ABSTRACT PÉREZ, DAMARIS ESTHER. Synthesis, Characterization and Reactivity of Rhenium(III) Complexes for Catalytic, Stoichiomet

ABSTRACT

PÉREZ, DAMARIS ESTHER. Synthesis, Characterization and Reactivity of Rhenium(III) Complexes for Catalytic, Stoichiometric and Insertion Reactions. (Under the direction of Elon A. Ison).

This document focuses on the synthesis and characterization of new rhenium(III) complexes bearing tridentate pincer ligands. These organometallic complexes were subject to comprehensive studies of their reactivity towards catalytic, stoichiometric and insertion reactions. Mechanistic studies of these transformations will lead to the development of more efficient catalytic systems. In Chapter 2, cationic bisacetonitrile diamidoamine (DAAm) rhenium(III) and cationic diamineamido (DAmA) rhenium(III) complexes were synthetized, and characterized. Their reactivity towards the hydrosilylation reaction of aldehydes was studied. It was proposed through experimental and kinetic data that the most likely pathway for this transformation follows a non- hydride ionic hydrosilylation mechanism. Various aliphatic and aromatic aldehydes were tested. Excellent yields were achieved at ambient temperature in neat conditions using dimethylphenylsilane. The reaction affords TONs of up to 9,200 and a TOF of up to 126 h-1. In Chapter 3, a computational mechanistic study of the hydrosilylation reaction of aldehydes by novel cationic rhenium(III) bisacetronitrile complex is presented. DFT analysis supports a non-hydride ionic hydrosilylation mechanism which was previously proposed based on experimental and kinetic data.

In Chapter 4, the reactivity of novel DAAm-C6F5 rhenium(III) acetate complex towards Lewis acidic boron reagents was investigated. The role of water as an oxygen source was studied.

Novel rhenium(III) and rhenium(V) complexes bearing a C6F5 ligand were synthesized, characterized, and proposed as intermediates. Mechanistic transformations were proposed base on experimental data and previous reports in the literature. In Chapter 5, new diamidoamine (DAAm) rhenium(III) aryl complexes were synthetized and characterized. Kinetic studies suggest a pseudo-first order overall reaction. The mechanism for the CO insertion with rhenium(III) complexes was proposed. Comparison of rhenium(III) versus rhenium(V) for CO activation was established.

Synthesis, Characterization and Reactivity of Rhenium(III) Complexes for Catalytic, Stoichiometric and Insertion Reactions

by Damaris Esther Pérez

A dissertation submitted to the Graduate Faculty of North Carolina State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Chemistry

Raleigh, North Carolina 2018

APPROVED BY:

______Dr. Elon A. Ison Dr. Reza Ghiladi Committee Chair

______Dr. Elena Jakubikova Dr. Joshua Pierce

DEDICATION

To my husband, Hernán

To my baby girl, Sofía Esther

To my parents, Zory and Fidel

To my sisters, Sandra and Vivian

To my grandparents, Mama Candy (1928 - 2012) and Papa Tato

For your support and unconditional love.

BIOGRAPHY

Damaris Esther Pérez (formerly Rivera-Santos) was born on September 13th, 1986 in

Aibonito, Puerto Rico. She and her sisters, Sandra and Vivian were raised by their parents, Fidel and Zoraida. From an early age, she had a passion for math and science. Damaris graduated from

Bonifacio Sánchez Jiménez high school in May 2004 and moved to San Juan, Puerto Rico to complete her undergraduate degree at University of Puerto Rico at Río Piedras (UPR-RP). She graduated with a Bachelor of Science degree in Chemistry in May 2009. Upon graduation, she stayed at UPR-RP took graduate courses and worked as a teaching assistant. Then, in August 2010, she moved to Oswego, New York to pursue a Master’s Degree in Chemistry at the State University of New York (SUNY) at Oswego. She conducted graduated research under the supervision of Dr.

Fehmi Damkaci in organic chemistry and worked as a teaching assistant. She graduated with a

M.S. in chemistry in May 2012 and then moved to Lincoln, Nebraska. In August 2012, she started graduated research under the supervision of Dr. James Takacs in organic chemistry and worked as a teaching assistant at University of Nebraska Lincoln (UNL). In May 2013, she moved to Raleigh,

North Carolina to pursue her Ph.D. in Chemistry at North Carolina State University. Since arriving to NC State University, she has been conducting her doctoral research in organometallic chemistry under the direction of Dr. Elon A. Ison. Upon graduation in 2018, Damaris plans to pursue a professional career in chemical industry.

iii

TABLE OF CONTENTS

LIST OF TABLES ...... vi LIST OF FIGURES ...... vii LIST OF SCHEMES ...... ix Chapter 1: General Introduction ...... 1 1.1 Organometallic Chemistry ...... 2 1.2 Scope of this dissertation ...... 3 1.3 References ...... 5 Chapter 2: Cationic Rhenium(III) Complexes: Synthesis, Characterization, and Reactivity for Catalytic Hydrosilylation Reaction of Aldehydes ...... 7 2.1 Abstract ...... 8 2.2 Introduction ...... 8 2.3 Results and Discussion ...... 10 2.4 Conclusions ...... 34 2.5 Experimental Section ...... 35 2.6 References ...... 42 Chapter 3: Mechanistic Investigation of the Hydrosilylation Reaction of Aldehydes by a Bisacetonitrile Cationic Rhenium(III) Complex: A Computational Study ...... 44 3.1 Abstract ...... 45 3.2 Introduction ...... 45 3.3 Results and Discussion ...... 49 3.4 Conclusions ...... 59 3.5 Experimental Section ...... 60 3.6 References ...... 61 Chapter 4: Reactivity of Novel DAAm Rhenium(III) Acetate Complex Towards Lewis Acidic Boron Reagents ...... 64 4.1 Abstract ...... 65 4.2 Introduction ...... 65 4.4 Conclusions ...... 83 4.5 Experimental Section ...... 84 4.6 References ...... 91 Chapter 5: Synthesis and Characterization of (DAAm)ReIII(CO)(R) Complexes, and Study of CO Insertion into Rhenium(III)-Carbon Bond ...... 93

5.1 Abstract ...... 94 5.2 Introduction ...... 94 5.3 Results and Discussion ...... 97 5.5 Conclusions ...... 115 5.6 Experimental Section ...... 116 5.7 References ...... 122

LIST OF TABLES

Table 2.1. Catalytic hydrosilylation of benzaldehyde according to equation 1 ...... 18 Table 2.2. Catalytic hydrosilylation of benzaldehyde with various silane substrates ...... 19 Table 2.3. Catalytic hydrosilylation of benzaldehyde according to equation 3 ...... 20 Table 2.4. Catalytic hydrosilylation of benzaldehyde with various silane substrates under neat conditions ...... 21 Table 2.5. Selected Crystallographic Data and Collection Parameters for 5a, 6b, and 8...... 41 Table 3.1 Comparison of the calculated DFT (B3PW91-D3) bond lengths (Å) for 4a with the X-ray crystallography ...... 50 Table 3.2 Comparison of the calculated DFT (B3PW91-D3) bond angles (°) for 4a with the X-ray crystallography.………...………………………………………………………. 51 Table 4.1 Selected Crystallographic Data and Collection Parameters for 17 and 18...... 90 Table 4.2 Selected Crystallographic Data and Collection Parameters for 19 and 21...... 90 Table 5.1 Results for the CO Insertion Reaction of (DAAm)ReIII(CO)(R) complexes ...... 104 Table 5.2 Results for the CO Insertion Reaction with 26 ...... 112 Table 5.3 Selected Crystallographic Data and Collection Parameters for 23a and 23b...... 121 Table 5.4 Selected Crystallographic Data and Collection Parameters for 25 and 26...... 121

LIST OF FIGURES

Figure 2.1. 1H NMR spectrum of cationic rhenium(III) bisacetronitrile complex

6a at room temperature in acetonitrile-d3 ...... 12 -1 Figure 2.2. FTIR spectrum (KBr Pellets cm ) of [(DAAm)Re(CO)(NCCH3)2][BF4]

(aryl = C6F5) (6a)…………..…………………………………………………………………… 13 Figure 2.3. X-ray structure of 5a…………………………...…………………………………... 14 Figure 2.4. X-ray structure of 6b ...... 14 Figure 2.5. Thermal ellipsoid plot of 8 ...... 16 1 Figure 2.6. H NMR of DAmA rhenium(III) dichloride complex (8) in acetonitrile-d3 ...... 17 Figure 2.7. Time profile for the catalytic hydrosilylation of benzaldehyde with dimethylphenylsilane and 2-bromomesitylene using 6a as the catalyst...... 23 Figure 2.8. Rate law dependence on rhenium concentration. Plot kobs (s-1) versus Rhenium concentration (mM) ...... 25 Figure 2.9. Kinetic plot of the concentration of benzaldehyde vs reaction time (s)...... 26 Figure 2.10. Hammett plot for the competition experiments between para-substituted benzaldehyde ...... 30 Figure 2.11. Eyring plot for the hydrosilylation of benzaldehyde with 6a ...... 31 Figure 2.12. Calibration curve to calculate the concentration of benzaldehyde by Gas Chromatography (GC) ...... 39 Figure 3.1 X-ray crystallography determined structure of 4a ...... 49 Figure 3.2 B3PW91-D3 calculated structure of intermediate 13 ...... 52 Figure 3.3 B3PW91-D3 calculated structure of intermediate 14 ...... 53 Figure 3.4 B3PW91-D3 calculated structure of TS4 ...... 55 Figure 3.5 B3PW91-D3 calculated structure of ion pair 15 ...... 56 Figure 3.6 B3PW91-D3 calculated structure of TS5 ...... 57 Figure 3.7 B3PW91-D3 calculated structure of 16 ...... 58 Figure 3.8 DFT (B3PW91-D3) calculated reaction progress for the hydrosilylation reaction of benzaldehyde by cationic rhenium(III) complex...... 59 Figure 4.1 Thermal ellipsoid plot of 17 ...... 68 1 Figure 4.2 H NMR (400 MHz, CD2Cl2) spectrum of

[(DAAm)Re(O)(OC(CH3)OB(C6F5)3] (17)…………………………………………………….. 69 Figure 4.3 X-ray structure of 18 ...... 70

vii

Figure 4.4 Decomposition of complex 17 monitored by 1H NMR spectroscopy overtime...... 72 Figure 4.5 X-ray structure of 19 ...... 73 1 Figure 4.6 H NMR (300 MHz, C6D6) spectrum of [(DAAm-C6F5)Re(CO)(C6F5)] (19)...... 74 19 Figure 4.7 F NMR (376 MHz, C6D6) spectrum of tris(pentafluorophenyl)borane monohydrate (20) ...... 76 1 Figure 4.8 H NMR (300 MHz, C6D6) spectrum of

[DAAmRe(CO)(OB(C6F5)2OC(CH3)O] (21) ...... 78 Figure 4.9 X-ray structure of complex 21 ...... 79 19 Figure 4.10 F NMR (376 MHz, C6D6) spectrum of bispentafluoroborinic acid (22)...... 81 1 Figure 5.1 H NMR spectrum of complex 23a in methylene chloride-d2 ...... 98 Figure 5.2 Thermal ellipsoid plot of 23a ...... 99 1 Figure 5.3 H NMR spectrum of complex 23b in methylene chloride-d2 ...... 100 Figure 5.4 Thermal ellipsoid plot of 23b ...... 101 1 Figure 5.5 H NMR spectrum of complex 24 in methylene chloride-d2 ...... 102 1 Figure 5.6 H NMR spectrum of complex 25 in methylene chloride-d2 ...... 102 Figure 5.7 Thermal ellipsoid plot of 25 ...... 103 Figure 5.8 Spectral comparison: (a) 1H NMR spectrum of the crude reaction of 23b 1 with 60 psi CO at 80 °C after 30 min in CD2Cl2. (b) H NMR spectrum of complex

23b in CD2Cl2 ...... 106 Figure 5.9 Integrated 1H NMR spectrum of the reaction of 23b with 60 psi CO at 80 °C after 30 minutes ...... 107 Figure 5.10 FTIR spectrum of the acyl-product ...... 108 Figure 5.11 FTIR spectrum of the complex 23b ...... 109 Figure 5.12 1H NMR spectrum of complex 26 in methylene chloride-d2 ...... 110 Figure 5.13 Thermal ellipsoid plot of 26 ...... 111 Figure 5.14 Time Profile for the disappearance of 23b ...... 113

viii

LIST OF SCHEMES

Scheme 1.1 Variety of rhenium complexes synthesized in the Ison group...... 3 Scheme 2.1. Catalytic comparison between neutral rhenium(V) complex and cationic rhenium(V) complexes for hydrosilylation reaction by Abu Omar group...... 9 Scheme 2.2. Catalytic competency of DAAm rhenium(III) complex versus rhenium(V) complexes in hydrosilylation reactions by Ison group...... 10 Scheme 2.3. Synthesis of novel cationic rhenium(III) bisacetronitrile derivatives ...... 11 Scheme 2.4. Synthesis of DAmA rhenium(III) dichloride complex 8...... 15 Scheme 2.5. Substrate scope using the optimized conditions according to equation 5...... 22 Scheme 2.6. Stoichiometric reaction of 6a and dimethylphenylsilane...... 27 Scheme 2.7. Proposed mechanism for the formation of 9 from 3a by Ison group...... 28 Scheme 2.8. Proposed mechanism for the formation of 9 from 6a...... 28 Scheme 2.9. Stoichiometric reaction of 6a and benzaldehyde...... 29 Scheme 2.10. Competition experiments between para-substituted benzaldehyde...... 29 Scheme 2.11. Variable temperature kinetic experiments...... 31 Scheme 2.12. Proposed mechanism for the hydrosilylation reaction of aldehydes catalyzed by 6a...... 32 Scheme 3.1 Hydride mechanisms for the hydrosilylation reaction of carbonyl compounds...... 46 Scheme 3.2 Proposed mechanism for catalytic ionic hydrosilylation by Bullock et al...... 47 Scheme 3.3 Catalytic Hydrosilylation Reaction of Aldehydes by 6a...... 47 Scheme 3.4 Proposed mechanism for the hydrosilylation reaction of aldehydes catalyzed by 6a...... 48 Scheme 4.1 Synthesis of DAAm rhenium(III) acetate complexes 3a and 3b previously reported by our group...... 65 Scheme 4.2 Synthesis of Lewis acid/base adducts previously reported by the Ison group...... 66 Scheme 4.3 Preliminary investigation of the reactivity of 3a towards Lewis acidic boron reagents...... 67

Scheme 4.4 Synthesis of [(DAAm)Re(O)(OC(CH3)OB(C6F5)3)], 17...... 69 Scheme 4.5 Attempts to reproduce previous results by our group...... 70 Scheme 4.6 Proposed reaction pathway for the formation of 18...... 71

Scheme 4.7 Synthesis of [(DAAm-C6F5)Re(CO)(C6F5)] (aryl = C6F5), 19...... 73

Scheme 4.8 Synthesis of [(DAAm-C6F5)Re(CO)(C6F5)], 19...... 74

Scheme 4.9 Proposed decarbonylation for the formation of 18 from 19...... 75 Scheme 4.10 Synthesis of tris(pentafluorophenyl)borane monohydrate, 20...... 75

Scheme 4.11 Reactivity of 19 with [H2O·B(C6F5)3] ...... 76 Scheme 4.12 Reactivity of 19 in the presence of oxygen ...... 77 Scheme 4.13 Probing 20 as the active species in the reaction of 3a towards Lewis acidic boric reagents...... 77 Scheme 4.14 Proposed mechanism pathway for the transformation of 21 to 19...... 80 Scheme 4.15 Synthesis of bis(pentafluorophenyl)borinic acid, 22...... 81 Scheme 4.16 Probing 22 as the active species in the reaction of 3a towards Lewis acidic boron reagents...... 82 Scheme 4.17 Equilibrium between the κ1 and κ2 isomers of 3a...... 82 Scheme 4.18 Proposed transformation of 3a’ in the presence of the conjugated base - [HOB(C6F5)3] to lead 21...... 83 Scheme 4.19 Intermediates identified in this investigation and proposed mechanism for each transformation...... 83 Scheme 5.1 Synthesis of DAAm rhenium(III) acetate complexes 3a and 3b...... 95 Scheme 5.2 Carbonylation Reaction of DAAm Oxorhenium(V) Complexes by the Ison Group ...... 95 Scheme 5.3 Carbonylation Reaction of DAP Oxorhenium(V) Complex by the Ison Group ...... 96 Scheme 5.4 Carbonylation Reaction of SSS Oxorhenium(V) Complex by the Ison Group ...... 96 Scheme 5.5 Carbonylation Reaction of PNP Nitridorhenium(V) Complex by the Ison Group ...... 96 Scheme 5.6 Synthesis of DAAm rhenium(III) derivatives ...... 97 Scheme 5.7 General CO insertion reaction with (DAAm-aryl)ReIII(CO)(R) complexes...... 104

Scheme 5.8 Synthesis of complex (DAAm-Mes)Re(O)(CH2Ph), 26...... 109 V Scheme 5.9 General CO insertion reaction with (DAAm-Mes)Re (O)(CH2Ph)(26) complex...... 111 Scheme 5.10 General Mechanism for Direct CO Insertion...... 113 Scheme 5.11 General Mechanism for CO Adduct formation...... 114 Scheme 5.12 Proposed CO Adduct formation mechanism for 23b...... 114

Scheme 5.13 Proposed Direct CO Insertion mechanism for 23b...... 114

Chapter 1: General Introduction

Department of Chemistry, North Carolina State University Raleigh, North Carolina, 27695-8204

1.1 Organometallic Chemistry

Organometallic chemistry is the field that combines inorganic and organic chemistry. It involves the interaction of a metal center surrounded by an organic ligand to generate an organometallic complex. Transition metal complexes have been shown to be of great importance in the catalytic conversion of various chemical transformations, increasing reactivity, selectivity, and creating new pathways for the formation of organic molecules.1,2 Unlike typical metal complexes, organometallic complexes are often easily prepared and have a good solubility in a variety of solvents, which makes them attractive species for homogenous catalysis. Also, their contribution to the field of green chemistry includes their capacity to: increase reaction rates, efficiency, selectivity of reactions, maximize atom economy and minimize waste of synthetic processes.3 Rhenium catalysts are attractive due to their unique properties. One of the benefits is that they possess properties of both early and late transition metals. The chemical properties of rhenium resemble the metals in the manganese group (Group 7) of the Periodic Table. However, the physical properties of rhenium are much more similar to those of the refractory metals of Groups 5 and 6, particularly molybdenum and tungsten. Also, it has access to a variety of oxidation states from -1 to +7. Rhenium easily change from one valence to another, a property which makes it ideal for use as a catalyst.4 Recently, our group reported the synthesis of a series of rhenium complexes that exhibit catalytic activity and the mechanisms of these transformations were described.5-18 Examples of rhenium(V) and rhenium(III) complexes synthesized in the Ison group are shown in Scheme 1.1. The stability and reactivity of organometallic compounds are associated with the nature of the ligands coordinated to the metal center of the complex. Tridentate ligands like DAAm

(diamidoamine), DAP (diamidopyridine) and PNP ([2-P(CHMe2)2-4-MeC6H3]2N) are of interest and attractive due to the ability to tune the electronic and steric properties at different sites on the ligand framework.

Scheme 1.1 Variety of rhenium complexes synthesized in the Ison group.

Oxo (O2-) Nitrido (N3-) Re(III) -Mes, -C6F5

R Alkyl (-R) R Re(V) CO X R Aryl (-Ph) Alkyl (-R) L Hydride (-H) R N Aryl (-Ph) Re Re R Halogens (-X) L L L R N

Tridentate Ligand Tridentate Ligand mono or dianionic chelate -Mes, -C6F5 mono or dianionic chelate (DAAm, DAP, SSS, PNP) -Dipp, -Dbhp, -iPr (DAAm, DAmA)

The use of rhenium complexes as catalysts has numerous advantages. Rhenium complexes have a large functional group tolerance,1,2 and the ability to tolerate a variety of ligands. Rhenium can generally form various stable complexes, which is beneficial for the study of reaction mechanisms. The synthesis of these complexes serves as an inexpensive and air and moisture stable alternative to low-valent late transition metals. These characteristics make the development of new rhenium complexes a promising option for current organometallic systems. Rhenium catalysts have been the subject of many studies, and subsequently, a variety of rhenium complexes bearing diverse ligand frameworks have been synthesized by many research groups.19

1.2 Scope of this dissertation

The work in this thesis explores the reactivity of rhenium(III) complexes. As shown in Scheme 1.1 (right) a variety of rhenium(III) complexes can be synthesized. The goal of this dissertation is to study new rhenium(III) complexes and discover potential catalytic applications based on experimental observations and data analysis. These reactivity studies will help to tune, improve and understand the behavior of these organometallic complexes. The majority of the investigations in this dissertation are based on the reactivity of [(DAAm-aryl)Re(CO)(OAc)] (aryl

= C6F5, Mes) complexes and their analogues. Previous studies on rhenium(III) acetate complexes helped us to improve and design a better and more efficient cationic DAAm rhenium(III) bisacetonitrile catalyst for the hydrosilylation reaction of aldehydes (Chapter 2). The mechanism for this reaction is explored computationally utilizing Density Functional Theory (DFT) and is presented in Chapter 3. The reactivity of DAAm-C6F5 rhenium(III) acetate complex with Lewis acids is investigated in Chapter

4. These studies yield interesting findings regarding unique C6F5 group transfer and the role of water in these transformations as an oxidizing agent. Our group has studied the reactivity of high valent rhenium(V) mono-oxo and nitrido complexes bearing tridentate ligands for the activation of CO. New DAAm rhenium(III) alkyl/aryl complexes were isolated and characterized. With these complexes in hand, their reactivity towards CO insertion was studied and compared to their rhenium(V) analogues (Chapter 5).

1.3 References

(1) Hartwig, J. F. Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 2010. (2) Crabtree, R. H. The Organometallic Chemistry of Transition Metals; John Wiley & Sons, Inc.: Hoboken, NJ, 2009. (3) Lancaster, M., Green Chemistry: An Introductory Text. The Royal Society of Chemistry: Cambridge, UK, 2002. (4) Millensifer, T.; Sinclair, D.; Jonasson, I.; Lipmann, A. Rhenium. In Critical Metals Handbook; Gunn, G., Ed.; Wiley: New York, 2014; p 340. (5) Feng, Y.; Aponte, J.; Houseworth, P. J.; Boyle, P. D.; Ison, E. A. Inorg. Chem., 2009, 48 (23), 11058-11066. (6) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc., 2011, 133, 13288-13291. (7) Lilly, C. P; Boyle, P. D.; Ison, E. A. Dalton Trans., 2011, 40, 11815-11821. (8) Lilly, C. P; Boyle, P. D.; Ison, E. A. Organometallics, 2012, 31, 4295-4301. (9) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. Organometallics. 2012, 31, 5994–5997. (10) Smeltz, J. L.; Lilly, C. P.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc. 2013, 135, 9433- 9441. (11) Robbins, L. K.; Lilly, C. P.; Smeltz, J. L.; Boyle, P. D.; Ison, E. A. Organometallics 2015, 34, 3152-3158. (12) Lambic, N. S.; Sommer, R. D.; Ison, E. A. J. Am. Chem. Soc., 2016, 138, 4832-42 (13) Lambic, N. S.; Lilly, C. P.; Robbins, L. K.; Sommer, R. D.; Ison, E. A. Organometallics 2016, 35, 2822-2829. (14) Lambic, N. S.; Lilly, C. P.; Sommer, R. D.; Ison, E. A. Organometallics 2016, 35(17), 3060-3068. (15) Pérez, D. E.; Smeltz, J. L.; Sommer, R. D.; Boyle, P. D.; Ison, E. A. Dalton Trans. 2017, 46(14), 4609-4616. (16) Lambic, N. S.; Sommer, R. D.; Ison, E. A. ACS Catalysis 2017, 7, 1170-1180. (17) Lambic, N. S.; Brown, C. A.; Sommer, R. D.; Ison, E. A. Organometallics 2017, 36, 2042-2051. (18) Lambic, N. S.; Sommer, R. D.; Ison, E.A. Dalton Trans. 2018, 47, 758-768. (19) (a) Owens, G. S.; Aries, J.; Abu-Omar, M. M. Catal. Today 2000, 55, 317-363; (b) Arias, J.; Newlands, C. R.; Abu-Omar, M. M. Inorg. Chem. 2001, 40, 2185-2192; (c)

Gangopadhyay, J.; Sengupta, S.; Bhattacharyya, S.; Chakraborty, I.; Chakravorty, A. Inorg. Chem. 2002, 41, 2616-2622; (d) Koshino, N.; Espenson, J. H. Inorg. Chem. 2003, 42, 5735-5742; (e) Du, G.; Abu-Omar, M. M. Curr. Org. Chem. 2008, 12, 1185-1198; (f) Ison, E. A.; Cessarich, J. E.; Du, G.; Fanwick, P. E.; Abu-Omar, M. M. Inorg. Chem. 2006, 45, 2385-2387; (g) Du, G.; Fanwick, P. E.; Abu-Omar, M. M. Inorg. Chim. Acta 2008, 361, 3184-3192; (h) Sousa, S. C.; Cabrita, I.; Fernandes, A. C. Chem. Soc. Rev. 2012, 41, 5641-5653; (i) Ahmad, I.; Chapman, G.; Nicholas, K. M. Organometallics 2011, 30, 2810- 2818; (j) Wang, Y.; Espenson, J. H. Org. Lett. 2000, 2, 3525-3526; (k) Schröckeneder, A.; Traar, P.; Raber, G.; Baumgartner, J.; Belaj, F.; Mösch-Zanetti, N. C. Inorg. Chem. 2009, 48, 11608-11614; (l) Kirillov, A. M.; Haukka, M.; Kirillova, M. V.; Pombeiro, A. J. L. Adv. Synth. Catal. 2005, 347, 1435-1446; (m) Abu-Omar, M. M.; Appelman, E. H.; Espenson, J. H. Inorg. Chem. 1996, 35, 7751-7757; (n) Abu-Omar, M. M.; Espenson, J. H. Organometallics 1996, 15, 3543-3549; (o) Abu-Omar, M. M.; Khan, S. I. Inorg. Chem. 1998, 37, 4979-4985; (p) Du, G.; Abu-Omar, M. M. Organometallics 2006, 25, 4920-4923; (q) Ziegler, J. E.; Zdilla, M. J.; Evans, A. J.; Abu-Omar, M. M. Inorg. Chem. 2009, 48, 9998-10000.

Chapter 2: Cationic Rhenium(III) Complexes: Synthesis, Characterization, and Reactivity for Catalytic Hydrosilylation Reaction of Aldehydes

Portions of this chapter were published in: Pérez, D. E.; Smeltz, J. L.; Sommer, R. D.; Boyle, P. D.; Ison, E. A. Dalton Trans., 2017, 46, 4609-4616

Department of Chemistry, North Carolina State University Raleigh, North Carolina, 27695-8204

2.1 Abstract

The catalytic efficiency for the hydrosilylation of benzaldehyde with organosilanes with the cationic rhenium(III) complex [(DAAm-aryl)Re(CO)(NCCH3)2][BF4] (DAAm = N,N-bis(2- arylaminoethyl)methylamine; aryl = C6F5) (6a) has been demonstrated. In addition, the catalytic competency of the complexes [(DAAm-aryl)Re(CO)(NCCH3)2][BF4] (DAAm = N,N-bis(2- arylaminoethyl)methylamine; aryl = Mes) (6b) and (DAmA-aryl)Re(CO)(Cl)2 (DAmA = N,N- bis(2-arylamineethyl)methylamino; aryl = C6F5) (8) were compared with 6a. Data suggest that electron-withdrawing substituents at the diamidoamine ligand increases the catalytic reactivity of the complexes. Various aliphatic and aromatic aldehydes were tested. Excellent yields were achieved at ambient temperature in neat conditions using dimethylphenylsilane. The reaction affords TONs of up to 9,200 and a TOF of up to 126 h-1. Kinetic and mechanistic studies were performed and possible reaction mechanisms for the hydrosilylation of benzaldehyde are proposed.

2.2 Introduction

Hydrosilylation reactions are useful due to the production of a protected alcohol in a single step under mild conditions.1 Thus, hydrosilylation is more attractive than the conventional two- step methodology of reduction by a metal hydride followed by silyl protection. In recent years several examples of mono-oxorhenium, dioxorhenium, and mono-oxomolybdenum complexes were shown to be highly active for the hydrosilylation of aldehydes and ketones. Abu-Omar and coworkers have shown several examples where cationic rhenium(V) mono-oxo complexes exhibit higher reactivity towards catalytic hydrosilylation reactions than neutral rhenium(V) complexes (Scheme 2.1). 2-4

Scheme 2.1. Catalytic comparison between neutral rhenium(V) complex and cationic rhenium(V) complexes for hydrosilylation reaction by Abu Omar group.

OSiEt O 5 mol% [Re] 3

1.5 equiv HSiEt3 ,CD2Cl2, rt H

n O N N Re O O B(C6F5)4 O O B(C6F5)4 N N O L O Re Re O O N O L O O N Re N N O L = H2O or CH3CN n n = 0 or 1

NO REACTION 100% conversion

However, recently our group reported a rhenium(III) complex that is significantly more active that its rhenium mono-oxo precursors. The results show that the rhenium(III) complex 3 is more reactive towards the hydrosilylation of benzaldehydes than rhenium(V) complexes 1 and 2 (Scheme 2.2) under similar reaction conditions.5 We hypothesized that when oxorhenium complexes are employed as catalysts, the active species are not high-valent rhenium intermediates, but species that are generated upon deoxygenation with the organosilane. Just a few examples of low-valent rhenium complexes not bearing an oxo ligand have been shown to be efficient catalysts for hydrosilylation reaction of carbonyl compounds.6,7

Scheme 2.2. Catalytic competency of DAAm rhenium(III) complex versus rhenium(V) complexes in hydrosilylation reactions by Ison group.

O OSiMe2Ph 0.1 mol% [Re] H H 1.2 equiv HSiMe2Ph, neat, 80 °C H

C F C F C6F5 6 5 CO 6 5 O O CH3 N N O C6F5 N Re Re Re OAc N N N C6F5 N N C6F5 N

1 2 3

3 d, 94 % 2 d, 98 % 2.5 h, 95 %

In this chapter we report the synthesis and catalytic competency of low-valent cationic DAAm rhenium(III) complexes for the hydrosilylation reaction of aldehydes. We demonstrate that novel cationic DAAm rhenium(III) complexes are more efficient hydrosilylation catalysts than the neutral DAAm rhenium(III) and DAAm rhenium(V) complexes previously reported in the literature by our group. We examine how changing the electronics and sterics of the aryl substituents on the diamidoamine nitrogens affects the catalytic reactivity of the complex. Also, in this chapter we reported the synthesis of a novel diamineamido (DAmA) ligand and investigated how the hybridization in one of the nitrido in the ligand framework affect the reactivity of the complex towards hydrosilylation.

2.3 Results and Discussion

2.3.1 Synthesis and Characterization of Cationic DAAm Rhenium(III) Complexes and Cationic DAmA Rhenium(III) Complex

The synthesis of cationic DAAm rhenium(III) complexes by the reaction of 3 with acids bearing non-coordinating anions in acetonitrile was investigated. A series of cationic [(DAAm- F aryl)Re(CO)(NCCH3)2][X] (aryl = C6F5, Mes; X = OTf, BAr , BF4, PF6) were synthetized from the novel DAAm rhenium(III) complexes 3a and 3b (Scheme 2.3). The corresponding DAAm rhenium(III) acetate complex was treated with 1 equiv of the non-coordinating acid at ambient

10 temperature. A color change from brown (3a) to yellow, or dark red (3b) to light orange, was observed immediately after the addition of acetonitrile to the reaction mixture, which was indicative of the cationic rhenium(III) bisacetonitrile complex formation.

Scheme 2.3. Synthesis of novel cationic rhenium(III) bisacetronitrile derivatives

R CO R CO X

R N 1 equiv HX R N NCCH3 Re OAc Re NCCH3 + HOAc N CH3CN, rt, 5 min N N N

(3a) R = C6F5 (4a) R = C6F5, X = OTf; (4b) R = Mes, X = OTf F F (3b) R = Mes (5a) R = C6F5, X = BAr 4; (5b) R = Mes, X = BAr 4 (6a) R = C6F5, X = BF4; (6b) R = Mes, X = BF4 (7a) R = C6F5, X = PF6; (7b) R = Mes, X = PF6

The representative 1H NMR spectrum of 6a is shown in Figure 2.1. The methylene protons from the ligand backbone resonate as 3 distinct multiplets at δ 4.08 (2H), δ 3.73 (2H) and δ 3.04 (4H) ppm. A singlet at δ 2.89 ppm (3H) is representative of the methyl group on the amine from the ligand backbone. The singlet corresponding to the methyl groups from the acetonitrile ligands resonate at δ 2.00 ppm (6H).

Figure 2.1. 1H NMR spectrum of cationic rhenium(III) bisacetronitrile complex 6a at room temperature in acetonitrile-d3.

The FTIR spectrum of 6a exhibited a carbonyl band at 1882 cm-1 (Figure 2.2). This stretch

-1 5,8 is at higher frequency than the two found in 3a (�CO 1868 and 1876 cm ). The electrophilic nature of the cationic complex decreases the p back-bonding consequently increasing the stretching frequency of the carbonyl ligand in the FTIR spectrum.

-1 Figure 2.2. FTIR spectrum (KBr Pellets cm ) of [(DAAm)Re(CO)(NCCH3)2][BF4] (aryl = C6F5) (6a).

Isolation and crystallization of complexes 5a (Figure 2.3) and 6b (Figure 2.4) confirmed the formation of the DAAm rhenium(III) bisacetronitrile cationic complex. The geometry about rhenium in both complexes is best described as a distorted octahedron with the carbonyl ligand and amine nitrogen occupying the axial plane. The X-ray structures show that the acetate ligand is replaced by two acetonitrile molecules in the equatorial plane of the complexes. The Re1-N4 (5a, 2.168 Å; 6b, 2.152 Å) and Re1-N5 (5a, 2.120 Å; 6b, 2.139 Å) bond lengths are comparable with typical cationic Re-acetonitrile bond lengths.6,9,10

Figure 2.3. X-ray structure of 5a. Thermal ellipsoids are at 50%. H atoms and counter ion have been omitted and the pentafluorophenyl groups on the ligand are shown in wireframe format for clarity. Selected bond lengths (Å) and angles (deg): Re1-C1, 1.867(4); Re1- N1, 1.932(3); Re1- N2, 2.220(3); Re1-N3, 1.928(1); Re1-N4, 2.167(3); Re1-N5, 2.120(2); N1-Re1-N3, 105.04(15); N1-Re1-N2, 80.79(13); N1-Re1-C1, 94.26(16); N2-Re1-C1, 172.91(16); N1-Re1-N4, 85.09(13); N1-Re1-N5, 160.96(15); Re1-N1-C8, 119.2(2); Re1-N3-C12, 116.9(3).

Figure 2.4. X-ray structure of 6b. Thermal ellipsoids are at 50%. H atoms and counter ion have been omitted for clarity. The mesitylene substituents in the diamido ligand framework are depicted in wireframe for clarity. Selected bond lengths (Å) and angles (deg): Re1-C24, 1.864(3); Re1-N1, 1.933(2); Re1-N2, 2.244(2); Re1-N3, 1.942(2); Re1-N4, 2.139(2); Re1-N5, 2.152(2); N1-Re1-N3, 108.48(9); N1-Re1-N2, 81.13(8); N1-Re1-C24, 93.44(10); N2-Re1-C24, 173.48(9); N1-Re1-N4, 88.32(9); N1-Re1-N5, 159.41(9).

The synthesis of 4a-b, 5a-b, 6a-b, and 7a-b provided a diverse set of cationic rhenium(III) complexes bearing either electron withdrawing (R = C6F5) or electron donating (R = Mes) analogues of the DAAm ligand framework. Some of the benefits of this series of cationic rhenium(III) complexes include: (1) the rhenium(III) precursors are air and moisture stable; (2) the synthesis is facile; (3) different non-coordinating acids can be used; and (4) a direct comparison of the reactivity could be made based on the electronics of the ligand framework. From previous reports, we hypothesized that in order to design a highly active rhenium catalyst for hydrosilylation reactions, the metal center must have an oxidation state number of three combined with a cationic charge. The main two characteristics necessary for a more efficient catalyst are present in these novel cationic DAAm rhenium(III) complexes. We proposed that cationic DAAm rhenium(III) complexes will be more efficient hydrosilylation catalysts than its precursors and the cationic rhenium(V) complex reported by Abu-Omar.2-4 The DAAm analogue, diamineamido (DAmA) ligand was also synthetized. Complex 3a was treated with an excess of hydrochloric acid (HCl) in methylene chloride at room temperature (Scheme 2.4). X-ray crystallography of the green crystals confirmed the formation of 8 in solution (Figure 2.5). The geometry about rhenium in 8 is best described as a distorted octahedral. The acetate ligand is protonated and replaced by two chloride ligands in the equatorial plane of the complex. X-ray studies also provided evidence for the protonation of one amide nitrogen and a change in hybridization at nitrogen from sp2 to sp3. Evidence for this protonation and the change in hybridization from sp2 to sp3 is provided by the analysis of the angles around N1 and N3 in 8. A comparison between 3a and 8 reveals a decrease in the Re-N1-C8 angle from 118° to 108°. Also, in 8 the Re-N1 bond length (2.2 Å) is longer than Re-N3 bond length (1.92 Å).

Scheme 2.4. Synthesis of DAmA rhenium(III) dichloride complex 8.

C6F5 CO C6F5 H CO Cl C6F5 N C6F5 N Re OAc excess HCl Re Cl N CH2Cl2, 24 h N N N

3a 8

Figure 2.5. Thermal ellipsoid plot of 8. Thermal ellipsoids are at 50%. H atoms have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Re1-C1, 1.862(3); Re1-N1, 2.200(2); Re1- N2, 2.262(2); Re1-N3, 1.920(2); Re1-Cl1, 2.4361(6); Re1-Cl2, 2.3685(6); N1-Re1-N3, 93.97(8); N1-Re1-N2, 78.18(7); N1-Re1-C1, 98.15(9); N2-Re1-C1, 174.44(9); N1-Re1-Cl1, 80.01(6); N1- Re1- Cl2, 163.79(5); Re1-N1-C8, 107.95(14); Re1-N3-C12, 120.44(16).

The 1H NMR spectrum of 8 is shown in Figure 2.6. In the 1H NMR spectrum of 8 a singlet at 3.36 ppm is observed for the methyl group on the amine nitrogen. The methylene protons from the ligand backbone are observed as three distinct multiplets at δ 4.16 (2H), δ 3.45 (2H) and δ 3.31 (4H) ppm. The amine proton was not observed by 1H NMR spectroscopy either at room temperature or -60 °C indicating that the proton exchange cannot be observed on the NMR time scale.

CH3

C6F5 H CO Cl C6F5 N Re Cl N N CH H 3 H H H 2 x CH2

CH2 CH2

1 Figure 2.6. H NMR of DAmA rhenium(III) dichloride complex (8) in acetonitrile-d3.

2.3.2 DAAm Rhenium(III) Complexes for the Catalytic Hydrosilylation of Aldehydes

The catalytic competency of complexes 6a, 6b, and 8 for the hydrosilylation of benzaldehyde with dimethylphenylsilane were examined according to Equation 1. Reactions were conducted at different temperatures and catalyst loadings in acetonitrile. The results are summarized in Table 2.1.

Table 2.1. Catalytic hydrosilylation of benzaldehyde according to equation 1.

O OSiMe2Ph x mol% [Re] H H (1) H 1.2 equiv HSiMe2Ph, CH3CN, T

Entrya [Re] mol% T (°C) time (h) %Conversionb TON

1c 8 1 80 95 79 79

2d 6a 0.1 rt 165 95 1125

3d 6a 0.1 40 118 41 513

4d 6a 0.1 60 48 46 575

5d 6a 0.1 80 48 49 613

6e 6a 0.5 rt 8 95 190

7d 6b 0.1 rt 24 NR -

8d 6b 0.1 40 34 NR - a General Reaction Conditions: Under a N2 atmosphere, 2 mg of catalyst, 1.2 equiv silane, and 0.5 b 1 equiv 2-bromomesitylene (internal standard) in 0.35 mL CH3CN. Calculated by H NMR spectroscopy of the crude reaction against 2-bromomesitylene as the internal standard. cRe (0.003 mmol), and aldehyde (0.3 mmol). dRe (0.003 mmol), and aldehyde (3 mmol). eRe (0.003 mmol), and aldehyde (0.6 mmol). TON = turnover number. NR = no reaction. rt = room temperature.

Complex 6a is the most efficient catalyst in the series, 95% conversion was attained at room temperature with 0.1 mol% catalyst loading achieving a turnover number (TON) of 1125 (Entry 2). Higher temperatures did not improve the catalytic competency of 6a (Entries 3-5). The low percent conversion at high temperatures is attributable to catalyst decomposition. Complex 8 is not as active as 6a (Entry 1) and the reaction must be heated due to the poor solubility of the complex in the solvent. This suggests that changing the hybridization in one of the amido nitrogens in the ligand framework has a notable effect on the reactivity of the complex. Hydrosilylation was not observed when 6b was utilized as catalyst (Entries 7 and 8). The results demonstrate that the electronics of the diamidoamine ligand dramatically affects the reactivity of the complex. We hypothesized that electron-donating groups decrease the electrophilicity and stabilize the metal

18 center making it less susceptible to react and activate the silane reagent. The effect of various silane substrates in the catalytic hydrosilylation of benzaldehyde was investigated according to Equation 2. Reactions were conducted at room temperature in acetonitrile. The results are summarized in Table 2.2.

Table 2.2. Catalytic hydrosilylation of benzaldehyde with various silane substrates.

O OSiR3 x mol% 6a H H (2) 1.2 equiv HSiR3, CH3CN, rt H

Entrya Silane mol% time (h) %Conversionb

c 1 HSiMe2Ph 0.1 165 95

c 2 HSiPh3 0.1 114 NR

d 3 HSiEt3 0.5 48 51

d 4 HSiPh3 0.5 46 12 a General Reaction Conditions: Under a N2 atmosphere, 2 mg of catalyst, and 0.5 equiv 2- b 1 bromomesitylene (internal standard) in 0.35 mL CH3CN. Calculated by H NMR spectroscopy of the crude reaction against 2-bromomesitylene as the internal standard. cRe (0.003 mmol), and aldehyde (3 mmol). dRe (0.0024 mmol), and aldehyde (0.554 mmol). NR = no reaction.

Data in Table 2.2 suggest that the catalytic hydrosilylation of benzaldehyde with 6a is more efficient with dimethylphenylsilane (Entry 1). Hydrosilylation was not observed when

HSiPh3 was utilized as the substrate (Entry 2), suggesting that steric bulk present in the silane inhibited the activation of the substrate by 6a. Similar results using HSiPh3 were observed 5 previously by our group. Increasing the catalyst loading in the presences of HSiPh3 does not have any notable improvement in the conversion of benzaldehyde to the silyl ether (Entry 4). The reaction in the presence of HSiEt3 just affords a 51% conversion (Entry 3). Even when the reaction was stirred for more than 48 h, no change in conversion was observed. The catalytic reactivity in the presences of HSiEt3 is slower and less efficient in comparison to the reactivity using HSiMe2Ph as the substrate. In previous reports by our group, comparison between HSiEt3 and HSiMe2Ph, had a similar trend.5

The catalytic reactivity for the hydrosilylation of benzaldehyde with 6a was investigated and optimized under neat conditions according to Equation 3. Reactions were conducted at room temperature with varying catalyst loadings. The results are summarized in Table 2.3.

Table 2.3. Catalytic hydrosilylation of benzaldehyde according to equation 3.

O OSiMe2Ph x mol% 6a H H (3) H 1.2 equiv HSiMe2Ph, neat, rt

Entrya mol% time (h) % Conversionb TON

1c 0.01 72 92 8,586

2d 0.03 24 100 3,000

3e 0.05 16 100 2326

4f 0.10 9 100 1163

5g 0.50 2 100 233

6h 1 1 100 115 a General Reaction Conditions: 2 mg [Re], 1.2 equiv silane, under a N2 atmosphere, 0.5 equiv 2- bromomesitylene (internal standard), and under neat conditions. bCalculated by 1H NMR spectroscopy of the crude reaction against 2-bromomesitylene as the internal standard. cAldehyde (28 mmol) and Re (0.003 mmol). dAldehyde (9 mmol) and Re (0.003 mmol). eAldehyde (6 mmol) and Re (0.003 mmol). fAldehyde (3 mmol) and Re (0.003 mmol). gAldehyde (0.6 mmol) and Re (0.003 mmol). hAldehyde (0.7 mmol) and Re (0.007 mmol).

The catalytic efficiency of 6a drastically improved when the reaction was performed in dimethylphenylsilane as the solvent. Complex 6a quantitatively converts benzaldehyde to the silyl ether product in less than 24 h using low catalyst loadings at room temperature (Entries 2-6). Also, 6a is able to afford TONs of up to 8,586 (Entry 1) using 0.01 mol% catalyst loading without any additional solvent at room temperature. In contrast, the catalytic hydrosilylation reaction using complex 3a must be heated to afford comparable results. At room temperature 6a (0.03 mol%) is a more efficient hydrosilylation catalyst than its precursor 3a (0.01 mol%).5

The effect of various silane substrates on the catalytic hydrosilylation of benzaldehyde was investigated according to Equation 4. The reaction was carried at room temperature under neat conditions at the same catalyst loading. The experimental results are summarized in Table 2.4.

Table 2.4. Catalytic hydrosilylation of benzaldehyde with various silane substrates under neat conditions.

O OSiR3 0.03 mol% 6a H H (4) 1.2 equiv HSiR3, neat, rt H

Entrya Silane time (h) %Conversionb

1 HSiMe2Ph 24 100

2 HSi(OEt)3 18 100

3 HSiMePh2 240 71

4 HSi(Et)3 240 30 a General Reaction Conditions: Re (0.003 mmol), and benzaldehyde (9 mmol); under a N2 atmosphere and 0.5 equiv 2-bromomesitylene (internal standard). bCalculated by 1H NMR spectroscopy of the crude reaction against 2-bromomesitylene as the internal standard.

Catalyst 6a performed efficiently for the hydrosilylation of benzaldehyde with HSiMe2Ph

(Entry 1) and HSiOEt3 (Entry 2). The reaction with HSiMePh2 (Entry 3) was slower than the reactions using HSiMe2Ph (24 h) or HSiOEt3 (18h), and 71% conversion was observed over 240 h. We attribute this to steric effects of the two benzene rings in the substrate. Our group observed similar effects when experimental results for the reaction using HSiMePh2 were compared to 5 HSiMe2Ph. Under neat conditions, hydrosilylation of benzaldehyde with HSiEt3 is slow and afforded low conversions (Entry 4). The different reactivity of HSiOEt3 (Entry 2) and HSiEt3 (Entry 4) is notable. This suggests that a more electron-donating silane is favorable for the hydrosilylation reaction. We chose 0.03 mol% catalyst loading (Table 2.3, Entry 2) as the optimized conditions to investigate the catalytic reactivity of 6a towards different aldehyde substrates. The reaction was performed according to Equation 5. The results are summarized in Scheme 2.5. The results suggest that catalyst 6a performed efficient hydrosilylation with a variety of aldehydes bearing different steric and electronic effects.

Scheme 2.5. Substrate scope using the optimized conditions according to equation 5.

O 0.03 mol% 6a OSiMe2Ph (5) 1.2 equiv HSiMe Ph, neat, rt, time R H R H 2 H

O O O O Cl H H H H Cl

24 h, 100% 24 h, 100% 18 h, 100% 24 h, 100%

O O O

H H H O (H C) N H CO 3 2 3 H OCH3

24 h, 92% 24 h, 100% 24 h, 65% 24 h, 100% H O O O O H H

24 h, 90% 48 h, 73% 24 h, 65%

The optimized reaction conditions had to be adapted based on the solubility of the aldehydes in dimethylphenylsilane. In general, most liquid aldehydes were treated under neat conditions and most solid aldehydes were dissolved in a minimum amount of acetonitrile. The percent conversion was determined by 1H NMR spectroscopy of the crude reaction mixture using 2-bromomesitylene as the internal standard. Hence, these results demonstrate that the cationic DAAm rhenium(III) complex 6a is a more effective hydrosilylation catalyst than the neutral DAAm rhenium(III) complex 3a and 3a DAAm rhenium(V) 1 and 2 precursors. Evidenced suggested that cationic DAAm rhenium(III) complex 6a even surpasses previous cationic rhenium(V) catalysts reported by the Abu-Omar2-4 group. Some of the specific benefits of this particular catalytic system include: (1) The reaction can be performed with low catalysts loading; (2) The reaction does not require high temperatures; (3) The reaction can be performed under neat conditions; (4) The catalyst precursor is air and moisture stable; and (5) The catalyst effectively performed hydrosilylation with variety aldehydes.

2.3.3 Kinetics Studies

Several kinetic experiments were performed in order to determine the dependences of each reagent involved in this reaction. The time profile for the hydrosilylation of benzaldehyde with dimethylphenysilane using catalyst 6a was investigated under pseudo-first order conditions (Figure 2.7). Product formation using 6a at different catalyst loadings (0.279 mM, 0.01 mol%; 0.829 mM, 0.03 mol%; 1.385 mM, 0.5 mol%; 2.770 mM, 1 mol%) was monitored with dimethylphenylsilane (3.324 M) and benzaldehyde (2.770 M). The reaction was performed under neat conditions at room temperature. The percent conversion of benzaldehyde was calculated by 1H NMR spectroscopy with 2-bromomesitylene as the internal standard.

Figure 2.7. Time profile for the catalytic hydrosilylation of benzaldehyde with dimethylphenylsilane and 2-bromomesitylene using 6a as the catalyst. Reaction conditions: [benzaldehyde] = 2.770 M; [silane] = 3.324 M; [2-bromomesitylene] = 1.385 M; [6a] = 0.279 mM (0.01 mol%), 0.829 mM (0.03 mol%), 1.385 mM (0.5 mol%), 2.770 (1 mol%). Reactions were performed in a vial under stirring at room temperature in a N2 atmosphere. At a fixed time, a representative aliquot from the reaction mixture was dissolved in CDCl3. The percent conversions were determined by 1H NMR spectroscopy by integrating the product peak against the internal standard.

The overall reaction dependence was obtained by monitoring product formation over time with 6a as the catalyst, and benzaldehyde and dimethylphenylsilane in approximately a 1:1 ratio. As shown in Figure 2.7, the reaction follows clean pseudo-first order kinetics with observed rate -6 -1 -5 -1 -5 -1 -4 -1 constants (kobs) of 6.2 x 10 s , 2.3 x 10 s , 4.31 x 10 s , and 1.0 x 10 s . The results suggest a rate law that is first-order overall.

x y z Rate = k[cat] [Benzaldehyde] [HSiMe2Ph]

�� y z Rate = kobs[Benzaldehyde] [HSiMe2Ph] (6) x where kobs = k[cat] From the kinetic plot the reaction is 1st order overall; i.e. Either y =1 and z = 0 or y = 0 and z = 1

The order with respect to rhenium was determined by plotting the observed rate constants

(kobs) at varying concentration of rhenium. As shown in Figure 2.8 a linear fit was obtained which suggests a first order dependence on [Re] i.e term x in the kobs expression above is 1.

Plot: k versus [Re] obs 1.2 10-4

-5 -1 -1 rate = 3.8(1) x 10 mM s -5 8 10 2 R = 0.9964 -1 / s obs k 4 10-5

0 0 1 2 3 [Re] / mM

Figure 2.8. Rate law dependence on rhenium concentration. Plot kobs (s-1) versus Rhenium concentration (mM). Reaction conditions: [benzaldehyde] = 2.770 M; [silane] = 3.324 M; [2- bromomesitylene] = 1.385 M; [6a] = 0.279 mM (0.01 mol%), 0.829 mM (0.03 mol%), 1.385 mM (0.5 mol%), 2.770 (1 mol%). Reactions were performed in a vial under stirring at room temperature in a N2 atmosphere. At a fixed time, a representative aliquot from the reaction mixture was 1 dissolved in CDCl3. The percent conversion was determined by H NMR spectroscopy by integrating the product peak against the internal standard.

The rate law obtained from the experiments in Figure 2.7 is first order overall. This means that the rate exhibits zero order dependences with respect to one of the substrates (benzaldehyde or silane) and a first order dependence with respect to the other substrate. The order with respect to benzaldehyde concentration was determined by the method of flooding. The decay in [benzaldehyde] (0.713 M) with 6a (0.713 mM) as the catalyst was monitored in an excess of dimethylphenylsilane (5.71 M, solvent). The reaction mixture was divided in 10 vials and each one was stirred at room temperature. At a fixed period of time benzaldehyde decay was measured by GC-MS, utilizing 2-bromomesitylene as the internal standard. A linear decay over time was observed which suggests a pseudo-zeroth order dependence with respect to [benzaldehyde] (Figure 2.9).

Plot: Benzaldehyde Decar Over Time 1

-5 -1 Rate = -3.3(3) x 10 Ms 2 R = 0.9681

0.6 [Benzaldehyde] / M [Benzaldehyde]

0.2

0 10000 20000 30000 Time / s

Figure 2.9. Kinetic plot of the concentration of benzaldehyde vs reaction time (s). Reaction conditions: [Rhenium] = 0.921 mM; [benzaldehyde] = 0.921 M; [silane] = 5.48 M. Benzaldehyde concentration was determined by GC-MS by integrating the benzaldehyde peak versus the internal standard, 2-bromomesitylene (0.438 M). The reaction was performed under neat conditions at room temperature in 10 different 4 mL screw cap vials for a fixed period of time. Given that the overall reaction is first order overall (Figure 2.7), a zeroth order dependence on benzaldehyde (Figure 2.9), suggests that the reaction is first order with respect to dimethylphenylsilane i.e. in the equation above, y = 0 and z = 1. Thus, the rate equation for the catalytic reaction is:

Rate = kobs[HSiMe2Ph] 1 where kobs = k[cat]

d[PhCH OSiMe Ph] 2 2 = k[cat][HSiMe Ph] dt 2

2.3.4 Reaction Mechanism

2.3.4.1 Mechanistic Studies

An investigation of the mechanism of the hydrosilylation of benzaldehyde with dimethylphenylsilane with 6a began with the examination of stoichiometric reactions between the cationic complex and the substrates. Attempts to identify and isolate any intermediates of substrate activation started with the stoichiometric reaction of 6a with dimethylphenylsilane in the absence of benzaldehyde (Scheme 2.6).

Scheme 2.6. Stoichiometric reaction of 6a and dimethylphenylsilane.

C6F5 CO BF4 CO C F NCCH 6 5 N 3 HSiMe2Ph Re H NCCH3 (DAAm)Re Re(DAAm) + 2 N rt N CO

6a 9

The reaction of 6a with dimethylphenylsilane described in Scheme 2.6 resulted in the formation of a DAAm dirhenium(II) complex (9). The formation of complex 9 has been reported and characterized by our group in the stoichiometric reaction of 3a with dimethylphenylsilane at 80 °C.5 This result suggests that dimethylphenylsilane is activated, in the absence of benzaldehyde, in a similar manner by 6a and 3a. The formation of 9 was confirmed by X-ray crystallography. The catalytic activity of 9 in hydrosilylation reactions was reported by our group. When 9 was used as a catalyst benzaldehyde was converted to the silyl ether product in < 5% yield in 5 h.5 A mechanism for the formation of 9 from 3a was proposed (Scheme 2.7). In this mechanism, complex 3a activates dimethylphenylsilane to produce the rhenium(III) hydride complex 10. In the absence of benzaldehyde, two molecules of 10 react to produce 9 and dihydrogen.

Scheme 2.7. Proposed mechanism for the formation of 9 from 3a by Ison group.

C6F5 CO C6F5 CO CO C F C F 6 5 N HSiMe2Ph 6 5 N Re OAc Re H (DAAm)Re Re(DAAm) N 80 °C N N -Me2PhSiOAc N CO

3a 10 9 Identification of complex 9 in both investigations, lead us to propose a similar pathway for the formation of 9 from 6a (Scheme 2.8).

Scheme 2.8. Proposed mechanism for the formation of 9 from 6a.

BF4 C6F5 CO CO C6F5 N NCCH3 HSiMe Ph Re 2 NCCH3 (DAAm)Re Re(DAAm) + H2 N rt N CO

6a 9

Proposed Mechanism

C F 6 5 CO BF4 C6F5 CO NCCH C6F5 N 3 HSiMe Ph C6F5 Re 2 N Re H NCCH3 N rt N N -BF4SiR3 -2 NCCH3 N

6a 10

C6F5 CO CO C6F5 N Re H 2 (DAAm)Re Re(DAAm) + H2 N rt N CO

10 9

On the other hand, consistent with kinetic data, no reactivity was observed in the stoichiometric reaction of 6a with benzaldehyde in the absence of dimethylphenylsilane (Scheme 2.9). The latter observation is consistent with studies reported by our group with 3a as the catalyst under similar reaction conditions.5

Scheme 2.9. Stoichiometric reaction of 6a and benzaldehyde.

C6F5 CO BF4 O NCCH C6F5 N 3 Re NCCH3 + H No Reaction N rt, 1 d N

We also examined the effect on the catalytic reaction of changing the electronics of the aldehyde substrate. Competition experiments between benzaldehyde and the corresponding para- substituted benzaldehyde were performed according to Scheme 2.10. 1H NMR spectroscopy was used to calculated product formation after 1 d. From the Hammett plot (Figure 2.10), it was observed that electron-withdrawing substituents accelerate the product forming step while electron-donating substituents retard it. This is indicated by the positive ρ value (ρ = 0.36). The last observation is consistent with previous reports by Ison group (ρ = 0.52).5

Scheme 2.10. Competition experiments between para-substituted benzaldehyde.

O OSiMe2Ph

H H H 0.5 mol% [6a] + + 0.5 equiv HSiMe2Ph, CH3CN, 1 d O OSiMe2Ph H H H X X X = OCH3, Cl, CF3, NO2

Hammet Plot 0.4

p-NO 2

0.2 ) H /P x p-CF (P 3

log p-Cl 0 p-H

2 R = 0.9252 p-OCH 3

-0.4 0 0.4 0.8 spara

Figure 2.10. Hammett plot for the competition experiments between para-substituted benzaldehyde. Reaction was carried at room temperature in CD3CN. Reaction conditions: [Re] = 0.00241 mmol, [benzaldehyde] = 0.554 mmol, [p-benzaldehyde] = 0.554 mmol, [dimethylphenylsilane] = 0.277 mmol and [2-bromomesitylene] = 0.274 mmol. The 1H NMR integration against the internal standard (2-bromomesitylene) of the methylene protons of the para-substituted silyl ether and the benzyloxydimethylphenylsilane were used to determine the product ratio value (PH/PX).

Variable temperature kinetic experiments were performed to investigate the activation parameters in a temperature range from 25 to 80 °C (Scheme 2.11). The enthalpy of activation (ΔH‡) was found to be 13.1(4) kcal/mol. The entropy of activation (ΔS‡) was found to be -34(1) cal/mol·K (Figure 2.11). The free energy of activation (ΔG‡) was calculated using the experimental activation parameters obtained as 23.5(4) kcal/mol.

Scheme 2.11. Variable temperature kinetic experiments.

O OSiMe2Ph 0.5 mol% [6a] H H H 1.2 equiv HSiMe2Ph, CH3CN, T

Figure 2.11. Eyring plot for the hydrosilylation of benzaldehyde with 6a. The rate of hydrosilylation was determined at different temperatures (25 - 80 °C). Reactions were performed in acetonitrile-d3.

2.3.4.2 Proposed Mechanism

The following mechanism has been proposed for the catalytic hydrosilylation of benzaldehyde by 6a based on kinetic data, the Hammett correlation, and mechanistic observations (Scheme 2.12).

Scheme 2.12. Proposed mechanism for the hydrosilylation reaction of aldehydes catalyzed by 6a.

CO NCCH (DAAm)Re 3 NCCH3

NCCH3

OSiR 3 CO H RDS: Silane Activation H (DAAm)Re NCCH3 Rate = k[Re][HSiMe2Ph] HSiR3

NCCH3

H- Transfer

CO CO OSiR3 CO SiR3 (DAAm)Re H SiR3 or (DAAm)Re H (DAAm)Re H + H 10 O “Ion Pair” H SN2-Si

Kinetic data suggest that the turnover-limiting step is the activation of the silane as no dependence was observed on [benzaldehyde] (Figure 2.9) and a first order dependence on [silane] was observed. Activation of the silane could occur via ƞ1 or ƞ2 coordination. Schubert previously discussed a possible reaction mechanism for the activation of a Si-H bond by an unsaturated metal center that can involve the generation of a M-H-Si bond as key intermediate. In the mechanism, coordination of the hydride to the metal center can lead to the formation of a ƞ1(H)-silane ligand. Then the coordination mode can change to a ƞ2-coordination which possess a stronger metal-silicon interaction compared to a ƞ1 coordination.11 Abu-Omar2,3 and Chan12 proposed activation of silane via ƞ2 coordination in their mechanistic studies for catalytic hydrosilylation. Attempts to isolate either a ƞ1 or a ƞ2 coordination intermediate were unsuccessful. Theoretical DFT studies were conducted to propose a silane activation pathway and will be discuss in Chapter 3.

In recent years, an ionic hydrosilylation mechanism has been proposed by Oestreich13,14 and computationally studied by Wei15-21 for catalysis by oxorhenium and oxomolybdenum complexes. In this mechanism aldehyde does not coordinate to the metal center; instead, silane is activated by coordination to rhenium. Our experimental data is consistent with these findings. The activation is followed by nucleophilic SN2-Si attack of the carbonyl substrate at the activated silicon leading to the cleavage of the Si-H bond and formation of an ion pair. Lastly, product formation results from hydride transfer from complex 10 to the silylcarbonium ion. Hydride transfer should be accelerated when electron-withdrawing groups are utilized in the para position of benzaldehyde. This is consistent with the Hammett data (Figure 2.10). Based on the experimental and kinetic data presented, the mechanisms above have been proposed by the following reasons: (1) Kinetic data suggest that the rate determine step (RDS) is the activation of the silane and no dependence was observed with respect to the concentration of benzaldehyde (Figure 2.9); (2) Treatment of 6a with dimethylphenylsilane in the absence of benzaldehyde results in the formation of complex 9 through the formation of an active Re-H complex (Scheme 2.8); (3) Electronic dependence on the benzaldehyde substrate, where electron- withdrawing group accelerate the reaction rate (Figure 2.10), is consistent to the nucleophilic attack of a rhenium-hydride to the a more electrophilic carbonyl; (4) Lastly, a ƞ2 coordination is a viable mechanism due to the activation of the organosilane previously proposed to occur in cationic rhenium complex2,3 but ƞ1 coordination cannot be discarded as we were not able to experimentally isolated any key intermediate.

2.4 Conclusions

We have shown that cationic DAAm rhenium(III) complexes can act as efficient catalysts for hydrosilylation reaction of aldehydes. Some of the features of this novel system are: (1) hydrosilylation occurs at room temperature; (2) low catalyst loadings (0.01 mol%) were utilized; (3) high percent conversions were observed; (4) a solvent free system was utilized; (5) facile complex synthesis; and (6) a variety of aldehyde substrates can be used. Experimental data suggest that 6a is significantly more active in comparison to 3a and previous rhenium catalysts reported in the literature. We successfully explored the kinetics and proposed mechanisms for the hydrosilylation reaction. Kinetic data revealed that the reaction rate is 1st order in [silane] and [6a], and it is independent of aldehyde concentration. A Hammett correlation suggests that electron- withdrawing substituents on the para position of benzaldehyde accelerate the product forming step. Based on this data, a non-hydride ionic outer-sphere hydrosilylation mechanism has been proposed.

2.5 Experimental Section

5 - + F 22 General Considerations. 3a-b and [(3,5-(CF3)2C6H3)4B] [H(OEt2)2] (HBAr 4) were prepared according to previous procedures. All reactions were carried out in a nitrogen filled glove box unless otherwise noted. All reagents were purchased from commercial sources, placed in a nitrogen filled glove box and used as received without further purification. 1H, 13C and 19F spectra were acquired on a Varian Mercury 700 MHz, Varian Mercury 400 MHz or Varian Mercury 300 MHz spectrometer. NMR chemical shifts are listed in parts per million (ppm) and are referenced to residual protons or carbons of the deuterated solvents, respectively at room temperature unless otherwise noted. The FTIR spectra were obtained in KBr thin films on a JASCO FT/IR-4100 instrument. Gas Chromatography was performed on an Agilent 7820A GC/FIC using HP-5MS columns. Elemental analyses were performed by Atlantic Micro Laboratories Inc. X-ray crystallography was performed at the X-ray Structural Facility at North Carolina State University by Dr. Roger D. Sommer and Paul D. Boyle.

General Synthesis [(DAAm-aryl)Re(CO)(NCCH3)2][X] (aryl = C6F5, Mes) (X = OTf, F BAr 4, BF4, PF6); 4a-b, 5a-b, 6a-b, 7a-b.

In a nitrogen filled glove box, the corresponding [(DAAm-aryl)Re(CO)(OAc)] (aryl = C6F5 (3a), Mes (3b)) (3a = 50 mg, 0.07 mmol; 3b = 50 mg, 0.08 mmol) was added to a screw cap vial and dissolved in 0.35 mL of acetonitrile. Then, the corresponding acid HX (X = F HOTf, HBAr 4, HBF4·OEt2, HPF6·H2O; 1 equiv) was added to the solution. The solvent was removed under pressure. The resulting oil was dissolved in diethyl ether and the corresponding bisacetonitrile product was precipitated with excess of hexanes. The final powder was collected via vacuum filtration.

[(DAAm-C6F5)Re(CO)(NCCH3)2][OTf], 4a. Following the general synthesis, complex 1 4a was synthetized in quantitative yield. H NMR (300 MHz, CD3CN) δ: 4.05 (m, 2H), 13 3.71 (m, 2H), 3.01 (m, 4H), 2.85 (s, 3H), 1.96 (s, 6H). C NMR (101 MHz, CD3CN) δ: 188.2, 172.9, 143.5, 142.7, 142.1, 141.3, 140.3, 139.0-138.2, 137.8, 137.5, 65.9, 59.1, 55.2. 19 F NMR (376 MHz, CD3CN) δ: -79.4 (s, 3F), -149.8 (dd, J = 22.1, 5.7 Hz, 2F), -150.8 (dd, J = 14.0, 8.0 Hz, 2F), -162.9 (t, J = 20.9 Hz, 2F), -165.8 (m, 2F), -166.7 (m, 2F). IR (FTIR, -1 cm ): �(CO) 1882. Anal. Calc. for C23H17F13N5O4ReS: C, 30.95; N, 7.85; H, 1.92. Found: C, 30.39; N, 7.74; H, 1.90.

[(DAAm-Mes)Re(CO)(NCCH3)2][OTf], 4b. Following the general synthesis, complex 1 4b was synthetized in quantitative yield. H NMR (400 MHz, CD3CN) δ: 6.87 (s, 2H), 6.76 (s, 2H), 4.03 (m, 2H), 3.56 (m, 2H), 3.07 (m, 2H), 2.91 (m, 5H), 2.33 (s, 6H), 2.25 (s, 6H), 13 1.96 (m, 6H), 1.84 (s, 6H). C NMR (126 MHz, CD3CN) δ: 193.4, 159.2, 135.3, 132.1, 19 131.6, 130.4, 129.7, 66.5, 60.2, 49.4, 20.4, 20.3, 20.0. F NMR (376 MHz, CD3CN) δ: - -1 79.6 (s, 3F). IR (FTIR, cm ): �(CO) 1890. Anal. Calc. for C29H39F3N5O4ReS·H2O: C, 43.71; N, 8.79; H, 4.93. Found: C, 42.50; N, 8.69; H, 5.00. F [(DAAm-C6F5)Re(CO)(NCCH3)2][BAr 4], 5a. Following the general synthesis, 1 complex 5a was synthetized in quantitative yield. H NMR (300 MHz, CD3CN) δ: 7.69 (m, 12H), 4.03 (m, 2H), 3.72 (m, 2H), 2.99 (m, 4H), 2.84 (s, 3H), 1.96 (s, 6H). 13C NMR (175

MHz, CD3CN) δ: 188.1, 163.0, 162.7, 162.4, 162.1, 143.7, 142.9, 142.2, 141.4, 140.4, 139.2-138.6, 137.9, 137.6, 130.2, 130.0, 129.8, 129.6, 127.7, 126.2, 124.6, 123.1, 66.2, 19 59.3, 47.5. F NMR (376 MHz, CD3CN) δ: -63.25 (s, 24F), -149.8 (m, 2F), -150.8 (m, 2F), -162.9 (m, 2F), -165.8 (t, J = 20.7 Hz, 2F), -166.2 (m, 2F). IR (FTIR, cm-1): �(CO) 1925.

Anal. Calc. for C54H29BF34N5ORe·H2O: C, 39.92; N, 4.31; H, 1.92. Found: C, 39.11; N, 4.80; H, 2.02. F [(DAAm-Mes)Re(CO)(NCCH3)2][BAr 4], 5b. Following the general synthesis, complex 1 5b was synthetized in quantitative yield. H NMR (376 MHz, CD3CN) δ: 7.69 (m, 12H), 6.87 (s, 2H), 6.75 (s, 2H), 4.02 (m, 2H), 3.55 (m, 2H), 3.07 (m, 2H), 2.91 (m, 5H), 2.33 (s, 19 6H), 2.25 (s, 6H), 1.96 (m, 6H), 1.84 (s, 6H). F NMR (376 MHz, CD3CN) δ: -63.4 (s, 24F). IR (FTIR, cm-1): �(CO) 1890. The complex was not characterized by 13C NMR due to its very poor stability. Elemental Analysis was not attempted on this complex because of its instability.

[(DAAm-C6F5)Re(CO)(NCCH3)2][BF4], 6a. Following the general synthesis, complex 1 6a was synthetized in quantitative yield. H NMR (300 MHz, CD3CN) δ: 4.05 (m, 2H), 13 3.72 (m, 2H), 3.01 (m, 4H), 2.85 (s, 3H), 2.14 (s, 6H). C NMR (176 MHz, CD3CN) δ: 188.2, 143.5, 142.7, 142.1, 141.3, 140.2, 139.0-138.4, 137.8, 137.6-137.3, 66.0, 59.1, 47.3. 19 F NMR (376 MHz, CD3CN) δ: -149.8 (dd, J = 22.1, 5.8 Hz, 2F), -150.8 (m, 2F), -151.7 (d, J = 20.2 Hz, 4F), -162.9(t, J = 20.9 Hz, 2F), -165.8 (m, 2F), -166.2 (m, 2F). IR (FTIR, -1 cm ): �(CO) 1882. Anal. Calc. for C22H17BF14N5ORe: C, 31.82; N, 8.43; H, 2.06. Found: C, 31.40; N, 7.90; H, 2.12.

[(DAAm-Mes)Re(CO)(NCCH3)2][BF4], 6b. Following the general synthesis, complex 1 6b was synthetized in quantitative yield. H NMR (300 MHz, CD3CN) δ: 6.87 (s, 2H), 6.75 (s, 2H), 4.03 (m, 2H), 3.55 (m, 2H), 3.07 (m, 2H), 2.91 (m, 5H), 2.33 (s, 6H), 2.25 (s, 6H), 13 1.84 (d, 12H). C NMR (126 MHz, CD3CN) δ: 193.4, 159.3, 135.3, 132.1, 131.6, 130.4, 19 129.7, 66.5, 60.2, 49.42, 20.43, 20.30, 20.02. F NMR (376 MHz, CD3CN) δ: -151.3 (s, -1 4F). IR (FTIR, cm ): �(CO) 1878. Anal. Calc. for C28H39BF4N5ORe·H2O: C, 44.68; N, 9.30; H, 5.49. Found: C, 43.80; N, 8.99; H, 5.38.

[(DAAm-C6F5)Re(CO)(NCCH3)2][PF6], 7a. Following the general synthesis, complex 1 7a was synthetized in quantitative yield. H NMR (300 MHz, CD3CN) δ: 4.07 (m, 2H), 13 3.74 (m, 2H), 3.05 (m, 4H), 2.87 (s, 3H), 1.99 (s, 6H). C NMR (101 MHz, CD3CN) δ: 188.1, 144.1, 143.3, 141.6, 140.8, 139.3, 138.5, 137.8, 136.9, 66.0, 59.2, 47.3. 19F NMR

(376 MHz, CD3CN) δ: -72.0 (s, 3F), -73.9 (s, 3F), -150.0 (dd, J = 22.1, 5.6 Hz, 2F), -150.9 (dt, J = 16.1, 4.7 Hz, 2F), -163.0 (t, J = 20.9 Hz, 2F), -165.8 (t, J = 20.0 Hz, 2F), -166.2 (m, -1 2F). IR (FTIR, cm ): �(CO) 1882. Anal. Calc. for C22H17F16N5OPRe: C, 29.74; N, 7.88; H, 1.93. Found: C, 29.62; N, 7.19; H, 2.03.

[(DAAm-Mes)Re(CO)(NCCH3)2][PF6], 7a. Following the general synthesis, complex 1 7b was synthetized in quantitative yield. H NMR (300 MHz, CD3CN, δ): 6.88 (s, 2H), 6.75 (s, 2H), 4.03 (m, 2H), 3.55 (m, 2H), 3.07 (m, 2H), 2.97 (m, 5H), 2.33 (s, 6H), 2.25 (s, 13 6H), 1.84 (m, 6H), 1.86 (s, 6H). C NMR (101 MHz, CD3CN, δ): 193.37, 159.31, 135.36, 132.15, 131.60, 130.40, 129.70, 66.52, 60.21, 49.47, 20.48, 20.34, 20.07. 19F NMR (376 -1 MHz, CD3CN, δ): -71.96 (s, 3F), -73.84 (s, 3F). IR (FTIR, cm ): �(CO) 1878. Elemental Analysis was not attempted on this complex because of its instability.

Synthesis of complex [(DAAm-C6F5)Re(CO)(Cl)2], 8. In a screw cap vial, complex 3a (50 mg, 0.07 mmol) was dissolved in a minimal amount of methylene chloride. Excess 12 M HCl (5.0 µL, 60.0 µmol) was added to the screw cap vial. The solution was shaken and let to stand overnight. The product precipitates out of solution 1 as green crystals (21.9 mg, 43.0% yield). H NMR (400 MHz, CD3CN) δ: 4.16 (m, 2H), 3.45 (m, 2H), 3.36 (s, 3H), 3.31 (m, 4H). The amine proton in the ligand backbone was not observed either at room temperature or -60 °C indicating that the proton is too fast to be 19 observed on the NMR time scale. F NMR (376 MHz, CD3CN) δ: -150.1 (m, 2F), -150.5 (dd, J = 23.5, 6.6 Hz, 2F), -162.9 (m, 2F), -167.2 (m, 4F). The complex was not

13 characterized by C NMR due to its very poor solubility in acetonitrile-d3 or methylene -1 chloride-d2. IR (FTIR, cm ): �(CO) 1873. Anal. Calc. for C18H13Cl2F10N3ORe: C, 29.48; N, 5.73: H, 1.65. Found: C, 29.79; N, 6.01: H, 1.59.

General Procedure for Catalytic Hydrosilylation of Aldehydes.

In a nitrogen filled glove box, the corresponding [(DAAm-aryl)Re(CO)(OAc)] (aryl = C6F5 (3a), Mes (3b)) (3a = 2 mg, 0.00277 mmol; 3b = 2 mg, 0.00320 mmol) was added to a screw cap vial and dissolved in 0.35 mL of acetonitrile. Then HBF4·OEt2 (1 equiv) was added to the solution and mixed. The solvent was removed under reduced pressure to afford an oil. The aldehyde (27.7 mmol, 0.01 mol%; 9.23 mmol, 0.03 mol%; 5.54 mmol, 0.05 mol%; 2.77 mmol, 0.1 mol%; 0.554 mmol, 0.5 mol%; 0.277 mmol, 1 mol%), silane (1.2 equiv) and 2-bromomesitylene (0.5 equiv) were sequentially added to the vial. The reaction mixture was stirred at room temperature for the designated amount of time. Then, an aliquot of the crude reaction was dissolved in CDCl3 or

CD3CN. The yield was determined by the proton NMR ratio of the product and starting material ((integration of methylene protons/2)/((integration of aldehyde peak) + (integration of methylene protons/2))) against 2-bromomesitylene as the internal standard.

General Procedure for Time Profile Experiments with 6a. In a nitrogen filled glove box complex 3a (2.0 mg, 0.00277 mmol) was added to a 4 mL screw cap vial. Followed by HBF4·OEt2 (0.3 µL, 0.00277mmol, 1 equiv), 0.35 mL CH3CN, and a stir bar. Then, benzaldehyde (27.7 mmol, 0.01 mol%; 9.23 mmol, 0.03 mol%; 5.54 mmol, 0.05 mol%; 2.77 mmol, 0.1 mol%); dimethylphenylsilane (33.24 mmol, 0.01 mol%; 11.08 mmol, 0.03 mol%; 6.648 mmol, 0.05 mol%; 3.324 mmol, 0.1 mol%), and 2-bromomesitylene (0.5 equiv) were sequentially added. The reaction mixture was stirred at room temperature. At a fixed time, an aliquot was placed in CDCl3 or CD3CN. The yield was determined by the proton NMR ratio of the product and starting material ((integration of the methylene protons/2)/((integration of the aldehyde peak) + (integration of the methylene protons/2))).

General Procedure to Obtain Order with Respect to Benzaldehyde by Gas Chromatography (GC). In nitrogen filled glove box 6a (0.00241 mmol, 2.0 mg), benzaldehyde (2.77 mmol, 0.28 mL), dimethylphenylsilane (16.62 mmol, 2.5 mL), and 2-bromomesitylene (0.2 mL) were mixed in a small screw cap vial. The reaction mixture was divided into 10 small screw cap vials equipped with a stir bar. The reactions were stirred at room temperature for a designated amount of time. At that time, the ratio benzaldehyde:2-bromomesitylene was determined by GC using the calibration curve method. The concentration of benzaldehyde was calculated solving for the term x using the equation of the line, y = 1.25(5)x + 0.04(3). The calibration curve was obtained by calculating the ratio of benzaldehyde:2-bromomesitylene at various benzaldehyde concentrations (Figure 2.12).

Figure 2.12. Calibration curve to calculate the concentration of benzaldehyde by Gas Chromatography (GC). Plot of Benzaldehyde Concentration (M) versus Ratio Benzaldehyde:2- bromomesitylene. Stock Solutions = 0.98 M, 0.74 M, 0.50 M, 0.30 M, 0.15 M benzaldehyde and 2-bromomesitylene (0.44 M) in dichloromethane.

General Procedure for Competition Reaction for Hammett Plot.

In nitrogen filled glove box 3a (2.0 mg, 0.00277 mmol), HBF4·OEt2 (0.37 µL, 0.00277 mmol), benzaldehyde (0.554 mmol, 56 µL), the indicated para-substituted benzaldehyde (0.554 mmol), dimethylphenylsilane (0.277 mmol, 0.5 equiv, 42 µL), and 2-bromomesitylene (0.274 mmol, 0.5 equiv, 42 µL) were sequentially added to a small screw cap vial. The reaction mixture was dissolved in CD3CN (0.35 mL) and stirred for 24 h at room temperature. An aliquot of the reaction 1 mixture was placed in CD3CN. The H NMR integration against the internal standard (2- bromomesitylene) of the methylene protons of the para-substituted silyl ether and the benzyloxydimethylphenylsilane were used to determine the product ratio value (PH/PX).

General Procedure for Eyring Plot data. In a nitrogen filled glove box 3a (2 mg, 0.00277 mmol) was added to a screw cap vial and dissolved in 0.35 mL of acetonitrile. Then HBF4·OEt2 (0.3 µL, 0.00277 mmol, 1 equiv) was added to the solution and mixed. The benzaldehyde (0.57 µL, 0.554 mmol), dimethylphenylsilane (0.1 mL, 0.665 mmol, 1.2 equiv) and 2-bromomesitylene (42 µL, 0.277 mmoles, 0.5 equiv) were sequentially added to the vial. The reaction was stirred at the respective temperature (25 – 80 °C).

An aliquot at a fixed time was dissolved in either in CDCl3 or CD3CN. The yield was determined by the proton NMR ratio of the product and starting material ((integration of methylene protons/2)/((integration of aldehyde peak) + (integration of methylene protons/2))) against 2- bromomesitylene as the internal standard.

X-ray Crystallographic Procedures and Data General Procedure for X-ray Determination X-Ray structural determination was performed at the X-Ray Structural Facility of North Carolina State University by Dr. Roger Sommer and Paul D. Boyle. The samples were mounted on a nylon loop with a small amount of Paratone N oil. The frame integration was performed using SAINT.22 The resulting raw data were scaled, and absorption corrected using a multi-scan averaging of symmetry equivalent data using SADABS.23 Most non-hydrogen atoms were obtained from the initial solution. The remaining non-hydrogen atom positions were recovered from a subsequent difference Fourier map. The structural model was fit to the data using full matrix least-squares based on F2. The calculated structure factors included corrections for anomalous dispersion from the usual tabulation. The structure was refined using the XL program from

SHELXTL24, graphic plots were produced using the NRCVAX crystallographic program suite. Additional information and other relevant literature references can be found in the reference section of the Facility's Web page (http://www.xray.ncsu.edu). Selected Crystallographic Data and Collection Parameters are shown in Table 2.5.

Table 2.5. Selected Crystallographic Data and Collection Parameters for 5a, 6b, and 8. Parameter 5a 6b 8 Empirical Form C54H29BF34N5ORe C28H40.50BF4N5.50ORe C18H12Cl2F10N3ORe Formula weight (g/mol) 1691.76 755.18 732.40 crystal system monoclinic monoclinic orthorhombic space group P 21/c P1 21/c 1 P b c a a, Å 9.777(2) 22.3852(5) 12.2754(4) b, Å 16.855(4) 14.7691(3) 18.8021(6) c, Å 36.649(8) 19.6678(5) 18.9382(6) Volume, (Å3) 6021(2) 6305.8(3) 4371.0(2) Z 4 8 8 ρ (g/cm3) 1.866 1.591 2.226 crystal size (mm) 0.39 × 0.13 × 0.09 0.043 × 0.089 × 0.268 0.13 × 0.10 × 0.04 R1, w R2 0.0509, 0.1101 0.0270, 0.0542 0.0323, 0.0609 GOF 1.030 1.009 1.021

2.6 References

(1) Marciniec, B. Comprehensive Handbook on Hydrosilylation, Elsevier, 2013. (2) Ison, E. A.; Trivedi, E. R.; Corbin, R. A.; Abu-Omar, M. M. J. Am. Chem. Soc. 2005, 127, 15374-15375. (3) Ison, E. A.; Cessarich, J. E.; Du, G.; Fanwick, P. E.; Abu-Omar, M. M. Inorg. Chem. 2006, 45, 2385-2387. (4) Du, G.; Abu-Omar, M. M. Organometallics 2006, 25, 4920-4923. (5) Smeltz, J. L.; Boyle P. D.; Ison, E. A. Organometallics, 2012, 31, 5994-5997. (6) Choualeb, A.; Maccaroni, E.; Blacque, O.; Schmalle, H.; Berke, H. Organometallics, 2008, 27, 3474-3481. (7) Dong, H.; Berke, H. Adv. Synth. Catal., 2009, 351, 1783-1788. (8) Note: These two CO stretches are attributed to two isomeric forms of 1a, where the acetate

ligand is bound in either a κ1 (1a') or a κ2 (1a) fashion to the rhenium metal center. (9) Chan, L. Y. Y.; Isaacs, E. E.; Graham, W. A. G. Can. J. Chem. 1977, 55, 111–114. (10) Perera, T.; Abhayawardhana, P.; Fronczek, F.R.; Marzilli, P.A.; Marzilli, L.G. Eur. J. Inorg. Chem. 2012, 618-627. (11) Schubert, U. Advances in Organometallic Chemistry Vol. 30, Academic Press, Inc., 1990, 151-187. (12) Zheng, G. Z.; Chan, T. H. Organometallics 1995, 14, 70. (13) Metsänen, T. T.; Hrobμrik, P.; Klare, H. F. T.; Kaupp, M.; Oestreich, M. J. Am. Chem. Soc. 2014, 136, 6912–6915. (14) Rendler, S.; Oestreich, M. Acid Catalysis in Modern Organic Synthesis, 2008. (15) Gu, P.; Wang, W.; Wang, Y.; Wei, H. Organometallics 2013, 32, 47-51. (16) Gu, P.; Wang, W.; Wang Y.; Wei, H. Organometallics, 2012, 32, 47-51. (17) Huang, L.; Wang, W.; Wei, X.; Wei, H. J. Phys. Chem. A, 2015, 119, 3789-3799. (18) Huang, L.; Zhang, Y.; Wei, H. Eur. J. Inorg. Chem., 2014, 2014, 5714-5723. (19) Wang, J.; Wang, W.; Huang, L.; Yang, X.; Wei, H. ChemPhysChem, 2015, 16, 1052-1060. (20) Huang, L.; Wang, W.; Wei, H. J. Mol. Catal. A: Chem., 2015, 400, 31–41. (21) Wang, J.; Huang, L.; Yang, X.; Wei, H. Organometallics, 2014, 34, 212-220. (22) Bruker-Nonius, SAINT version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA. (23) Bruker-Nonius, SADABS version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA.

(24) Bruker-AXS, XL version 2009.9, 2009, Bruker-AXS, Madison, WI 53711, USA.

Chapter 3: Mechanistic Investigation of the Hydrosilylation Reaction of Aldehydes by a Bisacetonitrile Cationic Rhenium(III) Complex: A Computational Study

Unpublished results

Department of Chemistry, North Carolina State University Raleigh, North Carolina, 27695-8204

3.1 Abstract

The catalyzed hydrosilylation reaction of benzaldehyde by the cationic rhenium(III) complex

[(DAAm-aryl)Re(CO)(NCCH3)2][BF4] (DAAm = N,N-bis(2-arylaminoethyl)methylamine; aryl =

C6F5) (6a) was studied theoretically to elucidate the mechanistic pathway for this transformation. Based on DFT (B3PW91-D3) calculations an ionic hydrosilylation mechanism was confirmed. Calculations revealed that silane activation by cationic rhenium(III) complex occurs via ƞ1- coordination to the rhenium center. A rhenium-silane adduct 13 was identified. The activation energy for the rate determine step (RDS), which is the activation of silane, was calculated to be 21.4 kcal/mol. Formation of the silylated product is a high exergonic step that has an overall free energy of -24.1 kcal/mol.

3.2 Introduction

Hydrosilation reactions represent an important class of chemical transformations.1 The importance of this type of transformation has led to an extensive study of the reaction mechanism.2 A better understanding of the catalytic cycle may lead to the development of more efficient catalysts. A wide variety of transition metal complexes show catalytic activity toward hydrosilylation reactions of unsaturated organic molecules.3 In recent years, the mechanisms of the catalytic hydrosilylation reaction of aldehydes has been widely studied. The majority of these pathways are proposed based on experimental and kinetic observations and some are supported by theoretical calculations. The most accepted mechanism, proposed by Chalk and Harrod4-6 for alkenes and alkynes and later modified for carbonyl compounds by Ojima7, follows a 3-step pathway: (1) oxidative addition of a Si-Hi bond to the metal center; (2) insertion of the unsaturated substrate into the metal hydride bond; and finally (3) reductive elimination to form the product (Scheme 3.1 left). In 2003, Toste et al. demonstrated that hydrosilylation of carbonyls by (PPh3)Re(O)2I could take place via an initial addition of the Si-H bond across one Re=O bond in a [2+2] manner (Scheme 3.1 right).8,9 This unprecedent mechanism was confirmed to be preferred by theoretically studies.10 The Ojima mechanism and [2+2] addition involve coordination and migration insertion of the substrate to the metal center. These pathways are classified as hydride mechanisms due to the involvement of a metal hydride intermediate in each catalytic cycle.

Scheme 3.1 Hydride mechanisms for the hydrosilylation reaction of carbonyl compounds.

PPh3 OSiR3 OSiR3 O Rʹ [M] I Re H O Rʹ H Rʹ H PPh3 H-SiR3 H OSiR3 H [M] [M] PPh3 Rʹ O Rʹ SiR3 I PPh3 Re O OCH Rʹ I O R3SiO 2 Re PPh3 R SiO H Rʹ 3 H Rʹ Rʹ PPh3 O [M] Rʹ SiR3 O

H Rʹ

Ojima’s Modified [2+2] Addition by Toste Chalk-Harrod Mechanism

In 1994, Bullock et al. introduced a non-hydride ionic hydrogenation pathway of alkenes11 and a similar mechanism for the ionic hydrogenation of ketones by cationic complexes of the + - form [CpM(CO)2(IMes)] [B(C6F5)4] (Cp = cyclopentadienyl, IMes = 1,3-bis(2,4,6- trimethylphenyl)- imidazol-2-ylidene, M = Mo, W) was proposed.12,13 A modified version for the hydrosilylation reaction of ketones using similar molybdenum and tungsten catalysts was also proposed by Bullock in 2003 (Scheme 3.2).14 Since then, ionic hydrosilylation mechanism have been reported in the literature using various metals such as rhenium15-17, iridium18-21, and ruthenium22. Density functional studies supported an ionic hydrosilylation pathway rather than classical Chalk-Harrod mechanism for these catalytic systems.23-25 In oxorhenium chemistry, the Wei group have reported theoretical studies proving ionic hydrosilylation mechanism over [2+2] + activation of silane across the terminal oxo ligand when Re(O)HCl2(PPh3), [Re(O)(hoz)2] (hoz =

2-(2′- hydroxyphenyl)-2-oxazoline(1−)), Re(O)Cl3(PPh3)2 (hydrosilylation of carbonyls) or

ReO2I(PPh3)2 (hydrosilylation of imines) are use as the catalysts, and both pathways compete when 26-30 ReO2I(PPh3)2 catalyzes the hydrosilylation of carbonyl compounds. The ionic hydrosilylation mechanism proceeds via the activation/coordination of the silane rather than an oxidative addition to the metal center like in the Chalk-Harrod mechanism. The catalytic cycle involved an heterolytic cleavage of the Si-H bond and features a SN2-Si attack at the activated silicon by the carbonyl oxygen, and the formation of an ion pair consisting of a metal hydride complex and a silylcarbonium ion (Scheme 3.2). The ionic outer-sphere mechanism involves a silyl cation and then a hydride transfer to a polar double bond. 46

Scheme 3.2 Proposed mechanism for catalytic ionic hydrosilylation by Bullock et al.

OSiEt3 H [M] Rʹ H H Rʹ H-SiEt3

H [M] H [M] + SiEt3

OSiEt3 O Rʹ Rʹ Rʹ Rʹ

Catalytic Ionic Hydrosilylation

Low-valent rhenium complexes have been shown to be excellent catalysts for the hydrosilylation reaction of aldehydes.31 The most recently example was published by our group (Scheme 3.3). In Chapter 2, we presented the catalytic efficiency of novel cationic DAAm rhenium

(III) [(DAAm-aryl)Re(CO)(NCCH3)2][BF4] (DAAm = N,N-bis(2-arylaminoethyl)methylamine; aryl = C6F5) (6a) for hydrosilylation reaction of aldehydes. Catalyst 6a can be use with low catalyst loadings (0.03 mol% - 0.01 mol%) to obtain complete conversion of a wide variety of aldehydes under neat conditions at room temperature within 24 h.32

Scheme 3.3 Catalytic Hydrosilylation Reaction of Aldehydes by 6a.

O 0.03 mol% [Re] OSiMe2Ph

1.2 equiv HSiMe Ph, neat, rt R H R H 2 H

C6F5 CO BF4

C6F5 N NCCH3 Re NCCH3 N N

TONs > 8500

Experimental and kinetic data suggest that the reaction rate is 1st order in [silane] and [6a], and independent of aldehyde concentration. A Hammett correlation suggests that electron- withdrawing substituents on the para position of benzaldehyde facilitate the product forming step. Based on this data and similarities to previous reports, a non-hydride ionic outer-sphere hydrosilylation mechanism was proposed. In the mechanism described in Scheme 3.4, aldehyde does not coordinate to the metal center; instead, silane is activated by an initial coordination to rhenium metal center in contrast to previous proposed mechanism using low-valent rhenium catalyst by Berke31 and Ison.32,33

Scheme 3.4 Proposed mechanism for the hydrosilylation reaction of aldehydes catalyzed by 6a.

CO NCCH (DAAm)Re 3 NCCH3

NCCH3

OSiR 3 CO H RDS: Silane Activation H (DAAm)Re NCCH3 Rate = k[Re][HSiMe2Ph] HSiR3

NCCH3

H- Transfer

CO CO OSiR3 CO SiR3 (DAAm)Re H SiR3 or (DAAm)Re H (DAAm)Re H + H 10 O “Ion Pair” H SN2-Si

In this chapter, Density Functional Theory34 (DFT) studies of the reaction depicted in Scheme 3.3 were performed. Our goal was to determine a reasonable mechanism pathway for the hydrosilylation reaction of aldehydes by 6a, to support computationally the mechanism previously proposed based on experimental data32 and to gain insights with regards to the structure and thermodynamics.

3.3 Results and Discussion

Density Functional Theory (DFT) calculations were conducted in order to gain additional understanding into the mechanism of the hydrosilylation reaction by cationic DAAm rhenium(III) complexes previously reported by our group. Complete computational details are outlined in the experimental section. First, we examine if the functional and basis set were suitable for describing the geometry of the cationic DAAm rhenium(III) complex. The geometry of 4a (Figure 3.1) was optimized using functionals B3PW9135-D336 and M0637. Results are summarized in Table 3.1 and Table 3.2. The deviation in the bond length (Å) and bond angles (°) from the X-ray crystallography structure for 4a were calculated. The average percent of deviation was less than 1.0% for B3PW91- D3 functional. Therefore, B3PW91-D3 functional was used for all other calculations.

Figure 3.1 X-ray crystallography determined structure of 4a.

Table 3.1 Comparison of the calculated DFT (B3PW91-D3) bond lengths (Å) for 4a with the X- ray crystallography.

% Deviation % Deviation Entry Bond X-ray B3PW91-D3a,b M06a (B3PW91-D3) (M06) 1 Re-C1 1.867 1.895 1.89 1.50 1.23

2 Re-N1 1.932 1.939 1.94 0.36 0.41

3 Re-N2 2.22 2.250 2.28 1.35 2.70

4 Re-N3 1.928 1.931 1.93 0.16 0.26

5 Re-N4 2.167 2.155 2.19 0.55 1.15

6 Re-N5 2.120 2.125 2.17 0.24 2.17

Average 0.69 1.32 aGeometries were calculated in the gas phase using the specified functional with the SDD38 basis set and an added � polarization function for rhenium and 6-31G(d,p) for all other atoms as implemented in Gaussian 0939. bThe dispersion correction was assessed using Grimme’s D3 parameter set.

Table 3.2 Comparison of the calculated DFT (B3PW91-D3) bond angles (°) for 4a with the X- ray crystallography. B3PW91- % Deviation % Deviation Entry Angle (°) X-ray M06 a D3 a,b (B3PW91-D3) (M06) 1 N1-Re1-N3 105.04 106.0 105.60 0.91 0.53

2 N1-Re1-N2 80.79 79.9 79.50 1.10 1.60

3 N1-Re1-C1 94.26 95.3 95.00 1.10 0.79

4 N2-Re1-C1 172.91 172.9 171.70 0.01 0.70

5 N1-Re1-N4 85.09 85.0 86.10 0.11 1.19

6 N1-Re1-N5 160.96 163.0 163.50 1.27 1.58

7 Re1-N1-C8 119.20 119.9 120.00 0.59 0.67 Re1-N3- 8 116.88 117.4 117.40 0.44 0.44 C12 Average 0.69 0.94 aGeometries were calculated in the gas phase using the specified functional with the SDD38 basis set and an added � polarization function for rhenium and 6-31G(d,p) for all other atoms as implemented in Gaussian 0939. bThe dispersion correction was assessed using Grimme’s D3 parameter set.

Part A: Silane Activation (RDS)

Ligand Dissociation: The starting point for the catalytic cycle for the hydrosilylation reaction catalyzed by the + six-coordinate octahedral complex [(DAAm-C6F5)Re(CO)(NCCH3)2] (11) is the dissociation of one acetonitrile. Ligand dissociation leads to the formation of an open coordination site on the rhenium center. Computational DFT results suggest that ligand dissociation and formation of + complex [(DAAm-C6F5)Re(CO)(NCCH3)] (12) is an overall endergonic process, ΔG° = 6.3 kcal/mol, with respect to reactant 11.

Coordination of Silane: The open coordination site in complex 12 can be occupied by silane. DFT calculations were performed using HSi(Me)2Ph as the silane substrate, similar to the reported optimized 32 conditions (Scheme 3.3). Addition of HSi(Me)2Ph results in the rhenium-silane adduct 13 with an energy of -1.1 kcal/mol (Figure 3.2). In adduct 13 the Re∙∙∙H distance is 2.18 Å. While this intermediate was identified by DFT studies it was not possible to identified or isolated experimentally. In the stoichiometric reaction of complex 6a with HSi(Me)2Ph at room temperature we identified complex 9 by X-ray crystallography (Chapter 2). Adduct 13 represents the intermediate prior to the second NCCH3 ligand dissociation.

Figure 3.2 B3PW91-D3 calculated structure of intermediate 13. Aryl rings are shown in wireframe for clarity.

The formation of intermediate 13 is followed by dissociation of NCCH3 ligand leading to the formation of complex 14. In structure 14, the silane resides on the equatorial plane cis to the diamidoamine ligand framework (Figure 3.3). The calculated Re∙∙∙H distance is 2.03 Å and the Re∙∙∙Si distance is 2.99 Å. The Si-H distance is 1.56 Å which is 0.07 Å longer that Si-H distance in free silane (1.49 Å). Distance of H-Si is indicative that silane is only weakly η1 coordinated to the rhenium center. The angle between Re-H-Si is 112.1°.

Figure 3.3 B3PW91-D3 calculated structure of intermediate 14. Aryl rings are shown in wireframe for clarity.

Part B: SN2-Si Attack

Nucleophilic Attack: In this step of the catalytic cycle, the nucleophilic oxygen of the benzaldehyde substrate attacks the activated silicon atom to prompt the heterolytic cleavage of the Si-H bond. Cleavage of the Si-H bond occurs via transition state TS4 (Figure 3.4). The length of the Si-H bond is 1.83 Å, which is elongated by 0.27 Å compared to the Si-H bond in 14. Also, in TS4 the Re-H distance

53 is 1.73 Å, which is 0.30 Å shorter than in 14. The distances suggest the formation of the Re-H bond and cleavage of the Si-H bond take place through TS4. The calculated vibrational motion of the imaginary frequency associated with TS4 represents an SN2-Si reaction, which features an attack of the nucleophilic aldehyde on the activated silicon, and at the same time the approach of the hydride to the rhenium center. The nucleophilic attack occurs syn to the Re-H bond in 14 (∠Re- H-Si-O = 117.7°). The approach syn to the Re–H bond is consistent with other theoretical reports in the literature.27 The cleavage of the Si-H bond to generate an electron-positive silicon atom and a metal hydride is promoted by σ-coordination of the Si-H bond to an electron-deficient or highly electrophilic metal center.40 This is because back-donation from the metal to the Si-H bond is less likely from an more electron-deficient metal center. In this case Si-H activation occurs by the coordination of the weakly basic Si-H bond to the high electrophilic rhenium metal center in 6a. Analysis of TS4 could explain why complex 6b (Chapter 2) does not show any reactivity for the catalytic hydrosilylation of aldehydes. Complex 6b possess electron-donating mesitylene groups in the ligand framework which make the complex less electron-deficient in comparison to 6a which possess pentafluorophenyl substituents in the ligand backbone. If the metal center is more electron-rich, in this case rhenium in 6b, the Si-H bond cleavage would be less likely to happen.

We hypothesized that a TS of an SN2-Si attack where the metal center is more electron-rich, like in 6b, will be higher in energy than TS4 and therefore more unlikely to happen.

Figure 3.4 B3PW91-D3 calculated structure of TS4. Aryl rings are shown in wireframe for clarity.

Ion Pair Formation: TS4 is followed by the formation of the ion pair 15 (Figure 3.5). The free energy of intermediate 15 was calculated at 19.9 kcal/mol below TS4. Intermediate 15 consist of an anionic rhenium hydride complex 10 (Chapter 2) paired with a silylcarbonium ion. The Si-H bond in ion pair 15 has a distance of 3.12 Å, significantly elongated in comparison to TS4.

Figure 3.5 B3PW91-D3 calculated structure of ion pair 15. Aryl rings are shown in wireframe for clarity.

Part C: Hydride Transfer & Product Formation

The last step of the catalytic cycle requires the formation of a C-H bond on the silylcarbonium ion to form the silylated product. The calculated vibrational motion of the imaginary frequency associated with TS5 represents a hydride transfer, which features the hydride ligand approaching to the carbon atom of the silylcarbonium ion (Figure 3.6). In TS5 the Re∙∙∙H distance is elongated by 0.14 Å compared to 15 (1.67 Å). The CH+∙∙∙H- distance is 1.36 Å, drastically shorter and closer to the anionic rhenium center in comparison to 15 (2.60 Å). The free energy value in this transition state is 5.1 kcal/mol.

Figure 3.6 B3PW91-D3 calculated structure of TS5. Aryl rings are shown in wireframe for clarity.

Following TS5 is the silylated product formation and product release of the catalytic cycle. Complex 16 shows the hydride transfer product (Figure 3.7). Intermediate 16 has a free energy of 2.1 kcal/mol. The ΔG° of 16 is comparable to ion pair 15 (2.20 kcal/mol). This suggests that these intermediates have similar stability and are under equilibrium. It is worth noting that in 16 the phenyl group next to the methylene is coordinated to the rhenium center. The Re∙∙∙phenyl distance is 2.45 Å. To close the catalytic, intermediate 2 is regenerate and then the silylated product is released.

Figure 3.7 B3PW91-D3 calculated structure of 16. Aryl rings and part of the silylated product are shown in wireframe for clarity.

In summary, the free-energy profile for the catalyzed hydrosilylation reaction of aldehydes by cationic rhenium(III) complex proceeds via an ionic hydrosilylation mechanism shown in Figure 3.8. The mechanism can be divided into three parts: (1) Silane activation (in red) subdivided in 2 steps (a) ligand dissociation, 11 → 12; and (b) coordination of silane through a rhenium-silane adduct, 12 → 13 → 14; (2) SN2-Si attack (in green) subdivided in 2 steps (a) nucleophilic attack, 14 → TS4; and (b) ion pair formation, TS4 → 15; and (3) Hydride transfer and product formation (in blue), 15 → TS5 → 16 → → product. Theoretical DFT studies are consistent with experimental data and suggest that the rate determining step (RDS) is the activation of silane. Experimentally the reaction rate where the reaction is first order with respect to silane and zero order with respect to benzaldehyde. Computationally, the highest activation energy in the reaction progress belongs to the silane activation (TS3) with ΔG⧧ = 22.5 kcal/mol upfield (Figure 3.8). Experimentally, the overall free energy of activation (ΔG‡ (298 K)) was 23.5 kcal/mol.32

Figure 3.8 DFT (B3PW91-D3/6-31G(d,p)) calculated reaction progress for the hydrosilylation reaction of benzaldehyde by cationic rhenium(III) complex.

3.4 Conclusions

The reaction mechanism for the catalytic hydrosilylation reaction of benzaldehyde by cationic rhenium(III) complex 11 was investigated using DFT studies. Computational studies suggest that the silane activation is via a ƞ1-coordination. DFT studies suggest the formation of intermediate 13, even though experimentally has not been isolated and characterized. From the comparison of this investigation to recent studies, we can conclude that 6a follows a non-hydride ionic outer-sphere hydrosilylation mechanism.

3.5 Experimental Section

Computational Details. Geometry and transition state optimizations were performed with the 6- 31G(d,p)41 basis set on light atoms and the SDD38 basis set with an added f polarization function on rhenium.42 Each optimization involved tight optimization criteria implemented in Gaussian 0939 (opt = tight) with an ultrafine integral grid (int = ultrafine) and the B3PW9135 functional. The use of Grimme’s dispersion correction36 was also employed in all calculations. All structures were fully optimized, and analytical frequency calculations were performed on all structures to ensure either a zeroth-order saddle point (a local minimum) or a first-order saddle point (a transition state). Energetics were calculated at 298 K with the 6-311++G(d,p)43 basis set for C, H, N, O, F and Si atoms and the SDD38 basis set with an added f polarization function42 on Re with the B3PW9135 functional. All energies are in the gas phase and reported in kcal/mol.

3.6 References

(1) Lipke, M. C.; Liberman-Martin, A. L.; Tilley, T. D. Angew. Chem. Int. Ed. 2017, 56, 2260 – 2294. (2) Some recent studies on the mechanism: (a) Singh, V.; Sakaki, S.; Deshmukh, M. M. Organometallics 2018, 37, 1258-1270. (b) Yang, X.; Wang, C. Angew. Chem. Int. Ed. 2018, 57, 923-928. (c) Fang, S.; Chen, H.; Wang, W.; Wei, H. New J. Chem. 2018, 42, 4923-4932. (d) Lam, Y. C.; Nielsen, R. J.; Goddard, W. A.; Dash, A. K. Dalton Trans. 2017, 46, 12507-12515. (e) Mukhopadhyay, T. K.; Rock, C. L.; Hong, M.; Ashley, D. C.; Groy, T. L.; Baik, M-H.; Trovitch, R. J. J. Am. Chem. Soc. 2017, 139(13), 4901-4915. (3) Recent examples of metal-catalyzed hydrosilylation: (a) Wu, H.; Zheng, C.; Chen, N.; Zhu, J.; Gao, D. Tetrahedron Lett. 2017, 58, 1576-1578 (alkene). (b) Skrodzki, M.; Witomska, S.; Pawluc, P. Dalton Trans. 2018, 47, 5948-5951 (alkyne). (c) Kumon, T.; Binti Mohd. Sari, S. A.; Yamada, S.; Konno, T. J. Fluorine Chem. 2018, 206, 8-18 (ketones). (d) Li, B.; Zhang, S.; Wu, W.; Liang, L.; Jiang, S.; Chen, L.; Li, Y. RSC Advances 2017, 7, 31795-31799 (imines). (e) Pérez, D. E.; Smeltz, J. L.; Sommer, R. D.; Boyle, P. D.; Ison, E. A. Dalton Trans., 2017, 46, 4609-4616 (aldehydes). (4) Chalk, A. J.; Harrod, J. F. J. Am. Chem. Soc. 1965, 87,16−21. (5) Harrod, J. F.; Chalk, A. J. J. Am. Chem. Soc. 1965, 87, 1133−1133. (6) Duckett, S. B.; Perutz, R. H. Organometallics 1992, 11, 90−98. (7) Ojima, I.; Kogure, T.; Kumagai, M.; Horiuchi, S.; Sato, Y. J. Organomet. Chem. 1976, 122, 83. (8) Kennedy-Smith, J. J.; Nolin, K. A.; Gunterman, H. P.; Toste, F. D. J. Am. Chem. Soc. 2003, 125, 4056-4057. (9) Nolin, K. A.; Krumper, J. R.; Pluth, M. D.; Bergman, R. G.; Toste, F. D. J. Am. Chem. Soc. 2007, 129, 14684. (10) Chung, L. W.; Lee, H. G.; Lin, Z. Y.; Wu, Y. D. J. Org. Chem. 2006, 71, 6000– 6009. (11) Song, J. S.; Bullock, R. M. J. Am. Chem. Soc. 1994, 126, 8602-8612. (12) Voges, M.; Bullock, R. M. J. Am. Chem. Soc. 2000, 122, 12594-12595. (13) Voges, M.; Bullock, R. M. J. Chem. Soc. Dalton Trans., 2002, 759-770. (14) Dioumaev, V. K.; Bullock, R. M. Nature 2003, 424 , 530. (15) Du, G.; Franwick, P. E.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007, 129, 5180-5187.

(16) Du, G.; Abu-Omar, M. M. Organometallics 2006, 25, 4920-4923. (17) Ison, E. A.; Trivedi, E. R.; Corbin, R. A.; Abu-Omar, M. M. J. Am. Chem. Soc. 2005, 127, 15374-15375. (18) Metsänen, T. T.; Hrobμrik, P.; Klare, H. F. T.; Kaupp, M.; Oestreich, M. J. Am. Chem. Soc. 2014, 136, 6912–6915. (19) Rendler, S.; Oestreich, M. Acid Catalysis in Modern Organic Synthesis, 2008. (20) Yang, J.; Brookhart, M. J. Am. Chem. Soc. 2007, 129, 12656-12657. (21) Park, S.; Brookhart, M. Organometallics 2010, 29, 6057-6064. (22) Gutsulyak, D. V.; Vyboishchikov, S. F.; Nikonov, G. I. J. Am. Chem. Soc. 2010, 132, 5950-5951. (23) Wang, J.; Huang, L.; Yang, X.; Wei, H. Organometallics, 2015, 34, 212-220. (24) Yang, Y-F.; Chung, L. W.; Zhang, X.; Houk, K.; Wu, Y-D. J. Org. Chem. 2014, 79, 8856−8864. (25) Wang, P. G.; Gu, P.; Wang, Y.; Wei, H. Organometallics 2014, 33, 847−857 (26) Gu, P.; Wang, W.; Wang, Y.; Wei, H. Organometallics, 2013, 32, 47-51. (27) Huang, L.; Wang, W.; Wei, X.; Wei, H. J. Phys. Chem. A, 2015, 119 (16), 3789-3799. (28) Huang, L.; Zhang, Y.; Wei, H. Eur. J. Inorg. Chem., 2014, 2014, 5714-5723. (29) Wang, J.; Wang, W.; Huang, L.; Yang, X.; Wei, H. ChemPhysChem, 2015, 16, 1052- 1060. (30) Huang, L.; Wang, W.; Wei, H. J. Mol. Catal. A: Chem., 2015, 400, 31–41. (31) (a) Choualeb, A.; Maccaroni, E.; Blacque, O.; Schmalle, H. W.; Berke, H. Organometallics 2008, 27, 3474−3481. (b) Dong, H.; Berke, H. Adv. Synth. Catal. 2009, 351, 1783−1788. (c) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. Organometallics 2012, 31, 5994−5997. (32) Pérez, D. E.; Smeltz, J. L.; Sommer, R. D.; Boyle, P. D.; Ison, E. A. Dalton Trans., 2017, 46, 4609-4616. (33) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. Organometallics 2012, 31, 5994−5997. (34) Parr, R. G.; Yang, W., Density Functional Theory of Atoms and Molecules; Oxford University Press: New York, 1989. (35) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-52. (b) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B: Condens. Matter 1992, 46, 6671-6687. (c) Burke, K.; Perdew, J.; Wang, Y., Derivation of

a Generalized Gradient Approximation: The PW91 Density Functional. In Electronic Density Functional Theory, Dobson, J.; Vignale, G.; Das, M., Eds. Springer US: 1998; pp 81-111. (36) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H., J. Chem. Phys. 2010, 132 (15), 154104. (37) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215-241. (38) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chem. Acc. 1990, 77, 123−141. (39) Gaussian 09, Revision A.2, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc., Wallingford CT, 2009. (40) Nagaraja, C. M.; Parameswaran, P.; Jemmis, E. D.; Jagirdar, B. R. J. Am. Chem. Soc. 2007, 129, 5587-5596. (41) Rassolov, V. A.; Pople, J. A.; Ratner, M. A.; Windus, T. L. J. Chem. Phys. 1998, 109, 1223−1229. (42) Ehlers, A. W.; Bö hme, M.; Dapprich, S.; Gobbi, A.; Höllwarth, A.; Jonas, V.; Kö hler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G. Chem. Phys. Lett. 1993, 208, 111−114. (43) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654−3665.

Chapter 4: Reactivity of Novel DAAm Rhenium(III) Acetate Complex Towards Lewis Acidic Boron Reagents

Unpublished results

Department of Chemistry, North Carolina State University Raleigh, North Carolina, 27695-8204

4.1 Abstract

The reactivity of the rhenium(III) complex (DAAm-aryl)Re(CO)(OAc) (DAAm = N,N- bis(2-arylaminoethyl)methylamine; aryl = C6F5) (3a) with Lewis acidic boron reagents was investigated. The reaction of 3a with tris(pentafluorophenyl)boron (B(C6F5)3) resulted in the formation of new DAAm rhenium(III) and DAAm rhenium(V) complexes [(DAAm-

C6F5)Re(O)(C6F5)] (18) and [DAAm-C6F5Re(CO)(C6F5)] (19). Also [(DAAm-

C6F5)Re(CO)(OB(C6F5)2OC(CH3)O)] (21) and [(DAAm)Re(O)(OC(CH3)OB(C6F5)3)] (17) were isolated and may be potential intermediates in the formation of 18 and 19. The role and reactivity of water in these transformations are described. Experimental data and previous reports are used to propose a mechanism for the acetate ligand activation.

4.2 Introduction

Since our group reported1,2 the synthesis of the rhenium(III) acetate complex (DAAm- aryl)Re(CO)(OAc) (DAAm = N,N-bis(2-arylaminoethyl)methylamine; aryl = C6F5 (3a), Mes (3b)) (Scheme 4.1) we have focused in the investigation of the effect of the valence on the metal center on the reactivity of these rhenium complexes. The synthesis of complexes 3a-b have allowed us to study and compare the reactivity of Re(V) and Re(III) complexes with organic substrates. The knowledge and new insight gained from these studies can enable future work that will lead to the design of more efficient catalysts. Scheme 4.1 Synthesis of DAAm rhenium(III) acetate complexes 3a and 3b previously reported by our group.

Direct Insertion 1, 2- Acyl Migration

R R R CO O CH O N 3 CO, 60 psi C CO, 60 psi N R N Re Re O Re OAc C6D6, 3 d C6D6, 80 °C, 8h N N N R N N R N

(3a) R = C6F5 (3b) R = Mes

The first examples of DAAm (DAAm = N,N-bis(2-arylaminoethyl)methylamine; aryl = 2 C6F5, Mes) rhenium(V) mono-oxo d complexes where the oxo ligand exhibits ambiphilic reactivity were reported by our group.3 The synthesis of Lewis acid-base adducts of tris(pentafluorophenyl)boron (B(C6F5)3) with [DAAmRe(O)(X)] (X = CH3, COCH3, Cl) in which the mono-oxo ligand is activated were described (Scheme 4.2 right). The terminal oxo ligand acts as nucleophile towards a Lewis acid. In addition, it was shown that the stoichiometric reaction of

[DAAmRe(O)(Cl)] (aryl = Mes) with triarylphospines (PAr3) in the presence of CO results in the formation of triarylphosphine oxide (OPAr3) (Scheme 4.2 left). The terminal oxo in this case acts as an electrophile towards the Lewis base. The surprising reactivity of [DAAmRe(O)(X)] complexes towards B(C6F5)3 inspired us to study the reactivity of the novel DAAm rhenium(III) acetate complex 3a with Lewis acidic boron reagents.

Scheme 4.2 Synthesis of Lewis acid/base adducts previously reported by the Ison group.

R O X N Re N N R B(C 3 6 F R = Mes, C F 5 PAr 6 5 )3 X = Me, COCH3, Cl B(C F ) Mes CO CO 6 5 3 C F 6 5 O Mes N X Re Cl N Re N

N N N C6F5

III (C F ) B OReVCH (CO)Re Cl + OPPh3 6 5 3 3

In this chapter we investigated the reactivity of 3a with the Lewis acid B(C6F5)3 in order to understand the mechanism. In addition, rhenium(V) and rhenium(III) complexes bearing a Re-

C6F5 bond are reported. The role and reactivity of water in these transformations are described. Experimental data and previous reports are used to propose a mechanism for the acetate ligand activation. Also, we identify possible intermediates in these transformations.

4.3 Results and Discussion

The most nucleophilic (basic) site on 3a was anticipated to be the acetate ligand. Therefore, activation of a Lewis acid at this location was anticipated. The overnight reaction of complex 3a with 1.2 equiv of B(C6F5)3 at room temperature in methylene chloride is shown in Scheme 4.3 (Complete procedure in Experimental Section). The formation of [(DAAm- aryl)Re(O)(OC(CH3)OB(C6F5)3)] (aryl = C6F5; 17) was confirmed by X-ray crystallography (Figure 4.1). In this reaction, activation of the acetate ligand in 3a was observed. Unexpectedly, the X-ray structure of complex 17 also showed the decarbonylation and oxidation of the metal center from rhenium(III) to rhenium(V).

Scheme 4.3 Preliminary investigation of the reactivity of 3a towards Lewis acidic boron reagents.

H C C F 3 6 5 CO OB(C6F5)3 C6F5 C F O 6 5 N 1.2 equiv B(C6F5)3 O Re OAc N Re N CH2Cl2, rt, 1 d

N N N C6F5

3a 17

Identified by X-ray

The geometry around rhenium in complex 17, is best described as a distorted square pyramid with the oxo ligand (O1) in the apical position and the acetate-borane ligand (O2-C18- O3-B1) in the basal plane. The Re-O1 bond length (1.6821 Å) is consistent with the assignment of a triple bond and rhenium(V) mono-oxo complexes reported by our group, which range between 1.68-1.69 Å.4 The C18-O3-B1 bond angle in this complex is 129 ̊, the distortion of the angle is likely due to the steric hindrance of the three pentafluorophenyl substituents on the boron ligand.

Figure 4.1 Thermal ellipsoid plot of 17. Thermal ellipsoids are at 50%. H atoms have been omitted for clarity. Pentafluorophenyl rings on B(C6F5)3 and at the ligand framework are depicted in wireframe for clarity. Selected bond lengths (Å) and angles (deg): Re1-O1, 1.6821(12); Re1-N1, 1.9424(14); Re1-Ne, 1.943(3); Re1-N2, 2.1212(13); Re1-O2, 2.0622(11); O2-C18, 1.273(4); O3- B1, 1.5447(19); N1-Re1-N3, 117.00(6); N1-Re1-N2, 80.45(5); N1-Re1-O2, 86.68(5); O1-Re1- N1, 122.01(12); O1-Re1-N2, 97.23(5); O1-Re1-O2, 105.05(5); C18-O3-B1, 129.07(13).

Complex 17 was independently synthesized according to the reaction in Scheme 4.4. 1 Complex [(DAAm-C6F5)Re(O)(OAc)], previously synthesized in our lab, was treated with five equivalents of B(C6F5)3 in methylene chloride at room temperature in a nitrogen filled glove box. After 30 min, excess hexanes were added to precipitate complex 17 as a dark blue powder.

Scheme 4.4 Synthesis of [(DAAm)Re(O)(OC(CH3)OB(C6F5)3)], 17.

H3C C6F5 OAc OB(C6F5)3 C F 6 5 O C6F5 N 5 equiv B(C F ) O Re O 6 5 3 N Re N CH2Cl2, rt, 30 min N N N C6F5

The 1H NMR spectrum of 17 is shown in Figure 4.2. The methylene protons from the ligand backbone in 17 resonate as three distinct multiplets: 4.20 (2H), 3.55 (4H) and 3.32 ppm (5H). A distinctive singlet at 1.76 ppm is representative of the methyl group on the acetate ligand in 17.

1 Figure 4.2 H NMR (400 MHz, CD2Cl2) spectrum of [(DAAm)Re(O)(OC(CH3)OB(C6F5)3] (17). Residual unidentified impurity peaks observed at 2.68 ppm, 2.30 ppm, and 2.19 ppm.

Attempts to reproduce the reaction in Scheme 4.3 were unsuccessful (Scheme 4.5). Instead, we identified by X-ray crystallography the DAAm rhenium(V) mono-oxo complex 18 (Figure 4.3). The geometry around rhenium in 18 is best described as distorted square pyramidal with the oxo ligand (O15) in the apical position and the diamidoamine ligand in the equatorial plane. The Re-O15 bond length (1.6849 Å) is consistent with rhenium(V) mono-oxo values 4 V III 5-8 previously reported. Complex 18 contains a new O≡Re -C6F5 bond. Examples of Re -C6F5 , V 4b VII 9 Re -C6F5 , and Re -C6F5 complexes have been previously reported. The Re1-C7 (2.107 Å) bond length in 18 is consistent with rhenium(V) carbon bond (2.113 Å) reported by our group.4

Scheme 4.5 Attempts to reproduce previous results by our group.

C6F5 CO C F 6 5 O C6F5 C6F5 N 1.2 equiv B(C F ) N Re OAc 6 5 3 Re + HC6F5 N CH2Cl2, rt N N C F N 6 5

3a 18

Figure 4.3 X-ray structure of 18. Ellipsoids are at the 50% probability level. Hydrogen atoms were omitted and the pentafluorophenyl substituents on the diamido ligand are depicted in wireframe for clarity. Selected bond lengths (Å) and bond angles (deg.): Re-O15, 1.6849(1); Re-N1, 1.9633(1); Re-N2, 2.1576(1); Re-N3, 1.9664(1); Re-C7, 2.1070(1).

We hypothesize that 17 is the precursor to complex 18 (Scheme 4.6). The transformation v involves the cleavage of two bonds, the B-C6F5 and Re -O, and the corresponding formation of V two new bonds, the B-O and Re -C6F5. Transfer reactions involving the C6F5 group are well known in transition-metal chemistry.10-12 The activation of the boron-carbon linkage is promoted by bond polarization. As outlined in Scheme 4.6, we propose a six-membered transition state prior to transfer of one C6F5 group on the boron to the rhenium(V) metal center. The B-O bond is stronger than the B-C6F5 bond. Therefore, thermodynamically there is a significant driving force for this reaction. Contrary to a previous literature report where a large activation barrier for 1,2- 12 transfer of C6F5 group in a silylene-borane adduct to silylborane in three months was observed , in this investigation, the C6F5 group transfer was observed in 14 d.

Scheme 4.6 Proposed reaction pathway for the formation of 18.

TS H3C OB(C F ) 6 5 3 C F C O C6F5 6 5 O C6F5 O O O O C F N N B(C6F5)2 N 6 5 Re Re Re C6F5 N N C6F5 N N N N C6F5 C6F5

17 18

The independent synthesis of complex 17 (Scheme 4.4) allowed us to investigate the proposed transformation shown in Scheme 4.6. The decomposition of complex 17 was monitored by 1H NMR spectroscopy for four days (Figure 4.4). Complex 17 was dissolved in 0.35 mL of 1 methylene chloride-d2 in a screwcap NMR tube. H NMR spectra were taken at a fixed period of time. Comparison of the 1H NMR spectra showed the disappearance of 17 (blue arrows) and appearances of new set of multiplets overtime (red arrows). We identified these new multiplets as the formation of 18 in solution. Also, we identified the formation of pentafluorobenzene (HC6F5) around 7 ppm as a side product. A distinct singlet around 2.07 ppm was identified as – 13 [MeCO2(B(C6F5)3)2] as previously reported by Mitu et. al.

Figure 4.4 Decomposition of complex 17 monitored by 1H NMR spectroscopy overtime.

The reaction of 3a was investigated over shorter time periods and with increased equivalents of the Lewis acid reagent, B(C6F5)3 (Scheme 4.7). Slow diffusion of pentane into a concentrated reaction mixture in methylene chloride afforded brown-red crystals. The formation of complex [(DAAm-C6F5)Re(CO)(C6F5)] (19) was confirmed by X-ray crystallography. As III shown in the X-ray structure, 19 contains a new Re -C6F5 bond (Figure 4.5). Also, the structure revealed that the carbonyl ligand is still bound to the metal center, indicating that rhenium has not been oxidized. The geometry around rhenium is best described as distorted trigonal bipyramidal with the carbonyl ligand (C24-O1) in the apical position and the pentafluorophenyl ligand in the equatorial position. The Re1-C24 (1.86 Å) bond length is comparable to the bond length value of complex 3a.2 The Re1-C18 (2.1090 Å) bond length is quite shorter (approximately 0.1 Å shorter) III 5-8 than previous reported Re -C6F5 values, which range from 2.174-2.203 Å. We attribute this to

72 the ligands present in each of the previous examples reported in the literature. For example, in 7 complex [Re(η-C5H5)(η-C5H4-{B(C6F5)2}(C6F5)] strong binding interactions of the cyclopentadienyl ligands with the metal center contribute to the elongation of the Re-carbon bond in comparison to complex 19.

Scheme 4.7 Synthesis of [(DAAm-C6F5)Re(CO)(C6F5)] (aryl = C6F5), 19.

C6F5 CO C6F5 CO F F C6F5 N 5 - 10 equiv B(C F ) C6F5 N Re OAc 6 5 3 Re F N C6H6, rt, 1 h N F F N N

3a 19

Figure 4.5 X-ray structure of 19. Ellipsoids are at the 50% probability level. Hydrogen atoms were omitted and the pentafluorophenyl substituents on the diamidoamine ligand are depicted in wireframe for clarity. Selected bond lengths (Å) and bond angles (deg.): Re1-C24, 1.8657(14); Re1-N1, 1.9121(11); Re1-N3, 1.9189(12); Re1-C18, 2.1090(13); Re1-N2, 2.2353(11); C24-Re1- N1, 94.98(5); C24-Re1-N3, 96.54(5); N1-Re1-N3, 120.23(5); C24-Re1-C18, 90.49(5); N1-Re1- C18, 115.89(5); N3-Re1-C18, 122.40(5); C24-Re1-N2, 173.10(5); N1-Re1-N2, 81.09(5); N3-Re1- N2, 80.72(5); C18-Re1-N2, 96.32(5).

Complex 19 was independently synthesized according to the reaction in Scheme 4.8. The 1H NMR spectrum of 19 revealed the presence of a new set of multiplets (Figure 4.6). A distinctive singlet at 1.64 ppm is representative of the methyl group on the amine in 19. The methylene protons from the ligand backbone in 19 resonate as three distinct multiplets: 3.10 (2H), 2.91 (2H), and 1.83 (4H) ppm.

Scheme 4.8 Synthesis of [(DAAm-C6F5)Re(CO)(C6F5)], 19.

C6F5 CO C6F5 CO C F C F 6 5 N 1 equiv Zn(C6F5)2 6 5 N Re OAc Re C6F5 N C6D6, rt, 30 min N N N

3a 19

1 Figure 4.6 H NMR (300 MHz, C6D6) spectrum of [(DAAm-C6F5)Re(CO)(C6F5)] (19).

We propose that complex 19 is a precursor to 18 via a decarbonylation/oxidation mechanism (Scheme 4.9). Pentafluorophenylboronic acid (C6F5B(OH)2) and pentafluorobenzene

(HC6F5) were also identified in reaction mixture with 19 by X-ray crystallography. The synthesis of boronic acids occurs by hydrolysis. Thus, the identification of (pentafluorophenyl)boronic acid suggests that the presence of water is influential in this reaction. Water is proposed to play a two-

part role in the reaction shown in Scheme 4.3 and Scheme 4.9. First its reaction with B(C6F5)3 will form the active species in this reaction and second, it is the origin of the oxygen atom for the oxo ligand in 17 and 18.

Scheme 4.9 Proposed decarbonylation for the formation of 18 from 19.

C6F5 CO C F F F 6 5 O C6F5 C6F5 N [O] N Re F Re N F F N N C F N 6 5

19 18

Tris(pentafluorophenyl)borane monohydrate [H2O·B(C6F5)3] (20) was identified by X-ray crystallography from a crude reaction mixture of the reaction between 3a and two equivalents of

B(C6F5)3. Compound 20 is known and there are many reports in the literature describing its reactivity.14,15 This observation suggests that 20 plays an essential role in this reaction. Compound 20 was independently synthesized according to the reaction in Scheme 4.10. Tris(pentafluorophenyl)borane was treated with one equivalent of distilled water at room temperature. After 2 h, 20 precipitated as a colorless powder in quantitative yields. The 19F NMR spectrum of 20 in benzene-d6 displays three signals at -135.3, -155.9 and -163.6 ppm corresponding to the ortho-, para- and meta- fluorine respectively (Figure 4.7). The chemical shifts are in agreement with literature reports.7,16,17

Scheme 4.10 Synthesis of tris(pentafluorophenyl)borane monohydrate, 20.

F5 1 equiv H2O H B O B F F5 pentanes, rt, 2 h H 5 F5 F5

19 Figure 4.7 F NMR (376 MHz, C6D6) spectrum of tris(pentafluorophenyl)borane monohydrate (20).

To confirm the transformation proposed on Scheme 4.9, complex 19 was treated with one equivalent of tris(pentafluorophenyl)borane monohydrate [H2O·B(C6F5)3] (20) (Scheme 4.11). Slow diffusion of acetonitrile into a concentrated reaction solution in benzene afforded X-ray quality crystals. Crystallographic studies confirmed the formation of complex 18.

Scheme 4.11 Reactivity of 19 with [H2O·B(C6F5)3]

C6F5 CO C F 6 5 O C F C6F5 6 5 N 1 equiv H2O·B(C6F5)3 N Re C6F5 Re N C D , rt 6 6 N N C F N 6 5

19 18

Confirmed by X-ray

To discard any other possible oxygen source, complex 19 was pressurized in the presence of oxygen (Scheme 4.12). No reactivity was observed after four days. These results suggest that complex 19 is the precursor of complex 18 in the presence of [H2O·B(C6F5)3] (20) as proposed on Scheme 4.9.

Scheme 4.12 Reactivity of 19 in the presence of oxygen

C6F5 CO

C6F5 N 40 psi O Re C F 2 6 5 No Reaction N C6D6, rt, 4 d N

The reactivity of 20 with 3a was investigated (Scheme 4.13). Addition of five equivalents of 20 to 3a resulted in the formation of a new rhenium(III) complex [(DAAm-

C6F5)Re(CO)(OB(C6F5)2OC(CH3)O)] (21). Three multiplets corresponding to the ligand backbone at 3.11, 2.91 and 1.83 ppm and two singlets corresponding to the methyl groups at 1.64 and 1.56 1 ppm were observed in the H NMR spectrum in benzene-d6 (Figure 4.8). X-ray quality green- brown crystals of 21 were obtained by slow diffusion of hexanes into a concentrated solution of the crude reaction mixture in benzene (Figure 4.9).

Scheme 4.13 Probing 20 as the active species in the reaction of 3a towards Lewis acidic boric reagents.

C6F5 C F C6F5 CO C6F5 CO 6 5 B O C6F5 N 5 equiv H O·B(C F ) C6F5 N O Re 2 6 5 3 Re OAc O N C6H6, rt, 1 d N N N

21 3a

1 Figure 4.8 H NMR (300 MHz, C6D6) spectrum of [(DAAm-C6F5)Re(CO)(OB(C6F5)2OC(CH3)O] (21).

The geometry about rhenium in 21 is best described as a distorted octahedron where the carbonyl ligand (C6-O1) occupies the axial position and the bidentate ligand (O3-B1-O3-C19-O4) lies in the equatorial plane. The Re-O2 bond length is 2.199 Å, while Re-O4 bond length (2.094 Å) is shorter by 0.105 Å. The differences can be attributed to the bulkiness of the two pentafluorophenyl substituents bound to the borane. The O3-C19 (1.286 Å) and O4-C19 (1.239 Å) bond lengths are in agreement with C-O double bond, suggesting resonance through O3-C19-

O4 characteristic of an acetate ligand. Pentafluorobenzene (C6F5H) was also identified from the crude reaction mixture by X-ray crystallography. The metal center in 21 was not oxidized and activation of the acetate ligand in 3a was observed.

Figure 4.9 X-ray structure of complex 21. Ellipsoids are at the 50% probability level. Hydrogen atoms were omitted and the pentafluoro substituents on the diamido ligand are depicted in wireframe for clarity. Selected bond lengths (Å) and bond angles (deg.): Re1-C6, 1.863(3); Re1- N3, 1.920(2); Re1-N1, 1.947(2); Re1-O4, 2.0946(19); Re1-O2, 2.1996(17); Re1-N2, 2.223(8); Re1-N2A, 2.26(3); O1-C6, 1.168(3); O2-B1, 1.489(3); O3-C19, 1.286(3); O3-B1, 1.506(4); O4- C19, 1.239(3); C6-Re1-N3, 93.86(11); C6-Re1-N1, 96.25(10); N3-Re1-N1, 107.51(10); C6-Re1- O4, 95.88(10); N3-Re1-O4, 90.91(9); N1-Re1-O4, 157.13(8); C6-Re1-O2, 97.30(9); N3-Re1-O2, 162.89(9); N1-Re1-O2, 84.17(8); O4-Re1-O2, 75.11(7); C6-Re1-N2, 173.56(14); N3-Re1-N2, 80.70(14); N1-Re1-N2, 82.22(13); O4-Re1-N2, 87.69(12); O2-Re1-N2, 88.79(13); C6-Re1-N2A, 168.4(4); N3-Re1-N2A, 80.4(4); N1-Re1-N2A, 76.1(4); O4-Re1-N2A, 94.3(4); O2-Re1-N2A, 90.7(4); B1-O2-Re1, 125.14(16); C19-O3-B1, 124.4(2); C19-O4-Re1, 140.17(19); O1-C6-Re1, 175.8(2).

Complexes 19 and 21 both possess the carbonyl ligand and the metal center is not oxidized. Experimental observations suggest that these are the two characteristics for intermediates present in an early stage of the reaction. The proposed mechanism for the transformation of 21 to 19 is shown in Scheme 4.14. The transformation consists of (1) rearrangement and C6F5 group transfer into the Re-O bond, and (2) formation and hydrolysis of an oxoborane to lead bis(pentafluorophenyl)boronic acid as previously reported by Ito et al18.

Scheme 4.14 Proposed mechanism pathway for the transformation of 21 to 19.

C6F5 C6F5 O C6F5 C F C6F5 CO C6F5 CO C F 6 5 CO B B 6 5 B O O C6F5 C6F5 C F C6F5 N O C6F5 N N O 6 5 Re Re O Re O O O N N N N N N

O C6F5 CO B C6F5 CO C F C F 6 5 N 6 5 O C6F5 Re C6F5 N O Re C F N 6 5 N N N

19 HO OH B O O O OH O H2O H2O F F B B + O C6F5 O C6F5 F F OH F

Wanglee et. al. reported the borane-induced dehydration of silica by water-catalyzed 15 grafting of B(C6F5)3. The Wanglee group reported that 20 undergoes a hydrolysis reaction in solution to afford pentafluorobenzene (C6F5H) and bis(pentafluorophenyl)borinic acid

[(C6F5)2BOH] (22). Compound 22 was proposed to be the active species for the formation of

≡SiOB(C6F5)2. To investigate whether 20 behaves in a similar manner in our investigation, we synthetized 22 as shown in Scheme 4.15.19 One equivalent of distilled water was added to tris(pentafluorophenyl)borane at room temperature in benzene. The reaction mixture was stirred and heated for 4 h at 100 °C. Compound 22 was obtained as a white precipitate in quantitative yields. Three resonances at -133.1, -147.9 and -161.1 ppm were observed in the 19F NMR spectrum in benzene-d6, which corresponds to the ortho-, para- and meta- fluorines (Figure 4.10). The chemical shifts agree with literature values.15

Scheme 4.15 Synthesis of bis(pentafluorophenyl)borinic acid, 22.

F5 F5 1 equiv H2O B HO B F5 C6H6, 100 °C, 4 h F5 F5

19 Figure 4.10 F NMR (376 MHz, C6D6) spectrum of bispentafluoroborinic acid (22).

No reaction was observed when five equivalents of 22 were added to a solution of 3a in benzene at ambient temperature after a day (Scheme 4.16). The observation suggests that 22 is not a relevant species in this reaction.

Scheme 4.16 Probing 22 as the active species in the reaction of 3a towards Lewis acidic boron reagents.

C6F5 CO C F 6 5 N 5 equiv HOB(C6F5)2 Re OAc No Reaction N C6H6, rt, 1 d N

In 1997 the Puddephatt group reported the first structure of a cationic palladium hydroxytris(pentafluorophenyl)boronate complex.14 They proposed that 20 acts as a strong acid - that forms the conjugate base, [HOB(C6F5)3] , which then coordinates to platinum. Bergquist et. - + al. calculated a pka of 8.6 for the equilibrium of H2O·B(C6F5)3 à HOB(C6F5)3 + H3O in acetonitrile. Such a value indicates that H2O·B(C6F5)3 must be regarded as a strong acid, with a 16 - strength comparable to that of HCl in acetonitrile. These findings suggest that [HOB(C6F5)3] is likely an important species in the transformation. Previously our group reported that the acetate ligand in 3a is bound in either a κ2 (3a) or a κ1 (3a') fashion to the rhenium metal center (Scheme 4.17).20 The isomerization involved in this reaction 3a’, affords an open coordination site.

Scheme 4.17 Equilibrium between the κ1 and κ2 isomers of 3a.

C F C F 6 5 CO O 6 5 CO O C6F5 N C6F5 N Re O Re N N O N N

3a' 3a

- The preliminary proposed mechanism for the reaction of 3a’ with [HOB(C6F5)3] is depicted in Scheme 4.18. First, the conjugate base of 20 coordinates to the rhenium(III) metal center. Then, release of pentafluorobenzene and activation of the acetate ligand occurs to lead complex 21.

- Scheme 4.18 Proposed transformation of 3a’ in the presence of the conjugated base [HOB(C6F5)3] to lead 21.

H2O·B(C6F5)3 C F C6F5 H 6 5 C6F5 C6F5 CO O C6F5 CO C6F5 CO B(C F ) B + - O 6 5 2 O C6F5 N H3O + [HOB(C6F5)3] C6F5 N C6F5 N O Re O Re O Re O O N N -HC6F5 N N N N

3a' 21

A summary of all intermediates isolated and proposed mechanism for each transformation is shown in Scheme 4.19. The early stage is characterized by rhenium(III) complexes and presence of the Re-CO bond. A mechanism corresponding to the late stage is still not yet fully understood and investigation of the reaction mechanism from 3a to 17 is still ongoing. Scheme 4.19 Intermediates identified in this investigation and proposed mechanism for each transformation.

C6F5 CO

C6F5 N Re OAc N N A D C B 3a

C6F5 H3C C F C F C F 6 5 CO 6 5 OB(C6F5)3 C6F5 6 5 CO B C F O O 6 5 C6F5 C6F5 N O O O N C6F5 N Re N Re Re C6F5 O N Re N N N C6F5 N N N C6F5 N

17 18 19 21

A = Insertion/ Decarbonylation B = Coordination C = Coordination/C6F5 group transfer D = C6F5 group transfer/Rearrangment/Decarbonylation

4.4 Conclusions

The reactivity of 3a towards B(C6F5)3 was investigated. Complexes 19 and 21 represent new examples of rhenium(III) and rhenium(V) complexes bearing a Re-C6F5 bond. Complexes 17, 19, and 21 were identified and proposed as intermediates in the transformation of complex 3a to 18. Our results offer a new perspective on the rearrangement and activation of boron-carbon bonds and insertion and transfer of the C6F5 group. 83

4.5 Experimental Section

General Considerations. [(DAAm-C6F5)Re(CO)(OAc)] and [(DAAm-C6F5)Re(O)(OAc)] were prepared according to previous published procedures.1 All reactions were carried out in a nitrogen filled glove box unless otherwise noted. All reagents were purchased from commercial sources, placed in a nitrogen filled glove box and used as received without further purification. 1H and 13C NMR spectra were acquired on a Varian Mercury 400 MHz or Varian Mercury 300 MHz spectrometer. NMR chemical shifts are listed in parts per million (ppm) and are referenced to residual protons or carbons of the deuterated solvents, respectively at room temperature unless otherwise noted. The FTIR spectra were obtained in KBr thin films on a JASCO FT/IR-4100 instrument. Elemental analyses were performed by Atlantic Micro Laboratories Inc. X-Ray crystallography was performed at the X-Ray Structural Facility of North Carolina State University by Dr. Roger Sommer.

Synthesis of complex [DAAmRe(O)[(OC(CH3)O)(B(C6F5)3)] (aryl = C6F5), 17.

In a nitrogen filled glove box, complex [(DAAm-C6F5)Re(O)(OC(CH3)O)] (25 mg, 0.035 mmol) was added to a screw cap vial and dissolved in 2 mL of methylene chloride. Then, B(C6F5)3 (90.1 mg, 0.176 mmol, 5 equiv) was added to the solution. The reaction mixture was allowed to stir at room temperature for 30 min. Excess of hexanes were added to the mixture and the resulting dark blue powder was filtered, washed with hexanes, and collected via vacuum filtration (30.5 mg, 71 1 % yield). H NMR (300 MHz, CD2Cl2) δ: 4.20 (m, 2H), 3.55 (m, 4H), 3.33 (m, 5H), 1.76 (s, 3H). 13 C NMR (150 MHz, CD2Cl2) δ: 181.01, 149.81, 149.09, 147.82, 147.17, 145.95, 145.09, 144.59, 143.86, 143.14, 142.60, 141.32, 141.02, 139.13, 138.40, 137.40, 137.13, 136.42, 132.60, 118.68, 114.19, 67.65, 67.42, 66.80, 65.91, 65.61, 58.65, 58.53, 57.82, 54.43, 54.22, 54.00, 53.78, 53.57, 19 43.25, 42.12, 23.41, 22.00. F NMR (376 MHz, CD2Cl2) δ: -134.64 (m, 6F), -135.90 (m, 2F), - 148.70 (m, 2F), -149.10 (m, 2F), -158.72 (t, J = 21.3 Hz, 2F), -159.12 (t, J = 20.2 Hz, 3F), -164.04 -1 (m, 2F), -165.31 (d, J = 23.1 Hz, 6F). FTIR (cm in KBr pellet) νC=O 1519 and 1469, νB-O 987. The complex was not characterized by elemental analysis due to poor stability.

Synthesis for [DAAmRe(CO)(C6F5)] (aryl = C6F5), 19.

Complex 3a (150 mg, 0.208 mmol) and Zn(C6F5)2 (83 mg, 0.208 mmol, 1 equiv) were added to a screw cap vial and dissolved in benzene. The reaction was stirred for 30 min at room temperature. The solvent was reduced under pressure. The resultant oil was dissolved in the minimum amount

84 of methylene chloride and excess of pentanes was added to the mixture. The final product precipitated as a brown powder, and then washed with hexanes, and collected via vacuum filtration 1 (122 mg, 71 % yield). H NMR (300 MHz, C6D6) δ: 3.18 (m, 2H), 2.98 (m, 2H), 1.90 (m, 4H), 19 1.71 (s, 3H). F NMR (376 MHz, C6D6) δ: -117.25 (ddd, J = 26.7, 20.2, 11.4 Hz, 2F), -149.85 (tdd, J = 23.4, 15.5, 6.4 Hz, 2F), -152.08 (m, 2F), -158.55 (t, J = 19.9 Hz, 1F), -158.97 (t, J = 22.2 Hz, 2F), -162.62 (m, 2F), -163.01 (td, J = 23.0, 3.4 Hz, 2F), -163.83 (td, J = 15.0, 7.2 Hz, 2F). IR -1 (KBr in cm ): νC=O 1905. Anal. Calc. for C24H11F15N3ORe: C, 34.79; H, 1.34; N, 5.07. Found: C,

33.77; H, 1.31; N, 4.84; crystalized with ½ equivalent of CH2Cl2.

15-17,21 Synthesis of Tris(pentafluorophenyl)borane Monohydrate [H2O∙B(C6F5)3], 20. Tris(pentafluorophenyl)borane (200 mg, 0.3906 mmol) was dissolved in 12 mL of pentane. DI

H2O (7 µL, 0.3906 mmol, 1 equiv) was added to the solution at room temperature and then stirred vigorously for 2 h. The white precipitate was separated, washed with hexanes and dried in vacuo overnight. The product was obtained as a colorless powder (208 mg, 100%) and stored in a nitrogen 19 filled glove box at -43 °C. F NMR (376 MHz in ppm, C6D6): -135.3 (m, 4F, ortho), -155.9 (m, 2F, para), -163.6 (m, 4F, meta). Results are consistent with previous reports in the literature.

Synthesis for [DAAmRe(CO)(OB(C6F5)2OC(CH3)O)] (aryl = C6F5), 21.

Complex 3a (5 mg, 0.007 mmoles) and B(C6F5)3 (18.0 mg, 0.0347 mmoles, 5 equiv; 71 mg, 0.1388 mmoles, 20 equiv) were added to a small crew cap vial and dissolved in benzene-d6. The reaction was stirred for 2 h up to 2 days at room temperature. The crude was filtered through celite and then silica. Slow diffusion of hexanes into a concentrated solution of the reaction mixture in a vial led 1 to the formation of green-brown X-ray quality crystals of the complex. H NMR (C6D6 in ppm, δ) 3.11 (m, 2H), 2.91 (m, 2H), 1.83 (m, 4H), 1.64 (s, 3H), 1.56 (s, 3H). Full characterization is still ongoing.

18 Synthesis of Bis(pentafluorophenyl)borinic Acid [HOB(C6F5)2], 22. Into a reaction vessel equipped with a stir bar, tris(pentafluorophenyl)borane (200 mg, 0.3906 mmol) and 1 equivalent of distillated water (7 µL, 0.3906 mmol) were added at room temperature and dissolved in benzene. The reaction mixture was heated at 100 °C for 4 hours. The resultant mixture was reduced under pressure and then dissolved in a minimal amount of benzene. A white precipitated crash out after the addition of hexanes. The product was obtained as a white powder

19 (208 mg, 100 %) and stored in a nitrogen filled glove box at -43 °C. F NMR (376 MHz, C6D6) δ: -133.1 (m, 4F, ortho), -147.9 (m, 2F, para), -161.1 (m, 4F, meta). Results in agreement with previous reports in the literature.

Procedure for the Reaction of [DAAmRe(CO)(OAc)] (aryl = C6F5) (3a) with B(C6F5)3: Reaction depicted in Scheme 4.3. In a nitrogen filled glove box complex 3a (100.1 mg, 0.170 mmol) was added to a screw cap vial and dissolved in a minimal amount of methylene chloride. Then, B(C6F5)3 (89.1 mg, 0.340 mmol, 1.2 equiv) was added to the solution. Slow diffusion of pentane into the reaction vial led to the formation of blue X-ray quality crystals of complex 17 overnight.

Procedure to Study the Decomposition of [DAAmRe(O)[(OC(CH3)O)(B(C6F5)3)] (aryl =

C6F5) (17) to 18: Reaction depicted in Scheme 4.6.

Complex 17 (10 mg, 0.008 mmol) was dissolved in 0.35 mL of methylene chloride-d2 in a screwcap NMR tube. The decomposition was monitored by 1H NMR spectroscopy for four days. 1H NMR spectra were taken every 2 h for the first 8 h.

Procedure to Study the Reactivity of [DAAmRe(CO)(C6F5)] (aryl = C6F5) (19) with

H2O∙B(C6F5)3 (20): Reaction depicted in Scheme 4.11.

In a nitrogen filled glove box complex 19 (10 mg, 0.012 mmol) and H2O∙B(C6F5)3 (20) (6.40 mg, 1 equiv) were added to a screw cap vial and dissolved in 0.35 mL benzene. The reaction was stirred for 1 d at room temperature. The reaction mixture was filtered through celite and then silica. The solvent was reduced under pressure. The resultant oil was dissolved in a minimum amount of benzene. Slow diffusion of acetonitrile into a concentrated solution of the reaction mixture in a small vial in benzene led to the formation of purple X-ray quality crystals of complex 18.

Procedure to Study Reactivity of [DAAmRe(CO)(C6F5)] (aryl = C6F5) (19) with oxygen: Reaction depicted in Scheme 4.12.

Complex 19 (5 mg, 0.006 mmol) was added to a J. Young NMR tube and dissolved in C6D6 (0.35 mL) at room temperature and was degassed. The reaction was pressurized with O2 (40 psi). The reaction was monitored by 1H NMR spectroscopy for up to 4 d.

General Procedure for X-ray Determination.

X-Ray structural determination was performed at the X-Ray Structural Facility of North Carolina State University by Dr. Roger Sommer. The samples were mounted on a nylon loop with a small amount of Paratone N oil. The frame integration was performed using SAINT.22 The resulting raw data were scaled, and absorption corrected using a multi-scan averaging of symmetry equivalent data using SADABS.23 Most non-hydrogen atoms were obtained from the initial solution. The remaining non-hydrogen atom positions were recovered from a subsequent difference Fourier map. The structural model was fit to the data using full matrix least-squares based on F2. The calculated structure factors included corrections for anomalous dispersion from the usual tabulation. The structure was refined using the XL program from SHELXTL24, graphic plots were produced using the NRCVAX crystallographic program suite. Additional information and other relevant literature references can be found in the reference section of the Facility's Web page (http://www.xray.ncsu.edu). Summary of the following crystal data is shown in Table 4.1 and Table 4.2.

X-ray Structure Determination for [DAAmRe(O)[(OC(CH3)O)(B(C6F5)3)] (aryl = C6F5), 17.

A blue-green plate-like specimen of C37H14BF25N3O3Re, approximate dimensions 0.122 mm x 0.184 mm x 0.367 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 14.51 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a triclinic unit cell yielded a total of 61812 reflections to a maximum θ angle of 30.50° (0.70 Å resolution), of which 12215 were independent (average redundancy 5.060, completeness 2 = 99.8%, Rint = 4.38%, Rsig = 4.04%) and 10589 (86.69%) were greater than 2σ(F ). The final cell constants of a = 11.1937(9) Å, b = 12.0125(11) Å, c = 17.4243(16) Å, α = 80.301(2)°, β = 72.607(2)°, γ = 64.245(3)°, volume = 2011.7(3) Å3, are based upon the refinement of the XYZ- centroids of 137 reflections above 20 σ(I) with 11.81° < 2θ < 54.79°. Data were corrected for absorption effects using the numerical method (SADABS). The ratio of minimum to maximum apparent transmission was 0.633. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.3880 and 0.6980. The final anisotropic full-matrix least- squares refinement on F2 with 820 variables converged at R1 = 3.37%, for the observed data and wR2 = 8.16% for all data. The goodness-of-fit was 1.081. The largest peak in the final difference electron density synthesis was 3.466 e-/Å3 and the largest hole was -1.386 e-/Å3 with an RMS

87 deviation of 0.143 e-/Å3. On the basis of the final model, the calculated density was 2.015 g/cm3 and F(000), 1172 e-.

X-ray Structure Determination for [DAAmRe(O)(C6F5))] (aryl = C6F5), 18.

A brown block-like specimen of C25.50H17F15N3ORe, approximate dimensions 0.087 mm x 0.104 mm x 0.156 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 34.42 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 75955 reflections to a maximum θ angle of 30.64° (0.70 Å resolution), of which 8810 were independent (average redundancy 8.621, completeness 2 = 99.5%, Rint = 5.02%, Rsig = 3.09%) and 7261 (82.42%) were greater than 2σ(F ). The final cell constants of a = 26.0646(9) Å, b = 12.0914(4) Å, c = 21.4328(12) Å, β = 122.1170(10)°, volume = 5721.0(4) Å3, are based upon the refinement of the XYZ-centroids of 904 reflections above 20 σ(I) with 22.80° < 2θ < 65.26°. Data were corrected for absorption effects using the numerical method (SADABS). The ratio of minimum to maximum apparent transmission was 0.936. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.5490 and 0.7020. The structure was solved and refined using the Bruker SHELXTL Software Package, using the space group C 1 2/c 1, with Z = 8 for the formula unit, C25.50H17F15N3ORe. The final anisotropic full-matrix least-squares refinement on F2 with 436 variables converged at R1 = 2.79%, for the observed data and wR2 = 6.17% for all data. The goodness-of-fit was 1.064. The largest peak in the final difference electron density synthesis was 4.317 e-/Å3 and the largest hole was -0.763 e-/Å3 with an RMS deviation of 0.130 e- /Å3. On the basis of the final model, the calculated density was 1.980 g/cm3 and F(000), 3272 e-.

X-ray Structure Determination for [DAAmRe(CO)(C6F5)] (aryl = C6F5), 19.

A brown-red block-like specimen of C24H11F15N3ORe, approximate dimensions 0.160 mm x 0.210 mm x 0.250 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 4.72 hours. The frames were integrated with the Bruker SAINT21 software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 99781 reflections to a maximum θ angle of 36.49° (0.60 Å resolution), of which 12345 were independent (average redundancy 8.083, completeness = 99.7%, 2 Rint = 3.25%, Rsig = 2.03%) and 10830 (87.73%) were greater than 2σ(F ). The final cell constants

88 of a = 14.0874(7) Å, b = 11.2861(6) Å, c = 15.8562(8) Å, β = 92.086(2)°, volume = 2519.3(2) Å3, are based upon the refinement of the XYZ-centroids of 9928 reflections above 20 σ(I) with 4.431° < 2θ < 72.50°. Scaling Data were corrected for absorption effects using the multi-scan method (SADABS22). The ratio of minimum to maximum apparent transmission was 0.696. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.3700 and 0.5040. The final anisotropic full-matrix least-squares refinement on F2 with 398 variables converged at R1 = 1.83%, for the observed data and wR2 = 4.26% for all data. The goodness-of-fit was 1.034. The largest peak in the final difference electron density synthesis was 1.387 e-/Å3 and the largest hole was -0.462 e-/Å3 with an RMS deviation of 0.106 e-/Å3. On the basis of the final model, the calculated density was 2.184 g/cm3 and F(000), 1576 e-.

X-ray Structure Determination for [DAAmRe(CO)(OB(C6F5)2OC(CH3)O)] (aryl = C6F5), 21.

A green-brown block-like specimen of C38H17BF20N3O4Re, approximate dimensions 0.101 mm x 0.124 mm x 0.297 mm, was used for the X-ray crystallographic analysis. Data collection The total exposure time was 18.57 hours. The frames were integrated with the Bruker SAINT21 software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 62258 reflections to a maximum θ angle of 28.32° (0.75 Å resolution), of which

9760 were independent (average redundancy 6.379, completeness = 99.8%, Rint = 3.39%, Rsig = 2.56%) and 8096 (82.95%) were greater than 2σ(F2). The final cell constants of a = 16.7553(5) Å, b = 10.8204(3) Å, c = 22.5120(7) Å, β = 106.0550(10)°, volume = 3922.2(2) Å3, are based upon the refinement of the XYZ-centroids of 102 reflections above 20 σ(I) with 3.940° < 2θ < 32.18°. Scaling Data were corrected for absorption effects using the numerical method (SADABS22). The ratio of minimum to maximum apparent transmission was 0.673. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.4460 and 0.7360. The final anisotropic full-matrix least-squares refinement on F2 with 655 variables converged at R1 = 2.47%, for the observed data and wR2 = 5.84% for all data. The goodness-of-fit was 1.058. The largest peak in the final difference electron density synthesis was 1.043 e-/Å3 and the largest hole was - 0.859 e-/Å3 with an RMS deviation of 0.097 e-/Å3. On the basis of the final model, the calculated density was 1.959 g/cm3 and F(000), 2232 e-.

Table 4.1 Selected Crystallographic Data and Collection Parameters for 17 and 18.

17 18

Emp. Form. C37H14BF25N3O3Re C25.50H17F15N3ORe Form. wt (g/mol) 1220.52 852.62 crystal sys triclinic monoclinic space group P -1 C 1 2/c 1 a (Å) 11.1937(9) 26.0646(9) b (Å) 12.0125(11) 12.0914(4) c (Å) 17.4243(16) 21.4328(12) V (Å3) 2011.7(3) 5721.0(4) Z 2 8 ρ (g/cm3) 2.015 1.980 crystal size (mm) 0.122 × 0.184 × 0.367 0.087 × 0.104 × 0.156 R1, w R2 0.0337, 0.0816 0.0279, 0.0617 GOF 1.081 1.064

Table 4.2 Selected Crystallographic Data and Collection Parameters for 19 and 21.

19 21

Emp. Form. C24H11F15N3ORe C38H17BF20N3O4Re Form. wt (g/mol) 828.56 1156.55 crystal sys monoclinic monoclinic space group P 1 21/n 1 P 1 21/c 1 a, Å 14.087(7) 16.755(5) b, Å 11.286(6) 10.820(3) c, Å 15.856(8) 22.512(7) V, (Å3) 2519.3(2) 3922.2(2) Z 4 4 ρ (g/cm3) 2.184 1.959 crystal size (mm) 0.160 × 0.210 × 0.250 0.101 × 0.124 × 0.297 R1, w R2 0.0183, 0.0426 0.0247, 0.0584 GOF 1.034 1.058

4.6 References

(1) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc. 2011, 133, 13288-13291. (2) Smeltz, J. L.; Webster, C. E.; Ison, E. A. Organometallics 2012, 31, 4055-4062. (3) Smeltz, J. L.; Lilly, C. P.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc. 2013, 135, 9433- 9441. (4) a) Feng, Y.; Aponte, J.; Houseworth, P. J.; Boyle, P. D.; Ison, E. A. Inorg. Chem., 2009, 48(23), 11058-11066.; b) Lambic, N. S.; Sommer, R. D.; Ison, E. A. J. Am. Chem. Soc. 2016, 138, 4832-4842. (5) Klahn, A. H.; Moore, M. H.; Perutz, R. N. J. Chem. Soc., Chem. Commun. 1992, 23, 1699- 1701. (6) Klahn, A. H.; Oelckers, B.; Godoy, F.; Garland, M. T.; Vega, A.; Perutz, R. N.; Higgitt, C. J. Chem. Soc., Dalton Trans., 1998, 3079-3086. (7) Doerrer, L. H., Graham, A. J.; Haussinger, D.; Green, M. L. H. J. Chem. Soc., Dalton Trans., 2000, 813-820. (8) Zhu, B.; Huang, X.; Hao, X. Dalton Trans., 2014, 43, 16726-16736. (9) Lai, Y-Y.; Bornand, M.; Chen, P. Organometallics 2012, 31, 7558-7565. (10) Kowalewski, M.; Krumm, B.; Mayer, P.; Schulz, A.; Villinger, A. Eur. J. Inorg. Chem. 2007, 5319–5322. (11) Wondimagegn, T.; Xu, Z.; Vanka, K.; Ziegler, T. Organometallics 2005, 24, 2076–2085. (12) Metzler, N.; Denk, M. Chem. Commun. 1996, 2657–2658. (13) Mitu, S.; Baird, M.C. Can. J. Chem., 2006, 84, 214-224. (14) Hill, G. S.; Manojlovic-Muir, L.; Muir, K. W.; Puddephatt, R. J. Organometallics. 1997, 16, 525–530. (15) Wanglee, Y. J.; Hu, J.; White, R. E.; Lee, M. Y.; Stewart, S. M.; Perrotin, P.; Scott, S. L. J. Am. Chem. Soc. 2012, 134, 355-366. (16) Bergquist, C.; Bridgewater, B. M.; Harlan, J.; Norton, J. R.; Friesner, R. A.; Parkin, G. J. Am. Chem. Soc, 2000, 122, 10581-10590. (17) Liesener, F. P.; Jannsen, M. K. Synthesis. 2006, 15, 2590-2602. (18) Ito, M.; Tokitoh, N.; Okazaki, R. Tetrahedron Lett., 1997, 38, 4451-4454. (19) Toshiya, I.; Ikuyo, I.; Hitoshi, M.; Toshimitsu, M. Process for preparing bis(fluoroaryl)boron derivatives. Patent WO2000037476A1, June 2, 2002. (20) Smeltz, J. L. Ph. D. Dissertation, North Carolina State University, 2013.

(21) Doerrerm L. H.; Green, M. L. H. J. Chem. Soc., Dalton Trans. 1999, 4325-4329l. (22) Bruker-Nonius, SAINT version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA. (23) Bruker-Nonius, SADABS version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA. (24) Bruker-AXS, XL version 2009.9, 2009, Bruker-AXS, Madison, WI 53711, USA.

Chapter 5: Synthesis and Characterization of (DAAm)ReIII(CO)(R) Complexes, and Study of CO Insertion into Rhenium(III)-Carbon Bond

Unpublished results

Department of Chemistry, North Carolina State University Raleigh, North Carolina, 27695-8204

5.1 Abstract

The complexes [(DAAm-aryl)Re(CO)(R)] [DAAm = N,N-bis(2- arylaminoethyl)methylamine; aryl = C6F5 or Mes], 23a (aryl = C6F5; R = CH2Ph), 23b (aryl = Mes;

R = CH2Ph), 24 (aryl = C6F5, R = phenyl), and 25 (aryl = C6F5, R = 1,3- bis(trifluoromethyl)benzene) were successfully synthetized and characterized. The synthesis and characterization of complex (DAAm-Mes)Re(O)(CH2Ph) (26) is also reported in this document. The CO insertion into the ReIII-carbon bond was investigated with complexes 19, 23a, 23b, 24, 25 and 26. A fast reaction was observed in complexes bearing electron-withdrawing groups in the ligand framework. We were able to identify an acyl product when the complex possesses electron- donating substituents in the ligand backbone. Kinetic studies suggest a first-order overall reaction for CO insertion with 23b. Observed rate constants 1.658(2) × 10-3 s-1 (23b) and 1.612(2) × 10-3 s-1 (Acyl Product) were obtained. Previous reports in the literature were used to propose two possible insertion mechanisms for the CO insertion reaction with 23b.

5.2 Introduction

The insertion reaction of carbon monoxide (CO) into a metal-alkyl bond is a well-known transformation in organometallic chemistry.1 Examples of insertion of CO into metal ligand bonds are known for most transition metals,2 and the insertion of CO into metal alkyl and aryl bonds to generate metal acyl complexes and intermediates is generally proposed as the carbon-carbon bond forming step. Because of the utility of carbonylation as a source for new carbon-carbon bond forming reactions, the design of new catalysts capable of these reactions is critical. The synthesis of the rhenium(III) acetate complex 3, introduced in chapter 2, was reported by our group.3 In this study we reported a new pathway for C-O and C-C bond formation. The mechanism of CO insertion with high-valent oxorhenium complexes of the form

(DAAm)Re(O)(CH3) was investigated. The reaction resulted in the formation of complex 3. The mechanism was studied experimentally and computationally. First, there is a direct CO insertion into a Re-alkyl bond followed by a 1,2-acyl migration to form 3 (Scheme 5.1).

Scheme 5.1 Synthesis of DAAm rhenium(III) acetate complexes 3a and 3b.

Direct Insertion 1,2- Acyl Migration

R CO R R CH3 O CH O N 3 60 psi CO N 60 psi CO R N Re Re O Re OAc C6H6, rt, 72 h C6H6, 80 °C, 8 h N N N R N N R N

1 2 3 R = (a) -C6F5, (b) -Mes

Our group had previously reported on the mechanism of carbonylation reactions with oxorhenium complexes that contain tridentate chelating ligands. They are diamidoamine DAAm3- 4 (DAAm = N,N-bis(2-arylaminoethyl)-methylamine; aryl = -C6F5 or -Mes) (Scheme 5.2), diamidopyridine DAP5 (DAP = 2,6-bis((mesitylamino)methyl)pyridine) (Scheme 5.3) and SSS6 (SSS = 2-mercaptoethylsulfide) (Scheme 5.4). More recently we reported the carbonylation reaction of neutral nitrido complexes analogous to the previously reported oxorhenium species of 7 the form (PNP)Re(N)X (PNP = [2-P(CHMe2)2-4-MeC6H3N and X = Cl and CH3) (Scheme 5.5).

Scheme 5.2 Carbonylation Reaction of DAAm Oxorhenium(V) Complexes by the Ison Group

C6F5 C6F5 O CH O COCH N 3 60 psi CO N 3 Re Re C6H6, rt, 72 h N N C6F5 N N C6F5

C6F5 C6F5 O CH R O COCH R N 2 60 psi CO N 2 Re Re CH2Cl2, rt, 16 h N N C6F5 N N C6F5

R = -Ph, 4-methylphenyl, 4-fluorophenyl or 4-methoxyphenyl

Mes Mes O H OH CO N 40 psi CO N Re Re CH2Cl2, rt, 16 h N N Mes N N Mes

Scheme 5.3 Carbonylation Reaction of DAP Oxorhenium(V) Complex by the Ison Group Carbonylation Reaction of DAP Oxorhenium(V) Complex by Ison Group

Mes Mes N O R N O COR 60 psi CO Re Re N C H , 80 °C, 2.5 - 24 h N N 6 6 N Mes Mes R‘ R‘ R = H, Me, Ph, 4-OMePh, 4-ClPh R’ = H, OMe

O BR OH R BR3 R 3 C O N O H 40 psi CO N C H Re Re N rt, 1 h N N N R R H C R= -Mes; -Dipp; O BR3 = BF3, B(C6F5)3

Scheme 5.4 Carbonylation Reaction of SSS Oxorhenium(V) Complex by the Ison Group

O R O COR S 400 psi CO S Re Re Toluene, 50 °C, 3 - 8 h S S S S R = -Me, -Et, -Ph

Scheme 5.5 Carbonylation Reaction of PNP Nitridorhenium(V) Complex by the Ison Group

iPr iPr iPr iPr Me P N Me P N C O Re 50 psi CO Re iPr iPr N P N P iPr C6H6, 80 °C, 1 h iPr

From mechanistic, kinetic, and computational studies it was shown that the most likely mechanism for the insertion reaction of CO was a direct insertion mechanism and not the typical two-step intramolecular mechanism. In this chapter, the synthesis and characterization of several III new rhenium(III) aryl and alkyl complexes of the form [(DAAm-aryl)Re (CO)(R)] (aryl = C6F5 or Mes; R = Bn, Ph, 1,3-bis(trifluoromethyl)benzene, C6F5) are reported, and their reactivity with CO is examined.

5.3 Results and Discussion

5.3.1 Synthesis and Characterization of (DAAm-aryl)ReIII(CO)(R) (aryl = C6F5 or Mes)

Complexes [(DAAm-aryl)Re(CO)(R)] [DAAm = N,N-bis(2-arylaminoethyl)methylamine; aryl = C6F5 or Mes], 23a (aryl = C6F5; R = CH2Ph), 23b (aryl = Mes; R = CH2Ph), 24 (aryl = C6F5,

R = phenyl), and 25 (aryl = C6F5, R = 1,3-bis(trifluoromethyl)benzene) were synthesized from the rhenium(III) acetate complex [(DAAm-aryl)Re(CO)(OAc)] [DAAm = N,N-bis(2- arylaminoethyl)methylamine; aryl = C6F5 (3a) or Mes (3b)] and the corresponding Grignard reagents (Scheme 5.6).

Scheme 5.6 Synthesis of DAAm rhenium(III) derivatives

R CO R CO R N X equiv R’MgBr R N Re OAc Re R’ N CH2Cl2, rt, 2 h N N N

R’ = CH2Ph (23); Ph (24); R = C6F5 (3a), Mes (3b) 1,3-bis(trifluoromethyl)benzene (25)

The 1H NMR spectrum of 23a is shown on Figure 5.1. A singlet at 3.27 ppm is observed for the two benzylic protons. The methylene protons from the ligand backbone are observed as three distinct multiplets at 3.76 (4H), 3.15 (2H), and 3.05 (2H) ppm. The aromatic benzylic protons are observed at 7.15 (2H), 6.99 (2H), and 6.88 (1H). The methyl group on the amine nitrogen resonate as a singlet at 3.09 ppm.

1 Figure 5.1 H NMR spectrum of complex 23a in methylene chloride-d2. Residual diethyl ether peak at 3.43 ppm.

Thermal ellipsoid plots for 23a is shown in Figure 5.2. X-ray quality crystals for this complex were obtained by slow diffusion of pentane into a concentrated solution of 23a in methylene chloride. The geometry about rhenium in 23a is best described as a distorted trigonal bipyramid with the carbonyl ligand and the amine nitrogen occupying the axial plane. In Figure 5.2, the acetate ligand is replaced by one benzyl in the equatorial plane of the complex. The Re1- C19 bond length is 2.112 Å, which is comparable to other Re-benzyl bond length reported by our group (2.155 Å).4b

Figure 5.2 Thermal ellipsoid plot of 23a. Thermal ellipsoids are at 50%. Hydrogen atoms have been omitted and the pentafluorophenyl substituents on the diamido ligand framework are depicted in wireframe for clarity. Selected bond lengths (Å) and angles (deg): Re1-C18, 1.860(2); Re1-N1, 1.9224(19); Re1-N3, 1.9391(18); Re1-C19, 2.112(2); Re1-N2, 2.2483(17); N1-C5, 1.411(3); N1- C1, 1.491(3); N2-C2, 1.485(3); N2-C17, 1.487(3); N2-C3, 1.489(3); N3-C11, 1.417(3); N3-C4, 1.482(3); C1-C2, 1.518(3); C19-C20, 1.499(3); C18-Re1-N1, 98.91(8); C18-Re1-N3, 96.70(8); N1-Re1-N3, 121.38(8); C18-Re1-C19, 92.00(9); N1-Re1-C19, 117.73(8); N3-Re1-C19, 117.72(8); C20-C19-Re1, 122.14(15).

The 1H NMR spectrum of 23b is shown on Figure 5.3. The aromatic benzylic protons are observed at 7.10 (2H), 6.98 (2H), and 6.87 (1H). Two methylene protons from the ligand backbone are observed at 3.64 (4H) and 3.14 (2H) ppm. The benzylic protons, one methylene proton from the ligand backbone and the methyl group on the amine nitrogen resonate as a multiplet at 3.03 (8H) ppm. The methyl groups from the mesitylenes in the ligand framework resonate at 2.27 (6H), 2.15 (6H), and 2.06 (6H).

1 Figure 5.3 H NMR spectrum of complex 23b in methylene chloride-d2.

Thermal ellipsoid plots for 23b is shown in Figure 5.4. X-ray quality crystals for this complex were obtained by slow diffusion of pentane into a concentrated solution of 23b in methylene chloride. The geometry about rhenium in 23b is best described as a distorted trigonal bipyramid with the carbonyl ligand and the amine nitrogen occupying the axial plane. The X-ray structure show that the acetate ligand is replaced by one benzyl in the equatorial plane of the complex. The Re1-C24 bond length is 2.134 Å, which is similar to 23a (2.112 Å).

100

Figure 5.4 Thermal ellipsoid plot of 23b. Thermal ellipsoids are at 50%. Hydrogen atoms have been omitted and the mesitylene substituents on the diamido ligand framework are depicted in wireframe for clarity. Selected bond lengths (Å) and angles (deg): Re1-C31, 1.8628(19); Re1-N1, 1.9240(18); Re1-N2, 2.2635(16); Re1-N3, 1.9454(17); Re1-C24, 2.134(2); O1-C31, 1.176(2); N1- C6, 1.445(3); N1-C1, 1.491(2); N2-C5, 1.488(3); N2-C3, 1.492(3); N2-C2, 1.495(3); N3-C15, 1.441(2); N3-C4, 1.490(3); C24-C25, 1.500(3); C31-Re1-N1, 97.18(8); C31-Re1-N3, 96.41(8); N1-Re1-N3, 126.23(7); C31-Re1-C24, 93.78(8); N1-Re1-C24, 113.27(8); N3-Re1-C24, 117.37(8); C31-Re1-N2, 175.32(7); N1-Re1-N2, 81.37(6); N3-Re1-N2, 81.01(6); C24-Re1-N2, 90.87(7).

Complex 24 and its analogue 25 were also synthesized according to the reaction in Scheme 5.6. In the 1H NMR spectrum of complex 24 the methylene protons from the ligand backbone are observed as three distinct multiplets at 3.85 (4H), 3.14 (2H), and 2.97 (2H) ppm (Figure 5.5). The aromatic protons are observed at 7.21 (2H), 6.99 (2H), and 6.81 (1H) ppm. The methyl group on the amine nitrogen resonate as a singlet at 2.50 ppm. The 1H NMR spectrum of complex 25 is shown in Figure 5.6.

101

1 Figure 5.5 H NMR spectrum of complex 24 in methylene chloride-d2.

1 Figure 5.6 H NMR spectrum of complex 25 in methylene chloride-d2.

102

The thermal ellipsoid plots for 25 is shown in Figure 5.7. X-ray quality crystals for this complex were obtained by slow diffusion of pentanes into a concentrated solution of 25 in methylene chloride. The geometry about rhenium in 25 is best described as a distorted trigonal bipyramid with the carbonyl ligand and the amine nitrogen occupying the axial plane. The acetate ligand is replaced by one 1,3-bis(trifluoromethyl)benzene group in the equatorial plane of the complex. The Re1-C18 bond length is 2.085 Å, which is similar to Re-Csp2 bond lengths found in 24.

Figure 5.7 Thermal ellipsoid plot of 25. Thermal ellipsoids are at 50%. Hydrogen atoms have been omitted and the pentafluorophenyl substituents on the diamido ligand framework are depicted in wireframe for clarity. Selected bond lengths (Å) and angles (deg): Re1-C26, 1.877(4); Re1-N1, 1.922(3); Re1-N3, 1.932(3); Re1-C18, 2.085(3); Re1-N2, 2.231(3); C26-Re1-N1, 97.64(13); C26- Re1-N3, 95.05(13); N1-Re1-N3, 124.55(12); C26-Re1-C18, 92.13(14); N1-Re1-C18, 111.99(13); N3-Re1-C18, 121.21(13); C26-Re1-N2, 172.65(13); N1-Re1-N2, 80.95(11); N3-Re1-N2, 80.01(11); C18-Re1-N2, 95.09(11); C6-N1-C1, 113.1(3); C6-N1-Re1, 127.5(2); C1-N1-Re1, 119.0(2); C12-N3-C4, 114.3(3); C12-N3-Re1, 125.8(2); C4-N3-Re1, 118.8(2).

103

5.3.2 Reactivity of (DAAm-aryl)ReIII(CO)(R) (aryl = C6F5 or Mes) with CO

The CO insertion reaction into a ReIII-carbon bond was investigated using complexes 19, 24a, 24b, 25 and 26. The reaction was studied at different pressures of CO, and temperatures (Scheme 5.7). Results of the screening reactions are summarized in Table 5.1.

Scheme 5.7 General CO insertion reaction with (DAAm-aryl)ReIII(CO)(R) complexes.

R CO R CO R N X psi CO R N R Re R Re N CD2Cl2, T, 30 min - 24 h N O N N

III Table 5.1 Results for the CO Insertion Reaction of (DAAm-aryl)Re (CO)(R) (aryl = C6F5 or Mes) complexes

Entrya Complex CO Pressure (psi) T (°C)c Result

1 19 60 80 D

2 19 60 rt D

3 19 25 rt D

4 19 10 rt D

5 23a 60 80 D

6 23a 60 rt D

7 23a balloon rt NR

8 23b 60 rt R

9 23b 60 80 Rb

10 23b 40 rt R

11 23b 40 80 Rb

12 23b 20 rt R

104

Table 5.1 (continued) 13 23b 20 80 Rb

14 23b 5 rt NR

15 24 60 80 D

16 24 60 rt D

17 24 25 rt D

18 24 10 rt D

19 25 60 80 D

20 25 60 rt D

21 25 40 80 D

22 25 40 rt D

23 25 20 80 D

24 25 20 rt D a General Reaction Conditions: 5 mg [Re] dissolved in CD2Cl2 (0.35 mL). After degas, the reaction mixture was pressurized with CO. The reactions were monitored in a 400 MHz NMR spectrometer. b c Decomposition after 30 minutes. Reactions heated at 80 °C were run in C6D6. NR = no reaction. rt = room temperature. D = decomposition. R = reactivity. Balloon = balloon pressure.

Base on the data shown in Table 5.1, complexes bearing electron withdrawing substituents in the ligand backbone decomposed almost instantly in the presence of CO. Complexes 19, 23a, 24, and 25 reacted instantly after the addition of CO. When complex 23a was reacted under balloon pressure at room temperature no reaction was observed (Entry 7). This suggest pressures higher than 1 atmosphere are necessary for complexes 23a to show any reactivity towards CO. The reactivity of these complexes can be identified as a change in color after the J Young tube was pressurized with CO. We did not observe any trend in terms of the R group substituents in these complexes. All complexes bearing electron withdrawing substituents in the ligand backbone showed similar reactivity. On the other hand, complexes bearing electron donating substituents in the ligand backbone 23b were more stable in the presence of CO. We were able to identify by 1H NMR spectrometry the formation of a new complex in solution (Figure 5.8). The 1H NMR

105 spectrum at the top (a) in Figure 5.8 belongs to the reaction of 23b with 60 psi CO at 80 °C after 30 minutes (Entry 9). A direct comparison of the 1H NMR spectra suggest almost complete transformation of 23b to a proposed acyl product. As is shown in Figure 5.8, a new set of peaks appears at 7.11, 6.98, 6.87, 6.82, 3.63, 3.14, 3.01, 2.27, 2.15, and 2.06 ppm (top). The integration of the 1H NMR spectrum after 30 minutes suggest the formation of an acyl-product (Figure 5.9). When 23b was treated with lower pressures of CO, no reactivity was observed (Entry 14). The reactions heated at 80 °C resulted in the formation of a new complex in solution after 30 min, but after an hour the 1H NMR spectrum of the reaction showed signs of decomposition (Entries 9, 11, and 13). This suggests that the new complex formed in solution is susceptible to high temperatures. When reactions were run at room temperature, we were able to identify the new complex in solution for up to 3 h (Entries 8, 10, and 12). From all 23b reactions at room temperature, the reaction using 20 psi of CO (Entry 12) was chosen as the conditions to perform kinetic studies in this reaction.

Figure 5.8 Spectral comparison: (a) 1H NMR spectrum of the crude reaction of 23b with 60 psi 1 CO at 80 °C after 30 min in CD2Cl2. (b) H NMR spectrum of complex 23b in CD2Cl2.

106

Figure 5.9 Integrated 1H NMR spectrum of the reaction of 23b with 60 psi CO at 80 °C after 30 minutes. Acyl Product and 23b peaks are identified.

Isolation of the acyl product from the reaction depicted in Scheme 5.7 by X-ray was unsuccessful. FTIR spectroscopy was used as an attempt to identify acyl-product formation in solution. The FTIR spectrum of the acyl-product from the reaction of 23b with 60 psi CO at room temperature after 2 h is shown in Figure 5.10. The FTIR spectrum in Figure 5.10 was directly compared to the FTIR spectrum of 23b (Figure 5.11). Typical terminal CO stretching frequency range from 2115 to 1850 cm-1.9 CO stretches for acyl rhenium complexes reported in the literature range from 1520 to 1630 cm-1.3-7,10 In the FTIR spectrum of 23b the CO stretch appears at 1856 cm-1 and the stretch at 1514 cm-1, based on Coates’11 spectral interpretations, was assigned to the sp3 CH bend of the benzylic protons. In the FTIR spectrum depicted in Figure 5.10 two CO stretches were identified at 1869 and 1843 cm-1, the CO acyl ligand stretch was identified at 1625 cm-1, and the sp3 CH bend stretch of the benzylic protons was assigned to the band at 1495 cm-1. The higher CO stretch for the acyl ligand is similar to the 2-mercaptoethylsulfide (SSS) rhenium acyl complex (1620 cm-1) previously reported by our group.4b The higher CO acyl stretching observed in Figure 5.10 is due to a more electron poor rhenium center in comparison to acyl rhenium complexes reported in the literature. For example, CO acyl ligand stretch for 107

(PNP)Re(N)(MeC(=O)) reported by our group was identified at 1520 cm-1.7 The low value of the CO acyl stretch is because the nitrido ligand is a stronger electron donor and therefore the rhenium metal center is more electron rich which results in an increased π-backbonding to the π* orbital of the CO acyl fragment. The acyl complex in this investigation has less π-backbonding into the π* orbital of the acyl ligand from the metal center in comparison to the ReVN complex previously reported. Data in Figure 5.10 suggest the presence in solution of a complex that possesses two terminal CO ligands. This piece of evidence can be used to propose the formation of a

(DAAm)Re(CO)2 intermediate previous to the acyl-product formation. Full characterization of the acyl-product will be investigated further in our lab in the future.

-1 - Figure 5.10 FTIR spectrum of the acyl-product. Identified stretches: �CO = 1869 cm & 1843 cm 1 -1 3 -1 ; �CO(acyl) = 1625 cm and �CH(sp ) = 1490 cm .

108

-1 3 Figure 5.11 FTIR spectrum of the complex 23b. Identified stretches: �CO =1856 cm and �CH(sp ) = 1514 cm-1.

To have a better understanding and insight in how the oxidation state of the metal center V might affect the reaction rate for this reaction, complex (DAAm-Mes)Re (O)(CH2Ph) (26) was synthetized following the reaction in Scheme 5.8.

Scheme 5.8 Synthesis of complex (DAAm-Mes)Re(O)(CH2Ph), 26.

Mes Cl Mes O CH2Ph Mes N 15 equiv BnMgBr N Re O Re N CH Cl , rt, 1 h 2 2 N N Mes N

The 1H NMR spectrum of complex 26 is shown in Figure 5.12. A singlet at 5.50 ppm is observed for the two benzylic protons. The two benzylic protons in 26 are more downfield in 4b comparison to the DAAm-C6F5 analogue previously reported by our group (4.55 ppm). The methylene protons from the ligand backbone are observed as three distinct multiplets at 4.20 (2H), 3.01 (2H), and 2.74 (2H) ppm. The aromatic protons from the benzyl ring are observed at 6.91 (2H), 6.57 (2H), and 5.96 (1H) ppm. The methyl group on the amine nitrogen and one methylene from the ligand backbone resonate as a multiplet at 3.34 (5H) ppm. Finally, the methyl groups

109 from the mesitylene substituents in the ligand backbone resonate as singlets at 2.26 (6H), 2.10 (6H), and 1.99 (6H).

Figure 5.12 1H NMR spectrum of complex 26 in methylene chloride-d2.

The thermal ellipsoid plot for 26 is shown in Figure 5.13. X-ray quality crystals for this complex were obtained by slow diffusion of pentane into a concentrated solution of 26 in methylene chloride. The geometry about rhenium in 26 is best described as a distorted square pyramid with the oxo ligand and the amine nitrogen occupying the axial plane. The chlorine ligand is replaced by one benzyl group in the equatorial plane of the complex.

110

Figure 5.13 Thermal ellipsoid plot of 26. Thermal ellipsoids are at 50%. Hydrogen atoms have been omitted and the mesitylene substituents on the diamido ligand framework are depicted in wireframe for clarity. Note: Complete X-ray crystallography report will be added upon completion.

Complex 26 was treated of CO as shown in Scheme 5.9. Preliminary results of the CO insertion reaction with 26 are shown in Table 5.2. Reactions were monitored by 1H NMR spectroscopy. The results in Table 5.2 suggest that complex 26 is unreactive towards CO under the same reaction conditions as used with 23b even for more than a day. Also, these results might suggest that harsh conditions are needed for the CO insertion reaction with 26. More studies need to be done to have a better understanding of the effect of the oxidation state on this reaction. A good starting point will be redoing the reaction at higher temperatures and CO pressures. This will be investigated further in our lab in the future.

V Scheme 5.9 General CO insertion reaction with (DAAm-Mes)Re (O)(CH2Ph) (26) complex.

Mes Mes CH2Ph O CH Ph O N 2 X psi CO N Re Re O solvent, T N N Mes N N Mes

111

Table 5.2 Results for the CO Insertion Reaction with 26

Entrya CO Pressure (psi) T (°C)c Result 1 20 rt NR 2 40 rt NR 3 60 rt NR 4 20 80b NR 5 40 80b NR 6 60 80b NR a General Reaction Conditions: 5 mg [Re] dissolved in CD2Cl2 (0.35 mL). After degassing, the reaction mixture was pressurized with CO. The reactions were monitored in a 300 MHz NMR b spectrometer. Reactions heated at 80 °C were run in C6D6. NR = no reaction. rt = room temperature.

5.3.3 Kinetic Experiments

Kinetic experiments were performed in order to determine the overall reaction rate. The reaction of 23b with excess CO at 20 psi in methylene chloride at room temperature was monitored by 1H NMR spectroscopy with hexamethylcyclotrisiloxane as the internal standard. The rate disappearance of 23b were monitored overtime for 1 hour (Figure 5.14). The total rhenium concentration ([24b] + [Product]) remains relatively unchanged during the course of the reaction. As shown in Figure 5.14, the reaction follows clean pseudo-first order kinetics with observed with -3 -1 -3 -1 rate constants (kobs) of 1.658(2) × 10 s (24b) and 1.612(2) × 10 s (Product). The results suggest a rate law that is first-order overall.

112

Figure 5.14 Time Profile for the disappearance of 23b. Reaction was performed with [Re(CO)(Bn)(DAAm-Mes)] (23b) (0.0159 mmol; 0.045M) and CO at 20 psi (0.0195 mmol) in a J. Young NMR tube in CD2Cl2 (0.35 mL) at room temperature for 60 minutes in a Varian 400 MHz NMR. The concentration was calculated using an internal standard, hexamethylcyclotrisiloxane (0.0108 mmol; 0.031 M) which was added to the reaction mixture. Observed rate constants of 1.658(2) × 10-3 s-1 (23b) and 1.612(2) × 10-3 s-1 (Product) were obtained.

5.4 Proposed Mechanism There are two possible reaction pathways for the acyl product formation. The first one is the direct insertion of CO into the Re-carbon bond (Scheme 5.10). This mechanism has been proposed and reported in the literature by our group with (DAAm)ReV(O)3,4, (DAP)ReV(O)5 and (PNP)ReV(N)7 complexes.

Scheme 5.10 General Mechanism for Direct CO Insertion. Intermolecular Mechanism

O O R CO Ln Re R Ln Re O

113

The second possible reaction pathway is the CO adduct formation prior to the CO migration into the Re-carbon bond (Scheme 5.11). This type of mechanism has been proposed and reported by our group in the CO insertion reaction of ReV complexes bearing an SSS6 ligand framework.

Scheme 5.11 General Mechanism for CO Adduct formation. Intramolecular Mechanism O O O R CO Ln Re R Ln Re CO Ln Re O R

Based on previous mechanisms reported by our group, we can propose the same two possible mechanisms for the CO insertion reaction with 23b. The reaction can either take place by the formation of a CO adduct (Scheme 5.12) or by the direct insertion of the CO into the ReIII- carbon(sp3) bond (Scheme 5.13).

Scheme 5.12 Proposed CO Adduct formation mechanism for 23b.

Mes CO Mes CO Mes CO Ph Ph CH Ph Mes N Mes N Mes N 2 Re Re Re CO N N N O N CO N N

Scheme 5.13 Proposed Direct CO Insertion mechanism for 23b.

Mes CO Mes CO Ph CH Ph Mes N Mes N 2 Re Re N N O N CO N

Preliminary data show in Figure 5.10 suggest the formation of a complex bearing two terminal CO ligands. Identification of this intermediate suggest that the reaction mechanism follow a CO adduct formation (Scheme 5.12). Still there is not enough data to confirm either of the two possible mechanisms proposed for the CO insertion into the ReIII-C(sp3) bond in 23b. Theoretical

114

DFT studies, and complete characterization of a CO adduct, and the acyl complex need to be perform in order to fully understand the mechanism for this transformation.

5.5 Conclusions

A series of Re(III) aryl complexes of form [(DAAm-aryl)Re(CO)R] (aryl = C6F5 or Mes) were synthetized, characterized, and their reactivity with CO was examined. Complex [(DAAm- Mes)ReV(O)(Bn)] (26) was also synthetized, characterized and its reactivity towards CO insertion was studied and compared to 23b. Based on the results we can conclude that Re(III) alkyl complexes are more reactive towards carbonylation than Re(V) complexes previously reported by our group. The CO insertion reaction with DAAm-Re(III) complex 23b can be performed at room temperature, low CO pressure (20 psi) and the reaction is complete in 30 min in comparison to DAAm-ReV, DAP-ReV, SSS-ReV and PNP-ReV. Experimental data suggest that complexes with the DAAm-C6F5 ligand framework is more reactive than DAAm-Mes ligand framework in the presence of CO. The acyl Re(III) product that possess a ligand framework with electron donating groups (DAAm-Mes) is more stable in solution than complexes with electron withdrawing

(DAAm-C6F5) groups in their ligand framework. Kinetic studies show a first-order overall reaction. Observed rate constants 1.658(2) × 10-3 s-1 (23b) and 1.612(2) × 10-3 s-1 (Product) were obtained from the time profile for the disappearance of 23b.

115

5.6 Experimental Section

General Considerations. 3a-b3 and (DAAm-Mes)Re(O)(Cl)10 were prepared according to previous procedures. All reactions were carried out in a nitrogen filled glove box unless otherwise noted. All reagents were purchased from commercial sources, placed in a nitrogen filled glove box and used as received without further purification. 1H, 13C and 19F spectra were acquired on a Varian Mercury 700 MHz, Varian Mercury 400 MHz or Varian Mercury 300 MHz spectrometer. NMR chemical shifts are listed in parts per million (ppm) and are referenced to residual protons or carbons of the deuterated solvents, respectively at room temperature unless otherwise noted. The FTIR spectra were obtained in KBr thin films on a JASCO FT/IR-4100 instrument. Elemental analyses were performed by Atlantic Micro Laboratories Inc. X-ray crystallography was performed at the X-ray Structural Facility at North Carolina State University by Dr. Roger D. Sommer and Caleb Brown.

General Procedure Synthesis [(DAAm-aryl)Re(CO)(R)] (aryl = C6F5, Mes); 23a-b, 24, and 25

In a nitrogen filled glove box, the corresponding [(DAAm-aryl)Re(CO)(OAc)] (aryl = C6F5 (3a), Mes (3b)) (3a = 250 mg, 0.347 mmol; 3b = 250 mg, 0.399 mmol) was added to a screw cap vial and dissolved in 0.5 mL of dichloromethane. Then, the corresponding

Grignard reagent (RMgX: R = CH2Ph, 3 equiv; Ph, 3 equiv; 1,3- bis(trifluoromethyl)benzene, 15 equiv) was added dropwise to the stirred solution. The solution was allowing to stir for 1-2 h (R = CH2Ph, Ph) or 24 h (R = 1,3- bis(trifluoromethyl)benzene) at room temperature. Water was added (50 mL × 3), and the organic layer was extracted and dried over Na2SO4. The mixture was filtered, and solvent was removed under reduced pressure. The resultant residue was dissolved in a minimal amount of dichloromethane and precipitated upon the addition of hexanes. The resultant precipitate was filtered to yield the final product as a powder.

[(DAAm-C6F5)Re(CO)(CH2Ph)], 23a. Following the general synthesis, complex 23a was 1 obtained in 63% yield as an orange powder. H NMR (400 MHz, CD2Cl2) δ: 7.15 (t, 2H), 6.99 (d, 2H), 6.88 (t, 1H), 3.76 (m, 4H), 3.27 (s, 2H), 3.15 (m, 2H), 3.09 (s, 3H), 3.05 (m, 2H). 13C NMR

(151 MHz, CD2Cl2) δ: 199.48, 130.13, 129.99, 129.57, 129.20, 129.00, 128.73, 128.14, 127.97, 19 127.75, 127.63, 127.43, 124.11, 59.50, 54.36, 49.03, 46.21, 21.35. F NMR (376 MHz, CD2Cl2) δ: -151.62 (m, 2F), -152.30 (dd, J = 23.8 Hz, 2F), -158.52 (m, 2F), -162.28 (t, J = 21.4 Hz, 2F), -

116

-1 164.39 (d, J = 24.5 Hz, 2F). IR (FTIR, cm ): �C=O 1885. Anal. Calc. for C25H18F10N3ORe: C, 39.90; N, 5.58; H, 2.41. Found: C, 39.77; N, 5.52; H, 2.36.

[(DAAm-Mes)Re(CO)(CH2Ph)], 23b. Following the general synthesis, complex 23b was 1 obtained in quantitative yield as an orange powder. H NMR (300 MHz, CD2Cl2) δ: 7.10 (t, 2H), 6.98 (d, 2H), 6.87 (t, 1H), 6.82 (s, 4H), 3.64 (m, 4H), 3.14 (m, 2H), 3.03 (m, 8H), 2.27 (s, 6H), 13 2.15 (s, 6H), 2.06 (s, 6H). C NMR (151 MHz, CD2Cl2) δ: 201.36, 157.81, 153.22, 134.48, 132.45, 130.93, 129.21, 129.09, 128.68, 127.57, 123.75, 62.17, 59.61, 46.16, 45.28, 20.91, 19.77, 19.43. -1 IR (FTIR, cm ): �C=O 1856. Anal. Calc. for C31H40N3ORe: C, 56.68; N, 6.40; H, 6.14. Found: C, 56.83; N, 6.41; H, 6.11.

[(DAAm-C6F5)Re(CO)(Ph)], 24. Following the general synthesis, complex 24 was 1 obtained in 81% yield as a brown powder. H NMR (300 MHz, CD2Cl2) δ: 7.21 (t, 2H), 6.99 (d, 2H), 6.81 (t, 1H), 3.85 (m, 4H), 3.14 (m, 2H), 2.97 (m, 2H), 2.50 (s, 3H). 13C NMR

(151 MHz, CD2Cl2)) δ: 133.76, 129.88, 129.55, 129.41, 129.10, 129.00, 128.86, 128.79, 127.90, 127.64, 126.24, 126.09, 60.81, 58.39, 41.98. CO peak was not identified by 13C 19 NMR. F NMR (376 MHz, CD2Cl2) δ: -151.71 (m, 2F), -151.72 (dd, J = 23.5, 6.3 Hz, 2F), -1 -161.66 (t, J = 21.6 Hz, 2F), -164.2 (m, 4F). IR (FTIR, cm ): �C=O 1859. Anal. Calc. for

C24H16F10N3ORe: C, 38.97; N, 5.68; H, 2.32. Found: C, 38.89; N, 5.65; H, 2.24.

[(DAAm-C6F5)Re(CO)(1,3-bis(trifluoromethyl)benzene)], 25. Following the general 1 synthesis, complex 25 was obtained in 34% yield as a brown powder. H NMR (300 MHz, CD2Cl2) δ: 7.45 (s, 2H), 7.32 (s, 1H), 3.91 (m, 4H), 3.17 (m, 2H), 3.00 (m, 2H), 2.52 (s, 3H). 13C NMR

(151 MHz, CD2Cl2) δ: 187.40, 184.75, 183.48, 143.54, 142.93, 141.92, 141.22, 140.16, 138.49, 19 137.06, 131.06, 130.85, 130.39, 128.85, 63.80, 58.92, 43.24. F NMR (376 MHz, CD2Cl2) δ: - 63.29 (s, 6F), -151.28 (m, 2F), -152.27 (dd, J = 23.4, 6.5 Hz, 2F), -160.41 (t, J = 21.5 Hz, 2F), - -1 163.49 (m, 4F). IR (FTIR, cm ): �C=O 1833. Anal. Calc. for C26H14F16N3ORe: C, 35.71; N, 4.80; H, 1.61. Found: C, 35.60; N, 4.74; H, 1.61.

Synthesis of complex [(DAAm-Mes)Re(O)(CH2Ph)], 26. In a nitrogen filled glove box complex [(DAAm-Mes)ReOCl] (220 g, 0.321 mmol) was added to a screw cap vial and dissolved in 0.5 mL of dichloromethane. Then, benzyl magnesium chloride (2 M in THF, 15 equiv) was added dropwise to the stirred solution. The reaction mixture was allowing to stir for 1 h at room temperature. Water was added (50 mL × 3), and the organic layer was extracted and dried over Na2SO4. The mixture was filtered, and solvent was removed under

117 reduced pressure. The resultant residue was dissolved in a minimal amount of dichloromethane and precipitated upon the addition of hexanes. The resultant precipitate was filtered to yield the 1 final dark green product (110 mg, 50% yield). H NMR (300 MHz, CD2Cl2) δ: 6.91 (s, 2H), 6.70 (s, 4H), 6.57 (m, 2H), 5.96 (d, 1H), 5.50 (s, 2H), 4.20 (m, 2H), 3.34 (m, 5H), 3.01 (m, 2H), 2.74 13 (m, 2H), 2.26 (s, 6H), 2.10 (s, 6H), 1.99 (s, 6H). C NMR (101 MHz, CD2Cl2) δ: 156.05, 155.63, 136.91, 134.52, 134.34, 129.79, 129.33, 129.13, 126.56, 122.27, 72.24, 66.71, 51.68, 36.60, 21.06,

19.10, 18.94. Anal. Calc. for C30H40N30ORe·2H2O: C, 52.92; N, 6.17; H, 6.51. Found: C, 53.20; N, 6.11; H, 6.35.

General Procedure for CO Insertion Reactions

The corresponding [(DAAm-aryl)Re(CO)(R)] (aryl = C6F5 or Mes) or [(DAAm-

Mes)Re(O)(CH2Ph) complex was added to a J. Young NMR tube and dissolved in CD2Cl2 (0.35 mL) at room temperature and was degassed. The reaction was pressurized with CO. At a fixed time, the J. Young NMR tube was placed in a 400 MHz NMR spectrometer and percent of conversion was determined by the 1H NMR spectroscopy.

General Procedure for CO Insertion Kinetic Studies with 23b Complex [(DAAm-Mes)Re(CO)(CH2Ph)] (23b) (10 mg, 0.0159 mmol) and hexamethylcyclotrisiloxane (0.0108 mmol, 1.5 equiv) were added to a J. Young NMR tube and dissolved in CD2Cl2 (0.35 mL) at room temperature and was degassed. The reaction was pressurized with CO (20 psi). The J. Young NMR tube was placed in a 400 MHz NMR spectrometer. The reaction was monitored by 1H NMR spectroscopy for 3 h. Every 5 minutes a 1H NMR was taken, and the peaks were integrated. The concentration was calculated using an internal standard. The first 65 min of product formation or reactant disappearance were plotted against time in seconds to obtain the time profile of the reaction.

X-ray Crystallographic Procedures and Data

General Procedure for X-ray Determination X-Ray structural determination was performed at the X-Ray Structural Facility of North Carolina State University by Dr. Roger Sommer and Caleb Brown. The samples were mounted on a nylon loop with a small amount of Paratone N oil. The frame integration was performed using SAINT.12 The resulting raw data were scaled, and absorption corrected using a multi-scan

118 averaging of symmetry equivalent data using SADABS.13 Most non-hydrogen atoms were obtained from the initial solution. The remaining non-hydrogen atom positions were recovered from a subsequent difference Fourier map. The structural model was fit to the data using full matrix least-squares based on F2. The calculated structure factors included corrections for anomalous dispersion from the usual tabulation. The structure was refined using the XL program from SHELXTL14, graphic plots were produced using the NRCVAX crystallographic program suite. Additional information and other relevant literature references can be found in the reference section of the Facility's Web page (http://www.xray.ncsu.edu). Summary of the following crystal data is shown in Table 5.3 and Table 5.4.

X-ray Structure Determination for [(DAAm-C6F5)Re(CO)(CH2Ph)], 23a.

A brown-orange block-like specimen of C25H18F10N3ORe, approximate dimensions 0.114 mm x 0.192 mm x 0.307 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 1.19 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 60219 reflections to a maximum θ angle of 35.09° (0.62 Å resolution), of which 10737 were independent (average redundancy 5.609, completeness = 99.9%, 2 Rint = 4.51%, Rsig = 3.41%) and 8865 (82.56%) were greater than 2σ(F ). The final cell constants of a = 11.0291(3) Å, b = 10.9894(2) Å, c = 20.4619(5) Å, β = 102.0950(10)°, volume = 2424.99(10) Å3, are based upon the refinement of the XYZ-centroids of 5521 reflections above 20 σ(I) with 3.834° < 2θ < 73.73°.Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.557. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.3030 and 0.5930. The final anisotropic full-matrix least-squares refinement on F2 with 362 variables converged at R1 = 2.59%, for the observed data and wR2 = 5.85% for all data. The goodness-of-fit was 1.024. The largest peak in the final difference electron density synthesis was 3.719 e-/Å3 and the largest hole was -1.616 e-/Å3 with an RMS deviation of 0.147 e- /Å3. On the basis of the final model, the calculated density was 2.061 g/cm3 and F(000), 1448 e-.

X-ray Structure Determination for [(DAAm-Mes)Re(CO)(CH2Ph)], 23b.

An orange block-like specimen of C31H40N3ORe, approximate dimensions 0.080 mm x 0.100 mm x 0.180 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 3.24 hours. The frames were integrated with the Bruker 119

SAINT software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 66279 reflections to a maximum θ angle of 33.15° (0.65 Å resolution), of which 10692 were independent (average redundancy 6.199, completeness = 99.9%, 2 Rint = 3.46%, Rsig = 2.44%) and 9097 (85.08%) were greater than 2σ(F ). The final cell constants of a = 10.2831(19) Å, b = 13.059(2) Å, c = 21.034(4) Å, β = 97.025(8)°, volume = 2803.4(9) Å3, are based upon the refinement of the XYZ-centroids of 9872 reflections above 20 σ(I) with 5.013° < 2θ < 66.45°. Data were corrected for absorption effects using the Multi-Scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.877. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.4343 and 0.4954. The structure was solved and refined using the Bruker SHELXTL Software Package, using the space group P 1 21/n 1, with Z = 4 for the formula unit, C31H40N3ORe. The final anisotropic full-matrix least-squares refinement on F2 with 332 variables converged at R1 = 2.31%, for the observed data and wR2 = 5.56% for all data. The goodness-of-fit was 1.031. The largest peak in the final difference electron density synthesis was 4.707 e-/Å3 and the largest hole was -0.905 e-/Å3 with an RMS deviation of 0.127 e- /Å3. On the basis of the final model, the calculated density was 1.556 g/cm3 and F(000), 1320 e-.

X-ray Structure Determination for [(DAAm- C6F5)Re(CO)(1,3 bis(trifluoromethyl)benzene)], 25.

An orange block-like specimen of C26H14F16N3ORe, approximate dimensions 0.152 mm x 0.235 mm x 0.322 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 1.38 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 41544 reflections to a maximum θ angle of 25.68° (0.82 Å resolution), of which 5180 were independent (average redundancy 8.020, completeness = 100.0%, 2 Rint = 7.43%, Rsig = 3.16%) and 4343 (83.84%) were greater than 2σ(F ). The final cell constants of a = 11.2085(8) Å, b = 16.6456(13) Å, c = 15.6542(12) Å, β = 110.759(2)°, volume = 2731.0(4) Å3, are based upon the refinement of the XYZ-centroids of 713 reflections above 20 σ(I) with 4.958° < 2θ < 67.39°. Data were corrected for absorption effects using the Numerical Mu from Formula method (SADABS). The ratio of minimum to maximum apparent transmission was 0.680. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.3200 and 0.5420. The structure was solved and refined using the Bruker SHELXTL

120

Software Package, using the space group P2(1)/n, with Z = 4 for the formula unit, C26H14F16N3ORe. The final anisotropic full-matrix least-squares refinement on F2 with 425 variables converged at R1 = 2.13%, for the observed data and wR2 = 4.55% for all data. The goodness-of-fit was 1.002. The largest peak in the final difference electron density synthesis was 1.286 e-/Å3 and the largest hole was -0.665 e-/Å3 with an RMS deviation of 0.105 e- /Å3. On the basis of the final model, the calculated density was 2.127g/cm3 and F(000), 1672 e-.

Table 5.3 Selected Crystallographic Data and Collection Parameters for 23a and 23b.

23a 23b Empirical Form C25H18F10N3ORe C31H40N3ORe Formula weight (g/mol) 752.62 656.86 crystal system monoclinic monoclinic space group P 1 21/c 1 P1 21/c 1 a, Å 11.0291(3) 10.2831(19) b, Å 10.9894(2) 13.059(2) c, Å 20.4619(5) 21.034(4) Volume, (Å3) 2424.99(10) 2803.4(9) Z 4 4 ρ (g/cm3) 2.061 1.556 crystal size (mm) 0.114 × 0.192 × 0.307 0.080 × 0.100 × 0.180 R1, w R2 0.0379, 0.0585 0.0321, 0.0556 GOF 1.024 1.031

Table 5.4 Selected Crystallographic Data and Collection Parameters for 25 and 26. 25 26 Empirical Form C26H14F16N3ORe C30H40N3ORe Formula weight (g/mol) 874.59 644.87 crystal system monoclinic monoclinic space group P2(1)/n P2(1)/n a, Å 11.2085(8) 10.423(2) b, Å 16.6456(13) 15.837(5) c, Å 15.6542(12) 16.566(3) Volume, (Å3) 2731.0(4) 2681.4(17) Z 4 4 ρ (g/cm3) 2.127 1.619 crystal size (mm) 0.152 × 0.235 × 0.322 0.116 × 0.140 × 0.424 R1, w R2 0.0328, 0.0455 0.0158, 0.0358 GOF 1.0002 1.045

121

5.7 References

(1) (a) Basolo, F. J. Organomet. Chem. 1990, 383, 579; (b) Anderson, G. K.; Clark, H. C.; Davies, J. A. Organometallics 1982, 1, 64; (c) Berdini, V.; Bassetti, M.; Mancini, G.; Mandolini, L.; Monti, D. J. Organomet. Chem. 1998, 551, 331; (d) Bi, S.; Zhao, J.; Fan, W.; Li, P. J. Theor. Comput. Chem. 2012, 11, 1; (e) Cao, Z.; Niu, S.; Hall, M. B. J. Phys. Chem. A 2000, 104, 7324; (f) Gonsalvi, L.; Adams, H.; Sunley, G. J.; Ditzel, E.; Haynes, A. J. Am. Chem. Soc. 1999, 121, 11233; (g) Gonsalvi, L.; Gaunt, J. A.; Adams, H.; Castro, A.; Sunley, G. J.; Haynes, A. Organometallics 2003, 22, 1047; (h) Huo, C. F.; Li, Y. W.; Beller, M.; Jiao, H. Organometallics 2003, 22, 4665; (i) Luo, X.; Tang, D.; Li, M. J. Mol. Struct.: THEOCHEM 2006, 765, 21; (j) Mawby, R. J.; Basolo, F.; Pearson, R. G. J. Am. Chem. Soc. 1964, 86, 3994; (k) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem. Soc. 1996, 118, 4746; (l) Schultz, B. E.; Gheller, S. F.; Muetterties, M. C.; Scott, M. J.; Holm, R. H. J. Am. Chem. Soc. 1993, 115, 2714; (m) Whitesides, G. M.; Boschetto, D. J. J. Am. Chem. Soc. 1969, 91, 4313; (n) Bassetti, M.; Capone, A.; Mastrofrancesco, L.; Salamone, M. Organometallics 2003, 22, 2535; (o) Cavell, K. J. Coord. Chem. Rev. 1996, 155, 209; (p) Conradie, J.; Swarts, J. C. Organometallics 2009, 28, 1018; (q) Donovan, B. T.; Geiger, W. E. Organometallics 1990, 9, 865; (r) Fantacci, S.; De Angelis, F.; Sgamellotti, A.; Re, N. Organometallics 2001, 20, 4031; (s) Monti, D.; Bassetti, M. J. Am. Chem. Soc. 1993, 115, 4658; (t) Scheuermann, M. L.; Rheingold, A. L.; Williams, B. S. Organometallics 2009, 28, 1613; (u) van Leeuwen, P. W. N. M.; Roobeek, C. F.; van der Heijden, H. J. Am. Chem. Soc. 1994, 116, 12117. (2) Wojcicki, A. Adv. Organomet. Chem. 1973, 11, 87-145. (3) Smeltz, J. L.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc. 2011, 133, 13288-13291. (4) (a) Smeltz, J. L.; Webster, C. E.; Ison, E. A. Organometallics 2012, 31, 4055-4062. (b) Robbins, L. K.; Lilly, C. P.; Smeltz, J. L.; Boyle, P. D.; Ison, E. A. Organometallics 2015, 34, 3152-3158. (c) Lambic, N. S.; Lilly, C. P.; Sommer, R. D.; Ison, E. A. Organometallics 2016, 35, 3060-3068. (5) (a) Lambic, N. S.; Lilly, C. P.; Robbins, L. K.; Sommer, R. D.; Ison, E. A. Organometallics 2016, 35, 2822-2829. (b) Lilly, C. P.; Boyle, P. D.; Ison, E. A. Organometallics 2012, 31, 4295-4301. (6) Robbins, L. K.; Lilly, C. P.; Sommer, R. D.; Ison, E. A. Organometallics 2016, 35, 3530- 3537.

122

(7) Lambic, N. S.; Sommer, R. D.; Ison, E. A. Dalton Trans., 2018, 47, 758-768. (8) Hartwig, J. F. Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 2010. (9) (a) Goldberg, K. I.; Bergman, R. G. J. Am. Chem. Soc. 1989, 111, 1285.; (b) Buhro, W. E.; Wong, A.; Merrifield, J. H.; Lin, G. Y.; Constable, A. C.; Gladysz, J. A. Organometallics 1983, 2, 1852. (10) Smeltz, J. L.; Lilly, C. P.; Boyle, P. D.; Ison, E. A. J. Am. Chem. Soc. 2013, 135, 9433- 9441. (11) Coates, J. Interpretation of Infrared Spectra, a Practical Approach. In Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed.; John Wiley & Sons Ltd: Chichester, 2000; 1-23. (12) Bruker-Nonius, SAINT version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA. (13) Bruker-Nonius, SADABS version 2009.9, 2009, Bruker-Nonius, Madison, WI 53711, USA. (14) Bruker-AXS, XL version 2009.9, 2009, Bruker-AXS, Madison, WI 53711, USA.

123