UC Riverside Electronic Theses and Dissertations

UC Riverside UC Riverside Electronic Theses and Dissertations

Title Carboranes: Building Blocks for Materials and Ligand Development

Permalink https://escholarship.org/uc/item/2vp9m2z6

Author Estrada, Jess Steven

Publication Date 2017

Peer reviewed|Thesis/dissertation

eScholarship.org Powered by the California Digital Library University of California

UNIVERSITY OF CALIFORNIA RIVERSIDE

Carboranes: Building Blocks for Materials and Ligand Development

A Dissertation submitted in partial satisfaction of the requirements for the degree of

Doctor of Philosophy

Chemistry

Jess Steven Estrada

September 2017

Dissertation Committee: Dr. Vincent Lavallo, Chairperson Dr. Richard Hooley Dr. Pingyun Feng

The Dissertation Jess Steven Estrada is approved:

Committee Chairperson

University of California, Riverside

Acknowledgements

I would like to thank Dr. Vincent Lavallo for giving me the opportunity to join his lab and for all of his help and support throughout graduate school. I owe a lot of my success to you. I would also like to thank the amazing faculty at UCR for all of their help and knowledge they have provided. I truly enjoyed every class I took here at UCR. In addition to the faculty, the staff in the chemistry department has been amazing over the last five years and played a huge role in my success as well so I would specifically like to thank Dr.

Borchardt, the NMR genius, Dr. Fook for all of his help with my great looking X-ray structures, Christina Youhas for answering literally every question I ever had and being so kind about answering. My lab mates, who created the most unique group of people probably in the history of UCR and that I’ve ever had the pleasure of working with. You guys (and girl) are all awesome friends and co-workers and I wish you all nothing but the best in your future careers. I would like to thank my family and friends who have put up with me being MIA over the last five years and who have provided the encouragement and inspiration for me to be successful. Finally, saving the best and MOST important for last,

I would like to thank my wife for putting up with all of the long hours (especially while finishing up those last few weeks) and god knows what else you put up with. I honestly could not have done this without you and I hope you know that. I dedicate this body of work to you and our beautiful son Miles.

iv The text, figures, and schemes for the following chapters have been reproduced, in part or in their entirety, from the following published or submitted manuscripts.

Chapter 2: “Resisting B-H Oxidative Addition: The Divergent Reactivity of the o- Carborane and Carba-closo-dodecaborate Ligand Substituents” J. Estrada, S.E. Lee, S. McArthur, A. El-Hellani, F.S. Tham, V. Lavallo, J. Organomet. Chem., 2015, 798, 214- 217

Chapter 3: “The Inductive Effects of 10 and 12-Vertex closo-Carborane Anion Ligand Substituents: Cluster Size and Charge Make a Difference” J. Estrada, C.A. Lugo, S.G. McArthur, V. Lavallo, Chem. Commun., 2016, 52, 1824-1826.

Chapter 4: “Changing the Charge: Electrostatic Effects in Pd Catalyzed Cross Coupling” A.L. Chan, J. Estrada, C.E Kefalidis, V. Lavallo, Organometallics, 2016, 35, 3257–3260.

Chapter 5: “Synthesis and Reactivity of a Zwitterionic Pd Allyl Complex Supported by a Perchlorinated Carboranyl Phosphine Ligand Substituents” J. Estrada, D.H. Woen, F.S. Tham, G.M. Miyake, V. Lavallo, Inorg. Chem., 2015, 54, 5142–5144.

Chapter 6: “Fusing Dicarbollide Ions with N-Heterocyclic Carbenes” J. Estrada, V. Lavallo, to be submitted.

Chapter 7: “Isolation of a Carborane-Fused Triazole Radical Anion” M. Asay, C. E. Kefalidis, J. Estrada, D. S. Weinberger, J. Wright, C. E. Moore, A. L. Rheingold, L. Maron, V. Lavallo, Angew. Chem. Int. Ed., 2013, 52, 11560-11563.

ABSTRACT OF THE DISSERTATION

Carborane Anions: Building Blocks for Materials and Ligand Development

Jess Steven Estrada

Doctor of Philosophy, Graduate Program in Chemistry University of California, Riverside, September 2017 Dr. Vincent Lavallo, Chairperson

Carborane anions are prized molecules due to their unique structure and bonding, as well as their anomalous stability and resistance to thermal and chemical decomposition. Since their discovery over half a century ago, they have been exploited as weakly coordinating anions stabilizing the most reactive cationic species known, but applications in catalysis by appending carborane anions as ligand substituents had never been attempted. We report a number of carborane-supported zwitterionic and anionic late transition metal complexes to serve as proof of concept for the application of anionic carborane ligand substituents in transition metal catalysis. The results from our studies confirm the anionic carborane is indeed inherently more stable than its isoelectronic neutral cousin, making it a suitable candidate for applications in catalysis. The donor properties of the 12 and 10 vertex anionic carboranes were probed to allow for the possibility of logical tuning of a given transition metal catalyst. Applications in catalysis and studies on the effect of charge as well as the development of a novel N-dicarbollide N-heterocyclic carbene are also reported. The work herein, should mitigate the development of novel single component zwitterionic and

vi anionic transition metal complexes for applications including, but not limited to hydrogenation, dehydrogenation, α-olefin polymerization, as well as various cross coupling reactions. In addition to the aforementioned, the isolation of the first triazole radical anion is disclosed as well as the methodology to develop a library of potential isolable triazole radical anions for applications in functional materials.

vii

Table of Contents Acknowledgements ...... iv List of Figures ...... x List of Schemes ...... xii Chapter 1: Introduction ...... 1 1.1 Background ...... 1

1.2 Synthesis and functionalization of H2C2B10H10 ...... 4

1.3 Synthesis of nido-carborane 7,8-C2B9H12 ...... 6

- 1.4 Synthesis and functionalization of HCB11H11 ...... 9 - 1.5 Synthesis and functionalization of HCB9H9 ...... 11 1.6 References ...... 15 Chapter 2: Resisting B-H Oxidative Addition: The Divergent Reactivity of the o-Carborane - (C2B10H10) and Carba-closo-dodecaborate (CB11H11 ) Ligand Substituents ...... 19 2.1 Abstract ...... 19 2.2 Introduction ...... 19 2.3 Results and Discussion...... 21 2.4 Summary and Conclusion ...... 24 2.5 Experimental ...... 24 2.6 References ...... 25 Chapter 3: The Inductive Effects of 10 and 12-Vertex closo-Carborane Anion Ligand Substituents: Cluster Size and Charge Make a Difference ...... 29 3.1 Abstract ...... 29 3.2 Introduction ...... 29 3.3 Results and Discussion...... 31 3.4 Summary and Conclusion ...... 36 3.5 Experimental ...... 36 3.6 References ...... 37 Chapter 4: Changing the Charge: Electrostatic Effects in Pd Catalyzed Cross-Coupling .. 41 4.1 Abstract ...... 41 4.2 Introduction ...... 41 4.3 Results and Discussion...... 43 4.5 Experimental ...... 52

viii

4.6 References ...... 53 Chapter 5: Synthesis and Reactivity of a Zwitterionic Pd Allyl Complex Supported by a Perchlorinated Carboranyl Phosphine Ligand Substituent...... 56 5.1 Abstract ...... 56 5.2 Introduction ...... 56 5.3 Results and Discussion...... 58 5.4 Summary and Conclusion ...... 62 5.5 Experimental ...... 62 5.6 References ...... 63 Chapter 6: Fusing Dicarbollide Ions with N-Heterocyclic Carbenes ...... 67 6.1 Abstract ...... 67 6.2 Introduction ...... 67 6.3 Results and Discussion...... 68 6.4 Summary and Conclusion ...... 73 6.5 Experimental ...... 73 6.6 References ...... 92 Chapter 7: Isolation of a Carborane-Fused Triazole Radical Anion ...... 94 7.1 Abstract ...... 94 7.2 Introduction ...... 94 7.3 Results and Discussion...... 95 7.4 Summary and Conclusion ...... 100 7.5 Experimental ...... 100 7.6 References ...... 101 Conclusion ...... 106

List of Figures Figure 1-1. Dicarba-closo-dodecaborane...... 1

- - Figure 1-2. Structure and numbering of nido-7,8 H2C2B9H10 , HCB11H11 , and the

- HCB9H9 anions...... 2

Figure 1-3. Orientation of px, py, and spz orbitals displaying the orientation of the MO’s in carborane clusters.1 ...... 3

- Figure 1-4. Carborane anion HCB11H11 ...... 9

- Figure 2-1. H2C2B10H10 1, HCB11H11 anion 2, and zwitterionic Ir complex 3...... 20

Figure 2-2. Synthesis of the cyclometalated Ir complexes 6 ...... 22

Figure 2-3. Synthesis of anionic Ir(I) complex 7...... 23

- Figure 3-1. Representations of H2C2B10H12 1, the isoelectronic HCB11H11 anion 2, and

- the 10-vertex CB9H10 anion 8 ...... 30

Figure 3-2. Synthesis of 9[Li+] and complex 10[Li+] ...... 32

Figure 3-3. Synthesis of complexes 12[Li+] and 8[Li+]...... 35

Figure 4-1. Synthesis of isosteric/electronic dianionic and neutral Pd(0) complexes 2 and 4 ...... 42

Figure 4-2. Solid-state structures of 2 and 4...... 44

Figure 4-3. (A) Rapid reaction of complex 2 with Cl-Ph at room temperature to afford a

9:1 ratio of species 5 and 6, respectively. (B) Lability of the phosphines bound to 2, as demonstrated by the stoichiometric addition of P(Cy)3 to afford a distribution of mono- and disubstitution products (B)...... 46

Figure 4-4. Plausible reaction profiles for the formation of the complexes 5 (black) and 6 (blue) ...... 49

- - Figure 5-1. (top) H2C2B10H10 1, nido-7,8-C2B10H12 2, and HCB11H11 3 Representations of anionic carboranyl phosphine 4, phosphine sulfonates 5, and trifluoroborate phosphines 6...... 56

Figure 5-2. Synthesis of complex 7...... 58

Figure 5-3. Solid-state structure of complex 7...... 60

Figure 6-1. The dicarbollide ion 1, a generic bis(dicarbollide) sandwich complex 2, and a generic NHC 3...... 68

Figure 6-2. Synthesis of the novel anionic nido-amine 5 and its subsequent reaction with

N-mesityl oxazolinium tetrafluoroborate to produce the N-nido-imidazolium zwitterion 6.

...... 69

Figure 6-3. Solid-state structure of N-dicarbollide imidazolium monoanion 7 ...... 70

Figure 6-4. Synthesis of the novel dicarbollide monoanion 7...... 71

Figure 6-5. Solid-state structure of the N-dicarbolliide NHC dianion 8...... 72

Figure 6-6. 1H{11B} NMR of Compound 5 (300 MHz, Acetonitrile-d3, 25 ˚C) ...... 75

Figure 6-7. 11B{1H} NMR of Compound 5 (96 MHz, Acetonitrile-d3, 25 ˚C) ...... 75

Figure 6-8. 13C{1H} NMR of Compound 5 (126 MHz, THF, 25 ˚C) ...... 76

Figure 6-9. 1H NMR of Compound 6 (500 MHz, THF-d8, 25 ˚C)...... 77

Figure 6-10. 1H{11B} NMR of Compound 6 (300 MHz, Acetonitriel-d3, 25 ˚C) ...... 78

Figure 6-11. 11B{1H} NMR of Compound 6 (96 MHz, Acetonitrile-d3, 25 ˚C) ...... 78

Figure 6-12. 11B NMR of Compound 6 (96 MHz, Acetonitrile-d3, 25 ˚C) ...... 79

Figure 6-13. 13C{1H} NMR of Compound 6 (126 MHz, THF-d8, 25 ˚C) ...... 79

Figure 6-14. 1H NMR of Compound 7 (500 MHz, THF-d8, 25 ˚C)...... 81

Figure 6-15. 11B{1H} of Compound 7 (96 MHz, THF-d8, 25 ˚C) ...... 81

Figure 6-16. 13C{1H} NMR of Compound 7 (96 MHz, DME, 25 ˚C) ...... 82

Figure 6-17. 1H NMR of Compound 8 (500 MHz, THF-d8, 25 ˚C)...... 83

Figure 6-18. 1H{11B} NMR of Compound 8 (300 MHz, THF-d8, 25 ˚C) ...... 84

Figure 6-19. 11B{1H} NMR of Compound 8 (96 MHz, THF-d8, 25 ˚C) ...... 84

Figure 6-20. 13C{1H} NMR of Compound 8 (126 MHz, THF-d8, 25 ˚C) ...... 85

Figure 6-21. HSQC (13C-1H) NMR of Compound 8 (126 MHz, THF-d8, 25 ˚C)...... 85

List of Schemes

Scheme 1-1. Synthesis of o-carborane from decaborane...... 5

Scheme 1-2. Substitution of o-carborane with iPr2PCl ...... 6

- Scheme 1-3. Structure and numbering of nido-7,8-C2B9H12 and depiction of dynamic bridging BHB bond...... 7

- Scheme 1-4. General synthesis of nido-7,8-H2C2B9H10 and the ensuing dicarbollide...... 7

Scheme 1-5. Formation of a dicarbollide with a strong base...... 8

- Scheme 1-6. Synthesis of HCB11H11 from decaborane B10H14...... 10

Scheme 1-7. Functionalization of HCB11H11 ...... 11

- - Scheme 1-8. Synthesis of 2-HCB9H9 and the 1-HCB9H9 ...... 12

Scheme 1-9. Functionalization of HCB9H9 ...... 13

xii

Chapter 1: Introduction

1.1 Background

The development of novel organometallic ligands bearing carborane clusters as ligand substituents, and the study of their unique chemical and magnetic properties is the focus of this dissertation. Carboranes are polyhedral clusters composed of boron and carbon varying in size and shape as well as carbon and boron content. The dicarba-closo- dodecaborane, more commonly known as the ortho-carborane, or simply o-carborane

(Figure 1-1), was prepared in 1957 by Reaction Motors, Inc. and first reported in literature jointly by Reaction Motors and Olin-Mathieson Corporation in 1963. O-carborane resulted as a byproduct from attempts to develop high-energy rocket fuels during the Cold War. At the time, boron hydride combustion yielded much higher energy outputs than available hydrocarbon fuels,2 however, further development on this front proved to be futile as combustion products of these fuels were detrimental to jet engine function. Large stockpiles of precursory boron rich materials were left Figure 1-1. Dicarba-closo-dodecaborane. More commonly known as o-carborane behind once this industry diminished. This has or H2C2B10H10. resulted in an influx of boron chemistry developments over the last few decades in attempts to prepare useful materials. Along with chemical developments in carborane chemistry came an inherent scientific curiosity to better understand these anomalously stable clusters, a stability often attributed to their structure and bonding. This curiosity brought about the discovery of a number of carborane clusters. Of specific interest to my dissertation

- - - research includes the nido-7,8 H2C2B9H10 , HCB11H11 , and the HCB9H9 anions and their functionalized derivatives (Figure 1-2).

- - - Figure 1-2. Structure and numbering of nido-7,8 H2C2B9H10 , HCB11H11 , and the HCB9H9 anions. Each

maroon number represents a boron vertex.

Note that the bonding framework of these carborane clusters is atypical of normal organic compounds. Each bond drawn between the carbon and boron atoms of the cluster does not represent a 2 centered, 2 electron bond, but is in fact used to illustrate the overall shape of the carborane clusters; in the case of o-carborane, an icosahedral geometry. The numbering of cluster vertices is based on the highest order symmetry axis of the polyhedra and starts with the highest atomic numbered heteroatom, which is carbon for the clusters discussed in this work. After the heteroatoms, the successive belts are numbered in a clockwise manner as shown in Figures 1-1 and 1-2. For clarity throughout this document, the chemical formulas, rather than formal names, will be used and are always written from left to right starting with carbon, for example, C2B10. Also note that any substituents on carbon and boron will be indicated to the left of carbon and the right of boron respectively, such as 1,2-H2C2B10H10 for the o-carborane pictured in Figure 1. The substituents bound

to boron and carbon atoms are not considered part of the skeletal structure and are deemed exo-bonded substituents. Bonds between carbon and boron atoms throughout the cluster are deemed endo-bonds and contribute to the cluster stability.

The stability of carborane clusters comes from their unique bonding framework.

Contrary to classic organic chemistry, the skeletal carbon and boron atoms in carboranes can have as many as six adjacent atoms in the cluster.

This hyper-coordination provides for exceptionally stable molecular structures. Counting electrons in clusters has become more simplified due to the development of polyhedral skeletal electron pair theory. This theory lays out a set of rules, known as

Figure 1-3. Orientation of px, py, and Wade’s rules, for counting electrons within carborane spz orbitals displaying the orientation of the MO’s in carborane clusters.1 clusters.3 Wade’s rules provides a formula to determine the number of skeletal electron pairs in relation to geometry and the number of vertices i.e. n + 1 pairs, where n is the number of vertices. Therefore, in the case of closo- carboranes having 12 vertices, the number of molecular bonding orbitals, or electron pairs, is 13. Each vertex provides a pair of px and py orbitals tangential to the cluster surface,

1 together with an spz orbital directed toward the center of the cluster (Figure 1-3 ). This results in the formation of n px,y bonding MOs in the polyhedral surface, plus a unique bonding MO inside the cluster that is formed by in-phase overlap of the spz hybrid orbitals directed toward the center. The total number of bonding MOs is therefore n + 1, as stated above.

The electron-delocalized covalent bonding within the endo-bonding framework has earned carboranes the reputation of being 3-dimensional aromatics and as a result, are often called benzene analogs. Nucleus-independent chemical shift (NICS) values, used to quantify aromaticity, are determined using density functional theory and provide a negative numerical measurement indicative of electron delocalization. The more negative the number, the more aromaticity a compound exhibits. The H2C2B10H10 has a NICS value of

-34.1 ppm4 as compared to -9.7 ppm for benzene.5 Contrary to benzene, however, studies revealed that the hydrogen exo-bonded substituents of the carbon vertices of the cluster are acidic enough to be removed with a strong base (pKa = is 22.0).6 This allows for further functionalization of the cluster at the carbon vertices, leading to vast opportunities for tuning of the steric and chemical properties, as well as covalently linking the clusters as ligand substituents, the basis of my research.

1.2 Synthesis and functionalization of H2C2B10H10

In 1963, Reaction Motors and Olin-Mathieson Corporation first reported the synthesis of H2C2B10H10 from decaborane, an open cluster composed of ten borons and fourteen hydrogen substituents, four of which are bridging hydrides (Scheme 1-1). Typical reaction conditions require the presence of a weak Lewis base, such as acetonitrile, to induce the loss of H2 gas, resulting in the formation of a reactive Lewis base adduct.

Refluxing this adduct in the presence of a wide variety of acetylene groups affords the corresponding H2C2B10H10 (Scheme 1-1). The reaction with acetylenes is quite robust and can be performed in the presence of a wide variety of functional groups, including

carbamates, ethers, esters, nitro groups and many more, but will not proceed in the presence of other strong Lewis bases as this will decompose the starting decaborane cluster.

Scheme 1-1. Synthesis of o-carborane from decaborane. Unlabeled vertices are BH bonds and omitted for clarity.

As stated above, the pKa of the C-H bonds in H2C2B10H10 is 22.0 and when R and

R’ are hydrogen, the carborane can be deprotonated with a strong base, such as n- butyllithium. Subsequent substitution with the electrophile of choice, for example iPr2PCl, results in the desired functionalized carborane (Scheme 1-2). The ease of functionalization of H2C2B10H10 has led to the development of a large library of clusters, specifically tailored to many applications such as medicinal chemistry, with studies in boron neutron capture therapy (BNCT) for cancer treatment, amino acid substitution using carborane as a benzene analog, and magnesium electrolyte applications,7 as well as many other fields of chemical research.1 However, its utility in transition metal catalysis has not been successful.8

Scheme 1-2. Substitution of o-carborane with iPr2PCl. Unlabeled vertices are BH bonds and omitted for

clarity.

1.3 Synthesis of nido-carborane 7,8-C2B9H12

Although H2C2B10H10 is quite robust and relatively inert toward acidic media and strong oxidants, it is susceptible to decomposition in basic media. Deboronation, or removal of a boron vertex, in the presence of a nucleophile, affords an open bowl like

- structure known as the nido-carborane (Figure 1-4). The nido-carborane 7,8-C2B9H12 anion was first prepared by Frederick Hawthorne in 1964 via the addition of potassium

9 hydroxide to a methanol solution of the H2C2B10H10. Nucleophilic attack of the B(3) or

B(6) position results in the loss of a boron vertex and the formation of the desired nido-

- carborane, 7,8-C2B9H12 anion. Upon degradation of H2C2B10H10, only one boron atom is removed and leaves as a “B+” leaving its hydrogen behind (Scheme 1-3). The resulting nido-carborane contains the C2B9 skeletal structure bearing 11 exo-bonded hydrogens as well as a dynamic bridging hydride that moves across B(9), B(10) and B(11). Diffraction studies show this hydrogen is observed above the open face of the polyhedron almost equidistant to all five atoms of the pentagonal open face, but is experimentally observed as

a dynamic bridging BHB bond commonly seen in 1H NMR as a broad resonance at approximately -3 ppm (Scheme 1-3).10

- Scheme 1-3. Structure and numbering of nido-7,8-C2B9H12 and depiction of dynamic bridging BHB bond.

- Scheme 1-4. General synthesis of nido-7,8-H2C2B9H10 and the ensuing dicarbollide. Unlabeled vertices are BH bonds omitted for clarity. Removal of a boron vertex leaves the number of bonding MO’s unchanged, giving it n+2, or 13, skeletal electron pairs. Furthermore, the bridging hydride has a pKa of 13.5 and can easily be removed by a strong base,11 for example, n-butyllithium or sodium hydride. The resulting cluster is an open faced dianionic bowl like structure, known as a dicarbollide, which can react further with various metal centers, creating a metallocarborane (Scheme

1-4).12

Scheme 1-5. Formation of a dicarbollide with a strong base followed by the formation of a metallocarborane. Unlabeled vertices are BH bonds omitted for clarity.

- Since the discovery of nido-carborane, 7,8-C2B9H12 in 1964, the majority of research has focused on metallocarboranes and their use as cyclopentadienyl analogs.9, 13 Depending on the conditions, the nido-carborane also acts as a coordinating anion and can have anywhere from 1 to 5 binding modes with various metal centers. More recently, interest has shifted with the discovery of nido-phosphines14 and bidentate phosphines, which luminesce when bound to gold and silver metal centers.15 A pyridine substituted nido- carborane was also found to possess luminescent properties with high quantum yields, and is a promising source for the development of blue organic light emitting diodes.16

Metallocarboranes have also been used for ethylene polymerization and hydrogenation catalysis.17 The discovery of the nido-carborane was quite a revelation as it provided a synthetically versatile substituent from the fairly inexpensive, commercially available

H2C2B10H10. However, its applications in catalysis have yet to surpass those of ubiquitous alkyl or aryl transition metal ligands.

- 1.4 Synthesis and functionalization of HCB11H11

In addition to the neutral H2C2B10H10 and nido-carborane, another carborane of

- interest is carba-closo-dodecaborate (HCB11H11 , Figure 1-4), a closed-cage icosahedral anionic carborane. Similar to H2C2B10H10, the

- HCB11H11 carborane anion has 12 vertices, but rather than 2 carbons and 10 borons within the skeletal structure, it has 1 carbon and 11 borons.

The parent carborane anion has similar features as the neutral H2C2B10H10, having 12 vertices, each with a 3 centered 2 electron bonding

Figure 1-4. Carborane anion HCB H - framework, hypercoordination of 6 bonds to each 11 11 skeletal atom, 26 delocalized skeletal electrons and 3-dimensional aromaticity (NICS value

-34.36 ppm). The electron count of an icosahedron must be 26 within the endo-bonding framework of the cluster. By swapping a carbon vertex containing 4 valence electrons (as seen in the neutral H2C2B10H10), with a boron vertex only containing 3 valence electrons

- (as seen in the HCB11H11 anion) the cluster requires one more electron to maintain the 26

- skeletal electron count, rendering the cluster anionic. The anionic nature of the HCB11H11 anion provides interesting properties that our lab is currently exploring for applications in various fields of chemistry, including catalysis, battery applications, materials, and

- bioorthogonal chemistry. Most importantly, the HCB11H11 anion is significantly more stable than its isoelectronic “cousin” H2C2B10H10. Bearing a negative charge, the cluster is not susceptible to nucleophilic attack and subsequent boron extrusion, but instead acts

as a weakly coordinating anion.18 Its derivatives have been found to stabilize the most reactive cationic intermediates, such as protonated fullerene,19 protonated benzene, and more recently, the stabilization of aryl cations for the activation of a wide variety of

20 - 21 hydrocarbons. The HCB11H11 anion was first synthesized by Knoth in 1967 from decaborane. Further improvements on the chemical synthesis were made by Reed in

201018q providing the below synthesis (Scheme 1-5) used in our laboratory.

- Scheme 1-6. Synthesis of HCB11H11 from decaborane B10H14. Unlabeled vertices are B omitted for clarity. The stability of the carborane can be enhanced by substitution of the B-H bonds for

B-X bonds, where X is Cl, Br, or I.18q This process is reminiscent of electrophilic aromatic substitution resulting in larger, more stable, less nucleophilic, and more weakly coordinating ions. This process is conveniently sequential starting at the most hydridic antipodal position B(12).22 This allows for the isolation of the mono-halogenated, hexa- halogenated and per-halogenated derivatives (Scheme 1-6). In addition to functionalization of the boron vertices, the carbon vertex can be functionalized as well, in

- 6 a manner similar to H2C2B10H10. The CH bond in HCB11H11 has a pKa of 21.8 and can be deprotonated using n-BuLi. Following deprotonation, the most common addition performed in our laboratory is the incorporation of a phosphine chloride, such as iPr2PCl,

affording the desired carbon functionalized anionic carborane. (Scheme 1-6).

Functionalization at the carbon vertex as well as the boron vertices creates endless opportunities for tuning the electronic and steric properties of the ensuing carboranyl ligand.

Scheme 1-7. Functionalization of HCB11H11 at boron (top) and at carbon (bottom). Unlabeled vertices are B omitted for clarity. X = I, Br, or Cl and R = H, I, Br, or Cl.

- 1.5 Synthesis and functionalization of HCB9H9

- Analogous to the 12 vertex carborane anion, HCB11H11 , the 10 vertex carborane

- anion, a gyroelongated square antiprism of the formula HCB9H9 , is showing increasing interest as its functionality is quite similar to the 12 vertex anion. It is a fairly weakly coordinating anion maintaining a closo structure, and has 2n+2, or 22, delocalized skeletal electrons. Its vertices consist of hypervalent carbon and boron atoms bonded via 3 centered two electron bonding and 2 centered 2 electron exo-bonded hydrogens. As in, the

- carboranes discussed above, the delocalized sigma bonding in HCB11H11 provides stability

and is deemed 3 dimensionally aromatic with a NICS value of -29.91.4 Furthermore, it has found its place in a vast array of chemical industries including liquid crystals, nonlinear optics, and cancer treatment drugs by delivery of radioactive halogens, and similar to the

- parent anion, HCB11H11 , it has been somewhat exploited as a weakly coordinating anion.

The first preparation of the 10 vertex anionic cluster was discovered by Knoth in 197123 through the reduction of nido-6-(Me3N)CB9H11 with sodium metal. However, a much simpler preparation has been developed using a Brellochs reaction (Scheme 1-7),24 and is the sole method of synthesis in our laboratory. Interestingly, due to the symmetry of the

- - 10 vertex carborane, two isomers exist, the 2-HCB9H9 and the 1-HCB9H9 (Scheme 1-7).

In our synthesis, the 2 isomer is isolated, however, the 1 isomer is 20.63 kcal/mol lower in energy and can be obtained by thermal isomerization.

- - Scheme 1-8. Synthesis of 2-HCB9H9 and the 1-HCB9H9 . Unlabeled vertices are B omitted for clarity.

- - The reactivity of the HCB9H9 is very similar to the HCB11H11 anion and can be functionalized at the boron and carbon vertices in the same manner. The process is sequential starting at the antipodal boron B(9) and moving up each belt. This allows for the isolation of the penta-halogenated and nona-halogenated derivatives (Scheme 1-8).

- Functionalization at the carbon vertex is done in the same manner as HCB11H11 using n-

BuLi followed by the addition of an electrophile of choice, affording the desired carbon functionalized anionic carborane (Scheme 1-8). Having a 12 vertex and a 10 vertex ligand scaffold provides even more opportunities for ligand development not only by having

- differing steric profiles, but also differing electronics with HCB9H9 having a more localized negative charge.

Scheme 1-9. Functionalization of HCB9H9 at Boron (top) and at Carbon (bottom). Unlabeled vertices are B omitted for clarity. X = I, Br, or Cl and R = H, I, Br, or Cl. The purpose of my dissertation research was to use the various carboranes

- - discussed above, H2C2B10H10, nido-7,8 H2C2B9H10, HCB11H11 , and HCB9H9 , as ligand substituents in the development of novel transition metal complexes and investigate their inherent properties as well as catalytic activity. The following chapters will discuss the pursuits and successes of this research with the preparation of: two isoelectronic iridium(I)

carboranyl phosphine complexes displaying starkly different stabilities, the preparation of two rhodium(I) carbonyl complexes for the determination of the inductive effects of

- - HCB11H11 and HCB9H9 , a palladium(0) complex bearing the C2B10H10 carborane as a ligand substituent displaying the role of charge in the oxidative addition of chlorobenzene, a novel zwitterionic palladium olefin polymerization catalyst, the preparation of a novel nido-carboranyl imidazolium and exploration of its reactivity, and finally the formation of the first isolable triazole radical.

1.6 References

1. A2 - Grimes, Russell N. In Carboranes (Second Edition), Academic Press: Oxford, 2011; pp 1107-1139.

2. R. L. Hughes, I. C. S., E. W. Lawless, In Production of the Boranes and Related Research. Academic Press: New York, 1967.

3. Wade, K., The structural significance of the number of skeletal bonding electron- pairs in carboranes, the higher boranes and borane anions, and various transition-metal carbonyl cluster compounds. Journal of the Chemical Society D: Chemical Communications 1971, (15), 792-793.

4. Schleyer, P. v. R.; Najafian, K., Stability and Three-Dimensional Aromaticity of closo-Monocarbaborane Anions, CBn-1Hn-, and closo-Dicarboranes, C2Bn-2Hn. Inorganic Chemistry 1998, 37 (14), 3454-3470.

5. Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Hommes, N. J. R. v. E., Nucleus-Independent Chemical Shifts: A Simple and Efficient Aromaticity Probe. Journal of the American Chemical Society 1996, 118 (26), 6317-6318.

6. Valliant, J. F.; Guenther, K. J.; King, A. S.; Morel, P.; Schaffer, P.; Sogbein, O. O.; Stephenson, K. A., The medicinal chemistry of carboranes. Coordination Chemistry Reviews 2002, 232 (1–2), 173-230.

7. Carter, T. J.; Mohtadi, R.; Arthur, T. S.; Mizuno, F.; Zhang, R.; Shirai, S.; Kampf, J. W., Boron Clusters as Highly Stable Magnesium-Battery Electrolytes. Angewandte Chemie International Edition 2014, 53 (12), 3173-3177.

8. Kumada, M.; Sumitani, K.; Kiso, Y.; Tamao, K., Silicon hydrides and nickel complexes. Journal of Organometallic Chemistry 1973, 50 (1), 319-326.

9. Wiesboeck, R. A.; Hawthorne, M. F., Dicarbaundecaborane(13) and Derivatives. Journal of the American Chemical Society 1964, 86 (8), 1642-1643.

10. Plesek, J., Potential applications of the boron cluster compounds. Chemical Reviews 1992, 92 (2), 269-278.

11. Farràs, P.; Teixidor, F.; Branchadell, V., Prediction of pKa Values of nido- Carboranes by Density Functional Theory Methods. Inorganic Chemistry 2006, 45 (19), 7947-7954.

12. Hawthorne, M. F.; Young, D. C.; Wegner, P. A., Carbametallic Boron Hydride Derivatives. I. Apparent Analogs of Ferrocene and Ferricinium Ion. Journal of the American Chemical Society 1965, 87 (8), 1818-1819.

13. (a) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Stable Silver Complexes with C2B9H12- derivatives. Inorganic Chemistry 1994, 33 (9), 1756-1761; (b) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Modulation of the B(3)-H.fwdharw.Ru Distances in 7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1994, 13 (7), 2751-2760; (c) Teixidor, F.; Ayllòn, J. A.; Viñas, C.; Kivekäs, R.; Sillanpää, R.; Casabò, J., Mercury coordination to Exo-dithio-7,8- dicarba-nido-undecaborate derivatives. Journal of Organometallic Chemistry 1994, 483 (1), 153-157; (d) Viñas, C.; Nuñez, R.; Teixidor, F.; Kivekäs, R.; Sillanpää, R., Modulation of Agostic B−H⇀Ru Bonds in exo-Monophosphino-7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1996, 15 (18), 3850-3858; (e) Teixidor, F.; Flores, M. A.; Viñas, C.; Kivekäs, R.; Sillanpää, R., Influence of S-Aryl Groups in the Coordination and Reactivity of (nido-Thiocarborane)ruthenium Complexes. Organometallics 1998, 17 (21), 4675-4679; (f) Teixidor, F.; Flores, M. A.; Viñas, C.; Sillanpää, R.; Kivekäs, R., exo-nido- Cyclooctadienerhodacarboranes: Synthesis, Reactivity, and Catalytic Properties in Alkene Hydrogenation. Journal of the American Chemical Society 2000, 122 (9), 1963-1973; (g) Teixidor, F.; Núñez, R.; Flores, M. A.; Demonceau, A.; Viñas, C., Forced exo-nido rhoda and ruthenacarboranes as catalyst precursors: a review. Journal of Organometallic Chemistry 2000, 614–615, 48-56; (h) Poater, J.; Solà, M.; Viñas, C.; Teixidor, F., π Aromaticity and Three-Dimensional Aromaticity: Two sides of the Same Coin? Angewandte Chemie International Edition 2014, 53 (45), 12191-12195.

14. Popescu, A. R.; Teixidor, F.; Viñas, C., Metal promoted charge and hapticities of phosphines: The uniqueness of carboranylphosphines. Coordination Chemistry Reviews 2014, 269, 54-84.

15. Crespo, O.; Dı́az, C.; O’Dwyer, C.; Gimeno, M. C.; Laguna, A.; Ospino, I.; Valenzuela, M. L., Luminescent Gold and Silver Complexes with the Monophosphane 1- (PPh2)-2-Me-C2B10H10 and Their Conversion to Gold Micro- and Superstructured Materials. Inorganic Chemistry 2014, 53 (14), 7260-7269.

16. Axtell, J. C.; Kirlikovali, K. O.; Djurovich, P. I.; Jung, D.; Nguyen, V. T.; Munekiyo, B.; Royappa, A. T.; Rheingold, A. L.; Spokoyny, A. M., Blue Phosphorescent Zwitterionic Iridium(III) Complexes Featuring Weakly Coordinating nido-Carborane- Based Ligands. Journal of the American Chemical Society 2016, 138 (48), 15758-15765.

17. Yinghuai, Z.; Hosmane, N. S., Carborane-based transition metal complexes and their catalytic applications for olefin polymerization: Current and future perspectives. Journal of Organometallic Chemistry 2013, 747, 25-29.

18. (a) Finze, M., Kohlenstoff-Extrusion/Cluster-Kontraktion: Synthese des fluorierten Cyano-closo-undecaborats K2[3-NC-closo-B11F10]. Angewandte Chemie 2007, 119 (46), 9036-9039; (b) Finze, M., Carbon Extrusion/Cluster Contraction: Synthesis of the Fluorinated Cyano-closo-Undecaborate K2[3-NC-closo-B11F10]. Angewandte Chemie International Edition 2007, 46 (46), 8880-8882; (c) Molinos, E.; Brayshaw, S. K.; Kociok- Kohn, G.; Weller, A. S., Cationic rhodium mono-phosphine fragments partnered with carborane monoanions [closo-CB11H6X6]- (X = H, Br). Synthesis, structures and reactivity with alkenes. Dalton Transactions 2007, (42), 4829-4844; (d) Douvris, C.; Ozerov, O. V., Hydrodefluorination of Perfluoroalkyl Groups Using Silylium-Carborane Catalysts. Science 2008, 321 (5893), 1188-1190; (e) Geis, V.; Guttsche, K.; Knapp, C.; Scherer, H.; Uzun, R., Synthesis and characterization of synthetically useful salts of the weakly-coordinating dianion [B12Cl12]2. Dalton Transactions 2009, (15), 2687-2694; (f) Gu, W.; Haneline, M. R.; Douvris, C.; Ozerov, O. V., Carbon−Carbon Coupling of C(sp3)−F Bonds Using Alumenium Catalysis. Journal of the American Chemical Society 2009, 131 (31), 11203-11212; (g) Knapp, C.; Schulz, C., How to overcome Coulomb explosions in labile dications by using the [B12Cl12]2- dianion. Chemical Communications 2009, (33), 4991-4993; (h) Bolli, C.; Derendorf, J.; Keßler, M.; Knapp, C.; Scherer, H.; Schulz, C.; Warneke, J., Synthese, Kristallstruktur und Reaktivität des starken Methylierungsmittels Me2B12Cl12. Angewandte Chemie 2010, 122 (20), 3616- 3619; (i) Bolli, C.; Derendorf, J.; Keßler, M.; Knapp, C.; Scherer, H.; Schulz, C.; Warneke, J., Synthesis, Crystal Structure, and Reactivity of the Strong Methylating Agent Me2B12Cl12. Angewandte Chemie International Edition 2010, 49 (20), 3536-3538; (j) Derendorf, J.; Ke; Knapp, C.; Ruhle, M.; Schulz, C., Alkali metal-sulfur dioxide complexes stabilized by halogenated closo-dodecaborate anions. Dalton Transactions 2010, 39 (37), 8671-8678; (k) Douvris, C.; Nagaraja, C. M.; Chen, C.-H.; Foxman, B. M.; Ozerov, O. V., Hydrodefluorination and Other Hydrodehalogenation of Aliphatic Carbon−Halogen Bonds Using Silylium Catalysis. Journal of the American Chemical Society 2010, 132 (13), 4946- 4953; (l) Duttwyler, S.; Douvris, C.; Fackler, N. L. P.; Tham, F. S.; Reed, C. A.; Baldridge, for Incipient Phenyl Cation Reactivity. Angewandte Chemie 2010, 122 (41), 7681-7684; (m) Duttwyler, S.; Douvris, C.; Fackler, N. L. P.; Tham, F. S.; Reed, C. A.; Baldridge, K.

Compounds of Group 13 Elements (E = Al, Ga, In) Stabilized by the Weakly Coordinating Dianion [B12Cl12]2–. Organometallics 2011, 30 (14), 3786-3792; (t) Bolli, C.; Köchner, T.; Knapp, C., [NO][HCB11Cl11] – Synthesis, Characterization, Crystal Structure, and Reaction with P4. Zeitschrift für anorganische und allgemeine Chemie 2012, 638 (3-4), 559-564; (u) Ibad, M. F.; Langer, P.; Reiß, F.; Schulz, A.; Villinger, A., Catalytic Trimerization of Bis-silylated Diazomethane. Journal of the American Chemical Society 2012, 134 (42), 17757-17768; (v) Ramirez-Contreras, R.; Ozerov, O. V., Convenient C- alkylation of the [HCB11Cl11]- carborane anion. Dalton Transactions 2012, 41 (26), 7842-7844.

19. Reed, C. A., Carborane acids. New "strong yet gentle" acids for organic and inorganic chemistry. Chemical Communications 2005, (13), 1669-1677.

20. Shao, B.; Bagdasarian, A. L.; Popov, S.; Nelson, H. M., Arylation of hydrocarbons enabled by organosilicon reagents and weakly coordinating anions. Science 2017, 355 (6332), 1403-1407.

21. Knoth, W. H., 1-B9H9CH- and B11H11CH. Journal of the American Chemical Society 1967, 89 (5), 1274-1275.

22. Hoffmann, R.; Lipscomb, W. N., Theory of Polyhedral Molecules. III. Population Analyses and Reactivities for the Carboranes. The Journal of Chemical Physics 1962, 36 (12), 3489-3493.

23. Knoth, W. H., B10H12CNH3, B9H9CH-, B11H11CH-, and metallomonocarboranes. Inorganic Chemistry 1971, 10 (3), 598-605.

24. Franken, A.; Kilner, C. A.; Thornton-Pett, M.; Kennedy, J. D., Monocarbaborane anion chemistry. An interesting encapsulation of the Pd2I2{P(C6H4-4-Me)3}4]2+ cation by a pair of [PhCB9H4I(C6H4Me)4]- anions. Chemical Communications 2002, (18), 2048-2049.

Chapter 2: Resisting B-H Oxidative Addition: The Divergent Reactivity of the o- - Carborane (C2B10H10) and Carba-closo-dodecaborate (CB11H11 ) Ligand Substituents

2.1 Abstract

Here, we report a study of two isoelectronic Ir(I) complexes supported by different

- carboranyl phosphines, bearing either C2B10H10 or CB11H11 ligand substituents. The neutral Ir(I) complex containing the C2B10H10 phosphine ligand is not isolable and undergoes spontaneous B-H cyclometalation to afford an Ir(III) hydride. In contrast, the

- anionic Ir(I) complex supported by a phosphine with a CB11H11 ligand R-group is stable towards B-H activation. This divergent reactivity has important implications for the design of carborane containing ligands for catalysis. Both compounds are fully characterized by multinuclear NMR spectroscopy, HRMS spectrometry, and single crystal x-ray diffraction studies.

2.2 Introduction

The proliferation of contemporary homogenous catalytic methods is largely due to the availability of a diverse array of ligand architectures. Due to their shape and 3- dimensional aromaticity, icosahedral carboranes1 are an interesting alternative to classical

2 alkyl and aryl ligand substituents. Soon after the discovery of H2C2B10H10 1 (Figure 2-1, left) in 1963 a carboranyl phosphine ligand3 and complex4 were reported by Smith.

Kumada5 was the first to implement such ligands in homogenous catalysis, but these and

6 all subsequent systems containing C2B10H10 1 fragments have not yielded catalysts that surpass the activity of systems supported by ubiquitous trialkyl or triaryl phosphines.

Perhaps this non-competitive catalytic behavior is due to the decomposition of the cluster

via well-known B-vertex extrusion1h, 7 and B-H cyclometalation reactions. Hawthorne8 was the first to observe intramolecular B-H oxidative addition of an o-carboranyl phosphine, as well as analogous intermolecular reactions with unsubstituted H2C2B10H10 icosahedra at an

Ir(I) center.

- Figure 2-1. H2C2B10H10 1, HCB11H11 anion 2, and zwitterionic Ir complex 3. Unlabeled vertices = B-H.

- In contrast to H2C2B10H10 1, the isoelectronic and weakly coordinating HCB11H11 anion1b 2 (Figure 2-1, center) is not susceptible to B-vertex extrusion reactions and is renowned for its inert properties, particularly when polyhalogenated.1i, 9 We recently

- reported the first utilization of HCB11H11 as a ligand substituent for a transition metal- based catalyst.10 Notably, this zwitterionic Au(I) complex, featuring an anionic carboranyl phosphine, is far more active than all known systems for the hydroamination of alkynes.

In addition, we reported an unusually stable pseudo-low coordinate zwitterionic Ir(I) compound 3 (Figure 2-1, right) that does not undergo spontaneous intramolecular B-H oxidative addition.11 Although not identical, 3 is reminiscent of Hawthorne's unstable Ir(I)

8 - 12 complexes and its behavior suggests that ligands bearing a HCB11H11 substituent are inherently more resistant to B-H activation. Given our interest in catalyst design we are

- seeking stronger evidence that the HCB11H11 ligand substituent has enhanced chemical

stability towards B-H oxidative addition. Here, we report such a study by the preparation and analysis of two isoelectronic iridium (I) complexes containing phosphines with

- H2C2B10H10 and HCB11H11 moieties, respectively. Where the former H2C2B10H10 containing complex is not isolable and spontaneously undergoes cyclometalation at one of

- the B-H vertices, the HCB11H11 analogue is stable.

2.3 Results and Discussion

13 Reacting the known C2B10H10 phosphine 4 with ½ an equivalent of (ClIr(COD))2 in hexane solvent results in a color change from light orange to red orange and subsequent formation of a precipitate (Figure 2-2, below). Based on Hawthorne's prior observations8, it was predicted that the ensuing Ir(I) complex 5 would be unstable towards cyclometalation and produce the corresponding Ir(III) complex 6. Indeed, analysis of the precipitate by 1H NMR spectroscopy shows clean formation of an iridium hydride at -16.42

2 ppm (d, JP-H = 13.1 Hz), suggesting cyclometalation of the C2B10H10 cluster. The coupling

2 constant is consistent with an iridium (III) hydride disposed cis to a phosphine (range JP-

8 1 H from 10 to 20 Hz). In addition to the expected H signals for the phosphine iPr-groups and COD protons, two carborane C-H resonances appear in a 4:1 ratio. The presence of two carborane C-H signals suggests the formation of a diastereomeric mixture (6’, 6’’), resulting from competitive B-H oxidative addition at the B3 and B6 positions (Figure 2-2).

Such competitive cyclometalations have been observed previously14 with other metals, and occur preferentially at B3/B6 positions rather than at B4/B5, since the B3/B6 boron atoms are more activated (attached to 2-carbon atoms). All other ligand 1H resonances from the two diastereomers 6’ and 6’’ are superimposed by coincidence. Analysis of the 31P NMR

spectra also shows a 4:1 mixture of compounds and is in agreement with our assignment.

The 11B NMR spectrum is uninstructive due to the multitude of broad overlapping resonances, which cannot be resolved. However, the solution and solid-state infrared

- spectra of complexes 6 show bands typical for B-H stretches of C2B10H10 (2460-2625 cm

1), as well as the presence of an Ir-H absorbance at 2223 cm-1.8

Figure 2-2. Synthesis of the cyclometalated Ir complexes 6. Solid-state structure of diastereomer 6’’. Color code: Gray = C, white = H, violet = P, red = Ir, brown = B, green = chlorine. Unlabeled vertices = B-H. Notable bond lengths (Å): Ir-P = 2.320(5); Ir-B = 2.094(2); Ir-Cl = 2.498(5); C1-C2 = 1.384(3); C3-C4 = 1.362(3). Selected bond angles (˚): B-Ir-P = 72.1; P-Ir-Cl = 90.2. A single-crystal X-ray diffraction study of one of the diastereomers 6’’ confirms that cyclometalation occurs adjacent to the two carborane carbon atoms (Figure 2-2, bottom right). Compound 6’’ adopts a distorted octahedral geometry with the hydride, which was located from the difference electron density map, cis to the phosphine ligand. The Ir-P, Ir-

B and Ir-Cl bond lengths are 2.320(5), 2.094(2) and 2.498(5) Å, respectively. The cyclooctadiene C-C double bond lengths trans to the phosphine (C1-C2 = 1.384(3) Å) and

metalated B-vertex (C3-C4 = 1.362(3) Å) are in the typical range (1.35-1.44) for olefins coordinated to Ir.

We next turned our attention to preparing a complex bearing a phosphine with a

- CB11H11 substituent, which is isoelectronic with the unstable complex 5. We predicted that treating the previously reported15 zwitterion 3 with a nucleophilic chloride source should produce the anionic isoelectronic complex 7 (Figure 2-3). Thus, treatment of solution of 3 dissolved in C6H5F with five equivalents of NMe4Cl resulted in the formation of an orange precipitate 7. Analysis of the precipitate by 31P NMR shows the formation of a single new product 7 (s, 32.95 ppm). Importantly, the 1H NMR spectrum of 7 shows no Ir-H

- resonance, suggesting that the CB11H11 substituent is not cyclometalated. Moreover, as indicated by 11B NMR spectroscopy (three resonances 1:5:5 ratio; 5:5 overlapping) the local C5v symmetry of the cluster is retained, confirming that cyclometalation has not occurred. In addition, solution I.R. shows no Ir-H absorbance, ruling out the possibility of a rapid reversible cyclometalation process faster than the NMR time scale.

Figure 2-3. Synthesis of anionic Ir(I) complex 7. Solid-state structure of 7 (note: most hydrogens and + NMe4 countercation omitted for clarity). Color code: gray = C, white = H, violet = P, red = Ir, brown = B, green = Cl. Unlabeled vertices = B–H. Notable bond lengths (Å): Ir–P = 2.371(7); Ir–Cl = 2.394(8); C1– C2 = 1.401(5); C3–C4 = 1.422(5). Selected bond angle (°): P–Ir–Cl = 91.4.

A single-crystal X-ray diffraction structure confirms the identity of the square planar Ir (I) complex 7 (Figure 2-3, previous page). While the Ir-P bond length (2.371(7)

Å) is similar to complex 6, the Ir-Cl (2.394(8) Å) and olefin bond lengths (C1-C2 =

1.401(5) Å, C3-C4 = 1.422(5) Å) are comparatively contracted and elongated, respectively.

These observations can be explained by the decrease in coordination number and oxidation state, which allows for stronger M-L σ and ힹ interactions. The closest B-H approach to the

Ir center is 2.83 (Ir-H1) which is outside the range of typical Ir B-H agostic interactions.

For comparison, the zwitterionic precursor 3 displays two strong agostic interactions with the metal center (Ir-H distances = 1.93(4) and 1.92(4) Å).8

2.4 Summary and Conclusion

- The study above provides strong evidence that the CB11H11 ligand substituent is far less susceptible to intramolecular B-H oxidative addition reactions compared to its neutral cousin, C2B10H10. This observation is important since it supports the notion that C2B10H10 may not be a suitable ligand substituent for catalysts that mediate reactions involving

- oxidative addition/reductive elimination sequences. In addition, the ability of the CB11H11 ligand R-group to resist intramolecular B-H oxidative addition suggests that ligands containing this group should be useful for a variety of catalytic processes.

2.5 Experimental

The previous experimental data has been published and may be obtained online at DOI: org/10.1016/j.jorganchem.2015.05.008

2.6 References

1. (a) Asay, M.; Kefalidis, C. E.; Estrada, J.; Weinberger, D. S.; Wright, J.; Moore, C. E.; Rheingold, A. L.; Maron, L.; Lavallo, V., Isolation of a Carborane-Fused Triazole Radical Anion. Angewandte Chemie International Edition 2013, 52 (44), 11560-11563; (b) Douvris, C.; Michl, J., Update 1 of: Chemistry of the Carba-closo-dodecaborate(−) Anion, CB11H12–. Chemical Reviews 2013, 113 (10), PR179-PR233; (c) Farras, P.; Juarez-Perez, E. J.; Lepsik, M.; Luque, R.; Nunez, R.; Teixidor, F., Metallacarboranes and their interactions: theoretical insights and their applicability. Chemical Society Reviews 2012, 41 (9), 3445-3463; (d) Grimes, R. N., Carboranes in the chemist's toolbox. Dalton Transactions 2015, 44 (13), 5939-5956; (e) Li, Y.; Carroll, P. J.; Sneddon, L. G., Ionic-Liquid-Promoted Decaborane Dehydrogenative Alkyne-Insertion Reactions: A New Route to o-Carboranes. Inorganic Chemistry 2008, 47 (20), 9193-9202; (f) Olid, D.; Nunez, R.; Vinas, C.; Teixidor, F., Methods to produce B-C, B-P, B-N and B-S bonds in boron clusters. Chemical Society Reviews 2013, 42 (8), 3318-3336; (g) Poater, J.; Solà, M.; Viñas, C.; Teixidor, F., π Aromaticity and Three-Dimensional Aromaticity: Two sides of the Same Coin? Angewandte Chemie International Edition 2014, 53 (45), 12191- 12195; (h) Popescu, A. R.; Teixidor, F.; Viñas, C., Metal promoted charge and hapticities of phosphines: The uniqueness of carboranylphosphines. Coordination Chemistry Reviews 2014, 269, 54-84; (i) Reed, C. A., H+, CH3+, and R3Si+ Carborane Reagents: When Triflates Fail. Accounts of Chemical Research 2010, 43 (1), 121-128; (j) Scholz, M.; Hey-Hawkins, E., Carbaboranes as Pharmacophores: Properties, Synthesis, and Application Strategies. Chemical Reviews 2011, 111 (11), 7035-7062; (k) Wright, J. H.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Click-Like Reactions with the Inert HCB11Cl11– Anion Lead to Carborane-Fused Heterocycles with Unusual Aromatic Character. Inorganic Chemistry 2013, 52 (10), 6223-6229.

2. Heying, T. L.; Ager, J. W.; Clark, S. L.; Mangold, D. J.; Goldstein, H. L.; Hillman, M.; Polak, R. J.; Szymanski, J. W., A New Series of Organoboranes. I. Carboranes from the Reaction of Decaborane with Acetylenic Compounds. Inorganic Chemistry 1963, 2 (6), 1089-1092.

3. Smith, H. D.; Knowles, T. A.; Schroeder, H., Chemistry of Decaborane- Phosphorus Compounds. V. Bromocarboranes and Their Phosphination. Inorganic Chemistry 1965, 4 (1), 107-111.

4. Smith, H. D., Complexes of Bisphosphinocarboranes with Nickel(II) Chloride. Journal of the American Chemical Society 1965, 87 (8), 1817-1818.

5. Tamao, K.; Kiso, Y.; Sumitani, K.; Kumada, M., Alkyl group isomerization in the cross-coupling reaction of secondary alkyl Grignard reagents with organic halides in the presence of nickel-phosphine complexes as catalysts. Journal of the American Chemical Society 1972, 94 (26), 9268-9269.

6. (a) Joost, M.; Estévez, L.; Mallet-Ladeira, S.; Miqueu, K.; Amgoune, A.; Bourissou, D., Enhanced π-Backdonation from Gold(I): Isolation of Original Carbonyl and Carbene Complexes. Angewandte Chemie International Edition 2014, 53 (52), 14512-14516; (b) Joost, M.; Estévez, L.; Miqueu, K.; Amgoune, A.; Bourissou, D., Oxidative Addition of Carbon–Carbon Bonds to Gold. Angewandte Chemie International Edition 2015, 54 (17), 5236-5240; (c) Joost, M.; Zeineddine, A.; Estévez, L.; Mallet−Ladeira, S.; Miqueu, K.; Amgoune, A.; Bourissou, D., Facile Oxidative Addition of Aryl Iodides to Gold(I) by Ligand Design: Bending Turns on Reactivity. Journal of the American Chemical Society 2014, 136 (42), 14654-14657.

7. (a) Farràs, P.; Olid-Britos, D.; Viñas, C.; Teixidor, F., Unprecedented B–H Activation Through Pd-Catalysed B–Cvinyl Bond Coupling on Borane Systems. European Journal of Inorganic Chemistry 2011, 2011 (16), 2525-2532; (b) Farràs, P.; Teixidor, F.; Rojo, I.; Kivekäs, R.; Sillanpää, R.; González-Cardoso, P.; Viñas, C., Relaxed but Highly Compact Diansa Metallacyclophanes. Journal of the American Chemical Society 2011, 133 (41), 16537-16552; (c) Popescu, A.-R.; Laromaine, A.; Teixidor, F.; Sillanpää, R.; Kivekäs, R.; Llambias, J. I.; Viñas, C., Uncommon Coordination Behaviour of P(S) and P(Se) Units when Bonded to Carboranyl Clusters: Experimental and Computational Studies on the Oxidation of Carboranyl Phosphine Ligands. Chemistry – A European Journal 2011, 17 (16), 4429-4443; (d) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., A novel B-H[right harpoon up]Ru agostic bond. Crystal structure of [RuCl{7,8-[small micro]-S(CH2CH2)S- C2B9H10}(PPh3)2][middle dot]Me2CO. Journal of the Chemical Society, Chemical Communications 1992, (18), 1281-1282; (e) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Stable Silver Complexes with C2B9H12- derivatives. Inorganic Chemistry 1994, 33 (9), 1756-1761; (f) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Modulation of the B(3)-H.fwdharw.Ru Distances in 7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1994, 13 (7), 2751-2760; (g) Teixidor, F.; Ayllòn, J. A.; Viñas, C.; Kivekäs, R.; Sillanpää, R.; Casabò, J., Mercury coordination to Exo-dithio-7,8-dicarba-nido-undecaborate derivatives. Journal of Organometallic Chemistry 1994, 483 (1), 153-157; (h) Teixidor, F.; Flores, M. A.; Viñas, C.; Kivekäs, R.; Sillanpää, R., Influence of S-Aryl Groups in the Coordination and Reactivity of (nido-Thiocarborane)ruthenium Complexes. Organometallics 1998, 17 (21), 4675-4679; (i) Teixidor, F.; Flores, M. A.; Viñas, C.; Sillanpää, R.; Kivekäs, R., exo-nido-Cyclooctadienerhodacarboranes: Synthesis, Reactivity, and Catalytic Properties in Alkene Hydrogenation. Journal of the American Chemical Society 2000, 122 (9), 1963-1973; (j) Teixidor, F.; Núñez, R.; Flores, M. A.; Demonceau, A.; Viñas, C., Forced exo-nido rhoda and ruthenacarboranes as catalyst precursors: a review. Journal of Organometallic Chemistry 2000, 614–615, 48-56; (k) Teixidor, F.; Vinas, C.; Mar Abad, M.; Lopez, M.; Casabo, J., Synthesis of [7,8-(PPh2)2-7,8-C2B9H10]-: a ligand analogous to 1,2-bis(diphenylphosphino)ethane with a "built-in" negative charge. Organometallics 1993, 12 (9), 3766-3768; (l) Viñas, C.; Nuñez, R.; Teixidor, F.; Kivekäs, R.; Sillanpää, R., Modulation of Agostic B−H⇀Ru Bonds in exo-Monophosphino-7,8-Dicarba-nido- undecaborate Derivatives. Organometallics 1996, 15 (18), 3850-3858.

8. Hoel, E. L.; Hawthorne, M. F., Preparation of B-.sigma.-carboranyl iridium complexes by oxidative addition of terminal boron-hydrogen bonds to iridium(I) species. Journal of the American Chemical Society 1975, 97 (22), 6388-6395.

9. (a) Douvris, C.; Ozerov, O. V., Hydrodefluorination of Perfluoroalkyl Groups Using Silylium-Carborane Catalysts. Science 2008, 321 (5893), 1188-1190; (b) Khandelwal, M.; Wehmschulte, R. J., Deoxygenative Reduction of Carbon Dioxide to Methane, Toluene, and Diphenylmethane with [Et2Al]+ as Catalyst. Angewandte Chemie International Edition 2012, 51 (29), 7323-7326; (c) Ramírez-Contreras, R.; Bhuvanesh, N.; Zhou, J.; Ozerov, O. V., Synthesis of a Silylium Zwitterion. Angewandte Chemie International Edition 2013, 52 (39), 10313-10315; (d) Wehmschulte, R. J.; Laali, K. K.; Borosky, G. L.; Powell, D. R., Synthesis and Structure of the First Bridgehead Silylium Ion. Organometallics 2014, 33 (9), 2146-2149; (e) Wehmschulte, R. J.; Saleh, M.; Powell, D. R., CO2 Activation with Bulky Neutral and Cationic Phenoxyalanes. Organometallics 2013, 32 (22), 6812-6819.

10. Lavallo, V.; Wright, J. H.; Tham, F. S.; Quinlivan, S., Perhalogenated Carba- closo-dodecaborate Anions as Ligand Substituents: Applications in Gold Catalysis. Angewandte Chemie International Edition 2013, 52 (11), 3172-3176.

11. El-Hellani, A.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Structure and Bonding of a Zwitterionic Iridium Complex Supported by a Phosphine with the Parent Carba-closo-dodecaborate CB11H11– Ligand Substituent. Organometallics 2013, 32 (23), 6887-6890.

12. (a) Himmelspach, A.; Finze, M.; Raub, S., Tetrahedral Gold(I) Clusters with Carba-closo-dodecaboranylethynido Ligands: [{12- -closo-1- CB11H11}2]. Angewandte Chemie International Edition 2011, 50 (11), 2628-2631; (b) El-Hellani, A.; Lavallo, V., Fusing N-Heterocyclic Carbenes with Carborane Anions. Angewandte Chemie International Edition 2014, 53 (17), 4489-4493; (c) Asay, M. J.; Fisher, S. P.; Lee, S. E.; Tham, F. S.; Borchardt, D.; Lavallo, V., Synthesis of unsymmetrical N-carboranyl NHCs: directing effect of the carborane anion. Chemical Communications 2015, 51 (25), 5359-5362.

13. Núñez, R.; Viñas, C.; Teixidor, F.; Sillanpää, R.; Kivekäs, R., Contribution of the o-carboranyl fragment to the chemical stability and the 31P-NMR chemical shift in closo- carboranylphosphines. Crystal structure of bis(1-yl-2-methyl-1,2-dicarba-closo- dodecaborane)phenylphosphine. Journal of Organometallic Chemistry 1999, 592 (1), 22- 28.

14. Fey, N.; Haddow, M. F.; Mistry, R.; Norman, N. C.; Orpen, A. G.; Reynolds, T. J.; Pringle, P. G., Regioselective B-Cyclometalation of a Bulky o-Carboranyl Phosphine and the Unexpected Formation of a Dirhodium(II) Complex. Organometallics 2012, 31 (7), 2907-2913.

15. Crabtree, R. H., The Organometallic Chemistry of the Transition Metals (fifth ed.), vol. 5, John wiley & sons Ltd., New Jersey. 2009.

Chapter 3: The Inductive Effects of 10 and 12-Vertex closo-Carborane Anion Ligand Substituents: Cluster Size and Charge Make a Difference

3.1 Abstract

- A phosphine containing a CB9H9 anion substituent and its subsequent ligation to a

Rh(I) carbonyl complex is reported. The complex is characterized by NMR spectroscopy and a single crystal X-ray diffraction study. In addition, the inductive effects of both 10 and 12 vertex C-functionalized closo-carborane anions are elucidated via I.R. analysis of the CO stretching frequencies of two Rh carbonyl complexes. Unlike C-functionalized neutral C2B10H10 the 10 and 12-vertex carborane anions are both strong electron donor substituents.

3.2 Introduction

An intriguing alternative to ubiquitous hydrocarbon ligand R-groups are the organomimetic1 closo-carboranes,2 which can be thought of as 3-dimensional analogues of

Hückel aromatics.2e The most common closo-carboranes utilized in ligand design are

2a 3 derived from neutral C2B10H12. Due to its facile synthesis, the H2C2B10H12 isomer 1 (Figure 3-1, top left) has been the most frequently utilized. When functionalized at a C-vertex 1 acts as a strong electron withdrawing group, more so than a benzene ring.4 Spokoyny,1, 5 has elegantly shown that functionalization at B-vertices renders such clusters strong donor substituents. Although 1 offers many distinct characteristics, such as a unique steric profile and the ability to form H–H hydrogen bonds,2a it exhibits reactivity that is perhaps undesirable for catalysis, such as facile B–H cyclometalation6 and cluster opening reactions7 to afford nido-carboranes. The latter reactions have been exploited by

Teixidor and Viñas to produce novel anionic ligands, featuring nido-cluster substituents.2b,

2d, 2f, 4b, 4c, 8 These nido-cluster substituents have also been shown to be strong donors when attached to ligands via the B-vertices.8h

- Figure 3-1. Representations of H2C2B10H12 1, the isoelectronic HCB11H11 anion 2, and the 10-vertex - CB9H10 anion 8. Unlabeled vertices = B–H.

- 2c The HCB11H11 anion 2 is isoelectronic with 1. As a result of the negative charge of 2 being delocalized over the 12 cage atoms, this carborane is rendered a weakly coordinating anion. This characteristic has led to the utilization of 2 and its derivatives as spectator anions to stabilize exotic cations.9 We recently reported the first examples of C- functionalized carborane anions 2 in ligand and transition metal-based catalyst design.10 It was demonstrated that 2 and its polyhalogenated variants are competent substituents for phosphine10b-e, 11 and N-heterocyclic carbene ligands,10c, 11 which produce unusual zwitterionic and anionic complexes. Importantly, it was shown that the introduction of a closo-carborane anion substituent can lead to dramatically improved catalytic performance compared to systems containing traditional ligands.12 In addition, we have shown that as a ligand substituent 2 is intrinsically more resistant to B–H cyclometalation compared to 1.10d

- Another closo-carborane anion is the ten vertex species HCB9H9 8, which has a slightly less delocalized charge. In contrast to 2, cluster 8 and its derivatives have rarely13 been exploited as weakly coordinating anions. Recently, improved synthetic protocols have made 8 readily accessible on a reasonable scale.14 We became interested in the possibility of utilizing 8 as a ligand substituent,15 which could complement our investigations into 2 by providing a smaller group with distinct electronic properties. Here, we report the first example of the utilization of cluster 8 as a ligand substituent. In addition, a comparative study on the inductive effects of closo-carborane anions 2 and 8 reveals that in contrast to C-functionalized C2B10H10 1 these anionic clusters are strong electron donors.

3.3 Results and Discussion

To begin this investigation we chose to target a simple phosphine 9[Li+] (Figure 3-

2) with a 10-vertex carborane anion substituent 8. Similar to 1 and 2, 8 contains a mildly acidic C–H vertex that can be deprotonated with strong bases and subsequently

+ functionalized with electrophiles. Thus, reaction of 8[HNMe3 ] with two equivalents of n-

BuLi followed by the addition of ClP(iPr)2 results in rapid formation of the desired ligand 9[Li+] as evidenced by multinuclear NMR spectroscopy (Figure 3-2). The 31P and 11B NMR spectra of the anionic carboranyl phosphine 9[Li+] display a single peak at

22.0 ppm and a set of three broad singlets (11B{1H}: 32.6, −16.4, −23.4 ppm; 1 : 4 : 4 ratio), respectively. Similar to the known carboranyl phosphine containing the 12-vertex cluster 2, which is air sensitive when dissolved in organic solvents,16 THF solutions of ligand 9[Li+] are oxidized to the corresponding phosphine oxide in two hours.

Figure 3-2. Synthesis of 9[Li+] and complex 10[Li+]. Solid-state structure of 10[Li+]. Notable bond lengths (Å): Rh–P1 = 2.3534(3), Rh–P2 = 2.3554(3), Rh–C3 = 1.9024(11), Rh–C4 = 1.9166(12), C4–O2 = 1.1321(15), C3–O1 = 1.1354(14), H1–Li1 = 1.997. Selected bond angles (°): C3–Rh–C4 = 154.4, P1–Rh– P2 = 171.9. Color code: C = grey, B = brown, Rh = blue, Li = green, O = red. Unlabeled vertices = B–H. To gain insight into the inductive effects of anionic carborane substituent 8, we targeted the carbonyl complex 10[Li+] for I.R. analysis. Although anionic because of the presence of two charged phosphine ligands, 10[Li+] has an iso-electronic coordination

+1 environment with cationic species RhL2(CO)2 for which there is a significant amount of

I.R. data available for comparison. Thus, reaction of 9[Li+] with a mixture of

(ClRh(CO)2)2 and AgBF4 dissolved in CH2Cl2 results in rapid formation of the anionic diphosphine complex 10[Li+]. The 31P and 11B NMR spectra of 10[Li+] display a doublet at 56.1 ppm (31P{1H}; d, 1J(Rh–P) = 102.4 Hz) and a set of three broad singlets (11B{1H}:

35.9, −15.7, −23.5 ppm; 1 : 4 : 4 ratio), respectively. The 11B proton coupled spectrum shows three doublet resonances, which indicates the presence of hydrides at each cluster vertex and rules out the possibility of cyclometalation at Rh. A single crystal X-ray diffraction study confirms the identity of 10[Li+], which displays a slightly distorted square planar geometry (sum of the L–M–L angles = 363.6°). The cone angle of ligand 9[Li+], as determined from the crystallographic data, is 168°. Notably, the Li+ counter cation in 10[Li+] is coordinated to the most electron rich B–H vertex that is antipodal to carbon.

Two THF molecules and an additional B–H interaction from a different molecule in the unit cell complete the tetrahedral coordination environment at Li+. The solution I.R. spectrum of 10[Li+] shows a prominent absorbance for the B–H architecture (I.R. = 2553

−1 −1 cm ; CH2Cl2) as well as an intense stretch for the carbonyl ligands (I.R. = 1997 cm ;

+1 CH2Cl2). For comparison, the isoelectronic complexes Rh(P(iPr)3)2(CO)2 and

+1 −1 Rh(P(Ph)3)2(CO)2 display CO stretching frequencies at 2010 cm and 2047

−1 17 cm (CH2Cl2), respectively. Therefore, in contrast to C-functionalized C2B10H10 1, 8 is a potent electron donor substituent, more so than an isopropyl group.

For further comparison we next sought to elucidate the inductive effects of the larger 12-vertex carborane anion 2. We hypothesized that because the negative charge in 2 is more delocalized it should be a weaker donor than 8 but still a stronger donor than

+ neutral C-functionalized C2B10H10 1. Reaction of phosphine 11[Li ], which is isostructural

+ with 9[Li ], with a mixture of (ClRh(CO)2)2 and AgBF4 dissolved in CH2Cl2 results in the formation of the anionic diphosphine complex 12[Li+] as determined by multinuclear

NMR spectroscopy (Fig. 3-3). However, 12[Li+] rapidly extrudes a CO ligand to afford the

mono-CO complex 8[Li+]. The difference in stability between 12[Li+] and isoelectronic 10[Li+] can perhaps be explained by the greater steric bulk of the 12-vertex carborane anion as well as electronic effects, vide infra. The structure of 8[Li+] was confirmed by NMR spectroscopy as well as a single crystal X-ray diffraction study. In the solid-state 8[Li+] displays a square planar geometry with a B–H “agostic- like”10b interaction with the coordination site where CO was liberated. The cone angle for phosphine ligand 11[Li+] is slightly larger (171°) compared to the 10-vertex ligand 9[Li+], vide supra. Interestingly, in contrast to 10[Li+] the Li+ cation in 8[Li+] is sequestered by four THF molecules and shows no interaction with the 12-vertex carborane anion. This feature highlights the reduced coordinative ability, due to enhanced charge

+ delocalization, of 2 compared to 3. The I.R. spectrum of a CH2Cl2 solution of 8[Li ] shows a single CO stretch at 1991 cm−1.

Figure 3-3. Synthesis of complexes 12[Li+] and 8[Li+]. Solid-state structure of 8[Li+]. Notable bond lengths (Å): Rh–P1 = 2.3017(4), Rh–P2 = 2.3686(4), Rh–C1 = 1.8113(19), C1–O1 = 1.145(2), H1–Rh = 1.878. Selected bond angles (°): H1–Rh–C1 = 176.3, P1–Rh–P2 = 175.2. Color code: C = grey, B = brown, Rh = blue, O = red. Unlabeled vertices = B–H. Seeking to generate 12[Li+] in pure form and obtain I.R. data for an isoelectronic comparison we charged a J-young tube, containing a solution of 8[Li+], with one atmosphere of CO. Gratifyingly, under these conditions 8[Li+] is completely converted back to 12[Li+], which is stable under a CO atmosphere. Analysis of the I.R. spectrum

+ −1 of 12[Li ] dissolved in CH2Cl2 reveals a strong CO stretch at 2012 cm . The increased

CO stretching frequency relative to 8[Li+], is consistent with the formation of 12[Li+], since both of the trans-CO ligands are engaged in competitive π-backbonding with the same Rh d-orbital. This data also suggests that the lability of CO in 7[Li+] compared to 10[Li+] is not only the result of increased steric pressure, but also perhaps a result of

reduced π-backbonding from the less electron rich Rh-center of 12[Li+]. From this analysis we can conclude that the 12-vertex carborane anion substituent 2 is an electron donor, nearly as strong as an isopropyl group, but not as potent as 8. The observed electronic differences between 2 and 8 mirror classical Hammet studies by Zakharkin,18 which showed that isoelectronic 10-vertex neutral dicarbaboranes are less electron withdrawing than their 12-vertex analogues.

3.4 Summary and Conclusion

The manuscript above demonstrates for the first time that 10-vertex closo- carborane anion 8 is a viable ligand substituent with distinct electronic and steric properties. In addition, the inductive effects of both the parent hydrido 10 and 12-vertex C- functionalized closo-carborane anions 8 and 2 have been elucidated. Unlike C- functionalized neutral C2B10H10 1, both 2 and 8 are strong electron donor substituents. The fact that C-functionalized 2 is a much stronger donor than 1 highlights the effect that cluster charge has on the inductive effects of 12-vertex closo-carboranes. Likewise, the increase in donor ability of 8 relative to 2 shows that implementing smaller clusters with less delocalized charge dramatically influences the inductive effects of carborane ligand substituents. These results are not only fundamentally important, but should be instructive for future investigations into ligand design for implementation in homogeneous catalysis.

3.5 Experimental

The previous experimental data has been published and may be obtained online at DOI: 10.1039/C5CC08377J

3.6 References

1. Spokoyny Alexander, M., New ligand platforms featuring boron-rich clusters as organomimetic substituents. In Pure and Applied Chemistry, 2013; Vol. 85, p 903.

2. (a) Scholz, M.; Hey-Hawkins, E., Carbaboranes as Pharmacophores: Properties, Synthesis, and Application Strategies. Chemical Reviews 2011, 111 (11), 7035-7062; (b) Farras, P.; Juarez-Perez, E. J.; Lepsik, M.; Luque, R.; Nunez, R.; Teixidor, F., Metallacarboranes and their interactions: theoretical insights and their applicability. Chemical Society Reviews 2012, 41 (9), 3445-3463; (c) Douvris, C.; Michl, J., Update 1 of: Chemistry of the Carba-closo-dodecaborate(−) Anion, CB11H12–. Chemical Reviews 2013, 113 (10), PR179-PR233; (d) Olid, D.; Nunez, R.; Vinas, C.; Teixidor, F., Methods to produce B-C, B-P, B-N and B-S bonds in boron clusters. Chemical Society Reviews 2013, 42 (8), 3318-3336; (e) Poater, J.; Solà, M.; Viñas, C.; Teixidor, F., π Aromaticity and Three-Dimensional Aromaticity: Two sides of the Same Coin? Angewandte Chemie International Edition 2014, 53 (45), 12191-12195; (f) Popescu, A. R.; Teixidor, F.; Viñas, C., Metal promoted charge and hapticities of phosphines: The uniqueness of carboranylphosphines. Coordination Chemistry Reviews 2014, 269, 54-84; (g) Zhang, J.; Xie, Z., Synthesis, Structure, and Reactivity of 13- and 14-Vertex Carboranes. Accounts of Chemical Research 2014, 47 (5), 1623-1633.

3. Li, Y.; Carroll, P. J.; Sneddon, L. G., Ionic-Liquid-Promoted Decaborane Dehydrogenative Alkyne-Insertion Reactions: A New Route to o-Carboranes. Inorganic Chemistry 2008, 47 (20), 9193-9202.

4. (a) Röhrscheid, F.; Holm, R. H., Zero-valent nickel complexes of bis(phosphino)- o-carboranes. Journal of Organometallic Chemistry 1965, 4 (4), 335-338; (b) Teixidor, F.; Núñez, R.; Viñas, C.; Sillanpää, R.; Kivekäs, R., The Distinct Effect of the o- Carboranyl Fragment: Its Influence on the I−I Distance in R3PI2 Complexes. Angewandte Chemie International Edition 2000, 39 (23), 4290-4292; (c) Núñez, R.; Farràs, P.; Teixidor, F.; Viñas, C.; Sillanpää, R.; Kivekäs, R., A Discrete P⋅⋅⋅ ⋅⋅⋅P Assembly: The Large Influence of Weak Interactions on the 31P NMR Spectra of Phosphane–Diiodine Complexes. Angewandte Chemie International Edition 2006, 45 (8), 1270-1272.

5. (a) Spokoyny, A. M.; Machan, C. W.; Clingerman, D. J.; Rosen, M. S.; Wiester, M. J.; Kennedy, R. D.; Stern, C. L.; Sarjeant, A. A.; Mirkin, C. A., A coordination chemistry dichotomy for icosahedral carborane-based ligands. Nat Chem 2011, 3 (8), 590-596; (b) Spokoyny, A. M.; Lewis, C. D.; Teverovskiy, G.; Buchwald, S. L., Extremely Electron-Rich, Boron-Functionalized, Icosahedral Carborane-Based Phosphinoboranes. Organometallics 2012, 31 (24), 8478-8481.

6. Hoel, E. L.; Hawthorne, M. F., Preparation of B-.sigma.-carboranyl iridium complexes by oxidative addition of terminal boron-hydrogen bonds to iridium(I) species. Journal of the American Chemical Society 1975, 97 (22), 6388-6395.

7. Wiesboeck, R. A.; Hawthorne, M. F., Dicarbaundecaborane(13) and Derivatives. Journal of the American Chemical Society 1964, 86 (8), 1642-1643.

8. (a) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Stable Silver Complexes with C2B9H12- derivatives. Inorganic Chemistry 1994, 33 (9), 1756-1761; (b) Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., Modulation of the B(3)-H.fwdharw.Ru Distances in 7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1994, 13 (7), 2751-2760; (c) Viñas, C.; Nuñez, R.; Teixidor, F.; Kivekäs, R.; Sillanpää, R., Modulation of Agostic B−H⇀Ru Bonds in exo- Monophosphino-7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1996, 15 (18), 3850-3858; (d) Teixidor, F.; Flores, M. A.; Viñas, C.; Kivekäs, R.; Sillanpää, R., Influence of S-Aryl Groups in the Coordination and Reactivity of (nido- Thiocarborane)ruthenium Complexes. Organometallics 1998, 17 (21), 4675-4679; (e) Núñez, R.; Viñas, C.; Teixidor, F.; Sillanpää, R.; Kivekäs, R., Contribution of the o- carboranyl fragment to the chemical stability and the 31P-NMR chemical shift in closo- carboranylphosphines. Crystal structure of bis(1-yl-2-methyl-1,2-dicarba-closo- dodecaborane)phenylphosphine. Journal of Organometallic Chemistry 1999, 592 (1), 22- 28; (f) Teixidor, F.; Flores, M. A.; Viñas, C.; Sillanpää, R.; Kivekäs, R., exo-nido- Cyclooctadienerhodacarboranes: Synthesis, Reactivity, and Catalytic Properties in Alkene Hydrogenation. Journal of the American Chemical Society 2000, 122 (9), 1963- 1973; (g) Teixidor, F.; Núñez, R.; Flores, M. A.; Demonceau, A.; Viñas, C., Forced exo- nido rhoda and ruthenacarboranes as catalyst precursors: a review. Journal of Organometallic Chemistry 2000, 614–615, 48-56; (h) Teixidor, F.; Barberà, G.; Vaca, A.; Kivekäs, R.; Sillanpää, R.; Oliva, J.; Viñas, C., Are Methyl Groups Electron-Donating or Electron-Withdrawing in Boron Clusters? Permethylation of o-Carborane. Journal of the American Chemical Society 2005, 127 (29), 10158-10159; (i) Farràs, P.; Olid-Britos, D.; Viñas, C.; Teixidor, F., Unprecedented B–H Activation Through Pd-Catalysed B–Cvinyl Bond Coupling on Borane Systems. European Journal of Inorganic Chemistry 2011, 2011 (16), 2525-2532; (j) Farràs, P.; Teixidor, F.; Rojo, I.; Kivekäs, R.; Sillanpää, R.; González-Cardoso, P.; Viñas, C., Relaxed but Highly Compact Diansa Metallacyclophanes. Journal of the American Chemical Society 2011, 133 (41), 16537- 16552; (k) Popescu, A.-R.; Laromaine, A.; Teixidor, F.; Sillanpää, R.; Kivekäs, R.; Llambias, J. I.; Viñas, C., Uncommon Coordination Behaviour of P(S) and P(Se) Units when Bonded to Carboranyl Clusters: Experimental and Computational Studies on the Oxidation of Carboranyl Phosphine Ligands. Chemistry – A European Journal 2011, 17 (16), 4429-4443; (l) Brusselle, D.; Bauduin, P.; Girard, L.; Zaulet, A.; Viñas, C.; Teixidor, F.; Ly, I.; Diat, O., Lyotropic Lamellar Phase Formed from Monolayered θ- Shaped Carborane-Cage Amphiphiles. Angewandte Chemie International Edition 2013, 52 (46), 12114-12118.

9. Reed, C. A., H+, CH3+, and R3Si+ Carborane Reagents: When Triflates Fail. Accounts of Chemical Research 2010, 43 (1), 121-128.

10. (a) Himmelspach, A.; Finze, M.; Raub, S., Tetrahedral Gold(I) Clusters with Carba-closo-dodecaboranylethynido Ligands: [{12- -closo-1- CB11H11}2]. Angewandte Chemie International Edition 2011, 50 (11), 2628-2631; (b) El-Hellani, A.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Structure and Bonding of a Zwitterionic Iridium Complex Supported by a Phosphine with the Parent Carba-closo-dodecaborate CB11H11– Ligand Substituent. Organometallics 2013, 32 (23), 6887-6890; (c) El-Hellani, A.; Lavallo, V., Fusing N-Heterocyclic Carbenes with Carborane Anions. Angewandte Chemie International Edition 2014, 53 (17), 4489-4493; (d) Estrada, J.; Lee, S. E.; McArthur, S. G.; El-Hellani, A.; Tham, F. S.; Lavallo, V., Resisting B–H oxidative addition: The divergent reactivity of the o-carborane and carba- closo-dodecaborate ligand substituents. Journal of Organometallic Chemistry 2015, 798, Part 1, 214-217; (e) Estrada, J.; Woen, D. H.; Tham, F. S.; Miyake, G. M.; Lavallo, V., Synthesis and Reactivity of a Zwitterionic Palladium Allyl Complex Supported by a Perchlorinated Carboranyl Phosphine. Inorganic Chemistry 2015, 54 (11), 5142-5144.

11. Asay, M. J.; Fisher, S. P.; Lee, S. E.; Tham, F. S.; Borchardt, D.; Lavallo, V., Synthesis of unsymmetrical N-carboranyl NHCs: directing effect of the carborane anion. Chemical Communications 2015, 51 (25), 5359-5362.

12. Lavallo, V.; Wright, J. H.; Tham, F. S.; Quinlivan, S., Perhalogenated Carba- closo-dodecaborate Anions as Ligand Substituents: Applications in Gold Catalysis. Angewandte Chemie International Edition 2013, 52 (11), 3172-3176.

13. (a) Xie, Z.; Liston, D. J.; Jelinek, T.; Mitro, V.; Bau, R.; Reed, C. A., A new weakly coordinating anion: approaching the silylium (silicenium) ion. Journal of the Chemical Society, Chemical Communications 1993, (4), 384-386; (b) Xie, Z.; Jelinek, T.; Bau, R.; Reed, C. A., New Weakly Coordinating Anions. III. Useful Silver and Trityl Salt Reagents of Carborane Anions. Journal of the American Chemical Society 1994, 116 (5), 1907-1913.

14. Bryan Ringstrand, D. B., Richard K. Shoemaker, Zbynek Janousek, Improved synthesis of closo-1-CB9H10- anion and new C-substituted derivatives. Collection of Czechoslovak Chemical Communications 2009, 74 (3), 419-431.

15. Finze, M.; Sprenger, J. A. P., Anionic Gold(I) Complexes—Twelve- and Ten- Vertex Monocarba-closo-borate Anions with Carbon–Gold σ Bonds. Chemistry – A European Journal 2009, 15 (38), 9918-9927.

16. Drisch, M.; Sprenger, J. A. P.; Finze, M., Carba-closo-dodecaborate Anions with Cluster Carbon–Phosphorous Bonds. Zeitschrift für anorganische und allgemeine Chemie 2013, 639 (7), 1134-1139.

17. (a) Nürnberg, O.; Werner, H., Vinyliden-Übergangsmetallkomplexe XXVI. Synthese und struktur kationischer Carbonyl-, Alken-, Vinyliden-, Alkinyl- und Alkin- Rhodiumkomplexe mit der Baueinheit trans-[Rh(PiPr3)2]. Journal of Organometallic Chemistry 1993, 460 (2), 163-175; (b) Evans, E. W.; Howlader, M. B. H.; Atlay, M. T., Synthesis, stability and reactivity of cationic carbonyl complexes of rhodium(I). Transition Metal Chemistry 1994, 19 (1), 37-40.

18. (a) Zakharkin, L. I.; Kalinin, V. N.; Snyakin, A. P.; Kvasov, B. A., Effect of solvents on the electronic properties of 1-o-, 3-o- and 1-m-carboranyl groups. Journal of Organometallic Chemistry 1969, 18 (1), 19-26; (b) L. I. Zakharkin, V. N. K., e. G. Rys, B.A. Kvasov, Izv. Akad. Nauk SSSR 1972, 507.

Chapter 4: Changing the Charge: Electrostatic Effects in Pd Catalyzed Cross-

Coupling

4.1 Abstract

A stable dianionic 14-electron Pd(0) complex supported by two monoanionic

- carboranyl phosphines (iPr2P-CB11H11 ) is reported. This complex rapidly undergoes the oxidative addition of Cl-C6H5 at room temperature and is a competent catalyst for Kumada cross-coupling. The isosteric PdL2 complex, supported by neutral o-carboranyl phosphines

(iPr2P-C2B10H11), does not display the same reactivity. The high reactivity of the dianionic

Pd(0) complex toward chloroarenes can be explained by electrostatic effects that promote both formation of monophosphine-ligated LPd0 and stabilization of the transition state during oxidative addition. This mode of stabilization is distinct from the well-known π- arene interactions of biaryl phosphines in that it occurs both on and off cycle.

4.2 Introduction

Pd-catalyzed cross-coupling is an indispensable method in the chemist’s tool box.1 The most common ligands utilized in cross-coupling protocols are neutral phosphines appended with alkyl or aryl groups. To achieve high activity, particularly with challenging substrates such as aryl chlorides, bulky electron-rich phosphines are implemented. With respect to electronics, such ligands render the metal more electron rich, promoting oxidative addition. The steric influence of the ligand promotes the formation of monophosphine-ligated species that are required to effect efficient C–X oxidative addition.

Among the state-of-the-art phosphine ligands,1c, 2 biaryl phosphines,1d, 1e, 3 such as S-phos, are often the go-to ligands for the cross-coupling of aryl chlorides and other difficult

substrates. It is believed that the high activity of these systems arises from metal π-arene interactions, which stabilize the low-coordinate resting state of the catalysts.4

Over the past few years our lab has developed a variety of phosphine5 and N-heterocyclic carbene6 ligands, such as 1 (Figure 4-1),

- featuring weakly coordinating CB11H11 substituents.7 Attaching these primarily inorganic clusters bestows L-type ligands with an anionic charge. The Figure 4-1. Synthesis of isosteric/electronic dianionic and neutral Pd(0) complexes 2 and 4, respectively. Legend: charge is delocalized throughout the (i) (TMS-CH2)2Pd(COD) 0.5 equiv, C6D6, room temperature, 5 min. Unlabeled vertices denote B–H closed polyhedral structure, which disposes the charge close to the metal center regardless of conformation. This fact, coupled

- 7b, 7c, 8 with the chemical stability of CB11H11 , makes our systems distinct from ligands appended with tetracoordinate borate moieties.9 We are particularly interested in utilizing the weak coordinative abilities and charge of the cluster to induce electrostatic effects in the coordination sphere of metals. In gold catalysis, we have demonstrated that it is possible to create highly active and well-defined single component systems for C–N bond forming reactions.5b Herein, we report our initial findings relevant to Pd-catalyzed cross-coupling chemistry by preparing the stable dianionic 14-electron Pd(0) species 2 that readily reacts with chloroarenes at room temperature. Implementing the isosteric ligand 3, featuring a neutral C2B10H10 ligand substituent, yields an isoelectronic, but neutral, complex 4 that is

not competent for the activation of chloroarenes or catalysis. We provide both experimental and computational evidence that the high reactivity of 2 is the result of electrostatic effects that facilitate both phosphine dissociation and stabilization of the transition state during the oxidative addition process. This mode of stabilization is distinct from the well-known π- arene interactions of biaryl phosphines,4 in that it occurs both on and off cycle.

4.3 Results and Discussion

We have recently determined the σ-inductive effects of both 10- and 12-

5e 7g, vertex closo-carborane anions. In contrast to neutral C-functionalized C2B10H10,

10 these anionic clusters are strong donors similar to alkyl groups. Ligand 1 is just slightly less donating than P(i-Pr)3 but significantly more bulky (cone angles: 1 (171°); P(i-

Pr)3 (160°)). Given our understanding of the donor ability of 1 and its steric parameters, we chose to utilize it to investigate the possibility of preparing the dianionic two-coordinate

Pd(0) species 2 for reactivity studies (Figure 4-1). At the same time, we wished to investigate the sterically identical, but neutral ligand 3, featuring a C2B10H11 group, and its

PdL2 complex 4 (Figure 4-1). Analogous neutral derivatives supported by classical phosphines are convenient for solution-based reactivity and mechanistic studies because they contain no additional ligands, which might complicate interpretation of the spectroscopic data. Thus, ligands 1 and 3 were reacted independently with 1/2 an equivalent of (TMS-CH2)2Pd(COD) in C6D6 (Figure 4-1). Analysis of the reaction mixtures by 31P NMR spectroscopy showed the rapid consumption of the ligands (1, 45.0 ppm; 3, 54.2 ppm) and the clean formation of new products (2, 65.5 ppm; 4, 81.1 ppm) containing symmetrical phosphines. 1H NMR spectroscopy showed resonances for the new

phosphine complexes as well as the expected signature for 1,2-bis(trimethylsilyl)ethane formed via reductive elimination. For the dianionic complex 2 the 11B{1H} NMR spectrum shows three resonances (1:5:5 ratio), demonstrating that the cluster is intact and retains its

11 1 local C5v symmetry. The B{ H} NMR spectrum of the neutral complex 4 is more complicated due to the lower symmetry of C2B10H10. Complex 2 can be isolated in pure form by precipitation of the mixture in hexane. Complex 4 can be isolated by recrystallization from acetonitrile. Interestingly, the dianionic complex 2 is soluble and stable in degassed D2O (Figure S2 in the Supporting Information), while the neutral complex 4 is insoluble.

The structures of 2 and 4 were unambiguously determined by single-crystal X-ray diffraction studies (Figure 4-2). In the solid state both complexes are dicoordinate and essentially linear (P–Pd–P angles: 2, 172.1°; 4, 170.8°). Aside from the two spectator lithium countercations tetrahedrally coordinated by four

THF molecules, the Pd fragment in 2 has very similar geometric parameters in

comparison Figure 4-2. Solid-state structures of 2 and 4. Selected bond lengths in Å: 2, to 4 (Figure 4-2). With Pd1a–P1a = 2.2615(10), Pd1a–P2a = 2.2688(10), P1a–C1a = 1.875(4), P2a– C2a = 1.872(3); 4, Pd1b–P1b = 2.2661(17), Pd1b–P2b = 2.2774(17), P1b– these two charge C1b = 1.869(7); P2b–C3b = 1.880(7). Hydrogens and countercations of 2 are omitted for clarity. Color code: B, brown; C, gray; Pd, blue; P, violet.

differentiated but isosteric/electronic complexes in hand, we sought to probe their reactivity with Cl-C6H5. Hence, we independently dissolved complexes 2 and 4 in neat Cl-

31 C6H5 and monitored the reactions by P NMR spectroscopy. Within 9 min complex 2 was completely consumed, to afford two new products in a ratio of 9:1 (Figure 4-3A). The primary product, which displays a singlet resonance, was assigned as the monoanionic bisphosphine Pd(II) aryl complex 5. The structure of isolated 5 was corroborated by multinuclear NMR spectroscopy as well as a single-crystal X-ray diffraction study (Figure

S41 in the Supporting Information). The formation of 5 can be explained by an oxidative addition of Cl-C6H5 with subsequent elimination of LiCl (recall complex 2 has two

Li+ countercations). The minor product 6 displays two coupled doublets (50.2 and 8.9 ppm and 323.5 Hz) in the 31P NMR spectrum, indicating the formation of an unsymmetrical bisphosphine complex. Complex 6 was isolated and determined by multinuclear NMR spectroscopy as well as a single-crystal X-ray diffraction study to be the B-cyclometalated

Pd(II) complex 6 (Figure S42 in the Supporting Information). The formation of 6 can be rationalized by an oxidative addition/σ-bond metathesis sequence which results in the elimination of benzene and LiCl (vide infra). In contrast complex 4 reacted very sluggishly, with approximately 90% of complex 4 intact after 24 h at room temperature. Aside from the starting material 4 an intractable mixture of numerous 31P-containing complexes was detected (Figure S13 in the Supporting Information). Even after 4 was refluxed in neat Cl-

C6H5 for 24 h a significant amount of starting material remained (Figure S14 in the

Supporting Information). The fast oxidative addition of 2 with Cl-C6H5 at room temperature is remarkable. Comparable dicoordinate PdL2 systems featuring standard

11 phosphines such as P(Cy)3 and P(t-Bu)3, which we independently prepared and subjected to identical reaction conditions, react sluggishly (Pd[P(Cy)3]2 (8): 70% conversion/24 h) or not at all (Pd[(P(t-Bu)3]2 (9): no reaction) (Figures S25 and S27 in the Supporting

Information). That being said, state of the art biaryl-phosphines such as S-phos can form

4 transient L2Pd complexes in solution, which show comparable rates of oxidative addition.12

Figure 4-3. (A) Rapid reaction of complex 2 with Cl-Ph at room temperature to afford a 9:1 ratio of species 5 and 6, respectively. (B) Lability of the phosphines bound to 2, as demonstrated by the stoichiometric addition of P(Cy)3 to afford a distribution of mono- and disubstitution products (B).

5e Considering that ligand 1 is less electron donating than P(i-Pr)3 and more electron rich ligands such as P(Cy)3 and P(t-Bu)3 do not promote the rapid oxidative addition of Cl-

C6H5 at ambient temperature, it is clear that the observed reactivity cannot be explained by the electron donor ability of 1. Likewise, the fast oxidative addition cannot be explained by steric arguments, since the isosteric complex 4, featuring the neutral C2B10H10, does not behave the same as complex 2. Therefore, we propose that the rapid oxidative addition is the result of electrostatic effects. Perhaps, the binding of two ligands that bear a pendant negative charge to the same transition-metal center favors ligand dissociation, on the basis of a Coulombic argument. If 2 behaves similarly to standard L2Pd systems, this effect would favor the formation of the required L-Pd(solvent) complexes. Indeed, the addition of excess phosphine 1 to solutions of 2 strongly retards the rate of oxidative addition, which is in line with a dissociative pathway prior to oxidative addition. To further probe the lability of phosphine 1, we treated a solution of 2 with 1 equiv of P(Cy)3. Indeed, ligand 1 is quite labile, as we observe a distribution of 1, 2, monosubstituted product 7, and the double ligand substitution product Pd(P(Cy)3)2 (8) (Figure 4-3B).

To probe the possibility that the fast oxidative addition was the result of interactions

+ between the Li cations and the Cl-C6H5 substrate, we examined the behavior of system 2 with Na+ and K+ countercations. These complexes were formed in situ by

+ + implementing Na and K phosphine salts of 1 and (TMS-CH2)2Pd(COD) (page S23 in the

Supporting Information). The presence of Na+ and K+ does not change the rate of oxidative addition, suggesting that the countercations do not participate in the ligand dissociation or oxidative addition steps of the reaction sequence.

In order to gain support for our hypothesis, we investigated the ligand substitution and oxidative addition process computationally (page S40 in the Supporting Information).

As depicted in Figure 4-4, complex 2 can undergo a dissociative phosphine substitution with Cl-C6H5 to afford intermediate II. The latter undergoes a low barrier (TSII–III: ΔH⧧ =

7.4 kcal/mol) oxidative addition to afford the monoanionic Pd(II) intermediate III, which is in line with our experimental conditions. Interestingly, in intermediate II, TSII–III, and III there is a weak interaction between the cluster and the Pd center. This interaction stabilizes both the intermediates and transition state during the oxidative addition process.

An atoms in molecules (QTAIM) analysis of both intermediates as well as the transition state confirms that this interaction is essentially purely electrostatic (Page S48 in the

Supporting Information). Such stabilizing interactions are unique and distinct from the π- arene interactions of biaryl phosphines, which occur only during the catalyst resting state.

For comparison, the isoelectronic and neutral o-carborane supported Pd(0) complex 4 displays an endothermic (12.2 kcal/mol) ligand substitution with Cl-C6H5 (Page

S46 in the Supporting Information). In addition, the activation barrier for the oxidative addition is +10.8 kcal/mol, with the sum of the two processes being +23.1 kcal/mol uphill.

The latter observation serves as a direct explanation of the absence of such reactivity experimentally.

Figure 4-4. Plausible reaction profiles for the formation of the complexes 5 (black) and 6 (blue). The values in parentheses correspond to the relative Gibbs free energies. Continuing along the reaction pathway, intermediate III has two options.

(1) III can isomerize to III′ followed by reassociation of the liberated anionic carboranyl phosphine and simultaneous extrusion of Cl– to afford complex 5 (Figure 4-4). These processes are barrierless and exothermic (sum of isomerization and ligand reassociation:

ΔH⧧ = −6.2 kcal/mol). (2) Alternatively intermediate III can cyclometalate (TSIII–IV:

ΔH⧧ = 9.0 kcal/mol) via a σ-bond metathesis pathway to afford intermediate IV (ΔH⧧ =

−10.6 kcal/mol), which subsequently binds a liberated phosphine ligand exothermically

(ΔH⧧ = −13.6 kcal/mol) to afford the thermodynamic product 6. These calculated competing reaction pathways are in agreement with the observed formation of 5 and 6 as the major and minor products, respectively. Moreover, adding excess phosphine to the reaction mixture not only slows down the rate of oxidative addition but also retards cyclometalation, which is in complete agreement with what one would expect from the calculations above.

In order to show that complex 2 is a competent catalyst for Pd catalyzed cross- coupling, we examined its efficiency for several simple Kumada cross-couplings (Table 4-

1). No catalysis was observed at room temperature; however, heating the mixtures at 65 °C for 15–24 h resulted in moderate to excellent yields. Since we know that the oxidative addition of chlorobenzene to 2 occurs within minutes at room temperature, the requirement of heat suggests that the slow step in the catalytic cycle is either transmetalation or reductive elimination. Both small (entries A–D) and sterically encumbered (entries E–H)

Grignard reagents are effective coupling partners. With respect to electronic effects, the activity of 2 follows the typical trend observed in Pd catalyzed cross-coupling, with unactivated or deactivated aryl chlorides being more efficient substrates.

Table 4-1. Kumada Coupling of Simple Aryl Chlorides

a10 mol % of 2. bThe catalyst 2 was formed in situ by premixing ligand 1 (2 equiv) and (TMS-CH2)2Pd(COD) (1 equiv). Legend: (i) 5 mol % of 2, 65 °C, THF (1 mL). Yields without brackets are calculated by integration of the 1H NMR spectra with durene as an internal standard. Yields in brackets are isolated yields.

4.4 Summary and Conclusion This study introduces a new paradigm in Pd cross-coupling catalyst design. We have demonstrated that ligands appended with isoelectronic, but charge- differentiated closo-carboranes can induce dramatically different reactivities at metal centers. The enhanced activity of complex 2 in comparison to 4 can be explained by electrostatic effects that facilitate both ligand dissociation and oxidative addition.

Importantly, the stabilizing electrostatic interactions between the carborane cage and the

metal center occur both on and off cycle, which is distinct from systems based on neutral biarylphosphines. We are currently investigating the transmetalation and reductive elimination steps in this unique catalytic system.

4.5 Experimental

The previous experimental data has been published and may be obtained online at DOI: 10.1021/acs.organomet.6b00622

4.6 References

1. (a) Amatore, C.; Jutand, A., Anionic Pd(0) and Pd(II) Intermediates in Palladium- Catalyzed Heck and Cross-Coupling Reactions. Accounts of Chemical Research 2000, 33 (5), 314-321; (b) Goossen, L. J.; Koley, D.; Hermann, H. L.; Thiel, W., Mechanistic Pathways for Oxidative Addition of Aryl Halides to Palladium(0) Complexes: A DFT Study. Organometallics 2005, 24 (10), 2398-2410; (c) Hartwig, J. F., Evolution of a Fourth Generation Catalyst for the Amination and Thioetherification of Aryl Halides. Accounts of Chemical Research 2008, 41 (11), 1534-1544; (d) Martin, R.; Buchwald, S. L., Palladium-Catalyzed Suzuki−Miyaura Cross-Coupling Reactions Employing Dialkylbiaryl Phosphine Ligands. Accounts of Chemical Research 2008, 41 (11), 1461- 1473; (e) Surry, D. S.; Buchwald, S. L., Dialkylbiaryl phosphines in Pd-catalyzed amination: a user's guide. Chemical Science 2011, 2 (1), 27-50; (f) García-Melchor, M.; Braga, A. A. C.; Lledós, A.; Ujaque, G.; Maseras, F., Computational Perspective on Pd- Catalyzed C–C Cross-Coupling Reaction Mechanisms. Accounts of Chemical Research 2013, 46 (11), 2626-2634.

2. Chen, L.; Ren, P.; Carrow, B. P., Tri(1-adamantyl)phosphine: Expanding the Boundary of Electron-Releasing Character Available to Organophosphorus Compounds. Journal of the American Chemical Society 2016, 138 (20), 6392-6395.

3. Strieter, E. R.; Buchwald, S. L., Evidence for the Formation and Structure of Palladacycles during Pd-Catal Bulky Monophosphinobiaryl Ligands. Angewandte Chemie 2006, 118 (6), 939-942.

4. Barder, T. E.; Biscoe, M. R.; Buchwald, S. L., Structural Insights into Active Catalyst Structures and Oxidative Addition to (Biaryl)phosphine−Palladium Complexes via Density Functional Theory and Experimental Studies. Organometallics 2007, 26 (9), 2183-2192.

5. (a) El-Hellani, A.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Structure and Bonding of a Zwitterionic Iridium Complex Supported by a Phosphine with the Parent Carba-closo-dodecaborate CB11H11– Ligand Substituent. Organometallics 2013, 32 (23), 6887-6890; (b) Lavallo, V.; Wright, J. H.; Tham, F. S.; Quinlivan, S., Perhalogenated Carba-closo-dodecaborate Anions as Ligand Substituents: Applications in Gold Catalysis. Angewandte Chemie International Edition 2013, 52 (11), 3172-3176; (c) Estrada, J.; Lee, S. E.; McArthur, S. G.; El-Hellani, A.; Tham, F. S.; Lavallo, V., Resisting B–H oxidative addition: The divergent reactivity of the o-carborane and carba- closo-dodecaborate ligand substituents. Journal of Organometallic Chemistry 2015, 798, Part 1, 214-217; (d) Estrada, J.; Woen, D. H.; Tham, F. S.; Miyake, G. M.; Lavallo, V., Synthesis and Reactivity of a Zwitterionic Palladium Allyl Complex Supported by a Perchlorinated Carboranyl Phosphine. Inorganic Chemistry 2015, 54 (11), 5142-5144; (e) Estrada, J.; Lugo, C. A.; McArthur, S. G.; Lavallo, V., Inductive effects of 10 and 12-

vertex closo-carborane anions: cluster size and charge make a difference. Chemical Communications 2016, 52 (9), 1824-1826.

6. (a) El-Hellani, A.; Lavallo, V., Fusing N-Heterocyclic Carbenes with Carborane Anions. Angewandte Chemie International Edition 2014, 53 (17), 4489-4493; (b) Asay, M. J.; Fisher, S. P.; Lee, S. E.; Tham, F. S.; Borchardt, D.; Lavallo, V., Synthesis of unsymmetrical N-carboranyl NHCs: directing effect of the carborane anion. Chemical Communications 2015, 51 (25), 5359-5362.

7. (a) Li, Y.; Carroll, P. J.; Sneddon, L. G., Ionic-Liquid-Promoted Decaborane Dehydrogenative Alkyne-Insertion Reactions: A New Route to o-Carboranes. Inorganic Chemistry 2008, 47 (20), 9193-9202; (b) Reed, C. A., H+, CH3+, and R3Si+ Carborane Reagents: When Triflates Fail. Accounts of Chemical Research 2010, 43 (1), 121-128; (c) Douvris, C.; Michl, J., Update 1 of: Chemistry of the Carba-closo-dodecaborate(−) Anion, CB11H12–. Chemical Reviews 2013, 113 (10), PR179-PR233; (d) Olid, D.; Nunez, R.; Vinas, C.; Teixidor, F., Methods to produce B-C, B-P, B-N and B-S bonds in boron clusters. Chemical Society Reviews 2013, 42 (8), 3318-3336; (e) Spokoyny Alexander, M., New ligand platforms featuring boron-rich clusters as organomimetic substituents. In Pure and Applied Chemistry, 2013; Vol. 85, p 903; (f) Popescu, A. R.; Teixidor, F.; Viñas, C., Metal promoted charge and hapticities of phosphines: The uniqueness of carboranylphosphines. Coordination Chemistry Reviews 2014, 269, 54-84; (g) Zhang, J.; Xie, Z., Synthesis, Structure, and Reactivity of 13- and 14-Vertex Carboranes. Accounts of Chemical Research 2014, 47 (5), 1623-1633.

8. Douvris, C.; Ozerov, O. V., Hydrodefluorination of Perfluoroalkyl Groups Using Silylium-Carborane Catalysts. Science 2008, 321 (5893), 1188-1190.

9. (a) Thomas, C. M.; Peters, J. C., Coordinating Anions: (Phosphino)tetraphenylborate Ligands as New Reagents for Synthesis. Inorganic Chemistry 2004, 43 (1), 8-10; (b) Kim, Y.; Jordan, R. F., Synthesis, Structures, and Ethylene Dimerization Reactivity of Palladium Alkyl Complexes That Contain a Chelating Phosphine–Trifluoroborate Ligand. Organometallics 2011, 30 (16), 4250- 4256; (c) Kronig, S.; Theuergarten, E.; Daniliuc, C. G.; Jones, P. G.; Tamm, M., Anionic N-Heterocyclic Carbenes That Contain a Weakly Coordinating Borate Moiety. Angewandte Chemie International Edition 2012, 51 (13), 3240-3244; (d) Gutsulyak, D. V.; Gott, A. L.; Piers, W. E.; Parvez, M., Dimerization of Ethylene by Nickel Phosphino– Borate Complexes. Organometallics 2013, 32 (11), 3363-3370.

10. Spokoyny, A. M.; Lewis, C. D.; Teverovskiy, G.; Buchwald, S. L., Extremely Electron-Rich, Boron-Functionalized, Icosahedral Carborane-Based Phosphinoboranes. Organometallics 2012, 31 (24), 8478-8481.

11. Barrios-Landeros, F.; Carrow, B. P.; Hartwig, J. F., Effect of Ligand Steric Properties and Halide Identity on the Mechanism for Oxidative Addition of Haloarenes to

Trialkylphosphine Pd(0) Complexes. Journal of the American Chemical Society 2009, 131 (23), 8141-8154.

12. Biscoe, M. R.; Fors, B. P.; Buchwald, S. L., A New Class of Easily Activated Palladium Precatalysts for Facile C−N Cross-Coupling Reactions and the Low Temperature Oxidative Addition of Aryl Chlorides. Journal of the American Chemical Society 2008, 130 (21), 6686-6687.

Chapter 5: Synthesis and Reactivity of a Zwitterionic Pd Allyl Complex Supported by a Perchlorinated Carboranyl Phosphine Ligand Substituent

5.1 Abstract

- A zwitterionic palladium complex of a phosphine bearing CB11Cl11 as a ligand substituent is reported. A single-crystal X-ray diffraction study reveals that, in the solid state, one of the chlorides of the carborane cage occupies a coordination site of the square- planar complex. However, in solution, the P–carborane bond of the ligand is rapidly rotating at temperatures as low as −90 °C, which demonstrates the carborane substituent’s weak coordinative ability even though this anion is covalently linked to the phosphine ligand. The complex is thermally stable and catalyzes the vinyl addition polymerization of norbornene.

5.2 Introduction

The availability of a diverse array of ligand frameworks is critical for the development of effective homogeneous catalysts. One of the most useful classes of ligands is monodentate tertiary phosphines. Such ancillary ligands are typically constructed with combinations of ubiquitous alkyl and aryl groups. An

- interesting alternative to classical alkyl and Figure 5-1. (top) H2C2B10H10 1, nido-7,8-C2B10H12 2, - and HCB11H11 3 (unlabeled vertices = B–H). aryl ligand R groups is polyhedral (bottom) Representations of anionic carboranyl phosphine 4 (unlabeled vertices = B), phosphine sulfonates 5, and trifluoroborate phosphines 6. carboranes.1 The most common

1d, 1e, 2 carboranes used in phosphine design are derived from C2B10H10. Due to its ready availability, the H2C2B10H10 isomer 1 (Figure 5-1) of these clusters is most often implemented. While this cluster offers a variety of interesting characteristics, such as an essentially spherical steric profile and the ability to form H–H hydrogen bonds,1e it exhibits reactivity that may not be desirable for catalysis, such as facile B–H cyclometalation and vertex extrusion reactions to afford nido-carboranes 2 (Figure 5-1). The latter property has been utilized by Teixidor et al.2a-h, 3 to create a family of anionic phosphines, bearing nido- carborane substituents, that exhibit unique coordination chemistry. Compared to

- 1b H2C2B10H10, isoelectronic HCB11H11 clusters 3 (Figure 5-1) are far more inert, particularly when polyhalogenated. Furthermore, these clusters are among the weakest coordinating anions known and have allowed for the preparation of many exotic cations4 and extraordinarily active systems for silylium catalysis.5

Recently, we reported the first transition-metal complexes [gold(I) and iridium(I)]

- 6 of phosphines containing carbon-functionalized CB11H11 as ligand substituents. When these phosphines react with metal precursors, ligand coordination as well as salt metathesis to produce zwitterionic species is observed. The previously reported zwitterionic gold

6b - complex, which is supported by the phosphine 4 (Figure 5-1), bearing CB11Cl11 , catalyzes the hydroamination of alkynes with amines with unprecedented activity. More recently, we have developed the synthetic methodology to access both dianionic symmetrical and monoanionic unsymmetrical carboranyl N-heterocyclic carbenes

(NHCs).7

One of the applications we envision utilizing such ligands for is the design of single-component late-metal olefin polymerization catalysts analogous to systems supported by phosphine sulfonates 5 (Figure 5-1).8 Initially discovered by Drent et al.8a,

8b and extensively investigated by Nozaki et al.8c, 8f, 8g, 8j, 8l and Jordan et al.,8d, 8e, 8h, 8i,

8k palladium-based olefin polymerization catalysts supported by ligands 5 produce unusual highly linear polyethylene as well as copolymers with high polar monomer content. This unique behavior is proposed to arise from pairing a weakly coordinating sulfonate moiety with a cis-binding phosphine. Kim and Jordan9 and Piers et al.10 have independently reported related systems that implement phosphines 6 that contain a tetracoordinate trifluoroborate moiety in lieu of a sulfonate group (Figure 5-1). The structure of carboranyl phosphine 4 is similar to those of 5 and 6 in that it contains a weakly coordinating anion proximal to the phosphine center. Here, we report our initial investigation into the synthesis of a zwitterionic palladium complex supported by 4 and its reactivity with several simple olefin substrates.

5.3 Results and Discussion

For this first investigation, we chose to target the palladium allyl complex 7 (Figure 5-2) because the allyl moiety should bring stability to the complex, compared to a more reactive metal alkyl. Hence, Figure 5-2. Synthesis of complex 7. ligand 4 was reacted with (ClPd allyl)2 in F–C6H5, whereupon a precipitate formed

(7/LiCl) over 30 min. After the precipitate was collected, the solid was dried and extracted

31 with CH2Cl2 (to remove insoluble LiCl) to afford 7 in 97% yield. Analysis of 7 by P

NMR spectroscopy shows the clean formation of a single new phosphorus-containing species (complex, +117.6 ppm; ligand, +77.3 ppm). At ambient temperature, the 11B NMR spectrum of 7 shows a 1:5:5 ratio of resonances, indicating that the clusters local C5v symmetry is not disrupted. This suggests that the carborane anion is rotating rapidly about the P–carborane bond and is weakly coordinating even though it is projected into the coordination sphere of the metal. Indeed, variable-temperature 11B NMR spectroscopy shows that the cluster is freely rotating down to temperatures as low as −90

°C. This behavior is in contrast to a recent report6a of a zwitterionic iridium complex,

- bearing the parent CB11H11 ligand substituent, that shows coordinative interactions in solution and in the solid state. The 1H NMR spectrum of 7 displays the expected signals for diasteriotopic isopropyl phosphine substituents as well as three broad resonances for the fluxional allyl group. Variable-temperature NMR analysis (−50 °C) allows resolution of the allyl resonances into five distinct multiplets.

A single-crystal X-ray diffraction study of 7 (Figure 5-3) confirms its identity and reveals that, in the solid state, one of the chlorides of the carborane anion is occupying a coordination site of the square-planar palladium center. The Pd–Cl distance [2.401(7) Å] is in the range reported for common Pd–Cl bonds in allyl complexes (terminal Pd–Cl,

2.350–2.410 Å;11 bridging Pd–Cl, 2.380–2.420 Å11-12 and shorter than that reported for

palladium complexes with weakly bound chlorocarbons (>2.425 Å).12a Although the metal center interacts with the halogen in the solid state, the B–Cl bond is only slightly elongated

[1.794(3) Å; average other B–Cl bonds in the pentagonal belt adjacent to carbon =

1.766(3) Å], indicating little perturbation of this exo-cluster bond. The P–Pd bond length of 6 [2.354(7) Å] is comparable

(2.319–2.363 Å)13 to that of palladium complexes containing o-carboranyl Figure 5-3. Solid-state structure of complex 7 (hydrogen atoms and disorder about phosphines. The allyl group is coordinated in the allyl group omitted for clarity). Color code: brown, B; green, Cl; gray, C; violet, P; blue, Pd. 3 an η fashion [Pd–Callyl bond lengths for C1 =

2.206(3) Å, C2 = 2.126(5) Å, and C3 = 2.138(5) Å] and is disordered over two positions.

We next sought to probe the ability of 7 to polymerize simple olefins. Interestingly, complex 7 does not react at all with ethylene (1 atm, 80 °C, CD2Cl2). However, the addition of 1 mol % complex 7 to a solution of 1-hexene results in very rapid and quantitative isomerization to 2-hexene (Table 1, entry 1). The ability of 7 to isomerize 1-hexene suggests the formation of intermediate palladium hydrides likely formed via facile β- elimination, which likely prevents chain propagation. We next examined the reactivity of 7 with styrene because isomerization is impossible with this substrate. Treatment of a

CH2Cl2 solution of styrene with 1 mol % 7 at room temperature results in the formation of traces (<1%) of a styrene head-to-tail dimer over 4 h. However, heating the mixture at 50

°C for 16 h improves the yield to 75% (Table 1, entry 2). In contrast, at ambient temperature

(solvent = C6H6 or F–C6H5), the more reactive norbornene instantly polymerizes when contacted with 7. However, the yield is low

(<25%), and the polymer is completely insoluble in all solvents. The insolubility of the polymer also suggests that a very high molecular weight material is formed, which results from poor initiation of the catalyst. In Table 5-1. support of this assertion, the primary complex observed in the crude mixture is 7, as indicated by 31P NMR analysis. Given that the initial insertion step of the polymerization likely occurs via an η1-bound ally species, we reasoned that the addition of the monomer to a heated solution of 7 might improve the performance. Indeed, the addition of norbornene to a solution of 7 (0.1 mol %) at 80 °C in F–C6H5 results in a higher yield of the polymer with improved solubility (Table 1, entry 3). Analysis of the material by 1H

NMR shows the absence of olefinic resonances, which suggests that the product forms via a vinyl addition polymerization pathway.14 Analysis of the polymer’s molecular weight by gel permeation chromatography coupled with light scattering revealed that the polymer had a weight-average molecular weight (Mw) of 44 kDa and a polydispersity index of 1.71.

Only a handful of palladium catalysts can produce soluble polynorbornene14 because the high reactivity of the monomer usually prohibits controlled polymerization. Monitoring the

reaction by 1H NMR shows that the polymerization is continuous throughout heating of the reaction, which demonstrates that the catalyst is thermally stable. The thermal stability of catalyst 7 is unusual because most palladium olefin polymerization systems begin to rapidly decompose above 70 °C.15

5.4 Summary and Conclusion

This paper introduces a novel approach to the design of zwitterionic olefin

- polymerization catalysts. Interestingly, the presence of a CB11Cl11 ligand substituent allows the polymerization of norbornene even though the anion is buried in the coordination sphere of the metal. This highlights the extraordinary weak coordinative ability of these R groups and paves the way for the development of other systems that will be effective for the polymerization of unstrained olefins. We are currently investigating the design of related Pd–Me catalysts, which should display enhanced initiation relative to the allyl system 7.

5.5 Experimental

The previous experimental data has been published and may be obtained online at DOI: 10.1021/acs.inorgchem.5b00576

5.6 References

1. (a) Scholz, M.; Hey-Hawkins, E., Carbaboranes as Pharmacophores: Properties, Synthesis, and Application Strategies. Chemical Reviews 2011, 111 (11), 7035-7062; (b) Douvris, C.; Michl, J., Update 1 of: Chemistry of the Carba-closo-dodecaborate(−) Anion, CB11H12–. Chemical Reviews 2013, 113 (10), PR179-PR233; (c) Olid, D.; Nunez, R.; Vinas, C.; Teixidor, F., Methods to produce B-C, B-P, B-N and B-S bonds in boron clusters. Chemical Society Reviews 2013, 42 (8), 3318-3336; (d) Spokoyny Alexander, M., New ligand platforms featuring boron-rich clusters as organomimetic substituents. In Pure and Applied Chemistry, 2013; Vol. 85, p 903; (e) Popescu, A. R.; Teixidor, F.; Viñas, C., Metal promoted charge and hapticities of phosphines: The uniqueness of carboranylphosphines. Coordination Chemistry Reviews 2014, 269, 54-84.

2. (a) Teixidor, F.; Casabo, J.; Romerosa, A. M.; Vinas, C.; Rius, J.; Miravitlles, C., Simultaneous conversion of Pd-PPh3 and B-H to B-PPh2 under exceedingly mild conditions. Crystal and molecular structure of PdPPh3Cl{7-SMe-8-Me-11-PPh2-7,8- C2B9H10}. Journal of the American Chemical Society 1991, 113 (26), 9895-9896; (b) Teixidor, F.; Vinas, C.; Mar Abad, M.; Lopez, M.; Casabo, J., Synthesis of [7,8-(PPh2)2- 7,8-C2B9H10]-: a ligand analogous to 1,2-bis(diphenylphosphino)ethane with a "built- in" negative charge. Organometallics 1993, 12 (9), 3766-3768; (c) Viñas, C.; Nuñez, R.; Teixidor, F.; Kivekäs, R.; Sillanpää, R., Modulation of Agostic B−H⇀Ru Bonds in exo- Monophosphino-7,8-Dicarba-nido-undecaborate Derivatives. Organometallics 1996, 15 (18), 3850-3858; (d) Teixidor, F.; Núñez, R.; Viñas, C.; Sillanpää, R.; Kivekäs, R., The Distinct Effect of the o-Carboranyl Fragment: Its Influence on the I−I Distance in R3PI2 Complexes. Angewandte Chemie International Edition 2000, 39 (23), 4290-4292; (e) Núñez, R.; Farràs, P.; Teixidor, F.; Viñas, C.; Sillanpää, R.; Kivekäs, R., A Discrete P⋅⋅⋅ ⋅⋅⋅P Assembly: The Large Influence of Weak Interactions on the 31P NMR Spectra of Phosphane–Diiodine Complexes. Angewandte Chemie International Edition 2006, 45 (8), 1270-1272; (f) Farràs, P.; Olid-Britos, D.; Viñas, C.; Teixidor, F., Unprecedented B–H Activation Through Pd-Catalysed B–Cvinyl Bond Coupling on Borane Systems. European Journal of Inorganic Chemistry 2011, 2011 (16), 2525-2532; (g) Farràs, P.; Teixidor, F.; Rojo, I.; Kivekäs, R.; Sillanpää, R.; González-Cardoso, P.; Viñas, C., Relaxed but Highly Compact Diansa Metallacyclophanes. Journal of the American Chemical Society 2011, 133 (41), 16537-16552; (h) Popescu, A.-R.; Laromaine, A.; Teixidor, F.; Sillanpää, R.; Kivekäs, R.; Llambias, J. I.; Viñas, C., Uncommon Coordination Behaviour of P(S) and P(Se) Units when Bonded to Carboranyl Clusters: Experimental and Computational Studies on the Oxidation of Carboranyl Phosphine Ligands. Chemistry – A European Journal 2011, 17 (16), 4429-4443; (i) Joost, M.; Estévez, L.; Miqueu, K.; Amgoune, A.; Bourissou, D., Oxidative Addition of Carbon–Carbon Bonds to Gold. Angewandte Chemie International Edition 2015, 54 (17), 5236-5240.

3. Teixidor, F.; Ayllon, J. A.; Vinas, C.; Kivekas, R.; Sillanpaa, R.; Casabo, J., A novel B-H[right harpoon up]Ru agostic bond. Crystal structure of [RuCl{7,8-[small

micro]-S(CH2CH2)S-C2B9H10}(PPh3)2][middle dot]Me2CO. Journal of the Chemical Society, Chemical Communications 1992, (18), 1281-1282.

4. Reed, C. A., H+, CH3+, and R3Si+ Carborane Reagents: When Triflates Fail. Accounts of Chemical Research 2010, 43 (1), 121-128.

5. (a) Douvris, C.; Ozerov, O. V., Hydrodefluorination of Perfluoroalkyl Groups Using Silylium-Carborane Catalysts. Science 2008, 321 (5893), 1188-1190; (b) Douvris, C.; Nagaraja, C. M.; Chen, C.-H.; Foxman, B. M.; Ozerov, O. V., Hydrodefluorination and Other Hydrodehalogenation of Aliphatic Carbon−Halogen Bonds Using Silylium Catalysis. Journal of the American Chemical Society 2010, 132 (13), 4946-4953; (c) Ramírez-Contreras, R.; Bhuvanesh, N.; Zhou, J.; Ozerov, O. V., Synthesis of a Silylium Zwitterion. Angewandte Chemie International Edition 2013, 52 (39), 10313-10315.

6. (a) El-Hellani, A.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Structure and Bonding of a Zwitterionic Iridium Complex Supported by a Phosphine with the Parent Carba-closo-dodecaborate CB11H11– Ligand Substituent. Organometallics 2013, 32 (23), 6887-6890; (b) Lavallo, V.; Wright, J. H.; Tham, F. S.; Quinlivan, S., Perhalogenated Carba-closo-dodecaborate Anions as Ligand Substituents: Applications in Gold Catalysis. Angewandte Chemie International Edition 2013, 52 (11), 3172-3176.

7. (a) El-Hellani, A.; Lavallo, V., Fusing N-Heterocyclic Carbenes with Carborane Anions. Angewandte Chemie International Edition 2014, 53 (17), 4489-4493; (b) Asay, M. J.; Fisher, S. P.; Lee, S. E.; Tham, F. S.; Borchardt, D.; Lavallo, V., Synthesis of unsymmetrical N-carboranyl NHCs: directing effect of the carborane anion. Chemical Communications 2015, 51 (25), 5359-5362.

8. (a) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I., Palladium catalysed copolymerisation of ethene with alkylacrylates: polar comonomer built into the linear polymer chain. Chemical Communications 2002, (7), 744-745; (b) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I., The first example of palladium catalysed non-perfectly alternating copolymerisation of ethene and carbon monoxide. Chemical Communications 2002, (9), 964-965; (c) Kochi, T.; Noda, S.; Yoshimura, K.; Nozaki, K., Formation of Linear Copolymers of Ethylene and Acrylonitrile Catalyzed by Phosphine Sulfonate Palladium Complexes. Journal of the American Chemical Society 2007, 129 (29), 8948-8949; (d) Luo, S.; Vela, J.; Lief, G. R.; Jordan, R. F., Copolymerization of Ethylene and Alkyl Vinyl Ethers by a (Phosphine- sulfonate)PdMe Catalyst. Journal of the American Chemical Society 2007, 129 (29), 8946-8947; (e) Vela, J.; Lief, G. R.; Shen, Z.; Jordan, R. F., Ethylene Polymerization by Palladium Alkyl Complexes Containing Bis(aryl)phosphino-toluenesulfonate Ligands. Organometallics 2007, 26 (26), 6624-6635; (f) Ito, S.; Munakata, K.; Nakamura, A.; Nozaki, K., Copolymerization of Vinyl Acetate with Ethylene by Palladium/Alkylphosphine−Sulfonate Catalysts. Journal of the American Chemical Society 2009, 131 (41), 14606-14607; (g) Nakamura, A.; Ito, S.; Nozaki, K.,

Coordination−Insertion Copolymerization of Fundamental Polar Monomers. Chemical Reviews 2009, 109 (11), 5215-5244; (h) Shen, Z.; Jordan, R. F., Copolymerization of Ethylene and Vinyl Fluoride by (Phosphine-bis(arenesulfonate))PdMe(pyridine) Catalysts: Insights into Inhibition Mechanisms. Macromolecules 2010, 43 (21), 8706- 8708; (i) Conley, M. P.; Jordan, R. F., cis/trans Isomerization of Phosphinesulfonate Palladium(II) Complexes. Angewandte Chemie International Edition 2011, 50 (16), 3744-3746; (j) Nakamura, A.; Munakata, K.; Ito, S.; Kochi, T.; Chung, L. W.; Morokuma, K.; Nozaki, K., Pd-Catalyzed Copolymerization of Methyl Acrylate with Carbon Monoxide: Structures, Properties and Mechanistic Aspects toward Ligand Design. Journal of the American Chemical Society 2011, 133 (17), 6761-6779; (k) Cai, Z.; Shen, Z.; Zhou, X.; Jordan, R. F., Enhancement of Chain Growth and Chain Transfer Rates in Ethylene Polymerization by (Phosphine-sulfonate)PdMe Catalysts by Binding of B(C6F5)3 to the Sulfonate Group. ACS Catalysis 2012, 2 (6), 1187-1195; (l) Ito, S.; Ota, Y.; Nozaki, K., Ethylene/allyl monomer cooligomerization by nickel/phosphine-sulfonate catalysts. Dalton Transactions 2012, 41 (45), 13807-13809; (m) Nakamura, A.; Anselment, T. M. J.; Claverie, J.; Goodall, B.; Jordan, R. F.; Mecking, S.; Rieger, B.; Sen, A.; van Leeuwen, P. W. N. M.; Nozaki, K., Ortho-Phosphinobenzenesulfonate: A Superb Ligand for Palladium-Catalyzed Coordination–Insertion Copolymerization of Polar Vinyl Monomers. Accounts of Chemical Research 2013, 46 (7), 1438-1449.

9. Kim, Y.; Jordan, R. F., Synthesis, Structures, and Ethylene Dimerization Reactivity of Palladium Alkyl Complexes That Contain a Chelating Phosphine– Trifluoroborate Ligand. Organometallics 2011, 30 (16), 4250-4256.

10. Gutsulyak, D. V.; Gott, A. L.; Piers, W. E.; Parvez, M., Dimerization of Ethylene by Nickel Phosphino–Borate Complexes. Organometallics 2013, 32 (11), 3363-3370.

11. (a) Viciu, M. S.; Navarro, O.; Germaneau, R. F.; Kelly, R. A.; Sommer, W.; Marion, N.; Stevens, E. D.; Cavallo, L.; Nolan, S. P., Synthetic and Structural Studies of (NHC)Pd(allyl)Cl Complexes (NHC = N-heterocyclic carbene). Organometallics 2004, 23 (7), 1629-1635; (b) Crabtree, R. H., The Organometallic Chemistry of the Transition Metals (fifth ed.), vol. 5, John wiley & sons Ltd., New Jersey. 2009.

12. (a) Chernyshova, E. S.; Goddard, R.; Pörschke, K.-R., Mononuclear NHC−Pd−π- Allyl Complexes Containing Weakly Coordinating Ligands. Organometallics 2007, 26 (13), 3236-3251; (b) Szuromi, E.; Shen, H.; Goodall, B. L.; Jordan, R. F., Polymerization of Norbornene and Methyl Acrylate by a Bimetallic Palladium(II) Allyl Complex. Organometallics 2008, 27 (3), 402-409.

13. (a) Lee, T.; Lee, S. W.; Jang, H. G.; Kang, S. O.; Ko, J., Synthesis and Reactivity of Organotin Compounds Containing the C,P-Chelating o-Carboranylphosphino Ligand [o-C2B10H10PPh2-C,P](CabC,P). X-ray Structures of (CabC,CH2P)SnMe2Br, [(CabC,P)SnMe2]2Pd, and [(CabC,P)SnMe2]Pd(PEt3)Cl. Organometallics 2001, 20 (4), 741-748; (b) Lee, Y.-J.; Lee, J.-D.; Kim, S.-J.; Keum, S.; Ko, J.; Suh, I.-H.; Cheong, M.;

Kang, S. O., Synthesis, Structure, and DFT Calculation of (Phosphino-o-carboranyl)silyl Group 10 Metal Complexes: Formation of Stable trans-Bis(P,Si-chelate)metal Complexes. Organometallics 2004, 23 (2), 203-214; (c) Fey, N.; Haddow, M. F.; Mistry, R.; Norman, N. C.; Orpen, A. G.; Reynolds, T. J.; Pringle, P. G., Regioselective B- Cyclometalation of a Bulky o-Carboranyl Phosphine and the Unexpected Formation of a Dirhodium(II) Complex. Organometallics 2012, 31 (7), 2907-2913.

14. (a) Mehler, C.; Risse, W., The Pd2+ -catalyzed polymerization of norbornene. Die Makromolekulare Chemie, Rapid Communications 1991, 12 (5), 255-259; (b) Mehler, C.; Risse, W., Addition polymerization of norbornene catalyzed by palladium(2+) compounds. A polymerization reaction with rare chain transfer and chain termination. Macromolecules 1992, 25 (16), 4226-4228; (c) Seehof, N.; Mehler, C.; Breunig, S.; Risse, W., Pd2+ catalyzed addition polymerizations of norbornene and norbornene derivatives. Journal of Molecular Catalysis 1992, 76 (1), 219-228; (d) Haselwander, T. F. A.; Heitz, W.; Krügel, S. A.; Wendorff, J. H., Polynorbornene: synthesis, properties and simulations. Macromolecular Chemistry and Physics 1996, 197 (10), 3435-3453; (e) Blank, F.; Scherer, H.; Janiak, C., Oligomers and soluble polymers from the vinyl polymerization of norbornene and 5-vinyl-2-norbornene with cationic palladium catalysts. Journal of Molecular Catalysis A: Chemical 2010, 330 (1–2), 1-9.

15. (a) Tempel, D. J.; Johnson, L. K.; Huff, R. L.; White, P. S.; Brookhart, M., Mechanistic Studies of Pd(II)−α-Diimine-Catalyzed Olefin Polymerizations1. Journal of the American Chemical Society 2000, 122 (28), 6686-6700; (b) Camacho, D. H.; Salo, E. V.; Ziller, J. W.; Guan, Z., Cyclophane-Based Highly Active Late-Transition-Metal Catalysts for Ethylene Polymerization. Angewandte Chemie International Edition 2004, 43 (14), 1821-1825; (c) Popeney, C. S.; Rheingold, A. L.; Guan, Z., Nickel(II) and Palladium(II) Polymerization Catalysts Bearing a Fluorinated Cyclophane Ligand: Stabilization of the Reactive Intermediate. Organometallics 2009, 28 (15), 4452-4463; (d) Pan, H.; Zhu, L.; Li, J.; Zang, D.; Fu, Z.; Fan, Z., A thermal stable α-diimine palladium catalyst for copolymerization of ethylene with functionalized olefins. Journal of Molecular Catalysis A: Chemical 2014, 390, 76-82.

Chapter 6: Fusing Dicarbollide Ions with N-Heterocyclic Carbenes

6.1 Abstract

Discovered by Hawthorne in 1965, dicarbollide ions are an intriguing class of nido- carboranes that mimic the behavior of the cyclopentadienyl anion. Here, we show that it is possible to directly link the dicarbollide ion to an N-Heterocyclic Carbene (NHC), to form an isolable N-dicarbollide NHC dianion. This molecule can be accessed via the sequential double deprotonation of a mono-nido-carboranyl imidazolium zwitterion. As revealed by a single crystal x-ray diffraction study, the first deprotonation leads to a monoanionic dicarbollide ion that adopts a bis(dicarbollide) structure in the solid-state. Subsequent deprotonation of this monoanion, leads to the first N-dicarbollide NHC, which was fully characterized by multinuclear NMR spectroscopy as well as a single crystal X-ray diffraction study.

6.2 Introduction

Carboranes are an unusual class of boron cluster compounds that contain at least one hypercoordinate carbon center. In 1965, Hawthorne reported the synthesis of a novel

2- nido-carborane anion [H2C2B9H9 ], the so called dicarbollide ion 1 (Figure 6-1). This fascinating molecule mimics the coordination chemistry of the cyclopentadienyl anion, and thus enabled the discovery of a new class of sandwich complexes, namely bis(dicarbollides) 2 (Figure 6-1), as well as other metallocarboranes. Aside from their fundamental importance such complexes have found applications in medicine, redox active materials, and catalysis.

Another class of molecules that contain an unusual carbon center are stable carbenes. While the first stable carbene was reported by Bertrand,

Arduengo discovered an exceedingly Figure 6-1. The dicarbollide ion 1, a generic useful class of carbenes, namely the N- bis(dicarbollide) sandwich complex 2, and a generic NHC 3. Unlabeled vertices of 1 and 2: B-H; M = metal ion. Heterocyclic Carbenes (NHCs) 3

(Figure 6-1). NHCs have found wide applications as ligands for catalysts/materials and can serve as catalysts in their own right. Recently, we reported the fusion of N-heterocyclic

1- carbenes with the chemically robust 12-vertex closo-carborane anion [HCB11H11 ]

(technically carbenoids, since alkali metal cations remain coordinated). We were interested in targeting these molecules to probe the compatibility of the NHC with such closo- carborane anions, and design a new class of super bulky charged ligands for applications

1- in coordination chemistry and catalysis. Given the observed compatibility of [CB11H11 ] closo-cluster as an N-substituent for NHCs, we became curious if certain nido-carboranes might also be viable carbene R-groups. Here we report the synthesis and full characterization of a dicarbollide monoanion that can be deprotonated to afford an isolable

N- dicarbollide NHC.

6.3 Results and Discussion

Previously we utilized condensation chemistry with the known 12-vertex closo-

1- carborane amine [H2NCB11H11 ] to access N-closo-carboranyl imidazolium NHC precursors. We envisioned using analogous methodology to target appropriate precursors

N-dicarbollide NHCs, however no such nido-amines have been reported. This was rather surprising given that ortho-carboranyl amine H2NC2B10H11 4 (Figure 6-2) has been known since 1965, and related compounds are well known to undergo B-vertex extrusion reactions with base to produce

1- nido anions [RCCHB9H11 ]. These species Figure 6-2. Synthesis of the novel anionic nido- amine 5 and its subsequent reaction with N-mesityl are the conjugate acids of dicarbollide ions, oxazolinium tetrafluoroborate to produce the N- nido-imidazolium zwitterion 6. which are formed by deprotonation of the acidic hydrogen bridging the open pentagonal face of the nido-structure. Thus we prepared the o-carborane amine 4 and subjected it to vertex extrusion conditions, utilizing KOH as the base. Gratifyingly the novel anionic nido- amine 5 was produced in 90% yield (Figure 6-2). For solubility reasons we decided to next target the mono-nido-carboranyl imidazolium zwitterion 6, in lieu of a symmetrical di- carborane substituted N-N-substituted derivative, since the ensuing NHC would be tetraanionic and likely insoluble. Thus, 5 was reacted with Furstner’s N-mesityl oxazolinium salt and subsequently treated with HBF4 to produce the N-nido imidazolium zwitterion 6 in 75% yield (Figure 6-2).

The zwitterion 6 contains two acidic protons with varying acidities (Ha, Hb) (Figure 6-

1- 3). For the parent unsubstituted cluster [H3C2B10H10 ] approximate pKa values in DMSO have been reported for Ha ( pKa = 13.5). Likewise, imidazolium salts have pKa values in

DMSO ranging from 16-24. Therefore, we predicted it would be possible to selectively and

sequentially deprotonate Ha and Hb of 6. Thus, 6 was treated with one equivalent of sodium hexamethyldisilazide (NaHMDS) in THF and the reaction was monitored by NMR spectroscopy. Analysis of the 1H NMR spectrum showed a similar set of resonances compared to 6 except the characteristic broad signal at -2.58 ppm for the bridging hydrogen

Ha had vanished, which is consistent with the selective formation of the dicarbollide monoanion 7.

The structure of 7 was unambiguously confirmed via a single crystal x-ray diffraction study (Figure

6-3). Interestingly, in the solid state two molecules of 7 have sequestered a single Na1+ ion to form a bis(dicarbollide) complex that Figure 6-3. In the solid-state the N-dicarbollide imidazolium monoanion 7 adopts a bis(dicarbollide) structure with one contains a single Na1+ countercation. Na1+ ion sequestered in the sandwich complex and the other Na1+ ion functioning as a solvent separated countercation. 1+ The bond lengths of each N- Note: the THF solvated Na countercation is omitted for clarity. Color code: grey = carbon, brown = boron, blue = nitrogen, white = hydrogen, orange = sodium. dicarbollide imidazolium ion are similar, thus we will highlight select distances from only one half of the complex. The carborane-nitrogen bond length (N1-C4 = 1.4651(18) Å) is in the range for a normal C-N single bond and thus their is no evidence for exo-ힹ-conjugation between the cluster and the heterocycle. Similarly, the heterocycles bond lengths (C1-N1 = 1.3295(18), C1-N2

=1.3326(18), C2-N2 = 1.3794(18), C3-N1 = 1.3861(17), C2-C3 = 1.338(2) Å) are in the expected range for an imidazolium salt.

We next turned our attention to the possibility of deprotonating Hb of

7 to form the N-dicarbollide NHC dianion 8 (Figure 6-4). Thus, 7 was treated with one equivalent of

NaHMDS in THF and the reaction Figure 6-4. Synthesis of the novel dicarbollide monoanion 7 and its subsequent reaction with an additional equivalent of was monitored by NMR NaHMDS to produce the N-dicarbollide NHC dianion 8. spectroscopy. The 1H NMR spectrum showed the disappearance of the characteristic doublets of the imidazolium ring of 7 and the formation of two new singlet H resonances at 6.94 and 6.45 ppm, which is consistent with the formation of 8. The 13C spectrum of the material confirmed the formation of 8 as indicated by the appearance of a resonance at

210.0 ppm, corresponding to the formation of an NHC. Compared to the analogous N-

1- Mesityl/N-[CB11H11 ] closo-carboranyl NHC, this resonance is shifted downfield by 10 ppm, which is likely the result of the electron withdrawing nature of the C-substituted nido- cluster.

The structure of 8 was unambiguously determined by a single crystal X-ray diffraction study (Figure 6-5). Unlike 7, which forms a bis(dicarbollide) structure in the solid-state, 8 maintains a dicarbollide

5 1+ structure with η coordination to Na . The isFigure depicted. 6-5. ColorSolid- code:state structure grey = carbon, of the Nbrown- = boron,dicarbolliide blue = NHC nitrogen, dianion orange 8. Note: = sodium, the red = second Na+ countercation is coordinated to a oxygen,molecule hydrogens is a dimer omitted in the unit for cellclarity. with both units linked via H-B-B-H edge coordination to Na2. For simplicity only one half of the dimer is H-B-B-H edge of the pentagonal face of the depicted. Color code: grey = carbon, brown = boron, blue = nitrogen, orange = sodium, red = dicarbollide. The NHC is chelating Na1 and oxygen, hydrogens omitted for clarity. displays the greatest yaw distortion (30˚) ever observed (previous largest yaw distortion =

22˚). Although slightly shorter compared to 7 the carborane-nitrogen bond length (N1-C4

= 1.4479(14) Å) of 8 is still in the range for a normal C-N single bond indicating no exo-

ힹ-conjugation. In addition, the NHC ힹ-system is orthogonal to the cyclopentadiene like ힹ-orbitals of the dicarbollide ion, precluding the possibility of ힹ-conjugation.

Compared to 7 the NHC ring bond lengths (C1-N1 = 1.3577(15), C1-N2 =1.3645(15), C2-

C3=1.3405(18) Å) are slightly elongated and shortened (C3-N1 =1.3817(15) Å), which is consistent with typical bond length changes upon transformation of an imidazolium to an

NHC. However, the C2-N2 bond length (1.3845(16) Å) is somewhat unusual, since it is elongated with respect to 7, and normally analogous bonds in classical alkyl/aryl substituted NHCs shorten. This observation might be the result of the dramatic yaw distortion induced by the chelating nature of the ligand.

6.4 Summary and Conclusion

Over 50 years after Hawthorne’s seminal discovery of the dicarbollide ion we have shown that this classical cyclopentadienyl anion mimic can be directly fused to stable carbenes. The marriage of these two unusual classes of carbon containing molecules introduces a new paradigm in ligand design. We are currently exploring the implementation of such N-dicarbollide NHCs in catalysis as well as investigating the possibility of preparing redox active N-bis(dicarbollide) NHCs

6.5 Experimental

Unless otherwise stated, all manipulations were carried out using standard Schlenk or glovebox techniques (O2, H2O < 1ppm) under a dinitrogen or argon atmosphere.

Solvents were dried on NaK, K or CaH2, distilled under argon before use. Ortho-carborane amine and the 1-mesityl-3-acetoxyoxazolinium tetrafluoroborate were prepared by literature methods[1,2]. Reagents were purchased from commercial vendors and used without further purification. NMR spectra were recorded on Bruker Avance 300MHz,

Bruker Avance 600MHz, Varian Inova 300MHz, Varian Inova 400MHz or Varian Inova

500MHz spectrometers. 1H NMR and 13C NMR chemical shifts were referenced to residual

11 31 solvent. B NMR chemical shifts were externally referenced to BF3OEt2. P NMR chemical shifts were externally referenced to 80% H3PO4 in H2O. HRMS was recorded on

Agilent Technologies 6210 (time of flight LC/MS) using ESI technique. IR spectra were recorded on Attenuated Reflectance (ATR) and were run on ABB MB3000 spectrometer in the 525-4000 cm-1 frequency range using a diamond crystal.

Synthesis and Spectroscopic Data of Compound 5

1.0 gram of NH2C2B10H11 was dissolved in 20 mL of MeOH and 20 mL of deionized water.

After adding 3 equivalence of KOH (1.05 g) and refluxed overnight the reaction mixture was concentrated in vacuo and the resulting solid was recrystallized in deionized water containg 2 grams of CsCl (2 eq.) to afford the desired compound 5 in 90 % yield (1.59 g,

1 0.56 mmol). H NMR (500 MHz, Acetonitrile-d3, 25 ˚C): 2.01 (bs, 2H, NH2), 1.83 (bs,

1H, CH), -2.74 ppm (bs, 1H, bridging-H); 11B{1H} NMR (96 MHz, Acetonitrile-d3, 25

˚C): -5.6, -7.3, -8.8, -11.1, -14.4, -17.8, -28.6, -31.8 ppm; 13C{1H} NMR (126 MHz,

- Acetonitrile-d3, 25 ˚C): 74.7, 49.3 ppm. HRMS: calculated [M−H]- {H2NC2B9H12 }:

- -1 149.1929 m/z; Found [M−H] {H2NC2B9H12 }: 149.1932 m/z. IR: 2484 cm (B-H stretch)

Figure 6-6. 1H{11B} NMR of Compound 5 (300 MHz, Acetonitrile-d3, 25 ˚C)

Figure 6-7. 11B{1H} NMR of Compound 5 (96 MHz, Acetonitrile-d3, 25 ˚C)

Figure 6-8. 13C{1H} NMR of Compound 5 (126 MHz, THF, 25 ˚C)

Synthesis and Spectroscopic Data of Compound 6

1 equivalent of compound 5 (1.0 gram, 3.5 mmol) was combined with the known 1-mesityl-

3-acetoxyoxazolinium tetrafluoroborate (1.17 g, 3.5 mmol) in 30 mL of dry acetonitrile and heated under inert atmosphere overnight. To the resulting brown-yellow solution,

HBF4 was added (1 eq., 0.55 mL) and the resulting mixture was heated in a closed thick walled Teflon schlenk tube at 120 ˚C overnight. Upon completion of the reaction, the

brown-yellow solution was filtered, then concentrated in vacuo, extracted with methylene chloride and filtered. The resulting product was recrystallized in acetonitrile to afford the desired product compound 6 in 75 % yield (0.836 g). 1H NMR (500 MHz, THF-d8, 25

˚C): 9.14 (dd, 1H, CH), 7.85 (dd, 1H, CH), 7.52 (dd, 1H, CH), 7.11 (s, 2H, meta-CH) 2.34

(s, 3H, CH3) 2.03 (s, 6H, CH3) 1.73 (bs, 1H, CHcarborane) -2.58 ppm (bs, 1H, bridging-H);

11B{1H} NMR (96 MHz, Acetonitrile-d3, 25 ˚C): -4.9, -7.1, -9.7, -17.4, -27.7, -32.0 ppm;

13C{1H} NMR (126 MHz, THF-d8, 25 ˚C): 141.8, 137.6, 135.2, 132.1, 130.1, 124.6, 123.9,

71.6, 42.1, 20.9, 17.0 ppm. HRMS: calculated [M−H]- {N2C14B9H24}: 318.2829 m/z;

-1 Found [M−H]- {N2C14B9H24}: 318.2888 m/z. IR: 2519 cm (B-H stretch)

Figure 6-9. 1H NMR of Compound 6 (500 MHz, THF-d8, 25 ˚C)

Figure 6-10. 1H{11B} NMR of Compound 6 (300 MHz, Acetonitriel-d3, 25 ˚C)

Figure 6-11. 11B{1H} NMR of Compound 6 (96 MHz, Acetonitrile-d3, 25 ˚C)

Figure 6-12. 11B NMR of Compound 6 (96 MHz, Acetonitrile-d3, 25 ˚C)

Figure 6-13. 13C{1H} NMR of Compound 6 (126 MHz, THF-d8, 25 ˚C)

Synthesis and Spectroscopic Data of Compound 7

1.1 equivalents of NaHMDS was added to a 10 mL THF solution of compound 6 (0.50 g,

1.56 mmol) , stirred for 15 minutes and concentrated to approximately 5 mL at which point the desired compound begins to precipitate. The reaction mixture was placed in a -30 ˚C freezer for 15 minutes to promote further precipitation of the desired product at which point the supernatant was decanted off and the white precipitate was washed 3 times with cold

THF (3 x 2 mL) affording the desired compound 7 quantitatively. A single crystal suitable for X-ray was obtained by layering a dimethoxyethane solution of compound 7 with hexanes. 1H NMR (300 MHz, THF-d8, 25 ˚C): 8.43 (bs, 1H, CH), 7.56 (s, 1H, CH), 7.08

(s, 1H, CH), 7.00 (s, 2H, meta-CH), 2.30 (s, 3H, CH3), 1.99 (s, 6H, CH3), 1.34 (bs, 1H,

11 1 13 1 CHcarborane); B{ H} NMR (96 MHz, Acetonitrile-d3, 25 ˚C): -17.2, -42.9 ppm; C{ H}

NMR (126 MHz, THF-d8, 25 ˚C): 141.1, 135.6, 133.1, 130.2, 123.7, 122.9, 72.6, 31.4,

21.2, 17.5 ppm.

Figure 6-14. 1H NMR of Compound 7 (500 MHz, THF-d8, 25 ˚C)

Figure 6-15. 11B{1H} of Compound 7 (96 MHz, THF-d8, 25 ˚C)

Figure 6-16. 13C{1H} NMR of Compound 7 (96 MHz, DME, 25 ˚C)

Synthesis and Spectroscopic Data of Compound 8

2.1 equivalents of NaHMDS was added to a 10 mL THF solution of compound 6 (0.50 g,

1.56 mmol) , stirred for 25 minutes and concentrated to approximately 5 mL volume at which point 5 mL of diethyl ether was added and the resulting mixture was placed in a -30

˚C freezer overnight for crystallization. The resulting crystals were filtered and washed with (2 x 1 mL) cold diethyl ether affording the desired compound 8 in 85% yield (0.86 g,

1.33 mmol) including 4 coordinated THF molecules bound to the sodium cations. 1H NMR

(300 MHz, THF-d8, 25 ˚C): 6.94 (d, 1H, CH), 6.89 (s, 2H, meta-CH), 6.45 (d, 1H, CH),

11 1 2.27 (s, 6H, CH3), 1.96 (s, 3H, CH3), 1.91 (s, 3H, CH3), 1.29 (bs, 1H, CHcarborane); B{ H}

NMR (96 MHz, THF-d8, 25 ˚C): -14.5, -18.4, -20.1, -23.6, -41.9 ppm; 13C{1H} NMR

(126 MHz, THF-d8, 25 ˚C): 210.0, 139.5, 137.3, 135.7, 128.9, 120.5, 116.8, 37.9, 20.9,

17.7 ppm.

Figure 6-17. 1H NMR of Compound 8 (500 MHz, THF-d8, 25 ˚C)

Figure 6-18. 1H{11B} NMR of Compound 8 (300 MHz, THF-d8, 25 ˚C)

Figure 6-19. 11B{1H} NMR of Compound 8 (96 MHz, THF-d8, 25 ˚C)

Figure 6-20. 13C{1H} NMR of Compound 8 (126 MHz, THF-d8, 25 ˚C)

Figure 6-21. HSQC (13C-1H) NMR of Compound 8 (126 MHz, THF-d8, 25 ˚C)

X-Ray Structure Determination of Compound 7

A colorless prism fragment (0.487 x 0.337 x 0.256 mm3) was used for the single crystal x-ray diffraction study of [C28H48B18N4Na].[C12H30O6Na] (sample vL267JE_0m). The crystal was coated with paratone oil and mounted on to a cryo-loop glass fiber. X-ray intensity data were collected at 200(2) K on a Bruker APEX2 platform-CCD x-ray diffractometer system (fine focus Mo-radiation,

λ = 0.71073 Å, 50KV/30mA power). The CCD detector was placed at a distance of 5.0600 cm from the crystal.

A total of 2400 frames were collected for a sphere of reflections (with scan width of 0.3o in ω, starting ω and 2θ angles of –30o, and φ angles of 0o, 90o, 120o, and 240o, for every 600 frames, 40 sec/frame exposure time). The frames were integrated using the Bruker SAINT software package and using a narrow-frame integration algorithm. Based on a monoclinic crystal system, the integrated frames yielded a total of 38367 reflections at a maximum 2θ angle of 56.564o (0.75 Å resolution), of which 6987 were independent reflections (Rint = 0.0290, Rsig = 0.0204, redundancy

= 5.5, completeness = 100%) and 5162 (73.9%) reflections were greater than 2σ(I). The unit cell parameters were, a = 30.6054(18) Å, b = 12.3100(7) Å, c = 15.4481(9) Å, β = 104.7754(10)o, V =

3 3 5627.7(6) Å , Z = 4, calculated density Dc = 1.123 g/cm . Absorption corrections were applied

-1 (absorption coefficient μ= 0.081 mm ; max/min transmission = 0.980 /0.962 ) to the raw intensity data using the SADABS program.

The Bruker SHELXTL software package was used for phase determination and structure refinement. The distribution of intensities (E2-1 = 0.941) and systematic absent reflections indicated two possible space groups, Cc and C2/c. The space group C2/c (#15) was later determined to be correct. Direct methods of phase determination followed by two Fourier cycles

of refinement led to an electron density map from which most of the non-hydrogen atoms were identified in the asymmetric unit of the unit cell. With subsequent isotropic refinement, all of the non-hydrogen atoms were identified. There was half a molecule of[C28H48B18N4Na] and half a disordered molecule of [C12H30O6Na] present in the asymmetric unit of the unit cell. The

[C28H48B18N4Na] molecule was located at the diagonal glide plane perpendicular to the b-axis. The

[C12H30O6Na] molecule was located at the 2-fold rotation axis parallel to the b-axis. All the DME molecules coordinated to Na2-atom were modeled with disorder (disordered site occupancy ratios were 59%/41% and 50%/25%/25%).

Atomic coordinates, isotropic and anisotropic displacement parameters of all the non-hydrogen atoms were refined by means of a full matrix least-squares procedure on F2. The H-atoms were included in the refinement in calculated positions riding on the atoms to which they were attached, except the H2, H3, H4, and H5 atoms bonded to C2B, B3, B4, and B5, respectively were refined unrestrained. The refinement converged at R1 = 0.0543, wR2 = 0.1538, with intensity

I>2σ(I). The largest peak/hole in the final difference map was 0.266/-0.480 e/Å3.

Table 6-1. Crystal data and structure refinement for vL267JE_0m. Identification code vL267JE_0m Empirical formula C40 H78 B18 N4 Na2 O6 Formula weight 951.62 Temperature 200(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group C 2/c Unit cell dimensions a = 30.6054(18) Å = 90°. b = 12.3100(7) Å = 104.7754(10)°. c = 15.4481(9) Å  = 90°. Volume 5627.7(6) Å3 Z 4 Density (calculated) 1.123 Mg/m3 Absorption coefficient 0.081 mm-1 F(000) 2024 Crystal size 0.487 x 0.337 x 0.256 mm3 Theta range for data collection 1.792 to 28.282°. Index ranges -40<=h<=40, -16<=k<=16, -20<=l<=20 Reflections collected 38367 Independent reflections 6987 [R(int) = 0.0290] Completeness to theta = 25.242° 100.0 % Absorption correction Semi-empirical from equivalents Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 6987 / 502 / 487 Goodness-of-fit on F2 1.030 Final R indices [I>2sigma(I)] R1 = 0.0543, wR2 = 0.1538 R indices (all data) R1 = 0.0738, wR2 = 0.1701 Extinction coefficient n/a Largest diff. peak and hole 0.266 and -0.480 e.Å-3

X-Ray Structure Determination of Compound 8

A colorless prism fragment (0.331 x 0.302 x 0.114 mm3) was used for the single crystal x- ray diffraction study of [C28H46B18N4Na4][C4H8O]8 (sample vL196LE_0m). The crystal was coated with paratone oil and mounted on to a cryo-loop glass fiber. X-ray intensity data were collected at 200(2) K on a Bruker X8APEX2 platform-CCD x-ray diffractometer system (fine focus Mo-radiation,  = 0.71073 Å, 50KV/30mA power). The CCD detector was placed at a distance of 5.0000 cm from the crystal.

A total of 4400 frames were collected for a sphere of reflections (with scan width of 0.3o in  and starting  and 2 angles of –28o, and  angles of 0o, 72o, 144o, 216o, and 288o for every 400 frames, 1200 frames with starting  and 2 angles of –28o and -scan from

0-360o, 1200 frames with starting  and 2 angles of 0o and -scan from 0-360o, 60 sec/frame exposure time). The frames were integrated using the Bruker SAINT software package and using a narrow-frame integration algorithm. Based on a triclinic crystal system, the integrated frames yielded a total of 40581 reflections at a maximum 2 angle

o of 56.564 (0.75 Å resolution), of which 9422 were independent reflections (Rint = 0.0182,

Rsig = 0.0167, redundancy = 4.3, completeness = 99.7%) and 7468 (79.3%) reflections were greater than 2(I). The unit cell parameters were, a = 10.4317(4) Å, b = 11.8803(5) Å, c =

15.5883(6) Å,  = 99.2304(15)  = 92.0280(16)o,  = 90.5167(15) V = 1905.46(13) Å3, Z

3 = 1, calculated density Dc = 1.135 g/cm . Absorption corrections were applied (absorption

-1 coefficient  = 0.088 mm ; max/min transmission = 0.990/0.971) to the raw intensity data using the SADABS program.

The Bruker SHELXTL software package was used for phase determination and structure refinement. The distribution of intensities (E2-1 = 0.992) and no systematic absent reflections indicated two possible space groups, P-1and P1. The space group P-1 (#2) was later determined to be correct. Direct methods of phase determination followed by two

Fourier cycles of refinement led to an electron density map from which most of the non- hydrogen atoms were identified in the asymmetric unit of the unit cell. With subsequent isotropic refinement, all of the non-hydrogen atoms were identified. There was half a molecule of [C28H46B18N4Na4][C4H8O]8 present in the asymmetric unit of the unit cell, where three of the four THF molecules were modeled with disorder (disordered site occupancy ratios were 57%/43%, 55%/45% and 52%/48%). The

[C28H46B18N4Na4][C4H8O]8 molecule was located at the inversion center.

Atomic coordinates, isotropic and anisotropic displacement parameters of all the non- hydrogen atoms were refined by means of a full matrix least-squares procedure on F2. The

H-atoms were included in the refinement in calculated positions riding on the atoms to which they were attached, except the H2B, H3B, H4B, and H5B atoms bonded to C2B,

B3, B4, and B5, respectively were refined unrestrained. The refinement converged at R1 =

0.0466, wR2 = 0.1370, with intensity I>2 (I). The largest peak/hole in the final difference map was 0.336/-0.221 e/Å3.

Table 6-2. Crystal data and structure refinement for vL196JE_0m. Identification code vL196JE_0m Empirical formula C60 H110 B18 N4 Na4 O8 Formula weight 1302.05 Temperature 200(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P -1 Unit cell dimensions a = 10.4317(4) Å = 99.2304(15)°. b = 11.8803(5) Å = 92.0280(16)°. c = 15.5883(6) Å  = 90.5167(15)°. Volume 1905.46(13) Å3 Z 1 Density (calculated) 1.135 Mg/m3 Absorption coefficient 0.088 mm-1 F(000) 696 Crystal size 0.331 x 0.302 x 0.114 mm3 Theta range for data collection 2.008 to 28.282°. Index ranges -13<=h<=13, -10<=k<=15, -20<=l<=20 Reflections collected 40581 Independent reflections 9422 [R(int) = 0.0182] Completeness to theta = 25.242° 99.6 % Absorption correction Semi-empirical from equivalents Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 9422 / 524 / 577 Goodness-of-fit on F2 1.076 Final R indices [I>2sigma(I)] R1 = 0.0466, wR2 = 0.1370 R indices (all data) R1 = 0.0593, wR2 = 0.1463 Extinction coefficient n/a Largest diff. peak and hole 0.336 and -0.221 e.Å-3

6.6 References

1. a) R. N. Grimes, Daltons Trans. 2015, 44, 5939-5956; b) A. R. Popescu, F. Teixidor, C. Viñas, Coord. Chem. Rev. 2014, 269, 54- 84; c) J. Zhang, Z. Xie, Acc. Chem. Res. 2014, 47, 1623-1633;d) D. Olid, R. Núñez, C. Viñas, F. Teixidor, Chem. Soc. Rev. 2013, 42, 3318-3336; e) C. Douvris, J. Michl, Chem. Rev. 2013, 113, PR179- PR233; f) A. M. Spokoyny, Pure Appl. Chem. 2013, 85, 903-919; g) P. Farras, E. J. Juarez-Perez, M. Lepsik, R. Luque, R. Núñez, F. Teixidor, Chem. Soc. Rev. 2012, 41, 3445-3463; h) P. Farras, E. J. Juarez-Perez, M. Lepsik, R. Luque, R. Nunez, F. Teixidor, Chem. Soc. Rev. 2012, 41, 3445-3463; i) M. Scholz, E. Hey-Hawkins, Chem. Rev. 2011, 111, 7035-7062; j) C. A. Reed, Acc. Chem. Res. 2009, 43, 121-128.

2. M. F. Hawthorne, D. C. Young, P. A. Wegner, J. Am. Chem. Soc. 1965, 87, 1818- 1819.

3. a) A. M. Spokoyny, T. C. Li, O. K. Farha, C. W. Machan, C. She, C. L. Stern, T. J. Marks, J. T. Hupp, C. A. Mirkin, Angew. Chem. Int. Ed. 2010, 49, 5339-5343; b) T. C. Li, A. M. Spokoyny, C. She, O. K. Farha, C. A. Mirkin, T. J. Marks, J. T. Hupp, J. Am. Chem. Soc. 2010, 132, 4580-4582.

4. R. T. Baker, M. S. Delaney, R. E. King Iii, C. B. Knobler, J. A. Long, T. B. Marder, T. E. Paxson, R. G. Teller, M. F. Hawthorne, J. Am. Chem. Soc. 1984, 106, 2965-2978.

5. a) E. Peris, Chem. Rev. 2017; b) A. Nasr, A. Winkler, M. Tamm, Coord. Chem. Rev. 2016, 316, 68-124; c) M. H. Wang, K. A. Scheidt, Angew. Chem. Int. Ed. 2016, 55, 14912-14922; d) S. Wang, X. Wang, Angew. Chem. Int. Ed. 2016, 55, 2308-2320; e) M. Soleilhavoup, G. Bertrand, Acc. Chem. Res. 2015, 48, 256-266; f) P. Chauhan, D. Enders, Angew. Chem. Int. Ed. 2014, 53, 1485- 1487; g) K. F. Donnelly, A. Petronilho, M. Albrecht, Chem. Commun. 2013, 49, 1145-1159; h) X. Bugaut, F. Glorius, Chem. Soc. Rev. 2012, 41, 3511-3522; i) T. Dröge, F. Glorius, Angew. Chem. Int. Ed. 2010, 49, 6940- 6952; j) F. E. Hahn, M. C. Jahnke, Angew. Chem. Int. Ed. 2008, 47, 3122-3172; k) N. Marion, S. Díez-González, S. P. Nolan, Angew. Chem. Int. Ed. 2007, 46, 2988-3000.

6. A. Igau, H. Grutzmacher, A. Baceiredo, G. Bertrand, J. Am. Chem. Soc. 1988, 110, 6463-6466.

7. A. J. Arduengo, R. L. Harlow, M. Kline, J. Am. Chem. Soc. 1991, 113, 361-363.

8. a) M. J. Asay, S. P. Fisher, S. E. Lee, F. S. Tham, D. Borchardt, V. Lavallo, Chem. Commun. 2015, 51, 5359-5362; b) A. El-Hellani, V. Lavallo, Angew. Chem. Int. Ed. 2014, 53, 4489-4493.

9. S. P. Fisher, A. El-Hellani, F. S. Tham, V. Lavallo, Daltons Trans. 2016, 45, 9762- 9765.

10. Willans and Coworkers have recently reported the insitu formation of NHC transition metal metalacarborane complexes featuring an ethylene tether linking the NHC to the cluster; see,

11. T. Jelinek, J. Plesek, S. Hermanek, B. Stibr, Collect. Czech. Chem. Commun. 1986, 51, 819-829.

12. L. I. Zakharkin, V. N. Kalinin, Zh. Obshch. Khim. 1965, 35, 1882- 1884.

13. A. Furstner, M. Alcarazo, V. Cesar, C. W. Lehmann, Chem. Commun. 2006, 2176- 2178.

14. A similar triazolium zwitterion was recently reported as a catalyst precursor for organocatalysis, but no spectroscopic evidence was provided for the formation of a dicarbollide ion, NHC, or N-dicarbollide NHC; see,

15. P. Farràs, F. Teixidor, V. Branchadell, Inorg. Chem. 2006, 45, 7947-7954.

16. a) A. M. Magill, K. J. Cavell, B. F. Yates, J. Am. Chem. Soc. 2004, 126, 8717-8724; b) Y.-J. Kim, A. Streitwieser, J. Am. Chem. Soc. 2002, 124, 5757-5761; c) R. W. Alder, P. R. Allen, S. J. Williams, J. Chem. Soc., Chem. Commun. 1995, 1267-1268.

17. a) M. Poyatos, J. A. Mata, E. Peris, Chem. Rev. 2009, 109, 3677- 3707; b) C. H. Leung, C. D. Incarvito, R. H. Crabtree, Organometallics 2006, 25, 6099-6107.

18. S. A. Mungur, S. T. Liddle, C. Wilson, M. J. Sarsfield, P. L. Arnold, Chem. Commun. 2004, 2738-2739.

Chapter 7: Isolation of a Carborane-Fused Triazole Radical Anion

7.1 Abstract

- A change in the redox properties of a triazole fused to CB11Cl11 through methylation to form a zwitterion enabled facile chemical reduction of the compound to an isolable triazole radical anion. The radical anion is stabilized by kinetic protection by the chlorinated carborane and the delocalization of spin density throughout the exo-cluster π system.

7.2 Introduction

Radicals play important roles as reactive intermediates in biological systems, in materials science, and in the construction of conducting polymers, and as spin carriers for solar-cell applications.1 With the exception of simple inorganic molecules, such as NO

2 and O2, most radicals are transient species that defy isolation. In 1900, Gomberg, in an attempt to isolate hexaphenylethane, discovered the first persistent organic radical species, triphenylmethylene. Following the pioneering studies of Gomberg, a variety of persistent and stable3 radical species of the second-row elements were reported.1b, 1c, 3 These species are often stabilized by kinetic protection of the radical center and/or by significant delocalization of the unpaired electron. Further stability can also be bestowed by the incorporation of a heavier main-group element into the spin system.

Among the radicals of interest are those derived from aromatic species. Aromatic

N heterocycles, for example, have been the subject of numerous oxidation/reduction studies; however, very few neutral radical species have been isolated.4 Cationic N- heterocyclic radicals are similarly rare, with only a limited number of isolable species

reported.5 Isolable N-heterocyclic radical anions are even more elusive: only a handful of examples of crystalline solids have been described, most of which are stabilized by adjacent heavy chalcogens (S or Se).6 Another class of radicals derived from aromatic

1d, 1e 2− 7 species that have recently drawn attention are those of icosahedral boranes (B12 ) and

− 7-8 carboranes (CB11 ). All of these species are produced by one-electron oxidation of the three-dimensional aromatic cluster by the removal of a skeletal electron. Notable examples

.− 9 .− 10 of isolated species are the stable radical anions B12Me12 and B12Cl12 , as well as the

. 11 neutral carborane radical HCB11Me11 .

Recently, we reported the synthesis of a family of unusual 1,2,3-triazoles that are fused with icosahedral carborane anions.12 Interestingly, both the carborane and five- membered-triazole portions of these molecules independently feature aromatic characteristics. Herein, we report the investigation of the electrochemical properties of one such molecule and its subsequent derivatization, which led to the first isolable triazole radical anion.13

7.3 Results and Discussion

It was postulated that because N-heterocyclic carborane anions, such as the N- phenyl-substituted derivative 1, display some degree of exo-cluster delocalization,14 these species might possess interesting redox properties (Figure 7-1, top). Cyclic voltammetry

(CV) experiments indicated that compound 1 undergoes no observable oxidation processes

and only irreversible reduction processes at large negative potentials (Figure 7-1, bottom,

A). We postulated that the large potential required to reduce the heterocyclic anion 1 was due to unfavorable electrostatic repulsion associated with the Figur e 7-1. Top: Synthesis of the zwitterionic production of a radical dianion. heterocycle 2 and chemical reduction to the radical anion 3 (unlabeled vertices: B Cl; We reasoned that if the negative Tf=trifluoromethanesulfonyl). Bottom: Cyclic voltammograms of A) 1 and B) 2 in THF (Eo=0.87 V; −1 supporting electrolyte: 0.1 M Bu4NPF6; scan rate: 100 mVs ; charge of 1 were masked by the working electrode: Pt wire; reference: Fc/Fc+=0.0 V). formation of a zwitterion, its electrochemical properties would be altered in such a way as to enable facile reduction. Hence, anion 1 was alkylated with methyl triflate to afford the zwitterionic species 2 in excellent yield (90 %; Figure 7-1, top). Heterocycle 2 was air- stable and showed no propensity to act as a methylating agent towards nucleophilic solvents, such as water, methanol, or acetonitrile. This behavior contrasts starkly with that of zwitterionic methylating agents derived from simple halogenated carborane anions,8a,

15 which are strong enough to abstract hydrides from alkanes, and highlights the unusual nucleophilicity of 1.

To compare the electrochemical properties of the anionic carborane 1 with those of the zwitterionic heterocycle 2, we carried out CV experiments. In marked contrast to 1, 2 undergoes a clearly reversible reduction at −0.87 V versus Fc/Fc+ (Figure 7-1,

bottom, B; Fc=ferrocene). However, at reduction potentials less than −2.50 V, irreversible one-electron-transfer steps are evident. Encouraged by these results, we set out to chemically reduce 2 and isolate the radical species 3.13 The CV data indicate that the choice of reducing agent may be extremely important inasmuch as over-reduction appears to be a distinct possibility. With a reduction potential of −1.33 V, which is comfortably higher than the required −0.87 V and well below that of the first irreversible reduction step

(−2.50 V), cobaltocene appeared to be an ideal choice of reducing agent. Furthermore, the cobaltocenium cation is generally weakly coordinating, robust, and often good for crystallization. Thus, a solution of cobaltocene in diethyl ether was added to the zwitterionic heterocycle 2. The mixture was stirred for 12 h, after which time a green/brown precipitate was isolated. NMR spectroscopic analysis of the material revealed the formation of the cobaltocenium cation and thus indicated that electron transfer had occurred. There were no signals for the heterocycle in the NMR spectrum. A radical species was detected by EPR spectroscopy: one broad signal was observed at g=2.003 (the peak- to-peak width of the signal was 28.8 G, or 28.2 G at 120 K), which is in very good agreement with the value calculated for the radical anion 3 (g=2.004). The hyperfine coupling constants could also be calculated (see Figure S18 in the Supporting Information) but were not observed experimentally because of the broad nature of the signal. The broadness of the signal may be attributed to the inherent asymmetry of the five-membered ring.

To confirm the formation of the radical anion 3 and compare its structural features with those of the unreduced zwitterion 2, we obtained crystals suitable for single-crystal

X-ray diffraction studies of both compounds. In the solid state, the five- membered heterocyclic portions of both 2 and 3 (Figure 7-2) are perfectly planar (sum of the internal angles: 540°), and the phenyl group remains coplanar

(dihedral angle between the two rings:

5.9° in 2 and 1.8° in 3). However, an examination of the bond lengths of the radical anion 3 reveals a dramatic Figure 7-2. X-ray crystal structure of the radical anion 3. Solvent of crystallization (THF) in the unit cell of 3 has elongation of the N1-N2-N3 portion of been omitted for clarity. Thermal ellipsoids are drawn at the 50 % probability level (H white, C gray, B brown, Cl the heterocycle (N1-N2 1.386(5), N2- green).

N3 1.391(5) Å; Figure 7-2) with respect to the zwitterionic carborane 2 (N1-N2 1.307(3),

N2-N3 1.302(3) Å; see Figure S11). This elongation is consistent with the addition of an electron into an N-N-N π antibonding molecular orbital of 2. Whereas the N1-C1 distances in 2 and 3 are identical (2: 1.413(3), 3: 1.413(5) Å), a bond contraction between N3 and

B1 (N3-B1 1.514(4) in 2, 1.472(6) Å in 3) is an indication of enhanced π donation14 from

N3 to the cluster. Additionally, the N3-C2 bond of 3 (1.413(5) Å) is shorter than that in the zwitterion 2 (1.430(4) Å), which suggests some delocalization of the spin density onto C2 of the benzene ring. However, the internal benzene-ring bond lengths are essentially identical in 2 and 3. Likewise, all bond lengths in the icosahedral core of the radical

anion 3 are similar to those in the zwitterion 2 and thus suggest no disruption of the skeletal electrons of the cage.

To gain further insight into the electronic structures of the redox couple 2 and 3, we carried out DFT calculations at the B3PW91 level of theory.

The LUMO of 2 and the

SOMO of 3 are nearly Figure 7-3. Calculated lowest occupied molecular orbital (LUMO) of 2 and singly occupied molecular orbital (SOMO) of 3. identical and are composed of Calculations were carried out at the B3PW91 level of theory. orbital contributions from the three nitrogen atoms and the pendent benzene ring, with very slight mixing of the icosahedral core (Figure 7-3). A π-bonding interaction between N3 and C2 is evident in the SOMO of 3, in agreement with the observed bond contraction in the X-ray diffraction study. The spin density is primarily located on the N3 fragment

(86 %), with 75 % centered on N2 (see Figure S17). The majority of the remaining spin density is distributed through the π system of the pendent benzene ring. The computed and experimental UV/Vis spectra of 2 and 3 (λmax(2)=385/360 nm; λmax(3)=347/346 nm) are in good agreement and thus allow an accurate discussion of the electronic transitions associated with these absorptions. The two absorptions are analogous π→π* transitions from the HOMO to the LUMO of 2 and from the SOMO-1 to the SOMO of 3 (see Figures

S13–S16). The blue-shifted absorbance of radical 3 relative to that of 2 is in line with a higher-energy electron transition to a half-filled antibonding orbital.

7.4 Summary and Conclusion

Radical 3 is stabilized by a combination of kinetic protection by the chlorinated porcupine-like carborane portion of the molecule and delocalization of the unpaired electron throughout the exo-cluster π system. Since the synthesis of precursors of type 1 is fairly general,12 the isolation of 3 paves the way for access to an entirely new series of stable radical heterocycles. Furthermore, our strategy of masking the charge of the carborane anion to render the reduction process more favorable can probably be applied to the isolation of other radical anions that contain the CB11 structural motif. We are currently investigating electronically distinct derivatives of 3 and exploring the possibility of applying these principles to the preparation of other novel heterocyclic radical anions.

Additionally, we are actively exploring the application of these radical anions in functional materials.

7.5 Experimental

The previous experimental data has been published and may be obtained online at DOI: 10.1002/anie.201306764

100

7.6 References

1. (a) Baumgarten, M.; Müllen, K., Radical ions: Where organic chemistry meets materials sciences. In Electron Transfer I, Mattay, J., Ed. Springer Berlin Heidelberg: Berlin, Heidelberg, 1994; pp 1-103; (b) Hicks, R. G., What's new in stable radical chemistry? Organic & Biomolecular Chemistry 2007, 5 (9), 1321-1338; (c) Zard, S. Z., Recent progress in the generation and use of nitrogen-centred radicals. Chemical Society Reviews 2008, 37 (8), 1603-1618; (d) Kaim, W.; Hosmane, N. S.; Záliš, S.; Maguire, J. A.; Lipscomb, W. N., Boratome als Spinträger in zwei- und dreidimensionalen Systemen. Angewandte Chemie 2009, 121 (28), 5184-5193; (e) Kaim, W.; Hosmane, N. S.; Záliš, S.; Maguire, J. A.; Lipscomb, W. N., Boron Atoms as Spin Carriers in Two- and Three- Dimensional Systems. Angewandte Chemie International Edition 2009, 48 (28), 5082- 5091; (f) Ratera, I.; Veciana, J., Playing with organic radicals as building blocks for functional molecular materials. Chemical Society Reviews 2012, 41 (1), 303-349.

2. (a) Gomberg, M., AN INSTANCE OF TRIVALENT CARBON: TRIPHENYLMETHYL. Journal of the American Chemical Society 1900, 22 (11), 757- 771; (b) Gomberg, M., ON TRIVALENT CARBON. Journal of the American Chemical Society 1901, 23 (7), 496-502; (c) Gomberg, M., ON TRIVALENT CARBON. Journal of the American Chemical Society 1902, 24 (7), 597-628.

3. Griller, D.; Ingold, K. U., Persistent carbon-centered radicals. Accounts of Chemical Research 1976, 9 (1), 13-19.

4. (a) Neugebauer, F. A., Kristallisiertes Tetrazolinyl-Radikal. Angewandte Chemie 1969, 81 (13), 496-496; (b) Neugebauer, F. A., Crystalline Tetrazolinyl Radical. Angewandte Chemie International Edition in English 1969, 8 (7), 520-520; (c) Neugebauer, F. A., Hydrazidinyl-Radikale: 1,2,3,5-Tetraazapentenyle, Verdazyle und Tetrazolinyle. Angewandte Chemie 1973, 85 (11), 485-493; (d) Neugebauer, F. A., Hydrazidinyl Radicals: 1 2,4,5-Tetraazapentenyls, Verdazyls, and Tetrazolinyls. Angewandte Chemie International Edition in English 1973, 12 (6), 455-464; (e) Neugebauer, F. A.; Fischer, H.; Krieger, C., 2,3-Dihydro-1H-1,2,4-triazol-1-yle, eine neue Gruppe stabiler Radikale. Angewandte Chemie 1989, 101 (4), 486-488; (f) Neugebauer, F. A.; Fischer, H.; Krieger, C., 2,3-Dihydro-1H-1,2,4-triazol-1-yls, a Novel Group of Stable Radicals. Angewandte Chemie International Edition in English 1989, 28 (4), 491-492; (g) Barth, T.; Kanellakopulos, B.; Krieger, C.; Neugebauer, F. A., Solid- state interactions in radicals: crystal structures of [small alpha]- and [small beta]-10- phenyl-5(10H)-Phenazinyl. Journal of the Chemical Society, Chemical Communications 1993, (21), 1626-1628; (h) Neugebauer, F. A.; Fischer, H., 2,5-Dihydro-1H-1,2,4- triazol-2-yl radicals: Syntheses and properties. Tetrahedron 1995, 51 (47), 12883-12898; (i) Barr, C. L.; Chase, P. A.; Hicks, R. G.; Lemaire, M. T.; Stevens, C. L., Synthesis and Characterization of Verdazyl Radicals Bearing Pyridine or Pyrimidine Substituents: A New Family of Chelating Spin-Bearing Ligands. The Journal of Organic Chemistry 1999, 64 (24), 8893-8897; (j) Risto, M.; Assoud, A.; Winter, S. M.; Oilunkaniemi, R.;

101

Laitinen, R. S.; Oakley, R. T., Heavy Atom Analogues of 1,2,3-Dithiazolylium Salts: Preparation, Structures and Redox Chemistry. Inorganic Chemistry 2008, 47 (21), 10100- 10109; (k) Chemistruck, V.; Chambers, D.; Brook, D. J. R., Structure−Property Relationships of Stable Free Radicals: Verdazyls with Electron-Rich Aryl Substituents. The Journal of Organic Chemistry 2009, 74 (5), 1850-1857; (l) Quast, H.; Philipp, G.; Ross, K.-H.; Doht, U. M., Stable 3-Imino-2,3-dihydroindazol-1-yl Radicals. European Journal of Organic Chemistry 2011, 2011 (10), 1891-1895.

5. (a) Schulz, A.; Kaim, W.; Hausen, H.-D., Crystalline heterocyclic radical cation salts as stable intermediates of new redox-mediating systems. Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases 1988, 84 (10), 3207-3214; (b) Schulz, A.; Kaim, W., Elektrochemische und spektroskopische Charakterisierung von N,N′-Dialkylchinoxalinium-Redoxsystemen. Chemische Berichte 1991, 124 (1), 129-139; (c) Hausen, H. D.; Kaim, W.; Schulz, A.; Moscherosch, M.; Jordanov, J., Erstmaliger Nachweis von π–π Dimerenbildung bei stabilen 1,4- Dialkylchinoxalinium-Radikalkationen. Struktur, Spektroskopie und Magnetismus / First Evidence for π-π-Dimerization of the Stable 1,4-Dialkylquinoxalinium Radical Cations. Structure, Spectroscopy and Magnetism. In Zeitschrift für Naturforschung B, 1993; Vol. 48, p 1181; (d) Hausen, H.-D.; Schulz, A.; Kaim, W., Flavosemiquinone model systems. Part 3. Molecular and crystal structure of the cation radical salt 1,4,6,7- tetramethylquinoxalinium triiodide. Journal of the Chemical Society, Perkin Transactions 2 1993, (3), 343-345; (e) Hilgers, F.; Kaim, W.; Schulz, A.; Zalis, S., Electronic absorption spectroscopy of some exceptionally stable 1,4-dialkyl-1,4-dihydro-1,4- diazinium radical cations: assignment of transitions, vibrational structure and effects of [small pi]-[small pi] dimerization. Journal of the Chemical Society, Perkin Transactions 2 1994, (1), 135-138; (f) Nishinaga, T.; Komatsu, K., Persistent [small pi] radical cations: self-association and its steric control in the condensed phase. Organic & Biomolecular Chemistry 2005, 3 (4), 561-569.

6. (a) Bock, H.; John, A.; Näther, C.; Ruppert, K., Strukturen und Molekül- Eigenschaften ladungsgestörter Moleküle. 52. Mitteilung. 2, 3-Diphenylchinoxalin- Radikalanionen in Lösung und im Kristall. Helvetica Chimica Acta 1994, 77 (6), 1505- 1519; (b) Makarov, A. Y.; Irtegova, I. G.; Vasilieva, N. V.; Bagryanskaya, I. Y.; Borrmann, T.; Gatilov, Y. V.; Lork, E.; Mews, R.; Stohrer, W.-D.; Zibarev, A. V., [1,2,5]Thiadiazolo[3,4-c][1,2,5]thiadiazolidyl: A Long-Lived Radical Anion and Its Stable Salts. Inorganic Chemistry 2005, 44 (20), 7194-7199; (c) Bagryanskaya, I. Y.; Gatilov, Y. V.; Gritsan, N. P.; Ikorskii, V. N.; Irtegova, I. G.; Lonchakov, A. V.; Lork, E.; Mews, R.; Ovcharenko, V. I.; Semenov, N. A.; Vasilieva, N. V.; Zibarev, A. V., [1,2,5]Selenadiazolo[3,4-c][1,2,5]thiadiazole and [1,2,5]Selenadiazolo[3,4- c][1,2,5]thiadiazolidyl – A Synthetic, Structural, and Theoretical Study. European Journal of Inorganic Chemistry 2007, 2007 (30), 4751-4761; (d) Maxim, C.; Matni, A.; Geoffroy, M.; Andruh, M.; Hearns, N. G. R.; Clerac, R.; Avarvari, N., C 3 symmetric tris(phosphonate)-1,3,5-triazine ligand: homopolymetallic complexes and its radical anion. New Journal of Chemistry 2010, 34 (10), 2319-2327.

102

7. Hawthorne, M. F.; Pushechnikov, A., Polyhedral borane derivatives: Unique and versatile structural motifs. In Pure and Applied Chemistry, 2012; Vol. 84, p 2279.

8. (a) Reed, C. A., H+, CH3+, and R3Si+ Carborane Reagents: When Triflates Fail. Accounts of Chemical Research 2010, 43 (1), 121-128; (b) Farràs, P.; Olid-Britos, D.; Viñas, C.; Teixidor, F., Unprecedented B–H Activation Through Pd-Catalysed B–Cvinyl Bond Coupling on Borane Systems. European Journal of Inorganic Chemistry 2011, 2011 (16), 2525-2532; (c) Himmelspach, A.; Finze, M.; Raub, S., Tetrahedral Gold(I) Clusters with Carba-closo-dodecaboranylethynido Ligands: [{12- -closo- 1-CB11H11}2]. Angewandte Chemie International Edition 2011, 50 (11), 2628-2631; (d) Popescu, A.-R.; Laromaine, A.; Teixidor, F.; Sillanpää, R.; Kivekäs, R.; Llambias, J. I.; Viñas, C., Uncommon Coordination Behaviour of P(S) and P(Se) Units when Bonded to Carboranyl Clusters: Experimental and Computational Studies on the Oxidation of Carboranyl Phosphine Ligands. Chemistry – A European Journal 2011, 17 (16), 4429- 4443; (e) Scholz, M.; Hey-Hawkins, E., Carbaboranes as Pharmacophores: Properties, Synthesis, and Application Strategies. Chemical Reviews 2011, 111 (11), 7035-7062; (f) Lavallo, V.; Wright, J. H.; Tham, F. S.; Quinlivan, S., Perhalogenated Carba-closo- dodecaborate Anions as Ligand Substituents: Applications in Gold Catalysis. Angewandte Chemie International Edition 2013, 52 (11), 3172-3176; (g) Olid, D.; Nunez, R.; Vinas, C.; Teixidor, F., Methods to produce B-C, B-P, B-N and B-S bonds in boron clusters. Chemical Society Reviews 2013, 42 (8), 3318-3336.

9. Peymann, T.; B. Knobler, C.; Frederick Hawthorne, M., An unpaired electron incarcerated within an icosahedral borane cage: synthesis and crystal structure of the blue, air-stable {[closo-B12(CH3)12][middle dot]}- radical. Chemical Communications 1999, (20), 2039-2040.

10. (a) Boeré, R. T.; Kacprzak, S.; Keßler, M.; Knapp, C.; Riebau, R.; Riedel, S.; Roemmele, T. L.; Rühle, M.; Scherer, H.; Weber, S., Oxidation von closo-[B12Cl12]2− zum Radikalanion [B12Cl12].− und zum neutralen B12Cl12. Angewandte Chemie 2011, 123 (2), 572-575; (b) Boeré, R. T.; Kacprzak, S.; Keßler, M.; Knapp, C.; Riebau, R.; Riedel, S.; Roemmele, T. L.; Rühle, M.; Scherer, H.; Weber, S., Oxidation of closo- [B12Cl12]2− to the Radical Anion [B12Cl12].− and to Neutral B12Cl12. Angewandte Chemie International Edition 2011, 50 (2), 549-552.

11. King, B. T.; Noll, B. C.; McKinley, A. J.; Michl, J., Dodecamethylcarba-closo- dodecaboranyl (CB11Me12•), a Stable Free Radical. Journal of the American Chemical Society 1996, 118 (44), 10902-10903.

12. Wright, J. H.; Kefalidis, C. E.; Tham, F. S.; Maron, L.; Lavallo, V., Click-Like Reactions with the Inert HCB11Cl11– Anion Lead to Carborane-Fused Heterocycles with Unusual Aromatic Character. Inorganic Chemistry 2013, 52 (10), 6223-6229.

103

13. (a) Treinin, A.; Hayon, E., Spectroscopic Observation of the Azide Radical in Solution. The Journal of Chemical Physics 1969, 50 (1), 538-539; (b) Hayon, E.; Simic, M., Absorption spectra and kinetics of the intermediate produced from the decay of azide radicals. Journal of the American Chemical Society 1970, 92 (25), 7486-7487; (c) Roberts, B. P.; Winter, J. N., Electron spin resonance studies of radicals derived from organic azides. Journal of the Chemical Society, Perkin Transactions 2 1979, (10), 1353- 1361; (d) Ram, M. S.; Stanbury, D. M., Electron-transfer reactions involving the azidyl radical. The Journal of Physical Chemistry 1986, 90 (16), 3691-3696; (e) Alfassi, Z. B.; Harriman, A.; Huie, R. E.; Mosseri, S.; Neta, P., The redox potential of the azide/azidyl couple. The Journal of Physical Chemistry 1987, 91 (8), 2120-2122; (f) DeFelippis, M. R.; Faraggi, M.; Klapper, M. H., Redox potentials of the azide and dithiocyanate radicals. The Journal of Physical Chemistry 1990, 94 (6), 2420-2424.

14. (a) Boyd, L. A.; Clegg, W.; Copley, R. C. B.; Davidson, M. G.; Fox, M. A.; Hibbert, T. G.; Howard, J. A. K.; Mackinnon, A.; Peace, R. J.; Wade, K., Exo-[small pi]- bonding to an ortho-carborane hypercarbon atom: systematic icosahedral cage distortions reflected in the structures of the fluoro-, hydroxy- and amino-carboranes, 1-X-2-Ph-1,2- C2B10H10[space](X = F, OH or NH2) and related anions. Dalton Transactions 2004, (17), 2786-2799; (b) Batsanov, A. S.; Clegg, W.; Copley, R. C. B.; Fox, M. A.; Gill, W. R.; Grimditch, R. S.; Hibbert, T. G.; Howard, J. A. K.; MacBride, J. A. H.; Wade, K., Preparative and structural studies on sulfur-linked carborane icosahedra: 2-Phenyl-ortho- carboranyl-sulfur systems (2-Ph-1,2-C2B10H10)2X (X = S, S2 or SO), and ortho- carboran-di-yl systems (1,2-C2B10H10Y)2 (Y = S or SO). Polyhedron 2006, 25 (2), 300- 306; (c) Fox, M. A.; Nervi, C.; Crivello, A.; Batsanov, A. S.; Howard, J. A. K.; Wade, K.; Low, P. J., Structural, spectroscopic, electrochemical and computational studies of C,C′-diaryl-ortho-carboranes, 1-(4-XC6H4)-2-Ph-1,2-C2B10H10 (X = H, F, OMe, NMe2, NH2, OH and O−). Journal of Solid State Electrochemistry 2009, 13 (10), 1483- 1495; (d) Fox, M. A.; Peace, R. J.; Clegg, W.; Elsegood, M. R. J.; Wade, K., Trends in ortho-carboranes 1-X-2-R-1,2-C2B10H10 (R = Ph, Me) bearing an exo-CN-bonded substituent group (X = NO, NNR′ or NHR′′). Polyhedron 2009, 28 (12), 2359-2370.

15. (a) Finze, M., Kohlenstoff-Extrusion/Cluster-Kontraktion: Synthese des fluorierten Cyano-closo-undecaborats K2[3-NC-closo-B11F10]. Angewandte Chemie 2007, 119 (46), 9036-9039; (b) Finze, M., Carbon Extrusion/Cluster Contraction: Synthesis of the Fluorinated Cyano-closo-Undecaborate K2[3-NC-closo-B11F10]. Angewandte Chemie International Edition 2007, 46 (46), 8880-8882; (c) Molinos, E.; Brayshaw, S. K.; Kociok-Kohn, G.; Weller, A. S., Cationic rhodium mono-phosphine fragments partnered with carborane monoanions [closo-CB11H6X6]- (X = H, Br). Synthesis, structures and reactivity with alkenes. Dalton Transactions 2007, (42), 4829- 4844; (d) Douvris, C.; Ozerov, O. V., Hydrodefluorination of Perfluoroalkyl Groups Using Silylium-Carborane Catalysts. Science 2008, 321 (5893), 1188-1190; (e) Geis, V.; Guttsche, K.; Knapp, C.; Scherer, H.; Uzun, R., Synthesis and characterization of synthetically useful salts of the weakly-coordinating dianion [B12Cl12]2. Dalton Transactions 2009, (15), 2687-2694; (f) Gu, W.; Haneline, M. R.; Douvris, C.; Ozerov,

104

O. V., Carbon−Carbon Coupling of C(sp3)−F Bonds Using Alumenium Catalysis. Journal of the American Chemical Society 2009, 131 (31), 11203-11212; (g) Knapp, C.; Schulz, C., How to overcome Coulomb explosions in labile dications by using the [B12Cl12]2- dianion. Chemical Communications 2009, (33), 4991-4993; (h) Bolli, C.; Derendorf, J.; Keßler, M.; Knapp, C.; Scherer, H.; Schulz, C.; Warneke, J., Synthese, Kristallstruktur und Reaktivität des starken Methylierungsmittels Me2B12Cl12. Angewandte Chemie 2010, 122 (20), 3616-3619; (i) Bolli, C.; Derendorf, J.; Keßler, M.; Knapp, C.; Scherer, H.; Schulz, C.; Warneke, J., Synthesis, Crystal Structure, and Reactivity of the Strong Methylating Agent Me2B12Cl12. Angewandte Chemie International Edition 2010, 49 (20), 3536-3538; (j) Derendorf, J.; Ke; Knapp, C.; Ruhle, M.; Schulz, C., Alkali metal-sulfur dioxide complexes stabilized by halogenated closo- dodecaborate anions. Dalton Transactions 2010, 39 (37), 8671-8678; (k) Douvris, C.; Nagaraja, C. M.; Chen, C.-H.; Foxman, B. M.; Ozerov, O. V., Hydrodefluorination and Other Hydrodehalogenation of Aliphatic Carbon−Halogen Bonds Using Silylium Catalysis. Journal of the American Chemical Society 2010, 132 (13), 4946-4953; (l) Duttwyler, S.; Douvris, C.; Fackler, N. L. P.; Tham, F. S.; Reed, C. A.; Baldridge, K. K.;

Incipient Phenyl Cation Reactivity. Angewandte Chemie 2010, 122 (41), 7681-7684; (m) Duttwyler, S.; Douvris, C.; Fackler, N. L. P.; Tham, F. S.; Reed, C. A.; Baldridge, K. K.;

Incipient Phenyl Cation Reactivity. Angewandte Chemie International Edition 2010, 49 (41), 7519-7522; (n) Finze, M.; Sprenger, J. A. P.; Schaack, B. B., Salts of the 1- cyanocarba-closo-dodecaborate anions [1-NC-closo-1-CB11X11]- (X = H, F, Cl, Br, I). Dalton Transactions 2010, 39 (10), 2708-2716; (o) Kessler, M.; Knapp, C.; Sagawe, V.; Scherer, H.; Uzun, R., Synthesis, Characterization, and Crystal Structures of Silylium Compounds of the Weakly Coordinating Dianion [B12Cl12]2−. Inorganic Chemistry 2010, 49 (11), 5223-5230; (p) Nava, M. J.; Reed, C. A., High Yield C-Derivatization of Weakly Coordinating Carborane Anions. Inorganic Chemistry 2010, 49 (11), 4726-4728; (q) Valášek, M.; Štursa, J.; Pohl, R.; Michl, J., Lithium Salts of [1,12-Dialkyl- CB11Me10]− Anions. Inorganic Chemistry 2010, 49 (22), 10255-10263; (r) Kessler, M.; Knapp, C.; Zogaj, A., Cationic Dialkyl Metal Compounds of Group 13 Elements (E = Al, Ga, In) Stabilized by the Weakly Coordinating Dianion [B12Cl12]2–. Organometallics 2011, 30 (14), 3786-3792; (s) Bolli, C.; Köchner, T.; Knapp, C., [NO][HCB11Cl11] – Synthesis, Characterization, Crystal Structure, and Reaction with P4. Zeitschrift für anorganische und allgemeine Chemie 2012, 638 (3-4), 559-564; (t) Ibad, M. F.; Langer, P.; Reiß, F.; Schulz, A.; Villinger, A., Catalytic Trimerization of Bis-silylated Diazomethane. Journal of the American Chemical Society 2012, 134 (42), 17757-17768; (u) Ramirez-Contreras, R.; Ozerov, O. V., Convenient C-alkylation of the [HCB11Cl11]- carborane anion. Dalton Transactions 2012, 41 (26), 7842-7844.

105

Conclusion

Since the discovery of carboranes in the 1950’s and 1960’s, chemists have been exploring the inherent properties of boron clusters, initially as a means to utilize the stockpiles of material left behind from the cold war era. After studying their structure, bonding, and unique magnetic and chemical properties, chemists were keen to take advantage of the burgeoning field of boron cluster chemistry. Their unique properties quickly sparked interest for applications in catalysis, organomimetics, energy storage devices, potential cancer treatment, and specifically for the anionic carboranes, exploitation of their weakly coordinating nature for the stabilization of reactive ions. The focus of my research was to further explore the various 12, 11, and 10 vertex carborane’s inherent properties as novel ligand R groups. To use anionic carboranes as ligand R groups, we needed to determine a few key properties of the clusters. Most importantly, was it susceptible to decomposition as had been shown with the neutral C2B10H10 substituent?

Second, do the anionic carboranes behave as electron donating substituents as expected, or electron withdrawing which had long been known for the neutral C2B10H10? Third, how might the charge affect the reactivity of the metal center? Lastly, when substituting the carborane as a ligand substituent, did it promote unique reactivity and allow for the preparation of a competent catalyst?

The use of anionic carboranes as ligand R groups was a novel approach for the development of transition metal catalysts and was dependent on the greater stability of the anionic carboranes compared to the already studied ineffective neutral carborane. Proof of this stability was presented in the development of two isoelectronic Iridium COD

106

complexes. Through the isolation of an Iridium I phosphine complex bearing the anionic

- carborane as a ligand substituent, we showed that the CB11H11 is inherently more stable and less susceptible to intramolecular B-H oxidative addition as a ligand substituent compared to its neutral cousin C2B10H10. This observation was important as it provides

- strong evidence that the CB11H11 is a suitable ligand substituent for catalysts that mediate

- a variety of catalytic processes. In addition to the inherent stability of the CB11H11 as a ligand substituent, the inductive properties were studied and it was found that the 12 vertex

- CB11H11 is in fact a strong donor and comparable to an isopropyl group, in contrast to the

- neutral 12 vertex C2B10H10. In this same work, it was noted that the 10 vertex CB9H9 was also a very strong donor, in fact, much more than an isopropyl group. Another question that needed to be answered was how the charge might affect the reactivity of a metal center with a ligand bearing a carborane anion. Through the preparation of two isoelectronic

- Pd(0) complexes we demonstrated that ligands appended with CB11H11 induced dramatically different reactivity than the analogous Pd(0) complex with the neutral

C2B10H10. This was directly related to the charge of the cluster and the stabilization of the transition state during oxidative addition via electrostatic interaction. In addition to the study of the carboranes magnetic and chemical properties, our goal of further developing unique transition metal catalysts was achieved with the preparation of the previously mentioned Pd(0) complex, which was a competent catalyst for Kumada cross-coupling. In

- addition, the preparation of a PdAllyl complex, bearing the CB11Cl11 as a ligand substituent, for the polymerization of norbornene showed that the halogenated carborane is indeed weakly coordinating enough to allow polymerization to occur with relatively low

107

polydispersity, even though the anion is buried in the coordination sphere of the metal.

Furthermore, its thermal stability was much higher than typical palladium cations and can be attributed to the carborane anions ability to stabilize reactive metal centers. In addition

- to stabilizing reactive metal centers, the CB11Cl11 anion was found to stabilize the formation of the first triazole radical. Facile synthesis of the precursor paves the way for the development of additional stable radical heterocycles including the analogous

- compound with the CB9Cl9 anion. The conclusion of my dissertation research was with the development of a novel unsymmetrical nido-imidazolium bearing a mesityl R-group on one nitrogen in the heterocycle and a nido-carborane on the other. Although a transition metal complex was not obtained with the novel nido-NHC-dicarbollide, evidence suggests that sequential formation of the dicarbollide, followed by the NHC will allow for the possible formation of unique heterobimetallic transition metal complexes, potentially useful for cooperative bimetallic catalysis. Furthermore, the synthetic route for the preparation of the nido-carborane amine will provide a useful precursor to a library of synthetically useful compounds. The work presented in this dissertation provides insight for the development of transition metal complexes and paves the way for the development of other systems that will be effective for cross-coupling, the polymerization of unstrained olefins, as well as a variety of other catalytic pathways. The versatility of carboranes and their potential impact on all fields of chemistry, provides great opportunity and excitement.

Although they are exciting compounds, their anomalous characteristics create a formidable complexity to this particular field of cluster chemistry.

108