 , B) CC PtPE2 , and C) COC PtPE2 . ................................ ................................ ................................ ....... 81 3 13 Representative drawings of the anti anti , syn anti , and anti anti rotational isomers of COC PtPE2 . ................................ ................................ ..................... 82 4 1 The four combinations of platinum aryl/acetylide complexes featuring DPAF and BTF chr omophores. ................................ ................................ ..................... 94 4 1 Synthesis of monosubstituted platinum(II) aryl precursors. ................................ 97 4 2 Hagihara reaction to generate the four mixed aryl/acetylide Pt(II) complexes. ... 98 4 4 Ground state absorption spectra of A) the four mixed ligand platinum complexes containing the BTF and DPAF chromophores 99 4 5 Fluorescence spectra of the four mixed ligand platinum complexes containing the BTF and DPAF chromophores. ................................ ................................ .. 101 4 6 Fluorescence spectra of DPAF Pt CCBTF with excitation at 378 and 330 nm. 102 4 7 Emission spectra after at least five freeze pump thaw cycles of A) the four mixed ligand platinum complexes containing the BTF and DPAF chromophores ................................ ................................ ................................ ... 104

PAGE 14

14 4 8 Principal component transient absorption spectra with excitation at 355 nm, absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles. 106 4 9 DFT optimized structures of the singlet ground state and charge difference densities for the computed S 0 1 vertical transition for A) BTFCC Pt , B) DPAFCC Pt , C) BTF Pt , and D) DPAF Pt . ........................... 110 4 10 DFT optimized structures of the singlet ground state and charge difference densities for the computed S 0 1 vertical transition for A) BTF Pt and B) DPAF Pt .. ................................ .......................... 112 4 11 DFT optimized structure of the singlet ground state and charge difference density for the vertical transition predicted at 376 nm of BTF Pt . .... 114 4 12 DFT optimized structure of A) the singlet ground state, and charge difference densities for the vertical transitions predicted at B) 401 nm and C) 350 nm of DPAF Pt . . ................................ ................................ ........................... 115 A 1 Spectra Physics software showing A) the instrument has warmed up and is ready for to be used. B) Pump laser is operational, but has not come up to the requested power. C) Pump laser is operational ................................ .......... 125 A 4 Absorption (solid colored lines) and emission (dotted lines) of two complexes showing excitation and registration wavelength selections (solid black lines), respectively. ................................ ................................ ................................ ...... 129 A 6 Defeating the safety shutter on the Tsunami Laser. ................................ ......... 134 B 1 General workflow of computations. ................................ ................................ ... 137 B 2 A) 2 D Chemdraw model of (ThPt) 2 BTD . B) 3 D model of (ThPt) 2 BTD built and visualized in GaussView. C: Gray, H: White, N: Blue, P: Orange, S: Yellow, Pt: Cerulean. ................................ ................................ ........................ 13 9 B 3 GaussView main window showing options of the Calculate tab. ...................... 140 B 4 Blank calculation setup window with notable areas labeled. ............................ 140 B 5 Major conformations of (ThPt) 2 BTD . ................................ ................................ 141 B 6 Appended input file for the geometry optimization/frequency calculation of (ThPt) 2 BTD . ................................ ................................ ................................ ..... 142 B 7 Example input file for the f requency calculation of (ThPt) 2 BTD . ....................... 146 B 8 Example input file for the TDDFT calculation of (ThPt) 2 BTD . ........................... 147 B 9 Example input fi le for a stability job on a platinum aryl compound. ................... 150

PAGE 15

15 B 10 Example input file for generation of Natural Atomic Orbitals. ........................... 150 B 11 Example input file for a volume job of a gold acetylide compound. .................. 152 B 12 Example PBS script for submitting jobs to the HPC. ................................ ........ 152 B 13 Routine information given in every Gaussian output file. ................................ .. 155 B 14 Successful optimization of (ThPt) 2 BTD to a stationary point. ........................... 156 B 15 N pole moments for (ThPt) 2 BTD . ................................ ................................ ..... 156 B 16 Final lines of the optimization of (ThPt) 2 BTD . ................................ .................. 157 B 17 Intermediate steps of a frequency job showing the number of vectors yet to converge. ................................ ................................ ................................ .......... 158 B 18 Frequency output of (ThPt) 2 BTD with negative frequencies, indicating that the geometry is not at an energy mini mum. ................................ ...................... 159 B 19 Frequency output of (ThPt) 2 BTD with no negative frequencies, indicating that the geometry is at an energy minimum. ................................ ............................ 159 B 20 TDDFT output of (ThPt) 2 BTD showing relevant vertical transitions. ................ 160 B 21 Output of a stability job indicating that an instability is present in the wavefunction. ................................ ................................ ................................ .... 162 B 22 Output of a stability job indicating that the wavefunction is now stable. ............ 162 B 23 Multiwfn startup screen with input file path typed in the last line of text. ........... 163 B 24 Multiwfn orbital composition analysis options. ................................ .................. 164 B 25 Percent contributions of a specified fra gment to a pair of specified MOs. ........ 165 B 26 Output of a volume job with the suggested radius boxed. ................................ 166 B 27 MO Editor window in GaussView displaying (ThPt) 2 BTD with the HOMO and LUMO highlighted for MO generation. ................................ .............................. 169 B 28 MO Editor window in GaussView displaying the HOMO of (ThPt) 2 BTD visualized with an isovalue of 0. 02. ................................ ................................ .. 170 B 29 Chemcraft program displaying the HOMO of (ThPt) 2 BTD visualized with an isovalue of 0.02. Useful buttons are labeled. ................................ .................... 171 B 30 ................................ 173

PAGE 16

16 B 31 CDD for the lowest energy computed transition of (ThPt) 2 BTD . Blue lobes indicate electron density being lost, while red lobes indicate electron density being gained. Image visualized at an isovalue of 0.0004. ................................ 173 B 32 MO editor window with the highest numbered alpha orbital of (ThPt) 2 BTD pointed out. Singly occupied MOs are also boxed. ................................ ........... 175 B 33 The triplet of (ThPt) 2 BTD in Chemcraft with the highest energy MO of both the alpha and beta electrons loaded. Note that the numbering in this program is sequential. ................................ ................................ ................................ ..... 175 C 1 ORTEP diagram of trans (ICy) 2 PtCl 2 ( 1 ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. ................................ ................................ ................................ ............... 178 C 2 ORTEP diagram of trans (ICy) 2 Pt(PE2) 2 ( 2a ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. ................................ ................................ ............................. 179 C 3 ORTEP diagram of trans (ICy) 2 Pt(BTF) 2 ( 2b ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. ................................ ................................ ............................. 179 C 4 ORTEP diagram of trans (ICy) 2 Pt(DPAF) 2 ( 2c ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. ................................ ................................ ............................. 180 C 5 1 H NMR (500 MHz, CDCl 3 ) of 1 . ................................ ................................ ....... 181 C 6 13 C NMR (125.7 MHz, CDCl 3 ) of 1 . ................................ ................................ ... 181 C 7 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2a . ................................ ................................ .... 182 C 8 13 C NMR (125.7 MHz, CD 2 Cl 2 ) of 2a . ................................ ............................... 182 C 9 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2b . ................................ ................................ ... 183 C 10 Insets for selected peaks from the 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2b . ............. 183 C 11 13 C NMR (125.7 MHz, CD 2 Cl 2 ) of 2b . ................................ ............................... 184 C 12 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2c . Residual hexanes are present. .................. 184 C 13 13 C NMR (125.7 MHz, CD 2 Cl 2 ) of 2c . ................................ ............................... 185 C 14 FTIR spectrum of 1 . ................................ ................................ .......................... 185 C 15 FTIR spectrum of 2a . ................................ ................................ ........................ 186

PAGE 17

17 C 16 FTIR spectrum of 2b . ................................ ................................ ........................ 186 C 17 FTIR spectrum of 2c . ................................ ................................ ........................ 187 C 18 Fluorescence lifetime decay for 2a (light blue) with instrument response function (dark blue). ................................ ................................ .......................... 187 C 19 Phosphorescence lifetime decay for 2a . ................................ ........................... 188 C 20 Phosphorescence lifetime decay for 2b . ................................ ........................... 188 C 21 Phosphorescence l ifetime decay for 2c . ................................ ........................... 189 C 22 Transient absorption decay for 2a . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, Q switch delay: 380 µs, 180 µJ per pulse. ................................ ................................ .............................. 189 C 23 Transient absorption decay for 2b . Initial camera delay: 100ns, camera delay increment: 20 µs, 100 images averaged per trace, Q switch delay: 382 µs, 180 µJ per pulse. ................................ ................................ .............................. 190 C 24 Transient absorption decay for 2c . Initial camera delay: 100ns, camera delay increment: 20 µs, 100 images averaged per trace, Q switch delay: 379 µs, 180 µJ per pulse. ................................ ................................ .............................. 190 C 25 Transient absorption decay profile for 2a . Conditions: 200 µJ per pulse, 380 µJ Q switch delay, 128 scans averaged, decay measured at 584 nm. ............. 191 C 26 Transient absorption decay profiles for 2a at varying laser energies. Conditions: 200 µJ 18 mJ per pulse, 380 200 µJ Q switch delay, 128 scans averaged per trace, decays measured at 584 nm. ................................ ........... 191 C 27 DFT optimized structure of . ................................ ................................ ........ 192 C 28 373.8 nm CDD of . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in the electronic transition. ................................ ................................ ................................ .......... 194 C 29 Normalized overlay of the experimental absorption spectrum of 2a with the TDDFT computed line spectra of . Only vertical excitations with f > 0.1 ar e shown. ................................ ................................ ................................ ........ 194 C 30 LUMO+1 (orbital 168) of compound . ................................ .......................... 194 C 31 LUMO (orbital 167) of compound . ................................ .............................. 195 C 32 HOMO (orbital 166) of compound . ................................ .............................. 195 C 33 HOMO 1 (orbital 165) of compound . ................................ .......................... 195

PAGE 18

18 C 34 DFT optimized structure of . ................................ ................................ ........ 195 C 35 429.3 nm CDD of . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in the e lectronic transition. ................................ ................................ ................................ .......... 196 C 36 Normalized overlay of the experimental absorption spectrum of 2a with the TDDFT computed line spectra of . Only vertical excitations with f > 0.1 are shown. ................................ ................................ ................................ ........ 196 C 37 LUMO+1 (orbital 262) of compound . ................................ .......................... 196 C 38 LUMO (orbital 261) of compound . ................................ .............................. 197 C 39 HOMO (orbital 260) of compound . ................................ ............................. 197 C 40 LUMO+1 (orbital 259) of compound . ................................ .......................... 197 C 4 1 DFT optimized structure of . ................................ ................................ ........ 197 C 42 381.5 nm CDD of . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in the electroni c transition. ................................ ................................ ................................ .......... 198 C 43 Normalized overlay of the experimental absorption spectrum of 2c with the TDDFT computed line spectra of . Only vertical excitations with f > 0.1 are shown ................................ ................................ ................................ ......... 198 C 44 LUMO+1 (orbital 282) of compound . ................................ .......................... 198 C 45 LUMO (orbital 281) of compound . ................................ .............................. 199 C 46 HOMO (orbital 280) of compound . ................................ .............................. 199 C 47 HOMO 1 (orbital 279) of compound . ................................ .......................... 199 D 1 1 H NMR ( 500 MHz, CDCl 3 ) of COC Br . ................................ ............................ 200 D 2 13 C NMR (126 MHz, CDCl 3 ) of COC Br . ................................ .......................... 200 D 3 1 H NMR (500 MHz, CDCl 3 ) of CC TMS . ................................ ........................... 201 D 4 13 C NMR (126 MHz, CDCl 3 ) of CC TMS . ................................ .......................... 201 D 5 1 H NMR (500 MHz, CD 2 Cl 2 ) of COC TMS . ................................ ....................... 202 D 6 13 C NMR (126 MHz, CD 2 Cl 2 ) of COC TMS . ................................ ...................... 202 D 7 1 H NMR (500 MHz, CD 2 Cl 2 ) of CC H . ................................ ............................... 203

PAGE 19

19 D 8 13 C NMR (126 MHz, CD 2 Cl 2 ) of CC H . ................................ ............................. 203 D 9 1 H NMR (500 MHz, CD 2 Cl 2 ) of COC H . ................................ ............................ 204 D 10 13 C NMR (126 MHz, CD 2 Cl 2 ) of COC H . ................................ ........................... 204 D 11 1 H NMR (500 MHz, CDCl 3 ) of C PtPE2 . ................................ ........................... 205 D 12 13 C NMR (126 MHz, CDCl 3 ) of C PtPE2 . ................................ ......................... 205 D 13 31 P NMR (121 MHz, CDCl 3 ) of C PtPE2 . ................................ .......................... 206 D 14 1 H NMR (500 MHz, CD 2 Cl 2 ) of CC PtPE2 . ................................ ....................... 206 D 15 13 C NMR (126 MHz, CD 2 Cl 2 ) of CC PtPE2 . ................................ ...................... 207 D 16 31 P NMR (121 MHz, CD 2 Cl 2 ) of CC PtPE2 . ................................ ...................... 207 D 17 1 H NMR (500 MHz, CD 2 Cl 2 ) of COC PtPE2 . ................................ .................... 208 D 18 13 C NMR (126 MHz, CD 2 Cl 2 ) of COC PtPE2 . ................................ ................... 208 D 19 31 P NMR (121 MHz, CD 2 Cl 2 ) of COC PtPE2 . ................................ ................... 209 D 20 1 H NMR (500 MHz, CDCl 3 ) of tol PtPE2 . ................................ ......................... 209 D 21 13 C NMR (126 MHz, CDCl 3 ) of tol PtPE2 . ................................ ........................ 210 D 22 31 P NMR (121 MHz, CDCl 3 ) of tol PtPE2 . ................................ ........................ 210 D 23 Phosphorescence lifetime decay for C PtPE2 . ................................ ................. 211 D 24 Phosphorescence lifetime decay for CC PtPE2 . ................................ .............. 211 D 25 Phosphorescence lifetime decay for COC PtPE2 . ................................ ........... 212 D 26 Phosphorescence lifetime decay for tol PtPE2 . ................................ ............... 212 D 27 Normalized transient absorption for R PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse.. ................................ ................................ ................................ ............... 213 D 28 Transient absorption decay for C PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse.. .... 213 D 29 Transient absorption decay for CC PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ................ 214

PAGE 20

20 D 30 Transient absorption decay fo r COC PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse.. ................................ ................................ ................................ ............... 214 D 31 Transient absorption decay for tol PtPE2 . Initial camera del ay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ................ 215 D 32 Representative molecular geometry of C PtPE2 during the molecular dynamics simulation proces s. ................................ ................................ ........... 215 D 33 Representative molecular geometry of CC PtPE2 during the molecular dynamics simulation process. ................................ ................................ ........... 216 D 34 Representat ive molecular geometry for the primary distribution of COC PtPE2 during the molecular dynamics simulation process. .............................. 216 D 35 Representative molecular geometry for the secondary distribution of COC PtPE2 during the molecular dynamics simulation process. .............................. 216 E 1 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCBTF . ................................ .. 217 E 2 A) Aromatic region of the 1 H NMR spectra of BTF Pt CCBTF . B) Aliphatic region of the 1 H NMR spectra of BTF Pt CCBTF . ................................ ............ 217 E 3 13 C NMR (126 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCBTF . ................................ . 218 E 4 Aromatic region of the 13 C NMR spectra of BTF Pt CCBTF . ........................... 218 E 5 31 P NMR (121 MHz, CD 2 Cl 2 ) of BTF Pt CCBTF . ................................ ............. 219 E 6 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCDPAF . ............................ 219 E 7 13 C NMR (126 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCDPAF . ........................... 220 E 9 31 P NMR (121 MHz, CD 2 Cl 2 ) of DPAF Pt CCDPAF . ................................ ........ 221 E 10 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCDPAF . ............................... 221 E 11 Aromatic region of the 1 H NMR spectra of BTF Pt CCDPAF . .......................... 222 E 12 13 C NMR (126 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCDPAF . .............................. 222 E 13 Aromatic region of the 13 C NMR spectra of BTF Pt CCDPAF . ......................... 223 E 14 31 P NMR (121 MHz, CD 2 Cl 2 ) of BTF Pt CCDPAF . ................................ ........... 223 E 15 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCBTF . ............................... 224

PAGE 21

21 E 16 Aromatic region of the 1 H NMR spectra of DPAF Pt CCBTF . .......................... 224 E 17 13 C NMR (126 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCBTF . .............................. 225 E 18 Aromatic region of the 13 C NMR spectra of DPAF Pt CCBTF . ......................... 225 E 19 31 P NMR (121 MHz, CD 2 Cl 2 ) of DPAF Pt CCBTF . ................................ ........... 226 E 20 Phosphorescence lifetime decay for BTF Pt CCBTF . ................................ ...... 226 E 21 Phosphoresce nce lifetime decay for DPAF Pt CCDPAF . ................................ 227 E 22 Phosphorescence lifetime decay for BTF Pt CCDPAF . ................................ ... 227 E 23 Phosphorescence lifet ime decay for DPAF Pt CCBTF . ................................ ... 228 E 24 Transient absorption decay for BTF Pt CCBTF . Initial camera delay: 100ns, camera delay increment: 10 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ................ 228 E 25 Transient absorption decay for DPAF Pt CCDPAF . Initial camera delay: 100ns, camera delay increment: 10 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ..... 229 E 26 Transient absorption decay for BTF Pt CCDPAF . Initial camera delay: 100ns, camera delay increment: 10 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ..... 229 E 27 Transient absorption decay for DPAF Pt CCBTF . Initial camera delay: 100ns, camera delay increment: 10 µs, 100 images averaged per trace, 180 µJ per pulse. ................................ ................................ ................................ ..... 230 E 28 LUMO of BT FCC Pt . ................................ ................................ .................. 230 E 29 HOMO of BTFCC Pt . ................................ ................................ .................. 230 E 30 LUMO of DPAFCC Pt . ................................ ................................ ................ 231 E 31 HOMO of DPAFCC Pt . ................................ ................................ ............... 231 E 32 LUMO of BTF Pt . ................................ ................................ ........................ 231 E 33 HOMO of BTF Pt . ................................ ................................ ....................... 231 E 34 LUMO of DPAF Pt . ................................ ................................ ..................... 232 E 35 HOMO of DPAF Pt . ................................ ................................ .................... 232 E 36 LUMO+1 of BTF Pt CCBTF . ................................ ................................ ............ 232

PAGE 22

22 E 37 LUMO of BTF Pt CCBTF . ................................ ................................ ................ 233 E 38 HOMO of BTF Pt CCBTF . ................................ ................................ ............... 233 E 39 HOMO 1 of BTF Pt CCBTF . ................................ ................................ ............ 233 E 40 CDD of the predicted transition at 360 nm for BTF Pt CCBTF . ....................... 233 E 41 LUMO+1 of DPAF Pt CCDPAF . ................................ ................................ ...... 234 E 42 LUMO of DPAF Pt CCDPAF . ................................ ................................ ........... 234 E 43 HOMO of DPAF Pt CCDPAF . ................................ ................................ .......... 234 E 44 HOMO 1 of DPAF Pt CCDPAF . ................................ ................................ ....... 234 E 45 CDD of the predicted transition at 350 nm for DPAF Pt CCDPAF . .................. 235 E 46 LU MO+1 of BTF Pt CCDPAF . ................................ ................................ ......... 235 E 47 LUMO of BTF Pt CCDPAF . ................................ ................................ ............. 235 E 48 HOMO of BTF Pt CCDPAF . ................................ ................................ ............. 235 E 49 HOMO 1 of BTF Pt CCDPAF . ................................ ................................ ......... 236 E 50 HOMO 2 of BTF Pt CCDPAF . ................................ ................................ ......... 236 E 51 CDD of the predicted CT tran sition at 438 nm for BTF Pt CCDPAF . This transition is not observed in the experimental UV Vis spectrum. ...................... 236 E 52 CDD of the predicted transition at 350 nm for BTF Pt CCDPAF . ..................... 236 E 53 LUMO+1 of DPAF Pt CCBTF . ................................ ................................ ......... 237 E 5 4 LUMO of DPAF Pt CCBTF . ................................ ................................ ............. 237 E 5 5 HOMO o f DPAF Pt CCBTF . ................................ ................................ ............. 237 E 5 6 HOMO of DPAF Pt CCBTF . ................................ ................................ ............. 237 E 5 7 Normalized UV Vis absorption spectrum of A) BTF Pt CCBTF , B) DPAF Pt C CDPAF , C) BTF Pt CCDPAF , and D) DPAF Pt CCBTF with the TDDFT predicted vertical transitions overlaid. ................................ ............................... 238

PAGE 23

23 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial F ulfillment of the Requirements for the Degree of Doctor of Philosophy LOOK IN TO THE LIGHT: TUNING PLATINUM ACETYLIDE COMPLEXES FOR ENHANCED PHOTOPHYSICAL PROPERTIES By Russell W. Winkel August 2015 Chair: Kirk S. Schanze Major: Chemistry My research fo cuses on enhancing the photophysical properties of trans p latinum (II) acetylide compounds by altering the properties of the ligand environment around the metal center. Platinum acetylide compounds have been studied for decades due to their rich photophysic al properties and potential material applications. The synthetic work described herein has three main facets. The first involves replacing the typical tribuylphosphine (PBu 3 ) ancillary ligands on the platinum atom with N heterocyclic carbenes (NHCs) to red uce the overall spin orbit coupling of the system and extend the excited state lifetime. Second, a series of platinum acetylide dimers, tethered by one, two, or three flexibilizing units was used to study excited state dynam ics. Specific properties include inter versus intramolecular excimer formation and triplet triplet annihilation. Finally, a series of complexes featuring one aryl chromophore and one aryl acetylide chromophore, with the connectivity being trans (Ar Pt CCAr), wa s studied, as a method to monosubstitute an aryl ligand to platinum was recently developed by our Group. These mixed chromophore compounds show dual emission in the singlet manifold and potentially the triplet manifold, giving rise to emission across most of the visible regio n. Add itionally, the ligands may be tuned such that the emission

PAGE 24

24 intensity of the fluorescence and phosphorescence are similar in intensity, so the emitted light appears white. Last ly, two manual s are presented . One is laser , t he other for use of Gaussian on the UF HPC . The laser manual covers basic operation, trouble shooting, two photon induced fluorescence measurements, and two photon absorption cross section measurements. The Gaussian/HPC manual lays out instructions for cre ation, execution, and interpretation of computational results.

PAGE 25

25 CHAPTER 1 INTRODUCTION Overview Electronic excitation of a molecule is induced when a photon of sufficient energy is absorbed, causing the molecule to be promoted from the ground state to the excited state. The decay of th e molecule back to the ground state can follow a variety of pathways, and tuning these pathways is essential for modern applications. This introduction will assume that the reader knows the fundamentals of photophysics, but w ill elaborate on the specific properties of the core system that were tuned, as well as how and why they were tuned. Square planar platinum(II) arylacetylide complexes have been known and intensely studied for several decades due to their wide range of app lications that include non linear absorption, polymer light emitting diodes, and polymer photovoltaic devices. 1 11 conjugated aryl acetylide ligands are coupled to a heavy metal with large spin orbit coupling, intersystem crossing to the triplet excited state can be very efficient. 1 , 6 This allows for strong triplet triplet absorption which can contribute to non linear absorption (NLA). Many variations of platinum acetylide complexes have been synthesized and their properties explored. 6 , 12 The following details of this chapter wil l introduce the concepts and motifs of the chapters to come . Platinum Carbene Complexes In Non chromophoric Systems database (to remove duplicate citations ) returns 571 journal article s. The earliest synthesis of a platinum carbene was in 1969 and was derived from a platinum

PAGE 26

26 isocyanide. 13 Most of the non chromophoric complexes are synthesized with the intention of imparting some form of reactivity on an outside system (485 citations ), with a fair subset of this being catalysis (160 citations ) . As Chromophoric Ligands W hen the above emission or only 49 articles remain. The earliest report of a luminescent platinum carbene is from 1999, 14 tinum complexes really took off ( 4 3 of 49 articles published since 2010). A vast majority of these complexes have two things in common . First, is that the form of the carbene is typically (90% of reports) a N heterocyclic carbene (NHC). This carbene struct ure is popular because the empty p orbital on the carbene carbon can be stabilized by resonance with lone pairs from adjacent nitrogen atoms. A generic structure of a NHC, with resonance structure s shown, can be found in Figure 1 1. Second , is that the car bene ligand had never been monodentate. It has always been bonded to another c helating moiety and used in bi or trid entate motifs. The bidentate structure is exemplified in a couple of the earliest reports, where platinum(II) tetracarbene dihalide complex es were synthesized using a bis(dicarbene) motif. 15 , 16 Tridentate ligands featuring NHCs followed not long after. 17 While a few of these compounds have been synthesized for medicinal purposes, 18 , 19 a majority tend to be for used in organic light emitting diodes (OLEDs) . Wh ile development of red and green OLEDs has advanced quickly, efficient, deep blue OLEDs have lagged considerably behind. 20 Platinum NHC s (Pt NHCs) have proven their potential in this field through a variety of cyclometallated complexes. Zhang and co workers used a carbene phenyl carbene, or CCC (based on the atoms that bond to

PAGE 27

27 Figure 1 1. Structure of a generic NHC showing resonance from nitrogen lone pairs. the me tal center), tridentate motif which shows excellent air stability under photoexcitation, but low emission quantum yields. 21 Both of these factors are attributed to excited state distortion around the platinum center in the excited state. Strassner and co workers have been the primary contrib utors to the field of blue phosphore scent platinum carbene compounds. While some of their work was me ntioned earlier, the recent focus has been on modifying a core structure consisting of a platinum center cyclometallated by both a phenyl NHC and an acetylacetonate (acac) ligand. These investigations focus primarily on the solid state properties, both in poly(methyl methacrylate) (PMMA) films and neat films. A core structure for this series of work is shown in Figure 1 2. The first modification made to this core structure was to look at the effect of phenyl connectivity in a 1,2,4 triazole 5 ylidene and su bstituent Figure 1 2. Core structure of the platinum NHCs explored by the Strassner Group. Figure 1 3. Structures of 1 aryl and 4 aryl 1,2,4 triazole 5 ylidenes studied in Reference 2 7. Triazole rings are numbered .

PAGE 28

28 effect on the aryl grou p . 22 Specifically, the effect of whether the phenyl is bonded to the NHC through the nitrogen at the 1 or 4 position. These structures are shown in Figure 1 3. The authors found that the 1 aryl complexes were up to three times more emissive than the 4 aryl complexes (~40% vs. ~13%), but no explanation wa s proposed for this phenomenon. The next report described the photophysics of four 1 (4 cyanoaryl)imidazole 2 ylidenes, with modifications to both the imidazolylidene and acac ligands. 23 The two complexes with a plain acac ligand showed quantum yields around 60%, while those with a dimesitoylmethane (mesacac) were in excess of 80%. The crystal packing, giv en by single crystal X ray crystallography, lends some insight as the cause of this observation. The packing of Pt NHCs with the minimal acac ligand show Pt Pt distances slightly shorter than the sum of the van der Waals radius. Thus, if a small percent of the complexes were to pack this way in the PMMA film, a reduction in quantum yield would be expected. However, when th e bulky mesacac ligand is used, no Pt Pt interaction is observed in the crystal packing. Interestingly, the emission lifetimes of the mes acac complexes are about half that of the acac compounds (~16 µ s vs. ~ 8.5 µ s ) . The compound with the highest quantum yield from this report (85%) is shown in Figure 1 4. Figure 1 4. Nitrile containing Pt NHC , featuring the mesacac ligand, with an 85% phosphorescence quantum yield in a 2 wt% PMMA film.

PAGE 29

29 Continuing with the theme of altering the acac ligand, one carbonyl moiety was ketoimine ligand. 24 Substitu e nts on the 4 position of the phenyl group were also tested opposite ketoimine ligand. The use of this NO bidentate ligand results in a mixture of isomers, both of which can be isolated. There is a preference for the imine nitrogen to be opposite of the carbene carbon, as evidenced by higher yields of this conforme r and DFT calculations showing it to be of a lower energy . Unfortunately, a majority of these complexes show quantum yields of 15% or less. DFT calculations again explained this phenomenon, as they showed significant distortion of the platinum square plane for these complexes in the triplet state. One ligand, however, stood out from the others, and it is illustrated in Figure 1 5. Modifying the aryl group attached to the NHC from phenyl to 1 dibenzo[b,d]furan 4 ylidene brings the quantum yield back up to 74 %. It was also shown that this is purely ketoimine back to acac pushes the quantum yield to 90%. Figure 1 5. Dibenzofuran containing Pt NHC with a 90% quantum yield in a 2 wt% PMMA film. The next report strays slightly from the blue emission region by extending the pi conjugation of the phenyl NHC ligand while holding acac constant. 25 The authors find that , as expected, by stra tegically fusing phenyl groups to the 4,5 bond of either ring of the phenyl NHC unit, decent quantum yields (> 40%) could be retained while shifting the

PAGE 30

30 emission color to the blue green and green domain. Addition of phenyl groups to the N 3 position had on ly a minimal effect. Returning to the success of the mesacac ligand, the Strassner Group probed the role of methyl groups at various locations of acac and dibenzolylmethane, the latter being the unmethylated parent compound of mesacac. 26 The group demonstrates in a step wise manner, that as steric bulk is increased from acac to d ipivaloylmethane to dibenzolylmethane the quantum yield increases (41 to 54 to 78%, respectively) . As explained earlier, this is due to inhibited Pt Pt interaction between complexes. Adding methyl groups to the dibenzolylmethane only increases the quantum yield a few percent, but gives the benefit of added solubility. Using these complexe s, devices were fabricated, and peaked at an external quantum efficiency (EQE) of 12.6%. The most recent report continues to use mesacac, but returns to the triazole NHC. 27 Intere stingly, most of the complexes reported show broad, structureless emissions that have a maxima shifted to the blue green region. This shift is attributed to a mesacac centered emission. Further support of this claim stems from a minimal change in emission maxima when substituents are changed on the phenyl NHC ligand. Although the emitting ligand has changed, emission quantum yields still tended to be greater than 80%. Taking the recent work of the Strassner Group as a whole, a couple key conclusions can be drawn. The most prominent trend is that of the bulkiness of the ancillary ligand. Evidence for this is provided by the systematic increase in bulk of the base acac system, leading to successively higher quantum yields. Rational of this trend is given by ex amining the solid state packing. Specifically, if the molecules are allowed

PAGE 31

31 to stack cofacially, interaction of platinum centers leads to non radiative deactivation of the excited state. This is illustrated in Figure 1 6, where two crystal packing diagrams are shown. The complex showing dimeric interactions has a quantum yield of 35% (2 wt% PMMA film); whereas the complex with the mesacac ligand shows no close packing, and has a quantum yield of 72%. Figure 1 6. Representative crystal packing diagrams o f Pt NHCs with A) low steric bulk of the ancillary ligand allowing close packing and dimeric interactions, and B) high steric bulk of the ancillary ligand preventing any dimeric or Pt Pt interaction. M. Tenne, S. Metz, G. Wagenblast, I. Munster and T. Stra ssner, Dalton Trans. , 2015, 44 , 8444 8455. Published by The Royal Society of Chemistry. The next point regards the ketoamine ancillary ligands used, and the detrimental effects on the quantum yields of the compounds featuring it. Not only did the ketoami ne ligands studied lack the steric bulk necessary to inhibit Pt Pt interactions, they are also capable of forming hydrogen bonds when dimerized. Unfortunately no ketoamine complex with mesityl side groups was modeled to test if the quantum yield returns to previously observed levels. An interesting observation , and insight to the electronic structure, stems from two complexes which show significantly stronger emission than the rest ( 20% QY) . Figure 1 7 shows the nitro functionalized emissive complex, and

PAGE 32

32 Figure 1 7. Nitro functionalized Pt NHCs which are modestly emissive despite bearing a ketoamine ancillary ligand. the acac congener of the other strongly emissive complex from this report was shown in Figure 1 5. As mentioned earlier, many of the complexes with the ketoamine ligand show distortion of the square plane around the platinum center in the triplet state. This is caused by the lowest ene rgy state being metal centered, and will be discussed in more detail in the next section. For current purposes, metal centered excited states are readily deactivated via non radiative decay, thus the excited state of the organic chromophore needs to be of lower energy for the complex to be emissive. Based on the weakly emissive complexes with the ketoamine ligand that were emissive with acac, the lowest metal centered excited state of these ketoamine complexes has been lowered further. While the emissive co mplexes may not seem to have much in common, the organic moieties have had their excited state energies lowered via two different routes. First, for the complex from Figure 1 5 with the dibenzofuran ligand , this is achieved by extending sys tem. Unfortunately there is no solid state packing information to make assertions about Pt Pt interactions for this complex, but the assumption based on the reports presented here would be that these interactions do not occur for this complex. The nitro functionalized Pt NHC shown in Figure 1 7 has lowered its excited acceptor. Despite this, the crystal

PAGE 33

33 packing shows significant stacking, which may explain why only a modest quantum yield is achieved. Many of the other complexes studied by the Strassner Group follow a similar trend, in that the complexes give better emission properties when the excited state energy of the organic chromophores have some separation from the metal centered excited state. The nitrile functionalized complexes (see Figure 1 4) illustrate this well, as they all have quantum yields above 60%. Polycyclic rings have the same effect on the quantum yield, but push the emission color towards the green region. Finally, a few tetraden tate (but not tetracarbene) platinum complexes using an OCCO 28 and CCCN 29 bondin g motif were synthesized by Li and Fleetham, respectively, and showed double digit EQEs. The structures of the best emitter from each paper referenced above are shown in Figure 1 8 . The tetradentate motif allows not only for greater quantum yields, but imp roved stability as well. This is because if one part of the without breaking three more bonds. Thus the disengaged ligand can come back and reform a bond with the metal cente r. Both of the complexes shown in Figure 1 6 have OCCO NCCC complex has a very narrow Figure 1 8 . Pt NHCs featuring tetradentate bonding motifs.

PAGE 34

34 Just as noteworthy is the EQE of an optimized device, which gives a peak efficiency of 24.8% , while holding at 22.7% at 100 cd m 2 , which is a practical luminance for display applications. The aforementioned device is, at the time of this writing, the reco rd holder for EQE of a pure blue emitting device. As Ancillary Ligands to Acetylene Chromophores R eport s of using a carbene as an ancillary ligand to other, lower energy, chromophoric ligands c ome primarily from the Venkatesa n group, and first appeared in 2011. 30 Th e first series of complexes again relied on a bidentate bis (NHC). The neutral NHC systems are good donors , which destabilize the metal centered states, and allow for efficie nt luminescent decay from the triplet state even when high bandgap acetylide chromophores are ligated. This is in direct contrast to the platinum phosphine (Pt PR 3 ) systems, where decay with the same or similar ligands is non radiative unless frozen in a g lass at 77 K. 31 Figure 1 9 shows the s tructures of the platinum carbene and platinum phosph ine complexes compared in the previous statement. Additionally, Figure 1 10 shows a qualitative orbital energy diagram illustrating the differences be tween Pt Figure 1 9 . P latinum carbene a nd platinum phosphine complexes with phenylacetylene chromophores. T he carbene complex shows room temperature phosphorescence in deaerated solution, while the phosphine complex do es not.

PAGE 35

35 Figure 1 10. Qualitative d orbital splitting diagram for a square planar platinum(II) phosphine complex versus a square planar platinum(II) carbene complex. Shown for comparison are the frontier orbitals of a hig h energy aryl acetylide ligand , such as phenylacetylene. NHCs and Pt PR 3 s that allows high energy emission at room temperature. Venkatesan and co workers continued to explore these high energy acetylide chromophores, and in 2014 described how altering the bidentate carbene ligand affected the emission energy of the system. 32 The first report of a monodentate NHC, with trans stereochemistry, as an ancillary ligand came in 2013 , also from the Venkatesa n group . 33 This study was published shortly before the work reported here in Chapter 2. 34 The authors chose to use a benzimidazole functionalized with dodecyl, or in one case isopropyl, chains to improve solubility. A generic structure is shown in Figure 1 11.

PAGE 36

36 Figure 1 11. A Pt NHC complex featuring monodentate carbene ligands and trans stereochemistry . The synthesis of these Pt NHC fin al products stemmed from a platinum acetylide precursor functionalized with cyclooctadiene (COD) as the ancillary. The COD was replaced by NHC ligands by refluxing the platinum COD, t hree equivalents of imidazolium salt, and t hree equivalents of potassium tert butoxide (KO t Bu) to give the cis Pt NHC in good yield. The trans Pt NHC complexes were then made from the cis complexes via thermal isomerization at 200°C, as direct synthesis of the trans Pt NHCs were very poor. The mechanism for this isomerization w as studied and found to be catalyzed by a Pt 0 NHC complex formed by reductive elimination of a diacetylide. The platinum of the Pt 0 NHC then coordinates to the platinum of a Pt II NHC, allowing for pseudorotation of the ligands, which is then followed by di ssociation of the Pt Pt interaction to give the trans Pt NHC in good yield . Continuing with their pursui t of blue phosphorescent OLEDs, high energy aryl acetylides chromophores were again used as the emitting species. Phenyl and 4 fluorophenylacetylide pe rformed the best in PMMA films, with a maximum quantum yield of 65%. Interestingly, when the dodecyl solubilizing groups were replaced with isopropyl groups (structure shown in Figure 1 12) , the quantum yield jumpe d to 80% , suggesting that the long alkyl c hains have a deleterious effect on emission efficiency.

PAGE 37

37 Figure 1 1 2 . Structure of the Pt NHC featuring monodentate carbene ligands demonstrating an 80% emission quantum yield in a PMMA film. In a change of pace, a Pt NHC extended chromophore. Coincidentally, it is the same chromophore ( 1 Ethynyl 4 (phenylethynyl)benzene (PE2)) as one that was used in Chapter 2, and the structure is shown in Figure 1 13. The reported soluti on photophysics of the Pt NHC complex featuring PE2 (again, shown in Figure 1 13) are significantly different from that which is studied in this manuscript. Possible explanations for this discrepancy may be the solvent used for photophysics, detrimental ef fects of benzamidazole versus imidazole, or that the alkyl chains provide minimal protection of the Pt center relative to the bulkier cyclohexyl groups used in Chapter 2. Figure 1 13. Structure of the Pt NHC featuring mo nodentate carbene ligands demonstrating an 80% emission quantum yield in a PMMA film.

PAGE 38

38 Excimers and Dimer ic Complexes Tethered by Flexible Linking Unit s Overview Excimer formation occurs when a molecule in the excited state interacts with a molecule in the ground state and the two associate to form a complex with a lower energy excited state. A Jablonski diagram illustrating this process is shown in Figure 1 14. One of the most well known systems that displays this b ehavior is pyrene . At low concentrations i n solution, pyrene exhibits only a single , structured emission. However, as the concentration of pyrene increases, a new, broad, low energy emission band starts to form. This process is concentration dependent because the key rate limiting step is the diff usional encounter of excited pyrene interacting with ground state pyrene. Thus, as the concentration of pyrene increases, the number of excited molecules and interactions increase, leading to the excimer emission band becoming over an order of magnitude st ronger than the monomeric emission. 35 Figure 1 14. A Jablonski diagram illustrating the energy levels of monomeric and excimeric excited states. If two pyrene molecules are covalently linked (tethered) through a molecule with three flexibilizing units (i.e. 1,3 (dipyrenyl)propane , structure shown in Figure 1 1 5 ), the

PAGE 39

39 excimer emission dominates the spectra regardless of concentration. 35 Three linking units are necessary for intramolecular excimer formation, because it allows for a cofacial overlap of the chromophores. Any shorter linke r will force the chromophores to diverge from each other , and will impede excimer formation. Figure 1 1 5 . The structure of 1,3 (di 2 pyrenyl)propane. Applications The utility of excimer formation and detection in organic molecules has led to a variety of applications for determining the kinetics and dynamics of a system . One of the earliest uses was to determine the lateral diffusion rate in the hydrophobic region of lipid bilayers. 36 This was accomplished by dispersing pyrene in an a queous lipid solution, and observing the ratio of monomeric emission to excimeric emission. Because of the low solubility of pyrene in water, all of the excimer emission can be assigned to interac tions within the lipid bilayer. Cyclization dynamics of poly mer systems can be explored by using excimeric chromophores as end groups. An early report by Redpath and Winnik described the cyclization rate constants of variable lengths of polystyrene with pyrene end groups. 37 Once again, detection of the desi red property, in this case cyclization of the polymer, relies on detection of excimer emission. If the chain does not cyclize, then no excimer emission should be detected. Dilute solutions were used to minimize intermolecular excimer interacti ons.

PAGE 40

40 Excimeri c Complexes Containing Platinum The earliest report of a platinum complex displaying excimeric properties came from Hiraga and co workers, using crystals of K 4 [Pt 2 (P 2 O 5 H 2 ) 4 ]·2H 2 O under variable pressure. 38 As pressure was increased from 1 atmosphere (atm) to 2.0 gigapascals (GPa), a new emission band formed between the fl uorescence and phosphorescence. Observation of an excimeric band in this system is attributed to reduced Pt Pt distance as p ressure increases, as observed by X ray diffraction and IR spectroscopy. Additionally, only well grown crystals were suitable for these experiments, as microcrystalline material did not show the same behavior. Organometallic platinum complexes are also cap able of excimer formation in solution. The compound Pt(5dpb)Cl, where 5dpb = 1,3 di(5 methyl 2 pyridyl)benzene ( shown in Figure 1 1 6 ) , shows typical excimeric behavior in solution, with a broad band emission at ~670 nm when dissolved at high concentrations . 39 Interestingly, this complex shows solid state emission properties in direct contrast to the platinum salt described in the previous report. Crystalline Pt(5dpb)Cl shows yellow luminescence, but when ground into a fine powder the emission became orange in color and had a similar max as the excimer in solution. X ray diffraction (XRD) data shows nearly identical packing in both the crystalline and powder samples, so stronger Pt Pt interactions wer e ruled out. However, the surface area had been greatly increased by creating the powder, which generates a high number of labile surface molecules. These surface molecules may move to for a dimer or aggregate, giving rise to low energy emission. Observati on of excimer formation is most commonly observed as fluorescence; however, it can also be observed in the form of phosphorescence 40 or triplet triplet absorption. 41 43 A repor t from 2007 suggested the ability of a platinum acetylide to form

PAGE 41

41 Figure 1 1 6 . The structure of Pt(5dpb)Cl. an excimer in the triplet state , which was then probed by transient absorption . 44 The PE2 , was designed to be an liquid at room temperature, and is shown in Figure 1 1 7 . A highly soluble oil was needed for this experiment because very high loading levels were necessary. A 57 mM PE2 octyl solution in benzene was required to see this behavior, and the transient absorption spectrum is reproduced in Figure 1 18. Additionally, no emission was observed from this excimer, thus transient absorption was the only method that could be used to study it. Figure 1 1 7 . The structure of PE2 octyl. Looking closer at Figure 1 18, one can see the absorption of a new species taking form, which is attributed to the triplet excimer. Interestingly, the absorptio n of this excimer is relatively narrow and shows a semblance of vibronic structure, suggesting localization of the triplet exciton on a single PE2 octyl unit, which has been previously shown for the PE2 chromophore in a platinum complex. 45 Thus it is reasonable to assess that excimer formation is primarily of ligand ligand interaction. Kinetic studies of of the PE2 octyl TA using free PE2 ligand as a triplet qu encher showed almost an order of magnitude greater quenching than excimer formation. The preference for quenching via PE2 ligand over excimer formation implies that the potential energy barrier to

PAGE 42

42 Figure 1 18. Transient absorption spectrum of 57 mM PE2 octyl in deoxygenated benzene after 476 nm laser excitation. Decay traces start at t=0 and were taken in increments of 150 ns. Reprinted from Chemical Physics Letters, Vol. 447, Slagle, J. E. et al. Triplet excimer formation in a platinum acetylide, 65 68 , Copyright 2007, with permission from Elsevier. excimer formation is affected by the sterics of the alkyl phosphine ligands around the platinum center. The concentration of PE 2 octyl in solution is noteworthy on its own, because it opens the door to poten tial use in non linear optical applications. There are many potential uses for non linear materials including, but not limited to goggles, windscreens, and optical coatings. Despite this, there are very few reports in which these molecules have been incorp orated into a matrix to study the solid state properties for potential commercialization. 46 48 Practically speaking, the concentration required to reach a useful saturation level should be as high as possible. Saturation is dependent on the deactivation of the excited state, which is directly proportional to the number of collision events between separate molecules, and thus the concentration. Monosubstituted Plat inum Aryl Compounds as Precursors to Mixed Ligand Platinum Aryl/Acetylene Systems As noted earlier, platinum acetylide complexes have been, and continue to be, intensely studied for their photophysical properties. However, the study of platinum

PAGE 43

43 complexes w ith monodentate aryl ligands leaves a considerable amount to be desired. Platinum aryl (Pt aryl) complexes of higher denticity are indeed well known, and were repeatedly mentioned earlier in this manuscript in various cyclometallating ligands. The forced c o planar conformation of ligand and metal presumably has a large effect on the photophysical properties of the system compared to if the aryl ligand were allowed to rotate freely . Platinum systems, with phosphine ancillary ligands, containing monodentate a ryl ligands typically are bound through a nitrogen atom. 49 The result though, is a non neutral system with labile ligands. When the literature is searched for platinum aryl systems, with the aryl group bonded through carbon, a vast majority of the aryl groups are monocyclic with only simple functional groups (e.g. phenyl and its derivatives) . This lack of variety stems from the harsh conditions that had been required for platinum carbon (Pt C) bond formation. 50 , 51 Additionally, these harsh conditions led to primar ily disubstitution of the aryl ligand. The monoaryl platinum complex can be obtained from the diaryl complex, but this requires another harsh reaction (HCl in Et 2 O). 50 A second route to forming this platinum carbon bond lies with oxidative addition. This route too is not without its drawbacks, as it requires the use of zero valent metal prec ursors, 52 54 which are air sensitive. Recently, a new platinum carbon bond formation reaction was devised by a member of the Schanze Group. 55 This new reaction draws on the methodology o f the Stille reaction, in that is uses an aryl stannane in the presence of 10 mol% Cu(I) to form the Pt C bond under mild conditions. When the aryl stannane is reacted with cis (PBu 3 ) 2 PtCl 2 at any ratio, the monosubstituted Pt aryl product is strongly favo red . With a

PAGE 44

44 facile synthetic pathway in hand , the route to chromophoric mixed ligand platinum aryl/acetylide (not to be confused with aryl acetylide) complexes has been paved. It would be remiss to not point out that many mixed aryl acetylide platinum comp lexes (mixed ligand complexes) have already been synthesized. However, a vast majority of these complexes use a phenyl platinum moiety (Ph)(PBu 3 ) 2 Pt(CCR) as a non chromophoric end group. 56 58 Another recent report from the Low Group details the synthesis of a set of mixed ligand complexes featuring triaryl amine chromophores , all of which are shown and given reference numbers in Figure 1 1 3 . 59 Th is report describes the synthesis, electrochemical, and spectro electrochemical properties of these compounds. The authors use the synthetic route of oxidative addition, proceeding in high yield, to generate their monosubstituted platinum aryl complex. They then use Hagihara coupling with a mono, di, or trifunctional a cetylide ligand to synthesize a set of mixed aryl/acetylide ligand platinum complexes. The monofunctional acetylide ligands were chosen with electrochemistry experiments in mind, as one of the ligands should not show an oxidation wave ( CCPh 4 OMe), while the other ligand should show a strong oxidation ( CCPhN(4 (OMe)Ph) 2 ). Specifically, the aforementioned ligands will serve two purposes. First, the mixed ligand platinum complex functionalized with the redox inactive acetylide , 1 , will allow for characteriz ation of the oxidation wave of the triphenyl amine moiety that is bonded to the platinum center through an aryl carbon. Second, the mixed complex functionalized with the redox active triarylamine acetylide ligand, 2 , will give insight to how strongly the a ryl and acetylide electrophores are coupled through the platinum center. A platinum dimer ( 3 ) and trimer ( 4 ) were also synthesized to study the redox properties when

PAGE 45

45 Figure 1 1 3 . The four mixed ligand platinum aryl/acet ylide complex es synthesized by the Low Group. conjugated organic linking units are used. Upon examination of the cyclic voltammograms (CVs) of all four complexes, the authors note that distinct oxidation waves are present for each non identical triarylamin e moiety. Thus, the individual electrophores are acting independently, as they are isolated from each other by the platinum center(s). Further experimentation using UV Vis NIR spectroelectrochemistry shows the formation of weak intervalence charge transfer (IVCT) bands in the 8500 cm 1 region upon oxidation of 2 and 4 ( 2 [ 2 ] + and 4 [ 4 ] 3+ , respectively). Upon further oxidation

PAGE 46

46 ([ 2 ] + [ 2 ] 2+ and [ 4 ] 3+ [ 4 ] 4+ ), the IVCT band collapses, allowing assignment of the band to 2 NArCC} {PtArN + 2 }. From these data, a coupling of 170 cm 1 for the ligands of 2 , and 130 cm 1 for the ligands of 4 were calculated. Thus these complexes show very weak interactions between aryl and acetylide chromophores.

PAGE 47

47 CHAPTER 2 P HOTOPHYSICAL P ROPERTIES OF TRANS P LATIN UM A CETYLIDE C OMPLEXES F EATURING N H ETEROCYCLIC C ARBENE L IGANDS Background Aryl acetylide ligands have been tuned by altering conjugation length, 6 , 12 v arying the internal and external functionality, 6 , 12 , 60 , 61 including donor acceptor motifs, 6 and cross conjugation. 7 , 62 A square planar platinum(II) complex featuring two aryl acetylide ligands can adopt either cis and trans geometries; 12 however, the trans configuration is prevalent in the complexes that have been reported. Unlike most platinum acetylides that contain monodentate ligands, bi and tr identate ligands use a variety of atoms for bonding to the metal center and altered structural motifs to achieve the desired chelating and electronic effects. 12 , 30 , 63 . The focus of the work reported here was to directly compare the properties of platinum phosphine complexes studied previously in the Schanze Group 64 to the newly synthesized Pt NHCs. The working theory was that replacement of the phosphine moieties with NHCs would reduce the spin orbit coupling of t he system, allowing for longer excited state lifetimes and enhanced triplet properties. In this chapter , we report the synthesis and photophysical properties of a series of trans platinum acetylide complexes that feature monodentate N heterocyclic carbene ligands in place of phosphine ligands. The aryl acetylide ligands used in this work were chosen for their substantial two photon absorption (2PA) cross sections and proven non linear optical R. W. Winkel, G. G. Dubinina, K. A. Abboud and K. S. Schanze, Dalton. Trans. , 2014, 43 , 17712 17720. Reproduced by permission of The Royal Society of Chemistry. http://pu bs.rsc.org/en/content/ articlelanding/2014/dt/c4dt01520g

PAGE 48

48 properties. 2 , 64 Finally, DFT computations were used to gain insight into the exact nature of the dominant electronic transitions for each complex. Synthesis and Structu re Platinum NHC complexes are generally prepared by first generating and isolating the silver carbene precursor, then in a separate reaction transmetallating to platinum . 65 It has been shown that the needed silver carbenes can be generated in high yield i n situ , but upon isolation material was lost . 66 To alleviate this problem in the present work, a one pot method was developed as shown in Figure 2 1 , and by using this approach trans (ICy) 2 PtCl 2 ( 1 ) was prepared in a 95% yield. This one pot reaction is quite robust, as there is no detectable formation of any byproducts even after extended reaction times, which is consistent with previously Figure 2 1. One pot synthesis of trans (ICy) 2 PtCl 2 ( 1 ). Figure 2 2 . ORTEP diagram of trans (ICy) 2 PtCl 2 ( 1 ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity.

PAGE 49

49 reported reacti ons . 65 The 1 H NMR spectrum could not confirm the trans nature of the complex, although it did differ as expected from the spectrum of the known cis (ICy) 2 PtCl 2 . 67 Sin gle crystal X ray crystallography was employed to confirm the structure of 1 , and the structure is shown in Figure 2 2 . Once the structure of 1 was confirmed, the complex was subjected to Hagihara reaction conditions as depicted in Figure 2 3 to introduce three different aryl acetylide ligands ( 1 ethynyl 4 (phenylethynyl)benzene ( PE2 ), 2 (9,9 diethyl 9 H fluoren 7 yl)benzo[ d ]thiazole ( BTF ), and 9,9 diethyl 7 ethynyl N , N diphenyl 9 H fluoren 2 amine ( DPAF )). Single crystal X ray crystallography was again used in conjunction with 1 H NMR to confirm the structures of trans (ICy) 2 Pt(PE2) 2 ( 2a ), trans (ICy) 2 Pt(BTF) 2 ( 2b ), and trans (ICy) 2 Pt(DPAF) 2 ( 2c ). Figure 2 3 . Hagihara reaction to g enerate the platinum acetylide c ompounds 2a c . ORTEP diagrams of 2a c are shown in Figure 2 4 . Important bond lengths and angles are presented in Table 2 1 and compared to their respective tributyl phosphine (PBu 3 ) analogues. The trans phosphine complexes (PBu 3 ) 2 Pt(PE2) 2 , (PBu 3 ) 2 Pt(BTF) 2 , and (PBu 3 ) 2 Pt(DPAF) 2 will be referred to as 3a c respectively. A full table of crystallographic data, and the structures of 2a c with the asymmetric unit fully labeled , can also be found in Appendix C .

PAGE 50

50 Fig ure 2 4. ORTEP diagrams of the molecular structures of 2a c , with thermal ellipsoids given at the 50% probability level. The hydrogen atoms are omitted for clarity. C: Black, N: Blue, S: Yellow, Pt: Pink, Chlorine: Green Table 2 1. Key i nteratomic distances (Ã…) and a ngles (deg) for p latinum acetylide c omplexes. Complex Pt1 C1 Pt1 C16(1) a a C1 Pt1 C16 N C1 Pt1 C16 b 2a 2.033(3) 2.016(3) 1.198(4) 88.64(10) 63.3 2b 2.012(5) 2.012(5) 1.180(6) 88.88(17) 74.3 2c 2.029(2) 2.032(2) 1.175(3) 89.73(8) 65.9 3a 68 1.981(6) 1.229(8) 3b 2 2.000(3) 1.208(4) 3c 2 2.023(8) 1.144(12) a Numbers in parentheses denote atomic numbering used in the crystal structures of the phosphine complexes. b N C1 Pt1 C16 dihedral ang les were selected to be acute and positive for ease of comparison. Two major observations can be made from the crystal structures of 2a c . The first is a correlation of the N C1 Pt C16 dihedral angle (i.e., twisting of the carbene ligand) and the length o f the Pt1 C1 bond. In particular, as the dihedral angle becomes smaller, the platinum carbene bond length increases. Computations have shown that C1 bonding interaction . 69 As the N C1 Pt C16 dihedral angle decreases, the overlap of the Pt d xy and C1 p x orbitals responsible for the backbonding also decreases, resulting in a longer bond. Additionally, the Pt1 ngths lend insight into the -

PAGE 51

51 degree of interaction (conjugation and spin orbit coupling) between the platinum center and the aryl acetylide ligands. Compounds 2a and 2b lengths than 3a and 3b and compound 2a also has a longer Pt1 C16 bond than 3a , both Unfortunately, conclusions cannot be drawn about the Pt1 C16 distance of 2b nor either property of 2c , because the values are within experimental error. As a further measure of the degree of metal ligand conjugation and extent of platinum backbonding, IR s pectra were recorded for 2a c . An expanded view of the l IR spectra can be found in Appendix C . Viewing the platinum center as a donor , increasing holds true in the experimental data, as the BTF ligand in 2b is a good acceptor, giving the complex the 1 . Complexes 2a and 2c , which contain at slightly higher Fig ure 2 5. 2a c .

PAGE 52

52 frequency ~2087 cm 1 . The second triple bond of the PE2 ligand in 2a does not display a strong vibrational mode and cannot be dir ectly assigned to a peak in the recorded spectrum. In summary, the X ray structural data suggest that there is a slightly greater interaction between the metal center and the alkyne ligand in the phosphine complexes relative to the carbene congeners. Photo physical Properties The ground state absorption spectra for the platinum acetylide series in THF solution are shown in Figure 3 along with fluorescence and phosphorescence spectra of the complexes in aerated solution and after at least three freeze pump th aw vacuum degassing cycles, respectively. (Phosphorescence spectra were initially attempted by degassing with argon; however, due to their long triplet lifetimes, vide infra , compounds 2a c are very sensitive to oxygen and significantly reduced phosphoresc ence was observed for argon deareated solutions.) Absorption, fluorescence, and phosphorescence maxima, as well as extinction coefficients, emission quantum yields, and emission lifetimes are given in Table 2 2 . The absorption maximum of 2b is red shifted relative to that of 2a due to the increased conjugation length offered by the electron accepting benzothiazole moiety. In 2c , when benzothiazole is substituted for the electron donating diphenylamino moiety, the absorption maximum lies between 2a and 2b . A bsorption maxima of 2a c are within 6 nm of 3a c , with 2a and 2b being red shifted and 2c slightly blue shifted. The extinction coefficients can be used as a qualitative measure of conjugation through the platinum center . 64 When the extinction coefficients of 2a c are compared to 3a c , 2a shows similar ligand interaction with the platinum center, whereas 2b and 2c show significantly less interaction Like t he ground state absorption, the fluorescence is red shifted for 2a and 2b and blue shifted for 2c

PAGE 53

53 Figure 2 6 . The A) g round state absorption, B) normalized fluorescence , a nd C) normalized phosphorescence spectra of 2a c in THF. Excitation was at the ground state absorption maxima , and the plot s w ere normalized. Table 2 2 . Summary of p hotophysical d ata for platinum a cetylides 2a c and 3a c . Complex Abs max (nm) (M 1 cm 1 ) Fl max (nm) Fl a Fl b (ns) Ph max (nm) Ph e Ph b (µs) TA max T1 Tn b (µs) 2a 357 83000 399 0.002 0.710 c 531 0.541 575 f 584 80.8 2b 408 122000 452 0.043 < 0. 1 00 d 576 0.220 355 749 95.9 2c 380 128000 393 0.041 < 0. 1 00 d 534 0.450 1080 g 606 111 3a 64 353 83000 391 0.0005 < 0. 1 00 d 527 0.0108 48.9 577 15.4 3b 64 402 154000 436 0.0092 < 0. 1 00 d 567 0.1752 75.6 660 13.3 3c 64 384 160000 401 0.0059 < 0. 1 00 d 533 0.3552 61.3 612 8.9 a Measured at RT using Ru(bpy) 3 Cl 2 in aerated water fl = 0.0379) 70 fl = 0.33) 71 as standards. b All decay profiles can be found in Appendix C . c 1 1 ) = 0.309 ns (66.1%), 2 2 ) = 1.37 ns (31.8%), 3 3 ) = 3.33 (2.1%). d Lifetime is shorter than the lower threshold of the PicoQuant instrument, 1 00 ps. e Measured at RT using 9,10 fl = 0.75) as a standard. 72 f 1 1 ) = 635 µs (87.9%), 2 ( 2 ) = 143 µs (12.1%). g 1 1 ) = 1693 µs (48.9%), 2 2 ) = 495 µs (51.1%).

PAGE 54

54 relative to the respective PBu 3 analogues. The phosphorescence of 2a c all show less than a 10 nm red shift relative to 3a c , with 2b again showing the greatest difference. Cons istent with reduced metal ligand interaction, the fluorescence quantum yields of 2a c are 4 7 times greater than 3a c , but nonetheless are comparatively low on an absolute scale (<4.5%). Only the fluorescence lifetime of 2a was long enough to be measurable with our apparatus, with the primary component being ~310 ps and two longer lived components greater than 1 ns with comparatively low amplitude. Interestingly, the phosphorescence lifetimes are enhanced ~5 times for 2b and greater than an order of magnitu de for 2a and 2c in the carbene complexes compared to the PBu 3 congeners. The phosphorescence quantum yields are similarly enhanced in the carbene complexes relative to the phosphines. Nanosecond transient absorption (TA) data were collected on 2a c to gai n insight into the triplet excited states, and the absorption difference spectra, assigned to triplet triplet absorption, are shown in Figure 2 7 . All of the spectra show bleaching in the region of the ground state absorption, with a strong, broad, and lon g lived absorption in the visible region. The triplet triplet spectra are qualita tively similar to those for the phosphine analogues, consistent with the notion that the triplet is localized on the arylacetylide ligands . 64 The TA absorption maximum of 2b (749 nm) was significantly red shifted relative to 3b (660 nm), while 2a (584 nm) and 2c (606 nm) remained almost unchanged as in 3a (577 nm) and 3c (612 nm). By analogy to the phosphorescence decay lifetimes, the lifetimes of the triplet triplet absorption of 2a c are significantly enhanced relative to 3a c . However, the transient absorption decay lifetimes were shorter than expected on the basis of the ph osphorescence lifetimes (Table 2 2 ), and it

PAGE 55

55 Figure 2 7. Principal component transient absorption spectra of 2a c , with excitation at absorptivity valu e of 0.62 at 355 nm after four freeze pump thaw cycles. was suspected that they were attenuated by triplet triplet annihilation, which would be power dependent. A power dependence study of the transient absorption decay kinetics was carried out on 2a and t he results showed that the kinetics are power dependent ( see Appendix C ). Nanosecond open aperture z scan studies were carried out on aerated solutions of 2a c to determine their relative non linear absorption response with respect to 3c . Complex 3c was us ed as a reference because of its well established non linear response under similar laser excitation conditions . 2 , 64 When this experiment is done using near infrared nanosecond laser pulses, the combined effects of 2 photon absorption and triplet triplet excited state absorption combine to give rise to the overall optical attenuation. An excitation waveleng th of 606 nm was chosen due to the lack of any ground state absorption at this wavelength. Figure 5 shows the relative z scan response of each complex; the order of response of this series of complexes is 2a < 2b < 2c 3c . Complexes 3a c follow the same t rend . 64

PAGE 56

56 Figure 2 8. Open aperture z scan transmittance using 606 nm light. Solutions were 1 mM in aerated THF and the laser 3c was used as the reference. Electronic Structure Calculations To further probe the nature of the metal chromophore interaction, DFT and TD DFT calculations were performed in the Gaussian 09, revision C.01, 73 suite of programs at the B3LYP level with the 6 31G(d) basis set for nonmetals and the SDD basis set for Pt. Carbene cyclohexane groups were replaced by methyl groups to improve compu tational efficiency, and these truncated complexes are denoted as . The property of interest is the nature of the electronic excitation corresponding to the ground state absorption maximum of each complex. TD DFT confirms that these transitions are p HOMO 2 3. As expected, all HOMOs are orbitals with the occupied Pt d xy giving rise to interaction between the two ligands. In turn, the contribution from the platinum. The predicted energies of these theoretical electronic transitions all match well with the experimental data. Computed vertical excitation energies for compounds ar e red shifted by 17 , 21 , and 2 nm, respectively,

PAGE 57

57 relative to the experimental data. Charge difference densities for the dominant singlet singlet transitions of are shown in Figure 2 9 . CDDs show how the charge moves in an electronic transition, and illustrates the charge delocalization across the ligands in and and charge buildup on the fluorene moiety of . Relevant molecular orbitals , enlarged CDDs , and a table of all significant vertical excitations for each compound can be found in Appen dix C . Table 2 3 . Summary of TD DFT c omputations for the S 0 S 1 transition of . Complex Wavelength (nm) Orbital Transitions (% contribution) Oscillator Strength, f 373.8 ( 2.4%) (97.6%) 2.3830 429.3 ( 5.3%) (94.7%) 2.2796 381.5 (18.3%) (81.7%) 2.6813 Figure 2 9. DFT optimized structures (top row), and charge difference densities (bottom row) for t he calculated transitions corresponding to the experimental ground state absorption maxima. Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in an electronic transition. CDDs were imaged at an isovalue of 0.0004. Summary and Conclusions In summary, three platinum acetylide complexes ( 2a c ) featuring N heterocyclic carbenes with the PE2, BTF, and DPAF chromophores, respectively, were synthesized from the platinum carbene precursor ( 1 ), via a Hagi hara reaction to generate the desired products. These complexes were characterized by NMR spectroscopy and

PAGE 58

58 single crystal X ray crystallography to confirm that the ligands were in a trans geometry with respect to the platinum center. In order to compare th e effect of substituting carbene ligands for the tributyl phosphine ligands which have been used in previous related platinum acetylide complexes, photophysical studies were performed and X ray data were analyzed. Optical transitions of the carbene complex es are very similar to those of the conjugated ligand chromophores. However, there is evidence for slightly less metal ligand interaction (and consequent reduced spin orbit coupl ing) in the carbene complexes. Lower extinction coefficients imply weaker conjugation between the two chromophores relative to their phosphine congeners. Fluorescence quantum yields are increased for all carbene complexes, and unlike any of the phosphine c omplexes, the fluorescence lifetime of 2a was long enough to be measurable. The phosphorescence lifetimes are also significantly enhanced, as reduced spin orbit coupling decreases the probability of the spin forbidden transition back to the singlet ground state. The longer phosphorescence lifetimes result in enhanced phosphorescence quantum yields for all carbene complexes, as there is more time for radiative decay to occur. The triplet triplet absorptions also followed the same trend, exhibiting significan t increases in the triplet lifetimes. In the solid state, single crystal X ray crystallography generally showed that the carbene complexes have longer Pt1 clear trends cannot be established since in some cases the bond lengths for the carbene and phosphine complexes are within experimental error. Decreased

PAGE 59

59 C16 bonds due to reduced metal evels, which is generally consistent with the experimentally observed trends. FTIR was also used to probe the degree of backbonding in the in 2b , which contains the electron accepting BTF ligand. The neutral PE2 ligand and electron donating DPAF ligand show approximately equivalent stretching frequencies. A non linear absorption by open aperture, nanosecond z scan revealed that the carbene complexes display strong non linear a bsorption, with a similar response compared to their phosphine analogs. The non linear absorption response is dependent on the acetylide chromophore, following the trend ( 2a (PE2) < 2b (BTF) < 2c (DPAF) ~ 3c (DPAF)). This is likely due to the response bein g triggered by two photon absorption, which has the same trend of increasing response as the z scan experiment . 2 This work shows that platinum acetylide complexes featuring carbene co ligands show promising photophysical and non linear optical properties. The properties of the complexes suggest that they may find use in material applications such as optical power limiting or photosensitization of singlet oxygen with near infrared, two photon excitation. Given the wide range of carbene li gands that are available , 74 their use can allow tuning of the electronic and ste ric properties of the Pt acetylide complexes, affording systems with tunable properties that can be optimized for specific applications. Experimental General Remarks All reactions were carried out under argon atmosphere. All starting chemicals used for the synthesis of the trans platinum acetylide complexes were purchased from

PAGE 60

60 commercial suppliers, reagent grade, and used without additional purification. K 2 PtCl 4 was purchased from Strem Chemicals. Solvents were of reagent grade unless otherwise noted. PE2, BTF, and DPAF ethynyl ligands were synthesized according to literature procedures . 2 , 64 trans (ICy) 2 PtCl 2 was synthesized by a one pot synthesis via modified transmetallation procedures . 65 , 75 Silica gel (230 400 mes h, 60 Å, Silicycle Inc.) was used for column and flash chromatography. 1 H (500 MHz) and 13 C (125.7 MHz) NMR spectra were recorded on a Varian Inova spectrometer. The chemical shifts were reported in ppm relative to tetramethylsilane (TMS) or residual proto nated solvent peaks in 1 H NMR spectra. Mass spectrometry and elemental analysis were performed by Mass Spectrometry Services and CHN Elemental Analysis Services respectively, both of which are located in house at the University of Florida. Fourier Transfor m Infrared Spectroscopy (FTIR) FTIR spectra were taken on a Bruker Vertex 80V FTIR using a Pike Technologies GladiATR ATR accessory and the OPUS 6.5 software package. The spectra were taken neat with the apparatus open to the atmosphere. Data was acquired in the range of 4500 400 cm 1 with a precision of ±4 cm 1 . X ray Structure Determinations A CHCl 3 solution of 1 and a CH 2 Cl 2 solution of 2b were layered with hexanes and refrigerated. After two days, colorless prisms of 1 and yellow prisms of 2b were isola ted. CH 2 Cl 2 solutions of 2a and 2c were placed in a vial, which was then placed in a beaker of hexanes and covered for vapor diffusion. After two days colorless needles of 2a and colorless prisms of 2c were isolated. X ray intensity data were collected at 100 K on either a Bruker SMART 1 and 2c ) and on a Bruker

PAGE 61

61 DUO 2b and 2a , respectively), from an ImuS power source and both using an APEXII CCD area detector. Raw data fra mes were read by program SAINT 76 and integrated using 3D profiling algorithms. The resulting data were reduced to produc e hkl reflections and their intensities and estimated standard deviations. The data were corrected for Lorentz and polarization effects and numerical absorption corrections were applied based on indexed and measured faces. The structure was solved and refi ned in SHELXTL6.1, using full matrix least squares refinement. The non H atoms were refined with anisotropic thermal parameters and all of the H atoms were calculated in idealized positions and refined riding on their parent atoms. Further details about th e asymmetric units of all reported complexes are given in Appendix C Absorption and Emission Spectroscopy Steady state absorption spectra were recorded on a Varian Cary 50 or a Cary 100 dual beam spectrophotometer. Corrected steady state emission measureme nts were performed on a Photon Technology International spectrophotometer (QuantaMaster). Absorption and fluorescence samples were run in aerated solution, while phosphorescence samples were deaerated with at least four freeze pump thaw cycles. Optically d ilute samples with O.D. < 0.1 at the excitation wavelength were used. Fluorescence quantum yields were determined by relative actinometry, with Ru(bpy) 3 Cl 2 Fl = 0.0379 in air saturated water) 70 for (ICy) 2 Pt(BTF) 2 , and Fl = 0.33 in THF) 71 for (ICy) 2 P t(PE2) 2 and (ICy) 2 Pt(DPAF) 2. Phosphorescence quantum yields were determined in the same manner using 9,10 Fl = 0.75 in cyclohexane) 72 as an actinometer.

PAGE 62

62 Fluorescence lifetimes were obtained with a PicoQuant FluoTime 100 Compact Fluorescence counting technique (TCSPC) with a PicoQuant FluoTime 100 Compact Fluorescence Lifetime Spectrophotometer. A UV pulsed diode laser provided the excitation at 375 nm (power < 10 mW). The laser was pulsed using an external BK Precision 4011A 5MHz function generator. The decay of 2a was obtained using tri expone ntial fitting parameters (Fluo Fit software). The decays of 2b and 2c were faster than the instrument response time (<1 00 ps). Phosphorescence lifetimes were obtained with a multichannel scaler/photon counter system with a NanoQuant FluoTime 100 Compact Pho sphorescence Lifetime Spectrophotometer. A UV pulsed diode laser provided the excitation at 375 nm (power < 10 mW). The laser was pulsed by a PDL800 B, which is a pulsed diode laser driver. Optically dilute solutions were freeze pump thawed a minimum of fo ur times. The decay of 2b was obtained using single exponential fitting parameters, and the decays of 2a and 2c were obtained using biexponential fitting parameters (FluoFit software). Nanosecond Transient Absorption (TA) Spectroscopy Measurements were per formed on an in house apparatus that is described in detail elsewhere . 77 355 nm, 10 ns fwhm, 180 µJ pulse 1 ) was used as the excitation source. Pr obe light was produced by a xenon flash lamp and the transient absorption signal was detected with a gated intensified CCD mounted on a 0.18 M spectrograph (Princeton PiMax/Acton Pro 180). Solutions had a matching optical density of 0.62 after a minimum of four freeze pump thaw cycles. An initial CCD image capture delay of 100 ns following the laser pulse was used to ensure full conversion to the triplet state before observation.

PAGE 63

63 An average of 100 images were acquired and the laser energy was 180 µJ/pulse, which was established as being sufficiently low to minimize triplet triplet power dependence. Open Aperture Z Scan Nonlinear transmission measurements were performed via an open aperture z scan apparatus . 78 The excitation wavelength (606 nm) was generated by a Continuum Surelite OPO Plus pumped with the third harmonic (355 nm) of a Continuum Surelite II 10 Nd:YAG laser. The laser beam was split with a 50:50 beam spl itter to two OPH PE10 SH V2 pyroelectric detectors, which measured the transmitted pulse energy as a function of the input pulse energy using an Ophir Laserstar dual channel optical laser energy meter. The beam was focused with a 25.4 mm diameter, 50.8 mm focal length concave lens. A ThorLabs motorized translation stage (Z825B and TDC001) allowed mm movement along the z axis. Computational Details DFT and TD DFT calculations were performed in the Gaussian 09, revision C.01 , 73 suite of programs at the B3LYP level with the 6 31G(d) basis set for nonmetals and the SDD basis set for Pt. Carbene cyclohexane groups were replaced by methyl groups to improve computational e fficiency. The ground state structures were optimized in the gas phase from idealized starting configurations without symmetry constraints. The optimized structures were confirmed to be minima by the lack of imaginary frequencies . Structures and orbitals w ere visualized using Chemcraft Version 1.7 , 79 which was also used to generate charge difference density (CDD) plots. CDDs were imaged at an isovalue of 0.0004.

PAGE 64

64 Synthesis of trans (ICy) 2 PtCl 2 (1) To a flask charged with 60 mL of argon degassed dichloromethane, 1 , 3 bis ( cy clohexyl ) imidazolium tetrafluoroborate ( 1 .0 g, 3.1 mmol, 1 eq.) and AgCl (0.492 g (3.4 mmol, 1.1 eq.) were added , and the solution was bubbled with argon for an additional 15 minutes. On the side, 0.125g of NaOH (3.1 mmol, 1 eq.) and ~1 eq of [Bu 4 N]Cl were added to 10 mL of argon degassed deionized water. The aqueous solution was then added to the reaction mixture via cannula. The mixture was bubbled with argon for an additional 15 minutes before being allowed to react for 15 hours at room temperature with vigorous stirring in the dark. The reaction mixture was then filtered and washed with deionized water (3 x 5 mL). The organic layer containing 1 , 3 b is ( cyclohexyl ) imidazole 2 yl silver chloride was then separated and used for the next step without identific ation or further purification. K 2 PtCl 4 (0.615 g, 1.48 mmol, 0.475 eq.) was added to the resulting dichloromethane solution of 1 , 3 bis(cyclohexyl)imidazole 2 yl silver chloride and the mixture was stirred vigorously for 48 hours at room temperature under ar gon in the dark. The reaction mixture was then filtered to remove the precipitate and the filtrate was reduced in vacuum. This was then purified by column chromatography (CH 2 Cl 2 ) resulting in product 1 as a white powder. Yield, 1.08 g, 95% Mp: 275 °C (dec) . Anal. Calc. for C 30 H 48 N 4 PtCl 2 : C, 49.31; H, 6.62; N, 7.67. Found: C, 49.20; H, 6.73; N, 7.44. 1 H NMR (500 MHz, CDCl 3 CH=CH ), 5.53 (tt, J = 11.6 Hz, 4.0 Hz, 4H, CH), 2.39 (d br , J = 10.9 Hz, 8H, CH 2 ), 1.94 (d br , J = 12.5 Hz, 8H, CH 2 ), 1. 79 (d br , J = 13.2 Hz, 4H, CH 2 ), 1.64 1.45 (m, 16H, CH 2 ), 1.25 (qt, J = 12.8 Hz, 6.3 Hz, 4H, CH 2 ). 13 C { 1 H} NMR (126 MHz, CDCl 3 -

PAGE 65

65 TOF MS (found) m/z. 753.2809 [M+Na] + , 1483.5698 [2M+Na] + . (calcd) 753.27 87 [M+Na] + , 1483.5683 [2M+Na] + . General Procedure for the Hagihara Coupling Reaction (2a c) To a flask charged with 10 mL of Et 2 N H and 20 mL of dichloromethane, 0.146 of 1 (0.146 g, 0.2 mmol, 1 eq.) and 0.4 mmol ( 2 eq. ) of the corresponding deprotected ary l acetylide (PE2, BTF, DPAF) were added. The solution was then bubbled with argon for 15 m inutes before addition of 5 mol % CuI (1.9 mg, 0.01 mmol, 0.0 25 eq). The reaction mixture was then stirred for 40 hours at room temperature , under argon , in the dark. The final mixture was diluted with 40 mL of dichloromethane, washed with deionized water (3 x 20 mL), dried over sodium sulfate and filtered. The filtrate was reduced in vacuum until a significant amount of precipitate formed. The precipitate was filtered, washed with cold methanol and hexanes and dried under vacuum. The obtained products 2a c showed a single spot in TLC and were pure by NMR. Additional fractions were obtained from the filtrate and purified by flash chromatography (eluent: 90:10 ethyl aceta te/hexanes gradient shifted to 10:90 ethyl acetate/hexanes). Complexes 2a and 2c were obtained as off white solids, while 2b was obtained as a light yellow solid. Note: Any chromatography that did not pass the product through in a short amount of time sign ificantly decreased the yield of the products. The same trend was observed for TLC: additional side product spots appeared if the product was allowed to sit for a few minutes on the TLC plate before chromatography. trans (ICy) 2 Pt(PE2) 2 (2a) Yield, 0.149 g, 70%. Mp: 272 °C (dec). Anal. Calcd for C62H66N4Pt: C, 70.10; H, 6.26; N, 5.27. Found: C, 69.78; H, 6.50; N, 5.20. 1 H NMR (500 MHz, CD 2 Cl 2 4H, Ph H terminal), 7.36 7.30 (m, 6H, Ph H terminal), 7.25 (d, J = 8.5 Hz, 4H, Ph H

PAGE 66

66 internal), 7.03 (d, J = 8.5 Hz, 4H, Ph H internal), 7.00 (s, 4H, CH=CH ), 5.39 (tt, J = 11.5 Hz, 3.7 Hz, 4H, CH), 2.38 (m, 8H, CH 2 ), 1.93 (m, 8H, CH 2 ), 1.78 (m, 4H, CH 2 ), 1.63 1.51 (m, 16H, CH 2 ), 1.28 (qt, J = 12.6 Hz, 3.5 Hz, 4H, CH 2 ). 13C NMR (126 MHz, CD 2 Cl 2 31.9, 131.5, 131.2 130.5, 128.9, 128.6, 124.1, 118.7, 117.0, 113.1, 106.3, 90.6, 89.9, 60.1, 34.3, 26.7, 26.1. APCI TOF MS (found) m/z. 1062.4967 [M] + (calcd) 1062.4950 [M] + . trans (ICy) 2 Pt(BTF) 2 (2b) Yield, 0.207 g, 73%. Anal. Calcd for C 82 H 88 N 6 PtS 2 : C, 6 9.51; H, 6.26; N, 5.93. Found: C, 69.59; H, 6.72; N, 5.83. Mp: 294°C (dec). 1 H NMR (500 MHz, CD 2 Cl 2 (dd, J = 1.6 Hz, 0.4 Hz 2H, Ar H fluorene), 8.03 7.99 (m, 4H, Ar H fluorene, Ar H benzothiazole), 7.94 (ddd, J = 8.0 Hz, 1.2 Hz, 0.6 Hz, 2H, Ar H benzothiazole), 7.70 (dd, J = 7.9 Hz, 0.4 Hz, 2H, Ar H fluorene,), 7.53 (dd, J = 7.8 Hz, 0.4 Hz, 2H, Ar H fluorine), 7.49 (ddd, J = 8.3 Hz, 7.2Hz, 1.2 Hz, 2H, Ar H benzothiazole), 7.39 (ddd, J = 8.3 Hz, 7.2 Hz, 1.2 Hz, 2H, Ar H benzothiazole), 7.12 (m, 2H, Ar H fluorene), 7.10 (dd, J = 7.8 Hz, 1.4 Hz, 2H, Ar H fluorine), 7.03 (s, 4H, CH=CH ), 5.51 (m, 4 H, CH), 2.45 (m, 8H, CH 2 ), 2.11 (dq, J = 14.6 Hz, 7.4 Hz, 4H, CH 2 ethyl), 2.04 1.92 (m, 12H, CH 2 ethyl, CH 2 cyclohexyl), 1.81 (d br , J = 13.5 Hz, 4H, CH 2 ), 1.69 1.55 (m, 16H, CH 2 ), 1.36 1.27 (m, 4H, CH 2 ), 0.30 (t, J = 7.3 Hz, 12H, CH 3 ). 13 C NMR (126 MHz, CD 2 Cl 2 154.9, 151.2, 150.8, 145.6, 137.2, 135.6, 132.1, 130.3, 130.1, 127.6, 126.8, 125.9, 125.5, 123.3, 122.2, 122.0, 120.2, 120.0, 117.1, 110.8, 107.1, 60.1, 56.7, 34.3, 33.3, 26.7, 26.2, 8.8. APCI TOF MS (found) m/z. 1417.6271 [M+H] + (ca lcd) 1417.6253. trans (ICy) 2 Pt(DPAF) 2 (2c) Yield, 0.139 g, 47% Mp: 222°C (dec). Anal. Calcd for C 92 H 100 N 6 Pt: C, 74.41; H, 6.79; N, 5.66. Found: C, 74.37; H, 7.34; N, 5.58. 1 H NMR (500 MHz, CD 2 Cl 2

PAGE 67

67 J = 8.2 Hz, 2H, Ar H fluorene), 7.37 (dd, J = 8 .6 Hz, 1.3 Hz, 2H, Ar H fluorene), 7.23 (m, 8H, Ph H), 7.09 6.94 (m, 24H, Ph H, Ar H fluorine, CH=CH imidazole), 5.51 (m, 4H, CH), 2.44 (m, 8H, CH 2 ), 1.94 (m, 8H, CH 2 ), 1.88 1.75 (m, 12H, CH 2 ethyl, CH 2 cyclohexyl), 1.59 (m, 16H, CH 2 ), 1.28 (m, 4H, CH 2 ), 0.29 (t, J = 7.3 Hz, 12H, CH 3 ), residual hexanes are present in the spectrum. 13 C NMR (126 MHz, CD 2 Cl 2 151.7, 149.9, 148.7, 147.0, 138.0, 137.9, 130.0, 129.6, 128.4, 125.7, 124.4, 124.1, 122.8, 120.3, 119.0, 117.0, 109.0, 106.8, 60.0, 56.3, 34.3, 33.2, 26.7, 26.2, 8.8. MALDI TOF MS (found) m/z. 1484.7692 [M] + (calcd) 1484.7677.

PAGE 68

68 CHAPTER 3 PLATI NUM DIMERS LINKED BY ONE, TWO, OR THREE FLEXIBIL E GROUPS FOR STUDY OF EXCITED STATE DYNAMICS Background For most platinum acetylide materials, it is not possible to achieve the high loading levels necessary to reach an energy saturation level suitable for commercial non linear optical applications . Despite this, very few studies have looked at intramolecular deactivation, i.e. self quenching via triplet triplet annihilation (TTA), as a method of achieving saturation. 80 Thus a series of platinum(II) aryl acetylide dimers co ntaining one, two, or three sp 3 hybridized linking units were synthesized. The saturation levels of these dimers should ideally be half that of a system containing only a monomeric platinum acetylide chromophore, and so a model complex containing only a si ngle platinum center was synthesized for comparison purposes . The structures of the four target complexes are shown in Figure 3 1. Figure 3 1. Structure of the target platinum acetylide dimers and mono platinum model. By using a series of dimers with varying tether length, inter versus intra molecular processes can be compared. The rational e being that intermolecular events are diffusion mediated, with no dependence on the tether. Conversely, intramolecular events will r equire that the chromophoric interaction with each other, whether this be through bond, through space, or co facial systems. Because the dimers are tethered via sp 3 hybridized

PAGE 69

69 atoms, through bon d interactions can presumably be ruled out. Furthermore, the dimers tethered with only one or two linking units are forced to have the chromophoric arms diverge from each other due to geometry restrictions, eliminating the possibility of a co facial overla p and leaving through space interactions as the only possibility for these two compounds. Transient absorption spectroscopy will be the primary probe for observing these deactivation processes. As explained earlier, chromophores tethered by three linking u nits are capable of adopting a co facial overlap. So i n addition to studying TTA behavior, the platinum dimer that is tethered with three linking units may be able to adopt this conformation which would allow for the routine formation and study of a triple t excimeric state. Excimeric behavior has been observed in both solution 39 and solid state 81 photophysics of platinum(II) complexes , and the nature of the excimer is not confined to solely ligand ligand or metal metal interaction s . 44 , 81 Contrary to most of the previous reports of excimeric Pt(II) species, the complexes studied in this report conta in bulky tributylphosphine ancillary ligands. While the report of a Pt(II) aryl acetylide complex which used trioctylphosphine ancillary ligands 44 may allow one to believe that the steric effects of these phosphines are negligible , the nature of the reported excimer is thought to be highly ligand localized with only some hindrance by phosphine ligands, rather than a system featuring interaction of the Pt d orbitals. Additional evidence for the bulk of phosphine ligands inhibiting P t Pt interactions is shown by the Che Group, where a tridentate cyclometallated Pt(II) complex with triphenylphosphine (PPh 3 ) as the final ligand shows a Pt Pt distance of > 4Ã… in the solid state, as the PPh 3 groups do not allow the platinum

PAGE 70

70 atoms to appro ach each other. 81 Interestingly, excimeric behavior is seen from the aforementioned system in the solid state beca use of weak interactions between the cyclometallating ligands. Again, this may lead one to believe that the system reported herein has a chance of intramolecular excimer formation through interactions of the aryl acetylide ligands. However, the bulk of the phosphine ligands would prevent chromophore interaction as readily as it would prevent Pt Pt interactions. Thus an excimer from the systems reported here are unlikely, but would represent a notable result. Synthesis Synthesis of the platinum dimers was st arted by making a series of three diarylacetylides of the form (4 (HCC)Ph) 2 R, wh ere R is CH 2 , CH 2 CH 2 , or CH 2 OCH 2 . The synthetic scheme for these three targets is shown in Figure 3 2 . The full synthesis of bis(4 ethynylphenyl)methane ( C H ), 82 and the individual compounds 1,2 bis(4 bromophenyl)ethane ( CC Br ), 83 1,2 bis(4 ethynylphenyl)ethane ( CC H ), 84 and bis(4 bromobenzyl)ether ( COC Br ) 85 have been previously reported. However, COC Br was made by modification of a different method , 86 and reports and characterization data of CC H in the journal literature are sca rce, so both will be reported here. In general, the diaryl halide is formed, followed by Sonogashira coupling with TMS acetylene, and finally removal of the silyl protecting group to give the diarylacetylide. This was accomplished in various ways depending on the linking unit required. Bis(4 iodophenyl)methane ( C I ) was made from bis(4 aminophenyl)methane via a Sandmeyer reaction. CC Br was synthesized from 4 bromobenzyl bromide by inducing homocoupling of the methylene using 0.5 equivalents of iron powder and catalytic CuCl in hot water. Finally, COC Br was formed by the S N 2 reaction of 4

PAGE 71

71 Figure 3 2 . Synthesis of diarylacetylides tethered by 1 3 flexibilizing units. bromobenzyl bromide with 4 bromobenz yl alcohol. This re action was done in the melt, using 5 equivalents of 4 bromobenzyl alcohol, relative to 4 bromobenzyl bromide, as the solvent and with 1 equivalent of FeSO 4 as a Lewis acid mediator. The yield of this reaction is reported as 72% ; however , this is with respe ct to all organic compounds added to the reaction. This is because once all 4 bromobenzyl bromide has been consumed, excess 4 bromobenzyl alcohol proceeds to condense with itself to form additional ether product. Additionally, a monoarylacetylide , p tolyla cetylene, was synthesized in order to make a mono platinum complex to be used as a model. Once the linking units were in hand, PE2 Pt Cl was synthesized by a modified l iterature procedure . 87 The copper free Hagihara reaction gives predominately monosubstituted platinum acetylide complexes. Thus, 1 equivalent of cis (PBu 3 ) 2 PtCl 2 , and 0.95 equivalents of PE2 were stirred overnight in diethylamine. The reactio n does give a mixture of mono and disubstituted products, which are difficult to separate for this system. The impure solid can be washed with methanol, as this will dissolve primarily the desired monosubstitued product, while leaving most of the disubsti tuted product behind. The filtrate can then be fully purified by chromatography with hexanes/CH 2 Cl 2 ,

PAGE 72

72 making sure to exclude the initial fractions which will be contaminated with the disubstituted impurity. A final Hagihara coupling was then done to form th e dimers and model complex in good yield. The Hagih ara syntheses and model complex formation are shown in Figure 3 3 . N ovel platinum complexes were characterized by elemental analysis or mass spectrometry , as well as 1 H, 13 C, and 31 P NMR. Novel aryl acetyl ide ligands were characterized by high resolution mass spectrometry (HRMS) rather than CHN analysis. The dimeric complexes are named by their linking unit, thus C PtPE2 , CC PtPE2 , and COC PtPE2 correspond to dimers with one, two, and three units, respectiv ely. The model is named tol PtPE2 , and the entire series is R PtPE2 . Figure 3 3 . Synthesis of platinum diarylacetylides and a mono platinum model complex. Photophysical Properties The normalized absorption, fluorescence, and phosphorescence spectra of R PtPE2 are shown in Figure 3 4 . As expected, the normalized absorption spectra are virtually superimposable with maxima at 348 nm , and the extinction coefficients of the dimers are all within experimental error of each othe r. The extinction coefficients of the dimers are approximately double the extinction coefficient of tol PtPE2 , signifying that there is no electronic communication through the linking unit. The fluorescence spectra all show a peak around 379 nm. However, C PtPE2 shows an additional structured

PAGE 73

73 Figure 3 4 . The A) g round state absorption, B) normalized fluorescence , a nd C) normalized phosphorescence spectra of R PtPE2 in THF. Ex citation was at the ground state absorption maxima , and all plot s w ere normalized. Table 3 1 . Summary of p hotophysical d ata for R PtPE2 . Complex Abs max (nm) (M 1 cm 1 ) Fl max (nm) Fl a Fl (ns) Ph max a,b (nm) Ph a,b Ph b (µs) TA max b T1 Tn b,c (µs) C 348 116 000 40 9 0.00 09 < 0. 1 0 d 5 27 0. 30 174 5 96 69.1 CC 348 1 07 000 379 0.0 005 < 0. 1 0 d 527 0.2 6 168 598 65.5 COC 348 1 09 000 3 79 0.0 004 < 0. 1 0 d 527 0.29 18 5 598 62 .4 tol 348 57 000 3 79 0.000 3 < 0. 1 0 d 526 0.34 16 9 595 61.7 a Measured at RT using Ru(bpy) 3 Cl 2 in aerated water fl = 0.0379) as a standard. b Samples degassed by five freeze pump thaw cycles on a high vacuum line. c All decay profiles can be found in A ppendix D. d Lifetime is shorter than the lower threshold of the PicoQuant instrument, 1 00 ps.

PAGE 74

74 peak at longer wavelengths. These spectra are all very noisy, as the fluorescence quantum yields of these complexes are less than 0.1%, and the fluorescence lif etimes are all less than 100 ps. Conversion to the triplet state is very efficient in these complexes, and the intersystem crossing yield is nearly unity as shown by the minimal fluorescence quantum yields. 1 , 6 The normalized phosphorescence spectra of R PtPE2 are also nearly superimposable in both peak maximum and vibro nic structure. The maxima for the dimers appear at 527 nm, while the maximum for the model at is at 526 nm. No observation of a broad emission to the red of the phosphorescence, which would indicate an excimeric state, is observed. Phosphorescence quantum yields are just under 30% for the dimers, and just over 30% for the model. Finally, the phosphorescence lifetimes are all single exponential and fall within the range of 175 ± 10 µ s. All photophysical data for R PtPE2 can be found in Table 3 1. A variety of transient absorption (TA) studies were done on solutions of R PtPE2, with matched absorbances of 0.58 at 355 nm, in order to study the higher triplet excited states, determine energy saturation levels, and further probe potential excimer behavior. Figure 3 5 shows the principal components of the transient absorption for R PtPE2 . Like most platinum acetylide compounds, the TA shows a strong, broad feature across most of the visible region, and like the other R PtPE2 spectra, the line shape is virtually supe rimposable for all of the compounds. A normalized plot of the transient absorption is shown in Appendix D to illustrate this more clearly. Interestingly , the model complex shows approximately 35% less TA response at 180 input energy, even though the chromophore concentration is identical to that of the dimers. The lifetimes recorded

PAGE 75

75 Figure 3 5 . Principal component transient absorption spectra of R PtPE2 , with excitation at 355 nm, 10 Solutions had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles. with time slices recorded in 15 s intervals can be found in Appendix D. However, with time slices of 15 s, any potential excimer formation may not be detected as the lifetime of the PE2 octyl excimer was 1.4 s. 44 The spectra for COC PtPE2 was recorded o n a new solution, with 400 ns time slices, in an attempt to see any excimer formation, an expansion of the maxima is shown in Figure 3 6. While this spectrum appears to decay normally, the experimental noise actually moves the 3 rd and 4 th time slices above the initial measurement. This is also shown by the lack of an isobestic point, indicating that there is only one species present. Additionally, the experiment was rerun with even shorter time slices. Spectra recorded with 200 and 100 ns time slices confir med that this observation was purely noise of the detector. The next transient absorption experiment was a power dependence study to determine saturation levels, and is shown in Figure 3 7. Looking at the full graph, it is apparent that the mono platinum m odel complex has a significantly higher saturation

PAGE 76

76 Figure 3 6. First five time slices of the transient absorption spectra of COC PtPE2 with a camera delay interval of 400 ns. E xcitation was at 355 nm, with a 10 ns pulse width, and 1 olution had an absorptivity value of 0. 56 at 355 nm after f ive freeze pump thaw cycles. Figure 3 7 . Transient absorption power dependence of A) R PtPE2 monitored at 550 nm, with B) an expansion of the low fluence region to show initial slopes. Solutions had a concentration of 5. 5 x10 6 M for the dimers, and 1.1x10 5 M for the model complex, resulting in an absorptivity value of 0.58 at 355 nm after five freeze pump tha w cycles. level than the dimers. It is expected that a mono platinum complex will have a higher saturation level than a dimeric complex because dimeric complexes may have both chromophores excited, which will then undergo decay via intramolecular triplet t riplet annihilation before triplet triplet absorption can take place. What is unexpected, is that

PAGE 77

77 C PtPE2 has an approximately 15% lower saturation level than CC PtPE2 and COC PtPE2 . Also shown in Figure 3 7 is an expansion of the lowest fluenc e region of the power dependence study . The low energy region supports the previous data, shown in Figure 3 5 , in that tol PtPE2 initially has a weaker T A response than the dimers . Finally, the lifetime decay as a function of time and fluence for each complex was plot ted to visually assess if there are any differences between the complexes. These data are shown in Figure 3 8 , and upon initial inspection show similar decays to one another The trend in lifetime decays was deconvolut ed by normalizing the decay of each com pound at the strongest fluence prior to decomposition ( 61 ± 7 mJ/(cm 2 pulse)) and plotting them on the same graph. The normalized data are shown in Figure 3 9 , along with an expansion of the fast timescale data. The plots show CC PtPE2 and COC PtPE2 to have a nearly identical TA decay. This can be rationalized by the lack of any notable intramolecular interaction by the two chromophoric moieties in these two dimers. The fastest to decay is tol PtPE2 , and it has the same exponential decay properties as the af orementioned dimers with no sign of any perturbation. Contrary to the other compounds, the decay of C PtPE2 shows a faster decay component in the first 10 µ s, before the primary, long, decay seen in the other complexes. Further comparison is achieved by pl otting the decays of R PtPE2 at both low and high fluence (1.7 ± 0.3 and 61 ± 7 mJ/(cm 2 pulse), respectively) on a logarithmic scale. These decays are shown in Figure 3 10. The logarithmic scale will help determine if any secondary processes are occurring i n the system, as the traces will deviate from linearity if it is not a single exponential decay. At low fluence, the decays are nearly purely single exponential, meaning that the only process occurring is the

PAGE 78

7 8 Figure 3 8. Transient absorption lifetime decays measured at 550 nm with variable fluence. A) C PtPE2 . B) CC PtPE2 . C) COC PtPE2 . D) tol PtPE2 . Figure 3 9 . A) Normalized transient absorption lifetime decays measured at 550 nm of the strongest fluence measured for R PtPE2 ( 61 ± 7 mJ/(cm 2 pulse)), with B) an expansion of the early timescale region to show initial slopes.

PAGE 79

79 unimolecular decay of the excited triplet state. At high fluence, the plotted decay s deviate from linear ity , indicating contributions from bimolecular processes. Because all of the complexes in the R PtPE2 series display this behavior, it is believed to be diffusion mediated i ntermolecular triplet triplet annihilation. Due to the smaller size of tol PtPE2 , it should have a greater diffusion rate, which leads to its diffusion rate being the fastest of the series. Having assigned the intermolecular TTA process to a slow, diffusio n mediated process, one may then want to assign the tertiary decay component of C PtPE2 (observed in the first 10 µ s) to intramolecular TTA. A structurally similar system was previously synthesized by the Schanze Group featuring a tetrakis(Pt(II) arylacetl ide)methane motif, 88 which showed strong TTA via intramol ecular (through space) pathways, and saturation levels of just over 25% of a mono platinum model nanoseconds, rather than microseconds. Figure 3 10 . Normalized transient absorption lifetime decays measured at 550 nm for R PtPE2 on a logarithmic scale at A) low fluence ( 1.7 ± 0. 3 mJ/(cm 2 pulse)) , and B) high fluence ( 61 ± 7 mJ/(cm 2 pulse)) .

PAGE 80

80 Molecular Dynamics Simulations To p rovide further evidence for the hypothesis that COC PtPE2 does not show excimeric behavior due to steric factors, molecular dynamics (MD) simulations were invoked. The MD simulation places the optimized molecule in a solvent box at a fixed temperature (298 K for these simulations ) and allows the molecule to move freely for a user defined amount of time. Exact details of the geometry optimization and MD procedure are given in the experimental section. A histogram can then be made by recording the distance be tween platinum centers at each step, thus giving a probability function. Histograms for each dimer are given in Figure 3 1 1 , the DFT optimized structures are shown in Figure 3 12, and representative structures from the MD simulations are shown in Appendix D . The gas phase DFT optimized structures match the predictions of the histograms well for C PtPE2 and CC PtPE2 , but the structure of COC PtPE2 gives a Pt Pt bond distance 1 Ã… shorter than what MD simulations would suggest. The deviation between DFT optimi zed gas phase structure and the MD probability is likely due to stabilization of the anomeric effect in THF solution. Upon closer inspection of each histogram, it appears that C PtPE2 and CC PtPE2 maintain the divergent nature of their chromophoric arms, a s they were expected to, while COC PtPE2 seems incapable of overcoming the rotational energy barrier to adopt a co facial orientation. C PtPE2 shows a broad distribution of distances, from 13 to 16 Angstroms, which can be assigned to rotations of the chrom ophore arms and distortion of the angle between the chromophores, centered on the linking methylene. CC PtPE2 gives a much narrower distribution, centered at 18.5 Angstroms. This is due to the chromophores adopting an anti conformation about the ethylene l inking unit and

PAGE 81

81 Figure 3 1 1 . Histograms showing the probability of the internuclear Pt Pt distance for A) C PtPE2 , B) CC PtPE2 , and C) COC PtPE2 . Figure 3 12. DFT optimi zed gas phase structures of A) C PtPE2 , B) CC PtPE2 , and C) COC PtPE2 . deviating only slightly from it. Finally, COC PtPE2 shows a bimodal distribution , as the dimethylene ether tether allows for greater rotational freedom . The major distribution,

PAGE 82

82 centered just under a Pt Pt distance of 18 Angstroms, represents an anti configuration about both carbon oxygen bonds . A very minor distribution is also present, and in the vicinity of 16 Angstroms. This set of conformations is the result of one of the chromophore s rotating so that the bonds of the linking unit ha ve now adopted a syn anti motif. A simplified drawing of th e three possible rotational isomers for the COC linking unit are shown in Figure 3 1 3 . It should be noted here that the COC linking unit may b e affected by the anomeric effect, but the overall contribution is minimal at most based on the photophysical results and dihedral angles centered on the C O bonds. Despite being able to adopt conformations where one chromophore has rotated to the syn posi tion, the MD simulation does not generate any structure where the Pt Pt distance is less than 14 Angstroms. Thus , even an approach to the syn syn conformer is not found to be populated in the ground state , which implies that co facial overlap would be diff icult to achieve in the excited state as well . Figure 3 1 3 . Representative drawings of the anti anti , syn anti , and anti anti rotational isomers of COC PtPE2 . Summary and Conclusions In summary, a series of three platinu m(II) aryl acetylide dimers with one, two, or three linking units were synthesized and their photophysics were fully characterized. Additionally, a mononuclear platinum complex was synthesized and characterized to serve as a model complex. The absorption s pectra of all four complexes were qualitatively identical, which shows that there is no electronic communication through

PAGE 83

83 the sp 3 hybridized linking unit(s). This is further supported by the extinction coefficient of the dimers being double that of the mode l compl ex (within experimental error). Looking at the emission spectra, all four complexes show very weak structure of this emission is broad and featureless for CC PtPE2 , COC PtPE2 , and tol PtPE2 , which is common for the PE2 chromophore . However, C PtPE2 shows an emission with pronounced vibronic structure. A possible explanation for this phenomenon is that the chromophoric arms of C PtPE2 are not able to rotate near ly as easily as the other dimers, leading to a more well defined emission. Moving to the phosphorescence, the spectra of all four complexes are once again virtually superimposable and show strong vibronic structure. This is due to the localization of the t riplet on a single PE2 chromophore. Phosphorescence quantum yields of the dimers are 30% or slightly less, while the model complex shows a slightly stronger emission (34%). The lifetimes of the four complexes all fall within experimental error of each othe r (175 ± 10 µ s) ; which provid es further support for the localization of the triplet. No extra emission bands that could be assigned to a possible excimeric state were observed in aerated, nor freeze pump thaw degassed solution. This is not a surprise, as the PE2 octyl excimer did not show any radiative decay. 44 Following the one photon photophysics, transient absorption was then studied to observe triplet triplet and possible excimer absorption. Initially studied was the broad spectr um TA, measured at multiple time slices. Again, the band shape of all four complexes was indistinguishable from one another, but the intensity of the model complex was drastically weaker at the studied energy. Initially, the camera delay

PAGE 84

84 increment was set to 15 µ s so lifetime data could be obtained, and the triplet triplet lifetimes all fall within the 60 70 µ s regime. The camera delay increment was then set to take spectra every 400 ns, to search for any fast components that would be indicative of excimer formation, but no sign of any secondary species was observed. The steric environment generated by the alkyl phosphines is presumed to prevent any close Pt Pt interactions or co facial overlap that would allow an excimeric state to form in this complex . The previously studied PE2 octyl complex is capable of forming an excimer despite its bulky phosphine ligands because the PE2 ligands are still allowed to overlap significantly, and even interact with the platinum center. Regardless of the previous result, TA power dependence was studied, as these compounds should show good saturation behavior due to their dimeric structure. This behavior is indeed seen, as the dimers saturate at a OD of ~75% that of the model complex when pumped at a fluence of 61 ± 7 mJ/(cm 2 pulse). Further study of the power dependence relied on observing variances of the decay of each complex as fluence was increased. At low fluence, the decay is nearly single exponential, corresponding to the unperturbed spontaneous decay. As the fluence w as increased, triplet triplet annihilation became more pronounced in all complexes. The rate of this deactivation process is nearly identical for all complexes, leading it to be assigned to diffusion mediated intermolecular TTA. However, C PtPE2 also displ ays a fast decay component which also becomes more prominent as fluence increases, which is not found in any of the other complexes. This fast decay is still too slow to be intramolecular TTA, on the basis of previous studies within the Schanze Group showi ng the aforementioned process is completed within tens of nanoseconds .

PAGE 85

85 Molecular dynamics simulations were used to study the molecular geometry in solution conditions. Histograms of the internuclear Pt Pt distance for C PtPE2 and CC PtPE2 show a single dis tribution and minimal deviation from the optimized gas phase geometries. Simulations on COC PtPE2 give a bimodal distribution. The major distribution is centered slightly under the maximum achievable Pt Pt distance, representing a conformation where the tw o chromophoric arms are slightly deviated from the anti anti conformer. The minor distribution only brings the platinum centers a couple Angstroms closer and is assigned a half folded syn anti rotational isomer. No evidence of the syn syn isomer, and there fore co facial overlap, was observed at any time during the simulation. Although no excimeric species was observed, t his work demonstrates the ability of platinum dimers to serve as potential agents for power limiting applications. Experimental General Rem arks All reactions were carried out under argon atmosphere. All starting chemicals used for the synthesis of the trans platinum acetylide complexes were purchased from commercial suppliers, reagent grade, and used without additional purification. K 2 PtCl 4 w as purchased from Strem Chemicals. Solvents were of reagent grade unless otherwise noted. PE2 Pt Cl was synthesized by a modified literature method . 87 C hro matography was done on a CombiFlash Rf 1 50 Medium Pressure Liquid Chromotography (MPLC) system . 1 H (500 MHz) and 13 C (125.7 MHz) NMR spectra were recorded on a Varian Inova spectrometer. 31 P ( 121 MHz) NMR spectra were recorded on a Varian Mercury spectrome ter. The chemical shifts were reported in ppm relative to tetramethylsilane (TMS) or residual solvent peaks . Elemental analysis was

PAGE 86

86 performed by Complete Analysis Laboratories, Inc . Mass spectrometry was performed in house by Mass Spec t rometry Services at the University of Florida. Absorption and Emission Spectroscopy Steady state absorption spectra were recorded on a Varian Cary 50 spectrophotometer. Corrected steady state emission measurements were performed on a Photon Technology International spectropho tometer (QuantaMaster). Absorption and fluorescence samples were carried out in aerated solution, while phosphorescence samples were deaerated with five freeze pump thaw cycles. Optically dilute samples with O.D. < 0.1 at the excitation wavelength were use d. Fluorescence and phosphorescence quantum yields were determined by relative actinometry, with Ru(bpy) 3 Cl 2 Fl = 0.0379 in air saturated water) . 70 Phosphoresc ence lifetimes were obtained with a multichannel scaler/photon counter system with a NanoQuant FluoTime 100 Compact Phosphorescence Lifetime Spectrophotometer. A UV pulsed diode laser provided the excitation at 375 nm (power < 10 mW). The laser was pulsed by a PDL800 B, which is a pulsed diode laser driver. Optically dilute solutions were freeze pump thawed five times. All decays were obtained using single exponential fitting parameters (FluoFit software). Nanosecond Transient Absorption (TA) Spectroscopy M easurements were performed on an in house apparatus that is described in detail elsewhere . 77 355 nm, 10 ns fwhm, 180 µJ pulse 1 ) was used as the excitation source. Probe light was produced by a xenon flash lamp and the transient absorption signal was detected with a gated intensi fied CCD mounted on a 0.18 M spectrograph (Princeton PiMax/Acton Pro 180). Solutions had a matching optical density of 0. 58 after five freeze -

PAGE 87

87 pump thaw cycles. An initial CCD image capture delay of 100 ns following the laser pulse was used to ensure full c onversion to the triplet state before observation. An average of 100 images were acquired and the laser energy was 180 µJ/pulse, which was established as being sufficiently low to minimize triplet triplet power dependence. Transient Absorption Power Depend ence The third harmonic of a Surelite I 10 Nd:YAG laser ) was used as the excitation source . Probe light was produced by a Xenon arc lamp and the transient absorption signal was detected at a single wavelength (550 nm) using a Triax 180 monochromator and a Si amplified photodetector from Thorlabs (PDA8A). Computational Details Structure optimization DFT calculations were performed in the Gaussian 09, revision C.01 , 73 suite of programs at the B3LYP level with the 3 21G basis set for C, H, O, the 6 31G(d) basis set for P , and the SDD basis set for Pt. Phosphine butyl groups were replaced by methyl groups to improve computational efficiency. The ground state structures were optimized in the gas phase from idealized starting configurations without symmetry constraints. The optimized structures were confirmed to be minima by the lack of imaginary frequencies . Molecular dynamics calculations Fully at omistic molecular dynamics simulations were performed using the Forcite module in the Materials Studio 6.0 software package. The dimers were packed inside of a cell that was approximately 10 Å larger than each dimension of the gas phase dimer, while the re maining volume was packed with 200 tetrahydrofuran (THF) solvent molecules to give a total density of 0.889 g/cm 3 . Atomic charges (Natural Population

PAGE 88

88 Analysis) for each dimer and THF molecule were obtained from the Gaussian DFT calculations. The packed cel l was annealed for 5 temperature cycles ranging from 300 to 700 K for a total time of 250 ps using the Universal Force Field (UFF). The annealed simulation cells were used for molecular dynamics calculations with the canonical ensemble, NVT, at a temperatu re of 298 K controlled by the Nose thermostat with a Q ratio of 0.01. Dynamics were calculated for a total of 1 ns for each dimer and snapshots were collected every 5 ps. Synthesis of bis( 4 bromobenzyl ) ether ( COC Br ) 4 B romobenzyl alcohol (4.7 g, 25 mmol, 13.5 eq.) , 4 bromobenzylbromide (0.46 g, 1.8 mmol, 1 eq.) , and FeSO 4 · 7H 2 O (0.52 g, 1.8 mmol, 1 eq . ) were added to a 50 mL round bottom flask. The flask was equipped with a Vigreaux column, and the mixture was heated to 130 ° C. The melt was allowed to react overnight before being cooled to room temperature. The resulting off white solid was dissolved in CH 2 Cl 2 , insolubles were removed by filtration, and the solvent was evaporated . The solid was suspended in a minimum of methanol, and the slurry was filtered to afford pure COC Br . Yield 72%. Mp: 81 °C . 1 H NMR (500 MHz, CD 2 Cl 2 7 (d, J = 8. 0 Hz, 4H), 7. 21 (d, J = 8. 0 Hz, 4H), 4. 48 (s, 4H) . 13 C NMR (126 MHz, CD 2 Cl 2 7.2 , 13 1.8 , 12 9.6 , 12 1.8 , 71 .7. General Procedure for the Sonogashira Coupling Reaction T he dibromo linking unit ( R Br , 0.15 g , 1 eq. ) was added to 15 mL of Et 2 NH in a 25 mL seal tube. The solution was degassed with argon for 15 minutes before Pd(PPh 3 ) 4 ( 20 mol% , 0.10 eq) and CuI (20 mol%, 0.10 eq . ) was added. The solution was bubbled for an additional 10 minutes before trimethylsilylacetylene (0.5 mL , 3.5 mmol, 5 eq ) was added, and the tube was sealed. The solution was heated to reflux and allowed to react overnight. The solution was poured into water and extracted with

PAGE 89

89 CH 2 Cl 2 . The organic layer was separated, filtered through Na 2 SO 4 , the filtrate was reduc ed to a minimum, and it was loaded onto silica gel. Purification of both CC TMS and COC TMS , via MPLC using hexanes and hexanes/ethyl acetate, respectively, gave pure product. Additional product with ~10% impurity was also isolated via MPLC, and can be inc luded in further reactions. 1,2 bis(4 (trimethylsilylethynyl)phenyl)ethane ( CC TMS ) Mp: 141 °C . 1 H NMR (500 MHz, CDCl 3 ): 35 (d, J = 8.0 Hz, 4H ), 7. 02 ( d , J = 8.0 Hz, 4 H ) , 2.87 ( s, 4 H), 0.24 ( s , 18 H) . 13 C NMR (126 MHz, CDCl 3 1 42.1 , 13 2.2 , 128. 7 , 1 20.9 , 10 5.4 , 9 3.8 , 37.7 . Mass Spec. DART MS (found) m/z 392.2225 [M+N H 4 ] + . (calcd) 392.2224 [M+N H 4 ] + . B is(4 (trimethylsilyl ethynyl)benzyl) ether ( COC TMS ) Mp: 120 °C. 1 H NMR (500 MHz, CD 2 Cl 2 ): 43 (d, J = 8. 5 Hz, 4H ), 7. 02 ( d , J = 8.5 Hz, 4 H ) , 4.54 ( s, 4 H), 0.24 ( s , 18 H) . 13 C NMR (126 MHz, CD 2 Cl 2 1 39.6 , 13 2. 4 , 128. 1 , 1 2 2 .9 , 10 5. 3 , 9 4.7 , 72.4 . Mass Spec. DART MS (found) m /z 408.2178 [M+ NH 4 ] + . (calcd) 408.2173 [M+ NH 4 ] + . General Procedure for the TMS Deprotection Reaction The TMS protected diarylacetylides ( R TMS ) were dissolved in 50 mL of 1:1 methanol:diethyl ether, and the solution was degassed with argon for 25 minutes. K 2 CO 3 (1 g) was then added, and the mixture was stirred at in the dark and at room temperature for 5 hours. After this time, the mixture was poured into water and extracted with CH 2 Cl 2 . The organic layer was separated, filtered over Na 2 SO 4 , the filtrate wa s reduced to a minimum, and it was loaded onto silica gel. Purification of both CC H and COC H , via MPLC using hexanes and hexanes/ethyl acetate, respectively, afforded the pure products as white solids.

PAGE 90

90 1,2 bis(4 ethynylphenyl)ethane ( CC H ) Yield across t wo steps: 55%. 1 H NMR (500 MHz, CD 2 Cl 2 ): 38 (d, J = 8.0 Hz, 4H ), 7. 11 ( d , J = 8.0 Hz, 4 H ) , 3.09 (s, 2H), 2.92 ( s, 4 H ). 13 C NMR (126 MHz, CD 2 Cl 2 1 4 3.1 , 13 2. 6 , 12 9.1 , 1 20. 2 , 84.1 , 77.1 , 37. 9. Bis(4 ethynylbenzyl) ether ( COC H ) Yield across two steps: 62%. Mp: 68 °C. 1 H NMR (500 MHz, CD 2 Cl 2 ): 48 (d, J = 8.0 Hz, 4H ), 7. 33 ( d , J = 8.0 Hz, 4 H ) , 4.56 (s, 4H), 3.13 ( s, 2 H ). 13 C NMR (126 MHz, CD 2 Cl 2 139.9 , 13 2.6 , 12 8.1 , 1 21.8 , 83.9 , 77.6 , 72.4. Mass Spec. DART MS (found) m/z 264.1390 [M+ NH 4 ] + . (calcd) 264.1383 [M+ NH 4 ] + . Synthesis of mo nosubs tituted platinum precursor trans ( PBu 3 ) 2 Pt (PE2 )Cl ( PE2 Pt Cl ) To a flask charged with 30 mL of Et 2 N H, cis (PBu 3 ) 2 PtCl 2 (0.35 g, 0.52 mmol, 1 eq.) was added, and the solution was bubbled with argon for 25 minutes. PE2 (0.10 g, 0.49 mmol, 0.95 eq.) was added , and reaction mixture was then stirred overnight, at room temperature, in the dark. The final mixture was poured into water and extracted with dichloromethane. The organic layer was separated, filtered over Na 2 SO 4 , and the solvent was evaporated. The resu lting solid was washed with copious MeOH, the filtrate was collected, reduced to a minimum, and loaded onto silica gel. Purification via MPLC using hexanes/dichloromethane afforded pure PE2 Pt Cl as a clear oil. Yield 38% 1 H NMR matched that of the literat ure. 89 Synthesis of model complex trans ( PBu 3 ) 2 Pt (PE2)(CC p tolyl) ( tol PtPE2 ) To a flask charged with 30 mL of Et 2 N H , PE2 Pt Cl (0.135 g, 0.16 mmol, 1 eq.) was added . The solution was then bubbled with argon for 15 m inutes before addition CuI (0.005 g) and 0.026 g 4 (tolyl)acetylene (0.026 g, 0.22 mmol, 1.35 eq.) was added .

PAGE 91

91 The reaction mixture was then stirred overnight at roo m temperature , in the dark. The final mixture was poured into water and extracted with dichloromethane . The organic layer was separated , filtered over Na 2 SO 4 , t he filtrate was reduced to a minimum, and it was loaded on to silica gel . Purification via MPLC using hexanes/ethyl acetate afforded pure tol PtPE2 as an off white solid . Yield 80 % Mp: 100 °C. 1 H NMR (500 MHz, CDCl 3 ): 7. 54 ( d, J = 7.5 Hz, 2 H, Ph H , PE2 terminal ), 7. 42 7. 31 (m, 5 H, Ph H , PE2 terminal & tolyl), 7.2 5 (d, J = 8. 0 Hz, 2 H, Ph H , tolyl), 7.1 9 (d, J = 8.0 Hz, 2 H, Ph H , PE2 internal), 7.0 4 (d, J = 8.0 Hz, 2 H, Ph H , PE2 internal) , 2.29 (s, 3H, tolyl), 2.1 3 (m, 12 H, CH 2 ), 1.60 (m, 12 H, CH 2 ), 1.44 (sextet, J = 7.5 Hz, 12 H, CH 2 ), 0.92 (t, J = 7.5 Hz, 18H, CH 3 ) . 13 C NMR (126 MHz, CDCl 3 13 4. 7 , 131. 7 , 131. 4 , 13 0 . 89 , 130. 83 , 129. 5 , 128. 8 , 128. 5 , 128. 2 , 126. 2 , 123.8 , 119.3, 109.3, 109.1, 90.3, 89.9 , 26. 6 , 24. 6 (t , J Pt C = 6.8 Hz ), 24.1 (t , J Pt C = 17. 1 Hz ), 21.5, 14.1. 31 P NMR (12 1 MHz, CDCl 3 ) 3.07 ( J Pt P = 2350 Hz ). ESI TOF MS (found) m/z 916.4654 [M+H] + . (calcd) 916.4683 [M+H] + . General Procedure for the Hagihara Coupling Reaction for Platinum Dimers. PE2 Pt Cl (0. 100 g, 0.11 mmol, 1 eq.) was dissolved in 20 mL of Et 2 NH, and the solution was degassed with argon for 15 minutes. CuI (0.005 g) and 0.95 equivalents of the deprotected diarylacetylene w ere added. T he solution was allowed to react overnight at room tempera ture, in the dark. The resultin g mixture was poured into water and extracted with CH 2 Cl 2 . The organic layer was separated, filtered over Na 2 SO 4 , t he filtrate was reduced to a minimum, and it was loaded on to silica gel . Purification of R PtPE2 via MPLC usi ng hexanes/ethyl acetate afforded the pure products as a semi transparent oil. Microcrystalline product can be obtained by dissolving the compound in acetone and adding to methanol to give a n off white precipitate.

PAGE 92

92 trans [ ( PE2 )Pt( PBu 3 ) 2 (CC p Ph)] 2 CH 2 ( C Pt PE2 ) Yield 68 % Mp: 89 °C . Anal. Calcd for C 97 H 136 P 4 Pt 2 : C, 64.15 ; H, 7.55 . Found: C, 6 4.01; H, 7.32 . 1 H NMR (500 MHz, CDCl 3 J = 8.0 Hz, 4H, Ph H , PE2 terminal), 7.38 7.28 (m, 10H, Ph H , PE2 terminal & tether ), 7.22 (d, J = 8.5 Hz, 4H, Ph H t eth er ), 7.18 (d, J = 8.0 Hz, 4H, Ph H , PE2 internal ), 7.00 (d, J = 8.0 Hz, 4H, Ph H , PE2 internal), 3.86 ( s , 2 H, CH 2 , tether), 2.12 (m, 24 H, CH 2 ), 1.60 (m, 24 H, CH 2 ), 1.44 (sextet, J = 7.5 Hz, 24 H, CH 2 ), 0.92 (t, J = 7.5 Hz, 36 H, CH 3 ) . 13 C NMR (126 MHz, CDCl 3 138.0, 131.7, 131.4, 131.0, 130.9, 129.5, 128.7, 128.5, 128.2, 126.9, 123.8, 119.3, 109.3, 109.1, 90.3, 89.9, 41.8, 26.6, 24.6 (t, J Pt C = 6.8 Hz), 24.1 (t, J Pt C = 17.2 Hz), 14.1. 31 P NMR (12 1 MHz, CDCl 3 ) 3.05 ( J Pt P = 2350 Hz ) . trans [ ( PE2 )Pt( PBu 3 ) 2 (CC p Ph)] 2 1,2 ethane ( CC PtPE2 ) Yield 84 % Mp: 148 ° C. 1 H NMR (500 MHz, CDCl 3 2 (d, J = 8.0 Hz, 4H, Ph H, PE2 terminal), 7.38 7. 34 (m, 10H, Ph H, PE2 terminal & tether), 7.22 (d, J = 8.5 Hz, 4H, Ph H tether), 7.1 6 (d, J = 8.0 Hz, 4H, Ph H, PE2 inter nal), 7.0 2 (d, J = 8.0 Hz, 4H, Ph H, PE2 internal), 2.84 (s, 4 H, CH 2 , tether), 2.1 3 (m, 24 H, CH 2 ), 1.6 2 (m, 24 H, CH 2 ), 1.4 6 (sextet, J = 7.5 Hz, 24 H, CH 2 ), 0.9 4 (t, J = 7.5 Hz, 36 H, CH 3 ) . 13 C NMR (126 MHz, CDCl 3 13 9.2 , 131. 9 , 131. 7 , 131. 2 , 13 1.1 , 1 30.1 , 12 9.0 , 128. 7 , 128. 6 , 127.2, 12 4.1 , 119. 6 , 109.3 4 , 109. 28 , 90. 4 , 90.2 , 38.2 , 26. 9 , 2 5.0 (t, J Pt C = 6.8 Hz), 24. 5 (t, J Pt C = 17. 3 Hz), 14. 2 . 31 P NMR (12 1 MHz, CDCl 3 ) 3.27 ( J Pt P = 2340 Hz ). trans [ ( PE2 ) Pt ( PBu 3 ) 2 (CC p Benzyl)] 2 O ( COC PtPE2 ) Yield 60 % Mp: 83 °C . C 9 8 H 13 8 OP 4 Pt 2 : C, 63.76 ; H, 7.53 . Found: C, 6 3.66; H, 7.21 . 1 H NMR (500 MHz, CDCl 3 2 ( m , 4H, Ph H), 7.3 9 7. 3 2 (m, 10H ), 7.26 7. 18 ( m , 12H ), 4.47 (s, 4 H, CH 2 , tether), 2.1 3 (m, 24 H, CH 2 ), 1.6 2 (m, 24 H, CH 2 ), 1.4 6 (sextet, J = 7.5 Hz, 24 H, CH 2 ), 0.9 3 (t, J = 7.5 Hz, 36 H, CH 3 ) . 13 C NMR (126 MHz, CDCl 3

PAGE 93

93 13 5 . 7 , 131. 9 , 131. 7 , 131. 2 , 13 1.1 , 1 30. 0 , 12 9.0 , 128. 7 , 128. 1 , 12 4.1 , 119. 6 , 113.3 109. 4 , 109. 3 , 90. 4 , 90.2 , 72.5 , 26. 9 , 2 5.0 (t, J Pt C = 6.8 Hz), 24. 5 (t, J Pt C = 17. 3 Hz), 14. 2 . 31 P NMR (12 1 MHz, CDCl 3 ) 3. 32 ( J Pt P = 2340 Hz ).

PAGE 94

94 CHAPTER 4 PHOTOPHYSICAL PROPERTIES OF COMPLEXES FEAT URING A MIXED ARYL PLATINUM ACETYLIDE CHROMOPHORE MOTIF Background As the class of disubstituted platinum(II) chromophores featuring one aryl ligand ( RAr [Pt]) and one acetylide ligand ( Ar CC [Pt]) is highly unknown, the goal of the work presented in this c hapter is to synthesize and fully characterize the photophysical properties of a series of four mixed complexes using the aryl ligands 2 (9,9 Diethyl 9 H fluoren e )benzo[ d ]thiazole ( BTF ) and 9,9 Diethyl N , N diphenyl 9 H fluoren 2 amine ( DPAF ) as well as the a cetylide congeners of the aryl ligands: 2 (9,9 Diethyl 9H fluoren 7 yl)benzo[d]thiazole (CCBTF) and 9,9 Diethyl 7 ethynyl N,N diphenyl 9Hfl uoren 2 amine (CCDPAF). Structures of the four target complexes are shown in Figure 4 1. Figure 4 1. The four combinations of platinum aryl/acetylide complexes featuring DPAF and BTF chromophores. These chromophores were chosen as they have been well studied in platinum acetylide complexes, 2 and the because they sho w competing excited state properties, donor. system of each chromophore, the order of LUMO energies for these ligands, working from highest to lowest, is D PAF > CCDPAF > BTF > CCBTF.

PAGE 95

95 Overall, these complexes are not expected to be drastically different from those of the purely acetylide congeners, as they are structurally very similar. Previous studies on platinum acetylide complexes have shown that the trip let exciton is localized to the lowest energy chromophore. 45 However, due to the weak coupling of the ligands through the platinum center, interesting propert ies such as dual emission may arise when complexes containing both the BTF and DPAF chromophores are synthesized. Furthermore, emission spectra were seen in purely platinum aryl systems in which the quantum yields of the fluorescence and phosphorescence we re nearly equivalent. 55 This would lead to a white light emitting species and a new class of white light emitters. Synthesis Synthesis of monosubstituted platinum (II) aryl (Pt aryl) complexes via a Cu(I) assisted Stille type reaction was recently developed by a previous member of the Schanze group. 55 Th e s e product s w ere isolated, and addition of the acetylide chromophore was achieved via the classic Cu(I) assisted Hagihara reaction to give the mixed aryl/acetylide ligated platinum (II) complex. A discussion of nomenclature is warranted before further disc ussion of the synthetic details. Nomenclature of these compounds will be by chromophore and connectivity. The only nomenclature feature that can be held constant for these complexes is the position of the acetylide feature , and the acetylide chromophore wi ll be distinguished from the . Thus the complexes will always be named in the form: aryl Pt acetylide . For example, the complex using the aryl ligand 2 (9,9 diethyl 9 H fluorene)benzo [d] thiazole (BTF) and the acetylide 9,9 Diethyl 7 ethynyl N,N diphenyl 9Hfluoren 2 amine (CCDPAF) would be named BTF Pt CCDPAF . The other

PAGE 96

96 two ligands used are the acetylide congener of BTF: 2 (9,9 diethyl 9H fluore 2 yl)benzo[d]thiazole (CCBTF), and the aryl congener of CCDPAF: 9,9 Diethyl N,N diphenyl 9Hfluoren 2 amine (DPAF). It should be noted that the aryl groups are bonded to the metal in the 7 postion of the fluorene moiety. Additionally, the synthesized mono substituted platinum aryl and DFT modeled platinum acetylide complexes will always be named with the chromophore first, i.e: BTF Pt Cl . Finally, when referring to the presence of a chromophoric moiety that is not dependent on how it is bonded to the platinum center, it will be referred to as chromophore (i.e. if a complex contains eithe r As stated earlier, a novel preparation of monosubstituted platinum(II) aryl complexes has been developed, and this methodology will serve as the first step in the synthesis present ed here , and is shown in Figure 4 1 . The precursor organic ligands BTF Br , CCBTF , DPAF Br , and CCDPAF were prepared according to literature procedures. 2 Starting with the aryl bromide ( BTF Br or DPAF Br ) in argon purged THF solution at 78°C , n butyl lithium was added dropwise to generate the lithium salt of the organic anion . After an hour, tributyltin chloride was added to quench the anion and form an aryl stannane. The solution was stirred at 78°C for two hours before being allowed t o warm to room temperature. The solution was then washed with water, extracted with dichloromethane, and had the solvent removed in vacuo to give a crude oil. Beause aryl stannanes are not especially amenable to chromatography, the crude was used as is. NM R was employed to help approximate the conversion to the aryl stannane , typically 50% , as it would be a waste of platinum to assume quan titative yield. The crude oil was redissolved in 15 mL of dimethylformamide and degassed with argon

PAGE 97

97 Figure 4 1. Synthesis of monosubstituted platinum(II) aryl precursors. for 15 minutes before 10 mol% of CuI and 1 equivalent of cis (PBu 3 ) 2 PtCl 2 (assuming 50% conversion to the aryl stannane) were added, and the solution was allowed to rea ct overnight at 60°C. The resulting solution was washed with water, extracted with dichloromethane, separated and filtered through Na 2 SO 4 , before being purified by medium pressure liquid chromatography (MPLC). While the previously reported DPAF Pt Cl purif ied nicely, and was isolated in 35% yield, the isolation of BTF Pt Cl proved to be problematic. This compound requires nearly pure dichloromethane to elute , and even after multiple columns retains an impurity of ~5%. Furthermore, Pt aryl complexes of this nature tend to be either oils or waxy solids, so recrystallization cannot be employed. Although not suitable for photophysics, the BTF Pt Cl was pure enough to be used in further reactions. The two Pt aryl complexes were then subject to standard Hagihara r eaction conditions with either CCBTF or CCDPAF to generate four platinum complexes with a mixed aryl/acetylide chromophore motif. The synthesis and yields of these re actions are shown in Figure 4 2, with the structures previously shown in Figure 4 1 .

PAGE 98

98 Figure 4 2. Hagihara reaction to generate the four mixed aryl/acetylide Pt(II) complexes. I solation of a ll four mixed complexes proceeded smoothly by washing with water, extracting with dichloromethane, separation and filteri ng through Na 2 SO 4 , followed by purification via MPLC using a hexanes:CH 2 Cl 2 gradient to elute a mostly pure (>9 8 %) product. The solution can then have the solvent removed in vacuo to give an oil, which is easily recrystallized by dissol ution in a minimum o f warm acetone and precipitating with methanol. Isolated yields are around 75%; similar to that of the typical Hagihara reaction to form platinum(II) acetylide complexes . The exception to this is BTF Pt CCDPAF , which does not recrystallize as readily and i s isolated in lower yield. Photophysical Properties Absorption The UV Vis spectra of all four mixed ligand complexes show a strong UV absorption which, for the three complexes containing the BTF chromophore, trails into the visible region. The absorption s pectra of the mixed complexes are shown in Figure 4 4 . The individual chromophores seem to be acting nearly completely independent of each other, suggesting minimal to no coupling through the platinum center. The primary evidence for this stems from the ab sorption onset of BTF Pt CCBTF and DPAF Pt -

PAGE 99

99 CCBTF . As the CCBTF ligand has the lowest energy excited state of all four ligands, it is responsible for the onset wavelength in these two complexes. Regardless of if CCBTF has BTF or DPAF bonded across the plati num atom from it, no change is seen in the absorption onset. Absorption, as well as all other tabular photophysical characterization data is given in Table 4 1. Time dependent density functional theory computations were used to clarify the absorption featu res, and the results will be discussed in detail in a later section . Figure 4 4 . Ground state absorption spectra of A) the four mixed ligand platinum complexes containing the BTF and DPAF chromophores , with spectra shown in pairs for B) BTF Pt CCBTF an d DPAF Pt CCDPAF , C) BTF Pt CCBTF and DPAF Pt CCBTF , and D) BTF Pt CCBTF and BTF Pt CCDPAF to emphasize trends.

PAGE 100

100 Table 4 1 . Summary of p hotophysical d ata for mixed aryl/acetylide ligated Pt(II) complexes . Complex Abs max (nm) (M 1 cm 1 ) Fl max (nm) Fl a Fl (ns) Ph max a ,b (nm) Ph a,b Ph a ,b (µs) TA max a ,b T1 Tn b ,c (µs) BTF Pt CCBTF 391 117000 438 0.031 < 0.1 0 ns d 569 0.16 900 658 50.5 DPAF Pt CCDPAF 368 90000 392 0.022 < 0.1 0 ns d 531 0.10 170 600 53.6 BTF Pt CCDPAF 380 110000 432 0.013 < 0.1 0 ns d 536 0 .099 37 0 586 48.9 DPAF Pt CCBTF 378 73000 43 9 0.011 < 0.1 0 ns d 568 0.14 650 660 61.2 a Measured at RT using Ru(bpy) 3 Cl 2 in aerated water fl = 0.0379) as a standard. b Samples degassed by five freeze pump thaw cycles on a high vacuum line. c All decay profiles can be found in Appendix D. d Lifetime is shorter than the lower threshold of the PicoQuant instrument, 1 00 ps.

PAGE 101

101 Fluorescence All of the mixed ligand complexes are weakly fluorescent, with quantum yields between 1 and 3%. Thus the intersystem cr ossing yield is still very high in these mixed complexes, as it is for both pure platinum(II) aryl complexes and platinum(II) acetylide complexes. A high intersystem crossing yield also implies efficient intersystem crossing, for which evidence is provided in that the fluorescence lifetimes are all <100 ps. The fluorescence spectra of all four mixed ligand complexes, when excited at their respective absorption maxima, are shown in Figure 4 5 As DPAF Pt CCDPAF contains donating ligands, its effective conjugation length is shorter, and it s fluorescence is blue shifted relative to the rest of the complexes. Additionally, this is the only structure to show vibronic structure. This is most likely due to the BTF chromop hore having free rotation of the C C b ond connecting the benzothiazole moiety to the fluorene. Conversely, BTF Pt CCBTF and DPAF Pt CCBTF , both of which utilize the lowest Figure 4 5 . Fluorescence spectra of the four mixed ligand platinum complexes containing the BTF and DPAF chromophores.

PAGE 102

102 energy ligand in the series (CCBTF), show nearly identical emission and display the most red shifted fluorescence in the series. Finally, BTF Pt CCDPAF shows an emission that is slightly blue shifted relative to the lowest energy fluorescence, as these two ligands are both of intermediate energy relative to the other two. After a cursory look at the fluorescence, these complexes seem to be behaving as pure platinum aryl or platinum acetylide complexes would. However, excitation at higher energies than that of the absorption maxima reveals a new feature in the emission of DPAF Pt CCBTF . T he fluorescence spectra using an excitation wavelength of 330 nm is shown in a normalized plot alongside the fluorescence from excitation at the gr ound state absorption maximum of 378 nm in Figure 4 6 . The spectrum obtained f rom 330 nm excitation shows new vibronic bands and shifts the emission maximum from 432 nm to 394 nm. This appears to be a direct result of exciting at the absorption maximum of the DPAF platinum chromophore. 55 The observation of emission from both Figure 4 6 . Fluorescence spectra of DPAF Pt CCBTF with excitation at 378 and 330 nm .

PAGE 103

103 ligands implies minimal coupling of the ligands through the platinum center and inefficient in ternal conversion. Interestingly, none of the other complexes display this behavior. While more data is certainly needed to draw any solid conclusions about this effect, the data presented here leads one to believe that a large gap in the energy of the orbital of each chromophore is required to see dual emission from the singlet manifold. Phosphorescence Moving on to the triplet manifold, s olutions of the mixed ligand complexes were subject to at least five freeze pump thaw cycles on a vacuum line to obs erved the phosphorescence properties. All complexes show pronounced vibronic structure and fair quantum yields ranging from 10 to 16%. Figure 4 7 shows the normalized phosphorescence spectra of the mixed ligand complexes. Starting with an analogous observa tion to the fluorescence spectrum, the complexes BTF Pt CCBTF and DPAF Pt CCBTF have nearly identical phosphorescence spectrum. This is again due to the confined to the lowest energy ligand. 45 DPAF Pt CCDPAF also follows this trend, as its phosphorescence maximum is almost 40 nm blue shifted from that of the CCBTF containing complexes (531 nm vs 569 nm, respectively). The last complex, BTF Pt CCDPAF , shares its phosphorescence onset and vibronic energy levels with that of the aforementioned complex, but the individual peaks are significantly broadened. This energy than that of the acetylide ligand (CCDPAF). Given that the onset wavelength and v ibronic peaks are at the same energies as that of DPAF Pt CCDPAF , it is fair to tentatively assign the major features of this emission as being primarily localized on the

PAGE 104

104 Figure 4 7. Emission spectra after at least five freeze pump thaw cycles of A) th e four mixed ligand platinum complexes containing the BTF and DPAF chromophores, with spectra shown in pairs for B) BTF Pt CCBTF and DPAF Pt CCBTF , C) BTF Pt CCBTF and DPAF Pt CCDPAF , and D) DPAF Pt CCDPAF and BTF Pt CCDPAF to emphasize trends. CCDPAF liga nd. Broadness would then be caused by minor contributions from the BTF ligand. However, further resolution is required to make definitive assignments to the nature of this emission. A closer examination of the phosphorescence quantum yields and lifetimes r eveals a couple of patterns. First, with respect to the quantum yield, the two mixed ligand complexes containing the CCDPAF ligand have lower radiative emission (10%) than the either DPAF Pt CCBTF (14%) or BTF Pt CCBTF (16%). An explanation for

PAGE 105

105 this trend may lie with the known monosubstituted platinum complexes containing the DPAF and CCDPAF ligands. 3 , 55 Both of t hese compounds show phosphorescence quantum yields on the order of 1%. However, if trying to draw comparisons to the known platinum acetylide complexes which utilize the CCDPAF and CCBTF chromophores (using the nomenclature of this chapter, DPAFCC Pt CCDPA F and BTFCC Pt CCBTF, respectively), this trend is reversed, as the DPAFCC Pt CCDPAF has a higher phosphorescence quantum yield than BTFCC Pt CCBTF. 64 Th e second observed trend in the triplet state involves the phosphorescence lifetime correlating inversely to the number of DPAF chromophores in the complex and whether the DPAF is bonded as an aryl or acetylide unit . Thus the lifetimes increase for each com plex in the order of DPAF Pt CCDPAF (170 µ s) < BTF Pt CCDPAF (370 µ s) < DPAF Pt CCBTF (650 µ s) < BTF Pt CCBTF (900 µ s). The lifetimes of the known DPAF Pt Cl and DPAFCC Pt Cl were also measured to help deconvolut e this series , as they were not measured in previous reports. All lifetime decays can be found in Appendix E. Both DPAF Pt Cl and DPAFCC Pt Cl showed biexponential decay µ s and a slow component of 68 and 61 µ s , respectively . Additionally, the contribution of th component makes up only 4% of the decay of DPAF Pt Cl (average lifetime: 66 µ s), but is 75% of the decay of DPAFCC Pt Cl (average lifetime: 21 µ s). These data strongly suggest that the DPAF chromophore largely contributes to the deactivation of th e mixed ligand triplet state. Transient Absorption To further probe the nature of the triplet manifold , transient absorption was run on solutions with a matched O.D. of 0.58. As seen in previous chapters, all complexes show ground state bleaching in the re gion of their respective absorptions, and a strong ,

PAGE 106

106 broad triplet triplet absorption (TTA) across most of the visible region. These spectra are plotted in Figure 4 8. Staying true to the notion that the triplet exciton is localized on the lowest energy lig and, the spectra of DPAF Pt CCDPAF , DPAF Pt CCBTF , and BTF Pt CCBTF show absorption features that are qualitatively identical to those found in the purely platinum acetylide compounds DPAFCC Pt CCDPAF and BTFCC Pt CCBTF. 64 The TTA of DPAF Pt CCDPAF shows a single feature with a Figure 4 8. Principal component transient absorption spe ctra with excitation at 355 nm, ation energy. Solutions had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles. A) All four mixed ligand platinum complexes containing the BTF and DPAF chromophores, with spectra shown in pairs for B) BTF Pt CCBTF and DPAF Pt CCDPAF , C) BTF Pt CCBTF and DPAF Pt CCBTF , and D) DPAF Pt CCDPAF and BTF Pt CCDPAF to emphasize trends.

PAGE 107

107 maximum at 600 nm. DPAF Pt CCBTF and BTF Pt CCBTF complexes show two features, as is typical for the CCBTF ligand. The TTA maximum is found at ~660 nm, and a minor feature is seen at just under 500 nm. Once again, BTF Pt CCDPAF shows triplet state behavior that breaks with the trends that have been previously seen in purely platinum acetylide systems. A close examination of the TTA of BTF Pt CCDPAF relative to the other three mixed ligand complexes first shows that the absorption maximum is blue shifted from DPAF Pt CCDPAF . Additionally, the absorption of BTF Pt CCDPAF follows the minor feature of the complexes with the CCBTF ligand in the 425 500 nm region . The se two observations from the TA spectra , along with the phosphorescence data the data again point a mixing of the triplet states in BTF Pt CCDPAF . Electronic Structure Calculations To further probe the nature of the electronic structure, and to help deconv olute the absorption spectra of these mixed ligand complexes, DFT and TD DFT calculations were performed in the Gaussian 09, revision C.01, 73 suite of programs at the B3LYP level . The 6 31G(d) basis set was used for C, H, N, the 6 31+G(d) basis set was used for P, S, and the SDD basis set was used for Pt. Phosphine butyl groups and fluorene ethyl groups were truncated to methyl groups to improve computational effic iency. These truncated complexes are denoted by the addition of a prime ( ) to the end of their name. Thus, the synthesized complex BTF Pt CCBTF would have its computed congener named BTF Pt . In addition to studying the mixed complexes, the monosubs tituted Pt aryl and Pt acetylide complexes were studied to provide further deconvolution.

PAGE 108

108 It is beneficial to start with examination of the TD DFT results and electronic structure of the monosubstituted complexes, as they contain each ligand individually, allowing for unambiguous assignment of vertical excitations and orbitals unperturbed by a second organic ligand. A table providing data for the lowest energy electronic transition ( S 0 1 ) is given in Table 4 2. Summarized here is the predicted wavelength of the vertical excitation, the orbital transitions presented as both orbital number and frontier orbital nomenclature, the percent contribution of each individual transition, and finally the oscillator strength of the overall transition. The computed tra nsitions for the model complexes are quite straightforward, as they consist purely of the HOMO LUMO transition. A qualitative look at the predicted wavelengths for each monosubstituted platinum complex shows the expected energy trend of DPAF Pt > DPAFC C Pt BTF Pt Cl > BTFCC Pt Cl , is dangerous to examine the data quantitatively, as DFT methods are well known to over delocalize electron density. Regarding the composition of the orbitals, t he HOMO for each monosubstituted platinum complex is delocalized fully across the organic chromophore and is perturbed by the d xy of platinum. In complexes containing the BTF chromophore, the p y orbital of chlorine also makes a minor contribution. T he LUMO s on the other hand show Table 4 2. Summary of TD DFT computations for the S 0 1 transition of monosubstituted platinum aryl and platinum acetylide compounds . Complex Wavelength nm Orbital Transitions (% contribution) Oscillator Strength f BTFCC P t 395.1 151 152 HOMO ( 100 %) 1.4296 DPAFCC Pt 369.8 161 162 HOMO (100 %) 1.1782 BTF Pt Cl 367.2 ( 100 %) 1.3153 DPAF Pt Cl 349.7 (100 %) 0.8276

PAGE 109

109 localization of electron density. For platinum complexes containing the BTF chromophore, the system as a whole is an accep tor donor system, and localization is strongest in the vicinity of the C C bond connecting benzothiazole to fluorene. Platinum complexes functionalized with the DPAF chromophore however, represents a donor donor system, forcing the electron density to build up around the central C C bond of the fluorene moiety. Minimal perturbations are seen from platinum, and none from chloride. Figures of relevant orbitals are given in Appendix E. To determine the overall difference in electron density between ground and excited state , charge difference density (CDD) plots were utilized, and are shown in Figure 4 9 alongside the computed singlet ground state structure . In the CDD, blue coloring represents electron density being lost, or where the electron density origi nated (ground state). Red coloring in the CDD represents electron density being gained, or where the electron density finishes (excited state). For the systems containing the BTF chromophore, the charge originates primarily on the platinum and the phenyl r ing of fluore ne that is closest to platinum. Charge then moves towards the benzothiazole acceptor character of BTF. Complexes containing the DPAF chromophore show nearly the opposite effect, as the diphenylamine group is a strong donor. Platinum also donates charge to the system, which causes significant Now that the electronic properties of the monosubstituted complexes have been established, they can be used to deconvolu te effects observed in the mixed ligand platinum compounds. TD DFT data for the mixed compounds are given in Table 4 3, which is formatted in the same manner as Table 4 2. Focusing first on BTF Pt

PAGE 110

110 Figure 4 9. DFT optimized structures of the sing let ground state and charge difference densities for the computed S 0 1 vertical transition for A) BTFCC Pt , B) DPAFCC Pt , C) BTF Pt , and D) DPAF Pt . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in an electronic transition. CDDs were image d at an isovalue of 0.0004. and DPAF Pt , the red most vertical transition is dominated (>90% contribution) by the HOMO LUMO transition. The HOMO LUMO transition for DPAF Pt is localized almost exclusively on the CCDPAF ligand, with mino r perturbation from the platinum center and DPAF ligand. Interestingly, BTF Pt CCBTF shows nearly the exact opposite behavior. The HOMO is localized primarily on the

PAGE 111

111 Table 4 3 . Summary of TD DFT computations for transition s of oscillator strength > 0.1 an d for the mixed platinum aryl/acetylide complexes. Complex Wavelength nm Orbital Transitions (% contribution) Oscillator Strength f BTF Pt 408.4 359.8 227 230 HOMO +1 227 229 HOMO UMO HOMO 1 +1 (2. 5 %) (9 4 . 0 %) (3.5%) ( 43.7 %) ( 53 .7% ) (2.6%) 2. 0824 0.8753 DPAF Pt 372.4 3 49.5 2 47 50 HOMO 2 1 2 HOMO +1 (6.1 %) (9 3 . 9 %) ( 26.0 %) ( 50.5 %) ( 23.5 %) 2.0126 0.3309 BTF Pt CCDPAF 438.5 a 375 .6 3 50.3 2 HOMO HOMO +1 2 HOMO HOMO +1 ( 10 0 %) (12.0 %) ( 43.0 %) (45 .0 %) ( 86.3 %) (11 .6%) (2.1 %) 0.1089 2.5273 0. 2525 DPAF Pt CCBTF 421.9 a 401.4 350.0 HOMO HOMO 2 3 40 1 (7.9 %) ( 92.1 %) ( 91 .6%) ( 8 .4 %) ( 100 %) 0.5378 1.2439 0. 7 6 22 a These predicted vertical transitions are charge transfer bands which are not observed in the experimental absorption spectrum at concentrations ranging from 0. 4 mM to 0.4 µ M. Pt CCBTF unit, but the LUMO is completely d elocalized across both BTF chromophores with no contribution from the platinum core. These two processes are illustrated in the CDDs presented in Figure 4 10. The higher energy predicted transition for both of these complexes are mostly similar in that the ligand, with slight ligand ligand charge transfer (LLCT) from the aryl to the acetylide

PAGE 112

112 Figure 4 10. DFT optimized structures of the singlet ground state and charge difference densities for the comput ed S 0 1 vertical transition for A) BTF Pt CCBTF and B) DPAF Pt C CDPAF . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in an electronic transition. CDDs were imaged at an isovalue of 0.00 04. ligand. Molecular orbitals for these compounds and CDDs for the high energy predicted vertical transition are given in Appendix E. Moving on to the mixed ligand complexes utilizing both chromophores, BTF Pt and DPAF Pt are predicted to have three transitions relevant to this study. However, the lowest energy predicted transition for both of these compounds shows purely charge transfer character from either CCDPAF to BTF or DPAF to CCBTF, respectively. These charge transfer bands are not observed in the

PAGE 113

113 experimental UV Vis at concentrations between 0.4 mM and 0.4 µ M. Aside from overdelocalization and through space interactions, charge transfer processes are the other well known deficiency of the DFT methodology. With these two erroneous ba nds dismissed , each mixed ligand complex now has two predicted vertical transitions that are of interest. As stated earlier, the individual ligands in BTF Pt are expected to be quite close in energy. This is exemplified by a single strong absorptio n in the experimental UV Vis spectrum, and the claim is strengthened by the TDDFT computation predicting a high degree of mixing of orbital contribution to a transition with very high oscillator strength. The CDD for this transition reveals donor charact er system of the CCDPAF ligand as well as from the platinum acceptor character is seen on both ligands, but with preference for BTF. The CDD for BTF Pt CCD is shown in Figure 4 11. The final compound to be discussed is DPAF Pt CCBTF , and it has the largest energy gap between its two ligands. In looking at the experimental UV Vis spectra, this becomes apparent as vibronic structure of the [Pt] CCBTF absor ption can be plainly seen. The other three compounds do not show vibronic structure due to increased overlap of the absorption bands of the two ligands. Looking at the vertical transition with the lowest energy (that is not the erroneous CT band), the tran sition is dominated by the HOMO 1 analogous manner, the high energy TDDFT predicted transition of 100% HOMO e

PAGE 114

114 Figure 4 11. DFT optimized structure of the singlet ground state and charge difference density for the vertical transition predicted at 376 nm of BTF Pt . Blue coloring indicates electron density being lost, while red coloring indicates elect ron density being gained in an electronic transition. CDDs were imaged at an isovalue of 0.0004. contribution from platinum. This finding gives a great amount of insight as to the phenomenon of dual fluorescence in DPAF Pt CCBTF . The ligands of this compou nd other, and are allowed to decay independently of one another. The CDDs for both predicted transi tions are shown in Figure 4 12. Summary and Conclusions In summary, f our platinum(II) complexes featuring one aryl and one acetylide ligand were synthesized and photophysically characterized. Synthesis was achieved by first using a recently developed Stille type coupling reaction of an aryl stannane with catalytic CuI to ge nerate a pair of monosubstituted platinum(II) aryl precursors. These two compounds were each reacted with two acetylide ligands, containing the same two chromophores, via standard Hagihara reaction conditions to afford the mixed ligand platinum complexes i n good yield.

PAGE 115

115 Figure 4 12. DFT optimized structure of A) the singlet ground state , and charge difference densities for the vertical transition s predicted at B) 401 nm and C) 350 nm of DPA F Pt CC BT . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in an electronic transition. CDDs were imaged at an isovalue of 0.0004. The absorption spectra show minimal coupling of the ligands through the platinum center. This is evidenced by the absorption onset of the two compounds containing the lowest energy CCBTF ligand being identical . Additional evidence is given by TDDFT computations. Specifically, the absorption of the mixed ligand complexes can be reasonably well modeled by the sum of the absorptio ns of the respective monosubstituted platinum aryl and platinum acetylide constituents . When the mixed ligand platinum complexes are excited at their absorption maxima, weak fluorescence is observed for all species, with no unusual features. However, DPAF Pt CCBTF gives a very pronounced dual emission upon excitation with 330 nm light. The TDDFT calculations inadvertently shine light onto the potential nature

PAGE 116

116 of this process, as they show the two main absorption processes are localized mostly on a single li gand (one absorption per ligand). Thus it seems the two excited state potential wells do not intersect at a point where efficient internal conversion can occur, and radiative decay is seen from both states. All complexes show efficient intersystem crossing , as evidenced by low Additionally, all complexes show phosphorescence at room temperature with fair quantum yields between 10 and 16%. While DPAF Pt CCBTF does not show any unique emission in the triplet state, its structural isomer BTF Pt CCDPAF does appear to show more than one radiative process. The resolution at room temperature does not allow for any definitive assignment, but the broadness of the structured emission as well as the significant increase in the ratio of the 0 1 vibronic band relative to the 0 0 band suggests that a second emission is present. Transient absorption spectroscopy provides further evidence for a mixing of the triplet states in BTF Pt CCDPAF . The three other mixed ligand compounds in the series b ehave nearly identically to their platinum acetylide congeners in the TA spectrum, in that the characteristics are based solely on the lowest energy chromophore. However, BTF Pt CCDPAF shows characteristics of both the BTF and CCDPAF ligands in the TA expe riment. As the energy of the excited state for the BTF and CCDPAF ligands are the closest in this series, a couple of pathways for population of non T 1 states is possible. The most straightforward would be the thermal population of T 2 if it is close enough in energy. The other option would invoke a similar explanation to the rational e for dual

PAGE 117

117 fluorescence in DPAF Pt CCBTF . Excitation at the main absorption band of BTF Pt CCDPAF creates an excited state where charge is delocalized across a majority of the m olecule, according to TDDFT computations. This potential well most likely has overlap with the triplet state potential wells of both the BTF and CCDPAF ligands . Intersystem crossing would then occur before internal conversion in the singlet state could fin ish, allowing the exciton to decay to either ligand. The individual triplet potential wells of BTF and CCDPAF would have poor overlap at reasonable energies, allowing radiative decay from either triplet state. The class of mixed ligand platinum(II) aryl/ac etylide complexes marks a very interesting development in platinum chromophores. While more experiments are certainly necessary to pin down the exact nature of the dual emission phenomena, this class of compounds shows promise that these observations were more than just good luck in choice of chromophore. With further study, it may become routine to generate fully conjugated platinum chromophores that display predictable dual emission. Experimental General Remarks All reactions were carried out under argon atmosphere. All starting chemicals used for the synthesis of the trans platinum acetylide complexes were purchased from commercial suppliers, reagent grade, and used without additional purification. K 2 PtCl 4 was purchased from Strem Chemicals. Solvents were of reagent grade unless otherwise noted. BTF Br , BTFCC H , DPAF Br , and DPAFCC H were synthesized according to literature procedures . 2 DPAF Pt Cl was synthesized by a literature technique, 55 and BTF Pt Cl was synthesized by modification of this technique , but could not be purified past 95%, and was carried on at this purity without further

PAGE 118

118 characterization . Chromatography was performed on a CombiFlash Rf 1 50 Medium Pressure Liquid Chromotography system. 1 H (500 MHz) and 13 C (125.7 MHz) NMR spectra were recorded on a Varian Inova spectrometer. 31 P ( 121 MHz) NMR spectra were recorded on a Varian Mercury spectrometer. The chemical shifts were reported in ppm relative to tetramethylsilane (TMS) or residual solvent peaks . Absorption and Emission Spectroscopy Steady s tate absorption spectra were recorded on a Varian Cary 50 spectrophotometer. Corrected steady state emission measurements were performed on a Photon Technology International spectrophotometer (QuantaMaster). Absorption and fluorescence experiments were car ried out in aerated solution, while phosphorescence samples were deaerated with at least five freeze pump thaw cycles. Optically dilute samples with O.D. < 0.1 at the excitation wavelength were used. Fluorescence and phosphorescence quantum yields were det ermined by relative actinometry, with Ru(bpy) 3 Cl 2 Fl = 0.0379 in air saturated water) . 70 Phosphorescence lifetimes were obtained with a multichannel scaler/pho ton counter system with a NanoQuant FluoTime 100 Compact Phosphorescence Lifetime Spectrophotometer. A UV pulsed diode laser provided the excitation at 375 nm (power < 10 mW). The laser was pulsed by a PDL800 B, which is a pulsed diode laser driver. Optica lly dilute solutions were freeze pump thawed five times. All decays were obtained using single exponential fitting parameters (Fluo Fit software). Nanosecond Transient Absorption (TA) Spectroscopy Measurements were performed on an in house apparatus that is described in detail elsewhere . 77 The third harmonic of a Continuum 355 nm, 10 ns fwhm, 180 µJ pulse 1 ) was used as the excitation source. Probe light

PAGE 119

119 was produced by a xenon flash lamp and the transient absorption signal was detected with a gated intensified CCD mounted on a 0.18 M spectr ograph (Princeton PiMax/Acton Pro 180). Solutions had a matching optical density of 0. 58 after at least five freeze pump thaw cycles. An initial CCD image capture delay of 100 ns following the laser pulse was used to ensure full conversion to the triplet s tate before observation. An average of 100 images were acquired and the laser energy was 180 µJ/pulse, which was established as being sufficiently low to minimize triplet triplet power dependence. Computational Details DFT and TD DFT calculations were perf ormed in the Gaussian 09, revision C.01 , 73 suite of programs at the B3LYP level with the 6 31G(d) basis set for C, H, N, the 6 31+G(d) basis set for P, S, and the SDD basis set for Pt. Phosphorous butyl groups and fluorene ethyl groups were replaced by methyl groups to improve computational efficiency. The ground state structures were optimized in the gas phase from idealized starting configurations without symmetr y constraints. The optimized structures were confirmed to be minima by the lack of imaginary frequencies . Structures and orbitals were visualized at an isovalue of 0.02 using Chemcraft Version 1.7 , 79 which was also used to generate charge difference density (CDD) plots. CDDs were imaged at an isovalue of 0.0004. General Procedure for the Hagihara Coupling Reaction for Mixed Platinum Aryl Acetylide Compounds. A platinum aryl precursor ( BTF Pt Cl or DPAF Pt Cl ) (0.100 g, 0.11 mmol, 1 eq.) was dissolved in 25 mL of Et 2 NH, a nd the solution was degassed with argon for 25 minutes. CuI (0.005 g) and 1.1 equivalents of the deprotected arylacetylide ( BTFCC H or DPAFCC H ) were added. The solution was allowed to react overnight at room

PAGE 120

120 temperature, in the dark. The resulting mixture was poured into water and extracted with CH 2 Cl 2 . The organic layer was separated, filtered over Na 2 SO 4 , t he filtrate was reduced to a minimum, and it was loaded on to silica gel . Purification via MPLC using hexanes/dichloromethane afforded the pure produc ts as oils. Microcrystalline product s can be obtained by dissolving the compound s in a minimum of acetone and inducing precipitation with MeOH. The mixtures were cooled, decanted, washed with additional MeOH, and dried to afford the four pure mixed ligand platinum products. trans ( PBu 3 ) 2 Pt (BTF)(CCBTF) ( BTF Pt CCBTF ) Isolated as yellow microcrystals. Yield : 76 %. 1 H NMR (500 MHz, CD 2 Cl 2 8.1 2 ( s , 1 H), 7 . 98 8. 08 (m , 5 H), 7.9 4 ( m , 2H), 7. 76 (d, J = 7.5 Hz, 1H), 7.7 2 (d, J = 7.5 Hz, 1H), 7.6 4 (d, J = 7.5 Hz, 1H), 7. 27 7.55 (m, 9H), 2.1 1 (m, 8H), 1. 79 (m, 12H), 1. 58 ( m , 12 H), 1.4 1 (sextet, J = 7.5 Hz, 12 H), 0.9 4 (t, J = 7.5 Hz, 18H), 0. 38 (t , J = 7.5 Hz, 6 H) , 0. 36 (t, J = 7.5 Hz, 6H) . 13 C NMR (126 MHz, CD 2 Cl 2 ): 169.6, 169.2, 154.94, 154.90, 154.8, 151.3 151. 0, 150.2, 149.6, 147.5, 145.5, 138.4, 137.6, 135.6, 135.5, 135.2, 133.8, 132.2, 131.3, 130.3, 130.2, 127.6, 127.5, 126.8, 126.7, 125.8, 125.5, 125.3, 123.4, 123.2, 122.20, 122.15, 122.09, 122.02, 120.4, 120.1, 119.4, 119.1, 56.8, 56.4, 33.5, 33.3, 26.8, 24 .9 (t, J Pt C = 6. 6 Hz), 23.3 (t, J Pt C = 16. 8 Hz) , 14.2, 9. 2 , 8. 9 . 31 P NMR (12 1 MHz, CD 2 Cl 2 ) 1.67 ( J Pt P = 2 600 Hz ). trans ( PBu 3 ) 2 Pt (DPAF)(CCDPAF) ( DPAF Pt CCDPAF ) Isolated as a white powder. Yield: 6 0 %. 1 H NMR (500 MHz, CD 2 Cl 2 7. 46 7.55 ( m , 3 H), 7. 20 7.33 (m, 13 H), 7. 0 9 (m, 10 H), 6.93 7.04 ( m , 6 H), 1.90 (m, 8H), 1.7 7 (m, 12H), 1.5 6 (m, 12H), 1.4 0 (sextet, J = 7.5 Hz, 12 H), 0.9 2 (t, J = 7.5 Hz, 18H), 0.3 6 (t, J = 7.5 Hz, 6H), 0.3 4 (t, J = 7.5 Hz, 6H). 13 C NMR (126 MHz, CD 2 C l 2 151.8, 150.7, 150.1, 148. 9 , 148.7, 147. 1 , 146.0, 139.9, 138. 4 , 13 8.0 , 137. 8 , 135. 8 , 13 4.0 , 133.5,

PAGE 121

121 130.0, 129. 7 , 129. 6 , 125. 7 , 124.8, 124.4, 124. 2 , 123.7, 122.8, 122.4, 121.0, 120. 4 , 120. 3 , 119.4, 119.1, 118.5, 56.4, 56.1, 33. 5 , 33.2, 26. 8 , 24.9 (t, J Pt C = 6. 6 Hz) , 23.2 (t, J Pt C = 16. 6 Hz) , 14.2 , 9.1 , 8. 9 . 31 P NMR (12 1 MHz, CD 2 Cl 2 ) 1.66 ( J Pt P = 2 60 0 Hz ). trans ( PBu 3 ) 2 Pt (BTF)(CCDPAF) ( BTF Pt CCDPAF ) Isolated as dull yellow powder . Yield: 43 %. 1 H NMR (500 MHz, CD 2 Cl 2 8 . 05 (s, 1H), 8.02 (d, J = 8.0 Hz, 1H ) 7.9 9 ( d , J = 8.0 Hz, 1H ), 7.9 3 ( d , J = 8.0 Hz, 1H ), 7.7 1 (d, J = 8.0 Hz, 1H), 7. 53 (d, J = 8.0 Hz, 1H), 7 .46 7. 52 (m, 2 H), 7 .36 7. 45 (m, 4 H), 7 .21 7. 27 (m, 6 H), 7.10 (m, 5 H), 7.00 (m, 3 H), 2.1 0 (m, 4 H), 1.90 (m, 4 H), 1.7 8 (m, 12H), 1.5 7 (m, 12H), 1.4 0 (sextet, J = 7.5 Hz, 12 H), 0.9 2 (t, J = 7.5 Hz, 18H), 0.3 6 (t, J = 7.5 Hz, 6H), 0.3 5 (t, J = 7.5 Hz, 6H). 13 C NMR (126 MHz, CD 2 Cl 2 ): 6 , 154.9, 151.8, 150. 2 , 150.1, 149. 6 , 148.7, 147.5, 147.1, 138.4, 137. 8 , 135.5, 135. 2 , 133.8, 131. 3 , 130. 1 , 129. 7 , 127. 5 , 126.7, 125. 7 , 125.3, 124.4, 124. 2 , 123.2, 122. 9 , 122.1, 122.0, 120. 4 , 120. 3 , 119. 4 , 119.12, 119.08, 56.41, 56.38, 33. 5 , 33.2, 26 . 8, 24.9 (t, J Pt C = 6. 7 Hz), 23.2 (t, J Pt C = 16.8 Hz) , 14.2, 9.1, 8. 9 . 31 P NMR (12 1 MHz, CD 2 Cl 2 ) 1.59 ( J Pt P = 2 6 0 0 Hz ). trans ( PBu 3 ) 2 Pt (DPAF)(CCBTF) ( DPAF Pt CCBTF ) Isolated as light yellow microcrystals . Yield : 8 0 %. H NMR (500 MHz, CD 2 Cl 2 8.10 (s, 1H), 8.04 (m, 2 H), 7.95 (d, J = 8.0 Hz, 1H), 7.76 (d, J = 8.0 Hz, 1H), 7.6 3 (d, J = 8 .5 Hz, 1H), 7.50 (m, 2 H), 7.40 (td, J = 7.5 Hz, J = 1.0 Hz, 1H), 7.30 (m, 5H), 7.2 3 (m, 4 H), 7.08 (m, 5 H), 6.97 (m, 3 H), 2.1 0 (m, 4 H), 1.90 (m, 4 H), 1.7 7 (m, 12H), 1.5 7 (m , 12H), 1.41 (sextet, J = 7.5 Hz, 12 H), 0.9 3 (t, J = 7.5 Hz, 18H), 0.3 7 (t, J = 7.5 Hz, 6H), 0.3 5 (t, J = 7.5 Hz, 6 H ). 13 C NMR (126 MHz, CD 2 Cl 2 169.2, 154.9, 151.3, 151.0, 150.7, 148.9, 146.1, 145.6, 139.9, 137.9, 137.5, 135.8, 135.6, 133.5, 132.2, 130.3, 129.6, 127.6, 126.8, 125.8, 125.5, 124.8, 123.7, 123.4, 122.5, 122.2, 122.1, 121.0,

PAGE 122

122 120.4, 120.1, 119.4, 118.6, 56.8, 56.1, 33.5, 33.3, 26. 75, 24.9 (t, J Pt C = 6.4 Hz), 23.2 (t, J Pt C = 16.8 Hz), 14.23, 9.13, 8.86 . 31 P NMR (12 1 MHz, CD 2 Cl 2 ) 1.73 ( J Pt P = 2 6 0 0 Hz ).

PAGE 123

123 APPENDIX A TSUNAMI FEMTOSECOND LASER MANUAL Overview The Tsunami laser is a titanium doped sapphire (Ti:Saph) laser, with a usea ble emission from 700 to 1050 nm. It is pumped by a Millenia eV, which emits the 2 nd harmonic of a Nd:YAG source in a continuous wave at 532 nm. Th e pump laser has a maximum power of 10 W, with the standard operating range of 8 9 W. This gives an output la ser power from the Ti:Saph of 0.5 to 1.8 W depending on the output wavelength selected. The Ti:Saph repetition rate is 80 MHz, and the pulse width is less than 100 fs. Assuming a 1.5 W average output power, this results in an ~1.9 nJ/pulse energy. If one t hen assumes a 100 fs pulse width, the power delivered per pulse is 19 0 kW. Covered in this manual will be general operation of the laser, acquiring a basic emission spectrum from two photon excitation, determination of a two photon absorption cross section , and a troubleshooting guide. Sample concentration may vary between 10 4 and 10 5 M, depending on the fluorescence quantum yield s of the samples. If obtaining a two photon cross section, reference compounds with similar absorption/emission energies shoul d be selected, 90 and the exact concentration of all samples and references must be known so that corrections can be made. The basic absorption and emission properties of the sample should be well known prior to attempting to measure a two photon cross section. * ** Please note that i t is incredibly easy to detune the laser to the point where it no longer lases! Be extraordinarily careful when adjusting the colored knobs to optimiz e the laser power! * **

PAGE 124

124 Laser Start up 1. Turn on chiller underneath the laser table. 2. Remove all plastic bags from the optics. Additionally, on the fluorimeter, remove the black piece of plastic so the laser may enter the sample chamber. 3. If it is not already there, place the white Teflon block on the Erlenmeyer flask at the laser output. This will eventually act as a beam stop/diffuser. 4. Open the computer. It should already be logged in and at the desktop. 5. Plug the Ocean Optics detector into the side USB port on the laptop. 6. Ocean Optics detector you should see weak peaks in the visible region from the ambient lighting. 7. Once the chiller has reached 19 ° C, the laser may be turned on. First look at the back of the small laser (Nd:YAG pump laser). In the top left area, there is a key that needs to be turned 90 ° clockwise, and a power button that needs to be depressed. 8. Next, on th moments to connect to the instrument. 9. In the middle of the Spectra Physics software, there will be an indicator telling whether the hardware is ready or not. The laser will take 10 20 minu tes to warm up, as depicted by a yellow indicator. 10. Once the laser has warmed up the indicator will turn green and be ready to be activated. This is shown in Figure A 1. 11. At the bottom of the Spectra Physics software, the power should be set to it is not, change it to this setting. 12. At the top, toggle the switch to turn on the laser. A small window will 13. The laser will turn on and the power should increase to the power se tting selected at the bottom of the software. If the power does not reach the desired setting, or it is not stable, see the troubleshooting section. The software panel with the pump laser at both low power and requested power is also shown in Figure A 1. 14. T oggle the switch to open the shutter, this will allow the laser to leave the instrument.

PAGE 125

125 15. c The laser is now ready for operation. Figure A 1. Spectra Physics software showing A) the i nstrument has warmed up and is ready for to be used. B) Pump laser is operational, but has not come up to the requested power. C) Pump laser is opera tional, at power, and ready for use. Laser Wavelength Selection, Mode Locking, and Output Power Optimization Wavelength Selection and Mode Locking 16. On top of the laser there are 2 metal knobs, 2 green knobs, and 2 blue knobs. The two metal knobs are for wav elength adjustment (knob closest to user), and prism adjustment (knob furthest from user). These knobs must be used in tandem to set the output wavelength while keeping the laser mode locked. The top of the instrument is shown, with knobs labeled, in Figur e A 2. know the instrument is mode locked by the large bandwidth of the laser output (Full Width Half Maximum bandwidth: ~6 nm @ 700 nm output, ~16 nm @ >800 nm output) If th e instrument is not mode locked, the output bandwidth will be very not pulsing. 2PA experiments do not work if the laser is not pulsing. A representati ve spectra of the outp ut when mode locked and in continuous wave form is given in Figure A 3 **

PAGE 126

126 Figure A 2. Top view of the Tsunami laser with each knob labeled. Figure A 3. Shape of the emission band of the Tsunami laser when A) emitting as a continuous wave and B) wh en pulsing (mode locked). 17. To change the wavelength, have the Spectra Suite software displayed on the computer and turn the metal knob closest to you in small increments. Clockwise rotation will increase the output wavelength; counterclockwise rotation will decrease the output wavelength. **While adjusting the wavelength, it is important to simultaneously adjust the prisms using the other metal knob. When done properly this will keep the instrument mode locked throughout the entire process of changing wavele ngth. A B

PAGE 127

127 of an art than a scientific process. With practice this will become easier. A general tip is to try to keep the emission as broad as possible while adjusting the wave length. If the output bandwidth is narrowing while wavelength adjustment is being made, adjust the prisms to broaden the peak back out.** Output Power Optimization 18. Once the instrument has been tuned to the proper wavelength, and is mode locked, move the po wer meter into the path of the beam after the first mirror. Then remove the white Teflon block that is blocking the beam. 19. Optimum energies at specific wavelengths are given at the end of this document. Depending on the day, you may or may not be able to ma tch the powers given by the technician. 20. To optimize the output power, the four colored knobs on top of the laser will be used. They are VERY sensitive! You should never have to take your fingers off of the knob while using it. If twisted even a 1/4 of a re volution TOTAL, lasing can be lost. Use onl y very, very small adjustments. The green knobs adjust the mirrors horizontally, while the blue knobs adjust vertically. A ny knob can be chosen to start, but experience suggests starting with the blue knob closest to the output of the Tsunami (users right). S tart by maximizing the power using only that knob. 21. Once the power is maximized using the first knob, move to the other knob of the SAME color, and optimize the power again. Go back and forth with knobs of the s ame color , one knob at a time, until the power cannot go higher. 22. Repeat this process with the other pair of colored knobs. Ideally, but not always, the power will now match the one reported by the technician. **The mirrors controlled by these knobs should be reoptimized at every new wavelength to ensure maximum output power. If scanning a large region quickly, reoptimize every 50 nm. ** Two Photon Excited Fluorescence (2PEF) If you are obtaining a two photon cross section, skip to Step 31 . If not continue wi th Step 23 . 23. Before exposing the sample to the laser, make sure that the beam dump is in place in the fluorimeter. This will prevent the laser from being directed into the fluorimeter lamp, which will damage it. Also make sure the black plastic cap has been removed, so the laser may enter the sample chamber without burning through the plastic . 24. Once the beam stop is in place, insert the sample into the fluorimeter.

PAGE 128

128 25. Next, remove all obstructions (power meter/Teflon beam stop ) from the path of the beam, so it m ay pass through the sample. If the sample has 2PA at the current laser wavelength, you should see luminescence from the sample. 26. Before acquiring any data, scan the laser wavelength to see at what input w avelengths your sample has 2PA. **Remember, if you ca no point in collecting data at this laser wavelength. ** 27. Once you know the approximate laser wavelengths that will give 2PA, set the laser to either the highest or lowest wavelength. This will all ow you to work from one end to the other. 28. In the software, select your acquisition region, and set the integration time to 0.5 seconds, with a step size of 1.0 nm. Set the lamp wavelength to a value outside of the acquisition region. Be sure to turn off th operate properly under these conditions. 29. Give the data a name (use eight characters or less), and then run the acquisition. Repeat this for all samples and blanks before adjusting the laser wavelength. 30. Repeat this for all vi able laser wavelengths in 10 nm increments . Two Photon Cross Section Measurements While the data collection steps are the same as described above, some steps must be taken before measuring the two photon excited fluorescence. First, the samples should be p repared, preferably with an overlap in absorbance at a mutually beneficial wavelength. This wavelength will be the selected excitation wavelength to observe a one photon emission. The fluorescence observed from the one photon process determines a parameter called the differential quantum yield (DQY) . The DQY is effectively the emission intensity of a sample at a single wavelength (rather than the emission intensity of samples excited with two photons is still based on the fluorescence quantum yield of the sample. A correction is made in the data work up for the difference in DQY of the sample and the reference.

PAGE 129

129 31. Select an excitation wavelength that is approximately hal fway between the absorption maxima of both the sample and the reference. Ideally, the absorption maxima of these compounds will be close to each other. Unfortunately this is not always the case, and you may have to use an excitation wavelength that is not near the maxima. If at all possible, try to make the sample and reference have matching absorptivity values at the excitation wavelength. This ensures the number of photons be ing absorbed by both compounds are the same. An example of selecting a n excitatio n wavelength from the absorption is shown in Figure A 4. Additionally, this process should be repeated for each new sample. 32. Using the predetermined excitation wavelengths, take a one photon emission spectrum for all samples and references. Make sure that t he beam dump and black card blocking the lamp have been removed. Also replace the black piece of plastic in the side of the fluorimeter and enable the dark offset, as it works properly when the sample chamber is sealed. 33. Once the one photon emission of all samples and references has been taken, pick an emission wavelength approximately halfway between the emission maxima of both the samp le and the reference , preferably where the emission intensities are equivalent . An example of selecting a registration wave length is also shown in Figure A 4. The ratio of these intensities is the DQY for this sample/reference pair. This DQY is valid ONLY at this one wavelength. Take not e of the DQY wavelength. You will need to take emission measurements from the two photon ex cited samples at this wavelength. Figure A 4. Absorption (solid colored lines) and emission (dotted lines) of two complexes showing excitation and registration wavelength selections (solid black lines) , respectively .

PAGE 130

130 34. Once DQYs have been obtained for ea ch sample/reference pair, replace the beam dump and black card, remove the plastic cap from the side of the sample compartment, and turn off the dark offset in the software. Return to step 16 to acquire the emission data. Data Conversion 35. Once all data has been collected, it needs to be converted to a useable file type. 36. onvert your file , and click okay twice. This converts from the raw instrument data file to a text file. 37. The computer is not connected to the network, so it must be transferred with a flash drive. Instrument Shut Down 38. Block the laser beam with the Teflon block. 39. In the Spectra Physics software, close the shutte r and toggle the laser to off. You should see the power drop to 0 W. 40. On the back of the laser head, push the button and turn the key a ¼ turn counterclockwise. 41. After all of the data is converted (as mentioned in the last section) close the Fluoromax softwa re and turn the instrument off. Log the time in the notebook. 42. Five minutes after the laser has been turned off, it is safe to turn off the chiller. Laser Powers and Bandwidths at Given Wavelengths Table A 1. Laser powers and bandwidths at given wavelength s. Wavelength (nm) Input Pump Power (W) IR Output (W) FWHM Bandwidth (nm) 700 8.2 0.7 6 720 8.1 1.1 8 760 8.2 1.75 13 800 8.2 1.6 13 850 8.8 1.5 17 900 8.8 1.1 14 1000 9.0 0.5 14

PAGE 131

131 Data Workup The final step in a two photon cross sec tion experiment is to plot the data and run the calculations. If only 2PEF data was collected, simply plot the multiple emission spectra as a function of fluorescence wavelength, with additional curves for each laser wavelength used. The two photon cross s ection is calculated using the equation , where 2 is the two photon cross section, F 2 reg ) is the ifferential respectively. If the absorption spectra at the selected excitation wavelength were not matched, then a correction factor of can be multiplied throug h. The simplest way to plot these data is to do all manipulation s in different worksheets in an Excel workbook. If necessary, due to non identical absorbance of the sample and reference at the absorption selected excitation wavelength, the first worksheet should contain the absorbance data for both compounds. Take the ratio of the sample absorbance to reference absorbance at the selected excitation wavelength, and make note of this value. If solution and reference absorbance values were matched, the first w orksheet should contain the one photon excited emission data . Take the ratio of reference fluorescence intensity to sample fluorescence intensity, at the registration wavelength, and make note of this value. On the side of one of these initial worksheets, take the ratio of reference solution concentration to sample solution concentration, if they were not identical, and make note of this value. In the next worksheet, all of the 2PEF data can be added. Make sure to carefully label each data set so there is n o confusion as to what value belongs to what data set.

PAGE 132

132 Take the ratio of sample 2PEF intensity to reference 2PEF intensity at the registration wavelength, and repeat this for each laser wavelength. Make note of all of these values. Multiply the 2PEF ratio by the previous values that were noted, as well as the reference two photon cross section, to give the sample two photon cross section . Repeat this for all laser wavelengths, keeping in mind the reference two photon cross section will change at every wavel ength. Troubleshooting Pump Laser is not Coming Up to the Set Power 1. communication window. 2. The communication window has a command prompt as well as an input/output section for old commands/returns. 3. This pump laser power problem is caused by un optimized Second Harmonic A graphic showing the communication window with the SHG queried and the respons e from the instrument is shown in Figure A 5. Figure A 5. Spectra 4. A 5 digit value will be returned, this is the current setting. Typically (>90% of the time), the SHG will need to be adjusted down from the current value. Start in

PAGE 133

133 increments of 25. To adjust the SHG, type: shg:XXXXX where XXXXX is the desired value. 5. After a few seconds, you should notice the energy change . If the energy moves all the way to the desired value and stays stable, you are done and can close the communication window. If the energy increases, but not all the way to the desired value, continue to incrementally adjust the SHG until the desired powe r is reached. If the energy goes down, adjust the SHG in the other direction as described above . 6. If adjusting in either direction in small increments is not successful, the initial SHG value was probably way off to start. Reduce the value by 1000. This wil l effectively scan the SHG because the change is not instantaneous. Watch the power meter on the software, and if it spikes at all, the ideal SHG value is somewhere in this range. Work the SHG back towards the initial value to find the ideal value. Laser C an Not B e Tuned Below 780 nm. If the laser cannot be tuned to a wavelength below 780 nm, it is typically because the prisms and wavelength cutoff slit are misaligned. This is due to the user turning the wavelength adjustment knob too far, and the laser sho oting over the top of the internal slits. To remedy this problem, bring the height of the wavelength selection knob back to a central point. Next, adjust the height of the prism knob to match the height of the wavelength selection knob. Slowly adjust the w avelength back towards 700 nm, while moving the prisms simultaneously, and the 700 780 nm range should now be accessible. Advanced Troubleshooting If the system is no longer lasing , the most likely reason is that you have adjusted the knobs for the mirrors too much, and you have detuned the system. If you know exactly which knob you were using when this happened, and did not touch any of the other knobs trying to figure out what you did, then tune only that knob back to its original position. If the system still is not lasing, follow the steps below.

PAGE 134

134 1. Take off the cover of the Tsunami laser, by first releasing the four clasps in each of the corners. 2. Defeat the safety shutter. This will allow the laser to pass into the cavity of the Tsunami. Be careful as the radiation from the pump laser is very intense. Defeating the safety shutter is shown if Figure A 6. Figure A 6. Defeating the safety shutter on the Tsunami Laser. 3. Because the femtosecond system is not lasing, only a very weak beam will be seen in the c avity of the laser. The IR viewer may be needed, and is helpful regardless, for this process. The first thing to check is that the beam is passing through the wavelength selection slit going forwards talking about a beam that would be heading out of the instrument) . 4. Locate the wavelength selection slit. This will be directly below (attached to) the wavelength selection knob. 5. Ensure the be am is passing through the slit in the forward direction, keeping in mind that a large amount of the laser output is in the near IR range, and is not visible to the eye . Figure A 7 illustrates the area around the wavelength selection slit while also showing the laser beam passing through. 6. If the beam is missing the slit in the vertic al direction, adjust the wavelength selection knob until the beam passes through it. If this does not work, use the blue mirror adjustment knob on the users left (closest to the pump laser).

PAGE 135

135 Figure A 7. Area in the region of the wavelength selection sl it. The slit itself, the nearest prisms, and the laser beam passing through/being (properly) cutoff by the slit are pointed out. 7. If the beam is missing the slit in the horizontal direction, use the green mirror adjustment knob on the users left ( again, the set closest to the pump laser) to bring the beam back into alignment with the slit. 8. Once the beam is passing through the slit in the forward direction , check to see if the retroreflection of the beam is passing through the slit in the reverse direction. T his is best done with the IR viewer, especially if the retroreflection is very far off. 9. Do not adjust the wavelength selection knob. If the beam is missing the slit in the horizontal direction, use the green mirror adjustment knob on the users right (close st to the output port) to bring the beam back into alignment with the slit. 10. right (closest to the output port) 11. Reestablishing lasing takes very precise alignment of the las er (thus why it is so easy to lose lasing). It may take a little bit of very fine tuning to get this to happen. When lasing is reestablished, the beam will become significantly brighter (>100x). 12. Move the power meter into the beam path (outside of the laser cavity), and optimize the power as normal.

PAGE 136

136 APPENDIX B COMPUTATIONAL METHODS Overview Computational chemistry is a powerful tool that can enhance knowledge of existing compounds and help try to predict if a target compound will have desirable properties. This appendix will provide first time users with everything they should need to start running their computations from scratch, including but not limited to: a discussion of what job type gives what data, accessing the UF High Performance Computing Center (HPC), using the Gaussian suite of programs, writing input files/reading output files, and visualizing results. This section will discuss only the practical aspects, r ather than theory. If the user desires to familiarize themselves with the foundational pr inciples of computational chemistry, Essentials of Computational Chemistry 91 does a very good of explaining both fu ndamental and advanced principles . There are three main job types that are useful to the Schanze Group, with a few other types that are needed much less frequently. The three main calculation types are : geometry optim ization, frequency, and Time Dependent Density Functional Theory (TDDFT). The geometry optimization , as its name implies, tak es the atomic coordinates input by the user and finds an energy minimum by systematically altering the coordinates until the force constants on each atom reach a default threshold. The frequency job takes the geometry optimized molecule and computes all vibrational frequencies, effectively giving a simulated IR spectrum. If all frequencies are positive, the molecule is at an energy mi nimum, and can be used in a subsequent TDDFT job. If not, the molecule will need to have its geometry reoptimized, the method of which will be discussed later. Finally, TDDFT puts the molecule in a simulated electromagnetic

PAGE 137

137 field, and analyzes its response , to simulate a UV Vis absorption spectrum . A common work flow is shown in Figure B 1. Figure B 1. General workflow of computations. Less common job types include: stability, population, and volume. Of these , stability is the most common, and is invoked after a failed TDDFT job. During an optimization, the program may have inadvertently used a higher energy wavefunction to describe the molecule rather than the ground state wavefunction. Stability checks this and will reoptimize the molecule with the corr ect, ground state, wavefunction. The volume job is used to calculate the mo lecular volume of a system, and is typically used to find a value for the Onsager radius (symbol: a ) to be used in a manual calculation of the difference between ground and excited state dipoles . Finally, the population job will print contributions of specific atomic orbitals (AO) to a specified set of molecular orbitals (MO). This allows the user to determine the percent contribution of an atom or moiety to an MO. Accessing the HPC an account through the HPC and learn how to use UNIX. To open an HPC account, simply visit http://www.hpc.ufl.edu

PAGE 138

138 hours. An email will be sent to notify the user of account creation. The HPC is accessed using an SSH shell client at the address gator.hpc. ufl.edu . SSH uses a UNIX interface, and the VI editor word processor. Tutorials to both can be found here: http://www.ee. surrey.ac.uk/Teaching/Unix/ and here: http://www.unix manuals.com/tutorials/vi/vi in 10 1.html , respectively. Once connected to the HPC via SSH, the system will start the user username in the HPC file system. No computations should be run in this location. A large amount of storage for each user is allocated at the location /scratch/lfs/ username , and this is where all jobs should be created and stored. Please see the UF HPC website and their wiki page for additional details on proper use and technical details. General Considerations Before diving in to making files and running jobs, some general practices for computational chemistry will be laid out to ensure the user is aware of them. First, the DFT fu nctional and atomic basis set must be consistent throughout the entirety of the workflow. A frequency calculation using the 6 31G(d) basis set on a molecule that had its geometry optimized using 6 311+G(d) (or any other basis set that is not exactly 6 31G( d)) will be invalid. This goes for TDDFT jobs as well. The same principle also applies to the DFT functional that was chosen. For the work of the Schanze Group, the B3LYP functional tends to be superior. The 6 31G(d) basis set can be used for all light ato ms (Z < 10), the 6 31+G(d) basis set can be used for third row and lower non metals, and th e SDD basis set for all metals. Gaussian input files are not submitted directly to the HPC. The HPC uses a batch submission system and a scheduler to determine what jobs should be started to maximize allocation of resources. To achieve this , a submission script is required . This

PAGE 139

139 submission script contains all details about the hardware that the user is requesting for the job. Specific details and an example script wil l be provided in a later section. In order to use Gaussian, a confidentiality agreement must be signed. This is because Gaussian is restricted software. The form must be signed in person, is located in NPB 2335, and is available during any office hours of Dr. Deumens. See http://wiki.rc.ufl.edu/doc/Gaussian_License for further details. Usi ng Gaussian 09 and GaussView 5 via the HPC The user does not directly interact with the main Gaussian program. All interaction is through the SSH client and the GaussView 92 program. Gaussian simply accepts the input file, runs the computations, and generates an output file. Before GaussView can be used, an X Windows Remote display program must be used. Please see http://wiki.rc.ufl.edu/doc/Gaussview for details on how to install and configure the software. Once the software is set up, GaussView can be opened. The user should take some time and familiar ize themselves with the basic features and to ols that can be used to manipulate molecular structures. A reference guide for GaussView can be found here http://www.gaussian.com/g_tech/gv5ref/gv5ref_toc.htm . Figure B 2. A) 2 D Chemdraw model of (ThPt) 2 BTD . B) 3 D model of (ThPt) 2 BTD built and visualized in GaussView. C: Gray, H: White, N: Blue, P: Orange, S: Yellow, Pt: Cerulean. A B

PAGE 140

140 Throughout this appendix, the molecule (ThPt) 2 BTD will be use d as an example for all of the common computational methods , and its structure is shown in Figure B 2. Other molecules will be used for less common job types. Creating Input Files and Submitting Jobs This section will take the user through saving a created molecule as an input file (.com file type) for all different job types. Examples will be given, and specific keywords will be explicitly discussed. Figu re B 3. Figure B 4 shows a blank calculation setup window. The user will be able to specify all commands (keywords) through this window. Figure B 3. GaussView main window showing options of the Calculate tab. Figure B 4. Blank calculation setup wi ndow with notable areas labeled.

PAGE 141

141 Geometry O ptimization /Standard Input P arameters As stated earlier, the geometry optimization, or just optimization, is the first calculation run, and works to find an energy minimum on the potential energy surface of the mo lecule. However, the optimized geometry found may only be a local energy minimum, rather than the true ground state minimum. This is exemplified by the orientation of the thiophene rings relative to the BTD ring in (ThPt) 2 BTD . There are three major conform ations which can be defined by whether the sulfur atoms in each ring are cis or trans with each other. Thus the three conformations are S cis, S cis; S cis, S trans; S trans, S trans. These three conformatio ns are illustrated in Figure B 5 . Figure B 5 . Major conformations of (ThPt) 2 BTD . A single optimization will not be able to look at all three of these conformations because of the relatively high energy barrier to thiophene rotation. Three separate jobs must be created , one for each individual starting geometry. A n appended sample i nput file , with important lines and commands labeled, is shown in F igure B 6 . The input file contains a variety of information, and discussion of these parameters will begin with variables tha t are seen in multiple job types. The first two lines are commands to allocate the physical resources of processors and RAM. These values must match the values given in the HPC job submission script, which will be shown later. The third line is the name of the checkpoint file. All useful data created by every job is stored in the checkpoint file upon completion of the job. This allows the user

PAGE 142

142 Figure B 6 . Appended input file for the geometry optimization/frequency calculation of (ThPt) 2 BTD .

PAGE 143

143 to carry in formation between jobs so it does not have to be respecified by the user (atomic coordinates) or recalculated by the program (energies, force constants, etc.). The name of the file can be anything the user desires, as long as the filetype (.chk) is left in place. The user should be careful though, because when the checkpoint file is used in a subsequent job, the file will be overwritten if it has not been copied and given a different name, or if it has not been moved to a different folder. The fourth line i s the route section, or command line, and it will be discussed last, as it is the most complex. A blank line must follow the command line to signal termination of the command line to the program. The next line is t he title , and it may contain whatever the user desires, as long as there is something there. Another blank line should follow the title line. The next line contains the charge and multiplicit y, respectively. Most computed molecules in the Schanze Group are closed shell singlets or triplets, which would be Cartesian coordinates are given for every atom. The full list is appended to save space in Figure B 5 . This list is generated entirely by the program based on the molecule built by the user. After another blank line, connectivity information is written. T his is generated by for geometry optimizations . One more blank line sepa rates the final region of text. The last area of text specifies basis set information. Simple organic molecules can use a single basis set, which would be specified in the command line; however, molecules with heavier atoms require more specialized basis sets. In order for the

PAGE 144

144 must be specified in the command line. The format of this text first states a string of atomic symbols, followed at the end of the line by a zero (0) and a return. Below the atom string is the basis set, and below that is a series of four (4) asterisks. This is repeated until every atom in the molecule is assigned a basis set. After the last atom or set of atoms are specified, as described above, a blank line is given with the last string of atoms and basis set repeated. s set specification is only necessary when the molecule contains a metal atom. Finally, at least two blank lines must be given to signal the end of the input script to the program. Coming back to the command line , all of the keywords specific to individual jobs are specified here. Working from left to right in the file shown in Figure B 6 , the pound print extra output information, and it must immediately follow the poun d sign (#p) if this is desired. All commands after this point may be in whatever order the user prefers. The n ext comm ands shown are calculation. A convenience of Gaussian 09 is that a molecule may be optimized and have frequencies calculated in a single job. This is neither required nor necessary, but it can be convenient for systems with less than ~150 atoms or ~900 electrons. Specif ic commands to purely frequency jobs will be discussed in the next section. which represent a DFT functional and basis set, respectively. As mentioned earlier, the B3LYP functional tends to be superior to others f or platinum acetylide systems. Also if one is to use multiple basis

PAGE 145

145 sets for a single molecule. the user does not have any metal atoms, but still wants to use multiple basis sets because of other heavy atoms (for The o be used for an The molecule. Typically, molecules subject to symmetry constraints do not o ptimize to a minimum, and require a second optimization job to reach a minimum. Calculations done The last , and it correspo nds to the number of points used to integrate the electron density of the DFT optimized molecule . This specifies that an integration grid which uses more points than the default grid in order to generate a more precise result. While some computational chem ists say that problems for all optimizations that the user wishes to publish. Freque ncy Once a geometry optimization has finished, the next thing the user must do is ensure that the optimized geometry is a minimum on the potential energy surface. Again, this can be done in the same job as the optimization, but for large molecules this is not feasible due to time and hardware constraints of the HPC. Before creating an input file for a frequency job, the user should first create a copy of the checkpoint file generated by the geometry optimization . The copied checkpoint file can either have i ts

PAGE 146

146 name modified, or it can be copied to a new folder. Using new folders is recommended by the author, and will be assumed when discussing subsequent input files, but shown as if the fre quency was run separately from the geometry optimization) file is given in Figure B 7 . Figure B 7 . Example input file for the frequency calculation of (ThPt) 2 BTD . The first thing one should note is that there is no information about atoms, their coordi nates, or their connectivity. All of these data are contained within the checkpoint file taken from the geometry optimization. Using the checkpoint file for these data requires three things. First, the name of the checkpoint file in the frequency input fil e must match the name of the copied checkpoint file generated from the optimization. If the name in the input file is different, the program will not be able to read the file, and read l guess of the specifies that the molecular geometry is also read from the checkpoint file.

PAGE 147

147 Time Dependent Density Functional Theory (TDDFT) The final routine job used in the Sc hanze Group is TDDFT. This job generates vertical electronic transitions for a molecule, as well as the oscillator strengt h and molecular orbital contributions of the transition. A TDDFT job effectively simulates the UV Vis absorption spectrum of a molecul e. An example input file is shown in Figure B 8 . Figure B 8 . Example input file for the TDDFT calculation of (ThPt) 2 BTD . The user will again need to copy the checkpoint file and change its name or copy mber of the users choice. This number tells the program how many vertical transitions to calculate, and it will start with the lowest energy transitions. If only the absorption corresponding to the HOMO LUMO gap is required, then five (5) is an acceptable number. However, the molecules of the Schanze

PAGE 148

148 Group are typically complex and have more featu res in their absorption spectra. A value in the range of 15 30 is usual ly required to see all major electronic transitions at wavelengths longer than 300 nm in the se molecules. Be careful though, as the amount of time to run a TDDFT job gets exponentially larger with the number of transitions calculated. Triplet C omputations General considerations A major area of study in the Schanze Group is the photophysical prope rties of molecules in the triplet state. The main purpose of calculating molecules in the triplet state is to simulate the transient absorption (TA) experiment by running TDDFT on a triplet optimized molecule. This process is almost completely analogous to the process for singlet molecules, with one main difference. As alluded to in Figure B 1, the triplet geometry optimization comes after the singlet geometry optimization. This is because, depending on the starting singlet structure input by the user, the triplet may optimize in a completely different way based how it was built. A triplet optimization should start from the optimized singlet structure so tha t the computational triplet has the best chance to match the structure of the lowest energy experiment al triplet. For this job, it is easiest to have GaussView specify the atomic coordinates and connectivity, rather than trying to use a checkpoint file. The user will open the singlet optimized structure in GaussView and setup a new calculation, but now sp for the charge and multiplicity (no charge/triplet). The subsequent frequency and specification to signal to the program that the molecule is in the triplet st ate. The triplet TDDFT job should ask for at least 25 states to be calculated, as there are many low

PAGE 149

149 oscillator strength transitions that may not allow higher energy transitions with high oscillator strength to be calculated. Estimation of phosphorescence energy For molecules that are not expected to be phosphorescent, Gaussian can be used to approximate this energy. All that is needed are the optimized singlet and optimized triplet. Determination of the phosphorescence energy will be discussed in the Resul ts section. Stability On occasion, the wavefunction of an optimized triplet molecule will not be in the ground state. A molecule with this problem will not show any sign of error until the results of the TDDFT job are examined, where an electronic transiti on of negative energy will be predicted. An output illustrating this problem will be shown in a later section. To remedy this problem, a stability job can be run to find and optimize the lower energy wavefunction for the molecule. An input file for this ty pe of job is shown in Figure B 9 . A checkpoint file from any previous job is needed as a starting point, and the another frequency calculation to ensure that the new s tate is an energy minimum, and then run TDDFT again. Only in extremely rare cases will the stability job not fix the error in the wavefunction on the first try. Population A Gaussian population analysis gives coefficients for every atomic orbital (AO) in a user specified range of molecular orbitals (MO). This is useful for determining percent contributions of atoms or moieties to the overall MO. This job can be run at any time after a molecule is found to be at an energy minimum, and the appropriate checkpo int

PAGE 150

150 file is again required for proper execution of this job. An example is shown in Figure B 10. Figure B 9 . Example input file for a stability job on a platinum aryl compound. Figure B 10 . Example input file for generation of Natural Atomic Orbit als.

PAGE 151

151 Orbital (NBO) module (NBO version 3. 1 93 ) will be used, and that the NBO specific command(s) will be read at the end of the input script. The keyword string to give after the basis set the output so that it can be read by Multiwfn, 94 , 95 which is a program for MO composition analysis. ud basis set specification. Reversing the order of t hese two (putting the NBO string before the basis set info ) will result in immediate failure of the job. Volume If the need arises to calculate an Onsager radius, the volume job will give this value directly . An Onsager radius is one of the variables used in the estimation of the difference in dipole moments between ground and excited state. An example is shown in Figure B 1 1 . d. PBS Script Once the user has created their input file, they must now properly edit their PBS script to ensure all values here match between the two files. The PBS script is a normal text file, just like all the other files dealt with here, but it should be noted that all commands in this file are case sensitive. The HPC wiki page maintains an up to date version of the PBS script for Gaussian 09 job submission at http://wiki.rc.ufl.edu/doc/ Gaussian_P BS . A version that is slightly modified to remove extraneous lines is shown in Figure B 1 2 .

PAGE 152

152 Working from the top down, the first three lines should never need to be changed. The first line specifies that this is a Bourne shell script. It is not important to know what the Bourne shell is. The second line requests that the job is submitted to the Figure B 1 1 . Example input file for a volume job of a gold acetylide compound. Figure B 1 2 . Example PBS script for submitting jobs to the HPC.

PAGE 153

153 default queu e. Other queues are available, but not needed for the jobs run by Schanze Group members. The third line requests that an email is sent to the user when the job begins (b), when the job ends (e), or if the job is aborted (a). The fourth line is the email wh ich notifications will be sent to. The fifth line is a job name specified by the user. This name is included in the notification emails, and is very helpful when running multiple jobs at the same time. The sixth line is the amount of real time (as opposed to CPU time) requested by the user. For smaller organometallic optimization/frequency jobs, 72 hours tends to be sufficient. 144 hours is the maximum allowed, and will need to be invoked for large molecules. Any amount of time below 144 hours is permitted, and as the user becomes more comfortable they will be able to more accurately judge the time required for a job. The seventh line is the number of nodes and processors per node requested, respectively. The number of nodes will always be one due to licensi ng limitations, but the number of processors may range from 1 to 16. Eight (8) processors seem to give the best balance between job time and time spent in the queue for organometallic molecules. While having more processors will allow the job to run faster , it will spend more time in the queue before it starts because it is requesting a larger amount of resources. The final line of #PBS specification is the amount of physical memory (RAM) requested. While there are ways to optimize the amount of RAM per job , so that one is not requesting excess and wasting resources, a safe method is to use 1000MB of RAM per core. It is of utmost importance that the number of processors and amount of RAM requested in the input file and the PBS script is the same in both plac es. Adverse effects are expected if these values do not match.

PAGE 154

154 The next three lines of commands are for the system and do not need to be make it available to the submitted j ob. Gaussian 09, and it is typically the latest revision (revision D.01. as of this writing). The change directory to the place where the job was submitted from running are placed in the same folder as the input file. In the last line, the program to be used, the input file name, and the output file is is case sensitive, so using would immediately return an error. Next, the name of the input file is found between brackets, with a space between the brackets and the file name. The input file name should be intuitive to the user, so that they do no t have to open the file to know is the standard input file type read by Gaussian and GaussView. The output file will ideally have the same name as the input file, b to ensure compatibility with the software. The author recommends that a file is named cule and what job is being run. When the inpu t file and submission script are ready to go, type the command: qsub submission script file name . This will submit the job to run, and the HPC will return a job number. While a job is running, the output file can be read. This will let the user track the p rogress of the job; however, most of the useful data is written only when the job finishes.

PAGE 155

155 Reading and Interpreting Output Files The output file contains a multitude of data, only a few lines of which is useful to the user. An overview of some non essenti al, but good to know, data is shown in Figure B 1 3 . Every output starts with software initialization, copyright and citation information, repetition of the specified input, and molecular structure information. From there, the output moves into data that is specific to individual job types. Figure B 1 3 . Routine information given in every Gaussian output file.

PAGE 156

156 Geometry Optimization The geometry optimization will continue to iterate geometries until a stationary point has been found. Figure B 1 4 shows part of the output when a s tationary point has been found. Atomic forces and displacements make up the criterion for successful optimization. The output will then list internuclear distances, angles, dihedral angles, and orbital energies. N pole moments rangin g from dipole to hexadecapole, broken down into vectors along Cartesian axes, are then listed. This is shown in Figure B 1 5 . Figure B 1 4 . Successful optimization of (ThPt) 2 BTD to a stationary point. Figure B 1 5 . N pole moments for (ThPt) 2 BTD .

PAGE 157

157 Final ly, a very dense set of data is given summarizing the finding s of the job . The tail end of this is shown in Figure B 1 6 . A data point that can be extracted from this information is the total energy of the system. This value is useful in conformational anal ysis, as a larger negative number represents a more stable system. Additionally, if the triplet geometry of a molecule has been optimized, the difference in energies between the singlet and triplet represent the adiabatic triplet energy. 45 This energy difference is a good estimation of the triplet energy/ phosphorescence. The last few lines consist of a quote, CPU time, and sizes of files generated during the job. T he user must be cautious, as the values obtained here are only valid if the molecule is at an energy minimum . Figure B 1 6 . Final lines of the optimization of (ThPt) 2 BTD . Frequency The frequency output is significantly shorter than that of an optimizati on. This is because the intermediate steps are only printed as a single line (two if the SSH window is too narrow). An example of this is shown in Figure B 1 7 . A frequency job computes

PAGE 158

158 vectors for every vibrational frequency of the submitted molecule. As s ome of these vectors converge, there is no need to recompute t hem, and the time between steps Figure B 1 7 . Intermediate steps of a frequency job showing the number of vectors yet to converge. decreases. Eventually, all vectors will converge, and the pr ogram will write every frequency, with every unit vector for every atom. If all of the frequencies are positive, then the molecule is at an energy minimum. However, if any negative (imaginary) frequencies are present, the molecule is not at an energy minim um, and it will need to be reoptimized. Fortunately, the negative frequency distorts the molecule in a way that adjusts the geometry closer to the minimum on the current potential energy surface. GaussView has a feature which allows the user to distort the molecule along a frequency mode, and then save the modified geometry for use in a new geometry optimization. U se of this feature will be described in detail in a later section. Examples of (ThPt) 2 BTD jobs where the molecule is not at an energy minimum, an d where it is at an energy minimum , are shown in Figures B 1 8 and B 1 9 , respectively. Once the user

PAGE 159

159 has confirmed their molecule is at an energy m inimum, they can move on to TDDFT, population, or volume jobs. Figure B 1 8 . Frequency output of (ThPt) 2 BTD with negative frequencies, indicating that the geometry is not at an energy minimum. Figure B 1 9 . Frequency output of (ThPt) 2 BTD with no negative frequencies, indicating that the geometry is at an energy minimum.

PAGE 160

160 TDDFT The TDDFT output contains data i nvolving vertical electronic transitions. A list of excited states is generated, with each state displaying information about the MOs involved in the transition, the CI coefficient(s) corresponding to every pair of ground and excited MOs, the energy of the transition in both electron Volts and nanometers, and the oscillator strength. These data are illustrated in Figure B 20 . Lower energy transitions tend to be straightforward, and typically consist of only one ground state orbital and one excited state orb ital. Higher energy transitions typically involve many orbitals . Figure B 20 . TDDFT output of (ThPt) 2 BTD showing relevant vertical transitions. coefficients are invoked. To o btain the percent contribution of a single transition, square the CI coefficient corresponding to that transition and divide by the sum of the squares

PAGE 161

161 of all CI coefficients for the entire excited state: , where %C i is the percent contribution of the i th term. Finally, to tell if a transition is relevant, its oscillator strength ( f ) must be considered. The oscillator strength is directly proportional to the extin ction coefficient f corresponds to a stronger transition. For conjugated organic and organometallic molecules, it is rare to see the strongest transition be less than 0.1. However, for complexes with shorter conjugation lengths , such as (ThPt) 2 BTD , this will happen, and the user must use their best judgement to pick out relevant transitions. For chromophores with very long conjugation lengths, the value of f is overestimated . Stability The useful part of a stability output looks similar to that of TDDFT. The program looks at excitation energies, and upon convergence, if an excitation energy of negative energy is present, determines that an instability is present in the wavefunction. This is shown in Figure B 2 1 . Upon detection of an instability, the program will reoptimize the wavefunction, and compute excitation energies again. If these excitations are all positive, then the molecule should now be stable and the job is complete. This is shown in Figure B 22. After the stability j ob completes successfully, the user should take this checkpoint file and run a frequency job before rerunning TDDFT. Population The population output for NAOs is very data intensive, as the program prints all coefficients for all orbitals for all atoms for every MO. Thus it is best for the user to copy the output to a local machine (using an SSH client), and analyze the results using the

PAGE 162

162 Figure B 2 1 . Output of a stability job indicating that an instability is present in the wavefunction. Figure B 2 2 . Output of a stability job indicating that the wavefunction is now stable. program Multiwfn. Multiwfn is available for free download at the website https://multiwfn.codeplex.com/ . A very detailed and very hel pful manual is also located at the same website, but an overview of the specific steps the user must take to obtain

PAGE 163

163 population data is presented here. hard drive, but f or efficiency, it is helpful to crea te a separate folder on the first level of the hard drive (ex. C: \ multiwfn) to store the transferred Gaussian output files in. Multiwfn runs in the command prompt, and inputs are purely textual. To get started, open the program, type the input file path ( ex. C: \ multiwfn \ jobname.out ), and hit enter. The program startup screen with a sample input file path is shown in Figure B 2 3 . Figure B 2 3 . Multiwfn startup screen with input file path typed in the last line of text. Upon the file loading successfully, the user will be presented with a number of options with varying functionality to examine the molecule. Please note that while the Gaussian composition analysis module. Because the Gaussian job specified to use the NAO proceed. A few options ar e now available to the user. They, along with the previous two

PAGE 164

164 steps, are shown in Figure B 2 4 . Also note that the MOs are numbered rather than named. The equivalent MO number can be deduced from the TDDFT output. Atoms are also numbered. Opening the stru cture in GaussView and holding the cursor over the desired atom will show the number it has been assigned. The user has a few paths they can take at this point. If the user only needs information about a single orbital, then they can select option 0 , typ e the orbital Figure B 2 4 . Multiwfn orbital composition analysis options. number , and a list of atoms with percentages will be printed. If multiple atoms (ex. a molecule containing multiple metal atoms, or the charge on a donor/acceptor moiety) need to be taken into account, simply add the percentages to get the contributions from the desired group of atoms. If multiple atoms across multiple MOs need to be analyzed, it is usually faster to first define a fragment before running the analysis. To do this

PAGE 165

165 rs to be analyzed. Multiwfn manual for the definitions of these), and the total contribution of the defined fragment. This is depicted in Figure B 2 5 . F igure B 2 5 . Perc ent contributions of a specified fragment to a pair of specified MOs. Volume The last output to be discussed is that of the volume job, shown in Figure B 26. There is not much to this output. If the user scrolls almost all the way to the b ottom, a suggested radius is given. This radius can be used in the Onsager equation. Please note that the value is only accurate to two significant figures.

PAGE 166

166 Figure B 2 6 . Output of a volume job with the suggested radius boxed. Visualizing Results One of the most important aspects of computational chemistry in the Schanze Group is the ability to convey results visually. Specific questions that are best answered visually include: what conformation has the molecule adopted, what do the MOs look like, and wh ere is the charge in an electronic transition starting from/travelling to ? All of these questions can be answered graphically using either GaussView or Chemcraft. While GaussView provides the simplest route to visualizing molecular geometries and MOs, its ability to generate charge difference density (CDD) plots (the answer to the final question listed above) is subpar. Thus the author recommends that whenever structure, MO, or CDD graphics are generated for publication, that Chemcraft is used throughout al l of the images, for consistency. Chemcraft can be downloaded fo r free, as a fully functional 15 0 day trial version, from the website http://chemcraftprog.com/ . Methods for visualization in both programs will be di scussed, as using GaussView is much simpler if the user only needs to see the output without generating publication worthy graphics. GaussView will be discussed first.

PAGE 167

167 GaussView Before starting in on the discussion of visualization, an addition must be mad e to a system file in the users HPC account. To do this, have only one instance of SSH open, and go to the home folder. Alternately, one can sim ply close all instances of SSH and open a new instance, which will start in the home folder. Using the VI editor , open read: # Load modules . The second line should read: module load Gaussian/g09 d01. This will automatically load the Gaussian module to the users SSH shell up on logging in, rather than having to specify it every session. Molecular g eometry If the user only needs to see the molecular geometry of the molecule, they can simply open the output file in GaussView. If the user desires to see intermediate geometries, s Read I ntermediate G window. Once open ed , the view of the molecule can be manipulated in all of the ways one would expect. GaussView does allow image files to be saved; however, these images are sav ed to the HPC cluster, rather than to a local machine . Thus, the Snipping Tool , on a Windows machine, or Grab, on a Macintosh , will suffice. Molecular orbitals Before MOs can be visualized, a formatted checkpoint (fchk) file must be generated. This file is generated from a normal checkpoint file, and contains all of the same information, but it is able to be read by more than just the Gaussian program. A fchk file may be created from any checkpoint file on hand (ex. the checkpoint file from an optimization, frequency, or TDDFT job), but the author recommends using the checkpoint file from the TDDFT job. This is because in the event that a molecule needs

PAGE 168

168 to be reoptimized many times, due to not being at an energy minimum, many optimization files may be create d. Thus it is possible to accidentally format and use the wrong checkpoint file, resulting in invalid data being generated. If a TDDFT checkpoint file is used, the likelihood of this accident is drastically reduced. To format the desired checkpoint file, t he user should be in the same folder as the file in an SSH client. Type the command: formchk jobname.chk jobname.fchk directly into the users SSH shell, which allows them to do this directly from the command line. Once the f chk file has been generated, the user may open it in GaussView. The user In this window the user will see the structure of the molecule, a list of orbitals with occupancies and energi selected, and the HOMO and LUMO highlighted, are shown in Figure B 2 7 . The user T he isovalue controls the size of the generated MOs, and it should remain at 0.02. real gain in resolution. nd click on the desired MOs in the list. By default the HOMO and LUMO are selected. Once all MOs orbitals. The window will be inactive during this time. Users should also n ote the green dotted line between the MOs with negative and positive energy. This line represents the vacuum energy of an electron. Great care should be taken if MOs of higher energy than

PAGE 169

169 Figure B 27. MO Editor window in GaussView displaying (ThPt) 2 BTD with the HOMO and LUMO highlighted for MO generation. this needs to be used, as these orbitals are not always realistic. Once the program finishes rendering the image, the user can click on a box to the right of the energy to select the MO to display. Thi s is shown in Figure B 28, along with the HOMO of (ThPt) 2 BTD . All MOs that are available for viewing will have this box next to them. If the user desires to view additional MOs, they can highlight the new ones and update the list again. Depending on the si ze of the molecule, this process takes approximately one to six minutes per MO. Chemcraft Generating cube files In order to use Chemcraft for visualization, two additional steps must be taken. First, MO cube files (extension: .cub) must be generated from t he formatted checkpoint file. The cube file contains geometry information as well as the data needed to visualize

PAGE 170

170 Figure B 28. MO Editor window in GaussView displaying the HOMO of (ThPt) 2 BTD visualized with an isovalue of 0.02. MOs. To create a cube fi le, move to the same folder as the fchk file that will be used in the SSH shell. Type the command: cubegen 0 mo=XX jobname.fchk XX.cub 80, where XX is the number of the desired MO. Repeat this process for every orbital needed. Finally, transfer the cube fi les to a local machine so they may be used by Chemcraft. Molecular geometry/molecular orbitals Open C hemcraft and open the cube file of the desired MO. The molecule can be manipulated in space just as in GaussView. Before doing anything else, ensure the pr row of buttons directly above the displayed molecule and pointed out in Figure B 29. The perspective mode greatly distorts the molecule. Chemcraft also has functionalit y to change the color scheme and save an image of what it is displaying. To change the

PAGE 171

171 . Adjust the image settings as desired, image is of. the same . These areas are also shown in Figure B 29. Again, the user can manipulate the viewing angle of the molecule before image capture . Figure B 29. Chemcraft program displa ying the HOMO of (ThPt) 2 BTD visualized with an isovalue of 0.02. Useful buttons are labeled. Charge difference density To start generating a charge difference density (CDD) , load all of the needed MOs into the program. The first MO is loaded normally, but each subsequent MO is 29). Before manipulating the cube files, the equation for generating a CDD must be laid out. In its most simple form,

PAGE 172

172 , where ES = Excited State, and GS = Ground State. Thus the CDD can be as simple as: . This is the case for the first computed excited state of (ThPt) 2 BTD (see Figure B 20 ). F or more complex transiti ons, the percent contributions need to be figured in. Using the 16 th excited state of (ThPt) 2 BTD as an example (again, see Figure B 20 ), there are four ground state orbitals and three excited state orbitals, all with different contributions. Please note th at only the percent contributions are additive. If the user tries to add the CI coefficients before they are squared and divided by the sum of the squares, the values will come out incorrectly and the generated CDD will be invalid. Once the percent contrib utions are properly calculated, they are multiplied by the square of the MO : . This is repeated for every MO involved in the CDD. Once all MOs have been weighted according to their percent contribution, the excited state MOs can be s ummed, and the ground state MOs can be subtracted from this sum: . All mathematical orbital manipulations are , and the operations offe red therein , is shown in Figure B 30. It must also be note d that because the MOs themselves have been squared, the isovalue to properly image a CDD must also be squared, resulting in the proper value being 0.0004. Figure B 31 shows the CDD plot of the firs t computed excited state for (ThPt) 2 BTD visualized at an isovalue of 0.0004. Although a CDD resembles a MO, as it should, the fundamental difference comes in how the plot is colored. A CDD shows where electron density is lost (blue coloring) and gained (re d color) in an electronic transition. Once the CDD is generated, double check that all orbitals have been manipulated properly, and save the image.

PAGE 173

173 Figure B 30. the operators displayed. Figure B 31. CDD for the lowest energy computed transition of (ThPt) 2 BTD . Blue lobes indicate electron density being lost, while red lobes indicate electron density being gained. Image visualized at an isovalue of 0.0004.

PAGE 174

174 Considerations for Triplet Graphics Generation While all of the a bove still applies to triplet jobs, the user must keep a few key details in mind when working with triplets. The main concern is in the numbering of the alpha and beta orbitals. Because Gaussian does not pair electrons in a triplet computation, each electr on is assigned its own orbital. When the time comes to extract data about these orbitals, the user will find that they are numbered sequentially, alpha then beta, rather than being assigned a symbol, as the Gaussian manual would have one believe. Thus the user must determine the highest numbered alpha orbital, and add it to the desired beta orbital to get the correct MO number to use in the cubegen utility. The author suggests using the MO editor in GaussView to determine the highest numbered alpha or bital. dropdown list and use the high number there. Additionally, the user has the option of scrolling to the top of the orbital list . An example of this, using the triplet of (ThPt) 2 BTD , is shown in Figure B 32. Once the triplet cube files have been generated, they can be loaded into Chemcraft. It is helpful to retain the numbers of the beta MOs used in the cubegen u tility, as Chemcraft will use these same numbers. Figure B 33 shows the MOs of the highest energy electron in the alpha and beta manifold loaded in Chemcraft, with the numbering again being sequential.

PAGE 175

175 Figure B 32. MO editor window with the highest num bered alpha orbital of (ThPt) 2 BTD pointed out. Singly occupied MOs are also boxed. Figure B 33. The triplet of (ThPt) 2 BTD in Chemcraft with the highest energy MO of both the alpha and beta electrons loaded. Note that the numbering in this program is se quential.

PAGE 176

176 APPENDIX C SUPPORTING INFORMATION FOR CHAPTER 2 X ray Experimental and Data The asymmetric unit of 1 consists of a half Pt complex (located on an inversion center) and one chloroform solvent molecule in general position. The coordinated Cl ligan d and one of the solvent Cl atoms are disordered and each was refined in two parts. In the solvent, the disorder in one Cl atom is accompanied with a smaller disorder in the chloroform C atom but this was not significant enough to be resolved. In the final cycle of refinement, 4584 used to refine 225 parameters and the resulting R 1 , wR 2 and S (goodness of fit) were 1.42 %, 4.02 % and 1.044 , respectively. The asymmetric unit of 2a consists of a half Pt complex and a dichlorom ethane solvent molecule. In the final cycle of refinement, 5134 reflections (of which 4881 are 331 parameters and the resulting R 1 , wR 2 and S (goodness of fit) were 2.50 %, 6.29 % and 1.085 , respectively. The asym metric unit of 2b consists of a half Pt complex and a dichloromethane solvent molecule (located on an inversion center). The C9 cyclohexyl group is disordered and refined in two parts. The methyl group at C31 is also disordered and refined in two parts. Ea occupation factors dependently refined. In the final cycle of refinement, 8708 reflections 422 parameters and the resulting R 1 , wR 2 and S (goodness of fit) were 4.20 %, 10.37 % and 1.036 , respectively. The asymmetric unit of 2c consists of a half Pt complex (located on an inversion center) and a dichloromethane solvent molecule. The ethyl group on C21 is disordered

PAGE 177

177 Table C 1. Summary of crystallographic data 1 ·CHCl 3 2a ·CH 2 Cl 2 2b ·CH 2 Cl 2 2c ·CH 2 Cl 2 empirical formula C32H50Cl8N4Pt C63H68Cl2N4Pt C84H92Cl4N6PtS2 C94H104Cl4N6 Pt fw 969.45 1232.13 1586.65 1654.72 0.71073 1.54178 0.71073 0.71073 cryst syst Monoclinic Monoclinic Monoclinic Triclinic space group P 2 1 / c P 2 1 / c P 2 1 / c P 1 unit cell dimen a [Å] 10.7953(19) 9.0339(4) 8.2552(7) 9.9529(2) b [Å] 14.724(3) 20.7765(9) 27.997(2) 13.1998(2) c [Å] 12. 701(2) 15.4670(6) 16.4264(14) 15.8601(3) 90 90 90 105.207(1) 97.930 101.326(2) 92.455(2) 93.171(1) 90 90 90 93.064(1) V [Å 3 ] 1999.6(6) 2846.5(2) 3793.1(6) 2002.60(6) Z 2 2 2 1 calcd [Mg m 3 ] 1.610 1.438 1.389 1.372 [mm 1 ] 4.071 6.664 2.094 1.936 F (000) 968 1256 1632 856 crystal size, mm 0.21 x 0.20 x 0.19 0.25 x 0.15 x 0.07 0.24 x 0.10 x 0.07 0.18 x 0.14 x 0.08 range [deg] 1.90 to 27.50 5.16 to 68.00 1.45 to 27.50 1.79 to 27.50 index ranges ( h,k,l ) 13,14; 15,19; 16 ,16 10,10; 24,24; 18,18 10,8; 35,36; 21,21 12,12; 17,17; 20,20 no. reflns collected 23816 72981 34768 35226 no. indep reflns 4584 [ R (int) = 0.0248] 5134 [ R (int) = 0.0291] 8708 [ R (int) = 0.0320] 9171 [ R (int) = 0.0292] no. relfns [ I I )] 4224 4 881 6184 9118 max. and min. transmn 0.5118 and 0.4862 0.6454 and 0.2888 0.8724 and 0.6378 0.8605 and 0.7219 no. data/restraints/params 4584/0/225 5134/0/331 8708/36/422 9171/0/474 goodness of fit on F 2 1.044 1.085 1.036 1.040 final R indices [ I I )] R 1 = 0.0142, wR 2 = 0.0402 R 1 = 0.0250, wR 2 = 0.0629 R 1 = 0.0420, wR 2 = 0.1037 R 1 = 0.0253, wR 2 = 0.0633 R indices (all data) R 1 = 0.0158, wR 2 = 0.0416 R 1 = 0.0261, wR 2 = 0.0643 R 1 = 0.0673, wR 2 = 0.1111 R 1 = 0.0256, wR 2 = 0.0636 largest diff peak/hole [e Å 3 ] 0.764 and 1.192 1.083 and 0.928 1.607 and 0.764 1.151 and 1.116 a Data common to all structures: temp of collection, 100(2) K; refinement method: full matrix least squares on F 2 .

PAGE 178

178 and was refined in two parts. In the final cycle of refinement, 91 71 reflections (of which 474 parameters and the resulting R 1 , wR 2 and S (goodness of fit) were 2.53 %, 6.33 % and 1.040 , respectively. All refinements were carried out by minimizing the wR 2 function using F 2 rather than F values. R 1 is calculated to provide a reference to the conventional R value but its function is not minimized. Figure C 1. ORTEP diagram of trans (ICy) 2 PtCl 2 ( 1 ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity.

PAGE 179

179 Figure C 2. ORTEP diagram of trans (ICy) 2 Pt(PE2) 2 ( 2a ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity. Figure C 3. ORTEP diagram o f trans (ICy) 2 Pt(BTF) 2 ( 2b ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity.

PAGE 180

180 Figure C 4. ORTEP diagram of trans (ICy) 2 Pt(DPAF) 2 ( 2c ) with the asymmetric unit labeled. Ellipsoids are at the 50% probability level. Hydrogen atoms are omitted for clarity.

PAGE 181

181 NMR Spectra Figure C 5. 1 H NMR (500 MHz, CDCl 3 ) of 1 . Figure C 6. 13 C NMR (125.7 MHz, CDCl 3 ) of 1 .

PAGE 182

182 Figure C 7. 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2a . Figure C 8. 13 C NMR ( 125.7 MHz, CD 2 Cl 2 ) of 2a .

PAGE 183

183 Figure C 9. 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2b . Figure C 10. Insets for selected peaks from the 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2b .

PAGE 184

184 Figure C 11. 13 C NMR (125.7 MHz, CD 2 Cl 2 ) of 2b . Figure C 12. 1 H NMR (500 MHz, CD 2 Cl 2 ) of 2c . Residual hexanes are present.

PAGE 185

185 Figure C 13. 13 C NMR (125.7 MHz, CD 2 Cl 2 ) of 2c . IR Spectra Figure C 14. FTIR spectrum of 1 .

PAGE 186

186 Figure C 15. FTIR spectrum of 2a . Figu re C 16. FTIR spectrum of 2b .

PAGE 187

187 Figure C 17. FTIR spectrum of 2c . Emission Lifetime Decays Figure C 18. Fluorescence lifetime decay for 2a (light blue) with instrument response function (dark blue).

PAGE 188

188 Figure C 19. Phosphor escence lifetime decay for 2a . Figure C 20. Phosphorescence lifetime decay for 2 b .

PAGE 189

189 Figure C 21. Phosphorescence lifetime decay for 2 c . Transient Absorption Decays Figure C 22. Transient absorption decay for 2a . Initial camera delay: 100ns, came ra delay increment: 15 µs, 100 images averaged per trace, Q switch delay: 380 µs, 180 µJ per pulse.

PAGE 190

190 Figure C 23. Transient absorption decay for 2 b . Initial camera delay: 100ns, camera delay increment: 20 µs, 100 images averaged per trace, Q switch dela y: 38 2 µs, 180 µJ per pulse. Figure C 2 4 . Transient absorption decay for 2 c . Initial camera delay: 100ns, camera delay increment: 20 µs, 100 images averaged per trace, Q switch delay: 3 79 µs, 180 µJ per pulse.

PAGE 191

191 Triplet Triplet Absorption Power Dependen ce Figure C 25. Transient absorption decay profile for 2a . Conditions: 200 µJ per pulse, 380 µJ Q switch delay, 128 scans averaged, decay measured at 584 nm. Figure C 26. Transient absorption decay profiles for 2a at varying laser energies. Conditi ons: 200 µJ 18 mJ per pulse, 380 200 µJ Q switch delay, 128 scans averaged per trace, decays measured at 584 nm.

PAGE 192

192 Computational Studies DFT and TD DFT calculations were performed in the Gaussian 09, revision C.01, suite of programs at the B3LYP level with the 6 31G(d) basis set for nonmetals and the SDD basis set for Pt. Carbene cyclohexane groups were replaced by methyl groups to improve computational efficiency. These truncated complexes are denoted as . The ground state structures were optimized i n the gas phase from idealized starting configurations without symmetry constraints. The optimized structures were confirmed to be minima by the lack of negative modes in a frequency calculation. Structures and orbitals were visualized using Chemcraft Vers ion 1.7, which was also used to generate charge difference density (CDD) plots. Orbitals were visualized at an isosurface value of 0.02, while CDDs were visualized at an isosurface value of 0.0004. Relevant data is shown in Table C 2. Figure C 27. DFT optimized structure of 2a .

PAGE 193

193 Table C 2. Summary of TD DFT c omputations for v erti cal e xcitations of oscillator strength greater than 0.1 for c ompounds . Complex Wavelength nm Orbital Transitions (% contribution) Oscillator Strength f 373.8 334 .2 (2.4%) (97.6%) (97.3%) (2.7%) 2.3830 1.0319 429.3 381.6 316.4 +1 (5.3%) (94.7%) (13.7%) (81.8%) (4.5%) (14.7%) (75.6%) (9.7%) 2.2796 0.7184 0.7139 381.5 344.6 319.53 319.50 315.3 UMO+1 MO+6 (18.3%) (81.7%) (3.6%) (77.8%) (18.6%) (2.3%) (24.3%) (12.9%) (60.5%) (2.2%) (13.1%) (24.4%) (60.3%) (11.1%) (85.7%) (3.2%) 2.6813 0.1267 0.1648 0.1494 0.2115

PAGE 194

194 Figure C 28. 373.8 nm CDD of 2a . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in the electronic transition. Figure C 29. Normalized overlay of the experiment al absorption spectrum of 2a with the TDDFT computed line spectra of . Only vertical excitations with f > 0.1 are shown. Figure C 30. LUMO+1 (orbital 168) of compound 2a .

PAGE 195

195 Figure C 31. LUMO (orbital 16 7 ) of compound 2a . Figure C 32. HOMO (orbital 16 6 ) of compound 2a . Figure C 33. HOMO 1 (orbital 165 ) of compound 2a . Figure C 34. DFT optimized structure of 2 .

PAGE 196

196 Figure C 35. 429.3 nm CDD of 2 . Blue coloring indicates electron density being lost, while red coloring indicates e lectron density being gained in the electronic transition. Figure C 36 . Normalized overlay of the experimental absorption spectrum of 2a with the TDDFT computed line spectra of 2 b . Only vertical excitations with f > 0.1 are shown. Figure C 37. L UMO+1 (orbital 262 ) of compound 2 .

PAGE 197

197 Figure C 38. LUMO (orbital 261 ) of compound 2 . Figure C 39. HOMO (orbital 260 ) of compound 2 . Figure C 40. LUMO+1 (orbital 259 ) of compound 2 . Figure C 41. DFT optimized structure of .

PAGE 198

198 Figure C 42. 381.5 nm CDD of 2c . Blue coloring indicates electron density being lost, while red coloring indicates electron density being gained in the electronic transition. Figure C 43. Normalized overlay of the experimental absorption spectrum of 2c wit h the TDDFT computed line spectra of . Only vertical excitations with f > 0.1 are shown Figure C 44. LUMO+1 (orbital 282) of compound .

PAGE 199

199 Figure C 45. LUMO (orbital 281) of compound . Figure C 46. HOMO (orbital 280) of compound . Figure C 47. HOMO 1 (orbital 279) of compound .

PAGE 200

200 APPENDIX D SUPPORTING INFORMATION FOR CHAPTER 3 NMR Spectra Figure D 1. 1 H NMR (500 MHz, CDCl 3 ) o f COC Br . Figure D 2. 1 3 C NMR ( 12 6 MHz, CDCl 3 ) o f COC Br .

PAGE 201

201 Figure D 3. 1 H NMR (500 MHz, CDC l 3 ) o f CC TMS . Figure D 4. 1 3 C NMR ( 12 6 MHz, CDCl 3 ) o f CC TMS .

PAGE 202

202 Figure D 5. 1 H NMR ( 500 MHz, CD 2 Cl 2 ) o f COC TMS . Figure D 6. 1 3 C NMR ( 12 6 MHz, CD 2 Cl 2 ) o f COC TMS .

PAGE 203

203 Figure D 7. 1 H NMR (500 MHz, CD 2 Cl 2 ) o f CC H . Figure D 8. 1 3 C NMR ( 12 6 M Hz, CD 2 Cl 2 ) o f CC H .

PAGE 204

204 Figure D 9. 1 H NMR ( 500 MHz, CD 2 Cl 2 ) o f COC H . Figure D 10. 1 3 C NMR ( 12 6 MHz, CD 2 Cl 2 ) o f COC H .

PAGE 205

205 Figure D 11. 1 H NMR (500 MHz, CDCl 3 ) o f C PtPE2 . Figure D 12. 1 3 C NMR ( 12 6 MHz, CDCl 3 ) o f C PtPE2 .

PAGE 206

206 Figure D 13. 31 P N MR ( 121 MHz, CDCl 3 ) o f C PtPE2 . Figure D 14. 1 H NMR (500 MHz, CD 2 Cl 2 ) o f CC PtPE2 .

PAGE 207

207 Figure D 15. 1 3 C NMR ( 126 MHz, CD 2 Cl 2 ) o f CC PtPE2 . Figure D 16. 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f CC PtPE2 .

PAGE 208

208 Figure D 17. 1 H NMR (500 MHz, CD 2 Cl 2 ) o f COC PtPE2 . Figure D 18. 1 3 C NMR ( 12 6 MHz, CD 2 Cl 2 ) o f COC PtPE2 .

PAGE 209

209 Figure D 19. 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f COC PtPE2 . Figure D 20. 1 H NMR (500 MHz, CDCl 3 ) o f tol PtPE2 .

PAGE 210

210 Figure D 21. 1 3 C NMR ( 12 6 MHz, CDCl 3 ) o f tol PtPE2 . Figure D 22. 31 P NMR ( 121 MHz, CDCl 3 ) o f tol PtPE2 .

PAGE 211

211 Emission Lifetime Decays Figure D 23 . Phosphorescence lifetime decay for C PtPE2 . Figure D 24 . Phosphorescence lifetime decay for CC PtPE2 .

PAGE 212

212 Figure D 25 . Phosphorescence lifetime decay for COC PtPE2 . Figure D 26 . Phosphorescence lifetime decay for tol PtPE2 .

PAGE 213

213 Transient Absorption Decays Figure D 27 . Normalized t ransient absorption for R PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse . Solutions had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles . Figure D 2 8 . Transient absorption decay for C PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 image s averaged per trace, 180 µJ per pulse . The s olution had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles .

PAGE 214

214 Figure D 2 9 . Transient absorption decay for CC PtPE2 . Initial camera delay: 100ns, camera dela y increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse. The s olution had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles . Figure D 30 . Transient absorption decay for COC PtPE2 . Initial ca mera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse . The s olution had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles .

PAGE 215

215 Figure D 3 1 . Transient absorption d ecay for tol PtPE2 . Initial camera delay: 100ns, camera delay increment: 15 µs, 100 images averaged per trace, 180 µJ per pulse . The s olution had an absorptivity value of 0.58 at 355 nm after five freeze pump thaw cycles . Figure D 3 2 . Representative mo lecular geometry of C PtPE2 during the molecular dynamics simulation process.

PAGE 216

216 Figure D 3 3 . Representative molecular geometry of CC PtPE2 during the molecular dynamics simulation process. Figure D 3 4 . Representative molecular geometry for the prim ary distribution of COC PtPE2 during the molecular dynamics simulation process. Figure D 3 5 . Representative molecular geometry for the secondary distribution of COC PtPE2 during the molecular dynamics simulation process.

PAGE 217

217 APPENDIX E SUPPORTING INFORMAT ION FOR CHAPTER 4 NMR Spectra Figure E 1. 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCBTF . Figure E 2. A) Aromatic region of the 1 H NMR spectra of BTF Pt CCBTF . B) Aliphatic region of the 1 H NMR spectra of BTF Pt CCBTF . A B

PAGE 218

218 Figure E 3. 1 3 C NM R (12 6 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCBTF . Figure E 4. Aromatic region of the 1 3 C NMR spectra of BTF Pt CCBTF .

PAGE 219

219 Figure E 5. 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f BTF Pt CCBTF . Figure E 6. 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCDPAF .

PAGE 220

220 Figure E 7. 1 3 C NMR (12 6 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCDPAF . Figure E 8. Aromatic region of the 1 3 C NMR spectra of DPAF Pt CCDPAF .

PAGE 221

221 Figure E 9 . 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f DPAF Pt CCDPAF . Figure E 10 . 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of BTF Pt CC DPAF .

PAGE 222

222 Figure E 1 1 . Aromatic region of the 1 H NMR spectra of BTF Pt CCDPAF . Figure E 1 2 . 1 3 C NMR (126 MHz, CD 2 Cl 2 ) spectra of BTF Pt CCDPAF .

PAGE 223

223 Figure E 1 3 . Aromatic region of the 1 3 C NMR spectra of BTF Pt CCDPAF . Figure E 1 4 . 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f BTF Pt CCDPAF .

PAGE 224

224 Figure E 1 5 . 1 H NMR (500 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCBTF . Figure E 16 . Aromatic region of the 1 H NMR spectra of DPAF Pt CCBTF .

PAGE 225

225 Figure E 1 7 . 1 3 C NMR (126 MHz, CD 2 Cl 2 ) spectra of DPAF Pt CCBTF . Figure E 1 8 . A romatic region of the 1 3 C NMR spectra of DPAF Pt CCBTF .

PAGE 226

226 Figure E 1 9 . 31 P NMR ( 121 MHz, CD 2 Cl 2 ) o f DPAF Pt CCBTF . Emission Lifetime Decays Figure E 20 . Phosphorescence lifetime decay for BTF Pt CCBTF .

PAGE 227

227 Figure E 21 . Phosphorescence lifetime decay for DPAF Pt CCDPAF . Figure E 22 . Phosphorescence lifetime decay for BTF Pt CCDPAF .

PAGE 228

228 Figure E 23 . Phosphorescence lifetime decay for DPAF Pt CCBTF . Transient Absorption Decays Figure E 24 . Transient absorption decay for B TF Pt CCBTF . Initial camera delay: 100ns, camera delay increment: 1 0 µs, 100 images averaged per trace, 180 µJ per pulse.

PAGE 229

229 Figure E 25 . Transient absorption decay for DPAF Pt CCDPAF . Initial camera delay: 100ns, camera delay in crement: 1 0 µs, 100 images averaged per trace, 180 µJ per pulse. The starred feature is absorption from decomposition product generated during the experiment . Figure E 26 . Transient absorption decay for BTF Pt CCDPAF . Initial c amera delay: 100ns, camera delay increment: 1 0 µs, 100 images averaged per trace, 180 µJ per pulse. *

PAGE 230

230 Figure E 27 . Transient absorption decay for DPAF Pt CCBTF . Initial camera delay: 100ns, camera delay increment: 1 0 µs, 100 imag es averaged per trace, 180 µJ per pulse. Computational Results Monosubstituted Platinum Model Complexes Figure E 28. LUMO of BTFCC Pt . Figure E 29. HOMO of BTFCC Pt .

PAGE 231

231 Figure E 30. LUMO of DPAFCC Pt . Figure E 31. HOMO of DPAFCC P t . Figure E 32. LUMO of BTF Pt . Figure E 33. HOMO of BTF Pt .

PAGE 232

232 Figure E 34. LUMO of DPAF Pt . Figure E 35. HOMO of DPAF Pt . Mixed Ligand Platinum Aryl/Acetylide Complexes Figure E 36. LUMO+1 of BTF Pt CCBTF .

PAGE 233

233 Figure E 37. LUMO of BTF Pt CCBTF . Figure E 38. HOMO of BTF Pt CCBTF . Figure E 39. HOMO 1 of BTF Pt CCBTF . Figure E 40. CDD of the predicted transition at 360 nm for BTF Pt CCBTF .

PAGE 234

234 Figure E 41. LUMO+1 of DPAF Pt CCDPAF . Figure E 4 2. LUMO of DPA F Pt CCDPAF . Figure E 4 3. HO MO of DPAF Pt CCDPAF . Figure E 4 4. HO MO 1 of DPAF Pt CCDPAF .

PAGE 235

235 Figure E 45. CDD of the predicted transition at 350 nm for DPAF Pt CCDPAF . Figure E 46. LUMO+1 of BTF Pt CCDPAF . Figure E 47. LUMO of BTF Pt CCDPAF . Figure E 48. HOMO of BTF Pt CCDPAF .

PAGE 236

236 Figure E 49. HOMO 1 of BTF Pt CCDPAF . Figure E 50. HOMO 2 of BTF Pt CCDPAF . Figure E 51. CDD of the predicted CT transition at 438 nm for BTF Pt CCDPAF . This transition is not observed in the experiment al UV Vis spectrum. Figure E 52. CDD of the predicted transition at 350 nm for BTF Pt CCDPAF .

PAGE 237

237 Figure E 53. LUMO+1 of DPAF Pt CCBTF . Figure E 5 4 . LUMO of DPAF Pt CCBTF . Figure E 5 5 . HOMO of DPAF Pt CCBTF . Figure E 5 6 . HOMO of DPAF Pt CCB TF .

PAGE 238

238 Figure E 5 7 . Normalized UV Vis absorption spectrum of A) BTF Pt CCBTF , B) DPAF Pt CCDPAF , C) BTF Pt CCDPAF , and D) DPAF Pt CCBTF with the TD DFT predicted vertical transitions overlaid. Black vertical lines represent predicted transitions from the monosubstituted model complexes, while colored lines represent predicted transitions from the computationally efficient congeners of the parent compl ex. Vertical transitions are normalized by oscillator strength to the strongest ligand found in each specific complex. * *

PAGE 239

239 LIST OF REFERENCES 1. J. E. Rogers, T. M. Cooper, P. A. Fleitz, D. J. Glass and D. G. McLean, J. Phys. Chem. A. , 20 02, 106 , 10108 10115. 2. J. E. Rogers, J. E. Slagle, D. M. Krein, A. R. Burke, B. C. Hall, A. Fratini, D. G. McLean, P. A. Fleitz, T. M. Cooper, M. Drobizhev, N. S. Makarov, A. Rebane, K. Y. Kim, R. Farley and K. S. Schanze, Inorg. Chem. , 2007, 46 , 6483 64 94. 3. K. Y. Kim, A. H. Shelton, M. Drobizhev, N. Makarov, A. Rebane and K. S. Schanze, J. Phys. Chem. A. , 2010, 114 , 7003 7013. 4. E. Glimsdal, M. Carlsson, B. Eliasson, B. Minaev and M. Lindgren, J. Phys. Chem. A. , 2007, 111 , 244 250. 5. G. J. Zhou and W . Y. Wong, Chem. Soc. Rev. , 2011, 40 , 2541 2566. 6. C. Liao, A. H. Shelton, K. Y. Kim and K. S. Schanze, ACS Appl. Mater. Interfaces , 2011, 3 , 3225 3238. 7. G. G. Dubinina, R. S. Price, K. A. Abboud, G. Wicks, P. Wnuk, Y. Stepanenko, M. Drobizhev, A. Reban e and K. S. Schanze, J. Am. Chem. Soc. , 2012, 134 , 19346 19349. 8. J. S. Wilson, A. S. Dhoot, A. J. A. B. Seeley, M. S. Khan, A. Kohler and R. H. Friend, Nature , 2001, 413 , 828 831. 9. F. Q. Guo, Y. G. Kim, J. R. Reynolds and K. S. Schanze, Chem. Commun. , 2006, 1887 1889. 10. W. Y. Wong and C. L. Ho, Acc. Chem. Res. , 2010, 43 , 1246 1256. 11. J. Mei, K. Ogawa, Y. G. Kim, N. C. Heston, D. J. Arenas, Z. Nasrollahi, T. D. McCarley, D. B. Tanner, J. R. Reynolds and K. S. Schanze, ACS Appl. Mater. Interfaces , 200 9, 1 , 150 161. 12. R. Chakrabarty, P. Mukherjee and P. Stang, Chemical Reviews , 2011, 111 , 6810 6918. 13. E. M. Badley, J. Chatt, R. L. Richards and G. A. Sim, J. Chem. Soc. D , 1969, 1322 1323. 14. S. W. Lai, M. C. W. Chan, K. K. Cheung and C. M. Che, Orga nometallics , 1999, 18 , 3327 3336. 15. Y. Unger, A. Zeller, S. Ahrens and T. Strassner, Chem. Commun. , 2008, 3263 3265.

PAGE 240

240 16. Y. Unger, A. Zeller, M. A. Taige and T. Strassner, Dalton Trans. , 2009, 4786 4794. 17. C. S. Lee, S. Sabiah, J. C. Wang, W. S. Hwang and I. J. B. Lin, Organometallics , 2010, 29 , 286 289. 18. T. Zou, C. N. Lok, Y. M. E. Fung and C. M. Che, Chem. Commun. , 2013, 49 , 5423 5425. 19. K. Li, T. Zou, Y. Chen, X. Guan and C. M. Che, Chem. Eur. J. , 2015, 21 , 7441 7453. 20. Y. Wu, S. X. Wu, H. B. Li, Y. Geng and Z. M. Su, Dalton Trans. , 2011, 40 , 4480 4488. 21. X. Zhang, A. M. Wright, N. J. DeYonker, T. K. Hollis, N. I. Hammer, C. E. Webster and E. J. Valente, Organometallics , 2012, 31 , 1664 1672. 22. M. Tenne, S. Metz, I. Münster, G. Wagenblast an d T. Strassner, Organometallics , 2013, 32 , 6257 6264. 23. A. Tronnier, S. Metz, G. Wagenblast, I. Muenster and T. Strassner, Dalton Trans. , 2014, 43 , 3297 3305. 24. A. Tronnier, A. Poethig, E. Herdtweck and T. Strassner, Organometallics , 2014, 33 , 898 908. 25. A. Tronnier, A. Pöthig, S. Metz, G. Wagenblast, I. Münster and T. Strassner, Inorg. Chem. , 2014, 53 , 6346 6356. 26. A. Tronnier, U. Heinemeyer, S. Metz, G. Wagenblast, I. Muenster and T. Strassner, J. Mater. Chem. C , 2015, 3 , 1680 1693. 27. M. Tenne, S. Metz, G. Wagenblast, I. Munster and T. Strassner, Dalton Trans. , 2015, 44 , 8444 8455. 28. K. Li, G. Cheng, C. Ma, X. Guan, W. M. Kwok, Y. Chen, W. Lu and C. M. Che, Chem. Sci. , 2013, 4 , 2630 2644. 29. T. Fleetham, G. Li, L. Wen and J. Li, Adv. Mater. , 2 014, 26 , 7116 7121. 30. Y. Z. Zhang, J. A. Garg, C. Michelin, T. Fox, O. Blacque and K. Venkatesan, Inorg. Chem. , 2011, 50 , 1220 1228. 31. J. E. Rogers, B. C. Hall, D. C. Hufnagle, J. E. Slagle, A. P. Ault, D. G. McLean, P. A. Fleitz and T. M. Cooper, J. C hem. Phys. , 2005, 122 , 214708. 32. Y. Zhang, J. Clavadetscher, M. Bachmann, O. Blacque and K. Venkatesan, Inorg. Chem. , 2014, 53 , 756 771.

PAGE 241

241 33. Y. Zhang, O. Blacque and K. Venkatesan, Chem. Eur. J. , 2013, 19 , 15689 15701. 34. R. W. Winkel, G. G. Dubinina, K . A. Abboud and K. S. Schanze, Dalton Trans. , 2014, 43 , 17712 17720. 35. K. A. Zachariasse, G. Duveneck and W. Kühnle, Chem. Phys. Lett. , 1985, 113 , 337 343. 36. H. J. Galla and E. Sackmann, BBA Biomembranes , 1974, 339 , 103 115. 37. A. E. C. Redpath and M. A. Winnik, J. Am. Chem. Soc. , 1982, 104 , 5604 5607. 38. T. Hiraga, T. Uchida, N. Kitamura, H. B. Kim, S. Tazuke and T. Yagi, J. Am. Chem. Soc. , 1989, 111 , 7466 7469. 39. T. Abe, T. Itakura, N. Ikeda and K. Shinozaki, Dalton Trans. , 2009, 38 , 711 715. 40 . S. Okajima, P. C. Subudhi and E. C. Lim, J. Chem. Phys. , 1977, 67 , 4611 4615. 41. A. Tsuchida, T. Ikawa, T. Tomie and M. Yamamoto, J. Phys. Chem. , 1995, 99 , 8196 8199. 42. K. E. Brown, W. A. Salamant, L. E. Shoer, R. M. Young and M. R. Wasielewski, J. Ph ys. Chem. Lett. , 2014, 5 , 2588 2593. 43. Y. Tamai, H. Ohkita, J. Shimada, H. Benten, S. Ito, S. Yamanaka, K. Hisada, K. Tani, K. Kubono and T. Shinmyozu, J. Phys. Chem. A. , 2013, 117 , 7776 7785. 44. J. E. Slagle, T. M. Cooper, D. M. Krein, J. E. Rogers, D. G. McLean and A. M. Urbas, Chem. Phys. Lett. , 2007, 447 , 65 68. 45. T. M. Cooper, D. M. Krein, A. R. Burke, D. G. McLean, J. E. Rogers and J. E. Slagle, J. Phys. Chem. A. , 2006, 110 , 13370 13378. 46. R. S. Price, G. Dubinina, G. Wicks, M. Drobizhev, A. Re bane and K. S. Schanze, ACS Appl. Mater. Interfaces , 2015, 7 , 10795 10805. 47. D. Chateau, F. Chaput, C. Lopes, M. Lindgren, C. Brännlund, J. Öhgren, N. Djourelov, P. Nedelec, C. Desroches, B. Eliasson, T. Kindahl, F. Lerouge, C. Andraud and S. Parola, ACS Appl. Mater. Interfaces , 2012, 4 , 2369 2377. 48. R. Vestberg, R. Westlund, A. Eriksson, C. Lopes, M. Carlsson, B. Eliasson, E. Glimsdal, M. Lindgren and E. Malmström, Macromolecules , 2006, 39 , 2238 2246. 49. T. R. Cook and P. J. Stang, Chem. Rev. , 2015, D OI: 10.1021/cr5005666. 50. W. D. Müller, G. Schmidtberg and H. A. Brune, Chem. Ber. , 1985, 118 , 4653 4664.

PAGE 242

242 51. S. Shekhar and J. F. Hartwig, J. Am. Chem. Soc. , 2004, 126 , 13016 13027. 52. T. A. Perera, M. Masjedi and P. R. Sharp, Inorg. Chem. , 2014, 53 , 76 08 7621. 53. C. J. Kuehl, S. D. Huang and P. J. Stang, J. Am. Chem. Soc. , 2001, 123 , 9634 9641. 54. T. Megyes, H. Jude, T. Grósz, I. Bakó, T. Radnai, G. Tárkányi, G. Pálinkás and P. J. Stang, J. Am. Chem. Soc. , 2005, 127 , 10731 10738. 55. A. S. Gundogan, X . Meng, R. W. Winkel and K. S. Schanze, Dalton Trans. , 2015, DOI: 10.1039/C5DT00538H. 56. L. Li, C. L. Ho and W. Y. Wong, J. Organomet. Chem. , 2012, 703 , 43 50. 57. M. Younus, N. J. Long, P. R. Raithby and J. Lewis, J. Organomet. Chem. , 1998, 570 , 55 62. 5 8. H. Zhan, S. Lamare, A. Ng, T. Kenny, H. Guernon, W. D. Harvey and W. Y. Wong, Macromolecules , 2011, 44 , 5155 5167. 59. K. B. Vincent, M. Parthey, D. S. Yufit, M. Kaupp and P. J. Low, Polyhedron , 2015, 86 , 31 42. 60. Z. J. Li, E. Badaeva, D. P. Zhou, J. Bjorgaard, K. D. Glusac, S. Killina and W. F. Sun, J. Phys. Chem. A. , 2012, 116 , 4878 4889. 61. Y. P. Zhou, M. Zhang, Y. H. Li, Q. R. Guan, F. Wang, Z. J. Lin, C. K. Lam, X. L. Feng and H. Y. Chao, Inorg. Chem. , 2012, 51 , 5099 5 109. 62. M. H. Nguyen and J. H. K. Yip, Organometallics , 2011, 30 , 6383 6392. 63. V. W. W. Yam, R. P. L. Tang, K. M. C. Wong and K. K. Cheung, Organometallics , 2001, 20 , 4476 4482. 64. A. Hobbs Shelton, Conjugated Nonlinear Absorbing Organometallic Platinum Complexes , Ph.D. Dissertation, University of Florida, Gainesville, FL, 2011. 65. C. P. Newman, R. J. Deeth, G. J. Clarkson and J. P. Rourke, Organometa llics , 2007, 26 , 6225 6233. 66. C. A. Citadelle, E. Le Nouy, F. Bisaro, A. M. Z. Slawin and C. S. J. Cazin, Dalton Trans. , 2010, 39 , 4489 4491. 67. S. Fantasia, A. Pasini and S. P. Nolan, Dalton Trans. , 2009, 38 , 8107 8110. 68. M. I. Bruce, J. Davy, B. C. Hall, Y. J. van Galen, B. W. Skelton and A. H. White, Appl. Organometal. Chem. , 2002, 16 , 559 568.

PAGE 243

243 69. H. Jacobsen, A. Correa, A. Poater, C. Costabile and L. Cavallo, Coord. Chem. Rev. , 2009, 253 , 687 703. 70. A. Harriman, J. Chem. Soc., Chem. Commun. , 197 7, 777 778. 71. S. L. Murov, I. Carmichael and G. L. Hug, Handbook of Photochemistry , Marcel Dekker Inc., New York, 2nd edn., 1993. 72. M. Mardelli and J. Olmsted, J. Photochem. , 1977, 7 , 277 285. 73. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuse ria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. M ennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. J. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. D aniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, (Revision C.01) , Gaussian, Inc., Wallingford, CT, 2011. 74. P. de Frémont, N. M. Scott, E. D. Stevens, T. Ramnial, O. C. Lightbody, C. L. B. Macdonald, J. A. C. Clybu rne, C. D. Abernethy and S. P. Nolan, Organometallics , 2005, 24 , 6301 6309. 75. H. M. J. Wang and I. J. B. Lin, Organometallics , 1998, 17 , 972 975. 76. SHELXTL6 , Bruker AXS, Madison, WI, 2008. 77. R. T. Farley, Photophysics of Platinum and Iridium Organome tallic Materials: From Molecular Wires to Nonlinear Optics , Ph.D. Dissertation, University of Florida, Gainesville, FL, 2007. 78. M. Sheikbahae, A. A. Said, T. H. Wei, D. J. Hagan and E. W. Vanstryland, IEEE J. Quantum Electron. , 1990, 26 , 760 769. 79. G. A. Andrienko, Chemcraft , http://www.chemcraftprog.com , 2012. 80. E. Glimsdal, P. Norman and M. Lindgren, J. Phys. Chem. A. , 2009, 113 , 11242 11249. 81. T. C. Cheung, K. K. Cheung, S. M. Peng and C. M. Che, Jour nal of the Chemical Society, Dalton Transactions , 1996, DOI: 10.1039/DT9960001645, 1645 1651.

PAGE 244

244 82. N. B. William B. Austin, William J. Kelleghan, Kreisler S. Y. Lau, J. Org. Chem. , 1981, 46 , 2280 2286. 83. G. Bounos, S. Ghosh, A. K. Lee, K. N. Plunkett, K. H. DuBay, J. C. Bolinger, R. Zhang, R. A. Friesner, C. Nuckolls, D. R. Reichman and P. F. Barbara, J Am Chem Soc , 2011, 133 , 10155 10160. 84. Y. T. Wu, T. Labahn, A. Demeter, K. A. Zachariasse and A. de Meijere, Eur. J. Org. Chem. , 2004 , 4483 4491. 85. Y. Kikugawa and Y. Ogawa, Chem. Pharm. Bull. , 1979, 27 , 2405 2410. 86. G. Joshi and S. Adimurthy, Synth. Commun. , 2011, 41 , 720 728. 87. Y. Liu, S. Jiang, K. Glusac, D. H. Powell, D. F. Anderson and K. S. Schanze, J. Am. Chem. Soc. , 2002, 124 , 12412 12413. 88 . K. Y. Kim, A. Haefele, R. T. Farley and K. S. Schanze, unpublished work. 89. T. M. Cooper, D. M. Krein, A. R. Burke, D. G. McLean, J. E. Rogers, J. E. Slagle and P. A. Fleitz, J. Phys. Chem. A. , 2006, 110 , 4369 4375. 90. N. S. Makarov, M. Drobizhev and A . Rebane, Opt. Express , 2008, 16 , 4029 4047. 91. C. J. Cramer, Essentials of Computational Chemistry: Theories and Models , Wiley, 2nd edn., 2004. 92. R. Dennington, T. Keith and J. Millam, GaussView , Semichem Inc., Shawnee Mission, KS, 2009. 93. E. D. Glen dening, A. E. Reed, J. E. Carpenter and F. Weinhold, NBO Version 3.1 . 94. T. Lu and F. Chen, J. Comp. Chem. , 2012, 33 , 580 592. 95. T. Lu and F. Chen, Acta Chim. Sinica , 2011, 69 , 2393 2406.

PAGE 245

245 BIOGRAPHICAL SKETCH Russ Winkel was born and raised in Cedarbur g, WI. It was during his time at Cedarburg High School that his love of chemistry was discovered and nurtured. Upon graduation, five semesters of chemistry had already been taken. Russ then went on to the University of Wisconsin Stevens Point in the fall of 2006. He took every course offered by the Chemistry Department and graduated in the spring of 2010 with an ACS Certified Chemistry Degree with Polymer Option and a minor in mathematics. The fall of 2010 took Russ to the University of Florida, where he started research under the direction of Professor Kirk Schanze. His research entailed the synthesis, photophysical characterization, and computational study of platinum(II) arylacetylide compounds. Russ received his Ph.D. for this work in the summer of 201 5.