Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

X-ray absorption spectroscopy as a tool for characterizing sulfur based reactive intermediates Martin-Diaconescu, Vlad 2009

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

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

Item Metadata


24-ubc_2009_fall_martin-diaconescu_vlad.pdf [ 3.35MB ]
JSON: 24-1.0060650.json
JSON-LD: 24-1.0060650-ld.json
RDF/XML (Pretty): 24-1.0060650-rdf.xml
RDF/JSON: 24-1.0060650-rdf.json
Turtle: 24-1.0060650-turtle.txt
N-Triples: 24-1.0060650-rdf-ntriples.txt
Original Record: 24-1.0060650-source.json
Full Text

Full Text

X-RAY ABSORPTION SPECTROSCOPY AS A TOOL FOR CHARACTERIZING SULFUR BASED REACTIVE INTERMEDIATES by VLAD MARTIN-DIACONESCU B.Sc., The University of British Columbia, 2003 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) August 2009 © Viad Martin-Diaconescu, 2009 Abstract Sulfur K-edge X-ray absorption spectroscopy (XAS) has proven to be a great tool for the investigation of sulfur oxidation states and sulfur-metal ligand bonding. In this thesis, XAS has been applied in the detection and characterization of sulfur-based reactive intermediates and products of photo-reacted sulfur species, with applications in both bioinorganic and inorganic chemistry. Low molecular weight thiols and their derivatives have important protein modulation, signal transduction and antioxidant activities. This includes glutathione (GSH), nitrosoglutathione (GSNO), and lipoic acid (LA), which are involved in complex redox pathways resulting in a variety of intermediates and products that can be difficult to characterize. These compounds have been used as models for thiol nitrosation and oxidation reactions, and their reactivity was probed with sulfur K-edge XAS, which has been developed into a valuable tool for the investigation of sulfur-containing radical species and related non-radical intermediates. XAS was also applied to investigate the reactivity of p-toluene sulfonyl chloride, an initiator in metal catalyzed living radical polymerization, to explore the effect of hyperconjugation on the reactivity of the S-Cl bond. A series of model compounds of the form RSO2G (C = -Cl, -OH, -alkyl) were used to evaluate the effect of aryl versus alkyl R groups on the photo-reactivity and orbital mixing of the S-G bond. II Table of Contents Abstract ii Table of Contents iii List of Tables v List of Figures vi List of Schemes ix List of Equations x List of Abbreviations xi Acknowledgements xii Chapter 1: X-ray Absorption Spectroscopy (XAS) I 1.1 The Synchrotron Light Source (X-ray Source) I 1.2 Sulfur K-edge X-ray Absorption Spectroscopy 4 1.3 Applications of XAS 7 1.4 Thesis Overview 9 Chapter 2: Implications of Thiol Oxidation 10 2.1 Low Molecular Weight Sulfur Species In Vivo 10 2.2 Mechanism and Control of Thiol Oxidation 13 2.3 Experimental Rationale 18 Chapter 3: Experimental Section 21 3.1 X-ray Absorption Spectroscopy (XAS) 21 3.2 Density Functional Calculations 24 3.3 Nuclear Magnetic Resonance Spectroscopy (NMR) 25 3.4 Electron Paramagnetic Resonance Spectroscopy (EPR) 25 3.5 Materials 26 Chapter 4: Sulfur K-edge XAS as a Probe of Sulfur Centered Radical Intermediates 27 4.1 Background 27 4.2 Results and Discussion 28 4.3 Conclusion 37 Chapter 5: Perthiyl Radical and Disulfide Bond Formation in Photo-irradiated Nitrosoglutathione and Lipoic Acid 38 5.1 Background 38 5.2 Results and Discussion 43 III 5.3 Conclusion 54 Chapter 6: Effects of Hyperconjugation on the Electronic Structure and Photo-reactivity of Organic Sulfonyl Chlorides 55 6.1 Background 55 6.2 Results 58 6.3 Discussion 71 6.4 Conclusion 74 Chapter 7: Concluding Remarks and Outlook 77 References 80 Appendix I ADF Input Files 88 iv List of Tables Table 2.1.1 Biologically important low molecular weight sulfur species 11 Table 4.2.1 Principal component analysis of UV irradiated GSH XAS spectra 31 Table 4.2.2 TD-DFT calculated sulfur K-edge bound transitions for thiyl and perthiyl radicals 36 Table 5.2.1 Half power saturation values for the EPR of LA 45 Table 6.2.1 TD-DFT and IiSCF of p-toluene sulfonyl chloride 66 Table 6.2.2 TD-DFT and ASCF of methane sulfonyl chloride 66 Table 6.2.3 TD-DFT and ASCF of p-toluene sulfonic acid 67 Table 6.2.4 TD-DFT and iSCF of methane sulfonate 67 Table 6.2.5 TD-DFT and IISCF of ethyl phenyl sulfone 68 Table 6.2.6 TD-DFT and IiSCF of methionine sulfone 68 V List of Figures Figure 1.1.1 Schematic diagram of the main components foUnd in a synchrotron 2 Figure 1.1.2 Schematic diagram of the main components found in a beamline 4 Figure 1.2.1 XAS model spectra of possible S1 transitions 5 Figure 1.3.1 XAS fingerprint spectra of varying sulfur oxidation states 7 Figure 2.1.1 Thiol oxidation and cell life cycle 10 Figure 2.2.1 Thiol oxidation pathways 15 Figure 2.2.2 Reduction pathways of oxidized sulfur species 17 Figure 3.1.1 XAS experimental setup at SSRL beamline 6-2 23 Figure 4.2.1 X-band EPR of UV irradiated glutathione 29 Figure 4.2.2 Power saturation study of UV irradiated glutathione 30 Figure 4.2.3 Sulfur K-edge XAS of UV irradiated GSH 32 Figure 4.2.4 Difference XAS spectra of UV irradiated GSH 33 Figure 4.2.5 Time evolution of features at 2468.8eV and 2470.5eV in UV irradiated GSH 34 Figure 4.2.6 DFT density contour maps of thiyl and perthiyl radicals 35 Figure 4.2.7 Peak fitting of XAS features due to thiyl and perthiyl radicals 37 vi Figure 5.1.1 GSNO sulfur K-edge XAS spectrum 40 Figure 5.1.2 DFT density contour maps ofCH32SNO 40 Figure 5.1.3 LA sulfur K-edge XAS spectrum 42 Figure 5.2.1 LA sulfur K-edge XAS spectra under photo-irradiation 43 Figure 5.2.2 EPR spectra of irradiated LA 45 Figure 5.2.3 EPR power saturation spectra of irradiated LA 46 Figure 5.2.4 GSNO sulfur K-edge XAS spectra under photo-irradiation 48 Figure 5.2.5 Time evolution of features at 2470.2 and 2471.6 in irradiated GSNO 49 Figure 5.2.6 EPR spectra of irradiated GSNO 51 Figure 5.2.7 NMR of irradiated GSNO and GSSG 52 Figure 5.2.8 EPR spectra of irradiated GSSG 53 Figure 5.2.9 EPR power saturation study of irradiated GSSG 54 Figure 6.2.1 Structure of sulfonyl model compounds 59 Figure 6.2.2 Geometries of ethyl phenyl sulfone used for TD-DFT 60 Figure 6.2.3 XAS and simulated spectra of sulfonyl chlorides 61 Figure 6.2.4 XAS and simulated spectra of sulfonates 62 vii Figure 6.2.5 XAS and simulated spectra of sulfones 63 Figure 6.2.6 Molecular orbitals of benzene 65 Figure 6.2.7 Time evolution of the sulfur K-edge features due to the S-Cl bond with in-situ irradiation 69 Figure 6.2.8 XAS with in situ irradiation of sulfonyl chlorides 70 Figure 6.3.1 Mixing of ffL and cT into SCl* 72 Figure 6.3.2 Effect of turning off4n*/SCla* mixing 73 Figure 6.4.1 Fitted XAS of p-toluene sulfonyl chloride 75 Figure 7.1.1 XAS and complementary techniques 77 VIII List of Schemes Scheme 5.1.1 GSNO photo-reactivity induced reaction pathways 39 Scheme 5.1.2 LA photo-reactivity induced products 41 Scheme 6.1.1 Metal catalyzed polymerization with p-toluene sulfonyl chloride as an initiator 56 Scheme 6.1.2 Sulfonyl chloride photo-cleavage reaction 58 ix List of Equations Equation 1.1.1 Bragg’s law 3 Equation 1.2.1 Heisenberg uncertainty principle 6 Equation 6.4.1 Percent excited state hyperconjugation in p-toluene sulfonyl chloride 75 x List of Abbreviations A Angstrom °C Degrees Celsius CysS Cysteine CysSSCys Cystine DFT Density Functional Theory DHLA Dihydrolipoic Acid DNA Deoxyribonucleic Acid EPR Electron Paramagnetic Resonance Spectroscopy eV Electron Volt f Oscillator Strength GPx Glutathione Peroxidase CR Glutathione Reductase Grx Glutaredoxin GSH Glutathione GSNO Nitrosoglutathione GSSG Glutathione Disulfide hr Hours K Kelvin LA Lipoic Acid m Multiplet MIN Minutes NMR Nuclear Magnetic Resonance Spectroscopy P112 Power of Half Saturation (of EPR signal) Prx Peroxyredoxin RF Radio Frequency RNR Ribonucleotide Reductase RS Thiyl Radical RSNO Nitrosothiol RSS Perthiyl Radical ASCF Slater Transition State Self-consistent Field Method SSRL Stanford Synchrotron Radiation Lightsource t Triplet TD-DFT Time Dependent Density Functional Theory Trx Thioredoxin TrxR Thioredoxin Reductase UV-VIS UV-Visible Spectroscopy XANES X-ray Absorption Near Edge Structure XAS X-ray Absorption Spectroscopy Zeff Effective Nuclear Charge xi Acknowledgements First and foremost I would like to thank Dr. Pierre Kennepohl, my supervisor and mentor, whose encouragement, patience and support, have not only made this work possible but have made my time as a graduate student a wonderful experience. His door has always been open and his help was always gladly given no matter the circumstance. Secondly, I would like to thank my lab mates, in particular Anusha Karunakaran with whom I have spent countless hours acquiring data, Mario Delgado with whom I have argued endlessly about everything under the sun and Kendra Getty who still holds the record for the number of straight hours of beamline data acquisition. Needless to say that I consider the fore mentioned people my friends and they will be sorely missed. I would also like to thank Dr. Serena DeBeer George, who has always been ready to help at any time and who was of great help during our data acquisition trips at the Stanford Synchrotron Radiation Lightsource (SSRL). From experimental setup and running the data acquisition, to sample preparation, Dr. S. George was always there when needed. On a related note, I would also like to thank Dr. Matthiew Latimer for his help with experimental setup for our photo-irradiation studies. Last but not least I would like to thank the UBC Chemistry department support staff, in particular Maria Ezhova for her help with NMR data acquisition, Jane Cua for her help with running and maintaining the ADF program, John Ellis for his help with acquisition orders, Ken Love and Razvan Neagu for their help over the years with equipment trouble shooting, and Pat Olsthoorn for his help with sample shipments to SSRL. xii I X-ray Absorption Spectroscopy (XAS) This thesis relies heavily on the application of X-ray absorption spectroscopy (XAS) at the sulfur K-edge to investigate and characterize the bonding, electronic configuration, and reactivity of a series of biologically relevant low molecular weight thiols and their derivatives, as well as sulfur species with applications in inorganic polymer chemistry. Photochemistry was employed to follow the reactivity of these species, and investigate their reactive intermediates and products. XAS proved to be a useful technique in the detection and characterization of the elusive thiyl radical in UV irradiated glutathione. The perthiyl radical intermediate was detected in irradiated samples of lipoic acid and glutathione, as well as from irradiated nitrosoglutathione where formation of the perthiyl radical was a two step process. To perform the irradiation experiments, new methodologies of in situ photo-irradiation coupled with XAS spectra acquisition allowed the photo-reactivity to be observed. Last but not least, the relevance of excited state hyperconjugation was identified and its effect on the reactivity of p-toluene sulfonyl chloride explained as it pertains to the initiation step in the metal catalyzed polymerization of olefins. In each case, XAS was essential in the detection and characterization of these products. 1.1 THE SYNCHROTRON LIGHT SOURCE (X-RAY SOURCE) X-ray absorption spectroscopy was the primary method used in these investigations. The intensities and the wide energy ranges used for the X-rays in the following experiments require a synchrotron source. A synchrotron light source consists of a circuit of accelerated electrons that give off electromagnetic radiation in the form of X-rays. All electrons when accelerated give off energy, and this has many applications in everyday life, such as radio signal transmission. In a synchrotron, however, electrons are accelerated to speeds approaching the speed of light (99.9% speed of light). These relativistic electrons give off electromagnetic radiation in a parallel path to the direction of propagation of the relativistic electron. A synchrotron light source usually consists of a booster ring where electrons are accelerated to near the speed of light, from which they are injected into the storage ring (figure 1.1.1). Magnetic fields from “bend magnets” change the direction of the I electrons in the storage ring allowing them to maintain a closed path. The velocity of electrons is maintained by constructively interfering radiofrequency (RF) waves which takes place in the cavity. The RF waves are generated by a klystron (RF amplifier) similar to those used in TV broadcasting stations. The RF waves are guided to the cavity by a wave guide and timed to “synchronize” with the electrons entering the cavity. At various positions in the straight sections on the storage ring there are wigglers and undulators. These are a series of magnets with alternating polarities that cause the electrons to have small deviations in their linear trajectory, in a sense causing them to wiggle and undulate. Since the emitted X-rays have no charge they are not affected by the magnets and their trajectory is linear. X-rays emitted during a “wiggle” have a path that comes out of the storage ring and it can be delivered to various work stations (hutches) using beamlines. On any one storage ring there can be several wigglers and undulators allowing for multiple work stations or hutches. EMITTED X-RAYS BEND MAGNET ELECTRONS WAVE GUIDE CAVITY _- KLYSTRON (RF SOURCE) / t STORAGE RING WIGGLER Figure 1.1.1 Schematic diagram of the main components found in a synchrotron. 2 Beamlines screen the broad range of radiation energies (from radio frequencies to y-rays to X-rays) coming from the wiggler and select those needed for the particular experiment. They also shape the beam to suit the sample size and help control flux. This is achieved with the help of beamline optics as shown in figure 1.1.2. As the X-ray beam leaves the wiggler it may pass through a series of slits that shape it to have the needed dimensions. X-rays with the required experimental energies are then selected using a double crystal monochromator. X-rays of various energies are selected based on their wavelengths using Bragg diffraction. Bragg diffraction occurs when the incoming X-rays are scattered by the periodic structure of the crystal lattice according to equation 1.1.1. n2=2dsin6 n = harmonic (integer > 1) Equation 1.1.1 2 = wavelength d = distance between lattice layers 0 = angle of diffraction Therefore the crystal monochromators can be used to modulate the energies being diffracted to the experimental hutch by varying the angle of incidence of the X rays and by changing the type of crystal used (lattice structure). If only one crystal was used scanning over a range of X-ray energies would mean moving the sample to match the angle of diffraction of the incoming beam. In practice, two crystal monochromators are used and moved in concert so that the diffracted beam is parallel to the incident beam. It is also evident from Bragg’s equation that a series of harmonics or X-ray wavelengths can satisfy the equation for a particular angle of diffraction and crystal lattice structure. Therefore beamlines can be equipped with harmonic rejection mirrors which can efficiently refract X-rays of a certain energy range, but not those of its harmonics. These mirrors are coated with elements of higher atomic number because the refraction of X-rays of a particular energy is dependent on the material density of the coating. Mirrors can also be used to focus both the incident X-ray beam and the diffracted beam used for experiments. Once X-rays of the correct energy are isolated they are allowed to pass into the hutch. This is where the experimental setup for the various experiments is located (see Chapter 3 Methodology). 3 MONOCH ROMATOR I — — — — I I Figure 1.1.2 Schematic diagram of the main components found in a beamline. I .2 SULFUR K-EDGE X-RAY ABSORPTION SPECTROSCOPY X-RAYS TO HUTCH Sulfur K-edge (S K-edge) XAS involves detection and characterization of sulfur core electron excitations, specifically those from the Is core level (Si). Features due to these excitations, as well as electron ionizations from the Sis core, fall in the 2400- 2600eV energy range. The sulfur K-edge spectra is dominated by an ionization edge or “edge jump” due to excitations from the core Is orbital to the continuum (cc—Sis), resulting in ionization (figure 1.2.1 52.) Features due to Sis excitations into unoccupied (figure 1.2.1 RS.) or partially unoccupied (figure 1.2.1 RS) acceptor orbitals with sulfur character appear as absorption peaks. They fall within what is termed the X-ray Absorption Near Edge Structure (XANES) region of the spectrum. The intensities of these transitions are governed by Fermi’s Golden Rule which states that the intensity of a transition is proportional to the probability of it occurring1. The probability of a transition occurring is dependent on the coupling between the ground state and the final state, therefore transitions resulting from atomic core excitations can only involve excitations to other orbitals in the same atom. Furthermore, dipole allowed (s4—p; p—d) transitions have a higher probability of occurrence. This means that excitations of Is sulfur core electrons into acceptor orbitals with a higher S3p character will be more SLITS I MIRROR WIGG LER I I. — — — — .1 4 3p RS Is SCo 3p SCc * * 3p Figure 1.2.1 XAS model spectra showing the possible transitions and resulting XAS features from S2 (only ionization is possible), thiolate (ionization and bound transitions to unoccupied molecular orbitals is possible) and the thiyl radical (ionization and bound transitions to both unoccupied and half occupied orbitals possible). FINAL SPECTRUM FOR RS’ ENERGY ENERGY RS** Is SCa 3p SCa* ENERGY S2- Is 5 intense than transitions to acceptor orbitals having lower S3 character. In addition, the probability of a transition is also dependent on the number of ways it can occur. At its simplest this means that a transition to a half occupied acceptor orbital should be half as intense as a similar transition to a fully unoccupied acceptor orbital. These acceptor orbitals are usually the antibonding counterparts of each of the chemical bonds sulfur is involved in, within the species being investigated. The intensities of these bound-state transitions can provide information regarding the amount of sulfur character in a particular bonding scheme, as well as the electronic configuration. Equation 1.2.1 2 The spectral linewidth of the transitions is dependent on the resolution of the monochromator as well as the core-hole lifetime resulting from the particular transition. The core-hole lifetime is dependent on the energy of the transition according to the Heisenberg uncertainty principle (Equation 1.2.1). Therefore high energy transitions result in short core-hole lifetimes and broad features or linewidths, while low energy transitions such as those found at the sulfur K-edge result in long core-hole lifetimes and relatively sharp features. The line shapes for each of the transitions are a combination of a gaussian component as a result of the resolution of the monochromator and a lorentzian component due to the core-hole lifetime and can be described by a pseudo-voigt function which is a linear combination of the two components. Another important feature of XAS transitions is the energy at which they occur. This is dependent on both the energy level of the core electrons, as well as that of the acceptor orbital. The energy level of the core electrons can be a measure of effective charge on the sulfur atom, while the energy of the acceptor orbital can give information about bond strengths. For example, in the study of perthiyl radicals (RSS) there are two transitions from each of the sulfurs to the same SS acceptor orbital. If the two sulfurs were equivalent only one feature should be present because the two transitions would be overlapping. However, two transitions are visible corresponding to the difference in effective nuclear charge (Zeff) between the sulfurs which lowers the core electron energies in one of the sulfurs versus the other. 6 1.3 APPLICATIONS OF XAS R2SO 4 RSO3Na: II I I : 1RSONa’ : RSH : 1c 2 : \ : 2 / !\\ ! : I I ‘•.. , :I \i! \‘ .. . -‘r , — — — _. _. — 1 Ii ., ,.../ .-.‘ .. -..-... 2470 2472 2474 2476 2478 2480 2482 2484 2486 2488 2490 ENERGY (eV) Figure 1.3.1 XAS fmgerprint spectra of sulfur species with varying oxidation states. The nature of the transitions observed in sulfur K-edge XAS make this technique sensitive to the various oxidation states of sulfur (figure 1.3.1). In fact, much of the literature involving sulfur K-edge XAS use the “fingerprint” method to investigate sulfur speciation. In this approach, the composition of the sulfur K-edge spectrum from a particular sample is determined by fitting and summing up spectra from similar samples containing 100% of a particular sulfur oxidation state2’. This method can be applied to both non-biological and biological systems. For example, sulfur speciation as it pertains to the effectiveness of different coal desulfurization methods was investigated using least square fitting of the XANES region of spectra from various coal samples that were desulfurized using either biological, chemical or caustic leaching methods4. Sulfur K-edge XAS was also used to assess the impact of land use by characterizing sulfur oxidation in soil samples from natural forest, tea plantations and cultivated fields from various locations in Ethiopia. Natural forest were found to have the most reduced sulfur species, followed by plantations with cultivated fields being most oxidized5. Most notably, oxidation of reduced sulfur species in the seventeenth-century Swedish warship, Vasa, was found to be catalyzed by iron 5 Na2SO4 Cl) z w I z w N -J 0 z GLUTATHIONE — — METHIONINE SULFOXIDE SODIUM METHANE SULFINATE — ---SODIUM METHANE SULFONATE SODIUM SULFATE 7 from corroded iron bolts, leading to the formation of sulfate salts which threaten the preservation of the ship6. Researchers have also collected the spectra for a variety of biologically relevant sulfur species. The distinct features differentiating between thiol groups and disulfides were used in a number of studies investigating the thiol redox couple responsible for redox homeostasis, in blood, and separated plasma and erythrocyte samples2’ 3 Extracellular cysteine was found to be mostly in its oxidized cystine form while intracellular cysteine was found to be more reduced2’. Increased oxidation state of sulfurs in transthyretin investigated with sulfur K-edge XAS is seen in amyloid fibrils, the main component of amyloid deposits present in such disorders as Alzheimer’s disease and Creutzfeldt-Jakob disease7. Features due to S-H bonds (SHa**_Sis), S-C bonds (SC0*€_Si), and S-S bonds (SS*E-_Sls), in samples were identified by comparing the XANES features from H2S, H2S and reduced and oxidized cysteine2’8 Investigation of the dependence of the cysteine spectra on pH shows a dramatic change in the spectral features with deprotonation2’. Another advantage of this technique is its ability to tolerate a broad range of sample types and preparation techniques. For example, the studies discussed thus far have included solid samples in the form of coal, soil, and wood; liquid samples in the form of blood, plasma and buffered cysteine, and H2S and H2S gas samples. It is not surprising then, that samples consisting of various sulfur accumulating bacteria were prepared and quantitative sulfur speciation studies were carried out. At least three different forms of sulfur were found in bacterial sulfur globules with cyclooctasulfur being prevalent in Beggiatoa alba, polythionates in Acidithiobacillus ferrooxidans and sulfur chains in green and purple sulfur bacteria10. XAS becomes an even more powerful experimental technique when coupled with simulations of the electron excitation transitions along with calculations of the orbitals of the donor and acceptor states involved in the transitions. This was first recognized and explored in chlorine K-edge XAS spectroscopy” and then extended to sulfur K-edge XAS where the metal-ligand covalency was measured in transition metal tetrathiolate complexes’2. XAS coupled with density functional calculations (DFT) were used in the characterization of the Cu-S bond in the active sites of blue copper proteins which was found to have a high covalency (38% S3 character) and a unique single it-bond between the copper and the sulfur13. The methodology was then applied to the 8 investigation of oxidation, reduction, and bonding of iron sulfur clusters in ferredoxins and rubredoxins’3. XAS and DFT were further used to assign the spectral features for S-nitroso proteins where transitions to SC, SNOa* and S-NO acceptor orbitals were identified14. The dependence of XAS spectra of disulfides on the dihedral angle was investigated by comparing the experimental and simulated spectra of oxidized lipoic acid to those of less conformationally strained disulfides. The results showed distinct differences in both the intensity and peak width of the SS.+-S1 and SC0*+-Si transitions, emphasizing the importance of accounting for molecular conformations when analyzing spectra using the fingerprint method15. 1.4 THESIS OVERVIEW The above discussion is meant to show the range of applicability of this experimental technique. It has been shown that sulfur K-edge XAS is sensitive to sulfur oxidation regardless of sample preparation. Sulfur speciation and electronic structure can be gained from the XANES region of the spectra encompassing bound transitions to low-lying molecular orbitals with S character. A combination of the fingerprint method and density functional calculations can be used in the assignment of spectral features and characterization of the sulfur bonding manifold. The current study investigated in situ photo-reactivity of both biological and non-biological samples. In each case DFT calculations were applied to better assign the spectral features and understand the electronic configuration. Additional spectroscopic techniques were used in the characterization of starting materials, intermediates and products. Particularly electron paramagnetic resonance (EPR) spectroscopy was essential in the detection and characterization of radical intermediates generated during the photo-reactivity studies. The following chapters cover a discussion of the experimental setups used throughout the thesis, a brief overview of the importance of low molecular weight sulfur based antioxidants in redox homeostasis, followed by the studies involving the main representatives of these species chosen for their relevance, reactivity and interesting electronic structure. Last but not least, is the investigation of the bonding and electronic configuration of p-toluene sulfonyl chloride as it pertains to its role as an initiator in the metal catalyzed polymerization of styrene, methacrylates, and acrylates. 9 2 Implications of Thiol Oxidation Low molecular weight sulfur species are involved in a variety of processes that maintain the oxidative balance in vivo. These compounds (table 2.1.1) are known to have antioxidant properties which come into play particularly in times of oxidative stress, when increases in oxidizing species disrupt regular cell function. Their interaction with reactive oxygen (ROS) and nitrogen (RNS) species not only have antioxidative consequences but are also important in signaling pathways. Low molecular weight thiols and their derivatives could be described both as “middle men” and, due to their antioxidant functions, as modulators of these signaling pathways. Their antioxidant properties also make low molecular weight su’fur compounds potential therapeutic agents in various diseases with symptoms of oxidative stress. Of these the most important and abundant is glutathione (GSH), which behaves as a redox buffer within the cell16. Outside the cell the most important thiol for maintaining redox homeostasis is cysteine (CysS)16. 2.1 LOW MOLECULAR WEIGHT SULFUR SPECIES IN VIVO PROLIFERATION DIFFERENTIATION APOPTOSIS INTRACELLULAR (GSH) -2601-230 mV -220/-I 90 mV -1 7OmV [RSH] > LRSSR] [RSH) < [RSSR] EXTRACELLULAR (CysS) <-80 mV -80 mV ‘-80 mV Figure 2.1.1 Intracellular and extracellular disulfide/thiol redox state at various stages in the lifecycle of the cell as described by the GSSG/2GSH couple (intracellular) and CysSSCysI2CysS couple (extracellular)16’17 Thiol oxidation has been linked to several cellular processes with implications ranging from cell signaling to progression of disease. Both the glutathione (GSSG/2GSH) and cysteine (CysSSCys/2CysS) redox states vary over the life cycle of the cell (figure 2.1.1). One can see a progression to a more oxidized state going from proliferation to apoptosis’6’17 The extent of oxidation has been explored in several systems by induction of differentiation in proliferating cells. Application of the differentiating agent sodium butyrate to a HT29 cell line caused a +6OmV shift (from -26OmV in the proliferating cells to -200mV in the differentiating cells)18. Exposure of slime mold to differentiation stimuli resulted in a decrease in GSH 10 Table 2.1.1 Biologically important low molecular weight sulfur species. LOW MOLECULAR WEIGHT SULFUR SPECIES Cysteine (CysS) 0 Precursor to GSH and the major extracellular thiol redox buffer16’19 HS OH Methionine 0 In addition to being used for cysteine synthesis, methionine oxidation/reduction H3C OH is involved in metal ion channel gating and NH2 neurodegenerative diseases20. Glutathione (GSH) The most abundant low molecular weight thiol, has antioxidant properties, modulates protein function through S-glutathionylation and maintains redox homeostasis2’. Nitrosoglutathione (GSNO) Used in the storage, transport and delivery of N0 a molecule important in cell signaling22. Dihydrolipoic acid (DHLA) Lipoic acid (LA) is the oxidized form, has a redox couple of LA/DHLA (-32OmV) and can directly reduce GSH23. Can chelate metal ions and prevent lipid peroxidation24. It is an antioxidant in both hydrophilic and lipophilic environments25. 11 and an increase in the antioxidant enzyme manganese superoxide dismutase (MnSOD), indicating a shift to a more oxidized environment during differentiation26. Variation of thiol oxidation states also occurs within the cell. Compartmentalization allows a more reduced environment within the nucleus where the GSH pool is implicated in protection from oxidative stress of DNA27 and DNA binding motifs28, regulation of gene transcription29,DNA synthesis30,and DNA repair31’2 In the endoplasmic reticulum on the other hand, the thiol redox couple favors the more oxidized state. Here the redox state of the GSSGI2GSH couple is approximately -18OmV33, as opposed to -26OmV in the nucleus16. This facilitates disulfide bond formation and isomerization of incorrectly formed disulfide bonds, a process modulated in part by protein disulfide isomerase (P01), a thiol disulfide oxidoreductase34’ The disulfide/thiol redox state can also be affected by external factors such as disease. A prime example is the oxidative pressures associated with HIV (Human lmmunodeficiency Virus). In HIV there is a systemic decrease in reduced GSH339 and GSH synthesis4°along with a decrease in total cysteine and cystine amounts ([CysS] + 2[CysSSCys])38. Chronic increase in thioredoxin (Trx, a family of proteins involved in the reduction of disulfides) concentrations can further compromise the innate immune response by inhibition of Iipopolysaccharide induced chemotaxis, resulting in shorter life expectancy of AIDS victims41. Addition of NAC (N-acetylcysteine) increases survival16 and in vitro addition of GSH and NAC inhibits viral replication42,indicating the potential for therapeutic applications of low molecular weight thiols. Last but not least, thiol oxidation can also affect signaling pathways. Protein tyrosine phosphatase (PTP) has a catalytically active cysteine residue which can be inhibited if oxidized to a sulfenic acid by H2043. Increased thiol oxidation and GSSGIGSH ratios also seem to stimulate the activity of several kinases such as the mitogen activated protein kinases (MAPK), JNK (c-Jun N-terminal kinase) and p38, required for tumor necrosis factor (TNF-c) induced apoptosis45. Nuclear transcription factors can also be affected by GSH oxidation. For example, Nil 2 (NE-F2 related factor) which regulates expression of several genes involved in antioxidant response46’47, is translocated to the nucleus when its Keap-1 subunit dissociates from the main complex due to oxidation and conjugation of its cysteine residues48. However, within the nucleus, DNA binding activity is controlled by Trx49, which maintains Nil 2’s 12 DNA binding cysteine residue in its reduced form50, highlighting the importance of compartmentalization, as discussed earlier. 2.2 MECHANISM AND CONTROL OF THIOL OXIDATION Reactive Oxygen (ROS) and Nitrogen (RNS) Species and Thiol Oxidation Thiol oxidation can be driven by the reactive oxidant pathways of ROS and RNS coupled with disulfide reduction systems such as those of the thioredoxins (figure 2.2.1). Comprehensive reviews on the formation and consequences of reactive oxygen species (ROS) and reactive nitrogen species (RNS) are available5053. These reactive species can arise endogenously from cellular processes or be mediated by external factors such as chemotherapeutic agents, UV irradiation, and other environmental stimuIi5053. At low concentrations, ROS and RNS can be beneficial and many are involved in signaling pathways50,gene transcription29 and host defenses against infection. Examples are leukocyte release of ROS to attack infecting bacteria, NO’ induction of smooth muscle relaxation55’56, and N0 inhibition of platelet aggregation57. Thiols and thiol redox pathways play important roles both as antioxidants to maintain ROS and RNS at homeostatic levels and as intermediary mediators in some of the ROS/RNS signaling pathways. In eukaryotes reactive oxygen species arise mainly in mitochondria from aerobic respiration58’59 or enzymatic reactions of NADP(H) oxidases in leukocytes54. The product of such reactions is the superoxide anion (Oj) which is not membrane permeable and thus reacts within whichever cellular compartment it is created60. Oj is converted to hydrogen peroxide (H20) by superoxide dismutase (SOD)61. H20 is membrane permeable and can act as a signaling molecule50’62 In the presence of redox-active metal ions, H20 is rapidly converted to the hydroxyl radical (HO’) via the Fenton reaction6365. HO is highly reactive and unselective towards a host of biological molecules. H2O can also oxidize thiols, and in high enough concentrations, it can lead to the generally irreversible sulfur oxidation states of RSO2H and RS03H66. Three main enzyme systems are employed in H20 removal: catalases (CAT)67, glutathione peroxidase (GPx)68 and peroxyredoxins (Prx)69 will degrade H20 to yield H20 and 02. The latter two, GPx and Prx involve intermediate oxidation of thiols. GPx reduction of H20 results in GSH dimerization to GSSG68. Prx is initially oxidized to sulfenic acid 13 (RSOH), which in the case of 1-CysS Prx can be reduced by vitamin C70. In 2-CysS Prx (Prx containing two cysteines in its active site) the initial oxidation to RSOH is followed by internal disulfide bond formation which can be later reduced by thioredoxins7’. Prx can also be further oxidized by H20 to give sulfinic acid (RSO2H), which can be reduced back to the active form by sulfiredoxin7173. N0 production from arginine is facilitated by nitric oxide synthase and similar to 02, it is the first product in the RNS chain74. N0 can further react with 02 to give N203, a nitrosylating agent, and with 02 to give peroxynitrite (ONOO)52. The reaction of N0 with thiols is generally considered too slow to have biological relevance, however N203 can nitrosylate a variety of substrates including thiols to give S-nitrosothiols75.0N00 contributes to oxidative stress and is scavenged by thiols resulting in disulfide bonds76. Denitrosation is facilitated by dihydrolipoic acid (DHLA) and Trx, which undergo oxidation resulting in release of a free thiol and HN077. Similar to the denitrosation mechanism of DHLA and Trx, transnitrosation reactions, followed by S-thiolation and disulfide bond formation modulate protein activity as is the case of creatine kinase78. Release of N0 can also be achieved by decomposition of two nitrosothiols or nitrosothiol reduction of metals such as copper. Both cases result in formation of a disulfide79’8° Disulfide formation via reduction of metals such as vanadium, copper and iron is also possible and reported to proceed through a thiyl radical intermediate65’81-83 Thiols such as glutathione and dihydrolipoic acid can ligate metal ions and are involved in both metal detoxification84 and delivery85. Reduced metal ions such as arsenic (Ill), chromium (IV) and chromium (V), copper (I), and iron (II) can induce oxidative stress, and their ligation may help in their excretion which would prevent them from reacting to induce oxidative stress63 . Metal reduction by GSH and GSH derivatives is associated with lipid peroxidation, which stimulates oxidative stress rather than represses it. The efficiency of the reduction is modulated by the ligating abilities of the thiol species and their pKa’s87’8 Lipid peroxidation due to Cu2 is inhibited by DHLA which is believed to chelate the metal via its vicinal thiols. The stability of the complex however, is pH dependent and the complex becomes destabilized with increasing pH24. 14 RSH OH o= NOS N0 Arginine O OXIDASESI — SOD H2 M1TOCHONDRIA 2 2 2 Figure 2.2.1 Thiol oxidation pathways, in vivo. RSH NO 0N00 RSH RSNO HNOINO 15 Mechanisms of Disulfide Reduction From the discussion of ROS and RNS one can see that thiols tend to become oxidized to disulfides when exposed to oxidative pressures. Therefore, a mechanism for the reduction of disulfides to thiols must exist to rejuvenate the reduced thiol pool. The principal pathways for such processes are the thioredoxin and glutaredoxin enzyme systems (figure 2.2.2). Thioredoxins (Trx) are a family of thiol-disulfide oxidoreductases found within both prokaryotes and eukaryotes89. The active form found in the cytosol and nuclei of cells is thioredoxin-1 (Trx 1)°, while mitochondna has the thioredoxin-2 (Trx 2) isoform which contains an N-terminal mitochondrial translocation sequence91. A third form of thioredoxin is found in spermatozoa, called sperm-specific Trx (Sptrx)92. All isoforms contain a conserved active site sequence: Cys-Gly-Pro-Cys93. They interact with a variety of disulfide containing species, most notably ribonucleotide reductase (RNR) involved in DNA repair and synthesis31,transcription factors, PDI in the endoplasmic reticulum, and peroxyredoxins53’. The Trx mechanism of action involves nucleophilic attack of the disulfide species to be reduced by the Trx N-terminal cysteine, forming a mixed disulfide intermediate. This in turn is reduced by the C-terminal cysteine of Trx95. The end products are a reduced dithiol species and an oxidized Trx now containing a disulfide in its active site95. Trx is ultimately reduced by thioredoxin reductase (TrxR) coupled with oxidation of NADPH to NADP and formation of a Sec-SCys bond in TrxR96. TrxR is a homodimer selenocysteine (Sec) enzyme97. The Sec site is located near the C-terminal of the enzyme (Gly-Cys-Sec-Gly-COOH)98. The motif is conserved across species and isoforms of TrxR99. Enzymatic activity of oxidized TrxR is restored by donation of electrons from NADPH via a bound FAD to a thiol/disulfide site (Cys-Val-Asn-Val-Gly Cys) present in one subunit of TrxR reducing it to a dithiol. This site is similar in structure and reactivity to that of glutathione reductase (GR) and is in the proximity of the Sec residue on the second subunit of TrxR. This allows the thiolldisulfide site to donate electrons to the C-terminal selenocysteine-cysteine bond (Sec-SCys), reducing it and restoring TrxR reduction capabilities100’101 16 RSOH ROS RSO2H ROS RSO3H RSH RSSR / RSNO RNW\ RSH ROS GRx/TRx NADP —NADPH Figure 2.2.2 Reduction pathways of oxidized sulfur species, in vivo. 17 Glutaredoxins (Grx) are another essential family of thiol/disulfide oxidoreductases, which only differ slightly from thioredoxins in their mechanism of action. The major isoform, glutaredoxin 1 (Grx I), is found in the cytosol and has an active site sequence Cys-Pro-Tyr-Cys’°2 Mitochondria have two glutaredoxins, glutaredoxin 2 (Grx 2) with the active site sequence Cys-Ser-Tyr-Cys’°3 and glutaredoxin 5 (Grx 5) with only one cysteine in its active site implicated in iron homeostasis’°4. Grx 1 is associated with dithiolldisulfide exchanges93, cell differentiation105 and apoptosis’ 06, dehydroascorbate reduction107, transcription factor regulation108,and as an electron source for ribonucleotide reductase3”‘. Grx can also catalyze the formation and reduction of mixed disulfides between proteins and GSH109, which has implications in protein regulation and cellular responses to oxidative stress. Similar to thioredoxins, Grx reduces disulfide bonds by oxidizing its own active site cysteines to a cystine. However, the Grx active site is reduced by GSH to give GSSG. First one GSH forms a mixed disulfide with Grx, followed by formation of GSSG when a second GSH interacts with the complex”°. The resulting GSSG can be reduced to GSH at the expense of NADPH by glutathione reductase (GR)1’1’3. Similarly, GSSG resulting from the Gpx catalyzed reduction of H20 is reduced to GSH by glutathione reductase. CR is a homodimer’’4with binding sites for NADPH and GSH, which are opposite to each other but on the same subunit”5. The electrons are transferred from the NADPH binding site to the active site of the enzyme via an FAD prosthetic group next to the redox active cystine residues1’3115, 116 The active site has the Cys-VaI-Asn-Val-Gly-Cys sequence, which is conserved in humans and yeast, and is similar to lipoamide dehydrogenase and TrxR”3’117 Upon electron donation from NADPH the active site becomes reduced and catalytically active. 2.3 EXPERIMENTAL RATIONALE Evidently, thiol oxidation is central to several biological processes, ranging from redox homeostasis to DNA transcription and modulation of enzyme activity. The most encountered oxidation states are the reduced —SH thiol of cysteines essential for transcription factor binding to DNA, and its counterpart the disulfide bond with implications in modulation of protein function as outlined by its effect on translocation of transcription factors to the nucleus. Enzymatic activity can also be affected by thiol 18 oxidation, as is the case of tyrosine phosphatase, where formation of sulfenic acid inhibits its function. Therefore characterization of the oxidation states of sulfur centers is important in the exploration of enzyme mechanisms as well as metabolic pathways. Unfortunately, there are no conventional methods that can analyze sulfur oxidation. The most common way of detecting thiols is the use of Ellman’s reagent (DTNB, 5,5’-dithiobis-2-nitrobenzoic acid)118. DTNB is a disulfide that exchanges with free thiols in solution to give a yellow chromophore118 Other chemical derivatization methods for detecting thiols involve tagging thiols with maleimides, iodoacetaimides, iodoacetates or thiosulfates, all of which have to be linked either to fluorophores, radionucleotides, affinity labels such as biotin or labels that will change the overall molecular weight119. This is generally followed by gel electrophoresis and comparison between runs with tagged thiols and runs with untagged thiols to look for band profile changes. Similarly disulfide bonds are investigated via electrophoresis under oxidizing and reducing conditions followed by comparison of band shifts9. Analysis of other oxidation states is often further complicated by the short lifetime of the species. Sulfenic acids, are generally not very stable but have been detected in vitro by selective reduction with sodium arsenite or selectively reacting them with dimedone or NDB-Cl (7- Chloro-4-nitrobenzo-2-oxa-1 ,3-diazole) followed by biotinylatio&’9’120 A similar method involves detection of S-nitrosylated thiols by selective reduction with sodium ascorbate followed by biotinylation121. Both methods have the drawback of requiring denaturing conditions and alkylation of free thiols prior to treatment with the selective reducing agents119 Therefore, sulfur K-edge XAS is proposed as an alternative and direct method to look at thiol speciation, which may prove to be a powerful tool in elucidating the catalytic mechanisms of enzyme systems. It has the advantage of being sensitive to all sulfur oxidations states. Disulfide bonds and free thiols along with methionine, sulfenic, sulfinic and sulfonic acids are readily detected by XAS3’122 Furthermore, spectral fitting of a mixture of thiols with varying oxidation states allows for quantitative analysis of the sample2’. Most recently XAS was proven to be a useful technique for the detection of S-nitrosylated proteins14. This suggests XAS could be applied to detect intermediates in the catalytic cycle of enzymes which are difficult to see by other means. A prime example is ribonucleotide reductase (RNR). There are three classes of this enzyme and in each case the reduction of the ribose moiety of the nucleotide substrate is 19 believed to have a thiyl radical intermediate123 124 Thiyl radicals are notoriously difficult to characterize due to their large spin orbit coupling (=382cm1)25 and broad g11 component 126, 127 Sulfur K-edge XAS, on the other hand, should be very sensitive to the dipole allowedS3—S1transition from the thiyl radical. XAS can also prove to be a useful source of information on particular bonding manifolds around the sulfur atom. It was previously proposed based on theoretical calculations coupled with X-ray structural data that the sulfonyl moiety (SO2) in RaSO2b type compounds experiences hyperconjugation interactions from the Ra nonbonding orbitals into the S-Rb * antibonding orbital128. A similar effect was also inferred in the case of aryl sulfonyls, which experience a bathochromic shift in the UV-VIS spectra of their benzyl group when the sulfonyl moiety is present129. Because sulfur K-edge XAS consists of transitions to unoccupied or partially unoccupied molecular orbitals it could be a direct probe for the detection of hyperconjugative effects which result from molecular orbital mixing. Thus, this thesis involved development of in situ experimental methods to investigate the reactivity, intermediates and products of biologically relevant low molecular weight thiols. Furthermore, the electronic configuration around sulfonyl groups was probed to investigate the effects of hyperconjugation on the reactivity of these compounds. The findings reported here, and the experimental methods devised, provide steps toward expanding XAS identification of thiol species in more complex systems, such as enzyme catalysis and reaction pathways. 20 3 Experimental Section To study the reactivity of sulfur model compounds samples were photo-irradiated with a Xenon arc lamp and X-ray absorption data was collected. The experimental setup for in situ photo-reactivity of sulfur containing model compounds is described in the subsection below. Time-dependent density functional theory was employed to better understand and assign the transitions giving rise to the various features of the XAS spectra. Electron paramagnetic resonance spectroscopy (EPR) was used to help identify the sulfur based radical intermediates formed during photolysis while nuclear magnetic resonance (NMR) was used to characterize the starting materials and final products. The materials used in the following experiments are described in section 3.5. 3.1 X-RAY ABSORPTION SPECTROSCOPY (XAS) Experimental Setup The XAS data was collected at beam line 6-2 of the Stanford Synchrotron Radiation Lightsource (SSRL). The facility is a 3GeV ring with a current of 60-1 OOmA. Beam line 6-2 operates in a high magnetic field mode of 10kG and consists of a 54-pole wiggler followed by a Ni coated harmonic rejection mirror, and a fully tuned Si(1 II) double crystal monochromator. The beam-line optics are under vacuum and protected from the pressurized experimental hutch by a beryllium window followed by a 6.3511m polypropylene window and a He chamber13. The spot size of the incident beam is controlled by a pair of horizontal and vertical JJ X-ray exit slits followed by a He gas ionization chamber, which measures the intensity of the incident beam (lo). The sample chamber is isolated from the ionization chamber by another polypropylene window. The sample itself is 45° to the incident beam and 45° to a fluorescence ion chamber Stern Head-Lytle detector130, which was filled with argon gas and maintained at ambient temperatures. He flow from a liquid helium cooler provided by Cryo Industries (HFC 1645 LHE-Cryocool) maintains the sample below -20°C where it can be photo-reacted with a Ushio 75W Xenon arc lamp positioned 50cm from the sample. At these temperatures the samples did not react in the X-ray beam. The sample chamber encasing consists of a specially adapted transparent glove-bag providing an anaerobic He atmosphere for data collection (figure 3.1.1). By varying the point to point 21 increments, number of data points and collection time at each point, the scan rate was varied in order to better follow the rate of photo-reactivity131. The spectrometer resolution was -0.5eV’32 Solid samples were mounted as a finely ground powder dusted on the adhesive side of sulfur-free Kapton tape (polyimide film with silicone adhesive) to minimize fluorescence self absorption effects13. Methyl sulfonyl chloride was mounted as a neat solution on the sulfur-free Kapton tape and covered with a polypropylene window under a nitrogen atmosphere. Spectra were acquired while the samples were irradiated with a 75W Xenon arc lamp at 253K (-20°C) under a helium atmosphere with <1% oxygen content. Glutathione mounted on the Kapton tape was irradiated with UV light at ambient temperature under aerobic conditions. The UV irradiation source was a LonglifeTM Filter 254nm shortwave ultraviolet lamp from Spectroline®. XAS spectra processing Energies of the Sulfur K-edge XAS spectra were calibrated using hydrated sodium thiosulfate (Na2SO3.5H0)with the first pre-edge feature peak maximum being calibrated at 2472.02eV13. The value of 2472.02eV for the first pre-edge feature was arrived at by repeated experiments on different beamlines and checked against the inflection point of sulfur at 2471.3eV and the X-ray absorption spectra of molybdenum and copper foil132. An alternative value for the first pre-edge feature of sodium thiosulfate is 2469.2eV determined by calibrating the spectra of sodium thiosulfate to the first intense feature in the spectra of NiS powder assigned as occurring at 2469.8eV’33. Spectra were background subtracted using the method of energy summation of two linearly weighted background terms derived from the background before and the background after the edge jump of the XAS spectra. The magnitude of the edge jump was normalized to a total intensity of 1. Principal component analysis of transient spectra was carried out using SixPack version 0.53135. 22 — . I Thermocouple Cl) SampleI He Fi lle d G lo ve B ag T ab le PW PW = po ly pr op yl en e w in do w Io ni za tio n C ha m be r L PW — — I D et ec to r Sl its B e W in do w X e A rc La m p F’ .) c) 3.2 DENSITY FUNCTIONAL CALCULATIONS Molecular orbital calculations to simulate spectrometric parameters and deduce molecular bonding interactions were carried out for the various species investigated. Density functional theory calculations were carried out using the Kohn-Sham Self- consistent field methodology provided with the Amsterdam Density Functional Software Package 2007.01136 137 Full geometry optimizations were followed by single-point and time-dependent density functional theory (TD-DFT) calculations. Calculations were performed using the BP86 functional and a doubly polarized triple-c basis sets (TZ2P) using Slater-type orbitals (STO) basis functions. TZ2P was necessary to allow for the interaction between the sulfur valence (3s, 3p and 3d) orbitals resulting in correct geometries for the hypervalent sulfur atom. Previous researchers have shown this combination of functional and basis sets to give the best approximation for calculated spectra and relative experimental energies and intensities in ligand K-edge XAS1. To calculate the corresponding sulfur core Is excitations for each of the compounds, the ModifyExcitations key was used, along with the no core option for any of the atoms and no symmetry operations were applied. The resulting excitation energies for the sulfur Is electrons were shifted by 76-81eV to account for deviations in the calculated core excited-state energies from experiment. These energy shifts fall within range of those calculated by other researchers1. In the case of the aryl and alkyl sulfonyl compounds fragment calculations were necessary to better understand contributions to molecular orbitals from the different “fragments” such as the aryl and sulfonyl moieties. Fragment calculations consist of modeling the orbital interactions for each fragment individually and then combining them to give the final picture. To assist in the spectroscopic assignment of XAS features and to account for relaxation effects due to electron excitation,’3814° the transition energies were recalculated using the Slater transition state self-consistent field method (ASCF) for the first 10 transitions of each model compound of interest. XAS spectra were then simulated using the recalculated energies and the oscillation strengths calculated with TD-DFT. iXSCF was done by having 1.5 electrons in the S1 orbital and one-half of an 24 electron in the orbital of interest which gives a good estimate for the relaxation energy in most core ionization/excitation cases’41. The magnitude of the calculated oscillatory strengths were used for the peak areas needed to simulate the features of the various XAS spectra using a pseudo-voigt function with 80% lorentziari and 20% gaussian character. The intensities of the oscillatory strength were normalized to fit the normalized experimental spectra. Details of further molecular calculations and deviations from the above procedures are mentioned in the text wherever applicable. Sample input files are found in Appendix 1. 3.3 NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY (NMR) NMR data was collected on a Bruker AV 300 NMR spectrometer. ‘H NMR data was collected with a I 5ppm sweep width centered at 6ppm, each spectrum consisting of 128 scans with a time domain of 32K. Samples were dissolved in deuterium oxide (020) solvent with a 99.9% deuterium atom composition, having a proton peak maximum calibrated at 4.8oppm. 3.4 ELECTRON PARAMAGNETIC RESONANCE SPECTROSCOPY (EPR) X-band EPR data collection at 9.5GHz was carried out using a Bruker Elexsys E 500 series continuous wave EPR spectrometer running the Xepr software package. Photo-reactivity studies were carried out using a 75W Xenon arc lamp positioned 50cm from the sample. In situ irradiation experiments were carried out while collecting spectra at room temperature or, using a finger dewar filled with liquid nitrogen, at temperatures of 77K. Studies at 195K (-78 °C) were carried out by irradiating the sample while in an acetone-dry ice bath followed by data acquisition at 77K. In situ experiments were not viable in acetone-dry ice bath because of “lossyness” due to microwave absorption by acetone. Samples were run either under vacuum or under an argon gas atmosphere. DPPH (g=2.0036) was used as the standard for spectra calibration’42. Power saturation profiles were acquired by varying microwave powers between 63mW and 0.002mW. Further details and deviations from the above procedures are mentioned in the text when needed. Simulation of spectra was carried out using the Bruker WinEPR SimFonia software package. 25 3.5 MATERIALS Sulfur-free Kapton tape was purchased from Creative Global Services Inc. and checked for sulfur contamination (by sulfur K-edge XAS) before use. The irradiation sources consisted of a LonglifeTM Filter 254nm shortwave ultraviolet lamp from Spectroline® for UV irradiation and a Ushio 75W Xenon arc lamp for full spectrum irradiation. 707-SQ-250 EPR tubes and a large finger dewar flask l5Oml (Suprasil WG 853-B-Q) from Wilmad Labglass were used for EPR experiments. Polycrystalline powders of reduced L-glutathione, (±)-x-lipoic acid, p-toluene sulfonyl chloride, p toluene sutfonic acid, phenyl ethyl sulfone, methionine sulfone, and sodium methane sulfonate along with a neat solution of methane sulfonyl chloride were purchased from Sigma-Aldrich and stored at 4°C until use. S-nitrosoglutathione was synthesized by reacting glutathione with sodium nitrite under acidic conditions and characterized using NMR (chapter 5 figure 5.2.7 p.52) and by its absorption at 545nm using UV-V1S143. Glutathione, (±)-a-Iipoic acid and p-toluene sulfonyl chloride were also characterized using NMR and no contaminants were found. S-nitroso glutathione (GSNO) synthesis To a mixture of Sml of deionized water mixed with 5m1 IM HCI, 5mmol (1.5g) of glutathione (GSH) were added and dissolved. The solution was kept in an ice bath wrapped in aluminum foil to prevent product photo-degradation. To the colorless GSH solution, 5mmol (O.35g) of sodium nitrite was added. The reaction immediately turned red and was stirred for I hour. Precipitation of the GSNO product was initiated with I OmI of -20°C acetone followed by a further 30 minute of stirring. The product was then filtered by suction filtration and washed successively with 5 x O.5m1 of ice-cold water, 5 x I ml acetone and 6m1 of diethyl ether added drop-wise. The pale pink product was dried and stored in the dark at -20°C until use. 26 4 Sulfur K-edge XAS as a Probe of Sulfur-Centered Radical Intermediates 4.1 BACKGROUND Sulfur based radical intermediates are involved in various biological processes1”,most notably the enzymatic reduction of ribose sugars in ribonucleotide reductase (RNR)’45’146 Thiyl radicals are postulated to be essential in the catalytic mechanism of all classes of ribonucleotide reductases124 however their detection and characterization have been elusive. Paramagnetic species are generally detected using electron paramagnetic resonance spectroscopy (EPR), however this is somewhat complicated for sulfur because of its relatively large spin orbit coupling (=382cm)125 and broad g11 component of thiyl (RS) radicals126’127 Recently Lassman and coworkers have generated thiyl radicals using UV irradiation in polycrystalline glasses of both protein and cysteine samples. Integrated EPR spectra of UV irradiated bovine serum albumin (BSA), and the RI subunit of RNR were compared to that of UV irradiated 300mM cysteine samples146’147 The UV irradiated cysteine samples exhibit a weak signal at a g, of 2.30 attributed to the thiyl radical. When integrated this gives “a broad ascending slope” in the absorption spectrum leading to the main absorption features at lower g values146. Similar “slopes” were also detected in the protein samples and attributed to the thiyl radical even though the g, signal was absent in the 1st derivative spectra. Such analysis requires careful background subtraction and the researchers chose the 330mT point in the field domain as the cutoff for the thiyl “absorption feature” so that other features present would not superimpose with the thiyl radical gi signal. Although this approach has proven useful in the detection of free thiyl radicals, proper characterization of the thiyl radical using this method can be hindered by many factors such as incorrect baseline subtraction, imperfections in the EPR tubes, as well as the presence of other features which may overlap with the absorption signal of the integrated EPR spectrum of the thiyl radical. Other EPR approaches to the detection of thiyl radicals involve spin trapping agents. Unfortunately, spin trapping techniques cannot provide information about the g-tensor of the radical, hyperfine couplings, or radical lifetimes147’148 Therefore XAS is proposed as a new tool for detecting thiyl radical intermediates which has the added advantage of also detecting EPR silent sulfur intermediates such 27 as sulfenic, sulfinic, and sulfonic acids that generally rely on secondary methods of detection such as chemical methods involving thiol derivatization119.These species can often form as byproducts of the radical generating reactions and their identification and characterization is important in the overall reaction mechanism. These studies used glutathione as a model compound which has a redox couple (2GSH4->GSSG+2e) that is essential for proper redox balance and homeostasis. It has been shown that photochemical one-electron oxidation of GSH forms a multitude of radical products including the short-lived thiyl (GS) and long-lived perthiyl (GSS) radicals, yet the one- electron redox chemistry of GSH and cysteine is still somewhat unclear. Therefore, polycrystalline glutathione is used as a model system for evaluating the use of sulfur K- edge XAS in the study of sulfur-containing radicals. 4.2 RESULTS AND DISCUSSION Characterization of Irradiated Samples with EPR Experimental conditions were as described in the experimental section. Polycrystalline GSH was irradiated in the UV region (2=254nm) at room temperature (295±5K) giving an EPR spectrum that has been previously investigated and attributed to a complex mixture of radical species (i.e., GS, G, H, and GSS)146. XAS data was collected for samples irradiated from 2 to 48 hours. EPR of the UV-irradiated samples remained unchanged even after prolonged exposure to the X-ray beam (>3 hours) and were not affected by the Kapton tape. From the spectra in figure 4.2.1 one can see that the features of the “GSH” spectra and those of “GSH” are comparable. “GSH” was a sample prepared and used for the XAS experiment. It consisted of GSH on Kapton tape irradiated for 48 hours and its EPR spectrum was acquired at room temperature. “GSH” is glutathione polycrystalline powder irradiated for 4 hours and placed in an EPR tube with its EPR spectrum acquired at room temperature. Both samples exhibit features at g values of 2.030 and 2.057 that are due to the perthiyl radical, while the feature at 2.009 is in the thiyl radical g± region (g 2.01) as previously reported by Lassman et a!. and Neese et a! 127, 146 In addition, the features of the simulated perthiyl EPR spectrum (figure 4.2.1) correspond well to those of irradiated glutathione. The perthiyl radical EPR spectrum was simulated with g values at 2.057, 2.03 and 2.07 which match well those derived 28 from the density functional calculations of the CH3SS model for the perthiyl radical used to analyze the XAS spectra. The calculated g values for the CH3SS perthiyl radical are 2.058, 2.027 and 2.002. The difference in the lower g values at 2.002 and 2.07 between the actual and calculated spectra may arise because of the presence of additional EPR signals such as those from a possible thiyl radical. I I : : GSH I I I I I I 2.10 2.05 2.oO 1.95 Figure 4.2.1: X-band EPR spectra of {JV irradiated glutathione. “GSH” is GSH on Kapton sulfur free tape irradiated for 48 hours and used for XAS with EPR collected at room temperature (power 2mW, 3scans, modulation amplitude 5G, modulation frequency 100kHz); and “GSH” is GSH powder irradiated overnight and placed in an EPR tube and run at room temperature (power 2mW, 3 scans, modulation amplitude 5G, modulation frequency 100kHz). The simulated spectrum of perthiyl radical has g values of 2.062, 2.030 and 2.007. _____________ 09 ____________________ SPECTRUM F PERTHIYL RADICAL iGSH g VALUE 29 The presence of the perthiyl radical was further confirmed with power saturation studies (figure 4.2.2). The feature at g=2.057 was chosen for these studies because it does not overlap with other spectral features. It is found that the signal at 2.057 increases linearly with higher microwave powers in the range of 0.002mW to 20mW, which is consistent with the low saturation effects seen in perthiyl radicals146’149 The g11 of the thiyl radical was not observed in the EPR spectra of either the GSHs or the “GSH” samples but was identified in the XAS spectra. Characterization of Irradiated Samples with )(AS The XAS data of UV irradiated GSH, shows distinct changes in the GSH spectrum upon irradiation (figure 4.2.3 and 4.2.4). The GSH spectrum is initially dominated by an intense CS0—1at -2473.5eV With UV-irradiation, new pre-edge features appear at 2468.8eV and 2470.5eV and the CS—S18feature of GSH broadens and decreases in intensity. The reactivity of GSH can also be followed by the 1.0 C’) CD 1.0 Cl) z LU I— z -J z CD Cl) 0.5 LU N -J 0 z 0.0 2 SQUARE ROOT OF MICROWAVE POWER (mW1) Figure 4.2.2: Power saturation study of the EPR signals from “GSH’” and “GSH” samples. 0) I Cl) CD >- F— C,) z LU I— z -J z CD Cl) LU N -J 0 z 0.5 0 1 0.0 30 disappearance of the features at 2476.7eV and 2479.3eV. Second derivative analysis of the spectra shows the presence of an additional feature at 2471.5eV as indicated by an inflection point in this region (figure 4.2.4 B). These pre-edge features coincide with the appearance of radical species in the EPR spectra and therefore should correspond to transitions to the half-occupied sulfur orbitals of sulfur-centered free radicals. At longer irradiation times, a high energy feature at 2482.6eV appears indicative of a highly oxidized sulfur species. This species has not been characterized, however formation of stable sulfoxyl radical intermediates along with sulfinic and sulfonic species have been reported as a result of photo-irradiated sulfur containing amino acids1. Principal component analysis (table 4.2.1) on the transient spectra shows that 3 components are needed to obtain >0.99 cumulative variance for the low energy pre-edge region of the spectra (2465 - 2475eV) whereas 4 components are needed when the higher energy components are included (2465 - 2490eV). This indicates the formation of 3 species as a result of irradiation. Table 4.2.1. Principal component analysis of UV irradiated GSH XAS spectra from the various irradiation time points. Dotted lines show the number of components needed for >99% cumulative variance over the energy ranges described. COMPONENT ANALYSIS FOR 2465eV — 2475eV RANGE Component Eigenvalue Cumulative Variance 1 25.62 0.907 2 1.75 0.969 3 0.63 0.991 4 0.12 0.995 5 0.04 0.997 6 0.03 0.998 7 0.02 0.999 COMPONENT ANALYSIS FOR 2465eV — 2490eV RANGE 1 46.42 0.922 2 2.24 0.966 3 0.89 0.984 4 0.33 0.990 5 0.25 0.995 6 0.14 0.998 7 0.05 0.999 31 2.5 FIGURE 4.2.3 Sulfur K-edge XAS of UV irradiated GSH after various time intervals of irradiation. 2.0 0 —-—-—2 8 ----24 48 Schr hr hr hr hr I I 1’ t 2475 ENERGY (eV) 2480 2485 32 0.2 Figure 4.2.4 Difference XAS spectra of UV irradiated GSH at various times and GSH control (A) and the second derivative of the difference spectra showing an inflection point at 2471.3eV (B). 2469 2470 2471 ENERGY (eV) 0.1 0.0 2468 1.0 0.5 0.0 -0.5 2472 2468 2469 2470 2471 ENERGY (eV) 2472 33 • • • • • 3.0 Lii — Ui. - a- -.2.0 Lii C) 1.5 - -.- - (2470.5) GSS • (24688)GS 1.0 0 05 ________________ 0 • ••_.•__ 0.0. • I • I • I • I 0 10 20 30 40 50 TIME HOURS Figure 4.2.5 Time evolution of the features 2468.8eV and at 2470.5eV in terms of calculated percentage of thiyl and perthiyl radical after various intervals of irradiation. Based on differences in their rates of formation (figure 4.2.5) the two low-energy pre-edge features must correspond to different sulfur radical species, most likely GS (thiyl) and GSS (perthiyl), respectively. To better understand the electronic structure and characteristics of these transitions a combination of unrestricted ground state and time-dependent density functional theory (TD-DFT) calculations were carried out on simplified models of the expected thiyl (CH3S) and perthiyl (CH3SS) radical species. Electron density maps were generated and the important empty valance orbitals are shown in figure 4.2.6 with relevant transitions listed in table 4.2.2. Spectra at the sulfur K-edge are dominated by transitions with electric dipole allowed S*—S1 character; therefore empty orbitals with S3, character are of most importance. DFT results show that the unpaired electron in GS resides in a singly-occupied S orbital perpendicular to the S-C bond axis127. Two other orbitals, labeled CS. and HCa*, have significant S3p character and account for the two major bound-state transitions at 2473.7eV and 2474.5eV in the Sulfur K-edge spectrum of GS. The HCa* final state obtains much of its sulfur K-edge intensity through intensity borrowing from the CS0 state at -1eV higher 34 energy. The lowest energy transition at 24688eV corresponds to a nearly pure (-92%) S3,,+—S1transition attributed to the singly occupied sulfur 3p orbital. The unpaired electron in GSS is located in a SS orbital (figure 4.2.6). Two transitions should be observed to this orbital, one from each of the S1 orbitals. TD-DFT calculations indicate a significant splitting of the two SS transitions of 1.3eV suggesting a more positive effective charge for SA, which lowers its Is core orbital energy. The a* transitions for GSS include both CS and SSa* contributions that occur at about the same energy as in the GSH spectrum. It is important to note that the low-energy pre edge feature for GSS should occur at higher energy than that of GS. This is because the acceptor orbital in the perthiyl species is a higher energy antibonding orbital. In addition, a splitting of the pre-edge feature in GSS is expected from the DFT results. cHs css CSc, 850*” Figure 4.2.6 DFT calculated electron density contour maps of important valence orbitals for thiyl and perthiyl radicals (Isovalue = O.075e.A3). CSc,* s3p sst 35 Table 4.2.2. TD-DFT calculated sulfur K-edge pre-edge features of relevant model compounds. The resulting excitation energies for the sulfur is electrons were shifted by +76.5eV in all cases, using the main a’ feature of GSH as a calibration point. Species Assignment Energy (eV) f CS0*+HScr*4_Sis 2472.8 2.Ox1O CH3S CS,,*—S1 2473.7 i.ixiO3 CSc*+HSa*(Sis 2474.4 1.3x10 S—S1 2468.8 3.9x10 CH3S HCa4Ss 2473.7 8.1x10 CSa*Sjs 2474.5 9.3x10 SS,_SBi 2469.3 2.5x104 SS*,_SA1 2470.6 1.8x1O • SS&±SBs 2472.7 3.1x10CH3SS 2473.4 5.3x10 SS0*4Ai 2474.1 2.6x103 CS*+SAi 2474.8 1.6x103 The DFT data is in good agreement with the experimental sulfur K-edge spectra. The lowest energy pre-edge feature in the UV irradiated GSH spectra was assigned to a S3,÷-S1 transition in GS while the feature at 2470.5eV was attributed to the SS transitions of GSS. The GSS pre-edge feature was fitted using PeakFit v4. 12 (figure 4.2.7) and the splitting caused by differences in Zeff of the two sulfur atoms, is similar to that predicted by DFT (—1.1eV vs. 1.3ev). The kinetic behavior of the sulfur K-edge pre-edge features are also consistent with previously published EPR data, showing initial formation of the thiyl radical followed by subsequent formation of the longer-lived perthiyl species146. Radical yields were estimated using the ratios of OFT predicted oscillator strengths (t) for pre-edge features of the methyl thiyl (CH3S) and methyl perthiyl (CH3SS) radicals with that of methane thiol (CH3SH). This allowed the extrapolation of the value for the areas of the radical species corresponding to a 100% thiyl or perthiyl XAS spectrum. The calculated areas corresponding to 100% of a species were then compared to the actual areas achieved at the different irradiation times to calculate the percent radical yields. Maximum yields for the thiyl and perthiyl radical were achieved after irradiating the sample for 4 hours (—0.45%) and 48 hours (—3.1%), respectively (figure 4.2.5). 36 CH3SS (SS1* Ais) Figure 4.2.7 Peak fitting of the difference spectra for the pre-edge region of the GSH sample irradiated for 48 hours and control (0 hours). The peak ratios for the two perthiyl radical pre edge features were kept within 20% of the DFT calculated result. Pseudo-voigt fI.inctions with 50% lorentzian and 50% gaussian character were used for each of the fitted peaks. 4.3 CONCLUSION This study shows the usefulness of XAS as a probe to detect sulfur based radical intermediates and was the first study to detect and characterize isolated free sulfur radicals using XAS. Within the same model system it is possible to differentiate between two sulfur radical intermediates GS and GSS’, which had pre-edge features well resolved from each other and the intense a* contributions. An EPR silent byproduct of the photo-chemical reaction described by the peak at 2482.6eV was also identified. This feature is indicative of a highly oxidized sulfur species possibly a sulfoxyl radical intermediate or a sulfonic acid. In the following sections this technique is applied to increasingly more complicated systems and reactivity profiles, which prove the usefulness of this technique when used in conjunction with more main stream spectroscopies such as EPR and NMR. • 48-0 hr Sum of fits Thiyl GS Perthiyl GSSB ----PerthiylGS’S CH3S (S3—S1) /CH3SS (SS÷- SBi) / / , 2468 2469 2470 2471 2472 ENERGY (eV) 37 5 Perthiyl Radical and Disulfide Bond Formation in Photo-irradiated Nitrosoglutathione and Lipoic Acid 5.1 BACKGROUND The initial study involving glutathione (GSH, chapter 4) showed that XAS can be very useful in the detection of sulfur based radicals, which have low energy pre-edge features well separated from the typical spectroscopic features of non-radical species. Nitrosoglutathione (GSNO), the S-nitrosylated version of glutathione, and lipoic acid (LA), the oxidized disulfide of dihydrolipoic acid (DHLA), are low molecular weight thiols which are also involved in antioxidant defense as previously described. In fact, DHLA is involved in the denitrosation of GSNO to yield LA and the free thiol GSH77. The reaction is believed to proceed via transfer of the N0 moiety to DHLA followed by the formation of an internal disulfide and release of N0. Sulfur K-edge studies with in situ photo irradiation show formation of new disulfide bonds in GSNO and formation of an additional pre-edge feature attributed to the perthiyl radical. The transition assigned to the perthiyl radical is also present in LA. EPR and NMR techniques were applied to better understand the reactivity of these species and complement the XAS experiment. Even though these systems have been previously investigated there is still some debate to their mechanism of action. Furthermore, their reactivity seems to be very dependent on the reaction conditions1501. If XAS is to be applied to more complicated reaction profiles such as those of enzymes, the reactivity of simpler systems and the conditions that govern them must first be investigated. S-nitmso cylutathione (GSNO) The main function of GSNO is as an N0 carrier and delivery system, N0 being an essential signaling molecule. The mechanism of N0 delivery is still under investigation and most recent studies suggest that at neutral pH N0 is released from nitrosoglutathione by a one electron reduction mechanism involving a possible GSNO - intermediate155. Still, other research emphasizes the importance of disulfide bond formation during N0 release as a result of denitrosation reactions, such as those involving DHLA and thioredoxin to give the corresponding lipoic acid and oxidized 38 thioredoxin products77.Nitrosothiol (RSNO) decomposition accompanied by release of N0 and formation of disulfides is also catalyzed by the presence of copper (Cu), application of heat and photo-irradiation156.Based on previous experiments with GSH, GSNO decomposition can be observed with sulfur K-edge XAS and in situ photo- irradiation of the sample. Initial formation of the GS intermediate should be followed by disulfide bond formation (GSSG) and be proceeded by the disappearance of features due to the GSNO starting material (scheme 5jj)152 GSNO ‘ > GS + N0 GS + GSNO > GSSG + N0 2GS >GSSG Scheme 5.1.1 GSNO photo-reactivity induced reaction pathways. The sulfur K-edge spectrum of GSNO was previously investigated and assigned by Szilagyi and coworkers; our data are consistent with this previous work (figure 5.1.1)14. The dominant features are the SN—S1 at 2471.7eV and the SNa*Si transitions at 2473.4eV due to the S-NO bond and the SC---S1feature at 2474.8eV. This assignment is reinforced further by TD-DFT simulation of the core S bound transitions of CH32SNO model system which indicates two SC*—S1 transitions account for the feature at 2472.8eV, while a single transition is observed for the SN0*+_Si feature. The SNG acceptor orbital for the SN**.Si at 2471.7eV shows a significant interaction between the p orbitals of all atoms making up the S-N-O bonding manifold (figure 5.1.2). This is consistent with an S-NO bonding model stabilized by resonance as previously described, with the most significant contribution being due to the SNa resonance structure followed by S-Nt and SNO ion pair157. Therefore, in a series of 5-nitrosothiols the intensity of the SN**Si and SN0*Si transitions could be used to determine the relative importance of the S-Na resonance structure, a project which is currently ongoing in the Kennepohl group. 39 2.5 1.0 Figure 5.1.1 GSNO sulfur K-edge XAS spectrum (solid line) with the main S1 core excitation transitions assigned’4 The edge jump of the spectra was normalized to 1. TD-DFT simulated XAS spectrum of GSNO usingCH32SNO as a model system (dotted line). sca* SN0 SN Figure 5.1.2 DFT calculated electron density contour maps of important empty valence orbitals forCH32SNO (Isovalue = O.05e.A3) 2.0 1.5 SN scIy* SN. 0.5 2472 ENERGY (eV) 40 Lipoic acid (LA) Lipoic acid is an important cofactor and antioxidant. The LA/DHLA redox couple is distinguished from that of other low molecular weight thiols by its ability to function as an antioxidant in both hydrophilic and lipophilic environments and even reduce other antioxidants such as glutathione23’25 In addition, LA can be further oxidized to give a disulfide radical cation1. The photochemical reactivity of LA was previously investigated. Depending on the conditions the presence of thiyl radicals, perthiyl radicals, disulfide radical cations, and triplet states were detected, making LA a versatile system to observe by XAS. Photolysis of disulfides in solution results in the formation of perthiyl radicals149. This reaction is more favourable with increased substitution of the resulting carbon radical which further stabilizes it150’ 151 The presence of oxygen during photolysis seems to facilitate thiyl radical formation150. Cage effects also impact photolysis. The hexacyclic disulfide trans-4,5-d ihyd roxy-1 ,2-dith iacyclohexane when irradiated (wavelengths between 305-4lOnm) near its absorbance maximum at 280nm (the absorbance spans the 250-325nm range) forms a thiyl radical pair which is proposed to undergo H shifts to form more stable products, including the formation of the starting material153. Laser flash photolysis of LA at 266nm in aqueous solution results in the formation of the disulfide radical cation, while flash photolysis at 355nm yields a triplet state with a lifetime of 75ns1 4. Therefore, irradiation of LA under anaerobic condition with a Xe arc lamp would be expected to give a combination of the perthiyl radical, thiyl radical or disulfide radical cation intermediates (scheme 5.1.2). RSS RS hv RS coupled with H-shifts (giving RSH; R=S; RSSR) ‘ RSSR’ 3RSsR Scheme 5.1.2 LA photo-reactivity induced products. The transitions of the sulfur K-edge XAS spectrum (figure 5.1.3) of LA were previously investigated15. The first and most intense pre-edge feature corresponds to the 41 SS—S1at 2472.3eV followed by the SC—S1at 2473.9eV. TD-DFT simulated XAS spectra for the bound transitions of methyl disulfide (CHSSCH)is consistent with this assignment (figure 5.1.3). The feature labeled LA1 at 2475.4eV in figure 5.1.3 is attributed to a collapse in symmetry between the sulfurs of LA due to bonding to a primary carbon in one case and a secondary carbon in the other. This feature is not present in symmetric disulfide systems15. The steric strain in the pentacyclo LA disulfide is emphasized by a lowering in energy of both the SS0 and SC0* when compared to cystine where these features appear at -2472.6eV and -2474.2eV respectively15. This suggests that weaker bonds than generally found in disulfides result in LA due to steric strain. The XAS of LA also exhibits broader features which could be do to self-absorption effects from non-homogenous grinding of the solid sample158. However, the intensities of the features are consistent from run to run and comparable to those of other researchers15 indicating that self-absorption should be minimal. Furthermore the features of the spectra are well resolved and self-absorption should not pose a significant problem for determining the energy of the various transitions of the spectra. 1 1. ss* SC.. LA1 2465 2470 ENERGY (eV) Figure 5.1.3 LA sulfur K-edge XAS spectrum (solid line) with the main S1 core excitation transitions assigned’. The edge jump of the spectra was normalized to 1. TD-DFT simulated XAS spectrum of LA using CH3SSCH as a model system (dotted line). 42 Previous assignments of the features from the sulfur K-edge XAS of GSNO and LA give an idea of which features are most important to observe in the irradiation studies. Disappearance of the SN—S1 and SNa*-Sis transitions in GSNO would indicate breakage of the S-NO bond, while a decrease in intensity of the SS0*÷—S15 in LA would suggest bond homolysis. Previously determined reaction pathways helped to explain the reactivity observed. It was found that although LA and GSNO are “simple”, small sulfur containing compounds their reactivity can be quite complex, and under the experimental conditions, even related. 5.2 RESULTS AND DISCUSSION Photo-irradiation of LA c) 0.60 2.5 0.55 __—•—-—-——•-—----.-_ ______. ::: y 2.0 0.40 —.-—2470.O V : / I. 0 2 4 6 810121416 J TIME (MIN) I 0 I 10<I Al 0 Il 0.5 I . I . I . I . 0.0 2460 2462 2464 2466 2468 2470 2472 2474 2476 2478 2480 ENERGY(eV) Figure 5.2.1 LA sulfur K-edge XAS spectrum with in situ photo-irradiation. Dotted arrows are decreasing features, solid arrows are growing features. Subset: Formation of the feature at 2470.0eV attributed to the perthiyl radical. 43 Irradiation of LA with a Xe arc lamp under anaerobic conditions was followed with sulfur K-edge XAS. In this system the starting compound is a strained cyclic disulfide and as displayed in figure 5.2.1 a feature at 2470.0eV appears after only a few minutes of irradiation. For LA a shift is also seen coupled with a drop in intensity of the feature at 2472.3eV corresponding to the breaking of the cyclic disulfide. However, formation of a feature at 2472.5eV and an increase in intensity at 2473.9eV with a 1.5eV separation between the two is suggestive of new disulfide bonds forming, along with perhaps other minor products. The higher energies lead to the conclusion that the forming disulfides have stronger bonds resulting from a less strained conformation. The disappearance of the peak at 2475.3eV indicates that if new disulfides are being formed they have sulfurs with similar core excitation energies. Unfortunately, the product from the irradiation forms a white insoluble precipitate which does not lend itself to easy characterization by other spectroscopic techniques such as NMR, and confirmation and characterization of the final product formed from the irradiation is still unclear. To confirm the presence of the perthiyl radical, EPR spectra of anaerobically irradiated LA were collected and analyzed. The formation of the perthiyl radical was clearly visible (figure 5.2.2). A rhombic signal was observed with irradiation at 195K and 77K with g-values of 2.002, 2.026 and 2.062 matching well the simulated EPR spectrum of the perthiyl radical and the density functional theory calculated values for the CH3SS perthiyl radical (2.058, 2.027 and 2.002)146. Table 5.2.1 shows half power saturation values (P112) which represent the value at which the signal intensity divided by the square root of power drops by 50%. Analysis shows that the features between 2.026 and 2.002, as well as those below 2.000, saturate at low powers suggesting they are due to carbon centered radicals49. Perthiyl radicals however like most sulfur centered radicals saturate at higher powers and their intensity increases with microwave power over the ranges investigated, overwhelming the spectra from other species (figure 5.2.3)146 149• The feature due to the perthiyl radical at g=2.062 is well resolved from other peaks and has a P112 of 20mW, consistent with previously reported values146. The features due to the perthiyl at 2.026 and 2.002 overlap features with low saturation powers (2.026 to 2.002 and below 2.002), which might explain why they have lower P1,2 values (6mw vs. 20mW). P112 values of carbon radical species were found to lie in the 0.6mW range which is consistent with previously reported values (‘-1 mW)146. LA shows a similar reactivity profile when dissolved in D20 as in the solid state. 44 Table 5.2.1 Half power saturation values for EPR spectra of solid LA irradiated at 77K (as in figure 5.2.3). g VALUE RANGE P112 mW SATURATION 2.062 20 2.026 6 2.026—2.002 0.6 2.002 6 2.00-1.95 0.6 12.062 2.026 12.002 SIMULATED SPECTRUM OF PERTHIYL RAICALI I I I : I I I I I IRRADIATED 6OMIN AT 195K I I Ii I I I I I I I SOLID LIPOIC ACID 2.10 2.08 2.06 2.04 2.02 g VALUE 2.00 LIPOIC ACID IN D20 IRRADIATED 6OMIN AT 77K 1.98 1.96 1.94 Figure 5.2.2 EPR of irradiated (60mm) LA under different conditions (microwave power 0.6mW, modulation frequency 50MHz, modulation amplitude 3.OG, 5 scans) and simulation of the spectrum for the perthiyl radical with g values of 2.062, 2.026 and 2.002. 45 2.062 2.026 2.002 3 3050 3150 3150 3250 3250 3133 3550 3550 3650 3650 3700 • r - - 3000 3050 3100 3150 3200 3250 I 3300 I 335J 3450 3450 3500 3550 3600 3650 3700 __________________________ 0.0635mW I I I I I I I ‘I I IT I’’ I I I I I I I I I I I I 0 3050 3100 3150 3203 3250 I 3300 I 335 3400 3450 3500 3550 3600 3650 3700 0.201rii/( 2 3050 3100 3150 3200 3250 3 335( 3450 3450 3500 3550 3600 3650 3700 3 30150 3150 3450 3203 3250 3300 335I 3400 3450 8550 3555 3600 3650 3707 I I I I I’ 20C)rT,V’I 3 0 3100 3150 3 3200 I 3300 I 3450 3550 3550 3603 3650) 3703 —--—-1 6.33mW 30)) 36503100 31 I’ 20.0mW 3050 3150) 3150 3450 3450 3501) 3550 3600 0 3750) ZZr\..JEJE 3000 3050 3100 3150 3200 3250 3300 3350 3400 3450 3500 3550 3600 3650 3700 Gauss Figure 5.2.3 EPR power saturation study of irradiated (60 mm) LA solid at 77K. Microwave powers were increased between 0.002mW and 63.3mW. Perthiyl radical at g values of 2.062, 2.026 and 2.002 is found to increase with increasing microwave powers; (modulation frequency 50MHz, modulation amplitude 3G, lOscans). 46 Photo-irradiation of GSNO Over the duration of 1 hour, irradiation caused the features in GSNO due to the S-NO bond to disappear as highlighted by the dotted arrows in figure 5.2.4. Of particular importance is the peak at 2471.6eV, which is well resolved from other features and allows one to easily follow the breaking of the S-N bond. Under the conditions of the experiment some minor products indicative of oxidized sulfur species form in the 2476.5 — 2477.5eV region where transitions from R2SO and RSOj may occur and at 2480.8eV indicative of the formation of RS03 2 The transition at 2480.8eV due to RS03 is also present in the initial spectrum, showing that it is formed as a minor product during GSNO synthesis. Appearance of features at 2472.8eV and 2474.0eV during irradiation, suggest the formation of disulfide bonds and correspond to SS—S1 and SC*+-S1 excitations2’ 15 They are split by 1.2eV and the higher intensity of the first feature over the second further suggests the presence of disulfide bond formation. However, the SS—S1and SC—S1transitions in disulfide bonds are generally separated by 1.5eV. The poor resolution of the separation between the SS—S1 and SC—S1 transitions indicates the presence of other species whose features are overlapping those of the forming disulfides and could be due to minor products. A small shoulder forming at 2470.2eV is consistent with the formation of a perthiyl radical, and forms congruently with S-N bond cleavage reaching a steady state when about half the GSNO is reacted (figure 5.2.5). The rate of perthiyl radical formation and the fact that it reaches a steady state is consistent with previously observed behavior of perthiyl radicals generated in both GSH and LA experiments, It is not clear from the XAS data by what mechanism the perthiyl radical is formed, but the presence of features from disulfide bonds suggests a possible secondary process, such as disulfide bond formation followed by S-C bond homolysis resulting in a perthiyl radical. To further investigate this, EPR data was acquired to determine the time of perthiyl radical generation and NMR data was used to confirm disulfide bond formation. 47 Cl) z UI 1.5 UI N -J ln< Li:: 0 z Figure 5.2.4 GSNO sulfur K-edge XAS spectrum with in situ photo-irradiation. Dotted arrows are decreasing features, solid arrows are growing features. 2.5 (N LO 0co F— c.J Lq CD F (N 2.0 (N 0 F (N 0.5 ENERGY (eV) 48 0.15 >- F— Co z IJJ z. ci w N -J 0.05 z C’sJ 0 C” 0.00 1.6 0 z 1.2 w F— z U w N -J0.8 < 0 z CD 0.4 Figure 5.2.5 Time evolution of the intensities from sulfur K-edge XAS spectra of the feature at 2471.6, indicating disappearance of the S-NO bond, and the feature at 2470.2, indicating the appearance of the perthiyl radical, during in situ photo-irradiation of GSNO. —.— 2470.2 ---.---2471.6 .1 • I • I • I • I • I • I • I • I • I • I • I. -5 05 10 15 2025 30 35 40 45 50 55 60 TIME (MIN) 49 Irradiation of GSNO at 195K or 77K, both as a solid or dissolved in D20, yields a similar EPR signal, suggesting that S-NO bond photo-cleavage is not sensitive to sample preparation under the conditions of this experiment. The most striking characteristic of the EPR spectra of irradiated GSNO is the absence of features due to the perthiyl radical such as those that would be expected at a g—2.06 (figure 5.2.6). The inflection point at 2.012 may suggest the presence of thiyl radicals, but a clear identification is not possible because of overlapping features with similar power saturation profiles. However, formation of the GSSG disulfide from photo-irradiated GSNO under anaerobic conditions at 77K and 195K as well as GSNO irradiated at room temperature was detected by NMR and is in fact the major product. Low levels of GSSG are present in our GSNO samples due to sample decomposition as illustrated in figure 5.2.7. The NMR of GSNO has features at 2.l5ppm (2H, m), 2.47ppm (2H, t) and 3.85ppm (IH, t) due to the glutamyl 13, and a protons respectively. This is followed by a singlet at 3.98ppm due to the glycyl protons which sits on top of the signals at 4.l3ppm (1 H, m) and 3.99ppm (1 H, m) from the cysteine 13 protons, followed by the cysteine a proton at 4.68ppm (1 H, t)159. GSSG formation can be followed by observing the doublet of doublets at 3.Oppm and 3.3ppm due to the cysteine 13 protons as well as the triplet at 2.6ppm due to the glutamyl y protons (dotted square figure 5.2.7)160. After irradiation at 77K for 1 hour the GSNO sample whose EPR spectra is shown in figure 5.2.6 was brought to room temperature and the NMR was acquired. The initial control NMR shows a ratio of 10:3 GSNO to GSSG, derived from the relative peak integrations. After irradiation the GSSG features increase in intensity leading to a 2:1 ratio GSNO to GSSG. The sample was further irradiated at room temperature for 5 minutes to facilitate full conversion of GSNO to the disulfide GSSG. The newly formed GSSG was then used to determine if perthiyl radical generation would occur with further irradiation at 77K. 50 - - - CONTROL — IRRADIATED -Vç NITROSO GLUTATHIONE IN D20 IRRADIATED 6OMIN AT 77K --- SOLID NITROSO GUTATHIONE IRRADIATED 3OMIN AT 195K Figure 5.2.6 EPR of GSNO irradiated under different conditions. No clear detection of sulfur based radicals; (microwave power 0.6mW, modulation frequency 50MHz, modulation amplitude 3.0G, 5 scans). :2.060 :2.012 I I . I : - - - CONTROL : — IRRADIATED — — 4----. -A’I4.V-—. I : : : : --- CONTROL : : — IRRADIATED \_. —— I I HRADIATED6OM4AT77K 2.10 2.05 2.00 1.95 gVALUE 51 GSSG RADIATED JI F”-1., I 4.5 .0 I 3.5 3.0 2.0 1.5 I I ‘ GSNO IRRADIATED I I 300K 5 MIN (GSSG) I I I I I • I - I I I I I 4.5 4.0 3.5 3.0 L5 2.0 1.5 0.2 0.2 Figure 5.2.7 Comparison ofNMR spectra from samples of GSNO and GSSG in D20. Dotted rectangle encloses features due to the presence of GSSG. /2.01.5 1.0 GSNO IRRADIATED2.0 77K60M1N 0.9 4.5 2.0 0.6 2.8 52 I I I — I I I I I I I I a a I I I I I I I I a I I I I I a I I I a • LIPOIC ACID IN D20 IRRADIATED 60M(N AT 77K (2mW) 2.062 2.028 GSSG IN D20 IRRADIATED 60 MIN AT 77K (20mW) 2.10 2.05 2.00 1.95 g VALUE Figure 5.2.8 Comparison of EPR from irradiated GSSG and LA showing the presence of perthiyl radicals; (modulation frequency 50MHz, modulation amplitude 3.OG, 5 scans). Irradiation of the GSSG sample in D20 at 77K gives a rich EPR spectrum (figure 5.2.8). When the EPR spectra of irradiated GSSG in D20 is compared with that of LA in D20 the feature attributed to the perthiyl radical at g - 2.062 is present in both samples. Furthermore, a shoulder at 2.028 in the spectrum of irradiated GSSG corresponds to the peak maximum of the perthiyl feature in irradiated LA. This is also confirmed by power saturation studies that show the feature at 2.062 has a P112 greater than 20mW and a linear saturation profile in the 0.002mW to 6mW range (figure 5.2.9). This is similar to the perthiyl radical behavior seen in both irradiated GSH and LA. The perthiyl radical yield under the conditions investigated is very low as evidenced by the weak EPR signal. This is mirrored by the XAS experiment, where the peak at 2470.2eV has a very low intensity. It is important to note that after 60 minutes of irradiation the NMR of the GSSG sample is still dominated by the disulfide features (figure 5.2.7). 53 1.8. 1.6. 1.4. Cl) z LU I 1.0. -J 0.0- —0.2— • • • • • 0.0 0.5 1.0 1.5 2.0 2.5 SQUARE ROOT OF MICROWAVE POWER mW112 Figure 5.2.9 Power saturation study of irradiated GSSG feature at g2.O62 attributed to the perthiyl radical; (modulation frequency 50MHz, modulation amplitude 3.0G. 5 scans). 5.3 CONCLUSION XAS spectra of irradiated GSNO and LA under He atmosphere indicate the formation of the perthiyl radical at —2470.0eV. The presence of the perthiyl radical was clearly shown in the EPR spectra of irradiated LA; however it is proposed that GSNO must first react to form GSSG before being able to account for the perthiyl radical signal in its XAS spectrum. Formation of GSSG from irradiated GSNO is indicated by the XAS spectra as evidenced by the disappearance of the features due to the S-NO bond and formation of new peaks at 2472.8eV and 2474.0eV. This was confirmed with NMR where the transition from GSNO to GSSG (figure 5.2.7) occurs with little or no side products. Further irradiation of the sample resulted in formation of a small amount of perthiyl radical giving rise to a weak EPR signal. XAS of irradiated GSNO therefore shows both the formation of GSSG along with the weak signal due to the perthiyl radical S*+.Si transition. . . . g—2.062 FEATURE OF IRRADIATED GSSG 54 6 Effects of Hyperconjugation on the Electronic Structure and Photo-reactivity of Organic Sulfonyl Chlorides 6.1 BACKGROUND The electronic structure of p-toluene sulfonyl chloride and related organic sulfones has generated some attention because of its role as an initiator in living radical polymerization reactions (scheme 6.1.1)161. 162 Living radical polymerization is characterized by a faster initiation step than the following propagation reactions and a minimization of termination processes, resulting in a narrower polydispersity index162’ 163 This is achieved by the presence of actively propagating species due to the persistent radical effect1. Seen from the perspective of the polymerization reactions involving sulfonyl chloride, this effect can be explained as follows. The S-Cl bond of the sulfonyl chloride is catalytically broken by CuCI/bpy (bpy = 2,2’-bipyridine) to give the sulfonyl radical and the corresponding CuCI2 organometallic complex. The sulfonyl radical can further react with the olefinic substrates, which can in turn further polymerize. However, termination events are inhibited by a rise in the concentration of the CuCl2 complex. As mentioned, initiation results in the formation of the CuCI2 complex. These complexes can not react with each other so they accumulate. If the growing chains or initiators react with each other (“self terminate7disproportionate) the concentration of the CuCI2 complexes will still increase with more initiation events. Further termination steps would involve the reaction of growing polymers (which are free radicals) with CuCl2 to give a halogenated alkyl, rather than “self termination” simply due to the increase in the concentration of the CuCI2 complex. However, the now dormant alkyl halides can be reactivated by the CuCI/bpy catalyst. Systems following the persistent radical effect reach a steady state of growing radicals established between the activation and deactivation (propagation and termination) processes. This demands stoichiometric amounts of catalyst to be added for proper reactivity modulation of “capped” dormant alkyl halide chains163. The S-Cl bond cleavage in addition to metal reduction can also be initiated by thermolysis or photo irradiation162’165-167 55 Scheme 6.1.1 Metal catalyzed living radical polymerization with p-toluene sulfonyl chloride as an initiator. C) C) R + C-) C) + I + CN C) C) I ICNC) C) + H2C z 0 H2C c’J C-) C-) •[oR CI-44JR C) z0 z z 0 z w I— w -J (1) w 56 Both aryl and alkyl sulfonyl chlorides (RSO2CI) have been termed universal initiators of metal catalyzed living radical polymerization of styrene, methylacrylates and acryIates6”162 Regardless of the nature of their R group (alkyl/aryl) or electron withdrawing or donating effects of substituted aromatic R groups, the polymerization reactions involving RSO2CI result in faster initiation than propagation and narrow polydispersity indices161.This has been attributed to several factors such as the faster formation of sulfonyl radicals vs. carbon centered radicals168, and the low rate of sulfonyl radical dimerization as compared to carbon radicals161 167 Most notably they also attributed this universality to a lack of effect of the R group on the reactivity of the sulfonyl radical161’ 169 For the aryl compounds this was attributed to poor conjugation between the aromatic ring and the suifonyl moiety. However the rate of oxidation of CuCI2 by aryl sulfonyl chlorides was shown to be impacted by the substituent on the aromatic ring: electron withdrawing groups at the para-substituted position enhanced copper oxidation166. Furthermore there is both computational and experimental evidence for hyperconjugation in sulfonyl compounds128’ 129 The computational study coupled with X-ray diffraction data for a series of sulfate monoesters, sulfamates, and methanesulfonates shows that the sulfur bonding is highly polarized with the substituents around the sulfur acting both as donor and acceptors resulting in a sulfonyl bonding manifold composed of polar interactions with reciprocal hyperconjugative bonding128. Secondly, a bathochromic shift in the benzene UV-Vis absorption is observed when a sulfonyl group is attached to a benzene ring, indicating conjugative mixing between the orbitals of the two moieties129. Therefore, the electronic structure of p-toluene sulfonyl chloride (Ia) and related sulfonyl species (figure 6.2.1) was explored in order to better understand their electronic structure, the importance of hyperconjugation, as well as its impact on sulfonyl radical generation and subsequent radical polymerization. A series of compounds of the form RSO2G were probed using sulfur K-edge XAS spectroscopy to determine the effect on the SG bond (G = -Cl, -OH, -alkyl) due to the presence of a it system in the R group. In particular, aryl (a, R = p-XC6H4-, X = H/CH3) and alkyl (b) R groups were chosen to study the effect of orbital mixing in the sulfonyl centre. Photo-cleavage (scheme 6.1.2) of the S-Cl bond, via irradiation with a Xe arc lamp, was investigated for the sulfonyl chlorides (1ab) and the reactivity was correlated to the hyperconjugative effects 57 observed. Molecular orbital calculations were carried out to aid in the assignment of spectral features. RSO2C1 hv_> [Rso2cl]* > RSO + C1 Scheme 6.1.2. Sulfonyl chloride photo-cleavage reaction (R alkyl, p-tolyl). 6.2 RESULTS Suffur K-edge XAS spectroscopy The sulfur K-edge XAS spectra of 1-3 a,b (figure 6.2.1), are shown along with a detailed analysis of simulated spectra in figures 6.2.3 to 6.2.5. The XAS of sodium methane sulfone was used as the model spectrum for 2b to facilitate analysis of solid samples. The spectra of aqueous methane sulfone17°and sodium methane sulfone are comparable. The pre-edge regions of the model spectra exhibit clear differences in their features as a function of the R group and substituent G. The aryl compounds (l-3a) show additional features not present in the spectra of their alkyl counterparts (l-3b). Compound Ia has three features, two peaks at 2477.4eV and 2481.2eV and a shoulder at 2479.6eV, which is not seen in lb where only the peaks at 2477.6 and 2480.9eV are present. The sutfonate 2a has a main peak at 2481.7eV flanked by two shoulders at 2479.9eV and 2483.9eV. In contrast 2b has only one shoulder at 2483.1eV in addition to its main peak at 2481.3eV (see figure 6.2.4). Lastly, 3a exhibits a peak at 2478.6eV in addition to the main absorption feature of the spectrum seen in 3b. The main absorption feature is also at slightly higher energy in 3a (2480.6eV) compared to 3b (2480.1 eV). The XAS spectra show that the aromatic ring has a significant effect on the energy and sulfur 3p character of the valence orbitals. This could be due to energy redistribution of the existing transitions in the alkyl compounds, the presence of additional transitions or a combination of the two. 58 Ethyl Phenyl Sulfonep-Toluene Sulfonyl Chloride Ia p-Toluene Sulfonic Acid 2a 3a 0 CI——CH3 II 0 Methane Sulfonyl Chloride lb 0 II HO—S—CH3 0 Methane Sulfonate 2b H3C’ Methionine Sulfone Figure 6.2.1 Structures of la. p-toluene sulfonyl chloride; lb methane sulfonyl chloride; 2a p-toluene sulfonic acid; 2b methane sulfonate; 3a ethyl phenyl sulfone; 3b methiomne sulfone. 3b 59 DFT calculations and spectroscopic assignments Spectral features and electronic transitions were simulated using TD-DFT. Fragment calculations were also carried out on R, SO2 and G for a better description of the assigned transitions and to assess the effect of the aryl substituent. The relevant transitions and their descriptions are listed in Tables 6.2.1 to 6.2.6 for each of the compounds. The XAS of 3b was simulated using (CH3)2S0 as a model compound. It has been previously shown that TD-DFT transitions for (CH3)2Sare in good agreement with XAS spectra of methionine15. Two conformational models were evaluated for the simulation of the sulfur K-edge XAS spectra of 3a. The lower energy conformation (linear) is only —O.42kJ/mol lower in energy than the higher energy conformation (bent). This low energy barrier suggests that the molecule is able to interconvert between linear and bent conformations (figure 6.2.2). The XAS spectra were best simulated using the bent conformation and thus this geometry was used in our analysis. LINEAR BENT H3C CH2 H3C Figure 6.2.2 Geometries of ethyl phenyl sulfone conformations (3a). The TD-DFT simulated spectra (figures 6.2.3 - 6.2.5) for the alkyl compounds (1-3b) are in good agreement with experimental data. The spectra of the alkyl compounds are dominated by features due to the o orbitals of the S-C and S=O bonds in the RSO2 molecular fragment. A low energy feature corresponding to the SCl0*+Si transition is clearly visible in lb. For the aryl compounds (1-3a) only the calculated spectrum for 2a is in good agreement with experimental data. Spectra from TD-DFT for Ia predicts the feature due to SCla+—Sis transition at 2477.4 to have a shoulder, while there is no accounting for the actual shoulder at 2479.6eV in the experimental data. For 3a, TD-DFT completely fails to predict the feature at 2478.6eV. 60 2472 2476 2480 2484 2488 ENERGY (eV) Figure 6.2.3 Comparison of simulated spectra using TD-DFT and ASCF, and experimental spectra for compounds la,b the sulfonyl chlorides. >- H (I) z w I z w N -J 0 z >- I— Cl) z Ui I— z 0 Ui N -J 0 z 2472 2476 ENERGY (eV) 2488 61 ENERGY (eV) ENERGY (eV) Figure 6.2.4 Comparison of simulated spectra using TD-DFT and zSCF, and experimental spectra for compounds 2a,b the sulfonates. 62 Cl) z LU I z U LU N -J 0 z Cl) z Ui I— z Lii N -j 0 z (I) z LU I— z LU N -J 0 z Cl) z LU I z LU N -J 0:: 0 z Figure 6.2.5 Comparison of simulated spectra using TD-DFT and tSCF, and experimental spectra for compounds 3a,b the sulfones. 63 2472 2476 2480 2484 2472 2476 2480 2484 ENERGY (eV) ENERGY (eV) However, TO-OFT generates a spectrum using the ground state calculation as its only reference, and thus does not account for possible electronic relaxation in the excited state. To gauge the importance of this effect, the transition energies were recalculated using the Slater transition state ASCF approach by populating the acceptor orbital with half an electron and removing the same half electron from the S1 core orbital14”171 ASCF can overestimate the relaxation shifts; however, it can be very effective in determining which transitions are most susceptible to relaxation effects. Generally, the alkyl sulfonyl compounds were not affected by relaxation effects as much as the aryl sulfonyls. But even for the alkyl compounds the ASCF spectra show broadening of the main spectral features and even a splitting into two peaks as is the case of 3b. For the aryls the relaxation effects seem least important in the case of 2a where there is not much change in the SCF simulation from that of the TD-DFT. Relaxation effects become more pronounced in 3a. Here, IXSCF accounts for the peak at 2478.6eV by shifting the energies for TO-OFT calculated transitions 5 and 6 to lower energy by -1.1eV. In Ia there is a dramatic splitting between transition I and transitions 2 and 3 (table 6.2.1), pushing the latter transitions to higher energy. This results in a single peak for the transition at 2477.4eV and suggests that transitions 2 and 3 are responsible for the shoulder at 2479.6eV in the XAS spectra, which TD-DFT calculations failed to account for. It is apparent that orbitals with iu antibonding character, especially those of the aryl rings are most affected by electronic relaxation. Together, the TD-DFT and tSCF results allow us to assign the features of the sulfur K-edge XAS spectra for each species as illustrated in figures 6.2.3 to 6.2.5. The sulfonyl chlorides exhibit low energy features corresponding to the energy of the SCI—S18 transition and higher energy features due to the SO0—S1and SCa*#Sis transitions. The shoulder at 2479.6eV, which is present only in the aryl (Ia), corresponds to the4*4_Sj transition. cI* and cJ are the two lowest energy it orbitals of the aryl ring (figure 6.2.6). i1. has the largest electron density on the carbon bound to the sulfonyl moiety allowing it to mix with the SCIa* final state. The mixing gives r sulfur p character allowing this transition to be visible in the sulfur K-edge XAS spectrum (figure 6.3.1). The XAS spectra of the sulfonic acids (2a,b) do not show a distinct pre-edge feature relating to the SG*, because the SOH0* orbital is higher in energy than SCla*.and overlaps with the main feature of 2a,b at 2481.7eV. The lower energy shoulder in 2a is attributed to the SOH0*_Si, which is lowered in energy going 64 from the alkyl to the aryl by mixing with the orbital of the aryl group. Similarly the main features of 3a and 3b correspond to a combination of SO—S1and SC—S1 transitions. These transitions are lower in energy than in either the case of the sulfonyl chlorides (la,b) or the sulfonic acids (2a,b). Like in the previous examples, the presence of the aryl group is accompanied by additional features in the spectra. 3a has a lower energy feature not present in 3b attributed to the cI* final state mixing with SO orbital. This interaction is particularly favorable in 3a because the two energy states are closer in energy than in any of the other aryl compounds. ,J2 LI Figure 6.2.6 Molecular orbitals of benzene. 65 Table 6.2.1. Sulfur K-edge DFT calculated transitions for la with major contributors listed first. ASCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary + Others) Eneiies(eV) ASCF (eV) f 1 SCl0 , , so 2476.99 0.0 2.3x10 2 2477.85 +2.3 1.1x104 3 , so, sci 2477.94 +3.2 1.3x10 4 c,S , so 2479.74 +1.8 3.6x 104 so 2480.13 +1.4 7.9x104 6 so, , so, ci 2480.59 +1.3 1.Ox 104 2480.69 +2.5 9.6x104 8 2480.76 +2.9 1.4x103 9 SO* 5O cJ 2481.14 +2.6 2.4x103 Table 6.2.2. Sulfur K-edge DFT calculated transitions for lb with major contributors listed first. ASCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary, Others) Eneiies(eV) zSCF (eV) f 1 SC1a* , 2477.50 0.00 3.8x1W 2 SCa* , SOa* 2479.72 +0.66 7.2x 104 3 SC , SOa* 2480.44 +0.18 1.3x10 4 SCcy* 2480.72 +1.12 3.5x104 5 SO CHa* 2480.92 +0.70 1.6x103 6 CH , SO0 2481.48 +1.84 2.4x 104 7 CHa* , SOC. 2481.67 +1.68 7.9x 104 66 Table 6.2.3. Sulfur K-edge DFT calculated transitions for 2a with major contributors listed first. ASCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary, Others) EnV) ASCF (eV) f 1 ci , SO* , SOH0 2479.16 0.00 1.3x10 2 2479.87 +0.16 l.0x104 3 , SOHy*, SOa* 2479.97 +0.49 3.6x10 4 SO,rjo* 2480.93 +1.27 4.3x10 5 so , 2481.71 +0.11 3.2x10 6 SC* , SO* 2481.92 -0.28 8.7x 104 7 SCo , OHa* 2482.25 +1.21 1.2x103 8 SC* , SO,, OH0 2482.43 +1.76 3.0x10 2482.78 +1.86 7.4x105 Table 6.2.4. Sulfur K-edge DFT calculated transitions for 2b with major contributors listed first. zSCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary, Others) Energies(eV) zSCF (eV) f 1 SOHa* , SCy* , SOa* 2479.50 0.00 2.0xl0 2 SO , SCa* 2480.35 -0.21 2.7x103 3 SCa* , SOHa* 2480.78 +0.02 1.6x1W3 4 SC ,SO 2481.18 -0.30 1.5x103 5 SO7rJcy* , SO 2481.31 +0.37 1.6x103 6 CHa* 2481.61 +0.05 2.3x10 7 CHa* ,SO 2481.83 +1.04 7.1x104 67 Table 6.2.5. Sulfur K-edge DFT calculated transitions for 3a with major contributors listed first. ASCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary, Others) Energies(eV) bSCF (eV) f 1 2476.95 0.00 1.8x104 2 2477.36 +0.85 3.6x10 3 SC,* , 2479.04 -0.86 4.1x105 4 SCa* ,CSa* 2479.53 -1.15 1.4x10 5 SO , SO , SC, C1$0 2479.81 -1.20 1.0x103 6 , SO0., SC0., 2480.01 -1.19 6.8x104 7 SO* , SO,. 2480.17 -0.57 l.6x1O 8 SO* , SO* 2480.34 -1.03 6.7x104 9 CH0. 2480.42 -0.49 8.8x10 Table 6.2.6. Sulfur K-edge DFT calculated transitions for 3b with major contributors listed first. tSCF calculated energy shifts are referenced to the lowest TD-DFT calculated energy transition. Transition Assignment (Primary, Others) EnV) tSCF (eV) f 1 SC0. 2478.53 0.00 1.9x104 2 SC0* 2479.24 -0.62 2.4x103 3 CH,. , SO0. 2479.59 -0.30 6.2x104 4 SC0* , SO. 2479.69 -0.54 7.Ox 5 CH0. , SO0. 2479.87 -0.33 1.Ox 6 son. , CH0. 2480.31 +0.13 3.4x10 7 CH0. ,SOo 2480.52 +0.12 4.7x10 8 SO0. ,CH0. 2480.75 +0.16 1.7x103 68 Photo-cleavage Reactions 0 z w I z fD w N -J 0 z To examine the effect of the aryl ring on the generation of the sulfonyl radical and photo-cleavage of the S-Cl bond, p-toluene sulfonyl chloride (Ia) and methane sulfonyl chloride (1 b) were irradiated with a 75W Xe arc lamp and XAS spectra were collected as described in the experimental section (figure 6.2.8). Scans were acquired consecutively with continuous irradiation and changes in the intensity of the features attributed to the S-Cl bond, present at 2477.4eV in Ia and 2477.6eV in Ib, were recorded. Each scan lasted 5.5 minutes, and over the same time span of irradiation the intensities of both peaks decrease indicating cleavage of the S-Cl bond. The photo- cleavage rate of Ia containing the aryl moiety was much higher (figure 6.2.7) than in the alkyl containing compound lb. This is indicated by the larger decrease in intensity with irradiation attributable to the S-Cl bond feature of Ia. — — —.. — —.. — —B. — _•. — — .. SCIm— S INTENSITIESis —.—la — .— lb 2.2 2.0 1.8 1.6 1.4 1.2 1.0 I . I I I I — I — -5 0 5 10 15 20 25 3035 40 45 50 55 60 65 TIME (MIN) Figure 6.2.7 Time evolution of the sulfur K-edge features due to the S-Cl bond in la and lb with in-situ irradiation. 69 Figure 6.2.8 XAS spectra with in situ irradiation of la (top) and lb (bottom) with a 75W Xe arc lamp under anaerobic conditions. Scans were taken every 5.5min with continuous irradiation. 70 2470 2475 2480 2485 2490 ENERGY (eV) 6.3 DISCUSSION Assigning the sulfur K-edge XAS spectral features of these compounds is the first step in determining the effect of their electronic structure on their respective reactivities. Of particular interest are p-toluene sulfonyl chloride (Ia) and its role as an initiator in living radical polymerization reactions. The reactivities of the complexes studied should be affected by electronic coupling such as that due to an aromatic group bound directly to the sulfonyl moiety. These interactions have been confirmed by the presence of features in the sulfur K-edge XAS spectra of the aryl compounds which, are not present in their alkyl counterparts, and are attributable to the mixing of cJY (aryl 1* orbitals) with orbitals containing sulfur 3p character. The XAS data coupled with the molecular orbital calculations give insights into the nature of bonding in these systems. In the alkyl compounds (I-3b), there is a distinct ordering of the empty valence orbitals with the lowest energy attributed to the SG0 followed by SR0, SO, and SO0. The splittings between these states is generally small resulting in a single intense broad feature for these species. The exception is the methane sulfonyl chloride (I b) with a very low SG0 feature -3eV below the main peak attributed to theSCl0*÷_Sjtransition. In contrast, the ordering in the aryl compounds is switched such that the transitions due to SR0*4_Sls are higher in energy than those of SO—S1,resulting in an energy arrangement resembling E(SG0*) <E(SO*) < E(SO,*) < E(SRa*). The reordering of SR0 and SO0 final states in the aryl compounds is attributable to the stronger S-C bond due to the sp2 character of the aryl carbons172. This pushes the SC0 orbital to higher energy, which is consistent with the main sulfur K- edge feature for the aryl compounds being —0.4eV higher in energy than that of the alkyl species. Furthermore, the aryl group has two low-lying ir’ orbitals and that can mix and redistribute intensity in the Sulfur K-edge spectra. Since T* has no electron density on the carbon bound to the sulfur, t* is effectively non-bonding with respect to the sulfonyl moiety (figure 6.3.1). As already mentioned cF’ has good overlap and a strong interaction with the sulfur moiety. This interaction, however, is dependent on the nature of the G group and its bonding interactions. The energy of the SGa*+_Sis increases from Cl— OH —* CH3, in agreement with expected bond strengths172174. In 71 the case of Ia, mixing with cI results in a further lowering of the SCIa*. The higher energy SOHa* orbital in 2a can interact more fully with the aryl group resulting in a highly mixed lowest lying final state in the XAS data. In 3a however, it is the mixing of the aryl with the SO which takes precedence. The absence of a strong SG interaction with could be attributed to both energetic and overlap considerations. As mentioned earlier, the SO* is closer in energy to the aryl I* and the SO orbitals might more readily overlap with the aryl t’ than the sp3 hybridized orbitals of the ethyl G group in 3a. - SCIa* SCIa* Figure 6.3.1 Mixing of the I*, cJ and SCI0* orbitals resulting in transitions 1 and 3 in the sulfur K-edge spectrum of la (top). (Isovalues = 0.060e.A3). To test whether the effect of the aryl group could in fact be an inductive effect rather than a conjugative one, the aryl ring was rotated with respect to the SG bond to simulate “turning off’ aryl hyperconjugation yet maintaining an inductive effect. The result of the rotation about the S-C(Sp2)bond on the predicted sulfur K-edge spectrum of Ia as calculated by TD-DFT is shown in figure 6.3.2. The starting point for the calculation has the aryl ring at 900 to the S-Cl bond and is consistent with the geometry observed in the crystal structure of Ia where the aryl ring and the S-Cl bond are almost perpendicular (84.3°). As the aryl ring is rotated from a perpendicular plane to the CI \ S CI S it4 SCI + 72 SG bond, a situation allowing for maximum conjugation, to a plane parallel to the SG bond, resulting in no conjugation to the SC bond, the splitting between the SCIO*—Sis and cI ÷—S1 decreases by —1eV. At the same time the intensity of SCl0*÷—Si increases suggesting increased sulfur p character, while the intensity of I* *—S, decreases indicating a decrease in the sulfur p character. This leads to the conclusion that as hyperconjugation is “turned off” the mixing between cJi. and S is also turned off. The calculated energy stabilization of the hyperconjugative interaction on SCl* is —0.5eV which is equal to the experimental value for the difference in SCl0*.Si transition energies going from lb to Ia. Figure 6.3.2 Effect of turning off 3?4ISCl0.mixing on the intensities and energies of transitions 1 and 3, by rotating the aryl ring. . 2477 2478 2479 2480 2481 2482 ENERGY (eV) 73 It is important to note that the above mentioned hyperconjugative interaction occurs between two empty antibonding orbitals; hence it is not a typical hyperconjugative effect. This excited state hyperconjugation17178 enhances cleavage of the S-Cl bond in Ia over lb in accordance with the postulate that the aryl group should have a large effect on the S-Cl bond cleavage. Direct photolysis likely results from excitation into the SCI0 acceptor orbital with subsequent bond cleavage to form radical products. This study provides a direct assessment of the nature of the SCI0* orbital and the effect of the aryl group on the photolytic process. This is very beneficial for living radical polymerization since initiation has to be faster than propagation; therefore facile radical generation is key. The aryl group in Ia allows for partial delocalization of the excited state electron through mixing with the I* orbital. Mixing of the cIi with SCl orbital results in a decrease in the excitation energy as previously described, but also in a likely increase in the excited state lifetime through charge separation. An increased lifetime of the SCI0 excited state allows for a higher transmission coefficient for the overall reaction and a faster rate of photo-cleavage. It can also be argued that the rate of photo-cleavage is faster in toluene sulfonyl chloride because the aryl group itself can enhance the absorption of photons, however this is unlikely since the source of irradiation passes through a plastic window which should remove photons with wavelenghts below --350nm. Because toluene has an absorption maximum at —260nm179,no enhancement of photon absorption should occur due to the presence of the aryl group under the described experimental conditions. 6.4 CONCLUSION In the work on sulfonyl complexes of the type RSO2G interactions present in the aryl compounds but not seen in their alkyl counterparts were identified. The mixing of the cI%* aryl orbital into the sulfonyl moiety gives rise to new features in the sulfur K- edge spectra. These features were characterized. Of particular interest is the excited state hyperconjugation interaction between t* and SCl0, resulting in a faster photo cleavage rate for the S-Cl bond. Excited state hyperconjugation facilitates photo cleavage by lowering the SCI0* energy and allowing delocalization of the excited state, increasing its lifetime and enhancing the photo-cleavage reaction. It is then reasonable to assume that the magnitude of excited state hyperconjugation can be modulated by 74 % mixing = (i48 xlOO% 2478 2479 ENERGY (eV) Figure 6.4.1 Fitted XAS spectra of p-toluene sulfonyl chloride (la) with intensities calculated for the SCl0*Sitransition and the4*+—Sis transition. electron withdrawing or donating groups on the aromatic ring. Future experiments include plans to acquire XAS spectra for a series of para substituted aryl sulfonyl chlorides with a variety of electron withdrawing and electron donating substituents. Characterization and quantization of the features arising due to the aryl cI* mixing with SCI0* should be a good measure of the total excited state hyperconjugative interaction present. A preliminary quantization of this effect in p-toluene sulfonyl chloride Ia was carried out. The increase in intensity (I_s ) of thet4*4_Sl transition coupled with the decrease in SCl0**—S1 transition intensity (Isci4s) is a direct measure of cI ESC10 mixing and hence excited state hyperconjugation. Assuming that no other contributions are present in the features due to the SCl0 and t orbitals, the percent of excited state hyperconjugation can be calculated by fitting the sulfur K-edge spectra (figure 6.4.1) and applying equation 6.4.1 to the measured intensities. For Ia 10-15% mixing of the cI* into the SCI0* is estimated. Equation 6.4.1 2482 75 The question still remains as to why excited state hyperconjugation does not seem to affect the living radical polymerization reactions in which p-toluene sulfonyl chloride (Ia) acts as an initiator. By definition the initiation step in such reactions is the faster than the propagation step, so the formation of the aryl sulfonyl radical intermediate will not be the major factor to impact reactivity. Furthermore using EPR techniques, previous research has shown that the sulfur p orbital containing the unpaired electron in the aryl sulfonyl radical is in the plane of the phenyl ring180. This would preclude any interaction of the paramagnetic S3, orbital with the aryl cI4 orbitals. Since it is the half empty S3p orbital which is involved in the polymerization reaction, there would be no major effect on the propagation step due to excited-state hyperconjugation. Also, as the polymer chain grows one would expect the effect of the aryl sulfonyl moiety to diminish. Even if the orientation was optimal for mixing of the S3 and i% states in the aryl sulfonyl radicals, preliminary TD-DFT calculation predict this interaction to be minimal. 76 7 Concludini Remarks and Outlook In this thesis, the application of Sulfur K-edge XAS to investigate the photo- reactivity of a series of model sulfur containing compounds was discussed. It was shown that XAS can be applied to a wide variety of systems with applications both in biological and inorganic chemistry. XAS proved to be a useful tool in the detection of thiyl radical intermediates in UV irradiated GSH, which are difficult to characterize using other spectroscopic techniques such as EPR. Further research investigating thiyl radical generation and characterization via XAS may help in the elucidation of the mechanism of action of enzyme systems which form thiyl radical intermediates, as is proposed in the case of ribonucleotide reductase123. Additional sulfur based intermediates were also characterized in the form of the stable perthiyl radical, which is a product of photo-irradiation in all biologically relevant low molecular weight sulfur species investigated, Initially it was somewhat unclear what the mechanism of perthiyl radical formation in GSNO is; however, the XAS spectra indicated generation of a disulfide bond upon photo-irradiation. NMR of irradiated GSNO confirmed disulfide bond formation, and EPR showed that only after disulfide bond generation is the perthiyl radical formed during photo-irradiation. :: LL r!i\! / I I LIPOIC ACID ‘ NITROSOGLUTATHIONEIv XAS XAS 246.0 2482 2484 2464 2448 2470 2472 2474 2476 2478 ENERGY (eV) 20 1.5 1.5 0 z 24 isb’ 2468 2470 2412 Figure 7.1.1 XAS data from irradiated LA and GSNO complements well the information from both NMR and EPR experiments, filling in the gaps when needed. 2474 2470 2470 2480 2482 2484 ENERGY(eV) 77 Furthermore it is important to note that in both the case of LA and GSNO, XAS detected the intermediates and products being formed, while detection with EPR and NMR proved difficult (figure 7.1.1). In the case of LA, the intermediate was readily detected by EPR while the final major product, which is believed to also be a disulfide, could not be characterized by NMR due to its insolubility. For GSNO the story is somewhat the opposite. While the major product was easily characterized by NMR the formation of the perthiyl radical intermediate seen in the XAS spectra required extensive EPR characterization. The sensitivity of this technique to the bonding configuration is particularly evidenced in the case of LA, which has features due to SSa*4_Sls and SCa*Sis transitions at lower energy than “linear” disulfides suggesting weaker bonding in the sterically strained pentacyclo-disulfide. GSNO, on the other hand, exhibits sulfur core excitations to the SN* orbital consistent with theoretical calculations, indicating the presence of an S-Na resonance form in nitrosylated thiols157. The applicability of XAS to characterize electronic configurations could be used to further investigate the bonding interactions in S-nitrosothiols. Understanding the bonding interaction in these compounds could help in the synthesis of S-nitrosothiols in which N0 release can be modulated with the potential to target specific locations within the body, and stimulate N0 controlled signaling pathways with beneficial consequences. The power of XAS to elucidate electronic configuration was also exploited in the case of p-toluene sulfonyl chloride. Here, it was found that p-toluene sulfonyl chloride is more susceptible to photo-reactivity than its alkyl counterpart methane sulfonyl chloride. This was attributed to an excited hyperconjugation bonding interaction, directly detectable via XAS, found between the aromatic antibonding orbitals and the SCla* orbital in p-toluene sulfonyl chloride. Excited state hyperconjugation has also been observed in a series of ruthenium arene thiolates (Ru(p-cym)(en)S02— Ph) whose role as therapeutic agents to combat cancer is dependent on Ru-S bond dissociation181. Hyperconjugation could play an essential role in these species and may influence bioactivity of the aryithiolato version of these drugs. Modulating this effect could lead to better anticancer therapeutic agents. Future work includes measuring the excited state hyperconjugation effects in a series of para substituted phenyl sulfonyl chlorides to 78 determine the impact of electron withdrawing and donating groups on S-Cl bond strengths, the results of which could be applied to the ruthenium arylthiolato complexes. Coupled with molecular orbital calculations and other spectroscopic techniques XAS is a powerful tool that can offer a wealth of information on the bonding, reactivity and mechanism of reaction for a variety of systems. 79 References 1. George S 0, Petrenko T, and Neese F, lnorg Chim Acta, 2008, 361(4), p965. 2. Pickering T J, Prince R C, Divers T, et at, FEBS Lett, 1998, 441(1), p11. 3. Rompel A, Cinco R M, Latimer M J, et al., Proc NatlAcad Sci USA, 1998, 95(11), p6122. 4. Huffman G P, Shah N, Huggmns F E, et al., Fuel, 1995, 74(4), p549. 5. Solomon 0, Lehmann J, and Martinez C E, Soil Sci Soc Am J, 2003, 67(6), p1721. 6. Sandstrom M, Jalilehvand F, Persson I, et al., Nature, 2002, 415(6874), p893. 7. Gales L, Cardoso I, Fayard B, et al., J Biol Chem, 2003, 278(1 3), p1 1654. 8. Prange A, Dahl C, Truper H G, et al., EurPhys J D, 2002, 20(3), p589. 9. Bellacchio E, McFarlane K L, Rompel A, et al., J Synch Rad, 2001, 8, p1056. 10. Prange A, Chauvistre R, Modrow H, et al., Microbiology-Sgm, 2002, 148, p267. 11. Shadle S E, Hedman B, Hodgson K 0, et al., lnorg Chem, 1994, 33(19), p4235. 12. Williams K, Hedman B, Hodgson K 0, et al., Inorg Chim Acta, 1997, 263(1-2), p315. 13. Solomon E I, Hedman B, Hodgson K 0, et al., Coord Chem Rev, 2005, 249(1-2), p97. 14. Szilagyi R K and Schwab D E, Biochem Biophys Res Commun, 2005, 330(1), p60. 15. Sarangi R, Frank P, Hodgson K 0, et al., lnorg Chim Acta, 2008, 361(4), p956. 16. Moriarty-Craige S E and Jones D P, Annu Rev Nutr 2004, 24, p481. 17. Schafer F Q and Buettner G R, Free Radical Biol Med, 2001, 30(11), p1191. 18. Kirlin W G, Cai J, Thompson S A, et al., Free Radical Biol Med, 1999, 27(11-12), p1208. 19. Jones D P, Carlson J L, Mody V C, et al., Free Radical Biol Med, 2000, 28(4), p625. 20. Hoshi T and Heinemann S H, JPhysiol-London, 2001, 531(1), p1. 21. Pastore A, Federici G, Bertini E, et al., Clin Chim Acta, 2003, 333(1), p19. 22. Myers P R, Minor R L, Guerra R, et al., Nature, 1990, 345(6271), p161. 23. Jocelyn P C, EurJ Biochem, 1967, 2(3), p327. 24. Lodge J K, Traber M G, and Packer L, Free Radical Biol Med, 1998, 25(3), p287. 25. Moini H, Packer L, and Saris N-E L, ToxicolApplPharmacol, 2002, 182(1), p84. 80 26. Allen R G, Newton R K, Sohal R S, et al., J Cell Physiol, 1985, 125(3), p413. 27. Sandstrom B E and Markiund S L, Biochem J, 1990, 271(1), p17. 28. KIug A and Rhodes 0, Cold Spring Harb Symp Quant Biol, 1987, 52, p473. 29. Sen C K and Packer L, FASEB J, 1996, 10(7), p709. 30. Suthanthiran M, Anderson M E, Sharma V K, et al., Proc Nat! Acad Sd USA, 1990, 87(9), p3343. 31. Avval F Z and Holmgren A, J Biol Chem, 2009, 284(13), p8233. 32. Ho Y-F and Guenthner T M, Chinese Pharm J (Taipei), 1997, 49(5-6), p267. 33. Hwang C, Sinskey A J, and Lodish H F, Science, 1992, 257(5076), p1496. 34. Sevier C S and Kaiser C A, Nat Rev Mo! Cell Biol, 2002, 3(11), p836. 35. Anfinsen C and Haber E, J Biol Chem, 1961, 236(5), p1361. 36. Staal F J T, Ela S W, Roederer M, et al., Lancet, 1992, 339(8798), p909. 37. BuhI R, Holroyd K, Mastrangeli A, et al., Lancet, 1989, 334(8675), p1294. 38. Eck H P, Gmunder H, Hartmann M, et al., Biol Chem Hoppe-Seyler 1989, 370(2), p101. 39. Staal, EurJ Clin Invest, 1998, 28(3), p194. 40. Smith CV, Jones D P, Guenthner TM, et aL, ToxicolApplPharmaco!, 1996, 140(1), p1. 41. Nakamura H, De Ros S C, Yodoi J J, et al., Proc NatlAcad Sd USA, 2001, 98(5), p2688. 42. Lioy J, Ho WZ, Cutilli J R, et al., J Clin Invest, 1993, 91(2), p495. 43. Denu J M and Tanner KG, Biochemistry, 1998, 37(16), p5633. 44. Hehner S P, Breitkreutz R, Shubinsky G, et al., J Immuno!, 2000, 165(8), p4319. 45. Read M A, Whitley M Z, Gupta S, et a!., J Biol Chem, 1997, 272(5), p2753. 46. Moinova H R and Mulcahy R T, Biochem Biophys Res Commun, 1999, 261(3), p661. 47. Hansen J M, Watson W H, and Jones D P, ToxicolSci, 2004, 82(1), p308. 48. Dinkova-Kostova AT, Holtzclaw W D, Cole R N, et a!., Proc NatlAcad Sd USA, 2002, 99(18), pll908. 49. Bloom D, Dhakshinamoorthy 5, and Jaiswal A K, Oncogene, 2002, 21(14), p2191. 50. Droge W, Phys Rev, 2002, 82(1), p47. 51. Finkel T and Holbrook N J, Nature, 2000, 408(6809), p239. 81 52. Wink D A and Mitchell J B, Free Radical Biol Med, 1998, 25(4-5), p434. 53. Nordberg J and Amer E S J, Free Radical Blot Mcd, 2001, 31(11), p1287. 54. Thomas E L, Lehrer R I, and Rest R F, Rev Infect Dis, 1988, 10, pS450. 55. Ignarro L J and Kadowitz P J, Annu Rev Pharmacol Toxicol, 1985, 25, p171. 56. Ignarro L J, Pharmacol Toxicol, 1990, 67(1), p1. 57. Radomski M W, Palmer R M J, and Moncada S, BrJ Pharmacol, 1987, 92(3), p639. 58. Turrens J F, Biosci Rep, 1997, 17(1), p3. 59. lnoue M, Sato E F, Nishikawa M, et al., Curr Med Chem, 2003, 10(23), p2495. 60. McIntyre M, Bohr D F, and Dominiczak A F, Hypertension, 1999, 34(4), p539. 61. Fridovich I, Science, 1978, 201 (4359), p875. 62. Rhee S G, Exp Mol Med, 1999, 31(2), p53. 63. Halliwell B, FASEB J, 1987, 1(5), p358. 64. Chance B, Sies H, and Boveris A, Phys Rev, 1979, 59(3), p527. 65. Leonard S S, Harris G K, and Shi X L, Free Radical Biol Med, 2004, 37(12), p1921. 66. Reddie K C and Carroll K S, Cuff Opin Chem Biol, 2008, 12(6), p746. 67. Chelikani P, Fita I, and Loewen PC, Cell Mol Life Sci, 2004, 61(2), p192. 68. Epp 0, Ladenstein R, and Wendel A, Eur J Biochem, 1983, 133(1), p51. 69. Chae H Z, Kim H J, Kang S W, et al., Diabetes Res Clin Pract, 1999, 45(2-3), p101. 70. Monteiro G, Horta B B, Pimenta D C, et al., Proc NatlAcad Sd USA, 2007, 104(12), p4886. 71. Rhee S G, Chae H Z, and Kim K, Free Radical Blot Med, 2005, 38(12), p1 543. 72. Biteau B, Labarre J, and Toledano M B, Nature, 2003, 425(6961), p980. 73. Woo H A, Chae H Z, Hwang S C, et a!., Science, 2003, 300(5619), p653. 74. Palmer R M J, Rees D D, Ashton D S, et al., Biochem Biophys Res Commun, 1988, 153(3), p1251. 75. Wink D A, Nims R W, Darbyshire J F, et al., Chem Res Toxicol, 1994, 7(4), p519. 76. Coupe P J and Williams D L H, J Chem Soc-Perkin Trans 2, 2001(9), p1595. 77. Stoyanovsky D A, Tyurina Y Y, Tyurin VA, et al., JAm Chem Soc, 2005, 127(45), p158l5. 82 78. Konorev E A, Katyanaraman B, and Hogg N, Free Radical Biol Med, 2000, 28(11), p1671. 79. AI-Sa’doni H and Ferro A, Clin Sd, 2000, 98(5), p507. 80. Butler A R, Al-Sa’doni H H, Megson I L, et aL, Nitric Oxide, 1998, 2(3), p193. 81. Davis F J, Gilbert B C, Norman R 0 C, et at., J Chem Soc-Perkin Trans 2, 1983(11), p1763. 82. PrutzWA, ButterJ, and Land E J, Biophys Chem, 1994, 49(2), p101. 83. Laurie S H, Lund T, and Raynor J B, J Chem Soc-Dalton Trans, 1975(14), p1389. 84. Lima A I G, Corticeiro S C, and de Atmeida Paula Figueira E M, Enzyme and Microb Technol, 2006, 39(4), p763. 85. Da Costa Ferreira A M, Cirioto M R, Marcocci L, et al., Biochem J, 1993, 292(3), p673. 86. Valko M, Rhodes C J, Moncot J, et at., Chemico-Bio! Interact, 2006, 160(1), p1. 87. Spear N and Aust S D, Arch Biochem Biophys, 1994, 312(1), p198. 88. Misra H P, J Biol Chem, 1974, 249(7), p2151. 89. Holmgren A, Annu Rev Biochem, 1985, 54, p237. 90. Taniguchi Y, TaniguchiUeda Y, Mon K, et at., Nucleic Acids Res, 1996, 24(14), p2746. 91. Spyrou G, Enmark E, Miranda-Vizuete A, et at., J Biol Chem, 1997, 272(5), p2936. 92. Miranda-Vizuete A, Ljung J, Damdimopoulos A E, et al., J BioI Chem, 2001, 276(34), p31 567. 93. Holmgren A, J Biol Chem, 1989, 264(24), p13963. 94. Berndt C, Liltig C H, and Holmgren A, Am J Phisiol Heart Circ Physiol, 2007, 292(3), pH1227. 95. KaIlis GB and Holmgren A, JBiolChem, 1980, 255(21), p10261. 96. Amer E S J and Holmgren A, Eur J Biochem, 2000, 267(20), p6102. 97. Tamura T and Stadtman T C, Proc NatlAcad Sd USA, 1996, 93(3), p1006. 98. Gladyshev V N, Jeang K T, and Stadtman T C, Proc NatlAcadSci USA, 1996, 93(12), p6146. 99. Sun Q A, Zappacosta F, Factor V M, et at., J Biol Chem, 2001, 276(5), p31 06. 83 100. Zhong L W, Amer E S J, and Holmgren A, Proc Nat! Acad Sd USA, 2000, 97(11), p5854. 101. Arscott L D, Gromer S, Schirmer RH, et al., Proc Nat! Acad Sd USA, 1997, 94(8), p3621. 102. Klintrot l-M, Hoog J-O, Jomvall H, et aL, EurJ Biochem, 1984, 144(3), p417. 103. Gladyshev V N, Liu AM, Novoselov S V, et al., J Blo! Chem, 2001, 276(32), p30374. 104. Wingert R A, Galloway J L, Barut B, et aL, Nature, 2005, 436(7053), p1035. 105. Takashima Y, Hirota K, Nakamura H, et al., !mmuno! Left, 1999, 68(2-3), p397. 106. Daily D, Vlamis-Gardikas A, Offen D, et aL, J Biol Chem, 2001, 276(2), p1335. 107. WeIIsWW, Xu D P, Yang Y F, et al., JBiolChem, 1990, 265(26), p15361. 108. Hirota K, Matsui M, Murata M, et al., Biochem Biophys Res Commun, 2000, 274(1), p177. 109. Gravina S A and Mieyal J J, Biochemistry, 1993, 32(13), p3368. 110. Bushweller J H, Aslund F, Wuthrich K, et al., Biochemistty, 1992, 31(38), p9288. 111. Mize C E, Thompson T E, and Langdon R G, J Biol Chem, 1962, 237(5), p1596. 112. Staal G E J and Veeger C, Biochim BiophysActa-Enzymology, 1969, 185(1), p49. 113. Massey V and Williams C H J, J Bio! Chem, 1965, 240, p4470. 114. Schulz G E, Zappe H, Worthington D J, et al., FEBS Left, 1975, 54(1), p86. 115. Zappe HA, Krohne-Ehrich G, and Schulz GE, JMolBiol, 1977, 113(1), p141. 116. Worthington 0 J and Rosemeyer MA, EurJBiochem, 1976, 67(1), p231. 117. Krohneehrich G, Schirmer RH, and Untuchtgrau R, EurJBiochem, 1977, 80(1), p65. 118. ElIman G L, Arch Biochem Biophys, 1959, 82(1), p70. 119. Eaton P, Free Radical Biol Med, 2006,40(11), p1889. 120. Ellis H R and Poole L B, Biochemistry, 1997, 36(48), p15013. 121. Jaifrey SR and Snyder S H, Science, 2001, 2001 (86), ppll. 122. Jalilehvand F, Chem Soc Rev, 2006, 35(12), p1256. 123. Nordlund N and Reichard P, Annu Rev Biochem, 2006, 75, p681. 124. Sjoberg B M, Metal Sites in Proteins and Models, 1997, 88, p139. 125. Carrington A and McLach Ian D M, Introduction to Magnetic Resonance. 1969, New York: Harper & Row,. 84 126. Symons M C R, J Chem Soc-Perkin Trans 2, 1974(13), p1618. 127. van Gastel M, Lubitz W, Lassmann G, et al., JAm Chem Soc, 2004, 126(7), p2237. 128. Denehy E, White J M, and Williams S J, lnorg Chem, 2007, 46(21), p8871. 129. Fehnel E A and Carmack M, JAm Chem Soc, 1950, 72(3), p1292. 130. Lytle F W, Greegor R B, Sandstrom 0 R, et aL, Nuci lnstr Meth, 1984, 226(2-3), p542. 131. Kennepohl P, Wasinger E, and George SD, JSynch Rad, 2009, 16, p484. 132. Hedman B, Frank P, Gheller S F, et al., JAm Chem Soc, 1988, 110(12), p3798. 133. Sekiyama H, Kosugi N, Kuroda H, et al., Bull Chem Soc Jpn, 1986, 59(2), p575. 134. Delgado-Jaime M U, Conrad J C, Fogg D E, et at., lnorg Chim Acta, 2006, 359(9), p3042. 135. Webb SM, Phys Scr 2005, T115, p1011. 136. Guerra C F, Snijders J G, te Velde G, et at., Theor Chem Acc, 1998, 99(6), p391. 137. Velde G T, Bickethaupt FM, Baerends E J, et al., J Comput Chem, 2001, 22(9), p931. 138. Schwarz K, Chem Phys, 1975, 7(1), pIOO. 139. Sen K 0, JPhys B, 1978, 11(19), pL577. 140. Gopinathan M 5, J Phys B, 1979, 12(4), p521. 141. Triguero L, Pettersson L GM, and Agren H, Phys RevB, 1998, 58(12), p8097. 142. Krzystek J, Sienkiewicz A, Pardi L, et at., J Magn Reson, 1997, 125(1), p207. 143. Hart T W, Tetrahedron Lett, 1985, 26(16), p2013. 144. Giles N M, Giles G I, and Jacob C, Biochem Biophys Res Commun, 2003, 300(1), p1. 145. Silva D J, Stubbe J, Samano V, et al., Biochemist,y, 1998, 37(16), p5528. 146. Lassmann G, Kolberg M, Bleifuss G, et at., Phys Chem Chem Phys, 2003, 5(11), p2442. 147. Kolberg M, Bteifuss G, Grastund A, et at., Arch Biochem Biophys, 2002, 403(1), p141. 148. Kolberg M, Bleifuss G, Sjoberg B-M, et at., Arch Biochem Biophys, 2002, 397(1), p57. 149. Terryn H, Tilquin B, and Houee-Levin C, Res Chem Intermed, 2005, 31(7-8), p727. 85 150. Morine C H and Kuntz R R, Photochem Photobiol, 1981, 33(1), p1. 151. Grant D Wand Stewart J H, Photochem Photobiol, 1984, 40(3), p285. 152. Wood P D, Bulent Mutus, and Robert W. Redmond, Photochem Photobiol, 1996, 64(3), p518. 153. Barron L B, Waterman K C, Filipiak P, et aL, J Phys Chem A, 2004, 108(12), p2247. 154. Bucher G, Lu C Y, and Sander W, ChemPhysChem, 2005, 6(12), p2607. 155. Manoj V M, Mohan H, Aravind U K, et al., Free Radical Biol Med, 2006, 41(8), p1240. 156. Williams DL H, Org Biomol Chem, 2003, 1(3), p441. 157. Timerghazin Q K, Peslherbe G H, and English A M, Org Left, 2007, 9(16), p3049. 158. George G N, Gnida M, Bazylinski D A, et at., J Bacteriol, 2008, 190(19), p6376. 159. Cavero M, Hobbs A, Madge D, et al., Bioorg Med Chem Left, 2000, 10(7), p641. 160. Petzold H and Sadler P J, Chem Comm, 2008(37), p4413. 161. Percec V, Kim H J, and Barboiu B, Macromol, 1997, 30(26), p8526. 162. Percec V, Barboiu B, and Kim H J, JAm Chem Soc, 1998, 120(2), p305. 163. Braunecker W A and Matyjaszewski K, Prog Polym Sd, 2007, 32, p93. 164. Fischer H, Chem Rev, 2001, 101(12), p3581. 165. Kharasch MS and ZavistA F, JAm Chem Soc, 1951, 73(3), p964. 166. Orochov A, Asscher M, and Vofsi D, J Chem Soc B Phys Org, 1969(3), p255. 167. Correa CM M and Waters WA, J Chem Soc C-Org, 1968(15), p1874. 168. Asscher M and Vofsi D, J Chem Soc, 1964(DEC), p4962. 169. AsscherJ SaM, J Chem Soc Perkin Trans 1, 1972, p1543 170. Akabayov B, Doonan C J, Pickering I J, et al., J Synch Rad, 2005, 12, p392. 171. Triguero L, Pettersson L GM, and Agren H, JPhys ChemA, 1998, 102(52), p10599. 172. Herron J T, Thermochemistty of sulfoxides and sulfones, in The Chemistiy of Suiphones and Suiphoxides, S. Patai, Z. Rappoport, and C.J.M. Stirling, Editors. 1988, John Wiley & Sons Ltd.: New York. p. 95. 173. Chatgilialoglu C, Griller 0, Kanabus-Kaminska J M, et al., J Chem Soc Perkin Trans 2, 1994(2), p357. 174. Korth H G, Neville A G, and Lusztyk J, J Phys Chem, 1990, 94(25), p8835. 175. Brunvoll J and Hargittai I, J Mo! Struct, 1976, 30, p361. 86 176. Nakal H and Kawai M, J Chem Phys, 2000, 113(6), p2168. 177. Rao C N R, Goldman G K, and Balasubramanian A, Can J Chem, 1960, 38, p2508. 178. Wladislaw B, Viertler H, Olivato P R, et at., J Chem Soc Perkin Trans 2, 1980(3), p453. 179. Hai D, Ru-Chun Amy F, Junzhong L, et at., Photochem Photobiol, 1998, 68(2), p141. 180. Chatgilialoglu C, Gilbert B C, and Norman R, J Chem Soc Perkin Trans 2, 1979, p770. 181. SriskandakumarT, Petzold H, Bruijnincx P, etal., JAm Chem Soc, (submitted 2009). 87 CD a x - a ‘1 z 0 .1 ‘1 CD co A1.1 ADF input file for ground state and time-dependent DFT calculations on CH3S . UNITI length Angstrom angle Degree END ATOMS 1 C xl yl zi 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 6 H x6 y6 z6 END GEOVAR xl —0.9323645165 yl 0.7538453976 zi —0.2554338150 x2 —1.358906052 y2 1.765267821 z2 —0.2847699841 x3 —0.1640953568 y3 0.7242029073 z3 0.5300502049 x4 —1.707025428 y4 0.1481577134E—0]. z4 —0.1134618763E—01 x5 —0.7743971035E—0l y5 0.3446657011 z5 —1.785980422 x6 —1.154233121 y6 0.4311598236 z6 —2.607633668 END XC GGA Becke Perdew END SYMMETRY NOSYM tol=0.001 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 50 END SCF iterations 50 converge 1.Oe—6 1.Oe—3 mixing 0.2 ishift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 4 7 7 89 A1FIT 10.0 Modi fyExcitat ions UseOccupied Al SubEnd End Fragments H t21.H S t21.S C t21.C End Al .2 ADF input file for ground state and time-dependent DFT calculations on CH3S’ radical. UNITS length Angstrom angle Degree END ATOMS 1 C xl yl zi 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 END GEOVAR xl —0.8955097192 yl 0.7461104821 zl —0.2655950654 x2 —1.354738363 y2 1.747210887 z2 —0.3146675968 x3 —0.1716129346 y3 0.7262054524 z3 0.5651110213 x4 —1.692308581 y4 0.1402182236E—01 z4 —0.5039963098E—01 x5 —0.1075872236 y5 0.3483459909 z5 —1.799000867 END XC GGA Becke Perdew END CHARGE 0 1 UNRESTRICTED SYMMETRY NOSYM tol=0.00l 90 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 50 END SCF iterations 50 converge l.0e-6 l.Oe—3 mixing 0.2 lshift 0.0 dils n=lO ok=0.5 cyc=5 cx=5.0 cxx=i0.0 END INTEGRATION 4 7 7 A1FIT 10.0 ModifyExcitat ions UseOccupied Al SubEnd End Fragments H t21.H S t2l.S C t21.C End A1.3 ADF input file for ground state and time-dependent DFT calculations on CH3SS radical. UNITS length Angstrom angle Degree END ATOMS 1 C xl yl zi 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 6 S x6 y6 z6 END GEOVAR xl —0.8166129615 yl 0.7670805715 zl —0.2042560278E—01 x2 —1.256955743 y2 1.771247152 z2 —0.9940988876E—01 x3 —0.7116845060E—01 y3 0.7416107139 z3 0.7891296540 x4 —1.608232053 91 y4 0.2820444077E—0l z4 0.1690125105 x5 O.2966549980E—01 y5 0.3608788083 z5 —1.548454032 x6 —1.314040181 y6 0.4088199419 z6 —2.953060041 END XC GGA Becke Perdew END CHARGE 0.0 1 UNRESTRICTED SYMMETRY NOSYM tol=0.001 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 40 END QTENS ESR END SCF iterations 50 converge 1.Oe—6 1.Oe-3 mixing 0.2 ishift 0.0 diis n=l0 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 4 7 7 A1FIT 10.0 Modi fyExcitations UseOccupied Al SubEnd End Fragments H t21.H S t21.S C t2l.C End 92 A1.4 ADF input file for ground state and time-dependent DFT calculations of CH32SNO. * The Molecule “$ADFBIN/adf” <<eor ModlfyExcitatlons UseOccupied Al SubEnd End UNITS END length Angstrom angle Degree ATOMS ‘C 2S 3N 40 5C 6H 7H 8H 9H 10 Fl END 0.093299330556 1.801870998890 2.564608078090 3.750776088010 —0.541415135994 —1.580659823740 0.010874680875 —0.548506825515 —0.475693154085 0.157575073270 0.376389670055 —0.249839100900 1.463353251410 1.529149793600 0.331366484092 0. 688033224536 0.971075028574 —0.689497465788 —0.217296736284 1.407803333340 —0.070212789684 —0.035836671374 —0.011760165370 —0.030834631742 —1. 458391795540 —1.411767390600 —2.158717739280 —1.862015876210 0.656435439622 0.310912848049 GUIBONDS 1 2 1 1.0 2 3 2 1.0 3 4 3 1.0 4 6 5 1.0 5 7 5 1.0 6 8 5 1.0 7 5 1 1.0 8 9 1 1.0 9 10 1 1.0 END CHARGE 0.0 SYMMETRY NOSYM tol=0.001 BASIS type TZ2P core None END XC GGA Becke Perdew END SAVE TAPE21 TAPE13 93 EXCITATION Davidson ONLYS ING lowest 10 END SCF iterations 50 converge 1.De—6 1.Oe—3 mixing 0.2 lshift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 6 6 6 A1FIT 10.0 eor mkdir tapes my —f TAPE* tapes 2>/dev/null * DENSE’ cp tapes/TAPE21 TAPE21 “$ADFBIN/densf” <<eor Density fit trans Potential coul trans eor my TAPE41 tapes/TAPE41 rm -f TAPE* my -f tapes/TAPE* A1.5 ADF input file for ground state and time-dependent DFT calculations of CH3SSCH * The Molecule “$ADFBIN/adf” <<eor TITLE Methyl Disulfide SE’ SK UNITS length Angstrom angle Degree END ATOMS 1 C xl yl zi 94 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 6 S x6 y6 z6 7 C x7 y7 z7 8 H x8 y8 z8 9 H x9 y9 z9 10 I-I xlO yb zlO END GEOVAR xl —0.4008420982 yl 0.3905396080 zi —0.2322155253 x2 —0.5126493230E—01 y2 0.4627550644 z2 —1.268344630 x3 —0.4137836966E—01 y3 —0.5403383933 z3 0.2217839358 x4 —1.494155479 y4 0.4149537687 z4 —0.2059453443 x5 0.2480197890 y5 1.801338869 z5 0.7591664997 x6 —1.313268316 y6 3.138545139 z6 0.7997379556 x7 —2.222534719 y7 2.734816620 z7 2.344050187 x8 —2.188392019 y8 1.654074759 z8 2.524988378 x9 —3.264016048 y9 3.045193457 z9 2.193798612 xlO —1.790446556 yb 3.276290285 zlO 3.194541593 END BASIS type TZ2P core None END XC GGA Becke Perdew END SYMMETRY NOSYM tol=0.001 SAVE TAE’E21 TAPE13 EXCITATION Davidson 95 lowest 20 END SCF iterations 50 converge l.Oe—6 l.Oe—3 mixing 0.2 lshift 0.0 diis n=lO ok=0.5 cyc=5 cx=5.0 cxx=l0.0 END INTEGRATION 4.0 7 7 A1FIT 10.0 ModifyExcitations Us eOccupied Al SubEnd End eor mkdir -p tapes my TAPE* tapes my logfile tapes * DENSF * my tapes/TAPE21 TAPE21 “$ADFBIN/densf” <<eor Grid Coarse Density fit trans Potential coul trans UNITS length Angstrom angle Degree END eor my TAPE* tapes cat logfile >> tapes/logfile rm —f logfile A16 ADF input file for ground state and time-dependent DFT calculations of Ia, p toluene sulfonyl chloride. “$ADFBIN/adf” <<eor TITLE TosCL single point and sk UNITS length Angstrom angle Degree 96 N < F F )- i- I- N ) N ) N ) H F - 0 0 w CO 0 ) D CO 0 ) 01 0) cO 0 f- H N ) N ) N ) I— i c o . 0 CO CO CO CO . J 0 ‘ . 0 CO 0 ) H N ) I— i c o O ) . 0 I— i N ) N ) ‘ . 0 ‘ . 0 ‘ . 0 CO CO CO — J - J J CO CO CO 01 Cr 1 01 ‘ U ) U ) 0 ) N ) N ) N ) F - F - I- L i C ) 0 I N ) I I 0) I 0 0) N ) I 0 I I I I N ) I CD N ) - 0 0 I- 0 CD 0 < . F - I- • I— i C ) • 0 . • 0 • F- 0 • I— i 0 • 0 . • N ). • - • CO • 0 1 . • C O • • — J • N ) 0 0 ) CO 0 . N ) I— i i- — J N ) 0 ) ‘ . 0 01 — 1 H CO CO 01 N ) N ) 0 CO - J 0 N ) I- CO — I N ) CO — J CO F - If ) — J 01 CO U ) CO F - CD — J C ) CO ‘ . 0 - 0 ‘ . 0 J ‘ . 0 N ) Cr 1 0 ’ N ) 0 ’ ‘ . 0 0 — 3 ‘ 0 . — 3 Cr 1 CO CC ) - )- CO U ) CO U ) I- ’ 01 CO 01 0 ) Cr 1 CO N ) N ) ‘ . 0 F- 0) 0 CO N ) F - 0 I— ’ CO U ) 0) 0 0 ) CO F ’ I- ’ J F ’ 0 0’ C ) U ) CD — ) 0 0 U ) CO ‘ . 0 — 3 3 U ) ‘ . 0 ‘ . 0 — 3 I— ’ CO ‘ . 0 — 1 F ’ U ) F ’ ‘ . 0 — - 1 — J 01 — 3 U ) F ’ CO ‘ . 0 I- ’ 01 0 ) ‘ . 0 F ’ N ) N ) 0 — 3 CO ‘ . 0 — 1 U ) 01 F ’ 1. 0 Cr 1 ‘ . o N ) - . J I- ’ CO 0 1 0 F ’ — ) — 3 — 1 CO F ’ F ’ 0. — 3 1- ’ 1. 0 F ’ U ) CO N ) 0 1 0 ) N ) F ’ CO 1. 0 CO CO 0 1 0 - . J CO CO 0. I- C ) U ) CO 01 N ) U ) CO CO CO F ’ U ) I- ’ 1. 0 CO — 3 0 0 ) CT ) F ’ F ’ - 1. 0 U ) CX ) 0 . U ) N ) CO 0 CO I- ’ 1. 0 1. 0 CO CO 0 F ’ CD CO F ’ I- ’ F ’ 1. 0 U ) CO F ’ — 3 C r1 0 U ) N ) U ) 1. 0 F ’ F ’ F ’ U ) — I CO 0. N ) CD — 1 U ) CO U ) 0. CT ) Li Li Li I I I 0 0 0 F ’ F ’ F ’ F ’ I- ’ F- ’ F ’ F ’ F ’ 1. 0 CO — 3 CO 01 0. U ) N ) F ’ — 3 CO U ’ 0. U ) N ) F ’ 0 L iL iO C )Q fl fl fl fl 0 0 0 C n L iL iL iL iL i I- ’ F ’ CO < F ’ F ’ I- ’ F ’ F ’ F ’ F ’ I— ’ N N N N N N N N N F ’ — 3 CO 01 0 . U ) N ) F ’ 0 1. 0 CO — ) CO 01 0. 0 ) N ) F ’ CO N F ’ CO F ’ 1. 0 N ) N ) ‘ CO CO CO F ’ Li Li i— ’ Z C O ID , < x x p x > x x x < X X X X X X ‘ 0 C O — J C O U ’U )N )F ’ F ’ F ’ F ’ F ’ I- ’ F ’ F ’ I- ’ — ) CO CT ) U ) N ) I- ’ 0 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < CD CO — 3 CO 01 1)S U ) N ) F ’ Li ID ‘- .3 0 (1) N N N N N N N N F— ’ F ’ F ’ F ’ F ’ F ’ F ’ I- ’ — 3 CO U ’ ‘ 0. 0) N ) F ’ 0 Co -.4 x13 1.870581833 y13 0.5053014545 z13 1.437374330 x14 0.2301057166E—01 yl4 —1.019022097 z14 0.7796858473 x15 —2.394325646 y15 —0.5434618682 z15 —0.7173188957 x16 —3.414582372 y16 0.2173973014 z16 —1.405283391 x17 —2.702866410 yl7 —1.411087197 z17 0.3984018889 x18 —1.503222788 yl8 —1.793094903 z18 —2.168454350 END BAS IS type TZ2P core None END XC GGA Becke Perdew END SYMMETRY NOSYM tol=0.001 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 40 END QTENS ESR END SCF iterations 50 converge 1.Oe-6 1.Oe—3 mixing 0.2 ishift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 4 7 7 A1FIT 10.0 Modi fyExcitations UseOccupied A2 98 SubEnd End A1.7 ADF input file for ground state and time-dependent DFT calculations on 2a, toluene sulfonic acid. TITLE TosOH UNITS END length Angstrom angle Degree Mcdi fyExcitations UseOccupied End Al SubEnd ATOMS 1 C 0.061224810000 2 C 1.081284720000 3 C 1.037479710000 4 C —0.046414630000 5 C —1.071099420000 6 C —1.023243530000 7 C 2.190910880000 8 H —0.093306460000 9 H —1.898206170000 10 H 1.788561680000 11 H 2.729706760000 12 H 2.862575740000 13 H 1.903204030000 14 H 0.103904940000 15 S —2.339174880000 16 0 —3.596823160000 17 0 —2.945342060000 18 0 —1.686522570000 19 H —2.371345840000 END 0.049058130000 0.923871870000 2.267291990000 2.741469300000 1.866526730000 0.523363600000 3.211706140000 3.785070040000 2.228471400000 4.170745050000 2.781108080000 3.339415560000 0.557946880000 —0. 991976200000 —0.631943340000 0.172111760000 —1.370258650000 —1.821075810000 —2.445797750000 0.471369870000 0.857320420000 0.472526260000 —0.280452020000 —0.668321440000 —0.275673950000 0.873950130000 —0.562655630000 —1.264700250000 1.165946210000 1.707014720000 0.040980180000 1.454616190000 0.751537950000 —0.736751500000 —1.548604770000 0.670000590000 —1.758960180000 —2.010584760000 GUIBONDS 1 1 2 1.5 2 2 3 1.5 3 3 4 1.5 4 4 5 1.5 5 5 6 1.5 6 6 1 1.5 7 3 7 1.0 8 4 8 1.0 9 5 9 1.0 10 7 10 1.0 11 7 11 1.0 12 7 12 1.0 13 2 13 1.0 14 1 14 1.0 15 6 15 1.0 99 16 15 16 1.0 17 15 17 1.0 18 15 18 1.0 19 18 19 1.0 END CHARGE 0 SYMMETRY NOSYM tol=0.001 BASIS type TZ2P core None END XC GGA Becke Perdew END SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END SCF iterations 50 converge 1.Oe—6 1.Oe—3 mixing 0.2 lshift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=l0.0 END INTEGRATION 4.0 7 7 A1FIT 10.0 A1.8 ADF input file for ground state and time-dependent DFT calculations on 3a, ethyl phenyl sulfone. TITLE TOSET LT UNITS length Angstrom angle Degree END ModifyExcitations UseOccupied Al SubEnd End ATOMS ZMATRIX 1 C 0 0 0 0.000000000000 0.000000000000 0.000000000000 2 S 1 0 0 1.816278885140 0.000000000000 0.000000000000 100 3C 4C 5C 6C 7C 8C 9H 10 H 11 H 12 H 13 H 14 0 15 0 16 H 17 H 18 C 19 H 20 H 21 H END GUIBONDS 1 3 5 1.5 2 5 6 1.5 3 6 7 1.5 4 7 8 1.5 5 8 4 1.5 6 4 3 1.5 7 9 5 1.0 8 10 6 1.0 9 11 7 1.0 10 12 8 1.0 11 13 4 1.0 12 2 3 1.0 13 14 2 2.0 14 15 2 2.0 15 1 18 1.0 16 1 16 1.0 17 1 17 1.0 18 18 19 1.0 19 18 20 1.0 20 18 21 1.0 21 1 2 1.0 END 106. 899856501000 124.006129744000 115.271556776000 119. 412787694000 120.251289779000 119.986133910000 119.857998085000 119.580933997000 120.034098705000 120.255700276000 121.239580671000 108.344152765000 106.519271414000 106.785596764000 100.583213043000 115.165721573000 110. 337319935000 109.571867685000 112.272659792000 0.000000000000 z4 180.131305356000 180.265759188000 359. 676685818000 0.103609510658 0.038623816230 179.663108224000 180.049979527000 180. 003485166000 0.304966833300 243.861518702000 230.512938086000 309.937580338000 248.295464965000 239.721362093000 54.089810678900 119.408268473000 120.160925607000 CHARGE 0.0 GEOVAR z4 270 360 END BASIS type TZ2P core None END XC GGA END 2 1 0 1.803204634640 3 2 1 1.397354088510 3 2 4 1.400777567820 5 3 2 1.393574909180 6 5 3 1.398925676020 7 6 5 1.395128365020 5 3 2 1.088838482220 6 5 3 1.089368547890 7 6 5 1.089516467600 8 7 6 1.089224660170 4 3 2 1.088369531690 2 1 3 1.459896803570 2 1 14 1.460016297510 1 2 3 1.097665587900 1 2 16 1.099819709170 1 2 17 1.525074627330 18 1 2 1.096505050370 18 1 19 1.098590747870 18 1 20 1.097122054340 SYMMETRY NOSYM tol=0.001 Bec]ce Perdew 101 GEOMETRY lineartransit 7 iterations 30 optim All Internal step rad=0.15 angle=l0.0 hessupd BFGS converge e=1.Oe—3 grad=1.Oe—2 rad=1.Oe—2 angle=0.5 END SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END SCF iterations 50 converge l.Oe—6 l.Oe—3 mixing 0.2 ishift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=i0.0 END INTEGRATION 4 7 7 A1FIT 10.0 A1.9 ADF input file for ground state and time-dependent DFT calculations on Ib, methane sulfonyl chloride. TITLE MESO2CL S—K edge UNITS length Angstrom angle Degree END ATOMS 1 C xl yl zl 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 6 0 x6 y6 z6 7 0 x7 y7 z7 8 Cl x8 y8 z8 END GEOVAR xl 0.1262384994 yl 0.3387232576 zi 0.1027941566 x2 —0.5022336713E—0l y2 —0.7203779344 z2 —0.1223752814 x3 1.136879736 y3 0.6417103710 102 z3 —0.1854172113 x4 —0.6408014542 y4 0.9725787169 z4 —0.3512016347 x5 —0.1444290617E—O1 y5 0.4759333329 z5 1.887679646 x6 1.092015671 y6 —0.2175129557 z6 2.505788982 x7 —1.384380827 y7 0.2318915910 z7 2.276125538 x8 0.3374298739 y8 2.534150464 z8 2.119821050 END BASIS type TZ2P core None END XC GGA Becke Perdew END SYMMETRY NOSYM tol=0.001 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END SCF iterations 50 converge 1.Oe—6 1.0e-3 mixing 0.2 lshift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 4 7 7 A1FIT 10.0 ModifyExcitations tJseOccupied A2 SubEnd End 103 A1.1O ADF input file for ground state and time-dependent DFT calculations on 2b, methane sulfonate. TITLE MESO2oh S k-edge Modi fyExcitations Us eOccupi ed Al SubEnd End UNITS END length Angstrom angle Degree ATOMS 1C 2H 3H 4H 5S 60 70 80 9H END 0.003598347641 —0.107523086699 0.929680712750 —0.871516986985 0.084634418404 1.353420691120 —1.176559210820 0.138569281350 1.029709048170 0.257529142080 —0.831662249905 0.578373563993 0.754746954421 0.670027403716 0.224627580814 0.317825950051 2.303618960260 2.562385026450 0.031130043395 —0.022514917387 —0. 456644480656 —0.400080180944 1.770385545970 2.316662331780 2.376036994920 1.690173254830 2.000187685390 GUIBONDS 1 1 2 1.0 2 1 3 1.0 3 1 4 1.0 4 1 5 1.0 5 5 6 2.0 6 5 7 2.0 7 5 8 1.0 8 8 9 1.0 END CHARGE 0.0 SYMMETRY NOSYM tol=0.00l BASIS type TZ2P core None END XC GGA END Becke Perdew SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END 104 SCF iterations 50 converge 1.Oe—6 1.Oe-3 mixing 0.2 ishift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.D END INTEGRATION 5 7 5 A1FIT 10.0 Al .11 ADF input file for ground state and time-dependent OFT calculations on model for 3b, dimethyl sulfone. TITLE MESO2ME UNITS length Angstrom angle Degree END ATOMS 1 C xl yl zi 2 H x2 y2 z2 3 H x3 y3 z3 4 H x4 y4 z4 5 S x5 y5 z5 6 C x6 y6 z6 7 H x7 y7 z7 8 H x8 y8 z8 9 H x9 y9 z9 10 0 xlO ylO zlO 11 0 xli yli zil END GEOVAR xi 0.2044050125E—01 yl —0.9974534674E—01 zi 0.1700119418E—01 x2 0.9474393788E—01 y2 —0.9907425984 z2 —0.6171230463 x3 0.2855590129 y3 —0.3163664456 z3 1.058590674 x4 —0.9905850021 y4 0.3205604226 z4 —0.3825563074E—01 x5 1.107101128 y5 1.186052739 z5 —0.6261950291 x6 2.747257732 y6 0.4456510507 z6 —0.5275515740 x7 3.428762148 y7 1.192506370 105 z7 —0.9527077002 x8 2.999641240 yS 0.2546875808 z8 0.5220608037 x9 2.770874092 y9 —0.4718597335 z9 —1.126793044 xlO 1.067924023 yb 2.309656506 zlO 0.3023560416 xli 0.7877457037 yli 1.372914110 zil —2.036584888 END BAS IS type TZ2P core None END XC GGA Becke Perdew END SYMMETRY NOSYM tol=0.001 SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END 5CR iterations 50 converge i.Oe—6 i.Oe—3 mixing 0.2 ishift 0.0 diis n=10 ok=D.5 cyc=5 cx=5.O cxx=i0.D END INTEGRATION 4 7 7 A1FIT 10.0 ModifyExcitations UseOccupied Al SubEnd End 106 A1.12 Example of ADF input file for ASCF calculations on Ia, p-toluene sulfonyl chloride. The same parameters and geometry was used as for the TO-OFT simulated XAS spectra (A1.6—A1.11). TITLE T0sCL single point and sk UNITS END length Angstrom angle Degree Occupations A 2 1.5 94 0.5 End 000000000 ATOMS 1 Cl 2S 3c 4C 5C 6C 7C 8c 9C 10 H 11 H 12 H 13 H 14 H 15 H 16 H 17 0 18 0 END GUIBONDS 1 4 5 1.5 2 5 6 1.5 3 6 7 1.5 4 7 8 1.5 5 8 3 1.5 6 3 4 1.5 7 6 9 1.0 8 7 10 1.0 9 8 11 1.0 10 9 12 1.0 11 9 13 1.0 12 9 14 1.0 13 5 15 1.0 14 4 16 1.0 15 3 2 1.0 16 2 17 2.0 17 2 18 2.0 18 2 1 1.0 END 0.000000000000 0.000000000000 106.496853068000 118.557279801000 117.874299902000 121.352353616000 118.587827208000 121.377518051000 120. 876021882000 119.608032709000 121.505638841000 111.392730897000 111.560459348000 110.467356737000 119.050343813000 120.544307863000 107.426265138000 107.675809576000 0.000000000000 0.000000000000 0.000000000000 270.000000000000 178.725610588000 359.074761896000 0.374721592606 359.630027239000 181.516526858000 179.311819726000 179. 485630372000 216. 941805491000 121.386500522000 119.548767898000 178.717365091000 357.343771086000 114.941385738000 129.994024837000 CHARGE 0.0 SYMMETRY NOSYM tol=0.001 ZMATRIX 0 0 0 0.000000000000 1 0 0 2.351742143770 2 1 0 1.789042803590 3 2 1 1.393041039510 4 3 2 1.394084742420 5 4 3 1.404468707820 6 5 4 1.405926490160 7 6 5 1.392476689840 6 5 4 1.507409500710 7 6 5 1.090577182850 8 7 6 1.087468068080 9 6 5 1.097114288210 9 6 12 1.096117323840 9 6 13 1.100510952110 5 4 3 1.090285259870 4 3 2 1.087382310450 2 1 3 1.456461412990 2 1 17 1.456298745440 107 BASIS type TZ2P core None END xc GGA Becke Perdew END SAVE TAPE21 TAPE13 SCF iterations 50 converge 1.Oe—6 1.Oe-3 mixing 0.2 lshift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 5 7 7 A1FIT 10.0 A1.13 Example of ADF input file for fragment analysis calculations on Ia, p-toluene sulfonyl chloride. The same geometry was used as for the TD-DFT simulated XAS spectra (A1.6 — A1.1 1) # dependency: /home/kennepohl/vlad/ t dependency: /home/kennepohl/vlad/fatoscl2b. so2 fatosci2b. so2 .t21 * dependency: /home/kennepohl/vlad/fatoscl2b.tolu fatoscl2b.tolu.t21 41’! /bin/sh 41’ The Molecule “$ADFBIN/adf” <<eor ModifyExcitations UseOccupied A2 SubEnd End TITLE TosCL single point and sk UNITS length Angstrom angle Degree END ATOMS 1 C —0.000530877574 0.051525045458 0.442496070463 f=tolu 2 C 1.028289251250 0.914455274938 0.810372700763 f=tolu 108 3C 4C 5C 6C 7C 8H 9H 10 H 11 H 12 H 13 H 14 H 15 S 16 0 17 0 18 Ci END 1.010443793250 —0 .074152325904 —1. 107046027750 —1.057859289750 2.138166441250 —0.115140542054 —1.937350825500 1.763416906250 2.719966015250 2.827258743250 1.858633166250 0.011061904906 —2.435499990920 —3.492687950070 —2.693599655520 —1.480078122500 2.269776252340 2.743820883340 1.895051224340 0.554468677538 3.191992867340 3.796277105340 2.260164784870 4.182428403340 2.785804026340 3.337761148340 0.529916568838 —0.994406982662 —0.570881729301 0.179634794787 —1.481187979180 —1.789108297530 0.441239067363 f=toiu —0.313417372337 t=tolu —0.708044550237 f=toiu —0.319776804537 f=tolu 0.830117228063 f=toiu —0.596035678637 f=toiu —1.281690789420 f=toiu 1.117787076060 f=tolu 1.665544860060 f=tolu —0.015277594327 f=toiu 1.402674269060 f=toiu 0.744985786363 f=toiu —0.675445186853 f=so2 —1.317476475650 f=so2 0.418872444354 f=so2 —2.181365072020 f=cl GUIB0NDS 1 1 2 1.5 2 2 3 1.5 3 3 4 1.5 4 4 5 1.5 5 5 6 1.5 6 6 1 1.5 7 3 7 1.0 8 4 8 1.0 9 5 9 1.0 10 7 10 1.0 11 7 11 1.0 12 7 12 1.0 13 2 13 1.0 14 1 14 1.0 15 6 15 1.0 16 15 16 2.0 17 15 17 2.0 18 15 18 1.0 END CHARGE 0.0 SYMMETRY NOSYM toi=0.001 BASIS type TZ2P core None END XC GGA Becke Perdew END Fragments ci fatosci2b. ci. t21 so2 fatosci2b. so2 . t21 tolu fatoscl2b.toiu.t21 end SAVE TAPE21 TAPE13 SCF 109 iterations 50 converge l.Oe—6 1.Oe—3 mixing 0.2 ishift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 4.0 4.0 4.0 A1FIT 10.0 eor A1.14 ADF input file for linear transit calculation of Ia, p-toluene sulfonyl chloride. TITLE TosCL single point and sk UNITS END length Angstrom angle Degree Modi fyExcitat ions UseOccupi ed A2 SubEnd End ATOMS ZMATRIX id 0 2S 1 3C 2 4C 3 5C 4 6C 5 7C 6 8C 7 9C 6 1011 7 11H 8 12H 9 1311 9 14 H 9 i5H 5 1611 4 170 2 180 2 END 0 0 0.000000000000 0 0 2.112212378510 1 0 1.776812417670 2 1 1.397124047540 3 2 1.392283251910 4 3 1.404803563300 5 4 1.403771133710 6 5 1.393918915470 5 4 1.507802400740 6 5 1.090512375050 7 6 1.087719808600 6 5 1.097339376130 6 12 1.096092809560 6 13 1.100358577080 4 3 1.090028523240 3 2 1.088856385520 1 3 1.446763840340 1 17 1.446652682370 0.000000000000 0.000000000000 101.791283537000 119.314039695000 118.703047905000 121.275067833000 118.481451871000 121.271228463000 120. 851499036000 119. 621769687000 121.346627364000 111.360452967000 111.525362136000 110.564242778000 119. 197779723000 120.359365501000 106.376924592000 105. 367852768000 0.000000000000 0.000000000000 0.000000000000 z4 182.041504214000 0.642281384535 359.979719236000 359.183344159000 180.770071385000 178.992673533000 179.882358502000 219.215075202000 121.260757225000 119. 670736484000 180.173793042000 1.708737756580 114.588104452000 130.961738470000 GUIBONDS 1 4 5 1.5 2 5 6 1.5 3 6 7 1.5 4 7 8 1.5 5 8 3 1.5 6 3 4 1.5 7 6 9 1.0 8 7 10 1.0 9 8 11 1.0 10 9 12 1.0 11 9 13 1.0 110 12 9 14 1.0 13 5 15 1.0 14 4 16 1.0 15 3 2 1.0 16 2 17 2.0 17 2 18 2.0 18 2 1 1.0 END CHARGE 0.0 GEOVAR z4 270 360 END SYMMETRY NOSYM tol=0.001 BASIS type TZ2P core None END XC GGA Becke Perdew END GEOMETRY smooth conservepoints lineartransit 7 iterations 30 optim All Internal step rad=0.15 angie=10.0 hessupd BFGS converge e=1.0e—3 grad=1.Oe—2 rad=1.Oe—2 angle=0.5 END SAVE TAPE21 TAPE13 EXCITATION Davidson lowest 20 END SCF iterations 50 converge 1.Oe-6 1.Oe-3 mixing 0.2 lshift 0.0 diis n=10 ok=0.5 cyc=5 cx=5.0 cxx=10.0 END INTEGRATION 3.0 7 7 A1FIT 10.0 111


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items