NEW INSIGHTS INTO THE ROLES OF METALS IN DELIVERY

A Dissertation submitted to the Faculty of the Graduate School of Arts and Sciences of Georgetown University in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Chemistry

By

Marie M. Melzer, B.S.

University of North Carolina at Charlotte

Washington , DC April 29, 2009

NEW INSIGHTS INTO THE ROLES OF METALS IN NITRIC OXIDE DELIVERY Marie M. Melzer, B.S.

Thesis Advisor: Timothy H. Warren, Ph.D.

Abstract

The biochemistry of organic compounds (E-NO, E = RS, RO, R2N, R) is connected to that of nitric oxide. These organic derivatives can serve either as sources of

NO in vivo or produce similar effects as NO such as vasodilation. In contrast to free NO, these organic derivatives are stable towards dioxygen. Release of NO from many organonitroso compounds requires a reducing equivalent, implicating redox active metalloenzymes. Thus in many cases the “stored” NO in its carrier molecules has to go through certain release-and-capture cycles to execute its various biological functions.

The transport of NO in biological systems is via S-nitrosothiols, part of sulfur-containing proteins.

Employing monovalent nickel and copper beta-diketiminates and tris(pyrazolyl)- borates as models for these metal ions in metalloenzymes, we find that the E-NO bonds of O-nitroso (RO-NO), N-nitroso compounds (R2N-NO), and S-nitroso (RS-NO) are readily cleaved to form well defined metal complexes [Ni](II)-E. In the case of the most biologically relevant Cu complexes, we observe an important contrast to all previously studied systems in which [Cu](II)-E species are reductively cleaved by NO to give the organonitroso compounds E-NO. The NO released in the nickel system binds with an extra equivalent of Ni(I) to form [Ni]-NO. For copper chemistry the fate of the NO

ii depends on the system used. Copper beta-diketiminate products are decomposed

and copper tris(pyrazolyl)borate nitrosyls [Cu](NO) are observed.

Copper(I) N-heterocylic carbene complexes are less electron-rich and do not activate the S-NO bond. Rather, we have shown that copper can have a role in the formation of the S-NO bond through trans-s- as well as addition to the copper(I) thiolate [Cu]-SR. This work has provided of insight into the biochemistry of

NO and the role of metals in E-NO bond cleavage and formation.

iii Acknowledgments

My experience in graduate school has been one of intense learning and growth. Not only have I gained a wide array of scientific skills, I walk away with analytical and problem solving skills that can only come through many years of scientific research.

There are many people who have walked alongside me through this journey. I am forever thankful and in debt to their support (and too often patience!).

It was an honor and privilege to work for Professor Timothy Warren the past five years. Not only is he a brilliant scientist, gifted teacher, and probably the most patient man I have ever met, he genuinely cares about the well-being and success of each of his students on a personal level. As a person who values family, I could not have asked to work for a more supportive boss when it came to time with family. Over the past five years, I have had several family commitments and emergencies, and Tim always gave me his full support in dealing with these issues. I aspire to be the kind of leader and mentor that Tim was to me.

I would like to thank the members of my committee, Professor Bahram Moasser,

Professor Steven Metallo, and Professor Toshiko Ichiye for their assistance and dedication to my Ph.D. journey. Your thoughts, comments, questions, and suggestions during my Phase 2A and Phase 2B were most appreciated and helpful in my graduate school experience. I would especially like to thank Bahram for his assistance with my kinetic questions.Kim Yearick Spangler, Dr. Yosra Badiei, and Matthew Varonka were my daily sanity throughout these past five years. The company of good friends and laughter helped me not to take life and chemistry too seriously. I could not have made iv this journey without you there and I will miss each of you tremendously. To Stefan

Wiese, thank you for being a patient glovebox mate with my mess and vials everywhere.

You were truly someone enjoyable to work alongside. To my undergraduate student

Ashley Bartell and Project SEED student Corwin Ward, thank you for your enthusiasm

and assistance on the project.

I appreciate assistance from Professor K. Travis Holman (X-ray structures) and Dr.

Ercheng Li (VT 1H and 15N high field NMR studies). The office staff: Kay Bayne, Inez

Traylor and Travis Hall have made the business side of graduate school much more amenable.

Finally, I would like to thank my family, Mom, Dad, Suzi and Rich Melzer, my sisters Jessica, Jocelyn, Allie, Gabrielle, and Miriam, my brother Chuck, my sister in- laws Sarah and Joanna, my brother in-law Tom, my nephews Brent and Michael, and my precious nieces Chloe and Montana. The experience of graduate school is not one walked alone. I appreciate everything that you have done for me along the way to encourage me to reach my goals.

Last, but most importantly, I would like to thank my husband Jonathan Melzer for his unconditional love and friendship. You are the most important thing in my life and I could not have made it without you. Thank you for being there for me when I was sad and frustrated, thank you for your support during the busy final months of completion, and thank you for making me feel like the most loved woman in the world. I hope to have many more life experiences with you by my side.

v Respective Contributions

Dr. Susanne Meyer from Professor Karsten Meyer’s research at the University of

Erlangen-Nürnberg performed in depth EPR analyses and simulation studies for several

II I compounds, [Me2NN]Cu (ON-Me2NN)2Cu (1), [Me2NN]Cu-NPh2 (2), [Me2NN]Cu-

15 NPh2 (2'), [Me2NN]Cu-I (3) (Figures 2.8, 2.13, and 2.14).

vi Table of Contents

Title……………………………………………………………..…………………………i

Abstract…………………………………..……….………………………………....…...ii

Acknowledgements……………………………………………..…………………….....iv

Respective Contributions…………………………………………….…………………vi

Table of Contents……………...………………………………..……………………....vii

General Introduction Organonitroso Compounds: N-nitrosoamines, O-nitrosoalcohols,

S-nitrosothiols; Biological Activities, Chemical Properties, and Interactions with

Metals…………………………………………………………………………...…………1

Abstract….………...... …………….....……………..…..……1

Introduction…...... …………….....……………….....……1

I.1.a. N- (R2NNOs) - biological applications..…...……..…...... ….... 3

I.1.b. N-Nitrosamines – syntheses….………………….....…………..…...... ……4

I.1.c. N-Nitrosamines - chemical properties….……………………....…….…..…5

I.2.a. O-Nitrosoalcohols (RONOs) - biological applications….…...……..…..…...6

I.2.b. O-Nitrosoalcohols – syntheses….…………….…….....……...... …..……...8

I.2.c. O-Nitrosoalcohols – chemical properties….………..….……...……..……..8

I.3.a. S-Nitrosothiols (RSNOs) - biological applications……………...... …..…...11

I.3.b. S-Nitrosothiols – syntheses….…...…………...... ……………...…..……...13

I.3.c. S-Nitrosothiols - chemical properties….….…………….....………..……..13

I.4.a. N-nitrosamines - complexes with metals….…...…….…...……...…..…….19

vii I.4.b. O-Nitrosoalcohols and S-Nitrosothiols - complexes with metals……...…..21

Chapter 1 Reductive Cleavage of O-, S-, N-Organonitroso Compounds by Nickel(I) β-

Dketiminates...…………………………………………………………………………...30

Abstract………………….……….…………………………………….……...... …..30

Introduction………………..…………………………………..……….……...... 30

Results and Discussion……………...……………………………….……………...33

1.1. Reaction of [Me3NN]Ni(2,4-lutidine) with CyONO or AdSNO…..…...... 33

1.2. Reaction of [Me3NN]Ni(2,4-lutidine) with Ph2NNO……………...…..……34

1.3. Independent synthesis of [Me3NN]Ni-NPh2 and reactivity with 2,4-

lutidine………………..…………………………………..……….……...... 35

1.4. Electronic properties of [Me3NN]Ni-NPh2. …………...……………...... 40

1.5. Mechanistic considerations…………………………………….……...... 44

Summary…………………………………………………………………...….……..46

Experimental Procedures…………………………………………………..………...53

References…………………………………………………………………...…...... 66

Chapter 2 A Three Coordinate Copper(II)-Amide from Reductive Cleavage of a

Nitrosamine……………….……………...………………………………………………69

Abstract..………………………………………………………….……………...... 69

Introduction………………..…………………………………..……….……...... 69

Results and Discussion……………...……………………………….……………...74

II I 2.1. Synthesis and characterization of [Me2NN]Cu (ON-Me2NN)2Cu ...... …...74

2.2. Isolation of [Me2NN]Cu-NPh2 from reaction mixture………………...... 78 viii 2.3. Independent synthesis of [Me2NN]Cu-NPh2..………………...... …………..79

2.4. Electronic properties of [Me2NN]Cu-NPh2.……...…………...... …………..86

2.5. Reaction of {[Me2AI]Cu}2 + Ph2NNO..……...…………...... ……………….88

t 2.6. Reaction of {[Me2AI]Cu}2 + xs BuONO..……...…………...... ……………89

2.7. Electronic differences between {[Me2NN]Cu}2 and {[Me2AI]Cu}2….…….92

Summary…………………………………………………………………...….……..94

Experimental Procedures…………………………………………………..………...97

References…………………………………………………………………...…...... 114

Chapter 3 Release of NO from S-nitrosothiols and Organic at a β-Diketiminato

Copper(I) Complex ………….……………...……………………………………….....118

Abstract..………………………………………………………….……………...... 118

Introduction..………………………………………………………………..…...... 118

Results and Discussion……………...……………………………….………….....125

t 3.1. Reaction of BuONO with {[Me2NN]Cu}2. …..…………….…………...... 125

t 3.2. Reaction of BuONO with {[Me2NN]Cu}2, flushed with N2(gas)…………127

3.3. Reaction of Ph3CSNO with {[Me2NN]Cu}2……………………………….131

3.4. Reaction of Ph3CSNO with {[Me2NN]Cu}2, flushed with N2(gas)……….132

3.5. Isolation of decomposition products in reaction of Ph3CSNO with

{[Me2NN]Cu}2………………………………………………………………….134

3.6. Mechanistic considerations - comparison with related nickel(I) system…..139

Summary…………………………………………………………………...….……141

Experimental Procedures…………………………………………………..…….....142 ix References…………………………………………………………………...…...... 153

Chapter 4 Reaction of S-nitrosothiols and Nitric Oxide with Tris(pyrazolyl)borate

CopperComplexes…….…….……………...…………………………………………...157

Abstract..………………………………………………………….……………...... 157

Introduction………………..…………………………………..……….……...... 157

Results and Discussion……………...……………………………….………….....166

4.1. Synthesis and stability of TpiPrCuII-SR complexes…………….………...... 166

iPr2 I 4.2. Reactivity of Tp Cu + RSSR in the presence of O2…………………...... 171

4.3. Reactivity of TpiPrCuII-SR complexes with S-nitrosothiols………………..174

iPr II 4.4. Reactivity of Tp Cu -SR complexes with NOgas…………………………178

iPr I 4.5.Tp Cu reactivity with Ph3SNO……………………………………………181

iPr 4.6. Tp Cu(NO) reactivity with C6F5SNO………………………………….....183

4.7. Synthesis of Tp(CF3)2Cu complexes and initial reactivity studies………...... 185

Summary…………………………………………………………………...….……189

Experimental Procedures…………………………………………………..…….....190

References..………………………………………………………………...…...... 206

Chapter 5 S-Nitrosothiol and Nitrosonium Reactivity at Two Coordinate Copper(I)

Thiolates …...…………………………………………………………………………...211

Abstract..………………………………………………………….……………...... 211

Introduction………………..…………………………………..……….……...... 211

x Results and Discussion……………...……………………………….………….....220

5.1. Synthesis and characterization of copper(I) thiolates IPrCu(SR)……….....220

5.2. Trans-s-nitrosation reactions between IPrCu(StBu) and BnSNO……….....226

5.3. Trans-s-nitrosation reactions between IPrCuStBu and tBuSNO…………...229

5.4. Trans-s-nitrosation reactions between IPrCuSBn and BnSNO……………230

5.5. Variable temperature NMR spectroscopy exploring degenerate exchange

tBu tBu between IPrCuSCH2Ar and ArCH2SNO…………………………………..232

+ 5.6. Reactivity of copper-thiolates IPrCu(SR) with NOgas and NO ……………240

5.7. Reactivity of the one coordinate copper cation {IPrCu}+ with RSNOs…...246

Summary…………………………………………………………………...….……251

Experimental Procedures…………………………………………………..…….....253

References..………………………………………………………………...…...... 273

Appendix Synthesis of Copper β-Diketiminate and Anilidoimine Complexes……...... 277

Introduction………………..…………………………………..……….……...... 277

Preparation of compounds……………...……………………………….…………279

A.1. General experimental procedure for β -diketimine (L1H-L3H)…...279

A.2. General experimental procedure for β -diketimine ligand L4H……………281

A.3. General experimental procedure for anilineimine ligands (L5H-L7H)...…282

A.4. General experimental procedure for anilineimine ligand L8H……………283

A.5. General experimental procedures to prepare lithium salts of β -diketimate

ligands (L3-L4) and anilidoimine ligands (L7-L8)……………………………..284

xi A.6. General procedure for the synthesis of LCu(MeCN) complexes L3Cu –

L8Cu……………………………………………………………………………287

I A.7. General procedure for [Cu ]2 complexes {L1Cu}2, {L2Cu}2, {L5Cu}2... ..289

References..………………………………………………………………...…...... 291

xii General Introduction

Organonitroso Compounds:

N-nitrosoamines, O-nitrosoalcohols, S-nitrosothiols.

Biological Activities, Chemical Properties, and Interactions with Metals

Abstract

N-nitrosamines, O-nitrosoalcohols, and S-nitrosothiols are three classes of related nitric oxide donors. The biological activities, syntheses, chemical properties as well as interactions with metals are reviewed for this important class compounds closely connected to nitric oxide itself.

Introduction

Nitric oxide, a simple diatomic gas, has much richer biological chemistry than initially expected. Just over twenty yeas ago this diatomic radical was thought to be simply a toxic gas—one of the constituents of acid rain, smog, and tobacco smoke. By

uterine relaxation cytostasis neurotransmission

gastrointestinal antiplatelet motility NO action

vasorelaxation immune function bronchodilation Figure I.1. Bioeffector functions of NO.

1 1988, however, it became widely accepted that nitric oxide was the important signaling

molecule in smooth muscle relaxation induction, thus serving as a potent vasodilator.1-4

Following this discovery, interest in the biology of nitric oxide grew at an extremely rapid pace which led to the identification of a rich collection of roles for this stable, simple inorganic radical. NO has been connected with immune function, hypertension, male impotence, septic shock, insulin-dependent diabetes mellitus, macrophage mediated destruction of oncogenic cells, and various nervous system activities (Figure I.1).5

The NO produced from the oxidation of L- to L-citruline by various isoforms of (NOS) has a lifetime of only 3-5 s in the blood.6 NO reacts with oxygen and oxygen containing to produce and nitrites

(Scheme I.1). Thus, a stable source for NO delivery is needed in the absence of nitric oxide synthase.

L-Arginine N-hydroxy-arginine L-Citrulline

Scheme I.1. Synthesis of NOgas in vivo from L-arginine to L-citrulline in the presence of O2, nitric oxide synthase (NOS), and NADPH.

2 The biochemistry of organic nitroso compounds (E-NO, E = RS, RO, R2N, R)

(Figure I.2) is connected to that of nitric oxide. These organic derivatives can serve either as sources of NO in vivo or produce similar effects as NO such as vasodilation. In contrast to free NO, these organic derivatives do not readily react with dioxygen.

Figure I.2. Family of organonitroso compounds.

I.1.a. N-Nitrosamines (R2NNOs) - biological applications.

N-nitrosamines with α-C-H atoms are generally considered to be carcinogenic.7,8

The established mechanism for carcinogensis is α-hydroxylation catalyzed by a variety of oxidzases and oxygenases such as cytochrome P450 and related enzymes. Decomposition

+ of the α-hydroxy- N-nitroso compounds produces powerful alkylating agents RN2 which

9 + can lead to DNA damage (Scheme I.2a). Spontaneous formation of the RN2 alkylating agent by R2NNO in vivo can also occur (Scheme I.2b). The release of NO by N-nitroso compounds in vivo can occur by denaturization and is carried out by the same cytochrome P450-related which accounts for 10-20% of total nitrosamine metabolism.10 After one-electron oxidation of the α-carbon atom, a highly unstable α-

• nitro amino radical is generated, which readily releases NO (Scheme I.2c).11

3 Scheme I.2. a) DNA alkylation pathway for R2NNO by cytochrome P450. b) DNA + alkylation pathway via spontaneous generation of RN2 alkylating agent. c) Denaturization by cytochrome P450 to form an unstable α-nitroso amino radical that readily releases NO.

N-nitroso-N-methylaniline has been shown to be an inhibitor of protein tyrosine phosphatases, papain, and caspase.10 The nitrosamine nitrosates the critical cysteine residues in the active site forming an S-nitrosothiol that results in enzyme inhibition. N- nitrosoindole can cause depurination, deamination, and the formation of a novel guanine analogue, oxaninie.12

I.1.b. N-Nitrosamines – syntheses.

N-nitrosamines can be synthesized by N-nitrosation of amines. Typical nitrosating agents

t 13-16 include N2O4, ClNO, RONO (R = Bu, Bu), and acidified (Scheme I.3a). N-

4 nitrosamine formation from their lithium amide precursors has been suggested as an

alternative route to the preparation of some nitrosamines (Scheme I.3b).17

Scheme I.3. Synthetic routes to N-nitrosamines.

I.1.c. N-Nitrosamines - chemical properties.

N-nitrosamines are known to take on planar structures due to a high rotational barrier about the N-N bond. These barriers are of similar magnitude to amides R2NC(O)R

‡ but the actual magnitude depends on the structure (Gc = 14-25 kcal/mol, Tc = 36-170

b)N-N Rotational Barriers F3C NN NN

F3C O O ~ 5 kcal/mol 25 kcal/mol N O N NN O

16.5 kcal/mol < 6 kcal/mol

Figure I.3. a) Resonance structures of R2NNO. b) Several representative structures of R2NNO compounds with different values for the rotational barrier about the N-N bond.

5 °C) (Figure I.3a).18 A rotational barrier as low as ~ 5 kcal/mol has been reported for N- nitrosohexafluorodimethylamine19 and < 6 kcal/mol for N-nitroso-N-methylaniline

(Figure I.3b).20 These two examples exhibit a lower barrier to rotation due to the electron withdrawing groups present in the molecule. Bicyclic N-nitrosamines also exhibit lower values for the barrier to rotation than found in typical aliphatic compounds.21

• N-nitrosamines are potential NO /NO+ donors through homolytic or heterolytic cleavage of the N-NO bond (Scheme I.4). Lower barriers to rotation typically correspond to a lower N-NO hemolytic bond dissociation energies. Thus, aryl and bicyclic nitrosamines have higher NO-releasing potentials than the alkylnitrosamines. Studies have shown that the homolytic cleavage of N-NO bonds to give NO radical is thermodynamically more favorable by 23-45 kcal/mol than the heterolytic cleavage which generates NO+,9,22 though this would depend on the ability to solvate the resulting

- + R2N and NO charged species.

Scheme I.4. Heterolytic and homolytic decomposition of N-nitrosamines. I.2.a. O-Nitrosoalcohols (RONOs) - biological applications.

Organic nitrites RONOs are esters between alcohols and nitrous acid, some of which have been used as clinical vasodilators for a long time (Figure I.4).18 From a study comparing organic nitrates such as and organic nitrites, the enzymes 6 responsible for vascular bioconversion of organic nitrates and organic nitrites to NO are

different.23 Organic nitrites are converted enzymatically to NO in vascular smooth muscle and the primary NO-generating activity is associated with the cytosol. In contrast, organic nitrates are metabolized by an enzyme associated with the plasma membrane.

Figure I.4. Examples of organic nitrites used as vasodilators in the clinical setting.

Organic nitrites can generate NO in vivo,24 requiring a one-electron reduction

(Scheme I.5, bottom). They also undergo rapid hydrolysis to give nitrite ion and the corresponding alcohol. The nitrite ion is not an active intermediate in vascular metabolism of nitrite esters but can be reduced to NO (Scheme I.5, top).10 S- nitrosothiols can be formed from organic nitrites in vivo via an enzymatic process in the vascular smooth muscle. Cytostolic GSTs can catalyze the reaction of RONO with R΄SH

Scheme I.5. NO• generation from organic nitrites.

7 to produce R΄SNO and play a significant role in the metabolism of organic nitrites in biological membranes (Scheme I.5, middle). Xanthine oxidase (XO) can catalyse the anaerobic reduction of organic nitrites to NO• (Scheme I.5, bottom).

I.2.b. O-Nitrosoalcohols – syntheses.

O-nitrosoalcohols can be synthesized by O-nitrosation of the corresponding alcohol.

Typical nitrosating agents include HNO2 (generated in situ from HNO3/H2SO4 or

2- 13,25 acidified NO ) or ClNO (Scheme I.6a). The reaction using HNO2 is the most common. We found that for the synthesis of bulky RONOs (e.g. 1-AdONO; 1-Ad = 1- adamantyl), the reaction of the lithium alkoxide LiOR with NOBF4 led to the desired product (Scheme I.6b).

Scheme I.6. Synthetic routes to O-nitrosalcohols.

1.2.c. O-Nitrosoalcohols – chemical properties.

The IR spectra of organic nitrites generally display two N=O stretching

-1 frequencies (νNO) in the 1610 - 1685 cm range due to presence of anti and syn isomers which are nearly equal in energy. The anti isomer occurs at a higher frequency compared to the syn, but the actual values depend on the R group attached to the oxygen atom

26 (Figure I.5). The presence of electron withdrawing groups shifts νNO to higher values.

8 Figure I.5. Organic nitrite syn/anti conformational preferences along with IR values for two comparative compounds. NMR spectroscopy can be used to assign anti and syn conformations in organic nitrites

and determine gas-phase and liquid-phase thermodynamic parameters (H°, S°, G°).

The syn conformers of the linear Cl-C4 are more stable in both gas and liquid phases. For the sterically hindered isopropyl, isobutyl, and neopentyl nitrites, however, the anti conformer predominates.27

The homolytic bond dissociation energy of the RO-NO bond is between 36-41 kcal/mol.28 Organic nitrites are both light and heat sensitive causing homolysis of the O-

Scheme I.7. Nitroso dimer formation from photolysis of a tertiary alkyl nitrite.

9 NO bond. The proceeding steps differ in photolysis and thermolysis. The shape of the

compound also plays a role in the final product formation. The thermal decomposition of

alkyl nitrites usually yields predominantly the alcohol and carbonyl analogs. Photolysis,

however can lead to several different products. Tertiary organic nitrites form a nitroso

dimmer via an alkoxy radical rearrangement (Scheme I.7). Straight chained nitrites can

undergo δ-H-abstraction (Barton rearrangement)29-31 to form nitroso-alcohol products

(Scheme I.8). Cyclic nitrites can ring open to form linear nitroso aldehyde dimers32 or form rearranged nitroso-alcohol cyclic products.33

Scheme I.8. Formation of a C-nitroso alcohol product from photochemical induced δ-H atom abstraction of an organic nitrite.

10 1.3.a. S-Nitrosothiols (RSNOs) - biological applications.

S-nitrosothiols are biological metabolites of NO and were previously less well- known than their oxygen counterparts due to their general thermal and chemical instability. Special interest in S-nitrosothiols, however, was invigorated around 1990 when it was found that it is an endogenous carrier of NO involved in a range of physiological functions.34 RSNOs have been proposed to be NO storage, transfer, and delivery vehicles. Many NO related biological functions such as anti-platelet aggregation, vasodilation, and pre-eclampsia35 have been directly associated with RSNOs, either in the release of NO or as the RSNO itself.36

S-nitrosoglutathione (GSNO) and S-nitrosocysteine (CySNO) are examples of endogenous RSNOs and are found in plasma at 0.02 - 0.2 and 0.2-0.3 M, respectively

Figure I.6. Examples of S-nitrosothiols with in vivo human plasma concentrations given for endogenous RSNOs. 11 (Figure I.6).37-39 Protein S-nitrosothiols that contain a primary cysteine residue such as S- nitrosoalbumin are significantly more stable than small S-nitrosothiol molecules.36 The exact reasons for this extra stability as well as the processes by which NO promotes glutathione modification through transnitrosation are not fully understood.

Trans-s-nitrosation is the transfer of the NO group from one SR group to another SR' group. This process is reversible and found to be first order in both the thiol and nitrosothiol.40 This process is important in the protein regulation through the modification of cysteine by S-nitrosation. The intermediate for this reaction is proposed to be a disulfide [RS(NO)SR']- (Scheme I.9) from ESI-MS, DFT calculations, and 15N NMR studies.41-44

Scheme I.9. Trans-s-nitrosation reaction between RS- and RSNO procedes through a nitroxyl disulfide intermediate [RS(NO)RS]-.

Copper-zinc superoxide dismutase (CuZnSOD) has been considered as a possible

Cu-enzyme that could catalyze the S-NO cleavage in vivo.45 CuZnSOD is an abundant enzyme in cells and is the major source of copper in human red blood cells46 present at a concentration of ca. 20 µM.47 The active site contains a Cu2+ ion bound to four histidine resiudes and has been identified as an active catalyst for low molecular weight RSNO decomposition.41,48 12 1.3.b. S-Nitrosothiols – syntheses.

S-nitrosothiols are generally synthesized from the reaction of thiol RSH with various nitrosating reagents XNO such as NOCl, N2O4, NO2, N2O3, HNO2, and R'ONO

(Scheme I.10a).13 S-nitrosation from reaction of RSNO and R'ONO is quantitative in

RSNO generation. As thiols often have a pungent odor and a few lower molecular weight ones (e.g. tBuSH) have an intense odor nearly indistinguishable from EtSH contained in natural gas, we have employed the use of non-volatile thallium thiolates TlSRs with

49,50 NOBF4 for the synthesis of a range of RSNOs (Scheme 1.10b).

Scheme I.10. Synthetic routes to S-nitrosthiols.

1.3.c. S-Nitrosothiols - chemical properties.

Most S-nitrosothiols, especially primary and secondary RSNOs, are unstable towards thermal loss of NO to form the corresponding disulfides RSSR. Such primary species such as S-nitroso-N-acetylcysteine do not have long shelf-lives and are usually

Figure I.7. S-nitrosothiols exhibit partial S=N bond character due to resonance which results in anti and syn isomeric forms. 13 Figure I.8. The syn and anti isomers of S-nitrosothiols can be independently characterized by 15N NMR and UV-Vis at 183 K and 298 K, Houk et al J. Am. Chem. Soc.2000, 122, 5889-5890. characterized only spectroscopically. In contrast, several tertiary RSNOs, including that

derived from N-acetylpenicillamine (SNAP), have been isolated and crystallized (Figure

I.9).

The S-NO bond exhibits significant double bond character due to the

delocalization of a sulfur lone pair onto the nitroso group (Figure I.7). This delocalization

allows the RSNOs to exist as syn and anti isomers. The structural preference for one

conformation over the other is dependent on the aliphatic substitution. Syn is preferred

for primary and secondary groups while anti is preferred for tertiary groups. This

suggests that the syn rotamer is favored on electronic grounds and the anti rotamer

minimizes steric interactions.

Variable temperature 15N NMR was used to study the syn / anti orientation equilibria for EtSNO, iPrSNO, tBuSNO (Figure I.8).51 All spectra from these three 14 compounds exhibited one broad peak in their 15N NMR spectra at room temperature.

When the temperature was lowered to -50 °C two sharp signals were resolved and were assigned to the syn and anti orientations of the RSNO structure. The primary and secondary RSNOs showed a preference for the upfield signal (syn orientation) and the tertiary RSNO showed a preference for the downfield signal (anti orientation). There are only seven crystal structures known for RSNOs including the structure S- nitrosohemoglobin.52 The SNO group in S-nitrosohemoglobin was modeled after the

SNO group in SNAP.53 This might need to be remodeled since nitroso group is on the primary cysteine residue and primary RSNOs favor the syn orientation.

The first reported high-resolution solid-state structure of an S-nitroso derivative of a cysteine or cysteinyl-containing compound reveals the geometry about the C-S-N=O group to be in the syn orientation with a dihedral angle of 2.4(3)° (Figure I.9).54 The X- ray crystal structure of S-nitrosocaptopril, a primary RSNO, reveals the geometry about the CSNO linkage to be in the syn orientation with a dihedral angle of 0.68°.51 The X-ray

55 crystal structure of SNAP (dihedral angle = 176.3°), ONSC(Me)2NHC(O) (dihedral

56 57 angle = 178.51°), and Ph3CSNO (dihedral angle = 175.7°) (Figure I.9) all reveal the geometry about the CSNO linkage to be in the anti orientation. A sterically encumbered, bowl-shaped triarylmethyl S-nitrosothiol, Ar3CSNO (Ar = 3,5-di(3,5- dimethylphenyl)phenyl) reveals both the anti and syn ortientation 0.67:0.33 in its X-ray crystal structure (Figure I.9).58

15 CySNO (1°) TritylSNO (3°) SYN ANTI

Ar3CSNO (3°) Ar3CSNO (3°) SYN (33 %) ANTI (67%)

Figure I.9. X-ray structures for the S-nitrosothiol of L-cysteine ethyl ester hydrochloride (EtCySNO), Ph3CSNO, and Ar3CSNO reveal the geometry about the RSNO bond.

Depending on the carbon substitution at the sulfur atom, RSNOs can be green

(tertiary) or pink (primary and secondary) in color. Their UV-Vis spectra typically show three bands – two intense bands in the UV and one weak band in the visible (Figure

I.8).51,52 Absorbances at wavelengths between 225-261 nm (ε ~ 104 cm-1M-1) are

* 3 -1 -1 attributed to the π → π transition, 330-350 nm (ε ~ 10 cm M ) are attributed to n0 →

* -1 -1 * π transition, and 550-600 nm (ε ~ 20 cm M ) are attributed to the forbidden nN → π

* transition. It is this low energy, weak nN → π transition determines the compound’s color.

Thermal and photochemical decompositions of RSNOs produce both the disulfide and NO. Homolytic cleavage was found to be generally ~29 kcal/mol lower in energy

16 than the corresponding heterolytic cleavage in acetonitrile (Figure I.10).59 Most S- nitrosothiols are unstable at room temperature and will decompose to give the disulfide and NO over time. Clean first-order kinetics were obtained for the decomposition of

RSNOs in the presence of a large excess of the corresponding thiol RSH.60 Measured enthalpies of activation ∆H≠ for a series of primary, seconday, and tertiary alkyl substituted RSNOs were only slightly lower than the theoretically calculated (DFT) homolytic bond dissociation energies of 30 – 33 kcal /mol.60 The use of a large excess of thiol to obtain reproducible kinetics suggests that there may be a bimolecular component to S-nitrosothiol decomposition. This is also consistent with the greater thermal stability of S-nitrosothiols with greater steric hinderance around the S-NO bond. Nonetheless, the measured activation parameters are inconsistent with a mechanism involving simple thermolysis to effect S-NO bond cleavage in biology – it is too slow. Moreover, little selectivity would be expected from purely thermally controlled S-NO cleavage.

Figure 1.10. Homolytic cleavage of the RS-NO bond is more facile than the heterolytic cleavage of the RS-NO bond.

Photolytic decomposition of RSNOs differ in aerobic vs. anaerobic environments.

In an anaerobic environment the S-NO bond homolytically cleaves, NO is released along with the thiyl radical, similar to thermal decomposition. In an aerobic environment the thiyl radical as well as NO can react with oxygen leading to more complex reaction

17 Scheme I.11. Photochemical decomposition of GSNO or RSNO in aerobic conditions yields disulfide and NO•.

pathways. When GSNO is irradiated by light at either 340 or 545 nm in the presence of

oxygen, GSOO• radical is first generated. This GSOO• radical can further react with another GSNO to form GSSG and NO (Scheme I.11).61

Under aerobic conditions, nitric oxide can react with dioxygen to form nitrous

62 anhydride (N2O3) (Scheme I.8). This molecule is believed to be the chain carrier involved in the chain-decomposition reaction of RSNOs.63 Kinetic data show that the rate of the chain reaction is independent of the concentration of RSNO and decreases with increasing bulkiness of the alkyl group. Thus, steric effects are prominent in the propagation step.

18 I.4.a. N-Nitrosamines - complexes with metals.

The interaction of nitrosamines with -containing biomolecules can result in

activation and/or denitrosation of the nitrosamines.7,8 In small molecule model complexes, nitrosamines may form intact R2NNO adducts with metal centers and have been shown to crystallize in three binding modes (Figure I.11). These intact R2NNO- metal complexes are formed by either attack of the nitrosamine on the metal or attack of an amide nucleophile on a metal-nitrosyl complex. The bonding mode for A has been shown for the reaction of Ph(Me)NNO with Na2PdCl4 (Figure I.11A). Adducts with O- bonding (Figure I.11B) mode have been isolated from the reaction of Me2NNO with the

52 following metal halide complexes: PdCl2, CuCl2, NiCl2, CoCl2, CdCl2, ZnBr2, AlCl3.

Adducts with N-bonding from the N=O nitrogen atom have been isolated and

- characterized from the reaction of [IrCl5] with a variety of typically unstable primary nitrosamines (Figure I.11D).64

A B C D

Figure I.11. Bonding modes in metal—nitrosamine complexes established by single- crystal X-ray crystallography. . In the crystal structure of the complex (Me2NNO)CuCl2 the Cu atoms are bridged by two chlorine atoms and the nitrosamine binding occurs via O-binding mode to one Cu and N-bonding mode to another (Figure I.11C).65,66 N-nitrosopiperidine complexes with

19 CuCl2 in an O,N-bonding mode where half of the Cu atoms are axially bonded to two N

atoms and the other half are axially bonded to two O atoms.67

Richter-Addo studied metalloporphyrin adducts of N-nitrosamines,

68 69 70,71 (porph)M(ONNR2) (M = Fe , Ru , Os ) (Scheme I.12). The first isolable metalloporphyrin complex was isolated from the reaction of excess Et2NNO and ferric porphyrin and displays O-bonding from the Et2NNO and Fe. The average N-O and N-N distances of the Et2NNO ligands (1.273 and 1.282 Å, respectively) suggest a significant contribution from the N=N resonance structure. Ruthenium and osmium carbonyl and nitrosyl nitrosamine complexes were also prepared and isolated, characterized by crystallography and various spectroscopic methods.

1 Scheme I.12. η -O binding of Et2NNO in M = Fe, Ru, O TTP or OEPs compounds.

20 I.4.b. O-Nitrosoalcohols and S-Nitrosothiols - complexes with metals.

Organic nitrites have long been used as nitrosating agents in coordination

chemistry in order to form metal-NO complexes as they are synthetically useful sources

+ of NO . Metal [M](RONO) complexes are less common than [M](R2NNO) complexes.

+ The reaction of alcohols with [IrCl3(NO)L2] leads to the formation of isolable

72 IrCl3(RONO)L2 (Scheme I.13a). Another example of an isolable M(RONO) complex

2+ 72 comes from the reaction of alkoxide ions with [Ru(bpy)2Cl(NO)] (Scheme I.13b).

Thus far, there is only one crystallographically characterized complex. In this X-ray structure, the organic nitrite is chelated to the Pd center through a metal-carbon bond

(Scheme I.13c).73

Alkyl nitrites and S-nitrosothiols both exhibit net trans addition of the RO-NO and RS-NO bonds to Ru and Os metalloporphyrins to give the corresponding

(por)M(NO)(SR) complexes.74,75 Mechanistic studies supported initial S-binding of the

RSNO to the metalloporphyrin center which then results in S-N bond cleavage to release

Scheme I.13. a) Reaction of ROH with [Ir](NO) leads to [Ir](RONO) b) Reaction of ROH with [Ru](NO) leads to [Ru](RONO) c) a [Pd](RONO) complex. 21 the stable NO radical which could then react with the 17-electron (por)M(CO)(SR)

intermediate to generate the observed nitrosyl product (Scheme I.13).76,77 The reaction of

S-nitroso-N-acetyl-L-cysteine methyl ester (RSNO) with ferrous (TPP)Fe(THF)2 only led to the generation of the known five-coordinate (TPP)Fe(NO) in almost quantitative

74 + yield, while the reaction of ferric [(TPP)Fe(THF2)] with isoamyl nitrite (RONO) resulted in the formation of the cationic nitrosyl alcohol complex [(TPP)Fe(NO)(HO-- i + 74 C5H11)] .

The first metal bound S-nitrosothiol was isolated from the reaction of benzylthiol

78 with the iridium nitrosyl K[IrCl5NO] (Scheme I.14). Benzyl-S-nitrosothiol is very unstable and readily dissociates to benzyl disulfide, BnSSBn and NO in air within 2 hr and 72 h anaerobically. When the benzyl-S-nitrosothiol is coordinated to the Ir center through the nitrogen, however, it is stable in acteonitrile and water over periods of weeks.

This work was further extended to include variety of coordinated nitrosothiols consisting of cysteine derivatives, mercaptosuccinic acid, benzyl thiol, and phenyl thiol.79 There were two structures possible in the case of the benzyl and phenyl-S-nitrosothiol, one

+ where all of the Cl atoms from the [Ir(Cl5)NO] remain coordinated to the Ir (Scheme

Scheme I.14. Reaction of RSH with [Ir](NO) to yield [Ir](N(O)SR). a) all other ligands are Cl- b) one Cl- is displaced by MeCN.

22 I.14A) and a second where one labile Cl- dissociates and is replaced by a coordinated

MeCN (Scheme I.14B). In the case of cysteine derivative thiols and mercaptosuccinic acid complexes form in A only.

Aim of Project

Understanding the mechanism by which E-NO compounds deliver NO is crucial for the effective treatment of diseases associated with disrupted regulation of NO and the future of NO delivery drugs. We have concentrated our efforts in understanding the reactivity of R2NNO, RONO, and RSNO with Ni and Cu model complexes. In particular, the copper chemistry is of high biological relevance due to its numerous roles in NO metabolism prescribed to copper ions and enzymes. Through the use of well defined model complexes, we aim to outline factors responsible for the cleavage and formation of

N-NO, O-NO, and S-NO bonds in this family of organonitroso compounds E-NO.

23 References

(1) Ignarro, L. J.; Buga, G. M.; S., W. K.; Byrns, R. E.; Chaudhuri, G. Proc. Natl. Acad.

Sci. U. S. A. 1987, 84, 9265.

(2) Palmer, R. M. J.; Ferrige, A. G.; S., M. Nature 1987, 327, 524.

(3) Marletta, M. A.; Yoon, P. S.; Iyengar, R.; Leaf, C. D.; Wishnok, J. S. Biochemistry

1988, 27.

(4) Moncada, S.; Radomski, M. W.; Palmer, R. M. J. Biochem. Pharmacol. 1988, 37,

2495.

(5) Butler, A.; Nicholson, R. Life, Death and Nitric Oixde; Royal Society of Chemistry,

2003.

(6) Rassaf, T.; Kleinbongard, P.; Preik, M.; Dejam, A.; Gharini, P.; Lauer, T.;

Erckenbrect, J.; Duschin, A.; Schulz, R.; Husch, G.; Feelisch, M.; Kelm, M. Clin. Res.

2002, 91, 470.

(7) Lijinsky, W. Chemistry and Biology of N-nitroso Compounds; Cambridge University

Press: Cambridge, 1992.

(8) Nitrosamines: Toxicology and Microbiology; VCH Ellis Horwood Ltd.: Chichester,

England, 1988.

(9) Cheng, J.-P.; Xian, M.; Wang, K.; Zhu, X.; Yin, Z.; Wang, P. G. J. Am. Chem. Soc.

1998, 120, 10266.

(10) Wang, P. G.; Xian, M. X.; Xiaoping, T.; Xuejun, W.; Cai, T.; Janzuk, A. J. Chem.

Rev. 2002, 102, 1091.

24 (11) Feelisch, M.; Samler, J. S. Methods in Nitric Oxide Research; Wiley & Sons: New

York, 1996.

(12) Yu, L.; McGill, A.; Ramirez, J.; Wang, P. G. Bioorg. Med. Chem. Lett. 1995, 5,

1003.

(13) Williams, D. L. H. Nitrosation; Cambridge University Press: Cambridge, U.K.,

1988.

(14) Fridman, A. L.; Mukhametshinm, F. M.; Novikov, S. S. Russ. Chem. Rev (Engl.

Trans) 1971, 40, 34.

(15) Saavedra, J. E. Org. Prep. Proced. Int. 1987, 29, 905.

(16) Werner, W.; Depreux, P. J. Chem. Res-S. 1980, 11, 372.

(17) Nudelman, N. S.; Bonatti, A. E. Syn. Lett 2000, 1825.

(18) Lee, J.; Chen, L.; West, A. H.; Richter-Addo, G. B. Chem. Rev. 2002, 102, 1019.

(19) Andreades, S. J. Org. Chem. 1967, 27, 4163.

(20) Looney, C. E.; Phillips, W. D.; Reilly, E. L. J. Amer. Chem. Soc. 1957, 79, 6136.

(21) Miura, M.; Sakamoto, S.; Yamaguchi, K.; Ohwada, T. Tetrahedron Lett. 2000, 41,

3637.

(22) Zhu, X. Q.; He, J. Q.; Li, Q.; Xian, M.; Lu, J.; Cheng, J.-P. J. Org. Chem. 2000, 65,

6729.

(23) Kowaluk, E. K.; Fung, H. L. Pharm. 1991, 259, 519.

(24) McMillan, G. R.; Calvert, J. G.; Thomas, S. S. J. Phys. Chem 1964, 68, 116.

(25) Noyes, W. A. Organic Syntheses; Wiley: New York, 1943; Vol. III.

25 (26) Rao, C. N. R.; Bhaskar, K. R. The Chemistry of Nitro and Nitroso Groups. Part 1;

Interscience: New York, 1969.

(27) Conboy, C. B.; Chauvel Jr., J. P.; Moreno, P. O.; True, N. S.; Ott, C. M. J. Phys.

Chem. 1986, 90, 4353.

(28) McMillen, D. F.; Golden, D. M. Ann. Rev. Phys. Chem. 1982, 33, 493.

(29) Kabasakalian, P.; Townley, E. R.; Yudis, M. D. J. Am. Chem. Soc. 1962, 84, 2716.

(30) Barton, D. H. R.; Beaton, J. M.; Geller, L. E.; Pechet, M. M. J. Am. Chem. Soc.

1960, 82, 2640.

(31) Hesse, R. H. Adv. Free Radical Chem. 1969, 3, 83.

(32) Gray, P.; Williams, A. Chem. Rev. 1959, 59, 239.

(33) Kabasakalian, P.; Townley, E. R. J. Org. Chem. 1962, 27, 2918.

(34) Feldman, P. L.; Griffith, O. W.; Stuehr, D. J. In Chem. Eng. News 1993; Vol. Dec

20, p 20.

(35) deBelder, M.; Lees, C.; Martin, J.; Moncada, S.; Campbell, S. Lancet 1995, 345,

124.

(36) Williams, D. L. H. Acc. Chem. Res. 1999, 32, 869.

(37) Kelm, M. Biochem. Biophys. Acta 1999, 1411, 273.

(38) Stamler, J. S. Cir. Res. 2004, 94, 414.

(39) Orie, N. N.; Vallance, P.; P., D.; Jones, D. P.; Moore, K. P. Am. J. Physiol. Heart

Circ. Physiol. 2005, 289, H916.

(40) Dicks, A.; Li, E.; Munro, A.; H., S.; Williams, D. Can. J. Chem. 1998, 76, 789.

(41) Hough, M. A.; Hasnain, S. S. Structure 2003, 11, 937. 26 (42) Singh, P. S.; Wishnok, J. S.; Keshive, M.; Deen, W. M.; S., T. Proc. Natl. Acad.

Sci. U. S. A. 1996, 93, 14428.

(43) Srivastava, S.; Dixit, B. L.; Ramana, K. V.; Chandra, A.; Chandra, D.; Zacarias, A.;

Petrash, J. M.; Bhatnagar, A.; Srivastava, S. K. Biochem. J. 2001, 358, 111.

(44) Perissinotti, L. L.; Adrian, G.; Turjanski, A. G.; Estrin, D. A.; Doctorovich, F. J.

Am. Chem. Soc. 2005, 127, 486.

(45) Jourd'heuil, D.; Laroux, F. S.; Miles, A. M.; Wink, D. A.; Grisham, M. B. Arch.

Biochem. Biophys. 1999, 361, 323.

(46) Romeo, A. A.; Capobianco, J. A.; English, A. M. J. Am. Chem. Soc. 2003, 125,

14370.

(47) Gartner, A.; Weser, U. FEBS Lett. 1983, 155, 15.

(48) Hough, M. A.; Hasnain, S. S. J. Mol. Bio. 1999, 287, 579.

(49) Varonka, M. S.; Warren, T. H. Inorg. Chim. Acta 2007, 360, 317.

(50) Varonka, M. S.; Warren, T. H. Inorg. Chem. 2009, Submitted.

(51) Bartberger, M. D.; Houk, K. N.; Powell, S. C.; Mannion, J. D.; Lo, K. Y.; Stamler,

J. S.; Toone, E. J. J. Am. Chem. Soc. 2000, 122, 5889.

(52) Lee, J. C., L.; West, A. H.; Richter-Addo, G. B. Chem. Rev. 2002, 102, 1019.

(53) Chan, N. L.; Rogers, P. H.; Arnone, A. Biochemistry 1998, 37, 16459.

(54) Yi, J.; Khan, M. A.; Lee, J.; Richter-Addo, G. B. Nitric Oxide 2005, 12, 261.

(55) Field, L.; Dilts, R. V.; Ravichandran, R.; Lenhert, P. G.; Carnahan, G. E. J. Chem.

Soc. Chem. Commun. 1978, 249.

27 (56) Lee, J.; Yi, G. B.; Powell, D. R.; Khan, M. A.; Richter-Addo, G. B. Can. J. Chem.

2001, 79, 830.

(57) Arulsamy, N.; Bohle, D. S.; Butt, J. A.; Irvine, G. J.; Jordan, P. A.; E. Sagan, E. J.

Am. Chem. Soc. 1999, 121, 7115.

(58) Goto, K.; Hino, Y.; Kawashima, T.; Kaminaga, M.; Yano, E.; Yamamoto, G.;

Takagi, N.; Nagase, S. Tetrahedron Lett. 2000, 41, 8479.

(59) Lu, J. M.; Wittbrodt, J. M.; Wang, K.; Wen, Z.; Schlegel, H. B.; Wang, P. G.;

Cheng, J. P. J. Am. Chem. Soc. 123, 123, 2903.

(60) Bartberger, M. D.; Mannion, J. D.; Powell, S. C.; Stamler, J. S.; Houk, K. N.;

Toone, E. J. J. Am. Chem. Soc. 2001, 123, 8868.

(61) Sexton, D. J.; Muruganandam, A.; McKenny, D. J.; Mutus, B. Photochem.

Photobiol. 1994, 59, 463.

(62) Ignarro, L. J. Nitric Oxide: Biology and Pathology; Academic Press: San Diego,

2000.

(63) Gross, L.; Montevecchi, P. C. Chem. Eurp. J. 2002, 8, 380.

(64) Di Salvo, F.; Estrin, D. A.; Leitus, G.; Doctorovich, F. Organometallics 2007, 27,

1985–1995.

(65) Klement, U. Acta Crystallogr. 1969, B25, 2460.

(66) Klement, U.; Schmidpeter, A. Angew. Chem. Int. Ed. Engl. 1968, 7, 470.

(67) Schmidpeter, A.; Noth, H. Inorg. Chim. Acta 1998, 269, 7.

(68) Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. J. Am. Chem. Soc. 1995, 117, 7850.

(69) Yi, G. B.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1996, 35, 3453. 28 (70) Chen, L.; Yi, G. B.; Wang, L. S.; Dharmawardana, U. R.; Dart, A. C.; Khan, M. A.;

Richter-Addo, G. B. Inorg. Chem. 1998, 37, 4677.

(71) Richter-Addo, G. B. Acc. Chem. Res. 1999, 32, 529.

(72) Reed, C. A.; Roper, W. R. J. Chem. Soc. Dalton Trans. 1972, 1243.

(73) Andrews, M. A.; Chang, T. C. T.; Cheng, C. W. F.; Emge, T. J.; Kelly, K. P.;

Koetzle, T. F. J. Am. Chem. Soc. 1984, 106, 5913.

(74) Yi, G.-B.; Chen, L.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1997, 36,

3876.

(75) Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. Chem. Commun. 1996, 2045.

(76) Kadish, K. M.; Adamian, V. A.; Caemelbecke, E. V.; Tan, Z. T., P.; Bianco, P.;

Boschi, T.; Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1996, 35, 1343.

(77) Chen, L.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1998, 37, 533.

(78) Perissinotti, L. L.; Estrin, D. A.; Leitus, G.; Doctorovich, F. J. Am. Chem. Soc.

2006, 128, 2512.

(79) Perissinotti, L. L.; Leitus, G.; Shimon, L.; Estrin, D.; Doctorovich, F. Inorg. Chem.

2008, 47, 4723.

29 Chapter 1

Reductive Cleavage of O-, S-, and N-Organonitroso Compounds by

Ni(I) β-Diketiminates

Abstract

An electron-rich nickel(I) β-diketiminate cleaves the E-NO bond of O-, S-, and N- organonitroso species to give the nickel-nitrosyl [Me3NN]Ni-NO along with dimeric nickel(II)-alkoxide or thiolate complexes {[Me3NN]Ni}2(µ-E)2 or the mononuclear nickel(II)-amide [Me3NN]Ni-NPh2. This diamagnetic, three coordinate amide exhibits temperature dependent NMR spectra due to a low lying triplet state.

Introduction

Stamler and others recognized S-nitrosothiols (RSNOs) in the early 1990s to serve as potent sources of NO in vivo.1-3 In contrast to O- and N-nitroso derivatives, S- nitrosothiols can thermally expel NO with concomitant formation of the corresponding disulfide (Scheme 1.1) due to the relative weakness of the RS-NO bond (31-32 kcal/mol)4 and strength of RS-SR bond (65-66 kcal/mol).5 This leads to relatively short lifetimes for

RSNOs, though they circulate at ca. 0.2 µM in human plasma.6,7

Organic nitrites (RONOs) and N-nitrosamines (R2NNOs) are much more thermally stable than S-nitrosothiols and typically require a reducing equivalent or photolysis8 to release NO (Scheme 1.1). This reflects their higher E-NO homolytic

t 9 10 dissociation energies (e.g. BuO-NO = 40.9(8) and Ph2N-NO = 87.7 kcal/mol ). As release of NO from many organonitroso compounds requires a reducing equivalent, it

30 Scheme 1.1. NO formation from O-, S-, and N-organonitroso compounds.

suggests a role for metalloenzymes in their metabolism. Each of these three classes of

organonitroso substances is known to interact with heme-containing biomolecules11-14 and copper ions are implicated in the catalytic decomposition of S-nitrosothiols.15

While N-nitroso compounds typically form intact E-NO adducts with metal complexes, O-nitroso and S-nitroso compounds commonly undergo E-NO bond cleavage serving as nitrosating agents in metal-nitrosyl synthesis.11 For instance, metalloporphyrin

16 17 18,19 adducts of N-nitrosamines (porph)M(ONNR2) (M = Fe , Ru , Os ) are known while

RONOs and RSNOs react to give a net trans-addition to form trans-(porph)M(NO)(E) (E

= OR or SR) species (Schemes I.9 and 1.2).20-26

Mechanistic studies of NO transfer from the stable S-nitrosothiol Ph3CSNO to a

a) b)

Scheme 1.2. a) Contribution of the dipolar resonance form of diethyl nitrosamine in its binding to ferric porphyrins. b) Trans addition of RENO compounds to porphyrin. 31 Co(II) porphyrin suggests that Co-NO bond formation proceeds via concerted homolytic

14 S-NO bond cleavage (Scheme 1.3). In the case of nickel, polymeric [(RS)Ni(NO)]x

27 species have been obtained upon the reaction of Ni(CO)4 with RSNOs.

Scheme 1.3. A cleavage of the Ph3CS-NO bond to form a [Co](NO) complex. S-NO bond cleavage is first thought to go through an N-bound RSNO-Co complex. . Employing the β-diketiminate framework, Pooja Kapoor in the Warren group has found that related C-nitroso compounds such as 3,5-dimethylnitrosobenzene react with

8 I 6 two equivalents of the d [Me2NN]Co (η -toluene) to undergo a four-electron reductive

a)

b)

6 Scheme 1.4. a) Activation of the ArN=O bond by 2 equiv. [Me2NN]Co(η -toluene). b) Synthesis of [NN]Ni(NO). . 32 III 28 cleavage of the ArN=O bond to give {[Me2NN]Co }2(µ-NAr)(µ-O) (Scheme 1.4a).

9 I Since the addition of NOgas to related d β-diketiminato Ni complexes results in the formation of stable [β-diketiminato]Ni(NO) species (Scheme 1.4b),29 the reducing ability of the [β-diketiminato]NiI fragment coupled with its low-coordination number was tested for clean 1-electron reactions with organonitroso compounds.

Results and Discussion

1.1. Reaction of [Me3NN]Ni(2,4-lutidine) with CyONO and AdSNO.

Careful addition of 1.0 equiv CyONO or AdSNO to 2.0 equiv [Me3NN]Ni(2,4- lutidine) (1) in toluene at room temperature results in the formation of green solutions.

Green crystals also immediately precipitate from the solution with CyONO whereas purple crystals form after standing at room temperature overnight with AdSNO. X-ray characterization of each solid substance identifies them as {[Me3NN]Ni}2(µ-OCy)2 (2)

(83% yield) and {[Me3NN]Ni}2(µ-SAd)2 (3) (88% yield), respectively (Scheme 1.5).

1 When these reactions are performed in benzene-d6, H NMR analysis shows that

29 [Me3NN]Ni(NO) (4) is formed in > 90% yield based on integration of its characteristic

Scheme 1.5. Activation of the E-NO bond by 2 equiv [Me3NN]Ni(2,4-lutidine) to form [Me3NN]Ni(NO) (4) and ½ equiv {[Me3NN]Ni}2(-E)2 (2 or 3). . 33 β-diketiminato backbone C-H resonance at δ 4.40 ppm against an internal standard.

Despite the steric bulk of the cyclohexyl alkoxide and adamantyl thiolate ligands, the X-ray structures of 2 and 3 consist of [Me3NN]Ni-E (E = OCy, SAd) dimers related by inversion symmetry (Figures 1.9 - 1.11). The asymmetric unit in the X-ray structure of 2 consists of two unique monomeric [Me3NN]Ni-OCy units. Both 2 and 3 show pseudo- tetrahedral coordination at nickel (twist angles between N-Ni-N / E-Ni-E planes:

81.1° and 83.0° for 2, 70.7° for 3), but have distinct Ni…Ni separations of 3.050(1) and

3.072(1) Å (for 2) and 3.559(1) Å (for 3). This is partially a result of the considerably shorter Ni-O bond distances spanning 1.955(2) - 1.994(2) Å in 2 relative to the longer Ni-

S distances of 2.318(1) and 2.303(8) Å in 3.

1.2. Reaction of [Me3NN]Ni(2,4-lutidine) with Ph2NNO.

In contrast to reactions with CyONO and AdSNO which produce very insoluble dimeric products, the reaction between 1.0 equiv Ph2NNO and 2.0 equiv [Me3NN]Ni(2,4- lutidine) (1) in benzene-d6 results in a deep blue-violet, homogeneous solution. We anticipated that [Me3NN]Ni-NPh2 (5) would form along with the nickel nitrosyl 4

1 (Scheme 1.6). H NMR analysis shows [Me3NN]Ni-NO (4) in > 90% yield along with

Scheme 1.6. Activation of the Ph2N-NO bond by 2 equiv [Me3NN]Ni(2,4-lutidine) to form [Me3NN]Ni(NO) (4) and [Me3NN]Ni-NPh2 (5).

34 broad resonances for 2,4-lutidine. Attempts at crystallization failed to produce X-ray

quality crystals of any new product.

1.3. Independent synthesis of [Me3NN]Ni-NPh2 and reactivity with 2,4-lutidine.

The Ni(II)-amide 5 can be synthesized independently by addition of LiNPh2 to

[Me3NN]NiI(2,4-lutidine) (prepared from addition of NiI2 and 2,4-lutidine to

[Me3NN]Tl) in THF (Scheme 1.7). Crystallization of the product from pentane provided the nickel-amide [Me3NN]Ni-NPh2 (5) as deep blue crystals in 59% yield. The X-ray structure of three-coordinate 5 exhibits short Ni-Namido (1.823(1) Å) and Ni-Nβ-dik

(1.826(1) and 1.833(3) Å) distances. The Namido atom is planar (sum of angles about N =

358.1(1)°) with one of the two N-aryl rings coplanar with the other twisted out of the Ni-

N-C plane by 65.9° (Figure 1.12).

Scheme 1.7. Independent synthesis of 5 via [Me3NN]NiI(2,4-lutidine) with LiNPh2.

Following the addition of Ph2NNO to 1 by UV-Vis spectroscopy in Et2O

-1 -1 demonstrates that [Me3NN]Ni-NPh2 (5) (λmax = 375 nm; ε = 6770 M cm , λ = 598 nm; ε

= 2210 M-1cm-1) is formed in the presence of 2,4-lutidine (Figure 1.1, Scheme 1.6). In

-1 -1 addition, [Me3NN]Ni(NO) (4) is observed at λmax = 445 nm (ε = 2000 M cm ).

[Me3NN]Ni-NPh2 (5) exhibits rich solution behavior. Despite the diamagnetic

30 1 nature of benzene-d6 solutions of 5 (µeff = 0.0 B.M. at RT by the Evans method), H

NMR spectra of 5 prepared from [Me3NN]Ni(I)(2,4-lutidine) invariably show broad 35 Reaction of 2 [Me3NN]Ni(2,4-lutidine) + Ph2NNO Et2O, 25 °C

2 [Me3NN]Ni(2,4-lutidine) + Ph2NNO

Pure Compound λmax(nm) Amax -1 -1 ε[Me (cm3NN]Ni(NPhM ) 2) 375 1.653 5800 399 1.535 4000 [Me3NN]Ni-NO 598 0.163 2200 [Me3NN]Ni-NPh2 + 2 2,4-lutidine Absorbance (scaled)

[Me3NN]Ni-NO + 2 2,4-lutidine

Wavelength (λ, nm)

Figure 1.1. UV-Vis studies in Et2O to confirm presence of [Me3NN]Ni-NPh2 (5) and [Me3NN]Ni(NO) (4) in reaction of [Me3NN]Ni(2,4-lutidine) and Ph2NNO.

peaks devoid of coupling information (Figure 1.2). We believe this to be due to trace

amounts of 2,4-lutidine present in samples of 5 prepared from [Me3NN]NiI(2,4-lutidine) which can lead to a rapid equilibrium between [Me3NN]Ni-NPh2 5 its 2,4-lutidine adduct

[Me3NN]Ni(NPh2)(2,4-lutidine) 6 (Figure 1.2).

To avoid the possibility of any exchange broadening in NMR spectra of 5 by 2,4- lutidine, we prepared 5 in a one-pot procedure in 53% yield under conditions excluding

Scheme 1.8. Synthesis of 5 under 2,4-lutidine free conditions. 36 µeff~ 2.8 B.M. µeff = 0.0 B.M. (predicted) (H.S.-d8)

* [Me3NN]Ni-NPh2 + 2,4-lutidine (1 equiv)

[Me3NN]Ni-NPh2 + 2,4-lutidine (0.1 equiv)

1 [Me3NN]Ni-NPh2 H NMR benzene-d6 0.01.02.03.04.05.06.07.08.09.0

1 Figure 1.2. H NMR spectra (300 MHz, benzene-d6) of [Me3NN]Ni-NPh2 with 2,4- lutidine present in the solution in varying amounts (0 eq., 0.01 eq., 1 eq.). (* indicates residual solvent peak).

this Lewis base (Scheme 1.8). Addition of incremental amounts of 2,4-lutidine up to 1

equiv per 5 results in increased broadening followed by complete disappearance of all 1H

NMR resonances attributed to 5 (Figure 1.2). Moreover, the effective magnetic moment of this solution increases to 1.3 B.M. (based on 5) when a full equivalent of 2,4-lutidine is present. The decrease in the UV-vis band of 5 at λ = 598 nm in the presence of an excess of 2,4-lutidine (Figure 1.3) also supports the formation of a Lewis base adduct.

Presumably this sterically encumbered adduct 6 would be tetrahedral with an expected

37 spin-only magnetic moment of 2.8 B.M. (Figure 1.2). A Job plot experiment was

conducted in order to unambiguously determine the stoichiometry of the 2,4-lutidine

adduct of 5. These results suggest that the 2,4-lutidine binds in a 1:1 ratio to 5 (Figure

1.4).

38 0.45

0.4

0.35

0.3

0.25

0.2 Absorbance 0.15 1. [Me NN]Ni-NPh 2. [Me3NN]Ni-NPh2 3 2 + 0.1 1 equiv. 2,4-lutidine

0.05 3. [Me NN]Ni-NPh + 10 equiv. 2,4-lutidine 0 3 2 350 400 450 500 550 600 650 700 750 800 Wavelength (λ, nm)

Figure 1.3. UV-Vis spectra in Et2O at room temperature for the solution studies of [Me3NN]Ni-NPh2 (5) with added 2,4-lutidine (0, 1, 10 equiv).

2 3 Job Plot 1.75 2 3 1.75 1.5 1.5 1.25

1 4 1.25 0.75 2 4 0.5 1 1 0.25 5

Absorbance (396 nm) 0 Absorbance 0 0.25 0.5 0.75 1 0.75 2 Mole Fraction of [Me3NN]Ni-NPh2

0.5 1

0.25 5 0 350 400 450 500 550 600 650 700 750 800 Wavelength (λ, nm)

Figure 1.4. Job plot of 2,4-lutidine with [Me3NN]Ni-NPh2 (5) to determine the stoichiometery of the binding interaction between 2,4-lutidine and 5. 39 1.4. Electronic properties of [Me3NN]Ni-NPh2.

1 Room temperature H NMR spectra of lutidine-free 5 in toluene-d8 show sharp signals expected for a diamagnetic species, but many appear at chemical shifts considerably different than in other related diamagnetic complexes such as

[Me3NN]Ni(NO) (4). For instance, the backbone C-H resonance in 5 occurs at δ 1.83 ppm whereas it appears at δ 4.40 ppm in 4. Moreover, this and many other chemical

m-H (β-diketiminate) o-Me p-Me m-NPh2 25 °C

10 °C

-10 °C

-30 °C

-50 °C

-70 °C * o-NPh2 p-NPh2 CH-backbone CH3-backbone 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 Figure 1.5. Variable temperature H NMR spectra of [Me3NN]Ni-NPh2 (300 MHz, toluene-d8). 40 Figure 1.6. d-orbital splitting diagram for the low spin (5 – L.S.) and high spin (5 – H.S.) states of 5. The two highest levels are the dxz and the dyz.

shifts in 5 such as the o- and p-NPh2 resonances are particularly temperature sensitive

(Figure 1.5). Upon cooling to – 70 °C, the backbone C-H resonance migrates to δ 4.39 ppm and the p-NPh2 signal moves from δ 5.67 (RT) to 7.03 ppm (Figure 1.5)

We attribute this behavior to contact shifts from a minute contribution of the high- spin state of [Me3NN]Ni-NPh2 (5 – H.S.) which increases with increasing temperature

(Figure 1.6). Both low-spin and high-spin three-coordinate β-diketiminato nickel-amides

31 32 have been isolated such as [Me3NN]Ni-N(Cp2Co)Ad) and LNi-N(TMS)2, respectively. While density functional theory (DFT) studies on [Me3NN]Ni-NPh2 suggest that the low-spin form is favored by 8 kcal/mol (Figure 1.7), the β-diketiminato central backbone as well as the o- and p-NPh2 C atoms bear significant electron density in the two “singly occupied” Kohn-Sham molecular orbitals (MOs) of 5 – H.S. (Figure

1.8).

41 [Me3NN]Ni-NPh2 S = 0 [Me3NN]Ni-NPh2 S = 1, +8.1 kcal/mol (5- L. S.) (5- H. S.) Ni-N1 1.793 Å N1-Ni-N3 121.3° Ni-N1 1.833 Å N1-Ni-N3 131.3° Ni-N2 1.790 Å N1-Ni-N3 144.6° Ni-N2 1.833 Å N1-Ni-N3 126.8° Ni-N3 1.785 Å N1-Ni-N3 94.4° Ni-N3 1.814 Å N1-Ni-N3 97.1°

Figure 1.7. DFT structures of low-spin (5 – L.S.) and high-spin (5 – H.S.) [Me3NN]Ni-NPh2. (ADF ZORA BP/TZ2P(+)). The high-spin S = 1 form is calculated to be 8.1 kcal/mol higher in electronic energy. DFT optimized distances and angles are also collected.

42 5 – H.S. HOMO-1 (spin α: -4.33 eV) 5 – H.S. HOMO (spin α: -3.72 eV)

5 – L.S. HOMO (-4.05 ev) 5 – L.S. LUMO (-3.08 eV)

Figure 1.8. Contour plots of frontier KS - molecular orbitals for 5 – H.S. (HOMO-1 and HOMO, top) and 5 – L.S. (HOMO and LUMO, bottom) and along with orbital energies.

43 1.5. Mechanistic considerations.

We propose three possible routes by which the E-NO bond of N-, O-, and S- nitroso compounds E-NO can be reductively cleaved by 2 equiv. [Me3NN]Ni(2,4- lutidine) to give [Me3NN]Ni(NO) and [Me3NN]Ni-E (or ½ {[Me3NN]Ni}2(µ-E)). While we are unsure of the exact binding mode of a putative [NiI](ENO) adduct, it is likely that the E-NO compound comes into contact with the Ni(I) center prior to E-NO bond cleavage. First, we considered the formation of [Ni](NO) which leaves the E• radical to combine with a second equiv of [Ni](2,4-lutidine) (Scheme 1.9, Mechanism A). In the second route, [Ni]-E would first form to release free NO• which is then trapped by

Mechanism A

[Me3NN]Ni(2,4lutidine) + ENO [Me3NN]Ni(NO) + E 2,4lutidine II [Me3NN]Ni(2,4lutidine) + E [Me3NN]Ni (E) 2,4lutidine Mechanism B II [Me3NN]Ni(2,4lutidine) + ENO [Me3NN]Ni (E) + NO 2,4lutidine

[Me3NN]Ni(2,4lutidine) + NO [Me3NN]Ni(NO) 2,4lutidine Mechanism C E [Me3NN]Ni(2,4lutidine) + ENO [Me3NN]Ni 2,4lutidine NO

[Me3NN]Ni(2,4lutidine)

[Me3NN]Ni(NO) + II [Me3NN]Ni (E)

Scheme 1.9. Three possible mechanisms for E-NO bond cleavage by [Me3NN]Ni(2,4- lutidine).

44 [Ni](2,4-lutidine) to form [Ni](NO) (Scheme 1.9, Mechanism B). The third mechanism

features a longer lived [Ni](E-NO) intermediate that requires an additional [Ni] fragment

to activate the E-NO bond, forming a strict 1:1 ratio of [Ni](NO) and [Ni]-E (Scheme 1.9,

Mechanism C).

The reactions presented in Sections 1.1 and 1.2 feature a 2:1 [Ni] : E-NO ratio.

Subsequently, we analyzed the amount of [Ni](NO) that forms when the stoichiometery

of Ni(I) : E-NO was varied (1:2, 1:1, and 2:1) by 1H NMR using naphthalene as an internal standard for these three organonitroso compounds (E = OCy, SAd, NPh2). When

1 equiv [Me3NN]Ni(2,4-lutidine) reacts with 1 equiv of E-NO, [Me3NN]Ni(NO) forms in

>80% yield (based on the moles of Ni). This suggests that mechanism B and C are not the correct route for this reaction. Mechanism B is ruled out as in the absence of an extra equivalent of Ni(I) the [Me3NN]Ni(NO) forms first, consumes the majority of the Ni(I), and leaves behind the E• (not detected as major product). Mechansim C is ruled out because a 1 : 1 ratio of [Ni] : E-NO, we would expect both [Me3NN]Ni(NO) and

[Me3NN]Ni(E) to form in 25 % yield with the other 50% of [Ni] unreacted. This leaves mechanism A as the best option for the cleavage of the E-NO bond by Ni(I). Further support of mechanism A is shown by the reaction of 2 equiv of E-NO (E = OCy, SAd,

NPh2) with 1 equiv of Ni(I)-lutidine: [Me3NN]Ni(NO) is formed in 88-92 % yield (based on the moles of Ni).

Scheme 1.10. Further support of Mechanism A where [Ni](NO) forms preferentially over [Ni]-E. In the presence of 1 equiv of E-NO the only observed nickel containing species is the [Ni](NO). 45 Summary

I The electron-rich Ni β-diketiminate [Me3NN]Ni(2,4-lutidine) reductively cleaves the E-NO bond of O-, S-, and N-organonitroso compounds to give the nickel nitrosyl

[Me3NN]Ni(NO) and corresponding [Me3NN]Ni-E species. This behavior contrasts with

33 II 34,35 II 36 the reductive observed by Ford involving Cu -OR and Cu -NR2 species which react with NO to give the corresponding O- and N-organonitroso species along with CuI complexes. Such reactivity at related β-diketiminato CuI complexes will be explored in Chapters 2 and 3.

46 N2΄ O1΄ N1 Ni΄ Ni N1΄ O1 N2

Figure 1.9. X-ray structure of one of the two independent molecules of {[Me3NN]Ni}2(µ-OCy)2 found in the solid-state structure of 2 (all H atoms omitted). Selected bond distances (Å) and angles (deg): Ni1–O1 1.994(2), Ni1–O1´ 1.963(2), Ni1– N1 1.971(2), Ni1–N2 1.957(2), N1–Ni1–N2 92.70(9), O1–Ni1–O1´ 77.38(8), N2–Ni1– O1´ 116.60(9), N1–Ni1–O1´ 137.33(9), N2–Ni1–O1 118.55(9), N1–Ni1–O1 116.64(9).

47 N4΄ N3 O2΄ Ni2΄ Ni2 O2 N3΄ N4

Figure 1.10. X-ray structure of one of the two independent molecules of {[Me3NN]Ni}2(µ-OCy)2 found in the solid-state structure of 2 (all H atoms omitted). Selected bond distances (Å) and angles (deg): Ni2–O2 1.955(2), Ni2–O2´ 1.972(2), Ni2– N3 1.964(2), Ni2–N4 1.974(2), N3–Ni2–N4 92.50(10), O2–Ni2–O2´ 77.06(9), N4–Ni2– O2´ 120.25(9), N3–Ni2–O2´ 115.01(9), N4–Ni1–O1´ 137.38(9), N3–Ni2–O2 116.78(9). 48 N2΄ S1΄ N1 Ni Ni ΄ N1΄ S1 N2

Figure 1.11. X-ray structure of {[Me3NN]Ni}2(µ-SAd)2 (3) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Ni1–S1 2.318(8), Ni1–S1´ 2.303(8), Ni1– N1 1.953(2), Ni1–N2 1.950(2), N1–Ni1–N2 93.81(9), S1–Ni1–S1´ 79.26(3), N2–Ni1– S1´ 108.76(6), N1–Ni1–S1´ 144.00(7), N2–Ni1–S1 123.22(7), N1–Ni1–S1 111.79(6).

49 N1 C24 N3 C30 Ni N2

Figure 1.12. X-ray structure of [Me3NN]Ni-NPh2 (5) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Ni1–N1 1.833(2), Ni1–N2 1.826 (2), Ni1–N3 1.823(1), N3–C24 1.421(1), N3–C24 1.406(2), N1–Ni1–N2 94.28(6), N3–Ni1–N1 121.08(7), N3–Ni1–N2 144.61(7), Ni–N3–C24 113.5(2), Ni–N3–C30 126.6(2).

50 N1

Ni N3

N2 I

Figure 1.13. X-ray structure of [Me3NN]Ni(I)(2,4-lutidine)•(Et2O) (all H atoms and ether molecule in the crystal lattice omitted). Selected bond distances (Å) and angles (deg): Ni1–N1 1.953(2), Ni1–N2 1.949 (3), Ni1–N3 2.046(3), Ni1–I1 2.569(6), N2–Ni1–N1 94.86(9), I1–Ni1–N3 116.04(8).

51 Table 1.1 Crystallographic parameters for 2, 3, 5, and 7. Compd. 2 3 5 7

Formula C58H80 N4Ni2O2 C33 H44 N2NiS C35H39N3Ni C30H38I N3Ni,

C4H8O Mol. Wt. 982.68 559.47 560.40 700.36 Temp.(K) 100(2) 173(2) 173(2) 173(2) Crystal Block Block Block Plate description Crystal color Green Purple Blue Green Crystal size 0.0096 0.0392 0.0370 0.0017 (mm3) System Monoclinic Monoclinic Monoclinic Monoclinic Space group P2(1)/c P2(1)/n P2(1)/n P2(1)/n a (Å) 13.0521(12) 15.202(3) 11.144(2) 15.691(4) b (Å) 19.2644(17) 12.297(3) 16.469(4) 8.4554(19) c (Å) 20.7729(19) 15.767(4) 16.532(4) 26.077(6) α (deg) 90 90 90 90 β (deg) 95.5150(10) 92.331(3) 91.421(3) 100.823(4) γ (deg) 90 90 90 90 Volume (Å3) 5199.0(8) 2945.1(12) 3033.2(11) 3398.0(13) Z 4 4 4 4 θ range (deg) 1.44-28.00 2.10-27.00 2.18-27.00 1.59-27.00 Measd reflns 46529 24338 25619 27864 Unique reflns 12405 6429 6619 7391 GOF of F2 1.099 1.024 0.944 1.076

R1 ( I > 2σ(I)) 0.0506 0.0453 0.0350 0.0431 wR2 (all data) 0.1527 0.1180 0.0912 0.1036 Largest diff. 1.660 and - 0.971 and - 0.490and – 1.009 and - peak and hole 0.676 0.565 0.287 0.744 (e-.Å-3) 52 Experimental Procedures

General experimental details.

All experiments were carried out in a dry nitrogen atmosphere using an MBraun

glovebox and/or standard Schlenk techniques. 4A molecular sieves were activated in

vacuo at 180 ºC for 24 h. Dry toluene and dichloromethane were purchased from Aldrich

and were stored over activated 4A molecular sieves under nitrogen. Diethyl ether and

tetrahydrofuran (THF) were first sparged with nitrogen and then dried by passage through

1 activated alumina columns. Pentane was first washed with conc. HNO3/H2SO4 to remove olefins, stored over CaCl2 and then distilled before use from sodium/benzophenone. All deuterated solvents were sparged with nitrogen, dried over activated 4A molecular sieves and stored under nitrogen. 1H and 13C NMR spectra were recorded on either a Varian 300 MHz or 400 MHz spectrometer (300 or 400 and 75.4 or

100.4 MHz, respectively). All NMR spectra, including Evans method data, were recorded at room temperature unless otherwise noted and were indirectly referenced to TMS using residual solvent signals as internal standards. GC-MS spectra were recorded on a Varian

Saturn 2100T, elemental analyses were performed on a Perkin-Elmer PE2400 microanalyzer in our laboratories, and UV-Vis spectra were recorded on either an Agilent

8432 or a Cary 50 spectrophotometer.

Cyclohexanol, diphenylamine, tert-butylnitrite, 1.6 M n-butyllithium, sodium metal, and elemental iodine were obtained from Acros, anhydrous NiCl2 from Strem, and

31 2,4-lutidine from Aldrich; all were used as received. [Me3NN]Ni(2,4-lutidine),

31 37,38 Tl[Me3NN], and AdSH were synthesized according to published procedures. 53 Preparation and characterization of compounds.

CyONO. CyONO was synthesized according to the reported procedure for synthesis of

RONO (R = ethyl, tert-butyl, isopropyl) from an acidic solution of sodum nitrite.39 CyOH

(10.020 g, 0.100 mol) in 18 M H2SO4 (2.78 mL, 0.050 mol) and H2O (2.00 mL, 0.111 mol) was added dropwise to a stirring solution of NaNO2 (10.500 g, 0.152 mol) in 50 mL

H2O at 0 ºC. The solution was stirred for one hour and then allowed to separate into two

layers. 20 mL of pentane was added to the reaction mixture and the organic layer was

separated from the water layer. The organic layer was washed twice with saturated

aqueous Na2CO3 and 25 mL of brine. The organic layer was dried over Na2SO4 and filtered through silica gel to remove any remaining alcohol. The pentane was removed in

1 vacuo to yield a yellow oil, 10.45 g, 81% yield. H NMR (benzene-d6, 400 MHz, 25 °C):

5.039 (m, 1, C(1)H), 1.652 (m, 2, C(2)H), 1.494-1.440 (m, 2, C(3)H), 1.374-1.307 (m, 2,

C(2)H), 1.293-1.221 (m, 1, C(4)H), 1.145-1.043 (m, 2, C(3)H), 1.028-0.995 (m, 1,

13 C(4)H). CNMR (benzene-d6, 100.4 MHz, 25 °C): δ 77.95 (C1), 32.32 (C2), 25.32 (C4),

-1 23.84 (C3). IR (cm ): νNOsyn 1599, νNOanti 1635. Organonitrites typically exhibit two bands for vNO, with the anti (trans) conformation at higher energy than the syn (cis) conformation.11,40

41 t Ph2NNO. BuONO (30.50 g, 0.300 mol) was added to Ph2NH (5.00 g, 0.030 mol) until all of the amine dissolved. The solution was stirred for 30 minutes and the tBuOH and excess tBuONO were removed in vacuo leaving a yellow/orange solid (5.524 g, 93%

1 yield). H NMR (benzene-d6, 400 MHz, 25 °C): δ 7.248 (m, 4, m-ArH), 7.029-6.899 (m,

54 13 2, p-ArH), 6.684 (m, 2, o-ArH). CcNMR (benzene-d6): δ 143.24, 137.40, 129.65,

129.39, 129.07, 127.82, 126.61, 119.78.

AdSNO.42 tBuONO (2.09 g, 20.3 mmol) was added to the stirring solution of AdSH

(1.00 g, 5.08 mmol) in 8 mL CH3CN and at -35 ºC. The solution was allowed to stir for

15 minutes. The solvent was removed in vacuo and the residue was extracted with ether

(2 × 5 mL) and filtered. The solution was concentrated to ca. 2 mL. The solution was left to crystallize at –35 °C overnight. The mother liquor was removed and the crystals were dried in vacuo to afford at total yield of 876 mg (76 %) green crystals. The product was

1 stored in the dark at -35 ºC to avoid disulfide formation. H NMR (benzene-d6, 400 MHz,

13 1 25 °C): δ 2.300 (d, 6, C(2)H2), 1.923 (m, 3, C(3)H), 1.503 (d, 6, C(4)H2). C{ H}NMR

(benzene-d6): δ 55.51 (C1), 43.55 (C2), 36.31 (C4), 30.25 (C3).

Synthesis of {[Me3NN]Ni}2(µµµ-OCy)2 (2). CyONO (19 mg, 0.150 mmol) in 3 mL toluene was added dropwise to [Me3NN]Ni(2,4-lutidine) (150 mg, 0.301 mmol) in 5 mL toluene.

The red solution turned green immediately and a precipitate formed after the solution sat at room temperature for 30 min. X-ray characterization of the green crystals formed allows their identification as [Me3NN]Ni2(µ-OCy)2 (63 mg, 83% yield). The product is too insoluble for further spectroscopic characterization. Anal. Calcd for C58H80N4Ni2O2:

C, 70.89; H, 8.21; N, 5.70. Found: C, 70.89; H, 8.55; N, 5.70. Single crystal X-ray unit cell data (at 100 K): a = 13.052(2) Å, b = 19.264(2) Å, c = 20.773(2) Å, α = γ = 90°, β =

95.515(2)°, V = 5199.0(8) Å3; space group P2(1)/c.

Synthesis of {[Me3NN]Ni}2(µµµ-SAd)2 (3). AdSNO (30 mg, 0.150 mmol) in 3 mL toluene

55 was added dropwise to a solution of [Me3NN]Ni(2,4-lutidine) (150 mg, 0.301 mmol) in 5 mL toluene. The red solution turned purple/green immediately and a purple precipitate formed after the solution sat at room temperature overnight. The resulting solution is green. X-ray characterization of the purple crystals formed allowed their identification as

-1- -1 [Me3NN]Ni2(µ-SAd)2 isolated in 88% yield (74 mg). UV-Vis (CH2Cl2, nm (M cm )):

657 (2540), 542 (3570), 375 (9430). Anal. Calcd for C66H88N4Ni2S2: C, 70.84; H, 7.93;

N, 5.01. Found: C, 70.66; H, 7.89; N, 4.81. Single crystal X-ray unit cell data (at 173 K): a = 15.202(3) Å, b = 12.297(3) Å, c = 15.767(4) Å, α = γ = 90°, β = 92.331(3)°, V =

2945.1(12) Å3; space group P2(1)/n.

Reaction of [Me3NN]Ni(2,4-lutidine) with Ph2NNO. Ph2NNO (19 mg, 0.150 mmol) in

3 mL toluene was added dropwise to [Me3NN]Ni(2,4-lutidine) (205 mg, 0.411 mmol) in

5 mL toluene. The red solution turned blue immediately. No precipitate formed during this reaction. An independent synthesis of [Me3NN]Ni-NPh2 (5) was performed in order to confirm the fate of the “NPh2” fragment. UV-Vis data indicates formation of

[Me3NN]Ni-NPh2 and [Me3NN]Ni(NO) in the presence of 2,4-lutidine (Figure 1.1).

29 General procedure for quantification of [Me3NN]Ni(NO) from reaction of

[Me3NN]Ni(2,4-lutidine) with organonitroso (E-NO) substrates. Naphthalene (0.038 g, 0.029 mmol) was dissolved in 1 mL benzene-d6. [Me3NN]Ni(2,4-lutidine) (0.146 g,

0.030 mmol) was dissolved in 2 mL benzene-d6. These solutions were analytically transferred to a 5 mL volumetric flask and diluted to 5 mL with benzene-d6 to make 5 mL of a 60 mM stock solution of [Me3NN]Ni(2,4-lutidine).

56 Reaction of [Me3NN]Ni(2,4-lutidine) with CyONO to form [Me3NN]Ni(NO). CyONO

(0.026 g, 0.200 mmol) was added to 2 mL of benzene-d6 in to make a 100 mM stock solution. 150 µL of the 100 mM CyONO stock solution (0.015 mmol) was added dropwise to 500 µL of [Me3NN]Ni(2,4-lutidine) (60 mM, 0.030 mmol) in benzene-d6.

1 Analysis of the solution by H NMR shows the formation of [Me3NN]Ni(NO) (4) based on its characteristic β-diketiminato backbone C-H resonance at δ 4.4 ppm. The nickel nitrosyl 4 is formed in 97% yield based on the integration of its backbone C-H resonance against the naphthalene standard.

Reaction of [Me3NN]Ni(2,4-lutidine) with AdSNO to form [Me3NN]Ni(NO). AdSNO

(0.040 g, 0.200 mmol) were added to 2 mL of benzene-d6 in to make a 100 mM stock solution. 150 µL of the 100 mM AdSNO stock solution (0.015 mmol) was added dropwise to 500 µL of [Me3NN]Ni(2,4-lutidine) (60 mM, 0.030 mmol) in benzene-d6.

1 Analysis of the solution by H NMR shows the formation of [Me3NN]Ni(NO) (4) based on its characteristic β-diketiminato backbone C-H resonance at δ 4.4 ppm. The nickel nitrosyl 4 is formed in 93% yield based on the integration of its backbone C-H resonance against the naphthalene standard.

Reaction of [Me3NN]Ni(2,4-lutidine) with Ph2NNO to form [Me3NN]Ni(NO).

Ph2NNO (0.040 g, 0.200 mmol) was added to 2 mL of benzene-d6 in to make a 100 mM stock solution. 150 µL of the 100 mM Ph2NNO stock solution (0.015 mmol) wase added dropwise to 500 µL of [Me3NN]Ni(2,4-lutidine) (100mM, 0.030 mmol) in benzene-d6.

1 Analysis of the solution by H NMR shows the formation of [Me3NN]Ni(NO) (4) based

57 on its characteristic β-diketiminato backbone C-H resonance at δ 4.4 ppm. No other

diamagnetic compound was seen in the spectra other than broad signals for 2,4-lutidine.

The nickel nitrosyl 4 is formed in 90 % yield based on the integration of its backbone C-

H resonance against the naphthalene standard.

[Me3NN]Ni(I)(2,4-lutidine). A solution of [Me3NN]Tl (3.00 g, 5.58 mmol) in 100 ml

THF was added to a stirring solution of NiI2 (1.74 g, 5.58 mmol) in 60 mL THF with 3 equiv. 2,4-lutidine (1.95 mL, 16.7 mmol). After addition the solution remained dark green/black in color. The solution was allowed to stir overnight and a yellow precipitate of TlI formed. The TlI was removed by filtration over Celite. The volume of the solution was reduced to 50 mL and it was allowed to crystallize overnight at -35 °C. The mother liquor was removed and the residue was washed with cold pentane (3 × 1 mL) to yield

1.922 g of green/black crystals. The volume of the mother liquor was reduced to 15 mL and placed back in the freezer to yield a second crop of crystals (0.698 g) for a combined yield of 75 % as a mono-THF solvate. For analysis the crystals were crushed into a fine power and dried in vacuo to give an elemental analysis reading free of THF. 1H NMR

(benzene-d6, 300 MHz, 25 °C): δ 93.5 (s, 1, 2,4-lutidine-ArH), 47.5 (s, 6, p-Ar- Me), 34.8

(s, 12, Ar-o-Me2), 32.7 (s, 4, m-Ar-H), -4.2 (br s, 3, 2,4-lutidine-Me), -11.3 (br s, 3, 2,4- lutidine-Me), -46.6 (s, 6, backbone-Me), -118.1 (s, 1, 2,4-lutidine-ArH). The backbone-H and one 2,4-lutidine-ArH were not observed. µ eff (benzene-d6) = 2.38 B.M.; UV-Vis

-1 -1 (CH2Cl2, nm (M cm )) 527 (562) and 676 (507); Anal. Calcd for C30H38N3NiI: C,

57.55; H, 6.07; N, 6.71. Found: C, 57.82; H, 6.04; N, 6.65.

58 X-ray quality crystals were obtained from crystallization of [Me3NN]Ni(I)(2,4-lutidine) in ether. One molecule of ether appears in the crystal lattice.

LiNPh2. n-Butyllithium (5.0 mL, 8.0 mmol, 1.6 M solution in hexanes) was added

dropwise to a solution of Ph2NH (1.00 g, 6.911 mmol) in 8 mL toluene at -35 °C. The solution was allowed to warm to room temperature for 1 hr after which a white precipitate had formed. The mother liquor was removed and the precipitate was washed with 3 mL of cold pentane and dried in vacuo affording 1.11 g of product in 92 % yield.

Synthesis of [Me3NN]Ni-NPh2 (5) from [Me3NN]Ni(I)(2,4-lutidine). LiNPh2 (0.067 g,

0.385 mmol) in 3 mL of THF was added to a stirring solution of [Me3NN]Ni(I)(2,4- lutidine) (0.242 g, 0.385 mmol) in 5 mL of THF at -35 °C. The dark solution immediately turned royal blue and was allowed to stir for 1 hr. The volatiles were removed in vacuo and the solid was extracted with pentane (3 × 5 mL). The pentane extracts were filtered through Celite and the filtrate was concentrated to 5 mL. The solution was left to crystallize overnight at -35 °C and afforded 127 mg (59% yield) of X- ray quality crystals of 5.

X-ray diffraction of one of these crystals shows the structure to be [Me3NN]Ni-NPh2 free

1 of 2,4-lutidine. The H NMR spectrum (benzene-d6) of these crystals, however, show trace amounts of 2,4-lutidine as very broad peaks. The 2,4-lutidine cannot be completely removed from the reaction mixture and its presence causes a change in solution spectra of

[Me3NN]Ni-NPh2. (See Figures 1.2, 1.3, 1.4).

59 Synthesis of [Me3NN]Ni-NPh2 (5) from NiI2 under 2,4-lutidine free conditions. In a one pot synthesis, n-butyllithium (1.90 mL, 3.00 mmol, 1.6 M solution in hexanes) was added dropwise to a solution of [Me3NN]H (1.00 g, 3.00 mmol) in 25 mL of THF at -35

°C. The solution was allowed to warm to room temperature and stir for 1 hr. Anhydrous

NiI2 (0.936 g, 3.00 mmol) was added as to the reaction mixture which caused an immediate color change from pale yellow to dark purple/black. The reaction was allowed to stir overnight. LiNPh2 (0.524 g, 3.00 mmol) in 8 mL of THF was then added to the solution which caused an immediate color change to royal blue. The solution was allowed to stir at RT for 1.5 h. The volatiles were removed in vacuo and the solid was extracted with pentane (3 × 10 mL). The pentane extracts were filtered through Celite and the

Plot of Chemical Shift vs 1/T for [Me3NN]Ni-NPh2 (4) (toluene-d8, 300 MHz) δ (ppm)

1/T (K-1)

Figure 1.14. Plot of chemical shift (ppm) vs. 1/T for 5 in toluene-d8. This plot shows that there is not a linear relationship between the chemical shift and 1/T and therefore does not follow Curie-Weiss behavior. 60 filtrate was concentrated to 8 mL. The solution was left to crystallize overnight at -35 °C

1 and afforded 0.894 g (53% yield) of product. H NMR (toluene-d8, 25 °C) δ 7.524 (t, 4,

NPh2-m-Ar-H), 7.054 (s, 4, m-Ar-H), 6.162 (d, 4, NPh2-o-Ar-H), 5.685 (t, 2, NPh2-p-Ar-

H), ), 3.389 (s, 12, o-Ar-H), 2.651 (s, 6, p-Ar-Me), 1.829 (s, 1, backbone-H), 0.138 (s, 3,

13 backbone-Me). C NMR (toluene-d8, 25 °C) δ 146.97, 143.78, 141.71, 138.68, 137.68,

-1 - 131.72, 129.51, 121.10, 118.23, 66.75, 24.73, 19.42, 16.95. UV-Vis (Et2O, nm, (cm M ):

598 (2210), 399 (4070), 375 (5760). Anal. Calcd for C35H39N3Ni: C, 75.01; H, 7.01; N,

7.50. Found: C, 74.90; H, 7.19; N, 7.45. µ eff (benzene-d6, R.T.) = 0.0 B.M.

Solution behavior of [Me3NN]Ni-NPh2 in the presence of 2,4-lutidine. A 38 mM solution of [Me3NN]Ni-NPh2 (0.060 g, 0.113 mmol, 1.5 mL) was prepared in benzene-d6.

A 93 mM solution of 2,4-lutidine (0.010 g, 0.093 mmol, 1 mL) was prepared in benzene-

1 d6. Three H NMR spectra are collected where the final concentration of the

[Me3NN]Ni(II) fragment is 19 mM: (1) [Me3NN]Ni-NPh2 in the absence of 2,4-lutidine,

(2) [Me3NN]Ni-NPh2 + 0.01 eq 2,4-lutidine, and (3) [Me3NN]Ni-NPh2 + 1 equiv. 2,4- lutidine. (See Figure 1.2).

Table 1.2. Measured solution magnetic moment for [Me3NN]Ni-NPh2 in benzene-d6 as a function of the amount of 2,4-lutidine present in the solution.

Spectrum [Me3NN]Ni-NPh2 2,4-lutidine Benzene-d6 µeff (B.M.) (38 mM) (93 mM) 1 0.500 mL 0 mL 0.5 mL 0.0 2 0.500 mL 0.020 mL 0.480 mL 0.0 3 0.500 mL 0.204 mL 0.296 mL 1.3

61 Solution UV-Vis studies with [Me3NN]Ni-NPh2 (5) and 2,4-lutidine. [Me3NN]Ni-

NPh2 (5) (0.007 g, 0.0125 mmol) was dissolved in 2 mL of Et2O. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with Et2O to make 10.00 mL of a 1.25 mM stock solution of [Me3NN]Ni-NPh2. 2,4-lutidine (20.0

µL, 0.173 µmol) was added to a 10.00 mL volumetric flask and Et2O was added to make

10.00 mL of a 17.3 mM stock solution. Three UV-Vis spectra were collected where the final concentration of the [Me3NN]Ni(II) fragment is 0.063 mM: (1) [Me3NN]Ni-NPh2 in the absence of 2,4-lutidine, (2) [Me3NN]Ni-NPh2 + 1 eq. 2,4-lutidine, and (3)

-1 [Me3NN]Ni-NPh2 + 10 eq. 2,4-lutidine. (See Figure 1.3). Keq ~ 8000 M (calculated assuming that [Me3NN]Ni(NPh2)(2,4-lutidine) has A598 nm = 0).

Table 1.3 Experimental setup to calculate Keq for 5 + 2,4-lutidine to give 6.

Spectrum [Me3NN]Ni-NPh2 (1.25 mM) 2,4-lutidine (17.3 mM) Et2O 1 0.150 mL 0.000 mL 2.850 mL 2 0.150 mL 0.011 mL 2.839 mL 3 0.150 mL 0.110 mL 2.740 mL

62 Estimation of Keq for 2,4-lutidine binding to [Me3NN]Ni-NPh2. In the absence of 2,4- lutidine, a 0.063 mM solution of [Me3NN]Ni-NPh2 (5) exhibits an absorbance at A590 nm =

0.138. Upon addition of 1 equiv 2,4-lutidine (0.063 mM initial concentration), the absorbance A598 nm = 0.101. Assuming that the absorbance of [Me3NN]Ni(NPh2)(2,4- lutidine) at 598 nm is 0, we can assign the new concentration of [Me3NN]Ni-NPh2 (5) as

(0.101/0.138)(0.063 mM) = 0.046 mM. Given the 1:1 adduct (see Job’s plot in Figure

1.4), [2,4-lutidine] = 0.046 mM and [[Me3NN]Ni(NPh2)(2,4-lutidine)] = 0.017 mM

(0.063 – 0.046 mM). These final concentrations can be used to estimate the equilibrium constant for binding:

-5 [[Me3NN]Ni(NPh2)(2,4-lutidine)] (1.7×10 M) -1 Keq = = - 5 -5 ~ 8000 M [[Me3NN]Ni-NPh2][2,4-lutidine] (4.6×10 M)(4.6×10 M)

43 Job Plot for the binding of 2,4-lutidine to [Me3NN]Ni-NPh2 to confirm 1:1 stochiometry of adduct. A 0.400 mM stock solution of [Me3NN]Ni-NPh2 (5) in ether was prepared by diluting 3.2 mL of a 1.25 mM stock solution of 5 in ether to 10.0 mL using a volumetric flask. A 0.400 mM stock solution of 2,4-lutidine in ether was prepared by diluting 0.230 mL of a 17.3 mM stock solution of 2,4-lutidine in ether to

10.0 mL using a volumetric flask. To construct the Job plot, the mole fractions of 5 and

2,4-lutidine were varied while keeping constant the total molar concentration of these two were species at 0.400 mM. The UV-vis absorbance at 396 nm (a peak which grows as

[Me3NN]Ni(NPh2)(2,4-lutidine) (6) is formed) was plotted against the mole fraction of

63 [Me3NN]Ni-NPh2 (5) in the solution (Figure 1.4). A 1:1 stochiometery was determined for the binding of 2,4-lutidine to [Me3NN]Ni-NPh2.

Table 1.4 Experimental setup for Job plot analysis of 5 with 2,4-lutidine.

Spectrum Mole Fraction [Me3NN]Ni-NPh2 2,4-lutidine A400 nm 0.400 mM 0.400 mM 1 1.00:0.00 3.000 mL 0.000 mL 0.524 2 0.75:0.25 2.250 mL 0.750 mL 0.659 3 0.50:0.50 1.500 mL 1.500 mL 1.791 4 0.25:0.75 0.750 mL 2.250 mL 1.047 5 0.00:1.00 0.000 mL 3.000 mL 0.042

DFT calculation details

The DFT calculations employed the Becke-Perdew exchange correlation functional [1] using the Amsterdam Density Functional suite of programs (ADF 2007.01)

[2]. Slater-type orbital (STO) basis sets employed for H, C, and N atoms were of triple-ζ quality augmented with two polarization functions (ZORA/TZ2P) while an improved triple-ζ basis set with two polarization functions (ZORA/TZ2P+) was employed for the

Ni atom. Scalar relativistic effects were included by virtue of the zero order regular approximation (ZORA) [3]. The 1s electrons of C and N as well as the 1s – 2p electrons of Ni were treated as frozen core. The VWN (Vosko, Wilk, and Nusair) functional was used for LDA (local density approximation) [4]. Default convergence (∆E = 1 × 10-3

-2 -2 hartree, max. gradient = 1 × 10 hartree / Å, max. Cartesian step = 1 × 10 Å) and integration (4 significant digits) parameters were employed for geometry optimizations.

Experimental X-ray coordinates for [Me3NN]Ni-NPh2 (5) were used as the starting point for the geometry optimization of low-spin [Me3NN]Ni-NPh2 (5 – L.S.) in a 64 restricted (S = 0) calculation. The geometry optimization for high-spin [Me3NN]Ni-

NPh2 (5 – H.S.) employed the DFT optimized coordinates for 5 – L.S. as a starting point in an unrestricted (S = 1) calculation specifying 2 unpaired electrons (spin α – spin β).

ADFview [2a] was used to prepare the three-dimensional representations of the structures shown in Figure 1.7 as well as to render the frontier Kohn-Sham MOs shown in Figure

1.8.

DFT References

[1] (a) Becke, A. Phys. Rev. A 1988, 38, 3098. (b) Perdew, J. P. Phys. Rev. B 1986,

34, 7406. (c) Perdew, J. P. Phys. Rev. B 1986, 33, 8822.

[2] (a) http://www.scm.com – last accessed July 25, 2008. (b) te Velde, G.;

Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; Van Gisbergen, S. J. A.;

Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931. (c) Fonseca Guerra, C.;

Snijders, J. G.; te Velde, G.; Baerends, E. J.; Acc., T. C. Theor. Chem. Acc. 1998, 99,

391.

[3] (a) Snijders, J. G.; Baerends, E. J.; Ros, P. Mol. Phys. 1979, 38, 1909. (b) Ziegler,

T.; Tschinke, V.; Baerends, E. J.; Snijders, J. G.; Ravenek, W. K. J. Phys. Chem. 1989,

93, 3050. (c) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99,

4597.

[4] Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.

65 References

(1) Stamler, J. S.; Simon, D. I.; Osborne, J. A.; Mullins, M. E.; Jaraki, O.; Michel, T.;

Singel, D. J.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 444.

(2) Stamler, J. S.; Jaraki, O.; Osborne, J.; Simon, D. I.; Keaney, J.; Vita, J.; Singel, D.;

Valeri, C. R.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 7674.

(3) Stamler, J. S.; Simon, D. I.; Jaraki, O.; Osborne, J. A.; Francis, S.; Mullins, M.;

Singel, D.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 8087.

(4) Bartberger, M. D.; Mannion, J. D.; Powell, S. C.; Stamler, J. S.; Houk, K. N.; Toone,

E. J. J. Am. Chem. Soc. 2001, 123, 8868.

(5) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of Organic

Compounds; 2nd ed.; Chapman and Hall: New York, 1986.

(6) Stamler, J. S. Cir. Res. 2004, 94, 414.

(7) Kelm, M. Biochem. Biophys. Acta 1999, 1411, 273.

(8) Wang, P. G.; Xian, M. X.; Xiaoping, T.; Xuejun, W.; Cai, T.; Janzuk, A. J. Chem.

Rev. 2002, 102, 1091.

(9) Batt, L.; Christic, K.; Milne, R. T.; Summers, A. J. Int. J. Chem. Kinet 1974, 6, 877.

(10) Zhu, X. Q.; He, J. Q.; Li, Q.; Xian, M.; Lu, J.; Cheng, J.-P. J. Org. Chem. 2000, 65,

6729.

(11) Lee, J.; Chen, L.; West, A. H.; Richter-Addo, G. B. Chem. Rev. 2002, 102, 1019.

(12) Hung, C. H.; Ching, W. M.; Chang, G. F.; Chuang, C. H.; Chu, H. W.; Lee, W. Z.

Inorg. Chem. 2007, 46, 10941.

66 (13) Zhu, X. Q.; Zhang, J. Y.; Mei, L. R.; Cheng, J.-P. Org. Lett. 2006, 8, 3065.

(14) Zhu, X. Q.; Zhang, J. Y.; Cheng, J.-P. Inorg. Chem. 2007, 46, 592.

(15) Williams, D. L. H. Acc. Chem. Res. 1999, 32, 869.

(16) Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. J. Am. Chem. Soc. 1995, 117, 7850.

(17) Yi, G. B.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1996, 35, 3453.

(18) Chen, L.; Yi, G. B.; Wang, L. S.; Dharmawardana, U. R.; Dart, A. C.; Khan, M. A.;

Richter-Addo, G. B. Inorg. Chem. 1998, 37, 4677.

(19) Richter-Addo, G. B. Acc. Chem. Res. 1999, 32, 529.

(20) Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. Chem. Commun. 1996, 2045.

(21) Yi, G.-B.; Chen, L.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1997, 36,

3876.

(22) Yi, G. B.; Khan, M. A.; Powell, D. R.; Richter-Addo, G. B. Inorg. Chem. 1998, 37,

208.

(23) Chen, L.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1998, 37, 533.

(24) Lee, J.; Yi, G. B.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1999, 38, 4578.

(25) Lee, J.; Yi, G. B.; Powell, D. R.; Khan, M. A.; Richter-Addo, G. B. Can. J. Chem.

2001, 79, 830.

(26) Andreasen, L. V.; Lorkovic, I. M.; Richter-Addo, G. B.; Ford, P. C. Nitric Oxide

2002, 6, 228.

(27) Bladon, P.; Dekker, M.; Knox, G. R.; Willison, D.; Jaffari, G. A.; Doedens, R. J.;

Muir, K. W. Organometallics 1993, 12, 1725.

(28) Dai, X.; Kapoor, P.; Warren, T. H. J. Am. Chem. Soc. 2004, 126, 4798. 67 (29) Puiu, S. C.; Warren, T. H. Organometallics 2003, 22, 3974.

(30) Evans, D. F. J. Chem. Soc. 1959, 2003.

(31) Kogut, E.; Wiencko, H. L.; Zhang, L.; Cordeau, D. E.; Warren, T. H. J. Am. Chem.

Soc. 2005, 127, 11248.

(32) Eckert, N. A.; Bones, E. M.; Lachicotte, R. J.; Holland, P. L. Inorg. Chem. 2003,

42, 1720.

(33) Ford, P. C.; Fernandez, B. O.; Lim, M. P. Chem. Rev. 2005, 105, 2439.

(34) Tran, D.; Skeleton, B. W.; White, A. H.; Laverman, L. E.; Ford, P. C. Inorg. Chem.

1998, 37, 2505.

(35) Lim, M. D.; Capps, K. B.; Karpishin, T. B.; Ford, P. C. Nitric Oxide-Biol. Chem.

2005, 12, 244.

(36) Tsuge, K.; DeRosa, F.; Lim, M. D.; Ford, P. C. J. Am. Chem. Soc. 2004, 126, 6564.

(37) Khullar, K. K.; Bauer, L. J. Org. Chem. 1971, 36, 3038.

(38) Kokosa, J. M.; Bauer, L. J. Org. Chem. 1975, 40, 3196.

(39) Noyes, W. A. Organic Syntheses; Wiley: New York, 1943; Vol. III.

(40) Rao, C. N. R.; Bhaskar, K. R. The Chemistry of Nitro and Nitroso Groups. Part 1;

Interscience: New York, 1969.

(41) Werner, W.; Depreux, P. J. Chem. Res-S. 1980, 11, 372.

(42) Girard, P.; Guillot, N.; Motherwell, W. B.; Hay-Motherwell, R. S.; Potier, P.

Tetrahedron 1999, 55, 3573.

(43) Job, P. Ann. Chim. 1928, 9, 113.

68 Chapter 2

A Three Coordinate Copper(II)-Amide from Reductive Cleavage of a

Nitrosamine

Abstract

An electron-rich copper(I) β-diketiminate cleaves the N-N bond of the nitrosamine Ph2N-

NO to give a three ligand, two copper product where the backbone C-H of the ligand is

II nitrosated along with the novel, three coordinate [Me2NN]Cu -NPh2. Spectroscopic and theoretical studies provide insight into this extremely rare example of a structurally characterized three coordinate copper(II) amide. Use of a related anilidoimine copper(I)

+ complex prevents nitrosative (NO ) attack at the backbone by Ph2NNO. The copper(I) anilidoimine is less electron rich and does not promote cleavage of the Ph2N-NO bond.

+ + This ligand is not free from NO attack; [Me2AI]NO is formed from NO attack at the amide nitrogen in the presence of excess tert-. As the copper(II)-amide

[Me2NN]Cu-NPh2 reacts with NOgas to form Ph2NNO, both N-NO bond cleavage and formation are observed at a common copper center.

Introduction

Nitrosoamines were first described in the chemical literature over a hundred years ago by Geuther,1 and were occasionally used as solvents or synthetic intermediates in the preparation of hydrazines and diazoalkanes.2 It was not until 1956 that nitrosoamines such as dimethylnitrosamine were found to induce liver cancer in rats and can alkylate

69 proteins and amino acids.3-5 Thus the conversion of an amine to a nitrosamine leads to carcinogenic materials, especially when α-C-H atoms are present.6 Nitrosamines are found in common pollutants such as tobacco smoke, motivating approaches for their reduction to amines.

More recent studies show that nitrosamines have vasorelaxant activity7,8 and can be considered as a class of NO donors since the N-NO bond can be homolytically or

9 - heterolytically cleaved. The related diazeniumdiolates [R2NN2O2] formed by the pressurization of secondary amines with NO in a basic solution have been extensively employed as NO donors. Moreover, a new class N-aryl based nitrosamines have been shown to function as both NO and HNO donors.10

The interaction of nitrosamines with heme-containing biomolecules can result in activation and/or denitrosation of the nitrosamines.6,11 In small molecule model

Figure 2.1. Schematic representations of the molecular structures of a) (Me2NNO)CuCl2 in which all copper centers are identical having one N and one O binding from two adjacent nitrosamines. b) {(CH2)5NNO}-CuCl2 in which half of the copper centers exhibit N binding and the other have exhibit O binding from the nitrosamine. 70 complexes nitrosamines may form intact R2NNO adducts with metal complexes. Richter-

12 Addo studied metalloporphrin adducts of N-nitrosamines, (porph)M(ONNR2) (M = Fe ,

Ru13, Os14,15) (Scheme I.10). In non-heme complexes, two solid state structures of copper with nitrosamines show unique coordination modes of the nitrosamine ligands in an extended solid (Figure 2.1).16,17, 18

Delivery and storage of nitric oxide in vivo is often accomplished via reductive nitrosylation: reduction of an oxidizing metal center by NO with concomitant nitrosylation of a nucleophile. (Scheme 2a).19,20 Copper centers in cytochrome c oxidase and laccase,21,22 a few ferriheme proteins,23-25 as well as Cu(II)26 and Fe(III)27,28 model systems exhibit reductive nitrosylation through the use of water or an alcohol as the nucleophilic agents (NOTE: nitrosylation refers to attack by NO•, nitrosation refers to attack by NO+).

Scheme 2.1 Equilibrium of a) reductive nitrosylation (metal reduced) and b) reductive cleavage (R2N-NO bond reduced).

Reductive nitrosylation is also at the core of a small molecule copper

DAC/cyclam-based NO sensor. Reaction of NO to form a new N-NO bond with concomitant reduction to Cu(I) removes quenching present in the Cu(II) coordinated

71 Scheme 2.2. Intramolecular nitrosylation of a copper coordinated deprotonated amide by NO+ formed upon concomitant reduction of Cu2+ to Cu1+.

cyclam.29 This is the first well defined example of intramolecular nitrosylation of a deprotonated amine coupled to reduction of the metal center from Cu2+ to Cu1+. A mechanism was proposed in which the key step features attack of NO on the deprotonated amine with concerted electron transfer to the coordinated copper center.

Once the Cu2+ ion is reduced it is released from the cylcam owing to the preference of

Cu+ for tetrahedral coordination and the inability of the macrocylic cyclam ligand to

easily accommodate that geometry.

Depending on the nature of the metal center, the reverse process may be favored

(Scheme 2.1b). Copper modified zeolites have been shown to remove and trap volatile

nitrosamines in mainstream cigarette smoke due a strong interaction between the N-N=O

of the nitrosamine and the Cu in the zeolite.30,31 The exact structure and mechanism of the copper-nitrosamine interaction, however, remains unknown. We recently reported that an electron-rich nickel(I) β-diketiminate complex [Me3NN]Ni(2,4-lutidine) cleaves the E- 72 Figure 2.2. Copper(I) synthons used herein using the β-diketiminate and anilio imine ligands.

NO bond of O-, S-, and N-organonitroso species to give the nickel-nitrosyl [Me3NN]Ni-

NO along with dimeric nickel(II)-alkoxide or thiolate complexes {[Me3NN]Ni}2(-E)2

32 or the mononuclear nickel(II)-amide [Me3NN]Ni-NPh2 (see Chapter 1).

We have employed electron-rich β-diketiminate as well as anilidoimine ligands as supporting ligands to provide access to low-valent copper(I) and copper(II) complexes

(Figure 2.2 and Appendix Figure A.1) which could be used as precursors for studying the reactivity of E-NO compounds with electron-rich, well defined copper complexes. Herein we report the reactivity of diphenylnitrosamine Ph2NNO with the related copper(I) β- diketiminate {[Me2NN]Cu}2 and anilidoimine {[Me2NN]Cu}2.

73 Results and Discussion

II I 2.1. Synthesis and characterization of [Me2NN]Cu (ON-Me2NN)2Cu .

32,33 34 Addition of 2 equiv Ph2NNO to 3/2 equiv {[Me2NN]Cu}2 at -35 ºC in ether resulted in color change from light yellow to dark brown. Formation of a copper mirror indicated that this reaction followed a complex course. Crystallization of the reaction mixture from ether allowed the isolation of both dark brown and green crystals.

Single crystal X-ray characterization of the brown crystals isolated in ca. 24 %

II I yield (based on the moles of Cu) identifies them as [Me2NN]Cu (ON-Me2NN)2Cu (1)

(Figure 2.3). Formation of 1 requires reaction of 3 equiv [Me2NN]Cu (monomer in solution) with 2 equiv. Ph2NNO. Additional products of the reaction are 2 equiv Ph2NH

0 1 13 along with Cu . We confirmed the presence of Ph2NH by H and C NMR and GC/MS.

The formation of a copper mirror on the sides of the vial confirms the formation of Cu0

Scheme 2.3. Reaction of 3/2 equiv. {[Me2NN]Cu}2 with 2 equiv. Ph2NNO to form 1. The backbone of the ligand is nitrosated (NO+ for H+ exchange).

74 Ph O N N N Ph N Cu H Cu+ N N

O N N O N Cu+ N Cu+ H Ph N Ph N N NH Ph Ph

Scheme 2.4. Nitrosation of β-diketiminate backbone CH in [Me2NN]Cu by Ph2NNO.

(Scheme 2.3)

The resonance structure of the β-diketiminate ligand in which the negative charge

is localized at the backbone C-H may be helpful to explain why [Me2NN]Cu is susceptible to nitrosation at the β-diketimate backbone. The localized negative charge on

- the backbone C-H may attack Ph2N-NO to give Ph2N which may deprotonate the backbone C-H to form Ph2NH. While this mechanism is simply a proposal, it rationalizes the susceptibility of the ligand backbone to NO+ attack.

Complex 1 has two copper ions with vastly different coordination environments.

Cu1 is linear (N1-Cu1-N3= 176.61(15)º), a relatively uncommon geometry for Cu which

75 is only observed in the 1+ oxidation state (d10). The coordination environment of Cu2 is square planar completed by two O-donors resulting from the nitrosation of the former β- diketiminato backbone C-H sites in the two modified β-diiminate ligands now coordinated to the linear Cu1 center. The square planar configuration at Cu2 is consistent with assignment of the Cu(II) oxidation state for Cu2. Both the square planar configuration at Cu2 as well as the axially biased EPR spectrum of 1 in frozen toluene glass (gparallel = 2.20, gperp = 2.05 with Aparpallel(Cu) and Aperp(Cu) = 177 and 26 G ; Figure

2.13) are consistent with assignment of a Cu(II) center in 1.

76 N2 O1 N1 N5

N7 Cu1 Cu2 N6 N4 N8 N3 O2

II I Figure 2.3. X-ray structure of [Me2NN]Cu (ON-Me2NN)2Cu (1) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu1–N1 1.863(4), Cu1–N3 1.870(3), Cu2- N7 1.942(3), Cu2–N8 1.940(4), Cu2-O1 1.953(3), Cu2–O2 1.948(3), N7–Cu2–N8 95.37(14), O1–Cu2–O1 158.88(13), O2–Cu2–N8 91.82(13), N7–Cu2–O2 162.73(13), N7–Cu1–O1 92.24(14), N1–Cu1–N3 176.61(15).

77 2.2. Isolation of [Me2NN]Cu-NPh2 from {[Me2NN]Cu}2 / Ph2NNO reaction mixture.

Pentane recrystallization of the green product which forms upon the interaction of

Ph2NNO with {[Me2NN]Cu}2 allows for the isolation of the novel three coordinate copper(II)-amide [Me2NN]Cu-NPh2 (2) in 31 % yield, identified by a preliminary single- crystal X-ray diffraction from the reaction mixture which included 1. Product 2 may form

I in a more easily understood route: the Ph2N-NO bond is reductively cleaved by the Cu complex to form [Me2NN]Cu-NPh2 (2) and release NOgas (Scheme 2.5). This follows a

I similar route to our previous work with [Me3NN]Ni (2,4-lutdine) and Ph2N-NO where the

II [Me3NN]Ni -NPh2 forms along with [Me3NN]Ni(NO) (see Chapter 1). The copper chemistry differs from the nickel chemistry principally in that the copper does not form an easily isolable copper-nitrosyl [Me2NN]Cu(NO).

I Scheme 2.5. Reductive cleavage of the Ph2N-NO bond by [Me2NN]Cu to release NOgas and form [Me2NN]Cu-NPh2 (2). 78 2.3. Independent synthesis of [Me2NN]Cu-NPh2.

A higher quality X-ray structure of 2 was obtained through its independent synthesis. Reaction of {[Me2NN]Cu}2(-Cl)2 with 2 equiv. LiNPh2 gives [Me2NN]Cu-

NPh2 (2) (62 % yield) by X-ray crystallography and UV-Vis characterization. After working up the X-ray data, however, unexpected electron density (> 3 e- / Å3) was found on C2 (Figure 2.4). This residual was too large to be ignored. In a similar compound,

Tolman found residual electron density on the backbone of his ligand and modeled it as a chlorine in 10% occupancy.35 He attributed this rogue chlorine atom to the

CuCl2•0.8THF used in the reaction synthesis. We too use CuCl2•0.8THF in the synthesis of {[Me2NN]Cu}2(-Cl)2 and therefore model the residual electron density to a chlorine atom in 15% occupancy with the hydrogen atom at 85% occupancy (Figure 2.4).

Refining the data under these conditions removes the residual electron density from the backbone. There is no further disorder in the structure and an R1 value of 0.0660 results after the final refinement, suggesting otherwise good quality crystal data.

79 N1 N3 C2 Cu1 Cl N2

Figure 2.4. X-ray structure of [Me2NN]Cu-NPh2 (2) with 0.15:0.85 Cl:H on C2 (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.916(4), Cu–N2 1.888(4), Cu–N3 1.840(4), C2-Cl 1.539(6), N1–Cu–N2 97.12(16), N3–Cu–N1 122.76(16), N3–Cu–N2 139.66(17).

80 We then moved to the use of the copper(II)-iodide [Me2NN]CuI to synthesize

[Me2NN]Cu-NPh2 (2). Reaction of {[Me2NN]Cu}2 with 1 equiv. I2 gives [Me2NN]CuI (3,

36 37 % yield) as purple crystals. If any extra I2 is added to the reaction mixture, the solution turns from purple expected for 3 to rusty brown with immediate precipitation of an uncharacterized compound. While the yield from the reaction to form 3 is quite low (<

40%), further attempts to improve the yield were not successful. It could be that excess iodine in the reaction solution reacts at the backbone C-H position. For instance, Dr.

Yosra Badiei from the Warren group has observed H[Cl2NNI] in the reaction of

{[Cl2NN]Cu}2(benzene) with I2 to give the β-diketimine ligand H[Cl2NNI] free of copper with an iodine atom on the backbone C-H replacing the H atom at this position.37

Curiously, this [Cu]-I complex crystallizes as a mononuclear species [Me2NN]Cu-

I (3) [Cu-I = 2.4295(5) Å] from ether (Scheme 2.6, Figure 2.5) and a dinuclear species

… {[Me2NN]Cu}2(-I)2 (3-dinuclear) [Cu-I = 2.6128(8) and 2.6076(9) Å; Cu Cu separation 3.673(2) Å] from pentane (Scheme 2.6, Figure 2.6). Reaction of 3 with LiNPh2 in THF delivers 2 in 81 % yield as green crystals from pentane (Scheme 2.7).

Scheme 2.6. Synthesis of [Me2NN]Cu-I (3) from reaction of {[Me2NN]Cu}2 with I2, crystallized from ether. Crystallization from pentane gives {[Me2NN]Cu}2(-I)2 (3- dinuclear).

Scheme 2.7. Synthesis of [Me2NN]Cu-NPh2 (2) from [Me2NN]Cu-I (3) with LiNPh2.

81 N1 I Cu1

N1'

Figure 2.5. X-ray structure of [Me2NN]Cu-I (3) (all H atoms omitted) crystallized from ether. Selected bond distances (Å) and angles (deg): Cu–N1 1.8767(17), Cu–I 2.4295(5), N1–Cu–N1´ 97.07(10), N1–Cu–I 131.46(5).

82 N2' N1 I Cu1' Cu1 I' N1' N2

Figure 2.6. X-ray structure of {[Me2NN]Cu}2(-I)2 (3-dinuclear) (all H atoms omitted) crystallized from pentane. Selected bond distances (Å) and angles (deg): Cu–N1 1.930(4), Cu–N2 1.926(4), Cu–I 2.6128(8) Cu–I´ 2.6076(9), N2–Cu–N1 96.20(16), Cu– I–Cu´ 89.42(3), N2–Cu–I 137.40(11) N2–Cu–I´ 100.86(11), N1–Cu–I 99.45(11), N1– Cu–I´ 141.95(11), I–Cu–I´ 90.58(3).

83 The X-ray crystal structure of 2 shows a short Cu-Namide distance of 1.841(6) Å with Cu-Nβ-dik distances of 1.914(6) and 1.885(6) Å (Figure 2.7). Both the Cu atom and

Namido atoms are planar (sum of angles 359.7(3) and 357.2(5)°, respectively). One of the two N-aryl rings is coplanar with N center while the other is twisted out of the Cu-N-C plane by ca. 65°. The metal-nitrogen distances are longer in the copper(II)-amide system

(Cu-Namide distance of 1.841(6) and Cu-Nβ-dik distances of 1.914(6) and 1.885(6) Å) than

32 in the related β-dikeiminato nickel(II)-amide [Me3NN]Ni-NPh2 (Ni-Namide distance of

1.823(1) and Ni-Nβ-dik distances of 1.826(1) and 1.833(3) Å).

[Me2NN]Cu-NPh2 (2) represents a rare example of a crystallographically characterized three-coordinate copper(II)-amide. Peters has recently reported the

2 t structure of the diphosphinoborate κ -[Ph2B(CH2PBu 2)]Cu-N(p-tolyl)2 which exhibits a

Cu-N bond distance of 1.906(2) Å.38 Based on several spectroscopic measurements, this species was best described as a Cu(I) coordinated aminyl radical. Employing chelating ligands, X-ray structures of two square planar copper(II) amides with electron-deficient sulfonyl amido substituents39,40 as well as distorted tetrahedral phosphine-amide species41 have been reported. In copper(I) chemistry, Gunnoe described the structures of three and

t t 42 two coordinate amides such as ( Bu2PCH2CH2PBu 2)Cu-NHPh and the N-heterocyclic carbene-based IPrCu-NHPh.43

84 N1 N3 Cu1

N2

Figure 2.7. X-ray structure of [Me2NN]Cu-NPh2 (2) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.914(6), Cu–N2 1.885(6), Cu–N3 1.841(6), N1–Cu–N2 97.0(3), N3–Cu–N1 122.4(3), N3–Cu–N2 140.2(3).

85 2.4. Electronic properties of [Me2NN]Cu-NPh2.

The electronic spectrum of 2 in ether exhibits four bands in the visible to near IR at 426 nm (530 M-1cm-1), 532 nm (280 M-1cm-1), 774 nm (1200 cm-1M-1), and 908 nm

(1200 cm-1M-). Monitoring by UV-vis spectroscopy, toluene solutions of 2 (1.20 mM) have a half-life of approximately 7 hr at RT, turning colorless after a day. GC/MS studies reveal that Ph2NH is the primary organic product of decomposition.

Simulation of the X-band EPR spectrum of 2 in frozen toluene glass (Figure 2.8) indicates a rhombic signal (g1 = 2.145, g2 = 2.048, g3 = 2.010) with A1(Cu) = 100 G.

The hyperfine to Cu is modestly depressed relative to that found in other three-

i 44 coordinate β-diketiminato halide species such as [ Pr2NN]CuCl as well as [Me2NN]CuI

(3) (A1(Cu) = 120 and 139 G (Figure 2.14), respectively) though similar to that found in

i Tolman’s three coordinate phenoxide [ Pr2NN]CuOAr (Ar = p-MeOC6H4); A1(Cu) = 90

G).45 We have unfortunately been unable to simulate reliable A(N) values for the amido

15 N atom, including the use of Li NPh2 the synthesis of 2. The EPR spectrum of 2 is

2 t distinct from Peter’s Cu(I) aminyl complex κ -[Ph2B(CH2PBu 2)]Cu-N(p-tolyl)2 which exhibits g-values (2.030, 2.008, 2.008) close to that of the free electron with a

38 significantly depressed A1(Cu) of ca. 60 G.

Nonetheless, DFT analysis of 2 points to appreciable radical character on the amido N atom as a result of the 3-electron / 2-center π-interaction between the Cu d- orbital most destabilized by σ-interactions with the β-diketiminate ligand (dyz) and the

2 9 II filled lone pair of an sp -hybridized amido N atom (pz) (Figures 2.8 and 2.9). The d Cu center of the β-diketiminato fragment possesses one electron in this dyz orbital which 86 g-values N R 2.32.42.5 2.12.2 2.0 1.9 z R N Cu N

y x * N R R N Cu N

N R

N Cu N R 260 280 300 320 340 Magnetic field / mT Figure 2.8. Cu-amido 2-center, 3-electron π-orbital interactions (left) and X-band EPR spectrum and simulation for [Me2NN]Cu-NPh2 (2) (frozen toluene glass, 30 K, 8.982 GHz). Simulation provides g1 = 2.145, g2 = 2.048, g3 = 2.010 with A1(Cu) = 100 G and A2(Cu) ≈ A3(Cu) ≈ 0 G. Simulation employed Aiso(N) = 14.5 G for three equivalent N atoms. Data were collected and simulated by Dr. Susanne Mossin of Prof. Karsten Meyer’s research group at the University of Erlangen, Germany (right) interacts with the sp2-hybridized amido lone pair to singly populate the Cu-N π* orbital which represents the SOMO of 2. The DFT spin densities at Cu and N are 0.30 and 0.27 e-, respectively, with significant delocalization of unpaired electron density over the two aryl rings (Figure 2.9, right). This analysis indicates a rather covalent Cu-amido π- interaction in 2 consistent with the modest value of A(Cu) observed in the EPR spectrum of 2. This bonding description is qualitatively similar to that in related copper(II)- thiolates in which the thiolate S atom bears significant spin density.44,46,47

The presence of electron-spin at the amido N atom suggests the possibility of N- centered reactivity with other radical species such as NO. We observe the immediate formation of Ph2NNO when 1 equiv NOgas is added to isolated [Me2NN]Cu-NPh2. While 87 [Me2NN]Cu-NPh2 (2) Cu-N1 1.930 Å N1-Cu-N3 132.3° Cu-N2 1.935 Å N2-Cu-N3 130.2° Cu-N3 1.880 Å N1-Cu-N2 97.4°

Figure 2.9. DFT structure (left) and contour plot of SOMO (right) for [Me2NN]CuNPh2 along with selected bond distances and angles. DFT electron spin densities predicted: Cu - - - 0.30 e ; Namide 0.27 e ; Nβ-dik 0.06, 0.08 e .

we were unable to quantify the Ph2NNO formed since it can be consumed by any

I II I [Me2NN]Cu species formed in solution to give [Me2NN]Cu (ON-Me2NN)2Cu (1), this system nonetheless demonstrates both the reductive cleavage of Ph2N-NO as well as the reductive nitrosylation of the corresponding Cu(II)-amide 2.

2.5. Reaction of {[Me2AI]Cu}2 with Ph2NNO.

In order to discourage nitrosation of the β-diketiminate backbone which ultimately results in the formation of 1, we turned our we turned our attention to anilidoimine ligands48 because the backbone of the ligand is “protected” from nitrosation due to the incorporation of an aryl ring. Combination of 2 equiv. Ph2NNO and

88 {[Me2AI]Cu}2 in benzene-d6 led to broadening of the [Me2AI]Cu(arene) which suggests some interaction between the [Me2AI]Cu and Ph2NNO moieties (Scheme 2.8).

Importantly, ligand nitrosation does not take place.

Scheme 2.8. Interaction of Ph2NNO with ½ equiv. {[Me2AI]Cu}2 in arene solvents.

t 2.6. Reaction of {[Me2AI]Cu}2 with excess BuONO.

Even though ligand nitrosation does not take place in the reaction between ½ equiv. {[Me2AI]Cu}2 and Ph2NNO, this anilidoimine ligand is still susceptible to nitrosation with more powerful NO+ sources such as organonitrites. In the presence of

Scheme 2.9. Nitrosation of the anilioimine {[Me2AI]Cu}2 from the reaction of excess t BuONO to give [Me2AI]NO (4).

89 t + excess BuONO, {[Me2AI]Cu}2 is attacked by NO at the amido nitrogen to lose its copper ion and give [Me2AI]NO (4) which was isolated by crystallization from ether

(Figure 2.10, Scheme 2.9). The negative charge on the backbone is not completely delocalized and the amido nitrogen bears a significant amount of that charge, thus making it susceptible to NO+ attack. All attempts at synthesizing 4 in the absence of copper were unsuccessful.

90 N1 O1 N3

N2

Figure 2.10. X-ray structure of [Me2AI](NO) (4) (all H atoms omitted). Selected bond distances (Å) and angles (deg): N1–N3 1.356(2), N3–O1 1.226 (2), N1–N3–O1 114.12(10).

91 2.7. Electronic differences between {[Me2NN]Cu}2 and {[Me2AI]Cu}2.

It is interesting to note that the copper(I) β-diketiminate [Me2NN]Cu activates the nitrosamine towards N-NO cleavage while the related anilidoimine complex [Me2AI]Cu does not under analogous conditions. Comparison of the IR spectra of analogous 2,6- dimethylphenylisocyanide complexes [Me2NN]Cu(CNAr) (5) and [Me2AI]Cu(CNAr) (6) indicates stronger backbonding from a more electron-rich copper center in the the β-

-1 diketiminato complex (νCN = 2124 and 2131 cm , respectively) (Figures 2.11 and 2.12).

This extra electron-richness may contribute to its reactivity in the reductive cleavage of the Ph2N-NO bond. In this context, the clean reaction between 2 equiv. [Me3NN]Ni(2,4- lut) and Ph2NNO to give [Me3NN]Ni-NPh2 and [Me2NN]Ni(NO) is likely due to the significantly more rich NiI metal center coupled with the formation of a stable {NiNO}10 species.

Figure 2.11. The IR v(C≡N) stretching frequencies are dependent on the electronic nature of the β-diketiminate-like ligand and indicate that the β-diketiminate is more electron rich.

92 N1 Cu1 C22 N3 N2

Figure 2.12. X-ray structure of [Me2NN]Cu(NCAr) (5) (Ar = 2,6 dimethylbenzene) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.926(2), Cu– N2 1.940(2), Cu–C22 1.815(2), C22–N3 1.156(3), C23–N3 1.404(3), N1–Cu–N2 98.03(8), N1–Cu–C22 133.78(9), N2–Cu–C22 127.82(8), Cu–C22–N3 174.8(2), C22– N3–C23 174.4(2).

93 Summary

This model system allows for the observation of both cleavage and formation of

an N-NO bond at a common copper center. While the β-diketiminate ligand is not a

completely innocent spectator, it provides a framework to allow for the isolation of the novel three coordinate copper(II)-amide [Me2NN]Cu-NPh2. Electronic effects are important in determining the ability to release NO from the R2NNO substrate. This may be of particular relevance in the biologically important generation of NO at copper sites from S-nitrosothiols RSNOs such as S-nitrosoglutathione which serve as a circulating reservoirs of NO.49

94 Table 2.1. Crystallographic parameters for 1, 2, 2', and 3-mononuclear.

Compd. 1 2 2' 3

Formula C63H73Cu2N8O2, C33H35CuN3 C33H34.85Cl0.15 C21H25Cu I N2

C4H10O CuN3 Mol. Wt. 1175.49 537.18 542.35 495.87 Temp.(K) 173(2) 100(2) 100(2) 100(2) Crystal Block Block Plate Block description Crystal color Brown Green Green Purple Crystal size 0.0410 0.0108 0.0217 0.0010 (mm3) System Triclinic Monoclinic Monoclinic Orthorhombic Space group P-1 C2/c C2/c Pbcn a (Å) 11.4757(7) 17.384(4) 17.348(5) 8.3269(8) b (Å) 14.7928(9) 15.302(4) 15.293(4) 13.9138(14) c (Å) 19.6284(11) 20.982(5) 21.153(6) 18.1184(18) α (deg) 82.2820(10) 90 90 90 β (deg) 79.1630(10) 96.118(3) 96.041(4) 90 γ (deg) 74.2790(10) 90 90 90 Volume (Å3) 3138.0(3) 5550(2) 5581(3) 2099.2(4) Z 2 4 4 4 θ range (deg) 1.06-25.00 1.78-25.00 1.78-27.00 2.25-28.00 Measd reflns 28645 17768 23267 17703 Unique reflns 11036 4882 6088 2525 GOF of F2 0.981 1.107 0.990 1.058

R1 ( I > 2σ(I)) 0.0556 0.0854 0.0660 0.0304 wR2 (all data) 0.1548 0.2439 0.1742 0.0858 Largest diff. 0.524 and 1.270 and 0.621 and 1.168 and - peak and hole -0.706 -0.603 -0.627 0.897 (e-.Å-3) 95 Table 2.1. Crystallographic parameters for 3-dinuclear, 4, and 5.

Compd. 3-dinuclear 4 5

Formula C42H50Cu2I2N4 C23H23N3O C30H34CuN3 Mol. Wt. 991.76 358.45 500.14 Temp.(K) 100(2) 100(2) 173(2) Crystal Plate Block Plate description Crystal color Purple Yellow Green Crystal size 0.0700 0.0882 0.0082 (mm3) System Monoclinic Triclinic Monoclinic Space group P2(1)/n P-1 P2(1)/n a (Å) 12.802(5) 8.3787(3) 12.9345(13) b (Å) 8.998(3) 8.4058(3) 11.3086(11) c (Å) 17.664(5) 13.8050(5) 19.1873(18) α (deg) 90 78.8960(10) 90 β (deg) 103.511(3) 83.9890(10) 106.4630(10) γ (deg) 90 89.4670(10) 90 Volume (Å3) 1978.6(12) 948.78(6) 2691.5(5) Z 2 2 4 θ range (deg) 1.78-28.00 1.51-27.00 1.64-26.00 Measd reflns 17452 10922 20975 Unique reflns 4694 4099 5282 GOF of F2 1.045 1.075 1.063

R1 ( I > 2σ(I)) 0.0413 0.0363 0.0379 wR2 (all data) 0.1040 0.0976 0.1067 Largest diff. 1.288 and 0.282 and 0.521 and peak and hole -0.981 -0.211 -0.238 (e-.Å-3)

96 Experimental procedures

General experimental details.

All experiments were carried out in a dry nitrogen atmosphere using an MBraun

glovebox and/or standard Schlenk techniques. 4A molecular sieves were activated in

vacuo at 180 ºC for 24 h. Dry toluene and dichloromethane were purchased from Aldrich

and were stored over activated 4A molecular sieves under nitrogen. Diethyl ether and

tetrahydrofuran (THF) were first sparged with nitrogen and then dried by passage through

1 activated alumina columns. Pentane was first washed with conc. HNO3/H2SO4 to remove olefins, stored over CaCl2 and then distilled before use from sodium/benzophenone. All deuterated solvents were sparged with nitrogen, dried over activated 4A molecular sieves and stored under nitrogen. 1H and 13C NMR spectra were recorded on either a Varian 300 MHz or 400 MHz spectrometer (300 or 400 and 75.4 or

100.4 MHz, respectively). All NMR spectra, including Evans method data, were recorded at room temperature unless otherwise noted and were indirectly referenced to TMS using residual solvent signals as internal standards. GC-MS spectra were recorded on a Varian

Saturn 2100T, elemental analyses were performed on a Perkin-Elmer PE2400 microanalyzer in our laboratories, and VU-Vis spectra were recorded on either an Agilent

8432 or a Cary 50 spectrophotometer.

EPR measurements were performed by Dr. Susanne Mossin in Prof. Karsten

Meyer’s research group at the University of Erlangen, Germany. Measurements were made on samples in solution in quartz tubes with Young valves. Anisotropic simulation was performed using the program ESRSIM written by Høgni Weihe, University of 97 Copenhagen, Denmark. Solution EPR spectra were recorded on a JEOL continuous wave

spectrometer JES-FA200 equipped with an X-band Gunn oscillator bridge, a cylindrical

mode cavity, and a helium cryostat. The g-value in the solution spectrum was calibrated using the second and fourth line of the Mn2+ hyperfine spectrum from an external Mn marker. The marker consists of Mn2+ thermally dispersed in MgO placed on the tip of a probe which is inserted slightly into the cavity. The cavity was tuned with both sample and probe inserted. The Mn marker is provided with the JEOL JES-FA200 spectrometer.

Diphenyamine, iodine, and tert-butylnitrite were obtained from Acros and were

50 51 52 33 used as received. Copper tert-butoxide , CuCl2•THF0.8 {[Me2AI]Cu)2 , and Ph2NNO were synthesized according to published procedures.

Preparation of compounds.

{[Me2NN]Cu}2. {[Me2NN]Cu}2 was synthesized with modifications to the literature

34 procedure. [Me2NN]H (2.00 g, 6.536 mmol) in 10 mL of toluene was adding to a stirring solution of copper(I) tert-butoxide (1.071 g, 7.843) in 25 mL of toluene. The solution was allowed to stir for 2 hr, filtered over Celite, and the volatiles were removed in vacuo to yield brown oil. The brown oil was washed with pentane (3 × 8 mL) to afford

1 2.25 g (92 %) of {[Me2NN]Cu}2 as a yellow powder. H NMR (benzene-d6, 400 MHz, 25

°C): δ 7.100 (d, 4, m-ArH), 6.993 (t, 2, p-ArH), 4.776 (s, 1, CHbackbone), 2.016 (s, 12, o-

ArMe2), 1.643 (s, 6, Mebackbone).

98 Reaction of {[Me2NN]Cu}2 with Ph2NNO. At room temperature Ph2NNO (0.125 g,

0.633 mmol) in 3 mL Et2O was added to a slurry of {[Me2NN]Cu}2 (0.350 g, 0.475 mmol) in 5 mL Et2O. The solution immediately turned dark green/brown. The solution was allowed to stir for two hours and 35 mg of copper metal plated on the outside of the vial during the course of the reaction (58 % based on total moles of Cu) removed in vacuo. The dried solid was stirred with 5 mL of pentane for 5 minutes causing the solution to turn dark green. The insoluble components were allowed to settle and the solution is filtered through Celite. This pentane washing procedure was repeated two more times. The filtrates were combined and concentrated to 3 mL and allowed to crystallize overnight at -35 °C. X-ray quality crystals form (0.160 g) and were identified as [Me2NN]Cu-NPh2 (2) (0.160 g, 0.298 mmol, 31% yield based on moles of Cu) crystallized The pentane insoluble solids were then washed with Et2O (3 × 5 mL) causing the solution to turn reddish brown. The washings were combined, filtered through Celite, and concentrated to 3 mL. The solution was allowed to crystallize overnight at -35 °C. X- ray quality crystals formed (0.082 g, 16 % yield, based on moles of Cu) and were

II I -1 -1 identified as [Me2NN]Cu (ON-Me2NN)2Cu 1. UV-Vis (Et2O, 25 °C, nm(cm M ))

341(110000). Anal. Calcd for C63H73Cu2N8O2: C, 68.70; H, 6.68; N, 10.17. Found: C,

68.88; H, 6.65; N, 10.41.

99 Independent synthetic routes to [Me2NN]Cu-NPh2 (2).

We synthesized 2 via two different Cu(II) starting material routes. We initially used {[Me2NN]Cu}2(-Cl)2, but discovered that the backbone C-H of the ligand is subject to attack by a Cl atom. Use of [Me2NN]Cu-I prevents halogenation of the ancillary ligand and allows for synthesis of pure [Me2NN]Cu-NPh2 (2).

{[Me2NN]Cu}2(µ-Cl)2. This compound was synthesized according to the published

53 compound {[Me2NN]lCu}2(-Cl)2 (methine C-H on backbone replaced with Cl) with some minor changes. In a one pot synthesis, n-butyllithium (2.00 mL, 3.27 mmol, 1.6 M solution in hexanes) was added dropwise to a solution of [Me2NN]H (1.00 g, 3.27 mmol) in 25 mL of THF at -35 °C. The solution was allowed to warm to room temperature and

II I Figure 2.11. Beer’s Law plot for [Me2NN]Cu (ON-Me2NN)2Cu (1) in Et2O. 100 stir for 1 hr. The solution of [Me2NN]Li in THF was added to a slurry of CuCl2•THF0.8

(0.627 g, 3.27 mmol) in 10 mL of THF causing the formation of a dark green solution.

The reaction mixture was allowed to stir for 4 hr and the solvent was removed in vacuo.

The green residue was extracted with CH2Cl2 (3 × 5 mL) yielding a purple solution that was filtered through a pad of Celite. The volume of solvent was reduced to 2 mL, pentane

(2 mL) was added, and the solution was stored at 20 °C overnight, causing the deposition of dark crystals, 0.987 g (75 % yield). No other characterization was performed on this species and it was directly used as described below.

[Me2NN]Ni-NPh2 (2) using {[Me2NN]Cu}2(µ-Cl)2. A solution of Ph2NLi (0.200 g, 1.143 mmol) in 3 mL of THF was added to a stiring solution of [Me2NN]Cu2(µ-Cl)2 (0.461 g,

0.572 mmol) in 4 mL THF. The brown solution immediately turned dark green and the solution was allowed to stir for 1 hr. The solvent was removed in vacuo and the solids were extracted 3 times with 5 mL of pentane. The solution was filtered over Celite, concentrated to 4 mL, and allowed to crystallize overnight at -35 °C. The mother liquor was removed and the residue was washed with a few mL of cold pentane to yield 0.255 g of green crystals. The volume of the mother liquor was reduced to 2 mL and placed back in the freezer to yield a second crop of crystals (0.124 g) for a combined yield of 62%.

Crystals of X-ray quality were grown from this reaction mixture and allowed for the identification of this compound to be the same as from the reaction of {[Me2NN]Cu}2 with Ph2NNO.

101 X-ray of [Me2NN]Cu-NPh2 (2) from {[Me2NN]Cu}2(µµµ-Cl)2. After working up the X- ray data, unexpected electron density (> 3 e- / Å3) was found on the β-diketiminate backbone methine C atom. This number was too large to be considered residual. In a similar compound, Tolman found residual electron density on the backbone of his ligand and modeled it as a chlorine in 10% occupancy.35 He attributed this rogue chlorine atom to the CuCl2•0.8THF used in the reaction synthesis. We too, use CuCl2•0.8THF in synthesizing [Me2NN]Cu2(-Cl)2 and model the residual electron density to a chlorine atom in 15% occupancy with the hydrogen atom at 85% occupancy. The C(2)-Cl(1) has a bond distance of 1.539(6). Refining the data under these conditions removes the residual electron density from the backbone. There is no further disorder in the structure and an

R1 value of 0.0660 results after the final refinement, suggesting good quality crystal data.

We then moved to [Me2NN]Cu-I to synthesize [Me2NN]Cu-NPh2.

Synthesis of [Me2NN]Cu-I (3). At room temperature I2 (0.070 g, 0.276 mmol) in 3 mL

Et2O was added to a slurry of {[Me2NN]Cu}2 (0.200 g, 0.271 mmol) in 5 mL Et2O. The solution immediately turned dark purple. The reaction is allowed to stir for 1 hr. The ether was removed in vacuo and the solids were extracted pentane (3 × 5 mL). The solution was filtered over Celite, concentrated to 3 mL, and allowed to crystallize overnight at -35 °C. The mother liquor was removed and the residue was washed with a few mL of cold pentane to yield 0.096 g of purple crystals (37% yield). UV-Vis (Et2O, 25

°C, nm(cm-1M-1)) 280(6600), 328(27000) 425(1200), 515(3500), 592(2400), 646(2200),

102 755(1300). Anal. Calcd for C21H25Cu1N2I1: C, 50.86; H, 5.08; N, 5.65. Found: C, 50.40;

H, 4.87; N, 5.51.

Distinct X-ray structures of [Me2NN]CuI (3) from pentane and ether.

Single crystal X-ray diffraction of [Me2NN]CuI (3) from pentane and ether gave different structures corresponding to dinuclear and mononuclear forms of this β-diketiminato copper(II) iodide. {[Me2NN]Cu}2(-I)2 (3 – dinuclear) crystallized from pentane (unit cell data: space group P2(1)/c, a = 12.802(5) Å, b = 8.998(3) Å, c = 17.664(5) Å, α= β=

3 90°, γ = 103.511(3)°, V = 1978.60(12) Å ) whereas the three coordinate [Me2NN]CuI (3

– mononuclear) crystallized from ether (unit cell data: space group Pbcn, a = 8.235(1) Å, b = 13.9138(14) Å, c = 18.1184(18) Å, , α= β= γ = 90°, V = 2099.2(4) Å3).

- Synthesis of [Me2NN]Ni-NPh2 (2) using [Me2NN]Cu-I (3) (Cl free conditions). A solution of Ph2NLi (0.200 g, 1.143 mmol) in 3 mL of THF was added to a stiring solution of [Me2NN]Cu-I (0.566 g, 1.143 mmol) in 4 mL THF. The purple solution immediately

turned dark green and the solution was allowed to stir for 1 hr. The solvent was removed

in vacuo and the solids were extracted with pentane (3 × 5 mL). The solution was filtered

over Celite, concentrated to 4 mL, and allowed to crystallize overnight at -35 °C. The

mother liquor was removed and the residue was washed with a few mL of cold pentane to

yield 0.495 g of green crystals (81 % yield). Crystals of X-ray quality were grown from

this reaction mixture and allowed for the identification of this compound to be the same

-1 - as from the reaction of {[Me2NN]Cu}2 with Ph2NNO. UV-Vis (Et2O, 25 °C, nm(cm M

103 1 )) 426(530), 532(280), 774(1200), 908(1200). Calcd for C33H35Cu1N3: C, 73.78; H, 6.57;

N, 7.82. Found: C, 73.46; H, 6.79; N, 7.75.

15 54,55 Ph2 NH. This compound was synthesized using the room-temperature palladium

56 catalyzed amination of Ar-Br compounds with Pd(dba)2 with slight modifications. In

15 the dry box, bromobenzene (565 L, 5.375 mmol), Ph NH2 (500 mg, 0.532 mmol),

Pd(dba)3 (0.044 g, 0.054 mmol), tri-tert-butylphosphine (10 L, 0.041 mmol), and potassium tert-butoxide (0.908 g, 8.110 mmol) were weighed directly into a screw cap vial. A stir bar was added followed by 10 mL of toluene to a give a purple mixture. The vial was removed from the dry box, and the mixture was stirred at room temperature for 1 hr. After 1 hr, the reaction mixture was adsorbed onto silica gel and chromatographed to

15 5% ethyl acetate/hexanes to give 0.823 g (91 % yield) of pure Ph2 NH2 as a gray solid.

15 GC/MS was taken on the solid to show complete consumption of Ph NH2 and pure

15 Ph2 NH2; GC/MS (CI-MS: m/z = 171(M + 1)) with a negligible amount of Ph3N.

104 Figure 2.12. Beer’s Law plot for [Me2NN]Cu-NPh2 (2) in Et2O.

15 Li NPh2. n-Butyllithium (2.75 mL, 4.400 mmol, 1.6 M solution in hexanes) was added

15 dropwise to a solution of Ph2 NH (0.500 g, 2.911 mmol) in 8 mL toluene at -35 °C. The solution was allowed to warm to room temperature for 1 hr after which a white precipitate had formed. The mother liquor was removed and the precipitate was washed with 3 mL of cold pentane and dried in vacuo affording 0.456 g of product in 89 % yield.

15 15 [Me2NN]Cu- NPh2. A solution of Ph2 NLi (0.200 g, 1.136 mmol) in 3 mL of THF was added to a stiring solution of [Me2NN]Cu-I (0.563 g, 1.136 mmol) in 4 mL THF. The purple solution immediately turned dark green and the solution was allowed to stir for 1 hr. The solvent was removed in vacuo and the solids were extracted 3 times with 5 mL of

105 pentane. The solution was filtered over Celite, concentrated to 4 mL, and allowed to

crystallize overnight at -35 °C. The mother liquor was removed and the residue was washed with a few mL of cold pentane to yield 0.435 g of green crystals (71 % yield).

Characterization by UV-Vis reveals this compound to be spectroscopically identical to

-1 -1 [Me2NN]Cu-NPh2. UV-Vis (Et2O, 25 °C, nm(cm M )) 426(530), 532(280), 774(1200),

908(1200).

Reaction of [Me2NN]Cu-NPh2 with 1 equiv NOgas. A solution of [Me2NN]Cu-NPh2

(0.040 g, 0.072 mmol) in 600 L of benenze-d6 was added to a NMR tube and capped with a rubber septa. The sealed NMR tube was brought outside the glovebox and NOgas

(1.75 mL, 0.072 mmol) was added. A 1H and 13C NMR were immediately taken on the sample to show the formation of Ph2NNO along with some [Me2NN]H and other

1 paramagnetic peaks. NMR data for Ph2NNO: H NMR (benzene-d6, 400 MHz, 25 °C): δ

7.199 (d, 2, o-ArH), 6.924-6.857 (m, 6, ArH), 6.666 (d, 2, o΄ -ArH). 13C{1H}NMR

(benzene-d6): δ 143.22, 137.35, 129.65, 129.39, 129.06, 127.82, 126.59, 119.72. No attempt was made to quantify the amount of Ph2NNO produced from the reaction since

II the copper in the solution will react with the nitrosamine and form [Me2NN]Cu (ON-

I Me2NN)2Cu . Thus the longer the reaction of NOgas with [Me2NN]Cu-NPh2 is allowed to stand, the more Ph2NNO will be consumed by copper.

Reaction of {[Me2AI]Cu}2 with Ph2NNO. At -35 °C, Ph2CNNO (0.020 g, 0.102 mmol) in 400 L of benzene-d6 is added drop wise to a stirring solution of {[Me2AI]Cu}2 (0.040 g, 0.051 mmol) in 600 L benzene-d6. The solution immediately turns from 106 orange/yellow to light brown. Analysis of the reaction mixture by both 1H and UV-Vis shows that the Ph2NNO ligand is labile and that the Cu(I) does not reductively cleave the

1 Ph2N-NO bond. The H NMR spectra (Figure 2.13a) shows broadening of the peaks attributed to {[Me2AI]Cu}2 suggesting that the nitrosamine exchanges with the

{[Me2AI]Cu}2 on the NMR timescale. At ~500 nm a new species appears in the UV-Vis spectra (Figure 2.13b). This is due to a small binding constant between {[Me2AI]Cu}2 and 2 equiv of Ph2NNO.

t [Me2AI]NO (4) as a byproduct of reaction of {[Me2NN]Cu}2 with BuONO. At room

t temperature, when 10 equiv Cu of BuONO (0.132 g, 1.280 mmol) in 2 mL Et2O is added to a slurry of {[Me2AI]Cu}2 (0.200 g, 0.256 mmol) in 4 mL Et2O, the solution turns dark brown immediately. The reaction mixture is allowed to stir for 30 min, filtered through Celite, and concentrated to 3 mL. At -35 °C overnight, yellow crystals form and

1 are identified as [Me2AI]NO by X-ray crystallography characterization. H NMR

(benzene-d6): δ 8.696 (s, 1, imine-H), 8.653 (d, 1, Ar-H), 7.019-6.818 (m, 9, Ar-H),

13 1 6.763 (d, 2, Ar-H), 6.553 (d, 1, Ar-H), 2.305 (s, 6, Ar-CH3), 1.621 (s, 6, Ar-CH3) C{ H}

NMR (benzene-d6): δ 159.42, 151.98, 141.23, 136.35, 131.38, 130.03, 129.89, 129.84,

129.00, 128.15, 127.88, 127.36, 127.10, 124.21, 120.60, 18.50, 17.94. IR (cm-1): 2281,

1444 (νNO), 1129 (νNN). We could not get the reaction mixture free of copper and were unable to get elemental analysis on this compound. Attempts at synthesizing this molecule independently under copper-free condtions were unsuccessful.

107 a)

{[Me2AI]Cu)2

1 H NMR benzene-d6

8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1 n20 b)

1 Figure 2.13 a) H NMR and b) UV-Vis reaction of {[Me2AI]Cu}2 + 2 Ph2NNO in benzene-d6 or toluene.

108 [Me2NN]Cu(CNAr) (5). A solution of 2,6-dimethyphenyl isocyanide (0.054 g, 0.408 mmol) in 2 mL ether was added to a slurry of {[Me2NN]Cu}2 (0.150 g, 0.204 mmol) in 4 mL ether. The insoluble material went completely into the solution and the color remained light yellow. The solution was allowed to stir for 15 mins. The volatiles were removed in vacuo and the resulted residue was crystallized from cold pentane at –35°C overnight to give 0.190 g (93 %) of the product as yellow crystals suitable for X-ray

1 diffraction. H NMR (benzene-d6, 300 MHz, 25 °C): δ 6.884 (d, 4, m-ArH), 6.732 (t, 2, p-

ArH), 6.594 (t, 1, p-CNArH), 6.394 (d, 2, m-CNArH), 5.037 (s, 1, CHbackbone), 2.405 (s,

13 1 12, o-ArMe2), 1.810 (s, 6, Mebackbone), 1.628 (s, 6, o-CNArMe2) C{ H} NMR (benzene- d6): δ 165.89, 162.64, 153.03, 136.81, 135.65, 134.90, 130.27, 128.00, 127.49, 122.77,

-1 94.34, 22.39, 19.29, 18.00. IR (cm ): 2124 (νC≡N).

[Me2AI]Cu(CNAr) (6). A solution of 2,6-dimethyl isocyanide (0.050 g, 0.384 mmol) in

2 mL ether was added to a slurry of {[Me2AI]Cu}2 (0.150 g, 0.192 mmol) in 4 mL ether.

The insoluble material went completely into the solution and the color remained bright yellow. The solution was allowed to stir for 15 mins. The volatiles were removed in vacuo and the resulted residue was crystallized from cold pentane at –35°C overnight to

1 give 0.157 g (78 %) of the product as yellow crystals. H NMR (benzene-d6, 300 Mz, 25

°C ): δ 7.968 (s, 1, imine-H), 7.423-6.318 (m, 13, Ar-H overlapping CNArH), 2.482 (s, 6,

-1 o-Ar-Me2), 2.310 (s, 6, o-Ar-Me2), 1.576 (s, 6, o-CNAr-Me2). IR (cm ): 2131 (νC≡N).

109 DFT calculation details

The DFT calculations employed the Becke-Perdew exchange correlation

functional [1] using the Amsterdam Density Functional suite of programs (ADF 2007.01)

[2]. Slater-type orbital (STO) basis sets employed for H, C, and N atoms were of triple-ζ quality augmented with two polarization functions (ZORA/TZ2P) while an improved triple-ζ basis set with two polarization functions (ZORA/TZ2P+) was employed for the

Cu atom. Scalar relativistic effects were included by virtue of the zero order regular approximation (ZORA) [3]. The 1s electrons of C and N as well as the 1s – 2p electrons of Cu were treated as frozen core. The VWN (Vosko, Wilk, and Nusair) functional was used for LDA (local density approximation) [4]. Default convergence (∆E = 1 × 10-3

-2 -2 hartree, max. gradient = 1 × 10 hartree / Å, max. Cartesian step = 1 × 10 Å) and integration (4 significant digits) parameters were employed for geometry optimizations.

Experimental X-ray coordinates for [Me2NN]Cu-NPh2 (2) were used as the starting point for the geometry optimization of [Me2NN]Cu-NPh2 in an unrestricted (S =

1/2) calculation specifying 1 unpaired electron (spin α – spin β). ADFview [2a] was used to prepare the three-dimensional representations of the structures as well as to render the frontier Kohn-Sham MOs which appear in Figure 2.9.

110 References for calculations

[1] (a) Becke, A. Phys. Rev. A 1988, 38, 3098. (b) Perdew, J. P. Phys. Rev. B 1986,

34, 7406. (c) Perdew, J. P. Phys. Rev. B 1986, 33, 8822.

[2] (a) http://www.scm.com – last accessed July 25, 2008. (b) te Velde, G.;

Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; Van Gisbergen, S. J. A.;

Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931. (c) Fonseca Guerra, C.;

Snijders, J. G.; te Velde, G.; Baerends, E. J.; Acc., T. C. Theor. Chem. Acc. 1998, 99,

391.

[3] (a) Snijders, J. G.; Baerends, E. J.; Ros, P. Mol. Phys. 1979, 38, 1909. (b) Ziegler,

T.; Tschinke, V.; Baerends, E. J.; Snijders, J. G.; Ravenek, W. K. J. Phys. Chem. 1989,

93, 3050. (c) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99,

4597.

[4] Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.

111

II I Figure 2.13. X-band EPR spectrum and simulation for [Me2NN]Cu (ON-Me2NN)2Cu (1) (frozen toluene glass, 90 K). Simulation provides gparallel = 2.20 and gperp = 2.05 with Aparallel(Cu) = 545 MHz and Aperp(Cu) = 74 MHz. Aperp(N) = 31 MHz and Aparallel(N) = 36 MHz was used for 4 nitrogens with similar hyperfine couplings. Data were collected and simulated by Dr. Susanne Mossin of Prof. Karsten Meyer’s research group at the University of Erlangen, Germany.

112

Figure 2.14. X-band EPR spectrum and simulation for [Me2NN]CuI (3) (frozen toluene glass, 12 K). Simulation provides g1 = 2.175, g2 = 2.215, g3 = 2.030 with A1(Cu) = 410 MHz, A2(Cu) ≈ A3(Cu) ≈ 40 MHz and Aiso(N) = 43 MHz for 2 N atoms with similar hyperfine coupling. Data were collected and simulated by Dr. Susanne Mossin of Prof. Karsten Meyer’s research group at the University of Erlangen, Germany.

113 References

(1) Geuther, A. Liebigs Annalen 1863, 128, 151.

(2) Pechmann, H. V. Chem. Ber. 1985, 28, 855.

(3) Magee, P. N.; Barnes, J. M. Br. J. Cancer 1956, 10, 114.

(4) Magee, P. N.; Hultin, T. Biochem. J. 1962, 83, 106.

(5) Magee, P. N.; Farber, E. Biochem. J. 1962, 83, 114.

(6) Lijinsky, W. Chemistry and Biology of N-nitroso Compounds; Cambridge University

Press: Cambridge, 1992.

(7) Lippton, H. L.; Gruetter, C. A.; Ignarro, L. J.; Meyer, R. L.; Kadowitz, P. J. J. Phys.

Pharmacol. 1982, 60, 68.

(8) DeRubertis, F. R.; Craven, P. A. Science 1976, 193, 897.

(9) Zhu, X. Q.; He, J. Q.; Li, Q.; Xian, M.; Lu, J.; Cheng, J.-P. J. Org. Chem. 2000, 65,

6729..

(10) Pavlos, C. M.; Boppana, P. K.; Toscano, J. P. 2005, #WO 2005/074598.

(11) Nitrosamines: Toxicology and Microbiology; VCH Ellis Horwood Ltd.: Chichester,

England, 1988.

(12) Yi, G.-B.; Khan, M. A.; Richter-Addo, G. B. J. Am. Chem. Soc. 1995, 117, 7850.

(13) Yi, G. B.; Khan, M. A.; Richter-Addo, G. B. Inorg. Chem. 1996, 35, 3453.

(14) Chen, L.; Yi, G. B.; Wang, L. S.; Dharmawardana, U. R.; Dart, A. C.; Khan, M. A.;

Richter-Addo, G. B. Inorg. Chem. 1998, 37, 4677.

(15) Richter-Addo, G. B. Acc. Chem. Res. 1999, 32, 529.

(16) Klement, U. Acta Crystallogr. 1969, B25, 2460. 114 (17) Klement, U.; Schmidpeter, A. Angew. Chem. Int. Ed. Engl. 1968, 7, 470.

(18) Schmidpeter, A.; Noth, H. Inorg. Chim. Acta 1998, 269, 7.

(19) Chien, J. C. W. J. Am. Chem. Soc. 1969, 91, 2166.

(20) Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 6037.

(21) Torres, J.; Cooper, C. E.; Wilson, M. T. J. Biol. Chem. 1998, 273, 8756.

(22) Torres, J.; Svistunenko, D.; Karlsson, B.; Cooper, C. E. J. Am. Chem. Soc. 2002,

124, 963.

(23) Hoshino, M.; Maeda, M.; Konishi, R.; Seki, H.; Ford, P. C. J. Am. Chem. Soc.

1996, 118, 5702.

(24) Reichenbach, G.; Sabatini, S.; Palombari, R.; Palmerini, C. A. Nitric Oxide 2001, 5,

395.

(25) Cabail, M. Z.; Pacheco, A. A. Inorg. Chem. 2003, 42, 270.

(26) Tran, D.; Skeleton, B. W.; White, A. H.; Laverman, L. E.; Ford, P. C. Inorg. Chem.

1998, 37, 2505.

(27) Fernandez, B. O.; Lorkovic, I. M.; Ford, P. C. Inorg. Chem. 2003, 42, 2.

(28) Fernandez, B. O.; Ford, P. C. J. Am. Chem. Soc. 2003, 125, 10510.

(29) Tsuge, K.; DeRosa, F.; Lim, M. D.; Ford, P. C. J. Am. Chem. Soc. 2004, 126, 6564.

(30) Xu, Y.; Yun, Z.; Zhu, J.; Xu, J.; Liu, H.; Wei, Y.; Huib, J. Chem. Commun. 2003,

1984.

(31) Xu, Y.; Liu, H.; Zhu, J.; Yun, Z.; Xu, J.; Cao, Y.; Wei, Y. New J. Chem. 2004, 28,

244.

115 (32) Melzer, M. M.; Jarchow-Choy, S.; Kogut, E.; Warren, T. H. Inorg. Chem. 2008, 47,

10187.

(33) Werner, W.; Depreux, P. J. Chem. Res-S. 1980, 11, 372.

(34) Amisial, L. T.; Dai, X.; Kinney, R. A.; Krishnaswamy, A.; Warren, T. H. Inorg.

Chem. 2004, 43, 6537.

(35) Aboelella, N. W.; Gherman, B. F.; Hill, L. M. R.; York, J. T.; Holm, N.; Young Jr,

V. G.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2006, 128, 3445.

(36) Dai, X., Georgetown University, 2004.

(37) Badiei, Y. M., Georgetown University, 2009.

(38) Mankad, N. P.; Antholine, W. E.; Szilagyi, R. K.; Peters, J. C. J. Am. Chem. Soc.

2009, 129.

(39) Nanthakumar, A.; Miura, J.; Diltz, S.; Lee, C.-K.; Aguirre, G.; Ortega, F.; Ziller, J.

W.; Walsh, P. J. Inorg. Chem. 1999, 38, 3010.

(40) Han, H.; Park, S. B.; Kim, S. K.; Chang, S. J. Org. Chem. 2008, 73, 2862.

(41) Harkins, S. B.; Mankad, N. P.; Miller, A. J. M.; Szilagyi, R. K.; Peters, J. C. J. Am.

Chem. Soc. 2008, 130, 3478.

(42) Elizabeth D. Blue, E. D.; Davis, A.; Conner, D.; Gunnoe, T. B.; Boyle, P. D.;

White, P. S. J. Am. Chem. Soc. 2003, 125, 9435.

(43) Goj, L. A.; Blue, E. D.; Delp, S. A.; Gunnoe, T. B.; Cundari, T. R.; Pierpont, A.

W.; Petersen, J. L.; Boyle, P. D. Inorg. Chem. 2006, 45, 9032.

(44) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 1999, 121, 7270.

116 (45) Jazdzewski, B. A.; Holland, P. L.; Pink, M.; Young, J., V. G.; Spencer, D. J. E.;

Tolman, W. B. Inorg. Chem. 2001, 40, 6097.

(46) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 2000, 122, 6331.

(47) Randall, D. W. G., S. D.; Hedman, B.; Hodgson, K. O.; Fujisawa, K.; Soloman, E.

I. J. Am. Chem. Soc. 2000, 122, 11620.

(48) Hayes, P. G.; Welch, G. C.; Emslie, D. J. H.; Noack, C. L.; Piers, W. E.; Parvez, M.

Organometallics 2003, 22, 1577.

(49) Stamler, J. S. Cir. Res. 2004, 94, 414.

(50) Badiei, Y. M.; Dinescu, A.; Dai, X.; Palomino, R. M.; Heinemann, F. W.; Cundari,

T. R.; Warren, T. H. Angew. Chem. Int. Ed. Engl. 2008, 47, 9961.

(51) So, J. H.; Boudjouk Inorg. Chem. 1990, 29, 1592.

(52) Badiei, Y. M.; Krishnaswamy, A.; Melzer, M. M.; Warren, T. H. J. Am. Chem. Soc.

2006, 128, 15056.

(53) Spencer, D. J. E.; Reynolds, A. M.; Holland, P. L.; Jazdzewski, B. A.; Duboc-Toia,

C.; Le Pape, L.; Yokota, S.; Tachi, Y.; Itoh, S.; Tolman, W. B. Inorg. Chem. 2002, 41,

6307.

(54) Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1998, 37, 2046.

(55) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem., Int. Ed. Engl. 1995,

34, 1348.

(56) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roman,

L. M. J. Org. Chem. 1999, 64, 5575.

117 Chapter 3

Release of NO from S-Nitrosothiols and Organic Nitrites at a

β-Diketiminato Cu(I) Complex

Abstract

An electron-rich copper(I) β-diketiminate cleaves the O-NO bond of the organic nitrite

t BuONO and the S-NO bond of the S-nitrosothiol Ph3SCNO to release NO and give the

t three coordinate copper(II) complexes [Me2NN]Cu-O Bu and [Me2NN]Cu-SCPh3. These copper(II) complexes are unstable in the presence of NO – flushing the system with N2 immediately after the addition of the organonitroso compound to remove the generated

NO(gas) allows for their quantification in 91 and 73 % yield, respectively. [Me2NN]Cu-

t O Bu reacts with NO to give ½ equiv {[Me2NN]Cu}2(-OH)2 while [Me2NN]Cu-SCPh3 reacts with NO to give a new copper(II) complex possessing two β-diketiminate ligands, one of which has undergone nitrosation at the β-diketiminate backbone methine C-H.

Introduction

Many NO related biological functions such as anti-platelet aggregation, vasodilation, and pre-eclampsia1 have been directly associated with RSNOs, either via the release of NO or as the RSNO itself.2 These compounds are believed to be involved in the storage and transport of NO and are considerably less oxygen sensitive than NO itself.3-6

118 There is much evidence to support NO as the endothelium derived relaxing factor

(EDRF). New studies are emerging, however, that support the EDFR to be a more stable source of NO, namely an S-nitrosothiol.7 Provided that facile pathways to release NO from S-nitrosothiols are available, NO can be deemed to circulate in the blood primarily

as RSNO species. Free NO has a blood concentration of < 0.012 nM at aPm = 4.5 cm/s

8 and Hct = 45% (Pm-permeability of red blood cell to NO, Hct-hematocrit). Thus, it is unlikely that NO is directly exported or generated from the red blood cell as an intravascular signaling molecule because its concentration would be too low to exert a physiological role. The nitroso adduct of serum albumin (S-nitrosoalbumin, MWT =

66000 Da) has a blood concentration of 5 M and composes about 80 % of the blood

SNO O H HO2C N CO2H NO N H O NH2 Free NO Snitrosocysteineglutathione 4 nM 0.020.2 M

SNO SNO

H2N CO2H

Snitrosocysteine AbluminCys34SNO 0.20.3 M 5 M

Figure 3.1. Concentration of NO and endogenous S-nitrosothiols in blood.

119 plasma RSNO content (Figure 3.1).9 Two other endogenous sources of RSNO are S- nitrosocysteine and S-nitrosocysteingluathione and have a blood concentration of 0.2 –

0.3 and 0.02 - 0.2 M, respectively (Figure 3.1).10-12 Administration of exogenous S- nitroso-albumin leads to vasorelaxation, a decrease in mean arterial blood pressure, and inhibition of platelet aggregation.13,14 Nitric oxide synthase (NOS) inhibitors lead to a decrease in the concentration of plasma RSNOs, vasoconstriction, and an increase in mean arterial blood pressure.15 This suggests that S-nitroso-albumin has a role functioning as a NO donor. Moreover, its formation is also dependent on NOS activity.

Kinetic studies on the decomposition of RSNO compounds in aqueous solutions did not yield consistent results until Williams, Butler, and co-workers reported that even trace amounts of copper ions (even from distilled water) catalyzed the decomposition of

RSNOs to NO and ½ equiv RSSR (Scheme 3.1).16 Williams suggested in a later study the possible biological relevance of the Cu+-induced decomposition of RSNO with the observation that Cu2+ ions complexed to amino acids, peptides, and proteins also

+ 17 decompose RSNOs via initial reduction by thiolate to Cu ..

Scheme 3.1. Activation of the S-NO bond by trace amounts of Cu ions in solution yield NO and ½ equiv of RSSR. Copper-zinc superoxide dismutase (CuZn-SOD) (Figure 3.2)18 has been considered as a possible Cu-enzyme that catalyzes NO release from RSNOs in vivo.19

The decomposition of GSNO by CuZn-SOD resulted in the sustained production of NO.

It was proposed that GSH reduces enzyme-associated Cu2+ to Cu1+ which mediates the 120 His44 His69

Asp81 His118 Cu Zn

His78 His46 His61

Figure 3.2. X-ray structure of active site of CuZn-SOD showing coordination of Cu to four histines, and Zn to three histines and one aspartic acid. Figure drawn using Open Source PyMolTM Version 0.99rc6 2007, PDB 1Q0E, Hough, M. A.; Hasnain, S. S. Structure 2003, 11, 937-946. reductive decomposition of the S-nitrosothiol to yield free NO. Thus, copper(I) appears

to be the active oxidation state for NO release from RSNOs. In fact, neocuproine (2,9-

dimethyl-1,10-phenanthroline) which is a potent chelator for copper(I) halts this

enzymatic denitrosylation.19

CuZn-SOD may represent an important physiological modulator of steady-state concentrations of low molecular weight S-nitrosothiols in vivo.19 Specifically, when glutathione is in the presence of CuZnSOD, S-nitrosoglutathione formation can be detected.19 Thiols do not form S-nitrosothiols with NO under anaerobic conditions. This implicates copper in having a role in the cleavage and formation of the RS-NO bond

(Scheme 3.2). Further studies of the copper ion-induced RSNO decomposition have confirmed the role of Cu+ in decomposition kinetics and rat vascular smooth muscle relaxation.2,20

121 Recent data suggest that either excessive or deficient levels of protein S- nitrosylation may contribute to disease. Mutations in CuZn-SOD can lead to RSNO depletion and disrupt subcellular localization of proteins that are regulated by S- nitrosylation such as glyceraldehyde-3-phosphate dehydrogenase (GAPDH).21 These results suggest that abnormal depletion of intracellular RSNOs contributes to motor neuron death and thereby contribute to amyotrophic lateral sclerosis (ALS). These initial findings raise the possibility that deficient S-nitrosylation is a general mechanism of disease pathogenesis.22 For instance, Parkin is an E3 ubiquitin ligase involved in the ubiquitination of proteins that are important in the survival of dopamine neurons in

Parkinson’s disease (PD). Parkin is S-nitrosylated in vitro as well as in vivo in a mouse model of PD and in brains of patients with PD and diffuse Lewy body disease. Moreover,

S-nitrosylation inhibits Parkin’s ubiquitin E3 ligase activity and its protective function.

This inhibition of Parkin’s ubiquitin E3 ligase activity by S-nitrosylation could contribute to the degenerative process in these disorders by impairing the ubiquitination of Parkin substrates.23

Scheme 3.2. Role of CuZnSOD in formation of GS-NO bond (left) and activation of low molecular weight RSNOs to release NO (right). 122 Organic nitrites RONO are related to S-nitrosothiols and exhibit related vasodilation behavior. For instance, butyl nitrite, , tert-butyl nitrite, , and isoamyl nirite have been clinically used as vasodilators and can generate NO in vivo.24 RONOs are less prone to homolytic “self-decomposition” than RSNOs

(homolytic BDE for RO-NO = 36-41 kcal/mol)25 due to the relative weakness of the RS-

NO bond (31-32 kcal/mol)26 and strength of RS-SR bond (65-66 kcal/mol) (Scheme

3.3).27 Organic nitrites are good sources of NO+. In fact, S-nitrosthiols can be synthesized from the nitrosation of thiols RSH by RONO (Scheme 3.3). S-nitrosothiols can be formed from organic nitrites in vivo since the NO+ portion of RONO can be transferred to a thiol via transnitrosation.28,29 In addition, both organic nitrites and S- nitrosothiols are both susceptible to NO loss via a 1-electron reduction.30-32

Scheme 3.3. Release of NO from RONOs requires a 1-electron reductant while homolysis can release NO from RSNOs.

The uptake of NO as NO+ by a nucleophilic ligand - reductive nitrosylation - has been observed for copper complexes (Scheme 3.6). The reaction of the five-coordinate

2+ cupric complex Cu(dmp)2(H2O) with NO in methanol leads to the formation of the

+ four-coordinate cuprous species Cu(dmp)2 along with and (presumably)

33,34 H2O released in solution (Scheme 3.4). An N-bound RONO ligand in

+ [(dmp)2CuN(O)OR] has been proposed based on kinetic considerations as an

123 intermediate in the reaction of a Cu2+(ROH) complex with NO. Ford also reports an

2+ intramolecular pathway in which a secondary amine coordinated to Cu is nitrosylated to an N-nitroso secondary amine.35 In general, reductive nitrosylation of a metal-bound ligand bond with concomitant reduction of the metal is fairly common, especially for Fe3+ and Cu2+.36

2+ Scheme 3.4. Reductive nitrosylation of MeOH and NO to Cu(dmp)2(H2O) to yield + MeONO and Cu(dmp)2 .

Electron-rich metalloenzymes may serve as 1-electron reducing agents and motivate studies of model complexes to understand factors that result in cleavage of the

E-NO bond and concomitant formation of NO (Schemes I.12 and 3.5). For instance, we recently reported a β-diketiminate nickel(I) system to cleanly reductively cleave the E-

- - - NO bond (E= RS , RO , and R2N ) and form Ni(II)-E and Ni-NO in the presence of two

Ni(I) fragments.37 Understanding the process by which E-NO compounds deliver NO is crucial for the effective treatment of diseases associated with disrupted regulation of NO and the design of future NO delivery drugs. 124 Scheme 3.5. Activation of E-NO bond requires a 1-electron reduction to release NO and E-.

Scheme 3.6. Equilibrium of a) reductive nitrosylation (metal reduced) and b) reductive cleavage (E-NO bond reduced). Results and Discussion

t 3.1. Reaction of BuONO with {[Me2NN]Cu}2.

t 38 When 2 equiv. BuONO are added to yellow {[Me2NN]Cu}2 (1) in toluene at -

30 °C, the solution immediately turns red from the growth of a UV-vis band at λmax =

409 nm (Figure 3.3). Upon warming to room temperature, a brown precipitate of

39 {[Me2NN]Cu}2(-OH)2 is isolated in 90 % yield. Characterization of the reaction mixture by GC/MS and 1H NMR reveals tBuOH as a byproduct of the reaction. Since we

t anticipated the formation of [Me2NN]Cu-OBu (2) (Scheme 3.7), we sought its synthesis

t t by a related oxidative route. Addition of 1 equiv. BuOOBu to {[Me2NN]Cu}2 (1) in

t toluene results in a similar UV-Vis spectra (Scheme 3.8). [Me2NN]Cu-OBu may be isolated on a preparative scale employing tBuOOBut in 86 % as red crystals from pentane

-1 -1 (UV-Vis λmax = 409 nm (1500 cm M )).

125 t Scheme 3.7. [Me2NN]Cu-OBu (2) forms from reaction of ½ equiv {[Me2NN]Cu}2 (1) t t and BuONO. [Me2NN]Cu-OBu (2) may also be prepared from the reaction of t t {[Me2NN]Cu}2 and BuOO Bu.

t Scheme 3.8. [Me2NN]Cu-OBu (2) is not stable in the presence of NO and decomposes to form ½ equiv. {[Me2NN]Cu}2(-OH)2. This copper(II)-alkoxide species is three coordinate, related in structure to other

β-diketiminato Cu(II)-phenoxide complexes prepared by Tolman (Figure 3.5).40 The X- ray crystal structure of 2 shows that the bulky tert-butyl group forces the copper(II)- alkoxide into a planar three coordinate environment despite the affinity of oxygen to

39 bridge in the related hydroxide {[Me2NN]Cu}2(µ-OH)2. The copper center has a Cu-O

126 t t Figure S#.3.3 .Reaction Reaction of ½of {[Me½ {[Me2NN]Cu}2NN]Cu}2 + BuONO,2 with BuONO0.33 mM, (0.33 0 °C, mM)10 s scans in toluene for 30 mins.at 0 °C. t 1) {[Me2NN]Cu}2 before any BuONO is added. 2) Reaction of {[Me2NN]Cu}2 with t t BuONO, [Me2NN]Cu-OBu (2) is formed and decomposed due to the presence of NOgas. 3) After 30 min, 2 decomposed to{[Me2NN]Cu}2(-OH)2 (λmax = 598 nm)

distance of 1.788(2) Å and Cu-Nβ-dik distances of 1.879(2) and 1.890(2) Å. The Cu atom is planar (sum of angles about Cu = 359.8(2)°). The alkoxide atom is bent, with a Cu-O-C angle of 123.82(12)°. Tolman’s four β-diketiminate Cu(II)-phenoxide structures have an average Cu-Ophenoxide distance of 1.810(5) Å, which is longer than the the Cu-Oalkoxide

distance in 2.

t 3.2. Reaction of BuONO with {[Me2NN]Cu}2, flushed with N2(gas).

t If [Me2NN]Cu-OBu is stable in solution overnight at room temperature (Scheme

3.7, second reaction), why is it such short lived (t1/2 ≈ 10 - 30 s at 0 °C) as monitored by

127 t UV-Vis when formed in the reaction between BuONO and 1 (Figure 3.3)? The NOgas released during the reductive cleavage of tBuONO must have a role in the decomposition

t of [Me2NN]Cu-OBu to {[Me2NN]Cu}2(-OH)2. In a sealed UV-Vis cuvette we added 2

t equiv. BuONO to a solution of {[Me2NN]Cu}2 in toluene at 0 °C. After the reagent was added we immediately flushed the cuvette with nitrogen gas for 30 min and let the cuvette sit at 0 °C for 2 hr. We then took at UV-Vis spectra and found that [Me2NN]Cu-

t O Bu had formed in 91 % yield and had not decomposed to {[Me2NN]Cu}2(-OH)2 in the absence of NO (Figure 3.4, spectrum 2). Addition of 1 equiv NO to this solution results in the formation of the hydroxide in 95 % yield within 30 min (Figure 3.4,

t spectrum 3). Both addition of BuONO to 1 as well as addition of NO to [Me2NN]Cu-

t OBu (2) have the final fate of {[Me2NN]Cu}2(-OH)2. When we conducted UV-Vis experiments at -80 °C, we were unable to see any new intermediate formation in the

t addition of BuONO to ½ equiv.{[Me2NN]Cu}2.

128 t Figure 3.4. Reaction of ½ {[Me2NN]Cu}2 with BuONO (0.33 mM, concentrations of t each) in toluene at 0 °C, then flushed with N2 gas. 1) {[Me2NN]Cu}2 before any BuONO t is added. 2) Reaction of {[Me2NN]Cu}2 with BuONO followed by flushing of N2(gas) 10 t min, spectra taken after 2.5 hr. shows [Me2NN]Cu-O Bu (2) is formed and not t decomposed due to the absence of NOgas. 3) Addition of NOgas to [Me2NN]Cu-O Bu leading to the formation of {[Me2NN]Cu}2(-OH)2.

129 N1

C22 Cu O N2

t Figure 3.5. X-ray structure of [Me2NN]Cu-O Bu (2) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.890(2), Cu–N2 1.879(2), Cu–O 1.788(2), N1– Cu–N2 96.17(7), N1–Cu–O 128.70(7), N2–Cu–O 134.94(6), Cu-O-C22 122.82(12).

130 3.3. Reaction of Ph3CSNO with {[Me2NN]Cu}2.

Reaction of {[Me2NN]Cu}2 (1) with the S-nitrosothiol Ph3CSNO initially follows a conceptually related course. Addition of 2 equiv. Ph3CSNO to {[Me2NN]Cu}2 in toluene at 0 °C results in the rapid (< 10 sec) formation of a new species with λmax = 731 nm that decomposes over 30 min to a new species with a λmax = 647 nm. Independent synthesis of the copper(II)-thiolate [Me2NN]Cu-SCPh3 (3) via reaction of [Me2NN]Cu-

t OBu and Ph3CSH confirms the identity of this first species (UV-Vis λmax = 731 nm

(5800 cm-1M-1)) (Scheme 3.9, Figure 3.6)

X-ray characterization of 3 shows a planar, three coordinate copper center despite the propensity of thiolate ligands to form bridging complexes (Figure 3.8). The Cu-Sthiolate distance is 2.137(1) Å and the Cu-Nβ-dik distances are 1.896(2) and 1.907(2) Å. The Cu atom is almost planar (sum of angles about Cu = 356.6(2)°). The thiolate atom is bent,

Scheme 3.9. [Me2NN]Cu-SCPh3 (3) forms in reaction of ½ equiv {[Me2NN]Cu}2 and Ph3CSNO. [Me2NN]Cu-SCPh3 (3) can be independently prepared from [Me2NN]Cu- t OBu and Ph3CSH. [Me2NN]Cu-SCPh3 (3) is not stable in the presence of NO and decomposes to form ½ equiv Ph3CSSCPh3 and ½ equiv {[Me2NN]Cu}([Me2NN]NO) (4).

131 Figure 3.6. Reaction of ½{[Me2NN]Cu}2 with Ph3CSNO in toluene at 0 °C (each at 0.20 mM, 10 s scans collected for 30 mins). 1) {[Me2NN]Cu}2 before any Ph3CSNO is

added. 2) Reaction of {[Me2NN]Cu}2 with 2 Ph3CSNO demonstrating rapid formation of [Me2NN]Cu-SCPh3 (3) with subsequent decomposition owing to presence of NOgas. 3) Addition of NOgas to [Me2NN]Cu-SCPh3 led to the formation of [Me2NN]Cu(NO[Me2NN) (4) in 83 % yield. with a Cu-S-C angle of 116.8(7)°. The UV-vis spectrum of 3 possesses a strong charge

-1 -1 transfer band at λmax = 731 nm (5800 cm M ). Its X-ray structure is similar to Tolman’s

i 41 previously reported [ Pr2NN]Cu-SCPh3 with a Cu-Sthiolate distance of 2.1243(8) Å, slightly shorter than 3.

3.4. Reaction of Ph3CSNO with {[Me2NN]Cu}2, flushed with N2(gas).

[Me2NN]Cu-SCPh3 (3) is unstable in the presence of NO. If a UV-Vis cuvette is flushed with N2 gas for 10 min immediately following the addition of Ph3CSNO to

{[Me2NN]Cu}2 (1) in toluene at 0 °C, we found that [Me2NN]Cu-SCPh3 (3) formed in 73

% spectroscopic yield without the formation of the band at λmax = 647 nm (Figure 3.7, 132 Spectrum 2). Addition of NO to [Me2NN]Cu-SCPh3 leads to rapid loss of this species

(Figure 3.7, Spectrum 3). We were unable to see any new intermediate when monitored by UV-vis spectroscopy in the addition of Ph3CSNO to ½ equiv of {[Me2NN]Cu}2 at -80

°C.

Figure 3.7. Reaction of ½{[Me2NN]Cu}2 with Ph3CSNO in toluene at 0 °C ( each at 0.20 mM) then flushed with N2 gas. 1) {[Me2NN]Cu}2 before any Ph3CSNO was added.

2) Reaction of {[Me2NN]Cu}2 with Ph3CSNO followed by flushing of N2(gas) for 10 min[Me2NN]Cu-SCPh3 (3) was formed and not decomposed due to the absence of NOgas. 3) Addition of NOgas to [Me2NN]Cu-SCPh3 led to the disappearance of the peak at 625 nm.

133 3.5 Isolation of decomposition products in reaction of Ph3CSNO with

{[Me2NN]Cu}2.

The reaction of ½ equiv. {[Me2NN]Cu}2 with Ph3CSNO on a preparative scale leads to formation of a green solution with a white precipitate in Et2O. The precipitate

13 determined to be Ph3CSSCPh3 in 82 % yield based on C NMR characterization and its independent synthesis. Careful crystallization of the green solution using Et2O / TMS2O allows for the isolation of green crystals of [Me2NN]Cu(NO[Me2NN]) (4) in 70 % yield

(Figure 3.9).

X-ray characterization of 4 shows a distorted square planar geometry about the copper center with N1-Cu-N3 and N2-Cu-N5 angles of 107.30(11)° and 98.25 (11)°, respectively (Figure 3.9). The Cu-N5 and Cu-NO distances are 1.962(3) and 2.037(3) Å and the Cu-Nβ-dik distances are 1.928(3) and 1.940(3) Å. The dihedral angle between the two separate N-Cu-N planes is 52.54°. The NO and imine C24-N4 distances are 1.278(3) and 1.271(4) Å which supports delocalization about the entire π-system of the κ2- nitrosated β-diketiminate ligand.

During the isolation of the major product 4, trace amounts of the copper(I) thiolate [CuSCPh3]4 (5) (Figure 3.10) and {[Me2NN]Cu}2(-OH)2 are also identified by

X-ray crystallography. The X-ray structure of 5 shows a tetranuclear copper center where each copper bound to two tritylthiolate ligands. The Cu1-S1 and Cu1-S2' distances are

2.1710(7) and 2.1674(7) Å and the Cu2-S1 and Cu2-S2 distances are 2.1556(8) and

2.1684(8) Å. This complex is very similar to another homoleptic copper(I) thiolate where

134 the R group is 2,4,6-triisopropylphenyl.42 The Cu1-Cu2 and Cu1-Cu2' distances are

2.7493(5) and 3.1041(8) Å.

135 N1

Cu S C22 N2

Figure 3.8. X-ray structure of [Me2NN]Cu-SCPh3 (3) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.896(2), Cu–N2 1.907(2), Cu–S 2.1377(6), N1–Cu–N2 95.76(8), N1–Cu–S 128.04(6), N2–Cu–S 132.79(6), Cu-S-C22 116.80(7). 136 N1 N3 Cu N5 N2 N4

O

Figure 3.9. X-ray structure of [Me2NN]Cu(NO[Me2NN]) (4) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–N1 1.928(3), Cu-N2 1.940(3), Cu-N3 2.037(3), Cu-N5 1.962(3), O-N5 1.278(3), N1-Cu-N2 96.32(11), N1-Cu-N3 107.30(11), N1-Cu-N5 148.91(11), N2-Cu-N3 135.92(11), N2-Cu-N5 98.25(11), N3-Cu-N5 80.30(11).

137 S1 Cu1 S2'

Cu2 Cu2'

Cu1' S2 S1'

Figure 3.10. X-ray structure of [CuSCPh3]4 (5) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu1-S1 2.1710(7), Cu2-S1 2.1556(8), Cu2-S2 2.1684(8), Cu1-S2' 2.1674(7), Cu1-Cu2 2.7493(5), Cu1-Cu2' 3.1041(5), S1-Cu2-S2 161.95(3), S1-Cu2-Cu1 50.80(2), S2-Cu2-Cu1 127.95(2), S2'-Cu1-S1 171.83(3), S2- Cu1'-Cu2' 137.74(2), S1-Cu1-Cu2 50.30(2), Cu2-S1-Cu1 78.90(3), Cu1'-S2-Cu2 91.43(8). 138 3.6. Mechanistic considerations - comparison with related nickel(I) system

t In both the reaction of BuONO and Ph3CSNO with {[Me2NN]Cu}2, the

t corresponding copper(II) products [Me2NN]Cu-OBu (2) and [Me2NN]Cu-SCPh3 (3) form in 91 % yield and 73 % yield, respectively when NO is flushed out of the system.

This is suggestive of initial binding of the O or S to the copper with subsequent cleavage of the RO-NO and RS-NO bonds to release NO (Scheme 3.10).

This mode of reactivity stands in contrast to that observed in the nickel chemistry described in Chapter 1 in which we see that [Ni](NO) rather than [Ni]-E formation is preferred in reaction with a deficit of organonitroso reagent E-NO. Nonetheless, both CuI and NiI are good one-electron reductants for reductive cleavage of the E-NO bond. The difference lays in the ability of NiI to also serve as a good trap for NO to form the nitrosyl complex [Ni](NO). Copper is less electron-rich and therefore the corresponding

[Cu](NO) formation is not as favorable.

Scheme 3.10. Suggested mechanism of E-NO activation through E coordination to I [Me2NN]Cu to form [Me2NN]Cu(ENO). The [Me2NN]Cu then reductively cleaves the II E-NO bond to form [Me2NN]Cu -E.

139 Table 3.1. Crystallographic parameters for 2, 3, 4, and 5.

Compd. 2 3 4 5

Formula C25H34CuN2O C40H40CuN2S C42H50CuN5O C38H30Cu2S2 Mol. Wt. 442.08 644.34 704.41 677.82 Temp.(K) 100(2) 100(2) 100(2) 100(2) Crystal Block Block Plate Block description Crystal color Red Blue Green Colorless Crystal size 0.0364 0.0528 0.0026 0.0132 (mm3) System Orthorhombic Monoclinic Monoclinic Monoclinic Space group P212121 P2(1)/n P2(1)/c P1(1)/n a (Å) 8.2840(12) 8.4626(10) 14.8262(3) 13.428(10) b (Å) 15.010(2) 19.083(2) 16.682(3) 8.6253(7) c (Å) 19.406(3) 10.7638(13) 15.328(2) 25.8383(19) α (deg) 90 90 90 90 β (deg) 90 107.1400(10) 95.564(2) 93.5980(10) γ (deg) 90 90 90 90 Volume (Å3) 2413.1(6) 1661.1(3) 3738.4(10) 2986.8(4) Z 4 2 4 4 θ range (deg) 1.72-28.25 1.98-27.00 1.39-27.00 1.76-28.00 Measd reflns 22055 14364 32065 26262 Unique reflns 5842 7082 8144 7110 GOF of F2 0.998 0.998 0.948 1.105

R1 ( I > 2σ(I)) 0.0319 0.0296 0.0561 0.0384 wR2 (all data) 0.0751 0.0659 0.1392 0.1054 Largest diff. 0.421 and 0.493 and 0.592 and 0.475 and - peak and hole -0.287 -0.252 -0.598 0.428 (e-.Å-3)

140 Summary

While reductive nitrosylation has become accepted as the norm in the reaction of

copper with nitric oxide to generate nitrosated organic species such as organonitrites, we

demonstrate the first examples of the opposite reaction - reductive cleavage. The

biologically relevant Cu(I) ion contained in the electron rich β-diketiminate ligand and

t cleaves BuO-NO and Ph3CS-NO substrates to release NO and form the corresponding

II t II three coordinate [Cu ]-OBu and [Cu ]-SCPh3 species. The β-diketiminate ligand again is subject to nitrosation by NO at the backbone CH, rendering it non-ideal to explore both reductive cleavage and reductive nitrosylation at a common copper complex. Chapter 4 explores the use of a more robust ligand framework less susceptible to ligand nitrosation: tris(pyrazolyl)borates.

141 Experimental procedures

General experimental details.

All experiments were carried out in a dry nitrogen atmosphere using an MBraun

glovebox and/or standard Schlenk techniques. 4A molecular sieves were activated in

vacuo at 180 ºC for 24 h. Dry toluene and dichloromethane were purchased from Aldrich

and were stored over activated 4A molecular sieves under nitrogen. Diethyl ether and

tetrahydrofuran (THF) were first sparged with nitrogen and then dried by passage through

43 activated alumina columns. Pentane was first washed with conc. HNO3/H2SO4 to remove olefins, stored over CaCl2 and then distilled before use from sodium/benzophenone. All deuterated solvents were sparged with nitrogen, dried over activated 4A molecular sieves and stored under nitrogen. 1H and 13C NMR spectra were recorded on either a Varian 300 MHz or 400 MHz spectrometer (300 or 400 and 75.4 or

100.4 MHz, respectively). All NMR spectra, including Evans method data, were recorded at room temperature unless otherwise noted and were indirectly referenced to TMS using residual solvent signals as internal standards. GC-MS spectra were recorded on a Varian

Saturn 2100T, elemental analyses were performed on a Perkin-Elmer PE2400 microanalyzer in our laboratories, and UV-Vis spectra were recorded on either an Agilent

8432 or a Cary 50 spectrophotometer. EPR measurements were carried out by collaborator, Susanne Mossin of Karsten Meyer’s group at the University of Erlangen,

Germany.

142 Tritylthiol, tert-butylnitrite, and tert-butylperoxide were obtained from Acros and

were used as received. Copper tert-butoxide was synthesized according to the published

procedure.44

Preparation of compounds.

{[Me2NN]Cu}2 (1). {[Me2NN]Cu}2 was prepared according to a modified literature

38 procedure. [Me2NN]H (2.00 g, 6.536 mmol) in 10 mL of toluene was adding to a stirring solution of copper(I) tert-butoxide (1.071 g, 7.843) in 25 mL of toluene. The solution was allowed to stir for 2 hr, filtered over Celite, and the volatiles were removed in vacuo to yield brown oil. The brown oil was washed with pentane (3 × 8 mL) to afford

1 2.25 g (92 %) of {[Me2NN]Cu}2 as a yellow powder. H NMR (benzene-d6, 400 MHz, 25

°C): δ 7.100 (d, 4, m-ArH), 6.993 (t, 2, p-ArH), 4.776 (s, 1, CHbackbone), 2.016 (s, 12, o-

ArMe2), 1.643 (s, 6, Mebackbone).

39 Isolation of {[Me2NN]Cu}2(-OH)2 from reaction of {[Me2NN]Cu}2 (2) with

tBuONO. At -30 °C, tBuONO (0.100 g, 0.972 mmol) in 2 mL of toluene was added dropwise to a stirring solution of {[Me2NN]Cu}2.(0.300 g, 0.405 mmol) in 3 mL of toluene at -35 °C. The solution immediately turned from light tan to red and then to brown within 30 minutes. A brown precipitate formed after 30 min of stirring. This precipitate was isolated in 0.298 g and found to be [Me2NN]Cu2(µ-OH)2 (96 % yield).

UV-Vis (toluene, 25 °C, nm(cm-1M-1) 308(13000), 598(550). Analysis of the reaction mixture by 1H and 13C NMR and GC/MS revealed the fate of the tBuO group to be

143 Concentration (mM)

Figure 310. Beer’s law plot for {[Me2NN]Cu}2(-OH)2 in toluene.

t 1 BuOH. H NMR (benzene-d6): δ 2.132 (br s, 1, Me3CO-H), 1.099 (s, 9, Me3COH)

13 1 C{ H}NMR (benzene-d6): δ 168.20 (Me3COH), 34.42 (Me3COH). GC/MS (m/z): 75

(M+1).

t t t Independent synthesis of [Me2NN]Cu-OBu (2). BuOO Bu (0.159, 1.086 mmol) in 2 mL of Et2O was added to a stirring slurry of {[Me2NN]Cu}2 (0.400 g, 0.543 mmol) in 4 mL of Et2O. The solution slowly turned from tan to dark red over the course of an hour.

After stirring for 2 h, the solution was filtered over Celite and the volatiles were removed in vacuo. The solid residue was extracted with pentane (3 mL), filtered over Celite, and left to crystallize overnight at -30 °C. The red crystals that formed were collected and

144 dried under vacuum to afford a total yield of 0.412 g (86 % yield) of product. UV-Vis

-1 -1 (toluene, 25 °C, nm(cm M ), 409(1500). Anal. Calcd for C25H34CuN2O: C, 67.92; H,

7.75; N, 6.34. Found C, 67.61; H, 7.63; N, 6.38. EPR (X-band, frozen toluene glass, 30

K): Simulation provides g1 = 2.233, g2 = 2.057, g3 = 2.037 with A1(Cu) = 120 G, A2(Cu)

Concentration (mM)

t Figure 3.11. Beer’s law plot of [Me2NN]Cu-O Bu.

≈ A3(Cu) ≈ 0 G and Aiso(N) = 12.9 G.

Ph3CSNO. Trityl-S-nitrosothiol was synthesized according to a modified literature

45 procedure. Ph3SH (0.500 g, 1.812 mmol) was dissolved in 8 mL Et2O and 3 equiv

tBuONO (0.560 g, 5.435 mmol) was added to the stirring solution at -35 ºC. The solution

145 immediately turned dark green and was allowed to stir for 15 minutes. The solvent was

removed in vacuo and the residue was extracted with ether (2 × 5 mL) and filtered. The

solution was concentrated to ca. 2 mL. The solution was left to crystallize at –35˚C

overnight. The mother liquor was removed and the crystals were dried in vacuo to afford at total yield of 518 mg (94 % yield) green crystals. The product was stored in the dark at

1 -35 ºC to avoid disulfide formation. H NMR (benzene-d6): δ 7.368-7.289 (m, 9, Ar-H),

13 1 7.221-7.172 (m, 6, Ar-H) C{ H}NMR (benzene-d6): δ 143.78 (Ar-C1´), 130.00 (Ar-C),

-1 127.98 (Ar-C), 127.37 (Ar-C4´), 75.99 (Ph-C-SNO). IR (cm , 25 °C) υNO = 1490

15 46 (reported value for Ph3CS NO υ15NO = 1514 ). In the a previous report on the IR

-1 spectrum of Ph3CSNO a broad band was reported between 1490 and 1530 cm as being due to rotational isomers of the -SNO group.47 In a later study no evidence for rotational isomers was observed in the crystal structure and the IR spectrum from dark green

15 crystals of either the unlabeled or N-labeled compounds showed only a single υNO vibration.46

Reaction of {[Me2NN]Cu}2 (1) with Ph3CSNO. At -35 °C, Ph3CSNO (124, 0.408 mmol) in 2 mL of toluene was added dropwise to a stirring solution of {[Me2NN]Cu}2

(0.150 g, 0.204). The solution immediately turned from light tan to purple within 10 s.

The solution upon warming to room temperature and stirring for 15 mins turned green/brown. The reaction mixture was filtered over alumina to remove the copper metal containing species and the reaction mixture is analyzed by GC/MS and 13C NMR.

Ph3CSSCPh3 is found to be a product from the reaction mixture (0.184 g, 82% yield).

146 13 48 Ph3CSSCPh3 C NMR (acetone-d6) δ 149.89; 129.75, 129.04, 128.80, 70.20, UV-Vis studies were then conducted to identify the purple and green/brown species.

Independent synthesis of [Me2NN]Cu-SCPh3 (3). At -30 °C, Ph3CSH (0.187 g, 0.678

t mmol) in 2 mL of Et2O was added dropwise to a stirring solution of [Me2NN]Cu-OBu

(0.300 g, 0.678 mmol) in 3 mL of Et2O. The solution immediately turned from red to dark blue. The volatiles were removed in vacuo and the solid residue was dissolved in 3 mL of pentane. The solution was filtered over Celite and left to crystallize overnight at -

30 °C. X-ray characterization of these crystals identified them as [Me2NN]Cu-SCPh3.

After the solution sat in the freezer overnight there was an appreciable amount of white solid in the vial. This was from the decomposition of [Me2NN]Cu-SCPh3 to a Cu(I) species and Ph3CSSCPh3. The instability of the compound prevented elemental analysis or an isolated yield.

t Table 3.2. Reaction of [Me2NN]Cu-O Bu (2) and Ph3CSH to form [Me2NN]Cu-SCPh3 (3) in situ for Beer’s Law analysis. t [Me2NN]Cu-SCPh3 [Me2NN]Cu-O Bu Ph3CSH Toluene (10 mM) (10 mM) 0.28 mM 84 L 84 L 2.832 mL 0.21 mM 63 L 63 L 2.874 mL 0.13 mM 39 L 39 L 2.922 mL 0.05 mM 15 L 15 L 2.970 mL

t In situ UV-Vis spectroscopy of the reaction of [Me2NN]Cu-OBu with Ph3CSH was conducted on this reaction at room temperature to determine value of λmax and molar

t absorptivity for pure 3 (Figure 3.12). [Me2NN]Cu-OBu (0.044 g, 0.100 mmol) was 147 dissolved in 2 mL of toluene. The solution was analytically transferred to a 10.00 mL

volumetric flask and diluted to 10.00 mL with toluene to make 10.00 mL of a 10.00 mM

t stock solution of [Me2NN]Cu-O Bu. Ph3CSH (0.028 g, 0.100 mmol) was dissolved in 2 mL of toluene. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with toluene to make 10.00 mL of a 10.00 mM stock solution of

Ph3CSH. Four solutions of [Me2NN]Cu-SCPh3 were made in varying concentrations,

t 0.28 mM, 0.21 mM, 0.13 mM, 0.05 mM, by adding Ph3SH to [Me2NN]Cu-OBu and diluting to 3 mL in a quartz cuvette. The progress of the reaction was monitored by UV-

t Vis to observe the disappearance of [Me2NN]Cu-OBu and the appearance of

Figure 3.12. Beer’s law plot of [Me2NN]Cu-SCPh3.

148 [Me2NN]Cu-SCPh3. The concentration of [Me2NN]Cu-SCPh3 at maximum growth in the

t UV-Vis spectra was assumed to be equal to the concentration of [Me2NN]Cu-OBu. UV-

Vis (toluene, 25 °C, nm(cm-1M-1)) 421(2200), 483(1700), 571(1600), 731(5800).

UV-Vis reactions.

Stock solution preparation. {[Me2NN]Cu}2 (1) (0.037 g, 0.0500 mmol) was dissolved in 2 mL of toluene. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with toluene to make 10.00 mL of a 5.0 mM stock solution

t of {[Me2NN]Cu}2 and 10 mM of copper(I). BuONO (12.0 µL, 0.100 mmol) was added to a 10.00 mL volumetric flask and toluene was added to make 10.00 mL of a 10.00 mM

t stock solution of BuONO. Ph3CSNO (0.031 g, 0.100 mmol) was dissolved in 2 mL of toluene. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with toluene to make 10.00 mL of a 10.00 mM stock solution of

Ph3CSNO. The solutions were stored in at -30 °C to avoid decomposition.

t UV-Vis monintoring of the reaction of {[Me2NN]Cu}2 with 2 equiv BuONO. To a sealed quartz cuvette fitted with a rubber septa at 0 °C tBuONO (100 L, 10 mM) was added to a solution of {[Me2NN]Cu}2 (100 L, 5 mM, in 2.800 mL toluene) for a total concentration of copper to be 0.33 mM. Scans were taken every 10 s from 1000 nm to

200 nm for a total of 30 mins. A new peak at 409 nm immediately appeared and reached

t its maximum absorbance after 30 s. We attributed this speices to be [Me2NN]Cu-OBu

t t (independent synthesis with {[Me2NN]Cu}2 and BuOO Bu). This species is formed in 84

149 -1 -1 % yield, A409 nm = 0.414, 0.33 mM, ε = 1500 cm M . This peak began to disappear and a new species grew in at 598 nm. We attributed this species to be {[Me2NN]Cu}2(-

-1 OH)2. This species was formed in 85% yield, A598 nm = 0.155, 0.33 mM, ε = 550 cm M

(Figure 3.3).

t UV-Vis monintoring of the reaction of {[Me2NN]Cu}2 with 2 equiv BuONO, N2

Flush. To a sealed quartz cuvette fitted with a rubber septa at 0 °C tBuONO (100 L, 10 mM) was added to a solution of {[Me2NN]Cu}2 (100 L, 5 mM, in 2.800 mL toluene) for a total concentration of copper to be 0.33 mM. After the reactant was added the solution was flushed with nitrogen gas for 30 mins. We let the cuvette sit at 0 °C for 2 hours. We

t then took a UV-Vis spectra of this reaction mixture. We found that [Me2NN]Cu-O Bu

-1 -1 had formed in 91 % yield (A409 nm = 0.450, 0.33 mM, ε = 1500cm M ). This finding

t suggested to us that the NOgas released from the reductive cleavage of the BuO-NO bond

t by Cu(I) decomposed the [Me2NN]Cu-O Bu to {[Me2NN]Cu}2(-OH)2. We confirmed this by adding in 2.0 equiv of NOgas back into this cuvette. After 30 min we took a UV-

Vis spectra and noted that {[Me2NN]Cu}2(-OH)2 had formed (Figure 3.4).

UV-Vis monintoring of the reaction of {[Me2NN]Cu}2 with 2 equiv Ph3CSNO. To a sealed quartz cuvette fitted with a rubber septa at 0 °C Ph3CSNO (60 L, 10 mM) was added to a solution of {[Me2NN]Cu}2 (60 L, 5 mM, in 2.880 mL toluene) for a total concentration of copper to be 0.20 mM. Scans were taken every 10 s from 1000 nm to

200 nm for a total of 30 mins. A new peak at nm immeditaley appeared at 731 nm and 150 reached its maximum absorbance after 10 s (1st scan taken). We attributed this speices to

t be [Me2NN]Cu-SCPh3 (independent synthesis with [Me2NN]Cu-O Bu + Ph3CSNO). This

-1 -1 species was formed in 78 % yield, A731 nm = 0.911, 0.20 mM, ε = 5800 cm M . This peak begins to disappear and a new species grows in at 647 nm. We attributed this species to be [Me2NN]Cu(NO)[Me2NN] (4). This species formed in 83 % yield, A647 nm =

0.140, 0.10 mM, ε = 1700 cm-1M (Figure 3.6).

UV-Vis monintoring of the reaction of {[Me2NN]Cu}2 with 2 equiv Ph3CSNO, N2 flush. To a sealed quartz cuvette fitted with a rubber septa at 0 °C Ph3CSNO (60 L, 10 mM) was added to a solution of {[Me2NN]Cu}2 (60 L, 5 mM, in 2.880 mL toluene) for a total concentration of copper to be 0.20 mM. After the reactant was added the solution was flushed with nitrogen gas for 10 mins. We found that [Me2NN]Cu-SCPh3 had formed

-1 -1 in 73 % yield (A731 nm = 0.845, 0.20 mM, ε = 5800 cm M ). After standing at 0 °C for 2 hours, we then took a UV-Vis spectra of this reaction mixture and 37 % of [Me2NN]Cu-

-1 -1 SCPh3 remained in solution (A731 nm = 0.537, 0.20 mM, ε = 5800 cm M ). This finding suggested to us that the NOgas released from the reductive cleavage of the Ph3CS-NO bond by Cu(I) decomposed the [Me2NN]Cu-SCPh3. 2.0 equiv of NOgas were added tp pure [Me2NN]Cu-SCPh3. We monitored the dissapearance of the peak at 625 nm. A new species grows in with a broad peak around 500 nm and 610 nm. We have not identified this decomposition product. It is however not [Me2NN]Cu(NO[Me2NN]) (4) (Figure 3.6).

151 Synthesis of [Me2NN]Cu(NO[Me2NN]) (4). At -35 °C, Ph3CSNOs (0.248 g, 0.814 mmol) in 2 mL of Et2O was added to a stirring slurry of {[Me2NN]Cu}2 (0.300 g, 0.407 mmol) in 8 mL Et2O. A white precipitate immediately formed (Ph3CSSCPh3) and was removed by filtration through Celite. The volume of the solution was reduced to 3 mL and filtered through Celite once again. The solution was layered with 0.5 mL TMS2O and left to crystallize overnight at -35 °C. The crystals that formed were collected and dried under vacuum to afford a total yield of 0.200 g (70 % yield) of green crystals identified as

-1 -1 [Me2NN]Cu(NO[Me2NN]) (4) by X-ray. UV-Vis (toluene, 25 °C, nm(cm M ), 647

(1700).

Isolation of [CuSCPh3]4. From the reaction of described for the isolation of 4

(Ph3CSNOs (0.248 g, 0.814 mmol) + {[Me2NN]Cu}2 (0.300 g, 0.407 mmol) in 8 mL

Et2O) forms trace amounts (< 10 mg) of colorless crystals. Analysis by X-ray reveals these crysyals to be [CuSCPh3].

152 References

(1) deBelder, M.; Lees, C.; Martin, J.; Moncada, S.; Campbell, S. Lancet 1995, 345,

124.

(2) Williams, D. L. H. Acc. Chem. Res. 1999, 32, 869.

(3) Fujimori, K.; Nakajima, T. Rev. Heteroat. Chem. 2000, 22, 181.

(4) Kerwin, J. F. J.; Lancaster, J. R. J.; Feldman, P. L. J. Med. Chem. 1995, 38, 4343.

(5) Al-Sa'doni, H. H.; Ferro, A. Clin. Sci. 2000, 98, 507.

(6) deBelder, A. J.; MacAllister, R.; Radomski, M. W.; Moncada, S.; Valence, P. J.

Cardiovasc. Res. 1994, 28.

(7) Ignarro, L. J. Nitric Oxide: Biology and Pathology; Academic Press: San Diego,

2000.

(8) Liu, X.; Qingtao Yan, Q.; Baskerville, K. L.; Zweier, J. L. J. Biol. Chem. 2007, 282.

(9) Stamler, J. S.; Jaraki, O.; Osborne, J.; Simon, D. I.; Keaney, J.; Vita, J.; Singel, D.;

Valeri, C. R.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 7674.

(10) Kelm, M. Biochem. Biophys. Acta 1999, 1411, 273.

(11) Stamler, J. S. Cir. Res. 2004, 94, 414.

(12) Orie, N. N.; Vallance, P.; P., D.; Jones, D. P.; Moore, K. P. Am. J. Physiol. Heart

Circ. Physiol. 2005, 289, H916.

(13) Maalej, N.; Albrecht, R.; Loscalzo, J.; Folts, J. D. J. Am. Coll. Cardiol. 1999, 33,

1408.

(14) Hallström1, S.; Franz, M.; Gasser, H.; Vodrazka, M.; Semsroth, S.; Losert, U. M.;

Haisjackl, M.; Podesser, B. K.; Malinski, T. Cariovascular Res. 2008, 77, 506. 153 (15) Zhang, Y.; Hogg, N. Am. J. Physiol. Lung Cell Mol. Physiol. 2004, 287, L467.

(16) McAninly, J.; Williams, D. L. H.; Askew, S. C.; Butler, A. R.; Russell, C. J. Chem.

Soc., Chem. Commun. 1993, 1758.

(17) Dicks, A. P.; Beloso, P. H.; Williams, D. L. H. J. Chem. Soc., Perkin Trans. 2

1997, 1429.

(18) Hough, M. A.; Hasnain, S. S. Structure 2003, 11, 937.

(19) Jourd'heuil, D.; Laroux, F. S.; Miles, A. M.; Wink, D. A.; Grisham, M. B. Arch.

Biochem. Biophys. 1999, 361, 323.

(20) Al-Sa'doni, H. H.; Megson, I. L.; Bisland, S.; Butler, A. R.; Flitney, F. W. 1997,

121, 1047.

(21) Johnson, M. A.; Macdonald, T. L.; Mannick, J. B.; Conaway, M. R.; Gaston, B. J.

Biol. Chem. 2001, 276, 39872– 39878.

(22) Schonhoff, C. M.; Matsuoka, M.; Tummala, H.; Johnson, M. A.; Estevez, A. G.;

Rui Wu, R.; Kamaid, A.; Ricart, K. C.; Hashimoto, Y.; Gaston, B.; Macdonald, T. L.; Xu,

Z.; Mannick, J. B. Proc. Natl. Acad. Sci. USA 2006, 103.

(23) Cung, K. K. K.; Thomas, B.; Li, X.; Pletnikova, O.; Troncoso, J. C.; Marsh, L.;

Dawson, V. L.; Dawson, T. M. Science 2004, 304, 1328.

(24) Cedergvist, B.; Persson, M. G.; Gustafsson, L. E. Biochem. Pharmacol. 1994, 47,

1047.

(25) McMillen, D. F.; Golden, D. M. Ann. Rev. Phys. Chem. 1982, 33, 493.

(26) Bartberger, M. D.; Mannion, J. D.; Powell, S. C.; Stamler, J. S.; Houk, K. N.;

Toone, E. J. J. Am. Chem. Soc. 2001, 123, 8868. 154 (27) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of Organic

Compounds; 2nd ed.; Chapman and Hall: New York, 1986.

(28) Patel, H. M. S.; Williams, D. L. H. J. Chem. Soc. Perkin Trans. 2 1990, 37.

(29) Meloche, B. A.; O'Brien, P. J. Xenobiotica 1993, 23, 863.

(30) Doel, J. J.; Godber, B. L. J.; Goult, T. A.; Eisenthal, R.; Harrison, R. Biochem.

Biophys. Res. Commun. 2000, 270, 880.

(31) Trujillo, M.; Alvarez, M. N.; Peluffo, G.; Freeman, B. A.; Radi, R. J. Biol. Chem.

1998, 273, 7828.

(32) Aleryani, S.; Milo, E.; Rose, Y.; Kostka, P. J. Biol. Chem. 1998, 273, 6041.

(33) Tran, D.; Ford, P. C. Inorg. Chem. 1996, 35, 2411.

(34) Tran, D.; Skeleton, B. W.; White, A. H.; Laverman, L. E.; Ford, P. C. Inorg. Chem.

1998, 37, 2505.

(35) Tsuge, K.; DeRosa, F.; Lim, M. D.; Ford, P. C. J. Am. Chem. Soc. 2004, 126, 6564.

(36) Ford, P. C.; Fernandez, B. O.; Lim, M. P. Chem. Rev. 2005, 105, 2439.

(37) Melzer, M. M.; Jarchow-Choy, S.; Kogut, E.; Warren, T. H. Inorg. Chem. 2008, 47,

10187.

(38) Amisial, L. T.; Dai, X.; Kinney, R. A.; Krishnaswamy, A.; Warren, T. H. Inorg.

Chem. 2004, 43, 6537.

(39) Dai, X.; Warren, T. H. Chem. Commun. 2001, 1998.

(40) Jazdzewski, B. A.; Holland, P. L.; Pink, M.; Young, J., V. G.; Spencer, D. J. E.;

Tolman, W. B. Inorg. Chem. 2001, 40, 6097.

(41) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 1999, 121, 7270. 155 (42) Schröter-Schmid, I.; Strähle, J. Z. Naturforsch. Teil B 1990, 43, 1537.

(43) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J.

Organometallics 1996, 15, 1518.

(44) Badiei, Y. M.; Dinescu, A.; Dai, X.; Palomino, R. M.; Heinemann, F. W.; Cundari,

T. R.; Warren, T. H. Angew. Chem. Int. Ed. Engl. 2008, 47, 9961.

(45) Girard, P.; Guillot, N.; Motherwell, W. B.; Hay-Motherwell, R. S.; Potier, P.

Tetrahedron 1999, 55, 3573.

(46) Arulsamy, N.; Bohle, D. S.; Butt, J. A.; Irvine, G. J.; Jordan, P. A.; Sagan, E. J. Am.

Chem. Soc. 1999, 121, 7115.

(47) Kresze, G.; Uhlich, U. Chem. Ber 1959, 92, 1048.

(48) Cochran, J. C.; Friedman, S. R.; Frazier, J. P. J. Org. Chem. 1996, 61, 1533.

156 Chapter 4

Reaction of S-Nitrosothiols and Nitric Oxide with Tris(pyrazolyl)borate

Copper Complexes

Abstract

Copper proteins in biology have been shown to have an important role in the formation

and cleavage of the S-N bond of S-nitrosothiols. Herein we show that the S-nitrosothiol

RS-NO bond is activated by TpiPr2CuII-SR, TpiPr2CuI, and TpiPr2CuI(NO) model complexes to release NO which is trapped by TpiPr2CuI in solution to form TpiPr2Cu(NO).

The formation of RSSR' from the reaction of TpiPr2CuII-SR with R'SNO points to a nitroxyl disulfide intermediate TpCu[RS(NO)SR']. In the reaction of S-nitrosothiols with the TpiPr2CuI fragment, UV-vis studies indicate that TpiPr2CuII-SR formation precedes

TpiPr2Cu(NO) formation. The electron deficient model complex Tp(CF3)2CuI(THF) does not form Tp(CF3)2CuII-SR upon exposure to RSNO, suggesting that electron-rich copper centers are required to release NO from S-nitrosothiols.

Introduction

Importance of copper in release of NO from S-nitrosothiols.

Nitric oxide is implicated in numerous biological roles ranging from signaling in the respiratory system1 to vasodilation in the cardiovascular system to host defense against microbial pathogens.2 While the various nitric oxide synthases (NOSs) generate

NO under a variety of conditions, NO itself is unstable in the plasma, with an estimated

157 half-life of 3 – 5 seconds.3 On the other hand, oxygen stable S-nitrosothiols (RSNOs) such as S-nitrosocysteine and S-nitrosoglutathione circulate at micromolar levels in the blood.4,5 Capable of serving as NO and NO+ donors,6,7 S-nitrosothiols have been implicated in a wide variety of physiological functions which often mirror those observed for nitric oxide itself.8,9 The nature of the molecular species involved in the biological reactivity of S-nitrosothiols, however, is clouded by the facile decomposition of RSNOs into free NO and disulfides by a copper-catalyzed process (Scheme 4.1).7,10

Scheme 4.1. Catalytic decomposition of S-nitrosothiols by Cu ions to form disulfide and nitric oxide.

For the copper ion-catalyzed RSNO decomposition reaction, the active species is proposed to be Cu+, which is generated by the reduction of Cu2+ by thiolate anion. This has been supported by several studies by Williams and co-workers.11-14 Williams also demonstrated the possible biological relevance of the Cu+-induced decomposition of

RSNOs through Cu2+ complexed to amino acids, peptides, and proteins that can partake in RSNO decomposition via initial reduction by thiolate to Cu+ yield nitric oxide

(Scheme 4.2).15 From this study, Williams proposed what he deems a realistic pathway for the formation of nitric oxide in vivo from S-nitrosothiols using copper sources

Scheme 4.2. The active species Cu+ in RSNO decomposition is generated by the reduction of Cu2+ to Cu1+ by a thiolate anion.

158 present in the body. This pathway could be important mechanistically in providing

an explanation for the role of S-nitrosothiols in the possible storage and delivery of nitric oxide in vivo. These results also provide a mechanism for nitric oxide generation from S-nitrosothiols administered therapeutically, which may lead to the widespread development of S-nitrosothiols as nitric oxide-releasing drugs.

2+ 2+ Scheme 4.2. CuZn-SOD releases Cu in the presence of H2O2. This free Cu can bind two GSNO molecules to release NO and from GSSR.

Figure 4.1. The structure of the copper–nitrosoglutathione complex, Figure taken from Arch. Biochem. Biophys. 1996, 372, 8-15 without permission.

159 Copper-zinc superoxide dismutase (CuZn-SOD) (Chapter 3, Figure 3.2) may

represent an important physiological modulator of steady-state concentrations of low

molecular weight S-nitrosothiols in vivo.16 This enzyme is the major source of copper in red blood cells, present at a concentration of ca 20 M.17 Endothelial CuZn-SOD effectively catalyzes the release of NO from GSNO in the presence of GSH.18 CuZn-SOD regulates NO levels in cells, by the dismutation of superoxide anion to form hydrogen peroxide (Scheme 4.2). Mechanistic studies of the release of NO from GSNO shows that the copper ion released from the Cu,ZnSOD/H2O2 reaction decomposes GSNO and that the γ-glutamyl group of GSNO and GSSG controls NO release from GSNO (Scheme

4.2).19 Copper binding to GSNO is essential to the release of NO (Figure 4.1).

While transnitrosation between S-nitrosothiols and free thiols as well as the corresponding thiolate anions has been observed in a variety of media,20-22 few molecular level details are known for transnitrosation at biologically relevant metal thiolates. Such exchange reactions are thought to proceed via nitroxyl disulfide intermediates [RS

(ON)SR']- theoretically considered by Houk et al.23and observed via 15N NMR spectroscopy by Estrin et al. in the exchange between S-nitrosocycsteine ethyl ester and its thiolate anion in methanol (Scheme 4.3).20

Scheme 4.3. Trans-s-nitrosation for RSNO and RS- with the proposed nitroxyl disulfide intermediate [RSN(O)SR']-. 160 Copper-thiolates and copper-nitrosyls in biology and model systems.

Copper proteins are defined as a protein containing one or more copper atoms and can be classified according to their biological fucntion, Cu center coodination environment, type and number of prosthetic centers, and by sequence similarity. The biological functions are either classified as catalysis or electron transfer. The catalytic copper proteins are commonly oxidases, reductases, dismutases, or hydrolases.

Type 1 Type2 Type 3 Blue Copper CuZnSOD Hemocyanin

R L HisN NHis L L O Cu Cu NHis L L HisN Cu Cu NHis CysS O NHis L HisN NHis L = N, O or S ligands

Cu(N His)2S CysR Cu(N His)mRn O2[Cu(N His)3]2

Trinuclear Center CuA CuB (Type 2 + 3 NO2 Reductase Cytochrome C oxidase Laccase HisN NHis HisN Cys Glu NHis HisN HisN Cu Cu HisN S O NHis CuHO OH Cu Cu L Met S NHis HisN HisN Cu Cys HisN NHis Fe

Cu(N His)2OHOH Cu2(N His)2O Glu [Cu(N His)3]2 S Met(S Cys)2 CuN His(N His)2L Figure 4.2. Family of copper proteins classified by coordination environment around the copper center. 161

There are six types of structurally distinct binding modes — Type 1, Type 2, Type

3, Trinuclear Center (Type 2 + Type 3), CuA, and CuB (Figure 4.2). Of greatest significance to our modeling work described below are Type 1 and CuA proteins which contain copper-sulfur centers surrounded by histidine donor ligands. Type 1 species are often called blue copper proteins. Their blue color originates from intense electronic absorption bands centered at wavelengths from 600 - 625 nm arising from a charge transfer transition from a cysteine thiolate ligand bound to the active site.24 The coordination enviornment about the copper metal gives rise to the low energy transition involving nearly trigonal arragement of two imidazole N atroms and the thiolate S atom.

A methionine ligand can be found along the trigonal axis in azurin and plastocyanin. The geometry can be described as trigonal based, distored tethrahedral.

To study the unique properties of blue copper proteins, many model complexes have been reported.25 The first successful blue copper model was TpMe2Cu(S-p-

Me2 NO2C6H4) synthesized by Ibers and Thompson in 1977 (Tp = hydrotris(3,5-dimethyl-

Figure 4.3. Synthetic models for the Type 1 copper proteins using tris(pyrazylol)borate anionic ligands.

162 a) Cytochrome C oxidase (CuB) b) Cu-nitrite Reductase (end on NO) (side on NO)

Figure 4.4. Two examples of copper nitrosyls found in biological systems. a) Cytochrome C oxidase with a Cu-NO bound end on, Figure taken from Vos, et al. Biochemistry 2001, 40, 7806, used without permission. b) Cu-nitrite reductase with a Cu- NO bound side-on.Tocheva et al. Science, 2004, 304, 867-870.

1-pyrazolyl)borate anion) (Figure 4.3).26 This model complex was able to reproduce the

-1 -1 intense blue color (λmax = 588 nm (3900 M cm )), but no X-ray analysis was carried out on this model complex. In a later study, Kitajima reported two structurally characterized

iPr2 iPr2 Cu(II)-thiolate complexes Tp Cu(SC6F5) (1) and Tp Cu(SCPh3) (2) using the hydrotris(3,5-diisopropyl-1-pyrazolyl) borate anion (Figure 4.3).27 In each case copper centers exhibit a distorted tetrahedral geometry due to the tripodal ligand forcing a distored four coordinate configuration. Both 1 and 2 show an intense blue color owing to an absorbance maximum centered around 600 nm with λmax for 1 and 2 at 672 nm (5960

M-1cm-1) and 625 nm (6600 M-1cm-1), respectively.

Nitric oxide is involved in the regulation of respiration by acting as a competitive ligand for molecular oxygen at the at the iron center binuclear active site of cytochrome c

28 oxidase (CcO). CcO is a large transmembrane protien that serves to convert O2 to two

163 molecules of H2O. The complex contains two , one cytochrome a and one

29 cytochrome a3, and two copper center (CuA and CuB). The dynamics of NO in and near this site are not well understood. Simulated structures of the active site of cytochrome c oxidase have been proposed based on flash photolysis studies. The heme a3 active site can accommodate two NO molecules, one at the Fe and the other at the CuB (Figure 4.3).

This is only a proposed stucture which has not been isolated and unambiguously structurally characterized (Figure 4.4a).28

Nitrite reductase (NiR) is an enzyme that uses type 1 and type 2 copper sites to reduce nitrite to nitric oxide during bacterial denitrification. Few Cu-nitrosyl complexes, including that of NiR, are well characterized, presumably because of their instability and sensitivity to oxygen. A copper-nitrosyl intermediate is a proposed, yet poorly characterized feature of the NiR catalytic cycle. This intermediate is formally described as Cu(I)-NO+ and is proposed to be formed at the type 2 copper site after nitrite binding and electron transfer from the type I copper site.30 The crystal structure of a type 2 copper-nitrosyl complex of nitrite reductase reveals an unprecedented side-on binding mode in which the nitrogen and oxygen atoms are nearly equidistant from the copper cofactor (Figure 4.4b).30 This stable copper-nitrosyl complex is described as Cu(II)-NO- regardless of the starting oxidation state of the metal.31 Isolation of the proposed intermediate Cu(I)-NO+ has not been possible and may require development of single turnover techniques in crystallography.31 The authors have concluded that the isolated

Cu(II)-NO- bound end-on is not relevant for the catalytic cycle of the enzyme.

164 In order to understand the fundamental chemistry of CuNO species, small

molecule model complexes have been synthesized and some structurally characterized.32-

34 The tris(pyrazolyl)borate ligand system provides a good scaffold for isolation and structural characterization of mononuclear Cu-NO complexes, TpR,R'Cu(NO) (Figure

4.5).32-34 These complexes were studied by FT-IR, NMR, and/or UV-Vis in solution and serve to model a possible copper intermediate in NiR during nitrite reduction. The

TpCu(NO) complexes disproportionate in the presence of excess NO to N2O and

R,R' Tp Cu(NO)2. Manometery studies show that the NO binding is weak, reversible, temperature dependent, and that the NO ligand can be irreversibly replaced by MeCN or

CO. For TpPh2Cu(NO) and TptBuHCu(NO) the unpaired electron resides primarily in the

π*(NO) orbital. TpMesHCu(NO) is EPR silent and exhibits an 1H NMR spectrum with sharp diamagnetic signals. TpCF3,CH3Cu(NO) is EPR silent and is not amenable to 1H

NMR characterization as the only species seen in the 1H NMR spectrum is the

TpCF3CH3Cu(MeCN). The EPR spectrum of TptBu,iPrCu(NO) is in agreement with the copper(I)-nitrosyl electronic structure. It is interesting to note that subtle changes in the ligand can lead to significant differences in the stability of the Cu-NO linkage.

165 Figure 4.5. TpR,R'Cu(NO) model complexes.

In light of the rich literature using the tris(pyrazolyl)borate supporting ligands to

isolate Cu(II)-thiolates and Cu(I)-NO complexes, we sought to study how each of these

species would react with S-nitrosothiols.

Results and Discussion

4.1. Synthesis and stability of TpiPrCuII-SR complexes.

iPr2 Addition of 2 equiv. RSH (R = C6F5, CPh3) to {Tp Cu}2(-OH)2 at -20 °C results in an immediate color change from green to dark blue yield TpiPr2Cu-SR in 70 -

78% yields. These complexes are unstable toward air and result in the formation

iPr2 35 iPr2 27 iPr2 27 (Tp Cu}2(µ-CO3). While both the Tp Cu-SC6F5 (1) and Tp Cu-SCPh3 (2) are isolable and able to be characterized by X-ray crystallography, they are short lived in

Scheme 4.4. Synthesis of TpiPr2Cu-SR from reaction of TpiPr2Cu-OH with RSH. 166 iPr Figure 4.6. Decomposition of Tp Cu-SC6F5 (1) in CH2Cl2 at room temperature following second order plot kinetics with respect to 1.

solution and decompose to form the corresponding TpCuI species along with ½ equiv.

RSSR (Scheme 4.4). We believe that the mechanism for decomposition is different for 1 and 2.

iPr When Tp Cu(SC6F5) (1) is allowed to decompose at room temperature in CH2Cl2

(0.30 mM) the plot of 1/Concentration (M-1) vs. time (s) gives a straight line (Figure 4.6).

iPr2 iPr2 I Scheme 4.5. Bimolecular decomposition of Tp Cu-SC6F5 to form 2 equiv. Tp Cu and C6F5SSC6F5.

167 Monitoring the decrease in the band centered at λ = 666 nm in CH2Cl2, concentration data were able to be obtained from its reported molar absorptivity (ε = 5960 M-1 cm-

1 27,36 iPr2 ). These data suggest that Tp Cu-SC6F5 decomposes via a bimolecular pathway.

Formation of a dinuclear complex would facilitate electron transfer to yield 2 equiv.

iPr I Tp Cu and C6F5SSC6F5 (Scheme 4.5).

iPr2 The decomposition pathway for Tp Cu-SCPh3 (2) in CH2Cl2 does not follow the

iPr2 same bimolecular pathway of Tp Cu-SC6F5 (1). The peak at 625 nm decreases over time and a new peak at 353 nm grows in (Figure 4.7). There is an isobestic point at 610 nm indicating that the mechanism for conversion of 2 to the new product at 353 nm does not go through an observable intermediate. We have been unable to identify this new

λ = 625 nm iPr2 Tp Cu-SCPh3 λ = 353 nm TpiPr2CuI?

iPr2 Figure 4.7. Disappearance of band at λ = 625 nm for Tp CuSCPh3 in its thermal decomposition in CH2Cl2 with appearance of new species with λmax = 353 nm.

168 iPr I species with λmax = 353 nm. It does not have the same UV-vis spectrum as does Tp Cu

iPr2 37 (synthesized from Tp Cu(MeCN) followed by tituration with CH2Cl2 / pentane to

iPr2 32 likely form (Tp Cu)2 but is likely a copper(I) species owing to the lack of any strong bands in the visible. The trityl group is quite large compared to that of the pentafluorophenyl group and sterically discourages bimolecular interactions, forcing an

iPr2 alternative pathway for decomposition. When Tp Cu-SC6F5 decomposes it does not form a new peak at 353 nm. Rather, the solution decolorizes and has a UV-Vis spectra like that of TpiPr2CuI.

iPr2 Plots of concentration (M) vs. time (s) for the decomposition Tp Cu-SCPh3 in

CH2Cl2 of are linear over a temperature range of 0 - 40 ºC (Figure 4.8) as judged by

iPr2 Figure 4.8. Plots of concentration vs time for decomposition of Tp Cu-SCPh3 (1) in CH2Cl2 from 0 – 40 °C, zero order kinetics followed. 169 -1 -1 25 monitoring the loss of the band of 2 at λmax = 625 nm with ε = 6600M cm ).

Eyring analysis of the decomposition reaction allows the activation parameters

S‡ and H‡ to be determined from the temperature dependence of the rate constant. A plot of ln(k/T) versus 1/T produces a straight line with the familiar form y = -mx + b

‡ ‡ where x = 1/T, y = 1/T, m = H /R, b = (ln(kb/h) + S /R), (Figure 4.9). The Eyring analysis reveals H‡ = 18 kcal/molK and S‡ = -33 cal/mol. The assumption is made that the activation enthalpy and the activation entropy are constant within this temperature range; however the activation free energy and the rate constant are temperature dependent.

The linear absorbance vs time plots for the decomposition of 2 indicate zero-order kinetics. This suggests a zero order mechanism for decomposition. A zero-order rate law for a chemical reaction means that the rate of the reaction is independent of reactant concentration. We believe the zero order decomposition of 2 to be due to an unknown radical catalyst in the reaction mixture – perhaps dioxygen or some other UV-Vis-silent species.

The entropy of activation S‡ is the entropy difference between the transition state and the ground state of the reactants. A negative value for S‡ signifies that the transition state is more ordered than the ground state for the reactants. We obtained a value of S‡ =

-33 cal/molK. This is suggestive of a more ordered transition state for the decomposition

iPr2 of Tp Cu-SCPh3.

170 iPr2 ‡ Figure 4.9. Eyring plot for Tp Cu-SCPh3 decomposition in CH2Cl2, H = 18 kcal/mol and S‡ = -33 cal/molK.

Both 1 and 2 are literature compounds that have been spectroscopically and structurally characterized by Kitajima. Other than mention of their general instability, no kinetic analysis of their decomposition has been reported in the literature. This is the first study to be reported on the decomposition pathways. Although we believe we uncovered some interesting reactivity information, questions still remain as to the exact mechanism for disulfide formation and the fate of the copper. Of particular relevance is the identification of the catalyst that leads to zero-order kinetics in the decomposition of 2.

iPr2 I 4.2. Reactivity of Tp Cu with RSSR in the presence of O2.

iPr2 I There is no reaction with 2 equiv. of Tp Cu with RSSR in CH2Cl2.. When the same reaction mixture is exposed to air, the solution immediately turns dark blue. UV-

171 iPr2 Vis analysis reveals the identity of the blue solution to be Tp Cu-SR (1 with C6F5S-

SC6F5 or 2 with Ph3CS-SCPh3) (Scheme 4.6,Figure 4.10; Scheme 4.7 Figure 4.11). There

iPr2 38,39 is no evidence of formation of the copper-peroxo complex {Tp Cu}2(-O2) , which

iPr2 I form purple solutions (λmax = 551 nm) when Tp Cu is exposed to O2, 78 % spectroscopic yield.

iPr2 I Scheme 4.6. Reaction of Tp Cu with ½ equiv. C6F5SSC6F5 in the presence of air leads iPr2 to the formation of Tp Cu-SC6F5.

2) λ = 666 nm iPr2 Tp Cu-SC6F5

1) TpiPr2CuI

iPr2 I Figure 4.10. UV-Vis in CH2Cl2 at 25 °C of the reaction of Tp Cu with ½ C6F5SSC6F5 iPr2 in the presence of air leads to the formation of a new peak at 666 nm, Tp Cu-SC6F5.

172 iPr2 I Scheme 4.7. Reaction of Tp Cu with ½ equiv. Ph3CSSCPh3 in the presence of air leads iPr2 to the formation of Tp Cu-SCPh3.

λ = 625 nm iPr2 Tp Cu-SCPh3

iPr2 I Figure 4.11. UV-vis spectra taken in CH2Cl2 at 25 °C of the reaction of Tp Cu with ½ equivl Ph3CSSCPh3 in the presence of air leads to the formation of a new peak at 625 nm, iPr2 Tp Cu-SCPh3.

173 4.3. Reactivity of TpiPrCuII-SR complexes with S-nitrosothiols.

When 1 equiv of tBuSNO is added to benzene or dichloromethane solutions of blue 1, the blue color is quickly discharged. Employing a freshly prepared sample of 1 in

t benzene-d6 free from the disulfide C6F5S-SC6F5, we find that addition of 1 equiv BuSNO

t at room temperature provides the unsymmetrical disulfide C6F5S-SBu (4) as the sole fluorine-containing species in 65% yield (Scheme 4.8). The symmetric disulfide tBuS-

t iPr2 Scheme 4.8. NO release from BuS-NO by Tp Cu-SC6F5. A nitroxyl disulfide t intermediate is proposed based on the formation of C6F5S-S Bu as the major sulfur- containing species.

λmax = 666 nm iPr2 II Tp Cu -SC6F5 λmax = 495 nm TpiPr2CuI(NO)

t Figure 4.12. UV-Vis reaction in CH2Cl2 at 25 °C between 1 and BuS-NO to form t iPr2 C6F5S-S Bu and Tp Cu(NO) (3).

174 SBut accounts for the remainder of the added tBuSNO. Monitoring this reaction in

CH2Cl2 by UV-vis spectroscopy shows the loss of copper-thiolate 1 at λ = 666 nm along with growth of a band at λ = 495 nm which was shown through independent synthesis

iPr2 iPr2 (addition of excess NO to Tp Cu(NCMe) in CH2Cl2) to be Tp Cu(NO) (3) (νNO =

1704 cm-1 from IR analysis) (Figure 4.12).33 The TpiPr2Cu(NO) (3) is not formed in high yields, nor is it long lived. As evident in the UV-Vis spectrum (Figure 4.12), the peak that grows in at 495 nm decomposes faster than the decomposition of 1. Thus by our reaction conditions we cannot monitor the full growth of the TpiPr2Cu(NO) (3) as the TpiPr2Cu-

I SC6F5 (1) decomposes due to the instability of the new Cu -nitrosyl product. We do know that the reaction goes to completion because the color is fully leached from the solution over a period of 4 hr at 25 °C.

If trans-s-nitrosation at copper were competitive with disulfide formation, a

t t t mixture of C6F5S-SC6F5, BuS-SBu , and BuS-SC6F5 would be expected. Combined with

iPr2 t iPr2 clean transnitrosation observed between Tp Zn-S Bu and C6F5SNO to give Tp Zn-

t 40 SC6F5 and BuSNO, these results point to a common nitroxyl disulfide intermediate

iPr2 t t iPr2 TpM ( BuSN(O)SC6F5) which releases the S-nitrosothiol BuSNO to give Tp Zn-

40 t iPr2 SC6F5 in the case of Zn , but quickly decomposes to BuS-SC6F5 and Tp Cu(NO) (3)

(or iPr2TpCuI + NO) with Cu (Scheme 4.9) This is also consistent with the ease of oxidation of the nitroxyl disulfide anion [RSN(O)SR’]- to give RS-SR' and NO suggested

23 t in a theoretical study. A less than quantitative yield for BuS-SC6F5 is observed because the TpCu(NO) generated also catalyzes the decomposition of tBuSNO to give tBuS-SBut

175 and NO. Further studies are underway to outline the mechanism of catalytic

decomposition of S-nitrosothiols by TpCuI systems (see below).

Scheme 4.9. Comparison of redox active copper with redox inactive zinc in the reaction of TpiPr2MII-SR with R'SNO.

We attempted to understand the order of the reaction using the method of initial rates. Our studies, however, were unsuccessful to due the fast rate of the reaction which prevented obtaining the initial rate information at 0 °C for this particular combination of

[Cu]-SR and R’SNO.

176 We thought that we might have a better chance at studying the kinetics using 2 due to the bulky nature of the trityl group which could slow down rates of reaction. When

1 equiv. Ph3CSNO is added to a CH2Cl2 solution of blue 2, there is no reaction as monitored by UV-vis spectroscopy over (Scheme 4.10). The band with λmax = 625 nm of

t iPr2 Scheme 4.10. NO release from BuS-NO by Tp Cu-SC6F5. A nitroxyl disulfide t intermediate is proposed based on the formation of C6F5S-S Bu as the major sulfur containing species.

Figure 4.13. Initial rate of reaction between 2 and varying equivalents of tBuS-NO to t iPr2 form Ph3CS-S Bu and Tp Cu(NO) (3) as monitored by loss of 2. The linear dependence of the initial rate on [tBuSNO] indicates first order behavior in this reagent.

177 2 does not disappear at 0 °C and no evidence of formation of 3 is seen over a 2 hr period.

When 1 equiv of tBuSNO is added to benzene or dichloromethane solutions of blue 2, the blue color is quickly discharged. Employing a freshly prepared sample of 2 in benzene-

t d6 free from the disulfide Ph3CSSCPh3, we find that addition of 1 equiv BuSNO at room

t temperature provides the unsymmetrical disulfide PhC3S-SBu (5) in 65% yield based on the integration of the methyl of the tBu group relative to a naphthalene standard in the 1H

NMR (δ 1.262 ppm). The symmetric disulfide tBuS-SBut accounts for the remainder of

t the added BuSNO. Monitoring this reaction in CH2Cl2 by UV-vis spectroscopy shows the loss of copper-thiolate 1 at λ = 625 nm along with growth of a band at λ = 495 nm which was shown through independent synthesis to be iPr2TpCu(NO) (3) (Scheme 4.9).33

We were interested in the values for the order of the reation, rate = k[TpiPr2Cu-

x t y SCPh3] [ BuSNO] . We conducted two sets of experiments where the initial

iPr2 concentration of Tp Cu-SCPh3 was held constant (0.67 mM or 0.50 mM) and the amount of tBuSNO was varied (1, ¾, ½, ¼ equiv) with respect to the copper-thiolate

t concentration. A plot of rate0 vs [ BuSNO] is a straight line, thus he reaction is first order with respect to tBuSNO (Figure 4.13). Related future studies varying the concentration

iPr2 t of Tp Cu-SCPh3 while holding the BuSNO concentration constant are planned.

iPr II 4.4. Reactivity of Tp Cu -SR complexes with NOgas.

t We find that both 1 and 2 activate the BuS-NO bond by reductive cleavage to

iPr2 I t release NO•, Tp Cu and the corresponding disulfide BuSSR (R = C6F5 or CPh3). We

iPr2 were also interested in the reaction of Tp Cu-SR (1 or 2) with NOgas. The reaction of 1

iPr2 equiv. NOgas with Tp Cu-SC6F5 leads to an initial decrease of the peak at 666 nm (1) 178 and formation of a new peak at 495 nm (3) (Scheme 4.15). The peak at 495 nm attributed

to the copper(I)-nitrosyl increases during the course of the reaction (Figure 4.14).

Interestingly, 1 is not completely decomposed by the NOgas. Rather, the peak at 666 nm which initially decrease in intensity begins to grow in and actually increases in intensity

iPr2 Scheme 4.11. Reaction between 1 and NOgas to form Tp Cu(NO) 3. Thiyl radical that forms from the reaction binds with TpiPr2CuI impurity present in the reaction mixture to iPr2 form Tp Cu-SC6F5.

λmax = 666 nm iPr2 II λmax = 495 nm Tp Cu -SC6F5 TpiPr2CuI(NO) 1. Intial spectrum of 1 before and NOgas is added

2. First spectrum afterNOgas is added

Figure 4.14. UV-Vis spectrum of the reaction of 1 with NOgas in CH2Cl2 at 25 °C to form a new species λmax = 495 nm (attributed to formation of 3) and presumably C6F5S•. The C6F5S• can react with copper(I) species in the reaction mixture to form back 1 (λmax = 666 nm).

179 beyond the orginal absorbance of 1 in the reaction mixture.

This was very puzzling to us at first. How could both 1 and 3 continue to grow in intensity throughout the course of the reaction? We have seen from our previous studies

iPr2 iPr2 I that Tp Cu-SC6F5 is thermally unstable in solution and decomposes to Tp Cu and

iPr2 C6F5SSC6F5. Perhaps it is useful to consider the reaction of Tp Cu-SC6F5 with NO as

iPr2 forming Tp Cu(NO) and the thiyl radical C6F5S• which dimerizes to form the disulfide.

If the solution contains TpiPr2CuI (either from the copper-thiolate decomposition or loss of

iPr2 iPr2 I NO from Tp Cu(NO)), the thiyl radical C6F5S• could bind to Tp Cu before it

iPr2 dimerizes and could form the Tp Cu-SC6F5 (Scheme 4.11).

No evidence of TpiPr2Cu(NO) (3) formation is present in the reaction of TpiPr2Cu-

SCPh3 (2) with NOgas (Figure 4.15). It is interesting to note that 2 decomposes

λ = 625 nm iPr2 Tp Cu-SCPh3 λ = 353 nm TpiPr2CuI??

iPr2 Figure 4.15. Disappearance of the band at λ = 625 nm from Tp CuSCPh3 in CH2Cl2 at 0 °C upon addition of NOgas and appearance of λ = 353 nm from the reaction of 2 with NOgas. No evidence of the copper-nitrosyl 3 is present. 180 considerably faster in the presence of NOgas versus in the absence of NOgas. The same unidentified compound with λmax = 353 nm grows in during the thermal decomposition of

2 also grows in from the reaction of 2 with NOgas.

iPr I 4.5. Tp Cu reactivity with Ph3SNO.

In our previous work with β-diketimiate CuI complexes, we demonstrated that

{[Me2NN]Cu}2 reductively cleaves the Ph3CS-NO bond to form the Cu(II)-thiolate

[Me2NN]Cu-SCPh3 and NOgas. The released NOgas nitrosated the backbone CH of the

[Me2NN]Cu-SCPh3 forming [Me2NN]Cu(NO[Me2NN]) (Chapter 3).

Tris(pyrazolyl)borate copper(I) complexes TpCu are perhaps better suited to study the reactivity of electron-rich CuI complexes with RSNOs because the corresponding

TpCu(NO) are known to form.

In order to prevent any side reactions of RSNOs with TpCu-SR species that may

iPr2 I form, we conducted our reactivity studies with Ph3CSNO employing excess Tp Cu (10

iPr2 I or 20 equiv.) Ph3CSNO (0.50 mM) was added to either 10 or 20 equiv of Tp Cu in

iPr2 CH2Cl2 at 0 °C. Tp Cu-SCPh3 (2) immediately grows in after the first spectrum is taken

(10 s) and reaches its maximum absorbance at 0.488 or 0.619 AU, respectively (Figure

4.16). The absorbance values correspond to concentrations of 2 of 0.075 mM and 0.095 mM – complete conversion of Ph3CSNO to 2 would give 2 in a concentration of 0.50

iPr2 mM. Thus, we are only able to observe Tp Cu-SCPh3 growth in 15 or 19 % yield, respectively, under these conditions. The copper-thiolate formation shows activation of the RS-NO bond by the CuI to release NO and form the Cu(II)-thiolate (Scheme 4.14).

The fate NO released during the course of the reaction is TpiPr2Cu(NO). We are not sure 181 iPr2 iPr2 if the Tp Cu(NO) is formed from reaction of NO with the Tp Cu-SCPh3 or if free

NO in the solution reacts with the excess TpiPr2CuI.

iPr2 The initial yield of Tp Cu-SCPh3 depends on the amount of excess copper(I) present in the reaction mixture. When 50 equiv of TpiPr2CuI are present, we can observe

iPr2 formation of Tp Cu-SCPh3 in 29 % yield, 30 equiv in 25 % yield, 20 equiv in 23 % yield, and 10 equiv in 15 % yield (Figure 4.16). The extent of TpiPr2Cu(NO) production

iPr2 I Scheme 4.14. Reaction of Ph3CSNO in the presence of xs Tp Cu to initially give iPr2 iPr2 iPr2 Tp Cu-SCPh3. The Tp Cu-SCPh3 then forms Tp Cu(NO) in the presence of NO.

λ = 495 nm λ = 495 nm λ iPr2 λ iPr2 = 625 nm Tp Cu(NO) = 625 nm Tp Cu(NO) iPr2 iPr2 Tp Cu-SCPh3 Tp Cu-SCPh3

Figure 4.16. Rate and extent of formation of 2 and 3 from the reaction of TpiPr2CuI with Ph3CSNO depends on the amount of excess copper(I) in solution. 182 also depends on the equivalents of TpiPr2CuI in solution, 67 % (20 equiv) and 49 % (10 equiv). No specific kinetic data can be determined as this time. The formation of

iPr2 iPr2 st Tp Cu(NO) and the disappearance of Tp Cu-SCPh3 do not follow any simple 0, 1 , or 2nd order kinetics. Further studies are planned using stop-flow kinetic methodology.

iPr 4.6. Tp Cu(NO) reactivity with C6F5SNO.

Both TpiPr2CuI and TpiPr2CuII-SR can activate the S-NO bond of S-nitrosothiols to release NO. Since TpiPr2Cu(NO) is produced during the course of these reactions, it would be useful to know how this copper(I) nitrosyl reacts with S-nitrosothiols RSNOs

iPr2 and disulfides RSSR. When 1 equiv. C6F5S-NO is added to a solution of Tp Cu(NO),

iPr2 the S-NO bond is reductively cleaved to form Tp Cu-SC6F5 (λmax = 666 nm) and release NO (Scheme 4.15, Figure 4.17). This species reaches a maximum absorbance as soon as the first spectrum is taken on the sample after the addition of C6F5S-NO and forms the copper-thiolate in 14 % yield. The TpiPr2Cu(NO) has decomposed by 40 %

iPr2 iPr2 after the initial addition of C6F5S-NO. Both Tp Cu(NO) and Tp Cu-SC6F5 decompose in the solution over time. A small new peak grows in at 815 nm and is

iPr2 41 attributed to the formation of Tp Cu(NO2) (Figure 4.17).

183 iPr2 Scheme 4.15. Reaction of Tp Cu(NO) with C6F5SNO leads to initial formation of iPr2 iPr2 iPr2 Tp Cu-SC6F5 with Tp Cu(NO) still in solution. The final product is Tp Cu(NO2).

λ = 495 nm TpiPr2Cu(NO)

1) 2) λ = 666 nm iPr2 Tp Cu-SC6F5 3) 3)

λ = 814 nm iPr2 Tp Cu(NO2) 4)

iPr2 Figure 4.17. UV-vis spectra of Tp Cu(NO) (Spectrum 1) with C6F5SNO in CH2Cl2 at iPr2 iPr2 25 °C leading to initial formation of Tp Cu-SC6F5 with Tp Cu(NO) (Spectrum 2) still iPr2 iPr2 in solution. Tp Cu-SC6F5 with Tp Cu(NO) decompose (Spectrum 3) and the final iPr2 copper product is Tp Cu(NO2) (Spectrum 4).

184 4.7. Synthesis of Tp(CF3)2Cu complexes and initial reactivity studies.

In all of the reactivity studies above involving the electron-rich TpiPr2CuI fragment and S-nitrosothiols, the RS-NO bond was cleaved to form copper(II)-thiolates TpiPr2Cu-

SR. To probe the influence of the electronic environment at copper(I) in the loss of NO from RSNOs, a less electron rich TpCuI complex would be needed.

Dias has found that the Tp ligand based on 3,5-bis(trifluoromethyl)pyrazole

(Scheme 4.16) significantly reduces the electron richness of the copper center. His group was able to isolate and crystallographically characterize the copper bound 1-

(CF3)2 (CF3)2 I adamantylazide complex Tp Cu(N3Ad) using Tp Cu generated in situ from the

(CF3)2 reaction of Tp Na with ½ equiv. [CuOTf]2(benzene) in THF followed by the addition

42 of 1-azidoadamantane (AdN3) (Scheme 4.17). The solid state structure revealed that the terminal nitrogen atom of AdN3 coordinates to the copper. Since the RN-NN bonds of

(CF3)2 azides are often easily activated to release N2(gas), Tp Cu(N3Ad) represents an unusual

Scheme 4.16. Synthesis of 3,5-bis(trifluromethyl)pyrazole from the reaction of 1,1,1,5,5,5-hexafluoropentane-2,4-dione and hydrazine monohydrate. 185 Scheme 4.17. Coordination of 1-adamantylazide to the Tp(CF3)2CuI fragment (Dias et al Inorg. Chem. 2000, 39, 3894-3901).

example of azide coordination to a complex. In this context, several β- diketiminato copper(I) complexes {[NN]Cu}2 react with organoazides to give dinuclear

43,44 nitrenes {[NN]Cu}2(µ-NR) upon loss of N2.

We anticipated three different possible binding modes for an S-nitrosothiol such as AdSNO to the electron-poor Tp(CF3)2CuI fragment (Figure 4.18). We isolated

Tp(CF3)2CuI(THF) (4) in 63 % yield using a synthetic procedure related to Dias’s (Scheme

Figure 4.18. Possible binding modes of S-nitrosothiol to Tp(CF3)2CuI.

186 Figure 4.19. Synthesis and isolation of Tp(CF3)2CuI(THF) (4).

4.18). 1H NMR analysis reveals one THF molecule coordinated to the copper(I).

Treatment of this compound with a green solution of AdSNO in toluene results in no color change and no effervescence. Crystallization this reaction mixture from a toluene/pentane solution led to crystals that effervesce and decompose in mineral oil

1 during inspection for X-ray diffraction (Scheme 4.19). H NMR analysis in benzene-d6 of the crystals reveals very little shift in the C-H peak (δ 6.257 ppm) of the 3,5- bis(trifluoromethyl)pyrazolyl ring from the Tp(CF3)2CuI(THF) (4) starting material (δ

6.240 ppm). The crystals do have adamantyl CH peaks, but the three types of CH peaks that are clearly seen in AdSNO are an overlapping multiplet from δ 2.104-1.201 ppm.

13C{1H} NMR analysis of the crystals does not result in major shifts of the carbon peaks

from free AdSNO (δ 43.56, 36.30, 30.24 ppm) to AdSNO reacted with Tp(CF3)2CuI(THF)

(4) (δ 43.13, 35.98, 30.65 ppm). The 19F NMR spectrum of the reaction mixture has two

CF3 peaks (δ -59.8 and -62.1 for back and front CF3 groups, respectively) where the front

(CF3)2 I CF3 peak is very broad and slightly shifted upfield compared to the Tp Cu (THF) (4)

complex (δ -59.8 and -63.0 for back and front CF3 groups, respectively). It is difficult to make any conclusions about the nature of the AdSNO interaction with the copper(I) strictly based on NMR data.

187 Crystals of the reaction of BnSNO with Tp(CF3)2CuI(THF) (4) in toluene/pentane also leads to bubbling and decomposition in the mineral oil. Isolation of the crystals and

1 analysis by H NMR shows a new CH2 benzyl peak at δ 3.849 ppm, not attributed to

BnSNO (δ 4.62 ppm) or BnSSBn (δ 3.55 ppm). The CH2 peak integrates to 4 H atoms

(CF3)2 I compared to the CH peak from Tp Cu (THF) (4) (3 H atoms). This is suggestive of

(CF3)2 I BnSSBn binding to the Tp Cu (THF) (4) (Scheme 14.19). An independent reaction

(CF3)2 I of BnSSBn with Tp Cu (THF) (4) in benzene-d6 also leads to the same peak at δ

3.849 ppm (Scheme 14.19). The isolated crystals are not a copper bound BnSNO, rather they are a copper bound disulfide. BnSNO is not very stable by itself and if its coordination to Tp(CF3)2CuI is labile, it could decompose to BnSSBn during the work up of this compound.

The most important conclusion from the above work is that the electron-poor

Tp(CF3)2CuI complex not oxidized from copper(I) to copper(II) from the reaction with

RSNO. This stands in contrast to the corresponding reactivity of TpiPr2CuI species with

RSNO to form TpiPr2CuII-SR.

Scheme 4.19. Reaction of Tp(CF3)2CuI(THF) with RSNO. The isolated product is disulfide bound to copper. 188 Summary

The redox properties of the metal center have a dramatic influence fate of S- nitrosothiols in their reaction with structurally related metal thiolates. The redox-inactive

iPr2TpZn-SR system undergoes clean, reversible transnitrosation with S-nitrosothiols, with a preference for the formation of less sterically hindered and more electron poor thiolate residues at zinc.40 On the other hand, RSNOs reductively cleave the copper-thiolate bond in TpCuiPr2-SR' species with concomitant formation of the copper(I) nitrosyl

TpiPr2Cu(NO) along with the disulfide RS-SR'. Electron-rich TpiPr2CuI complexes react with RSNOs to give the corresponding copper(II)-thiolates TpiPr2Cu-SR as well as the copper-nitrosyl TpCu(NO) (in the case of R = trityl). Use of the electron-poor Tp(CF3)CuI, however, copper(I) is not oxidized to copper(II) and the RS-NO bond is not reductively cleaved. These observations clearly indicate an important role for the electronic environment at copper to promote reductive cleavage of RS-NO species.

189 Experimental procedures

General experimental details.

All experiments were carried out in a dry nitrogen atmosphere using an MBraun

glovebox and/or standard Schlenk techniques. 4A molecular sieves were activated in

vacuo at 180 ºC for 24 h. Dry toluene and dichloromethane were purchased from Aldrich

and were stored over activated 4A molecular sieves under nitrogen. Diethyl ether and

tetrahydrofuran (THF) were first sparged with nitrogen and then dried by passage through

45 activated alumina columns. Pentane was first washed with conc. HNO3/H2SO4 to remove olefins, stored over CaCl2 and then distilled before use from sodium/benzophenone. All deuterated solvents were sparged with nitrogen, dried over activated 4A molecular sieves and stored under nitrogen. 1H and 13C NMR spectra were recorded on an Inova Varian 300 MHz or 400 MHz spectrometer (300 or 400 and 75.4 or

100.46 MHz, respectively). All NMR spectra were recorded at room temperature unless otherwise noted and were indirectly referenced to TMS using residual solvent signals as internal standards. 19F NMR spectra were recorded at 282.34 or 375.9 MHz using an internal or external reference of C6F6 set to δ = -164.9 ppm. GC-MS spectra were recorded on a Varian Saturn 2100T and elemental analyses were performed on a Perkin-

Elmer PE2400 microanalyzer, and UV-Vis spectra were taken on a Cary 50 or Agilent

8453 spectrophotometer in our laboratories.

All thiols and KOH were obtained from Acros, anhydrous [Cu(MeCN)]PF6 and

NOBF4 from Strem, and TlOEt and Cu(ClO4)2⋅6H2O from Aldrich; all were used as received. The hydrotris(3,5-diisopropylpyrazol-1-yl)borate potassium salt TpiPr2K was 190 synthesized according to a literature procedure.39 The thallium thiolates TlStBu and

TlSC6F5 were synthesized following a modified literature procedure by the reaction of free thiol with TlOEt in ether followed by washing of the solids with pentane.46 The S-

t nitosothiols BuSNO and C6F5SNO were synthesized in situ by the reaction of the

47 corresponding thallium thiolates with 1 eq. NOBF4 in CDCl3 or C6D6. The S- nitrosothiol Ph3CSNO is more stable and can be prepared on a preparative scale by the

t 48 t reaction of BuONO and Ph3CSNO. Caution! BuSH is extremely volatile and possesses a pungent odor essentially indistinguishable from EtSH contained in natural gas.

Synthesis of starting materials.

iPr2 iPr2 (Tp Cu)2(µ-OH)2. (Tp Cu)2(-OH)2 was synthesized according to the literature

49 procedure with slight modifications. Care must be taken to remove all CO2 from all

iPr2 solutions employed in this synthesis to prevent formation of green (Tp Cu)2(µ-CO)3

TpiPr2K (0.640 g, 1.27 mmol) was dissolved in 15 ml of MeOH in a 50 mL Schlenk round bottom flask. Cu(ClO4)2⋅6H2O (0.471 g, 1.27 mmol) was dissolved in 10 mL of MeOH and added dropwise to the TpiPr2K solution. The solution immediately turned a deep blue green color was allowed to stir under N2(gas) for 2 h. Crushed KOH pellets (0.080 g, 1.43 mmol) were added to the reaction mixture. The solution was allowed to stir for another 2 h upon which a deep blue color formed. In air, the solvent was removed in vacuo and the residue was extracted (3 × 25 mL) with CH2Cl2 and filtered over Celite. The volume was reduced to 5 mL and 1 mL of pentane was added to help crystallization. Blue crystals

-1 formed overnight at -35 ºC in 86% yield (0.598 g). IR (CH2Cl2 film, cm ) 3674 (OH),

191 2967 (CH), 2931 (CH), 2538 (BH), 1537, 1462, 1392, 1382, 1300, 1177, 1052, 705; UV-

-1 -1 Vis (CH2Cl2, nm (cm M )) 700 (230).

TpiPr2Cu. TpiPr2Cu was synthesized according to the literature procedure with modifications.37 TpiPr2K (0.600 g, 1.19 mmol) was dissolved in 8 mL of MeCN and added to Cu(MeCN)BF4 (0.449 g, 1.43 mmol) in 8 mL of MeCN. The reaction was allowed to stir for 2 h. The solvent was removed in vacuo. The off-white residue was slurried in

CH2Cl2 (15 mL), the slurry was filtered through Celite, and the residue cake was washed with 5 mL of CH2Cl2. The filtrate was concentrated to 3 mL and layered with MeCN.

Colorless crystals formed overnight at -35 ºC 58% Yield (0.394 g). The crystals were collected and dried under reduced pressure. Characterization by 1H NMR shows broad

1 peaks with no MeCN present. H NMR (C6D6): δ 5.963 (s, 3, pz) 3.762 (m, 3, CHMe2),

13 1 3.409 (m, 3, CHMe2), 1.417 (d, 18, CHMe2) 1.282 (d, 18, CHMe2); C{ H} NMR

(C6D6): δ 163.24, 161.27, 160.33, 28.74, 26.60, 23.77, 23.47.

iPr2 iPr2 Tp Cu(NO) (3). Tp Cu (0.053 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2.

The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to

iPr2 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM stock solution of Tp Cu.

TpiPr2Cu (0.500 mL of 10 mM solution) was placed in a quartz cuvette with a rubber septa and 2.5 mL CH2Cl2 was added to bring the solution to a final volume of 3 mL, 1.67 mM. One equiv. NOgas (0.25 mL, 0.01 mmol) was syringed in the cuvette and the colorless solution immediately turned red. A UV-Vis spectra was taken. Additional 0.25 mL aliquots of NOgas were added until the absorbance reached a maximum value. UV-

192 -1 -1 -1 Vis (CH2Cl2, nm, (cm M )) 495 (960 - estimated)). IR (CH2Cl2 solution, cm ) 1704

(NO). This compound was always made in situ and never isolated due to its instability.

Synthesis and stability of TpiPrCuII-SR complexes (Section 4.1).

iPr2 iPr2 Tp CuSC6F5 (1). Tp CuSC6F5 was synthesized according to the literature procedure

27,36 iPr2 with slight modifications. (Tp Cu)2(µ-OH)2 (0.200 g, 0.183 mmol) was dissolved in

3 mL of CH2Cl2 with crushed molecular sieves at the bottom of the vial. C6F5SH (0.074 g, 0.367 mmol) was dissolved in 2 mL of CH2Cl2. Both solutions were chilled to -35 ºC.

iPr2 C6F5SH was added to (Tp Cu)2(µ-OH)2 and the reaction mixture was allowed to sit in the freezer at -35 ºC for 1 h. The reaction mixture was filtered over Celite and the volume was reduced to 2 mL and 1 mL of pentane was added to help crystallization. Dark blue crystals formed overnight at -35 ºC which were isolated in 72% yield (0.192 g). UV-Vis

-1 -1 (CH2Cl2, 25 °C, nm(cm M )) 360(1250), 666 (5900), 1028 (1300).

iPr2 iPr2 Kinetics of Tp CuSC6F5 decomposition. A 0.30 mM solution of Tp CuSC6F5 in

CH2Cl2 was placed in a quartz cuvette and allowed to decompose in the UV-Vis instrument. Spectra were taken every 240 s for 2 h. A plot of 1/[M] reveals a straight line pointing to a bimolecular route for decomposition. The rate law of this reaction is rate =

iPr2 2 -1 -1 k[Tp CuSC6F5] where k = 0.182 M s .

iPr2 iPr2 Tp CuSCPh3 (2). Tp CuSCPh3 was synthesized according to the literature procedure

27 iPr2 with slight modifications. (Tp Cu)2(µ-OH)2 (0.200 g, 0.183 mmol) was dissolved in 3 mL of CH2Cl2 with crushed molecular sieves at the bottom of the vial. Ph3CSH (0.102 g,

0.367 mmol) was dissolved in 2 mL of CH2Cl2. Both solutions were chilled to -35 ºC.

193 iPr2 Ph3CSH was added to (Tp Cu)2(µ-OH)2 and the reaction mixture was allowed to sit in the freezer at -35 ºC for 1 h. The reaction mixture was filtered over Celite and the volume was reduced to 2 mL and 1 mL of pentane was added to help crystallization. Dark blue crystals formed overnight at -35 ºC which were isolated in 78% yield (0.230 g). UV-Vis

-1 -1 (CH2Cl2, 25 °C) 625 nm (6600 M cm ), 910 (1230).

iPr2 iPr2 Kinetics of Tp CuSCPh3 decomposition. A 0.16 mM solution of Tp CuSC6F5 in

CH2Cl2 was placed in a quartz cuvette and allowed to decompose in the UV-Vis instrument at 0, 10, 20, 30, and 40 ºC. Spectra were taken every 30 s. A plot of [M] vs. time reveals a straight line pointing to a zero order rate for decomposition. The rate of decomposition is independent of the concentration. An Eyring plot analysis was performed by plotting ln(k/T) (s-1K-1) vs. 1/T (K), H‡ = 18 kcal/mol and S‡ =33 cal/mol.

iPr2 Reactions of Tp Cu with RSSR in presence of O2 (Section 4.2).

iPr2 Tp Cu with C6F5SSC6F5 in presence of O2. C6F5SSC6F5 (0.040 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM

iPr2 stock solution of C6F5SSC6F5. Tp Cu (0.040 mL, 10 mM, 0.40 mol) and C6F5SSC6F5

(0.020 mL, 10 mM, 0.20 M) was placed in a quartz cuvette with a rubber septa and 2.94

iPr2 mL CH2Cl2 was added to bring the solution to a final volume of 3 m with [Tp Cu] =

0.133 mM. A UV-Vis spectrum is taken and there is no reaction in the absence in an anaerobic environment. While the cuvette is in the UV-Vis, the solution is exposed to air while scans are taken every 30 s. A peak at 666 nm grows in over the course of 30 min 194 iPr2 and is attributed to the formation of Tp Cu-SC6F5. The peak reaches Amax = 0.651 corresponding to 81 % spectroscopic yield (Scheme 4.13, Figure 4.17)

iPr2 Tp Cu with Ph3CSSCPh3 in presence of O2. Ph3CSSCPh3 (0.055 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM

iPr2 stock solution of Ph3CSSCPh3. Tp Cu (0.040 mL, 10 mM, 0.40 mol) and

Ph3CSSCPh3 (0.020 mL, 10 mM, 0.20 M) was placed in a quartz cuvette with a screw top and a septa and 2.94 mL CH2Cl2 was added to bring the solution to a final volume of

3 m with [TpiPr2Cu] = 0.133 mM. A UV-Vis spectrum is taken and there is no reaction in the absence in an anaerobic environment. While the cuvette is in the UV-Vis, the solution is exposed to air while scans are taken every 30 s. A peak at 625 nm grows in over the

iPr2 course of 30 min and is attributed to the formation of Tp Cu-SCPh3. The peak reaches

Amax = 0.685 corresponding to 78 % spectroscopic yield (Scheme 4.14, Figure 4.18).

Reactions of TpiPR2Cu-SR with RSNO (Section 4.3).

iPr2 t iPr2 Reaction of Tp CuSC6F5 with BuSNO. Tp CuSC6F5 (0.073 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM

iPr2 t t stock solution of Tp CuSC6F5. BuSNO was made from the reaction of TlS Bu (0.030 g,

0.100 mmol) in 2 mL CH2Cl2 added to stirring crystals of NOBF4 (0.012 g, 0.100 mmol).

The reaction is allowed to stir for 10 min and the solution was analytically transferred to a 10.00mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a

t iPr2 10 mM stock solution of BuSNO. Tp CuSC6F5 (0.066 mL of 10 mM solution) and 195 2.434 mL CH2Cl2 were placed in a quartz cuvette with a rubber septa in an inert

t atmosphere. BuSNO (0.066 mL of 10 mM solution, diluted with 0.434 mL CH2Cl2) was syringed into the cuvette. UV-VIS spectra were taken every 120 s for 20 min. The peak at

iPr2 666 nm corresponding to Tp CuSC6F5 decreases over time along with a new peak at

495 nm growing in for a couple of scans and then decreasing over time. Confirmation by independent synthesis identifies this peak to be TpiPr2CuNO. GC-MS of the reaction

t mixture shows the fate of the thiyl radical to be C6F5SS Bu m/z= 283. C6F5SSC6F5 can

iPr2 also be identified in the GC-MS spectra m/z= 398. Tp CuSC6F5 decomposes by means of a second order reaction mechanism and there always seems to be C6F5SSC6F5 in the

II Cu SR starting material. Thus it is hard to determine whether the C6F5SSC6F5 in the GC-

MS is a product from the reaction or if it is present in the starting material. A low

iPr2 concentration in situ synthesis of Tp CuSC6F5 prevented any of the C6F5SSC6F5 from

t t forming and the reaction with BuSNO produced only C6F5SS Bu as confirmed by GC-

MS analysis (EI: m/z = 288 M+)

t Quantification of C6F5SS Bu. C6F5SH (0.018 g, 0.092 mmol) and internal standard C6F6

19 (0.017 g, 0.092 mmol) were mixed together in 1 mL of C6D6. F NMR was taken in

iPr2 order to determine the ratio of C6F5SH:C6F6. This solution was added to (Tp Cu)2(-

OH)2 (0.050 g, 0.046 mmol) in 3 mL of C6D6 with crushed molecular sieves at the bottom of the vial. The mixture was allowed to sit at room temperature for 15 minutes upon which the color changed from blue green to dark blue. The reaction mixture was filtered to remove the molecular sieves and divided into two vials. tBuSNO (0.016 g, 0.046 mmol) in 1 mL of C6D6 was added to one vial and nothing was added to the second vial. 196 19 F NMRs were taken of each. There was no C6F5SSC6F5 present in either reaction

t t mixture and C6F5SS Bu was present in 65 % yield in the reaction with BuSNO by quantification with C6F6 standard.

t Unsymmetrical disulfide test reaction C6F5SS Bu. In order to rule out the formation of

t iPr2 t C6F5SS Bu in the reaction of Tp CuSC6F5 with BuSNO due only to C6F5SNO and

tBuSNO decomposition in some sort of transnitosation mechanism we performed a test

reaction. The test reaction consisted of independently synthesizing the C6F5SNO and

t BuSNO in C6D6, mixing the two S-nitrosothiols together and letting them decompose

t t t over five days to form a mixture of C6F5SSC6F5, BuSS Bu, and C6F5SS Bu. Samples of independently prepared C6F5SNO (from reaction of C6F5STl 0.125 mg, 0.310 mmol with

t t NOBF4 0.036 mg, 0.310 mmol in C6D6) and BuSNO (from reaction of BuSTl 0.091 mg,

0.310 mmol with NOBF4 0.036 mg, 0.310 mmol in C6D6) were added to each other. The reaction mixture stirred for five days to allow the two S-nitrosothiols to decompose to

t t t 1 19 form a mixture of C6F5SSC6F5, BuSS Bu, and C6F5SS Bu, characterized by H NMR, F

Table 4.1. Characterization data for disulfide formation.

Compound 1H NMR 19F NMR GC/MS % in reaction

tBuSStBu 1.201 --- 178 m/z 25 % -134.026, C6F5SSC6F5 --- -150.400, 398 m/z 50 % -162.173 -134.030, t BuSSC6F5 1.064 -153.909, 288 m/z 25 % -163.187

197 NMR, and GC/MS. Quantification of the disulfides was conducted using CF3C6H5 as a

1H and 19F NMR internal standard.

iPr2 iPr2 Reaction of Tp CuSCPh3 with Ph3CSNO. Tp CuSCPh3 (0.080 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM

iPr2 stock solution of Tp CuSCPh3. Ph3CSNO (0.035g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 10 mM stock solution of

iPr2 Ph3CSNO. Tp CuSCPh3 (0.066 mL of 10 mM solution) and 2.434 mL CH2Cl2 were placed in a quartz cuvette with a rubber septa in an inert atmosphere. Ph3CSNO (0.066 mL of 10 mM solution, diluted with 0.434 mL CH2Cl2) was syringed into the cuvette at 0

°C. Analysis by UV-vis shows no reaction between 2 and Ph3CSNO.

iPr2 t iPr2 Reaction of Tp CuSCPh3 with BuSNO. Tp CuSCPh3 (0.066 mL of 10 mM solution) and 2.434 mL CH2Cl2 were placed in a quartz cuvette with a rubber septa in an inert atmosphere. tBuSNO (0.066 mL of 10 mM solution, diluted with 0.434 mL

CH2Cl2) was syringed into the cuvette. The peak at 625 nm corresponding to

iPr2 Tp CuSCPh3 decreases over time along with a new peak at 495 nm growing in for a couple of scans and then decreasing over time. Confirmation by independent synthesis identifies this peak to be TpiPr2CuNO. GC-MS of the reaction mixture shows the fate of

t the thiyl radical to be Ph3CSS Bu (parent ion peak not seen, but fragmentation pattern

1 shows a peak at m/z = 57 (Me3C•) and m/z = 243 (Ph3C•). H NMR analysis of the

198 t reaction mixture with a naphthalene internal standard shows the Ph3CSS Bu to be formed in 65 % yield by integration of its tBu signal at δ 1.262 ppm.

Table 4.2. Preparation of solutions for initial rate analysis of the reaction between iPr2 t Tp CuSCPh3 and BuSNO.

iPr2 t Ratio Tp CuSCPh3 BuSNO CH2Cl2 Rate [Cu]SR:R'SNO (10 mM) (10 mM) 1:1 0.075 mL 0.075 mL 2.850 mL 0.0153 1:3/4 0.075 mL 0.056 mL 2.869 mL 0.0116 1:1/2 0.075 mL 0.036 mL 2.889 mL 0.0067 1:1/4 0.075 mL 0.019 mL 2.906 mL 0.0044

iPr2 t Kinetics of reaction of between Tp Cu-SCPh3 and BuSNO. The rate law for the

iPr2 x t y reaction follows rate = k[Tp Cu-SCPh3] [ BuSNO] . We determined x = 1 using the method of initial rates. If amount of tBuSNO is greater the 1 equiv when compared to

iPr2 Tp CuSCPh3 the peak at 625 nm decreases too fast and we are unable to measure the

iPr2 initial rate of decomposition. We used the 10 mM stock solutions of Tp CuSCPh3 and

tBuSNO previously prepared.

iPr2 Reaction of Tp CuSC6F5 with NO (Section 4.4).

iPr2 Reaction of Tp CuSC6F5 with NO. To 3 mL of a 0.040 mM solution of

iPr2 Tp CuSC6F5 in CH2Cl2 at 25 °C was added 1 equiv. NOgas (3 L, 0.12 mol). The first spectrum after NOgas addition showed the peak at 666 nm immediately decreased from A

= 0.22 to A = 0.125 with formation of a new peak at A495 nm = 0.14. The second spectrum shows that the peak at 495 nm continues to increase in intensity. The peak at 666 nm also begins to increase in intensity. Both peaks reached a maximum absorbance after 30 min.

We attributed the growth of the new peak at 495 nm to be from TpiPr2Cu(NO) formed from the addition of NO to the copper(II) thiolate to activate the Cu-S bond and release 199 iPr2 I C6F5S•. The C6F5S• can then react with a Tp Cu impurity in the reaction mixture to

iPr2 reform Tp CuSC6F5.

iPr2 Reaction of Tp CuSCPh3 with NO. To 3 mL of a 0.100 mM solution of

iPr2 Tp CuSCPh3 in CH2Cl2 at 25 °C was added 1 equiv of NOgas (8 L, 0.30 mol). The first spectrum after NOgas addition showed the peak at 625 nm immediately decreased from A = 0.70 to A = 0.50 with formation of a new peak at A353 nm = 0.38. An isobestic point forms throughout the course of the reaction from the decrease at 653 nm and in the increase at 353 nm. Formation of TpiPr2Cu(NO) is not observed. We have not identified the peak at 353 nm.

iPr2 I Reactions of excess Tp Cu with Ph3CSNO (Section 4.5).

iPr2 I Reaction of 10 or 20 equiv of Tp Cu with Ph3CSNO. Ph3CSNO (0.031 g, 0.100 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a

10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a

iPr2 I 10 mM stock solution of Ph3CSNO. Tp Cu (0.101 g, 0.200 mmol) was dissolved in 2 mL of CH2Cl2. The solution was analytically transferred to a 10.00 mL volumetric flask and diluted to 10.00 mL with CH2Cl2 to make 10.00 mL of a 20 mM stock solution of

Table 4.3. Preparation of solutions for reaction of 10 or 20 equiv. TpiPr2Cu with Ph3CSNO.

iPr2 I t iPr2 iPr2 Ratio Tp Cu BuSNO CH2Cl2 [Tp CuSCPh3] [Tp Cu(NO)] [Cu]:R'SNO (20 mM) (10 mM) Max Formed Max Formed 10:1 0.750 mL 0.150 mL 2.100 mL 0.075 mM (15 %) 0.245 mM (49 %) 20:1 1.500 mL 0.056 mL 1.350 mL 0.095 mM (19 %) 0.335 mM (67 %)

200 iPr2 I Tp Cu . The first spectrum taken after addition of Ph3CSNO shows a peak at 625 nm with a maximum absorbance already reached. This peak decreases and a new peak at 495

iPr2 iPr2 nm grows in. The amount of Tp CuSCPh3 and Tp Cu(NO) formation increases with increasing copper concentration.

iPr2 Reaction of Tp Cu(NO) with C6F5SNO (Section 4.6).

iPr2 iPr2 Reaction of Tp Cu(NO) with C6F5SNO. Tp Cu(NO) was prepared in situ from the

iPr2 I reaction of Tp Cu (2.5 mL, 3.0 mM, 0.009 mmol) with addition of 5 equiv of NOgas

(1.00 mL, 0.038 mmol). The solution immediately turns red with A495 nm = 2.5 indicating

87 % conversion. A 10 mM solution of C6F5SNO was made in situ from the reaction of

C6F5STl (0.020 g, 0.050 mmol) in 5 mL of CH2Cl2 added to stirring crystals of NOBF4

(0.006 g, 0.050 mmol). The solution immediately turns red and is allowed to stir for 3

iPr2 mins. To the cuvette containing Tp Cu(NO) (2.5 mL, 2.6 mM) was added C6F5SNO

(0.650 mL, 10 mM). The first spectrum after addition of C6F5SNO showed a new peak at

iPr2 666 nm that reached its maximum absorbance, Tp CuSC6F5 (A666 nm = 2.40, 15 % yield). Both peaks at 495 nm and 666 nm decrease in intensity over the course of the reaction.

Synthesis of Tp(CF3)2CuI(THF) (4) and reactivity with RSNO (Section 4.7).

3,5-bis(trifluoromethyl)pyrazole. 1,1,1,5,5,5-hexafluoropentane-2,4-dione (10.00 g,

48.1 mmol) was added slowly at 0 °C to a solution of hydrazine hydrate (2.400 g, 48.1 mmol) in 500 mL EtOH and allowed to stir for 30 mins. The solvent was removed in vacuo at 40 °C to yield a white solid. 3,5-bis(trifluoromethyl)-4,5-dihydro-pyrazol-5-ol

1 1 was the product after the reaction as confirmed by H NMR analysis. H NMR (CDCl3, 201 300 MHz, 25 °C), δ 6.542 (s br, 1, OH), 3.346 (d, 1, CH'), 3.033 (d, 1, CH). The white solid was melted and heated with stirring as water distilled slowly. After cessation of water evolution, the temperature rose and a material boiling at 147-149 °C (1 atm) distilled and solidified. The solid was dissolved in CH2Cl2, was dried (MgSO4), filtered, and the solvent was removed in vacuo to give a pure white solid, 3,5-

1 bis(trifluoromethyl)pyrazole, 5.0 g, 50 % yield. H NMR (CDCl3, 300 MHz, 25 °C), δ

19 12.970 (s br, 1, NH), 6.928 (s, 1, CH). F{H} NMR (CDCl3, 300 MHz, 25 °C), δ -

82.368.

Tp(CF3)2Na(THF). Tp(CF3)2Na(THF) this compound was synthesized according to the

50 literature procedure with slight modifications. Under nitrogen, NaBH4 (0.265 g, 7.0 mmol) and freshly distilled 3,5-(CF3)2PzH (5.0g, 24.5 mmol) were placed in a 100mL round bottom flask equipped with a reflux condenser and a stir-bar. Dry kerosene (0.5 mL, dried by running through a short alumina column) was added to the mixture, and the mixture was gradually heated in an oil bath under nitrogen to about 200 °C. When the temperature reached about 90 °C, 1 mL of dry toluene was added from the top of the condenser. This was mainly to break the solid cake and to wash down the subliming 3,5-

(CF3)2PzH. The temperature was maintained at 195-205 °C for 6 h without going above this temperature range. The mixture was allowed to cool to room temperature. After removing toluene under vacuum, petroleum ether (10 mL) was added to the resulting solid, stirred for a few minutes, and the solution was filtered to obtain a white solid. This solid was extracted into diethyl ether (3 × 10 mL), and the combined ether extracts were

(CF3)2 filtered, and dried in vacuo. The resulting solid Tp Na(H2O) was dissolved in THF 202 (20 mL), and stirred overnight. The solvent was removed under vacuum and the solid was

recrystallizated from toluene/THF t -35 °C to yield 2.346 g of pure product (49 % yield).

(CF3)2 1 Tp Na(H2O): H NMR (C6D6, 300 MHz, 25 °C), δ 6.285 (3 , 1, CH), 0.830 (s, 2,

(CF3)2 1 H2O). Tp Na(THF): H NMR (C6D6, 300 MHz, 25 °C), δ 6.343 (3, 1, CH), 3.512 (s,

19 4, THF-CH2), 1.389 (s, 4, THF-CH2). F{H} NMR (CDCl3, 300 MHz, 25 °C), δ -59.683

(s, 9, CF3-back), -63.585 (s, 9, CF3-front).

Tp(CF3)2Cu(THF) (4). Tp(CF3)2Na(THF) (0.260 g, 0.363 mmol) and bis(copper(I)trifluoromethanesulfonate)benzene (0.091 g, 0.182 mmol) were combined in

THF (15 mL) at -35 °C and allowed to stir and warm up to room temperature for 1 hr.

The THF was removed in vacuo, the residue was extracted (3 × 5 mL) with toluene, and filtered over Celite. The volume was reduced to 3 mL and layered with 0.500 mL of pentane and allowed to crystallize in the freezer at -35 °C overnight. Tp(CF3)2CuI(THF)

1 was obtained as colorless crystals, 170 mg (62 % yield). H NMR (C6D6, 300 MHz, 25

19 °C), δ 6.240 (3, 1, CH), 3.910 (s, 4, THF-CH2), 1.453 (s, 4, THF-CH2). F{H} NMR

(CDCl3, 300 MHz, 25 °C), δ -59.832 (s, 9, CF3-back), -62.964 (s, 9, CF3-front).

Reaction of Tp(CF3)2Cu(THF) (4) with AdSNO. A green solution of AdSNO (0.026 mg,

0.132 mmol) in 1 mL of toluene and added to a colorless solution of Tp(CF3)2Cu(THF) (4)

(0.100 g, 0.132 mmol) in 5 mL of toluene at -35 °C. The solution was allowed to stir for

10 mins. No color change or effervescence was observed. The volume was reduced to 2 mL, layered with 0.500 mL of pentane and allowed to crystallize in the freezer at -35 °C overnight. Colorless crystals formed, 100 mg (86 % yield assuming crystals are

Tp(CF3)2Cu(AdSNO)). All attempts to isolate a crystal and analyze it by X-ray diffraction 203 were unsuccessful as the crystals bubble in the mineral oil and decompose. 1H NMR

19 (C6D6, 300 MHz, 25 °C), δ 6.257 (3, 1, CH), 2.104-1.201 (m, 15, AdSNO-CH). F{H}

NMR (CDCl3, 300 MHz, 25 °C), δ -59.76 (s, 9, CF3-back), -62.12 (s, 9, CF3-front).

Reaction of Tp(CF3)2Cu(THF) (4) with BnSNO. A red solution BnSNO (made in situ from the 5 min reaction of BnSTl (0.043 g, 0.132 mmol) in 3 mL of toluene added to stirring crystals of NOBF4 (0.016 g, 0.132 mmol)) is and added to a colorless solution of

Tp(CF3)2Cu(THF) (4) (0.100 g, 0.132 mmol) in 5 mL of toluene at -35 °C. The solution was allowed to stir for 10 mins. No color change or effervescence was observed. The volume was reduced to 2 mL, layered with 0.500 mL of pentane and allowed to crystallize in the freezer at -35 °C overnight. Colorless crystals formed, 80 mg (72 % yield assuming crystals are Tp(CF3)2Cu(BnSNO), 65 % yield assuming crystals are

Tp(CF3)2Cu(BnSSBn)). All attempts to isolate a crystal and analyze it by X-ray diffraction were unsuccessful as the crystals bubble in the mineral oil and decompose. 1H NMR 1H

19 NMR (C6D6, 300 MHz, 25 °C), δ 6.240(3, 1, CH), 3.849 (s, 4, PhCH2SSCH2Ph). F{H}

NMR (CDCl3, 300 MHz, 25 °C), δ -59.927 (s, 9, CF3-back), -62.177 (s, 9, CF3-front).

The integration of the CH2 from the benzyl species integrates to twice as much as expected if BnSNO is coordinated to the copper(I). We suspected that BnSSBn is the species that is coordinating. We confirmed this by independent synthesis of

Tp(CF3)2Cu(BnSSBn).

Independent synthesis of Tp(CF3)2Cu(BnSSBn). A colorless solution BnSSBn ( 0.032 g,

0.132 mmol) is and added to a colorless solution of Tp(CF3)2Cu(THF) (4) (0.100 g, 0.132 mmol) in 5 mL of toluene at -35 °C. The solution was allowed to stir for 10 mins. No 204 color change or effervescence was observed. The volume was reduced to 2 mL, layered

with 0.500 mL of pentane and allowed to crystallize in the freezer at -35 °C overnight.

1 1 Colorless crystals formed, 115 mg (94 % yield). H NMR H NMR (C6D6, 300 MHz, 25

19 °C), δ 6.238(3, 1, CH), 3.849 (s, 4, PhCH2SSCH2Ph). F{H} NMR (CDCl3, 300 MHz,

25 °C), δ -59.900 (s, 9, CF3-back), -62.156 (s, 9, CF3-front).

205 References

(1) Schonhoff, C. M.; Matsuoka, M.; Tummala, H.; Johnson, M. A.; Estevez, A. G.; Rui

Wu, R.; Kamaid, A.; Ricart, K. C.; Hashimoto, Y.; Gaston, B.; Macdonald, T. L.; Xu, Z.;

Mannick, J. B. Proc. Natl. Acad. Sci. USA 2006, 103.

(2) Butler, A.; Nicholson, R. Life, Death and Nitric Oixde; Royal Society of Chemistry,

2003.

(3) Rassaf, T.; Kleinbongard, P.; Preik, M.; Dejam, A.; Gharini, P.; Lauer, T.;

Erckenbrect, J.; Duschin, A.; Schulz, R.; GHusch, G.; Feelisch, M.; Kelm, M. Clin. Res.

2002, 91, 470.

(4) Romeo, A. A.; Capobianco, J. A.; English, A. M. J. Am. Chem. Soc. 2003, 125,

14370.

(5) Stamler, J. S. Cir. Res. 2004, 94, 414.

(6) Lee, J.; Chen, L.; West, A. H.; Richeter-Addo, G. B. Chem. Rev. 2002, 102, 1019.

(7) Williams, D. L. H. Acc. Chem. Res. 1999, 32, 869.

(8) Mellion, B. T.; Ignarro, L. J.; Myers, C. B.; Ohlstein, E. H.; Ballot, B. A.; Hyman,

A. L.; Kadowitz, P. J. Mol. Pharmacol. 1983, 23, 653.

(9) Ramachandran, N.; Root, P.; Jiang, X.; Hogg, P. J.; Mutus, B. Proc. Natl. Acad. Sci.

U.S.A. 2001, 98, 9539.

(10) Dicks, A. P.; Swift, H. R.; Williams, D. L. H.; Butler, A. R.; AlSadoni, H. H.; Cox,

B. G. Journal of the Chemical Society-Perkin Transactions 2 1996, 481.

(11) Williams, D. H. L. The Chemistry of Amino, Nitroso, Nitro, and Related Groups,

Supplement F2; John Wiley and Sons: Chichester, 1996. 206 (12) Williams, D. L. H. Chem. Commun. 1996, 1085.

(13) Williams, D. L. H. Methods Enzymol. 1996, 268, 299.

(14) Singh, R. J.; Hogg, N.; Joseph, J.; Kalyanaraman, B. J. Biol. Chem. 1996, 271,

18596.

(15) Dicks, A. P.; Williams, D. L. H. Chem. Biol. 1996, 3, 655.

(16) Jourd'heuil, D.; Laroux, F. S.; Miles, A. M.; Wink, D. A.; Grisham, M. B. Arch.

Biochem. Biophys. 1999, 361, 323.

(17) Gartner, A.; Weser, U. FEBS Lett. 1983, 155, 15.

(18) Gordge, M. P.; Hothersall, J. S.; Neild, G. H.; Dutra, A. A. J. Pharmacol. 1996,

119, 533.

(19) Singh, R. J.; Hogg, N.; Goss, S. P.; Antholine, W. E.; Kalyanaraman, B. Arch.

Biochem. Biophys. 1996, 372, 8.

(20) Perissinotti, L. L.; Turjanski, A. G.; Estrin, D. A.; Doctorovich, F. J. Am. Chem.

Soc. 2005, 127, 486.

(21) Barnett, D. J.; Rios, A.; Williams, D. L. H. J. Chem. Soc. Perkin Trans. 2 1995,

1279.

(22) Barnett, D. J.; McAninly, J.; Williams, D. L. H. J. Chem. Soc. Perkin Trans. 2

1994, 1131.

(23) Houk, K. N.; Hietbrink, B. N.; Bartberger, M. D.; R., M. P.; Choi, R. Y.; Voyksner,

R. D.; Stamler, J. S.; Toone, E. J. J. Am. Chem. Soc. 2003, 125, 6972.

(24) Solomon, E. I.; Szilagyi, R. K.; George, S. D.; Basumallick, L. Chem. Rev. 2004,

104, 419. 207 (25) Kitajima, N. Adv. Inorg. Chem. 1992, 39, 1.

(26) Thompson, J. S.; Marks, T. J.; Ibers, J. A. Proc. Natl. Acad. Sci. USA 1977, 74,

3114.

(27) Kitajima, N. F., K.; Tanaka, M.; Moro-oka, Y. J. Am. Chem. Soc. 1992, 114, 9232.

(28) Vos, M. H.; Lipowski, G.; Lambry, J. C.; Martin, J. L.; Liebl, U. Biochemistry

2001, 40, 7806.

(29) Scrivens, G.; Gilbert, B. C.; Lee, T. C. P. J. Chem. Soc. Perkin Trans. 2 1995, 955.

(30) Tocheva, E. I.; Rosell, F. I.; Mauk, A. G.; Murphy, M. E. P. Science 2004, 304,

867.

(31) Tocheva, E. I.; Rosell, F. I.; Mauk, A. G.; Murphy, M. E. P. Biochemistry 2007, 46,

12366.

(32) Ruggiero, C. E.; Carrier, S. M.; Antholine, W. E.; Whittaker, J. W.; Cramer, C. J.;

Tolman, W. B. J. Am. Chem. Soc. 1993, 115, 11285.

(33) Schneider, J. L.; Carrier, S. M.; Ruggiero, C. E.; Young Jr., V. G.; Tolman, W. B.

J. Am. Chem. Soc. 1998, 120, 11408.

(34) Fujisawa, K.; Tateda, A.; Miyashita, Y.; Okamoto, K.; Paulat, F.; Praneeth, V. K.

K.; Merkle, A.; Lehnert, N. J. Am. Chem. Soc. 2008, 130, 1205.

(35) Kitajima, N.; Hikichi, S.; Tanaka, M.; Morooka, Y. J. Am. Chem. Soc. 1993, 115,

5496.

(36) Matsunaga, Y.; Fujisawa, K.; Ibi, N.; Miyashita, Y.; Okamoto, K. Inorg. Chem.

2005, 44, 325.

208 (37) Fujisawa, K.; Tetsuya Ono, T.; Ishikawa, Y.; Amir, N.; Miyashita, Y.; Okamoto, K.

I.; Lehnert, N. Inorg. Chem. 2006, 45, 1698.

(38) Kitajima, N. F., K.; Moro-oka, Y. J. Am. Chem. Soc. 1990, 112, 3210.

(39) Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Morooka, Y.; Hashimoto, S.; Kitagawa,

T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am. Chem. Soc. 1992, 114, 1277.

(40) Varonka, M. S.; Warren, T. H. Inorg. Chem. 2009, Submitted.

(41) Nicolai Lehnert, N.; Cornelissen, U.; Neese, F.; Ono, T.; Noguchi, Y.; Okamoto, K.

I.; Fujisawa, K. Inorg. Chem. 2007, 46, 3916.

(42) Dias, H. V. R.; Polach, S. A.; Goh, S. A.; Archibong, E. F.; Marynick, D. S. Inorg.

Chem. 2000, 39, 3894.

(43) Badiei, Y. M.; Krishnaswamy, A.; Melzer, M. M.; Warren, T. H. J. Am. Chem. Soc.

2006, 128, 15056.

(44) Badiei, Y. M.; Dinescu, A.; Dai, X.; Palomino, R. M.; Heinemann, F. W.; Cundari,

T. R.; Warren, T. H. Angew. Chem. Int. Ed. Engl. 2008, 47, 9961.

(45) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.; Timmers, F. J.

Organometallics 1996, 15, 1518.

(46) Jenkins, D. M.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 1748.

(47) Varonka, M. S.; Warren, T. H. Inorg. Chim. Acta. 2007, 360, 317.

(48) Girard, P.; Guillot, N.; Motherwell, W. B.; Hay-Motherwell, R. S.; Potier, P.

Tetrahedron 1999, 55, 3573.

(49) Sun, Y. J.; Zhang, L. Z.; Cheng, P.; Lin, H. K.; Yan, S. P.; Sun, W.; Liao, D. Z.;

Jiang, Z. H.; Shen, P. W. Inorg. Chem. 2003, 42, 508. 209 (50) Dias, H. V. R.; Jin, W.; Kim, H. L.; Lu, H. L. Inorg. Chem. 1996, 35, 2317.

210 Chapter 5

S-Nitrosothiol and Nitrosonium Reactivity at Two Coordinate

Copper(I) Thiolates

Abstract

A small family of stable N-heterocyclic carbene Cu(I)-thiolates IPrCu(SR) were synthesized to explore their reactivity with S-nitrosothiols RSNOs and the nitrosonium

ion. Despite the propensity of copper ions in a variety of coordination environments to

catalyze the loss of NO from RSNOs, we found that IPrCu(SR) complexes readily engage

in trans-s-nitrosation reactions with negligible RSNO decomposition. The reaction of

IPrCu(StBu) and BnSNO are in equilibrium with IPrCu(SBn) and tBuSNO to give an equilibrium constant Keq = 0.25(5) in CDCl3 at RT. The rate of trans-s-nitrosation tracks with the size of the S-substituents. For instance, degenerate trans-s-nitrosation between

t t 1 IPrCu(S Bu) and BuSNO is slow on the H NMR timescale at RT in polar (CDCl3) and non-polar (C6D6) solvents. Degenerate exchange between the smaller IPrCu(SBn) and

1 -1 BnSNO, however, is fast on H NMR timescale (kobs > 100 s at 0.1 mM concentrations of each component) in C6D6 but is at the slow exchange limit in CDCl3. All attempts at identifying a possibly trans-s-nitrosation intermediate such as a copper(I) nitroxyl disulfide IPrCu[SR(NO)SR'] were unsuccessful. The reaction of the nitrosonium source

NO[BF4] with 2 equiv. IPrCu(SR) leads to formation of RSNO and the novel dinuclear

t copper(I) thiolates [{IPrCu}2(µ-SR)]BF4 (R = Bn or Bu). Importantly, loss of NOgas from

211 S-nitrosothiols does not readily take place owing to the less electron-rich nature of the

[IPrCu]+ fragment as compared to the β-diketiminates [NN]Cu or tris(pyrazolyl)borates

TpCu.

Introduction

S-nitrosothiols, first synthesized in 1909,1 were not recognized until the 1990s as a biological source of nitric oxide.2-4 They circulate through the blood and can serve as vasodilators5,6 or be used as anti-platelet agents.3,7 For decades, copper ions have been demonstrated to catalyze the decomposition of S-nitrosothiols RSNOs to NOgas and ½

RSSR (Scheme 5.1).8-13 There are also, clear indications that copper does play a part in the in vivo reactions of RSNOs, specifically with vasodilation14 responses and anti- platelet aggregation.15 Given the facility at which S-NO decomposition takes place in the presence of Cu ions, we wanted to explore the possibility of synthesizing copper complexes that would interact with RSNOs without decomposition to NOgas and disulfide

RSSR.

Scheme 5.1. RSNO decomposition by copper ions.

Another important reaction of S-nitrosothiols is trans-s-nitrosation, the direct transfer of the NO group from one thiol to another without the generation of free NO

(Scheme 5.2).16 Trans-s-nitrosation is generally facile under physiological conditions.17

Trans-s-nitrosation is thought to be the mechanism by which NO is transferred from low

212 Scheme 5.2. Nitroxyl disulfide intermediate proposed for trans-s-nitrosation.

molecular weight endogenous RSNOs such as S-nitrosocystein (CysNO) and S- nitrosoglutathione (GSNO) to S-nitroso-albumin.3 The nitroso adduct of serum albumin

(S-nitrosoalbumin, MWT = 66000 Da) has a blood concentration of 5 M and composes

3 about 80 % of the blood plasma RSNO content. S-nitroso-albumin is a large protein, too large to cross the plasma membrane. Any free NO released from S-nitroso-albumin would be easily scavenged by the iron in hemoglobin. Thus, trans-s-nitrosation from S- nitroso-albumin to low molecular weight endogenous RSNOs allows for the direct transfer of NO to intracellular targets.18,19

Trans-s-nitrosation has been reported to be a reversible reaction, first order in both the thiol and nitrosothiol and does not involve free NO as an intermediate.20 A nitroxyl disulfide intermediate has been proposed on the basis of ESI/MS experiments for

S-nitroso-N-acetylpenicillamine (SNAP), GSNO/GSH, and SNAP/AR (aldose redutase).21-23 A detailed computational study on trans-s-nitrosation between MeSNO and

MeS- describes the nitroxyl intermediate in detail. Methanethiolate (a) and S- nitrosomethane thiol (b) combine in the gas phase to form an ion-molecule complex (c).

A transition state (d) for nucleophilic attack at the nitrogen of b leads to the novel

213 a b

c

d e

Figure 5.1. Geometries for transnitrosation stationary points, selected bond lengths in Å, Figure taken from Houk, et al. J. Am. Chem. Soc. 2003 125, 6972-6976, used without permission. nitroxyl disulfide intermediate (e). A second transition state and dissociate completes the thiolate exchange (Figure 5.1).

In a later study, Doctorovich et al. characterized the nitroxyl disulfide intermediate of S-nitroso-L-cysteine ethyl ester (ECysSNO) with L-cysteine ethyl ester thiolate (ECyS-) by UV-Vis and NMR (15N, 13C DEPT, 2D HECTOR, and COSY)

24 15 experiments in polar solvents (Scheme 5.3). The N NMR spectrum in methanol-d4 shows a new peak at δ 41 ppm attributed to the nitroxyl disulfide intermediate which is significantly upfield shifted as compared to the S-nitrosothiol ECyS15NO which appears at δ 393 ppm. The 13C NMR spectra of the reaction mixture also shows a new signal assigned to –CH2S of the nitroxyl disulfide intermediate. Kinetic measurements were

214 obtained for this trans-s-nitrosation in water to give the second order rate constant k2 =

15.6(2) × 10-3 M-1s-1 (26 °C, pH 7.4). This is in reasonable agreement with other experimental values of NO transfer from RSNOs to RS- in aqueous solutions.20

Varying the stoichiometry between ECySNO and ECyS- led to different percentages of the nitroxyl disulfide intermediate formation in methanol-d4. It was found that when the ratio was 2.78:1 ECyS-:ECySNO, the intermediate was formed in 93 % yield. This is suggestive of an equilibrium existence among the different species involved. These experiments were all conducted in polar solvents—water or methanol.

The computational results suggest that the proposed trans-s-nitrosation mechanism should be less operative in less polar solvents. This is the first example of spectroscopic evidence to support the formation of the nitroxyl disulfide intermediate which helps to shed light on the mechanism of trans-s-nitrosation.

Scheme 5.3. Trans-s-nitrosation reaction between ECyS- and ECySNO is proposed to go through a nitroxyl disulfide intermediate [ECyS(NO)ECyS]-.

Copper ions in biology exist in a variety of coordination environments. As motivation for the model complexes employed in this thesis, we focus on cupredoxins involved in electron transfer (Blue, Type 1 and Purple, CuA) and enzymatic proteins

215 (Red, Type 2) (See Chapter 4). There is experimental evidence for a link between blue,

purple, and red copper transformations in nitrous oxide reductase.25 Cupredoxins contain copper-thiolate ligation and the differences in color arise from the subtle differences in the geometry and ligand donor set about the copper active site. Blue (Type 1) copper proteins have a mononuclear copper active site in which the copper-thiolate resides in a trigonal environment at Cu completed by two histidine donors.26 Red (Type 2) copper proteins have a mononuclear Cu active site where the single Cu-thiolate ligand is in a

27 distorted tetrahedral geometry with two Cu-histidine coordinating. Purple (CuA) copper proteins posses a binuclear copper cluster in which two bridging thiolates form a diamond Cu2S2 core. Each Cu is also coordinated by one histidine leading to trigonal coordination at each copper. This is an example of a delocalized Cu1.5Cu1.5 system28

I II which motivated the synthesis of a Cu Cu species to synthetically model the CuA copper protein site (Figure 5.3).29

Figure 5.2. Cupredoxin folds with blue, purple, and red copper. (A) Blue copper azurin (Pseudomonas aeruginosa),PDB ID code 4AZU. (J.Mol.Biol. 221: 765-772) (B) Purple CuA of N2O reductase (P. denitrificans), PDB ID code 1FWX. (J.Biol.Chem. 275: 41133-41136) (C) Red Type 2 Nitrosocyanin (Nitrosomonas europaea), PDB ID code 1IBY. (Biochemistry 40: 5674-5681) Figure216 taken from Savelieff, et al. Proc. Natl. Acad. Sci. USA 2008, 105, 7919-7924, used without permission.

To better understand the types of interactions available between copper complexes and S-nitrosothiols, we sought to prepare stable copper-thiolate complexes which do not readily decompose S-nitrosothiols. There are only a few examples of mononuclear copper-thiolates known using small molecule complexes (Figure 5.3). For instance, a small number of mononuclear copper(II) thiolates supported by tris(pyrazolyl)borate (Chapter 4) and β-diketiminate ligands (Chapter 3) are known.

Each of these copper(II)-thiolates is thermally unstable with respect to the loss of disulfide RSSR and formation of the corresponding copper(I) species. In addition,

TpCuII-SR species exhibit important reactivity with RSNOs to form the copper(I) nitrosyls TpCu(NO) and RSSR.

Figure 5.3. Examples of small molecule Cu-thiolates.

217 Examples of two coordinate Cu(I)-thiolates [NHC]Cu-SR using a solitary N- heterocyclic carbene supporting ligand have been recently reported by Gunnoe.30 These copper-thiolates are colorless and thermally stable due to the copper being in the +1 oxidation state rather than the +2 oxidation state. They also catalyze the addition of the

S−H bonds of thiols RSH across electron-deficient olefins to regioselectively produce anti-Markovnikov products.30 N-heterocyclic carbenes are attractive ancillary ligands for several reasons:31 (1) they are strong σ-donors resulting in metal-carbene bonds which are not particularly susceptible to metal-carbene dissociation, (2) the steric properties can be modified by altering the groups bound to the N-atoms, and (3) the activity of NHC ligands can be modified by the introduction of electron donating or withdrawing substituents.31

The NHC ligands are readily synthesized (Scheme 5.4) from the condensation of glyoxal with 2 equiv. substituted anilines to form the corresponding α-diimine. The α- diimine then reacts with paraformaldehyde in toluene with HCl/ether solution to form the imidazolium chloride.32,33 To make the copper(I) chloride derivative [NHC]CuCl, KOtBu and anhydrous CuCl are added to the imidaolium salt in THF to cleanly form the IPrCuCl in 90 % yield when crystallized from dichloromethane (Scheme 5.4).34 This was the starting material for [NHC]Cu-SR species as well as {[NHC]Cu}+ cations.35

218 Scheme 5.4. Synthesis of imidizolium salt precursors as well as the N-heterocyclic copper(I) chloride complex IPrCuCl.

Since nitric oxide generally reacts with higher oxidation state metals to reductively nitrosate them, we chose copper(I) as our oxidation state. Only a couple of examples of S-nitrosothiol coordination to metal centers have been observed.36,37 These complexes are kinetically inert due to their d6 and octahedral configuration. To encourage interactions between metals and S-nitrosothiols, poor binding ligands with a propensity to lose NOgas, we felt that a low coordinate environment would give us the best chance observe [M](RSNO) adducts which are likely to be rather labile.

219 Results and Discussion

5.1. Synthesis and characterization of copper(I) thiolates IPrCu(SR).

t tBu Reaction of IPrCuCl with TlSR (R = Bn (1), Bu (2), CH2Ar (3), CPh3 (4), or

NaSR (R = Me (5)) provides the two coordinate IPrCu(SR) complexes in good yields (70

- 77%) (Scheme 5.5). IPrCu(SBn) (2) was previously synthesized by the reaction of the copper(I) methyl complex IPrCuMe with BnSH.30 No elemental analysis was reported for this compound in this paper as the authors suggest that the compound is too unstable for this type of characterization. We found exactly the opposite to be true. Copper(II)- thiolates are particularly unstable species. This is due to the ease of reducibility of the

Cu2+ to Cu1+ with loss of the thiyl radical RS• from copper(II)-thiolates [CuII]-SR

(Scheme 5.6). Electron withdrawing R groups attached to the thiolate ligand in [CuII]-SR species help to stabilize the Cu-S bond and thus allow for isolable copper(II)-thiolates.

Ligands and R groups that are sufficiently bulky also help stabilize the copper(II)-thiolate

Scheme 5.5. Synthesis of IPrCu(SR)from reaction of IPrCuCl and TlSR (R = Bn (1), tBu tBu (2), CH2Ar (3) CPh3 (4) or NaSR (R = Me (5)).

220 because this prevents dimerization which facilitates the formation of the RSSR. These N-

heterocyclic copper(I)-thiolates are remarkably stable due to the low oxidation state of

the metal, since further reduction of Cu1+ to Cu0 in this system does not appear to be a favorable process. Thus the copper-thiolate is stable towards loss of RS• to form RSSR.

This family of Cu(I)-thiolates proves to be easy to synthesize, colorless, thermally stable, and convenient for solution NMR analysis.

Scheme 5.6. Electron movement illustrating decomposition of CuII-SR to form CuI and the RS•. CuI-SR species are considerably more stable to this decomposition pathway. tBu The X-ray structure of IPrCu(SCH2Ar ) (3) features a nearly linear C3-Cu-S linkage (angle = 172.80(7)º) with Cu-S and S-C28 and distances of 1.840(3) and 2.130(1)

Å, respectively (Figure 5.4). The Cu-S-C28 linkage has an angle of 105.80(7)°. The C3-

Cu-S linkage for the trityl species 4 is slightly more bent with an angle of 167.36(8)°

(Figure 5.5). The Cu-S-C28 linkage has an angle of 109.85(8)° with Cu-S and S-C28 distances of 2.130(1) and 1.871(3) Å, respectively.

1 The room temperature H NMR spectra of copper thiolates 1 – 5 either in CDCl3 or C6D6 are quite similar suggesting similar solution behavior among these complexes.

For instance, each gives a singlet near δ 7.1 ppm for the carbene backbone H atoms, a septet near δ 2.6 ppm corresponding to the isopropyl methine H atom CHMe2 along with

221

two sets of diastereotopic doublets for the isopropyl methyl groups CHMe2 near δ 2.3 and

2.2 ppm.

222 N1 C1 S Cu N2 C28

tBu Figure 5.4 X-ray structure of IPrCu(SCH2Ar )(3) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–C1 1.884(2), Cu–S 2.1304(7), S–C28 1.840(3), C1– Cu–S 172.80(7), Cu-S-C28 105.77(9), N2–C1–Cu 126.76(17), N1–C1–Cu 129.18(17), C29–C281–S 118.55(9). 223 S N1 C1 Cu C28 N2

Figure 5.5 X-ray structure of IPrCu(SCPh3) (4) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–C1 1.983(3), Cu–S 2.1304(10), S–C28 1.871(3), C1– Cu–S 167.36(8), Cu-S-C28 109.85(8), N2–C1–Cu 122.14 (18), N1–C1–Cu 133.95(19), C29–C28–S 109.74(17).

224 Table 5.1. Crystallographic parameters for 3 and 4.

Compd. 3 4

Formula C38H51CuN2S C46H51CuN2S, C4H10O Mol. Wt. 631.39 801.61 Temp.(K) 100(2) 100(2) Crystal Block Block description Crystal color Colorless Colorless Crystal size 0.2018 0.0184 (mm3) System Monoclinic Monoclinic Space group C2/c P2(1)/n a (Å) 29.689(3) 12.714(4) b (Å) 15.6315(13) 16.891(5) c (Å) 17.0794(14) 21.335(7) α (deg) 90 90 β (deg) 94.8040(10) 103.570(5) γ (deg) 90 90 Volume (Å3) 7898.4(11) 4454(2) Z 8 4 θ range (deg) 1.86-27.00 1.71-28.00 Measd reflns 32710 39851 Unique reflns 8618 10605 GOF of F2 1.076 1.045

R1 ( I > 2σ(I)) 0.0406 0.0523 wR2 (all data) 0.1087 0.1346 Largest diff. 0.367 and 0.919 and peak and hole -0.298 -0.505 (e-.Å-3)

225 5.2. Trans-s-nitrosation reactions between IPrCu(StBu) and BnSNO

Followed by 1H NMR spectroscopy, the addition of 1 equiv. BnSNO to

t IPrCu(S Bu) (2) in CDCl3 results in the formation IPrCu(SBn) (1) (δ = 3.28 ppm;

t t SCH2Ph) and BuSNO (δ = 1.94 ppm; S Bu) along with the partial disappearance of

t t IPrCu(S Bu) (δ = 1.07 ppm; S Bu) and BnSNO (δ = 4.67 ppm; SCH2Ph) (Figure 5.6a).

Thus, a trans-s-nitrosation equilibrium ensues (Scheme 5.7). Since all four species in the trans-s-nitrosation reaction are observable, we can calculate Keq = 0.25(5) for this equilibrium. Thus, the bulky, more electron-rich thiolate group StBu is favored at copper in 2 vs. “NO+” in tBuSNO.

t When BnSNO is added to IPrCu(SBu ) in C6D6, we also see evidence for trans-s- nitrosation but note extremely broad resonances for the corresponding species in exchange. The two types of 1H NMR methylene resonances expected for pure BnSNO (δ

= 4.08 ppm; SCH2Ph) and IPrCu(SBn) (δ = 3.81 ppm; SCH2Ph) are seen as one broad peak at δ 3.99 ppm rather than distinct signals (Figure 5.6b). Similarly, the methyl peaks for tBu are broaden and slightly shifted closer to each other; tBuSNO (downfield shift, δ

= 1.49 ppm; StBu) and IPrCu(StBu) (upfield shift δ = 1.51 ppm; StBu) appear as barely separated components. For comparison, the StBu groups of pure tBuSNO and

IPrCu(StBu) appear at δ = 1.48 and1.52 ppm, respectively.

t While the very broad SCH2Ph and overlapping S Bu resonances prevent the calculation of a reliable equilibrium constant, these broad peaks indicate that trans-s- nitrosation takes place on the NMR timescale. Importantly, NOgas loss from RSNOs does not readily take place. 226 Scheme 5.7. Trans-s-nitrosation between IPrCu(StBu) (2) and BnSNO. The equilibrium favors the bulky, more strongly electron-donating StBu group at copper vs. “NO+”.

PhCH2SSCH2Ph IPrCu(SCMe3)

THF Me3CSNO

PhCH2SNO IPrCu(SCH2Ph)

4.8 4.6 4.4 4.2 4.0 3.8 3.6 3.4 3.2

1.01.52.02.53.03.54.04.55.05.56.06.57.07.5 1 a) H NMR, CDC3

Me3CSNO

PhCH2SNO + IPrCu(SCMe3) IPrCu(SCH2Ph)

THF

1 a) H NMR, C6D6

1 Figure 5.6. H NMR spectra (300 MHz, 25 °C) in (a) CDCl3 or (b) C6D6 illustrating trans-s-nitrosation between IPrCu(StBu) (2) and BnSNO.

227 The reaction does not follow the same trend for thiol exchange. When 1 equiv.

t BnSH is added to IPrCu(S Bu) in CDCl3 the reaction goes completely to the right as monitored by 1H NMR spectroscopy forming tBuSH (δ = 1.79 and 1.47 ppm; HStBu and

t HS Bu) and IPrCu(SBn) (δ = 3.30 ppm; SCH2Ph) (1) quantitatively (Scheme 5.8).

Scheme 5.8 Trans-thiolation between IPrCu(StBu) (2) and BnSH. The reaction goes completely to the right favoring the less bulky IPrCu(SR) and tertiary RSH.

Me3CSH

Me3CSH IPrCu(SCH2Ar)

3.3 3.1 2.9 2.7 2.5 2.3 2.1 1.9 1.7 1.5

1.01.52.02.53.03.54.04.55.05.56.06.57.07.5

1 Figure 5.7. H NMR spectrum (300 MHz, 25 °C, CDCl3) illustrating trans-thiolation between IPrCu(StBu) 2 and BnSH.

228 5.3. Trans-s-nitrosation reactions between IPrCuStBu and tBuSNO

t In both polar (CDCl3) and nonpolar (C6D6) NMR solvents, IPrCu(S Bu) (2) does not exchange with tBuSNO on the 1H NMR timescale (Scheme 5.9). The StBu peaks of the S-nitrosothiol tBuSNO and the copper(I)-thiolate IPrCu(StBu) reaction are sharp and not shifted from the pure starting materials (Figure 5.8). Despite the fast exchange on he

NMR timescale seen in the non-degenerate exchange involving –SBn and –StBu species in C6D6 discussed above, perhaps the bulky tert-butyl significantly slow down trans-s- nitrosation.

Scheme 5.9. Degenerate trans-s-nitrosation of tBuSNO and IPrCu(StBu) 2.

IPrCu(SCMe3)

Me3CSNO

1.55 1.50 1.45

a) benzene-d6 Me3CSNO

IPrCu(SCMe3)

1.9 1.7 1.5 1.3 1.1 b) chloroform-d1

7.5 6.5 5.5 4.5 3.5 2.5 1.5 Figure 5.8. 1H NMR spectra (300 MHz, 25 °C) of degenerate trans-s-nitrosation t t between BuSNO and IPrCu(S Bu) (2) in (a) benzene-d6 and (b) chloroform-d1.

229 5.4. Trans-s-nitrosation reactions between IPrCuSBn and BnSNO

In the degeneration reaction of IPrCu(SBn) with BnSNO (Scheme 5.10), we see

1 differing H NMR behavior in CDCl3 and C6D6. In the polar solvent CDCl3, the BnSNO and IPrCu(SBn) resonances are sharp and unshifted from their normal positions (Figure

5.9a). This indicates that the rate of exchange between BnSNO and IPrCu(SBn) is not on

1 the H NMR timescale CDCl3. In a similar experiment performed in the nonpolar solvent

C6D6, the two SCH2Bn peaks of BnSNO and IPrCu(SBn) normally at δ 4.08 and 3.81 ppm combine to form one broad peak at δ 3.98 (Figure 5.9b). When different

Scheme 5.10. Degenerate trans-s-nitrosation of BnSNO and IPrCu(SBn) (2).

PhCH2SNO IPrCu(SCH2Ph)

a) chloroform-d1 IPrCu(SCH2Ph) + PhCH2SNO

b) benzene-d6

1 Figure 5.9. H NMR spectra (300 MHz, 25 °C) in: (a) CDCl3 of trans-s-nitrosation between IPrCu(SBn) and BnSNO, two SCH2Ph peaks present or (b) C6D6 of trans-s- nitrosation between IPrCu(SBn) and BnSNO, one SCH2Ph peak present. 230 PhCH2SNO

IPrCu(SCH2Ph) PhCH2SNO δ(ppm) 1: 3 1 0 3.816 1 ½ 3.867 1 1 3.979 1: 2 1 2 4.021 1 3 4.027 1: 1 0 1 4.082 1: 1/2

IPrCu(SCH2Ph)

1 Figure 5.10. H NMR spectra (300 MHz, 25 °C, C6D6) of mixtures containing different ratios of IPrCu(SBn) and BnSNO. Only the SCH2Ph region (δ 4.3 - 3.6 ppm) shown.

stochiometric amounts (1/2, 1, 2, and 3 eq.) of BnSNO are added to IPrCu(SBn) the

chemical shift of the broad peak moves toward the compound in greater excess (Figure

5.10) clearly indicating an equilibrium between BnSNO and IPrCu(SBn).

We can begin to estimate the lower limit to the rate of trans-s-nitrosation in C6D6 at room temperature by considering the minimum rate kc that results in coalescence of two NMR signals originally in equal amounts (eq 5.1):

kc = π ∆νo / √2 (5.1)

Considering the broad, coalesced signal observed in C6D6 spectra which represents both the SCH2Ph resonances of BnSNO and IPrCu(SBn) which normally appear at δ 4.08 and 231 3.81 ppm in this solvent, this represents a frequency separation ∆νo = 300 (4.08 – 3.81) =

81 Hz. Thus the minimum actual rate for coalescence would be 180 s-1. Assuming an overall second order rate law for trans-s-nitrosation that has been observed in other

- systems (rate = k2[RS ][RSNO]), the actual concentrations employed (0.055 M in each

4 -1 -1 component) would lead to a second order rate constant k2 of at least 5.9 × 10 M s .

This is orders of magnitude higher than the values of 0.6-262 × 10-3 M-1s-1 observed for trans-s-nitrosation between S-nitrosothiols RSNO and the corresponding thiolate anion

RS- in aqueous solution.20,24

5.5. Variable temperature NMR spectroscopy exploring degenerate exchange

tBu tBu between IPrCuSCH2Ar and ArCH2SNO

It was of great interest to study these trans-s-nitrosation reactions by NMR spectroscopy at low temperature to explore the possibility that we might observe the proposed intermediate in the reaction, a copper(I) nitroxyl disulfide IPrCu(RSNOSR’)

(see Schemes 5.3 and 5.12). Unfortunately, IPrCu(SBn) possesses poor solubility in toluene-d8 below room temperature. Thus, low temperature NMR spectroscopy

t tBu employed the analogue with a p- Bu substituent on the thiolate IPrCu(SCH2Ar ) (3) which has (only) slightly better solubility at low temperature.

tBu tBu Before studying mixtures of ArCH2SNO and IPrCu(SCH2Ar ) by variable temperature NMR spectroscopy, it is important to recognize that S-nitrosothiols exhibit syn / anti isomeric forms with a barrier of interconversion rotation of ca. 10 kcal/mol. For instance, Ph3CNO possesses an experimental barrier to S-NO bond rotation of 10.7(7) kcal/mol38 at which requires the use of low temperatures to observe discrete syn and anti 232 isomeric forms by 1H and 15N NMR spectroscopy – at room temperature these are typically in fast exchange. The syn isomer is favored over the anti isomer in the primary and secondary S-nitrosothiols. For instance, the S-nitrosothiols MeSNO, EtSNO, iPrSNO show syn : anti ratios of 4:1, 3:1, and 2:1 at -50 °C in toluene-d8. In tertiary S- nitrosothiols such as tBuSNO in which the steric influence of the S-substituent are more important, the anti isomer is favored 6:1 over the syn isomer under similar conditions.39

At -90 °C, Houk et al. were able to observe the syn and anti isomers in the 15N NMR spectra for EtS15SNO (δ 378.4 and 447.2 ppm), iPrSNO (δ 381.9 and 446.7 ppm), and

t 15 BuSNO (δ 403.3 and 462.7 ppm) in toluene-d8. N NMR chemical shifts were also calculated using B3LYP-GIAO approach and were in relative good agreement with the experimental values suggesting an approximate 60 ppm downfield shift for the anti 15N

NMR signals relative to that for the syn resonances. There was no mention in the experimental section of this paper as to how their 15N NMR referencing was conducted.

We first carried out variable temperature 1H and 15N NMR studies out on

tBu 1 ArCH2SNO. In CDCl3 at -60 °C the major and minor H NMR (500 MHz) signals are sharp and well resolved appearing at δ 6.207 and 4.402 ppm. In toluene-d8 at -70 °C the major and minor 1H NMR (500 MHz) signals are also sharp and well resolved appear at δ

5.468 and 3.695 ppm (Figure 5.11). On the basis of the previous work described above in which the major signal was assigned as the syn isomer for primary S-nitrosothiols, we calculate a syn / anti ratio of 14:1 in each case.

233 4.158

tBu ArCH2SNO

25 °C 5.468 3.695

tBu tBu ArCH2SNO ArCH2SNO anti syn -60 °C 0.07 1.00

1 a) H NMR, toluene-d8

tBu ArCH2SNO

25 °C

tBu tBu ArCH2SNO ArCH2SNO anti syn

-70 °C 6.3 6.2 6.1 6.0 5.9 5.8 5.7 5.6 5.5 5.4 5.3 5.2 5.1 5.0 4.9 4.8 4.7 4.6 4.5 4.4 4.3 1 b) H NMR, chloroform-d1

1 tBu Figure 5.11. a) H NMR spectra (300 MHz, toluene-d8) of ArCH2SNO at 25 °C and - 1 60 °C, CH2 region shown only. b) H NMR spectra (300 MHz, chloroform-d1) of tBu ArCH2SNO at 25 °C and -70 °C, CH2 region shown only.

234 15 tBu 15 tBu N-labeling of the S-nitrosothiol to form ArCH2S NO from ArCH2SH and

tBuO15NO (Scheme 5.11) allows independent confirmation of our assignment of the syn

15 rotomer as the major species by low temperature N NMR spectroscopy. In CDCl3 at -60

°C the syn : anti isomers cannot be resolved and the there is only one 15N NMR (30.428

15 MHz) signal appearing at δ -198.4 ppm. In toluene-d8 at -70 °C the major and minor N

NMR (30.428 MHz) signals are sharp and well resolved appearing at δ 360.6 and 195.9 ppm. As the temperature is raised, the 15N signal is broadens into the baseline due to this exchange process. At room temperature, the chemical shift of the average peak is slight downfield shifted to -196.4 ppm from the pure syn peak, -195.9 ppm (Figure 5.12).

tBu 15 t 15 tBu Scheme 5.11. Synthesis of ArCH2S NO from BuO NO with ArCH2SH.

235 tBu 15 a) 25 ° C ArCH2S NO

tBu 15 tBu 15 ArCH2S NO ArCH2S NO anti syn

b) -70 ° C 360 340 320 300 280 260 240 220 200

15 1 Figure 5.12. N{ H} NMR spectra (30.428 MHz, toluene-d8) at (a) 25 °C and (b) -70 t 15 °C, anti/syn 0.08:1. (Referenced to BuO NO, toluene-d8 at 196 ppm).

236 tBu tBu Monitoring a 1:1 mixture of IPrCuSCH2Ar (3) and ArCH2SNO in CDCl3 and

1 C6D5CD3 by low temperature H NMR spectroscopy showed separate SCH2Ar

tBu tBu resonances for the anti and syn isomers of ArCH2SNO as well as IPrCu(SCH2Ar ) at

-60 °C (Scheme 5.12, Figure 5.13; CDCl3) and -70 °C (Figure 5.14; C6D5CD3)). For the

CDCl3 solution, raising the temperature led to broadening of the CH2Ar peaks into the baseline at 0 °C. At 50 °C a new CH2 peak grows in at δ 4.704 ppm. We attribute this

tBu peak to the coalescence of the anti and syn isomers of ArCH2SNO and not a trans-s-

1 nitrosation intermediate. In toluene-d8, at 50 C°, one SCH2Ar H NMR signal appears at

tBu tBu δ 4.21 ppm due to coalescence of the ArCH2SNO and IPrCu(SCH2Ar ) peaks.

tBu Scheme 5.12. Trans-s-nitrosation reaction between 3 and ArCH2SNO.

We see no evidence of a trans-s-nitrosation intermediate when a 1:1 mixture of

tBu tBu 15 15 IPrCuSCH2Ar (3) and ArCH2S NO is monitored by low temperature N NMR spectroscopy. In both in CDCl3 at – 60 °C and toluene-d8 at -70 °C, only signals for the

tBu 15 syn and anti isomers of ArCH2S NO are observed. These findings suggest that the

tBu putative nitroxyl disulfide intermediate IPrCu(RS(NO)SR) (R = CH2Ar ) (Schemes 5.3 and 5.12) is higher in energy than both the corresponding IPrCuSR and RSNO species in this ligand system using our conditions.

237 Coalesced CH2 Peak tBu tBu tBu ArCH2SSCH2Ar ArCH2SNO

50 °C

40 °C

30 °C

20 °C

10 °C

0 °C

-10 °C -20 °C

-30 °C

-40 °C

tBu ArCH2SNO, anti tBu -50 °C ArCH2SNO, syn tBu IPrCuSCH2Ar

1 tBu Figure 5.13. H NMR spectra (500 MHz, CDCl3) of IPrCuSCH2Ar with 1 equiv. tBu CH2ArSNO at temperatures ranging from -50 °C to 50 °C.

238 tBu tBu Coalesced CH2 Peak ArCH2SSCH2Ar toluene tBu tBu IPrCuSCH2Ar + ArCH2SNO 50 °C 40 °C 30 °C 20 °C 10 °C 0 °C -10 °C -20 °C -30 °C

-40 °C -50 °C -60 °C tBu tBu ArCH2SNO, syn -70 °C ArCH2SNO, anti

tBu Ligand CHMe2 IPrCuSCH2Ar tBu IPrCuSCH2Ar 1 tBu Figure 5.14. H NMR spectra (500 MHz, C6D5CD3) of IPrCuSCH2Ar with 1 equiv. tBu CH2ArSNO at temperatures ranging from -70 °C to 50 °C.

239 + 5.6. Reactivity of copper-thiolates IPrCu(SR) with NOgas and NO

Solutions of copper(I) thiolates IPrCu(SR) (R = Bn (1) or tBu (2)) have no

1 reactivity with NOgas since they exhibit unchanged H NMR spectra after 10 - 60 min when 1 - 10 equiv. anaerobic NOgas are added to CDCl3 solutions of these species.

Interesting reactivity occurs, however, when a source of NO+ is added to IPrCu(SR).

Addition of NOBF4 to a stirring solution of IPrCu(SBn) (1) in CDCl3 results in an immediate color change from light tan to red. This color change suggests the formation of

t 1 BuSNO which is red and is confirmed by H NMR spectroscopy by its SCH2Ph signal at

t δ 4.69 ppm. Similarly, addition of NOBF4 to IPrCu(S Bu) (2) in CDCl3 results in a red solution owing to the presence of tBuSNO (δ 1.90 ppm) confirmed by 1H NMR spectroscopy along with a new StBu resonance at δ 0.55 ppm. (Scheme 5.13).

Crystallization of each reaction mixture reveals the novel cationic dinuclear

t thiolates [{IPrCu}2(µ-SR)]BF4 (R = Bn (6) or Bu (7) (Figures 5.16 and 5.17). Despite the bridging interaction, the Cu-S distances in 6 (2.146(2) and 2.157(2) Å) and 7

(2.182(2) and 2.169(2) Å) are not markedly different than found in the thiolates 1

t Scheme 5.13. Reaction of IPrCu(SR) (R = Bn (1) or Bu (2)) with NOBF4 gives t [{IPrCu}2(µ-SR)]BF4 (R = Bn (6) or Bu (7)).

240 (2.127(1) Å) and 3 (2.130(1) Å). The copper centers are separated by 3.575(1) and

3.731(1) Å with Cu-S-Cu angles of 112.34(7) and 119.48(13)°. 1H NMR spectra of these species in CDCl3 reveal upfield shifted thiolate resonances at δ 2.93 ppm (SCH2Ph) and δ

0.58 ppm (StBu), likely due to the ring current of the NHC ligands.

Based on these observations, it appears that reaction of NO+ with 1 and 2 initially generates an equivalent of S-nitrosothiol RSNO within the coordination sphere of copper which can be formally displaced by another equivalent of IPrCu(SR) to form dinuclear 6 and 5. Consistent with this suggestion, reaction of {IPrCu(NCMe)}BF4 35 with IPrCuSR in results in displacement of NCMe and formation of 4 and 5 in isolated yields of 86 and

89% (Scheme 5.14).

Scheme 5.14. Independent synthesis of {IPrCu}2(µ-SR)]BF4 (6 and 7) from t {IPrCu(MeCN}BF4 and IPrCu(SR) (R = Bn (1) or Bu (2)).

When NOBF4 is added to IPrCu(SCPh3) (4) in CDCl3 the solution turns bright,

+ fluorescent yellow reminiscent of the color of the trityl cation Ph3C . Analysis of the reaction mixture by 1H NMR, however, does not unambiguously reveal the formation of

Ph3CSNO which is somewhat hampered by the complexity of the aromatic region in the reaction mixture.

241 Scheme 5.15. Reaction of IPrCu(SCPh3) (4) with NOBF4 gives [{IPrCu}2(µ-SH)]BF4 (8). Crystallization of the reaction mixture reveals the dicopper hydrosulfide

[{IPrCu}2(µ-SH)]BF4 (8) in 68 % yield (Scheme 5.15, Figure 5.18). Despite the bridging interaction, the Cu-S distances in 8 (2.127(2) and 2.138(2) are not markedly different than found in the thiolate 4 (2.1304(10) Å. The copper centers are separated by 3.537(1)

Å with a Cu-S-Cu angle of 112.08(8)°.

242 C55

N1 S N3 C1 Cu1 Cu2 N2 C28

N4

Figure 5.16. X-ray structure of [{IPrCu}2(µ-SBn)]BF4 (6) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–C1 1.882(6), Cu1–S 2.1570(16), Cu2- C28 1.874(6), Cu2-S 2.1462(16), Cu1…Cu2 3.575(1), S-C55 1.858(5), C55–S–Cu1 103.70(18), C55-S-Cu2 105.89(19), C1-Cu1-S 162.41(18), C28-Cu2-S 176.01(17), N2– C1–Cu1 131.8(4), N1–C1–Cu 123.5(4), N3–C28–Cu2 127.5(4), N4–C28–Cu2 128.2(4), Cu1-S-Cu2 112.34(7).

243 C55A

Cu2 Cu1 SA N1 C1 C28 N3

N2 N4

t Figure 5.17. X-ray structure of [{IPrCu}2(µ-S Bu)]BF4 (7) (all H atoms omitted). Two similar conformations of the StBu group are present in a 53:47 ratio. Only the major occupancy is shown for clarity. Selected bond distances (Å) and angles (deg): Cu–C1 1.892(4), Cu1–SA 2.144(3), Cu2-C28 1.892(4), Cu2-SA 2.176(3), Cu1…Cu2 3.731(1), SA–C55A 1.831(13), C1–Cu1–SA 171.99(15), Cu1-SA-Cu2 119.48 (13), C28-Cu2-SA 173.68(14), C55A–SA–Cu1 110.7(4), C55A-SA-Cu2 110.0(4), N2–C1–Cu1 130.6(3), N1–C1–Cu1 25.5(3), N3–C28–Cu2 125.6(3), N3–C28–Cu2 130.9(3).

244 H61

S

N1 Cu1 Cu2 C1 N3 C28 N2 N4

Figure 5.18. X-ray structure of [{IPrCu}2(µ-SH)]BF4 (8) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu–C1 1.837(7), Cu1–S 2.138(2), Cu2-C28 1.860(6), Cu2-S 2.1274(19), Cu1…Cu2 3.537(1), C1-Cu1-S 166.36(19), C28-Cu2-S 174.5(2), N2–C1–Cu1 137.2(5), N1–C1–Cu 121.5(4), N3–C28–Cu2 127.0(5), N4–C28– Cu2 129.6(5), Cu1-S-Cu2 112.08(8).

245 5.7. Reactivity of the one coordinate copper cation {IPrCu}+ with RSNOs.

1 H NMR analysis of the reaction of [IPrCu(MeCN)]BF4 with 1 equiv. RSNO (R =

t Bn, Bu) in CDCl3 at room temperature indicates that the RSNO does not displace the

MeCN bound to copper (Scheme 5.18a). Nontheless, rapid decomposition of the S- nitrosothiol does not ensue, suggesting that this electron-poor copper fragment does not reductively cleave the RS-NO bond.

We sought the synthesis of a synthon to the {IPrCu}+ cation which possesses a

F F ligand more labile group than MeCN. Reaction of IPrCuCl with NaBAr 4 (Ar = 3,5-

F bis(trifluoromethyl)benzene) in ether leads to the formation of [IPrCu(Et2O)]BAr 4 (9) in

80 % yield (Scheme 5.16).

F Scheme 5.16. Synthesis of {IPrCu(Et2O)}BAr 4 (9).

F The [IPrCu(Et2O)]BAr 4 (9) species characterized by X-ray features (Figure 5.19) a Cu-C1 distance of 1.868(4) similar to that of the IPrCuSR and [{IPrCu}2(-

SR)]BF4.complexes. The CH3CH2-O-CH2CH3 unit is rather “flat” with a torsion angle

1 of 6.73°. Its H NMR spectrum in CDCl3 features an upfield shifted CH3 triplet signal 246 for bound Et2O (δ 0.824 ppm) versus the CH3 triplet signal in free Et2O (δ 1.21 ppm).

The CH2 quartet in bound Et2O (δ 3.439 ppm) versus the CH2 quartet signal in free Et2O

(δ 3.48 ppm) does not shift significantly.

t F Reaction of BuSNO with [IPrCu(Et2O)]BAr 4 (9) causes a downfield shift of the bound ether OCH2CH3 triplet from δ 0.824 to 1.045 ppm. For comparison, the chemical

t shift of this group for this resonance of free ether in CDCl3 is δ 1.21 ppm. The BuSNO peak does not shift, however, from its original position at δ 1.88 ppm. This suggests that the downfield shift All attempts at crystallizing 9 in the presence of tBuSNO did not result in the desired cation {IPrCu(tBuSNO)}+. Rather, the ether-bound 9 was always the species that crystallized out of the reaction mixture.

Removal of all the ether from the reaction mixture resulted in the appearance of a

t new S Bu singlet at δ 1.05 ppm when redissolved in CDCl3. We attribute this to the formation of the disulfide adduct {IPrCu(tBuSStBu)}+. This was confirmed by an

F t t independent synthesis in CDCl3 employing [IPrCu(Et2O)]BAr 4 with BuSS Bu. The free

tBuSStBu resonance at δ 1.27 ppm is replaced with a peak at δ 1.05 ppm is present

(Scheme 5.17).

F Similarly, the addition of BnSNO to [IPrCu(Et2O)]BAr 4 (9) slightly shifts the bound ether OCH2CH3 triplet downfield to δ 0.993 ppm. There is no evidence of BnSSBn or BnSNO in the reaction mixture. When the ether is removed from the reaction mixture in vacuo the new SCH2Ph peak remains and is a singlet at δ 3.44 ppm. For reference, the

SCH2Ph peak of BnSSBn in CDCl3 appears at δ 3.60. This new peak is attributed to

BnSSBn weakly binding to the [IPrCu]+ cation in solution. 247 Scheme 5.17. Reaction of RSNOs and RSSRs with 9.

248 N1 O C1 Cu B N2

F Figure 5.19. X-ray structure of [IPrCu(Et2O)]BAr 4 (9) (all H atoms omitted). Selected bond distances (Å) and angles (deg): Cu-C1 1.868(4), Cu-O 1.889(3), C1-Cu-O 178.75(15), N1-C1-Cu 127.3(3), N2-C1-Cu 127.9(3).

249 Table 5.2. Crystallographic parameters for 6, 7, 8, and 9.

Compd. 6 7 8 9

Formula C61H79BCu2F4 C58H81Cu2N4S, C54H72Cu2N4S, C63H58BCl0Cu

N4S 2(C4H8O), BF4 (C6H5Cl), BF4 F24N2O Mol. Wt. 1114.23 1224.43 1135.66 1389.46 Temp.(K) 100(2) 100(2) 173(2) 100(2) Crystal Plate Block Block Block description Crystal color Colorless Colorless Colorless Colorless Crystal size 0.0016 0.0684 0.01893 0.0218 (mm3) System Monoclinic Triclinic Monoclinic Triclinic Space group C2/c P-1 C2/c P-1 a (Å) 32.857(3) 12.348(3) 31.583(4) 13.6976(18) b (Å) 12.3716(12) 17.138(4) 10.4802(12) 14.6818(19) c (Å) 31.680(3) 17.187(4) 36.893(4) 16.793(2) α (deg) 90 75.961(3) 90 110.815(2) β (deg) 95.4190(10) 75.033(3) 99.072(2) 91.660(2) γ (deg) 90 72.937(3) 90 97.281(2) Volume (Å3) 12820(2) 3303.9(14) 12059(2) 3121.2(7) Z 8 2 8 2 θ range (deg) 1.25-25.00 1.25-25.00 1.12-24.25 1.30-25.00 Measd reflns 45684 32174 40691 30754 Unique reflns 11276 11602 9743 10985 GOF of F2 0.898 1.078 0.938 0.824

R1 ( I > 2σ(I)) 0.0684 0.0570 0.0785 0.0526 wR2 (all data) 0.1697 0.1578 0.2231 0.1448 Largest diff. 0.866 and 1.722 and 0.974 and - 0.675 and peak and hole -0.443 -1.046 1.381 -0.520 (e-.Å-3) 250 Summary

Despite the important biological role of copper ions in catalyzing the loss of

NOgas from S-nitrosothiols, this study employing two coordinate copper(I) thiolates

IPrCu(SR) clearly implicates a role for the metal’s coordination environment and oxidation state in this reaction Instead of catalyzing NOgas loss from RSNOs, clean trans-s-nitrosation equilibria are observed. The rate of thiolate exchange is significantly faster in non-polar solvents, orders of magnitude faster than simple RSNO / RS- trans-s- nitrosation in water. Conceptually related, the addition of NOBF4 to IPrCu(SR) leads to

RSNO loss and generation of novel two coordinate dicopper species [{IPrCu}2(µ-

+ SR)]BF4. RSNOs do not bind strongly to the one coordinate {IPrCu} fragment, being readily displaced by other donors.

Nontheless, rapid decomposition of the RSNO does not ensue. This is in contrast to what we observe in the β-diketiminate (Chapter 3, Scheme 5.18.b) and tris(pyrazolyl)borate (Chapter 4, Scheme 5.18c) ligand systems. The copper(I) ion in the

N-heterocyclic carbene ligand system apparently is not strong enough of a 1-electron reductant to readily activate the RS-NO to give NOgas and give the corresponding

Scheme 5.18. Comparison of RSNO interactions with three different types of copper(I) coordination environments. 251 {IPrCu(SR)}+ species.

While these two coordinate copper(I) thiolates represent a somewhat unusual coordination environment, they are not completely without precedence in biology.

+ Dinuclear purple CuA sites possess [{HisCu}2(µ-SR)] cores (Figure 5.2b) which may be conceptually disassembled into HisCu-SR and a {HisCu-SR}+ fragments linked through bridging thiolates to produce a valence delocalized Cu1.5Cu1.5 species capable of rapid electron transfer. Thus, oxidation of the two coordinate [NHC]CuSR species may lead to additional opportunities for the study of elusive targets for bioinorganic modeling.

252 Experimental procedures

General experimental details

All experiments were carried out in a dry nitrogen atmosphere using an MBraun

glovebox and/or standard Schlenk techniques. 4A molecular sieves were activated in

vacuo at 180 ºC for 24 h. Dry toluene and dichloromethane were purchased from Aldrich

and were stored over activated 4A molecular sieves under nitrogen. Diethyl ether and

tetrahydrofuran (THF) were first sparged with nitrogen and then dried by passage through

1 activated alumina columns. Pentane was first washed with conc. HNO3/H2SO4 to remove olefins, stored over CaCl2 and then distilled before use from sodium/benzophenone. All deuterated solvents were sparged with nitrogen, dried over activated 4A molecular sieves and stored under nitrogen. 1H and 13C NMR spectra were recorded on an Inova Varian 300 MHz, 400, or 500 MHz spectrometer (300, 400, or 500 and 75.4, 100.465, or 125.8 MHz, respectively). 19F NMR spectra were recorded at

282.344 MHz or 375.845using an internal or external reference of C6F6 set to δ = -164.9 ppm. All NMR spectra were recorded at room temperature unless otherwise noted and were indirectly referenced to TMS using residual solvent signals as internal standards.

Elemental analyses were performed on a Perkin-Elmer PE2400 microanalyzer in our laboratories.

t BuSH and BnSH were obtained from Acros, anhydrous CuCl and NOBF4 from

Strem, and TlOEt, KOtBu as well as (4-tert-butylphenyl)methanethiol from Aldrich; all were used as received. The thallium thiolates TlStBu, TlSBn, and TlS(4-tBu)Bn were synthesized following a modified literature procedure by the reaction of free thiol with 253 TlOEt in ether followed by washing of the solids with pentane.3 The S-nitosothiols

tBuSNO, BnSNO, and p-tBuBnSNO were synthesized in situ by the reaction of the corresponding thallium thiolates with 1 eq. NOBF4 in CDCl3 or C6D6. Use of an internal standard indicates > 95% purity of resulting RSNOs.40 Caution! tBuSH is extremely volatile and possesses a pungent odor indistinguishable from ethanethiol contained in natural gas. Glyoxal-bis-(2,6-diisopropylphenyl)imine32, 1,3-bis-

32 34 35 (diisopropylphenyl)imidazolium chloride, IPrCuCl, [IPrCu(NCMe)]BF4, and

F 41 NaBAr 4 were synthesized according to published literature procedures.

IPrCu(SBn) (1). A yellow slurry of TlSBn (0.135 g, 0.411 mmol) in 3 mL of THF was added to a stirring solution of IPrCuCl (0.200 g, 0.411 mmol) in 5 mL of THF upon which TlCl immediately precipitated. The solution was allowed to stir for 1 h. The white precipitate was filtered out and the solution was concentrated to 2 mL. Crystals suitable for X-ray analysis grew overnight at -30 ºC yielding 0.161 mg of product (68%). 1H

NMR (CDCl3, 25 °C) δ 7.486 (t, 2, p-ArH- NHC), 7.300 (d, 4, m-ArH-NHC), 7.112 (s, 2,

CH-NHC), 7.067-6.900 (m, 5, CH2Ph), 3.275 (s, 2, CH2Ph), 2.600 (sept, 4, CHMe2),

13 1 1.300 (d, 12, CHMe2), 1.226 (d, 12, CHMe2); C{ H} NMR (CDCl3, 25 °C) δ 181.72,

145.63, 144.83, 134.55, 130.32, 127.93, 127.74, 124.76, 124.03, 122.75, 28.68, 25.26,

24.77, 23.80. Anal. Calcd. for C34H43CuN2S: C, 70.98; H, 7.53; N, 4.87. Found C, 71.22;

H, 7.56; N, 4.87.

IPrCu(StBu) (2). A yellow solution of TlStBu (0.151 g, 0.514 mmol) in 3 mL of THF was added to a stirring solution of IPrCuCl (0.250 g, 0.514 mmol) in 5 mL of THF upon

254 which TlCl immediately precipitated. The solution was allowed to stir for 1 h. The

volatiles were removed in vacuo and the white residue was extracted with 3 x 5 mL of

CH2Cl2 and filtered through Celite. The solvent was removed in vacuo and the residue washed with 5 mL of n-pentane. The remaining solid was dried in vacuo to afford 0.206 g

1 (74%) of the product. H NMR (CDCl3, 25 °C) δ 7.451 (t, 2, p-ArH- NHC), 7.277 (d, 4, m-ArH-NHC), 7.123 (s, 2, CH-NHC), 2.615 (sept, 4, CHMe2), 1.328 (d, 12, CHMe2),

t 13 1 1.221 (d, 12, CHMe2), 1.040 (s, 9, Bu); C{ H} NMR (CDCl3, 25 °C) δ 182.73, 145.79,

134.88, 130.36, 124.20, 122.85, 39.27, 34.41, 28.88, 24.88, 24.22. Anal. Calcd. for

C31H45CuN2S: C, 68.78; H, 8.38; N, 5.18. Found C, 69.04; H, 8.37; N, 4.77.

tBu tBu IPrCu(SCH2Ar ) (3). A yellow solution of TlS(CH2Ar ) (0.330 g, 0.860 mmol) in 3 mL of THF was added to a stirring solution of IPrCuCl (0.419 g, 0.860 mmol) in 8 mL of

THF upon which TlCl immediately precipitated. The solution was allowed to stir for 1 h.

The white precipitate was filtered out and the solution was concentrated down to 2 mL.

Crystals suitable for X-ray analysis grew overnight at -30 ºC yielding 0.420 mg of

1 product (77%). H NMR (CDCl3, 25 °C) δ 7.490 (t, 2, p-ArH- NHC), 7.305 (d, 4, m-

tBu ArH-NHC), 7.118 (s, 2, CH-NHC), 7.069 (d, 2, o-ArH-CH2Ar ), 6.833 (d, 2, m-ArH-

tBu tBu CH2Ar ) 3.322 (s, 2, CH2Ar ), 2.615 (sept, 4, CHMe2), 1.328 (d, 12, CHMe2), 1.304

tBu t 13 1 (s, 9, CH2Ar ), 1.221 (d, 12, CHMe2), 1.040 (s, 9, Bu); C{ H} NMR (CDCl3, 25 °C) δ

182.24, 145.95, 143.82, 134.86, 130.61, 127.83, 127.82, 125.01, 124.33, 122.98, 34.42,

31.65, 28.99, 25.83, 25.05, 24.10. Anal. Calcd. for C38H51CuN2S: C, 72.28; H, 8.14; N,

4.44. Found C, 71.99; H, 7.80; N, 4.15.

255 IPrCu(SCPh3) (4). A tan solution of TlSCPh3 (0.246 g, 0.514 mmol) in 3 mL of THF was added to a stirring solution of IPrCuCl (0.250 g, 0.514 mmol) in 5 mL of THF upon which TlCl immediately precipitated. The solution was allowed to stir for 1 h. The white precipitate was filtered out and the solution was concentrated down to 2 mL and layered with 1 mL of Et2O. Crystals suitable for X-ray analysis grew overnight at -30 ºC yielding

1 0.350 mg of product in 88 % yield. H NMR (CDCl3, 25 °C) δ 7.514 (t, 2, p-ArH- NHC),

7.274 (d, 4, m-ArH-NHC), 7.152-7.120 (m, 6, m-ArH-CPh3), 7.024 (s, 2, CH-NHC),

6.923-6.886 (m, 9, o/p-ArH-CPh3), 2.504 (sept, 4, CHMe2), 1.169 (d, 12, CHMe2), 1.136

13 1 (d, 12, CHMe2); C{ H} NMR (CDCl3, 25 °C) δ 182.07, 145.75, 145.05, 134.56, 130.74,

129.62, 126.27, 125.15, 124.38, 123.36, 29.39, 28.91, 24.95, 24.07. Anal. Calcd. for

C38H51CuN2S: C, 72.28; H, 8.14; N, 4.44. Found C, 72.50; H, 8.01, N, 4.42.

IPrCu(SMe) (5). MeSSMe (0.500 g, 5.319 mmol) was stirred in 10 mL of THF overnight with Na(s) (0.250 g, 10.800). A tan precipitate was formed and attributed to be

NaSMe (0.700 g, 93 % yield). A tan suspension of NaSMe (0.036 g, 0.514 mmol) in 3 mL of THF was added to a stirring solution of IPrCuCl (0.250 g, 0.514 mmol) in 5 mL of

THF. The solution was allowed to stir for 1 h. The THF was removed in vacuo and the residue was brought back up in 10 mL of CH2Cl2, filtered over Celite. The CH2Cl2 was removed in vacuo and the solid was washed with 2 × 3 mL of Et2O. The solid was

1 collected on a frit to yield 0.240 mg of product in 94 % yield. H NMR (CDCl3, 25 °C)

δ 7.451(t, 2, p-ArH- NHC), 7.259 (d, 4, m-ArH-NHC), 7.096 (s, 2, CH-NHC), 2.333

(sept, 4, CHMe2), 1.528 (s, 3, SMe), 1.297 (d, 12, CHMe2), 1.209 (d, 12, CHMe2);

256 13 1 C{ H} NMR (CDCl3, 25 °C) δ 182.93, 145.69, 134.48, 130.35, 124.07, 122.72, 28.75,

24.78, 23.85, 6.61. Anal. Calcd. for C28H39CuN2S: C, 67.36; H, 7.87; N, 5.61. Found C,

67.61; H, 7.68; N, 5.44.

1H NMR Reactions of IPrCu(SR) with TlSR'

Scheme 5.18. Reaction of IPrCuSR with TlSR' to form IPrCu(RS(Tl)SR') complex in solution. IPrCu(SBn(Tl)StBu). A yellow solution of TlStBu (0.015 g, 0.052 mmol) in 0.5 mL of

1 CDCl3 was added to IPrCu(SBn) (1) (0.030 g, 0.052 mmol) in 1 mL CDCl3. A H NMR analysis of the sample and showed sharp peaks suggesting that the TlStBu forms an

1 adduct with IPrCu(SBn) (1) H NMR (CDCl3, 25 °C) δ 7.471 (t, 2, p-ArH- NHC), 7.316-

7.235 (m, 9, m-ArH-NHC and CH2Ph), 7.137 (s, 2, CH-NHC), 4.198 (s, 2, CH2Ph),

t 2.636 (sept, 4, CHMe2), 1.346 (d, 12, CHMe2), 1.242 (d, 12, CHMe2), 1.187 (s, 9, Bu).

IPrCu(SBn(Tl)SBn). A yellow slurry of TlSBn (0.017 g, 0.052 mmol) in 0.5 mL of

1 CDCl3 was added to IPrCu(SBn) (1) (0.030 g, 0.052 mmol) in 1 mL CDCl3. H NMR was analysis of the sample and showed one sharp CH2Ph peak suggesting that the TlSBn

1 forms an adduct with IPrCu(SBn) (1). H NMR (CDCl3, 25 °C) δ 7.513 (t, 2, p-ArH-

257 NHC), 7.328 (d, 4, m-ArH-NHC) 7.149-7.054 (m, 7, CH-NHC and CH2Ph), 3.769 (s, 4,

CH2Ph), 2.607 (sept, 4, CHMe2), 1.305 (d, 12, CHMe2), 1.230 (d, 12, CHMe2).

IPrCu(StBu(Tl)StBu). A yellow solution of TlStBu (0.016 g, 0.056 mmol) in 0.5 mL of

t 1 CDCl3 was added to IPrCu(S Bu) (2) (0.030 g, 0.055 mmol) in 1 mL CDCl3. H NMR analysis of the sample and showed sharp peaks suggesting that the TlStBu forms a adduct

t 1 with IPrCu(S Bu) (2) H NMR (CDCl3, 25 °C) δ 7.462 (t, 2, p-ArH- NHC), 7.287 (d, 4, m-ArH-NHC), 7.134 (s, 2, CH-NHC), 2.631 (sept, 4, CHMe2), 1.343 (d, 12, CHMe2),

t 1.245 (s, 18, Bu), 1.234 (d, 12, CHMe2).

tBu tBu tBu IPrCu(SCH2Ar (Tl)SCH2Ar ). A yellow solution of TlSCH2Ar (0.019 g, 0.049

tBu mmol) in 0.5 mL of CDCl3 was added to IPrCu(SCH2Ar ) (3) (0.030 g, 0.049 mmol) in

1 1 mL CDCl3. H NMR analysis of the sample and showed sharp peaks suggesting that the

t tBu 1 TlS Bu forms an adduct with IPrCu(SCH2Ar ) (3) H NMR (CDCl3, 25 °C) δ 7.494 (t,

tBu 2, p-ArH- NHC), 7.321-7.003 (d, 10, m-ArH-NHC, CH-NHC, and ArH-CH2Ar ), 3.811

tBu t (s, 4, CH2-Ar ), 2.614 (sept, 4, CHMe2), 1.309 (d, 12, CHMe2), 1.261 (s, 18, Bu), 1.211

(d, 12, CHMe2).

All attempts at crystallizing adducts which form in the reaction between IPrCu(SR) and

TlSR resulted in crystals of IPrCuSR and not the thallium thiolate adduct. This suggests that the lowest energy state for the reaction is the copper(I) thiolate and not the adduct with thallium thiolate.

1H NMR experiments for trans-s-nitrosation reactions of IPrCu(SR) with R'SNO

IPrCu(StBu) and BnSNO. BnSNO was generated from the addition of TlSBn (0.018 g,

0.055 mmol) in 0.500 mL CDCl3 to stirring NOBF4 (0.007 g, 0.055 mmol) crystals. The 258 colorless solution immediately turns bright red and the solution is allowed to stir for 5

min. This solution is added to a light tan solution of IPrCu(StBu) (0.030 g, 0.055 mmol) in 0.500 mL CDCl3. The color of the red solution lessens and a hint of green appears.

This sample is analyzed by 1H NMR and IPrCu(SBn) and tBuSNO are both seen as new products (δ 3.278 and 1.939 ppm) with IPrCu(StBu) (δ 1.065 ppm) and BnSNO (δ 4.670 ppm) reactants still remaining. We repeated this experiment three times and calculated the equilibrium constant of the reaction.

The experiment was similarly repeated in C6D6. The SCH2Ph peaks from IPrCu(SCH2Ph) and BnSNO form one new peak at δ 3.978 ppm, similar to what is seen in the degenerate

t reaction between IPrCu(SBn) and BnSNO in C6D6. The two types of S Bu peaks from

IPrCu(StBu) and tBuSNO overlap and equilibrium information cannot be taken from these spectra since the pure components appear at δ 1.480 and 1.506 ppm, respectively.

Transthiolation experiment between IPrCu(StBu) and BnSH. BnSH (0.007 g, 0.055

t mmol) in 0.100 mL CDCl3 was added to a light tan solution of IPrCu(S Bu) (0.030 g,

0.055 mmol) in 0.500 mL CDCl3. No color change is observed. This sample is analyzed by 1H NMR and IPrCu(SBn) (δ 3.278 ppm) and tBuSH (δ 1.469 and 1.786 ppm for tBu and SH resonances, respectively) are both seen as new products with complete consumption of IPrCu(StBu) and BnSNO reactants. Thus the reaction favors the primary copper thiolate and the tertiary thiol.

Degenerate trans-s-nitrosation experiments

IPrCu(StBu) with tBuSNO. tBuSNO was generated from the addition of TlStBu (0.016 g, 0.055 mmol) in 0.500 mL CDCl3 to stirring NOBF4 (0.007 g, 0.055 mmol) crystals. 259 The colorless solution turns green and the solution is allowed to stir for 5 min. This

solution is added to a light tan solution of IPrCu(StBu) (0.030 g, 0.055 mmol) in 0.500

1 mL CDCl3. No major color change is observed. This sample is analyzed by H NMR.

IPrCu(StBu) (δ 1.065 ppm) and tBuSNO (δ 1.939 ppm) are both seen as sharp unshifted peaks. This suggests that the trans-s-nitrosation reaction does not take place on the 1H

NMR timescale. The experiment was repeated in C6D6 with identical results to the CDCl3 experiment, . IPrCu(StBu) (δ 1.517 ppm; tBu) and tBuSNO (δ 1.485 ppm) appear at different chemical shifts in C6D6.

IPrCu(SBn) with BnSNO. BnSNO was generated from the addition of TlSBn (0.018 g,

0.055 mmol) in 0.500 mL CDCl3 to stirring NOBF4 (0.007 g, 0.055 mmol) crystals. The colorless solution immediately turns bright red and the solution is allowed to stir for 5 min. This solution is added to a light tan solution of IPrCu(SBn) (0.032 g, 0.055 mmol) in

1 0.500 mL CDCl3. No major color changes are observed. This sample is analyzed by H

NMR and the SCH2Ph peaks in IPrCu(SBn) (δ 3.814 ppm) and BnSNO (δ 4.670 ppm) are both broad, but remain at the same chemical shift.

Table 5.3. Experimental set up for reaction of IPrCuSBn (1) and BnSNO using varying equivalents of each reactant.

Spectrum RSNO:CuSR PhCH2SNO IPrCu(SCH2Ph) C6D6 δ (ppm) Ratio 200 mM 100 mM 1 0:1 0.200 mL 0 mL 0.800 mL 4.082 2 1:3 0.600 mL 0.400 mL 0 mL 4.027 3 1:2 0.400 mL 0.400 mL 0.200 mL 4.021 4 1:1 0.200 mL 0.400 mL 0.400 mL 3.979 5 1:1/2 0.100 mL 0.400 mL 0.500 mL 3.867 6 0:1 0 mL 0.400 mL 0.600 mL 3.816

260 1 Performing the reaction analogously in benzene-d6 revealed one H NMR peak at 3.978 ppm for the SCH2Ph protons. When different stochiometric amounts (1/2, 1, 2, and 3 eq.) of BnSNO are added to IPrCu(SBn), the chemical shift of the broad peak moves toward the compound in greater excess. A 200 mM stock solution of PhCH2SNO was generated from the addition of TlSBn (0.131 g, 0.400 mmol) in 0.200 mL CDCl3 to stirring NOBF4

(0.047 g, 0.400 mmol) crystals. A 100 mM stock solution of IPrCu(SBn) was prepared

1 by dissolving (0.115 g, 0.200 mmol) in 0.200 mL of CDCl3. H NMR spectra were taken in C6D6 varying the ratio PhCH2SNO to IPrCu(SBn): (1) 0:1, (2) 1:3, (3) 1:2, (4) 1:1, (5)

1:1/2, (6) 1:0. The final concentration of copper was kept constant throughout the reaction.

Assuming coalescence and based on the frequency difference between the SCH2Ph resonances in IPrCu(SBn) and BnSNO, this implies an actual rate of exchange in excess of 180 s-1.

Variable temperature 1H NMR spectroscopy for trans-s-nitrosation reactions of

tBu tBu IPrCu(SCH2Ar ) with 1 equiv. ArCH2SNO.

We attempted a variable temperature 1H NMR experiment using IPrCu-SBn and BnSNO in both the polar (CDCl3) and nonpolar (toluene-d8) NMR solvents. At low temperature

IPrCu(SBn) has very low solubility in the nonpolar solvent. This led us to the synthesis

tBu and use IPrCu(SCH2Ar ) (3), an electronically similar compound with higher solubility in the nonpolar solvent.

1 tBu Variable temperature H NMR spectra were taken on each component ArCH2SNO and

tBu IPrCu-SCH2Ar individually in CDCl3 and toluene-d8. 261 tBuArCH2SNO. S-nitrosothiols are known to exhibit syn / anti isomerization. The two isomers are in fast equilibrium at room temperature and only on peak can been seen in the corresponding 1H NMR spectra. At lower temperatures the isomerization between the syn

and anti conformations is locked out on the NMR timescale and the two isomers can be

1 clearly seen by H NMR in both CDCl3 and toluene-d8 (Figure 5.11.a and Figure 5.11.b)

Table 5.4. 1H NMR chemical shifts (500 MHz) for syn / anti rotamers of the methylene tBu peak of ArCH2SNO.

tBu tBu Spectrum ArCH2SNO ArCH2SNO -60 °C, CDCl3 -70 °C, C6D5CD3 Syn 4.402 3.695 Anti 6.207 5.488 Average (25 °C) 4.656 4.158

tBu IPrCu(SCH2Ar ) (3). The solubility of 3 is very low in toluene-d8 -20 °C resulting in the loss of signals due to 3 below this temperature (Figure 5.20). In contrast, the 1H NMR signals in this compound remain sharp and unshifted over the temperature range -60 to 50

°C in CDCl3 (Figure 5.21).

262 Ligand CHMe2 tBu tBu IPrCuSCH2Ar IPrCuSCH2Ar toluene 50 °C 40 °C 30 °C 20 °C 10 °C 0 °C -10 °C -20 °C -30 °C

-40 °C -50 °C -60 °C

-70 °C THF

1 Figure 5.20. Variable temperature H NMR spectra (500 MHz, toluene-d8) of tBu IPrCuSCH2Ar (3) from -70 °C to 50 °C. Compound 3 exhibits low solubility in the nonpolar solvent its1H NMR signals do not appear below -20 °C.

263 Ligand CHMe2 tBu tBu IPrCuSCH2Ar IPrCuSCH2Ar 50 °C

40 °C

30 °C

20 °C

10 °C

0 °C

-10 °C -20 °C

-30 °C

-40 °C

-50 °C THF -60 °C THF

1 Figure 5.21. Variable temperature H NMR spectra (500 MHz, CDCl3) of tBu IPrCuSCH2Ar (3) from -60 °C to 50 °C.

264 tBu tBu 1 Trans-s-nitrosation between IPrCu-SCH2Ar and ArCH2SNO followed by H

tBu tBu NMR. ArCH2SNO was generated from the addition of TlSCH2Ar (0.021 g, 0.055 mmol) in 0.500 mL CDCl3 to stirring NOBF4 (0.007 g, 0.055 mmol) crystals. The colorless solution immediately turns bright red and the solution is allowed to stir for 5

tBu min. This solution is added to a light tan solution of IPrCu(SCH2Ar ) (0.035 g, 0.055 mmol) in 0.500 mL CDCl3. No major color changes are observed. This sample was analyzed by 1H NMR from -70 °C to 25 °C.

tBu tBu 15 15 Trans-s-nitrosation between IPrCu(SCH2Ar ) and ArCH2S NO followed by N

NMR.

tBu 15 t 15 ArCH2S NO. BuO NO was first synthesized according to the reported procedure from an acidic solution of sodum nitrite with slight modifications.42 The tBuO15NO served as a source of 15NO+ for an RSH to form RS15NO.21 To an ice cold solution of 18

t M H2SO4 (1.04 ml, 19.55 mmol) and H2O (1.04 mL) was added BuOH (1.447 g, 19.55

15 mmol) dropwise followed by the addition of a solution of Na NO2 (1.500 g, 21.4 mmol) in 2 mL of H2O. The reaction was kept at 0 °C during the addition. The reaction mixture

was warmed to room temperature and allowed to stir for 2 hr. The aqueous layer was

separated and the organic portion was washed with aq NaHCO3 (0.500 g) and NaCl

(0.200 g). . The organic layer was dried over Na2SO4 and filtered through silica gel to remove any remaining alcohol to yield 1.580 g (78 % yield) of a yellow oil. 1H NMR

15 15 (toluene-d8, 500 MHz, 25 °C): 1.263 (s, 9, Me3CO NO). N NMR (toluene-d8, 500

15 MHz, 25 °C): 0 (s, 1, Me3CO NO).

265 tBu To 600 L of toluene-d8 was added ArCH2SH (112 L, 0.060 mmol), followed by

t 15 tBu 15 1 BuO NO (7 L, 0.060 mmol) to give concentrations of 100 mM of ArCH2S NO. H

NMR (toluene-d8, 500 MHz, 25 °C): 7.072 (d, 2, o-ArH), 6.778 (d, 2, m-ArH), 4.260 (s,

tBu 15 2, CH2Ar ), 1.128 (s, 9, Me3CAr). N NMR (toluene-d8, 500 MHz, 25 °C): 0.420 (s, 1,

15 15 t 15 Me3CO NO). N NMR (toluene-d8, 500 MHz, -70 °C): 0.42 (s, 1, BuO NO). -0.15 (s,

tBu 15 tBu 15 syn, ArCH2S NO), 164.54 (s, anti, ArCH2S NO).

tBu To 600 L of chloroform-d1 was added ArCH2SH (112 L, 0.060 mmol), followed by

t 15 tBu 15 1 BuO NO (7 L, 0.060 mmol) to give concentrations of 100 mM of ArCH2S NO. H

NMR (chloroform-d1, 500 MHz, 25 °C): 7.272 (d, 2, o-ArH), 7.073 (d, 2, m-ArH), 4.656

tBu 15 (s, 2, CH2Ar ), 1.288 (s, 9, Me3CAr). N NMR (toluene-d8, 500 MHz, 25 °C): 0.075 (s,

15 1, Me3CO NO). The syn/anti isomerization was not evident at -50 °C in chloroform-d1.

1 15 Table 5.5 H and N NMR data for syn and anti rotometers of PhCH2SNO and tBu t 15 ArCH2SNO in toluene-d8 at -70 °C,(Referenced to BuO NO, toluene-d8 at 196 ppm).

S-nitrosothiol 1H Shift 1H Shift Syn:Anti 15N Shift 15N Shift Anti:Syn Anti Syn 1H NMR Anti Syn 15N NMR PhCH2SNO 5.973 3.560 14:1 209.341 137.927 0.8:1 tBu ArCH2SNO 6.027 4.402 14:1 360.564 195.860 0.8:1

tBu 15 t Trans-s-nitrostation between ArCH2S NO and IPrCu(SCH2Ar Bu) observed by

15 tBu 15 tBu N NMR. ArCH2S NO was generated from the addition of ArCH2SH (112 L,

t 15 0.060 mmol) in 0.300 mL chloroform-d1 or toluene-d8 followed by BuO NO (7 L,

tBu 15 0.060 mmol) to give concentrations of 100 mM of ArCH2S NO. This solution is

tBu added to a light tan solution of IPrCu-SCH2Ar (0.038 g, 0.060 mmol) in 0.300 mL 266 chloroform-d1 or toluene-d8. No major color changes are observed. Theses sample is

15 analyzed by N NMR at -50 and 25 °C (chloroform-d1) or -70 °C to 25 °C (toluene-d8).

The only 15N containing molecule in this reaction mixture is the S-nitrosothiol in both solvents at both temperatures. No new species was formed. This confirms that we do not observe the nitroxyl disulfide intermediate using our ligand system in these reaction conditions.

+ NOgas and NO reactions with IPrCu(SR) complexes

t IPrCu(SR) with NOgas. When 1 equiv. NOgas is added to IPrCu(SR) (R = Bn, Bu, CPh3),

1 there was no reaction by H NMR in CDCl3 at 25 °C. Addition of 2, 5, or 10 equiv. NOgas

1 did not cause a change in the H NMR spectra. Thus NOgas does not react with IPrCu(SR) species under anaerobic conditions.

IPrCu(SBn) with NOBF4. To a stirring solution of IPrCu(SBn) (0.100 g, 0.174 mmol) in

1.5 mL CDCl3 is added NOBF4 (0.020 g, 0.174 mmol). The solution immediately turns from light tan to red. This color change suggests the formation of BnSNO. This formation

1 is confirmed by H NMR spectroscopy (δ 4.684 ppm, ArCH2SNO). Quantification of the formation of tBuSNO was unsuccessful as the amount of tBuSNO was not consistent over several experimental attempts. The IPrCu(SCH2Ar) peak is broad and almost completely disappears into the baseline. A broad peak is also detected from 6.8-6.3 ppm. This peak is attributed to the aromatic peaks on the benzyl ring. Crystallization of this reaction mixture from toluene layered with pentane reveals this peak to be from the formation of a dicopper thiolate cation, [{IPrCu}2(-SBn)]BF4 (6).

267 Independent synthesis of [{IPrCu}2(SBn)]BF4 (6). IPrCu(MeCN)BF4 (0.200 g, 0.352 mmol) in 5 mL THF is added to a stirring solution of IPrCu(SBn) (0.202 g, 0.352 mmol) in 5 mL THF. The reaction is allowed to stir for 10 minutes. The volatiles were removed in vacuo and the white residue was extracted with 3 × 5 mL of CH2Cl2 and filtered through Celite. The solvent was removed in vacuo and the residue washed with 5 mL of n-pentane. The remaining solid was dried in vacuo to afford 0.346 g (89 % yield)

1 t 1 Analysis of this solid H NMR confirms it to be [{IPrCu)2(S Bu)]BF4. H NMR (CDCl3,

25 °C) δ 7.480 (t, 4, p-ArH- NHC), 7.268 (d, 8, m-ArH-NHC), (s, 4, 7.191, CH-NHC),

6.988 (br S, 3, p-Ar-Bn, m-Ar-Bn), 6.632 (br s, 2, o-Ar-Bn), 2.929 (br s, 2, CH2-Bn),

13 1 2.512 (br sept, 8, CHMe2), 1.208 (d, 24, CHMe2), 1.137 (br s, 24, CHMe2). C{ H}

NMR (CDCl3, 25 °C) δ 180.26, 145.78, 134.51, 130.72, 128.28, 127.84, 124.35, 123.73,

19 1 29.24, 28.84, 25.05, 23.94, F{ H} NMR (CDCl3, 25 °C) δ -161.81. Several peaks in the

1 H NMR spectrum are broad (all of the Bn peaks, CH-NHC CHMe2 on the NHC ligand) and two carbon peaks were missing in the 13C NMR spectra. We attribute this to

+ exchange between IPrCuSBn and IPrCu in CDCl3. Anal Calcd. for C61H79BCu2F4N4S:

C, 65.75; H, 7.15; N, 5.03. Found C, 65.56; H, 7.42; N, 5.20.

t t IPrCu(S Bu) with NOBF4. To a stirring solution of IPrCu(S Bu) (0.100 g, 0.185 mmol) in 1.5 mL CDCl3 is added NOBF4 (0.022 g, 0.185 mmol). The solution immediately turns from light tan to green. This color change suggests the formation of tBuSNO. This

1 formation is confirmed by H NMR spectroscopy (δ 1.895 ppm, Me3CSNO).

Quantification of the formation of tBuSNO was unsuccessful as the amount of tBuSNO

268 was not consistent over several experimental attempts. A new StBu peak appears at δ

0.582 ppm. Crystallization of this reaction mixture from THF reveals this peak to be from

t the formation of a dicopper thiolate cation, [{IPrCu}2(-S Bu)]BF4 (7).

t Independent synthesis of [{IPrCu}2(µ-S Bu)]BF4 (7). IPrCu(MeCN)BF4 (0.200 g,

0.352 mmol) in 5 mL THF is added to a stirring solution of IPrCu(StBu) (0.190 g, 0.352 mmol) in 5 mL THF. The reaction is allowed to stir for 10 minutes. The volatiles were removed in vacuo and the white residue was extracted with 3 x 5 mL of CH2Cl2 and filtered through Celite. The solvent was removed in vacuo and the residue washed with 5 mL of n-pentane. The remaining solid was dried in vacuo to afford 0.325 g (86 % yield)

1 t 1 Analysis of this solid H NMR confirms it to be [{IPrCu)2(-S Bu)]BF4 (7). H NMR

(CDCl3, 25 °C) δ 7.486 (t, 4, p-ArH- NHC), 7.330 (d, 8, m-ArH-NHC), 7.241 (s, 4, CH-

NHC), 2.514 (sept, 8, CHMe2), 1.227 (d, 24, CHMe2), 1.090 (d, 24, CHMe2), 0.582 (s, 9,

t 13 1 Bu); C{ H} NMR (CDCl3, 25 °C) δ 178.36, 145.52, 134.36, 130.70, 124.46, 124.14,

19 1 46.68, 38.58, 28.99, 24.99, 24.03, F{ H} NMR (CDCl3, 25 °C) δ -165.03.

C58H83BCu2F4N4S: Anal. Calcd. for C, 64.37; H, 7.73; N, 5.18. Found C, 64.22; H, 7.64;

N, 5.25.

[{IPrCu}2(µ-SH)]BF4 (9). To a stirring solution of IPrCu(SCPh3) (0.126 g, 0.174 mmol) in 1.5 mL CDCl3 is added NOBF4 (0.020 g, 0.174 mmol). The solution immediately turns

+ from light tan to fluorescent yellow. This color change suggests the formation of Ph3C .

Crystallization of the reaction mixture reveals the fate of the Cu to be [{IPrCu}2(SH)]BF4

(9). The trityl group is lost as the trityl cation as evident by the fluorescent yellow color

269 of the solution. The crystals are colorless. All attempts at independently synthesizing this

compound were unsuccessful. We isolated 0.055 g of crystals (68 % yield assuming

crystals are free of trityl byproducts).

Reactions of sources of the {IPrCu}+ Cation with RSNO.

t [IPrCu(MeCN)]BF4 with RSNO. When 1 equiv. RSNO (R = Bu, Bn) is added to

1 [IPrCu(MeCN)]BF4 in CDCl3 there is no spectral changes indicated by H NMR as compared to the starting material.

F F Synthesis of [IPrCu(Et2O)]BAr 4. NaBAr 4 (2.360 g, 2.68 mmol) was dissolved in 10 mL of Et2O. This solution was added to a slurry of IPrCuCl (1.00 g, 2.06 mmol) in 20 mL

Et2O. The solution was allowed to stir overnight. The Et2O was removed in vacuo and brought back up in 10 mL of CH2Cl2, filtered over Celite, and redissolved in 10 mL of

Et2O. The solution was concentrated to 5 mL and allowed to crystallize overnight at -35

1 °C. This yielded 2.291 (80 % yield) of colorless crystals. H NMR (CDCl3, 25 °C) δ

F F 7.702 (s, 8, o-ArH,BAr 4), 7.518 (t, 2, p-ArH- NHC), 7.493 (s, 4, p-ArH,BAr 4), 7.300

(d, 4, m-ArH-NHC), 7.262 (s, 2, CH-NHC), 3.439 (q, 4, OCH2CH3), 2.473 (sept, 4,

13 1 CHMe2), 1.236 (d, 12, CHMe2), 1.188 (d, 12, CHMe2), 0.824 (t, 6, OCH2CH3); C{ H}

1 2 NMR (CDCl3, 25 °C) δ . δ 161.79 (q, JB-C = 50 Hz), 145.62, 145.14, 139.44 (q, JC-F =

19 1 F{ H} NMR (CDCl3, 25 °) δ -65.418. Anal. Calcd. for C63H58CuN2BF24 : C, 55.09; H,

4.21; N, 2.02. Found C, 54.85; H, 4.26; N, 2.04.

F t t Reaction of [IPrCu(Et2O)]BAr 4 with BuSNO. BuSNO was generated from the

t addition of TlS Bu (0.013 g, 0.043 mmol) in 0.300 mL CDCl3 to stirring NOBF4 (0.005 g,

0.055 mmol) crystals. The colorless solution turns green and the solution is allowed to 270 F stir for 5 min. This solution is added to a light tan solution of [IPrCu(Et2O)]BAr 4 (0.060 g, 0.043 mmol) in 0.300 mL CDCl3. No major color change is observed. This sample is

1 analyzed by H NMR. Bound Et2O shifts from 0.824 ppm to 1.045 ppm (Figure 5.22).

t t F Removing the Et2O in vacuo leads to [IPrCu( BuSS Bu)]BAr 4. All crystallization

1 F attempts were unsuccessful. H NMR (CDCl3, 25 °C) δ 7.698 (s, 8, o-ArH,BAr 4), 7.512

F (m, 6, overlapping p-ArH- NHC and p-ArH,BAr 4), 7.310 (d, 4, m-ArH-NHC), 7.259 (s,

2, CH-NHC), 2.463 (sept, 4, CHMe2), 1.238-1.153 (m, 24, CHMe2), 1.045 (s, 18,

Me3SSMe3).

1 F Figure 5.22 H NMR spectra (300 MHz, CDCl3) of (a) [IPrCu(Et2O)]BAr 4, (b) F t [IPrCu(Et2O)]BAr 4 with 1 equiv. BuSNO, Et2O moves downfield, and (c) [IPrCu(Et2- F t t t O)]BAr 4 with BuSNO, removal of all Et2O in vacuo leaves BuSS Bu weakly binding to {IPrCu+} fragment.

F Reaction of [IPrCu(Et2O)]BAr 4 with BnSNO. BnSNO was generated from the

t addition of TlS Bn (0.014 g, 0.043 mmol) in 0.300 mL CDCl3 to stirring NOBF4 (0.005 g,

0.055 mmol) crystals. The colorless solution turns red and the solution is allowed to stir

271 F for 5 min. This solution is added to a light tan solution of [IPrCu(Et2O)]BAr 4 (0.060 g,

0.043 mmol) in 0.300 mL CDCl3. No major color change is observed. This sample is

1 analyzed by H NMR. Bound Et2O shifts from 0.824 ppm to 0.993 ppm (Figure 5.23).

F Removing the Et2O in vacuo leads to [IPrCu(BnSSBn)]BAr 4. All crystallization attempts

1 F were unsuccessful. H NMR (CDCl3, 25 °C) δ 7.704 (s, 8, o-ArH,BAr 4), 7.510 (m, 6,

F overlapping p-ArH- NHC and p-ArH,BAr 4), 7.310 (d, 4, m-ArH-NHC), 7.264-7.240

(m, 5, CH2Ph), 6.980 (s, 2, CH-NHC), 3.442 (s, 4, PhCH2SSCH2Ph), 2.437 (sept, 4,

CHMe2), 1.216 (d, 12, CHMe2), 1.145 (d, 12, CHMe2).

a)

b)

c)

1 F Figure 5.23 H NMR spectra (300 MHz, RT, CDCl3) of (a) [IPrCu(Et2O)]BAr 4, (b) F [IPrCu(Et2O)]BAr 4 with 1 equiv. BnSNO, Et2O moves downfield, and (c) [IPrCu(Et2- F O)]BAr 4 with BnSNO, removal of all Et2O in vacuo leaves BnSSBn weakling binding to {IPrCu+} fragment.

272 References

(1) Tasker, H. S.; Jones, H. O. J. Chem. Soc. 1909, 95, 1910.

(2) Stamler, J. S.; Simon, D. I.; Osborne, J. A.; Mullins, M. E.; Jaraki, O.; Michel, T.;

Singel, D. J.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 444.

(3) Stamler, J. S.; Jaraki, O.; Osborne, J.; Simon, D. I.; Keaney, J.; Vita, J.; Singel, D.;

Valeri, C. R.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 7674.

(4) Stamler, J. S.; Simon, D. I.; Jaraki, O.; Osborne, J. A.; Francis, S.; Mullins, M.;

Singel, D.; Loscalzo, J. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 8087.

(5) Ignarro, L. J.; Gruetter, C. A. Biochem. Biophys. Acta 1980, 631, 221.

(6) Gruetter, D. Y.; Gruetter, C. A.; Barry, B. K.; Baricos, W. H.; Hyman, A. L.;

Kadowitz, P. J.; Ignarro, L. J. Biochem. Pharmacol. 1980, 29, 2943.

(7) Mellion, B. T.; Ignarro, L. J.; Myers, C. B.; Ohlstein, E. H.; Ballot, B. A.; Hyman,

A. L.; Kadowitz, P. J. Mol. Pharmacol. 1983, 23, 653.

(8) Askew, S. C.; Butler, A. R.; Flitney, F. W.; Kemp, G. D.; Megson, I. L. Bioorg.

Chem. 1995, 3, 1.

(9) Dicks, A. P.; Swift, H. R.; Williams, D. L. H.; Butler, A. R.; Al-Sa'doni, H. H.; Cox,

B. G. J. Chem. Soc. Perkin Trans. 2 1996, 481.

(10) Scrivens, G.; Gilbert, B. C.; Lee, T. C. P. J. Chem. Soc. Perkin Trans. 2 1995, 955.

(11) Gilbert, B. C.; Harrington, G.; Scrivens, G.; Silvester, S. Free Radicals in Biology and Enviornment; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1997.

(12) Dicks, A. P.; Williams, D. L. H. Chem. Biol. 1996, 3, 655.

273 (13) Schrammel, A.; Pfeiffer, S.; Schmidt, K.; Koesling, D.; Mayer, B. Mol. Pharmacol.

1998, 54, 207.

(14) Al-Sa'doni, H. H.; Megson, I. L.; Bisland, S.; Butler, A. R.; Flitney, F. W. 1997,

121, 1047.

(15) Gordge, M. P.; Meyer, D.; Hothersall, J. S.; Neild, G. H.; Payne, N. N.; Noronha-

Dutra, A. A. Br. J. Pharmacol. 1995, 114, 1083.

(16) Arnelle, D. R.; Stamler, J. S. Arch. Biochem. Biophys. 1995, 318, 279.

(17) Munro, A. P.; Williams, D. H. L. J. Chem. Soc. Perkins Trans. 2 2000, 1794.

(18) Hogg, N. Annu. Rev. Pharmacol. Toxicol. 2002, 42, 585.

(19) Ignarro, L. J. Nitric Oxide: Biology and Pathology; Academic Press: San Diego,

2000.

(20) Dicks, A.; Li, E.; Munro, A.; H., S.; Williams, D. Can. J. Chem. 1998, 76, 789.

(21) Houk, K. N.; Hietbrink, B. N.; Bartberger, M. D.; R., M. P.; Choi, R. Y.; Voyksner,

R. D.; Stamler, J. S.; Toone, E. J. J. Am. Chem. Soc. 2003, 125, 6972.

(22) Singh, P. S.; Wishnok, J. S.; Keshive, M.; Deen, W. M.; S., T. Proc. Natl. Acad.

Sci. U. S. A. 1996, 93, 14428.

(23) Srivastava, S.; Dixit, B. L.; Ramana, K. V.; Chandra, A.; Chandra, D.; Zacarias, A.;

Petrash, J. M.; Bhatnagar, A.; Srivastava, S. K. Biochem. J. 2001, 358, 111.

(24) Perissinotti, L. L.; Adrian, G.; Turjanski, A. G.; Estrin, D. A.; Doctorovich, F. J.

Am. Chem. Soc. 2005, 127, 486.

(25) Savelieff, M. G.; Wilson, T. D.; Elias, Y.; Nigles, M. J.; Garner, D. K.; Lu, Y.

Proc. Natl. Acad. Sci. USA 2008, 105, 7919. 274 (26) Nar, H.; Messerschmidt, A.; Huber, R.; van de Kamp, M.; Canters, G. W. J. Mol.

Biol. 1991, 221, 765.

(27) Lieberman, R. L.; Arciero, D. M.; Hooper, A. B.; Rosenzweig, A. C. Biochemistry

2001, 40, 5674.

(28) Brown, K.; Djinovic-Carugo, K.; Haltia, T.; Cabrito, I.; Saraste, M.; Moura, J. J.;

Moura, I.; Tegoni, M.; Cambillau, C. J. Biol. Chem. 2000, 275, 41133.

(29) Houser, R. P.; Young, J., V. G.; Tolman, W. B. J. Am. Chem. Soc. 1996, 118, 2101.

(30) Delp, S. A.; Munro-Leighton, C.; Goj, L. A.; Ramı´rez, M. A.; Gunnoe, B. T.;

Petersen, J. L.; Boyle, P. A. Inorg. Chem. 2007, 46, 2365.

(31) Hermann, W. A.; Köcher, C. Angew. Chem., Int. Ed. Engl. 1997, 36, 2162

(32) Huang, J.; Nolan, S. P. J. Am. Chem. Soc. 1991, 121, 9889.

(33) Lin, J. C. Y.; Huang, R. T. W.; Lee, C. S.; Bhattacharyya, A.; Hwang, W. S.; Lin, I.

J. B. Chem. Rev. 2009, 104, 10.1021.

(34) Jurkauskas, V.; Sadighi, J. P.; Buchwald, S. L. Org. Lett. 2003, 5, 2417.

(35) Fructos, M. R.; de Fremont, P.; Nolan, S. P.; Diaz-Requejo, M. M.; Perez, P. J.

Organometallics 2006, 25, 2237.

(36) Perissinotti, L. L.; Estrin, D. A.; Leitus, G.; Doctorovich, F. J. Am. Chem. Soc.

2006, 128, 2512.

(37) Perissinotti, L. L.; Leitus, G.; Shimon, L.; Estrin, D.; Doctorovich, F. Inorg. Chem.

2008, 47, 4723.

(38) Arulsamy, N.; Bohle, D. S.; Butt, J. A.; Irvine, G. J.; Jordan, P. A.; E. Sagan, E. J.

Am. Chem. Soc. 1999, 121, 7115. 275 (39) Bartberger, M. D.; Houk, K. N.; Powell, S. C.; Mannion, J. D.; Lo, K. Y.; Stamler,

J. S.; Toone, E. J. J. Am. Chem. Soc. 2000, 122, 5889.

(40) Varonka, M. S.; Warren, T. H. Inorg. Chim. Acta 2007, 360, 317.

(41) Brookhart, M.; Grant, B.; Volpe Jr, A. F. Organometallics 1992, 11, 3920.

(42) Das, J.; Patil, S. N.; Awasthi, R.; Narasimhulu, C. P.; Trehan, S. Synthesis 2005,

11, 1801.

276 Appendix

Synthesis of Copper β-Diketiminate and Anilidoimine Complexes

Introduction

β-diketiminates are bidentate ligands which typically coordinate through the two

nitrogen atoms of the ligand to late transition metal ions. They have been used as

ancillary ligands for cationic, early transition metal olefin polymerization catalysts1-3 as well as for the design of model complexes for the active sites of metalloenzymes.4,5

Having two donor N atoms at long distances from each other (2.8-3.0 Å), β-diketiminates can be used to model histidine sites present in many non-heme metalloenzymes6

The steric and electronic properties of anionic β-diketiminate ligands may be readily varied by a number of different synthetic approaches. Synthetically modifying the aryl rings can control whether the metal is mononuclear or dinculear.7 Furthermore, the donor ability of the ligands may be tuned without changing their steric profile by changing the substituents on the backbone.8,9 β-diketiminates are difficult to displace from the metal center because they are anionic and covalently bound through the nitrogen to the metal center.

β-Diketiminate ligands can stabilize three-coordination in main group and late transition metal complexes. A number of research groups have reported low-coordinate complexes of Ti,10, Rh, Ir,11 Mn,12 Fe,13-15 Co,4 Ni4, Cu16 and Zn17,18 with sterically encumbered β-diketiminates. For example, Holland et al. have reported electronically unsaturated, three-coordinate complexes of Fe, Co and Ni with a bulky diketiminate

277 bis-Histidine -Diketiminate Anilidoimine

+ R''' OOC NH3

N R' R R' R R' N N N R'' Cu Cu Cu N N N R R R' R R'' R

N L1 R''' L5 + R = R' = Me, R'' = R''' = H OOC NH3 R = R' = Me, R'' = H L2 L6 R = R' = R''' = Me, R'' = H R = iPr, R' = Me, R'' = H L3 L7 R = iPr, R' = Me, R'' = R''' = H R = R' = iPr, R'' = H L4 L8 R = iPr, R' = R''' = H, R'' = Ph R = R' = iPr, R'' = Me

Figure A.1. Similarity between bis(histidine)-copper coordination and β-diketiminate land anilidoimine copper complexes L1Cu – L8Cu.

ligand (R = iPr).4 Tolman et al. have used three-coordinate β-diketiminate Cu(I) complexes4,5 to mimic dioxygen activation in biological and industrial catalytical systems.

The central methine position of the β-diketiminate ligand is known to be susceptible to electrophilic attack which complicates reactivity studies with electrophiles.19 We hypothesized that these issues could be addressed by using related anilido imine ligands20 where the backbone C-H is “protected” from attack by an

278 electrophile. The anilidoimine ligands have similar features to the β-diketiminate ligands

in that they are bidentate, anionic ligands readily amenable to electronic and steric tuning.

We have employed electron-rich β-diketiminate as well as anilidoimine ligands as supporting ligands to provide access to low-coordinate copper(I) and copper(II) species which could be used as precursors for studying the reactivity of E-NO compounds with copper complexes (Figure A.1.). Compiled below are general procedures used to prepare different classes of β-diketiminate and anilidoimine ligands along with their lithium salts and copper(I) complexes. Detailed procedures for many specific compounds can be found in the literature references cited below. Representative 1H NMR data are presented for each substance.

Preparation of compounds

A.1. General experimental procedure for β- diketimine ligands (L1H-L3H).

To prepare the free β-diketimine ligands (L1H-L3H), 2 equiv. of the corresponding aniline (15-30 g), 1 equiv. 2,4-pentanedione and 1 equiv. p-toluenesulfonic acid are heated under reflux in toluene (250 mL) using a Dean-Stark trap for 4-6 h until about 5 mL of water is seen in the Dean-Stark trap. The ligand is worked up from a saturated solution of Na2CO3(aq) with CH2Cl2 followed by precipitation from cold methanol. This procedure was adapted from the method of Budzelaar et al.11,21,22 and works well for a wide variety of nitrogen aryl ring substituents. L1H-L3H can be conveniently prepared on a 15-20 g scale.

279 Scheme A.1. Preparation of L1H – L3H from the condensation of 2,4-pentanedione with 2 equiv. aniline.

[Me2NN]H (L1H). L1H can be obtained in 20 g quantities and is isolated as a light tan

1 powder from cold methanol (90 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ 12.217 (s,

1, NH), 7.019-6.929 (m, 6, m-ArH, p-ArH), 4.813 (s, 1, backbone-H), 2.130 (s, 12, Ar-

Me), 1.541 (s, 6, backbone-Me).

[Me3NN]H (L2H). L2H can be obtained in 20 g quantities and is isolated as a light tan

1 powder from cold methanol (88 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ 12.239 (s,

1, NH), 6.809 (m, 4, m-ArH), 4.854 (s, 1, backbone-H), 2.184 (s, 6, p-Ar-Me), 2.152 (s,

12, o-Ar-H) 1.597 (s, 6, backbone-Me).

i [ Pr2NN]H (L3H). L3H can be obtained in 20 g quantities and is isolated as a light tan

1 powder from cold methanol (92 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ 12.466 (s,

1, NH), 7.200-7.114 (m, 6, m-ArH, p-ArH), 4.874 (s, 1, backbone-H), 3.303 (sept, 4, Ar-

CHMe2), 1.657 (s, 6, backbone-Me), 1.202 (d, 12, Ar-CHMe2), 1.202 (d, 12, Ar-CHMe2),

1.150 (d, 12, Ar-CHMe2).

280 A.2. General experimental procedure for β -diketimine ligand L4H.

Preparation of L4H which possesses a phenyl group on the central backbone carbon of the ligand was prepared according to the literature procedure. The ligand synthesis first requires the synthesis of 2-phenyl-1,3-bis(dimethylamino)trimethinium hexafluorophosphate(Scheme A.2a).24 The 2-phenyl-1,3-bis(dimethylamino)trimethinium hexafluorophosphate was treated with NaOH under reflux to give 3-hydroxy-2- phenylacrylaldehyde. In a second step 2 equiv.2,6-diisopropylaniline are added to 3-

i hydroxy-2-phenylacrylaldehyde in the presence of acid (HCl) yield [ Pr2NN]PhL4H can be prepared on a 10-15 g scale.

i Scheme A.2. Preparation of [ Pr2NN]PhH (L4H).

i [ Pr2NN]PhH (L4H). L4H can be obtained in 15 g quantities and is isolated as a yellow

1 crystalline solid from hot ethanol (75 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ

281 12.418 (s, 1, NH), 7.821(s, 2, p-ArH), 7.291-7.263 (m, 6, Ar-H), 7.167 (s, 2, backbone-

H), 7.052 (m, 3, Ar-H), 3.443 (sept, 4, Ar-CHMe2), 1.199 (d, 24, Ar-CHMe2).

A.3. General experimental procedure for anilineimine ligands (L5H-L7H). The anilineimine ligand where R'' = H (Figure A.1) can be prepared in two steps following the literature procedure (Scheme A.3.).20 The starting materials are all commercially available which provides an easy route to the ligand. Imine formation occurs smoothly at room temperature from the reaction of the corresponding aniline and o-fluoro benzaldehyde. The “anilido” portion of the ligand was installed via a nucleophilic aromatic displacement of fluoride using LiNHAr. L5H-L7H can be prepared on a 15-20 g scale.

Scheme A.3. Preparation of anilineimine ligands (L5H-L7H).

[Me2AI]H (L5H). L5H can be obtained in 15-20 g quantities and is isolated as a bright

1 yellow crystalline solid from hot methanol (85 % yield). H NMR (C6D6, 300 MHz, 25

°C) δ 10.633 (s, 1, amide-NH), 7.996 (s, 1, imine-NH), 7.063-6.932 (m, 9, Ar-H),

6.588 (t, 1, Ar-H), 6.351 (d, 1, Ar-H), 2.161 (s, 6, Ar-Me2 ), 2.101 (s, 6, Ar-Me2).

[MeAiPrI]H (L6H). L6H can be obtained in 15-20 g quantities and is isolated as a bright

1 yellow crystalline solid from hot methanol (87 % yield). H NMR (C6D6, 300 MHz, 25

282 °C) δ 10.991 (s, 1, amide-NH), 8.353 (s, 1, imine-NH), 7.326-7.276 (m, 7, Ar-H), 7.233

(d, 1, Ar-H), 7.082 (t, 1, Ar-H), 6.710 (t, 1, Ar-H), 6.487 (d, 1, Ar-H), 3.337 (sept, 2, Ar-

CHMe2), 2.331 (s, 6, Ar-Me2), 1.290 (d, 12, Ar-CHMe2).

i [ Pr2AI]H (L7H). L7H can be obtained in 15-20 g quantities and is isolated as a bright

1 yellow crystalline solid from hot methanol (85 % yield). H NMR (C6D6, 300 MHz, 25

°C) δ 10.526 (s, 1, amide-NH), 8.363 (s, 1, imine-NH), 7.337-7.136 (m, 9, Ar-H),

6.701 (t, 1, Ar-H), 6.286 (d, 1, Ar-H), 3.212 (sept, 2, Ar-CHMe2), 3.106 (sept, 2, Ar-

CHMe2), 1.178 (d, 12, Ar-CHMe2), 1.142 (d, 12, Ar-CHMe2).

A.4. General experimental procedure for anilineimine ligand L8H.

The initial condensation of 2,6-diisopropylaniline with 2-fluoro-acetophenone to form the corresponding imine did not proceed to an appreciable amount by typical acid-catalyzed condensation procedures. Thus an alternative synthesis to form the imine was required

25 using TiCl4. The “anilido” portion of the ligand was installed via a nucleophilic aromatic displacement of fluoride using LiNHAr. L8H can be prepared on a 15-20 g scale.

i Scheme A.4. Preparation of [ Pr2AI]MeH (L8H).

283 i [ Pr2AI]MeH (L8H). L8H can be obtained in 15-20 g quantities and is isolated as a bright

1 yellow crystalline solid from hot ethanol (77 % yield). H NMR (C6D6, 300 MHz, 25 °C)

δ 11.582 (s, 1, amide-NH) 7.485 (d, 1, Ar-H), 7.308-7.160 (m, 8, Ar-H), 6.586 (t, 1, Ar-

H), 6.475 (d, 1, Ar-H), 3.447 (sept, 2, Ar-CHMe2), 3.057 (sept, 2, Ar-CHMe2), 1.964 (s,

3, imine-Me), 1.171 (d, 6, Ar-CHMe2), 1.152-1.120 (m, 18, Ar-CHMe2).

A.5. General experimental procedures to prepare lithium salts of β- diketimate ligands (L3-L4) and anilidoimine ligands (L7-L8).

In an inert atmosphere, chilled n-butyllithium (1.6 M, -35 °C) was added dropwise to a stirred solution of free ligand (L3H-L4H or L7H-L8H) (1.00-20.00 mmol) in toluene (5-

200 mL) at -35 °C. The solution is allowed to stir for 1.5 hr while warming to room temperature. The toluene is removed in vacuo to yield brown or yellow oils. The oil is washed with pentane (3 × 3-10 mL) to afford a colorless or yellow powder. The solution is filtered and the solid is collected on a frit to yield (70-90 %) pure lithium salt LLi.

Scheme A.5. Preparation of L3Li-L4Li, L6Li-L8Li from the reaction of free ligand (L3H-L4H, L6H-L8H) with n-butyllithium solution in nonpolar/non-coordinating solvents (toluene / pentane).

284 i [ Pr2NN]Li (L3Li). L3Li can be obtained in 20 g quantities and is isolated as a light tan

1 powder from pentane (77 % yield) H NMR (C6D6, 300 MHz, 25 °C) δ 7.234-7.123 (m,

6, m-ArH, p-ArH), 5.014 (s, 1, backbone-H), 3.423 (sept, 4, Ar-CHMe2), 1.913 (s, 6, backbone-Me), 1.276 (d, 12, Ar-CHMe2), 1.204 (d, 12, Ar-CHMe2), 1.567 (d, 12, Ar-

CHMe2).

i [ Pr2NN]PhLi (L4Li). L4Li can be obtained in 1 -5 g quantities and is isolated as a bright

1 yellow crystalline solid from pentane (90 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ

7.807 (s, 2, backbone-H), 7.326 (d, 2, Ar-H), 7.120-7.095 (m, 8, Ar-H), 6.298 (t, 1, Ar-

H), 3.031 (sept, 4, Ar-CHMe2), 1.198 (d, 12, Ar-CHMe2), 1.072 (d, 12, Ar-CHMe2).

[MeAiPrI]Li (L6Li). L6Li can be obtained in 1 – 5 g quantities and is isolated as a

1 bright yellow crystalline solid from pentane (82 % yield). H NMR (C6D6, 300 MHz, 25

°C) δ 8.082 (s, 1, imine-NH), 7.218-6.977 (m, 9, Ar-H), 6.515 (t, 1, Ar-H), 6.354 (d, 1,

Ar-H), 3.213 (sept, 2, Ar-CHMe2), 2.314 (s, 6, Ar-Me2), 1.167 (d, 6, Ar-CHMe2), 1.116

(d, 6, Ar-CHMe2).

i [ Pr2AI]Li (L7Li). L7Li can be obtained in 1 – 5 g quantities and is isolated as a bright

1 yellow crystalline solid from pentane (70 % yield). H NMR (C6D6, 300 MHz, 25 °C) δ

7.916 (s, 1, imine-NH), 7.315-7.234 (m, 7, Ar-H), 7.055 (d, 1, Ar-H), 6.945 (t, 1, Ar-H),

6.379 (t, 1, Ar-H), 6.270 (d, 1, Ar-H), 3.328 (sept, 2, Ar-CHMe2), 3.328 (sept, 2, Ar-

CHMe2), 2.857 (sept, 2, Ar-CHMe2), 1.222 (d, 6, Ar-CHMe2), 1.180 (d, 12, Ar-CHMe2),

1.040 (d, 6, Ar-CHMe2).

285 i [ Pr2AI]MeLi (L8Li). L8Li can be obtained in 0.25 – 2 g quantities and is isolated as a

1 bright yellow crystalline solid from pentane (86 % yield) ( H NMR (C6D6, 300 MHz, 25

°C) δ 7.691 (d, 1, Ar-H), 7.388-7.053 (m, 8, Ar-H), 6.606 (d, 1, Ar-H), 6.415 (t, 1, Ar-

H), 3.378 (sept, 2, Ar-CHMe2), 2.847 (sept, 2, Ar-CHMe2), 2.122 (s, 3, imine-Me), 1.268

(d, 6, Ar-CHMe2), 1.231 (d, 6, Ar-CHMe2), 1.124 (d, 6, Ar-CHMe2), 1.083 (d, 6, Ar-

CHMe2).

[Me2AI]Li (L4Li). L4Li cannot be prepared using n-butyllithum because it acts as a nucleophile as well as a simple base; a mixture of L4Li is produced along with the imine alkylated product. A bulkier alkyllithium reagent is thus necessary and we used tert- butyllithium. A cold solution of 1.7 M tert-butyllithium (10 mL, 17 mmol) was slowly added to a cold solution of [Me2AI]H (5 g, 15.2 mmol) in 100 mL of toluene causing the formation of a bright yellow solution. The solution was allowed to stir for two hours. The solvent was removed in vacuo and the residue was suspended in 30 mL of pentane and

Scheme A.6. Preparation of L5Li from the reaction of [Me2AI]H (L5H) free ligand and tert-butyllithium solution in nonpolar/non-coordinating solvent (toluene / pentane).

286 allowed to sit at -35ºC overnight. The solid was isolated, washed with several mL of cold

1 pentane and dried in vacuo to yield 4.60 g (91% yield). H NMR (C6D6, 300 MHz, 25 °C)

δ 7.701 (s, 1, imine-NH), 7.194-6.890 (m, 9, Ar-H), 6.434 (d, 1, Ar-H), 6.333 (t, 1, Ar-

H), 2.115 (s, 6, Ar-Me2), 1.920 (s, 6, Ar-Me2).

A.6. General procedure for the synthesis of LCu(MeCN) Complexes L3Cu –

L8Cu.4,23,25 In an inert atmosphere a solution of lithium salt (1.00-20.00 mmol) in THF

(5-200 mL) was added to a slurry of [Cu(MeCN)4]PF6 (1.0 equiv) in THF and stirred for

30 min. The solvent was removed in vacuo and the residue extracted with toluene (3 × 5-

50 mL). The mixture was stirred for 10 min and then filtered through Celite. The solvent was removed in vacuo to yield a brown oil. The oil was washed with pentane (3 × 5-50 mL) to precipitate out a yellow or orange solid. The solution was allowed to sit overnight at -35 °C in the freezer to encourage all of the solid to precipitate out of solution. The solution is filtered and the solid is collected on a frit to yield (75-90 %) LCu(MeCN).

Scheme A.7. Preparation of L3Cu(MeCN)-L5Cu(MeCN) from the reaction of L3Li- L8Li with [Cu(MeCN)4]PF4 in THF, worked up from pentane.

i [ Pr2NN]Cu(MeCN) (L3Cu(MeCN)). L3Cu(MeCN) can be obtained in 0.20 -1.0 g

quantities and is isolated as a pale yellow crystalline solid from pentane (90 % yield). 1H

NMR (C6D6, 300 MHz, 25 °C) δ 7.174 (d, 4, m-Ar-H), 7.075 (t, 2, p-Ar-H), 4.986 (s, 1, 287 backbone-H), 3.589 (sept, 4, Ar-CHMe2), 1.854 (d, 12, Ar-CHMe2), 1.234 (d, 12, Ar-

CHMe2), 0.045 (s, 3, MeCN).

i [ Pr2NN]PhCu(MeCN) (L4Cu(MeCN)). L4Cu(MeCN) can be obtained in 0.20-2.0 g quantities and is isolated as a pale yellow crystalline solid from pentane (86 % yield). 1H

NMR (C6D6, 300 MHz, 25 °C) δ 7.938 (s, 2, backbone-H), 7.321 (d, 2, Ar-H), 7.118-

7.024 (m, 8, Ar-H), 6.955 (t, 1, Ar-H), 3.189 (sept, 4, Ar-CHMe2), 1.153 (d, 12, Ar-

CHMe2), 1.029 (d, 12, Ar-CHMe2), 0.256 (s, 3, MeCN).

[Me2AI]Cu(MeCN) (L5Cu(MeCN)). L5Cu(MeCN) can be obtained in 0.20-0.50 g quantities and is isolated as a bright orange crystalline solid from pentane (75 % yield).

1 H NMR (C6D6, 300 MHz, 25 °C) δ 7.781 (s, 1, imine-NH), 7.219 (d, 2, Ar-H), 7.090-

6.901 (m, 7, Ar-H), 6.546 (d, 1, Ar-H), 6.336 (t, 1, Ar-H), 2.158 (s, 6, Ar-Me2), 1.999 (s,

6, Ar-Me2), 0.440 (s, 3, MeCN).

[MeAiPrI]Cu(MeCN) (L6Cu(MeCN)). L6Cu(MeCN) can be obtained in 0.50-2.0 g

quantities and is isolated as a pale yellow crystalline solid from pentane (78 % yield). 1H

NMR (C6D6, 300 MHz, 25 °C) δ 8.042 (s, 1, imine-NH), 7.221-6.879 (m, 9, Ar-H),

6.512 (t, 1, Ar-H), 6.317 (d, 1, Ar-H), 3.043(sept, 2, Ar-CHMe2), 2.064 (s, 6, Ar-Me2),

1.182 (d, 6, Ar-CHMe2), 1.003 (d, 6, Ar-CHMe2), 0.293 (s, 3, MeCN).

i [ Pr2AI]Cu(MeCN) (L7Cu(MeCN)). L7Cu(MeCN) can be obtained in 0.30-0.75 g quantities and is isolated as a yellow crystalline solid from pentane (83 % yield). 1H

NMR (C6D6, 300 MHz, 25 °C) δ 8.083 (s, 1, imine-NH), 7.345-7.130 (m, 4, Ar-H),

288 7.034 (m, 4, Ar-H), 6.866 (d, 1, Ar-H), 6.789 (t, 1, Ar-H), 3.534(sept, 2, Ar-CHMe2),

3.320 (sept, 2, Ar-CHMe2), 1.352 (d, 6, Ar-CHMe2), 1.280 (d, 12, Ar-CHMe2), 1.214 (d,

6, Ar-CHMe2), 1.145 (d, 6, Ar-CHMe2).

i [ Pr2AI]MeCu(MeCN) (L8Cu(MeCN)). L8Cu(MeCN) can be obtained in 0.50-1.50 g quantities and is isolated as a pale yellow crystalline solid from pentane (78 % yield). 1H

NMR (C6D6, 300 MHz, 25 °C) δ 7.676 (d, 1, Ar-H), 7.345-7.130 (m, 7, Ar-H), 6.789 (d,

1, Ar-H), 6.478 (t, 1, Ar-H), 3.703 (sept, 2, Ar-CHMe2), 3.220 (sept, 2, Ar-CHMe2),

2.139 (s, 3, imine-Me), 1.456 (d, 6, Ar-CHMe2), 1.331 (d, 6, Ar-CHMe2), 1.268 (d, 6, Ar-

CHMe2), 1.134 (d, 6, Ar-CHMe2), 0.903 (s, 3, MeCN).

I 26 26 27 A.7. General procedure for [Cu ]2 complexes {L1Cu}2, {L2Cu}2, {L5Cu}2 .

In an inert atmosphere, free ligand LH (1.00-20.00 mmol) in (5-100 mL) of toluene was adding to a stirring solution of copper(I) tert-butoxide (1.200-24.00) in (10-100 mL) of toluene. The solution was allowed to stir for 2 hr, filtered over Celite, and the volatiles were removed in vacuo to yield brown oil. The brown oil was washed with pentane (3 ×

26 5-50 mL) to afford {LCu}2 as a yellow or orange powder in (60-92 % yield). It should be noted that ligands with isopropyl substituents on the ortho positions of the aryl rings are too bulky to have successful results with the following synthesis. Lithium salts LLi instead should be used to transmetallate to copper.

289 Scheme A.8. Synthesis of base free copper complexes {L1Cu}2, {L2Cu}2, {L5Cu}2 from reaction of free ligand, L1H, L2H, or L5H with CuIOtBu.

{[Me2NN]Cu}2 ({L1Cu}2). {L1Cu}2 can be obtained in 0.50-10 g quantities and is

1 isolated as a yellow solid from pentane (92 % yield) H NMR (benzene-d6, 400 MHz, 25

°C): δ 7.100 (d, 4, m-ArH), 6.993 (t, 2, p-ArH), 4.776 (s, 1, backbone-CH), 2.016 (s, 12, o-ArMe2), 1.643 (s, 6, backbone-Me).

{[Me3NN]Cu}2 ({L2Cu}2). {L2Cu}2 can be obtained in 0.50-10 g quantities and is

1 isolated as a yellow solid from pentane (85% yield) H NMR (benzene-d6, 300 MHz, 25

°C): δ 6.924 (s, 4, m-ArH), 4.809 (s, 1, backbone-CH), 2.275 (s, 6, p-ArMe2), 2.033 (s,

12, o-ArMe2), 1.675 (s, 6, backbone-Me).

{[Me2AI]Cu}2 ({L5Cu}2). {L5Cu}2 can be obtained in 0.25-1.5 g quantities and is

1 isolated as an orange solid from pentane (60 % yield). H NMR (benzene-d6, 300 MHz,

25 °C): δ 7.742 (s, 1, imine-H), 6.888-7.233 (m, 8, Ar-H), 6.520 (d, 1, Ar-H), 6.326 (t, 1,

Ar-H), 2.077 (s, 3, Ar-Me2), 1.919 (s, 3, Ar-Me2).

290 References

(1) Kim, W.-K.; Fevola, M. J.; Liable-Sands, L. M.; Rheingold, A. L.; Theopold, K. H.

Organometallics 1998, 17, 4541.

(2) Gibson, V. C.; Maddox, P. J.; Newton, C.; Redshaw, C.; Solan, G. A.; White, A. J.

P.; Williams, D. J. Chem. Commun. 1998.

(3) Hayes, P. G.; Piers, W. E.; McDonald, R. J. Am. Chem. Soc. 2002, 124, 2132.

(4) Spencer, D. J. E.; Reynolds, A. M.; Holland, P. L.; Jazdzewski, B. A.; Duboc-Toia,

C.; Le Pape, L.; Yokota, S.; Tachi, Y.; Itoh, S.; Tolman, W. B. Inorg. Chem. 2002, 41,

6307.

(5) Aboelella, N. W.; Gherman, B. F.; Hill, L. M. R.; York, J. T.; Holm, N.; Young Jr,

V. G.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2006, 128, 3445.

(6) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996, 96, 2239.

(7) Bourget-Merle, L.; Lappert, M. F.; Severn, J. R. Chem. Rev. 2002, 102, 3031.

(8) Yokota, S.; Tachi, Y.; Nishiwaki, N.; Ariga, M.; Itoh, S. Inorg. Chem. 2001, 40,

5316.

(9) Laitar, D. S.; Mathison, C. J. N.; Davis, W. M.; Sadighi, J. P. Inorg. Chem. 2003, 42,

7354.

(10) Basuli, F.; Huffman, J. C.; Mindiola, D. J. Inorg. Chem. 2003, 42, 8003.

(11) Budzelaar, P. H. M.; Moonen, N. N. P.; Gelder, R. D.; Smits, J. M. M.; Gal, A. W.

Eur. J. Inorg. Chem. 2000, 753.

291 (12) Chai, J.; Zhu, H.; Fan, H.; Roesky, H. W.; Magull, J. Organometallics 2004, 23,

1177.

(13) Smith, J. M.; Lachicotte, R. J.; Holland, P. L. Chem. Commun. 2001, 1542.

(14) Smith, J. M.; Lachicotte, R. J.; Holland, P. L. Organometallics 2002, 21, 4808.

(15) Gibson, V. C.; Marshall, E. L.; Navarro-Llobet, D.; White, A. J. P.; Williams, D. J.

J. Chem. Soc., Dalton Trans. 2002, 4321.

(16) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 1999, 121, 7270.

(17) Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 1998, 120, 11018.

(18) Cheng, M.; Darling, N. A.; Lobkovsky, E. B.; Coates, G. W. Chem. Commun.

2000, 2007.

(19) Brown, E. C.; Aboelella, N. W.; Reynolds, A. M.; Aullo'n, G.; Alvarez, S.;

Tolman, W. B. Inorg. Chem. 2004, 43, 3335.

(20) Hayes, P. G.; Welch, G. C.; Emslie, D. J. H.; Noack, C. L.; Piers, W. E.; Parvez, M.

Organometallics 2003, 22, 1577.

(21) Budzelaar, P. H. M.; Oort, A. B. V.; Orpen, A. G. Eur. J. Inorg. Chem. 1998, 1485.

(22) Budzelaar, P. H. M.; Gelder, R.; Gal, A. W. Organometallics 1998, 17, 4121.

(23) Spencer, D. J. E.; Aboelella, N. W.; Reynolds, A. M.; Holland, P. L.; Tolman, W.

B. J. Am. Chem. Soc. 2002, 124, 2108.

(24) Davies, I. W.; Marcoux, J. F.; Wu, J.; Palucki, M.; Corley, E. G.; Robbins, M. A.;

Tsou, N.; Ball, R. G.; Dormer, P.; Larsen, R. D.; Reider, P. J. J. Org. Chem. 2000, 65,

4571.

292 (25) Reynolds, A. M. G., B. F.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2005,

44, 6989.

(26) Amisial, L. T.; Dai, X.; Kinney, R. A.; Krishnaswamy, A.; Warren, T. H. Inorg.

Chem. 2004, 43, 6537.

(27) Badiei, Y. M.; Krishnaswamy, A.; Melzer, M. M.; Warren, T. H. J. Am. Chem. Soc.

2006, 128, 15056.

293