Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigating the role of metal-ligand bonding on biological activity of metallotherapeutics Sriskandakumar, Thamayanthy 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_sriskandakumar_thamayanthy.pdf [ 5.28MB ]
Metadata
JSON: 24-1.0073949.json
JSON-LD: 24-1.0073949-ld.json
RDF/XML (Pretty): 24-1.0073949-rdf.xml
RDF/JSON: 24-1.0073949-rdf.json
Turtle: 24-1.0073949-turtle.txt
N-Triples: 24-1.0073949-rdf-ntriples.txt
Original Record: 24-1.0073949-source.json
Full Text
24-1.0073949-fulltext.txt
Citation
24-1.0073949.ris

Full Text

  INVESTIGATING THE ROLE OF METAL-LIGAND BONDING ON BIOLOGICAL ACTIVITY OF METALLOTHERAPEUTICS by Thamayanthy Sriskandakumar B.Sc., The University of Peradeniya, 2001 M.Sc., The University of Connecticut, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2013  © Thamayanthy Sriskandakumar, 2013 ii  ABSTRACT  Defining and understanding the structure activity relationship and mechanisms of action of pharmaceutical agents in a biological environment is vital for their faster and successful clinical development. The mode of activation is often complicated for metalloanticancer drugs due to their wide spectrum of activity and interactions with biomolecules, during and after transport to cancer cells. The research herein describes the investigation of thiolate ligand oxygenation and its potential role in the mechanism of action for a family of “half-sandwich RuII arene” anticancer complexes. X-ray absorption spectroscopy (XAS), in concert with density functional theory (DFT) calculations, has been used to describe the influence of thiolato oxygenation on the nature of the Ru-S bond with the effect on ancillary ligands modifications in the parent thiolato (M-SR) and oxidized sulfenato (M-SOR) and sulfinato (M-SO2R) species respectively. This study suggests that the sulfenato species are most susceptible to ligand exchange, but only via activation by protonation of the terminal oxo group. Perturbations of the sulfenato and sulfinato species can be achieved via either protonation or Lewis acid interaction; however the effect is greater in the sulfenato compared to that of sulfinato. DFT calculations are in agreement with the experimental data. Further studies of the electronic structure of a broader series of OsII and RuII arene complexes have been performed using a similar combined spectroscopic and computational approach. Results from these studies provide important new insights into the chemical and biochemical properties of these complexes. Finally, these studies uncover the correlation between the electronic structure and reactivity that is important and must be considered when investigating the transition metal complexes in medicinal chemistry. iii  PREFACE  The research described in this dissertation is a collaborative work with Prof. Peter J. Sadler, medicinal inorganic group from University of Warwick, Coventry, UK.  Ruthenium and Osmium arene complexes investigated here were prepared by Holm Petzold, Pieter C. A. Bruijnincx, Abraha Habtemariam, Nicolas P. E. Barry, Anna Louisa Noffke and Khatija Bhayat from the Sadler group.  I and my group members collected the X-ray absorption spectroscopic (XAS) data at Stanford Synchrotron Radiation Lightsource (SSRL) and Canadian Light Source (CLS).  XAS data was processed and analysed by myself using SIXpack and BluePrintXAS.  I performed all the density functional theory (DFT) calculations under the direct guidance from Pierre Kennepohl using Amsterdam density functional (ADF) and ORCA computational programs. A version of chapter 3 has been published in Sriskandakumar, T.; Petzold, H.; Bruijnincx, P. C. A.; Habtemariam, A; Sadler, P.; and Kennepohl, P. “Influence of Oxygenation on the Reactivity of Ruthenium-Thiolato Bonds in Arene Anticancer Complexes: Insights from XAS and DFT”. J. Am. Chem. Soc. 2009, 131, 13355-13361.  I did the experimental and theoretical work for this manuscript.  The first draft was written by me and then modified with input from co-authors.  I wrote all the chapters herein with feedback from Prof. Pierre Kennepohl.   iv  TABLE OF CONTENTS ABSTRACT  .......................................................................................................................... ii PREFACE  ......................................................................................................................... iii TABLE OF CONTENTS ............................................................................................................. iv LIST OF TABLES ..................................................................................................................... vi LIST OF FIGURES .................................................................................................................. vii LIST OF SCHEMES ................................................................................................................. xii LIST OF EQUATIONS ............................................................................................................ xiii LIST OF SYMBOLS AND ABBREVIATIONS ............................................................................... xiv ACKNOWLEDGEMENT ......................................................................................................... xvii Chapter 1. INTRODUCTION .................................................................................................. 1 1.1 Scope of the Thesis........................................................................................................... 1 1.2 Metal Complexes as Anticancer Drugs ............................................................................. 2 1.2.1 Platinum Drugs with Anticancer Properties ............................................................. 2 1.2.2 Ruthenium Complexes with Anticancer Activities ................................................... 4 1.2.3 OsII Arene Complexes with Anticancer Properties ................................................. 19 1.3 Cytotoxicity of Ruthenium and Osmium Arene Complexes ........................................... 23 1.4 Activation Mechanisms of Organometallic Anticancer Complexes ............................... 23 1.4.1 Mode of Action: Platinum Anticancer Drugs .......................................................... 24 1.4.2 Mode of Action: Ruthenium Complexes ................................................................ 26 1.4.3 Mode of Action: RuII-arene Complexes (RAPTA and RM175) ................................ 28 1.4.4 Mode of Action: OsII-arene Complexes .................................................................. 31 1.5 Applied Approach ........................................................................................................... 33 Chapter 2. METHODOLOGY ................................................................................................ 35 2.1 X-ray Absorption Spectroscopy (XAS) ............................................................................ 35 2.1.1 Basic Principles of X-ray Absorption Spectroscopy ................................................ 35 2.1.2 Metal K-edge X-ray Absorption Near Edge Structure ............................................ 38 2.1.3 Ligand K-edge X-ray Absorption Near Edge Structure ........................................... 41 2.1.4 Instrumentation and Experimental Setup .............................................................. 47 2.2 Density Functional Theory (DFT) .................................................................................... 56 2.2.1 Fundamentals of DFT ............................................................................................. 57 2.2.2 Utilized DFT Calculations ........................................................................................ 60 Chapter 3. EFFECT OF OXYGENATION ON THE REACTIVITY OF RUTHENIUM (II) ARENE THIOLATO COMPLEXES 65 3.1 Introduction ................................................................................................................... 65 3.2 Experimental and Computational Aspects ..................................................................... 68 3.2.1 Sample Preparation ................................................................................................ 68 3.2.2 XAS Data Collection ................................................................................................ 69 3.2.3 Data Processing and Analysis ................................................................................. 70 3.2.4 DFT Calculations ..................................................................................................... 71 3.3 Results and Analysis ....................................................................................................... 72 3.3.1 Effect of Oxygenation on Ru Metal Center ............................................................ 73 3.3.2 Effect of Oxygenation on S-Ligand ......................................................................... 74 v  3.3.3 DFT Analysis ............................................................................................................ 77 3.3.4 Ancillary Ligand Effects .......................................................................................... 81 3.4 Discussion ....................................................................................................................... 84 3.5 Conclusions .................................................................................................................... 88 Chapter 4. ACTIVATION OF OXIDIZED SULFUR LIGANDS AND ITS IMPACT ON THE REACTIVITY OF RUTHENIUM (II) ARENE COMPLEXES .................................................................................... 91 4.1 Introduction ................................................................................................................... 91 4.2 Experimental and Computational Aspects ..................................................................... 94 4.2.1 Sample Preparation ................................................................................................ 95 4.2.2 XAS Data Collection ................................................................................................ 96 4.2.3 Data Processing and Analysis ................................................................................. 96 4.2.4 DFT Calculations ..................................................................................................... 97 4.3 Results and Analysis ....................................................................................................... 98 4.3.1 Lewis acid Interaction and Brønsted Acid Protonation on Sulfinato ..................... 99 4.3.2 Lewis Acid Interaction and Brønsted Acid Protonation on Sulfenato .................. 104 4.4 Discussion ..................................................................................................................... 109 4.5 Conclusions .................................................................................................................. 110 Chapter 5. SPECTROSCOPIC INVESTIGATION OF RUTHENIUM (II) AND OSMIUM (II) ARENE THIOSEMICARBAZONE COMPLEXES .................................................................................... 112 5.1 Introduction ................................................................................................................. 112 5.2 Experimental and Computational Aspects ................................................................... 115 5.2.1 Experimental Aspects ........................................................................................... 115 5.2.2 DFT Calculations ................................................................................................... 115 5.3 Results and Analysis ..................................................................................................... 116 5.3.1 Effect of TSC Ligand Modification ........................................................................ 116 5.3.2 Effect of the Monodentate Ligand ....................................................................... 121 5.3.3 Effect of Metal (Ru vs Os) in Dimeric Complexes ................................................. 122 5.3.4 Monomer to Dimer formation using UV-Vis Spectroscopy and DFT ................... 123 5.4 Discussion ..................................................................................................................... 126 5.5 Conclusions .................................................................................................................. 127 Chapter 6. CONCLUSIONS AND FUTURE DIRECTIONS ........................................................ 129 REFERENCES ...................................................................................................................... 133 APPENDIX A: BLUEPRINT XAS FITTINGS AND tddft ANALYSIS............................................... 146 APPENDIX B: MOLECULAR ORBITAL DESCRIPTION AND INPUT FILES .................................... 161 APPENDIX C: XAS AND DFT ANALYSIS OF RUII/OSII-ARENE (S-S-) LIGATED COMPLEXES ......... 171 vi  LIST OF TABLES  Table 1.1: Cytotoxicity (IC50) of RuII and OsII complexes toward human ovarian cancer cells, A2780 ..   .......................................................................................................................................... 23 Table 2.1: Electron binding energies of the elements investigated in this thesis ............................ 36 Table 3.1: Experimental peak energies and transition assignments and experimentally determined S3p character in the acceptor orbital as estimated from normalized fit intensities* ....... 75 Table 3.2: Selected bond lengths DFT optimized structures of 3b with (structure 1) and without (structure 2) scalar ZORA relativistic effect are compared with parameters derived from the X-ray structure ........................................................................................................... 78 Table 3.3: Selected bond lengths of DFT optimized structure of 2a with scalar relativistic effect and solvent (structure 1) and without solvent (structure 2) .................................................. 79 Table 4.1: Selected bond length (Å) of 3a, 3aZn2+ and 3aH+ from the DFT analysis ......................... 99 Table 4.2: Experimentally determined S3p character* and calculated charge distribution of acceptor (Ru4dσ* ← S3p) orbital in sulfinato complexes ................................................... 100 Table 4.3: Selected bond length (Å) of 2a, 2aaq, 2a-Zn2+ and 2aH+ from the DFT analysis ............. 105 Table 4.4: DFT-calculated S3p contribution* (Ru4dσ* ← S3p) and the total charge distribution of Kohn-Sham acceptor orbital, Ru4dσ* of 2a, 2aZn2+ and 2aH+ complexes ........................ 105 Table 5.1: S K-edge XAS peak energies and transition assignments with experimentally estimated S3p character in the acceptor orbitals from normalized fit intensities compared with calculated DFT value ...................................................................................................... 120  vii  LIST OF FIGURES  Figure 1.1: Structure of globally approved PtII anticancer drugs in clinical use (a); PtIV prodrug undergoing Phase III clinical evaluation (b) ................................................................... 3 Figure 1.2: Chemical structures of first reported ruthenium complex fac-[RuCl3(NH3)3] with anticancer activity (a) and other RuIII complexes currently in clinical evaluation NAMI-A and KP1019(b) .................................................................................................. 6 Figure 1.3: General structure of some important RuII arene complexes with anticancer property 7 Figure 1.4: The general structure of half-sandwich RuII arene complexes with monodentate ligand (a), bidentate chelating ligand (b) and tethered ligand (c). Reprinted from Bruijnincx et. al.,  Adv. Inorg. Chem. 2009, 61, 1–62., Copyright (2009), with permission from Elsevier. ............................................................................................... 9 Figure 1.5: Arene ligands substituted with RuII-arene complexes developed by Sadler and Dyson.   ...................................................................................................................................... 10 Figure 1.6: Biologically active ethacrynic acid tagged to RAPTA complexes via arene (a) and imidazole (b) ligands (next generation RAPTA complexes) .......................................... 11 Figure 1.7: RuII-arene complexes with monodentate ligands (a) and tethered arene ligand (b) developed by Sadler ..................................................................................................... 12 Figure 1.8: Systematic chelate variation in RuII-arene complexes developed by Sadler. R denotes to arene ........................................................................................................................ 13 Figure 1.9: RuII-arene complexes developed by Keppler with biologically active chelating ligands ..   ...................................................................................................................................... 15 Figure 1.10: Novel RuII-arene anticancer complexes under investigation ....................................... 16 viii  Figure 1.11: RuII-arene complex [(η6-hmb)Ru(en)Z]+ studied for monodentate ligand (Z) substitution reactions. ................................................................................................. 17 Figure 1.12: Most potent OsII-arene anticancer complexes reported to date by Sadler ................. 20 Figure 1.13: Systematic chelating ligand variation in OsII-arene complexes prepared by Sadler. ... 21 Figure 1.14: Novel OsII-arene complex presently under investigation ............................................ 22 Figure 2.1: Energy transition and resulting spectrum of typical K-edge XAS experiment. Spectrum illustrating the key features (pre-edge, edge-jump and ionization edge) ................... 35 Figure 2.2: Mn K-edge XAS of rac-[MnIII(EHPG)]Na, (where EHPG - ethylenebis[(o-hydroxyphenyl) glycine]),  illustrating XANES, EXAFS regions and other spectroscopic features ......... 37 Figure 2.3: Ru K-edge XAS of Ru carbene complexes illustrating the pre-edge intensity decreases with increased centrosymmetry of the complexes. Reprinted from Getty et al., Inorg. Chim. Acta 2008, 361, 1059–1065. Copyright 2008, with permission from Elsevier.178  ...................................................................................................................................... 39 Figure 2.4: Mn K-edge XAS of Mn-transferrin complex showing the edge-jump shifts with Zeff during the photoreduction of the sample with multiple scans. .................................. 40 Figure 2.5: S K-edge XAS of series of sulfur redox species. Reprinted from Pickering et al. FEBS Lett. 1998, 441, 11–14, Copyright (1998) with permission from Elsevier .................... 43 Figure 2.6: Extracting the pre-edge intensity involved in each features of S K-edge XAS obtained from [(η6-hmb)Ru(en)(SOiPr)]+ complex (where hmb – hexamethyl benzene, iPr – isopropyl). ..................................................................................................................... 44 Figure 2.7: A series of S-ligands illustrating a linear relationship between dipole integral factor and S3p←S1s transition energy. Adapted with permission from Sarangi et al. J. Am. Chem. Soc. 2007, 129, 2316-2326. Copyright 2007 American Chemical Society ......... 46 ix  Figure 2.8: A schematic of the most important components of a 3rd generation synchrotron source. Adapted from reference 198 with permission from John Wiley and Sons, Copyright © 2011, John Wiley and Sons.206 ................................................................. 49 Figure 2.9: Schematic layout of beamlines configuration at SSRL ................................................. 50 Figure 2.10: A schematic layout of the XAFS station of the SXRMB beamline at CLS211 .................. 52 Figure 2.11: Schematic diagram for the detection methods used in this thesis work ..................... 53 Figure 3.1: Molecular structures of [(η6-ar)Ru(en)(SOnR)]+ complexes investigated using XAS and DFT.  Complexes 1, 2 and 3 are thiolates, sulfenates and sulfinates respectively.  In series a ar = p-cymene and in series b ar = hmb. In 3a, 3b and 1c R = Ph; in 3a″ R = cyclohexyl; Complexes 3a’ and 3b’ are in silico models only. ...................................... 67 Figure 3.2: Normalized Ru K-edge XANES spectra of complexes of 1a,b - 3a,b.  Inset shows the 1st derivative spectra indicating the edge jump position. ........................................... 73 Figure 3.3: S K-edge spectra of RuII arene A - thiolato 1a and 1b, B - sulfenato 2a and 2b, and C - sulfinato 3a and 3b complexes.  An asterisk (*) denotes features due to thiolato- based impurities in the sulfenato/sulfinato complexes (>5%)..................................... 74 Figure 3.4: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 1a, 2a and 3a. ................. 76 Figure 3.5: Geometry optimized structure of 3b with (structure 1) and without (structure 2) scalar ZORA relativistic effects included during optimization ...................................... 78 Figure 3.6: Geometry optimized structure of 2a with relativistic effect and solvent MeOH (structure 1) and without solvent (structure 2) ........................................................... 78 Figure 3.7: Breakdown of SOnR ligand donor contributions to Ru4dσ* orbital (in %) for 1a→2a→3a’ and for 1b→2b→3b’ (open symbols) including the total ligand charge donation (blue squares) as well as specific contributions from sulfur (green circles), x  oxygen (per atom, red circles), and isopropyl (black circles).  Lines are used to indicate general trends only. ..................................................................................................... 80 Figure 3.8: Normalized S K-edge XAS spectra of 1a (black) and 1c (red).  Inset shows second derivative spectra indicating presence of a shoulder at 2472.7 eV in 1c due to a low- lying φπ* orbital from the aryl moiety. .......................................................................... 82 Figure 3.9: Illustration of hyperconjugation in complex 1c due to S3p character mixing into aryl π* orbitals (ɸπ*) ................................................................................................................. 83 Figure 3.10: S K-edge XAS spectra of 3a″ (R = cyclo hexyl, black) and 3a (R = Ph, red).  Presence of a shoulder at ~2477.5 eV in 3a indicating a low-lying φπ* orbital from the aryl moiety.  ...................................................................................................................................... 84 Figure 3.11: Graphical representation of the Ru4dσ* Kohn Sham orbital from 2a. ........................... 85 Figure 3.12: Comparison of S-O bond distance and orbital mixing in RuII vs CdII. ........................... 86 Figure 4.1: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 3a, 3aaq, 3aZn2+ and 3aH+ ......   .................................................................................................................................... 101 Figure 4.2: Frontier MO diagram of complexes 3a, 3aaq, 3aZn2+ and 3aH+. Only α orbitals are shown but α and β orbitals are identical. Red and green arrows indicate the energy change in the SOσ* and Ru-Sσ* with protonation respectively. Orbitals are normalized to LUMO in each molecule. ........................................................................................ 102 Figure 4.3: (A) Normalized S K-edge XAS of complexes 3a, 3aZn2+, 3aCo2+ and 3aCr3+, (B) XAS and simulated TDDFT of complex 3aZn2+. ......................................................................... 103 Figure 4.4: The Ru-Sσ* orbital of 3aCo2+ and 3aCr3+ showing the interaction of metal 3d with S 3p and Ru 4d orbitals ...................................................................................................... 103 Figure 4.5: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 2a, 2aaq, 2aZn2+ and 2aH+ ......   .................................................................................................................................... 106 xi  Figure 4.6: Frontier MO diagrams for complexes 2a, 2aaq, 2aZn2+ and 2aH+. Only α orbitals are shown but α and β orbitals are identical. Red and green arrows indicate the energy change in the SOσ* and Ru-Sσ* with protonation respectively. Orbitals are normalized to LUMO in each molecule. ........................................................................................ 107 Figure 4.7: (A) S K-edge XAS of 2b and the first scan of 2b solutions at pH 2.5 and pH 7.4; (B) simulated TDDFT of 2b, 2baq and 2bH+. An energy shift of 60.0 eV applied to simulated spectra to calibrate them with experimental data. ................................................... 107 Figure 5.1: RuII/OsII-arene thiosemicarbazone complexes investigated in this Chapter.............. 113 Figure 5.2:  Normalized S K-edge XAS of unmodified (4a) and modified (4b) TSC ligands (a) and complexes 5a-c (b). * denotes the thiolate based impurities. TDDFT spectra of these complexes are provided in Appendix A.3.1. ............................................................... 116 Figure 5.3: MO diagram of the π-systems of Thiosemicarbarzone molecule described using model complex [CH2=NH-CS-NH-N=CH2]+ .................................................................. 118 Figure 5.4: Frontier MO diagram of complexes 5a-c. Only α orbitals are shown (α and β are identical). Red arrow indicates the energy shift from 5a→5b. Orbitals are normalized to LUMO in each complex. ......................................................................................... 119 Figure 5.5: S K-edge XAS of complexes 6a and 6b. Os M3-edge background is not subtracted from the spectrum 6b. ........................................................................................................ 122 Figure 5.6: S K-edge XAS of 5b and 6a .......................................................................................... 123 Figure 5.7: (a) UV-vis spectra of complexes 5b and 6a. (b) simulated spectra of 5b with functionals BP86 and B3LYP. (c) simulated spectra of 6a with functionals BP86 and B3LYP. (d) simulated UV-vis spectra of 5baq compared with 5a spectrum ............... 124 Figure 5.8: MO orbital 324 showing the S-S interaction in 6a. .................................................... 125 xii  LIST OF SCHEMES  Scheme 1.1: Direct Synthesis of sulfenato (2) and sulfinato (3) complexes from thiolato (1) complex, where R’ = p-cym or hmb; R = iPr or Ph ........................................................ 18 Scheme 1.2: Oxidation of RuII-arene thiolato complex in the presence of GSH and O2, where R = p- cym or hmb; R’ = iPr or Ph.......................................................................................... 18 Scheme 1.3: Activation mechanism of cisplatin ............................................................................... 25 Scheme 1.4: Hydrolysis mechanism of RAPTA-C complexes in pure water, where R is p-cym ........ 29 Scheme 1.5: M-Cl bond activation mechanism of RuII/OsII – arene bidentate complexes, where R denotes arene .............................................................................................................. 29 Scheme 1.6: Proposed mechanism for RuII-arene azopyridine, [(η6-ar)Ru(azpy-R)(X)]+, complexes 30 Scheme 1.7: Proposed mechanism for the oxidation of NADH to NAD+ catalysed by OsII-arene iminopyridine complex ................................................................................................. 32 Scheme 1.8: Systematic approach used to investigate the research problems described in this thesis ............................................................................................................................ 33 Scheme 3.1: Proposed pathway for DNA binding by RuII arene Thiolate, 1, [(η6-ar)Ru(en)(SR)]+ complexes 87 Scheme 3.2: Postulated global model for biological mechanism of primary RuII arene, [(η6-ar)Ru(en)(X)]+ complexes ...................................................................................... 88 Scheme 4.1: Protonation of complex 2 followed by hydrolysis under acidic condition (HCl).76 ...... 91 Scheme 4.2: Brønsted acid protonation (a)76 and Lewis acid interaction (b) of RuII arene sulfenato (2) complex, where R = p-cym, R’ = iPr......................................................................... 94 Scheme 4.3: Brønsted acid protonation (a) and Lewis acid interaction (b) of RuII arene sulfinato (3) complex, where R = p-cym, R’ = Ph .............................................................................. 94 Scheme 5.1: Conversion of 6a from 5b in methanol or water........................................................ 123 xiii  LIST OF EQUATIONS  Equation 2.1:  Ground state wave function of metal-sulfur ligand LUMO orbital, Ru4dσ* .............. 44 Equation 2.2:  Intensity of Ru4dσ* ←S1s transition ........................................................................... 45 Equation 2.3:  Relationship between dipole integral (3||1), S3p character (	) and  pre-edge intensity...................................................................................................................... 46 Equation 2.4:  Beer-Lambert law .................................................................................................... 54 Equation 2.5:  Schrödinger equation .............................................................................................. 57 Equation 2.6:  Electron correlation functional, E(ρ), describing the electron density of a system . 58 Equation 3.1:  Approximate Voigt lineshape .................................................................................. 71 xiv  LIST OF SYMBOLS AND ABBREVIATIONS  %  Percent ←  To (indicates transition) ~   Approximately =   Equal; Double bond ±   Plus or minus, Error bar -   Minus; Covalent bond ≃  Almost equal η2  Two-fold hapticity η6  C6 six-fold hapticity A  Ampere (units) Å  Angstrom β   Amount of S3p character in the acceptor orbital acac  Acetylacetate ar  Arene atm  Atmosphere (101 kPa) Azpy  Azopyridine bip  Biphenyl bipy  Bipyridine bz  Benzene dach  1,2- diaminocyclohexane dha  Dihydroanthracene en  Ethylenediamine Eo  Ionisation (threshold) energy eV  Electron volt(s) G  Giga (units) g   Gram (units) gly  Glycine hmb  Hexamethylbenzene hrs  hours I   Intensity I0   Incident X-ray intensity I1  Sample absorption I2  Reference absorption Im  Imidazole Impy  Iminopyridine Ind  Indole iPr  Iso-propyl K   Kelvin/ k   Kilo (units) k   Photoelectron wavenumber log P  log [Partition coefficient] or measure of hydrophobicity/lipophilicity xv  m  Milli (units) p-cym  p-cymene Ph  Phenyl Pico  Picolinate pKa  -log of the acid dissociation constant pm  Pico meter tha  Tetrahydroanthracene Zeff  Effective nuclear charge ζ   Zeta π, π*    Pi (bonding and anti-bonding interaction) σ, σ*    Sigma (bonding and anti-bonding interaction)  ADF  Amsterdam density functional (Software package) ADIT   6-amino-2,3-dimethyl-4-azahex-3-ene-2-thiolate BL   Beamline BluePrint XAS Matlab-based toolbox for the fitting and analysis of XAS data BN  Boron nitride BNCT  Boron neutron capture therapy BP86  Becke and Perdew exchange correlation functional BSA  Bovin serum albumin cGMP  Guanosine-3’,5’-cyclic monophosphate CIF  Crystallographic information file CLS  Canadian light source, Inc. COSMO  Conductor-like screening model DCM  Double crystal monochromator DFT  Density functional theory DNA  Deoxyribonucleic acid EPR  Electron paramagnetic resonance ESI-MS  Electron spray ionization mass spectrometry EXAFS  Extended X-ray absorption fine structure GGA  generalized gradient approximation GSH  Glutathione GSSG  Glutathione disulfide GTO  Gaussian type orbital HK  Hohenberg-Kohn HOMO  Highest occupied molecular orbital IR  Infrared KS  Kohn-Sham LDA  Local density approximation LMCT  Ligand to metal charge transfer LUMO  Lowest unoccupied molecular orbital MLCT  Metal-to-ligand charge transfer MO  Molecular orbital MOAnalyzer Matlab-based toolbox tailored for ADF & ORCA output file xvi  MOLDEN Pre- and post-processing program of molecular and electronic    structure NADH  β-Nicotinamide adenine dinucleotide NAMI-A  New anti-tumour metastasis inhibitor-A NBO  Natural bond orbitals NHase  Nitrile hydratase NMR  Nuclear magnetic resonance ORCA  An Ab initio, DFT and semiempirical electronic structure package PBS  Phosphate-buffered saline PIM-1  Proto-oncogene of the serine-threonine kinase family PTA  1,3,5-triaza-7-phosphoadamantane RAPTA  Ruthenium-arene-PTA RF  Radio frequency RDC  Ruthenium derived compound RNA  Ribonucleic acid ROS  Reactive oxygen species SCF  Self consistent field SIXpack  Graphical user interface for XAS data processing and analysis SSRL  Stanford synchrotron radiation lightsource STOs  Slater-type orbitals STXM  Scanning Transmission X-ray Microscopy SVP  Split valence polarization SXRMB  Soft X-ray micro-characterization beamline TDDFT  Time-dependent density functional theory TEY  Total electron yield TSC  Thiosemicarbazone TZP  Triple zeta with polarisation basis set TZVP  Triple zeta valence polarization UKS  Unrestricted Kohn-Sham UV-vis  Ultraviolet-visible XAFS  X-ray absorption fine structure XANES  X-ray absorption near edge structure XAS  X-ray absorption spectroscopy XPS  X-ray photoelectron spectroscopy XRD  X-ray diffraction ZORA  Zeroth order regular (relativistic) approximation     xvii  ACKNOWLEDGEMENT  Apart from the efforts of myself, the success of this thesis would not have been possible without the encouragement, guidance and support of many others. I would like to take this opportunity to express my gratitude to the people who have been contributed in the successful completion of this dissertation. First and foremost, I would like to express my greatest appreciation to Pierre for being an inspired supervisor. I can’t say thank you enough for his excellent guidance, flexibility, freedom, patience and caring which provided me with an excellent atmosphere for doing research. I feel motivated, encouraged and educated every time I attend his meeting, without his encouragement and enthusiastic approach this research would not have materialized. Above all, Pierre is very understanding and excellent human being which made me feel comfortable for balancing my personal and academic life throughout the years. Besides my advisor, I would like to extent my gratitude to the rest of my examining as well as supervisory committee: Prof. Craig Grapperhaus, Prof. Chris Orvig, Prof. Suzana Straus, Prof. Hongbin Li, Prof. Mark MacLachlan and Prof. Andrew MacFarlane for spending their valuable time to review my thesis, giving insightful comments and encouragement. I am very fortunate to be a part of the Kennepohl group and I really appreciate the support, help, encouragement and friendship I received from past and present group members: Anusha, Mario, Vlad, Kendra, Danielle, Wei, Tulin, Insun, Shirin and Chantal. I am very grateful to Mario, who wrote the Matlab based programs BluePrint XAS and MOAnalyzer which extensively used in this thesis. Mario was always willing to discuss, help and give his best suggestions. I really appreciate his help in my research and in preparation of this thesis. I also want to thank Anusha for being my best friend and companion in the lab until she left, which made my grad school experience more memorable and enjoyable. I was able to mentor a RISE student Joey and other two summer students Caroline and Cynthia. It was my pleasure to guide and help them with experimental and computational work involved in their projects. It was an amazing experience to work with our collaborators Prof. Peter Sadler and his coworkers Drs. Abraha Habtemariam, Holm Petzold, Pieter C. A. Bruijnincx, and Nicolas P. E. Barry, as well as graduate students Anna Louisa Noffke and Khatija Bhayat from University of xviii  Edinburgh and University of Warwick, UK. They provided with all the complexes studied in this thesis and I am extremely grateful for their collaboration. I am most grateful to the staff at the Stanford Synchrotron Radiation Lightsource (SSRL): Drs. Serena DeBeer, Mathew Latimer and Erik Nelson and Allyson Aranda; and Canadian Light Source (CLS): Dr. Yongfeng Hu; for their technical and scientific assistance in experimental setup and data collection. I would also like to express my gratitude to Tulin and Insun for reading my thesis and giving feedback to improve on grammar and format of this document. In addition, I am very grateful to Dr. Jacqueline Cawthray and Anna Louisa Noffke, who expressed interest in reading my thesis and giving valuable comments and suggestions on my scientific writings. I would like to extent my acknowledgement to the staff of UBC Chemistry department; first to the teaching instructors Dana Zendrowski and Sophia Nussbaum for having me as a teaching assistant, which provided me an opportunity to improve my teaching skills while being a graduate student; next to the graduate program secretary Sheri Harbour for her guidance and taking care of all the paper work throughout my graduate life and especially during my maternity leave. Other important person, I owe thanks is Jane Cua for her wonderful support with network and computers in the Chemistry department; following that to Pat Olsthoorn for his help with sample shipments to SSRL and University of Warwick, UK. Last, but by no means least, I thank my husband Kumar and my little daughter Swathi for their personal support and great patience at all times. Kumar and Swathi were always there cheering me up and stood by me through the good times and bad. I would also like to thank my parents and brothers for their support, encouragement and best wishes. The research work described in this thesis was funded by NSERC (Canada). Furthermore, academic, technical and infrastructure support were provided by UBC and University of Warwick. Most of the experimental work was carried out at SSRL, a national user facility operated by Stanford University on behalf of the United States Department of Energy and CLS, a national user facility supported by NSERC Canada, NRC Canada, CIHR, the Province of Saskatchewan, Western Economic Diversification Canada and the University of Saskatchewan. Additionally, the computational studies were performed through the Center for Higher Order Structure Elucidation (CHORSE) of the UBC Chemistry Department. 1  CHAPTER 1. INTRODUCTION 1.1 Scope of the Thesis This thesis provides an account of investigations into the electronic structure of a series of Group 8-arene complexes with anticancer properties. A series of ruthenium (RuII) and osmium (OsII) arene complexes have been studied using a combination of X-ray absorption spectroscopy (XAS) and density functional theory (DFT) calculations. Obtaining a detailed understanding of the mechanism of action of metallo-pharmaceuticals continues to be an ongoing challenge that limits broader development. X-ray absorption spectroscopy (XAS) is an element-specific probe of electronic structure that provides a unique perspective on metal-ligand bonding. This thesis utilizes XAS to provide the first in-depth investigations of RuII-arene anticancer drugs and provides novel insights into the mechanism of action of these complexes. This approach should be more generally applicable in the study of metallodrugs. The first chapter presents a historical perspective on metallo anticancer complexes followed by their present status in human clinical studies. It further describes the ancillary ligand modifications of RuII (section 1.2.2.2) and OsII arene (section 1.2.3) anticancer drugs followed by the currently accepted biological mechanisms of these complexes (section 1.3). Chapter 2 describes the basics of the experimental methodology, data analysis, and computational modelling used to interpret the experimental results. Chapters 3 and 4 lead through the XAS and DFT analysis of a series of RuII-arene thiolato, sulfenato and sulfinato complexes and conclude with a proposal for the biological mechanism of anticancer activity of RuII-arene thiolato complexes. Chapter 5 focuses on spectroscopic studies of RuII-arene complexes with chelating thiosemicarbazone ligands. Finally, the 2  conclusions and key findings of the thesis work are highlighted in Chapter 6. Some recommendations for future efforts in this research area are also presented therein. 1.2 Metal Complexes as Anticancer Drugs Medicinal inorganic chemistry is a relatively young and rapidly developing interdisciplinary research area. While the use of metals as pharmaceuticals has been identified in ancient ayurvedic (India), acupuncture (China), Arabic and Egyptian1–6 medicines, the use of well-defined metal complexes for medicinal purposes began in the 1950s with the discovery of the antibacterial and anticancer properties of ruthenium polypyridyl complexes by Dwyer.7 A decade later, the landmark discovery of the anti-tumour properties of cisplatin by Barnett Rosenberg and coworkers8–10 marked the advent of modern medicinal inorganic chemistry. Since its approval in 1978, cisplatin has continued to be the most widely used chemotherapeutic anticancer agent in clinical use.11,12 1.2.1 Platinum Drugs with Anticancer Properties Platinum anticancer drugs presently play a huge role in cancer chemotherapy. Together with the archetypical cisplatin, the 2nd generation Pt-drugs (Figure 1.1a) carboplatin, designed also by Rosenberg in the 1970s,13 and oxaliplatin, discovered by Yoshinori Kidani in 1978,14 are administered to cancer patients globally. They are presently amongst the most effective chemotherapeutic agents in clinical use. Cisplatin is mainly used in the treatment of testicular cancers, where the survival rate is ~ 100% with early detection. Carboplatin is considered to be a safer analogue to cisplatin and is mostly used in the treatment of ovarian cancer.15 Oxaliplatin is a more recently-developed treatment option, 3  which is used mainly for colorectal cancer in combination with organic drugs.16 In addition, nedaplatin,17 lobaplatin18,19 and heptaplatin20,21 are clinically approved drugs in Japan, China and Korea respectively. The most favourable candidate under development is satraplatin, Figure 1.1b, the only orally active PtIV prodrug due to kinetic inertness, which is in Phase III clinical trials for hormone refractory prostate cancer.16,22,23  Figure 1.1: Structure of globally approved PtII anticancer drugs in clinical use (a); PtIV prodrug undergoing Phase III clinical evaluation (b) Although cisplatin and other platinum drugs are arguably the most successful metallo anticancer drugs today, their clinical efficacy is diminished by a high general toxicity (neuro-, hepato- and nephrotoxicity). This leads to undesirable side-effects, which limit the administered dose. Additionally, it is inactive against a number of cancer cell lines including all metastatic (secondary) cancers. Most importantly, many tumors show intrinsic and acquired uni- and multicellular resistance against these drugs.24 These limitations have made it necessary to search for alternative chemotherapeutic strategies that are capable of overcoming the problems associated with Pt-drugs while still maintaining an equal or greater level of activity and broadening the spectrum of the therapeutic effect. In attempts to find new, metal-based anticancer drugs superior to cisplatin, some thousands of transition metal complexes, besides platinum, have recently been investigated 4  with varying degrees of success for their antitumor activity.23,25 Medicinal chemists are attracted by transition metal complexes as they offer a unique spectrum of reactivity and structural diversity which are not readily available in organic-based drugs. In particular, metallo-anticancer drugs based on ruthenium (RuII/III) appear to be the most promising and rapidly emerging in the last two decades.26 1.2.2 Ruthenium Complexes with Anticancer Activities Ruthenium stands out as an attractive alternative to platinum. The rich and well-established synthetic and coordination chemistry combined with several oxidation states (RuII, RuIII and RuIV) available under physiological conditions make ruthenium compounds well-suited for medicinal applications.27,28 Ruthenium complexes have also been investigated as immunosuppressants, nitric oxide scavengers, antimicrobial agents and antimalarials.29,30 Interestingly, ruthenium is able to mimic iron in binding to various biomolecules, such as human serum albumin and the iron-transport protein transferrin. This stimulates a more effective delivery of ruthenium complexes to cancer cells, because such rapidly dividing cells have a greater demand for iron and exhibit an over expression of transferrin receptors.27 This interaction can be part of the reason why these complexes are less toxic than their platinum counterparts.31 Moreover, their metal–ligand exchange kinetics ranges over many orders of magnitude and can be tuned via ligand variation. This is a key advantage for kinetic stability and drug development.   5  1.2.2.1 Ruthenium (III) Complexes with Anticancer Properties The first report of anticancer activity from a ruthenium complex was published in 1976 by Durig et al. for the simple coordination complex fac-[RuCl3(NH3)3] (Figure 1.2a).  This complex was also found to induce filamentous growth of Escherichia coli at the similar concentration as required for the same effect by cisplatin.32 Subsequently, this compound and analogues (e.g. cis-[Ru(NH3)4Cl2]) were evaluated for their anti-tumour properties by Clarke in 1980.33 Unfortunately, the poor solubility of these complexes prevented further pharmacological application.34 Since then, intense efforts have been made to develop ruthenium compounds as anti-cancer agents and many active agents are currently emerging as potential new drugs.26 To date, two RuIII complexes NAMI-A, trans-[RuIIICl4(DMSO)(Im)](ImH) (where Im = imidazole),  and KP1019, trans-[RuIIICl4(Ind)2](IndH) (where Ind = indole), Figure 1.2b, have entered clinical trials with promising results.23,26,35,36 Although both display octahedral geometry, NAMI-A and KP1019 show distinct antitumor behaviours. NAMI-A, developed by Mestroni, Alessio and coworkers,37,38 was the first ruthenium drug to reach human clinical development in 1999 and currently is in Phase I/II clinical evaluation as an antitumor and more importantly as an antimetastatic agent. Clinical studies showed that it inhibits matrix metalloproteinases and prevents tumor invasion of nearby tissues (antimetastic effect) but appears to lack direct cytotoxic effects (also see section 1.3.2).39 The reduced analogue of NAMI-A was interestingly more active than NAMI-A itself against metastatic growth.40 KP1019, developed by Keppler and coworkers,35 is in Phase II trials for the treatment of colorectal cancer. Phase I clinical studies on KP1019 revealed direct 6  antitumor activity against a wide range of primary explants of human tumors by inducing apoptosis (also see section 1.3.2). Advantageously, it is not associated with dose-limiting toxicity.41,42  Figure 1.2: Chemical structures of first reported ruthenium complex fac-[RuCl3(NH3)3] with anticancer activity (a) and other RuIII complexes currently in clinical evaluation NAMI-A and KP1019(b) 1.2.2.2 Ruthenium (II) Complexes with Anticancer Properties As initially proposed by Clarke,43 RuIII complexes are relatively inert toward ligand substitution, acting principally as prodrugs that undergo “activation by reduction” in vivo to become more labile RuII counterparts which are consequently responsible for the observed cytotoxicity (also see section 1.3.2). By considering this activation mechanism, a large number of RuII complexes have been systematically developed and tested for anticancer activity in cultured tumor cells and animal models.44–48 Among them, the most encouraging RuII arene complexes (RM175, RAPTA-T and DW1/2 - racemic mixture of DW1 and DW2) are discussed in this section. 7   RuII Arene Complexes with Anticancer Activities  Figure 1.3: General structure of some important RuII arene complexes with anticancer property The era of investigating the RuII arene complexes as anticancer drugs has first started with the compound [(η6-C6H6)Ru(metronidazole)Cl2] (metronidazole = 1-β-hydroxyethyl-2- methyl-5-nitroimidazole), Figure 1.3a, reported with anticancer activity by Tocher et al. in 1992.49 The complex exhibits a greater and more selective cytotoxicity than the ligand metronidazole, a widely used antibiotic and antifungal agent; however, further biological studies have not been disclosed.50 Presently, RAPTA (Ruthenium-Arene-PTA, Figure 1.3b), DW1/2 (Dwyer 1/2, 1.3c) and RM175 (Figure 1.3d) are the three families of organometallic half-sandwich RuII-arene complexes drawing increasing attention as antitumor and/or antimetastatic agents.51,52 Interestingly, these three complexes show some similarities in their structure but demonstrate distinctly different antitumor activities. RAPTA: In RAPTA complexes, developed by Dyson and coworkers in 2004,46,53 the piano stool shape is occupied by an arene and by three monomer ligands, one of them being a PTA (1,3,5-triaza-7-phosphoadamantane) ligand. These complexes are weakly cytotoxic but the toxicity is better expressed on tumor cells than in healthy cells due to the biological characteristic of the PTA ligand.46,54,55 In addition to antitumor activity,56 they also exhibit Ru Cl Cl P N N N RAPTA-T Ru Cl H2N NH2 Racemic mixture of DW1/2 (a) (b) Ru* C N N OH NH O O O RM175 + (c) (d) Ru Cl Cl N N NO2 OH [( 6-bz)Ru(metronidazole)Cl2] 8  antimetastatic,46 and antiangiogenic57 properties with negligible side effects. Interestingly, they only show a low cytotoxicity in vitro but a high metastatic activity in vivo.46 Thus, they are inactive against primary tumors but active towards secondary metastasis tumours (also see section 1.3.3 for activation mechanism).38 DW1/2: An intriguing example of RuII half-sandwich complexes which entered the field with a purported different mode of action, was developed by Meggers and coworkers in 2004. It has strong similarity to the natural product Staurosporine, isolated from Streptomyces Staurosporeus bacterium and possess anti-cancer properties.47,58,59 These complexes were studied as organometallic enzyme inhibitors58,60 and the relevant anticancer activity has only been reported recently.61 They specifically bind to PIM-1, proto-oncogene of the serine-threonine kinase instead of DNA.61,62 This is one of the first ruthenium anticancer drugs targeting a signal transduction pathway as a mechanism for anti-tumour activity.63 Therefore, a promising in vivo application is expected for highly selective organometallic protein kinases, PIM-1 inhibitors, such as DW1/2.64 RM175: The group of RM175 complexes, “piano stool” shaped RuII arene diamines developed by Sadler and coworkers in 2001,45 are enormously promising. The success of these RuII complexes is that they simply circumvent the activation by reduction mechanism. The specific design of these complexes, piano stool or half-sandwich, is to intentionally retain its antitumor activity in cisplatin-resistant cancer cells by a unique mode of binding: direct monofunctional coordination to the metal center accompanied with arene intercalation.65,66 In contrast to RAPTA, the RM175 family of complexes shows very high cytotoxicity in vitro as well as in vivo against human ovarian cancer cells (A2780)67,68 with IC50 values comparable to 9  that of carboplatin (6 µM).45 In addition, the more hydrophobic tetrahydroanthracene (tha) derivative [(η6-tha)Ru(en)Cl]+ shows equipotent activity to cisplatin (0.6 µM).67 The above examples show that the anticancer activity of these complexes highly depends on their structural design. Investigations of the ancillary ligand modification and mode of action of these complexes through their electronic structure are a major focus of this thesis. Therefore, these complexes are reviewed in detail below.  Rational Design of RuII-arene Anticancer Complexes    Figure 1.4: The general structure of half-sandwich RuII arene complexes with monodentate ligand (a), bidentate chelating ligand (b) and tethered ligand (c). Reprinted from Bruijnincx et. al.,  Adv. Inorg. Chem. 2009, 61, 1–62., Copyright (2009), with permission from Elsevier. Three different building blocks define the ligand set of the piano stool structure and can be fine-tuned to optimize the activity of these complexes. The anti-tumour and anti- metastatic properties of RuII and OsII η6-bound arene complexes of the type [MII(η6-ar)(X)(Y)(Z)] have recently been explored.29,45,67,69 The three legs of the ‘piano stool’ are designed in three different ways. As shown in Figure 1.4, it can be three monodentate ligands (X, Y and Z) or one bidentate ligand (X-Y) with one monodentate ligand (Z) or one of the ligands can be tethered to the arene ring (e.g. arene-Y). Combinations of the ligand sets 10  lead to a wide range of relatively labile or rather inert complexes depending on the nature of X, Y, Z and the arene ligand. In addition, the three legs can possibly be occupied by one tridentate ligand (X-Y-Z). They are rarely seen as potential candidates in Sadler and/or Dyson complexes due to their inertness towards substitution reactions.70 Arene Ligand:  The “seat” of the “piano stool” is the arene ligand, binding as a η6-electron donor and a π-acceptor. It stabilizes the metal in its +2 oxidation state.71  These complexes possess an excellent amphiphilic character (hydrophobicity provided by the arene ligand is counterbalanced by the hydrophilic metal centre),51 which is believed to increase their interaction with biomolecules. The hydrophobic face provided by the arene plays an important role during the transportation of ruthenium through the cell membranes.28 Most importantly, arene ligands can be excellent DNA intercalators, or form π-π arene-nucleobase stacks - a unique mode of DNA binding involved with RuII arene complexes.66 In addition, the synthetic diversity of the arene ligands is an excellent scaffold for the coupling of organic segments for targeted chemotherapy.27,51  Figure 1.5: Arene ligands substituted with RuII-arene complexes developed by Sadler and Dyson. Sadler and coworkers72 found a correlation between hydrophobicity (measured in log P, where P is partition coefficients in an octanol-water system), cellular uptake and cytotoxicity as follows, from less toxic to more toxic, tha (tetrahydroanthracene) > bip (biphenyl) > p-cym (p-cymene) > bz (benzene), Figure 1.5, in a series of OsII-arene [Os(η6-ar)(4-methyl- 11  picolinate)Cl] complexes. Therefore, log P seems to be a valuable tool for predicting the cytotoxicity of these complexes. On the other hand, the effect of different arene ligands on the cytotoxicity of RAPTA complexes follows the order of toluene > hmb (hexamethylbenzene) > benzene > p-cymene,46 Figure 1.5. Interestingly, there is no correlation reported between log P and cytotoxicity. A next generation of RAPTA complexes that are functionalized via their arene ligand with a biologically active organic group (e.g. ethacrynic acid (Figure 1.6a),73 human serum albumin (HSA) through hydrazone (Figure 1.6b),74 naphthalimide,75) display improved drug efficiency and delivery through multiple modes of activity. Thus, a new horizon has very recently been investigated for targeted chemotherapy.  Figure 1.6: Biologically active ethacrynic acid tagged to RAPTA complexes via arene (a) and imidazole (b) ligands (next generation RAPTA complexes) The above mentioned studies show that the arene ligand can play an important role in the properties and activity of such complexes. It suggests that arene modifications have a major impact on the electronic structure of the resulting complexes and mode of activity.76 Therefore, investigating the arene effect on the electronic structure of these complexes is essential to understand their actual role in their activity and mechanism. The effect of arene substitution (p-cym vs hmb) on the geometry and electronic structure of RuII-arene thiolato 12  complexes is investigated through the metal-ligand covalency of RuII-arene thiolato complexes studied in Chapter 3. Monodentate and Tethered Ligands: RAPTA-T (Figure 1.3b) and its analogues are the best example for a successful candidate belonging to the sub-family of RuII-arene complexes with three monodentate ligands, Figure 1.4a.27,51,77 The effect of ancillary ligand modifications of these complexes has been surveyed recently.48 The water soluble phosphine ligand PTA, highly influences the selectivity of RAPTA complexes towards cancer cells. Any modification on the PTA ligand, however, seems to lead to a loss of selectivity.46  Figure 1.7: RuII-arene complexes with monodentate ligands (a) and tethered arene ligand (b) developed by Sadler On the other hand, other complexes with three monodentate ligands, such as [(η6-ar)Ru(NH3)Cl2] (where ar = p-cym, bip), in Figure 1.7a, are not active towards cancer cells even at high concentrations (100 µM) and undergo rapid hydrolysis followed by loss of the arene ligand.78 For the same reason, the piano-stool RuII-tethered complex, [(η6:η1-C6H5(CH2)2NH2)RuCl2], Figure 1.7b, developed by Sadler et. al. in 2007 is also inactive.79 Structural modifications and activity relationship of monodentate ligands and tethered ligand complexes are beyond the scope of this thesis but have been recently reviewed by Noffke et al.70 Ru Cl Cl NH3 R R = p-cym or bip Ru Cl X N H2 n+ X = Cl or PTA (a) (b) 13  Bidentate Chelating Ligand: The most promising complexes developed by Sadler, including RM175, belong to the sub family of [(η6-ar)Ru(XY)Z]+, where XY is a bidentate chelating ligand and Z is a monodentate ligand. Chelate complexes are generally more stable towards ligand substitution; with an appropriate choice of the labile Z-ligand, the other building blocks in the structure, allow to selectively controlling aquation and substitution reactions.  Thereby the cytotoxicity of this class of complexes can be fine-tuned in cancer cells.67,80  Structure activity relationships showed that the most active complexes contain a stable bidentate N,N-donor ligand, a more hydrophobic arene ligand, and a halide as exchangeable ligand. The perfect example is RM175.45   Figure 1.8: Systematic chelate variation in RuII-arene complexes developed by Sadler. R denotes to arene  The first complex of a bidentate amino-phosphine (P,N-bidentate) chelating ligand, [(η6-bz)Ru(Me2NCH2CH2PPh2-N,P)Cl]+, Figure 1.8a, developed by Sadler showed cytotoxicity towards cancer cells but suffered from poor aqueous solubility.81,82 To increase solubility, a σ-donor N,N-bidentate, ethylenediamine (en), Figure 1.8b, was introduced. Later on, en became one of the most promising bidentate ligands, which leads to the discovery of a series of RuII-arene anticancer complexes with reasonable solubility and significant cytotoxicity (e.g. RM175).45 Potentially, primary and secondary amine N,N-ligands (e.g. en) are excellent 14  hydrogen bond (H-bond) donors and therefore enhance the selectivity for G-binding by H-bond formation to the C6O of the guanine nucleobase.40 A loss of cytotoxicity was observed in an effort to modify the en ligand.84 However, complexes containing 2,2’-bipyridine-3,3’-diol [bipy(OH)2] as chelating ligand showed a dramatic increase in the anticancer activity. In aqueous solution, only the neutral species [(η6-ar)Ru(N,N-bipyOHO)Cl], Figure 1.8c, is actively present (with deprotonated chelating ligand) over a pH range of 2-10 and strongly binds to 9-ethylguanine, a DNA model nucleobase.85 RuII-arene complexes with other bidentate systems, such as N,O- and O,O- chelating ligands, have also been studied and showed moderate cytotoxicity and coordination to guanine (N7 binding). Intriguingly, RuII-arene complexes with a para-substituted phenylazopyridine chelating ligand, and iodo monodentate ligand [(η6-ar)Ru(N,N-azpyR)I]+ (where R = NMe2 or OH), Figure 1.8d, do not undergo activation by hydrolysis but are highly cytotoxic toward cancer cells with an IC50 of 2-6 µM. Iodine is preferred as monodentate ligand over chloride, however the reason is not known. Recently, Keppler and coworkers have developed RuII-arene complexes with biologically active bidentate ligands (indoloquinolines as N,N- ligands,86 hydroxypyr(id)ones as O,O- ligands,87 thiomaltols as S,O- ligands,88 Figure 1.9)  a strategy to selectively enhance the cytotoxicity via a dual-action mechanism. OsII derivatives of indoloquinolines and hydroxypyrones as ligands are also prepared and those analogues seem more active than their Ru analogues. The mode of action is not known for [(η6-arene)RuII(indoloquinoline)Cl]+ complexes.86 A remarkable shift in cytotoxicity was observed with thiomaltol complexes, [(η6-arene)RuII(thiomaltol)Cl] when they were switched from maltol complexes. The shift in 15  cytotoxicity is correlated with the higher stability of the S,O-chelate ligands in aqueous solution.88 However, the later complex [(η6-arene)RuII(hydroxypyrone)Cl]+ undergoes hydrolysis of the Ru-Cl bond and forms aqua [(η6-arene)RuII(hydroxypyrone)OH2]2+ species. Those aqua species directly coordinate to N7 of GMP, in addition to react with amino acids upon loss of the hydroxypyrone ligand.87 Therefore, they show only low potency to inhibit the growth of the cancer cells but high potency to inhibit the CDK2/Cyclin A protein kinase.  Figure 1.9: RuII-arene complexes developed by Keppler with biologically active chelating ligands These studies suggest that bidentate chelating ligands play a critical role in recognition of the complex by biological targets83 besides altering the pKa of the aqua complex.89 Therefore, developing metallo-arene complexes combined with bioactive (such as antibiotic, antifungal, anticancer) chelating ligands is a very recent successful approach to make more efficient anticancer drugs with different modes of activities.86,87,90–93 In that effort, a series of biologically potent sulfur containing bidentate ligands (i.e. thiosemicarbazones, Figure 1.10) hybrid with RuII-arene complexes are currently under investigation by Sadler and coworkers. These complexes do not readily crystallise which limits the structure elucidation by X-ray crystallographic methods. Therefore, the geometric and electronic structure, metal-ligand 16  covalency and structure activity relationship of newly developed RuII-arene thiosemicarbazone complexes are investigated using XAS and DFT in Chapters 5.  Figure 1.10: Novel RuII-arene anticancer complexes under investigation The biological mechanism of water soluble complexes belonging to this class is activated by hydrolysis of the monodentate ligand (i.e. Ru-Z bond as in Figure 1.8).94 Therefore, an appropriate choice of the monodentate ligand is vital to optimize the cytotoxicity of these complexes. The rate and extent of hydrolysis of the Ru-Z bond are highly dependent on the nature of Z ligand. A study which investigated the hydrolysis rate on a series of monodentate ligands [(η6-hmb)Ru(en)Z]+, where Z = Cl, Br, I, N3, (Figure 1.11) shows a good correlation (experimentally Cl- ≈ Br- < I- < N3- as well as theoretically Br- < Cl- < I- < N3-) between hydrolysis rate, hydrolysis equilibrium and cytotoxicity. However, sterically- demanding ligands can significantly affect the rate of aquation.80 The effect of monodentate ligand (I vs Cl) on the electronic structure (covalency) of RuII-arene thiosemicarbazone complexes is investigated using XAS and DFT will be discussed in Chapter 5. Ru S N N H N HH I X R + X = Cl / I R = H / NMe2 17   Figure 1.11: RuII-arene complex [(η6-hmb)Ru(en)Z]+ studied for monodentate ligand (Z) substitution reactions. RuII-arene thiolates: Generally, but not always, complexes that spontaneously hydrolyze are cytotoxic.  Intriguing exceptions to this rule are biologically relevant RuII-arene thiolato, [(η6-arene)Ru(en)SPh]+, complexes prepared by Fuyi Wang from the Sadler group in 2005.80 They do not hydrolyze but undergo faster substitution reaction with guanosine-3’,5’-cyclic monophosphate, GMP. Therefore, the cytotoxicity of these complexes is believed to be activated by oxidation of the thiolato ligand to its sulfenato or sulfinato form, a similar mode of activation also observed for Ru-glutathione adducts, [(η6-bip)Ru(en)(GS-S)] formed during the reaction of RM175 and glutathione, GSH.95 This novel concept forms the counter hypothesis to the “activation by reduction” theory discussed earlier for RuIII complexes and is thus very important to explore in detail. Furthermore, the biochemical studies done by Dr. Holm Petzold show that RuII-arene thiolato (1) complexes, when reacted with one molar equivalent of hydrogen peroxide (H2O2), exclusively yield the corresponding sulfenato (2) complexes, whereas excess H2O2 leads to corresponding sulfinato (3) complexes, Scheme 1.1. The crystallographic data of these complexes shows that the Ru-S bond gets remarkably shorter with oxidation.76,96 A similar trend is observed in other RuII,97,98 NiII 99–101 and CoIII 102 sulfinato complexes; in contrast, elongation is also observed in PtII 103 and NiII 100 sulfenato complexes. Therefore, shortening 18  of the M-S bond upon oxidation is an apparent contradiction to the conventional wisdom that the S-donor strength decreases during the oxidation.101  Scheme 1.1: Direct Synthesis of sulfenato (2) and sulfinato (3) complexes from thiolato (1) complex, where R’ = p-cym or hmb; R = iPr or Ph   Scheme 1.2: Oxidation of RuII-arene thiolato complex in the presence of GSH and O2, where R = p-cym or hmb; R’ = iPr or Ph Interestingly, Holm also noticed that thiolato complexes undergo oxidation reactions in the presence of GSH (Scheme 1.2),104 a biologically important activation reaction. This reaction of RuII-arene thiolato complexes partially elucidates that these complexes would be activated through ligand oxidation. The intra molecular hydrogen bonds within the molecule and inter molecular hydrogen bond presence in the crystallized structures of sulfenato (with solvent molecule -CH3OH) and sulfinato (with neighbouring molecule) might also play a role in the shorter Ru-S bond.76,96 In addition, the key species responsible for the mechanism (i.e. either sulfenato or sulfinato or both), geometric and electronic structures of these complexes 19  and the effect of thiolato ligand oxidation on the metal center was not clear from the analysis of nuclear magnetic resonance (NMR), X-ray crystallography and mass spectrometry. Therefore, a detailed investigation of the electronic and geometric properties of RuII arene thiolato, sulfenato and sulfinato complexes will be discussed in Chapters 3 and 4. 1.2.3 OsII Arene Complexes with Anticancer Properties Following the growing success of ruthenium complexes as potential anticancer drugs, organometallic osmium(II) complexes, [(η6-arene)Os(XY)Z]+, have recently been investigated and show a wide range of in vitro and in vivo cytotoxic activity. Their activity varies all the way from inactive105 to an order of magnitude more potent than either cisplatin or their Ru counterparts with negligible toxicity.91,92,106,107 In the last five years, most of the Ru-complexes as anticancer agents in the front line (i.e. in clinical and preclinical trials) have been reported with their osmium analogues (vide infra). Importantly the osmium analogue of NAMI-A exhibits better in vitro anticancer activity.108 The osmium analogue of RAPTA shows similar cytotoxicity,109 while, in contrast that of KP1019110 and RM175, AFAP51111 reveal more and less potent activity in different cancer cell lines. It appears now that the world leading researchers (Sadler,105,107,112–114 Dyson109,115 and Keppler86,93,110,116,117) in this field explore their new strategies for both RuII and OsII equivalents. However, the aqueous chemistry and pharmacological properties of osmium complexes have been explored in only a limited way. Sadler and coworkers have recently reported a series of OsII-arene complexes with promising new and unusual features.28,72,91,92,105–107,111–114,118–120 Third row transition metals are usually considered to be more inert than those of the 1st and 2nd rows. Consequently, although RuII and its heavier 20  congener OsII arenes are isostructural, their chemical reactivity is different which reflects in the biological activity. The rate of aquation is up to 100x slower (more inert),105,109 and the resulting aqua complexes are ca. 1.5 pKa units more acidic for OsII.114 Therefore, at physiological pH, the osmium complexes can exist largely in the less reactive [Os-OH]+ form114 which might explain why they are negligibly toxic to normal cells.  Figure 1.12: Most potent OsII-arene anticancer complexes reported to date by Sadler To date, two of the most active OsII-arene anticancer complexes are isoelectronic complexes, of azopyridine, [(η6-p-cym)Os(Azpy-NMe2)I]+, FY026,92,106,107 and iminopyridine, [(η6-p-cym)Os(Impy-NMe2)I]+, NSC755639,91 Figure 1.13, which are even more active than the clinical drug cisplatin in vitro. Even though, these bioisosteres show similar activity, interestingly their mode of action is completely different. OsII-arene azopyridine complexes are stable and inert toward aquation. Their cytotoxic activity appears to involve redox mechanisms of GSH by azopyridine ligand.107 Conversely, iminopyridine complexes show a dual mode of action; direct interaction of DNA by hydrolysis and increasing the level of reactive oxygen species (ROS) via oxidizing the NADH to NAD+.91 Therefore, the mode of activation is different in nature and suggests that subtle changes in structure can lead to dramatic changes in chemical reactivity and biological activity. Os I N N N N + (b) NSC755639 Os I N N CH N + (a) FY026 21  1.2.3.1 Rational Design of OsII-arene Complexes with Anticancer Properties Most of the OsII-arene complexes have a half-sandwich “piano-stool” geometry with the general formula [(η6-ar)Os(XY)(Z)]n+, Figure 1.4b, except the OsII analogue of RAPTA developed by Dyson. As with the analogous Ru(II) arene complexes, this scaffold is attractive for drug design as it allows for fine tuning of the three main building blocks, the arene, the XY bidentate chelating ligand and the monodentate leaving group Z, to optimize the cytotoxicity. The nature of the arene can influence cell uptake and DNA intercalation. The lability of the leaving group Z (Cl or I) is important in hydrolysis activation. The bidentate chelating ligand (XY) controls the stability and the ligand exchange rates of these complexes. The effect of different arene ligands and monodentate leaving group (Z) are analogous to that of its Ru counterparts as discussed in Section 1.2.2.2. However, the effect of different bidentate ligands shows some variations in the stability and kinetics of these complexes as discussed here.  Figure 1.13: Systematic chelating ligand variation in OsII-arene complexes prepared by Sadler. The lack of information on the aqueous chemistry of OsII-arene complexes under biologically-relevant conditions has been a major obstacle in current attempts to design 22  anti-cancer drugs. Therefore, current research has focused on the effects of the bidentate ligand XY on the overall reactivity of OsII-arene complexes.  Changing the bidentate ligand from O,O- to N,O- to N,N- , Figure 1.13, decreases the hydrolysis rate and increases the acidity.105,122 Therefore the resulting OsII-arene aqua complexes are more acidic and hydrolyse more slowly. The complexes with acac and gly ligands lack cytotoxicity due to their rapid hydrolysis reaction, whereas picolinate complexes, Figure 1.13c, hydrolyse much more slowly, form kinetically stabile DNA adducts and exhibit equipotent cytotoxicity to cisplatin in human ovarian cancer cells.112,113 Recently reported N,N-chelated ligand iminopyridine complexes, Figure 1.13e, fascinatingly show excellent cytotoxicity even higher than cisplatin via dual mode of action, DNA binding via substitution and induce level of ROS.91  Figure 1.14: Novel OsII-arene complex presently under investigation As discussed above, the newly developed series of OsII-arene organometallic complexes (Figure 1.13) are good examples of how to tune a family of organometallic complexes all the way from a lack of cytotoxic activity to even more potent than the clinical used drugs cisplatin and carboplatin. Rational design of the chelating ligands and structure of the OsII-arene 2+ M S S M N N N N N HN I N NH I 23  complexes play an important role in controlling the reactivity (i.e. hydrolysis, acidity of aqua adducts) and biological behaviour (oxidation of GSH and NADH) of these anticancer drugs. To extend this series further, OsII arene complexes with biologically related S,N- (thiosemicarbazone) ligands (Figure 1.15) are presently under development by Sadler and coworkers. Investigation of their electronic and geometric structures of these newly developed complexes is described in Chapters 5. 1.3 Cytotoxicity of Ruthenium and Osmium Arene Complexes Sadler’s RuII and OsII arene complexes generally show very high cytotoxicity against human ovarian cancer cells (A2780) with IC50 values comparable to cisplatin and carboplatin. The anticancer activity of these complexes is compared with that of carboplatin and cisplatin in Table 1.1. Table 1.1: Cytotoxicity (IC50) of RuII and OsII complexes toward human ovarian cancer cells, A2780 RuII complex IC50/µM OsII complex IC50/µM [(η6-p-cym)Ru(en)Cl]+ 10 ±1.1 67 [(η6-p-cym)Os(pico)Cl]+ 4.2 113 [(η6-bip)Ru(en)Cl]+ 5 ±0.4 67 [(η6-bip)Os(pico)Cl]+ 4.5 113 [(η6-tha)Ru(en)Cl]+ 0.5 ±0.1 67,80 [(η6-hmb)Ru(en)SPh]+ 22 80 Carboplatin 6 ±0.7 67,80 Cisplatin 0.6 ±0.06 67,80  1.4 Activation Mechanisms of Organometallic Anticancer Complexes Organometallic anticancer drugs are often prodrugs, i.e. precursors to medicinally-active molecules that must undergo either chemical or enzymatic transformation in vivo to generate the active drug.123 Metal prodrugs are commonly activated by ligand 24  substitution, ligand oxidation, a change in oxidation state (redox process), a photochemical process, or combinations of these.28 Unlike most organic-based drugs, the metal centre in organometallic chemotherapeutics allows for a wide range of possible coordination numbers, geometries, accessible redox states, and thermodynamic and kinetic properties.124 Along with this greater landscape of chemical possibilities provided by the metal centre, the intrinsic properties of the different building blocks (cationic metal ion and ligands) offer different modes of interaction. Combinations of these features provide a wide spectrum of reactivity and mechanisms by which to interact with biological molecules and DNA. Anticancer properties are achieved via inhibiting the cell growth either by direct interaction with DNA or indirectly by inhibition of DNA synthesis and replication, and cell death (apoptosis). The cytotoxic effects of complexes are experienced by the rapidly dividing tumor cells, but this mode of action invariably affects the normal cells. Therefore, selective cytotoxicity and (multi)drug resistance are currently the biggest challenges in cancer chemotherapy. Promisingly, new candidates have advanced to clinical trials to possibly overcome these problems.26 Accepted and proposed mechanisms for metallo anticancer complexes and some challenges in investigating those mechanisms are discussed in this section. 1.4.1 Mode of Action: Platinum Anticancer Drugs Cisplatin is a prototypical example of a prodrug that has been extensively studied. The prodrug is activated through hydrolysis of the Pt-Cl bonds. The chloride ligands remain attached to the Pt(II) centre in the blood stream, where the Cl- ion concentration is relatively 25  high (≃100 mM).15,16 Inside cancer cells, a decreased Cl- ion concentration (4–20 mM) favours drug aquation with the loss of one or both of the chloride ligands, Scheme 1.3.15   Scheme 1.3: Activation mechanism of cisplatin The aquated active cisplatin adducts binds to DNA at the N7 position of guanine, and to a lesser extent adenine, through the formation of a covalent bond with the nitrogen atom.15 Ring closure through the formation of a second DNA bond forms intrastrand and interstrand crosslink bis-adducts. These adducts cause distortions in DNA, including unwinding and bending, and activate various signal-transduction pathways. However, the pathway(s) from Pt-DNA binding to apoptosis remains incompletely elucidated. Carboplatin was designed to improve the clinical performance of cisplatin.13 Aquation of carboplatin yields the same active component as cisplatin and forms the same DNA adducts in a slower rate.125,126 Therefore, it is considered as a clinically safer analogue to cisplatin, although it has its own side effect and showed similar drug resistance.127 Oxaliplatin is the first approved drug that was capable of overcoming cisplatin/carboplatin resistance.15,128 The specific ligand set of oxaliplatin, bidentate ligand 26  1,2-diaminocyclohexane (dach) combined with another bidentate oxalate ligand, appears to play a crucial role in the non-cross-resistant character with cisplatin. Oxaliplatin predominantly forms oxliplatin-induced-1,2-intrastrand crosslink DNA (GpG intrastrand)129,130 adduct with the bulky hydrophobic dach ligand pointing into the DNA major groove, which prevents binding of DNA repair proteins.129 The oxalate ligand also greatly reduces the severity of the side effects of the drug compared with cisplatin.131 1.4.2 Mode of Action: Ruthenium Complexes Initially ruthenium anticancer drugs were hypothesized to exert their anticancer activity by direct interaction with DNA as observed with platinum. However, it is now evident that ruthenium shows a number of differences with platinum.132 Prominently, ruthenium appears to accumulate favourably in neoplastic masses rather than normal tissues perhaps because it is transported by transferrin into tumors.133 Neoplastic tissues are rich in transferrin receptor therefore transferrin–ruthenium complexes can be actively transported, liberated and then internalized by the tumor.134 Intriguingly, some ruthenium anticancer complexes demonstrate greater antimetastatic properties in addition to anticancer activity.46,111,135 This antimetastatic effect is likely to be mediated  through interactions with extracellular matrix components.136,137 A study on biological strengths of Ru and Os analogues suggests that ruthenium seems to play a key role in antimetastatic activity;111 however, further studies should be conducted to confirm this property. Due to the above mentioned properties, ruthenium is predicted to show patterns of antitumor activity and clinical toxicity that are unique in nature and distinct from those of platinum. 27  1.4.2.1 RuIII complexes (NAMI-A and KP1019)  Octahedral RuIII-complexes are relatively inert towards ligand substitution. Therefore, they are essentially prodrugs and remain in their relatively inactive RuIII state until they reach the tumor cells. In a tumor, under lower oxygen content and acidic extracellular pH, the appreciable amounts of cellular reducing agents such as glutathione and ascorbic acid provide physiologically accessible RuIII/RuII redox potential to selectively reduce the drugs from inactive RuIII to more reactive RuII species.138 This is called the “activation-by-reduction” mechanism.33 These RuIII complexes therefore show selective cytotoxicity towards tumor and even hypoxic tumors that are more likely to be resistant to chemotherapy and radiation.139 The mechanism of action of NAMI-A is recognized by substitution of Cl- ligand140,141 followed by multiple interactions with biological molecules outside and inside the cells.20,30 This leads to a direct coordination to nucleic acids and tumor cell DNA.143 However, molecular level investigation of these interactions is not well documented. The mode of activation of KP1019 is initiated with its binding to transferrin by substitution of Cl- ligand.144 Initially, it was suggested that it would undergo a redox reaction followed by apoptosis via the mitochondrial pathway,138,145 but this suggestion is not well supported by recent experimental findings and it appears that the adducts of both species (RuIII and RuII) are found to be active.144,146 However, formation of ROS species and DNA interactions are also reported.147,148 Unfortunately, KP1019 also has no clear molecular level mechanism of action and/or on the main molecular target responsible for the anticancer activity.26 28  Therefore, it appears that lack of understanding of not only their structure or behaviour in biological environment but also their exact molecular mechanisms of action, is a major disadvantage in their further clinical developments. 1.4.3 Mode of Action: RuII-arene Complexes (RAPTA and RM175) Half-sandwich, pseudo-octahedral RuII/OsII-arene complexes of the three main groups do not possess similarities in their activity (as discussed in Section 1.2.2.2) or their mechanism. The mode of activity of these complexes is unique in their nature and different from that of NAMI-A or KP1019. 1.4.3.1 RAPTA Family complexes Studies on the mode of action of these RAPTA-type complexes are limited.38 However, available studies suggested that the major intracellular targets of RAPTA compounds appear to be proteins, although DNA binding is also observed.53 In the RAPTA-C complex, aquation seems to be the activation step for their cytotoxicity, Scheme 1.4. Hydrolysis of this complex is suppressed in the blood plasma where Cl- concentration is about 100 mM, but occurs once it enters the cell cytoplasm where Cl- concentration is much lower.51 Then the labile aqua species undergo substitution by biomolecules.51 However, the hydrolysis-resistant (RAPTA-oxalato and RAPTA-diketonato) complexes, which are also equally cytotoxic, undergo a different mode of mechanism called “ring slippage” of the arene to create a vacant coordination site.149,150 29   Scheme 1.4: Hydrolysis mechanism of RAPTA-C complexes in pure water, where R is p-cym 1.4.3.2 Sadler’s RM175-type complexes The biological activity of these complexes appears to be initiated by Ru-Z (Z = Cl, I or N3) ligand hydrolysis, Scheme 1.5.89,94 The hydrolysis of the prodrug, [(η6-ar)Ru(en)Cl]+, is suppressed outside cells by the high concentration of chloride ions, a common suppression mechanism applied for Pt and Ru prodrugs. However, in the nucleus and cytoplasm, where the Cl- concentration is lower, the hydrolysed species are predominant.83 The aqua species, [(η6-ar)Ru(en)OH2]2+, is more reactive and it appears that the predominant species at that particular pH might influence the cytotoxicity. The pKa value of the corresponding aqua complexes of this series is ≃ 8.94 Therefore, at physiological pH 7.4, the more active aqua adducts should prevail.70 The rate and extent of hydrolysis is highly depend on the nature of the monodentate ligand and sterically-hindered ligands can significantly affect the rate of aquation.80    Scheme 1.5: M-Cl bond activation mechanism of RuII/OsII – arene bidentate complexes, where R denotes arene Ru Cl H2N NH2 R + +H2O -Cl- Ru H2O H2N NH2 R 2+ Ru HO H2N NH2 R + +H2O -H3O + 30  The hydrolysis reaction appears to be very similar to that of cisplatin but these complexes are active toward cisplatin-resistant cells. This suggests that the mode of interaction to DNA is different.151,152 In fact, the [Ru-ar-en] species strongly interact with DNA through a dual mode of binding: monofunctional coordination to N7 of guanine,66 which is different from the bifunctional adduct of cisplatin,151 accompanied by arene intercalation in addition to strong stereospecific hydrogen bonding (between an en NH group and the C6 carbonyl group of the guanine system).65,153 The intercalation ability, hydrophobic π-π arene-nucleobase stacking interaction, increases with hydrophobicity (bz < p-cym <bip < tha)66 and length (p-terp > o, m-terp)154 of the arene ring. Further, bioanalytical studies on these interactions show that the [Ru-ar-en] species bind preferentially to guanine residues in double-helical DNA.151  Scheme 1.6: Proposed mechanism for RuII-arene azopyridine, [(η6-ar)Ru(azpy-R)(X)]+, complexes Intriguingly, an unusual mechanism was recently reported for azopyridine complexes, [(η6-ar)Ru(azpy-R)(X)]+ where X = Cl or I, which are substitution inert but highly cytotoxic. The 31  iodido complex is equipotent to cisplatin in ovarian and lung cancer cells. In contrast, the chlorido complexes are an order of magnitude less potent and less inert. They slowly loose the arene in aqueous solution. The role of the iodide ligand in their cytotoxicity is not known. However, cytotoxicity appears to involve an increase in ROS, formed during activation by reduction of the ligand by GSH, Scheme 1.6. The reduction potential for azopyridine ligands is high but it is biologically achievable when the ligand coordinates to RuII.155 However, the effect of the metal center has not been investigated. 1.4.4 Mode of Action: OsII-arene Complexes The activation mechanism of osmium analogues, prepared by Dyson, Sadler and Keppler, appears to follow a very similar mechanism to that proposed and accepted (hydrolysis followed by DNA binding or protein interaction) for their ruthenium analogues.109,116,122,156 However the Os complex AFAP51 binds to DNA in a different mode of interaction (unwinding without bending) which induces non-repairable damage to DNA by causing a large degree of DNA unwinding (21-27°)156 compared to that of RuII complexes (7-14°)151 or cisplatin (6° and 13° for mono- and bi-functional respectively).157 A novel mechanism has recently been proposed for OsII-arene iminopyridine complexes that are more potent than cisplatin with nanomolar IC50 values. They undergo ligand substitution in aqueous solution to bind DNA, in the meantime controlling the modulation of redox pathways by catalysing the oxidation of NADH → NAD+ in cancer cells (Scheme 1.7). Exceptionally, the driving force for both mechanisms comes from the activation of the Os-Z bond, where Z = I/Cl.91 These studies further suggest that OsII-arene complexes own novel mechanisms of action that can be finely tuned through the choice of the ligands. 32   Scheme 1.7: Proposed mechanism for the oxidation of NADH to NAD+ catalysed by OsII-arene iminopyridine complex The mechanism described in this section clearly indicates that these organometallic complexes exhibit a wide spectrum of anticancer activity and mode of actions with biological molecules which are very unique in their nature. It is evident that a more fundamental understanding of the mechanism and the structure activity relationship of these metallodrugs is essential for a more rational design of alternatives. That should help in finding ways to reduce their side effects and also to speed up the clinical drug development process. With this goal in mind, this thesis uses an interdisciplinary approach, XAS in concert with DFT, to investigate the electronic structure of the RuII and OsII arene complexes with anticancer properties. The approach is described in detail in section 1.4. 33  1.5 Applied Approach  Scheme 1.8: Systematic approach used to investigate the research problems described in this thesis   As illustrated in Scheme 1.8, the work presented in this thesis makes use of XAS as a primary experimental technique. XAS is an element-specific, powerful probe to study metal-ligand covalency. The edge jump is sensitive to Zeff, the effective nuclear charge, and therefore well suited to resolve the research problems described in this thesis. The main experiments carried out are: (a) metal K-edge XAS, to examine the Zeff on the metal center; (b) ligand (S) K-edge XAS, to examine the Ru-S and Os-S covalency and the oxidation state. The basic principles of these techniques, experimental set up and the data analysis are discussed in Section 2.1. 34  All the Ru and Os complexes investigated herein were prepared and characterized for purity by members of Peter Sadler’s group in the United Kingdom.* Interpretation of XAS data was obtained through extensive multistep computational processing of the data combined with theoretical approaches to extract the accurate results. Several DFT calculations were performed to support the experimental XAS results.  Details of these calculations are described in Section 2.2. In recent years, XAS studies, combined with other bioanalytical techniques, have effectively been used to investigate and corroborate the postulated mechanism of Pt, Ru and other metallodrugs.158,159 An interesting study on octahedral PtIV compounds in cultured cancer cells shows that PtIV undergo activation by reduction to form active PtII species and the redox speciation in cancer cells.160,161 Another study characterized the RuIII antimetastatic agent NAMI-A with bovin serum albumin (BSA) shows that Cl- ligand is substituted upon binding the protein; however, the metal center remains its initial oxidation state (RuIII).140,141 An investigation of the electronic structure of a series of RuIII and RuII complexes related to NAMI-A and KP1019 was accomplished with the help of XAS and DFT calculations.162 A very recent study on KP1019 shows that the mode of activation of the complex appears to involve with two different active species: (i) RuIII species with one Cl-substituted (ii) RuII species with three Cl- ligand substituted; formed upon binding to transferrin and GSH. Both species potentially target the cancer cells.144 However further studies need to be conducted to identify the key species accountable for the anticancer activity of KP1019.  * All the complexes studied in this thesis were provided by members of Peter Sadler’s group from University of Edinburgh and University of Warwick, United Kingdom 35  CHAPTER 2. METHODOLOGY 2.1  X-ray Absorption Spectroscopy (XAS) 2.1.1 Basic Principles of X-ray Absorption Spectroscopy XAS is an element specific spectroscopy sensitive to the local chemical and structural environment of the absorbing atom(s). It is an effective tool for studying most elements of the periodic table and can be particularly useful for so-called “spectroscopically silent” species such as ZnII, FeII and CuII.163,164 XAS is extremely useful for studying transition metal complexes in biological systems.165–167 The electronic and geometric structure of a site can be investigated from the perspective of component elements within an active site of interest, without interference from other elements. A major advantage of this approach is that it can be performed on virtually any type of sample (gas, crystalline, amorphous or solution even in µM concentration) with generally minimal sample preparation.164,168  Figure 2.1: Energy transition and resulting spectrum of typical K-edge XAS experiment. Spectrum illustrating the key features (pre-edge, edge-jump and ionization edge) An X-ray absorption spectrum is obtained when an incident monochromatic X-ray beam has sufficient energy to excite a core electron to an orbital, either partially filled or empty, or to the continuum as shown in Figure 2.1. An abrupt increase in the absorption coefficient 36  gives rise to the dominant feature of the K-edge XAS spectrum known as “absorption edge”, “rising edge” or “edge jump”.163,164 The edge jump corresponds to transitions that have adequate energy to eject a core electron from an atom to the continuum via the photo- electric effect. Given that core electrons have well-defined binding energies, the corresponding absorption edges also occur in predictable energy regions. These edges are typically labelled according to the orbital/state from which the core electron is ejected: i. e., K = 1s, L1 = 2s, L2 = 2p1/2, L3 = 2p3/2, M1 = 3s… etc. The uppercase letter represents the principal quantum number of the orbital and the subscript number denotes a specific absorption line.169 Table 2.1: Electron binding energies of the elements investigated in this thesis Element K-edge (eV) L3-Edge (eV) L2-Edge (eV) Ruthenium 22117.0* 2838.0 2967.0 Sulfur 2472.0* 162.5 163.6 Chlorine 2822.4* 200.0 202.0 Osmium 73871.0 10871.0 12385.0 *experimental results presented only from K-edge of Ru, S and Cl in this thesis The exact energy of the absorption edge also depends upon the effective nuclear charge (Zeff) of the absorbing atom and has been widely used as an oxidation state marker.164,170,171 Table 2.1 shows the absorption edge of the elements investigated in the works herein.  The absorption energy of a particular edge (e.g. K-edge) is well separated and unique for each element and edges even in the nearest neighbours S and Cl.169 However, Ru L3-edge (2838 eV) is always interfered by Cl K-edge (2822.4 eV) since their corresponding edge energies are very close. In that case, multiple steps of data processing is required.172 37   Figure 2.2: Mn K-edge XAS of rac-[MnIII(EHPG)]Na, (where EHPG - ethylenebis[(o-hydroxyphenyl) glycine]), illustrating XANES, EXAFS regions and other spectroscopic features XAS spectra are generally subdivided into two main regions: the X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS), see Figure 2.2. The XANES region is generally considered to range from a few electron-Volts (eV) below the element’s absorption edge to ~50 eV above the edge. Key features included in this region are the aforementioned absorption edge and in some cases just below the edge, minor peaks visible are denoted as pre-edge features. In first row transition metal K-edge spectra, the pre-edge features originate from weak quadrupole allowed transition (1s→3d). The pre-edge intensity and position provide information about the electronic environment of the absorbing atom. Therefore, the XANES region of the spectrum is sensitive to the oxidation state, the immediate coordinating atom, and electronic structure of absorbing atom.163,164,173 In this thesis, XANES of Ru, S and Cl K-edges have been investigated to get oxidation state, covalency and geometry and electronic structures of the complexes studied. The EXAFS region, typically defined from ~50 eV to up to 1000 eV above the edge, is dominated by oscillations in the intensity (Figure 2.2) as a function of incident photon energy. 38  These oscillations result from constructive and destructive interference of the outgoing photoelectron waves due to back-scattering from neighbouring atoms.  Therefore, EXAFS provides direct information regarding the local geometric structure around the element of interest.163 The analysis of reasonable quality EXAFS can give accurate bond lengths (±0.02 Å), as well as information on coordination number (±1) and the chemical identity of neighbouring atoms. In many respects, EXAFS analysis is a complementary technique to single-crystal X-ray diffraction (XRD) for substances in amorphous or solution state.168 Like every other technique, XAS also has its limitations: it requires structural models and lacks chemical selectivity for nearest neighbouring atoms in the same row of the periodic table (e.g.: C/ N / O) in the EXAFS analysis. Recent advances, such as high energy resolution fluorescence detection (HERFD)174, can address some of these limitations although the required instrumentation is not generally accessible.168 The EXAFS region of the XAS spectrum is not a major focus of the work described in this thesis; however details of the theory and information content of this region of the spectrum can be found in recent reviews.158,163,175,176 2.1.2 Metal K-edge X-ray Absorption Near Edge Structure Metal K-edge XANES occurs when a 1s core electron is excited to the valence and conduction bands (nd←1s),164 Figure 2.1. Therefore, it consists of two important features of relevance to ligand effects on the metal center, i.e. bound state transitions in the pre-edge region and the ionization of 1s electrons, IE1s, resulting in the absorption edge.177,178 The pre-edge features of the 2nd row transition metal (e.g. Ru) complexes are usually attributed to electric quadrupole-allowed 4d←1s transitions (analogues to 3d←1s pre-edge transitions of 1st row transition metal complexes),178 as a result the intensity of the pre-edge is very 39  weak.  However, mixing of the metal 5p orbitals into the valence 4d orbitals can occur in non-centrosymmetric complexes. The intensity of pre-edge features can be greatly enhanced by even small amounts of 5p mixing, as 5p←1s transitions are formally electric dipole allowed.163  Figure 2.3: Ru K-edge XAS of Ru carbene complexes illustrating the pre-edge intensity decreases with increased centrosymmetry of the complexes. Reprinted from Getty et al., Inorg. Chim. Acta 2008, 361, 1059–1065. Copyright 2008, with permission from Elsevier.178 In cases where there can be no 5p mixing, such as in complexes with octahedral geometry, the pre-edge features gain very low-intensity and are often buried under the edge. Such issues have recently been investigated for a series of well-defined RuII carbene complexes.178 This study clearly demonstrated the correlation between the intensity of the 4d←1s pre-edge features and the presence of distortions in the geometric structure that could lead to Ru 4d−5p orbital mixing (see Figure 2.3). 40   Figure 2.4: Mn K-edge XAS of Mn-transferrin complex showing the edge-jump shifts with Zeff during the photoreduction of the sample with multiple scans. The second and most dominant feature of this region is the intense absorption-edge. It corresponds to the electric dipole allowed ionization of the core 1s electron to continuum (e.g. Ru∝ ←1s). The energy at the 1st inflection point of the edge is typically used as a measure of the ionization energy of the K-shell (1s) electron (IE1s), and it is sensitive to the effective nuclear charge of the metal center (Zeff),177 Figure 2.4.  Generally, as the Zeff of the absorbent increases, the absorption-edge shifts to higher energy.  The change in Zeff is felt strongly on the core 1s electron compared to that of weakly bound valence electrons. Consequently, the energy gap between the core and the valence level increases and thus the absorption-edge shifts to higher energy.164 The energy of the absorption-edge is widely used as a marker of metal oxidation state.170,171,179 It has been noted that there are some exceptions to this rule171 and that other factors such as ligand type, 5p←1s transition and multiple scattering features can also affect the energy position. Therefore, a shift in edge energy with the oxidation state is relative and 41  studies should be done with a group of complexes containing the same spin state, similar ligand systems and geometry.163,164,180 All the Ru complexes studied herein are mostly pseudo-octahedral complexes with similar ligand fields, therefore the Ru K-edge can be used quite reliably to probe the oxidation state of the metal, as well as the geometry of these complexes. 2.1.3 Ligand K-edge X-ray Absorption Near Edge Structure In transition metal complexes, an important factor in bonding is the electron delocalisation between the metal and the ligand (e.g. mixing of Ru4d with S3p to form a Ru-S bonding interaction), often referred to as covalency.167 The covalency of metal-ligand bonds is an important factor in defining chemical and physical properties.167,181 It can therefore be highly informative to experimentally determine the metal-ligand covalency in complexes for which the electronic structure is of interest. Ligand K-edge XAS  involves the ligand core electron (1s) excited to the valence (np) and conduction bands, np←1s.167 It is a powerful tool to address the electronic structures of metal-ligand interactions in transition metal complexes.182,183 The intense pre-edge features result from dipole allowed transitions (np←1s) from ligand 1s core electrons into delocalized antibonding molecular orbitals, which are formed with significant contribution from metal and the ligands.184 The intensity of the ligand pre-edge feature is proportional to the mixing of ligand orbitals into the metal d orbitals.185 Therefore, ligand pre-edge intensity is a direct probe of the covalency of a metal-ligand bond, and the energies of ligand pre-edge transitions reflect the relative ligand field strength experienced by the metal in a ligand system.184 42  Ligand K-edge XAS is often considered a better probe than metal K-edges to measure covalency due to the increased resolution of transitions in the soft X-ray region obtained by the combination of reduced core hole lifetime and improved monochromator resolution,182,184 and the increased electric dipole-allowed intensity of the ligand pre-edge features. Above all, the main advantage of XAS is it does not require a nuclear spin whereas this is necessary for analysis of covalency using EPR spectroscopy.184 The quantification of sulfur K-edge to investigate the nature of metal-sulfur bonding is used extensively in this thesis and is discussed in detail in section 2.1.4.1.  In addition, Cl K-edge is also used wherever possible as a secondary probe to study the ancillary ligand effect on the metal center. However it is limited to the qualitative analysis herein. Quantitative ligand K-edge XAS as a tool to measure covalency has been pioneered by Solomon, Hedman, Hodgson and coworkers.182,184  It has now been extensively used to determine metal-ligand covalency of S and Cl ligands in transition metal complexes.99,167,183,186–190 The success of this approach has created an impetus to expand the methodology to other elements – such as P,191–193 Br,194 C,195,196 and O.197 However, more studies should be conducted to corroborate these ligand K-edges for their wide applicability. 2.1.3.1 Sulfur K-edge X-ray Absorption Near Edge Structure Sulfur is one of the classic examples of a spectroscopically silent element with poor NMR characteristics (small 33S nuclear magnetic moment, low natural abundance, and large line widths)198 and a few alternative spectroscopic approaches. However, sulfur K-edge XAS exhibits relatively sharp linewidths and a large chemical shift range over its range of oxidation states as shown in Figure 2.5.198 Interestingly, the energy of the edge jump (consequently the 43  pre-edge features) increases with an increase in the formal oxidation state (Zeff) of the sulfur.198,199 In addition, when a sulfur species is coordinated to a metal, pre-edge features corresponding to Mnd-Sσ* ← S1s transitions appear in the spectrum. The energy of such features depend on the Zeff of the metal. Therefore, S K-edge XAS is sensitive to the chemical nature and Zeff of the ligand, and the Zeff of the coordinated metal.200,201  Figure 2.5: S K-edge XAS of series of sulfur redox species. Reprinted from Pickering et al. FEBS Lett. 1998, 441, 11–14, Copyright (1998) with permission from Elsevier Pre-edge Intensity and Covalency: The intensity of each pre-edge feature results predominantly from redistribution of the localized electric dipole-allowed atomic S3p ← S1s transitions to formally forbidden Mnd-Sσ* ← S1s transitions, through covalent S3p/Mnd mixing (see Figure 2.6). The intensity of the transition is thus directly proportional to the extent of S3p mixing in each antibonding orbital.185 However, the studies discussed herein concentrate 44  on the metal-sulfur (M-S) valence orbital which is closely related to the reactivity of the complexes. In XAS, the M-S bond is studied indirectly through its M-S antibond, represented by the lowest energy feature corresponding to Mnd←S1s transition in the spectrum, Figure 2.6. The intensity therefore reflects the amount of S3p character in the Ru4dσ* acceptor orbital (β) as shown in equation 2.1.  It is assumed that the transition originates from a localized core orbital, therefore contributions from other atomic orbitals (i.e., α, χI in eq 2.1) do not significantly impact the total intensity of intense electric dipole transitions.167,200,202  Equation 2.1:  Ground state wave function of metal-sulfur ligand LUMO orbital, Ru4dσ*  Where,  |∗〉 - LUMO wave function    |〉 – Ru metal atomic orbital wave function    - Amount of Ru4d character in Ru4dσ* orbital   	|〉 - Ligand atomic orbital wave function   	 - Amount of S3p character in Ru4dσ* orbital   ∑ !"" |#"〉 - Other ligands atomic orbital wave function   Figure 2.6: Extracting the pre-edge intensity involved in each features of S K-edge XAS obtained from [(η6-hmb)Ru(en)(SOiPr)]+ complex (where hmb – hexamethyl benzene, iPr – isopropyl). |4%&∗〉 =  α|4% 〉 −  β)3〉 − * χ+ |#++ 〉 45   Thus, the amount of S3p character (	) involved in that transition can be quantitatively determined from two parameters, as shown in equation 2.2: the total integrated area under the S K-edge peaks, (,-./01∗←345), and the dipole integral, 〈||78〉, which gives the intensity of the parent S3p← S1s electric-dipole-allowed transition. The total integrated intensity can be obtained from the experimental data by systematic peak fitting. The reference dipole integral, 〈||78〉, has been extensively studied and standardized experimentally184,187,203 and theoretically185 by Solomon and coworkers. Equation 2.2:  Intensity of Ru4dσ* ←S1s transition  Where  〈||78〉 – Dipole integral                   β   - S3p character   S3p Character Quantification: The amount of S3p-character involved in each antibonding orbital is reflected by the pre-edge intensity and hence, indirectly measures the S-covalency of the bonding orbitals. The dipole integral, 〈||78〉 of each sulfur atom involved with a transition is consequently required to extract the S3p character from the pre-edge intensity (equation 2.3).  The dipole integral of a series of S-based ligands is calculated from the analysis of 33S EPR superhyperfine factor187 and X-ray photoelectron spectroscopy (XPS).184 These studies have demonstrated that the inherent dipole integral for a S3p ← S1s transition depends on the charge on the S ligand and the nature of the M-S overlap (σ vs π), which modulates the radial distribution function of both the S1s and S3p orbitals. ,4%& ∗←1 =  β2〈3 ||1〉 46  Equation 2.3:  Relationship between dipole integral (〈;<=|>|;?@〉), S3p character (A) and  pre-edge intensity BC = D E3FG 	HI||78〉 Where, BJ – total area under the pre-edge transition    E – number of holes (either in metal nd or ligand np)    F – number of absorbing S atoms   Figure 2.7: A series of S-ligands illustrating a linear relationship between dipole integral factor and S3p←S1s transition energy. Adapted with permission from Sarangi et al. J. Am. Chem. Soc. 2007, 129, 2316-2326. Copyright 2007 American Chemical Society An accurate linear correlation is obtained between the dipole integral for the S3p←S1s transition and its energy as shown in Figure 2.7. This valuable experimental conclusion is supported by DFT calculations that show a nearly linear increase in the dipole integral when the charge on the absorbent S atom is increased.185 This is due to the fact that an increase in charge on the S atom shifts the core 1s orbital to deeper binding energy relative to the 47  valence (3p) orbitals. Therefore, as the energy gap of S3p (valence) ←S1s (core) transition increases, the dipole integral consequently increases.187 The studies described herein involve a range of RuII and OsII arene S-ligand complexes. The S-ligand on each complex shows unique Zeff and mode of interaction with metal ions. Therefore, the dipole integral of each complex is directly obtained from the extrapolated graph of Figure 2.7 based on their S3p←S1s transition energy. The estimated values used to extract the S3p character of each complex are mentioned in the relevant chapters and show reasonable agreement with literature values reported for these type of complexes.187,204 2.1.4 Instrumentation and Experimental Setup XAS requires an intense and polychromatic X-ray source. Thus it needs the use of synchrotron radiation. Synchrotron radiation is the name given to the broad spectrum of electromagnetic radiation emitted by relativistic accelerated charged particles, i.e. electrons or positrons, travelling in polygon accelerators. Synchrotron sources produce very intense, collimated and polarized X-rays which contain a continuous band of wavelengths spanning a wide range, from µm (infrared) to pm (hard X-ray). Such sources can be tuned to any desired wavelength using appropriate optical devices, monochromators, and filters. In addition, 3rd generation synchrotron sources produce highly intense, parallel and narrow beams of X-rays,205,206 optimized for photon flux and brilliance. The data presented in this thesis were collected at two 3rd generation synchrotron radiation facilities: the Stanford Synchrotron Radiation Lightsource (SSRL, Stanford, California) and the Canadian Light Source (CLS, Saskatoon, Saskatchewan). 48  2.1.4.1 Synchrotron Radiation Components Figure 2.8 illustrates the main components of a synchrotron source. A hot filament in an electron-gun produces free electrons. Those electrons are accelerated using a linear accelerator (linac) into an evacuated booster ring, where they further accelerate to the energy of the electrons in the main storage ring and then are periodically injected into the storage ring. The storage ring is built with arced sections containing bending magnets and straight sections containing insertion devices such as wigglers and undulators. Bending magnets accelerate the electrons around the arced sections (into the next straight section). The energy lost by electrons due to emission of synchrotron radiation is replenished by the right amount of extra energy every time they pass through a radio frequency (RF) supply.206,207 The beamline uses the radiation emitted from either insertion devices or the bending magnets, and therefore is positioned tangentially to the storage ring, along the axes of the insertion devices and bending magnets. A beamline is divided into three sections: the front end, the optics and the experimental hutch.206 The front end isolates the beamline from the storage ring, monitors the position of the photon beam, blocks the X-ray radiation when required, and filters out the low-energy tail of the synchrotron radiation spectrum. The photon beam is focused and monochromatized in the optics section before it enters the experimental hutch. Beamlines are designed specifically for a particular purpose, optimized for a particular energy range and a particular experimental setup. The work presented in this thesis was performed at three different beamlines of two different synchrotron facilities. Metal (Ru) K-edge was done at high energy beamline 7-3 of SSRL, and ligand K-edge was 49  performed at low and mid energy beamlines 4-3 of SSRL and soft X-ray micro-characterization beamline (SXRMB) of CLS. Details of each beamline setup are individually described in the following sections.   Figure 2.8: A schematic of the most important components of a 3rd generation synchrotron source. Adapted from reference 198 with permission from John Wiley and Sons, Copyright © 2011, John Wiley and Sons.206 2.1.4.2 Stanford Synchrotron Radiation Lightsource Beamlines 4-3 and 7-3 SSRL beamlines 4-3 and 7-3 are installed on a wiggler insertion device (Figure 2.9) therefore they produce significantly higher fluxes and brilliance than bending magnet beamlines. In both beamlines, the beam coming from the wiggler is collimated to a double crystal monochromator (DCM) by the collimating mirror (1st mirror). The monochromator (Si (111) φ=0° set for BL 4-3; and Si (220) φ=0° set for BL 7-3) is used to fine-tune the incident energy around the threshold energy (E0) of the probed element and keeps the beam 50  propagating in a direction parallel to the initial direction following two reflections. The focusing mirror (2nd mirror) directs the beam to the sample. These mirrors also suppress higher order harmonics.   Figure 2.9: Schematic layout of beamlines configuration at SSRL Beamline 7-3: This beamline operates at 4.6-37 keV and is dedicated to XAS of biological samples. It produces an intense and high energy beam that can be used to study dilute solid and frozen solution samples. Ru K-edge XAS data presented in Chapter 3 were collected from this beamline. In the experimental hutch, the incident beam is directed from the focusing mirror and is detected by an ionization chamber (I0) filled with argon (Ar) gas. The sample is kept in a continuous-flow liquid He cryostat and the temperature of the cryostat is controlled at ~ 5-10 K during the experiments to minimize the photoreduction and disorder of the samples at low temperature. The emitted fluorescence from the sample is measured using a 30- element germanium (HPGe) detector array filled with liquid N2.177 The transmitted radiation coming from the sample, measured using the 2nd ionization chamber (I1) filled with Ar gas, penetrates through the reference sample (e.g. Ru metal foil). 51  The transmittance from the reference sample is measured using the 3rd ionization chamber (I2) filled with Ar gas. Here, the reference sample is used as an internal calibrant.177 Beamline 4-3: Beamline 4-3 covers the energy range from 2.4-14 KeV, however it has special capabilities for soft-energy (2.4-6 KeV) studies in addition to hard X-rays. Hence, S K-edge XAS of solid state data presented in Chapter 3 and solution data presented in Chapter 4 were collected from this beamline. In the experimental hutch, the beam coming from the focusing mirror enters the N2 gas-filled ionization chamber (I0), where the intensity of the incident beam is measured. Then it penetrates through the sample placed in the sample box, filled with He gas, at a 45° angle to the incident beam to get the maximum fluorescence signals.208 A Lytle detector,209 which is operated with N2 as an ionizing gas and installed perpendicular to the sample box, collects the fluorescence from the sample (If).210 In principle, the intensity of the light transmitted through the sample can be measured using the second ionization chamber (I1). However, this cannot be used for internal calibration since there is typically insufficient radiation to penetrate through the reference sample to provide adequate signal at these photon energies. An external calibration method is used as a common practise to calibrate the lower energy beamlines. This is done by collecting XAS data from a reference sample (e.g. Na2S2O3 for S K-edge) in between the experimental sample. 2.1.4.3 Canadian Light Source Beamline SXRMB SXRMB beamline at the CLS is installed on a dipole bending magnet (Figure 2.10).211 Although, SXRMB covers a wide energy range (1.7-10 KeV), it is optimized for the intermediate energy regime (2 - 5 KeV). This covers the K-edges of many important elements 52  (Si, P, S, Cl, and Ca), the L-edges and M-edges of 4d and 5d transition metals respectively.211 Thus, S and Cl K-edges data presented in Chapters 4 and 5 were collected from this beamline. In SXRMB, two sets of crystals (Si-111 and InSb-111) monochromator is used to cover the desired energy range. Therefore, in addition to the expanded spectral range, it also provides capabilities to perform X-ray absorption fine structure (XAFS), microprobe, and hard X-ray photoemission with improved radiation in terms of flux, brilliance, photon and spatial resolution.211 All the XAS experiments described herein were performed using the XAFS station. The experimental setup at this beamline is presently limited to solid state samples.  Figure 2.10: A schematic layout of the XAFS station of the SXRMB beamline at CLS211 The beam coming from the bending magnet goes through an acceptance aperture and carbon filter before it is deflected by the collimating mirror. The bendable cylindrical mirror collimates the beam to the monochromator (DCM). In addition, it reflects the higher order harmonics. The monochromator scans the energy by rotating around the first crystal while adjusting the second crystal to track the beam and to maintain a fixed exit beam.211 Then the toroidal mirror focuses the beam in both directions on to the sample. In the experimental hutch, the incident beam goes through an ion chamber, kept in vacuum, for in-line I0 monitoring.211 A major advantage provided by the SXRMB end station is that data can be 53  collected in two different ways. As with the previous beamlines, the intensity of the fluorescence from the sample (If) can be measured. However, SXRMB also allows for Total Electron Yield (TEY) detection (see below). A bias ring placed in front of the sample improves the detection efficiency. 2.1.4.4 Detection Methods  Figure 2.11: Schematic diagram for the detection methods used in this thesis work As discussed in the above section, the XAS experiments involved with this thesis work were conducted at three different beamlines – with varying data collection and acquisition opportunities. Data were collected using three different detection methods based on the sample conditions and X-ray energies: transmittance (used for reference samples at BL 7-3), fluorescence yield (used for sample at BL 7-3 and BL 4-3) and total electron yield (used at SXRMB). General schematics of these different detection modes are shown in Figure 2.11. Transmittance: Transmittance (Figure 2.11a) is the simplest method to measure XAS. According to the Beer-Lambert law, equation 2.4, the log of the ratio of the intensity of the 54  transmitted (I1) and incident (I0) X-ray beams yields the absorbance, A. XAS experiments detected by transmittance are thus essentially analogous to UV-vis spectroscopy. Even though transmittance is the most direct method of detection, it suffers from “transmittance thickness” or pinhole effects, spatial variations in absorption due to inhomogeneity in the thickness of the sample.212 For hard X-ray (> 6 KeV) experiments, these distortions are generally insignificant. However, these effects become significant for softer X-rays (< 6 KeV) where the absorption coefficient is much higher.212 Therefore, sample preparation is extremely important and care must be taken to make samples of uniform thickness. However, in this thesis work, transmittance was only used to measure the XAS of reference Ru foil on BL 7-3. Equation 2.4:  Beer-Lambert law K = LM (OP/O?) Where, A – Absorbance  I0 – Intensity of incident X-rays   I1 – Intensity of transmitted X-rays  Fluorescence: In dilute samples, the element of interest is generally at low enough concentration that transmittance measurements are extremely challenging. However, fluorescence detection can be used to indirectly measure the absorbance. This approach yields essentially the same results as transmission with more than one order of magnitude improvement in sensitivity.213 The fluorescence results from secondary processes, i.e. when an electron from a higher-energy orbital fills the hole in the core shell with concomitant emission of a photon.163,175 The energy of the emitted photon is characteristic for a given 55  element and edge, thus allowing for selective detection of fluorescence resulting from a specific excitation process. The fluorescence data is collected using an energy resolving solid state multi-channel X-ray detector positioned at 90° with respect to the incoming beam (see Figure 2.11b). For diluted samples, the detected X-ray fluorescence (If) is proportional to the incident X-ray absorption (I0) and used to obtain the absorbance, A ∝ If/I0. Therefore, it is analogue to an optical fluorescence excitation spectrum.212 Unfortunately, fluorescence detection carries several experimental challenges, most notably self-absorption that may distort the experimental spectrum. Self-absorption occurs in thick and/or more concentrated samples when emitted photons are reabsorbed by the sample. This effect distorts the shape of the emission spectra by diminishing intense features and shifts the maximum.199 Therefore, the samples investigated herein were diluted with BN and homogeneously mixed and spread in a thin uniform layer to minimize the self-absorption effect. The fluorescence method was used at SSRL beamline 4-3 using a N2 ionizing chamber known as a Lytle detector. By contrast, at beamline 7-3, fluorescence was measured using a 30 element Ge array detector. Total Electron Yield (TEY): Another common detection method, particularly in the soft X-ray region, is TEY.212,214 Electron ejection – either via direct photoelectrons (ejected from core orbitals and have a kinetic energy that is the difference between the X-ray energy and their binding energy) and Auger electrons (emitted as part of the relaxation process as a higher-orbital electron fills the hole left behind by the photoelectron) – is a direct result of core excitation processes. Therefore, measuring the total number of electrons ejected provides a measure of the degree of photon absorption.212 56  At beamline SXRMB at the CLS, the sample drain current (Ie) is measured as a means of detecting TEY. The drain current can be used to determine the absorbance, A ∝ Ie/I0. A limitation associated with this detection mode is that the sample must be conductive enough to allow for proper charge flow.212,214 The complexes studied in this thesis were found to be conductive enough to yield good quality data. 2.2 Density Functional Theory (DFT) The use of density functional theory (DFT) in chemistry has dramatically been increased since Walter Kohn and John Pople received the Nobel Prize in chemistry in 1998 for their pioneering efforts in the development of the density functional theory and computational methods in quantum chemistry respectively.215,216 DFT is now considered to be the predominant method for evaluating the electronic structure of molecular systems containing almost all the elements in the periodic table.216 This has been achieved by combination of computational speed and accuracy of the method in describing molecular systems.217 DFT has gained enormous popularity as a means to gain insight in the areas of physics, chemistry, biochemistry, nanochemistry, and material science.216 DFT has become an invaluable complementary and supporting tool to assist in the assignment and interpretation of spectroscopic data.165,166 Within the context of this thesis, DFT methods were used as an important complementary tool to assist in the interpretation of XAS data for RuII and OsII-arene complexes. Two different software packages were used to perform the DFT calculations discussed in this thesis. The Amsterdam Density Functional (ADF, version 2007.01)218,219 was used to perform the calculations described in Chapter 3 and the ORCA (version 2.9) program220,221 was used for all other calculations. 57  2.2.1 Fundamentals of DFT 2.2.1.1 Functionals The theoretical foundation for DFT is provided by the Hohenberg-Kohn (HK) theorems. It proves that the exact ground-state energy of a molecular system is uniquely determined by its total electron density ρ(r).222 This simple relationship provides a powerful alternative to solutions of the Schrödinger equation, equation 2.5, which technically requires knowledge of all electrons in the system.217 The first HK theorem describes the basic relationship and defines a functional, E[ρ], which determines the exact energy from the exact electron density of the system. It further states that the functional can be written by the terms in equation 2.6.222 Notably, the theorem does not state the form or the exact nature of the functional that satisfies the theorem. However, the HK theorems indicate that the problem is variational such that the predicted electron correlation functional to the ground state, (E[RS]), is greater or equal to the exact ground state electron density functional, (E[ρ]), i.e. (E[RS] ≥ E[ρ]).223  Equation 2.5:  Schrödinger equation TѰ =  ĤѰ Where, Ѱ – Wavefunction,  Ĥ – Hamiltonian operator  W – Energy of the state Ѱ  In principle, finding the exact electron density of a system requires that appropriate terms to the generalized functional in Equation 2.6 be determined. Of greatest concern, is the kinetic energy functional (XYRZ). A formalism to address this has been achieved by the Kohn-Sham (KS) construction,224 which treats ρKS(r) = ρ(r); the density of the fictitious system 58  is equal to that of the real or exact system. Therefore, the non-interacting kinetic energy can be calculated from the Kohn-Sham orbitals.223 This construction allows the second Hohenberg–Kohn theorem to be applied and yield through variation of the density the single-particle (Kohn–Sham) equations.  Equation 2.6:  Electron correlation functional, E(ρ), describing the electron density of a system WYRZ = [\\ +  [̂ \YRZ + _YRZ + XYRZ + W`ab YRZ Where, VNN – nuclear-nuclear repulsion term  VeN [ρ] - nuclear-electron attraction term             J[ρ] – electron-electron repulsion term  T[ρ] – kinetic energy functional (universal unknown)   W`ab YRZ – exchange correlation functional (universal unknown) The second so-called “universal unknown” is the exchange correlation functional. One of the simplest functionals used is the local density approximation (LDA),225 which assumes that the electron density is uniform throughout the system. This method is very straight-forward, but leads to an overestimation of the correlation energy and generally overestimates bonding interactions.223 Generalized gradient approximations (GGA), first developed by Becke226 and Perdew227 provide a significantly better model for the exchange-correlation functional. The Becke-Perdew226,227 gradient-corrected functional, BP86, has been widely utilized in the literature and is used in all the DFT calculations described in this thesis. The functional is built by combining the Vosko-Wilk-Nusair local density approximation228 in conjunction with the gradient-corrected functional model for exchange by Becke226 and the gradient-corrected model for correlation by Perdew.227 59  2.2.1.2 Basis Sets Within the Kohn-Sham formalism, a suitable basis set – which allows for the construction of appropriately flexible electron density distributions – is required.223 There are two types of basis sets commonly used in electronic structure calculations: Slater type orbitals (STO)229 and Gaussian type orbitals (GTO).230 STOs are implemented in ADF219 whereas GTOs are used in ORCA,221 and most other computational chemistry programs.223 STOs more properly represent the component atomic orbitals (AOs) near the nucleus with their cusp (discontinuous derivative) behaviour and long range decay. By contrast, GTOs has a zero slope at the nucleus and falls off too rapidly therefore three times as many GTOs as STOs are required to achieve the same level of accuracy.230 On the other hand, the integrals that involve GTOs may be solved analytically,231 for which computational techniques are well developed. Therefore, double hybrid functionals can be efficiently calculated with GTOs.223 For aforementioned reasons both type of basis sets are widely used and well recognized by the chemistry community. In addition to the simple form of the basis functions, the size of the basis set is absolutely critical.230 The smallest basis sets are known as single-zeta (ζ) basis sets, constructed from a single STO. Greater flexibility is afforded by describing each component orbital with a larger number of basis functions; the accuracy of the calculation increases with increasing basis set size with a concomitant increase in computational time and effort. The triple-zeta polarization (TZP) basis set is the present standard basis set used in computational chemistry to get reasonable results.223 TZP basis set is constructed with 4 STOs, the first three STOs to optimize the size of the orbital and the 4th STO optimize the shape of the orbital by 60  adding a polarization function (P).230 In addition, the recently reported Ahlrichs basis sets232,233 introduce diffusion (V) and polarization (P) functions to GTOs, e.g. def2-TZVP (where TZVP - triple zeta valence polarization) to get more accurate results. The def2 basis sets are designed to provide consistent accuracy across the whole periodic table. In this thesis, the calculations performed using ADF utilized the TZP basis set and the calculations performed using ORCA utilized def2-TZVPP (double polarization function) for Ru and Os atoms and TZVP for rest of the molecule. 2.2.2 Utilized DFT Calculations The DFT calculations performed in this thesis are intentionally designed to support and interpret the experimental spectroscopic data by elucidating the electronic structure of the complexes of interest. Therefore, geometry optimizations, single point calculation and time-dependant density functional calculation were performed. The detail of each calculation is discussed below. 2.2.2.1 Geometry Optimization Geometry optimization is a procedure that attempts to find the molecular geometry of minimum energy using an iterative procedure. During this process, the derivative (1st) of the energy is analytically calculated with respect to nuclear displacements of a particular geometry at the end of the self-consistent field (SCF) procedure.219 Geometry optimization is normally the initial procedure for any DFT analysis and often initiated from the available crystal structure of the molecule, considered as a reference point. In all cases presented in this thesis, the optimized geometry was in good agreement with the crystal structure (where 61  available). In cases where crystal structures are not available, it was assumed that DFT- optimized structures would provide a reasonable estimate of the overall molecular structure of the complexes. It has been shown by other researchers that the accuracy of the DFT optimized structures of transition metal complexes is appreciably better than that of medium resolution crystal structures.217,234 To achieve more accurate and faster convergence, special corrections and approximations which account for physical effects that are not directly included in the functional and basis sets were incorporated in the geometry optimization calculations. Such corrections include frozen core approximation, scalar relativistic effect and solvent effects using conductor-like screening model (COSMO). Frozen Core Approximation:  It was applied for the calculations involved with heavy transition metal (Ru) complexes in Chapter 3 using ADF.218 The frozen core orbitals are taken from very accurate and large STO basis sets and treated together with nucleus of the atom (Ru) of interest. Therefore, the core electrons are not included in the perturbation treatment, because deep-core atomic orbitals change very little upon bond formation and insignificantly effect relative energies or geometry parameters. Thus, the core functions do not add variational freedom to the (valence) basis set, and hence, reduce the size of the variational basis set and consequently cut down the computational time without sacrificing the quality. Therefore, a frozen core calculation provides the total charge density and potential in the valence and core region, by excluding the deep-core orbitals.219 Frozen virtual orbitals are not allowed in the geometry optimization calculation done by ORCA.220 62  Relativistic Effects: Physical phenomena observed when the speeds of particles, ν, are comparable with the speed of light, c, are referred to as relativistic effects.235 Relativistic effects can strongly influence the chemical and physical properties of heavy elements and their complexes (e.g. Ru and Os).235 Therefore, scalar relativistic effect was employed for all the Ru and Os complexes studied in this thesis using the zero-order regular approximation (ZORA) method. At the zeroth order this expansion contains relativistic corrections to the non-relativistic energy. This relatively simple correction has been shown to provide an excellent description of valence and sub-valence electrons in heavy elements.219,236 Solvent Effects: The effect of solvation for complexes in solution was achieved using the conductor-like screening model (COSMO) developed by Klamt.237 This method is an efficient way to calculate the energies, structures and properties of molecules in solution.220 Here the environment effects are treated in an approximate fashion. The solvent is represented as a dielectric medium that induces charge polarization on a suitably defined surface around the molecule, the cavity, immersed in the solution is called dielectric polarizable continuum. COSMO (aqueous) model calculations are performed to obtain the geometry optimization of particular Ru complexes studied in Chapters 4 and 5. 2.2.2.2 Single Point Calculation Single point calculation often follows the geometry optimization because it utilizes the optimized geometry of the complex of interest. Single point calculation provides important information (e.g. energy, electron occupancy, % contribution from each atomic orbital) about molecular orbitals.218,220 These results are used to construct the valence MO diagram, MO pictures and electronic structure of the molecule. Single point analysis using ADF calculates 63  the atomic charges for each atom based on Mulliken population,238 Voronoi and Hirshfeld239 analyses; additionally Löewdin charges240 can be calculated in ORCA. 2.2.2.3 Time Dependent Density Functional Theory (TDDFT) DFT is a ground-state based methodology (since its foundations are based on HK theorems), which approaches the calculation of excited states and their properties indirectly. In this respect, DFT has greatly profited from the analogous developments in the ab initio area where time-dependent linear response theory has been worked out in great detail.234 The most straightforward treatment in which the basis functions do not depend on the perturbation and the exchange correlation potential is assumed to be time-independent (adiabatic approximation).223 TDDFT calculations are used to simulate and rationalize the XAS and UV-vis (excited-state techniques) spectra of Ru and Os complexes, discussed in Chapters 4 and 5. 2.2.2.4 Natural Bond Orbital (NBO) Analysis In contrast to DFT, natural bond orbital analysis describes the hybridization and bonding of a system using polyatomic wavefunctions (Ψ), based on one particle density matrix.241 A set of bonding and antibonding orbitals are constructed from self-consistent field (SCF) quantum mechanical calculations and referred as NBOs. The calculated orbitals are best possible and/or most rapidly convergent orbitals with maximum electron density.242 NBO is also referred as “chemist’s basis set” as it illustrates the molecular structures with the picture of localized bonds and lone pairs as basic units. 64  The NBO analysis, presented in this thesis, is done using natural bond orbital program (NBO 5.0)242 combined with an as yet unreleased version of ORCA (version 2.9*). The NBO 5.0 program not only satisfies the Pauli exclusion principle (maximum-occupancy 2.000) but also includes coefficients for factors such as polarization, symmetry adaption, delocalization and atomic hybridization to achieve the highest possible percentage of the electron density. Therefore, it provides the most accurate possible natural Lewis structure of Ψ.242 NBO analysis is herein used to obtain the natural Lewis structure of metal-ligand (i.e. Ru-S, Ru-X) bonds to show the extent of metal-ligand interaction in a series of Ru-arene complexes in addition to DFT analysis. 65  CHAPTER 3. EFFECT OF OXYGENATION ON THE REACTIVITY OF RUTHENIUM (II) ARENE THIOLATO COMPLEXES 3.1 Introduction† In recent years, there has been rapid growth in the design of organometallic anticancer complexes and studies focusing on biological mechanisms to selectively increase their reactivity in cancer cells.12,29,46,243–246 RuII arene complexes have attracted significant interest because they have shown to bind strongly to DNA151 and induce conformational changes, including denaturation. This damage can lead to further downstream effects, such as apoptosis.3,152 However, the biological specificity of these complexes has, to date, not been adequately defined. Clearly, ligand substitution at these coordinatively saturated octahedral complexes is required in order for DNA binding. However, some of the complexes with the greatest therapeutic potential – the RuII arene thiolate prodrugs (see Figure 3.1) are highly inert towards substitution in their initial form. It has been shown that these complexes are quite susceptible to ligand oxidation at the sulfur atom. Interestingly, recent work has shown that reversible post-translational oxygenation of cysteine thiolate (SR¯) to sulfenato (SOR¯) and sulfinate (SO2R¯) functions as a redox-based signal transduction mechanism, which plays an important role in cellular regulatory processes.247–249 In addition, these thiolate redox processes in biological systems are sometimes controlled by metal coordination (ZnII, FeIII or  † This chapter is based on collaborative work with Prof. Peter J. Sadler and co-workers Drs. Holm Petzold, Pieter C. A. Bruijnincx and Abraha Habtemariam from University of Edinburgh and University of Warwick, UK.  Most of the work presented herein is already published and reprinted with permission from J. Am. Chem. Soc. 2009, 131, 13355-13361. Copyright 2009 American Chemical Society. 66  CoIII) to the S atom.250,251 One might thus postulate that sulfur oxygenation might play a role in the biological fate of Ru(II) arene thiolato complexes. As discussed in section 1.2.2.2, the cytotoxic half-sandwich [(η6-ar)Ru(en)(SR)]+ complexes (where ar = arene (p-cym or hmb); en = ethylenediamine) seem to be activated via thiolate oxygenation towards DNA binding. Furthermore, the primary class of RuII arene complexes (X = Cl) form a thiolate adduct when they react with glutathione (GSH),95,252 an abundant thiol in cells, and human albumin,253 and also readily undergo thiolate oxygenation. Therefore, redox activation may play a general role in the mechanism of action of a wide variety of these RuII arene complexes. The work presented in this chapter focuses on defining the properties of Ru-S bonds as a function of (i) the extent of thiolate oxygenation, (ii) the nature of the thiolate substituent (alkyl or aryl) and (iii) the arene (p-cym or hmb). In order to understand the relevance of ligand-centered oxidation on its mechanism of action, it is necessary to study the influence of electronic effects on the bonding of the sulfenato and sulfinato complexes. The synthesis of metal-sulfenato complexes has been a great challenge. However, our collaborators have recently succeeded in synthesizing the RuII complexes containing a monodentate sulfenato ligand by direct oxidation of thiolato ligand using H2O2 (Scheme 1.1).76 This result allowed for comparison of a series of complexes containing monodentate thiolato, sulfenato, and sulfinato ligands (Figure 3.1).  Previous studies have explored the effect of thiolate oxygenation in transition metal complexes.201,251,254–268 In general, such studies have involved complexes with multiple thiolato-based ligands, specifically in relation to the study of nitrile hydratase (NHase)-like 67  model systems with FeIII,201,257,259,260,264,266,267 CoIII,254–256,262,265,266,268 NiII,251,261,265 ZnII 261 and RuII 263 centres. By contrast, this study allows for a direct comparison of the electronic effects of a single sulfur-containing ligand and its impact on the nature of the resultant sulfenate and sulfinate metal ligands.   Figure 3.1: Molecular structures of [(η6-ar)Ru(en)(SOnR)]+ complexes investigated using XAS and DFT. Complexes 1, 2 and 3 are thiolates, sulfenates and sulfinates respectively.  In series a ar = p-cymene and in series b ar = hmb. In 3a, 3b and 1c R = Ph; in 3a″ R =  cyclohexyl; Complexes 3a’ and 3b’ are in silico models only.  In RuII arene complexes, oxidation of the thiolato ligand in complex 1 to sulfenato (2) and/or sulfinato (3) ligands plays an important role in controlling its stability and reactivity.76 Studies of these complexes with biologically relevant molecules such as GSH,95,104,252 cGMP (guanosine-3’,5’-cyclic monophosphate)95 and human serum albumin253 have implied that 68  sulfur-centered oxidation labilizes the Ru-S bond. However, available crystallographic data indicate that a shortening of the Ru-S bond occurs upon the oxidation of the ligand (e.g., 1→2).76,268 A similar observation was also made in RuII arene sulfinato complexes by White and coworkers.97 It was postulated that the lability of the Ru-S bond should be quite sensitive to H-bonding and ancillary ligand effects, however this has not been verified.80,104,252 X-ray absorption spectroscopy (XAS)167,198,200 in concert with density functional theory (DFT) calculations was used to probe the above mentioned issues of the Ru-S bond in a series of [(η6-ar)Ru(en)(SOnR)]+ complexes (n=0-2) in greater detail. Herein we have shown that ligand oxidation seems to be necessary but not sufficient for biological activation. Therefore, it is postulated that protonation of the sulfenate complex, [(η6-ar)Ru(en)(SOR)]+, would allow for ligand dissociation and DNA binding. The specific electronic factors that contribute to observed variations in reactivity of these complexes are also discussed. 3.2 Experimental and Computational Aspects This section describes sample preparation and data collection for S K-edge and Ru K-edge XAS followed by data processing and fitting with specifically designed software programs. The chosen computational model utilized for DFT calculations of these systems is also discussed. 3.2.1 Sample Preparation All complexes studied in this chapter were prepared by Drs. Holm Petzold, Pieter C. A. Bruijnincx and Abraha Habtemariam using reported procedures76 except complexes 3a and 3b.  Complexes 3a,b were synthesized by oxidation of the corresponding thiolate (20 mg, 69  0.035 mmol), with an excess amount of hydrogen peroxide (10 mol) in methanol (10 ml). Then, the reaction mixture was stirred for 16 hours. The solvent was removed to dryness under vacuum, and the residue was washed with diethyl ether. The pure products were obtained by crystallization from methanol. All the complexes were fully characterized by 1H NMR spectroscopy, ESI-MS spectrometry and elemental analysis.96 S K-edge XAS samples were mounted as a finely ground 1:1 mixture of the complex to boron nitride (BN) dusted on sulfur-free Kapton tape across the window of an Al plate. All the samples with a 1:4 ratio of the complex to BN for Ru K-edge XAS were finely ground and then compactly pressed in between a 0.5 mm thick Al spacer, tightly sealed on both sides with Kapton tape. 3.2.2 XAS Data Collection XAS measurements for S K-edge were gathered at SSRL beamline 4-3 using a modified “low Z” setup allowing for low temperature data acquisition under ring conditions of 3.0 GeV and 100 mA current. The specific beamline setup for these experiments were described in section 2.1.5.2 and further detail about data acquisition is described in the literature.210,269 Energies were calibrated using sodium thiosulfate (Na2S2O3) as a reference standard, with the 1st derivative of the first pre-edge feature being calibrated at 2472.02 eV.184 Reference samples remained throughout the run and scanned before and after each data set to ensure stability of the monochromator and beam. Signal was detected with a Lytle fluorescence detector filled with N2 gas at ambient temperature (298 K) and pressure (~1 atm). XAS Collect software270 was used to collect the data in the energy range from 2430-2550 eV. At least three scans were taken for each sample and the experiment was repeated at least twice to confirm reproducibility. 70  Ru K-edge XAS data were collected at SSRL on beamline 7-3 under ring conditions of 80-100 mA at 3.0 GeV. This beamline has a 20-pole, 2 T wigglers, 0.8 mrad beam, and a Si (220) double-crystal monochromator that was detuned by 50% intensity to attain harmonic rejection. Detail of the experimental configuration of the beamline was described in section 2.1.5.2. Energy calibration was done using the 1st derivative of the first lowest energy inflection point of Ru metal foil assigned at 22117 eV. The incident X-ray intensity (I0), sample absorption (I1), and Ru reference absorption (I2) were measured as transmittance using argon-filled ionization chambers. Six to eight sweeps were taken for each sample, and all data were measured to k = 15 Å-1 at 13 ± 3 K within an Oxford Instruments CFI208 continuous-flow liquid helium cryostat.  Each measurement was repeated twice to ensure the reproducibility of the results. 3.2.3 Data Processing and Analysis XAS data reduction, i.e. averaging, calibration and background removal of S K-edge was done using SIXPack.271 Prior to normalization, all acceptable scans were averaged and a linear background was subtracted from the entire energy range. Normalization of the data was accomplished by fitting cumulative pseudo-Voigt functions to the post edge region and normalizing the edge jump to 1.0 at 2490 eV. A Matlab based toolbox for the fitting and analysis of XAS spectra, BluePrint XAS version 0.5,272,273 was used to fit experimental pre-edge features by using pseudo-Voigt functions (sum of Lorentzian and Gaussian functions, Equation 3.1). A Monte Carlo based method is used to simultaneously fit the background and the spectroscopic features. Experimental line shapes are expected to be a convolution of a Lorentzian transition envelope and the beamline 71  spectrometer optics (monochromator resolution) imposed the Gaussian line shape.274,275 Therefore, a fixed 1:1 ration of Lorentzian : Gaussian (pseudo-Voigt) contribution successfully reproduced the pre-edge features. The ionization edge jump was fit with cumulative pseudo-Voigt line shapes.      Equation 3.1:  Approximate Voigt lineshape ,c  = H dη f gh2 i 2 (W−j)2+gh2 i2 k + (1 − η)l−1/2(W−jhm )²o Where, η – Lorentzian fraction     Γ – Full Width at Half Maximum (FWHM) of the Voigt profile     H – Maximum value of the function     P – Position of the peak    E – Energy     C – Constant [c = 2√qF4 = 2.355  Ru K-edge XAS data were analyzed using SIXPack.271 Scans were averaged and energy calibration was reconfirmed using the 1st derivative of the first lowest energy inflection point of reference Ru foil assigned as 22117 eV. Background subtraction and normalization were performed simultaneously using a linear pre-edge function and a quadratic post edge function. 3.2.4 DFT Calculations Starting geometries used for the DFT calculations were obtained directly from the crystallographic information file (CIF) of these complexes. All the DFT calculations discussed 72  in this chapter were performed using the Amsterdam Density Functional (ADF) modeling suite 2007.01.218,219 The model used for calculations is a Becke-Perdew, BP86,226,227 gradient-corrected functional with a triple-zeta polarization (TZP) basis set. The relativistic correction for Ru metal was applied using scalar ZORA,235,276 unless otherwise stated. The small frozen core basis set approximation was applied with no molecular symmetry and a high numerical integration (6.0) was used to maintain numerical accuracy in the optimizations. The initial geometry optimization was followed by single-point calculations and fragment analysis. Fragment analysis calculations were carried out on the [(η6-ar)Ru(en)(SOnR)]+ complexes with labelled fragments of Ru, ar (cym and hmb), en, S, On (where n=0-2) and R (iPr, Ph) to explain the ligands to metal contributions and transition assignments. ADF input files can be obtained from the supporting information of the related publication.96 Natural bond order (NBO) analysis is done on geometry optimized structures using NBO 5.0 program242 combined with ORCA (version 2.9*)‡. Single point calculations using ADF and ORCA yield similar results with only minor differences attributable to minor methodological differences between the two software programs. The above model with the smaller split valence polarization, SVP, basis set is used to illustrate the extent of Ru-S bond interaction in thiolato, sulfenato and sulfinato complexes. 3.3 Results and Analysis The effects of sulfur ligand oxygenation on the Ru metal center and S ligand are investigated through Ru K-edge XAS and S K-edge XAS respectively. In addition, modifications  ‡ ORCA (version 2.9*) is a modified version of ORCA 2.9, which allows for NBP analysis. This version is as yet unavailable for general users. 73  to the ancillary ligand are also investigated using S K-edge XAS. Since XAS is an element specific technique, this methodology seems to be appropriate for investigating the crucial mechanism of RuII arene complexes discussed herein. 3.3.1 Effect of Oxygenation on Ru Metal Center  Figure 3.2: Normalized Ru K-edge XANES spectra of complexes of 1a,b - 3a,b.  Inset shows the 1st derivative spectra indicating the edge jump position. Ru K-edge XAS spectra for solid state thiolato (1a,b), sulfenato (2a,b), and sulfinato (3a,b) complexes are shown in Figure 3.2. The XANES region of the spectrum clearly indicates that the edge region is nearly superimposable, which is confirmed by comparison of first derivatives of the spectra. The striking similarities between these spectra imply that there is no change in metal oxidation state upon ligand oxidation. Each complex is best described as a RuII with a low spin 4d6 configuration; oxygenation of the ligand does not directly affect the metal center and its electronic structure. The absence of pre-edge features in the Ru K-edge data shows that there is little or no mixing of Ru5p character into the empty Ru4dσ* orbitals. 74  Therefore, the ligand environment around the metal centre is well described as pseudo-octahedral in all complexes,178,277 and the acceptor orbitals are a set of two gerade, eg (4dx²-y² and 4dz²) orbitals. This further strengthens the conclusion that ligand oxidation causes only minor perturbations at the metal center. 3.3.2   Effect of Oxygenation on S-Ligand  Figure 3.3: S K-edge spectra of RuII arene A - thiolato 1a and 1b, B - sulfenato 2a and 2b, and C - sulfinato 3a and 3b complexes.  An asterisk (*) denotes features due to thiolato-based impurities in the sulfenato/sulfinato complexes (>5%).  75  As shown in Figure 3.3, the S K-edge spectra of these complexes differ markedly as the nature of the sulfur-containing ligand and its degree of oxidation change. Sulfur ligand oxygenation shifts the absorption edge to higher energy by ~3 eV from 1→2 as well as from 2→3, due to a decrease of the S1s orbital energy with increasing Zeff. DFT calculations predict that the lowest energy feature in each of the complexes corresponds to a Ru4dσ* ←S1s transition. The intensity of the transition derives from the electric dipole-allowed character of the atomic S3p←S1s transition. Therefore the intensity can be correlated with the amount of S3p character in the Ru4dσ* acceptor orbital as shown in Equation 2.2. The quantification methodology is explained in section 2.1.4. Table 3.1: Experimental peak energies and transition assignments and experimentally determined S3p character in the acceptor orbital as estimated from normalized fit intensities* Complex S K-edge XAS Transition % S3p in Acceptor Energy (eV) Acceptor ← Donor XAS DFT 1a 2471.2 2472.8 Ru4dσ*←S1s SCσ* ← S1s 22.5 ± 1.5% 29.0 ± 2.0% 20 33 1b 2471.2 2473.0 Ru4dσ*←S1s SCσ* ← S1s 24.0 ± 1.6% 17.1 ± 1.1% 19 29 1c 2471.4 2472.7 2473.5 Ru4dσ*←S1s ϕπ* ← S1s SCσ* ← S1s 17.1 ± 1.4% 11.7 ± 0.8% 27.1 ± 1.8% 16 8 a 2a 2474.2 2475.0 2476.2 Ru4dσ*←S1s SOσ*/SCσ*← S1s SOσ*/SCσ*← S1s 22.2 ± 1.2% 24.8 ± 1.2% 21.4 ± 1.1% 17 a a 2b 2474.3 2475.1 2476.2 Ru4dσ*←S1s SOσ*/SCσ*← S1s SOσ*/SCσ*← S1s 22.8 ± 0.9% 17.8 ± 0.7% 20.6 ± 0.8% 18 a a 3a 2477.1 2478.1 2479.2 Ru4dσ*←S1s ϕπ* ← S1s SOσ*/SCσ*← S1s 11.2 ± 1.3% 5.8 ± 1.0% 27.7 ± 2.9% 6 (6)b a a 3b 2477.2 2478.4 2479.4 Ru4dσ*←S1s ϕπ* ← S1s SOσ*/SCσ*← S1s 13.2 ± 0.4% 8.8 ± 0.3% 33.2 ± 1.0% 7 (4)b a a *DFT-calculated S3p contributions to the acceptor orbitals as determined from a Mulliken charge decomposition of the Kohn-Sham orbitals are also included for comparison and included in Appendix B.  aExact values in these cases cannot be extracted since S3p character is dispersed over a wide energy range and multiple acceptor orbitals.  CValues in parentheses are for 3a’ and 3b’ where R = iPr.  76  Systematic peak fitting272,273 and assignment of the observed spectroscopic features were performed for each complex and is presented in Appendix A. The dipole integral, 〈S3p|r|S1s〉 as described in Equation 2.2, of thiolato and sulfinato complexes used to estimate the contributions of the experimental S3p character to the Ru4dσ* have been obtained from the literature; the values of 〈SR¯3p|r|S1s〉 ≃ 8184,187 and 〈SO2R¯3p|r|S1s〉 ≃ 18185,187 were used, respectively. The value of the dipole integral for the sulfenato complexes was interpolated from that of the thiolato and sulfinato complexes since it has been previously shown that the relationship between the dipole integral and Zeff is effectively linear184,185,187 and i.e. 〈SOR¯3p|r|S1s〉 ≃ 13.  Figure 3.4: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 1a, 2a and 3a. Table 3.1 summarizes the results with estimates of uncertainties obtained from the fitting procedure. Notably, there is little change in the S3p contributions to the Ru4dσ* upon oxidation from 1→2 (e.g., 1a vs 2a), however, there is a substantial drop in S3p character in the fully oxygenated species (3a,b). A natural bond orbital (NBO)242 analysis of the DFT results provides a useful way of visualizing localized bonding in these complex systems. The localized NBO Ru-Sσ* antibond for each of the complexes is shown in Figure 3.4), illustrating S3p 77  contributions to the Ru4dσ*. The NBO derived bond order§ of the Ru-Sσ* bond is found to be 0.78, 0.74 and 0.28 in 1a, 2a and 3a respectively. These results are in agreement with the experimental S K-edge XAS results. Slight variations of the arene ligand (i.e., a, where ar = p-cym vs b, where ar = hmb) generate only minor differences in the spectroscopic data that cannot be distinguished within experimental error. 3.3.3 DFT Analysis To explore the influence of oxygenation from the oxygen atom(s) end, we have investigated our DFT results in greater detail. Earlier studies on Fe-sulfinato complexes suggested that contributions from the oxygen atoms in defining the overall charge donation from oxidized ligands cannot be neglected.201,251 The ADF input files and the obtained molecular orbital descriptions derived from the output are included in Appendix B. 3.3.3.1 Relativistic Effect and Hydrogen Bonding on Geometry Optimization To study the factors influencing the geometric structure of these species, systematic changes to the computational model were performed. Most notably, the effect of relativistic effects (Figure 3.5) and the potential influence of explicit H-bonding to the sulfur-containing ligand (Figure 3.6) were explored. Important bond distances from the resulting molecular structures are tabulated in Tables 3.2 and 3.3. Geometries calculated without including relativistic effects or inter molecular hydrogen bonding tended to yield M-L bond distances that were significantly longer relative to experiment. However, geometries obtained with  § NBO derived bond order is calculated using the formula: (tCu"uv Cww.xyuz"{Cu"uv Cww.) H  ; It is anticipated that the Ru-S bond is single. 78  scalar ZORA relativistic corrections and H-bonding (wherever appropriate based on crystallographic coordinates) are in better agreement with available crystallographic data.76  Figure 3.5: Geometry optimized structure of 3b with (structure 1) and without (structure 2) scalar ZORA relativistic effects included during optimization Table 3.2: Selected bond lengths from DFT optimized structures of 3b with (structure 1) and without (structure 2) scalar ZORA relativistic effect are compared with parameters derived from the X-ray structure Bond (Å) Structure 1 Structure 2 X-ray S-Ru 2.407 2.423 2.3144(7) S-O 1.511/1.504 1.508/1.504 1.472(3)/1.470(3) S-C 1.819 1.818 1.806(3)   Figure 3.6: Geometry optimized structure of 2a with relativistic effect and solvent MeOH (structure 1) and without solvent (structure 2) 79  Table 3.3: Selected bond lengths of DFT optimized structure of 2a with scalar relativistic effect and solvent (structure 1) and without solvent (structure 2) Bond (Å) Structure 1 Structure 2 X-ray S-Ru 2.396 2.429 2.3790(8) S-O 1.581 1.566 1.552(2) S-C 1.885 1.891 1.823(4)  3.3.3.2 Effect of Ligand Oxidation As seen in the x-ray crystallographic structures, Ru-S bond distances decrease with ligand oxygenation from 1→2→3. The resultant valence MO descriptions are also in reasonable agreement with the overall qualitative trends derived from the S K-edge XAS spectra (see Table 3.1 and Appendix B). Calculations suggested that although S3p⟹Ru4d charge donation decreases upon oxidation, the overall ligand donor contributions remain essentially unchanged (20-23%, Figure 3.7). This is partly due to a compensatory effect of the oxygen atoms of the oxidized ligands. These consequences are consistent with those obtained on model systems of iron-and cobalt-containing nitrile hydratases active site.201,251,267,278 Figure 3.7 summarizes the effects of ligand oxidation on charge donation to the metal center and large effects are observed as a function of oxidation. The observed trends are consistent for both hmb and p-cym series of complexes. To probe ligand oxidation exclusively, in silico model complexes 3a’ and 3b’ (where R = iPr) are included in the comparison study, due to fully oxidized alkylsulfinato complexes akin to 3a,b have not as yet been isolated. 80   Figure 3.7: Breakdown of SOnR ligand donor contributions to Ru4dσ* orbital (in %) for 1a→2a→3a’ and for 1b→2b→3b’ (open symbols) including the total ligand charge donation (blue squares) as well as specific contributions from sulfur (green circles), oxygen (per atom, red circles), and isopropyl (black circles).  Lines are used to indicate general trends only. Although the overall charge donation from the sulfur-containing ligand remains similar in all three oxidation states, the origin of the charge donation changes rather remarkably upon oxidation. For example, in the thiolato complexes, the Ru-S coordination bond is dominated by direct S3p⟹Ru4d charge donation. Sulfur donation changes little in the sulfenato complexes, but a new and relatively large contribution from the sulfenato oxygen emerges, which increases overall charge donation from the ligand. By contrast, sulfur donation in the sulfinato complexes is dramatically decreased, with some compensatory effects by the appearance of contributions from the alkyl R group. Importantly, the addition of a second terminal oxo group causes a net decrease in charge donation by each of these terminal oxygen atoms. This effect is attributed to a loss of S-Oπ character in the S-O bonds on going from the sulfenato to the sulfinato species, switching off an efficient mechanism for charge delocalization in the ligand. A significant conclusion drawn from these data is that 81  charge donation from the terminal oxo groups is largest in 2, implying that the sulfenato species should be most susceptible to chemical perturbations by having a direct interaction with Ru4d orbital(s). 3.3.4 Ancillary Ligand Effects 3.3.4.1 Arene Ligand Effects In principle, ancillary ligands may also have an impact on the Ru-S bond. However, in these complexes, changes in the arene ligand (i.e., when changing from p-cym to hmb) do not result in spectroscopically significant changes in the Ru4d*←S1s transition. DFT calculations also similarly predict only a very little change (~1%) in the S3p contribution to the antibonding, which is within the experimental error of the measurements (see Table 3.1). However, the SCσ*←S1s transition, is predicted to be more sensitive to the nature of the arene ligand, and we do observe a marked change in the relative intensities of the Ru4d*←S1s and SCσ*←S1s transitions for 1b (ar=hmb) as compared to 1a (ar=p-cym) as well as a drop in the S3p contribution for 1b (see Table 3.1); this is consistent with greater electron donation from hmb279 to the Ru center as compared to p-cym. A similar effect is also observed in the S K-edge data for the sulfenato complex (see 2a vs 2b in Table 3.1). It has formerly been suggested that observed differences in the pKa between protonated 2a (3.37) and 2b (3.61) result from increased charge donation from hmb in 2b vs p-cym in 2a. Both XAS data and DFT calculations support this interpretation.   82  3.3.4.2 Sulfur Ligand Substitution Effect  Figure 3.8: Normalized S K-edge XAS spectra of 1a (black) and 1c (red).  Inset shows second derivative spectra indicating presence of a shoulder at 2472.7 eV in 1c due to a low-lying φπ* orbital from the aryl moiety. It has previously been noted that arylthiolato complexes of the type discussed herein are less susceptible to oxidation104 and generally more inert than their alkylthiolato counterparts. The S K-edge XAS spectrum of the arylthiolato complex 1c (R = Ph in Figure 3.8) differs from 1a in that a new feature appears as a low-energy shoulder on the intense SCσ*←S1s feature as clearly shown in the inset of Figure 3.6. Similar features has been observed in organic sulfones due to S3p character mixing into aryl π* orbitals (ɸπ*) through excited state hyperconjugation.280 Therefore the intensity of this feature results from redistribution of intensity from the main Ru4d*←S1s transition as pictured in Figure 3.9, which must be considered when interpreting the transition intensities. 83   Figure 3.9: Illustration of hyperconjugation in complex 1c due to S3p character mixing into aryl π* orbitals (ɸπ*) From 1a to 1c, the intensity of the lowest energy transition, Ru4d*←S1s, would predict a moderate decrease in charge donation to the metal center (from 22% to 17%, as in Table 3.1); however the intensity redistributed through hyper-conjugation (see ɸπ*←S1s in Table 3.1) must be included for the total donation from the S ligand. This indicates that the overall S3p⟹Ru4d charge donation is, in fact, greater in 1c (~29%) than 1a (~22%). The slightly increased energy of the Ru4d*←S1s transition is also consistent with greater Ru-S covalency in 1c. Both the increased charge donation from the ligand and available delocalization over the aryl substituent should contribute to a modest strengthening of the Ru-S bond, which is consistent with decreased susceptibility to oxidation in the aryl complex. This higher covalency of the Ru-S bond appears to be carried through even in oxidized species. A shoulder at ~2477.5 eV in the S K-edge spectrum of complex 3a (see Figure 3.10) results from excited state hyperconjugation (ɸπ*←S1s as shown in Table 3.1), i.e., S3p mixing into aryl π* orbitals (ɸπ*), which concomitantly redistributes the antibonding character of the Ru-S bond. This is also supported by the slightly increased energy of the Ru4d*←S1s transition. 84   Figure 3.10: S K-edge XAS spectra of 3a″ (R = cyclo hexyl, black) and 3a (R = Ph, red).  Presence of a shoulder at ~2477.5 eV in 3a indicating a low-lying φπ* orbital from the aryl moiety. 3.4 Discussion Results obtained from XAS and DFT indicate that the Ru-S bond in these species is quite sensitive to ligand oxygenation and subtle modifications in ancillary ligands. Ru K-edge XAS is consistent with the fact that only minor perturbations occurred to the overall pseudo-octahedral geometry of these complexes, and the oxidation state of metal center remains as Ru2+, low spin 4d6, regardless of the thiolate ligand oxidation state. This is also in agreement with the observed total charge donation from the S-containing ligand which changes only slightly upon oxygenation due to the compensation between S atom, terminal oxygen(s) and the R group. This implies that S-oxygenation should have very little impact on the inherent strength and/or lability of the Ru-S bond in sulfenate. It should be noted here that the second oxygenation drastically decreases the Ru-S covalency in sulfinate but the formation of such species under the biological condition is very difficult.Therefore, these studies suggest that involvement of the terminal oxo group is particularly important in the sulfenato derivative, particularly through a S-Oπ bonding interaction. In addition, 85  modifications to the ancillary arene ligand, as well as the thiolato substituent group have a quantifiable impact on the details of the Ru-S interaction.  Figure 3.11: Graphical representation of the Ru4dσ* Kohn Sham orbital from 2a. The most important take home message from our study is that ligand oxygenation alone should not have a major impact on the lability of the Ru-S bond; however the oxygenated sulfenato species does have an impact on the reactivity of this family complexes. A close investigation of the Ru4dσ* (see Figure 3.11), clearly showed that the LUMO has both RuSσ* and SOπ* character. Here the charge donation from the SOπ orbital helps to preserve the highly covalent Ru-S bond even after ligand oxidation. As mentioned in the literature the interaction certainly withdraws electron density from the SOπ orbital, and concomitantly strengthens the S-O bond.281,282 This postulation is also supported by DFT model calculations on RuIISOiPr and CdIISOiPr species (Figure 3.12), confirming that orbital mixing with the Ru 4d orbitals causes a decrease in the S-O bond distance. The nature of this orbital suggests that protonation of the terminal oxo group would have a dramatic impact on Ru-S bonding. As 86  observed in the previous study, protonation would also weaken the H-bond (specified in Figure 3.11) between the sulfenato ligand and the NH2 of the adjacent en-ligand under acidic conditions.76   Figure 3.12: Comparison of S-O bond distance and orbital mixing in RuII vs CdII.  Based on our studies, a mechanism of activity is postulated for the parent RuII arene thiolato complexes as depicted in Scheme 3.1. The oxidation step may involve reactive oxygen species such as peroxides formed by the oxidation of GSH or NADH ( β-nicotinamide adenine dinucleotide) by oxygen.104,283 First, ligand oxygenation would make the prodrug susceptible to protonation at the terminal oxo group, which is followed by hydrolysis and DNA binding under acidic conditions. In cancer cells, the pH of intracellular and extracellular fluids can be acidic (pH 6-7); however lysosomal compartments in cells can attain more acidic condition, such as pH 4-5.284,285 Fundamentally, to activate these complexes, even relatively low conversion to the protonated form (2→2-H+) should be sufficient for ligand substitution and subsequent DNA binding. The Ru-S bond could also be weakened by Lewis acid activation as has been noticed by Kovacs and coworkers in nitrile hydratase (NHase) model complexes, where ZnII binding to the terminal oxo position has a significant impact on the Fe-S bond and the observed effect is very similar to that of proton addition in a smaller extent.267 In cancer cells, protonation would be the primary source for activating the sulfenato ligand, however 87  recent studies shows a higher accumulation of mobile ZnII ion in breast cancer cells.286,287 Therefore, Lewis acid activation of sulfenato species cannot be neglected. Nevertheless in either case, activation toward ligand substitution should occur through the mono-oxygenated sulfenato species of the complex.   Scheme 3.1: Proposed pathway for DNA binding by RuII arene Thiolate, 1, [(η6-ar)Ru(en)(SR)]+ complexes  It should be pointed out here that the reactivity of sulfenato and sulfinato adducts are clearly important for understanding the factors which affect the metabolism and distribution of the primary class of RuII arene [(η6-ar)Ru(en)(X)]+ complexes, where X = Cl, Br or N3, generally a labile ligand. Interestingly, the chloride complexes of this family have been shown to form adducts with natural thiols such as cysteine,288 glutathione95 and albumin253 outside and inside cells. Ru-S coordination appears to induce the oxidation of thiolates to sulfenates or sulfinates95,252,253 and subsequently stabilize the sulfenato ligands,76,95 often unstable free ligands and highly reactive.281,282 Taken together these biochemical studies in combination with our results, allowed us to develop a global model (Scheme 3.2) that explains the biochemical mechanism of action for the primary class of RuII arene complexes (in addition to hydrolysis reaction). The Ru-S bond strength is remarkably similar in complexes 1, 2 and 3, 88  which explains why protein sulfenic and sulfinic acids289,290 can compete with thiols as the preferred metal binding site in such complexes.   Scheme 3.2: Postulated global model for biological mechanism of primary RuII arene, [(η6-ar)Ru(en)(X)]+ complexes 3.5 Conclusions The effect of thiolato oxygenation on the nature of the Ru-S bonds, and the influence of arene and thiolato ligand modification on this reaction have been effectively investigated using X-ray absorption spectroscopic and density functional theory computational analysis, in a series of Ru arene anticancer complexes. Metal complexes used for therapeutics are often prodrugs. They frequently undergo ligand exchange or redox reactions before they reach the cancer cells. Detailed knowledge about such activation processes is vital to the drug 89  design process as well as the construction of meaningful structure-activity relationships for metallodrugs. In this study, we propose a modified mechanism of activation that is ligand-centered, and it is therefore important to understand how oxidation of ar-Ru-SR (thiolate) bonds to ar-Ru-S(O)R (sulfenato) and ar-Ru S(O2)R (sulfinate) affects the reactivity of these complexes. Herein we have recognized that the total charge donation from the sulfur ligand to the metal center is not altered by the thiolato ligand oxygenation. Therefore, the ligand oxygenation should have little impact on the lability of the ligand directly. Our work therefore suggests a need for an additional activation step for biological activity. Insight analysis of the Ru-S bonding in the sulfenato species indicates that this bond should strongly be affected by protonation (H+ addition) and/or Lewis acid activation of the terminal oxo group (Figure 3.11). Sulfur ligand oxidation also provides an additional pro-drug activation pathway to the primary RuII arene complexes. This is further supported by the competitive reaction of GSH/cGMP with parent [(η6-ar)Ru(en)Cl]+ complex, where the complex is directly coordinated to GSH followed by oxidation of thiolato lignd in the coordinated GSH. The oxygenation appears to provide a facile route for displacement of S-bound glutathione by G N7, leading to formation of the cGMP adduct [(η6-ar)Ru(en)(cGMP-N7)]+ as the major product even in the presence of a 250-fold molar excess of GSH.95 We further noticed that the slight variation of arene (p-cym vs hmb) and thiolato ligand (iPr vs Ph) on the thiolato complexes seem to be amplified in the oxidized complexes as evidenced by larger changes in charge donation of individual atoms besides the energies of the empty valence orbitals. Thus, small modifications in the parent complexes can lead to 90  major effects in the oxidized species, a finding which is significant in optimizing the design of this family of antitumor complexes in the future. 91  CHAPTER 4. ACTIVATION OF OXIDIZED SULFUR LIGANDS AND ITS IMPACT ON THE REACTIVITY OF RUTHENIUM (II) ARENE COMPLEXES 4.1 Introduction Ruthenium (II) arene complexes are attractive targets for non-Pt metallodrugs.  They offer a versatile platform for anticancer drug design and exhibit a distinct mode of biological activity. In earlier work (Chapter 3), we have shown that sulfur oxidation is an essential but insufficient step for the activation of RuII arene thiolate, [(η6-ar)Ru(en)(SR)]+, prodrugs.96 This proposal was developed based on the electronic structure of these complexes upon sulfur oxygenation and supported by the observed hydrolysis chemistry of the sulfenate complexes under acidic conditions (Scheme 4.1).76 It is therefore highly feasible that combined oxidation and protonation are required for efficient substitution of thiolato ligands by guanine N776 and DNA binding.95   Scheme 4.1: Protonation of complex 2 followed by hydrolysis under acidic condition (HCl).76 Generally sulfenates are too reactive to be isolated291 and readily disproportionate to sulfinates and thiolates282 unless coordinated to a transition metal. Interestingly, the only Ru complex reported with protonated sulfenato on the oxygen atom (sulfenic acid) is stabilized by strong hydrogen bonds on the OH group.292 Petzold et. al. showed that protonation of 2b, [(η6-hmb)Ru(en)(SOiPr)]+, is possible in aqueous HCl solution at pH 2.11 (pKa of the 2 H -H H2O/Cl - -HOS-R' Ru H2N NH2 S R' O R Ru H2N NH2 S R' O R H 2 Ru H2N NH2 X R 2+/+ X=H2O/Cl2H 92  protonated sulfenic acid is 3.61). The reaction is completed in 16 hours with partial hydrolysis of 2b. The reaction yields the reactive aqua adduct, [(η6-hmb)Ru(en)(H2O)]2+and chloride complex, [(η6-hmb)Ru(en)(Cl)]+ (Scheme 4.1).76 As shown in Figure 3.11, the DFT model calculations of complex 2a indicates that protonation of the oxygen should lead to a decrease in the Ru-S covalency, which presumably weakens the Ru-S bond. We wished to test this hypothesis as a means of confirming the proposed activation mode of oxidized sulfur in RuII-arene complexes. Unfortunately, the protonated species have not been isolable due to instability of the resulting complexes.76 Therefore, isolating the protonated sulfenato species for further spectroscopic studies is highly impractical. However, the influence of Lewis acids on the metal-sulfur bonding has been used as an indirect method of evaluating the effect of perturbations at the sulfur-oxygen bond in such complexes. There has been a growing interest in studying redox-inactive metal ions (such as ZnII) serving as Lewis acids to evaluate their potential impact on the reactivity of biologically-relevant metal complexes.293 Kovacs, Solomon and co-workers201,267 investigated the effect of direct interaction of ZnII on an iron-sulfenato complex, [FeIII(ADIT)(ADIT-O)]+, where ADIT is 6-amino-2,3-dimethyl-4-azahex-3-ene-2-thiolate. Crystallographic data of [FeIII(ADIT)(ADIT-O-ZnCl3)] confirmed that sulfenato oxygen is bound to the ZnII centre and due to this interaction the S-O bond is slightly elongated. Furthermore, the electronic absorption spectrum of the [FeIII(ADIT)(ADIT-O)]+ complex is perturbed in the presence of ZnII. Specifically, the lowest energy ligand to metal charge transfer (LMCT) band experiences a bathochromic/red shift – a similar result is also observed upon protonation of that 93  complex.267 The effect of other Lewis acidic metal ions such as Na+,268,294 Li+,295 and Pt2+,296 with the metal-sulfenato oxygen has also been explored. As discussed in Section 3.4, we have developed a model for the biological mode of action of RuII arene complexes, which suggests that the mono-oxygenated sulfenato species is most susceptible to ligand exchange under acidic conditions.96 We thus seek to explore the effect of exogenous metal ions on the bonding in RuII arene complexes as a means of indirectly probing the effect of protonation on these species. Particular attention is paid to sulfenato species due to their importance in the proposed mode of action, although the effect on sulfinato species is also explored for comparative purposes. In addition, sulfinato species is also reported as a co-product with sulfenato species during the biological reaction of parent RuII-arene chlorido complex, [(η6-ar)Ru(en)(Cl)]+, with glutathione (GSH)252 and human albumin.253 This chapter explores the effect of protonation and Lewis acid complexation on complexes 2a ([(η6-p-cym)Ru(en)(SOiPr)]+) and 3a ([(η6-p-cym)Ru(en)(SO2Ph)]+ as shown in Scheme 4.2 and Scheme 4.3 respectively. Synthesis of the sulfenato-Lewis acid adducts has been more challenging, and as a result, the Lewis acid effect of sulfenates has been investigated using in silico models. However, an analogous series of Lewis acid (ZnII, CoII and CrIII) adducts of sulfinato complexes have been successfully synthesized,** which made it possible for both experimental and computational studies on these systems to be conducted.  **This chapter is based on the work collaborated with Prof. Sadler and co-workers Drs. Petzold and Habtemariam from Department of Chemistry, University of Warwick, Coventry, UK. 94  Importantly, data available for the sulfinato complexes provide a useful means to evaluate the computational models used to probe the more reactive sulfenato species.   Scheme 4.2: Brønsted acid protonation (a)76 and Lewis acid interaction (b) of RuII arene sulfenato (2) complex, where R = p-cym, R’ = iPr   Scheme 4.3: Brønsted acid protonation (a) and Lewis acid interaction (b) of RuII arene sulfinato (3) complex, where R = p-cym, R’ = Ph 4.2 Experimental and Computational Aspects Herein, the preparation of solid and solution samples and data collection at the CLS beamline SXRMB as well as BL 4-3, SSRL are described. Data processing and analysis are accomplished using the following software packages: SIXpack271 and BluePrint XAS.272,273 All DFT and TDDFT calculations in this Chapter are performed using the ORCA220 program. MOAnalyzer297 is used to analyze the results from the DFT and TDDFT calculations. Furthermore, it is also used to compare the XAS and simulated TDDFT spectra.  95  4.2.1 Sample Preparation Complexes [(p-cym)Ru(en)(SOiPr)]+- 2a, [(hmb)Ru(en)(SOiPr)]+ - 2b and [(p-cym)Ru(en)(SO2Ph)]+ - 3a was initially prepared by Dr. Holm Petzold using a previously published procedure.76 Lewis adducts of 3a, i.e. 3aZn2+, 3aCo2+, and 3aCr3+ were prepared by Dr. Abraha Habtemariam as follows. A solution of 1 mM concentration of 3a was prepared in doubly deionized water. Then, 1 mL of aliquots  was added to 0.6 mM (3 mole equivalent) of the 5 ml of Zn(NO3)2, CoBr2 and CrCl3 solutions to prepare 3aZn2+, 3aCo2+, and 3aCr3+ respectively. They were mixed well by shaking for couple of minutes to get a clear solution and freeze dried immediately. XAS data at the S K- edge were collected on these freeze-dried samples. XAS spectra were not collected at the Zn K-, Co K-, or Cr K-edges due to their presence in large excess. Samples were mounted as finely ground powders homogeneously diluted with 25% BN and spread evenly on sulfur-free double sided carbon tape attached to a long copper sample holder. Reference samples were placed on the top and bottom of the holder to check the monochromator readings. A solution sample of 2b prepared in 1mM concentration, was buffered (with PBS - phosphate-buffered saline) at pHs 7.4 and 2.5. 1 M HCl solution was used to adjust the pH of the buffer. The solution was mixed with ~40% glycerol, a glassing agent, to reduce diffraction produced by ice crystals in XAS. Freshly prepared samples were pre-equilibrated in a water-saturated He atmosphere for an hour to minimize bubble formation in the sample cell. Then, the solution was injected into XAS cells sealed with a polypropylene window on the front, and rapidly frozen in liquid nitrogen.  96  4.2.2 XAS Data Collection XAS measurements of S K-edge were performed at soft X-ray micro characterization beamline (SXRMB) of the Canadian Light Source (CLS), using a Si(111) double-crystal monochromator with the higher-order harmonics rejected with Pt-coated mirrors under standard operating conditions of 2.9 GeV and 250 mA beam current.  Detailed descriptions of beamline configuration and data acquisition are described in section 2.1.5.3 and are also found in literature.211  Energy calibration of S K-edge was performed using sodium thiosulfate (Na2S2O3) with the first pre-edge feature being calibrated at 2472.02 eV. Reference samples were scanned at the beginning and end of each sample to ensure stable monochromator readings.  Signal was recorded under vacuum with the surface sensitive total electron yield (TEY) which was complemented by the bulk sensitive total fluorescence yield (FLY) detected with a channel plate detector.  Three scans were taken for each sample and the experiment was repeated twice to confirm reproducibility.  Solution sample data of 2b was collected at BL 4-3 at SSRL according to the procedure described in Section 3.2.2. 4.2.3 Data Processing and Analysis Sulfur K-edge XAS data were analysed using SIXpack.271  Energy calibration was reconfirmed via the inflection point of the first derivative of the lowest energy pre-edge peak of Na2S2O3 (=2472.02 eV).  All acceptable scans were averaged and fit to a linear background and subtracted from the entire spectrum and normalized to an edge jump of 1.0.  BluePrint XAS version 1.1,272,273 was used to simultaneously fit the background, edge jump, and spectroscopic features. Peaks in the spectrum were modelled using pseudo-Voigt functions, 97  whereas edges are modelled using cumulative pseudo-Voigt functions. Restrictions are set for each fit parameter but actual fit parameters are assigned using a Monte Carlo algorithm to reduce the user bias in analysis. For each spectrum, 100 sets of initial fit parameters were generated, yielding 100 unique fits. Fit results can be investigated for robustness (relative to initial fit parameters) and the best and most representative fits can be filtered. 4.2.4 DFT Calculations Initial structures for geometry optimization of all complexes were obtained by modifying the solid-state molecular structures of 2a76 and 3a96 using MOLDEN298, a general pre- and post-processing program for computational chemistry. DFT calculations were performed using ORCA220 version 2.9, a combined ab Initio, DFT and semiempirical electronic structure package. Geometry optimizations were obtained using the BP86226,227 exchange correlation functional and a TZVP + def2-TZVP/J basis set was used for all light atoms; the RI-J approximation was employed for faster convergence. The TZVPP+TZVPP/J basis set was used for ruthenium metal. Scalar relativistic effects were evaluated using the built-in ZORA module within ORCA 2.9.235,276 Relativistic effects were applied to the auxiliary (/J) basis sets. A dense integration grid (grid4) was employed for good numerical accuracy and tight geometry convergence criteria (TightSCF) were enforced. Due to the overall size of the computational model, convergence was accomplished with damped convergence (using the SlowConv keyword). Solvation was included (where applicable) with COSMO, as implemented within ORCA.237 The molecular structures obtained from geometry optimization were utilized for “single point” (i.e., with fixed atomic positions) TDDFT simulations of sulfur K-edge XAS spectra.  MO 98  Analyzer version v1.1beta297 (a Matlab-based numerical and graphical analysis toolbox for ORCA output file) was used to visualize simulated TDDFT spectra and to compare these with experimental data.  A systematic energy shift of ~ 60.4 eV was applied to the spectra to align with experimental XAS. This shift is consistent with the calibration value (60.01 ± 0.11 eV) reported by Debeer et. al. for TDDFT of the S K-edge calculation done with BP86 functional + ZORA scalar relativistic effect.299 Natural bond orbitals (NBO) analysis was performed using the NBO version 5.0 program.242  Due to limitations within the NBO program, full BP86 calculations using a smaller SVP basis set was used.  Although quantitative differences exist between SVP and TZVP results, the qualitative bonding description and trends are the same. TightSCF was applied for single point convergence and VeryslowConv applied for SCF convergence. Examples of input files for each type of computation performed in this Chapter are given in Appendix B.2. 4.3 Results and Analysis Given the experimental challenge of directly probing the protonated forms of 2 (2H+) and 3 (3H+), computational studies have been used to evaluate the effect of terminal oxo-group protonation in these species. Attempts were made to experimentally evaluate the effect of an acidic environment on 2b (vide infra). As the preparation of Lewis acid adducts of complex 2 has been unsuccessful to date, both the effect of protonation and that of Lewis acids on RuII arene sulfenato complexes has been explored using computational studies. By contrast, Lewis acid adducts of RuII arene sulfinato (3) complexes have been investigated both experimentally and computationally. In addition, solvation effects of water are investigated 99  in silico for sulfenato, 2a → 2aaq, and sulfinato, 3a → 3aaq, complexes. Results obtained for the sulfinato complexes are described first, followed by that for the sulfenato species. 4.3.1 Lewis acid Interaction and Brønsted Acid Protonation on Sulfinato The effect of solvation and protonation on the sulfinato complexes is investigated using in silico models of 3aaq and the RuII arene sulfinic acid derivative, 3aH+ using DFT/COSMO calculations respectively. In addition, the effect of Lewis acids on sulfinato complexes is evaluated using a combination of spectroscopic and computational data. Table 4.1: Selected bond length (Å) of 3a, 3aZn2+ and 3aH+ from the DFT analysis Complex Ru-S Bond length  S-O Bond length 3a 2.408 1.531, 1.535 3aaq  2.383 1.539, 1.542 3aZn 2+  2.297 1.594 3aH+  2.321 1.516, 1.689  The computational results for solvated (3aaq), protonated (3aH+) and ZnII-perturbated (3aZn2+) complex 3a show that solvation and perturbation result in slight structural changes to those complexes. Elongation of the S-O bond and concomitant shortening of the Ru-S bond occur as shown in Table 4.1. Importantly, either one or both of the S-O bonds can undergo perturbation. In the case of protonation, only the first protonation is explored, which elongates the S-O bond directly attached to proton. This concomitantly shortens the other S-O bond which also mitigates the effect on the Ru-S bond. On the other hand, ZnII can, in principle, bind to either one or both oxygens. Overall, it is found that a single Zn(II) binding to both terminal oxo groups (3aZn2+) yield results that are most consistent with the 100  experimental XAS data.†† In this model, both S-O bonds are elongated whereas the Ru-S bond is shortened. It has been postulated that perturbation of these oxidized species would directly affect the Ru-Sσ* bond. Therefore, the charge contribution of Kohn-Sham acceptor orbital, Ru4dσ*, is investigated in detail to study the solvation and protonation effect of sulfinato complexes as shown in Table 4.2. Table 4.2: Experimentally determined S3p character* and calculated charge distribution of acceptor (Ru4dσ* ← S3p) orbital in sulfinato complexes Complex % S3p in acceptor SO 2 R(%) H/M(H 2 O) 2 (%) Ru(%) ar(%) en(%) XAS DFT 3a 6.3 ± 1.3% 6.2 23.1 - 45.7 21.5 6.0 3aaq - 5.5 16.9 - 45.9 23.4 9.7 3aZn 2+  5.8 ± 1.5% 5.2 12.7 1.3 45.6 25.1 11.4 3aH+ - 5.8 11.3 0.0 43.7 25.7 15.6  * DFT-calculated S3p contribution to the acceptor orbital, Ru4dσ* ← S3p, is determined from a Mulliken charge decomposition of the Kohn-Sham orbital. Zn2+ accompanied with two H2O molecules and bound to both oxygen of the sulfinato to complete the tetrahedral coordination sphere. Löwdin analysis of the Kohn-Sham orbitals shows that solvation and perturbation of the sulfinato oxygen via either protonation or a Lewis acid directly affects the S-O bonding. The observed electronic changes also reflect changes in bonding due to solvation and perturbations of the S-O bond. Surprisingly, there is only a small decrease in the S3p contribution to the Ru4dσ* orbital, the overall contribution from the sulfinato ligand to Ru4dσ* orbital substantially decreases due to both solvation and perturbation, as in Table 4.2. These  †† TDDFT simulated XAS spectra of 3aZn2+.2H2O (where ZnII binds to both sulfinato oxygens) and 3aZn2+.3H2O (where ZnII binds to one of the sulfinato oxygens) were compared with experimental XAS of 3aZn2+. This indicates that the binding mode for ZnII with 3a appears to be 3aZn2+.2H2O 101  changes are mostly compensated by increased charge donation from the ethylenediamine (en) and arene ligand but not necessarily affected the charge of the metal (Ru) centre. In contrast, the slight decreases in the S3p contribution do not agree with the shortened Ru-S bonds in 3aaq, 3aZn2+ and 3aH+ during solvation, perturbation and protonation respectively. This would lead to suggest that the nature of the Ru-S bond in 3a is dominated by electrostatic interaction between the RuII and sulfinato (RSO2-) group rather than the covalent bonding interaction of S3p charge donation, Ru4dσ* ← S3p.  Figure 4.1: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 3a, 3aaq, 3aZn2+ and 3aH+ In addition, the NBO derived bond order of the Ru-S bond is found to be 0.28 in 3a; this indicates that the covalent Ru-S interaction in 3a is weaker than the Ru-S bond in 2a (where NBO bond order is 0.74). By contrast, the NBO analysis of Ru-S bonding in 3aaq, 3aZn2+ and 3aH+ predicts that there is no actual bonding between Ru and S and therefore the pictured Ru-Sσ* orbital in Figure 4.1 is listed as lone-pair with occupancy of 1.19, 1.35 and 1.32 respectively. This is also seen in Ru sulfinato complexes by Shearer et al.204 The slight increase in occupancy suggests increasing electrostatic contributions to a Ru-S bond that is already dominated by electrostatics. 102  As shown in Figure 4.2, an additional approach – comparing the relative energies of the relevant molecular orbitals – is applied to study the effect of solvation and perturbations of the sulfinato oxygen(s). The greatest perturbation is observed in the relative MO energies when comparing 3a, 3aaq, 3aZn2+ and 3aH+ (Figure 4.2). Solvation and protonation leads to a dramatic decrease in the energy of the SOσ*, reflecting a weakening of the S-O bond. However, the relative energy of Ru-Sσ* remains essentially unchanged due to perturbation of acids, presumably because there is no significant SOπ* interaction found in the Ru4dσ* of sulfinato complexes (also see Figure 3.7). Overall, the effect of perturbation is greater than just aqueous solvation.  Figure 4.2: Frontier MO diagram of complexes 3a, 3aaq, 3aZn2+ and 3aH+. Only α orbitals are shown but α and β orbitals are identical. Red and green arrows indicate the energy change in the SOσ* and Ru-Sσ* with protonation respectively. Orbitals are normalized to LUMO in each molecule.  E n e rg y (e V ) 103  The effect of Lewis acids on the sulfinato complexes was evaluated both experimentally and computationally. The sulfinato-Lewis acid adducts 3aZn2+, 3aCo2+ and 3aCr3+ were investigated using Sulfur K-edge XAS as a means of determining the effect of these metal ions on the sulfur atom and its interactions. The S K-edge XAS data, Figure 4.3a, indicate that the RuII arene sulfinato complex 3a is dramatically influenced by the Lewis acid interaction. The pertubation effect on these complexes is distinct by the nature of Lewis acidity and coordination chemistry of Zn2+, Co2+ and Cr3+.300,301  Figure 4.3: (A) Normalized S K-edge XAS of complexes 3a, 3aZn2+, 3aCo2+ and 3aCr3+, (B) XAS and simulated TDDFT of complex 3aZn2+.  Figure 4.4: The Ru-Sσ* orbital of 3aCo2+ and 3aCr3+ showing the interaction of metal 3d with S 3p and Ru 4d orbitals 104  The relatively featureless high energy feature of 3a, assigned to the combined SC/SO←S1s transitions (2478.0 - 2479.0 eV), split into two well-resolved features upon addition of Zn2+. This effect is similar to that which is observed in DFT calculations, which suggest that Zn2+ binding lowers the energy of the SOσ*. A similar effect is observed with Co(II) and Cr(III), although the analysis is much more complex due to the presence of empty 3d orbitals, allowing for greater S 3p delocalization, shown in Figure 4.4. As discussed in Section 2.1.4, the intensity of the Ru4dσ*←S1s transition is proportional to the amount of S3p character (β) in the Ru4dσ* acceptor orbital, as given in equation 2.2. A systematic peak fitting and assignment (shown in Appendix A.2) were performed on the S K-edge XAS of 3aZn2+ to extract the S3p character using the dipole integral value of 〈SO2R¯3p|r|S1s〉 ≃ 18.185,187 As shown in Table 4.2, the analysis indicates that there is little change in the S3p contribution and therefore, consistent with that of DFT analysis. Furthermore, the TDDFT simulation of the 3aZn2+ XAS is in good correlation with the experimental S K-edge XAS, Figure 4.3b. 4.3.2 Lewis Acid Interaction and Brønsted Acid Protonation on Sulfenato The model system 2a was used for all computational studies on the sulfenato species, with a primary focus on the in silico solvated (2aaq), protonated (2aH+) complexes and the perturbated ZnII adduct (2aZn2+). The ZnII adduct was modelled as a tetrahedral ZnII ion with three explicit water ligands and the 4th coordination position being taken by the sulfenato oxygen from complex 2a (a tetrahedral coordination of ZnII with oxygen(s) is indicated by XAS and TDDFT results of 3aZn2+). The aqueous solvation and acidic perturbatory effect on the overall geometric structure of 2a is generally quite small and localized to 1-2 bonds from the 105  terminal oxygen atom. The interactions trigger elongation of the S-O bond and that results concomitantly in a slight contraction of the Ru-S bond (Table 4.3). A similar trend is also observed in the Lewis acid adduct of FeIII-sulfenato complex, [FeIII(ADIT)(ADIT-O-ZnCl3)].201,267  Table 4.3: Selected bond length (Å) of 2a, 2aaq, 2a-Zn2+ and 2aH+ from the DFT analysis Complex Bond length Ru-S Bond length S-O 2a 2.442 1.589 2aaq  2.396 1.618 2aZn 2+  2.384 1.677 2aH+  2.359 1.724  Table 4.4: DFT-calculated S3p contribution* (Ru4dσ* ← S3p) and the total charge distribution of Kohn-Sham acceptor orbital, Ru4dσ* of 2a, 2aZn2+ and 2aH+ complexes Complex % S 3p in acceptor SOR (%) H/M(H2O)3 (%) Ru (%) Ar (%) En (%) 2a 12.7 20.9 - 41.1 20.0 6.1 2aaq  11.2   18.9  - 42.6  20.5   6.1 2aZn 2+  10.6 17.0 6.2 46.4 20.5 6.4 2aH+  9.2 14.9 0.4 48.3 19.5 6.1  *DFT-calculated S3p contribution to the acceptor orbital, Ru4dσ*←S3p, is determined from a Mulliken charge decomposition of the Kohn-Sham orbital. Cation Zn2+ is accompanied with three H2O molecules to complete the tetrahedral coordination sphere. As with sulfinato complexes, the Löwdin analysis of the Kohn-Sham orbitals shows that aqueous solvation of the complex and perturbation of the sulfenato oxygen via either protonation or a Lewis acid directly affect bonding. Solvation and both perturbations moderately decrease S3p contributions in the Ru-Sσ*, reflecting a decrease in donation into the metal centre. The effect is greater in 2aH+ as compared to 2aZn2+, which reflects the greater influence of protonation. The overall charge donation from the S-ligand (SOR) to that 106  acceptor orbital is also decreased, again with a larger effect being observed for 2aH+ than for 2aZn2+. These differences are consistent with protonation of the oxygen atom having a greater influence than ZnII coordination. Contrasting to sulfinato, the observed reduction in overall charge contribution from the sulfenato ligand does not result in compensatory charge donation from the other ligands (i.e. arene and en).  Figure 4.5: Natural bond orbital (NBO) picture of Ru-Sσ* in complexes 2a, 2aaq, 2aZn2+ and 2aH+  A natural bond orbital (NBO) analysis for 2a, 2aZn2+ and 2aH+ was used to estimate bond orders and visualize the localized Ru-S bonding (the NBO of Ru-Sσ* antibond is shown in Figure 4.5). For complexes 2aZn2+ and 2aH+, the NBO bond order analysis suggests no formal covalent bonding between the Ru and S atoms, as compared to 0.74 and 0.72 of a bond in the parent 2a (see Section 3.3.2) and 2aaq complexes respectively. This qualitatively suggests that solvation followed by protonation or the presence of a Lewis acid is sufficient to weaken the Ru-S interaction as predicted. Another approach to defining the effect of perturbations of the sulfenato oxygen is to compare the relative energies of the relevant molecular orbitals. Frontier molecular orbital diagrams for 2a, 2aaq, 2aZn2+ and 2aH+ are shown in Figure 4.6. Most notably, perturbation 107  of the sulfenato oxygen lowers the energy of both the Ru-Sσ* and SOσ* empty orbitals. As anticipated, the effect is significantly greater for the SOσ* than for the Ru-Sσ*.  Figure 4.6: Frontier MO diagrams for complexes 2a, 2aaq, 2aZn2+ and 2aH+. Only α orbitals are shown but α and β orbitals are identical. Red and green arrows indicate the energy change in the SOσ* and Ru-Sσ* with protonation respectively. Orbitals are normalized to LUMO in each molecule.      Figure 4.7: (A) S K-edge XAS of 2b and the first scan of 2b solutions at pH 2.5 and pH 7.4; (B) simulated TDDFT of 2b, 2baq and 2bH+. An energy shift of 60.0 eV applied to simulated spectra to calibrate them with experimental data. E n e rg y (e V ) 108  Attempts to probe the effect of protonation through experiment were more challenging. Sulfur K-edge XAS data of complex 2b were obtained at pH 7.4 and 2.5. The data are shown in Figure 4.7a in comparison to that which is obtained for 2b in the solid state. Photoreduction of the solution samples was a major problem, thus data obtained from single rapid scans are shown. Data are significantly poorer at the lower pH, but there is no indication of any major changes in the S K-edge XAS spectrum. However, protonation of the sulfenato oxygen should have a major effect on the spectrum, as seen from simulated spectra in Figure 4.7b. By contrast, however, a rather large difference between the solution and solid state spectra were observed. The lowest energy feature of the spectrum (the Ru4dσ* ← S1s transition) drops by ~1.5eV upon solvation. Such large effects are not observed in silico using the COSMO solvation model, suggesting that the solvation model does not adequately reflect changes in the localized bonding upon solvation. However, there is a slight decrease in the S3p character contribution, Table 4.4 and the energy of the Ru-Sσ* orbital (Figure 4.6) due to the solvation. The Frontier MO diagrams of these complexes, Figure 4.6, show that the energy of S-Oσ* and Ru-Sσ* orbitals decreases upon solvation/protonation. In the simulated TDDFT spectra, only the S-Oσ* shifted to lower energy. However, the energy of S-Cσ* orbital is not necessarily changed due to the same effects; thus in the S K-edge spectra the higher energy peak at 2476.2 eV, assigned to SCσ* ← S1s transition, does not affected by solvation. Simulated TDDFT spectrum of 2baq is in closer agreement with the solvated spectra 2b (pHs 7.4 and 2.5) as seen in Figure 4.7. Therefore complex 2b is solvated in pHs 7.4 and 2.5 but the protonated species is not identified in the lower pH. 109  4.4 Discussion DFT and XAS studies performed on the solvated, protonated and perturbed RuII arene sulfenato and sulfinato complexes show that both solvation and perturbation of the terminal oxo groups redistribute S3p character of the sulfur-containing ligands. Unsurprisingly, solvation and perturbation of the sulfenato oxygen affect both the Ru-S and S-O bonds without a strong effect on the S-C bond. The Ru-S bond in the sulfenato complexes is thus quite sensitive to relatively small perturbations. Moreover, protonation of the sulfenato ligand results in significant changes in bonding. By contrast, the effect of solvation and other perturbations on sulfinato complexes is more limited and only affects the S-O bond with little or no effect on the Ru-S bond covalency. It should be noted here that the effect of Lewis acid on 3a appears to be slightly higher than the effect observed in 2a; however in 3a the ZnII ion interacts with both oxygen of the sufinato ligand, thus the total effect seems to be higher but the average effect caused by a single oxygen would be slightly lower to that caused by 2a. These results are in agreement with the idea that protonation of the sulfenato should have a strong impact on the strength and lability of the Ru-S bond whereas protonation of the sulfinato has little or no impact on the strength and lability of the Ru-S bond. Interestingly, aqueous solvation partially activates the RuII arene sulfenato and sulfinato complexes by lowering the energy of the Ru-Sσ*. Complex 2 and 3 have intra (SO--H2N) and inter H-bonds between neighbouring en ligand and solvent MeOH (or in the complex 3 the neighbouring dimer molecule) respectively. These two H-bonds are removed or weakened due to solvation effects. Therefore, in the H-bond free-complex 2, the Ru-S bond is elongated which concomitantly decreases the covalency and makes the molecule more 110  labile. However, this effect is minimized in complex 3 due the nature of the Ru-S bond (vide infra). As discussed in Chapter 3, formation of the sulfenato species via ligand oxygenation is a necessary but insufficient step for biological activation of these complexes. However, the Ru-S interaction in the sulfenato complex is quite different due to the more pronounced pi contributions in the S-O bond, Figure 3.11. Due to this unique property, protonation of the sulfenato oxygen directly weakens the covalency of Ru-S bond and also the H-bond between the sulfenato group and the NH2 of the en ligand, as observed in the hydrolysis of sulfenato complexes.76 These results agreed with the postulated mechanism in Scheme 3.1. In contrast, the Ru-S bond in the sulfinato does not involve pi contributions to a terminal oxo. Therefore, protonation of the sulfinato species has only a minor effect on the covalency of Ru-S bond. Protonation of the sulfenato is therefore observed to be an effective method of labilizing the Ru-S bond, specifically by weakening the S-O π-bond in the ligand. In addition, a recent study on breast cancer tissues reveals a positive association between the amount of mobile ZnII and cancer.286,287 Therefore, ZnII can act as a Lewis acid for the protonation of the sulfenato complexes in cancer cells. 4.5 Conclusions The effect of solvation, protonation and perturbation of sulfenato and sulfinato complexes is investigated using X-ray absorption spectroscopy and density functional theory. Examining, understanding and controlling the biological mechanism of the metal prodrugs are important for their faster pre-clinical and clinical development. Our earlier study (described in Chapter 3) shows that sulfur ligand oxygenation does not necessarily change 111  the charge distribution to the metal center and therefore does not directly have impact on the lability of the ligand. The present study shows that the second and most important step, solvation followed by protonation of the sulfenato complexes, decreases the charge donation from the sulfur ligand to the metal center. Therefore, it does play a vital role on the lability of the ligand. Therefore, activation mechanism of Ru arene thiolato complexes is unique and leads through the ligand-centered oxidation followed by solvation and protonation.  112  CHAPTER 5. SPECTROSCOPIC INVESTIGATION OF RUTHENIUM (II) AND OSMIUM (II) ARENE THIOSEMICARBAZONE COMPLEXES 5.1 Introduction Thiosemicarbazones (TSCs) are one of the Schiff base motifs that have been extensively studied for a broad spectrum of chemotherapeutic properties over the past decade.302,303 In medicinal chemistry, TSCs have emerged as an important class of ligands due to their variable donor properties, structural diversity, and biological applications.303 Interestingly, it has been postulated that inside the cells, thiosemicarbazone (TSC) ligands coordinate to metal ions to form complexes and the redox properties of those complexes would be responsible for their biological properties.302 In addition, studies have shown that the biological properties of TSCs can be modified and in fact enhanced, by metal ion binding.302 Substituted and metal coordinated heterocyclic TSC complexes have shown a wide range of medicinal properties including cytotoxic and antitumor,304 antibacterial,305,306 antiviral,307 anti-trypanosomal308 and anti-alzheimer309 properties. One of the first concepts applied in developing organometallic RuII anticancer drugs is to combine metal-arene moieties with bioactive ligands. In fact, the first organometallic ruthenium complex, [Ru(η6-bz)Cl2(metronidazole)], tested for anticancer activity involved such a combination.49 Since then, the use of non-cytotoxic ligands, i.e. ethylenediamine and PTA, with metal-arene systems has led to complexes with significant anticancer activity such as the RM175-type and RAPTA-family complexes. In recent years, however, researchers have revisited the original strategy and generated highly efficient anticancer drugs with different modes of activity such as paullones,116 ethacrynic acid,73 and staurosporine.310 A series of 113  biologically potent TSC ligands hybrid with RuII and OsII arene complexes have been recently developed by Sadler and coworkers (see Figure 5.1).  The anticancer activity, cellular accumulation and mode of activity of these complexes are currently under investigation.311  Figure 5.1: RuII/OsII-arene thiosemicarbazone complexes investigated in this Chapter A number of bonding modes have been observed for TSCs in their neutral or anionic forms with metals.303 Recently, Beckford and coworkers reported RuII-arene complexes coordinated with neutral TSC as a bidentate ligand, which possess both anticancer and antimicrobial properties.312–316 Fluorescence studies of these complexes with human serum albumin (HSA) suggest their interactions with biological molecules;313 however, there is little discussion of structure activity relationships or plausible mechanisms of action. Other studies show that TSCs can also coordinate to the metal in a monodentate fashion through the sulfur atom.317,318 Newly developed half sandwich mononuclear and sandwich dinuclear RuII, OsII-p-cym-TSC complexes, Figure 5.1, studied herein lack good quality crystal structures. As a result, they are characterized using NMR, mass spectrometry, infrared, and UV-vis spectroscopy.311 It is therefore necessary to investigate their electronic structure to confirm 114  the geometry and the coordination fashion of the TSC ligand. This investigation will further be useful to investigate the biological mechanism of these complexes in detail. Ru(II) and Os(II) arene complexes with bidentate chelating ligands, developed by Sadler and co-worker, show promising anticancer activity in a range of cancer cell-lines.3,28,52,69 The mode of action of these complexes has been comprehensively investigated and a concensus mechanism has been postulated.28,96,121 However, cellular speciation and distribution of such complexes is challenging. Ru metal accumulation in cells and their compartments can be determined by ICP-MS, but there is no confirmation on the ligand set of the complex as it was first administered. Therefore, the complexes studied herein have been synthesized with iodine-substituted thiosemicarbazone ligands, Figure 5.1. Here, the iodine (-I) label can be used for ligand cellular uptake studies. Thus, both the metal and the ligand can be traced and quantified inside cells. The ratio of metal to ligand can be used as an indicator about the state of the complex. The investigation of structure-activity relationships for Ru and Os arene complexes have shown that changing the metal centre, making a small modification on the chelate ligand91 or exchanging the monodentate ligand, e.g. from chloride to iodide, can have a major effect on biological activity.90,319 Therefore, in this study the effects of monodentate ligand (Cl vs I), metal centre (Ru vs Os) and the modification of the TSC ligand are investigated through their electronic structure. X-ray absorption spectroscopy combined with DFT and TDDFT calculations is used to investigate the TSC ligands as well as their Ru/Os-p-cym-TSC complexes.  115  5.2 Experimental and Computational Aspects 5.2.1 Experimental Aspects All the complexes studied here were prepared and characterized by Anna Louisa Noffke from the Sadler group‡‡ according to the procedure to be published in our collaborative manuscript.311 These complexes were characterized using NMR, electron spray ionization mass spectrometry, infrared spectroscopy and elemental analysis. XAS experiments on S K-edge were performed at the soft X-ray micro characterization beamline (SXRMB) of the Canadian Light Source (CLS).211 Samples for the experiments were prepared according to the procedure mentioned in Section 4.2.1. S K-edge XAS data was collected and analysed using the methods described in Section 4.2.2 and 4.2.3, respectively. 5.2.2 DFT Calculations The initial coordinates for the geometry optimization of TSCs 4a and 4b were obtained from ChemBio3D structures and that of complexes 5a-c, [(p-cym)RuII(TSC)X]+ (X= I/Cl) were obtained by modifying the X-ray crystal structure of simplified alkyl substituted [(p-cym)RuII(TSC)Cl] complex using MOLDEN.298 The coordinates for complexes 6a and 6b were obtained from the crystal structures. DFT calculations were performed using the ORCA220 quantum chemistry program, version 2.9. Geometry optimization was obtained using BP86 (Becke-Perdew 86) gradient-corrected functional,226,227 in combination with the TZVP basis set and def2-TZVP/J auxiliary basis sets.233 Combined TZVPP and TZVPP/J auxiliary basis sets were applied on the metal center Ru and Os. Scalar relativistic effects were  ‡‡ This chapter is based on collaborative work with Prof. Sadler and co-worker Anna Louisa Noffke from Department of Chemistry, University of Warwick, Coventry, UK 116  introduced using ZORA.320,321 A dense integration grid (grid4) and tight convergence criteria were enforced unless otherwise stated. The geometry optimized structures were utilized to perform single point TDDFT calculations of the XAS spectra. An energy shift of ~ 60.5 eV was applied to the spectra when it was constructed with experimental XAS for comparison using MO Analyzer (version 1.1 beta).297 Simulation of UV-vis spectra of 5b and 6a were done using BP86 and B3LYP +VDW10 (Van der Waals corrections) functionals. 5.3 Results and Analysis S K-edge XAS in concert with DFT and TDDFT calculations are extensively used to probe the electronic structure of complexes 4 - 6. Additionally, effects of substitution of the NMe2 group on the TSC pro-ligand, changing the metal centre from Ru to Os and the monodentate ligand from Cl to I are also investigated. In addition, an attempt was made to simulate the UV-vis spectra of 5b and 6a. 5.3.1 Effect of TSC Ligand Modification   Figure 5.2:  Normalized S K-edge XAS of unmodified (4a) and modified (4b) TSC ligands (a) and complexes 5a- c (b). * denotes the thiolate based impurities. TDDFT spectra of these complexes are provided in Appendix A.3.1. 117  The effect of thiosemicarbazone ligand modification is investigated in proligands (4a and 4b) and in metal-ligated complexes (5a and 5b) using S K-edge XAS. As shown in Figure 5.2a, the S K-edge XAS spectra of unmodified (4a) and modified (4b) proligands exhibit two dominant features in the near edge between 2470 - 2475 eV. Based on the literature280 and results from DFT calculations, these two features are attributed to SCπ* ← S1s (at 2471.5 eV) and SCσ* ← S1s (at 2472.3 eV) transitions. Furthermore, DFT also predicts there are three empty π* orbitals with sulfur character in the TSC proligand (also see the π* orbitals of the model complex in Figure 5.3). Transitions to all three of these π* orbitals contribute to the first features as shown in the TDDFT spectra of 4a and 4b (see Appendix A.1.3). S K-edge XAS data indicate that NMe2 substitution on the TSC proligand results in larger contributions from the S 3p into the π-system. In addition, S 3p contributions to the SCπ* and SCσ* final states also increase, in agreement with DFT calculations. The S K-edge XAS of unmodified (5a) and modified (5b) RuII metal complexes also exhibit two clearly resolved features in that region. The lowest energy feature in these spectra are much more intense than for the corresponding proligand due to the presence of the Ru4dσ* ← S1s transition. A closer look at the spectrum for 5b indicates the presence of low energy shoulder at 2471.5eV, which likely corresponds with one component of the SCπ* ← S1s features. This assignment is in good agreement with DFT calculations (see the Frontier MO diagram in Figure 5.4) Interestingly, our experimental and computational data suggest that the TSC possess an extended π-system, running through the backbone of the ligand. As anticipated, the π-system is preserved in the metal complexes as well. A simplified model complex, 118  [CH2=NH-CS-NH-N=CH2]+, of TSC is used to illustrate those π-orbitals as shown in Figure 5.3. The three π* orbitals listed in the MO diagram are the ones which contribute to the SCπ* ← S1s transitions in the S K-edge XAS. The simulated XAS spectra of TSCs show that these three transitions are added together in the first lower energy peak.   Figure 5.3: MO diagram of the π-systems of Thiosemicarbarzone molecule described using model complex [CH2=NH-CS-NH-N=CH2]+ 119   Figure 5.4: Frontier MO diagram of complexes 5a-c. Only α orbitals are shown (α and β are identical). Red arrow indicates the energy shift from 5a→5b. Orbitals are normalized to LUMO in each complex. As mentioned in earlier Chapters, the intensity of each of the experimental features in the S K-edge XAS spectra results predominantly from redistribution of the localized electric dipole-allowed S3p ← S1s transition to each of the final states, as defined by equation 2.2. Thus, the intensity of a particular spectroscopic feature depends on the amount of S3p character in that final state. Systematic fits to the data in each case allow for the extraction of experimentally derived S3p contributions to each of the features in the spectrum. The dipole integral (as in equation 2.2 〈S3p|r|S1s〉) used to estimate the S3p character from complex 5a, b, c is 120  〈SR¯3p|r|S1s〉 ≃ 13 and that of complexes 6a, b is ≃ 14. The results from these fits are given in Table 5.1. For comparison, results from density functional calculations are also included in the table. The S3p character, extracted from the XAS spectra for Ru4dσ* ← S1s and SCσ* ← S1s transitions, agrees well with that obtained from DFT; however, the S3p character, extracted from XAS for the SCπ* ← S1s transition is lower than that obtained from DFT. That might imply that the SCπ* ← S1s transition is weaker in these complexes.  Table 5.1: S K-edge XAS peak energies and transition assignments with experimentally estimated S3p character in the acceptor orbitals from normalized fit intensities compared with calculated DFT value Comp S K-edge XAS Transition % S3p in Acceptor orbital Energy (eV) Assignment XAS DFT SCπ* Ru4dσ* 5a 2472.3 SCπ* + Ru4dσ*  16.0 ± 1.4%  3.1 12.9 2473.3 SCπ*  4.0 ± 0.3%  12.9 2474.1 SCσ*  22.5 ± 2.8%  16.6 5b 2472.3 SCπ* + Ru4dσ*  18.1 ± 0.8% 2.9 13.2 2473.3 SCπ*  8.9 ± 0.8%  12.4 2474.1 SCσ*  11.8 ± 1.0%  12.1 5c  2472.3 SCπ* + Ru4dσ*  18.5 ± 1.2%  3.3 12.7 2473.2 SCπ*  6.8 ± 1.6%  12.9 2474.1 SCσ*  16.5 ± 1.7%  16 6a 2472.4 Ru4dσ* 15.4 ±  19.3 2473.2 SCπ* / SCσ* 21.8 ±  a 6b 2473.2 Os5dσ* 10.7 ±  12.2  2474.2 SCπ* / SCσ* 29.8 ±  a  Detail fit result for each complex is provided in Appendix A.3.1. a S3p character is dispersed over a wide energy range and multiple acceptor orbitals The NMe2 group substitution on the TSC ligand appears to have a mild influence on bonding in the metal complex as seen from the slightly increased intensity of the Ru4dσ* ← S1s 121  transition in complex 5b compared to that of 5a. In addition, the whole spectrum is down shifted in energy by ~0.6 eV (shown by arrow in Figure 5.2) in 5b. A similar shift is also observed in 4b. This can be attributed to the more electron rich environment of the TSC due to the electron donating property of alkylamino (NMe2) group.  It is also consistent with a slightly higher charge donation from S1s to Ru4dσ* which leads to a higher covalency of the Ru-S bond in complex 5b. The increased charge donation from S concomitantly decreases the charge donation from the Cl ligand (13.5% in 5a and 12% in 5b) as noticed from the DFT results. In addition, it also reduces the energy gap between the Ru4dσ* orbitals. Therefore, NMe2 group substitution in 5b might lead to a more covalent Ru-S bond and more labile Ru-Cl bond compared to a less covalent Ru-S bond in 5a. Interestingly, the bond strength and lability of these complexes appear to correlate well with their solution chemistry. i.e. the complex 5a forms fragments in solution with no distinct species. In contrast, the complex 5b forms dimeric species (6a) through an intermediate (may be aqua) species311 (this is further investigated in Section 5.3.4). 5.3.2 Effect of the Monodentate Ligand The effect of the monodentate halide ligand is investigated using S K-edge XAS by exchanging the chloride ligand in 5a to iodide ligand in 5c. As shown in Error! Reference source not found..2b, when the ligand is exchanged from chloride to iodide (from 5a to 5c), there is a noticeable increase (16% → 18.5%) in the intensity of the first feature, which correlates to the Ru4dσ* ← S1s transition. It suggests that the covalency of the Ru-S bond is slightly greater in 5c compare to that of 5a. Our DFT results are consistent with this 122  interpretation as the calculated covalency of the Ru-I bond (I5p character is 16.9%) in 5c is also more covalent than the Ru-Cl bond (Cl3p character is 13.5%) in complex 5a. These complexes are only poorly soluble in water, but they exhibit cytotoxicity in µM concentration.311 The biological activity of complexes 5a and 5c at the same concentration is under investigation. However, preliminary studies suggest that less soluble complexes might be more active. Based on these studies and prior studies, one might predict that the substitution of monodentate ligand may not be the only activation mechanism for these complexes. 5.3.3 Effect of Metal (Ru vs Os) in Dimeric Complexes  Figure 5.5: S K-edge XAS of complexes 6a and 6b. Os M3-edge background is not subtracted from the spectrum 6b. S K-edge XAS spectra of the sandwich dinuclear RuII and OsII-p-cym-TSC complexes are shown in Figure 5.5. The near edge region of the spectra is dominated by two intense features. The lower energy feature is assigned to the Mndσ* ← S1s transition. The higher energy feature is assigned to the SCπ*/SCσ* ← S1s transition. Importantly, the π-system of the TSC ligand is strongly affected by deprotonation of 2N on the TSC ligand (Figure 5.1). Metal 123  exchange from ruthenium (6a) to osmium (6b) results in a large shift in the energy of the Mndσ* ← S1s transition to higher energy. This change reflects the higher energy of the Os5d orbitals compared to the Ru4d orbitals. The Mndσ* ← S1s transition in 6a is thus shifted to higher energy (2472.4 → 2473.2 eV ) in 6b. General estimates for the M-S covalency have also been extracted from the data for complexes 6a and 6b. Results are tabulated in Table 5.1 with available DFT analysis. Both XAS and DFT suggest that the Ru-S covalency in 6a is higher than Os-S covalency in 6b. 5.3.4 Monomer to Dimer formation using UV-Vis Spectroscopy and DFT  Scheme 5.1: Conversion of 6a from 5b in methanol or water  Figure 5.6: S K-edge XAS of 5b and 6a 124  Interestingly, it is noticed that 5b converts into 6a in methanol or water in 14 hours, as shown in Scheme 5.1. During conversion, deprotonation occurs on 2N (hydrazinic nitrogen) of the TSC ligand. The reaction has been followed by NMR and UV-vis spectroscopy and the results from the analysis suggest that the electronic environment of the TSC ligand in dimer 6a is different from that of monomer 5b. S K-edge XAS results further suggest a rather significant change in the electronic environment of the sulfur atoms, as shown in Figure 5.6.   Figure 5.7: (a) UV-vis spectra of complexes 5b and 6a. (b) simulated spectra of 5b with functionals BP86 and B3LYP. (c) simulated spectra of 6a with functionals BP86 and B3LYP. (d) simulated UV-vis spectra of 5baq compared with 5a spectrum NMR studies suggest that dimerization may proceed through initial solvolysis of the monoemeric species to form a dicationic aqua monomer. However, such species have not been isolated or identified by any analysis. DFT calculations have been performed on in silico 125  models of 5b, 6a and the putative aqua species (5baq) to generate simulated electronic absorption spectra of these complexes using valence TDDFT calculations. The calculations were performed using both the BP86 and B3LYP functionals and the resulting spectra are compared with experimental spectra in Figure 5.7. The simulated UV-vis spectra of 6a are in reasonable agreement with experiment, reproducing the main features of the experimental spectrum as seen in Figure 5.7c. The two functionals yield differences in the specific energies of the main features, with a somewhat better agreement using the B3LYP functional. By contrast, simulated spectra for 5b are in much poorer agreement with the experimental spectrum (Figure 5.7b). The simulated spectra are far more sensitive to the nature of the functional, but in both cases, additional features are observed that are not observed in the experimental data. The simulated spectra of 5baq (Figure 5.7d) are somewhat simpler (due to the loss of intense Ru ← Cl LMCT transitions) and yield a somewhat better overall agreement with the experimental spectra. However, these results are sufficiently ambiguous to require further study.   Figure 5.8: MO orbital 324 showing the S-S interaction in 6a. 126  Interestingly, a systematic change is observed in the UV-vis spectroscopic features in the region between 300 – 500 nm (Figure 5.7a) during the dimer formation, i.e. when 5b converts into 6a. The feature at 406 nm in 5b is divided into two new features (at 345 and 435 nm) in 6a. TDDFT calculations predict that these features arise from Ru ← L charge transfer (LMCT). The nature of the splitting of the feature indicates that the electronic environment around the Ru-S bond is divided in the dimer molecule. As anticipated, the MO picture of Ru-S bond in 6a shows that the electronic environment of the Ru-S bonds in dimer 6a is divided by S3p-S3p interaction through Ru 4d as shown in figure 5.8. 5.4 Discussion A series of half sandwich [(p-cym)RuII(TSC)X]+ complexes and sandwich RuII and OsII-p-cym-TSC complexes are investigated using S K-edge XAS with the help of DFT calculations. The effects of the ancillary ligand modification and metal are investigated in those complexes. Most importantly, the experimental and computational data are in strong agreement when comparing the different complexes in the series. DFT optimized geometries reasonably reproduce the available crystallographic data of 5b and 6a. The valence molecular orbital descriptions are in good agreement with the trend observed in the XAS spectra, Figure 5.4. Overall, the experimental and computational results are in reasonable agreement. NMe2 substitution of TSC directly influences its π-system as seen when comparing the spectra of proligands 4a and 4b. These differences are also observed in complexes 5a/5b, where the electron donating NMe2 appears to strengthen the Ru-S bond and concomitantly weaken the Ru-Cl bond in complex 5b. It is tempting to speculate that such a modification could strongly influence the chemical and biochemical behaviour of these complexes; the 127  potency of these complexes as anticancer agents is currently under investigation. However, there is a universal agreement in the literature that dialkylation of the 4N of the TSC ligand increases in vitro cytotoxicity of both the free ligands and metal complexes.302,322 At this point, it is not clear that exchanging the monodentate ligand from Cl to I should have a major influence on the biological activity of these complexes; however, the chloro complex 5a rapidly dissociates in solution and leads to decomposition. Decreasing the lability of the M-X bond would thus likely be very beneficial. This is achieved in the iodo complex (5c) whose solution chemistry is still as yet unexplored. Interestingly, it has recently been noticed that the substitutionally inert iodo osmium azopyridine complex, [Os(η6-bip)(azpy-NMe2)I]+ and the related iodo iminopyridine complex, [Ru(η6-p-cym)(p-Impy-NMe2)I]+ are more potent than their chloride analogues.90,106,107 A drastic increase in biological activity is also observed in the Os complex 6b compared to its Ru analogue, 6a. Generally, osmium complexes are more potent than their ruthenium equivalents,70 which indicates that the metal center plays a major role in the biological activity of these types of complexes. 5.5 Conclusions Structural modification of metal-arene-thiosemicarbazone complexes is investigated through their electronic structure using XAS and DFT calculation. An electron donating group (NMe2), substituted on the TSC ligand, directly interacts with the π-system of the TSC ligand and increase the Ru-S covalency. This effect also appears to stabilize the complexes in solution and possibly moderates their mechanism of action. It was long believed that the metal coordination can increase the cellular uptake of these complexes, however the studies 128  herein show that metal coordination to TSC ligand can play a huge role in their mode of activity. Above all, the studies indicate that deeper investigation of the electronic structure of these complexes is a crucial factor in the rational design and development of improved metal arene complexes for medicinal applications.  129  CHAPTER 6. CONCLUSIONS AND FUTURE DIRECTIONS  A combination of X-ray absorption spectroscopic methods have been exploited to investigate the electronic structure of ruthenium(II) and osmium(II) arene complexes that have shown promise as metalloanticancer drugs; many of these complexes are currently under preclinical evaluation. The studies described herein have contributed to testing and evaluating postulated biological modes of action. In addition, these studies have provided a systematic approach to evaluating the effect of ancillary ligand modifications and their impact on both their chemical and biochemical properties. In these studies, computational methods have also been utilized to support the experimental results and provide further insights – especially in cases where appropriate experimental data are unavailable. Results from this thesis clearly demonstrate the importance of both the metal centre and ancillary ligands in determining the biological activity of metalloanticancer drugs.  Furthermore, the work provides an example where defining the mode of activation necessitates an understanding of the fundamental chemical properties of the prodrug and biological intermediates. Such fundamental aspects must be considered when designing and optimizing new metallopharmaceutical agents in the future. The electronic structure of RuII-arene thiolato complexes has been successfully investigated through the thiolato, 1, sulfenato, 2, and sulfinato, 3, complexes using Ru K-edge and S K-edge XAS in concert with computational methods. Analysis indicates that the sulfur ligand oxygenation have little impact on the lability of the oxidized ligand directly. However, this process creates a sulfenato species in which the Ru-S bond is highly correlated with the 130  nature of the adjoining S-O bond. The nature of the Ru-S=O bonding in these species provides for a specific pathway for prodrug activation in RuII arene thiolato complexes. These ideas are tested by exploring the effect of protonation and/or Lewis acid activation on the sulfenato species as well as the sulfinato species (where the S-O π-bonding is observed to be less important). Results suggest that perturbation at the terminal oxo of the sulfenato has a larger impact on the degree of charge donation from the sulfur ligand to metal centre; directly affecting the lability of the ligand. The impact is significantly smaller in the sulfinato species, which supports the importance of the SO π-bonding in the observed effects. Altogether, these results suggest that the activation mechanism of RuII-arene thiolato complexes proceeds through the ligand-centered oxygenation followed by solvation and protonation. This thesis further explores the effect of relatively small modifications to ancillary ligands – noting that they can lead to rather large effects in the oxidized species, a finding which is particularly important to the design and structure optimization of this family of complexes. It is notable, however, that current studies have been limited to in vitro experiments. It would be of great interest to extend these studies to determine the fate of RuII-arene thiolato complexes in vivo using the actual cancer cells model. Data from this thesis provide a firm foundation for the application of XAS-based imaging methods (e.g. STXM-scanning transmission X-ray microscopy, and/or microprobe XAS imaging) to treated tumour cells. Such studies would allow for direct speciation in tumour cells, both from the perspective of the metal centre as well as the sulfur-containing ligand in the appropriate biological environment. 131  Recent studies from Sadler group show that the thiolato ligand formed during the reaction between [(η6-ar)Ru(en)Cl]+ complex and glutathione (GSH) yield an interesting [Ru-ar-en-SG] adduct which can be subsequently oxidized by dioxygen to afford an unique sulfenate intermediate. The competitive reaction between GSH and cGMP with [(η6-ar)Ru(en)Cl]+ complex showed that the oxidation of coordinated GSH in the thiolato complex appears to provide a facile route for displacement of S-bound glutathione by G N7, leading to formation of the cGMP adduct [(η6-ar)Ru(en)(cGMP-N7)]+ as the major product even in the presence of a 250-fold molar excess of GSH. Surprisingly, protonation is not observed during this reaction. Therefore, it appears to be an interesting aspect to investigate the electronic structure of Ru-arene-glutathione adduct, [Ru-ar-en-SG]+ and its mono-oxygenated product ([Ru-ar-en-SOG]+) to study the insight changes in Ru-S bond in those species. This investigation would further provide an understanding of the correlation between structure-activity relationships of ligand design around the metal centre. Beyond the parent metal arene complexes that form the foundation of this thesis, several divergent ligand design motifs continue to be explored to improve efficacy. The initial work lends itself to broader studies of the electronic structure of RuII/OsII-arene complexes using a wide range of ancillary ligands. For example, thiosemicarbazone complexes 4-6 have been studied using a similar approach applied for RuII-arene thiolato complexes. In these cases, our studies on the free TSC proligand indicate that the conjugated π-system is quite sensitive to slight modifications, which translates into changes in the properties of the metal complexes themselves. Additional insights from detailed spectroscopic studies have lead to directed modifications of the ligand structure that are currently being explored. In such 132  complexes, it has further been observed that monomer/dimer equilibria may play an important role in defining the biologically-relevant species. Initial studies have been performed, but a better understanding of the factors that control this equilibrium is an area of current intense exploration. In addition, preliminary studies on RuII/OsII-arene-TSC complexes show prominent differences between the cytotoxicity of RuII and OsII complexes. It appears to suggest that metal centre might also play a role in anticancer activity, therefore further investigation should have been focused with more RuII and OsII-arene counterparts to investigate the actual role of the metal.  Overall, the work presented herein provides a particular approach to exploring the nature of metal complexes of interest as therapeutics. The application of element-specific spectroscopies provides a unique perspective that can lead to important insights into chemical and biochemical properties of such complexes. The initial methodology is being applied to a number of related species – providing useful information in developing our understanding of such biologically-active metal-based drugs. This general approach should be able to significantly contribute to our understanding of other metallopharmaceuticals - as a correlation between their electronic structure and their biological mode of action is explored. 133  REFERENCES (1)  Orvig, C.; Abrams, M. J. Chem. Rev. 1999, 99, 2201–4. (2)  Thompson, K. H.; Orvig, C. Dalton Trans. 2006, 761–764. (3)  Ronconi, L.; Sadler, P. J. Coord. Chem. Rev. 2007, 251, 1633–1648. (4)  Gaynor, D.; Griffith, D. M. Dalton Trans. 2012, 41, 13239–13257. (5)  Sarkar, P. K.; Das, S.; Prajapati, P. K. Ancient Sci. Life 2010, 29, 1–6. (6)  Koch, I.; Moriarty, M.; House, K.; Sui, J.; Cullen, W. R.; Saper, R. B.; Reimer, K. J. Sci. Total Environ. 2011, 409, 4545–4552. (7)  Dwyer, F. P.; Gyarfas, E. C.; Rogers, W. P.; Koch, J. H. Nature 1952, 170, 190–191. (8)  Rosenberg, B.; VanCamp, L.; KRIGAS, T.; Van Loretta, C. Nature 1965, 205, 698–699. (9)  Rosenberg, B.; VanCamp, L.; Trosko, J. E.; Mansour, V. H. Nature 1969, 222, 385–386. (10)  Rosenberg, B.; VanCamp, L. Cancer Res. 1970, 30, 1799–1802. (11)  Lippert, B.; Editor. Cisplatin: Chemistry and Biochemistry of a Leading Anticancer Drug.; Verlag Helvetica Chimica Acta, 1999; p. 563 pp. (12)  Bruijnincx, P. C. A.; Sadler, P. J. Curr. Opin. Chem. Biol. 2008, 12, 197–206. (13)  Di Pasqua, A. J.; Goodisman, J.; Dabrowiak, J. C. Inorg. Chim. Acta 2012, 389, 29–35. (14)  Kidani, Y.; Inagaki, K.; Iigo, M.; Hoshi, A.; Kuretani, K. J. Med. Chem. 1978, 21, 1315– 1318. (15)  Kelland, L. Nat. Rev. Cancer 2007, 7, 573–584. (16)  Wheate, N. J.; Walker, S.; Craig, G. E.; Oun, R. Dalton Trans. 2010, 39, 8113–8127. (17)  Alberto, M. E.; Lucas, M. F. A.; Pavelka, M.; Russo, N. J. Phys. Chem. B 2009, 113, 14473–14479. (18)  Drugs in R&D 2003, 4, 369–372. (19)  Gietema, J. A.; Guchelaar, H. J.; de, V. E. G.; Aulenbacher, P.; Sleijfer, D. T.; Mulder, N. H. Anti-cancer drugs 1993, 4, 51–55. (20)  Kim, D.-K.; Kim, G.; Gam, J.; Cho, Y.-B.; Kim, H.-T.; Tai, J.-H.; Kim, K. H.; Hong, W.-S.; Park, J.-G. J. Med. Chem. 1994, 37, 1471–1485. (21)  Kim, N. K.; Im, S.-A.; Kim, D.-W.; Lee, M. H.; Jung, C. W.; Cho, E. K.; Lee, J. T.; Ahn, J. S.; Heo, D. S.; Bang, Y.-J. Cancer 1999, 86, 1109–1115. (22)  Simpson, D.; Dunn, C.; Curran, M.; Goa, K. L. Drugs 2003, 63, 2127–2156. (23)  Farrer, N. J.; Sadler, P. J. Medicinal Inorganic Chemistry : State of the Art, New Trends, and a Vision of the Future; Alessio, E., Ed.; WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2011; pp. 1–47. (24)  Wang, D.; Lippard, S. J. Nat. Rev. Drug Discov. 2005, 4, 307–320. (25)  Ott, I.; Gust, R. Arch. Pharm. 2007, 340, 117–126. (26)  Bergamo, A.; Gaiddon, C.; Schellens, J. H. M.; Beijnen, J. H.; Sava, G. J. Inorg. Biochem. 2012, 106, 90–99. (27)  Ang, W. H.; Dyson, P. J. Eur.J. Inorg. Chem. 2006, 2006, 4003–4018. (28)  Bruijnincx, P. C. a; Sadler, P. J. Adv. Inorg. Chem. 2009, 61, 1–62. (29)  Yan, Y. K.; Melchart, M.; Habtemariam, A.; Sadler, P. J. Chem. Commun. 2005, 4764– 4776. (30)  Clarke, M. J. Coord. Chem. Rev. 2002, 232, 69–93. 134  (31)  Jakupec, M. A.; Galanski, M.; Arion, V. B.; Hartinger, C. G.; Keppler, B. K. Dalton Trans. 2008, 183–194. (32)  Durig, J. R.; Danneman, J.; Behnke, W. D.; Mercer, E. E. Chem. Biol. Interact. 1976, 13, 287–94. (33)  Clarke, M. J. Met. Ions Biol. Syst. 1980, 11, 231–283. (34)  Clarke, M. J.; Zhu, F.; Frasca, D. R. Chem. Rev. 1999, 99, 2511–2533. (35)  Lentz, F.; Drescher, A.; Lindauer, A.; Henke, M.; Hilger, R. a; Hartinger, C. G.; Scheulen, M. E.; Dittrich, C.; Keppler, B. K.; Jaehde, U. Anti-cancer drug 2009, 20, 97– 103. (36)  Antonarakis, E. S.; Emadi, A. Cancer chemother. Pharmacol. 2010, 66, 1–9. (37)  Alessio, E.; Mestroni, G.; Bergamo, A.; Sava, G. Curr. Top. Med. Chem. 2004, 4, 1525– 1535. (38)  Dyson, P. J.; Sava, G. Dalton Trans. 2006, 1929–33. (39)  Bergamo, A.; Gagliardi, R.; Scarcia, V.; Furlani, A.; Alessio, E.; Mestroni, G.; Sava, G. J. Pharmacol. Exp. Ther. 1999, 289, 559–564. (40)  Sava, G.; Bergamo, A.; Zorzet, S.; Gava, B.; Casarsa, C.; Cocchietto, M.; Furlani, A.; Scarcia, V.; Serli, B.; Iengo, E.; Alessio, E.; Mestroni, G. Eur. J. Cancer 2002, 38, 427– 435. (41)  Hartinger, C. G.; Jakupec, M. A.; Zorbas-Seifried, S.; Groessl, M.; Egger, A.; Berger, W.; Zorbas, H.; Dyson, P. J.; Keppler, B. K. Chem. Biodivers. 2008, 5, 2140–2155. (42)  Jakupec, M. A.; Arion, V. B.; Kapitza, S.; Reisner, E.; Eichinger, A.; Pongratz, M.; Marian, B.; Graf N., von K.; Keppler, B. K. Int. J. Clin. Pharmacol. Ther. 2005, 43, 595– 596. (43)  Clarke, M. J. M. J.; Bitler, S.; Rennert, D.; Buchbinder, M.; Kelman, A. D. J. Inorg. Biochem. 1980, 12, 79–87. (44)  Gaiddon, C.; Jeannequin, P.; Bischoff, P.; Pfeffer, M.; Sirlin, C.; Loeffler, J. P. J. Pharmacol. Exp. Ther. 2005, 315, 1403–1411. (45)  Morris, R. E.; Aird, R. E.; Murdoch, P. D. S.; Chen, H. M.; Cummings, J.; Hughes, N. D.; Parsons, S.; Parkin, a; Boyd, G.; Jodrell, D. I.; Sadler, P. J. J. Med. Chem. 2001, 44, 3616–21. (46)  Scolaro, C.; Bergamo, A.; Brescacin, L.; Delfino, R.; Cocchietto, M.; Laurenczy, G.; Geldbach, T. J.; Sava, G.; Dyson, P. J. J. Med. Chem. 2005, 48, 4161–4171. (47)  Kilah, N. L.; Meggers, E. Aust. J. Chem. 2012, 65, 1325–1332. (48)  Süss-Fink, G. Dalton Trans. 2010, 39, 1673–1688. (49)  Dale, L. D.; Tocher, J. H.; Dyson, T. M.; Edwards, D. I.; Tocher, D. A. Anti-Cancer Drug Des. 1992, 7, 3–14. (50)  Hartinger, C. G.; Metzler-Nolte, N.; Dyson, P. J. Organometallics 2012, 31, 5677–5685. (51)  Dyson, P. J. Chimia 2007, 61, 698–703. (52)  Dougan, S. J.; Sadler, P. J. Chimia 2007, 61, 704–715. (53)  Allardyce, C. S.; Dyson, P. J.; Ellis, D. J.; Heath, S. L. Chem. Commun. 2001, 1396–1397. (54)  Scolaro, C.; Geldbach, T. J.; Rochat, S.; Dorcier, A.; Gossens, C.; Bergamo, A.; Cocchietto, M.; Tavernelli, I.; Sava, G.; Rothlisberger, U.; Dyson, P. J. Organometallics 2006, 25, 756–765. 135  (55)  Scolaro, C.; Chaplin, A. B.; Hartinger, C. G.; Bergamo, A.; Cocchietto, M.; Keppler, B. K.; Sava, G.; Dyson, P. J. Dalton Trans. 2007, 5065–5072. (56)  Chatterjee, S.; Kundu, S.; Bhattacharyya, A.; Hartinger, C. G.; Dyson, P. J. J. Biol. Inorg. Chem. 2008, 13, 1149–1155. (57)  Nowak-Sliwinska, P.; Van Judy R., B.; Casini, A.; Nazarov, A. A.; Wagnieres, G.; Van Hubert, den B.; Dyson, P. J.; Griffioen, A. W. J. Med. Chem. 2011, 54, 3895–3902. (58)  Bregman, H.; Williams, D. S.; Atilla, G. E.; Carroll, P. J.; Meggers, E. J. Am. Chem. Soc. 2004, 126, 13594–13595. (59)  Zhang, L.; Carroll, P.; Meggers, E. Org. Lett. 2004, 6, 521–523. (60)  Meggers, E.; Atilla-Gokcumen, G. E.; Bregman, H.; Maksimoska, J.; Mulcahy, S. P.; Pagano, N.; Williams, D. S. Synlett 2007, 1177–1189. (61)  Smalley, K. S. M.; Contractor, R.; Haass, N. K.; Kulp, A. N.; Atilla-Gokcumen, G. E.; Williams, D. S.; Bregman, H.; Flaherty, K. T.; Soengas, M. S.; Meggers, E.; Herlyn, M. Cancer Res. 2007, 67, 209–217. (62)  Bullock, A. N.; Russo, S.; Amos, A.; Pagano, N.; Bregman, H.; Debreczeni, J. E.; Lee, W. H.; Von Frank, D.; Meggers, E.; Knapp, S. PLoS One 2009, 4, No pp. given. (63)  Xie, P.; Streu, C.; Qin, J.; Bregman, H.; Pagano, N.; Meggers, E.; Marmorstein, R. Biochemistry 2009, 48, 5187–5198. (64)  Merkel, A. L.; Meggers, E.; Ocker, M. Expert Opin. Investig. Drugs 2012, 21, 425–436. (65)  Chen, H. M.; Parkinson, J. A.; Parsons, S.; Coxall, R. a; Gould, R. O.; Sadler, P. J. J. Am. Chem. Soc. 2002, 124, 3064–3082. (66)  Chen, H. M.; Parkinson, J. A.; Morris, R. E.; Sadler, P. J. J. Am. Chem. Soc. 2003, 125, 173–186. (67)  Aird, R. E.; Cummings, J.; Ritchie, A. A.; Muir, M.; Morris, R. E.; Chen, H. M.; Sadler, P. J.; Jodrell, D. I. Br. J. Cancer 2002, 86, 1652–1657. (68)  Guichard, S. M.; Else, R.; Reid, E.; Zeitlin, B.; Aird, R. E.; Muir, M.; Dodds, M.; Fiebig, H.; Sadler, P. J.; Jodrell, D. I. Biochem. Pharmacol. 2006, 71, 408–415. (69)  Melchart, M.; Sadler, P. J. In Bioorganometallics; Wiley-VCH Verlag GmbH & Co. KGaA, 2006; pp. 39–64. (70)  Noffke, A. L.; Habtemariam, A.; Pizarro, A. M.; Sadler, P. J. Chem. Commun. 2012, 48, 5219–46. (71)  Pizarro, A. M.; Melchart, M.; Habtemariam, A.; Salassa, L.; Fabbiani, F. P. A.; Parsons, S.; Sadler, P. J. Inorg. Chem. 2010, 49, 3310–3319. (72)  Van Sabine H., R.; Mukherjee, A.; Pizarro, A. M.; Sadler, P. J.; Van Rijt, S. H. J. Med. Chem. 2010, 53, 840–9. (73)  Ang, W. H.; Parker, L. J.; De Anastasia, L.; Juillerat-Jeanneret, L.; Morton, C. J.; Lo Mario, B.; Parker, M. W.; Dyson, P. J. Angew. Chem. Int. Ed. 2009, 48, 3854–3857. (74)  Ang, W. H.; Daldini, E.; Juillerat-Jeanneret, L.; Dyson, P. J. Inorg. Chem. 2007, 46, 9048–9050. (75)  Kilpin, K. J.; Clavel, C. M.; Edafe, F.; Dyson, P. J. Organometallics 2012, 31, 7031–7039. (76)  Petzold, H.; Xu, J. J.; Sadler, P. J. Angew. Chem. Int. Ed. 2008, 47, 3008–3011. (77)  Ang, W. H.; Casini, A.; Sava, G.; Dyson, P. J. J. Organomet. Chem. 2011, 696, 989–998. (78)  Betanzos-Lara, S.; Habtemariam, A.; Clarkson, G. J.; Sadler, P. J. Eur.J. Inorg. Chem. 2011, 2011, 3257–3264. 136  (79)  Melchart, M.; Habtemariam, A.; Novakova, O.; Moggach, S. A.; Fabbiani, F. P. A.; Parsons, S.; Brabec, V.; Sadler, P. J. Inorg. Chem. 2007, 46, 8950–8962. (80)  Wang, F. Y.; Habtemariam, A.; Van der Geer, E. P. L.; Fernandez, R.; Melchart, M.; Deeth, R. J.; Aird, R. E.; Guichard, S.; Fabbiani, F. P. A.; Lozano-Casal, P.; Oswald, I. D. H.; Jodrell, D. I.; Parsons, S.; Sadler, P. J.; Fernández, R. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 18269–18274. (81)  Guo, Z.; Habtemariam, A.; Sadler, P. J.; James, B. R. Inorg. Chim. Acta 1998, 273, 1–7. (82)  Morris, R. E.; Habtemariam, A.; Guo, Z.; Parsons, S.; Sadler, P. J. Inorg. Chim. Acta 2002, 339, 551–559. (83)  Melchart, M.; Habtemariam, A.; Parsons, S.; Moggach, S. a.; Sadler, P. J. Inorg. Chim. Acta 2006, 359, 3020–3028. (84)  Habtemariam, A.; Melchart, M.; Fernandez, R.; Parsons, S.; Oswald, I. D. H.; Parkin, A.; Fabbiani, F. P. A.; Davidson, J. E.; Dawson, A.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J. J. Med. Chem. 2006, 49, 6858–6868. (85)  Bugarcic, T.; Habtemariam, A.; Stepankova, J.; Heringova, P.; Kasparkova, J.; Deeth, R. J.; Johnstone, R. D. L.; Prescimone, A.; Parkin, A.; Parsons, S.; Brabec, V.; Sadler, P. J. Inorg. Chem. 2008, 47, 11470–86. (86)  Filak, L. K.; Muehlgassner, G.; Bacher, F.; Roller, A.; Galanski, M.; Jakupec, M. A.; Keppler, B. K.; Arion, V. B. Organometallics 2011, 30, 273–283. (87)  Hanif, M.; Henke, H.; Meier, S. M.; Martic, S.; Labib, M.; Kandioller, W.; Jakupec, M. A.; Arion, V. B.; Kraatz, H.-B.; Keppler, B. K.; Hartinger, C. G. Inorg. Chem. 2010, 49, 7953–7963. (88)  Kandioller, W.; Hartinger, C. G.; Nazarov, A. A.; Bartel, C.; Skocic, M.; Jakupec, M. A.; Arion, V. B.; Keppler, B. K. Chem.-Eur. J. 2009, 15, 12283–12291, S12283/1– S12283/32. (89)  Fernandez, R.; Melchart, M.; Habtemariam, A.; Parsons, S.; Sadler, P. J. Chem.-Eur. J. 2004, 10, 5173–5179. (90)  Romero-Canelon, I.; Pizarro, A. M.; Habtemariam, A.; Sadler, P. J. Metallomics 2012, 4, 1271–1279. (91)  Fu, Y.; Romero, M. J. M. J.; Habtemariam, A.; Snowden, M. E.; Song, L.; Clarkson, G. J.; Qamar, B.; Pizarro, A. M.; Unwin, P. R.; Sadler, P. J. Chem. Sci. 2012, 3, 2485–2494. (92)  Shnyder, S. D.; Fu, Y.; Habtemariam, A.; Van Rijt, S. H.; Cooper, P. a.; Loadman, P. M.; Sadler, P. J. Med. Chem. Commun. 2011, 2, 666. (93)  Muehlgassner, G.; Bartel, C.; Schmid, W. F.; Jakupec, M. A.; Arion, V. B.; Keppler, B. K. J. Inorg. Biochem. 2012, 116, 180–187. (94)  Wang, F. Y.; Chen, H. M.; Parsons, S.; Oswald, I. D. H.; Davidson, J. E.; Sadler, P. J. Chem. Eur. J. 2003, 9, 5810–20. (95)  Wang, F. Y.; Xu, J. J.; Habtemariam, A.; Bella, J.; Sadler, P. J. J. Am. Chem. Soc. 2005, 127, 17734–17743. (96)  Sriskandakumar, T.; Petzold, H.; Bruijnincx, P. C. A.; Habtemariam, A.; Sadler, P. J.; Kennepohl, P. J. Am. Chem. Soc. 2009, 131, 13355–61. (97)  Lindsay, C.; Cesarotti, E.; Adams, H.; Bailey, N. A.; White, C. Organometallics 1990, 9, 2594–2602. (98)  Grapperhaus, C. a; Poturovic, S.; Mashuta, M. S. Inorg. Chem. 2005, 44, 8185–8187. 137  (99)  Dey, A.; Green, K. N.; Jenkins, R. M.; Jeffrey, S. P.; Darensbourg, M.; Hodgson, K. O.; Hedman, B.; Solomon, E. I. Inorg. Chem. 2007, 46, 9655–60. (100)  Buonomo, R. M.; Font, I.; Maguire, M. J.; Reibenspies, J. H.; Tuntulani, T.; Darensbourg, M. Y. J. Am. Chem. Soc. 1995, 117, 963–973. (101)  Farmer, P. J.; Reibenspies, J. H.; Lindahl, P. A.; Darensbourg, M. Y. J. Am. Chem. Soc. 1993, 115, 4665–4674. (102)  Tyler, L. A.; Noveron, J. C.; Olmstead, M. M.; Mascharak, P. K. Inorg. Chem. 2000, 39, 357–362. (103)  Petzold, H.; Braeutigam, S.; Goerls, H.; Weigand, W.; Romanski, J.; Mloston, G. Eur. J. Inorg. Chem. 2007, 5627–5632. (104)  Petzold, H.; Sadler, P. J. Chem. Commun. 2008, 37, 4413–4415. (105)  Peacock, A. F. A.; Habtemariam, A.; Fernández, R.; Walland, V.; Fabbiani, F. P. a; Parsons, S.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J. J. Am. Chem. Soc. 2006, 128, 1739– 48. (106)  Fu, Y.; Habtemariam, A.; Pizarro, A. M.; Van Rijt, S. H.; Healey, D. J.; Cooper, P. a; Shnyder, S. D.; Clarkson, G. J.; Sadler, P. J. J. Med. Chem. 2010, 53, 8192–6. (107)  Fu, Y.; Habtemariam, A.; Basri, A. M. B. H.; Braddick, D.; Clarkson, G. J.; Sadler, P. J.; Editor, G.; Orvig, C.; Board, E.; Edwards, P. G.; Hahn, F. E.; Champness, N. R.; Dempsey, J. L.; Winkler, J. R.; Gray, H. B.; Stephenson, A.; Ward, M. D. Dalton Trans. 2011, 40, 10553–62. (108)  Cebrian-Losantos, B.; Krokhin, A. A.; Stepanenko, I. N.; Eichinger, R.; Jakupec, M. A.; Arion, V. B.; Keppler, B. K. Inorg. Chem. 2007, 46, 5023–5033. (109)  Dorcier, A.; Ang, W. H.; Bolano, S.; Gonsalvi, L.; Juillerat-Jeannerat, L.; Laurenczy, G.; Peruzzini, M.; Phillips, A. D.; Zanobini, F.; Dyson, P. J. Organometallics 2006, 25, 4090–4096. (110)  Buchel, G. E.; Stepanenko, I. N.; Hejl, M.; Jakupec, M. A.; Keppler, B. K.; Arion, V. B. Inorg. Chem. 2011, 50, 7690–7697. (111)  Bergamo, A.; Masi, A.; Peacock, A. F. A.; Habtemariam, A.; Sadler, P. J.; Sava, G. J. Inorg. Biochem. 2010, 104, 79–86. (112)  Van Rijt, S. H.; Peacock, A. F. A.; Johnstone, R. D. L.; Parsons, S.; Sadler, P. J. Inorg. Chem. 2009, 48, 1753–62. (113)  Peacock, A. F. A.; Parsons, S.; Sadler, P. J. J. Am. Chem. Soc. 2007, 129, 3348–57. (114)  Peacock, A. F. a; Habtemariam, A.; Moggach, S. a; Prescimone, A.; Parsons, S.; Sadler, P. J. Inorg. Chem. 2007, 46, 4049–59. (115)  Dorcier, A.; Dyson, P. J.; Gossens, C.; Rothlisberger, U.; Scopelliti, R.; Tavernelli, I. Organometallics 2005, 24, 2114–2123. (116)  Schmid, W. F.; John, R. O.; Arion, V. B.; Jakupec, M. A.; Keppler, B. K. Organometallics 2007, 26, 6643–6652. (117)  Hanif, M.; Nazarov, A. A.; Hartinger, C. G.; Kandioller, W.; Jakupec, M. A.; Arion, V. B.; Dyson, P. J.; Keppler, B. K. Dalton Trans. 2010, 39, 7345–7352. (118)  Rijt, S. H. van; Kostrhunova, H.; Brabec, V.; Sadler, P. J. Bioconjugate Chem. 2011, 22, 218–26. (119)  Peacock, A. F. a; Melchart, M.; Deeth, R. J.; Habtemariam, A.; Parsons, S.; Sadler, P. J. Chem. Eur. J. 2007, 13, 2601–13. 138  (120)  Van Rijt, S. H.; Hebden, A. J.; Amaresekera, T.; Deeth, R. J.; Clarkson, G. J.; Parsons, S.; McGowan, P. C.; Sadler, P. J. J. Med. Chem. 2009, 52, 7753–64. (121)  Pizarro, A. M.; Habtemariam, A.; Sadler, P. J. Top. Organomet. Chem. 2010, 32, 21– 56. (122)  Rijt, S. H. van; Peacock, A. F. A.; Sadler, P. J.; Bonetti, A.; Leone, R.; Muggia, F. M.; Howell, S. B. In Platinum and Other Heavy Metal Compounds in Cancer Chemotherapy; Bonetti, A.; Leone, R.; Muggia, F. M.; Howell, S. B., Eds.; Humana Press: Totowa, NJ, 2009; pp. 73–79. (123)  Rautio, J.; Kumpulainen, H.; Heimbach, T.; Oliyai, R.; Oh, D.; Jaervinen, T.; Savolainen, J. Nat. Rev. Drug Discov. 2008, 7, 255–270. (124)  Guo, Z.; Sadler, P. J. Angew. Chem. Int. Ed. 1999, 38, 1512–1531. (125)  Neidle, S.; Ismail, I. M.; Sadler, P. J. J. Inorg. Biochem. 1980, 13, 205–212. (126)  Frey, U.; Ranford, J. D.; Sadler, P. J. Inorg. Chem. 1993, 32, 1333–1340. (127)  Laurie, S. A.; Siu, L. L.; Winquist, E.; Maksymiuk, A.; Harnett, E. L.; Walsh, W.; Tu, D.; Parulekar, W. R. Cancer 2010, 116, 362–368. (128)  Simpson, D.; Dunn, C.; Curran, M.; Goa, K. L.; Beale, P.; Oncology, M.; Centre, C.; Wales, N. S.; Chatelut, E.; Regaud, I. C. Drugs 2003, 63, 2127–2156. (129)  Kasparkova, J.; Vojtiskova, M.; Natile, G.; Brabec, V. Chem. -Eur. J. 2008, 14, 1330– 1341. (130)  Spingler, B.; Whittington, D. A.; Lippard, S. J. Inorg. Chem. 2001, 40, 5596–5602. (131)  Boulikas, T.; Vougiouka, M. Oncol. Rep. 2003, 10, 1663–1682. (132)  Heffeter, P.; Jungwirth, U.; Jakupec, M.; Hartinger, C. G.; Galanski, M.; Elbling, L.; Micksche, M.; Keppler, B. K.; Berger, W. Drug Resist. Updat. 2008, 11, 1–16. (133)  Sava, G.; Zorzet, S.; Giraldi, T.; Mestroni, G.; Zassinovich, G. Eur. J. Cancer Clin. Oncol. 1984, 20, 841–847. (134)  Sava, G.; Bergamo, A. Int. J. Oncol. 2000, 17, 353–365. (135)  Gagliardi, R.; Sava, G.; Pacor, S.; Mestroni, G.; Alessio, E. Clin. Exp. Metastasis 1994, 12, 93–100. (136)  Bergamo, A.; Masi, A.; Dyson, P. J.; Sava, G. Int. J. Oncol. 2008, 33, 1281–1289. (137)  Levina, A.; Mitra, A.; Lay, P. A. Metallomics 2009, 1, 458–470. (138)  Schluga, P.; Hartinger, C. G.; Egger, A.; Reisner, E.; Galanski, M.; Jakupec, M. A.; Keppler, B. K. Dalton Trans. 2006, 1796–1802. (139)  Rockwell, S.; Dobrucki, I. T.; Kim, E. Y.; Marrison, S. T.; Vu, V. T. Curr. Mol. Med. 2009, 9, 442–458. (140)  Ascone, I.; Messori, L.; Casini, A.; Gabbiani, C.; Balerna, A.; Dell’Unto, F.; Castellano, A. C. Inorg. Chem. 2008, 47, 8629–8634. (141)  Liu, M.; Lim, Z. J.; Gwee, Y. Y.; Levina, A.; Lay, P. A. Angew. Chem. Int. Ed. 2010, 49, 1661–1664, S1661/1–S1661/9. (142)  Bergamo, A.; Sava, G. Dalton Trans. 2007, 1267–1272. (143)  Pluim, D.; Van Robert C. A. M., W.; Beijnen, J. H.; Schellens, J. H. M. Cancer chemother. Pharmacol. 2004, 54, 71–78. (144)  Hummer, A. a; Heffeter, P.; Berger, W.; Filipits, M.; Batchelor, D.; Büchel, G. E.; Jakupec, M. A.; Keppler, B. K.; Rompel, A. J. Med. Chem. 2013. 139  (145)  Reisner, E.; Arion, V. B.; Keppler, B. K.; Pombeiro, A. J. L. Inorg. Chim. Acta 2008, 361, 1569–1583. (146)  Bergamo, A.; Masi, A.; Jakupec, M. A.; Keppler, B. K.; Sava, G. Metal-based drugs 2009, 2009, 681270. (147)  Hartinger, C. G.; Zorbas-Seifried, S.; Jakupec, M. A.; Kynast, B.; Zorbas, H.; Keppler, B. K. J. Inorg. Biochem. 2006, 100, 891–904. (148)  Kapitza, S.; Jakupec, M. A.; Uhl, M.; Keppler, B. K.; Marian, B. Cancer Lett. 2005, 226, 115–121. (149)  Ang, W. H.; Daldini, E.; Scolaro, C.; Scopelliti, R.; Juillerat-Jeannerat, L.; Dyson, P. J. Inorg. Chem. 2006, 45, 9006–9013. (150)  Vock, C. A.; Renfrew, A. K.; Scopelliti, R.; Juillerat-Jeanneret, L.; Dyson, P. J. Eur. J. Inorg. Chem. 2008, 1661–1671. (151)  Novakova, O.; Chen, H. M.; Vrana, O.; Rodger, A.; Sadler, P. J.; Brabec, V. Biochemistry 2003, 42, 11544–11554. (152)  Novakova, O.; Kasparkova, J.; Bursova, V.; Hofr, C.; Vojtiskova, M.; Chen, H. M.; Sadler, P. J.; Brabec, V. Chem. Biol. 2005, 12, 121–129. (153)  Liu, H.-K.; Sadler, P. J. Acc. Chem. Res. 2011, 44, 349–59. (154)  Bugarcic, T.; Novakova, O.; Halamikova, A.; Zerzankova, L.; Vrana, O.; Kasparkova, J.; Habtemariam, A.; Parsons, S.; Sadler, P. J.; Brabec, V.; Halámiková, A.; Zerzánková, L.; Vrána, O.; Kaspárková, J. J. Med. Chem. 2008, 51, 5310–5319. (155)  Dougan, S. J.; Habtemariam, A.; McHale, S. E.; Parsons, S.; Sadler, P. J. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 11628–11633. (156)  Kostrhunova, H.; Florian, J.; Novakova, O.; Peacock, A. F. A.; Sadler, P. J.; Brabec, V. J. Med. Chem. 2008, 51, 3635–3643. (157)  Keck, M. V; Lippard, S. J. J. Am. Chem. Soc. 1992, 114, 3386–3390. (158)  Aitken, J. B.; Levina, A.; Lay, P. A. Curr. Top. Med. Chem. 2011, 11, 553–71. (159)  P, G. M. J. D.; Groessl, M.; Dyson, P. J. Curr. Top. Med. Chem. 2011, 11, 2632–2646. (160)  Hall, M. D.; Foran, G. J.; Zhang, M.; Beale, P. J.; Hambley, T. W. J. Am. Chem. Soc. 2003, 125, 7524–7525. (161)  Hall, M. D.; Dillon, C. T.; Zhang, M.; Beale, P.; Cai, Z.; Lai, B.; Stampfl, A. P. J.; Hambley, T. W. J. Biol. Inorg. Chem. 2003, 8, 726–732. (162)  Harris, T. V; Szilagyi, R. K.; McFarlane Holman, K. L. J. Biol. Inorg. Chem. 2009, ASAP. (163)  Penner-Hahn, J. E. Coord. Chem. Rev. 2005, 249, 161–177. (164)  Sarangi, R. Coord. Chem. Rev 2013, 257, 459–472. (165)  Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee, S.-K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y.-S.; Zhou, J. Chem. Rev. 2000, 100, 235–349. (166)  Solomon, E. I.; Lowery, M. D. Science 1993, 259, 1575–1581. (167)  Solomon, E. I.; Hedman, B.; Hodgson, K. O.; Dey, A.; Szilagyi, R. K. Coord. Chem. Rev. 2005, 249, 97–129. (168)  Ortega, R.; Carmona, A.; Llorens, I.; Solari, P. L. J. Anal. At. Spectrom. 2012, 27, 2054. (169)  Thompson, A.; Attwood, D.; Gullikson, E.; Howells, M.; Kim, K.-J.; Kirz, J.; Kortright, J.; Lindau, I.; Liu, Y.; Pianetta, P.; Robinson, A.; Scofield, J.; Underwood, J.; Williams, G.; Winick, H. X-ray Data Booklet; Thompson, A. C., Ed.; 3rd editio.; Lawrence Berkeley National Laboratory, University of California, Berkeley: California, 2009. 140  (170)  Ressler, T.; Brock, S. L.; Wong, J.; Suib, S. L. J. Synch. Rad. 1999, 6, 728–30. (171)  Colpas, G. J.; Maroney, M. J.; Bagyinka, C.; Kumar, M.; Willis, W. S.; Suib, S. L.; Mascharak, P. K.; Baidya, N. Inorg. Chem. 1991, 30, 920–928. (172)  Delgado-Jaime, M. U.; Conrad, J. C.; Fogg, D. E.; Kennepohl, P. Inorg. Chim. Acta 2006, 359, 3042–3047. (173)  Dillon, C. T. Aust. J. Chem. 2012, 65, 204–217. (174)  Glatzel, P.; Jacquamet, L.; Bergmann, U.; De Frank M. F., G.; Cramer, S. P. Inorg. Chem. 2002, 41, 3121–3127. (175)  Levina, A.; Armstrong, R. S.; Lay, P. A. Coord. Chem. Rev. 2005, 249, 141–160. (176)  Rehr, J. J.; Albers, R. C. Rev. Mod. Phys. 2000, 72, 621–654. (177)  Getty, K. J.; Delgado-Jaime, M. U.; Kennepohl, P. J. Am. Chem. Soc. 2007, 129, 15774– 15776. (178)  Getty, K. J.; Delgado-Jaime, M. U.; Kennepohl, P. Inorg. Chim. Acta 2008, 361, 1059– 1065. (179)  Yano, J.; Kern, J.; Irrgang, K.-D.; Latimer, M. J.; Bergmann, U.; Glatzel, P.; Pushkar, Y.; Biesiadka, J.; Loll, B.; Sauer, K.; Messinger, J.; Zouni, A.; Yachandra, V. K. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 12047–12052. (180)  De Frank., G.; De Groot, F. Chem. Rev. 2001, 101, 1779–1808. (181)  Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996, 96, 2239–2314. (182)  Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1990, 112, 1643–1645. (183)  Glaser, T.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. Acc. Chem. Res. 2000, 33, 859– 868. (184)  Shadle, S. E.; Penner-Hahn, J. E.; Schugar, H. J.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1993, 115, 767–776. (185)  Neese, F.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. Inorg. Chem. 1999, 38, 4854– 4860. (186)  Shadle, S. E.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. Inorg. Chem. 1994, 33, 4235– 4244. (187)  Sarangi, R.; George, S. D.; Rudd, D. J.; Szilagyi, R. K.; Ribas, X.; Almeida, M.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Synchrotron, S.; Linear, S.; Rovira, C. J. Am. Chem. Soc. 2007, 129, 2316–2326. (188)  George, S. D.; Brant, P.; Solomon, E. I. J. Am. Chem. Soc. 2005, 127, 667–74. (189)  Lugo-Mas, P.; Dey, A.; Xu, L.; Davin, S. D.; Benedict, J.; Kaminsky, W.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Kovacs, J. A. J. Am. Chem. Soc. 2006, 128, 11211–11221. (190)  Dey, A.; Jenney, F. E.; Adams, M. W. W.; Johnson, M. K.; Hodgson, K. O.; Hedman, B.; Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 12418–31. (191)  Yoon, W.-S.; Chung, K. Y.; Nam, K.-W.; McBreen, J.; Wang, D.; Huang, X.; Li, H.; Chen, L.; Yang, X.-Q. J. Power Sources 2008, 183, 427–430. (192)  Yoon, W.-S.; Chung, K. Y.; McBreen, J.; Zaghib, K.; Yang, X.-Q. Electrochem. Solid St. 2006, 9, A415–A417. (193)  Shirakawa, J.; Nakayama, M.; Wakihara, M.; Uchimoto, Y. J. Phys. Chem. B 2006, 110, 17743–17750. (194)  Rehder, D.; Schulzke, C.; Dau, H.; Meinke, C.; Hanss, J.; Epple, M. J. Inorg. Biochem. 2000, 80, 115–121. 141  (195)  Jeong, H. K.; Park, B.-G.; Kim, J.-Y.; Noh, H.-J. EPL 2011, 95, 47005/1–47005/4. (196)  Shimada, H.; Imamura, M.; Matsubayashi, N.; Sato, T.; Hayakawa, T.; Takehira, K.; Kure, S.; Kinoshita, A.; Nishijima, A. J. Phys. IV 1997, 7, 919–920. (197)  Bradley, J. A.; Yang, P.; Batista, E. R.; Boland, K. S.; Burns, C. J.; Clark, D. L.; Conradson, S. D.; Kozimor, S. A.; Martin, R. L.; Seidler, G. T.; Scott, B. L.; Shuh, D. K.; Tyliszczak, T.; Wilkerson, M. P.; Wolfsberg, L. E. J. Am. Chem. Soc. 2010, 132, 13914–13921. (198)  Pickering, I. J.; Prince, R. C.; Divers, T.; George, G. N. FEBS Lett. 1998, 441, 11–14. (199)  Pickering, I. J.; George, G. N.; Yu, E. Y.; Brune, D. C.; Tuschak, C.; Overmann, J.; Beatty, J. T.; Prince, R. C. Biochemistry 2001, 40, 8138–45. (200)  Williams, K. R.; Hedman, B.; Hodgson, K. O.; Solomon, E. I.; Rose Williams, K. Inorg. Chim. Acta 1997, 263, 315–321. (201)  Dey, A.; Chow, M.; Taniguchi, K.; Lugo-Mas, P.; Davin, S.; Maeda, M.; Kovacs, J. A.; Odaka, M.; Hodgson, K. O.; Hedman, B.; Solomon, E. I. J. Am. Chem. Soc. 2006, 128, 533–541. (202)  Solomon, E. I.; Gorelsky, S. I.; Dey, A. J. Comput. Chem. 2006, 27, 1415–1428. (203)  Rose, K.; Shadle, S. E.; Glaser, T.; De Simon, V.; Cherepanov, A.; Canters, G. W.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1999, 121, 2353–2363. (204)  Shearer, J.; Callan, P. E.; Masitas, C. a; Grapperhaus, C. a Inorg. Chem. 2012, 51, 6032–45. (205)  Clarke, R. Incoherent sources: Synchrotron, in Encyclopedia of Modern Optics; Elsevier, 2005; pp. 217–224. (206)  Willmott, P. Introduction to Synchrotron Radiation : Techniques and Applications; Wiley: Hoboken, NJ, USA, 2011. (207)  Duke, P. J. Synchrotron Radiation: Production and Properties; Oxford University Press: Oxford, 2000. (208)  Cramer, S. P.; Peng, G.; Christiansen, J.; Chen, J.; Van J., E.; George, S. J.; Young, A. T. J. Electron. Spectrosc. Relat. Phenom. 1996, 78, 225–229. (209)  Lytle, F. W.; Greegor, R. B.; Sandstrom, D. R.; Marques, E. C.; Wong, J.; Spiro, C. L.; Huffman, G. P.; Huggins, F. E. Nucl. Instr. & Meth. in Phys. Res. A 1984, 226, 542–548. (210)  Kennepohl, P.; Wasinger, E. C.; George, S. D. J. Synch. Rad. 2009, 16, 484–488. (211)  Hu, Y. F.; Coulthard, I.; Chevrier, D.; Wright, G.; Igarashi, R.; Sitnikov, A.; Yates, B. W.; Hallin, E. L.; Sham, T. K.; Reininger, R. In THE 10TH INTERNATIONAL CONFERENCE ON SYNCHROTRON RADIATION INSTRUMENTATION; Garrett, R; Gentle, I; Nugent, K; Wilkins, S., Ed.; 2010; Vol. 1234, pp. 343–346. (212)  George, G. N.; Gnida, M.; Bazylinski, D. A.; Prince, R. C.; Pickering, I. J. J. Bacteriol. 2008, 190, 6376–6383. (213)  Jaklevic, J.; Kirby, J. A.; Klein, M. P.; Robertson, A. S.; Brown, G. S.; Eisenberger, P. Solid State Commun. 1977, 23, 679–682. (214)  Vlachos, D.; Craven, A. J.; McComb, D. W. J. Synchrotron Rad. 2005, 12, 224–233. (215)  Kohn, W. Electronic structure of matter - wave functions and density functionals 1999, 213–237. (216)  Kepp, K. P. Coord. Chem. Rev. 2013, 257, 196–209. (217)  Ryde, U. Dalton Trans. 2007, 607–625. 142  (218)  Baerends, E. J. ADF 2007.01ed.; SCM, Theoretical Chemistry, Vrije Universiteit: Amsterdam, The Netherlands 2007. (219)  Velde, G. T.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca C., G.; Van S. J. A., G.; Snijders, J. G.; Ziegler, T.; Fonseca Guerra, C.; Van Gisbergen, S. J. a. J. Comput. Chem. 2001, 22, 931–967. (220)  Neese, F. ORCA, an Ab Initio, Density Functional and Semiempirical Electronic Structure Program Package, version 2.9, Universitat Bonn, Germany 2012. (221)  Neese, F. WIREs Comput. Mol. Sci 2012, 2, 73–78. (222)  Hohenberg, P.; Kohn, W. Phys. Rev. B 1964, 136, B864–B871. (223)  Neese, F. Coord. Chem. Rev 2009, 253, 526–563. (224)  Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, 1133–&. (225)  Eriksson, L. A.; Pettersson, L. G. M.; Siegbahn, P. E. M.; Wahlgren, U. J. Chem. Phys. 1995, 102, 872–878. (226)  Becke, A. D. Phys. Rev. A: At. Mol. Opt. Phys. 1988, 38, 3098–3100. (227)  Perdew, J. P. Phys. Rev. B: Condens. Matter 1986, 33, 8822–8824. (228)  Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200–1211. (229)  Slater, J. C. Phys. Rev. 1930, 36, 57–64. (230)  Jensen, F. Introduction to computational chemistry; Second Edi.; John Wiley & Sons, Ltd., 2007; pp. 193–231. (231)  Delgado-Jaime, M. U. Electronic structure studies of ruthenium-based catalysts for olefin metathesis: an x-ray absorption spectroscopy perspective, 2009. (232)  Weigend, F.; Furche, F.; Ahlrichs, R. J. Chem. Phys. 2003, 119, 12753–12762. (233)  Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297–3305. (234)  Neese, F. J. Biol. Inorg. Chem. 2006, 11, 702–711. (235)  Van E., L.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99, 4597–4610. (236)  Filatov, M.; Cremer, D. Mol. Phys. 2003, 101, 2295–2302. (237)  Klamt, A.; Schueuermann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799–805. (238)  Mulliken, R. S. J. Chem. Phys. 1962, 36, 3428–3440. (239)  Hirshfeld, F. L. Theor. Chim. Acta 1977, 44, 129–138. (240)  Lowdin, P. O. Adv. Quantum Chem. 1970, 5, 185–199. (241)  Reed, A. E.; Curtiss, L. a.; Weinhold, F. Chem. Rev. 1988, 88, 899–926. (242)  Weinhold, F.; Landis, C. R. Chem. Edu. Res. Pract. Eur. 2001, 2, 91–104. (243)  Bregman, H.; Carroll, P. J.; Meggers, E. J. Am. Chem. Soc. 2006, 128, 877–884. (244)  Gras, M.; Therrien, B.; Suss-Fink, G.; Stepnicka, P.; Renfrew, A. K.; Dyson, P. J. J. Organomet. Chem. 2008, 693, 3419–3424. (245)  Nguyen, A.; Top, S.; Pigeon, P.; Vessières, A.; Hillard, E. A.; Plamont, M. A.; Huché, M.; Rigamonti, C.; Jaouen, G. Chem.-Eur. J. 2009, 15, 684–696. (246)  Novakova, O.; Nazarov, A. A.; Hartinger, C. G.; Keppler, B. K.; Brabec, V. Biochem. Pharmacol. 2009, 77, 364–374. (247)  Roos, G.; Messens, J. Free Radic. Biol. Med. 2011, 51, 314–326. (248)  Charles, R. L.; Schroder, E.; May, G.; Free, P.; Gaffney, P. R. J.; Wait, R.; Begum, S.; Heads, R. J.; Eaton, P. Mol. Cell. Proteomics 2007, 6, 1473–1484. (249)  Jacob, C.; Knight, I.; Winyard, P. G. Biol. Chem. 2006, 387, 1385–1397. (250)  Maret, W. Antioxid. Redox Signaling 2006, 8, 1419–1441. 143  (251)  Dey, A.; Jeffrey, S. P.; Darensbourg, M.; Hodgson, K. O.; Hedman, B.; Solomon, E. I. Inorg. Chem. 2007, 46, 4989–4996. (252)  Wang, F. Y.; Weidt, S.; Xu, J. J.; Mackay, C. L.; Langridge-Smith, P. R. R.; Sadler, P. J. J. Am. Soc. Mass. Spectrom. 2008, 19, 544–549. (253)  Hu, W.; Luo, Q.; Ma, X.; Wu, K.; Liu, J.; Chen, Y.; Xiong, S.; Wang, J.; Sadler, P. J.; Wang, F. Y. Chem. Eur. J. 2009, 15, 6586–94. (254)  Yano, T.; Arii, H.; Yamaguchi, S.; Funahashi, Y.; Jitsukawa, K.; Wawa, T.; Masuda, H. Eur.J. Inorg. Chem. 2006, 3753–3761. (255)  Yano, T.; Wasada-Tsutsui, Y.; Arii, H.; Yamaguchi, S.; Funahashi, Y.; Ozawa, T.; Masuda, H. Inorg. Chem. 2007, 46, 10345–53. (256)  Rat, M.; De Sousa, R. A.; Tomas, A.; Frapart, Y.; Tuchagues, J. P.; Artaud, I. Eur.J. Inorg. Chem. 2003, 759–765. (257)  O’Toole, M. G.; Kreso, M.; Kozlowski, P. M.; Mashuta, M. S.; Grapperhaus, C. a J. Biol. Inorg. Chem. 2008, 13, 1219–1230. (258)  Mascharak, P. K. Coord. Chem. Rev. 2002, 225, 201–214. (259)  Lee, C. M.; Hsieh, C. H.; Dutta, A.; Lee, G. H.; Liaw, W. F. J. Am. Chem. Soc. 2003, 125, 11492–11493. (260)  Kung, I.; Schweitzer, D.; Shearer, J.; Taylor, W. D.; Jackson, H. L.; Lovell, S.; Kovacs, J. A. J. Am. Chem. Soc. 2000, 122, 8299–8300. (261)  Grapperhaus, C. a; Mullins, C. S.; Kozlowski, P. M.; Mashuta, M. S. Inorg. Chem. 2004, 43, 2859–2866. (262)  Bourles, E.; De Sousa, R. A.; Galardon, E.; Giorgi, M.; Artaud, I. Angew. Chem. Int. Ed. 2005, 44, 6162–6165. (263)  Begum, F. A.; Farah, A. A.; Hunter, H. N.; Lever, A. B. P. Inorg. Chem. 2009, 48, 2018– 2027. (264)  Rose, M. J.; Betterley, N. M.; Mascharak, P. K. J. Am. Chem. Soc. 2009, 131, 8340– 8341. (265)  De Sousa, R. A.; Artaud, I. Tetrahedron 2008, 64, 2198–2206. (266)  Hopmann, K. H.; Himo, F. Eur.J. Inorg. Chem. 2008, 1406–1412. (267)  Lugo-Mas, P.; Dey, A.; Xu, L.; Davin, S. D.; Benedict, J.; Kaminsky, W.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Kovacs, J. A. J. Am. Chem. Soc. 2006, 128, 11211–21. (268)  Heinrich, L.; Mary-Verla, A.; Li, Y.; Vaissermann, J.; Chottard, J.-C. C. Eur.J. Inorg. Chem. 2001, 2001, 2203–2206. (269)  Hedman, B.; Frank, P.; Gheller, S. F.; Roe, A. L.; Newton, W. E.; Hodgson, K. O. J. Am. Chem. Soc. 1988, 110, 3798–3805. (270)  George, M. J. J. Synchrotron Rad. 2000, 7, 283–286. (271)  Webb, S. M. Phys. Scripta 2005, T115, 1011–1014. (272)  Delgado-Jaime, M. U.; Kennepohl, P. J. Synchrotron Rad. 2010, 17, 119–128. (273)  Delgado-Jaime, M. U.; Mewis, C. P.; Kennepohl, P. J. Synchrotron Rad. 2010, 17, 132– 137. (274)  Stohr, J. Springer Series in Surface Sciences;25 NEXAFS Spectroscopy;; Ertl, Gerhard; Gomer, Robert; Mills, Douglas, L. ., Ed.; 1st Ed.; Springer-Verlag: Berlin, Heidelberg, 1992; p. 213 pp. 144  (275)  Tyson, T. A.; Roe, A. L.; Frank, P.; Hodgson, K. O.; Hedman, B. Phys. Rev. B 1989, 39, 6305–6315. (276)  Van Christoph., W. J. Chem. Phys. 1998, 109, 392–399. (277)  Westre, T. E.; Kennepohl, P.; DeWitt, J. G.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1997, 119, 6297–6314. (278)  Kennepohl, P.; Neese, F.; Schweitzer, D.; Jackson, H. L.; Kovacs, J. A.; Solomon, E. I. Inorg. Chem. 2005, 44, 1826–1836. (279)  Govindaswamy, P.; Mozharivskyj, Y. A.; Kollipara, M. R. Polyhedron 2004, 23, 3115– 3123. (280)  Martin-Diaconescu, V.; Kennepohl, P. Inorg. Chem. 2008, 338, 1038–1044. (281)  Ishii, A.; Komiya, K.; Nakayama, J. J. Am. Chem. Soc. 1996, 118, 12836–12837. (282)  Goto, K.; Holler, M.; Okazaki, R. J. Am. Chem. Soc. 1997, 119, 1460–1461. (283)  Votyakova, T. V; Reynolds, I. J. Arch. Biochem. Biophys. 2004, 431, 138–144. (284)  Vaupel, P.; Kallinowski, F.; Okunieff, P. Cancer Res. 1989, 49, 6449–6465. (285)  Tannock, I. F.; Rotin, D. Cancer Res. 1989, 49, 4373–4384. (286)  Grattan, B. J.; Freake, H. C. Nutrients 2012, 4, 648–675. (287)  Cui, Y.; Vogt, S.; Olson, N.; Glass, A. G.; Rohan, T. E. Cancer Epidemiol. Biomarkers Prev. 2007, 16, 1682–1685. (288)  Wang, F. Y.; Chen, H. M.; Parkinson, J. A.; Murdoch, P. D. S.; Sadler, P. J. Inorg. Chem. 2002, 41, 4509–4523. (289)  Claiborne, A.; Miller, H.; Parsonage, D.; Ross, R. P. FASEB J. 1993, 7, 1483–1490. (290)  Regulation, R.; Claiborne, A.; Yeh, J. I.; Mallett, T. C.; Luba, J.; Crane, E. J.; Parsonage, D.; Crane  III, E. J.; Charrier, V. Biochemistry 1999, 38, 15407–15416. (291)  Allison, W. S. Acc. Chem. Res. 1976, 9, 293–299. (292)  Dilworth, J. R.; Zheng, Y.; Lu, S.; Wu, Q. Transition Met. Chem. 1992, 17, 364–368. (293)  Leeladee, P.; Baglia, R. a; Prokop, K. a; Latifi, R.; De Visser, S. P.; Goldberg, D. P. J. Am. Chem. Soc. 2012, 134, 10397–400. (294)  Furukawa, N.; Konno, Y.; Tsuruoka, M.; Ogawa, S. Heteroat. Chem. 1992, 3, 495–503. (295)  O’Donnell, J. S.; Schwan, A. L. Tetrahedron Lett. 2003, 44, 6293–6296. (296)  Aucott, S. M.; Kilian, P.; Robertson, S. D.; Slawin, A. M. Z.; Woollins, J. D. Chem. Eur. J. 2006, 12, 895–902. (297)  Delgado-Jaime, M. U.; DeBeer, S. J. Comput. Chem. 2012, 33, 2180–2185. (298)  Schaftenaar, G.; Noordik, J. H. J. Comput.-Aided Mol. Des. 2000, 14, 123–134. (299)  Debeer, S.; Neese, F. Inorg. Chem. 2010, 49, 1849–53. (300)  Brown, I. D.; Skowron, a. J. Am. Chem. Soc. 1990, 112, 3401–3403. (301)  Brown, I. D. Acta Crystallogr., Sect. B: Struct. Sci 1988, 44, 545–553. (302)  Dilworth, J. R.; Hueting, R. Inorg. Chim. Acta 2012, 389, 3–15. (303)  Lobana, T. S.; Sharma, R.; Bawa, G.; Khanna, S. Coord. Chem. Rev. 2009, 253, 977– 1055. (304)  Quiroga, A. G.; Perez, J. M.; Lopez-Solera, I.; Masaguer, J. R.; Luque, A.; Roman, P.; Edwards, A.; Alonso, C.; Navarro-Ranninger, C. J. Med. Chem. 1998, 41, 1399–1408. (305)  Offiong, O. E.; Martelli, S. Farmaco 1994, 49, 513–518. (306)  Zhang, G.; Li, Y.; Feng, X.; Zhang, R.; Zhao, J. Asian J. Chem. 2013, 25, 2003–2006. (307)  Hadjipavlou-Litina, D. Pharmazie 1996, 51, 468–470. 145  (308)  Demoro, B.; Sarniguet, C.; Sánchez-Delgado, R.; Rossi, M.; Liebowitz, D.; Caruso, F.; Olea-Azar, C.; Moreno, V.; Medeiros, A.; Comini, M. a; Otero, L.; Gambino, D. Dalton Trans. 2012, 41, 1534–43. (309)  Crouch, P. J.; Barnham, K. J. Acc. Chem. Res. 2012, 45, 1604–1611. (310)  Debreczeni, J. E.; Bullock, A. N.; Atilla, G. E.; Williams, D. S.; Bregman, H.; Knapp, S.; Meggers, E. Angew. Chem. Int. Ed. 2006, 45, 1580–1585. (311)  Noffke, Anna Louisa; Sriskandakumar, T.; Clarkson, G. J.; Kennepohl, P.; Sadler, P. J. In Preparation 2013. (312)  Beckford, F. A.; Leblanc, G.; Thessing, J.; Shaloski Jr, M.; Frost, B. J.; Li, L.; Seeram, N. P. Inorg. Chem. Commun. 2009, 12, 1094–1098. (313)  Beckford, F. A. Int. J. Inorg. Chem. 2010, 2010, 1–7. (314)  Beckford, F. A.; Dourth, D.; Shaloski, M.; Didion, J.; Thessing, J.; Woods, J.; Crowell, V.; Gerasimchuk, N.; Gonzalez-Sarrías, A.; Seeram, N. P. J. Inorg. Biochem. 2011, 105, 1019–29. (315)  Beckford, F. A.; Thessing, J.; Shaloski Jr, M.; Mbarushimana, P. C.; Brock, A.; Didion, J.; Woods, J.; Gonzalez-Sarrías, A.; Seeram, N. P. J. Mol. Struct. 2011, 992, 39–47. (316)  Beckford, F. A.; Shaloski, J. M.; Leblanc, G.; Thessing, J.; Lewis-Alleyne, L. C.; Holder, A. A.; Li, L.; Seeram, N. P. Dalton Trans. 2009, 10757–10764. (317)  Grguric-Sipka, S.; Alshtewi, M. A. A. M.; Jeremic, D.; Kaluderovic, G. N.; Gomez-Ruiz, S.; Zizak, Z.; Juranic, Z.; Sabo, T. J. J. Serb. Chem. Soc. 2008, 73, 619–630. (318)  Garcia, G.; Solano, I.; Sanchez, G.; Santana, M. D.; Lopez, G.; Casabo, J.; Molins, E.; Miravitlles, C. J. Organomet. Chem. 1994, 467, 119–126. (319)  Romero-Canelon, I.; Salassa, L.; Sadler, P. J. J. Med. Chem. 2013, 56, 1291–1300. (320)  Van Lenthe, E; Baerends, E. J. and Snijders, J. G. J. Chem. Phys. 1993, 99, 4597. (321)  Van Wullen, C. J. J. Chem. Phys. 1998, 109, 392. (322)  Kowol, C. R.; Trondl, R.; Heffeter, P.; Arion, V. B.; Jakupec, M. A.; Roller, A.; Galanski, M.; Berger, W.; Keppler, B. K. J. Med. Chem. 2009, 52, 5032–5043. 146  APPENDIX A: BLUEPRINT XAS FITTINGS AND TDDFT ANALYSIS A.1. Peak fitting of S K-edge XAS for chapter 3  Complete assignment and peak fitting of S K-edge XAS of 1a,b – 3a,b and 1c 1a. Peak fitted results for complex 1a. Parameter Lower Upper Ave. (Prefits) Std. (Prefits) Ave. (Fits) Std. (Fits) Value (SingleFit) 95% cint I1 0.02 0.255 0.094347 0.036577 0.105296 0.001381 0.10558 0.002341 G1 0 1 0.51876 0.302984 0.999558 0.003455 1 NaN O1 2473 2474 2473.5 0.293554 2473.07 0.128322 2473.04 0.232907 W1 0 3.88 1.71634 1.13891 0.36752 0.178785 0.336127 0.130431 I2 0 3.56 1.71355 0.960464 1.79394 0.05905 1.80089 0.131502 G2 0 1 0.483722 0.294029 0.073894 0.116267 0.049247 0.242252 W2 0 1.19 0.808472 0.272398 0.600723 0.034358 0.593142 0.063677 O2 2470.8 2471.5 2471.16 0.199457 2471.16 0.010313 2471.16 0.041139 I3 0 4.34 2.03075 1.0892 2.3207 0.087642 2.32114 0.277292 O3 1.5 2.4 1.89714 0.282309 1.589 0.093275 1.56925 0.108282    147  1b. Peak fitted results for complex 1b. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.05 1.25 0.150495 0.0660733 0.118601 0.00014804 0.118005 0.00275812 G1 0 1 0.436011 0.273751 0.999984 0.00011912 1  NaN O1 2473 2475 2473.96 0.55839 2473 0.0197406 2473.5  NaN W1 0 2.5 1.36882 0.695706 1.26668 0.0149275 0.871153 0.279167 I2 0 4.36 2.38379 1.12097 1.96383 0.00454944 2.04508 0.186581 G2 0 1 0.5284 0.290112 0.0868087 0.00949577 0.191797 0.283205 W2 0 1 0.559092 0.260856 0.525365 0.00065762 0.574753 0.0500038 O2 2470.8 2471.4 2471.09 0.163915 2471.17 0.00221263 2471.19 0.0400388 I3 0 4.64 1.98229 1.25039 1.36858 0.0196129 1.87864 0.151654 O3 1.6 2.3 1.95152 0.208485 1.76459 0.00485936 1.7337 0.063654   148  1c. Peak fitted results for complex 1c. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.0688 0.0867 0.077145 0.00506018 0.0742052 0.00089529 0.0760551 0.00155182 G1 0 1 0.459855 0.296387 0.536967 0.104378 0.519819 0.245544 O1 2474 2475.5 2474.76 0.458292 2474 1.58E-08 2474  NaN W1 0.393 1.62 0.991814 0.378082 0.421225 0.111455 0.393  NaN I2 0.9 3 1.8191 0.551085 1.36351 0.0855311 1.23285 0.0478363 G2 0 1 0.414276 0.292035 0.648636 0.0752171 0.631519 0.126845 W2 0.1 0.752 0.597998 0.120106 0.533627 0.0126238 0.512746 0.0139934 O2 2471.1 2471.8 2471.46 0.182577 2471.32 0.043603 2471.36 0.0171232 I3 0.236 2.94 1.52416 0.708566 0.940767 0.0561907 0.92841 0.0636473 O3 1.159 1.585 1.38348 0.121728 1.30109 0.0209332 1.26814 0.0383745 I4 0.904 3.99 2.29293 0.841133 2.15045 0.0685482 2.05248 0.0985099 O4 0.68599 0.88603 0.799364 0.05288 0.822903 0.0291481 0.815665 0.0354149   149  2a. Peak fitted results for complex 2a. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.12 0.255 0.155141 0.0260575 0.124316 0.00326066 0.124286 0.00333189 G1 0 1 0.503641 0.256758 0.986857 0.0893549 1  NaN O1 2476 2478.6 2477.19 0.740609 2477.1 0.426467 2477.26 0.184795 W1 0 2.5 1.23302 0.633378 0.729058 0.216568 0.783444 0.245793 I2 0 3.56 1.71752 1.00499 2.86037 0.124814 2.89668 0.292063 G2 0 1 0.45288 0.285177 0.0271982 0.0481386 0.0129662 0.133925 W2 0 1.19 0.822904 0.238977 0.632716 0.0520398 0.637212 0.049569 O2 2473.8 2474.5 2474.19 0.193632 2474.17 0.0411194 2474.17 0.0592574 I3 0 4.34 2.2939 1.18926 3.17379 0.175754 3.22337 0.260321 O3 0.5 1.2 0.855323 0.206164 0.873168 0.0447447 0.878647 0.0550629 I4 0 3.11 1.62685 0.854773 2.64729 0.381874 2.77457 0.273186 O4 0.5 1.5 0.966931 0.27956 1.10675 0.126442 1.15095 0.0712536   150  2b. Peak fitted results for complex 2b. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.12 0.855 0.306338 0.0754835 0.29327 0.00507634 0.292413 0.00604917 G1 0 1 0.504747 0.282795 1 3.80E-09 1  NaN O1 2476 2478.6 2477.23 0.713192 2477.4 0.150895 2477.43 0.122369 W1 0 2.5 1.12051 0.699113 0.774849 0.156413 0.753549 0.173375 I2 0 3.56 1.90681 0.962012 2.92175 0.252446 2.96359 0.3829 G2 0 1 0.475225 0.274922 0.0956256 0.0641679 0.0859078 0.125486 W2 0 1.19 0.803103 0.267369 0.67574 0.013736 0.673722 0.0560822 O2 2473.8 2474.5 2474.22 0.195111 2474.27 0.0337253 2474.27 0.0637375 I3 0 4.34 2.27487 1.1978 2.30391 0.119486 2.30591 0.322586 O3 0.5 1 0.752735 0.139809 0.771694 0.0270685 0.775241 0.0735744 I4 0 3.11 1.55131 0.825494 2.65594 0.162205 2.6801 0.204692 O4 0.75 1.3 1.01587 0.153147 1.15465 0.0133195 1.15283 0.0860011   151  3a. Peak fitted results for complex 3a. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.068 0.191 0.162978 0.019141 0.174105 0.00329274 0.172434 0.0047692 G1 0 1 0.475496 0.278803 0.986457 0.0669339 1  NaN O1 2478.4 2479.3 2478.78 0.245627 2478.43 0.0839717 2478.4  NaN W1 0.00936 0.67 0.346422 0.195757 0.350154 0.118384 0.281058 0.104501 I2 0 3.13 1.5971 0.809429 2.01943 0.175363 2.08035 0.249136 G2 0 1 0.414492 0.265539 0.114988 0.0911067 0.0495228 0.097076 W2 0.125 0.823 0.665392 0.099387 0.770355 0.0223647 0.753933 0.0527629 O2 2476.5 2477.3 2476.93 0.214902 2477.14 0.0746859 2477.1 0.0758509 I3 0 4 2.58531 0.922087 2.07838 0.313944 2.28822 0.368455 O3 0.75 1.5 1.12664 0.20457 1.00366 0.091065 1.04403 0.0750988 I4 0 6 4.28231 1.32769 5.0038 0.392066 5.03086 0.3456 O4 1 1.7001 1.2955 0.176457 1.09822 0.0640549 1.11675 0.0906606   152  3b. Peak fitted results for complex 1a. Parameter  Lower  Upper  Ave. (Prefits)  Std. (Prefits)  Ave. (Fits)  Std. (Fits)  Value (SingleFit)  95% cint I1 0.024 0.134 0.113781 0.0111915 0.125719 6.52E-05 0.125746 0.00220012 G1 0 1 0.423544 0.30079 0.999994 4.12E-05 1  NaN O1 2479.5 2482 2480.1 0.506445 2480.01 0.00463287 2480.02 0.0869251 W1 0.00264 0.558 0.308928 0.153539 0.296912 0.0013961 0.301583 0.0632076 I2 0.09 3.71 2.05749 0.786637 2.37437 0.0049125 2.35782 0.131116 G2 0 1 0.488711 0.256238 0.413007 0.0027241 0.419465 0.076412 W2 0.065 0.817 0.692303 0.0994842 0.72529 0.00192617 0.730295 0.0308457 O2 2476.5 2477.2 2477.05 0.131518 2477.15 0.00262287 2477.16 0.0418552 I3 0.35 4.87 3.66294 0.954029 3.16857 0.00633512 3.15302 0.196136 O3 1 1.4262 1.26027 0.114531 1.2127 0.00144236 1.20367 0.0351816 I4 0 5.97 5.30023 0.500945 5.96801 0.00683777 5.97  NaN O4 0.7 1.2 1.00285 0.14082 0.987345 0.00137411 0.983731 0.0435188   153  A.2. Peak fitting of S K-edge XAS for chapter 4 3a. Peak fitted results for complex 3a. Coeff  Low  Up  Avg  Err I1 0.0008 0.005 0.003438 7.82E-05 G1 0 1 0.934698 0.140492 O1 2478.5 2484 2478.68 0.338813 W1 0.25 1.5 0.555725 0.182142 m1 1.00E-08 5.00E-06 1.06E-06 1.12E-06 YI 0.009 0.0806 0.017201 0.002747 m2 5.00E-07 0.0002 6.21E-06 2.04E-06 I2 0.9 3 1.1439 0.216804 G2 0 1 0.203669 0.102547 W2 0.1 0.7 0.474938 0.062813 O2 2476 2477.3 2476.93 0.074953 I3 1 4 2.00845 0.258904 W3 0.4 0.8 0.786062 0.042139 O3 0.7 1.2 1.01865 0.098823 I4 1 6 5.79798 0.326125 W4 0.5 1 0.841612 0.042523 O4 0.8 1.5 1.19057 0.123122      154  3aZn2+. Peak fitted results for complex 3aZn2+ Coeff  Low  Up  Avg  Err I1 0.0005 0.005 0.001038 7.11E-06 G1 0 1 0.994777 0.018938 O1 2478 2484 2479.28 0.561714 W1 0.25 1.5 0.34201 0.11951 m1 1.00E-10 1.50E-05 8.92E-09 8.35E-08 YI 0.009 0.0806 0.01529 0.000206 m2 5.00E-10 0.0002 5.34E-08 1.30E-07 I2 0.1 2 0.910188 0.550435 G2 0 1 0.890658 0.066586 W2 0.1 0.8 0.586554 0.083585 O2 2476 2477.3 2477.09 0.225226 I3 1 3 2.35504 0.384046 W3 0.4 1 0.629359 0.063246 O3 0.7 1 0.749108 0.064717 I4 2 4 3.25321 0.488368 W4 0.3 0.8 0.755317 0.046278 O4 0.5 0.9 0.832329 0.086293 I5 3 5 4.01951 0.680711 W5 0.5 0.9 0.890345 0.018289 O5 0.5 2 0.979501 0.148579     155  A.3.1. TDDFT spectra of complexes 4-6 and Peak fitting of S K-edge XAS for chapter 5 A.3.1a. TDDFT spectra of complexes 4-6      156  A.3.1b. Peak fitting of S K-edge XAS of complexes 4-6 5a. Peak fitted results for complex 5a. Coeff  Low  Up  Avg  Err  0.0008 0.005 0.001816 9.68E-05 1 0 1 0.839146 0.15472 O1 2475.5 2480 2475.81 0.03009 W1 0.25 1.5 1.42531 0.107836 m1 2.00E-06 5.00E-06 3.53E-06 9.30E-07 YI 0.009 0.0806 0.013711 0.002281 m2 5.00E-06 0.0002 8.54E-06 1.62E-06 I2 1.5 4 2.07707 0.159636 G2 0 1 0.460884 0.031948 W2 0.1 0.7 0.568023 0.008403 O2 2472 2472.5 2472.28 0.008984 I3 0.5 2.5 0.505111 0.038484 W3 0.4 0.8 0.799151 0.006201 O3 0.7 1.2 1.01603 0.178059 I4 1 4 2.9199 0.337419 W4 0.5 1 0.889426 0.020134 O4 0.8 1.5 0.894663 0.17079   157  5b. Peak fitted results for complex 5b. Coeff  Low  Up  Avg  Err I1 0.0008 0.005 0.001918 4.51E-05 G1 0 1 0.999283 0.004088 O1 2475.5 2480 2475.5 0.001837 W1 0.25 2 1.71451 0.167347 m1 1.00E-06 4.72E-06 1.01E-06 4.72E-08 YI 0.009 0.0806 0.018943 0.000117 m2 5.00E-06 0.0002 8.80E-06 6.71E-07 I2 1.5 5 2.35279 0.062741 G2 0 1 0.517588 0.019571 W2 0.4 0.7 0.672867 0.004059 O2 2472 2472.5 2472.19 0.001419 I3 0.5 2.5 1.1569 0.103983 W3 0.4 0.7 0.623371 0.012081 O3 0.8 1.785 1.49197 0.010745 I4 1.5 4 1.52631 0.108763 W4 0.5 0.8 0.799848 0.000977 O4 0.5 1.5 0.711385 0.009402      158  5c. Peak fitted results for complex 5c. Coeff  Low  Up  Avg  Err I1 0.0005 0.005 0.001592 7.68E-05 G1 0 1 0.982911 0.041207 O1 2474.5 2480 2475.96 0.036509 W1 0.25 1.5 1.39624 0.191471 m1 -4.72E-05 -4.72E-07 -8.18E-06 4.16E-06 YI 0.0206 0.0806 0.049062 0.010273 m2 6.00E-07 0.000196 5.42E-06 1.71E-06 I2 1 3 2.40103 0.115025 G2 0 1 0.414938 0.063225 W2 0.1 1 0.517129 0.00678 O2 2472.3 2472.4 2472.34 0.001552 I3 0.3 3 0.87869 0.196823 O3 0.8 1.5 1.44619 0.059105 W3 0.5 0.8 0.731353 0.042253 I4 0.5 3 2.15073 0.20933 W4 0.5 0.8 0.799492 0.003556 O4 0.5 1 0.62189 0.045914       159  6a. Peak fitted results for complex 6a. Coeff Low Up Avg Err I1 0.0005 0.005 0.001767 0.000106 G1 0 1 0.9998 0.001527 O1 2474 2480 2476.67 0.08184 W1 0.25 1.5 1.45791 0.148073 m1 -0.00047 -4.72E-07 -1.67E-05 7.24E-06 YI 0.0206 0.0806 0.054584 0.017858 m2 1.00E-08 0.000196 7.40E-07 1.77E-06 I2 1 6 2.15721 0.127548 G2 0 1 0.47571 0.054058 W2 0.1 1 0.482716 0.005518 O2 2472.2 2472.6 2472.41 0.001146 I3 1 6 3.05459 0.230126 O3 1 2 1.61176 0.006751 W3 0.5 1.5 0.995231 0.028027     160  6b. Peak fitted results for complex 6b. Coeff  Low  Up  Avg  Err I1 0.0001 0.09 0.000945 9.12E-05 G8 0 1 0.01978 0.073391 O1 2463 2467 2463.88 0.918916 W8 0.5 3.5 3.42322 0.214313 I2 0.0001 0.02 0.001697 0.000156 G9 0 1 0.995145 0.022986 O2 2476 2478 2476.53 0.056616 W9 0.5 3 1.29594 0.323296 m1 1.00E-08 0.001 3.43E-06 9.82E-07 YI 0.001 0.01 0.005445 0.002413 m2 1.00E-06 0.001 4.20E-06 2.61E-06 I3 10 18 17.7456 0.805753 G1 0 1 0.170415 0.215371 W1 2 5 4.37967 0.195036 O4 2460 2462 2460.9 0.161563 I4 2 8 7.8119 0.550408 G2 0 1 0.431663 0.234338 W2 0 2 0.39147 0.488092 O5 2462 2467 2462.08 0.150085 I6 1 3 1.49628 0.142153 G3 0 1 0.592478 0.036032 W3 0.1 1 0.522858 0.010343 O7 2473 2474.5 2473.17 0.015311 I7 1 3 2.18713 0.391832 W4 0 0.8 0.567699 0.047166 O8 0.5 1.1 0.93861 0.02155 I8 1 3 1.99216 0.473306 W5 0.5 1.5 0.742309 0.078488 O9 0.5 1.5 0.881694 0.057641   161  APPENDIX B: MOLECULAR ORBITAL DESCRIPTION AND INPUT FILES B.1. Molecular Orbital (MO) Description of complexes 1a,b-3a,b (1a) MO description of complex 1a. Desc orbi tal energy (eV) occ Ru SR ar en Expt energy (eV) Assignment 4d 5s 5p 3p 3s R enσ* -SCσ* 75a -1.667 0  5 3 4  16 23 42 2473.0  SCσ* ← S1s arπ* -SCσ* 74a -1.733     2  35 43 13 arπ* - SCσ* 70a -2.194 0    4  7 48 36 CHσ*-SCσ* 69a -2.505 0 -1 -1 -2 16 2 16 34 27 enσ* - SCσ*  66a -3.214 0  1 -3 7 1 5 17 63 arπ* 64a -4.263 0 20      73 3 arπ* 63a -4.423 0 20   1   75 Ru 4dσ* 62a -5.126 0 47   20   21 5 2471.1  Ru4d(σ*)←S1s Ru 4dσ* 61a -5.508 0 53      28 15 S 3pπ 60a -7.260 2 20  1 67  4 2 Ru4dπ+S3pπ 59a -8.240 2 50   18 1 3 15 2 Ru 4dπ 58a -8.495 2 61 2  9  1 16 1 Ru4dπ+S3pπ 57a -8.897 2 69   16  1 3 S3pπ + RuSσ 56a -9.234 2 41   33 2 5 7 1 SCσ 55a - 10.208 2 3   22  40 24 1  (1b) MO description of complex 1b. Desc orbi tal energy (eV) occ Ru SR ar en Expt energy (eV) Assignment 4d 5s 5p 3p 3s R arπ* - SCσ* 82a -1.509 0    4  9 52 24 2473.0 SCσ* ← S1s arπ* - SCσ* 81a -1.624 0  7 2 5  21 47 12 SCσ* - enσ*  75a -2.327 0    13  12 24 36 enσ* - SCσ* 72a -3.179 0 -1   7  6 18 60 Ru-arπ* 70a -3.942 0 18      76 2 162  Desc orbi tal energy (eV) occ Ru SR ar en Expt energy (eV) Assignment 4d 5s 5p 3p 3s R Ru-arπ* 69a -4.114 0 15   1   78 Ru-Sσ* 68a -4.902 0 46   19   20 5 2471.2 Ru4d(σ*)←S1s Ru4dσ* 67a -5.112 0 54      24 15 S 3pπ 66a -7.068 2 23  1 64  2 3 Ru4dπ+S3pπ 65a -7.961 2 51  1 18   18 Ru 4dπ 64a -8.162 2 69 1  1  3 14 3 Ru4dπ+S3pπ 63a -8.690 2 63   19   5 S3pπ + RuSσ 62a -8.969 2 32 1  41 2 5 8 SCσ 61a -9.904 2 6 1  15 1 20 43 4  (1c) MO description of complex 1c. Desc orbi tal energy (eV) occ Ru SR ar en Expt energy (eV) Assignment 4d 5s 5p 3p  3s R Rπ* - SCσ* 95a -0.093 0  2  3  30 49 9 2473.3 SCσ* ← S1s enσ* - SCσ* 89a -0.715 0  14 1 5  1 43 30 CHσ* - SCσ* 88a -0.816 0  6 2 1  24 57 6 Rπ* - SCσ* 86a -1.102 0    5 -1 41 34 14 CHσ* - SCσ* 85a -1.185 0    1  12 39 35 CHσ* - SCσ* 83a -1.371 0  26 2 2  5 42 18 CHσ* - SCσ* 82a -1.565 0  -1 2 3  9 52 27 CHσ* - SCσ* 80a -1.798 0    1  2 65 25 CHσ* - SCσ* 79a -1.939 0  7  11  8 42 29 CHσ* - SCσ* 78a -2.004 0 -1   1  8 42 41 arπ* - SCσ* 77a -2.113 0  1 1 2  9 54 23 enσ* - SCσ* 76a -2.299 0  5  1  4 32 49 CHσ* - SCσ* 75a -2.601 0  1 -4 9 1 12 44 34 enσ*- SCσ* 74a -2.739 0  4  3  6 17 59 Rπ*- Ru-S4dσ* 72a -3.543 0 4   8  67 4 10 2472.5 Rπ* ← S1s 163  Desc orbi tal Energy (eV) occ Ru SR ar en Expt energy (eV) Assignment 4d 5s 5p 3p 3s R Ru-arπ* 69a -4.250 0 22     1 72 2 Ru-arπ* 68a -4.316 0 16     3 75 1 Ru4dσ* 67a -5.169 0 44   16   18 5 2471.3 Ru 4dσ* ← S1s Ru4dσ* 66a -5.455 0 53   1   28 14 S 3pπ 65a -7.602 2 21   58  1 9 4 Ru 4dπ+S3pπ 64a -8.162 2 20   21  42 9 Ru 4dπ + Rπ 63a -8.659 2 69     20 2 Ru 4dπ (NB) 61a -8.785 2 70     10 10 S3pπ + RuSσ 60a -9.142 2 49   26   11 1 S3pπ + Rπ 59a -10.031 2 10   38 1 36 2 1  (2a) MO description of complex 2a. Desc orbi tal energy (eV) occ Ru SOR ar en M eO H Expt energy (eV) Assignment 4d 5s 5p 3p 3s O R CHσ*- enσ*-SCσ* 89a -1.118 0  6  2   7 45 30 4 2476.0 SCσ* ← S1s enσ*- SCσ*-SOσ* 83a -1.976 0  5  8 1 1 15 36 18 7 arπ*- SCσ* 81a -2.139 0  1  2   1 77 6 1 CHσ*- SCσ* 80a -2.305 0  -1 -1 5   3 55 32 SCσ*-SOσ* 79a -2.399 0 1 2 1 22 2 2 19 18 25 CHσ*-SCσ* 77a -2.981 0    4   7 38 41 SOσ* - enσ* 76a -3.185 0  13  30 2 5 3 21 15 2474.8 SOσ*←S1s enσ* - SOσ* 75a -3.513 0  -6  22  2 5 15 49 2 Ru-ar π* 74a -4.431 0 22       74 2 Ru-ar π* 73a -4.564 0 22   2    74 Ru4dσ*-SOπ* 72a -5.127 0 46   17 1 4 1 23 6  2474.2 Ru4dσ*←S1s Ru4dσ* 71a -5.579 0 53       27 17 SOπ* 70a -7.730 2 20   25 2 27 4 9 2 Ru 4dπ 69a -8.394 2 52  1 2  13 4 16  1 Ru 4dπ (nb) 68a -8.601 2 63  2   3  4  8 164  Desc Orbi tal Energy (eV) occ Ru SOR ar en M eO H Expt energy (eV) Assignment 4d 5s 5p 3p 3s O R MeOH 67a -8.732 2 25     2    66 Ru 4dπ 66a -8.902 2 59   2  10  12 1 5 SOπ (LP) 65a -9.077 2 18   4  46 11 2  6  (2b) MO description of complex 2b. Desc orbit al energy (eV) oc c Ru SOR ar en M e O H Expt energy Assign ment 4d 5s 5p 3p 3s O R enσ*-SCσ* 90a -1.841 0  1  1   9 46 32 2476.0 SCσ* ← S1s CHσ*-SCσ* 89a -1.959 0  -1  3   19 39 34 2 CHσ*-SCσ*-SOσ* 88a -1.985 0  2  13 2 2 18 44 8 arπ*- SCσ*- SOσ* 87a -2.113 0   2 6 1 2 10 52 15 arπ*- SCσ*- SOσ* 86a -2.283 0  2 2 4 1 2 8 63 6 CHσ*-SCσ* 85a -2.532 0 1 3  9 1  11 39 29 CHσ*-SOσ* 84a -2.653 0  5  2   9 40 37 2474.8 SOσ*←S 1s CHσ*-SOσ* 83a -2.771 0  9  8  3 5 30 33 SOσ* 82a -3.309 0 5  3 40  4 9 13 17 Ru-arπ* 80a -4.263 0 18  1 2    75 2 Ru-arπ* 79a -4.284 0 17   1    79 Ru4dσ* 78a -4.952 0 48   18  4  20 5  2474.3 Ru4dσ*← S1s Ru4dσ* 77a -5.219 0 55       26 17 SOπ* 76a -7.625 2 17   27 3 3 2 4 5 Ru 4dπ 75a -8.001 2 55  2   4 5 15 1 5 MeOH 74a -8.113 2 4         91 Ru 4dπ 73a -8.308 2 67     4  15 2 1 Ru 4dπ (nb) 72a -8.496 2 76 1    7  5  1 SOπ (LP) 71a -8.884 2 17   5 4 9 10 2 2 7   165  (3a) MO description of complex 3a. Desc orb ital energy (eV) occ Ru SO2R ar en Expt energy Assignment 4d 5s 5p 3p 3s O R CHσ* -SOCσ* 95a -0.860 0  3 -2 6 1 2 21 41 19  SCσ* ← S1s/ SOσ*←S1s CHσ* -SCσ* 94a -1.005 0  6  6 3 3 27 30 22 CHσ* -SCσ* 92a -1.281 0    3 1  16 47 24 CHσ* -SCσ* 91a -1.345 0  8 3 3 1  7 55 14 RΠ* - SOCσ* 88a -1.689 0  5  5 2 2 37 31 7 CHσ*-SCσ*+ O2pπ 87a -1.883 0  2  4  1 12 64 8 CHσ*-SCσ*+ O2pπ 86a -1.952 0 -1 2 2 5  1 10 42 27 arπ*-SCσ*+ O2pπ 83a -2.421 0 3  2 27  7 5 13 34 2479.1 arπ* - SCσ*+O2pπ 82a -2.495 0  1 -1 5  1 2 57 26 CHσ* -SOCσ* 81a -2.655 0  13 1 3   7 25 42 CHσ* - SOσ* 80a -3.041 0  2 4 12  5 6 18 34 2477.6 SOσ*←S1s CHσ* - SOσ* 79a -3.436 0 2   9  8 12 9 50 Rπ*-Ru4dσ*- SOσ* 76a -4.283 0 8   2 5 7 39 22 7  R π* ← S1s Ru-arπ* 75a -4.443 0 16      5 75 Ru-arπ* 74a -4.539 0 11   1  1 25 55 Ru4dσ* 73a -5.163 0 35 4 1 4 3 9 10 14 16 2476.9 Ru4d(σ*)←S1s Ru4dσ* 72a -6.098 0 52 2  1 1 1  23 11 Ru 4dπ 71a -7.707 2 58  2 1 1 4  14 8 O2pπ 70a -8.001 2 17 1  3 2 35 8 5 19 Ru 4dπ (NB) 69a -8.113 2 58     14 3 11 2 Ru 4d 68a -8.308 2 63     9 3 13 2 Rπ 66a -8.496 2 10   1  8 68 4 2 SOπ 65a -8.884 2 23   8  31 21 6     166  (3b) MO description of complex 3b. Desc orbit al energy (eV) oc c Ru SO2R ar en Expt energy Assignment 4d 5s 5p 3p 3s O R enσ*-SOCσ* 107a -0.523 0    3  3 19 38 24 SCσ* ← S1s / SOσ*←S1s CHσ*- SOCσ* 106a -0.560 0  1  2 1 1 3 79 4 enσ*-SOCσ* 103a -0.799 0   2 3 3 3 16 38 26 CHσ*- SCσ* 102a -0.848 0    1   9 52 22 CHσ*- SCσ* 101a -0.859 0    1   7 59 18 CHσ*- SOCσ* 97a -1.367 0    1  2 19 57 15 CHσ*- SOCσ* 96a -1.464 0   3 11  2 8 54 12 arπ*-SOCσ* 94a -1.692 0 1  3 18  5 19 34 9 2479.2 CHσ*-SCσ*-SOπ* 93a -1.718 0  1 3 16 2 4 8 41 15 CHσ*- SOCσ* 92a -1.772 0  13  12  3 22 39 2 2478.3 CHσ*- SOCσ* 91a -1.845 0  4  7   8 47 22 enσ*-SCσ*-SOπ* 89a -2.139 0  1  4   18 21 47 SOπ*←S1s arπ* - SOπ* 88a -2.316 0  3 1 2   4 77 3 Π*- Ru4dσ* 82a -4.086 0 10    3 3 48 25   Rπ* ← S1s Ru-arπ* 81a -4.191 0 19      4 71 3 Ru-arπ* 80a -4.348 0 12      26 55 2 Ru4dσ* 79a -5.226 0 45   7 3 4 10 21 5 2477.0 Ru4dσ* ← S1s Ru4dσ* 78a -5.301 0 50       30 16 Ru 4dπ 77a -8.233 2 59     10  17 4 SOπ* 76a -8.332 2 44 2 2 3 1 17 4 17 Ru 4dπ (NB) 75a -8.676 2 73     4 13 2 Ru4dπ+ O2pπ 74a -8.870 2 39   2  27 13 8   167  B.2: Selected Input files for each type of calculations: ADF Input files for Chapter 3 can be found in the supplementary information of the related publication.96 Selected ORCA files are given here. 1. Geometry optimization for scalar-relativistic effect of complex 2aZn ! RKS OPT BP86 TZVP def2-TZVP/J ZORA Split-RI-J SlowConv SCFCONV6 ! NormalSCF ! Normalprint ! Grid4 Nofinalgrid  % SCF MAXITER 500        end  % basis NewGTO 44 "TZVPP" end         NewAuxGTO 44 "TZVPP/J" end         End  * xyz 3 1 C    -0.605820     1.329045    -0.301372 C     0.019674     0.069592    -0.196767 C     1.433476    -0.048823     0.055612 C     2.162258     1.155241     0.240143 C     1.536400     2.438854     0.168769 C     0.142411     2.540092    -0.082346 Ru     0.468086     1.199897     1.686634 S     1.762016     1.908078     3.632934 O     1.417680     0.883062     4.797851 C     2.159066    -1.383199    -0.032845 C     1.504193    -2.535208     0.742545 C    -0.528322     3.883224    -0.183230 N    -1.090289     2.199998     2.852804 C    -1.613983     1.329132     3.962912 C    -1.514440    -0.119472     3.519908 N    -0.118840    -0.343211     3.042678 C     2.325892    -1.746891    -1.525717 C     3.623538     1.669218     3.349676 C     4.338108     2.451960     4.452675 C     3.977504     0.190291     3.343505 H     3.229102     1.090916     0.457483 H     2.120542     3.343366     0.340535 H    -1.679011     1.385798    -0.491970 H    -0.584132    -0.831837    -0.311700 H    -1.581702     3.847816     0.128937 H    -0.006813     4.645377     0.410092 H    -0.520639     4.217351    -1.233072 H     3.166643    -1.226974     0.386437 H     2.057732    -3.467320     0.563978 H     1.524990    -2.357394     1.827766 H     0.465988    -2.717409     0.425408 H     2.935735    -2.655660    -1.624808 H     2.820749    -0.942080    -2.086783 H     1.351778    -1.942375    -1.998841 H     3.801528     2.142309     2.371009 H     5.427048     2.369429     4.317169 H     4.088646     2.045752     5.442914 H     4.075079     3.518502     4.437719 H     5.066596     0.065236     3.250912 H     3.659284    -0.277184     4.285890 H     3.498026    -0.348775     2.514391 H    -1.754667    -0.789507     4.360639 H    -2.210888    -0.338654     2.696785 H    -0.967860     1.487652     4.835820 H    -2.646075     1.603663     4.223368 H    -0.019277    -1.291126     2.669740 H     0.539883    -0.198907     3.867610 H    -0.732076     3.076229     3.249739 H    -1.853526     2.457501     2.220622 Zn     2.471942     1.140174     7.027167 newgto "TZVPP" end *    2. COSMO solvation effect on complex 2aZn ! RKS OPT BP86 TZVP def2-TZVP/J ZORA Split-RI-J SlowConv SCFCONV6 COSMO(water) ! NormalSCF ! Normalprint ! Finalgrid6  % SCF MAXITER 500        end  % basis NewGTO 44 "TZVPP" end         NewAuxGTO 44 "TZVPP/J" end         end  * xyz 3 1 C      -0.815221      0.805373     -0.064177   C       0.141608     -0.252559      0.111665   C       1.539574     -0.000794      0.125374   C       1.920556      1.381748      0.213344   C       0.976799      2.435036      0.104111   C      -0.410444      2.157811     -0.164403 168    Ru      0.390831      1.144225      1.867884   S       1.891916      1.040909      3.583171   O       1.613767      0.023097      4.815817   C       2.575115     -1.090831      0.012322   C       2.195240     -2.420496      0.675975   C      -1.371859      3.254640     -0.480778   N      -0.639462      2.725643      2.957710   C      -1.438486      2.184861      4.113919   C      -2.072045      0.887133      3.662957   N      -0.973982      0.015733      3.104618   C       2.853403     -1.281890     -1.509699   C       3.834337      1.198331      3.435300   C       4.272327      2.283825      4.404937   C       4.476626     -0.153164      3.627407   H       2.976908      1.636943      0.317125   H       1.324289      3.469642      0.134125   H      -1.873388      0.552928     -0.164056   H      -0.215484     -1.283312      0.146594   H      -2.414482      2.993154     -0.257953   H      -1.104320      4.212879     -0.013218   H      -1.322520      3.432928     -1.572216   H       3.506399     -0.710597      0.465033   H       3.040709     -3.117725      0.610513   H       1.943897     -2.293135      1.742295   H       1.349123     -2.913512      0.176134   H       3.666607     -2.011295     -1.626669   H       3.164973     -0.348972     -1.998652   H       1.967084     -1.672853     -2.028927   H       3.903247      1.528220      2.388213   H       5.356932      2.431254      4.271321   H       4.104242      1.998063      5.452177   H       3.789032      3.250263      4.212248   H       5.566392     -0.033733      3.498769   H       4.373849     -0.528231      4.666327   H       4.146271     -0.914273      2.907161   H      -2.582026      0.377308      4.493489   H      -2.810228      1.053037      2.865993   H      -0.749251      2.023502      4.954224   H      -2.194874      2.916553      4.433880   H      -1.389708     -0.772754      2.595479   H      -0.444348     -0.395572      3.886027   H       0.004844      3.453062      3.290779   H      -1.272399      3.199259      2.300555   Zn      2.370864     -1.748915      5.347099  newgto "TZVPP" end *  3. Single point calculation for TDDFT-XAS of  3aZn  ! UKS BP86 TZVP def2-TZVP/J ZORA Split-RI-J SlowConv SCFCONV7 COSMO(water) ! TightSCF ! Normalprint ! Grid4 NoFinalgrid  % basis NewGTO 44 "TZVPP" end         NewAuxGTO 44 "TZVPP/J" end         end  % method SpecialGridAtoms 44          SpecialGridIntAcc 7          end  % tddft  NRoots 200          MaxDim 2000          OrbWin[0]=6,6,-1,-1          OrbWin[1]=6,6,-1,-1   DoQuad true          end  * xyz 3 1 C       7.020763      3.749904     15.596716 C       7.191419      3.068214     16.832147 C       6.393494      3.480206     17.948840 C       5.514838      4.584083     17.822126 C       5.383511      5.334620     16.605635 C       6.121912      4.860274     15.491661 Ru      4.982315      3.099538     16.239431 N       5.010056      0.959287     16.176530 C       4.314594      0.420744     14.962081 C       4.568783      1.371525     13.813520 N       4.203214      2.745113     14.281800 C       8.140253      1.917042     16.954106 C       4.529954      6.587471     16.564607 C       5.384469      7.774593     17.066091 S       2.774881      2.854324     16.976043 O       1.758478      2.880268     15.812230 Zn      0.907298      0.625918     18.080575 newgto "TZVPP" end O      -0.008059      1.182306     19.760741 C       2.124677      4.060163     18.186354 C       1.238698      5.042609     17.737788 C       0.731223      5.954252     18.670310 C       1.120727      5.881022     20.012126 C       2.013042      4.889128     20.438497 C       2.528275      3.967199     19.523033 O       2.649591      1.473898     17.777745 O      -0.143224      1.111441     16.464421 C       3.914778      6.886674     15.192387 H       3.208735      3.181488     19.854202 H       2.310605      4.828650     21.486031 H       0.722055      6.596805     20.732317 H       0.030624      6.723216     18.341591 H       0.934343      5.083149     16.692077 H       3.709665      6.438023     17.286458 H       9.105516      2.305255     17.317802 H       8.327622      1.433753     15.986620 H       7.796051      1.175458     17.686989 H       3.175170      2.837666     14.326603 H       4.533906      3.439964     13.605943 H       5.629476      1.381768     13.532613 169  H       3.975610      1.097887     12.930188 H       3.241821      0.366002     15.188428 H       4.672229     -0.591985     14.737007 H       5.978808      0.623565     16.195147 H       4.553532      0.592016     17.020731 H       5.996440      5.335971     14.518586 H       7.566724      3.416781     14.714507 H       6.458521      2.942783     18.893825 H       4.897865      4.860008     18.678550 H       5.795160      7.586028     18.068241 H       4.765025      8.681309     17.116228 H       6.222565      7.966610     16.379888 H       3.241123      7.751032     15.271978 H       3.328320      6.036765     14.812018 H       4.682919      7.138525     14.446440 H      -1.098758      1.302686     16.502224 H       0.324143      1.882773     16.031375 H       0.447264      1.021561     20.608821 H      -0.936018      0.911156     19.898052 O       1.209998     -1.343644     18.063277 H       1.616445     -1.757982     18.847861 H       0.464502     -1.920422     17.808433 *   4. Single point calculation for UV-vis simulation of 5b  ! RKS SP B3LYP VDW10 def2-TZVP def2-TZVP/J ZORA RIJCOSX SlowConv SCFCONV6 COSMO(Water) ! TightSCF ! Normalprint ! Grid4 Finalgrid6 % basis NewGTO 44 "def2-TZVPP" end         NewAuxGTO 44 "def2-TZVPP/J" end         end % tddft NRoots 100  MaxDim 500  end * xyz 1 1 C       4.138244      3.751845     15.344398   C       2.748812      3.624089     15.147494   C       2.010992      4.828880     14.955091   C       2.634128      6.086848     14.964961   C       4.028963      6.204797     15.244039   C       4.764392      5.024521     15.430748   Ru      3.630714      4.799121     13.492852   Cl      4.110517      6.488845     11.811755   C       2.074504      2.289004     15.109741   C       4.710897      7.544597     15.404748   C       3.997550      8.711357     14.722059   N       4.961735      3.611783     12.324986   N       4.327334      2.700637     11.485634   C       3.019543      2.744410     11.235192   S       2.053577      3.957406     11.936429   C       6.256636      3.499318     12.284715   C       7.268830      4.271146     12.947071   C       7.164940      5.621476     13.329483   C       8.206535      6.276743     13.943221   C       9.429186      5.613679     14.239581   C       9.551554      4.267416     13.804837   C       8.509172      3.635376     13.168072   N      10.439076      6.244635     14.889462   N       2.516624      1.797877     10.420967   C       1.121212      1.471261     10.395421   C       0.230840      2.227302      9.644740   C      -1.123296      1.905065      9.651448   C      -1.563808      0.819595     10.401349   C      -0.674572      0.050157     11.144522   C       0.674716      0.383729     11.138810   I      -3.631507      0.328260     10.424839   C       4.872410      7.820499     16.911236   H       6.644092      2.673288     11.689095   H       5.829945      5.080291     15.600563   H       4.746262      2.859571     15.417604   H       0.958587      4.768732     14.710567   H       2.043668      6.962960     14.746887   H       1.274396      2.275434     14.369997   H       1.634127      2.071405     16.086207   H       2.786638      1.499095     14.872619   H       5.708809      7.447088     14.971983   H       4.605506      9.611816     14.822385   H       3.033720      8.915479     15.194005   H       3.833683      8.513971     13.663462   H       5.422201      8.751299     17.062848   H       5.414517      7.016965     17.412093   H       3.894821      7.924399     17.387940   H       3.149035      1.081510     10.088835   H       4.914500      2.002510     11.043438   H       0.587905      3.067557      9.064924   H      -1.816118      2.498684      9.071837   H      -1.015890     -0.791266     11.730894   H       1.378338     -0.195034     11.723438   H       8.644825      2.610715     12.841252   H      10.470060      3.722406     13.960320   H       8.084662      7.321116     14.188785   H       6.265038      6.169443     13.097872   C      10.325738      7.653406     15.245862   C      11.705932      5.561709     15.126412   H      12.364061      6.220432     15.684912   H      11.560265      4.651695     15.713390   H      12.201036      5.290957     14.188912   H      11.216717      7.954221     15.788948   H      10.232243      8.286882     14.358561   H       9.462341      7.831963     15.891343   *   5. NBO calculation of 2a # NBO calculation of [Ru(p-cymene)(en)(SO-iPr)]+ in gas phase ! RKS BP86 SVP ZORA NORI ! TightSCF VerySlowConv 170  ! Grid4 NoFinalgrid ! Normalprint ! NBO % maxcore 1024 % scf maxiter 1000      end % pal nprocs 8      end * xyz 1 1 C       0.019674      0.069592     -0.196767   C       1.433476     -0.048823      0.055612   C       2.162258      1.155241      0.240143   C       1.536400      2.438854      0.168769   C       0.142411      2.540092     -0.082346   C      -0.605820      1.329045     -0.301372   Ru      0.468086      1.199897      1.686634   N      -1.090289      2.199998      2.852804   C      -1.613983      1.329132      3.962912   C      -1.514440     -0.119472      3.519908   N      -0.118840     -0.343211      3.042678   C       2.159066     -1.383199     -0.032845   C       2.325892     -1.746891     -1.525717   C      -0.528322      3.883224     -0.183230   S       1.762016      1.908078      3.632934   C       3.623538      1.669218      3.349676   C       3.977504      0.190291      3.343505   C       1.504193     -2.535208      0.742545   O       1.417680      0.883062      4.797851   C       4.338108      2.451960      4.452675   H       3.229102      1.090916      0.457483   H       2.120542      3.343366      0.340535   H      -1.679011      1.385798     -0.491970   H      -0.584132     -0.831837     -0.311700   H      -1.581702      3.847816      0.128937   H      -0.006813      4.645377      0.410092   H      -0.520639      4.217351     -1.233072   H       3.166643     -1.226974      0.386437   H       2.057732     -3.467320      0.563978   H       1.524990     -2.357394      1.827766   H       0.465988     -2.717409      0.425408   H       2.935735     -2.655660     -1.624808   H       2.820749     -0.942080     -2.086783   H       1.351778     -1.942375     -1.998841   H       3.801528      2.142309      2.371009   H       5.427048      2.369429      4.317169   H       4.088646      2.045752      5.442914   H       4.075079      3.518502      4.437719   H       5.066596      0.065236      3.250912   H       3.659284     -0.277184      4.285890   H       3.498026     -0.348775      2.514391   H      -1.754667     -0.789507      4.360638   H      -2.210888     -0.338654      2.696785   H      -0.967860      1.487652      4.835820   H      -2.646075      1.603663      4.223368   H      -0.019277     -1.291126      2.669740   H       0.539883     -0.198907      3.867610   H      -0.732076      3.076229      3.249739   H      -1.853526      2.457501      2.220622  * 171  APPENDIX C: XAS AND DFT ANALYSIS OF RUII/OSII-ARENE (S-S-) LIGATED COMPLEXES C.1: Effect of Chelating Ligand Modification in Os-arene-dithiol Complexes  Figure C.1: Structures of novel five-coordinated [(η6-Benzene)Os(dithiol)] complexes (7) Electronic structures of 7a, [(η6-Benzene)Os(1,2-benzenedithiol)] and 7b, [(η6-Benzene)Os(3,6- dichloro-1,2-benzenedithiol)] are examined using S K-edge XAS in concert with DFT, TDDFT and NBO calculations.  C.1.1: Sulfur K-edge X-ray Absorption Spectroscopy  Figure C.2: (a) Normalized S K-edge XAS of complexes 7a and 7b with peak assignments, (b) XAS and 1st       derivative spectrum of 7a, (c) XAS and its 1st derivative spectrum of 7b.  172  Table C.1: DFT calculated S3p contribution to Os-Sπ* and Os-Sσ* acceptor orbitals in 7a and 7b Complex % S3p in Os-Sπ* % S3p in Os-Sσ* Total % S3p in Os-S 7a 23.8 22.3 46.1 7b 22.4 21.7 44.1 Electronic structures of 7a, [(η6-Benzene)Os(1,2-benzenedithiol)] and 7b, [(η6-Benzene)Os(3,6- dichloro-1,2-benzenedithiol)] are examined using S and Cl K-edge XAS in concert with DFT and NBO calculations. As shown in Figure C.2, the S K-edge spectrum of 7a shows three main features in the near edge region. The lower energy weak feature at 2471.6 eV is assigned to Os5dπ* ← S1s, a feature at 2473.2 eV is assigned to Os5dσ* ← S1s and the higher energy feature at ~2475.0 eV is assigned to S- Cσ* ← S1s. However, the last two features cannot be distinguished in 7b. All the features are assigned based on literature and DFT calculations. The calculated S3p character from DFT is shown in Table C.1. DFT results indicate that the halide substitution on the (S, S-) chelating ligand slightly reduce the S3p character contribution to both orbitals. Therefore, the covalency of the Os-Sπ and Os-Sσ is decreased.  C.1.2: Chlorine K-edge X-ray Absorption Spectroscopy  Figure C.3: (a) Chlorine K-edge XAS of complex 7b. (b) XAS-TDDFT comparison for complex 7b.  Somewhat surprisingly, the pre-edge region of the Cl K-edge XAS spectrum of 7b (Figure C.3a) shows a weak shoulder at 2821.8 eV below the intense C-Clσ* ← Cl1s transition. Based on TDDFT simulations, this weak low-energy feature may correspond to the presence of some Cl 3p character in the empty Os 5d orbitals; specifically overlap of a filled Cl 3p orbitals with the adjacent S 3p orbital 173  involved in σ-bonding with the metal centre (see Figure C.3, Os5dσ* orbital). TDDFT simulations of the Cl K-edge XAS spectrum (Figure C.3b) is in good agreement with experimental XAS. Cl K-edge studies are consistent with the results obtained from S K-edge XAS.  C.1.3: Natural Bond Analysis  Figure C.4: Natural bond orbital pictures of Os-Sσ* orbitals in complexes 7a and 7b. The natural bond analysis of the Os-S bonds in 7a and 7b predicts that both Os-S bonds in each complex are equal and the bond order for 7a (0.833) is slightly higher than 7b (0.0831) as shown in Figure C.4. XAS, DFT, TDDFT and NBO studies on these complexes suggest that the covalency of Os-S bond in 7a is slightly greater than 7b. Therefore, Os-S bond is weakend by the halide substitution on the chelating ligand.  C.2: Different Arene Substitution in Os-arene-dithiane Complexes  Figure C.5: Six-coordinated [(η6-arene)Os(1,4-dithiane)Cl]+ complexes (8). Effect of arene (benzene, p-cymene and biphenyl) ligand substitution is investigated in a series of Os-arene-dithianel complexes, 8a-c using S and Cl K-edges combined with DFT and NBO calculations. 174  C.2.1: Sulfur and Chlorine K-edge X-ray Absorption Spectroscopy  Figure C.6: (a) Sulfur and (b) Chlorine K-edge XAS of complexes 8a-c.    Figure C.7: (a) First derivative spectra of S K-edge XAS and (b) simulated TDDFT spectra of complexes 8a-c.  S K-edge XAS of all these complexes show only a broad and intense feature at 2473.0 - 2476.0 eV, as shown in Figure C.6a. DFT calculation shows that this feature can be assigned to both Os5dσ* ← S1s and S-Cσ* ← S1s transitions. Unlike in Ru-arene complexes, in Os-arene complexes the energy of Os-Sσ* and S-Cσ* are very close (a similar tendency is also observed in Os-arene thiosemicarbazone complexes in Chapter 5); hence they cannot be distinguished in the experimental spectra. The first derivative spectra of the S K-edge XAS (Figure C.7a) also show that there are two features in each spectra. This is again supported by the simulated TDDFT spectra as shown in Figure C.7b. The S K-edge XAS combined with DFT calculations predicts that there is Os-S coordination 175  present in those complexes. However, the covalency of Os-S bond in those complexes cannot be extracted. Therefore, S K-edge XAS fails to investigate the effect of different arene substitution in those complexes. Therefore, Cl K-edge XAS was attempted as a way to investigate the effect of different arene substitution in those complexes. A sharp and well separated feature appears at 2822.5 eV is assigned to Os5dσ* ← Cl1s transition as shown in Figure C.6b. The different arene ligand effect could, in principle, be investigated by quantifying the Os-Cl covalency. However, due to the poor quality of the data, such a comparison is not made here and we require better quality data for quantitative analysis.  C.2.1: Natural Bond Analysis  Figure C.8: Natural bond analysis of Os-S and Os-Cl orbitals of 8a-c NBO analysis indicates that the Os-S covalency and Os-Cl covalency in all the complexes are very close in nature. Each complex has two Os-S bonds, which are slightly different from each other in 8a-c. The lowest covalency for Os-S bond is obtained in 8c, [(η6-bip)Os(dithiane)] whereas the lowest covalency for Os-Cl bond is obtained in 8b, [(η6-p-cym)Os(dithiane)].   176  C.3: Effect of geometry in Ru and Os Dodecarborane complexes  Figure C.9: Structures of five-coordinated, 10, and six-coordinated, 11, RuII/OsII arene dodecarborane       complexes  RuII-arene 16-electron (10a) and 18-electron (11a) complexes as well as OsII-arene 16-electron (10b) and 18-electron complexes (11b) have been investigated through their electronic structure using S K-edge XAS, DFT, and TDDFT analysis     Figure C.10: (a) S K-edge XAS of 10a and 11a are compared with their 1st derivative spectra; (b) S K-edge       XAS-TDDFT comparison of 10a and 11a.  Interestingly, as seen in Figure C.10a, there are 3 features appeared in the S K-edge XAS of 10a, whereas only two features in 11a. The simulated TDDFT spectra of complex 10a also shows 3 main features and those features are assigned to Ru4dπ* ←S1s, Ru4dσ* ← S1s and S-Cσ* ← S1s transitions; whereas in the simulated spectrum of 11a, there are two well separated features appeared and they can be assigned to Ru4dσ* ←S1s and S-Cσ* ← S1s transitions. These features are further resolved in the 1st derivative spectra as shown in Figure C.10b. 177  Therefore, in the five-coordinated complex ligand is coordinated through Ru-Sπ and Ru-Sσ interaction, however, in six-coordinated complex it is only coordinated through Ru-Sσ interaction. This can be attributed that in 16-electron, five-coordinated complex 10a, additional d-electron on the metal is coordinated through a π- interaction with S 3p lone pair in addition to σ-interaction. However, such interaction is not possible in 18-electron, six-coordinated complex 11a.    Figure C.11: (a) S K-edge XAS of OsII complexes 10b and 11b are compared with their 1st derivative spectra;         (b) XAS-TDDFT comparison of complexes 10b and 11b.   Table C.2: DFT calculated S3p contribution to Mndπ* and Mndσ* acceptor orbitals in 10 and 11 Complex % S3p in Mndπ* % S3p in Mndσ* DFT DFT 10a 21.5 20.8 10b 21.0 16.0 11a - 26.9 11b - 26.4  Contrastingly, the S K-edge XAS of five- and six- coordinated OsII-arene complexes, 10b and 11b, are dominated by two features, which is also consistent with the 1st derivative spectra of these complexes as shown in Figure C.11a. However, the simulated TDDFT spectrum of 10b shows two main features and a shoulder on the second feature and the simulation of 11b is very consistent with the experimental XAS results, Figure C.11b. The features are identified with the TDDFT simulation spectra, the Kohn-Sham molecular orbital picture and energy. Interestingly, the analysis indicates that in 16-electron, five-coordinated complex, all three interactions 178  such as Os5dπ* ← S1s, Os5dσ* ← S1s and S-Cσ* ← S1s are present as it is observed in the Ru analogue. However, Os5dσ* ←S1s and S-Cσ* ← S1s transitions are very close in energy, as a result they merge together and look like a single peak in XAS and shows a shoulder in the TDDFT spectra, Figure C.11b. To investigate the strength of M-S bond in five-coordinated vs six-coordinated complexes as well as in the Ru vs Os complexes, DFT calculated S3p-character involved in those orbitals are obtained as in Table C.2. DFT results suggest that the Ru-Sσ bond in 11a is more covalent than in 10a; however, the overall Ru-S bond strength in 10a is more covalent than in 11a due to π- and σ- interactions. This trend is also supported by the Frontier MO diagram (Figure C.12) showing a down shift in energy of Ru4dσ* orbital when it goes from 10a to 11a. Similarly, energy drop is also observed with OsII complexes between 10b and 11b but in smaller extent compare to the RuII analogues. Therefore, this implies that the M-S bonds in six-coordinated complex are less covalent than the M-S bonds in five-coordinated complexes. In between the five-coordinated RuII and OsII complexes (10a vs 10b), the OsII complex is less covalent compare to the RuII complex. However, the covalency of the M-S bond is almost equal in RuII and OsII six-coordinated complexes (11a and 11b). Systematic peak fitting should be performed to compare the DFT results with XAS.   Figure C.12: Frontier MO diagram of complexes 9-11. Only α orbitals are shown (α and β are identical). Red         arrow indicates the energy shift from 10a→11a and 10b → 11b. Orbitals are normalized to LUMO         in each complex E n e rg y (e V )

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0073949/manifest

Comment

Related Items