University of Florida Thesis Or Dissertation Formatting

DESIGN, SYNTHESIS, AND DEPOSITION FROM SINGLE SOURCE PRECURSORS

NATHANIEL ELBA RICHEY

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2018

To grokking in fullness

ACKNOWLEDGMENTS

First and foremost, I would like to thank my advisor, Professor McElwee-White for her guidance and support throughout graduate school. Her guidance has helped me grow into an independent and critical thinker and her rigorous edits of my documents has greatly improved my scientific writing. I will greatly miss the group potlucks at her home. I would also like to thank the rest of my committee for their time and feedback on my projects.

I would also like to thank my external collaborators and others who I have worked with. I would like to particularly Professor David Wei and his previous student Dr. Jingjing Qiu for their hard work on the SPMCSD of copper. I would also like to thank Dr. Eric Lambers for training me on X-ray photoelectron spectroscopy and assisting me with interpreting the data from it. I would like to thank Professor Daniel Talham and his student Sinha Khushboo for assistance in using their atomic force microscope. Finally, I would like to thank Katrina Pangilinan at the

University of Tennessee Knoxville for performing the thermogravimetric analysis coupled to mass spectrometry on several of my compounds.

I must also thank my many excellent group members, both past and present, for their continuous support. I would like to thank Dr. Kelsea Johnson for training me when I first joined in the summer of 2013. I would also like to acknowledge Alina Kilbert, who worked with over a summer while she was still in high school and was my first mentee as a graduate student. I would also like to thank Kevin Hamlin, Jessica Tami, and Chandler Haines for being awesome undergraduates. Duane Bock, Michelle Nolan, and Nathan Ou have greatly supported my research by helping to characterize my deposits by SEM, XRD, and Raman spectroscopy, respectively. I was privileged to work with Michelle Nolan and Alex Touchton on the deposition of tungsten carbonitride from a precursor they synthesized. I would also like to acknowledge

Zahra Ali who I worked with for six months while she was visiting from Pakistan, whose high

energy was an invigorating change of pace. Nathan Ou has been a great help in modifying our inhouse AACVD reactor into a conventional CVD reactor. Finally, I would like to thank Will

Carden, Chris Brewer, and Scot Matsuda for their stimulating discussions in lab and for simply being awesome lab mates.

I must also thank all of the other friends I have made while at UF, both those inside and outside of the chemistry department. They have provided a constant route to relaxing and enjoying life outside of the lab. Last, but not least, I thank my family who have been, and continue to be, the most supportive group of people.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 10

LIST OF ABBREVIATIONS ...... 13

ABSTRACT ...... 14

CHAPTER

1 INTRODUCTION ...... 16

Nanoscale Fabrication ...... 16 Precursors for Chemical Depositions ...... 19 Chemical Vapor Deposition ...... 21 Surface Plasmon Mediated Chemical Solution Deposition ...... 25 Importance of Precursor Design ...... 32

2 SURFACE PLASMON MEDIATED CHEMICAL SOLUTION DEPOSITION OF COPPER ...... 33

Copper Uses and Deposition ...... 33 SPMCSD from Adduct Stabilized Copper β-diketonates ...... 36 SPMCSD from Copper Borohydrides ...... 43 Summary of Copper Deposition from SPMCSD ...... 57

3 AEROSOL ASSISTED CHEMICAL VAPOR DEPOSITION OF TUNGSTEN DISULFIDE ...... 59

Transition Metal Dichalcogenides ...... 59 Tungsten Dithiocarbamate Precursor ...... 63 Dithiolene Precursors ...... 70 Tungsten Disulfide Monolayers ...... 76 Summary of WS2 Deposition...... 81

4 INTRAGROUP COLLABORATIONS ...... 83

Preface ...... 83 N,N-Disubstituted-N-acylthioureas for the Deposition of Metal Sulfides ...... 83 Deposition of WCxNy using Aerosol Assisted Chemical Vapor Deposition ...... 94 Chemical Vapor Deposition of WOx from Volatile Single Source Precursors ...... 101 Summary ...... 106

5 EXPERIMENTAL PROCEDURES ...... 107

General Considerations ...... 107 Copper Precursors ...... 108 Tungsten Sulfide Precursors ...... 112 N,N-disubstituted-Nʹ-acylthiourea Ligands and Precursors ...... 113 Tungsten Carbonitride Precursor ...... 116 Deposition Procedures ...... 117

APPENDIX

A COMPOUND STRUCTURES, FORMULAS, AND NUMBERS ...... 120

B FULL X-RAY CRYSTAL STRUCTURES ...... 123

C FULL X-RAY TABLES...... 125

X-ray Tables of 20 ...... 125 X-ray Tables of 22 ...... 131 X-ray Tables of 23 ...... 143 X-ray Tables of 24·NCCH3 ...... 155

D FULL TANDEM MASS SPECTRUM ...... 166

LIST OF REFERENCES ...... 167

BIOGRAPHICAL SKETCH ...... 183

LIST OF TABLES

Table page

2-1 Solid-state and solution phase decomposition temperatures of the synthesized copper borohydrides ...... 51

4-1 Fragments from the MS-MS of 20 ...... 88

4-2 Selected X-ray crystallographic data for 20, 22, 23, 24 ...... 90

4-3 Selected bond distances (Å) for 20, 22, 23, 24 ...... 91

4-4 Selected bond angles (degrees) for 20, 22, 23, 24 ...... 91

4-5 Elemental composition of the deposits from 29 as determined by XPS ...... 105

C-1 Crystal data and structure refinement for 20...... 125

C-2 Atomic coordinates (x 104) and equivalent isotropic displacement parameters (Å2x 103) for 20 ...... 126

C-3 Bond lengths [Å] and angles [°] for 20 ...... 127

C-4 Anisotropic displacement parameters (Å2x103) for 20 ...... 129

C-5 Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x 103) for 20 ...... 130

C-6 Crystal data and structure refinement for 22...... 131

C-7 Atomic coordinates (x104) and equivalent isotropic displacement parameters (Å2x 103) for 22 ...... 132

C-8 Bond lengths [Å] and angles [°] for 22 ...... 134

C-9 Anisotropic displacement parameters (Å2x 103) for 22 ...... 139

C-10 Hydrogen coordinates (x104) and isotropic displacement parameters (Å2x103) for 22 ...... 141

C-11 Crystal data and structure refinement for 23...... 143

4 C-12 Atomic coordinates (x10 ) and equivalent isotropic displacement parameters (Å2x 103) for 23 ...... 144

C-13 Bond lengths [Å] and angles [°] for 23 ...... 146

C-14 Anisotropic displacement parameters (Å2x 103) for 23 ...... 151

4 C-15 Hydrogen coordinates (x10 ) and isotropic displacement parameters (Å2x 103) for 23 .....153

C-16 Crystal data and structure refinement for 24·NCCH3 ...... 155

4 C-17 Atomic coordinates (x10 ) and equivalent isotropic displacement parameters (Å2x 103) for 24·NCCH3 ...... 156

C-18 Bond lengths [Å] and angles [°] for 24·NCCH3 ...... 158

2 3 C-19 Anisotropic displacement parameters (Å x 10 ) for 24·NCCH3 ...... 162

4 C-20 Hydrogen coordinates (x10 ) and isotropic displacement parameters (Å2x 103) for 24·NCCH3 ...... 164

LIST OF FIGURES

Figure page

1-1 Cartoon representation of top-down and bottom-up fabrication ...... 16

1-2 Examples of physical vapor deposition techniques ...... 18

1-3 Examples of dual source precursor deposition, coreactant deposition and single source precursor depositions ...... 20

1-4 General scheme of precursor transport and reaction process in CVD ...... 22

1-5 Aerosol-assisted CVD ...... 25

1-6 Surface plasmon resonance ...... 27

1-7 Raman spectra of the AgFON substrate before and after soybean oil polymerization ...... 27

1-8 The SPMSCD process ...... 29

1-9 Selected SPR effect in bowtie nanostructures ...... 29

1-10 Reaction coordinate diagram for the decomposition of PPh3AuCH3 to AuNPs on the AgFON during SPMCSD ...... 31

2-1 Selected copper(II) bis-(β-diketonate) and bis-(β-diketoesterate) precursors ...... 34

2-2 Structures of commercialized copper β-diketonates ...... 35

2-3 Proposed CVD mechanism for the decomposition of LCu(hfac) precursors ...... 36

2-4 Predicted mechanism for the SPMCSD of copper from adduct stabilized copper β- diketonates ...... 37

2-5 Synthesis of adduct stabilized copper β-diketonates ...... 38

2-6 SEM of the deposits from 2 ...... 39

2-7 Thermogravimetric analysis of compounds 3 and 4 ...... 40

2-8 UV-Vis spectra over time for compounds 3 and 4 in benzene solution ...... 41

2-9 SEM micrographs of the deposits from the SPMCSD of 3 ...... 42

2-10 XPS spectra of the SPMCSD deposits after 70 minutes of irradiation ...... 42

2-11 Synthesis of 5 and 6 by the Gysling method ...... 44

2-12 Structures of the copper complexes 5-12 ...... 45

2-13 IR spectra of compounds 5, 6, and 7 ...... 46

2-14 IR spectrum of compound 8 ...... 47

2-15 IR spectra of compounds 9, 10, and 11 ...... 48

2-16 IR spectra of compound 12 and the proposed polymeric [P(o-tolyl)3Cu(NCBH3)]n ...... 49

2-17 TGA traces of compounds 5, 7, 8, 10, and 11 ...... 50

2-18 Initial SPMCSD results from the deposition of 5 ...... 52

2-19 SPMCSD results from deposition of 5 with rigorous exclusion of air and water ...... 54

2-20 Morphology control of the copper deposits ...... 56

2-21 Surface Enhanced Raman Spectroscopy of 4-MBA using AgFON/Cu ...... 57

3-1 Crystal structures of the layered materials graphite and a generic transition metal dichalcogenide ...... 60

3-2 Band structure of bulk and monolayer WS2, black arrow indicates the fundamental band gap ...... 61

3-3 Synthesis of WS(S2)(S2CNEt2)2 ...... 63

3-4 TGA of 13 ...... 64

3-5 TGA-MS data from 13 ...... 65

3-6 Isothermal TGA and pyrolysis of WS(S2)(S2CNEt2)2 ...... 67

3-7 SEM images of the WS2 deposits grown at 350, 400, 450, and 500 °C ...... 68

3-8 W 4f and S2p regions of the XPS spectra from the WS2 deposits ...... 69

3-9 Characterization of the WS2 deposits ...... 70

3-10 Synthesis of compound 14 ...... 71

3-11 The TGA and TGA-MS traces of 14 ...... 73

3-12 1H NMR spectrum of the byproducts from the thermolysis of 14...... 74

3-13 Two possible decomposition pathways for 14 ...... 75

3-14 Raman spectra of the deposits from 14 ...... 76

3-15 Powder vaporization variants used for the deposition of TMD monolayers ...... 77

3-16 Representative diagram of the adapted powder vaporization deposition ...... 78

3-17 Optical microscopy images from the adapted powder vaporization method on various substrates from 13 ...... 79

3-18 Optical microscopy images for powder vaporization deposits from 13 using form gas ...... 80

3-19 AFM images from powder vaporization deposition of 13 under form gas ...... 82

4-1 Synthesis of the N,N-disubstituted-Nʹ-acylthioureas and their nickel compounds ...... 86

4-2 Thermal properties of 18, 19, 20, 21 ...... 87

4-3 X-ray crystallographic structures of 20, 22, 23, 24 ...... 89

4-4 Representative example of the dihedral angle calculation ...... 92

4-5 Thermal properties of 20, 22, 23, 24 ...... 92

4-6 Deposition results from 20, 23, 24 ...... 93

4-7 Synthesis of the tungsten guanidinato complexes ...... 96

4-8 Detected byproducts from the thermolysis of 25...... 97

4-9 The TGA plot of 25 ...... 97

4-10 Low- and high-resolution micrographs from the deposition of WCxNy from 25 ...... 98

4-11 XPS of the deposits from 25 ...... 101

4-12 Examples of some fluorinated alkoxide precursors with their reported sublimation temperatures and pressures ...... 102

4-13 TGA plot of 29 ...... 103

4-14 Full XPS spectra of the deposits from 29 ...... 104

4-15 High resolution XPS of the W 4f and O 1s peaks of the deposits from 29 ...... 106

B-1 Full X-ray crystal structure of 22 ...... 123

D-1 Fragments observed in the tandem mass spectrometry experiments of 20 ...... 166

LIST OF ABBREVIATIONS

4-MBA 4-Mercaptobenzoic acid

AACVD Aerosol assisted chemical vapor deposition

AgFON Silver film on nanosphere

CVD Chemical vapor deposition

DPA Diphenylacetylene

DTA Di-(p-tolyl)acetylene

DTG Derivative of thermogravimetric analysis

EDX Energy dispersive X-ray spectroscopy hfac Hexafluoroacetylacetonate

IR Infrared

MHY 2-methyl-1-hexen-3-yne

MSMS Tandem mass spectrometry

SCCM Standard cubic centimeters per minute

SEM Scanning electron microscopy

SERS Surface enhanced Raman spectroscopy

SPMCSD Surface plasmon mediated chemical solution deposition

SPR Surface plasmon resonance

TGA Thermogravimetric analysis

TGA-MS Thermogravimetric analysis coupled to mass spectrometry

TMD Transition metal dichalcogenide

VTMS Vinyltrimethylsilane

XPS X-ray photoelectron spectroscopy

XRD X-ray diffraction

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

DESIGN, SYNTHESIS, AND DEPOSITION FROM SINGLE SOURCE PRECURSORS

Nathaniel Elba Richey

May 2018

Chair: Lisa McElwee-White Major: Chemistry

Chemical vapor deposition is a powerful tool for the deposition of thin films. Single source precursors can be used in chemical vapor deposition to ideally achieve lower temperature depositions and better control the stoichiometry of the deposits, through the choice of ligands used. However, many potential single source precursors lack the volatility to be used in conventional chemical vapor deposition. Solution phase and aerosol assisted deposition techniques have been used to overcome the obstacle of volatility in order to make use of the otherwise latent precursors.

Surface plasmon mediated chemical solution deposition has been used to deposit copper nanoparticles and thin films. Precursors containing hexafluoroacetylacetonate were found to deposit copper with some fluorine and carbon contamination from the ligand.

Bis(triphenylphosphine)copper borohydride was found to deposit pure copper nanoparticles and thin films when deposited under an inert atmosphere. The morphology of these deposits could readily be controlled through the deposition parameters. Derivatives with an additional coordinating ligand did not undergo any decomposition, presumably due to their greater solution phase stability.

Tungsten disulfide is an inorganic, semiconducting, layered material and has been previously deposited through a wide variety of techniques. However, no single source precursor had previously been reported for the deposition of tungsten disulfide thin films. The compound

WS(S2)(S2CNEt2)2 was found to be suitable for the deposition of WS2 using aerosol assisted chemical vapor deposition. Thermogravimetric analysis showed that this precursor decomposed through a three-step process. Vertically aligned platelets of WS2 were deposited at temperatures above 350 °C. Several attempts to grow monolayers of WS2 from WS(S2)(S2CNEt2)2 resulted in spherical particles with heights much larger than that of typical WS2 monolayers. A dithiolene complex, W(CO)2(S2C4H6)2, has also been investigated as a single source precursor for the deposition of WS2. From thermogravimetric analysis coupled to mass spectrometry and thermolysis data, it appears that this precursor undergoes two simultaneous decomposition pathways which both result in WS2.

Deposition has been carried out with a variety of different precursors from collaborators within the McElwee-White group. Several nickel complexes of N,N-disubstituted-Nʹ- acylthioureas were investigated and it was found that the decomposition temperature was largely dependent on the substituents used. N,N-diisopropyl-Nʹ-cinnamoylthiourea complexes were used to deposit metal sulfides of nickel, cobalt, and zinc at 350 °C. The guanidinato complex,

i WN(NMe2)[(N Pr)2C(NMe2)]2, was used as a single source precursor for the deposition of

WCxNy at deposition temperatures above 200 °C. The volatile tungsten complex,

WO(OC(CF3)2CH3)3(hfac), has been investigated as a precursor for both AACVD and CVD of

WOx. The carbon contamination of both deposits was removed by ion sputtering, indicating no incorporation of carbon from the precursor or solvent in AACVD.

CHAPTER 1 INTRODUCTION

Nanoscale Fabrication

Over the last half century, we have seen our technological advances lead to smaller and smaller devices. This development is, in part, due to the creation of the integrated circuit which is composed of many transistors. Smaller, faster, and more efficient devices based on the integrated circuit are continuously developed by decreasing the size of the transistors, allowing for a greater density of transistors on an integrated circuit. In 1965, Gordon Moore of Intel predicted that the number of transistors on an integrated circuit would double every one to two years due to the decreasing size of the transistors, until such a time that transistors reach the single atom scale.1, 2 So far, this trend has held true and is commonly known as “Moore’s Law.”

Should this trend in minimization of transistor size continue, it is expected to reach its natural end in the late 2020s as it approaches the atomic limit.3 Attempts to keep up with Moore’s Law has led to some of the greatest advances in fabrication methods for materials on the nanoscale.

For instance, integrated circuits are currently made on a 5 nm node scale, but decreasing this node to smaller sizes becomes exponentially more difficult and costly.4 Two general approaches are typically considered in the fabrication of such small-scale components and are classified as

“top-down” or “bottom-up” (Figure 1-1).5

Figure 1-1. Cartoon representation of top-down and bottom-up fabrication.

Top-down fabrication techniques involve taking a large piece of the bulk material and removing some of the material until the desired size and shape is obtained. Photolithography,6 exfoliation,7 and etching8 are several very common top-down approaches used in materials chemistry. However, two major challenges currently face top-down fabrication methods.

Firstly, the size regime of the features that can be made with this fabrication technique can be limiting. For instance, photolithography will be inherently limited by the wavelength of the photons used, which is typically limited to features of tens or hundreds of nanometers.9 Other research areas, such as atomic layer etching, may provide an alternative route for decreasing the size regime available to top-down methods, but the scope of this method is still quite restricted.10

The second challenge is the inherent wastefulness of top-down methods, as much of the bulk material is discarded during processing. While unavoidable, this waste can be minimized by recovering the removed materials or by beginning with a bulk material which is very close in size to the desired final material.

Alternatively, bottom-up fabrication techniques build up to the final material through the use of smaller parts of that material. This assembly is most often done with single atoms or small clusters, called precursors, of the material which can allow for controlled growth at the atomic size regime. A wide variety of deposition and growth techniques are commonly used for the bottom-up fabrication of a plethora of materials. The mechanism behind these techniques can broadly be separated into two classes: physical or chemical.11, 12 Physical deposition techniques do not undergo a chemical reaction and deposit from single atoms of the desired material. Of course, single atoms are often quite reactive, so physical deposition techniques are often limited to vapor phase transport of the precursor in an ultrahigh vacuum environment. These restrictions help to prevent condensation of the desired material prior to its interaction with the substrate.

Because of this limit to the vapor phase transport, these techniques are generally considered as physical vapor deposition (PVD) techniques. Common techniques within PVD are evaporative or sublimation deposition, sputtering,13 and molecular beam epitaxy (

Figure 1-2).14 These techniques are generally suited for the deposition of high-purity thin films, but, due to the reactive nature of the transported species, are poor candidates for the deposition onto a non-flat substrate.

Figure 1-2. Examples of physical vapor deposition techniques.

On the other hand, chemical deposition and growth techniques are broadly defined as requiring a chemical reaction to occur which promotes the deposition or growth of the desired material.12 In general, these techniques use a precursor molecule which contains one or more element of the desired material stabilized by organic or inorganic ligands. An external source of energy is then used to promote the chemical decomposition of the precursor(s) to the final material. This external energy is most often heat, but photoassisted and plasma enhanced processes are also known. Because these methods require a chemical reaction to occur, chemical deposition can lend itself readily to both solution and vapor phase deposition. Common chemical deposition techniques include chemical vapor deposition (CVD),12 atomic layer deposition,15 chemical bath deposition,16 solvothermal nanoparticle growth,17 and chemical beam

epitaxy.18 From this variety of different techniques, a wide breadth of materials can be grown or deposited onto numerous substrates for use in a diverse range of applications.

Precursors for Chemical Depositions

One of the most important aspects of chemical depositions is in the choice of suitable precursor(s). In general precursors should be stable enough for handling and transport, but reactive enough for deposition and growth of the desired material at relatively low temperatures

(100-600 °C). Chemical depositions will often use more than one precursor or a precursor with a coreactant to deposit the desired material (Figure 1-3). For example, many compound semiconductor films are deposited from a dual source precursor route in which a mixture of a volatile metal salt or carbonyl (such as GaCl3, TiCl4, or W(CO)6) and a volatile nonmetal precursor (such as H2S, diethylselenide, NH3, or tributylphosphine) are used as the precursors for the two parts of the deposited semiconductor.19-21 Coreactants are commonly used to assist in the deposition of pure films from metal-organic precursors by helping to remove the organic fragments of the precursor.22 Common coreactants are mild reductants, such as hydrogen gas or ethanol, or oxidants, such as oxygen gas or ozone, which can react with impurities left behind by the decomposition of the metalorganic precursor. Often times, a secondary precursor, such as water or oxygen, will also be called a coreactant, but for the sake of clarity, a coreactant will only be described herein as a necessary component for deposition that does not incorporate into the final material. Occasionally, both coreactants and multiple precursors are needed to deposit a desired material.23

While these coreactant systems have been used to successfully deposit a wide range of films, there are several drawbacks to their uses, particularly in vapor phase deposition. A major drawback is the toxicity of the precursors or of the byproducts generated. Many metal carbonyls, which release several equivalents of carbon monoxide when they decompose, are used in CVD.24

Arsine (AsH3) is a common precursor for III-V films and has an LC50 (4 hours) of 5-40 ppm.

Also, many metal halide precursors that are used with a coreactant form the hydrogen halide gas as a byproduct, which is not only toxic but potentially damaging to the substrate and reactor equipment as well.12 There is also potential for gas phase reactions to occur before the precursors reach the substrate, which can result in nonconformal deposits or a clog in the transfer lines. One potential way around these issues is to use single source precursors, in which no secondary precursor or coreactant is needed.25

Figure 1-3. Examples of dual source precursor deposition, coreactant deposition and single source precursor depositions.

A single source precursor contains all of the desired elements of the film in one molecule and does not require another precursor or coreactant to decompose to the final material.26, 27

Most single source precursors are used in the deposition of semiconductors such as metal pnictogenides or chalcogenides in which the final anion of the film is often directly bound to the metal in the precursor as a ligand. Several metallic films have also been deposited from single source precursors, although this is case is rarer as it requires all of the ligands to be removed during decomposition with no additional coreactant.28 Single source precursors can also provide

more consistent stoichiometry in the final deposits, as the decomposition is inherent to the precursor and not dependent on the ratio of two or more precursors.

Each variation of chemical deposition differs slightly from one another, therefore a precursor that works well for one technique is not guaranteed to work well for another. For instance, vapor phase reactions typically require volatile compounds whereas solution based reactions do not. By determining the design parameters for the precursors of a given technique, we are able to better predict and synthesize precursors that result in lower deposition temperature without losing purity of the final material.

Chemical Vapor Deposition

As a technique, CVD is broadly defined as any chemical deposition method that involves vapor phase transport of the precursor(s) to a substrate for decomposition. There are a wide variety of CVD methods such as metal-organic (MO)CVD,29 plasma-enhanced (PE)CVD,30 photo-assisted (PA)CVD,31 atmospheric pressure (AP)CVD,32 and more. Many other techniques, such as atomic layer deposition or chemical beam epitaxy, can also be considered as specialized variants of CVD. All of these variants closely follow the same general mechanism for growth. In general, CVD is achieved by flowing a volatile chemical precursor in an inert carrier gas, such as argon or nitrogen, to a hot substrate where decomposition into the desired material is then thermally initiated.33 One of the most impressive features of CVD is its ability to promote conformal growth of a material on a nonuniform substrate.34 This control is achieved through careful control of the temperature and pressure of the CVD system to maximize surface selective growth.

Mechanism of CVD

A generic mechanism for CVD is shown in Figure 1-4. Step I is the adsorption and desorption of molecular precursors onto the substrate surface. This process can then be followed

by step II, in which the hot substrate begins to decompose the adsorbed precursor into intermediate fragments and volatile byproducts. These fragments form surface adsorbed, metastable intermediates or readily desorb into the gas phase. This surface mediated decomposition pathway is commonly referred to as heterogeneous decomposition. In step III, the precursor undergoes a partial gas phase decomposition, called homogeneous decomposition, before adsorbing on the surface where it decomposes to the similar byproducts and intermediates from step II. Step III is occasionally needed to promote growth when there is poor adsorption of the starting precursor. In step IV it is possible for the metastable intermediates to adsorb or desorb onto the substrate. Finally, in step V the adsorbed intermediate decomposes to the final material and is able to nucleate and grow on the surface to produce the desired films.

III Precursor

Partially decomposed intermediate I IV Volatile byproduct(s) II V Deposited material Substrate Surface

Figure 1-4. General scheme of precursor transport and reaction process in CVD.

Control of the temperature and pressure of the CVD is system is critical because these conditions can influence all of the above steps. In steps I and IV, low pressure will significantly increase the relative amount of gas phase species with respect to the adsorbed species. As a result, films grown at lower pressures will usually be more pure, due to better desorption of byproducts, but will grow at significantly hampered rates. Temperature must be carefully considered in both the gas phase and at the substrate surface. High substrate temperatures can help in the decomposition steps II and V, generally resulting in faster growth rates and higher

surface mobility of the final material, which can lead to more crystalline deposits. However, excessive heating of the substrate can also promote desorption of the precursor and intermediates, resulting in a decrease of growth rate, and can also cause thermal decomposition of normally stable byproducts, causing higher impurity levels. High gas phase temperatures will promote homogeneous decomposition in step III. This phenomenon is occasionally beneficial in deposition of certain materials, such as silicon carbide from methyltrichlorosilane,35 when the parent precursor needs to be activated through partial decomposition before it can readily adsorb onto a substrate and undergo final heterogeneous decomposition. However, full gas phase decomposition could result in transfer line clogging and ruined conformality of the film. The difference in the substrate and gas phase temperatures can be partially controlled by reactor type.

Cold-wall reactors are only locally heated as the substrate to minimize the temperature in the gas phase, while hot-walled reactors, such as tube furnaces, are heated throughout so that the gas phase temperature is nearly the same as the substrate.36 Temperature and pressure can be used to create a careful balance precursor adsorption, desorption, and decomposition, giving CVD its well-known surface selected growth.

Aerosol assisted chemical vapor deposition

One of the biggest drawbacks of conventional CVD is the requirement for the precursors to be volatile. This prerequisite limits the scope of available compounds and restricts potential ligand choices, potentially eliminating ligands that undergo alternative decomposition pathways and lead to precursors with lower decomposition temperatures. To overcome this obstacle, several solution based CVD techniques have been implemented in which a precursor is dissolved prior to transport.37 Of these methods, aerosol-assisted (AA)CVD is one of the most well- known. In AACVD the precursor is dissolved in a suitable solvent and then nebulized into an aerosol, which is simply liquid droplets small enough to act as a gas. This solution which is

aerosolized carries with it the dissolved precursor molecule and transports it to the substrate in a similar manner to conventional CVD. Eventually, the aerosol solvent evaporates, leaving behind just the precursor which can undergo surface adsorption and decomposition as normal (Figure 1-

5). One slight difference from conventional CVD is that the desorption of the precursor in step I is hampered due to the precursor’s lack of volatility, which may hinder conformal growth on complex substrates.

Because of this modification, AACVD allows for a much wider range of potential precursors and most AACVD processes utilize single source precursors with ligand sets that make them nonvolatile and unusable for conventional CVD. Dual precursor set ups can still be used to achieve a variety of materials or doped semiconductors, so long as care is taken to ensure the precursors are both adequately soluble in the chosen solvent.38, 39 Typical solvents of choice for AACVD are stable and noninteracting organic solvents such as heptane, toluene, or tetrahydrofuran. However, a variation of AACVD, called Mist-CVD, utilizes water as both the solvent and as a coreactant to deposit metal oxide films.40, 41 Careful consideration must be given when using lower boiling point solvents, particularly if reduced pressure is used, as evaporation can compete with aerosol formation and result in poor transport of the precursor.

Precursors for AACVD must meet several requirements: 1) adequate solubility in a suitable solvent, 2) sufficient stability for transport to the substrate, 3) clean decomposition with no contamination from the ligands. The first criterion ensures adequate transfer of the precursor to the substrate. The second criterion ensures no decomposition occurs during transport to the surface, which leads to loss of materials and clogging of the transfer lines. This criterion must also be balanced with the benefit of low decomposition temperature. The final criterion is necessary for any good precursor but is particularly important for single source precursors which

often have a high organic content from their ligands. Additionally, an ideal precursor would be readily and safely synthesized and handled.

Figure 1-5. Aerosol-assisted CVD. (a) Modified mechanism of AACVD. (b) Simplified diagram of an AACVD reactor.

Surface Plasmon Mediated Chemical Solution Deposition

Another possibility to avoid the need for volatile precursors is to use chemical solution depositions.42 Two main types of chemical solution depositions exist.43 The first involves coating the substrate with a solution of the precursor which is then followed by post-deposition processing to achieve the desired film, such as the processes of spin and dip coating. The second involves submerging the substrate in the solution and then initiating growth while still submerged. This second process can be quite difficult to perform under traditional thermal

deposition practices as localized heating of the substrate becomes difficult in a solution. Even when localized heating is achieved, heat transfer from the substrate to the solution is oftentimes inevitable. A hot solution can lead to evaporation of the solvent and potentially lead to solution phase decomposition of the precursor which can be wasteful and disrupt the conformality of the deposit.

A new chemical solution deposition technique which is driven by visible light has recently been developed by the Wei group.44 This technique is called surface plasmon mediated chemical solution deposition (SPMCSD) and utilizes the inherent surface plasmon resonance

(SPR) which can be induced in metallic nanostructures. The SPR of a nanoparticle is the collective oscillation of the surface electrons which can be excited by certain wavelengths of light (Figure 1-6). This excitation causes an enhanced electromagnetic field around the metal nanostructure, which has proven to be useful for various spectroscopic and sensing techniques.45

The enhanced electromagnetic field also results in the generation of “hot” electrons, which are conduction electrons promoted to an excited state that release energy as heat upon relaxation.

This heat results in ‘hotspots’ at the nanoparticles’ surfaces which are often quickly quenched by the nearby environment.46

The SPMCSD experiment utilized a silver film on nanosphere (AgFON) as a substrate and visible light from a xenon lamp to generate the hot electrons. Silver was chosen as the metal since it has the most intense SPR effect in the visible region of the electromagnetic spectrum.

Studies on the AgFON show that it can reach at least 230 °C when irradiated by a xenon lamp with a power density of 2 W/cm2 and a 515 nm long pass filter. This temperature was determined experimentally through the thermal polymerization of soybean oil, which is known to occur at 230 °C in the absence of a catalyst.47 Polymerization was observed through the decrease

in the ratio of the vinyl C-H to the aliphatic C-H stretch in the Raman spectrum from 0.37 to 0.34 after irradiation of a AgFON substrate coated in soybean oil (Figure 1-7). The exact temperature of the substrate is not known. However, using a lower power setting for the xenon lamp did not result in polymerization. It was also determined that the bulk solution of benzene in these experiments did not rise above 40 °C after three hours of irradiation.

Figure 1-6. Surface plasmon resonance. (a) SPR excitation on a spherical nanoparticle. Reprinted with permission from reference 48. Copyright (2011), American Chemical Society. (a) sequence of events and approximate time scales. Reprinted with permission from reference 49. Copyright (2011), American Chemical Society

Figure 1-7. Raman spectra of the AgFON substrate before and after soybean oil polymerization. Reprinted with permission from reference 50. Copyright (2013), American Chemical Society.

This technique has been used to successfully deposit gold nanoparticles (AuNPs) from

50 PPh3AuCH3 (Figure 1-8). This precursor was shown to decompose during SPMCSD to free triphenylphosphine, ethane, and surface bound AuNPs. In these SPMCSD experiments the triphenylphosphine can be washed away with benzene after deposition and the gaseous ethane escapes to the headspace of the solution upon formation. Thus, the deposited AuNPs were found to be pure gold and free of stabilizing ligands. These AuNPs grew between 3-10 nm and were found to be self limiting in size. This self limitation could be due to the low thermal conductivity of small AuNPs,51 or from the decreased SPR effect of gold with visibile light which would result in less heating. This lack of thermal conductivity was also noted by irraditing a soybean coated AuNPs-AgFON sample with light and, unlike the bare AgFON, no polymerization of the soybean oil was observed. A control test was performed by submerging the AgFON substrate in a solution of PPh3AuCH3 in the dark and no deposition occurred, indicating that this was in fact an SPR mediated process.

While the deposition of AuNPs can be useful for surface enhanced Raman spectroscopy

(SERS) and possible catalysis, a more controlled and directed growth could be even more useful for a wider range of applications. By using a silver bowtie structure, in which the tips of two nanoprisms of silver are placed adjacent to one another, directed growth has been achieved.

Under irradiation with specific polarized lasers, the SPR enhancement occurs most intensely at the apex between the two prisms and growth of AuNPs occurs preferentially at this location

(Figure 1-9).52 This may hopefully lead to the formation of more complex nanostructures, such as nanowires, or selective deposition of nanoparticles through a movable SPR active tip, such as in tip enhanced Raman spectroscopy.53

Figure 1-8. The SPMSCD process. (a) Decomposition of the PPh3AuCH3 precursor. (b) Cartoon representation of the process. (c) AgFON before deposition of AuNPs, (d) AgFON after deposition of 10 nm AuNPs. Reprinted with permission from reference 50. Copyright (2013), American Chemical Society.

Figure 1-9. Selected SPR effect in bowtie nanostructures. (a) Calculated enhanced EM field. (b) SEM image of the directed deposition of AuNPs between two silver prisms by SPMCSD.

The mechanism of the SPMCSD experiment has been investigated with density functional theory (DFT) calculations.54 The original mechanism was proposed to go through a dissociation of the triphenylphosphine from surface-bound precursor followed by two methylgold fragments undergoing bond homolysis to form ethane and gold(0), similar to the previously predicted CVD mechanism of the same precursor.55 However, this mechanism was calculated to be spontaneous at room temperature by DFT and should have proceeded without any SPR heating of the AgFON, which disagreed with the dark control reaction performed previously. Instead, the benzene was calculated to form a passivating monolayer on the AgFON surface. This monolayer prevents the gold precursor from adsorbing onto the AgFON and decomposing through a surface attached pathway. This barrier forces a partial solution-phase decomposition pathway, in which of the gold precursor dissociates in solution to the free triphenylphosphine and the methylgold fragment (Figure 1-10). The methylgold fragment is then small enough to adsorb in the space between the benzenes which form the monolayer.

Multiple adsorbed methylgold fragments can then come together to form ethane and the AuNPs.

The phosphine dissociation step then becomes the rate limiting step and prevents spontaneous decomposition at room temperature. This mechanism can be considered approximately like a

CVD mechanism in which homogeneous decomposition must occur before the adsorption of active intermediates.

Precursors for SPMCSD must meet three requirements: 1) no light absorption in the range of the SPR absorption, 2) undergo a thermally decomposition in solution below the hot spot temperature of the AgFON, and 3) be soluble in a suitable solvent. The first requirement is to prevent competitive light absorption between the substrate and the precursor, which could lower the surface temperature of the substrate or cause photolytic decomposition of the

Figure 1-10. Reaction coordinate diagram for the decomposition of PPh3AuCH3 to AuNPs on the AgFON during SPMCSD. Reprinted with permission from reference 56. Copyright (2015), American Chemical Society. precursor. The exact SPR range is partially controlled through a long pass filter which blocks light of lower wavelengths. Secondly, a precursor with high thermal stability would result in no deposition. The previously reported hot spot temperatures reached at least 230 °C, but this temperature can be adjusted depending on the power density of the xenon lamp and on the long pass filter used. Finally, because this is a solution based method, the precursor must be soluble in a suitable organic solvent. Benzene was previously used as the solvent, as it has no absorption in the SPR range and is believed to help control the deposition of the AuNPs.54 Additionally, an ideal precursor should also decompose to a pure material, be safe to handle, and be readily synthesized.

Importance of Precursor Design

As briefly described above, precursors need to be carefully considered for each different chemical deposition technique. Oftentimes, new deposition techniques will start with the same precursors of older and more well-known techniques. However, problems often arise due to small, or sometimes large, differences in the mechanism of deposition. Even methods which utilize a similarly activated decomposition pathway can undergo slightly different mechanisms.

For instance, while both CVD and SPMCSD are thermally activated process, CVD undergoes a gas phase or surface bound mechanism while AACVD undergoes a partial solution phase fragmentation. This small difference in mechanism could greatly influence the difference in precursor design between the two techniques. While some design criteria of precursors can be predicted prior to any deposition, the more nuanced criteria are typically only reveled through failed depositions or theoretical calculations.

CHAPTER 2 SURFACE PLASMON MEDIATED CHEMICAL SOLUTION DEPOSITION OF COPPER*

Copper Uses and Deposition

Copper is a cheap and abundant metal and has one of the lowest resistances of any pure element, second only to silver. Because of this, thin films and nanoparticles of copper have been utilized in a wide variety of applications such as catalytic oxidation/reduction,57, 58 SERS,59 optoelectronics,60 and, since 1997, has been used almost exclusively as the interconnect material of integrated circuits.61 Several methods have been used to deposit pure copper films such as

CVD,62 sputtering,63 electroless plating,64 and electrolytic plating.65 Of these, CVD is one of the more common techniques since it provides conformal and controllable growth of a copper seed layer, which can then be further modified as needed.28, 61, 66 Nanoparticles of copper have been grown using a variety of solvothermal,67 microwave assisted,68 and sol-gel processes.69

However, these methods often require a coreactant, typically a reductant, and result in a ligand stabilized nanoparticle surface which can hinder its desired application. Due to the similar surface-mediated thermal decomposition pathways of CVD and SPMCSD, precursors for copper

CVD are good potential candidates for SPMCSD.

Copper Precursors for CVD

Precursors of copper can be broadly separated by the oxidation state of copper in the precursor complex.70-72 Many of the studied copper(II) precursors are based on copper(II) bis-(β- diketonates) and related structures (Figure 2-1).73 In this family of complexes, copper(II) bis-

(hexafluoroacetylacetonate) (Cu(hfac)2) is perhaps the most well studied, as its heavily

74 fluorinated ligand makes it readily volatile. Typical depositions using Cu(hfac)2 are performed

* SPMCSD experiments and characterization of the deposits were carried out by Jingjing Qiu and Professor David Wei at the University of Florida

at substrate temperatures above 250 °C when using hydrogen gas as a coreactant.75-77 Higher deposition temperatures are needed to obtain pure copper deposits when no coreactant is used.

However a deposition temperature that is too high results in C, O, and F contamination from decomposition of the ligand.78 Several other ligands have also been employed in copper(II) precursors, including β-diketiminates,79 carboxylates,80 and aminoalkoxides.81 Unfortunately, a d9 copper(II) ion exhibits a d-d transition often located in the visible region. While this transition energy changes slightly depending on the coordination around copper, nearly all copper(II) species exhibit absorption above 500 nm. This absorption of light could interfere with light absorption by the SPR active substrate, which is needed for substrate heating in SPMCSD.

Figure 2-1. Selected copper(II) bis-(β-diketonate) and bis-(β-diketoesterate) precursors.

Fortunately, many copper(I) precursors are also known for CVD and, as a d10 ion, exhibit no d-d transition and very rarely have absorption above 500 nm. Several different families of copper(I) precursors exist.73 However, the most utilized precursors are neutral complexes of

Cu(β-diketonate), which take on the general formula of LCu(β-diketonate), and are favored due to their often high volatility and low deposition temperatures.27, 73 A plethora of neutral ligands,

L, have been used and include alkenes,82 dialkenes,83 alkynes,84 phosphines,85 and isocyanides.86

Similar to the copper(II) precursors, the β-diketonate is often hfac which imparts both volatility and some stability to prevent rapid room temperature decomposition. Several volatile and thermally robust complexes with nonfluorinated β-diketonate ligands have also been reported in

recent years.87, 88 Two of the most well-known commercialized precursors for the CVD of copper are from this family of complexes and are (VTMS)Cu(hfac) (1), trade name Cupra- select®, and (MHY)Cu(hfac) (2), trade name Gigacopper® (Figure 2-2).89 These precursors are generally referred to as Lewis base stabilized copper β-diketonate complexes, which ignores the contribution of the π-acidity of the neutral ligand since copper is often considered a fairly poor π- donor.90, 91 However, π-donation is observed in several of these compounds and is perhaps most notable when L = alkyne through the distortion of the C-C≡C angle away from 180°.92, 93

Because of this π-donation from the copper, this family of precursors will simply be referred to as copper β-diketonates.

Figure 2-2. Structures of commercialized copper β-diketonates.

In general, copper(I) precursors are promising for low temperature CVD due to their ability to undergo low temperature disproportionation into pure copper(0) and copper(II).94 This disproportionation reaction allows for very low temperature deposition without the need for a coreactant, such as the hydrogen gas used for deposition from Cu(hfac)2. Typically, deposition temperatures of copper from 1 are between 100-250 °C.95 However, this readily accessed decomposition pathway makes many of these precursors, including 1, mildly unstable at room temperature due to the weak coordination of the ligand.96, 97 The mechanism of the deposition from these copper β-diketonates has been thoroughly investigated.82, 98, 99 This mechanism is often heterogeneous, starting with adsorption of the precursor followed by surface assisted

dissociation of the neutral ligand resulting in a surface bound Cu(hfac) fragment (Figure 2-3). A homogeneous dissociation of the neutral ligand has also been considered.94 The surface adsorbed Cu(hfac) fragment can then undergo disproportionation reaction with another Cu(hfac) fragment to form the copper film and Cu(hfac)2, which should then desorb into the gas phase and not decompose at the deposition temperatures.

Figure 2-3. Proposed CVD mechanism for the decomposition of LCu(hfac) precursors.

One potential concern about using this family of precursors for SPMCSD is that it produces Cu(hfac)2 as a byproduct, which would remain in solution. As previously noted,

Cu(hfac)2 absorbs light within the SPR range and could result in competitive absorption of light with the substrate. However, this was ruled out as a restrictive issue, as the concentration of

Cu(hfac)2 should never reach a point where it absorbs a significant enough portion of the incoming light to compete with the SPR response and thus limit the temperature of the hot spot on the substrate. Also, the only known photochemistry of Cu(hfac)2 uses ultraviolet light with wavelengths of 230-250 nm so competing photoreactions were not of concern.100

SPMCSD from Adduct Stabilized Copper β-diketonates

Mechanistically, these compounds can be expected to decompose in a similar fashion to the PPh3AuCH3 compounds, with solution-phase dissociation of the neutral ligand needing to

occur before adsorption of the Cu(hfac) fragment (Figure 2-4). Two Cu(hfac) fragments should then disproportionate into pure metallic copper on the substrate and Cu(hfac)2 which then can desorb from the substrate into solution. This is very similar to the reported CVD mechanism of these precursors. However, an important difference in these mechanisms is the solution phase dissociation of the ancillary ligand L. As noted above, several of the neat precursors decompose at room temperature due to weak bonding with the neutral ligand which can lead to ready disproportionation of the complex.96 This dissociation would likely be more favored in a dilute solution which SPMCSD previously has utilized (< 0.02 M). Because of this, stronger adducts than VTMS should be used. Compound 2 has been reported as moderately more stable than 1 due to the conjugated alkyne ligand, MHY, and was reported as stable for up to four hours at 65

°C.89, 101

Figure 2-4. Predicted mechanism for the SPMCSD of copper from adduct stabilized copper β- diketonates.

Several conjugated alkyne stabilized copper β-diketonates have been synthesized and used as SPMCSD precursors. These precursors were synthesized through minor variations of a known procedure (Figure 2-5).101 In brief, a suspension of cuprous oxide and L were stirred in

an organic solvent, Hhfac was then added slowly to the stirring solution, resulting in a color change of the solution to a yellow or chartreuse color depending on the complex. Purification was performed by filtering off the excess cuprous oxide and followed by column chromatography, crystallization, or solvent washings as needed. Compound 2 is a yellow liquid when purified, but slowly oxidized at room temperature as noted by a color change to chartreuse.

This could be minimized by storing at -20 °C. Compounds 3 and 4 were pale-yellow solids that showed greater stability and did not undergo noticeable oxidation after several months when stored under ambient conditions. The syntheses of several other neutral adduct complexes were attempted, such as cyclooctatriene (COT), cyclooctadiene (COD), and 1,2- divinyltetramethyldisiloxane (DVTMSO). However, these were found to be relatively more unstable and decomposed in solution during workup, and therefore were not considered for use as SPMCSD precursors.

Figure 2-5. Synthesis of adduct stabilized copper β-diketonates.

Attempts to deposit copper using SPMCSD were first carried out with 2, as it was previously used to deposit pure copper by CVD with substrate temperatures between 170-300

°C.101 After 30 minutes of irradiation with a xenon lamp (2.0 W/cm2 power density, 515 nm longpass filter), a benzene solution of 2 resulted in a change of the AgFON surface, which is consistent with the deposition of a copper film (Figure 2-6a). However, a control experiment, in which the AgFON was submerged in a benzene solution of 2 and not irradiated, also resulted in

the formation of nanoparticles after 30 minutes (Figure 2-6b). This was likely due to the lability of the MHY ligand in a dilute benzene solution, resulting in disproportionation at room temperature. It is worth noting that more deposits were observed with irradiation, which is indicative of a partial surface plasmon mediated mechanism. The stability of 2 has previously been increased by the addition of free MHY ligand.101 However, no depositions were attempted with additional MHY to help stabilize 2 in solution.

Figure 2-6. SEM of the deposits from 2. a) After 30 minutes of irradiation with a xenon lamp. b) After 30 minutes in the dark.

The adduct precursors (DTA)Cu(hfac) (3) (DTA = di-p-tolylacetylene) and

(DPA)Cu(hfac) (4) (DPA = diphenylacetylene) were then synthesized as an increase in the conjugation of the alkyne was reported to increase the stability of the adduct.89 Compound 3 had not been previously reported, but was readily synthesized as described above with no decomposition during workup or storage. Initial attempts to synthesize 4 resulted in a green solid, which was paramagnetic by NMR. Eventually, it was found that quickly filtering the reaction solution, adding hexanes, and storing at -20 °C would crystallize 4 out of solution as a stable, pale yellow solid. Previously, 4 had been reported and found to exist with an pseudo- trigonal planar geometry exhibiting a distorted C−C≡C angle of approximately 160°.92 This

weakening of the triple bond is consistent with the observed decrease in stretching frequency from 2217 cm-1 in free DPA to 1990 cm-1 which was measured in 4.102 Likewise, 3 exhibits an

C≡C stretching frequency at 1993 cm-1 and so is expected to have a comparable distortion in the

C-C≡C angle to 4 and other alkyne adducts.84

Neither 3 or 4 had previously been used in the deposition of copper, so the thermal stability of these compounds was first investigated by thermogravimetric analysis (TGA) (Figure

2-7). The onset of decomposition is noted for both complexes between 100-130 °C and mass loss ceased at 180 and 200 °C for 3 and 4, respectively. The residual masses from TGA experiments match well with the expected mass of metallic copper formed through the thermally-initiated disproportionation pathway. The solution stability of these complexes was qualitatively determined by tracking the change of their UV-Vis spectra over time in a benzene solution (Figure 2-8). A small increase in absorption of visible light was noted for both complexes, which was attributed to an increase of copper(II) in solution. However, the change in the spectrum of 3 appeared to be slower than that of 4, which was taken as better stability in solution. Because of this, 3 was preferentially chosen as the next precursor for SPMCSD over 4.

100 (DTA)Cu(hfac) (DPA)Cu(hfac)

40 Weight%

0 0 100 200 300 400 500 Temperature (C)

Figure 2-7. Thermogravimetric analysis of compounds 3 and 4.

0.2 0.2 0 min. 0 min. 60 min. 10 min. 120 min. 20 min.

180 min 30 min.

Absorbance(a.u.) Absorbance (a.u.)

0.0 0.0 500 600 700 500 600 700 Wavelength (nm) Wavelength (nm)

Figure 2-8. UV-Vis spectra over time for compounds 3 and 4 in benzene solution. (a) Compound 3 over 180 minutes. (b) Compound 4 over 30 minutes.

Irradiation of an AgFON submerged in a benzene solution of 3 resulted in the growth of large copper islands on the AgFON after 30 minutes (Figure 2-9a). A film of copper covered the

AgFON when the sample was irradiated for 70 minutes (Figure 2-9b). X-ray photoelectron spectroscopy (XPS) was used to determine the chemical composition of the deposits (Figure 2-

10). Two major peaks are observed at 932.6 and 952.6 eV which could correspond to the 2p3/2

0 +1 and 2p1/2 peaks of either Cu or Cu . A small shoulder was also observed at slightly higher biding energies which can be attributed to Cu2+. However, XPS of these deposits also revealed significant fluorine and carbon contamination. A fluorine 1s peak was observed at 688.3 eV and is attributed to fluorine bound to carbon. Several peaks are observed in the carbon 1s region of the XPS. The major peak at 284.9 eV can be attributed to adventitious carbon from the atmosphere, which has adsorbed and decomposed on the deposits. Two minor peaks can also be observed at 288.3 eV, which is attributed to a carbon oxygen double bond, and at 292.8 eV, which is likely due to carbon bound to fluorine. From this XPS data, we believed that the surface was being passivated either by adsorbed Cu(hfac)2 or coordinating hfac ligands, which were unable to undergo disproportionation into the Cu(hfac)2.

Figure 2-9. SEM micrographs of the deposits from the SPMCSD of 3. (a) After 30 minutes of irradiation. (b) After 70 minutes of irradiation.

Figure 2-10. XPS spectra of the SPMCSD deposits after 70 minutes of irradiation. (a) Copper 2p region. (b) Fluorine 1s region. (c) Carbon 1s region

While a ligand-passivated surface is not inherently bad, many applications of copper films and nanostructures require direct contact with the copper surface to work efficiently.

While post deposition processing could potentially be used to remove this contamination, no such attempts were made as one of the goals of this deposition technique is to directly deposit

nanoparticles or films free of surface contamination without the need for post deposition processing. Because of this persistent contamination, stabilized β-diketonates, and many of the other previously known copper(I) CVD precursors, appear to be poor candidates for SPMCSD precursors. By comparing these results to the only other previously known SPMCSD precursor,

PPh3AuCH3, we predict that the ideal SPMCSD precursor would decompose to either neutral fragments which only weakly interact with the substrate or volatile byproducts that readily escape into the headspace gases.

SPMCSD from Copper Borohydrides†103

The synthesis and structure of transition metal borohydrides were heavily investigated throughout the 1960s and 1970s.104, 105 These compounds were primarily investigated as reductants for chemical synthesis or as hydrogen storage materials.106, 107 Several metal borohydride complexes have also been shown to be volatile enough to be used for the CVD of

108 metal borides. These precursors are typically homoleptic compounds, such as Zr(BH4)4 and

Hf(BH4)4, which is believed to be important for their weak intermolecular interaction and

109 volatility. Attempts to synthesize homoleptic copper borohydrides, such as CuBH4 or

110 Cu(BH4)2, result in rapid decomposition to copper metal at room temperature. However, many phosphine stabilized copper(I) borohydrides have been previously synthesized.111-114

These compounds were primarily of interest due to the varied binding modes of the borohydride to the copper center, but found uses as a weak reductant.115 Of these copper borohydrides, bis(triphenylphosphine)copper borohydride (5) was one of earliest synthesized and most well studied.116 Compound 5 was reported to decompose between 165 and 177 °C as a solid, but was

† “Surface Plasmon-Mediated Chemical Solution Deposition of Cu Nanoparticle Films.” Qiu, J.; Richey, N. E.; DuChene J. S.; Zhai, Y.; Zhang, Y.; McElwee-White, L.; and Wei, W. D. J. Phys. Chem. C, 2016, 120, 20775.

also noted to decompose in a benzene solution at 50 °C with byproducts of hydrogen gas, triphenylphosphine, and a triphenylphosphine borane adduct. These neutral byproducts and low temperature solution phase decomposition were promising for the use of these precursors in

SPMCSD experiments.

Synthesis and Characterization of Copper Borohydride Complexes

All of the phosphine stabilized copper borohydrides were synthesized through adaption of a two-step procedure previously reported by Gysling (Figure 2-11).117 In the original procedure, (PPh3)2Cu(NO3) (6) is synthesized through the reduction of Cu(NO3)2 by excess triphenylphosphine in refluxing methanol, though we have found that ascorbic acid could also be used as the reductant. Salt metathesis with NaBH4 in a mixture of DCM and ethanol is then used to produce 5 in good yields after recrystallization (>80% yield). Because both 5 and 6 are air stable solids and that this procedure begins with a copper(II) salt, these procedures can be performed without rigorous exclusion of air and water, unlike other methods of synthesis which start with copper(I) salts.113 This procedure has been adapted to the synthesis of similar derivatives with different phosphines and borohydrides.

Figure 2-11. Synthesis of 5 and 6 by the Gysling method.

In general, the synthesis of the nitrate and borohydride could be easily followed by IR spectroscopy. The N-O and B-H stretches are both well known for coordination compounds of copper and can be used to determine their binding mode to the metal. The purity of the borohydride product could be followed by the disappearance of the N-O stretches from the nitrate complex. All of the synthesized copper borohydrides and nitrates are given in Figure 2-12

and the details of their synthesis and characterization are discussed below. All compounds were white solids with no absorption of light in the visible region of the electromagnetic spectrum.

Figure 2-12. Structures of the copper complexes 5-12.

Compound 6 was first synthesized in 1960 by Goodgame and Cotton, though the nature of the nitrate bonding was not known until its crystal structure was determined by Palenik and

Messmer in 1969.118, 119 Two distinct bands from the symmetric and asymmetric stretch of the

-1 nitrate are observed at 1475 and 1275 cm and indicate a C2v symmetry of the nitrate (Figure 2-

13).120, 121 This was confirmed in the crystal structure as it was determined that the nitrate acts as a bidentate ligand, resulting in a tetrahedral coordination around the copper center.119 Metathesis of the nitrate for borohydride or cyanoborohydride can be easily followed by the disappearance of the nitrate peaks and the rise of new bands from the borohydride.

% Transmittance % % Transmittance % 5 7 6 6

2400 2200 2000 1400 1200 2400 2200 1400 1200 -1 -1 Wavenumber (cm ) Wavenumber (cm )

Figure 2-13. IR spectra of compounds 5, 6, and 7.

Compounds 5113, 114 and 7111, 122 along with their crystal structures were both previously reported by Lippard and coworkers, though from an alternative synthetic pathway. From the

- crystal structure, 5 was shown to coordinate through two of the hydrides of the BH4 in a bidentate fashion, similar to 6. Compound 5 exhibited several major peaks for the B-H vibrations at 2398, 1987, and 1926 cm-1, which agreed well with the previous reports (Figure 2-

13). The broad band at 2398 cm-1 is attributed to terminal B-H stretches while the bands at 1987 and 1926 cm-1 are attributed to the copper bridging B-H stretches. Similar IR spectra were reported for other bis(phosphine)copper borohydrides, which are therefore also expected to coordinate in a similar bidentate fashion. However, when cyanoborohydride is used in place of borohydride a unique dimeric structure is observed where the cyanoborohydride coordinates through both a bridging hydride and the nitrile functionality. The IR of this compound is similar to that of 5 with broad bands at 2372 and 2203 cm-1 for the terminal and copper bridging B-H stretches, respectively. A new band is also observed at 2187 cm-1 and is attributed to the C≡N stretch of the coordinated nitrile.123 Tris(triphenylphosphine)copper cyanoborohydride (8) was also previously reported by Lippard.111 Attempts to synthesize tris(triphenylphosphine)copper nitrate from the Gysling method failed. However, it was found that by adding one additional

equivalent of triphenylphosphine to a solution of 6 prior to the addition of sodium cyanoborohydride in ethanol would result in 8. In 8 the band from the bridging B-H disappears and only the terminal B-H band at 2338 cm-1 and the coordinated nitrile band at 2172 cm-1 remain (Figure 2-14). The third phosphine in 8 readily dissociates in organic solvents, such as

chloroform, to regenerate 7. % Transmittance %

2400 2200 1400 1200 -1 Wavenumber (cm )

Figure 2-14. IR spectrum of compound 8.

Tris(phosphine)copper nitrates/borohydrides can be obtained when less sterically hindered phosphines are used.124, 125 Tris(methyldiphenylphosphine)copper nitrate (9) was readily synthesized by the adapted Gysling method when using methyldiphenylphosphine in place of triphenylphosphine. The symmetric and asymmetric stretch of the nitrate in 9 are observed at higher wavenumbers, 1407 and 1300 cm-1, which agrees well with the previous synthesis of 9 (Figure 2-15). These bands are also significantly closer in wavenumber than the bidentate nitrate complex. This is believed to be caused by the unidentate binding of the nitrate, which was later confirmed by its X-ray crystal structure.126 From 9, compounds 10 and 11 could be readily synthesized through a similar salt metathesis as described above. Terminal and bridging B-H stretches are observed at 2319 and 2026 cm-1, respectively, in 10. The IR spectrum of 11 is very similar to that of 8 and has a band for the terminal B-H stretch at 2325 cm-1 and a

band for C≡N at 2183 cm-1. While 11 had not been previously synthesized, the similarity of the

IR spectrum of 11 and 8 was good indication of a similar solid-state structure.

% Transmittance % % Transmittance % 9 9 10 11

2400 2200 2000 1400 1200 2400 2200 1400 1200 -1 -1 Wavenumber (cm ) Wavenumber (cm )

Figure 2-15. IR spectra of compounds 9, 10, and 11.

Synthesis of complexes with other, more sterically hindered phosphines were also attempted. Tri(o-tolyl)phosphine has one of the largest Tolman cone angles of any phosphine.127

Bis(tri(o-tolyl)phosphine)copper nitrate (12) was synthesized through a modification of the

Gysling synthesis, though ascorbic acid had to be used as a reductant. The IR of 12 showed bands for the nitrate at 1436 and 1298 cm-1 (Figure 2-16). This IR was very close to that of 10, and so unidentate coordination of the nitrate is expected. To my knowledge, unidentate binding has not previously been observed for bis(phosphine)copper nitrates and is assumed to be from the large steric bulk of the tri(o-tolyl)phosphine. Previously, attempts to synthesize tris(tri(o- tolyl)phosphine)copper borohydride resulted in immediate decomposition to copper metal upon addition of the sodium borohydride, and so was not attempted.113 However, no previous attempts to synthesize the cyanoborohydride derivative has been reported. Salt metathesis of 12 with sodium cyanoborohydride in ethanol resulted in the precipitation of a white solid which was insolable in most solvents, though was readily dissolved in acetonitrile. IR of this solid revealed bands for both terminal and bridging B-H stretches at 2432 and 2097 cm-1, as well as a band

from the C≡N at 2205 cm-1. From this IR spectrum, it appears that the cyanoborohydride is coordinating through both the nitrile and hydride. The insolubility of this product likely indicates that it is polymeric in nature, though smaller clusters have not been ruled out. One equivalent of tri(o-tolyl)phosphine can also be isolated from the filtrate of this reaction and is displaced to allow for coordination of the cyanoborohydride. No trialkylphosphines were used, as they have previously been reported to be unsuitable for the stabilization of copper borohydrides and resulted in rapid decomposition during the transmetalation of the borohydride.113

% Transmittance % 12 [P(o-tolyl) Cu(NCBH )] 3 3 n

2400 2200 1400 1200 -1 Wavenumber (cm )

Figure 2-16. IR spectra of compound 12 and the proposed polymeric [P(o-tolyl)3Cu(NCBH3)]n.

Thermal Properties of Copper Borohydride Complexes

As stated previously, 5 is known to decompose as a solid at temperatures above 165 °C and in benzene solution at temperatures above 50 °C. However, the reported thermal properties of the other derivatives are quite limited and further study was needed to determine their suitability for SPMCSD. Solid state decomposition temperatures could be determined by TGA whereas solution phase decomposition temperatures were determined by heating the complex in a solvent and observing the temperature at which copper particles or a copper film formed.

Benzene was used for initial solution phase decomposition measurements, followed by a toluene

solution or a naphthalene melt for compounds stable in refluxing benzene. These solvents were chosen for their similarity to benzene so as to minimize potential solvent effects.

The TGA plots for compounds 5, 7, 8, 10, and 11 are given in Figure 2-17. Solid-state decomposition temperatures are noted when the weight percent drops below 95%. By TGA we see that 5 begins to decompose at 152 °C, although mass loss occurs most rapidly at temperature above 160 °C which explains the previous solid-state decomposition temperature being reported at higher temperatures. Compound 7 has the highest decomposition onset temperature by TGA, with significant decomposition occurring at 174 °C. Interestingly, compound 8 showed initial decomposition lower than that of 7 with decomposition beginning at 156 °C. Compound 10 has the lowest solid-state decomposition temperature with an onset close to 113 °C. This decrease may be due to the increased electron donation of the methyldiphenylphosphine, which could also be why previous attempts to synthesize copper borohydrides with trialkylphosphines have failed.

Compound 11 decomposes at 149 °C. The similarity between this and 8, and the disparity between 8 and 7, may indicate the importance of the coordination of the cyanoborohydride.

100% 5 7 8 80% 10 11

60%

40% Weight%

20%

0% 0 100 200 300 400 500 o Temperature ( C)

Figure 2-17. TGA traces of compounds 5, 7, 8, 10, and 11.

However, as previously described for 5, the solution phase decomposition temperature could be significantly different from the solid-state decomposition temperature. As previously reported, compound 5 was found to decompose in benzene between 50-55 °C. An additional equivalent of triphenylphosphine could be added to a toluene solution to stabilize 5 up to 80-90

°C, though addition of more triphenylphosphine to the solution did not further influence the stability. Both compounds 7 and 8 decomposed in a naphthalene melt between 160-170 °C, which is very close to their solid-state decomposition. This likely indicates that 8 dissociates the third phosphine in the naphthalene melt which would result in the formation of 7. This isnʹt surprising since this dissociation was noted above in organic solvents. Compound 10 also showed very similar in both solid and solution-phase decomposition temperatures and decomposed in toluene at 100-105 °C. Interestingly, 11 was the only compound that had a higher solution phase decomposition temperature than in its solid-state and no decomposition was observed in the naphthalene melt until 190 °C. The higher solution phase decomposition temperatures of 7, 8, 10, and 11 could potentially come from stability due to having an extra neutral coordinating ligand, either as the additional phosphine in 10 and 11 or the nitrile coordination in 7, 8, and 11. Increased stability could also come from having the weaker reducing strength of the cyanoborohydride in 7, 8, and 11.

Table 2-1. Solid-state and solution phase decomposition temperatures of the synthesized copper borohydrides. Compound Solid (°C) Solution (°C) 5 152 50-55a 7 174 160-170b 8 156 160-170b 10 113 100-105c 11 149 190-195b aBenzene solution, bnaphthalene melt, ctoluene solution.

Deposition of Copper Borohydrides

Due to having the lowest solution phase decomposition temperature, SPMCSD experiments with 5 were attempted first. The AgFON substrate was immersed in a benzene solution containing 5 at room temperature and irradiated with visible light for 60 minutes using a xenon lamp (435 nm longpass filter, 2.0 W/cm2 power density). The bulk solution temperature was monitored via a thermocouple throughout the duration of the SPMCSD reaction and was observed to reach 40 ºC. Inspection of the substrate with SEM after irradiation shows an obvious change in surface roughness on the silver surface with the occurrence of many nanoparticles on the AgFON substrate (Figure 2-18). However, XPS of these deposits reveals a significant amount of Cu2+ as indicated by the peaks at 935 and 955 eV and the broad satellite peaks at 943 and 963.0 eV. This oxidation likely occurred during the deposition process, as no care was taken to remove water or oxygen from the deposition solvent or atmosphere. This was interesting to note, since similar handling was used for 3 previously and no significant oxidation was observed in those copper deposits which indicates that the hfac contamination may passivate the surface from reaction with oxygen or water.

a) b)

Figure 2-18. Initial SPMCSD results from the deposition of 5. (a) SEM image of the deposits. (b) XPS of the Cu 2p region.

When the solution was prepared with anhydrous benzene and the substrate exposed to the solution under nitrogen gas, a change in morphology of the substrate was observed after just five minutes of irradiation with the xenon lamp (Figure 2-19a). No change in substrate morphology or appearance of nanoparticles was observed when a control experiment was conducted under dark conditions for 3 h, confirming that light irradiation is required to initiate nanoparticle deposition (Figure 2-19b). Comparison of the binding energies from the Cu 2p and Cu LMM regions of 5 and the copper deposits confirms a change in chemical state during the SPMCSD process (Figure 2-19c, d). The spectrum of the nanoparticles exhibits two peaks with binding energies of 932.6 eV and 952.5 eV for Cu 2p3/2 and Cu 2p1/2 regions, respectively. The satellite peak characteristic of Cu2+ at 943 eV was absent, indicating that no significant oxidation occurred during growth. As the binding energies of Cu0 and Cu+1 are very similar in the Cu 2p region, the Cu LMM Auger electron region was used to distinguish between these oxidation states. The dominant peak at 568.0 eV clearly shows that the majority of the deposited nanoparticles consists of metallic Cu0. The slight shoulder at 569.8 eV indicates the presence of

+1 Cu , likely from a thin layer of Cu2O on the nanoparticle surface which forms from post deposition handling of the material in atmosphere. No phosphorus or boron contamination was observed in the XPS spectrum, indicating clean removal of the precursor ligands (Figure 2-19e, f). Since the substrate itself does not induce the decomposition of 5 in the dark, and the precursor has no absorption in the irradiation range used for the deposition, these results are consistent with photothermal decomposition of 5 to yield copper nanoparticles via SPR excitation of the underlying AgFON substrate. This SPMCSD approach provides high-purity Cu

NPs on the AgFON substrate that could serve as a plasmonic platform for a variety of applications.

(a(a) ) (ba)(b) ) (cb)(c) )

100100 nm nm 100100 nm nm 100100 nm nm

c) d)

e) f)

Figure 2-19. SPMCSD results from deposition of 5 with rigorous exclusion of air and water. (a) SEM image of the deposits. (b) SEM of the AgFON after 3 hours of submersion in a benzene solution of 5. (c) XPS of the Cu 2p region. (d) XPS of the Cu LMM Auger region. (e) XPS of the P 2p region. (f) XPS of the B 1s region. Adapted with permission from reference 103. Copyright American Chemical Society.

The morphology of the copper deposit could be easily controlled by the parameters used

during the deposition. When using the same lamp power and longpass filter, longer irradiation

times resulted in total coverage of the AgFON with copper. After just 30 minutes of irradiation

an irregular film of copper was deposited and the underlying AgFON was no longer

distinguishable (Figure 2-20a). When the time and longpass filter are kept constant, an increase in the lamp’s power density increases the rate of deposition. After only 5 minutes with excitation from a 2.8 W/cm2 lamp total coverage of the AgFON was observed, although the underlying spherical morphology of the substrate was still seen (Figure 2-20b). By decreasing the power density of the lamp to 1.5 W/cm2 and increasing the cutoff wavelength of the longpass filter to 515 nm, a significant decrease in deposition rate is observed. After irradiating under these conditions for 10 minutes a thin and conformal film of copper is observed (Figure 2-20c).

The increase and decrease of the deposition rate is expected as it has been previously noted that that plasmonic heating efficiency is proportional to the excitation power density.128 The XPS of these deposits revealed that the nature of the copper deposited was unchanged by the different conditions used. Based on these observations, it appears that copper atoms nucleate to form nanoparticles in the initial stages of precursor decomposition and then eventually coalesce into a continuous film. This growth behavior agrees well with the Volmer-Weber model of film deposition.129 A similar Volmer-Weber growth mode was observed for the growth of Cu islands on a Ag surface under ultrahigh vacuum conditions.130

Compared to the initial AgFON substrate, the newly formed AgFON/Cu film substrate

(Figure 2-20b) exhibits significant extinction across the entire visible spectrum (400—700 nm) with a peak maximum (λmax) at ca. 500 nm (Figure 2-21a), making the hybrid plasmonic substrate an intriguing platform for SERS applications. Moreover, the native oxide (i.e. Cu2O) on the copper surface offers unique opportunities to interact with Raman-active molecules containing carboxylic acid moieties that would otherwise not adsorb onto gold or silver-based

SERS substrates.131 The Raman probe molecule 4-mercaptobenzoic acid (4-MBA) was chosen

Figure 2-20. Morphology control of the copper deposits. (a, d) SEM and XPS of the deposits grown when irradiated for 30 minutes with a 435 longpass filter and lamp power density of 2.0 W/cm2. (b, e) SEM and XPS of the deposits grown when irradiated for 5 minutes with a 435 longpass filter and lamp power density of 2.8 W/cm2. (c, f) SEM and XPS of the deposits grown when irradiated for 10 minutes with a 515 longpass filter and lamp power density of 1.5 W/cm2. Adapted with permission from reference 103. Copyright American Chemical Society. to illustrate the advantages of a AgFON/Cu substrate as a plasmonic SERS platform. Figure 2-

21b shows the SERS spectra of 4-MBA adsorbed onto the AgFON/Cu substrate along with the pure 4-MBA powder for reference collected under 532 nm laser excitation. The dominant features located at 1084 cm-1 and 1586 cm-1 were assigned to aromatic ring vibrations, while the signal at 1182 cm-1 arises from the C-H deformation mode.132 It is noted that the COO- vibrational mode observed at 1290 cm-1 in the pure 4-MBA powder shifted to ca. 1390 cm-1 upon adsorption on the Cu2O surface. Such a significant shift in this vibrational mode confirms that the carboxylate group preferentially anchors the Raman probe molecule to the native oxide on

the AgFON/Cu surface.133 From these experiments, the magnitude of the SERS enhancement factor (EF) from the AgFON/Cu hybrid structure was estimated to be on the order of ~105.

Figure 2-21. Surface Enhanced Raman Spectroscopy of 4-MBA using AgFON/Cu. (a) Diffuse- reflectance spectra of the AgFON substrate and the AgFON/Cu film. (b) Raman spectra of 4-MBA adsorbed onto the AgFON/Cu film and of pure 4-MBA powder. Adapted with permission from reference 103. Copyright American Chemical Society.

Attempts to deposit copper from both 7 and 10 both failed and resulted in no visible morphology change of the AgFON which is seen for successful depositions. This was initially surprising, considering that both of these compounds exhibited solution phase decomposition at temperatures lower than estimated hotspot temperature of the AgFON. Currently, it is expected that the additional coordinating ligand in both 7 and 10 helps prevent significant solution phase dissociation into a fragment that is small enough to adsorb between the adsorbed benzene molecules. However, no further attempts were made utilizing other solvents which may not have interacted strongly with the AgFON.

Summary of Copper Deposition from SPMCSD

Copper nanoparticles and thin films were successfully deposited from two precursor compounds, 3 and 5, onto a AgFON substrate using visible light. The choice of precursor was found to be very important for obtaining high purity of the final copper deposit. Compound 3 decomposed through a disproportionation pathway and resulted in contamination of the

nanoparticles with the β-diketonate ligand, hfac. Meanwhile, 5 decomposed through the formation of hydrogen gas which readily left the solution into the headspace gases and resulted in pure copper deposits when grown under an inert atmosphere. The morphology of the copper deposits from 5 could be easily controlled through the irradiation time, lamp power density, and longpass filter used. Morphologies ranging from nanoparticles to thick films could be deposited depending on the conditions used. The AgFON/Cu film was shown to be a promising candidate for SERS of carboxylate containing small molecules.

The solution phase stability of precursors was also shown to be very important for

SPMCSD experiments. When a ligand is too labile in solution, such as the MHY in 2, then nonselective deposition occurs without any external light. However, too many stabilizing ligands, such as in 7 and 10, would result in no deposition. This is likely due to the hindering of significant ligand dissociation to form a fragment that could then adsorb onto the benzene passivated AgFON and undergo decomposition.

CHAPTER 3 AEROSOL ASSISTED CHEMICAL VAPOR DEPOSITION OF TUNGSTEN DISULFIDE

Transition Metal Dichalcogenides

In 2004, it was found that graphite could be exfoliated to generate one atom thick sheets of graphene.134 Due to its 2-dimensionally confined features, graphene has been found to exhibit several physical and electronic phenomena that are absent in its bulk counterpart.135 For instance, single layer graphene can be considered as a zero bandgap semiconductor, as its valence and conduction bands meet at a Dirac point, but do not overlap.136 This results in graphene having extremely high charge mobility and low electrical resistivity.137 Graphene also exhibits several unique physical properties such as an extremely high thermal conductivity and is one of the strongest materials ever measured.138, 139 With these several unique properties, graphene has been dubbed a “wonder material” by many and has resulted in a significant increase in research interest of many other 2D materials.

While many other 2D materials are known, layered transition metal dichalcogenides

(TMD) are perhaps the most well studied after graphene.140-143 These TMD materials take on a general formula of ME2, where M is a tetravalent transition metal and E is a group 16 chalcogenide such as sulfur, selenium, or tellurium. While not all TMD materials are layered, many of the early groups (groups 4, 5, and 6) and some later metals (Pt, Pd, and Re) are, and so have been investigated as graphene analogues. The layers of TMDs are comprised of the metal atoms sandwiched between two sulfide layers which form weak Van der Waals interactions with other layers (Figure 3-1). The electronic properties of the TMD materials can range from metallic, semiconductor, semimetal, or superconducting depending on the metal and chalcogenide used and on the polymorph of the material.

Figure 3-1. Crystal structures of the layered materials graphite and a generic transition metal dichalcogenide. Adapted with permission from reference 144. Copyright (2014), Royal Society of Chemistry.

Applications of TMDs

Like graphene and graphite, a single layer of a TMD material can behave quite differently from the bulk material. Because of this, applications for the bulk material and monolayers of TMDs can differ significantly. Significant amounts of work on bulk TMD materials have been conducted since the 1960s.145 Due to the weak intermolecular forces between the layers, they are able to slide over one other with relatively little friction. This makes

TMDs promising and widely studied materials for tribological applications, such as solid lubricants.146 The spacing between layers also provides space for the intercalation of small atoms, such as lithium.147 This has resulted in many TMDs being used as in energy storage materials such as capacitors and batteries.148, 149 These materials have also been found to be promising candidates for electrocatalytic hydrogen evolution.150, 151

A wider variety of applications can be achieved when using monolayers of TMDs due to the confinement of their electrons. Applications such as hydrogen evolution, which do not rely on the layered structure of bulk TMDs, can also be utilized with monolayers of the TMD.152, 153

However, as atomically thin semiconductors, monolayers of TMDs have potential uses in many

nanodevices. Currently, one of the most explored applications of TMDs is their use as atomically thin field effect transistors, which graphene cannot be used for due to its lack of bandgap.154 Surface modified monolayers have also been employed for various biomedical applications such as photothermal therapy and drug delivery.155 Several TMDs go from an indirect bandgap in the bulk to a higher energy, direct bandgap for their monolayers (Figure 3-

2).156 This direct bandgap leads to a large photoluminescence and makes monolayers of those

TMDs useful in photonic applications such as in photodetectors or light emitting diodes.157-159

Figure 3-2. Band structure of bulk and monolayer WS2, black arrow indicates the fundamental band gap. Adapted with permission from reference 156. Copyright (2011), American Physical Society, doi.org/10.1103/PhysRevB.83.245213.

Deposition Techniques for TMDs

Of all the layered TMDs known, MoS2 and WS2 are the more widely studied, as they are naturally occurring and thus are air- and water-stable. Bulk crystals of TMDs are typically produced by chemical vapor transport, in which powdered metal and chalcogen, and often a transport agent such as I2, are placed in a sealed tube and heated to temperatures above 600

160, 161 °C. Recently, crystalline WS2 has been synthesized at room temperature by reaction of bis(cyclopentadienyl)tungsten dihydride and elemental sulfur in a benzene solution.162 Thin

films of TMDs can be deposited by CVD utilizing a dual source precursor approach with a metal carbonyl or halide and hydrogen or dialkyl chalcogenide as the metal and chalcogen source, respectively, at temperature above 150 °C.163, 164 Several single source precursors, such as

t t Ti(S Bu)4 and Mo(S Bu)4, have been used for the CVD of TMD materials at temperatures between 110-350 °C and are discussed in more detail below.165

Monolayers of TMDs have previously been produced from chemical and mechanical exfoliation of the corresponding bulk TMD material.166, 167 Within the last five years, direct growth of crystalline monolayers of various TMDs has been achieved through several variations of CVD. This was first achieved through the sulfurization of MoO3 or WO3 at 800 °C with a large excess of sulfur.157, 168 This method has now been applied to many TMD monolayers and

169 heterostructures. Metal organic CVD has also been used to deposit WSe2 when using W(CO)6 and dimethylselenium as the two precursors with growth temperatures between 600-900 °C.170

Thin films of ammonium tetrathiomolybdate and tetrathiotungstate have been spin coated onto a substrate and annealed at 900 °C under hydrogen gas and excess sulfur to produce monolayers of the corresponding metal disulfide.171

Single source precursors for TMDs

Single source precursors have been used for a variety of TMD materials, both in conventional CVD for sufficiently volatile precursors and AACVD for nonvolatile precursors.

As previously noted, thiolates of titanium and molybdenum have been previously used for deposition of their disulfides in low pressure CVD.165 Selenoates have been used with titanium, zirconium, and hafnium to deposit their corresponding diselenides in low pressure CVD.172

Many TMDs of group 4 and 5 metals have been deposited from chalcogenoether adducts of the metal halide using low pressure CVD.173-175 Dithiocarbamates of molybdenum have been used

176, 177 for both low pressure and AACVD of MoS2. Dithiocarbamates and thiolates have also

38 been used to deposite ReS2 using AACVD. However, despite its increasing popularity in the literature, no single source precursor had previously been developed for WS2.

Tungsten Dithiocarbamate Precursor‡178

Our prior use of oxo and dioxo single source precursors for deposition of WOx films and

179, 180 nanorods suggested use of sulfido complexes for the deposition of WS2. Although bridging sulfido complexes have been used successfully in the deposition of gallium and iron sulfide, to our knowledge, there was no precedent for the use of a terminal sulfido complex for

CVD of a metal sulfide.181, 182 For the ancillary ligands, dithiocarbamates have been found in

177 precursors for a variety of metal sulfides, including the TMD MoS2 from Mo(S2CNEt2)4.

While the tungsten analogue has been reported, the sensitivity of its solutions to oxygen and heat would make it difficult to handle for AACVD.183, 184 These considerations led to our choice of the air-stable complex, WS(S2)(S2CNEt2)2 (13), as a precursor for the AACVD of WS2. This compound was previously synthesized in a one-step reaction between ammonium tetrathiotungstate and tetraethylthiuram disulfide in acetonitrile (Figure 3-3).185

Figure 3-3. Synthesis of WS(S2)(S2CNEt2)2.

While 13 had been previously prepared, its thermal properties were not reported. From the TGA (Figure 3-4), we see an initial onset of decomposition around 200 °C followed by a period of slow decomposition which is complete just above 400 °C. Stepwise decomposition has

‡ “Aerosol‐assisted chemical vapor deposition of WS2 from the single source precursor WS(S2)(S2CNEt2)2.” Richey, N.E.; Haines, C.; Tami, J.L.; and McElwee-White, L., Chem. Commun., 2017, 53, 7728.

been observed previously for other dithiocarbamate complexes, with EtSH, EtNCS, CS2, and

183, 184 H2S expected as the volatile byproducts. The initial stage of decomposition ends with a residual mass of approximately 68%, which is consistent with formation of WS2(S2NCEt2)

(68.7% of the original mass). This is followed by a slow period of mass loss which eventually ends at 46.58% weight, near the calculated residual mass of WS2 (42.83%).

Figure 3-4. TGA of 13. Adapted from reference 178 with permission from The Royal Society of Chemistry.

Mass spectrometry coupled to TGA (TGA-MS) was used to identify the volatile byproducts from the thermal decomposition of 13 (Figure 3-5). As expected, there were strong

+ + + MS peaks at m/z values of 34.0, 76.0, and 87.0 which correspond to H2S , CS2 , and EtNCS , respectively. Small fragments from EtNCS were also observed at m/z 72.0, 58.0, and 27.0 which

+ + + + + + correspond to CH2NCS , NCS , and HCN respectively. Three sulfur species, S , S2 , and S6 were also detected by TGA-MS at m/z 32.0, 64.0, and 192.0, respectively. Interestingly, no peak for EtSH (m/z 62.0) was detected. However, the small peak at m/z 60.0 could come from the formation and fragmentation of EtSSEt+. The analogous process has been reported for the

186 decomposition of M(Se2CNEt)2 complexes (M = Zn or Cd).

Figure 3-5. TGA-MS data from 13. Adapted from reference 178 with permission from The Royal Society of Chemistry.

Many common fragments can be observed by TGA-MS at more than one decomposition

+ + temperature. For both CS2 and EtNCS a large peak is observed during the initial decomposition

+ at 200 °C followed by smaller peaks at 400 °C. The H2S formation follows a different profile, with a larger peak being observed at 400 °C. Peaks associated with sulfur species are observed

+ only at 200 °C. Peaks of C2H4S can be seen at both 200 and 400 °C. From these data, it appears that the majority of 13 decomposes at 200°C, but that some dithiocarbamate ligand remains, slowly decomposing, until a critical temperature that causes full decomposition is reached. It is plausible that the rapid, initial decomposition forms WS2(S2CNEt2) and the remaining dithiocarbamate may then slow the further growth of WS2 by terminating the reactive edge sites of WS2 where growth occurs most rapidly.

Isothermal TGA was also used to probe the thermal deposition pathway of 13 (Figure 3-

6a). Isothermal TGA was performed at several temperatures by heating the solid to the desired temperature at a ramp rate of 20 °C per minute and then maintaining that temperature for 50 minutes. The isothermal TGA at 400 °C appears similar to the normal TGA data, as the precursor has nearly fully decomposed before it reaches the isothermal temperature. At an isotherm of 350 °C, full precursor decomposition is observed within 15 minutes after it reaches the isothermal temperature. At 300 °C the precursor slowly decomposes over time, but does not fully decompose even after 50 minutes. Most interestingly, at 200 °C the mass loss was significantly slowed at 68% residual mass, indicating that this intermediate was stable. Pyrolysis of 13 at 200 °C resulted in a color change from green to brown. This solid was then washed with water, ethanol, and ether to remove other byproducts. The IR from the resulting solid was found to match well with the previously reported [W(S)2(S2CNEt2)2]2 complex, which is a dimer of the

187 proposed WS2(S2NCEt2) intermediate (Figure 3-6b).

Figure 3-6. Isothermal TGA and pyrolysis of WS(S2)(S2CNEt2)2. (a) Isothermal TGA, circles indicate the time when a run reaches the isotherm temperature. (b) IR of the brown solid after pyrolysis at 200 °C, inset the previously reported W2S4(S2NEt2)2 complex.

Deposition from 13 was performed at temperatures from 300 to 500 °C to investigate the material grown at different points in the temperature range where mass loss was observed by

TGA. Although precursor decomposition was evident by TGA at 200 °C, no deposits were obtained upon attempted deposition at 300 °C. This is likely a result of the slow decomposition rate of 13 at 300 °C, resulting in only a minimal amount of precursor decomposing on the substrate which was not detectable. At growth temperatures of 400-500 °C the resulting dark films covered the entire substrate, while at 350 °C dark spots were grown on some parts of the substrate but the coverage was sparse. SEM images of the material deposited at temperatures above 350 °C revealed plate-like structures (Figure 3-7), which are indicative of WS2 nano- sheets that have grown perpendicular to the substrate. This morphology is typical for the 2H phase of WS2 and, because more active edge sites are exposed, is preferred for catalytic and battery applications.152, 188

Figure 3-7. SEM images of the WS2 deposits grown at 350, 400, 450, and 500 °C. Adapted from reference 178 with permission from The Royal Society of Chemistry.

XPS was performed on the deposits to ascertain the tungsten and sulfur oxidation states

(Figure 3-8). The W 4f7/2 and 4f5/2 peaks were observed at binding energies (BEs) of 32.5 to

33.0 and 34.8 to 35.1 eV, respectively, which is consistent with W4+. A small peak at BE 38.5 eV is assigned as the W 5p peak. For material grown at 350 °C, small peaks are observed at BE

36.3 and 38.4 eV indicating slight oxidation of the tungsten, possibly due to exposure of the samples to the atmosphere prior to characterization. The S 2p3/2 and 2p1/2 peaks can be observed at BEs of 162.3 to 162.7 and 163.4 to 163.8 eV, respectively, which are within the ranges

2- 189 reported for the S oxidation state found in WS2. Small amounts of adventitious C and O, ascribed to handling of the samples in the atmosphere, were also observed.

350 °C 400 °C

Intensity (a.u) Intensity Intensity (a.u) Intensity

40 38 36 34 32 165 164 163 162 161 40 38 36 34 32 165 164 163 162 161 Binding Energy (eV) Binding Energy (eV) Binding Energy (eV) Binding Energy (eV)

450 °C 500 °C

Intensity (a.u.) Intensity Intensity (a.u.) Intensity

40 38 36 34 32 165 164 163 162 161 40 38 36 34 32 165 164 163 162 161 Binding Energy (eV) Binding Energy (eV) Binding Energy (eV) Binding Energy (eV)

Figure 3-8. W 4f and S2p regions of the XPS spectra from the WS2 deposits. Adapted from reference 178 with permission from The Royal Society of Chemistry.

Raman spectroscopy and X-ray diffraction (XRD) were used to identify the deposits as

2H-WS2. For material grown between 350-500 °C, the characteristic WS2 peaks at approximately 350 and 420 cm-1 were observed (Figure 3-9a). The peak at 350 cm-1 is a

1 -1 combination of the E 2g and 2LA(M) peaks with the A1g peak at 420 cm . In these spectra the

1 E 2g/2LA(M) peak was significantly more intense than the A1g, peak which is expected for irradiation with a 532 nm laser. The spectrum of the deposit at 350 °C was measured at a visibly dark spot and shows intense WS2 peaks. A Raman spectrum was also taken at a spot that appeared to be bare substrate and WS2 peaks were still observed, although at much lower intensity (Figure 3-9b). No change in the Raman spectrum was observed when the solvent was not rigorously dried and deoxygenated prior to AACVD (Figure 3-9c). The XRD plots for material grown at all temperatures exhibited a sharp peak around 14° 2θ, which is the characteristic (002) reflection of WS2 layers (Figure 3-9d). A small peak at 29° 2θ could also be

seen for the (004) reflection of WS2. Both Raman and XRD displayed only peaks for the silicon substrate after attempted growth of material at 300 °C.

Figure 3-9. Characterization of the WS2 deposits. (a) Raman spectrum at different deposition temperatures. (b) Raman spectrum from a visibly bare spot on the deposition at 350 °C. (c) Raman spectrum of a sample from deposition without rigorous exclusion of air and water. (d) XRD plots of the deposits. Adapted from reference 178 with permission from The Royal Society of Chemistry.

Dithiolene Precursors

Metal dithiolene complexes have been studied since the 1960s and are known for a variety of metals.190 Much of the initial interest in these complexes came from the “redox noninnocence” of the dithiolene ligand.191 These complexes have also found promise as

molecular superconductors and in conductive coordination polymers.192, 193 Dithiolenes also play an important role in biological systems in the form of the molybdopterin cofactor which many molybdenum and tungsten enzymes utilize.194, 195 However, despite the varied use of these metal dithiolenes, there have been no attempts to apply these ligands as a source of sulfide in single source precursors.

We hypothesized these complexes would be promising candidates as metal sulfide single source precursors due to their expected cycloreversion reaction which would produce the metal sulfide and a volatile alkyne. Many of the previously reported tungsten dithiolenes exist as ionic complexes,196 which could reduce their solubility in the organic solvents typically used for

AACVD. Because of this, our initial studies focused on the neutral complex dicarbonyl bis(dimethyldithiolene)tungsten (W(CO)2(S2C4H6)2) (14). Compound 14 was synthesized through the three step reaction previously used by Holm.197 In brief, ligand exchange was achieved by stirring W(CO)3(NCCH3)3 with Ni(S2C4H6)2 for several days in dichloromethane and then followed by purification using column chromatography (Figure 3-10).

Figure 3-10. Synthesis of compound 14.

The thermal properties of 14 were investigated by TGA and TGA-MS (Figure 3-11).

From TGA, 14 undergoes a rapid mass loss between 175-205 °C to approximately 67 weight %.

Decomposition continues slowly above 205 °C, eventually ending at 54.7 weight %. The final residual mass is in good agreement with the formation of WS2 from 14, which should be 52.1 weight %. The TGA-MS experiments gave several interesting mass fragments. A broad peak at

+ m/z 34.0 can be attributed to H2S and is primarily observed in the slow region of decomposition above 205 °C. A sharp peak at m/z 54.0 is observed during the rapid decomposition before 205

+ °C and is attributed to the radical cation of butyne, C4H6 . Another sharp peak before 205 °C is observed at m/z 60.0. However, this peak matches the weight of both carbonyl sulfide, COS+,

+ and of C2H4S , which could come from the cleavage of the dithiolene ligand, and so could not be conclusively assigned in a low resolution MS. Another broad peak is observed above 205 °C at

+ m/z 64.0 and is assigned as S2 . The highest weight fragment is observed as a sharp peak prior to

+ 205 °C at m/z 140.0 and is attributed to tetramethylthiophenium, C8H12S .

Thermolysis of 14 was also performed to investigate side products formed during thermal decomposition. Thermolysis was performed by heating 14 above 200 °C under a constant flow of nitrogen gas and condensing the byproducts downstream at 77 K. These byproducts were then quickly dissolved in deuterated chloroform in order to obtain a 1H NMR spectrum (Figure 3-12).

As expected from TGA-MS, peaks for both 2-butyne (15) and tetramethylthiophene (16) are observed at δ 1.75 ppm and 1.98 and 2.28 ppm, respectively. Another major peak observed at δ

2.17 ppm is assigned as 4,5-dimethyl-1,3-dithio-2-one (17).198 Compound 17 has previously been used in the synthesis of nickel and tin dimethyldithiolene under basic conditions through a reversible ring opening reaction.199, 200 A small amount of undecomposed 14 was also obtained

and may indicate some mild volatility of 14. Several other minor peaks were also observed in the

1H NMR spectrum, but were not identified.

Figure 3-11. The TGA and TGA-MS traces of 14.

From these data, it appears that 14 undergoes two separate, but simultaneous decomposition pathways, as 16 and 17 could not both be formed from 14 while still producing

WS2 (Figure 3-13). Pathway 1 describes a possible route to WS2 through the formation of 16. It

has previously been reported that nickel dithiolenes can undergo reactions with alkynes to produce tetrasubstituted thiophenes, including tetramethylthiophene, through a dithiadiene intermediate.201 A similar pathway is assumed for 14 in which one dithiolene undergoes cycloreversion into 2-butyne which can coordinate to the tungsten center. This 2-butyne can then react with the other dithiolene to form the dithiadiene intermediate, which then readily loses elemental sulfur to form 16. In pathway 2 both 15 and 17 are formed. In this pathway, one of the dithiolene inserts into one of the carbonyls and then forms 17 through reductive elimination and ring closure. This pathway also requires a cycloreversion of the remaining dithiolene to produce

15 and WS2. It is currently unknown at which stage in either of these reaction pathways that the carbonyl(s) dissociates.

Figure 3-12. 1H NMR spectrum of the byproducts from the thermolysis of 14.

Figure 3-13. Two possible decomposition pathways for 14.

With this decomposition data in hand, depositions of WS2 from 14 were attempted using

AACVD. Initial depositions from 14 were carried out prior to reactor optimization and without rigorous exclusion of air and water from the solvent. Deposition under these conditions resulted in deposition of WOx as determined by Raman spectroscopy (Figure 3-14a). This WOx contamination also persisted when dried and deoxygenated solvents were used if there was a minor leak in the reactor which allowed for atmospheric oxygen and water contamination. This indicates that the intermediates formed from the decomposition of 14 are significantly more sensitive to oxidation than the intermediates formed from 13. Once the reactor conditions were optimized and the solvent was rigorously dried and deoxygenated then the deposit showed the

1 -1 -1 expected E 2g/2LA(M) peaks at 350 cm and the A1g peak at 420 cm (Figure 3-14b). These initial results are promising for the use of metal dithiolenes as single source precursors for the deposition of metal sulfides and experiments are still ongoing. Further investigations with tungsten dithiolenes will also include the mono and tris dithiolene complexes, W(CO)4(S2C4H6)

202 and W(S2C4H6), respectively.

Figure 3-14. Raman spectra of the deposits from 14. (a) WOx deposits from initial deposition experiments. (b) WS2 deposits from after optimization of the deposition conditions.

Tungsten Disulfide Monolayers

As previously mentioned, there are many ways to deposit monolayers of WS2 and other

TMD materials. Of these methods, one of the most widely used methods is powder vaporization deposition from sulfur and MO3 (M = Mo or W) powders. Several common variants of this method are illustrated in Figure 3-15 and they share several common characteristics. These depositions are done in hot walled tube furnaces that include two heating zones: a low temperature furnace for the sulfur powder and a downstream high temperature furnace for the

MO3 powder and deposition. Variation A shows one of the early examples in which a pre-

157 deposited thin film of MO3 is then sulfurized by the sublimed sulfur powder. Variation B shows a variation that foregoes the pre-deposition of MO3 and places the substrate downstream

203 of a crucible containing the MO3 powder. A final common variation is shown as variation C

204 and utilizes a substrate placed face down onto a crucible containing the MO3 powder. One of the main drawbacks of these depositions is the need for high temperatures, typically above 800

°C. If this deposition could be performed at lower temperatures then monolayers of TMDs could be directly deposited on a wider variety of substrates.

Figure 3-15. Powder vaporization variants used for the deposition of TMD monolayers.

We have attempted to adapt these methods for use with compound 13 in our cold-wall reactor. Due to the sulfur rich nature of 13, no excess crucible of sulfur powder should be needed for the attempted depositions of monolayers. Several issues occur when trying to adapt some of the above variants to our vertical, cold-wall reactor and using a single source precursor.

Variant A should work regardless of reactor type. However, there are difficulties in quantifying the amount of 13 that has been deposited on to the substrate prior to heating and initial attempts at spin coating 13 onto a silicon substrate have appeared to have failed. A vertical reactor cannot be set up for variant B as it requires a clear pathway for volatile MOx fragments to flow downstream of the gas flow onto the substrate. Variant C can be easily adapted to our reactor

using 13 as a single source precursor. This was done by loading a small amount of 13 into a porcelain crucible, topping it with a cleaned, face-down substrate, and carefully placing them onto the heating stand of the reactor (Figure 3-16). In a typical setup, the reactor is purged with vacuum for one hour and then returned to atmospheric pressure under a flow of ultra-high purity nitrogen gas. A microliter syringe and a toluene solution of 13 (1 mg/mL) could be readily used to prepare crucibles with microgram quantities of 13. When this solution was used, the toluene solvent was evaporated under ambient conditions for two hours prior to loading the crucible into the reactor.

Figure 3-16. Representative diagram of the adapted powder vaporization deposition.

Initial attempts to deposit monolayers were performed using 20 µg of 13 at 300 °C under nitrogen gas for 150 minutes. These deposition conditions were tested with a variety of substrates (silicon, gallium nitride, sapphire, and silica) that have all been reported to promote

205-208 epitaxial growth of TMD monolayers. Deposition of WS2 from 13 appears to be consistent over the substrates used with circular particles between 1-5 µm observed for all substrates

(Figure 3-17). These morphologies are strikingly different than the triangular crystals often seen

140 by optical microscopy for deposits of WS2 monolayers in previous reports. However, it has been calculated that a dodecagonal morphology is favored over the typical trigonal morphology

209 for MoS2 when molybdenum rich deposition conditions are used. It is therefore possible that

that the five excess equivalents of sulfur in 13 are not enough to function in the same way as the large excess of sulfur powder used in typical powdered vaporization experiments to deposit trigonal morphologies.

Figure 3-17. Optical microscopy images from the adapted powder vaporization method on various substrates from 13. (a) Silicon with native silicon oxide layer. (b) Gallium nitride. (c) Sapphire. (d) Silica.

Deposition was also attempted under a flow of form gas (96% nitrogen gas and 4% hydrogen gas) as hydrogen gas has previously been noted to improve the growth of WS2 monolayers.210 These depositions were carried out onto silicon substrates at 350 °C for 90 minutes with between 5 – 40 μg of 13 loaded into the crucible (Figure 3-18). The deposits under these conditions when 20 μg of 13 was used appeared very similar to that of the deposits under the high purity nitrogen gas. Similar circular morphologies were also observed when 40 μg of 13

was used. However, these deposits were not monodisperse and two distinct size regimes of the deposits can be observed. When using smaller quantities of 13 (5 and 10 μg) the coverage of the deposits was significantly sparser than when larger amount of 13 were used. The overall size of the deposited particles appears to be similar in size to those observed when 20 μ of 13 is used.

This could potentially come from the toluene solution pooling as it evaporates, leaving behind localized high concentration areas of 13 in the crucible.

Figure 3-18. Optical microscopy images for powder vaporization deposits from 13 using form gas. (a) 5 μg of 13. (b) 10 μg of 13. (c) 20 μg of 13. (d) 40 μg of 13.

Atomic force microscopy (AFM) was used to determine the height of the deposits grown under form gas. Particle heights appear to be lowest in the 5 μg deposits and most dispersed in sizes for the 10 μg deposits. However, due to the sparse coverage of these deposits only a small number of the particles (<10 discrete particles) were surveyed, which is not enough to draw any

definitive conclusions. The particles from the 20 μg deposits had a mean height of 207 nm with a standard deviation of ± 71 nm. Particles from the 40 μg deposits have a lower mean height of

178 nm, but a much larger deviation of ± 115 nm, likely coming from the two size regimes of particles. All of these deposits had means much greater than the expected height of a single monolayer (approximately 1 nm) and attempts to deposit monolayers from 13 are ongoing.

Summary of WS2 Deposition

Compound 13 was studied as a single source precursor for the AACVD of WS2. The thermal properties of this compound were investigated using TGA and TGA-MS. It was determined that 13 decomposed in a three-step process. The first step of decomposition affords

W(S)2(S2CNEt2)2 which could be isolated as the previously reported dimeric structure

[W(S)2(S2CNEt2)2]2 upon thermolysis of 13 at 200 °C. This was followed by slow decomposition until 400 °C when the weight % became stable close to the expected residual mass of WS2. Several fragments from the dithiocarbamate ligand could be observed by TGA-

MS at both the initial (200 °C) and final (400 °C) decomposition steps. Pure WS2 was deposited by AACVD from 13 at temperatures above 350 °C from AACVD.

Compound 14 has been investigated as the first dithiolene complex to be used as a single source precursor for a metal sulfide deposition. From TGA-MS and thermolysis experiments, 14 is expected to undergo two simultaneous decomposition routes which can both lead to WS2.

Deposition of 14 using AACVD results in WS2 by Raman spectroscopy when air and water are carefully excluded. Further deposition studies with 14 are ongoing with the assistance of Frank

Scheffler.

Growth of WS2 monolayers from a variation of the common powder vaporization method has been attempted using 13. These depositions result in circular deposits severel microns in

diameter. AFM reveals these deposits to be much larger than a single monolayer. Studies will be continued to optimize conditions for monolayer growth.

Figure 3-19. AFM images from powder vaporization deposition of 13 under form gas. (a) 5 μg of 13. (b) 10 μg of 13. (c) 20 μg of 13. (d) 40 μg of 13.

CHAPTER 4 INTRAGROUP COLLABORATIONS

Preface

The above work on the deposition of WS2 from single source precursors has led to several other related projects with collaborators within the McElwee-White lab. Several other ligand sets were investigated for the AACVD of metal sulfides, particularly N,N-disubstituted-

N-acylthioureas.211 Deposition of tungsten carbonitride was performed using AACVD from a novel single source precursor.212 Finally, a volatile single source precursor that has been synthesized by Nathan Ou was deposited using both AACVD and conventional CVD and the nature of these deposits has been investigated with XPS. The sections of this chapter highlight these intragroup collaborations in detail.

N,N-Disubstituted-N-acylthioureas for the Deposition of Metal Sulfides§**213

Transition metal sulfides are a promising class of materials that show potential in a wide range of energy and photonic related applications.214, 215 The materials properties of the metal sulfide are important when considering appropriate applications. For example, crystalline materials are typically considered better for photonic applications.216 Meanwhile, amorphous metal sulfides have been shown to be promising for their use in lithium ion batteries and as electrocatalysts.217, 218 In attempts to control the properties of the materials, a multitude of strategies have been employed in their synthesis. Among the techniques that have been explored are variants of solvothermal processes and chemical vapor deposition (CVD).25 For these methods, single source precursors, in which both the metal and sulfide sources are the same

§ This work was in collaboration with Zahra Ali, who performed the synthesis of the precursors in this section.

** “N,N-dialkyl-Nʹ-acylthioureas as modular ligands for deposition of transition metal sulfides.” Ali, Z.;† Richey, † N.E.; Bock, D.C.; Abboud, K.A.; Akhtar, K.; Sher, M.; and McElwee-White, L., Dalton Trans., 2018, 47, 2719- 2726. †Authors contributed equally.

compound, are often able to produce the metal sulfide at lower temperatures and with better control of stoichiometry.219 Many of these single source precursors lack the volatility to be used in conventional CVD, where volatilization of neat precursor occurs in a bubbler. However, this limitation can be overcome by using AACVD, in which the precursor is dissolved in an appropriate solvent and nebulized for transport to the substrate.27, 220 Since volatility is not critical for AACVD precursors, the technique allows exploration of ligand chemistry that could not be used in conventional CVD. Study of new ligands can provide a means to tune the stoichiometry of the deposited material, increase atmospheric stability, and control the decomposition temperature of the precursor.221

Currently, there are several ligands that are commonly used in single source precursors for the AACVD of metal sulfides. Metal thiolates have been used to deposit metal sulfides at temperatures as low as 150 °C, however their reactivity with air and water makes them difficult to work with.222, 223 Dithiocarbamate ligands have been widely used in deposition of a wide range of metal sulfides, such as iron, copper, cobalt, nickel, zinc, molybdenum, tungsten, and tin.39, 177, 224, 225 These complexes are often stable to air and water, and so do not require special handling. However, this comes at the price of higher deposition temperatures, typically above

300 °C. Xanthate ligands have similarly found use with a variety of metals and often decompose at lower temperatures than their dithiocarbamate analogues.226-228 This lowered decomposition temperature of the xanthate precursors comes from the Chugaev elimination mechanism pathway by which these ligands can readily decompose to their corresponding metal sulfide.229

Thiobiuret and dithiobiuret complexes have also found use for the deposition of a wide variety of late transition metal sulfides and show no oxygen incorporation from the thiobiurets.230, 231

Several other ligands, such as thioureas and dithiophosphates, have also been used though less commonly.232, 233

N,N-disubstituted-Nʹ-acylthiourea ligands have long been known to coordinate to many transition metals in a chelating fashion similar to acetylacetonate.234, 235 The TGA of several of these complexes is consistent with thermal decomposition to the corresponding metal sulfide or pure metal.236, 237 However, these complexes are typically not sufficiently volatile for conventional CVD. There are a few reports of their use in AACVD but these studies do not explore the effects on decomposition temperature from varying the substituents on the ligand.238

Herein, we investigate the impact of the N-substituents, R and R’ on decomposition temperatures of Ni complexes and demonstrate that these ligands can be used for deposition of metal sulfides via AACVD.

Results and Discussion

To investigate the effect of substituents, we synthesized and compared four nickel complexes with ligands derived from two acid chlorides, benzoyl and cinnamoyl chloride, and two secondary amines, diphenyl- and diisopropylamine. The N,N-disubstituted-Nʹ-acylthiourea ligands are readily synthesized in a two-step, one-pot reaction by mixing the acid chloride and sodium thiocyanate, followed by addition of the secondary amine in dry acetone (Figure 4-1).239

After purification by recrystallization from an appropriate solvent mixture, the ligand was deprotonated with a mild base, such as sodium acetate, and reacted with a metal salt to form the

240 respective ML2 or ML3 complex in good yields. The wide range of commercially available acid chlorides and secondary amines make this family of ligands highly modular. This modularity can be used to tune the properties, such as thermal decomposition temperature, of these ligands through the use of select R and R′.

Figure 4-1. Synthesis of the N,N-disubstituted-Nʹ-acylthioureas and their nickel compounds. Adapted from reference 213 with permission from The Royal Society of Chemistry.

The thermal behavior of these complexes was investigated by both TGA and the first derivative of the TGA plot (DTG) (Figure 4-2). From these data, it was observed that Ni(L3)2

(20) (where R = PhCH=CH, R′ = iPr) has the lowest decomposition temperature, with mass loss occurring most rapidly at 223 °C. An increase of approximately 25 °C in decomposition temperature is observed for Ni(L1)2 (18), in which R is phenyl. A similar increase of approximately 25 °C in the decomposition temperature is observed in comparison of Ni(L4)2

(21) and Ni(L2)2 (19). Varying R′ from alkyl to aryl moiety also caused a significant increase in decomposition temperature, with a difference of approximately 45 °C between 18 and 19 as well as between 20 and 21. This resulted in a total difference of 70 °C in the decomposition of 20 and

19, even though the ligand backbone remained unchanged. This shows that decomposition temperature is directly correlated to the substituents and that both R and R′ need to be carefully considered when designing precursors based on this ligand framework.

Figure 4-2. Thermal properties of 18, 19, 20, 21. (a) TGA. (b) First derivative of the TGA plot. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Fragmentation from mass spectrometry has previously been used as a model for the gas phase decomposition of precursors in CVD, though some care needs to be taken in interpretation.241, 242 Tandem mass spectrometry (MS-MS) was used to elucidate possible decomposition pathways of 20. In this experiment the [M + H]+ ion was selectively trapped and fragmentation was brought on by collision induced dissociation until the parent peak was no longer visible. A list of abundant ions observed in positive mode appears in (Table 4-1). The base peak is observed at m/z 510.93 and is attributed to loss of diisopropylcyanamide ([M + H −

i + N( Pr)2CN] ). The second largest peak is observed at m/z 507.07 and is attributed to the loss of the cinnamoyl moiety ([M + H − PhCH=CHCO]+). In these two most abundant fragments the

Ni-S bond is unaffected. This may be indicative of the preference for these complexes to decompose to the metal sulfide. Another fragment is observed for the

Ni(L3)(diisopropylcyanamide) complex at m/z 473.17 (M + H − PhCH=CHCOSH]+) with a high relative abundance. If this fragment were also formed during the thermal gas phase reaction, the cyanamide would be expected to be labile, similar to what has been reported for acetonitrile adducts of tungsten imido precursors.241 This secondary fragmentation (loss of the cyanamide)

would provide another route to the ion observed at m/z 347.07 ([M + H − HL3]+), corresponding to overall loss of the L3 ligand. Loss of the HNRʹ2 moiety is also observed in good abundance at

i + m/z 535.87 ([M + H − NH( Pr)2] ).

Table 4-1. Fragments from the MS-MS of 20. m/z Fragment Relative abundance 637.07 [M + H]+ 0 i + 535.87 [M + H − NH( Pr)2] 44.3 i + 510.93 [M + H − N( Pr)2CN] 100 507.07 [M + H − PhCH=CHCO]+ 92.1 473.17 [M + H − PhCH=CHCOSH]+ 91.0 347.07 [M + H − HL3]+ 55.7

Since among all nickel complexes, 20 had the lowest decomposition temperature, HL3 was selected for the synthesis and study of several other ML2 and ML3 complexes (M = Ni, Zn,

Co, and Cu). The formation of all the metal complexes could be easily followed with IR spectroscopy during synthesis by observing the disappearance of the amide N-H and C=O stretches of the free ligand. For the diamagnetic complexes, 18, 19, 20, 21, 23, and 24 the disappearance of the N-H proton in 1H NMR could also be followed.

The X-ray crystal structures were determined for all M(L3)2 and M(L3)3 complexes

(Figure 4-3, Table 4-2). Compound 20 showed a square planar geometry with the sulfurs cis to one another and a S1-Ni1-S1A bond angle of 83.83(2)° (Figure 4-3a). The cis isomer has been noted for other nickel complexes of N,N-disubstituted-Nʹ-acylthioureas as well and by analogy,

238 we assign the predominant isomer for 18, 19, and 21 as cis. The crystal structure of Cu(L3)2

(22) contained two copper centers, one ordered and the other disordered, and two disordered tetrahydrofuran molecules from the crystallization process resulting in an overall formula of

Cu(L3)2·THF. Complex 22 also showed square planar geometry, however the coordination around copper was trans with respect to the sulfurs with a S1-Cu1-S1A angle of 180.0° (Figure

4-3b). The difference in cis and trans coordination around nickel and copper, respectively, has been observed previously for similar bis(thiobiuret) complexes.231 The cobalt(II) ion in the starting material CoCl2 was oxidized during the synthesis of Co(L3)3 (23), consistent with the diamagnetic 1H NMR spectrum. The crystal structure shows 23 as the facial isomer with S-Co-S angles ranging from 86.31(2) to 89.47(2)° (Figure 4-3c). Compound Zn(L3)2 (24) crystallized in a distorted tetrahedral geometry indicated by the S1-Zn-S2 angle of 120.76(4)° (Figure 4-3d). A noninteracting acetonitrile molecule was also present in the asymmetric unit from the crystallization process (Figure 4-3).

Figure 4-3. X-ray crystallographic structures of (a) 20, (b) 22, (c) 23, (d) 24. Selected solvent molecules and disorder removed for clarity. Full structures are given in appendix B. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Table 4-2. Selected X-ray crystallographic data for 20, 22, 23, 24.

20 22 23 24·NCCH3 Formula C32H42NiN4O2S2 C36H50CuN4O3S2 C48H63CoN6O3S3 C34H45ZnN3O2S2 crystal system Tetragonal Triclinic Monoclinic Monoclinic Space group I-4 P-1 P2I/c P2I/n a (Å) 14.662(2) 10.5648(7) 9.6733(4) 13.8922(16) b (Å) 14.662(2) 11.9358(8) 26.7438(10) 7.4030(9) c (Å) 15.487(2) 15.2248(10) 18.9117(7) 34.250(4) α (deg) 90 79.6830(10) 90 90 β (deg) 90 72.2510(10) 92.5405(9) 99.859(2) γ (deg) 90 84.0150(10) 90 90 volume (Å3) 3329.2(11) 1796.4(2) 4887.7(3) 3470.4(7) 3 Dcalc(Mg/m ) 1.272 1.173 1.260 1.312 total reflns 36001 36766 114933 34093 unique reflns 5829 8897 18743 8616 GOF of F2 1.029 1.025 1.016 0.829

While the coordination around the metal center varies significantly among these complexes, several common characteristics are observed (Table 4-3, Table 4-4). As expected, the metal-sulfide bonds were significantly longer, approximately 0.3 Å on average, than the metal-oxide bonds due to the increased ionic radius of the sulfur. The bond lengths between the core nitrogen (N1, N21, or N41) and the neighboring carbons (C1/C2, C21/C22, or C41/C42) do not differ significantly, indicating resonance delocalization through the ligand core. The ligand backbone was distorted from the planarity expected for the free ligand. The dihedral angle between the OCN and SCN planes (Figure 4-4), where N is the core nitrogen of the backbone

(N1, N21, or N41) is less pronounced in 20 and one of the ligands in 23 with angles of

13.536(161) and 8.980(360)°, respectively, but was larger in Zn(L3)2·NCCH3 with angles of

42.087 (0.413) and 50.317(416)°. This dihedral angle also varied significantly for ligands on the same metal center and range between 8.980(360), 34.806(211), and 20.638(373)° on the 23 complex. These factors also cause the C-N-C angle in the ligand to open up to values of

123.69(15) to 126.8(3)°.

Table 4-3. Selected bond distances (Å) for 20, 22, 23, 24. M-O bond M-S bond Backbone C-N Compound Bond lengths lengths Bond lengths 20 Ni1-O1/S1 1.8719(9) 2.1424(4) C1-N1 1.3283(16) C2-N1 1.3441(16) 22a Cu1-O1/S1 1.9093(13) 2.2709(4) C1-N1 1.323(2) C2-N1 1.354(2) 23 Co1-O1/S1 1.9080(14) 2.2106(6) C1-N1 1.335(3) Co1-O21/S21 1.9463(13) 2.2243(5) C2-N1 1.339(3) Co1-O41/S41 1.9255(13) 2.2032(6) C21-N21 1.337(2) C22-N21 1.340(2) C41-N41 1.329(3) C42-N41 1.336(4)

24·NCCH3 Zn1-O1/S1 1.968(3) 2.2706(11) C1-N1 1.324(5) Zn1-O21/S21 1.959(3) 2.3083(12) C2-N1 1.351(5) C21-N21 1.313(5) C22-N21 1.359(5) aLengths from the ordered molecule within the unit cell.

Table 4-4. Selected bond angles (degrees) for 20, 22, 23, 24. Backbone S-M-S Dihedral Dihedral Compound Bonds Bonds C-N-C angles intersect angles angles 20 S1-Ni1-S2 83.83(2) C1-N1-C2 124.36(11) N1 13.536(161)

22a S1-Ni1-S2 180.0 C1-N1-C2 125.10(16) N1 32.825(221) 23 S1-Co1-S42 89.47(2) C1-N1-C2 125.53(17) N1 8.980(360) S1-Co1-S42 86.76(2) C21-N21-C22 123.69(15) N21 34.806(211) S21-Co1-S42 89.31(2) C1-N1-C2 126.3(2) N41 20.638(373) 24·NCCH3 S1-Zn1-S2 121.76(4) C1-N1-C2 126.8(3) N1 42.087 (0.413) C21-N21-C22 126.4(3) N21 50.317(416) aAngles from the ordered molecule within the unit cell.

TGA and DTG of 22, 23, and 24 showed similar decomposition temperatures to that of

20 (Figure 4-5). From the DTG trace, both 23 and 22 appear to undergo a multistep decomposition pathway. With these relatively low decomposition temperatures, these compounds were precursor candidates for the AACVD of their corresponding metal sulfide.

Toluene solutions of 0.65 mM were prepared for the 20, 23, and 24 complexes and depositions onto silicon substrates were carried out at 350 °C. Compound 22 was inadequately soluble in toluene for these depositions and attempts to deposit 22 from a tetrahydrofuran solution at 350

°C were unsuccessful, as indicated by absence of the metal sulfide on the substrate by energy dispersive X-ray spectroscopy (EDX) or XRD.

Figure 4-4. Representative example of the dihedral angle calculation. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Figure 4-5. Thermal properties of 20, 22, 23, 24. (a) TGA. (b) First derivative of the TGA plot. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Deposition from 20 resulted in a mixture of individual nanorods and thin sheets, which appeared to be fused from several nanorods (Figure 4-6a). Growth from 23 resulted in the deposition of platelets perpendicular to the substrate with an underlying thin film (Figure 4-6b).

Compound 24 appeared to give the most conformal film, with minimal surface features (Figure

4-6c). EDX of these depositions resulted in an M:S ratio ranging from 1.0:0.84-1.06. This is within the error for the metal monosulfide deposition, however this is also within range of the sulfur deficient Ni9S8 and Co9S8 phases, which are also known. PXRD revealed mostly amorphous deposits, with only ZnS showing some crystallinity by the presence of a small peak at

28.5° which is indicative of the ZnS (111) reflection (Figure 4-6d). Several peaks from the silicon substrate were also observed in the NiS and ZnS deposits. No increase in crystallinity was observed after annealing at 650 °C for one hour.

Figure 4-6. Deposition results from 20, 23, 24. (a) SEM image of NixSy. (b) SEM image of CoxSy. (c) SEM image of ZnS. (d) XRD pattern of the deposits. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Summary

In summary, we have demonstrated that the thermal decomposition temperatures of bis(N,N-disubstituted-Nʹ-acylthiourea)nickel complexes vary with the substituents R and R′.

Compound 20 (where R is PhCH=CH and Rʹ is isopropyl) had the lowest decomposition temperature, with the most rapid mass loss occurring at 223 °C. The related complexes 22, 23, and 24 were all synthesized and found to crystalize in the trans-square planar, fac-octahedral, and distorted tetrahedral geometries, respectively. Compounds 20, 23, and 24 were found to deposit their corresponding metal sulfide from AACVD at temperatures of 350 °C but attempted depositions from solution of 22 were unsuccessful due its low solubility. These experiments demonstrate the viability of N,N-disubstituted-Nʹ-acylthiourea as modular ligands for the

AACVD of metal sulfides.

††‡‡212 Deposition of WCxNy from 25 using Aerosol Assisted Chemical Vapor Deposition

Tungsten carbonitride (WCxNy) has been considered as a candidate for copper diffusion barriers in integrated circuits due to its low electrical resistivity, high thermal stability, and minimal chemical reactivity with existing circuit materials.243 Previously reported single source precursors for deposition of WNxCy have utilized N-bound ligands such as nitrido, imido, hydrazido, amido, guanidinato, and amidinato moieties to provide the nitrogen in the resulting material.244-246 Optimization of the precursors has primarily targeted low temperature deposition

(<350 °C) to minimize and thermal damage to the substrate. Our studies of tungsten nitrido complexes of the type WN(NR2)3 demonstrated that the W≡N and W–NR2 moieties are

†† This work was performed in collaboration with Michelle Nolan and Alex Touchton, who performed the synthesis and characterization of the guanidinato complexes.

‡‡ “Synthesis and Characterization of Tungsten Nitrido Amido Guanidinato Complexes as Precursors for Chemical Vapor Deposition of WNxCy Thin Films.” Nolan, M.N.; Touchton, A.J.; Richey, N.E.; Ghiviriga, I.; Rocca, J.R.; Abboud, K.A.; and McElwee-White, L., Eur. J. Inorg. Chem. 2018, 46–53.

acceptable sources of N in depositions carried out at temperatures as low as 125 °C. In addition, variation of the substituents in the amido moieties of WN(NR2)3 provided a means to tune the precursor volatility.247, 248

A general strategy to improve the thermal stability of a CVD precursor while maintaining volatility is the addition of chelating ligands, such as guanidinato ligands.249 Guanidinato ligands are N-bound, bidentate ligands which have been extensively employed in CVD precursors due to their electronic versatility, thermal stability, and nitrogen contribution to deposits.250 Late transition metal guanidinates have been used in deposition of metal films, including Cu and Ni.251 However, early transition metal guanidinate precursors have yielded metal nitride and metal carbonitride thin films under CVD conditions. Mixed guanidinato amido complexes of Ta and Nb have been employed in the CVD of TaN and NbN thin films. Growth of

252 ZrNxCy and TiNxCy films was achieved from bis(guanidinato) complexes.

Our previous study of the tungsten imido guanidinato precursor

i i W(N Pr)Cl3[( PrN)2CNMe2] demonstrated that guanidinato ligands are suitable for use in precursors for the AACVD of WNxCy. We now report the synthesis of tungsten nitrido amido

i guanidinato precursors along with a demonstration of the use of WN(NMe2)[(N Pr)2C(NMe2)]2

(25) as a single source precursor for the AACVD of WNxCy thin films. Guanidinato complexes of a variety of metals have been prepared by insertion of N,N′-dialkylcarbodiimides

(RN=C=NR) into metal amide bonds. A relevant example is the preparation of group V metal imido amido guanidinato complexes by selective insertion of carbodiimides into Ta–amido and

Nb–amido bonds in the presence of an imido ligand.253 This synthetic strategy is attractive for preparation of derivatives of WN(NMe2)3 because it utilizes commercially available carbodiimides.

Results and Discussion

Reaction of WN(NMe2)3 with two equivalents of N,N′-diisopropylcarbodiimide (DIC)

i (Figure 4-7) afforded the nitrido amido guanidinato complex WN(NMe2)[(N Pr)2C(NMe2)]2

(25). Reducing the amount of DIC to one equivalent did not result in formation of a mono- guanidinato complex. Complex 26 was prepared analogously with N,N′- dicyclohexylcarbodiimide (DCC). The attempted synthesis of N-tert-butyl guanidinato derivatives of 25 by addition of N,N′-di-tert-butylcarbodiimide to solutions of WN(NMe2)3 resulted in recovery of starting material, even after extended reflux at temperatures up to 115 °C, suggesting steric limitations to the insertion reaction. An alternate route starting from

WN(NEt2)3 was utilized to synthesize complexes 27 and 28 through the insertion of DIC and

DCC.

Figure 4-7. Synthesis of the tungsten guanidinato complexes. Adapted from reference 212 with permission from John Wiley & Sons.

Because CVD is a thermal process, examining the pyrolysis of a precursor under non- deposition conditions can provide valuable insight into possible breakdown pathways which occur during deposition. The thermal decomposition of 25 was investigated by pyrolysis of neat samples and by TGA. Decomposition products from solid state pyrolysis were isolated by heating a neat sample of 25 to 200 °C under 1 atm of N2 and condensing the volatile components at 77 K. Analysis of the condensate by 1H NMR, 13C NMR, FTIR, and GC/MS revealed the

presence of dimethylamine and diisopropylcarbodiimide (Figure 4-8). Isolation of DIC indicates that the deinsertion of carbodiimide is an accessible decomposition pathway. This deinsertion would regenerate the original dimethylamido ligand, which could then decompose further through pathways previously established for WN(NMe2)3. For example, the presence of

247 dimethylamine is consistent with the thermolysis of WN(NMe2)3 under similar conditions.

Analogous deinsertion has been reported for other guanidinato complexes.251 The TGA of 25 indicates an initial decomposition step at 200 °C (Figure 4-9). The residual mass of 63% corresponds to the loss of one guanidinato ligand and one amido ligand, consistent with the products observed by GC/EI-MS.

Figure 4-8. Detected byproducts from the thermolysis of 25. Adapted from reference 212 with permission from John Wiley & Sons.

100%

90%

80%

70%

60% Weight%

50%

40%

0 100 200 300 400 500 600 Temperature (C)

Figure 4-9. The TGA plot of 25. Adapted from reference 212 with permission from John Wiley & Sons.

Compound 25 was tested as a single source precursor for the AACVD of WNxCy. The

AACVD was performed from heptane solutions of 25 at substrate temperatures between 200-400

°C, resulting in dark, iridescent deposited films. Films exhibited good adhesion to the Si substrates and passed a qualitative peel test using adhesive tape. Deposited material at all growth temperatures was found to be amorphous by XRD. The morphologies of the deposits were determined by SEM, as shown in Figure 4-10. At temperatures of 200 and 300 °C, films resembled Stranski–Krastanov type growth: a layer of material covered the substrate surfaces with islands of growth appearing on top. At 400 °C, these islands appeared to grow into more pronounced vertical features. Cross-sectional SEM indicated average film thicknesses of 29, 25, and 14 µm at deposition temperatures of 200, 300, and 400 °C, respectively.

Figure 4-10. Low- and high-resolution micrographs from the deposition of WCxNy from 25. (a,b) Deposited at 200 °C. (c,d) Deposited at 300 °C. (e,f) Deposited at 400 °C. Adapted from reference 212 with permission from John Wiley & Sons.

The compositions of the deposits were determined by XPS. Nitrogen content of the deposits ranged from 11.6 to 17.0 atomic %, while tungsten comprised approximately 20 atomic

%. Carbon accounted for approximately 20 to 30 atomic % of the deposited material. These deposits showed higher N and C content, but lower W content, than the deposits previously

247, 248, 254 grown from WN(NR2)3 complexes. Oxygen was present in significant amounts in all films and was attributed to post-deposition oxidation of the samples when handling in air prior to obtaining XPS. Due to the amorphous nature of the films, significant in-diffusion of oxygen is possible post-growth. Although depth profiling was not carried out for these samples, we have previously determined by Auger electron spectroscopy depth profiling that oxygen content in our

255 WNxCy films is highest in the upper layers, consistent with oxidation during handling.

High-resolution XPS was taken to determine the chemical environment of the different elements by comparison to known data.256 Deconvolution of the W 4f signal revealed two signals for deposits grown at 400 °C, and three peaks at 200 and 300 °C (Figure 4-11a). For all deposits, a W 4f7/2 peak was observed with a binding energy (BE) of 33.2 eV with the corresponding W 4f5/2 peak at a BE of 35.3 eV. Peaks were also observed in all deposits with

BEs of 35.4 eV, representing W 4f7/2, and 37.6 eV for the W 4f5/2. For deposits grown at 300 °C and below, W 4f7/2 and 4f5/2 peaks were observed with BEs of 34.2 and 36.4 eV, respectively.

The W 4f7/2 peaks at BEs of 33.2, 34.2, and 35.4 eV are consistent with values reported for W in the +4, +5, and +6 oxidation states, respectively. The W 4f7/2 peaks at 35.4 and 33.2 eV have previously been observed from deposits grown from WN(NR2)3 complexes. It is likely that the

W 4f7/2 peak at 35.4 eV and corresponding W 4f5/2 peak at 37.6 eV are from WO3 contamination from post growth oxidation. However, no deposit from 25 showed a peak at 31.5 eV which has

254 previously been reported for W metal observed in depositions from WN(NR2)3. It is possible that any W metal formed in these samples was oxidized during handling.

Deconvolution of the N 1s signal revealed two peaks at all temperatures (Figure 4-11b).

A peak with a BE of 400.0 eV was observed at all temperatures and has been assigned as CNx, which typically exhibits BEs of 398.5 to 400.6 eV. A second peak was observed at BE 397.4 eV, corresponding to WN. This peak comprised most of the N at deposits grown at 400 °C and was attributed to more extensive decomposition of the precursor than was seen in deposits from lower temperature growths. Likewise, the C 1s signal was deconvoluted into two distinct signals

(Figure 4-11c). Amorphous C is typically observed at a BE of 284.6 eV and can be observed at all growth temperatures. This peak was more intense in deposits grown at 400 °C, which is consistent with increased solvent decomposition during higher temperature growth. A peak at

BE 286.2 eV was also observed at all growth temperatures and was attributed to CNx in the deposits. Interestingly, no peaks were observed for carbidic C, which has a characteristic BE of

283.5 eV. Oxide was observed in samples from all growth temperatures due to the rapid oxidation of WNxCy deposits when handling the samples in air prior to analysis. This oxygen peak was assigned as WO3 with a BE of 530.7 eV and used as the internal reference for samples after sputtering (Figure 4-11d).

Summary

In conclusion, several tungsten nitrido amido guanidinato complexes of the type

WN(NR2)[(NR′)2C(NR2)]2 have been prepared and characterized. The thermal decomposition of

i WN(NMe2)[(N Pr)2C(NMe2)]2 (25) was studied and it was found to decompose through deinsertion of the diisopropylcarbodiimide and loss of dimethylamine. Compound 25 was found to be useful as a single source precursor for amorphous WCxNy thin films from AACVD at temperature above 200 °C.

100

Figure 4-11. XPS of the deposits from 25. (a) Region of W 4f peak. (b) Region of the N 1s peak. (c) Region of the C 1s peak. (d) Region of the O 1s peak. Adapted from reference 212 with permission from John Wiley & Sons.

§§ Chemical Vapor Deposition of WOx from Volatile Single Source Precursors

While AACVD provides a useful route to investigate a wide range of interesting ligand sets not compatible with conventional CVD, there are several drawbacks to industrializing the

AACVD process. One drawback is the increase in waste generated through the use of an aerosol solvent and another is challenges of the scalability of typical AACVD experiments.257 However, a variety of ligand types could be initially screened using AACVD and then those found to be most promising could be adapted for use in CVD. One strategy would be synthesizing volatile derivatives of those promising ligand types. In general, volatility is determined by intermolecular forces. One general method in reducing intermolecular forces is to use heavily fluorinated ligands which help to repel nearby molecules due to their high electron density.66

§§ This work was done in collaboration with Nathan Ou and Duane Bock. N.O. synthesized and characterized compound 29 and assisted with the CVD experiments. D.B. performed the AACVD experiments.

101

Therefore, a reasonable strategy of modifying promising AACVD precursors to be used in conventional CVD is to use fluorinated analogues of the selected ligand sets.

Recently it was shown that several oxo-fluoroalkoxide tungsten(VI) complexes could be used as single source precursors for deposition of tungsten oxide (WOx) at temperatures between

250-550 °C.258, 259 By TGA, these precursors were found to decompose before sublimation at atmospheric pressure. However, they were volatile enough to be sublimed at reduced pressures of between 300-400 millitorr with heating (Figure 4-12). This class of precursors was further improved with the addition of a κ2-β-diketonate in order to increase their air and water stability by creating a full octahedral coordination sphere around the tungsten center.260, 261 Some of the fluoalkoxide based complexes from this series were found to moderately volatile, with some undergoing thermal sublimation at atmospheric pressure during the TGA experiments.260 The high volatility of these compounds makes them promising precursors for conventional CVD.

This is particularly important for the above family of compounds, as their deposits were found to contain carbon contamination at deposition temperatures above 400 °C. It was not possible to determine the source of the carbon contamination in these deposits as being from the precursor itself or from the AACVD solvent used.259-261

85-90 °C 45 °C 165 °C 400 mTorr 300 mTorr 760 Torr

Figure 4-12. Examples of some fluorinated alkoxide precursors with their reported sublimation temperatures and pressures.

102

Recently, we have used hexafluoroacetylacetonate (hfac) to synthesize a more heavily fluorinated derivative of this family of compounds, WO(OC(CF3)2CH3)3(hfac) (29). This compound was found to sublime at atmospheric pressures beginning at approximately 65 °C during the TGA experiment. This was noted by a rapid, nearly complete loss of mass (Figure 4-

13). It could also be readily sublimed at room temperature when placed under vacuum. Due to the complete sublimation of the compound by TGA, the exact thermal decomposition temperature could not be determined. Depositions were attempted from 29 using both low pressure CVD and AACVD at 500 °C to investigate any difference between the two techniques.

Figure 4-13. TGA plot of 29. Inset is compound 29.

For deposition by conventional CVD, a stainless steel bubbler was loaded with 0.50 g of

29 and attached to the CVD reactor and heated to 75 °C. Deposition proceeded at 200 Torr with a flow rate of ultra-high purity nitrogen gas of 200 standard cubic centimeters per minute (sccm) for a total exposure time of 150 minutes. The AACVD deposition was carried out using a 0.057

M solution of 29 in anhydrous diglyme. A total of 10 mL of this solution was loaded into a gastight syringe and injected into the nebulizer at a rate of 4 mL per hour for a total deposition time of 150 minutes. For AACVD, the reactor was maintained at 350 Torr and a flow rate of

1000 sccm was used.

103

Previously, XPS has been used to determine the elemental composition of the WOx deposits and determine the amount of carbon contamination. The films deposited from both methods show approximately 20 atomic % carbon contamination prior to any sputtering (Figure

4-14). After ion sputtering the films for two minutes to remove the surface impurities of the deposits this carbon contamination was fully removed. This indicated that the carbon contamination was localized at the surface of the films and likely resulted from adventitious carbon deposits from atmospheric contaminants. These results indicate that the diglyme solution does not decompose during the deposition process at 500 °C and that the previously reported carbon contaminants likely came from the precursors themselves. The primary difference between these two deposits appears to be in the W and O composition of the films (Table 4-5).

The CVD deposits have a W:O ratio of 1:2.80 which is slightly more oxygen rich than the

AACVD deposits which had a ratio of 1:2.55. However, as these atomic % of the W and O are only 2% different, which may be within the error of the XPS. Because of this, further XPS studies are ongoing to investigate the nature of this difference.

Figure 4-14. Full XPS spectra of the deposits from 29. (a) Deposit from CVD. (b) Deposit from AACVD

104

Table 4-5. Elemental composition of the deposits from 29 as determined by XPS. W atomic % O atomic % C atomic % CVD as deposited 20.5 59.8 21.5 CVD sputtered 26.3 73.7 <1 AACVD as deposited 16.7 61.8 19.7 AACVD sputtered 28.2 71.8 <1

High resolution XPS scans were also performed on the W and O peaks after sputtering to investigate their nature in the deposits by comparison with previous reports of tungsten oxides.262, 263 Both deposits show a set of two doublets in the W4f spectra (Figure 4-15). The doublets at higher binding energies contain the W 4f7/2 peak at 35.6 and 35.4 eV, which is in good agreement with previous reports of W6+ bound to oxides. The lower energy doublet, with the W 4f7/2 peaks at 34.0 and 33.4 eV, corresponds well to lower oxidation state tungsten found in substoichiometric WOx species such as W19O49 or W20O2.9. This lower energy peak comprises more of the spectra for the AACVD deposit, which matches well with the lower W:O ratio also observed for this deposit. The O 1s peak of WOx does not change significantly regardless of the oxidation state of the tungsten.264 Due to the lack of adventitious carbon which is often used as the reference, after sputtering this peak was used as an internal reference and set to a value of

530.5 eV. A higher binding energy shoulder was also observed at values of 531.4 and 531.6 eV for the CVD and AACVD deposit, respectively. This peak is commonly observed in WOx XPS spectra and is considered to be from the surface oxides which are not fully coordinated to tungsten centers. However, it has previously been noted that significant sputtering of WO3 results in reduction of the WO3 to substoichiometric WO3-x and eventually to tungsten metal.

Further studies would be needed to determine the reliability of these high resolution scans and the W:O ratio after sputtering.273

105

Summary

Compound 29 was used to deposit WOx using either CVD or AACVD. After ion sputtering to remove surface contamination, these deposits were found to contain no significant amount of carbon within the deposits. Preliminary XPS of the deposits showed lower oxygen content of the deposit from AACVD than from CVD. This was also apparent in the high resolution XPS scan of the W 4f peaks, as an increase in the peak for lower oxidation state of W.

Further studies are ongoing to determine the effect of depositions at different temperatures as well as further characterization studies.

Figure 4-15. High resolution XPS of the W 4f and O 1s peaks of the deposits from 29. (a) W 4f region of the CVD deposit. (b) O 1S region of the CVD deposit. (c) W 4f region of the AACVD deposit. (d) O 1s region of the AACVD deposit.

106

CHAPTER 5 EXPERIMENTAL PROCEDURES

General Considerations

In general, synthesis was performed using standard Schlenk line techniques using dry, air-free solvents. Reagents were purchased from Sigma Aldrich or Fisher Scientific; deuterated solvents were purchased from Cambridge Isotopes. Nitrogen gas (99.999%) was purchased from

AirGas. Air-free and dry solvents were purified using an MBraun MB-SP solvent purification system and stored over activated 4 Å molecular sieves for at least 48 h prior to experiments. Dry acetone was distilled over CaSO4 and KMnO4 prior to use. All other reagents were used as received. NMR spectra were recorded on a Varian Mercury 300 (300 MHz) spectrometer using residual protons from deuterated solvents for reference. The UV-Vis spectroscopy was performed on a Shimadzu UV-1650PC. All absorptions were measured in benzene or hexanes in a quartz cell. The IR spectra were obtained on a Bruker Vertex 80V or a Perkin Elmer Spectrum

One equipped with an ATR diamond crystal stage. Thermogravimetric analysis (TGA, TA

Discovery5500) was performed under N2 gas with a heating rate of 10 °C/minute. The TGA coupled to a mass spectrometer (TGA-MS, Discovery Mass Spectrometer) was used to determine the masses of byproducts between 1-300 atomic mass units (AMU). Mass spectrometry was performed on an Agilent 6220 ESI-TOF with Agilent 1100 LC with electrospray ionization (ESI) or direct analysis in real time (DART) ionization. Elemental CHN analysis was performed by Robertson Microlit laboratories, New Jersey. X-Ray intensity data were collected at 100 K on a Bruker DUO diffractometer using MoKa radiation (l = 0.71073 Å) and an APEXII CCD area detector.

The elemental compositions of deposited materials were determined by X-ray photoelectron spectroscopy (XPS, ULVAC-PHI XPS) or by energy-dispersive X-ray

107

spectroscopy (FESEM, FEI Nova NanoSEM 430). When noted, samples were sputtered with an ion beam of Ar+ for up to two minutes. The crystallinities of the deposits were measured by X- ray diffraction (XRD, Panalytical X’pert Pro). Morphologies of deposits were determined by field emission scanning electron microscope (FESEM, FEI Nova NanoSEM 430). Raman spectroscopy (LabRAM Aramis) was performed using a 532 nm laser with 40× objective lens.

Copper Precursors

Compound 2 was synthesized as previously described and characterized by comparison with literature data.101 All of the copper borohydride compounds appear to be air stable and can be synthesized without any special handling. In general, the hydrogens bound to boron were not seen in the 1H NMR spectra due to quadrupolar broadening from the boron.265, 266

(DPA)Cu(hfac) (3).

Cu2O (62.1 mg, 0.43 mmol) and DTA (160.1 mg, 0.78 mmol) were stirred in 4 mL of dichloromethane. Hexafluoroacetylacetone (0.09 mL, 0.72 mmol) was then added dropwise to the solution. After stirring for 30 minutes the solution was mostly a transparent yellow color with some Cu2O remaining. The solid was filtered off and the solvent was removed in vacuo. The solid was washed with hexanes and a pale yellow solid was collected to obtain 3 in 68% yield.

1 H NMR (300 MHz, C6H6) δ 2.03 (s, 6H), 6.13 (s, 1H), 6.94 (d, 4H, J = 9 Hz), 7.64 (d, 4H, J = 9

19 -1 Hz). F NMR (282 MHz, CDCl3) δ -79.27 (s). FTIR (neat) 1642, 1993 cm .

(DTA)Cu(hfac) (4).

Cu2O (149.2 mg, 1.04 mmol) and DPA (364.3 mg, 2.04 mmol) were stirred in 5 mL benzene. Hexafluoroacetylacetone (0.28 mL, 1.98 mmol) was then added dropwise to the solution. After two minutes of stirring, the solution became clear green with some Cu2O remaining. The solution was quickly filtered and 5 mL hexanes was added to the filtrate, which was then allowed to stand in the freezer. After several hours, a yellow precipitate formed, which

108

was then filtered and washed with hexanes. The product was dried in vacuo to obtain 4 in 24%

1 yield. H NMR (300 MHz, CDCl3) δ 6.25 (s, 1H), 7.45 (m, 6H), 7.80 (m, 4H). FTIR (neat) 1640,

1990 cm-1. The spectroscopic data were comparable to the reported values.92

(PPh3)2Cu(BH4) (5).

113, 117 This synthesis was adapted from a previously reported procedure. NaBH4 (89.0 mg,

2.35 mmol) was dissolved in 10 mL absolute ethanol. In a separate flask, 6 (1.30 g, 2.01 mmol) was dissolved in 20 mL dichloromethane. The ethanol solution was then added to the dichloromethane solution and stirred for 30 minutes. This solution was then filtered and washed with 4 mL of dichloromethane followed by 8 mL hexanes. The filtrate was allowed to stand at 20

°C for several hours to produce crystals. This crude product was collected and recrystallized in 8 mL dichloromethane and 2 mL hexanes at -20 °C. The solid was then filtered and dried in vacuo

1 31 to obtain 7 in 56% yield. H NMR (300 MHz, C6D6) δ 6.93 (m, 18H), 7.47 (m, 12H). P NMR

-1 (121 MHz, C6D6) δ -1.19 (broad). FTIR (neat) 1140, 1936, 1989, 2265, 2378, 2399 cm . The spectroscopic data were comparable to the reported values.113

(PPh3)2Cu(NO3) (6). Procedure 1.

This synthesis was adapted from a previously reported procedure.117 In a flask equipped with a condenser, PPh3 (10.5 g, 40 mmol) was heated in 100 mL methanol until all of the solid completely dissolved. Cu(NO3)2·3H2O (2.45 g, 10.1 mmol) was added to the hot methanol solution in three approximately equal portions. The solution was then refluxed for five minutes.

The product precipitated out of solution as a white solid. The solid was then filtered and washed with ethanol and ether. Product was dried in vacuo to obtain 6 in 73% yield. 1H NMR (300 MHz,

109

31 C6D6) δ 6.91 (m, 12H), 7.42 (m, 18H). P NMR (121 MHz, CDCl3) δ -0.57 (broad). FTIR (neat)

1274, 1455 cm-1. The spectroscopic data were comparable to the reported values.117

(PPh3)2Cu(NO3) (6). Procedure 2.

117 This synthesis was adapted from a previously reported procedure. In a flask, PPh3

(5.32 g, 20.3 mmol), Cu(NO3)2·3H2O (2.46 g, 10.2 mmol), and L-ascorbic acid (1.78 g, 10.1 mmol) were stirred in 100 mL methanol. This solution was then heated at 50 °C for ten minutes.

The white solid was filtered, washed with ethanol and ether. The product was dried in vacuo to

1 31 obtain 6 in 83% yield. H NMR (300 MHz, C6D6) δ 6.91 (m, 12H), 7.42 (m, 18H). P NMR

-1 (121 MHz, C6D6) δ -0.68 (broad). FTIR (neat) 1274, 1456 cm . The spectroscopic data were comparable to the reported values.117

[(PPh3)2Cu(NCBH3)]2 (7).

This synthesis was adapted from a previously reported procedure.111 In a beaker,

Na(BH3CN) (63 mg, 1.0 mmol) was dissolved in 7.5 mL absolute ethanol. In a separate flask, 6

(643 mg, 0.99 mmol) was dissolved in 10 mL of chloroform. The ethanol solution was then added to the chloroform solution and stirred for 10 minutes. The solution was then allowed to stand for 24 hours and then filtered to remove any solid. The filtrate stood at -20 °C, then the solid was filtered and washed with 95% ethanol. The product was dried in vacuo to obtain 7 in

1 31 52% yield. H NMR (300 MHz, C6D6) δ 6.96 (m, 36H), 7.44 (m, 24H). P NMR (121 MHz,

-1 C6D6) δ -2.00 (broad). FTIR (neat) 1094, 2187, 2203, 2373 cm . Characterization was comparable to the reported values.111

(PPh3)3Cu(NCBH3) (8).

This synthesis was adapted from a previously reported procedure.111 In a beaker,

Na(BH3CN) (72.5 mg, 1.2 mmol) was dissolved in 10 mL of absolute ethanol. In a separate flask, 6 (644.7 mg, 1.0 mmol) and PPh3 (259.5 mg, 1.0 mmol) were stirred in 20 mL of absolute

110

ethanol. The solution of Na(BH3CN) was then added to the solution of 6 and stirred for 2 hours.

The white solid was filtered, washed with ether and dried in vacuo to obtain 8 in a 54% yield.

FTIR (neat) 1094, 2172, 2338 cm-1.

(PMePh2)3Cu(NO3) (9).

This synthesis was adapted from a previously reported procedure.117 In a flask equipped with a condenser, PPh2Me (1.5 mL, 8.1 mmol) was stirred in 20 mL methanol. Cu(NO3)2·3H2O

(485 mg, 2.0 mmol) was added to the flask, which was then heated to reflux. After refluxing for

30 minutes, the colorless solution was removed from heat. Deionized water was added to the methanol solution to precipitate a white solid, which was then filtered. The product was dried in

1 vacuo to obtain 9 in 83% yield. H NMR (300 MHz, C6H6) δ 1.56 (s, 9H), 6.95 (m, 18H), 7.28

31 -1 (m, 12H). P NMR (121 MHz, C6H6) δ -17.16 (broad). FTIR (neat) 1300, 1407 cm .

Characterization was comparable to the reported values.124

(PMePh2)3Cu(BH4) (10).

117 This synthesis was adapted from a previously reported procedure. In a beaker, NaBH4

(30.0 mg, 0.79 mmol ) was dissolved in 7.5 mL absolute ethanol. In a separate flask, 9 (349.9 mg, 0.48 mmol) was dissolved in 7.5 mL ethanol. The ethanol solution was then added to the dichloromethane solution and stirred for 2.5 hours. A white solid was filtered off and washed with 90% ethanol. The product was dried in vacuo to obtain 10 in 65% yield. FTIR (neat) 1064,

2061, 2316 cm-1.

(PMePh2)3Cu(NCBH3) (11).

This synthesis was adapted from a previously reported procedure.117 In a beaker,

Na(BH3CN) was dissolved in 5 mL absolute ethanol. In a separate flask, 9 (581.8 mg, 0.81 mmol) was dissolved in 10 mL dichloromethane. The ethanol solution was then added to the dichloromethane solution and stirred for 30 minutes. The solution was then filtered and the

111

filtrate was concentrated in vacuo. Water was added to precipitate a white solid, which was

1 collected. The product was dried in vacuo to obtain 11 in 70% Yield. H NMR (300 MHz, C6H6)

δ 1.35 (s, 9H), 6.94 (m, 18H), 7.22 (s, broad, 12H). FTIR (neat) 1110, 2123, 2274, 2325 cm-1.

(P(o-tolyl)3)2Cu(NO3) (12).

In a flask, Cu(NO3)2·3H2O (404.8 mg, 1.68 mmol), tri-o-tolylphosphine (1.008 g, 3.31 mmol), and L-ascorbic acid (291.7 mg, 1.66 mmol) were stirred in 20 mL of methanol. The flask was then added to an oil bath preheated to 50 °C. After 90 minutes, the solution was colorless with a white solid. The solid was then filtered and washed with methanol. The product was dried

1 in vacuo to obtain 12 in 55% yield. H NMR (300 MHz, C6H6) δ 2.36 (s, 18H), 6.92 (m, 24H).

FTIR (neat) 1298, 1435 cm-1.

Tungsten Sulfide Precursors

Compound 14, W(CO)3(NCCH3)3, and Ni(S2C4H6)2 were synthesized as previously described and characterized by comparison with known literature data.197, 267, 268

WS(S2)(S2CNEt2)2 (13)

This synthesis was adapted from a previously reported procedure.185 To a flame dried flask was added (NH4)2WS4 (500 mg, 0.144 mmol) and tetraethylthiuram disulfide (1.08 g, 3.64 mmol). The flask was then purged three times with N2 and anhydrous acetonitrile

(approximately 30 mL) was added by cannula while stirring. A green color appeared immediately and the solution was stirred for an additional two hours. The solution was chilled in an ice bath for 10 min and then filtered. The resulting green solid was washed with twice with 5 mL acetonitrile, 10 mL water, 5 mL 95% ethanol, and 10 mL pentane. The green solid was then dried under vacuum to afford the product in 86% yield. The product was identified by

185 1 comparison to literature data. H NMR (300 MHz, CDCl3) δ 1.12 (t, 3H, J = 7.2 Hz), δ 1.41

(m, 9H), δ 3.42 (m, 2H), δ 3.84 (m, 6H). FTIR 543.9 cm-1 (m), 496.6 cm-1 (s).

112

N,N-disubstituted-Nʹ-acylthiourea Ligands and Precursors

General Procedure for Ligands L1, L2, L3 and L4.

This procedure was adapted from a previously reported synthesis.269, 270 The acyl chloride

(34 mmol) was stirred with NaSCN (34 mmol) in dry acetone (25 mL) for 10 minutes, after which the solution was milky white. The dialkyl or diaryl amine (34 mmol) was dissolved in dry acetone (15 mL) and was then added to the stirring solution, which turned yellow. The solution was then added to 250 mL of ice cold water, at which point the crude product precipitated. The crude product was filtered, dried, and recrystallized from an acetonitrile:water (3:1) mixture.

N,N-diisopropyl-Nʹ-benzoylthiourea (HL1)

Benzoyl chloride (3.95 mL) and diisopropylamine (4.80 mL) were used as the acyl chloride and diamine, respectively. While previously prepared, no spectroscopic data were

270 1 reported. Yield: 7.0116 g, 78%. H NMR (300 MHz; DMSO-d6): ẟ = 10.32 ppm (s, 1H), 7.70

(m, 5H), 4.56 (broad, 1H), 4.30 (s, 1H), 1.36 (d, 12H). FTIR (neat): 3245 cm-1 (m), 1786 cm-1

(m), 1650 cm-1 (s), 1599 cm-1 (m), 1581 cm-1 (m), 1538 cm-1 (m).

N,N-diphenyl-Nʹ-benzoylthiourea (HL2)

Benzoyl chloride (3.95 mL) and diphenylamine (4.80 mL) were used as the acyl chloride and diamine, respectively. The compound was characterized by comparison to literature data.271

1 Yield: 10.7371 g, 95%. H NMR (300 MHz; DMSO-d6): ẟ = 11.18 ppm (s, 1H,), 7.40 (m, 15H).

FTIR (neat): 3210 cm-1 (m), 1690 cm-1 (s), 1591 cm-1 (m), 1501 cm-1 (s).

N,N-diisopropyl-Nʹ-cinnamoylthiourea (HL3)

Cinnamoyl chloride (5.6644 g) and diisopropylamine (4.80 mL) were used as the acyl

1 chloride and diamine, respectively. Yield: 8.0972 g, 82%. H NMR (300 MHz; DMSO-d6: ẟ =

10.13 ppm (s, 1H), 7.54 (m, 6H), 6.80 (d, 1H), 4.33 (broad, 1H), 1.39 (broad, 12H). FTIR (neat):

3237 cm-1 (m), 1661 cm-1 (m), 1626 cm-1 (s), 1577 cm-1 (w), 1526 cm-1 (s).

113

N,N-diphenyl-Nʹ-cinnamoylthiourea (HL4)

Cinnamoyl chloride (5.6644 g) and diphenylamine (4.80 mL) were used as the acyl

1 chloride and diamine, respectively. Yield: 10.6032 g, 87%. H NMR (300 MHz; DMSO-d6): ẟ =

11.02 ppm (s, 1H), 7.32 (m, 16H), 6.67 (d, 1H). FTIR (neat): 3178 cm-1 (m), 1704 cm-1 (s), 1684 cm-1 (w, sh), 1667 cm-1 (s), 1634 cm-1 (s), 1617 cm-1 (s), 1589 cm-1 (m), 1535 cm-1 (m).

General Procedure for Nickel Complexes.

This procedure was adapted from a previously reported synthesis.272 Nickel chloride hexahydrate (0.5940 g, 2.5 mmol) was dissolved in (10 mL) of water and was added to a stirring solution of L1-4 (5 mmol) in acetonitrile (35 mL). Sodium acetate (0.8200 g, 10 mmol) in water

(10 mL) was then added, resulting in precipitated complex that was filtered off and dried in vacuo.

Ni(L1)2 (18)

From 1.3220 g of HL1, yield: 1.0245 g, 70%. The compound was characterized by

272 1 comparison to literature data. H-NMR (300 MHz; DMSO-d6): ẟ = 8.18 (m, 4H), 7.61 (m,

6H), 4.85 (s, 2H), 4.10 (s, 2H), 1.55 (d, 12H), 1.34 (broad, 12H). FTIR (neat): 1587 cm-1 (w)

-1 + 1509 cm (m). MS (ESI) m/z 585.18 (M+H ). Elemental analysis: Calc. for C28H38N4O2S2Ni: C,

57.44; H, 6.54; N, 9.57; Found: C, 57.51; H, 6.60; N, 9.57%.

Ni(L2)2 (19)

From 1.6621 g of HL2, yield: 1.1183 g, 62%. The compound was characterized by

272 1 comparison to literature data. H-NMR (300 MHz; DMSO-d6): ẟ = 9.00 (m, 4H), 7.46 (m,

22H), 7.12 (m, 4H). FTIR (neat): 1588 cm-1 (w), 1508 cm-1 (s). MS (DART) m/z 721.12 (M+H+).

Elemental analysis: Calc. for C40H30N4O2S2Ni: C, 66.59; H, 4.19; N, 7.77. Found: C, 67.04; H,

3.84; N, 7.58%.

114

Ni(L3)2 (20)

From 1.4522 g of HL3, yield: 1.1165 g, 73%. X-ray quality crystals were obtained from

1 recrystallization from acetonitrile. H-NMR (300 MHz; DMSO-d6): δ 7.63 (m, 6H), 7.42 (m,

6H), 7.00 (d, 2H), 4.85 (s, 2H), 3.95 (s, 2H), 1.49 (broad, 12H), 1.26 (broad, 12H). FTIR (neat):

1638 cm-1 (m), 1577 cm-1 (w), 1506 cm-1 (m). MS (ESI) m/z 637.21 (M+H+). Elemental analysis:

Calc. for C32H42N4O2S2Ni: C, 60.29; H, 6.64; N, 8.79. Found: C, 60.29; H, 6.41; N, 8.72%.

Ni(L4)2 (21)

1 From 1.7923 g of HL4, yield: 1.2377 g, 64%. H-NMR (300 MHz; DMSO-d6): δ = 7.49

(m, 28H), 7.11 (m, 4H), 6.89 (m, 2H). FTIR (neat): 1629 cm-1 (m), 1589 cm-1 (w), 1576 cm-1

-1 + (w), 1501 cm (s). MS (ESI) m/z 773.15 (M+H ). Elemental analysis: Calc. for C44H34N4O2S2Ni:

C, 68.32; H, 4.43; N, 7.24. Found: C, 68.22; H, 4.18; N, 7.14%.

Cu(L3)2·THF (22)

Same procedure as for 20, substituting the nickel chloride with cupric nitrate trihydrate

(0.6040 g, 2.5 mmol) and recrystallized from THF. Yield: 1.4111g, 79%. FTIR (neat): 1636 cm-1

(m), 1577 cm-1 (w), 1501 cm-1 (m, sh). MS (ESI) m/z 642.21 ([M-THF+H]+). Elemental analysis: Calc. for C36H50N4O3S2Cu: C 60.52; H, 7.05; N, 7.84. Found: C, 60.41; H, 7.09; N,

7.78%.

Co(L3)3 (23)

Same procedure for 20, substituting the nickel chloride with cobalt chloride hexahydrate

(0.5948 g, 2.5 mmol) and 3 equivalents of HL3 (7.5 mmol, 2.1782 g). Yield: 1.738 g, 75%. 1H-

NMR (300 MHz; DMSO-d6): δ = 7.45 (m, 18H), 6.80 (d, 3H), 5.40 (broad, 3H), 1.33 (broad,

36H). FTIR (neat): 1633 cm-1 (m), 1575 cm-1 (w), 1501 cm-1 (m). MS (ESI) m/z 927.35 (M+H+).

Elemental analysis: Calc. for C48H63N6O3S3Co: C, 62.18; H, 6.85; N, 9.06. Found: C, 62.03; H,

6.44; N, 8.87%.

115

Zn(L3)2 (24)

Same procedure for 20, substituting the nickel chloride with zinc nitrate hexahydrate

1 (0.7437 g, 2.5 mmol). Yield: 1.4495 g, 90%. H-NMR (300 MHz; DMSO-d6): δ = 7.61 (m, 4H),

7.52 (d, 2H), 7.38 (m, 6H), 6.64 (d, 2H), 5.48 (s, 2H), 3.85 (s, 2H), 1.45 (m, 12H), 1.23 (m,

12H). FTIR (neat): 1636 cm-1 (m), 1575 (w), 1511 (m, sh). MS (ESI) m/z 643.21 (M+H+).

Elemental analysis: Calc. for C32H42N4O2S2Zn: C, 59.66; H, 6.57; N, 8.70. Found: C, 59.76; H,

6.36; N, 9.16%.

Tungsten Carbonitride Precursor

i WN(NMe2)[( PrN)2C(NMe2)]2 (25)

Inside the glovebox, WN(NMe2)3 (2.4 g, 7.3 mmol) was combined with 200 mL of pentane in a flask to form a light tan suspension.247 To this suspension, DIC (2.4 mL, 15 mmol) was added in a single aliquot. The reaction mixture quickly darkened to an amber color. This mixture was allowed to stir for 10 min and then was filtered through a fine frit to yield a yellow filtrate, which was concentrated under vacuum. The mother liquor was removed and the pure product was dried under vacuum, yielding 2.9 g (5.0 mmol, 69%) of 25 as yellow crystals. 1H

NMR (500 MHz, C6D6): δ 1.03 (d, 3H, CH(CH3)2), 1.16 (d, 3H, CH(CH3)2), 1.18 (d, 3H,

CH(CH3)2), 1.18 (d, 3H, CH(CH3)2), 1.37 (d, 3H, CH(CH3)2), 1.39 (d, 3H, CH(CH3)2), 1.63 (d,

3H, CH(CH3)2), 2.07 (d, 3H, CH(CH3)2), 2.48 (s, 6H, CN(CH3)2), 2.53 (s, 6H, CN(CH3)2), 3.60

(sept, 1H, CH(CH3)2), 3.71 (s, 3H, WN(CH3)CH3), 3.89 (sept, 1H, CH(CH3)2), 4.10 (sept, 1H,

13 CH(CH3)2), 4.12 (sept, 1H, CH(CH3)2), 4.82 (s, 3H, WN(CH3)CH3). C NMR (C6D6, 151

MHz): 22.9, 24.1, 24.6, 25.1, 25.6, 25.7, 25.9, 27.8 (CH(CH3)2), 40.1 (CN(CH3)2), 46.6

(CH(CH3)2), 47.1 (CH(CH3)2), 47.5 (CH(CH3)2), 48.9 (WN(CH3)CH3), 50.8 (CH(CH3)2), 68.0

15 (WN(CH3)CH3), 166.2, 166.7 (N3C). N NMR (C6D5CD3, 51 MHz): δ 139.2, 146.6, 183.2,

i 14 187.2 (W[( PrN)2C(NMe2)]2), 236.5 (WNMe2). N NMR (C6D6, 43 MHz): δ 133, 181

116

i + (W[( PrN)2C(NMe2)]2), 237 (WNMe2), 753 (W≡N). MS (DART-TOF): Calc for [M + H]

583.3428; found 583.3440. Elemental analysis: C20H46N8W: Calc C, 41.24; H, 7.96; N, 19.24; found C, 40.83; H, 7.80; N, 18.87.

Deposition Procedures

Depositions were carried out using a Blue Wave Semiconductors CVD reactor with a

Liquifog ultrasonic liquids atomizer from Johnson Matthey Piezoproducts.

General Procedure for AACVD

Silicon with native silicon dioxide (Si/SiO2, n-type, <100>) was cut into squares of approximately 1 cm2 and cleaned in boiling isopropanol, acetone, and methanol for three minutes each. The substrates were then placed onto the heating stand of the reactor, placed under vacuum

(<300 mTorr), and heated to the desired temperature. In a glovebox, the precursor was dissolved in 20 mL of a suitable, anhydrous solvent and added to a glass trap. The trap was then removed from the glovebox and connected to the N2 inlet of the reactor, N2 was flowed through the trap for 10 min before connecting to the transfer line. The pressure of the reaction chamber was increased to 350 Torr and the transfer line was heated to 50 °C. The trap was then opened to the reaction chamber and nebulization of the solution was started. During the course of the deposition (typically 60-75 min), N2 flow was maintained at 200 sccm and the pressure was maintained at 350 Torr. Once all of the solution had been nebulized, the pressure of the reaction chamber was increased to atmospheric pressure and the substrates were cooled to room temperature.

Deposition of WS2 from 13

The above procedure was followed using a toluene solution of 13 (250 mg, 0.43 mmol).

Depositions were carried out between 300-500 °C.

117

Deposition of WS2 from 13 Under Ambient Conditions

Using the same concentration of a solution of 13, deposition was performed at 400 °C without drying or deoxygenating the toluene and without any vacuum/nitrogen purges. Under these conditions, deposition still resulted in WS2 by Raman spectroscopy

Deposition of WS2 from 14

The above procedure was followed using a toluene solution of 14 (250 mg, 0.52 mmol).

Depositions were carried out at 400 °C.

Deposition of MSx (M = Ni, Co, or Zn) from 20, 23, 24

The above procedure was followed using toluene solutions of 20, 23, 24 (0.64 mmol).

Depositions were carried out at 350 °C.

Deposition of WCxNy from 25

The above procedure was followed using a heptane solution of 25 (596 mg, 1.02 mmol).

Depositions were carried out at between 200-400 °C.

Procedure for Low Pressure CVD of 29

Silicon with native silicon dioxide (Si/SiO2, n-type, <100>) was cut into squares of approximately 1 cm2 and cleaned in boiling methanol, acetone, and water for three minutes each.

The substrates were then placed onto the heating stand of the reactor and placed under vacuum for at least one hour (<300 mTorr). In a glovebox, 29 (0.50 g, 0.53 mmol) was loaded into the stainless steel bubbler. The bubbler was then removed from the glovebox and connected to the

N2 inlet of the reactor and to the inlet of the reactor and heated to 75 °C for at least one hour. The substrates were heated to 500 °C and and the transfer line was heated to 50 °C. The pressure of the reaction chamber was then increased to 200 Torr and the bubbler was opened to the reaction chamber. During the course of the deposition, N2 flow was maintained at 200 sccm and the pressure was maintained at 2000 Torr. After 150 minutes of exposure of the bubbler, the pressure

118

of the reaction chamber was increased to atmospheric pressure and the substrates were cooled to room temperature.

119

APPENDIX A COMPOUND STRUCTURES, FORMULAS, AND NUMBERS

(PPh3)Cu(NO3) 6

WS(S2)(S2CNEt2)2 (VTMS)Cu(hfac) 13 1

[(PPh3)2Cu(NCBH3)]2 7

W(CO)2(S2C4H6)2 (MHY)Cu(hfac) 14 2 (PPh)3Cu(NCBH3) 8 C4H6 15

(PMePh ) Cu(NO ) 2 3 3 9 (DTA)Cu(hfac) C8H12S 3 16

(PMePh2)3Cu(BH4) 10 C4H6S2(CO) 17

(DPA)Cu(hfac) 4 (PMePh2)3Cu(NCBH3) 11

N,N-diisopropyl-Nʹ- phenylthiourea

(PPh3)2Cu(BH4) HL1 5 (P-(o-tolyl)3)2Cu(NO3) 12

120

N,N-diphenyl-Nʹ- Zn(L3)2 phenylthiourea 24 HL2 Ni(L3)2 20

N,N-diisopropyl-Nʹ- cinnamoylthiourea WN(NMe )[(NiPr) C(N HL3 2 2 Me2)]2 25

Ni(L4)2 21

N,N-diphenyl-Nʹ- cinnamoylthiourea HL4 WN(NMe2)[(NCy)2C(N Me2)]2 26

Cu(L3)2 22

i WN(NEt2)[(N Pr)2C- Ni(L1)2 18 (NEt2)]2 27

Co(L3)3 23 WN(NEt2)[(NCy)2C- (NEt2)]2 Ni(L2)2 28 19

121

WO(OC(CF3)2CH3)3(hfac) 29

122

APPENDIX B FULL X-RAY CRYSTAL STRUCTURES

Figure B-1. Full X-ray crystal structure of 22. Adapted from reference 213 with permission from The Royal Society of Chemistry.

Figure B-2. Full X-ray crystal structure of 23. Adapted from reference 213 with permission from The Royal Society of Chemistry.

123

213 Figure B-3. Full X-ray crystal structure of 24·NCCH3. Adapted from reference with permission from The Royal Society of Chemistry.

124

APPENDIX C FULL X-RAY TABLES

X-ray Tables of 20

Table C-1. Crystal data and structure refinement for 20. Identification code 20 Empirical formula C32H42N4NiO2S2 Formula weight 637.52 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Tetragonal Space group I-4 Unit cell dimensions a = 14.662(2) Å = 90°  b = 14.662(2) Å = 90°  c = 15.487(2) Å  = 90° 3 Volume 3329.2(11) Å Z 4 3 Density (calculated) 1.272 Mg/m Absorption coefficient 0.742 mm-1 F(000) 1352 3 Crystal size 0.200 x 0.162 x 0.079 mm Theta range for data collection 1.913 to 32.828°. Index ranges -22≤h≤20, -22≤k≤21, -23≤l≤22 Reflections collected 36001 Independent reflections 5829 [R(int) = 0.0226] Completeness to theta = 25.000° 100.00% Absorption correction Integration Max. and min. transmission 0.9516 and 0.8848 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 5829 / 0 / 190 Goodness-of-fit on F2 1.029 Final R indices [I>2sigma(I)] R1 = 0.0200, wR2 = 0.0485 [5510] R indices (all data) R1 = 0.0225, wR2 = 0.0491 Absolute structure parameter -0.0091(19) -3 Largest diff. peak and hole 0.320 and -0.143 e.Å R1 = (||Fo| - |Fc||) / |Fo| 2 2 2 2 2 1/2 2 2 2 1/2 wR2 = [w(Fo - Fc ) ] / wFo  ]] S = [w(Fo - Fc ) ] / (n-p)] 2 2 2 2 2 w= 1/[ (Fo )+(m*p) +n*p], p = [max(Fo ,0)+ 2* Fc ]/3, m & n are constants

125

Table C-2. Atomic coordinates (x 104) and equivalent isotropic displacement parameters (Å2x 103) for 20. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. x y z U(eq) Ni1 5000 10000 5372(1) 10(1) S1 5961(1) 9830(1) 6401(1) 20(1) O1 5858(1) 9836(1) 4490(1) 12(1) N1 7113(1) 9118(1) 5168(1) 13(1) N2 7416(1) 8773(1) 6573(1) 14(1) C1 6649(1) 9472(1) 4512(1) 11(1) C2 6868(1) 9196(1) 6002(1) 12(1) C3 7156(1) 9405(1) 3688(1) 12(1) C4 6884(1) 9787(1) 2947(1) 13(1) C5 7395(1) 9773(1) 2132(1) 13(1) C6 6945(1) 10006(1) 1366(1) 18(1) C7 7400(1) 9993(1) 576(1) 22(1) C8 8316(1) 9747(1) 541(1) 21(1) C9 8772(1) 9516(1) 1297(1) 20(1) C10 8323(1) 9533(1) 2088(1) 16(1) C11 8283(1) 8320(1) 6299(1) 16(1) C12 8960(1) 8984(1) 5889(1) 24(1) C13 8109(1) 7468(1) 5757(1) 22(1) C14 7222(1) 8790(1) 7514(1) 17(1) C15 7272(1) 7840(1) 7916(1) 30(1) C16 7843(1) 9468(1) 7972(1) 33(1)

126

Table C-3. Bond lengths [Å] and angles [°] for 20. Ni1-O1 1.8719(9) C16-H16B 0.98 Ni1-O1#1 1.8719(9) C16-H16C 0.98 Ni1-S1 2.1424(4) O1-Ni1-O1#1 86.34(5) Ni1-S1#1 2.1424(4) O1-Ni1-S1 94.92(3) S1-C2 1.7370(13) O1#1-Ni1-S1 178.62(3) O1-C1 1.2771(15) O1-Ni1-S1#1 178.62(3) N1-C1 1.3283(16) O1#1-Ni1-S1#1 94.92(3) N1-C2 1.3441(16) S1-Ni1-S1#1 83.83(2) N2-C2 1.3458(16) C2-S1-Ni1 107.51(4) N2-C14 1.4852(18) C1-O1-Ni1 130.12(8) N2-C11 1.4963(17) C1-N1-C2 124.36(11) C1-C3 1.4806(17) C2-N2-C14 121.56(11) C3-C4 1.3376(17) C2-N2-C11 121.66(11) C3-H3A 0.95 C14-N2-C11 116.64(10) C4-C5 1.4672(17) O1-C1-N1 130.46(12) C4-H4A 0.95 O1-C1-C3 117.40(11) C5-C6 1.4002(18) N1-C1-C3 112.13(11) C5-C10 1.4077(19) N1-C2-N2 115.66(11) C6-C7 1.3927(19) N1-C2-S1 126.28(10) C6-H6A 0.95 N2-C2-S1 117.97(10) C7-C8 1.392(2) C4-C3-C1 124.25(12) C7-H7A 0.95 C4-C3-H3A 117.9 C8-C9 1.391(2) C1-C3-H3A 117.9 C8-H8A 0.95 C3-C4-C5 125.44(12) C9-C10 1.3911(18) C3-C4-H4A 117.3 C9-H9A 0.95 C5-C4-H4A 117.3 C10-H10A 0.95 C6-C5-C10 118.39(12) C11-C13 1.526(2) C6-C5-C4 119.03(12) C11-C12 1.529(2) C10-C5-C4 122.58(12) C11-H11A 1 C7-C6-C5 121.04(13) C12-H12A 0.98 C7-C6-H6A 119.5 C12-H12B 0.98 C5-C6-H6A 119.5 C12-H12C 0.98 C8-C7-C6 120.01(13) C13-H13A 0.98 C8-C7-H7A 120 C13-H13B 0.98 C6-C7-H7A 120 C13-H13C 0.98 C9-C8-C7 119.60(13) C14-C16 1.523(2) C9-C8-H8A 120.2 C14-C15 1.528(2) C7-C8-H8A 120.2 C14-H14A 1 C10-C9-C8 120.65(13) C15-H15A 0.98 C10-C9-H9A 119.7 C15-H15B 0.98 C8-C9-H9A 119.7 C15-H15C 0.98 C9-C10-C5 120.31(13) C16-H16A 0.98 C9-C10-H10A 119.8

127

Table C-3. Continued. N2-C11-C13 112.13(11) N2-C14-C16 110.74(12) N2-C11-C12 112.77(11) N2-C14-C15 112.01(12) C13-C11-C12 113.69(12) C16-C14-C15 112.11(13) N2-C11-H11A 105.8 N2-C14-H14A 107.2 C13-C11-H11A 105.8 C16-C14-H14A 107.2 C12-C11-H11A 105.8 C15-C14-H14A 107.2 C11-C12-H12A 109.5 C14-C15-H15A 109.5 C11-C12-H12B 109.5 C14-C15-H15B 109.5 H12A-C12-H12B 109.5 H15A-C15-H15B 109.5 C11-C12-H12C 109.5 C14-C15-H15C 109.5 H12A-C12-H12C 109.5 H15A-C15-H15C 109.5 H12B-C12-H12C 109.5 H15B-C15-H15C 109.5 C11-C13-H13A 109.5 C14-C16-H16A 109.5 C11-C13-H13B 109.5 C14-C16-H16B 109.5 H13A-C13-H13B 109.5 H16A-C16-H16B 109.5 C11-C13-H13C 109.5 C14-C16-H16C 109.5 H13A-C13-H13C 109.5 H16A-C16-H16C 109.5 H13B-C13-H13C 109.5 H16B-C16-H16C 109.5 Symmetry transformations used to generate equivalent atoms: #1 -x+1,-y+2,z

128

Table C-4. Anisotropic displacement parameters (Å2x103) for 20. The anisotropic displacement factor exponent takes the form: -22[ h2 a*2U11 + ... + 2 h k a* b* U12] U11 U22 U33 U23 U13 U12 Ni1 11(1) 13(1) 7(1) 0 0 3(1) S1 19(1) 31(1) 9(1) -5(1) -3(1) 13(1) O1 11(1) 16(1) 10(1) 0(1) 0(1) 2(1) N1 13(1) 16(1) 11(1) 0(1) 0(1) 2(1) N2 14(1) 16(1) 11(1) 2(1) -1(1) 3(1) C1 12(1) 10(1) 11(1) -1(1) 1(1) -1(1) C2 11(1) 12(1) 13(1) -1(1) -1(1) 1(1) C3 11(1) 13(1) 12(1) -1(1) 1(1) 1(1) C4 13(1) 14(1) 13(1) -1(1) 2(1) 0(1) C5 15(1) 12(1) 12(1) 0(1) 1(1) 0(1) C6 18(1) 20(1) 14(1) 2(1) 1(1) 5(1) C7 26(1) 27(1) 13(1) 4(1) 1(1) 6(1) C8 24(1) 26(1) 14(1) 3(1) 6(1) 0(1) C9 15(1) 27(1) 17(1) 1(1) 4(1) 0(1) C10 14(1) 22(1) 13(1) 1(1) 0(1) -1(1) C11 13(1) 18(1) 15(1) 2(1) -1(1) 4(1) C12 14(1) 28(1) 29(1) 8(1) 1(1) 1(1) C13 22(1) 20(1) 24(1) -3(1) -2(1) 8(1) C14 19(1) 22(1) 10(1) 2(1) 0(1) 4(1) C15 38(1) 31(1) 20(1) 11(1) 4(1) 4(1) C16 41(1) 38(1) 20(1) -10(1) -6(1) -2(1)

129

Table C-5. Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x 103) for 20. x y z U(eq) H3A 7709 9068 3685 15 H4A 6314 10094 2949 16 H6A 6320 10175 1385 21 H7A 7084 10153 62 26 H8A 8627 9737 3 25 H9A 9397 9344 1273 24 H10A 8645 9382 2601 20 H11A 8579 8101 6842 19 H12A 9561 8694 5848 35 H12B 9006 9534 6246 35 H12C 8749 9150 5309 35 H13A 8679 7126 5691 33 H13B 7881 7647 5187 33 H13C 7655 7084 6046 33 H14A 6582 9012 7587 21 H15A 7017 7858 8500 45 H15B 7910 7642 7943 45 H15C 6923 7410 7562 45 H16A 7675 9501 8584 49 H16B 7775 10071 7708 49 H16C 8478 9267 7920 49

130

X-ray Tables of 22

Table C-6. Crystal data and structure refinement for 22. Identification code 22

Empirical formula C36H50CuN4O3S2 Formula weight 634.31 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 10.5648(7) Å = 79.6830(10)°  b = 11.9358(8) Å  = 72.2510(10)°  c = 15.2248(10) Å  = 84.0150(10)° Volume 1796.4(2) Å3 Z 2 Density (calculated) 1.173 Mg/m3 Absorption coefficient 0.698 mm-1 F(000) 670 Crystal size 0.259 x 0.176 x 0.108 mm3 Theta range for data collection 1.737 to 28.287°. Index ranges -14≤h≤14, -15≤k≤15, -20≤l≤20 Reflections collected 36766 Independent reflections 8897 [R(int) = 0.0210] Completeness to theta = 25.242° 100.00% Absorption correction Multi Max. and min. transmission 0.5633 and 0.5089 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8897 / 0 / 450 Goodness-of-fit on F2 1.025 Final R indices [I>2sigma(I)] R1 = 0.0386, wR2 = 0.0950 [7375] R indices (all data) R1 = 0.0514, wR2 = 0.1040 Largest diff. peak and hole 1.162 and -0.518 e.Å-3 R1 = (||Fo| - |Fc||) / |Fo| wR2 = [w(Fo - Fc)] / wFo]] 1/2 S = [w(Fo - Fc)] / (n-p)]1/2

131

Table C-7. Atomic coordinates (x104) and equivalent isotropic displacement parameters (Å2x 103) for 22. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. x y z U(eq) Cu1 0 0 0 16(1) S1 896(1) -1632(1) -568(1) 21(1) O1 1147(1) -93(1) 770(1) 24(1) N1 2743(2) -1588(1) 364(1) 19(1) N2 3473(2) -2309(1) -1003(1) 20(1) C1 2053(2) -845(2) 903(1) 18(1) C2 2470(2) -1821(1) -395(1) 18(1) C3 2472(2) -848(2) 1745(1) 19(1) C4 1882(2) -163(2) 2374(1) 19(1) C5 2290(2) -31(2) 3185(1) 21(1) C6 1614(2) 790(2) 3735(2) 29(1) C7 2043(3) 1018(2) 4460(2) 41(1) C8 3145(3) 414(2) 4652(2) 39(1) C9 3804(2) -421(2) 4131(2) 32(1) C10 3388(2) -645(2) 3402(1) 26(1) C11 4828(2) -2576(2) -869(1) 22(1) C12 4850(2) -3409(2) 12(2) 28(1) C13 5547(2) -1499(2) -962(2) 26(1) C14 3309(2) -2660(2) -1854(2) 27(1) C15 4358(2) -2171(2) -2740(2) 35(1) C16 3281(2) -3957(2) -1726(2) 36(1) Cu2 5000 5000 5000 24(1) S21 5899(2) 6744(2) 4553(1) 8(1) O21 4660(20) 4980(18) 3843(16) 24(2) Cu2' 4774(6) 5106(3) 5103(4) 29(1) S21' 5917(5) 6749(4) 4545(3) 44(2) O21' 4450(30) 5130(20) 3895(19) 31(4) N21 5277(2) 6786(1) 2924(1) 20(1) N22 7144(2) 7680(1) 2802(1) 20(1) C21 4541(2) 5884(2) 3221(1) 20(1) C22 6129(2) 7059(2) 3355(1) 18(1) C23 3637(2) 5809(2) 2660(1) 21(1) C24 2975(2) 4879(2) 2765(1) 21(1) C25 2157(2) 4675(2) 2188(1) 21(1) C26 1788(2) 3566(2) 2245(2) 28(1) C27 1052(2) 3326(2) 1690(2) 32(1) C28 670(2) 4196(2) 1071(2) 30(1) C29 1023(2) 5306(2) 1010(2) 28(1) C30 1764(2) 5541(2) 1562(1) 22(1) C31 7443(2) 7932(2) 1762(1) 22(1) C32 7667(2) 6863(2) 1304(1) 28(1)

132

Table C-7. Continued. C33 6448(2) 8803(2) 1439(1) 29(1) C34 8111(2) 8106(2) 3182(1) 24(1) C35 9474(2) 7485(2) 2871(2) 36(1) C36 8187(2) 9399(2) 2930(2) 32(1) O41 1618(5) 5807(4) 6305(4) 77(2) C42 2159(6) 7196(4) 5658(4) 43(1) C43 1477(4) 7325(4) 4923(3) 32(1) C44 298(7) 6676(5) 5259(5) 56(2) C45 843(8) 5644(7) 5594(5) 71(2) O51 2563(4) 5998(4) 5828(3) 44(1) C52 1831(9) 7099(7) 6054(7) 64(2) C53 1136(12) 7466(10) 5411(9) 95(3) C54 838(9) 6391(7) 5092(6) 55(2) C55 1354(8) 5473(6) 5719(5) 52(2)

133

Table C-8. Bond lengths [Å] and angles [°] for 22. Cu1-O1 1.9093(13) C16-H16C 0.98 Cu1-O1#1 1.9093(13) Cu2-O21 1.91(2) Cu1-S1#1 2.2709(4) Cu2-O21#2 1.91(2) Cu1-S1 2.2709(4) Cu2-S21#2 2.283(2) S1-C2 1.7473(18) Cu2-S21 2.283(2) O1-C1 1.280(2) S21-C22 1.741(3) N1-C1 1.323(2) O21-C21 1.33(2) N1-C2 1.354(2) Cu2'-Cu2'#2 0.551(12) N2-C2 1.335(2) Cu2'-O21'#2 1.91(3) N2-C14 1.493(2) Cu2'-O21' 1.97(3) N2-C11 1.501(2) Cu2'-S21' 2.301(6) C1-C3 1.478(2) Cu2'-S21'#2 2.323(6) C3-C4 1.336(3) S21'-C22 1.732(5) C3-H3A 0.95 S21'-Cu2'#2 2.323(6) C4-C5 1.465(2) O21'-C21 1.22(3) C4-H4A 0.95 O21'-Cu2'#2 1.91(3) C5-C6 1.398(3) N21-C21 1.324(2) C5-C10 1.401(3) N21-C22 1.359(2) C6-C7 1.391(3) N22-C22 1.342(2) C6-H6A 0.95 N22-C34 1.488(2) C7-C8 1.385(3) N22-C31 1.497(2) C7-H7A 0.95 C21-C23 1.480(2) C8-C9 1.383(3) C23-C24 1.334(3) C8-H8A 0.95 C23-H23A 0.95 C9-C10 1.387(3) C24-C25 1.470(2) C9-H9A 0.95 C24-H24A 0.95 C10-H10A 0.95 C25-C30 1.396(3) C11-C13 1.523(3) C25-C26 1.399(3) C11-C12 1.524(3) C26-C27 1.390(3) C11-H11A 1 C26-H26A 0.95 C12-H12A 0.98 C27-C28 1.387(3) C12-H12B 0.98 C27-H27A 0.95 C12-H12C 0.98 C28-C29 1.393(3) C13-H13A 0.98 C28-H28A 0.95 C13-H13B 0.98 C29-C30 1.388(3) C13-H13C 0.98 C29-H29A 0.95 C14-C15 1.526(3) C30-H30A 0.95 C14-C16 1.527(3) C31-C32 1.525(3) C14-H14A 1 C31-C33 1.526(3) C15-H15A 0.98 C31-H31A 1 C15-H15B 0.98 C32-H32A 0.98 C15-H15C 0.98 C32-H32B 0.98 C16-H16A 0.98 C32-H32C 0.98

134

Table C-8. Continued C16-H16B 0.98 C33-H33A 0.98 C33-H33C 0.98 C1-N1-C2 125.10(16) C34-C36 1.526(3) C2-N2-C14 122.05(16) C34-C35 1.528(3) C2-N2-C11 122.78(15) C34-H34A 1 C14-N2-C11 115.15(15) C35-H35A 0.98 O1-C1-N1 129.50(17) C35-H35B 0.98 O1-C1-C3 116.35(15) C35-H35C 0.98 N1-C1-C3 114.00(15) C36-H36A 0.98 N2-C2-N1 115.81(16) C36-H36B 0.98 N2-C2-S1 119.55(14) C36-H36C 0.98 N1-C2-S1 124.43(14) O41-C45 1.591(9) C4-C3-C1 122.46(16) O41-C42 1.821(8) C4-C3-H3A 118.8 C42-C43 1.485(7) C1-C3-H3A 118.8 C42-H42A 0.99 C3-C4-C5 126.66(17) C42-H42B 0.99 C3-C4-H4A 116.7 C43-C44 1.443(8) C5-C4-H4A 116.7 C43-H43A 0.99 C6-C5-C10 118.18(18) C43-H43B 0.99 C6-C5-C4 119.03(17) C44-C45 1.383(10) C10-C5-C4 122.66(17) C44-H44A 0.99 C7-C6-C5 121.01(19) C44-H44B 0.99 C7-C6-H6A 119.5 C45-H45A 0.99 C5-C6-H6A 119.5 C45-H45B 0.99 C8-C7-C6 119.9(2) O51-C52 1.493(9) C8-C7-H7A 120.1 O51-C55 1.545(9) C6-C7-H7A 120.1 C52-C53 1.382(14) C9-C8-C7 119.8(2) C52-H52A 0.99 C9-C8-H8A 120.1 C52-H52B 0.99 C7-C8-H8A 120.1 C53-C54 1.543(14) C8-C9-C10 120.5(2) C53-H53A 0.99 C8-C9-H9A 119.7 C53-H53B 0.99 C10-C9-H9A 119.7 C54-C55 1.492(11) C9-C10-C5 120.53(19) C54-H54A 0.99 C9-C10-H10A 119.7 C54-H54B 0.99 C5-C10-H10A 119.7 C55-H55A 0.99 N2-C11-C13 111.98(15) C55-H55B 0.99 N2-C11-C12 114.51(16) O1-Cu1-O1#1 180 C13-C11-C12 112.36(17) O1-Cu1-S1#1 86.27(4) N2-C11-H11A 105.7 O1#1-Cu1-S1#1 93.73(4) C13-C11-H11A 105.7 O1-Cu1-S1 93.72(4) C12-C11-H11A 105.7 O1#1-Cu1-S1 86.27(4) C11-C12-H12A 109.5 S1#1-Cu1-S1 180 C11-C12-H12B 109.5

135

Table C-8. Continued. C2-S1-Cu1 104.71(6) C11-C12-H12C 109.5 C1-O1-Cu1 130.77(12) H12A-C12-H12C 109.5 H12B-C12-H12C 109.5 Cu2'-S21'-Cu2'#2 13.7(3) C11-C13-H13A 109.5 C21-O21'-Cu2'#2 132.9(17) C11-C13-H13B 109.5 C21-O21'-Cu2' 131.9(17) H13A-C13-H13B 109.5 Cu2'#2-O21'-Cu2' 16.3(4) C11-C13-H13C 109.5 C21-N21-C22 124.28(16) H13A-C13-H13C 109.5 C22-N22-C34 121.66(15) H13B-C13-H13C 109.5 C22-N22-C31 123.16(15) N2-C14-C15 111.94(17) C34-N22-C31 115.06(15) N2-C14-C16 109.97(18) O21'-C21-N21 130.9(11) C15-C14-C16 112.88(17) N21-C21-O21 128.2(10) N2-C14-H14A 107.2 O21'-C21-C23 115.1(11) C15-C14-H14A 107.2 N21-C21-C23 113.93(16) C16-C14-H14A 107.2 O21-C21-C23 117.0(10) C14-C15-H15A 109.5 N22-C22-N21 115.38(16) C14-C15-H15B 109.5 N22-C22-S21' 119.7(2) H15A-C15-H15B 109.5 N21-C22-S21' 124.7(2) C14-C15-H15C 109.5 N22-C22-S21 120.30(15) H15A-C15-H15C 109.5 N21-C22-S21 124.14(15) H15B-C15-H15C 109.5 C24-C23-C21 122.17(17) C14-C16-H16A 109.5 C24-C23-H23A 118.9 C14-C16-H16B 109.5 C21-C23-H23A 118.9 H16A-C16-H16B 109.5 C23-C24-C25 126.45(17) C14-C16-H16C 109.5 C23-C24-H24A 116.8 H16A-C16-H16C 109.5 C25-C24-H24A 116.8 H16B-C16-H16C 109.5 C30-C25-C26 118.41(18) O21-Cu2-O21#2 180.0(2) C30-C25-C24 122.58(17) O21-Cu2-S21#2 85.3(7) C26-C25-C24 118.98(17) O21#2-Cu2-S21#2 94.7(7) C27-C26-C25 121.03(19) O21-Cu2-S21 94.7(7) C27-C26-H26A 119.5 O21#2-Cu2-S21 85.3(7) C25-C26-H26A 119.5 S21#2-Cu2-S21 180.00(5) C28-C27-C26 119.9(2) C22-S21-Cu2 105.16(12) C28-C27-H27A 120.1 C21-O21-Cu2 126.0(14) C26-C27-H27A 120.1 Cu2'#2-Cu2'-O21'#2 87.9(13) C27-C28-C29 119.73(19) Cu2'#2-Cu2'-O21' 75.9(11) C27-C28-H28A 120.1 O21'#2-Cu2'-O21' 163.7(4) C29-C28-H28A 120.1 Cu2'#2-Cu2'-S21' 85.4(9) C30-C29-C28 120.22(19) O21'#2-Cu2'-S21' 87.5(8) C30-C29-H29A 119.9 O21'-Cu2'-S21' 91.1(8) C28-C29-H29A 119.9 Cu2'#2-Cu2'-S21'#2 80.9(8) C29-C30-C25 120.72(19) H12A-C12-H12B 109.5 O21'#2-Cu2'-S21'#2 92.0(8)

136

Table C-8. Continued. O21'-Cu2'-S21'#2 85.5(8) C25-C30-H30A 119.6 S21'-Cu2'-S21'#2 166.3(3) N22-C31-C32 113.33(16) C22-S21'-Cu2' 107.7(3) N22-C31-C33 113.13(16) C22-S21'-Cu2'#2 102.0(3) C32-C31-C33 112.66(17) N22-C31-H31A 105.6 H43A-C43-H43B 108.2 C32-C31-H31A 105.6 C45-C44-C43 99.3(6) C33-C31-H31A 105.6 C45-C44-H44A 111.9 C31-C32-H32A 109.5 C43-C44-H44A 111.9 C31-C32-H32B 109.5 C45-C44-H44B 111.9 H32A-C32-H32B 109.5 C43-C44-H44B 111.9 C31-C32-H32C 109.5 H44A-C44-H44B 109.6 H32A-C32-H32C 109.5 C44-C45-O41 110.9(6) H32B-C32-H32C 109.5 C44-C45-H45A 109.5 C31-C33-H33A 109.5 O41-C45-H45A 109.5 C31-C33-H33B 109.5 C44-C45-H45B 109.5 H33A-C33-H33B 109.5 O41-C45-H45B 109.5 C31-C33-H33C 109.5 H45A-C45-H45B 108 H33A-C33-H33C 109.5 C52-O51-C55 94.7(5) H33B-C33-H33C 109.5 C53-C52-O51 107.6(8) N22-C34-C36 111.49(17) C53-C52-H52A 110.2 N22-C34-C35 110.70(17) O51-C52-H52A 110.2 C36-C34-C35 112.23(17) C53-C52-H52B 110.2 N22-C34-H34A 107.4 O51-C52-H52B 110.2 C36-C34-H34A 107.4 H52A-C52-H52B 108.5 C35-C34-H34A 107.4 C52-C53-C54 106.8(9) C34-C35-H35A 109.5 C52-C53-H53A 110.4 C34-C35-H35B 109.5 C54-C53-H53A 110.4 H35A-C35-H35B 109.5 C52-C53-H53B 110.4 C34-C35-H35C 109.5 C54-C53-H53B 110.4 H35A-C35-H35C 109.5 H53A-C53-H53B 108.6 H35B-C35-H35C 109.5 C55-C54-C53 101.0(7) C34-C36-H36A 109.5 C55-C54-H54A 111.6 C34-C36-H36B 109.5 C53-C54-H54A 111.6 H36A-C36-H36B 109.5 C55-C54-H54B 111.6 C34-C36-H36C 109.5 C53-C54-H54B 111.6 H36A-C36-H36C 109.5 H54A-C54-H54B 109.4 H36B-C36-H36C 109.5 C54-C55-O51 103.4(6) C45-O41-C42 92.1(4) C54-C55-H55A 111.1 C43-C42-O41 100.5(4) O51-C55-H55A 111.1 C43-C42-H42A 111.7 C54-C55-H55B 111.1 O41-C42-H42A 111.7 O51-C55-H55B 111.1 C43-C42-H42B 111.7 H55A-C55-H55B 109.1 C29-C30-H30A 119.6 O41-C42-H42B 111.7

137

Table C-8. Continued. H42A-C42-H42B 109.4 C44-C43-C42 109.8(4) C44-C43-H43A 109.7 C42-C43-H43A 109.7 C44-C43-H43B 109.7 C42-C43-H43B 109.7 Symmetry transformations used to generate equivalent atoms: #1 -x,-y,-z #2 -x+1,-y+1,-z+1

138

Table C-9. Anisotropic displacement parameters (Å2x 103) for 22. The anisotropicdisplacement factor exponent takes the form: -22[ h2 a*2U11 + ... + 2 h k a* b* U12]. U11 U22 U33 U23 U13 U12 Cu1 16(1) 16(1) 17(1) -4(1) -6(1) 2(1) S1 18(1) 21(1) 27(1) -10(1) -10(1) 3(1) O1 27(1) 24(1) 26(1) -11(1) -15(1) 8(1) N1 17(1) 19(1) 22(1) -5(1) -6(1) 1(1) N2 19(1) 21(1) 23(1) -7(1) -7(1) 3(1) C1 15(1) 17(1) 20(1) -3(1) -5(1) -1(1) C2 18(1) 14(1) 21(1) -2(1) -6(1) 0(1) C3 16(1) 21(1) 22(1) -3(1) -8(1) 1(1) C4 16(1) 20(1) 20(1) -2(1) -6(1) 0(1) C5 19(1) 24(1) 18(1) -3(1) -5(1) -1(1) C6 29(1) 34(1) 28(1) -11(1) -13(1) 9(1) C7 43(1) 54(2) 35(1) -26(1) -20(1) 18(1) C8 41(1) 56(2) 29(1) -20(1) -20(1) 11(1) C9 29(1) 45(1) 27(1) -10(1) -15(1) 9(1) C10 25(1) 32(1) 24(1) -8(1) -10(1) 7(1) C11 17(1) 21(1) 27(1) -7(1) -6(1) 4(1) C12 23(1) 22(1) 35(1) -1(1) -9(1) 4(1) C13 21(1) 25(1) 32(1) -2(1) -8(1) -1(1) C14 25(1) 33(1) 28(1) -17(1) -10(1) 6(1) C15 40(1) 39(1) 26(1) -10(1) -10(1) 6(1) C16 34(1) 36(1) 41(1) -22(1) -4(1) -3(1) Cu2 39(2) 21(2) 19(2) 20(1) -26(1) -29(1) S21 10(2) 10(2) 3(2) 4(1) -1(1) -8(1) O21 30(4) 19(4) 24(2) 4(2) -15(3) -2(2) Cu2' 24(1) 33(2) 30(2) -13(2) -5(1) 13(2) S21' 43(3) 51(4) 42(3) -9(2) -15(2) -5(2) O21' 41(8) 26(7) 37(6) 9(4) -31(6) -15(5) N21 17(1) 23(1) 19(1) -1(1) -6(1) -3(1) N22 19(1) 22(1) 18(1) -2(1) -6(1) -4(1) C21 20(1) 23(1) 17(1) 1(1) -7(1) -3(1) C22 16(1) 17(1) 20(1) -1(1) -4(1) 0(1) C23 20(1) 26(1) 18(1) 0(1) -8(1) -3(1) C24 18(1) 25(1) 20(1) -1(1) -8(1) -1(1) C25 16(1) 25(1) 21(1) -5(1) -5(1) -1(1) C26 24(1) 27(1) 34(1) -2(1) -14(1) -2(1) C27 28(1) 30(1) 42(1) -12(1) -15(1) -2(1) C28 24(1) 41(1) 32(1) -14(1) -14(1) 1(1) C29 24(1) 36(1) 26(1) -5(1) -12(1) 3(1) C30 19(1) 25(1) 24(1) -3(1) -8(1) -1(1) C31 20(1) 25(1) 19(1) -1(1) -3(1) -6(1)

139

Table C-9. Continued. C32 26(1) 34(1) 25(1) -10(1) -3(1) -2(1) C33 31(1) 28(1) 24(1) 3(1) -8(1) -3(1) C34 22(1) 30(1) 22(1) -1(1) -7(1) -11(1) C35 24(1) 39(1) 47(1) -1(1) -16(1) -7(1) C36 37(1) 30(1) 29(1) -5(1) -6(1) -14(1)

140

Table C-10. Hydrogen coordinates (x104) and isotropic displacement parameters (Å2x103) for 22. x y z U(eq) H3A 3187 -1357 1843 23 H4A 1119 285 2286 23 H6A 850 1199 3612 35 H7A 1581 1585 4822 49 H8A 3447 574 5142 47 H9A 4548 -845 4273 39 H10A 3851 -1219 3048 31 H11A 5358 -2970 -1403 26 H12A 5761 -3727 -47 41 H12B 4263 -4027 94 41 H12C 4542 -3008 553 41 H13A 6461 -1708 -944 39 H13B 5080 -1083 -446 39 H13C 5561 -1015 -1557 39 H14A 2422 -2337 -1913 33 H15A 4133 -2312 -3287 52 H15B 5232 -2540 -2740 52 H15C 4384 -1348 -2761 52 H16A 3127 -4178 -2273 54 H16B 2563 -4220 -1167 54 H16C 4134 -4304 -1657 54 H23A 3526 6444 2211 25 H24A 3038 4289 3262 25 H26A 2044 2968 2670 33 H27A 813 2568 1735 38 H28A 169 4035 690 36 H29A 755 5904 591 33 H30A 2006 6299 1512 27 H31A 8314 8304 1530 27 H32A 8223 6295 1589 43 H32B 8114 7058 634 43 H32C 6807 6549 1393 43 H33A 6309 9456 1774 43 H33B 5600 8450 1571 43 H33C 6797 9063 767 43 H34A 7780 7923 3878 29 H35A 9379 6660 3043 53 H35B 10072 7722 3178 53 H35C 9846 7675 2192 53 H36A 7287 9759 3104 48 H36B 8599 9602 2257 48 H36C 8722 9665 3269 48

141

Table C-10. Continued. H42A 1857 7804 6053 52 H42B 3139 7193 5388 52 H43A 1229 8139 4753 38 H43B 2086 7051 4358 38 H44A -395 7005 5761 67 H44B -76 6608 4750 67 H45A 1465 5307 5065 85 H45B 126 5111 5912 85 H52A 1210 6987 6694 77 H52B 2466 7672 6017 77 H53A 1674 7979 4871 114 H53B 296 7887 5699 114 H54A -127 6343 5194 66 H54B 1320 6365 4426 66 H55A 1627 4771 5432 63 H55B 675 5291 6330 63

142

X-ray Tables of 23

Table C-11. Crystal data and structure refinement for 23. Identification code 23 Empirical formula C48H63CoN6O3S3 Formula weight 927.15 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P21/c Unit cell dimensions a = 9.6733(4) Å = 90°  b = 26.7438(10) Å = 92.5405(9)°  c = 18.9117(7) Å = 90° 3 Volume 4887.7(3) Å Z 4 3 Density (calculated) 1.260 Mg/m Absorption coefficient 0.525 mm-1 F(000) 1968 3 Crystal size 0.265 x 0.060 x 0.052 mm Theta range for data collection 1.523 to 34.023°. Index ranges -15≤h≤15, -42≤k≤38, -27≤l≤29 Reflections collected 114933 Independent reflections 18743 [R(int) = 0.0601] Completeness to theta = 25.000° 100.00% Absorption correction Multi Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 18743 / 0 / 560 Goodness-of-fit on F2 1.016 Final R indices [I>2sigma(I)] R1 = 0.0582, wR2 = 0.1133 [12079] R indices (all data) R1 = 0.1105, wR2 = 0.1310 Extinction coefficient n/a -3 Largest diff. peak and hole 0.970 and -0.805 e.Å

R1 = (||Fo| - |Fc||) / |Fo| wR2 = [w(Fo - Fc)] / wFo]]1/2

S = [w(Fo - Fc)] / (n-p)]1/2 w= 1/[(Fo)+(m*p)2+n*p], p = [max(Fo,0)+ 2* Fc]/3, m & n are constants.

143

Table C-12. Atomic coordinates (x104) and equivalent isotropic displacement parameters (Å2x 103) for 23. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. x y z U(eq) Co1 2118(1) 3247(1) 3019(1) 19(1) S1 891(1) 2619(1) 3421(1) 27(1) O1 642(1) 3488(1) 2408(1) 23(1) N1 -802(2) 2784(1) 2228(1) 28(1) N2 -1020(2) 2042(1) 2792(1) 30(1) C1 -358(2) 3249(1) 2117(1) 24(1) C2 -361(2) 2485(1) 2758(1) 25(1) C3 -1194(2) 3523(1) 1569(1) 29(1) C4 -894(2) 3988(1) 1378(1) 28(1) C5 -1644(2) 4302(1) 853(1) 31(1) C6 -2834(2) 4139(1) 472(1) 40(1) C7 -3496(3) 4443(1) -24(2) 49(1) C8 -2998(3) 4916(1) -153(1) 52(1) C9 -1831(3) 5088(1) 222(2) 48(1) C10 -1161(2) 4782(1) 722(1) 38(1) C11 -2319(2) 1940(1) 2355(1) 36(1) C12 -2069(3) 1880(1) 1570(2) 50(1) C13 -3464(2) 2312(1) 2504(2) 43(1) C14 -594(3) 1647(1) 3301(2) 43(1) C15 -1585(4) 1615(1) 3899(2) 68(1) C16 -406(3) 1143(1) 2944(2) 62(1) S21 3900(1) 3009(1) 3723(1) 24(1) O21 1410(1) 3676(1) 3752(1) 19(1) N21 3425(1) 3882(1) 4428(1) 16(1) N22 5677(2) 3700(1) 4222(1) 17(1) C21 2060(2) 3861(1) 4284(1) 16(1) C22 4336(2) 3579(1) 4136(1) 16(1) C23 1175(2) 4113(1) 4798(1) 18(1) C24 1629(2) 4284(1) 5430(1) 18(1) C25 837(2) 4556(1) 5954(1) 17(1) C26 1510(2) 4710(1) 6587(1) 22(1) C27 832(2) 5000(1) 7074(1) 25(1) C28 -536(2) 5138(1) 6942(1) 23(1) C29 -1226(2) 4975(1) 6324(1) 22(1) C30 -555(2) 4690(1) 5836(1) 21(1) C31 6805(2) 3400(1) 3918(1) 24(1) C32 7748(2) 3178(1) 4504(1) 39(1) C33 7593(2) 3707(1) 3394(1) 36(1) C34 6144(2) 4156(1) 4612(1) 18(1) C35 5641(2) 4635(1) 4256(1) 24(1) C36 5860(2) 4131(1) 5398(1) 23(1)

144

Table C-12. Continued. S41 2805(1) 2726(1) 2203(1) 32(1) O41 3163(1) 3804(1) 2679(1) 20(1) N41 3588(4) 3564(1) 1518(1) 63(1) C41 3662(2) 3859(1) 2084(1) 25(1) C42 3128(3) 3092(1) 1495(1) 45(1) C43 4395(3) 4334(1) 1957(1) 35(1) C44 4318(2) 4729(1) 2375(1) 22(1) C45 4982(2) 5216(1) 2288(1) 22(1) C46 5952(2) 5307(1) 1776(1) 30(1) C47 6570(3) 5771(1) 1726(1) 40(1) C48 6236(2) 6157(1) 2180(1) 38(1) C49 5283(2) 6072(1) 2690(1) 31(1) C50 4660(2) 5609(1) 2743(1) 26(1) N42 2572(4) 2832(1) 900(2) 22(1) C51 2114(4) 2298(2) 891(2) 25(1) C52 3414(9) 1986(3) 765(5) 49(2) C53 908(5) 2202(2) 371(3) 35(1) C54 2779(5) 3071(2) 207(2) 32(1) C55 4099(10) 3239(4) 109(5) 66(3) C56 1763(5) 3486(2) 27(3) 40(1) N42' 3389(4) 2929(1) 813(2) 26(1) C51' 2782(4) 2446(1) 572(2) 19(1) C52' 3436(7) 1973(2) 922(4) 31(2) C53' 1238(5) 2418(2) 618(3) 36(1) C54' 4332(7) 3182(3) 232(4) 36(1) C55' 3267(8) 3331(3) -357(4) 72(2) C56' 5459(7) 2893(3) 78(4) 59(2)

145

Table C-13. Bond lengths [Å] and angles [°] for 23. Co1-O1 1.9080(14) C16-H16A 0.98 Co1-O41 1.9255(13) C16-H16B 0.98 Co1-O21 1.9463(13) C16-H16C 0.98 Co1-S41 2.2032(6) S21-C22 1.7543(18) Co1-S1 2.2106(6) O21-C21 1.264(2) Co1-S21 2.2243(5) N21-C22 1.335(2) S1-C2 1.741(2) N21-C21 1.337(2) O1-C1 1.265(2) N22-C22 1.340(2) N1-C1 1.335(3) N22-C34 1.486(2) N1-C2 1.339(3) N22-C31 1.488(2) N2-C2 1.349(3) C21-C23 1.485(2) N2-C14 1.475(3) C23-C24 1.335(3) N2-C11 1.497(3) C23-H23A 0.95 C1-C3 1.481(3) C24-C25 1.472(2) C3-C4 1.332(3) C24-H24A 0.95 C3-H3A 0.95 C25-C26 1.398(2) C4-C5 1.466(3) C25-C30 1.402(2) C4-H4A 0.95 C26-C27 1.390(3) C5-C10 1.393(3) C26-H26A 0.95 C5-C6 1.401(3) C27-C28 1.386(3) C6-C7 1.379(4) C27-H27A 0.95 C6-H6A 0.95 C28-C29 1.389(3) C7-C8 1.379(4) C28-H28A 0.95 C7-H7A 0.95 C29-C30 1.382(3) C8-C9 1.384(4) C29-H29A 0.95 C8-H8A 0.95 C30-H30A 0.95 C9-C10 1.389(4) C31-C33 1.517(3) C9-H9A 0.95 C31-C32 1.525(3) C10-H10A 0.95 C31-H31A 1 C11-C12 1.523(4) C32-H32A 0.98 C11-C13 1.524(3) C32-H32B 0.98 C11-H11A 1 C32-H32C 0.98 C12-H12A 0.98 C33-H33A 0.98 C12-H12B 0.98 C33-H33B 0.98 C12-H12C 0.98 C33-H33C 0.98 C13-H13A 0.98 C34-C35 1.517(3) C13-H13B 0.98 C34-C36 1.525(3) C13-H13C 0.98 C34-H34A 1 C14-C15 1.517(4) C35-H35A 0.98 C14-C16 1.521(4) C35-H35B 0.98 C14-H14A 1 C35-H35C 0.98 C15-H15A 0.98 C36-H36A 0.98 C15-H15B 0.98 C36-H36B 0.98

146

Table C-13. Continued C15-H15C 0.98 C36-H36C 0.98 O41-C41 1.253(2) C51'-C53' 1.502(6) N41-C41 1.329(3) C51'-C52' 1.547(8) N41-C42 1.336(4) C51'-H51B 1 C41-C43 1.480(3) C52'-H52D 0.98 C42-N42' 1.395(4) C52'-H52E 0.98 C42-N42 1.409(4) C52'-H52F 0.98 C43-C44 1.324(3) C53'-H53D 0.98 C43-H43A 0.95 C53'-H53E 0.98 C44-C45 1.465(3) C53'-H53F 0.98 C44-H44A 0.95 C54'-C56' 1.378(9) C45-C46 1.398(3) C54'-C55' 1.536(10) C45-C50 1.401(3) C54'-H54B 1 C46-C47 1.383(3) C55'-H55D 0.98 C46-H46A 0.95 C55'-H55E 0.98 C47-C48 1.388(4) C55'-H55F 0.98 C47-H47A 0.95 C56'-H56D 0.98 C48-C49 1.383(3) C56'-H56E 0.98 C48-H48A 0.95 C56'-H56F 0.98 C49-C50 1.382(3) O1-Co1-O41 85.84(6) C49-H49A 0.95 O1-Co1-O21 87.45(6) C50-H50A 0.95 O41-Co1-O21 89.38(5) N42-C54 1.478(6) O1-Co1-S41 91.70(4) N42-C51 1.496(5) O41-Co1-S41 94.32(4) C51-C53 1.514(6) O21-Co1-S41 176.13(4) C51-C52 1.536(10) O1-Co1-S1 93.75(4) C51-H51A 1 O41-Co1-S1 178.85(4) C52-H52A 0.98 O21-Co1-S1 89.53(4) C52-H52B 0.98 S41-Co1-S1 86.76(2) C52-H52C 0.98 O1-Co1-S21 176.68(5) C53-H53A 0.98 O41-Co1-S21 90.93(4) C53-H53B 0.98 O21-Co1-S21 91.74(4) C53-H53C 0.98 S41-Co1-S21 89.31(2) C54-C55 1.374(10) S1-Co1-S21 89.47(2) C54-C56 1.513(7) C2-S1-Co1 106.06(8) C54-H54A 1 C1-O1-Co1 129.12(14) C55-H55A 0.98 C1-N1-C2 125.53(17) C55-H55B 0.98 C2-N2-C14 122.68(18) C55-H55C 0.98 C2-N2-C11 121.4(2) C56-H56A 0.98 C14-N2-C11 115.73(18) C56-H56B 0.98 O1-C1-N1 130.3(2) C56-H56C 0.98 O1-C1-C3 115.99(19) N42'-C51' 1.483(5) N1-C1-C3 113.71(17)

147

Table C-13. Continued. N42'-C54' 1.609(8) N1-C2-N2 115.15(18) N2-C2-S1 117.62(17) N2-C14-C15 111.0(2) C4-C3-C1 122.29(18) N2-C14-C16 112.3(2) C4-C3-H3A 118.9 C15-C14-C16 111.9(2) C1-C3-H3A 118.9 N2-C14-H14A 107.1 C3-C4-C5 127.66(19) C15-C14-H14A 107.1 C3-C4-H4A 116.2 C16-C14-H14A 107.1 C5-C4-H4A 116.2 C14-C15-H15A 109.5 C10-C5-C6 118.1(2) C14-C15-H15B 109.5 C10-C5-C4 119.2(2) H15A-C15-H15B 109.5 C6-C5-C4 122.7(2) C14-C15-H15C 109.5 C7-C6-C5 120.8(3) H15A-C15-H15C 109.5 C7-C6-H6A 119.6 H15B-C15-H15C 109.5 C5-C6-H6A 119.6 C14-C16-H16A 109.5 C8-C7-C6 120.4(3) C14-C16-H16B 109.5 C8-C7-H7A 119.8 H16A-C16-H16B 109.5 C6-C7-H7A 119.8 C14-C16-H16C 109.5 C7-C8-C9 119.9(3) H16A-C16-H16C 109.5 C7-C8-H8A 120.1 H16B-C16-H16C 109.5 C9-C8-H8A 120.1 C22-S21-Co1 100.51(6) C8-C9-C10 119.9(3) C21-O21-Co1 128.24(11) C8-C9-H9A 120.1 C22-N21-C21 123.69(15) C10-C9-H9A 120.1 C22-N22-C34 121.97(14) C9-C10-C5 120.9(2) C22-N22-C31 123.06(15) C9-C10-H10A 119.5 C34-N22-C31 114.94(14) C5-C10-H10A 119.5 O21-C21-N21 129.09(16) N2-C11-C12 113.01(18) O21-C21-C23 114.68(15) N2-C11-C13 112.31(18) N21-C21-C23 116.13(15) C12-C11-C13 113.4(2) N21-C22-N22 117.21(15) N2-C11-H11A 105.8 N21-C22-S21 124.13(13) C12-C11-H11A 105.8 N22-C22-S21 118.39(13) C13-C11-H11A 105.8 C24-C23-C21 124.37(16) C11-C12-H12A 109.5 C24-C23-H23A 117.8 C11-C12-H12B 109.5 C21-C23-H23A 117.8 H12A-C12-H12B 109.5 C23-C24-C25 127.71(16) C11-C12-H12C 109.5 C23-C24-H24A 116.1 H12A-C12-H12C 109.5 C25-C24-H24A 116.1 H12B-C12-H12C 109.5 C26-C25-C30 117.96(16) C11-C13-H13A 109.5 C26-C25-C24 119.04(15) C11-C13-H13B 109.5 C30-C25-C24 122.94(16) H13A-C13-H13B 109.5 C27-C26-C25 121.05(17) C11-C13-H13C 109.5 C27-C26-H26A 119.5 H13A-C13-H13C 109.5 C25-C26-H26A 119.5

148

Table C-13. Continued. H13B-C13-H13C 109.5 C28-C27-C26 120.26(18) C26-C27-H27A 119.9 C34-C36-H36C 109.5 C27-C28-C29 119.15(18) H36A-C36-H36C 109.5 C27-C28-H28A 120.4 H36B-C36-H36C 109.5 C29-C28-H28A 120.4 C42-S41-Co1 105.06(8) C30-C29-C28 120.86(17) C41-O41-Co1 128.31(13) C30-C29-H29A 119.6 C41-N41-C42 126.3(2) C28-C29-H29A 119.6 O41-C41-N41 130.1(2) C29-C30-C25 120.69(17) O41-C41-C43 116.92(18) C29-C30-H30A 119.7 N41-C41-C43 112.88(19) C25-C30-H30A 119.7 N41-C42-N42' 104.6(3) N22-C31-C33 111.07(17) N41-C42-N42 127.3(2) N22-C31-C32 110.77(17) N41-C42-S41 126.15(17) C33-C31-C32 112.60(18) N42'-C42-S41 126.5(2) N22-C31-H31A 107.4 N42-C42-S41 105.6(2) C33-C31-H31A 107.4 C44-C43-C41 123.30(19) C32-C31-H31A 107.4 C44-C43-H43A 118.3 C31-C32-H32A 109.5 C41-C43-H43A 118.3 C31-C32-H32B 109.5 C43-C44-C45 127.31(19) H32A-C32-H32B 109.5 C43-C44-H44A 116.3 C31-C32-H32C 109.5 C45-C44-H44A 116.3 H32A-C32-H32C 109.5 C46-C45-C50 117.93(19) H32B-C32-H32C 109.5 C46-C45-C44 122.95(19) C31-C33-H33A 109.5 C50-C45-C44 119.11(18) C31-C33-H33B 109.5 C47-C46-C45 120.6(2) H33A-C33-H33B 109.5 C47-C46-H46A 119.7 C31-C33-H33C 109.5 C45-C46-H46A 119.7 H33A-C33-H33C 109.5 C46-C47-C48 120.7(2) H33B-C33-H33C 109.5 C46-C47-H47A 119.6 N22-C34-C35 112.91(14) C48-C47-H47A 119.6 N22-C34-C36 112.50(15) C49-C48-C47 119.3(2) C35-C34-C36 113.73(16) C49-C48-H48A 120.4 N22-C34-H34A 105.6 C47-C48-H48A 120.4 C35-C34-H34A 105.6 C50-C49-C48 120.3(2) C36-C34-H34A 105.6 C50-C49-H49A 119.8 C34-C35-H35A 109.5 C48-C49-H49A 119.8 C34-C35-H35B 109.5 C49-C50-C45 121.1(2) H35A-C35-H35B 109.5 C49-C50-H50A 119.4 C34-C35-H35C 109.5 C45-C50-H50A 119.4 H35A-C35-H35C 109.5 C42-N42-C54 115.7(3) H35B-C35-H35C 109.5 C42-N42-C51 125.6(3) C34-C36-H36A 109.5 C54-N42-C51 117.0(3) C34-C36-H36B 109.5 N42-C51-C53 112.9(4)

149

Table C-13. Continued. H36A-C36-H36B 109.5 N42-C51-C52 106.0(4) C53-C51-C52 114.8(4) C51'-C52'-H52D 109.5 N42-C51-H51A 107.6 C51'-C52'-H52E 109.5 C53-C51-H51A 107.6 H52D-C52'-H52E 109.5 C52-C51-H51A 107.6 C51'-C52'-H52F 109.5 C51-C52-H52A 109.5 H52D-C52'-H52F 109.5 C51-C52-H52B 109.5 H52E-C52'-H52F 109.5 H52A-C52-H52B 109.5 C51'-C53'-H53D 109.5 C51-C52-H52C 109.5 C51'-C53'-H53E 109.5 H52A-C52-H52C 109.5 H53D-C53'-H53E 109.5 H52B-C52-H52C 109.5 C51'-C53'-H53F 109.5 C51-C53-H53A 109.5 H53D-C53'-H53F 109.5 C51-C53-H53B 109.5 H53E-C53'-H53F 109.5 H53A-C53-H53B 109.5 C56'-C54'-C55' 120.2(6) C51-C53-H53C 109.5 C56'-C54'-N42' 112.7(5) H53A-C53-H53C 109.5 C55'-C54'-N42' 102.8(5) H53B-C53-H53C 109.5 C56'-C54'-H54B 106.8 C55-C54-N42 115.1(5) C55'-C54'-H54B 106.8 C55-C54-C56 109.1(6) N42'-C54'-H54B 106.8 N42-C54-C56 113.8(4) C54'-C55'-H55D 109.5 C55-C54-H54A 106 C54'-C55'-H55E 109.5 N42-C54-H54A 106 H55D-C55'-H55E 109.5 C56-C54-H54A 106 C54'-C55'-H55F 109.5 C54-C55-H55A 109.5 H55D-C55'-H55F 109.5 C54-C55-H55B 109.5 H55E-C55'-H55F 109.5 H55A-C55-H55B 109.5 C54'-C56'-H56D 109.5 C54-C55-H55C 109.5 C54'-C56'-H56E 109.5 H55A-C55-H55C 109.5 H56D-C56'-H56E 109.5 H55B-C55-H55C 109.5 C54'-C56'-H56F 109.5 C54-C56-H56A 109.5 H56D-C56'-H56F 109.5 C54-C56-H56B 109.5 H56E-C56'-H56F 109.5 H56A-C56-H56B 109.5 C52'-C51'-H51B 105.9 C54-C56-H56C 109.5 H56A-C56-H56C 109.5 H56B-C56-H56C 109.5 C42-N42'-C51' 118.2(3) C42-N42'-C54' 129.1(4) C51'-N42'-C54' 112.7(4) N42'-C51'-C53' 114.0(3) N42'-C51'-C52' 115.6(4) C53'-C51'-C52' 108.8(4) N42'-C51'-H51B 105.9 C53'-C51'-H51B 105.9

150

Table C-14. Anisotropic displacement parameters (Å2x 103) for 23. The anisotropic displacement factor exponent takes the form: -22[ h2 a*2U11 + ... + 2 h k a* b* U12] U11 U22 U33 U23 U13 U12 Co1 14(1) 19(1) 25(1) -8(1) 3(1) -3(1) S1 22(1) 23(1) 35(1) -3(1) -1(1) -8(1) O1 18(1) 27(1) 24(1) -9(1) 1(1) -4(1) N1 22(1) 32(1) 30(1) -12(1) 3(1) -9(1) N2 22(1) 24(1) 45(1) -13(1) 5(1) -7(1) C1 18(1) 33(1) 22(1) -12(1) 4(1) -6(1) C2 18(1) 26(1) 34(1) -13(1) 7(1) -6(1) C3 21(1) 41(1) 26(1) -10(1) -2(1) -6(1) C4 18(1) 41(1) 24(1) -11(1) 2(1) -1(1) C5 24(1) 43(1) 26(1) -7(1) 5(1) 3(1) C6 32(1) 49(2) 39(1) -8(1) -7(1) 4(1) C7 38(1) 65(2) 44(2) -5(1) -7(1) 12(1) C8 44(1) 76(2) 37(1) 11(1) 10(1) 26(2) C9 43(1) 56(2) 47(2) 13(1) 20(1) 10(1) C10 29(1) 48(1) 37(1) 1(1) 10(1) 2(1) C11 19(1) 29(1) 59(2) -20(1) 4(1) -7(1) C12 35(1) 56(2) 57(2) -34(1) 1(1) -14(1) C13 23(1) 30(1) 76(2) -16(1) 7(1) -7(1) C14 36(1) 26(1) 66(2) -4(1) 1(1) -10(1) C15 83(2) 45(2) 76(2) 11(2) 25(2) -10(2) C16 37(1) 32(1) 114(3) -18(2) -6(2) 2(1) S21 17(1) 16(1) 40(1) -9(1) -2(1) 2(1) O21 14(1) 21(1) 22(1) -4(1) 1(1) -1(1) N21 13(1) 15(1) 21(1) -2(1) 2(1) 0(1) N22 13(1) 17(1) 22(1) -4(1) 1(1) 1(1) C21 15(1) 13(1) 20(1) 0(1) 3(1) -1(1) C22 15(1) 15(1) 18(1) 0(1) 2(1) -1(1) C23 13(1) 18(1) 25(1) -2(1) 4(1) 0(1) C24 13(1) 17(1) 25(1) -1(1) 3(1) 0(1) C25 14(1) 15(1) 21(1) -2(1) 3(1) 0(1) C26 16(1) 25(1) 23(1) -2(1) 1(1) 3(1) C27 23(1) 32(1) 21(1) -5(1) -1(1) 2(1) C28 22(1) 24(1) 22(1) -3(1) 6(1) 3(1) C29 16(1) 25(1) 26(1) -2(1) 4(1) 4(1) C30 15(1) 24(1) 23(1) -5(1) 0(1) 0(1) C31 16(1) 25(1) 32(1) -10(1) 4(1) 4(1) C32 28(1) 38(1) 52(2) 2(1) 3(1) 18(1) C33 26(1) 51(1) 31(1) -6(1) 11(1) 5(1) C34 14(1) 18(1) 21(1) -3(1) 0(1) -1(1) C35 26(1) 18(1) 27(1) 1(1) 1(1) -2(1) C36 22(1) 26(1) 19(1) -3(1) -2(1) 4(1)

151

Table C-14. Continued. S41 31(1) 21(1) 44(1) -17(1) 15(1) -8(1) O41 18(1) 25(1) 18(1) -6(1) 3(1) -6(1) N41 152(3) 20(1) 20(1) 0(1) 26(1) 16(1) C41 36(1) 23(1) 18(1) 3(1) 6(1) 12(1) C42 85(2) 28(1) 21(1) -11(1) -17(1) 30(1) C43 52(1) 25(1) 29(1) 10(1) 24(1) 12(1) C44 18(1) 30(1) 18(1) 5(1) 2(1) 4(1) C45 19(1) 26(1) 22(1) 5(1) 0(1) 4(1) C46 28(1) 33(1) 30(1) 5(1) 9(1) 4(1) C47 36(1) 37(1) 48(1) 11(1) 17(1) -1(1) C48 34(1) 29(1) 51(2) 11(1) 4(1) 0(1) C49 29(1) 28(1) 36(1) 2(1) -2(1) 6(1) C50 22(1) 31(1) 25(1) 2(1) 1(1) 4(1)

152

Table C-15. Hydrogen coordinates (x104) and isotropic displacement parameters (Å2x 103) for 23. x y z U(eq) H3A -1972 3361 1346 35 H4A -101 4135 1608 34 H6A -3190 3814 557 49 H7A -4300 4326 -280 59 H8A -3455 5124 -497 62 H9A -1489 5415 137 57 H10A -363 4903 978 45 H11A -2660 1608 2517 43 H12A -2939 1789 1317 74 H12B -1382 1616 1507 74 H12C -1726 2196 1382 74 H13A -4334 2200 2270 64 H13B -3219 2642 2321 64 H13C -3576 2334 3015 64 H14A 328 1746 3516 52 H15A -1271 1355 4234 102 H15B -2514 1532 3705 102 H15C -1612 1938 4143 102 H16A 14 907 3286 92 H16B 198 1183 2545 92 H16C -1309 1016 2771 92 H23A 223 4157 4668 22 H24A 2574 4222 5557 22 H26A 2446 4615 6685 26 H27A 1309 5103 7499 30 H28A -996 5341 7270 27 H29A -2170 5062 6237 27 H30A -1043 4583 5416 25 H31A 6361 3116 3651 29 H32A 8474 2979 4294 59 H32B 7203 2963 4807 59 H32C 8171 3448 4789 59 H33A 8325 3501 3201 53 H33B 8006 3999 3635 53 H33C 6956 3818 3008 53 H34A 7175 4161 4587 21 H35A 6111 4922 4483 35 H35B 4640 4668 4301 35 H35C 5849 4626 3753 35 H36A 6317 4412 5645 34 H36B 6220 3815 5595 34

153

Table C-15. Continued. H36C 4860 4149 5458 34 H43A 4949 4356 1556 42 H44A 3772 4693 2777 26 H46A 6187 5048 1460 36 H47A 7231 5827 1378 48 H48A 6658 6476 2140 45 H49A 5055 6332 3005 37 H50A 4002 5556 3094 32 H51A 1804 2215 1375 30 H52A 3187 1630 792 73 H52B 4135 2068 1127 73 H52C 3749 2063 296 73 H53A 494 1877 473 52 H53B 1235 2202 -112 52 H53C 213 2465 417 52 H54A 2598 2804 -156 38 H55A 4285 3227 -396 99 H55B 4764 3025 373 99 H55C 4190 3584 279 99 H56A 828 3378 136 60 H56B 1790 3565 -478 60 H56C 2013 3784 306 60 H51B 2967 2420 57 23 H52D 3104 1675 665 46 H52E 3171 1953 1415 46 H52F 4446 1993 907 46 H53D 883 2132 341 54 H53E 819 2726 430 54 H53F 1004 2376 1114 54 H54B 4703 3500 444 43 H55D 2446 3472 -146 108 H55E 3002 3034 -637 108 H55F 3674 3580 -665 108 H56D 6003 2820 516 88 H56E 6036 3074 -249 88 H56F 5141 2579 -141 88

154

X-ray Tables of 24·NCCH3

Table C-16. Crystal data and structure refinement for 24·NCCH3. Identification code 24·NCCH3 Empirical formula C34H45ZnN5O2S2 Formula weight 685.24 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P21/n Unit cell dimensions a = 13.8922(16) Å = 90°  b = 7.4030(9) Å = 99.859(2)°  c = 34.250(4) Å  = 90° 3 Volume 3470.4(7) Å Z 4 3 Density (calculated) 1.312 Mg/m Absorption coefficient 0.865 mm-1 F(000) 1448 3 Crystal size 0.304 x 0.036 x 0.035 mm Theta range for data collection 1.207 to 28.323°. Index ranges -15≤h≤18, -9≤k≤9, -45≤l≤43 Reflections collected 34093 Independent reflections 8616 [R(int) = 0.1359] Completeness to theta = 25.000° 100.00% Absorption correction Integration Max. and min. transmission 0.9788 and 0.8661 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8616 / 0 / 406 Goodness-of-fit on F2 0.829 Final R indices [I>2sigma(I)] R1 = 0.0498, wR2 = 0.0892 [3934] R indices (all data) R1 = 0.1435, wR2 = 0.1210 -3 Largest diff. peak and hole 1.188 and -0.545 e.Å R1 = (||Fo| - |Fc||) / |Fo| wR2 = [w(Fo - Fc)] / wFo]]1/2

S = [w(Fo - Fc)] / (n-p)]1/2 w= 1/[ (Fo)+(m*p)2+n*p], p = [max(Fo,0)+ 2* Fc]/3, m & n are constants.

155

Table C-17. Atomic coordinates (x104) and equivalent isotropic displacement parameters (Å2x 3 10 ) for 24·NCCH3. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. x y z U(eq) Zn1 8273(1) 4928(1) 4347(1) 18(1) S1 8665(1) 5853(2) 4986(1) 17(1) S2 8518(1) 1987(2) 4162(1) 19(1) O1 6888(2) 5639(4) 4245(1) 22(1) O2 8715(2) 6125(4) 3899(1) 19(1) N1 11116(3) 4921(6) 2138(1) 40(1) N2 6741(2) 6987(4) 4851(1) 16(1) N3 7837(2) 8813(4) 5233(1) 14(1) N4 8425(2) 3716(5) 3454(1) 16(1) N5 7247(2) 1558(4) 3478(1) 15(1) C1 12116(3) 4762(6) 1563(1) 29(1) C2 11554(3) 4851(7) 1886(1) 26(1) C3 6414(3) 6149(6) 4513(1) 17(1) C4 5338(3) 5917(5) 4426(1) 16(1) C5 4886(3) 5417(5) 4070(1) 15(1) C6 3837(3) 5243(5) 3923(1) 13(1) C7 3127(3) 5413(5) 4165(1) 17(1) C8 2145(3) 5290(5) 4010(1) 19(1) C9 1853(3) 4948(7) 3607(1) 24(1) C10 2539(3) 4757(6) 3364(1) 25(1) C11 3522(3) 4902(6) 3520(1) 22(1) C12 7684(3) 7314(5) 5009(1) 14(1) C13 8821(3) 9329(5) 5448(1) 14(1) C14 8835(3) 9447(6) 5896(1) 21(1) C15 9175(3) 11064(6) 5285(1) 23(1) C16 7040(3) 10101(6) 5278(1) 18(1) C17 6247(3) 9276(6) 5485(1) 21(1) C18 6613(3) 11029(6) 4887(1) 23(1) C19 8796(3) 5285(6) 3580(1) 16(1) C20 9319(3) 6193(6) 3295(1) 16(1) C21 9569(3) 7931(6) 3322(1) 18(1) C22 10100(3) 8915(6) 3053(1) 19(1) C23 10641(3) 8017(6) 2806(1) 21(1) C24 11144(3) 8955(7) 2558(1) 22(1) C25 11104(3) 10816(7) 2544(1) 22(1) C26 10568(3) 11742(6) 2783(1) 22(1) C27 10077(3) 10795(6) 3041(1) 19(1) C28 8005(3) 2493(5) 3669(1) 16(1) C29 6785(3) 80(6) 3672(1) 20(1) C30 6641(3) -1608(6) 3417(1) 26(1) C31 5831(3) 758(6) 3791(1) 28(1)

156

Table C-17. Continued. C32 6752(3) 2043(6) 3069(1) 17(1) C33 6396(3) 3995(6) 3036(1) 25(1) C34 7368(3) 1538(6) 2756(1) 24(1)

157

Table C-18. Bond lengths [Å] and angles [°] for 24·NCCH3. Zn1-O2 1.959(3) C14-H12 0.98 Zn1-O1 1.968(3) C15-H16 0.98 Zn1-S1 2.2706(11) C15-H15 0.98 Zn1-S2 2.3083(12) C15-H17 0.98 S1-C12 1.752(4) C16-C18 1.530(5) S2-C28 1.759(4) C16-C17 1.537(5) O1-C3 1.274(4) C16-H21 1 O2-C19 1.277(4) C17-H19 0.98 N1-C2 1.140(5) C17-H20 0.98 N2-C3 1.324(5) C17-H18 0.98 N2-C12 1.351(5) C18-H24 0.98 N3-C12 1.345(5) C18-H23 0.98 N3-C16 1.488(5) C18-H22 0.98 N3-C13 1.489(5) C19-C20 1.476(5) N4-C19 1.313(5) C20-C21 1.332(5) N4-C28 1.359(5) C20-H31 0.95 N5-C28 1.333(5) C21-C22 1.469(5) N5-C29 1.483(5) C21-H25 0.95 N5-C32 1.495(5) C22-C23 1.391(5) C1-C2 1.460(5) C22-C27 1.393(5) C1-H1 0.98 C23-C24 1.378(5) C1-H2 0.98 C23-H30 0.95 C1-H3 0.98 C24-C25 1.379(5) C3-C4 1.482(5) C24-H29 0.95 C4-C5 1.326(5) C25-C26 1.380(6) C4-H10 0.95 C25-H28 0.95 C5-C6 1.463(5) C26-C27 1.395(6) C5-H4 0.95 C26-H27 0.95 C6-C11 1.397(5) C27-H26 0.95 C6-C7 1.398(5) C29-C30 1.518(6) C7-C8 1.380(5) C29-C31 1.536(5) C7-H5 0.95 C29-H38 1 C8-C9 1.393(5) C30-H34 0.98 C8-H6 0.95 C30-H33 0.98 C9-C10 1.377(5) C30-H32 0.98 C9-H9 0.95 C31-H35 0.98 C10-C11 1.383(5) C31-H37 0.98 C10-H7 0.95 C31-H36 0.98 C11-H8 0.95 C32-C33 1.525(6) C13-C15 1.516(5) C32-C34 1.529(5) C13-C14 1.532(5) C32-H39 1 C13-H14 1 C33-H42 0.98 C14-H13 0.98 C33-H40 0.98

158

Table C-18. Continued. C14-H11 0.98 C33-H41 0.98 C34-H43 0.98 C7-C8-C9 119.6(4) C34-H45 0.98 C7-C8-H6 120.2 O2-Zn1-O1 99.70(11) C9-C8-H6 120.2 O2-Zn1-S1 124.32(9) C10-C9-C8 120.3(4) O1-Zn1-S1 98.94(8) C10-C9-H9 119.8 O2-Zn1-S2 97.58(8) C8-C9-H9 119.8 O1-Zn1-S2 112.89(9) C9-C10-C11 119.7(4) S1-Zn1-S2 121.76(4) C9-C10-H7 120.2 C12-S1-Zn1 99.71(13) C11-C10-H7 120.2 C28-S2-Zn1 90.45(13) C10-C11-C6 121.3(3) C3-O1-Zn1 124.2(2) C10-C11-H8 119.4 C19-O2-Zn1 122.2(3) C6-C11-H8 119.4 C3-N2-C12 126.8(3) N3-C12-N2 115.1(3) C12-N3-C16 122.7(3) N3-C12-S1 119.1(3) C12-N3-C13 122.4(3) N2-C12-S1 125.5(3) C16-N3-C13 114.9(3) N3-C13-C15 111.2(3) C19-N4-C28 126.4(3) N3-C13-C14 111.3(3) C28-N5-C29 122.1(3) C15-C13-C14 111.8(3) C28-N5-C32 122.1(3) N3-C13-H14 107.4 C29-N5-C32 115.6(3) C15-C13-H14 107.4 C2-C1-H1 109.5 C14-C13-H14 107.4 C2-C1-H2 109.5 C13-C14-H13 109.5 H1-C1-H2 109.5 C13-C14-H11 109.5 C2-C1-H3 109.5 H13-C14-H11 109.5 H1-C1-H3 109.5 C13-C14-H12 109.5 H2-C1-H3 109.5 H13-C14-H12 109.5 N1-C2-C1 179.9(5) H11-C14-H12 109.5 O1-C3-N2 128.7(4) C13-C15-H16 109.5 O1-C3-C4 116.7(4) C13-C15-H15 109.5 N2-C3-C4 114.4(3) H16-C15-H15 109.5 C5-C4-C3 121.2(4) C13-C15-H17 109.5 C5-C4-H10 119.4 H16-C15-H17 109.5 C3-C4-H10 119.4 H15-C15-H17 109.5 C4-C5-C6 128.9(4) N3-C16-C18 112.4(3) C4-C5-H4 115.5 N3-C16-C17 113.4(3) C6-C5-H4 115.5 C18-C16-C17 112.4(3) C11-C6-C7 117.9(3) N3-C16-H21 105.9 C11-C6-C5 118.7(3) C18-C16-H21 105.9 C7-C6-C5 123.3(3) C17-C16-H21 105.9 C8-C7-C6 121.1(4) C16-C17-H19 109.5 C8-C7-H5 119.4 C16-C17-H20 109.5 C6-C7-H5 119.4 H19-C17-H20 109.5

159

Table C-16. Continued. C16-C17-H18 109.5 C30-C29-H38 107.4 H19-C17-H18 109.5 C31-C29-H38 107.4 H20-C17-H18 109.5 C29-C30-H34 109.5 C16-C18-H24 109.5 C29-C30-H33 109.5 C16-C18-H23 109.5 H34-C30-H33 109.5 H24-C18-H23 109.5 C29-C30-H32 109.5 C16-C18-H22 109.5 H34-C30-H32 109.5 H24-C18-H22 109.5 H33-C30-H32 109.5 H23-C18-H22 109.5 C29-C31-H35 109.5 O2-C19-N4 128.6(4) C29-C31-H37 109.5 O2-C19-C20 117.9(4) H35-C31-H37 109.5 N4-C19-C20 113.3(3) C29-C31-H36 109.5 C21-C20-C19 123.1(4) H35-C31-H36 109.5 C21-C20-H31 118.4 H37-C31-H36 109.5 C19-C20-H31 118.4 N5-C32-C33 112.9(3) C20-C21-C22 126.1(4) N5-C32-C34 112.0(3) C20-C21-H25 117 C33-C32-C34 113.1(3) C22-C21-H25 117 N5-C32-H39 106 C23-C22-C27 118.2(4) C33-C32-H39 106 C23-C22-C21 121.7(4) C34-C32-H39 106 C27-C22-C21 120.1(4) C32-C33-H42 109.5 C24-C23-C22 121.1(4) C32-C33-H40 109.5 C24-C23-H30 119.4 H42-C33-H40 109.5 C22-C23-H30 119.4 C32-C33-H41 109.5 C23-C24-C25 120.2(4) H42-C33-H41 109.5 C23-C24-H29 119.9 H40-C33-H41 109.5 C25-C24-H29 119.9 C32-C34-H44 109.5 C24-C25-C26 119.9(4) C32-C34-H43 109.5 C24-C25-H28 120.1 H44-C34-H43 109.5 C26-C25-H28 120.1 C30-C29-H38 107.4 C25-C26-C27 119.9(4) C31-C29-H38 107.4 C25-C26-H27 120 C29-C30-H34 109.5 C27-C26-H27 120 C29-C30-H33 109.5 C22-C27-C26 120.6(4) H34-C30-H33 109.5 C22-C27-H26 119.7 C29-C30-H32 109.5 C26-C27-H26 119.7 H34-C30-H32 109.5 N5-C28-N4 117.3(3) H33-C30-H32 109.5 N5-C28-S2 121.0(3) C29-C31-H35 109.5 N4-C28-S2 121.3(3) C29-C31-H37 109.5 N5-C29-C30 112.0(3) H35-C31-H37 109.5 N5-C29-C31 109.7(4) C29-C31-H36 109.5 C30-C29-C31 112.8(3) H35-C31-H36 109.5 N5-C29-H38 107.4 H37-C31-H36 109.5

160

Table C-18. Continued. C32-C34-H45 109.5 H44-C34-H45 109.5 H43-C34-H45 109.5

161

2 3 Table C-19. Anisotropic displacement parameters (Å x 10 ) for 24·NCCH3. The anisotropic displacement factor exponent takes the form: -22[ h2 a*2U11 + ... + 2 h k a* b* U12] 11 22 33 23 13 12 U U U U U U Zn1 18(1) 23(1) 13(1) -4(1) 5(1) -1(1) S1 19(1) 20(1) 12(1) -5(1) 2(1) 3(1) S2 22(1) 20(1) 14(1) -1(1) 1(1) 1(1) O1 18(2) 39(2) 12(1) -5(1) 5(1) 2(1) O2 23(2) 21(2) 15(2) -6(1) 6(1) -4(1) N1 53(3) 32(2) 42(2) 2(2) 23(2) 2(3) N2 15(2) 20(2) 12(2) -5(2) 1(1) -2(2) N3 15(2) 14(2) 12(2) -2(2) 3(1) 1(2) N4 16(2) 18(2) 13(2) -1(2) 3(1) -3(2) N5 16(2) 16(2) 12(2) -2(2) 2(1) -1(2) C1 32(3) 35(3) 21(2) -3(2) 7(2) 0(2) C2 31(2) 22(3) 25(2) -2(2) 3(2) 1(2) C3 16(2) 18(2) 19(2) 6(2) 4(2) 3(2) C4 15(2) 17(2) 18(2) -2(2) 6(2) -1(2) C5 14(2) 16(3) 16(2) 1(2) 5(2) 2(2) C6 15(2) 9(2) 15(2) 3(2) 2(2) 1(2) C7 21(2) 17(3) 14(2) -2(2) 1(2) 1(2) C8 16(2) 17(3) 25(2) 3(2) 8(2) 1(2) C9 17(2) 30(3) 23(2) -1(2) 1(2) 0(2) C10 18(2) 41(3) 14(2) -6(2) -3(2) 2(2) C11 16(2) 29(3) 21(2) -1(2) 7(2) 2(2) C12 16(2) 17(2) 11(2) 2(2) 4(2) 1(2) C13 12(2) 15(2) 15(2) -4(2) 1(2) -2(2) C14 23(2) 23(3) 16(2) -4(2) 2(2) -4(2) C15 19(2) 26(3) 24(2) 4(2) 6(2) -3(2) C16 18(2) 15(2) 19(2) -7(2) 2(2) 3(2) C17 20(2) 27(3) 17(2) -3(2) 6(2) 3(2) C18 18(2) 29(3) 24(2) 1(2) 6(2) 7(2) C19 10(2) 20(3) 16(2) 1(2) -1(2) 2(2) C20 13(2) 26(3) 11(2) -2(2) 3(2) 2(2) C21 13(2) 31(3) 11(2) -1(2) 3(2) 0(2) C22 15(2) 28(3) 12(2) 0(2) 1(2) -3(2) C23 20(2) 20(3) 23(2) 1(2) 4(2) -1(2) C24 21(2) 29(3) 16(2) 1(2) 5(2) -1(2) C25 22(2) 28(3) 15(2) 0(2) 6(2) -6(2) C26 26(3) 16(2) 23(2) 0(2) 2(2) -5(2) C27 17(2) 28(3) 11(2) -5(2) 0(2) -1(2) C28 17(2) 15(2) 17(2) -6(2) 5(2) 0(2) C29 25(2) 18(2) 18(2) -2(2) 4(2) -6(2) C30 29(3) 20(3) 29(3) 0(2) 3(2) -4(2) C31 31(3) 29(3) 24(2) -3(2) 11(2) -6(2)

162

Table C-19. Continued. C32 19(2) 22(2) 10(2) 0(2) 1(2) -4(2) C33 25(3) 26(3) 23(2) -1(2) 1(2) 5(2) C34 31(3) 29(3) 11(2) -1(2) 1(2) 1(2)

163

Table C-20. Hydrogen coordinates (x104) and isotropic displacement parameters (Å2x 103) for 24·NCCH3. x y z U(eq) H1 12766 4265 1663 44 H2 12183 5978 1458 44 H3 11778 3982 1352 44 H10 4966 6130 4630 20 H4 5299 5134 3885 18 H5 3325 5618 4441 21 H6 1671 5438 4177 23 H9 1178 4846 3500 28 H7 2338 4526 3089 30 H8 3992 4768 3351 26 H14 9285 8349 5403 17 H13 8468 10516 5954 31 H11 9512 9541 6034 31 H12 8534 8361 5985 31 H16 9144 10945 4998 34 H15 9850 11299 5412 34 H17 8758 12068 5339 34 H21 7353 11079 5457 21 H19 5816 8509 5298 32 H20 5864 10245 5579 32 H18 6555 8546 5711 32 H24 7144 11527 4765 35 H23 6177 12008 4939 35 H22 6244 10148 4707 35 H31 9485 5505 3082 20 H25 9387 8597 3535 22 H30 10663 6734 2809 25 H29 11520 8319 2396 26 H28 11445 11459 2370 26 H27 10534 13023 2772 26 H26 9723 11439 3210 23 H38 7239 -240 3922 25 H34 6119 -1399 3190 40 H33 6461 -2620 3575 40 H32 7250 -1894 3322 40 H35 5965 1838 3957 41 H37 5558 -187 3940 41 H36 5361 1055 3552 41 H39 6152 1274 3014 21 H42 6083 4289 3264 37 H40 5923 4146 2791 37

164

Table C-20. Continued. H41 6952 4804 3033 37 H44 7945 2321 2784 36 H43 6979 1695 2492 36 H45 7576 276 2792 36

165

APPENDIX D FULL TANDEM MASS SPECTRUM

D:\Desktop\...\637w3_CID-35 12/14/2017 3:17:03 PM ZA-222_637w3_CID-35_50avg

637w3_CID-35 #1-50 RT: 0.01-1.23 AV: 50 NL: 1.33E5 T: + p ESI Full ms2 [email protected] [175.00-2000.00] 510.93 100 95 473.13 507.07 90 85 80 75 70 65 60 347.07 55 349.07 50 45 535.87

40 Relative Abundance Relative 595.00 35 364.87 451.13 30 483.07 25 20 511.93 15 489.07 287.13 321.13 423.20 448.07 10 536.87 596.00 332.93 5 319.13 365.93 406.00 452.13 431.07 532.93 538.00 561.07 0 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 620 640 m/z Figure D-1. Fragments observed in the tandem mass spectrometry experiments of 20.

166

LIST OF REFERENCES

1. G. E. Moore, Proceedings of the IEEE, 1998, 86, 82-85.

2. G. E. Moore, IEEE Solid-State Circuits Society Newsletter, 2006, 11, 36- 37.

3. M. M. Waldrop, Nature (London, U. K.), 2016, 530, 144-147.

4. H. Bu, 5 nanometer transistors inching their way into chips, https://www.ibm.com/blogs/think/2017/06/5-nanometer-transistors/, (accessed 01/03, 2018).

5. P. Iqbal, J. A. Preece and P. M. Mendes, Nanotechnology: the "top-down" and "bottom-up" approaches, 2012.

6. K. L. Berkowski, K. N. Plunkett, Q. Yu and J. S. Moore, J. Chem. Educ., 2005, 82, 1365-1369.

7. Y. Huang, E. Sutter, N. N. Shi, J. Zheng, T. Yang, D. Englund, H.-J. Gao and P. Sutter, ACS Nano, 2015, 9, 10612-10620.

8. N. J. Kadhim, S. H. Laurie and D. Mukherjee, J. Chem. Educ., 1998, 75, 840-843.

9. S. C. Singh and H. Zeng, Sci. Adv. Mater., 2012, 4, 368-390.

10. K. J. Kanarik, T. Lill, E. A. Hudson, S. Sriraman, S. Tan, J. Marks, V. Vahedi and R. A. Gottscho, J. Vac. Sci. Technol., A, 2015, 33, 020802/020801- 020802/020814.

11. D. M. Mattox, Handbook of Physical Vapor Deposition (PVD) Processing: Film Formation, Adhesion, Surface Preparation and Contamination Control, Noyes, 1998.

12. A. C. Jones, M. L. Hitchman and Editors, Chemical Vapour Deposition; Precursors, Processes and Applications, Royal Society of Chemistry, 2009.

13. J. E. Greene, J. Vac. Sci. Technol., A, 2017, 35, 05C204/201- 205C204/260.

14. S. A. Chambers, Surf. Sci. Rep., 2000, 39, 105-180.

15. J. Hämäläinen, M. Ritala and M. Leskelä, Chem. Mater., 2013, 26, 786- 801.

16. G. Hodes, Phys. Chem. Chem. Phys., 2007, 9, 2181-2196.

167

17. N. T. K. Thanh, N. Maclean and S. Mahiddine, Chem. Rev., 2014, 114, 7610-7630.

18. A. C. Jones, Chem. Soc. Rev., 1997, 26, 101-110.

19. C. S. Blackman and I. P. Parkin, Chem. Mater., 2005, 17, 1583-1590.

20. J. B. Mullin, S. J. C. Irvine and D. J. Ashen, J. Cryst. Growth, 1981, 55, 92-106.

21. G. B. Stringfellow, Annu. Rev. Mater. Sci, 1978, 8, 73-98.

22. A. Tuchscherer, C. Georgi, N. Roth, D. Schaarschmidt, T. Rueffer, T. Waechtler, S. E. Schulz, S. Oswald, T. Gessner and H. Lang, Eur. J. Inorg. Chem., 2012, 2012, 4867-4876.

23. M. Nagai and K. Kishida, Appl. Surf. Sci., 1993, 70–71, Part 2, 759-762.

24. E. P. Boyd, D. R. Ketchum, H. Deng and S. G. Shore, Chem. Mater., 1997, 9, 1154-1158.

25. A. N. Gleizes, Chem. Vap. Deposition, 2000, 6, 155-173.

26. C. E. Knapp and C. J. Carmalt, Chem. Soc. Rev., 2016, 45, 1036-1064.

27. P. Marchand, I. A. Hassan, I. P. Parkin and C. J. Carmalt, Dalton Trans., 2013, 42, 9406-9422.

28. P. G. Gordon, A. Kurek and S. T. Barry, ECS J. Solid State Sci. Technol., 2015, 4, N3188-N3197.

29. S. Mathur, A. P. Singh, R. Mueller, T. Leuning, T. Lehnen and H. Shen, Metal-organic chemical vapor deposition of metal oxide films and nanostructures, 2012.

30. S. R. Droes, T. T. Kodas and M. J. Hampden-Smith, in Carbide, Nitride and Boride Materials Synthesis and Processing, ed. A. W. Weimer, Springer Netherlands, Dordrecht, 1997, 579-603.

31. K. R. Johnson, P. Arevalo Rodriguez, C. R. Brewer, J. A. Brannaka, Z. Shi, J. Yang, B. Salazar, L. McElwee-White and A. V. Walker, J. Chem. Phys., 2017, 146, 052816/052811-052816/052819.

32. C. W. Dunnill, Z. A. Aiken, J. Pratten, M. Wilson and I. P. Parkin, Chem. Vap. Deposition, 2010, 16, 50-54.

33. J. L. Vossen, W. Kern and Editors, Thin Film Processes II, Academic, 1991.

168

34. S. M. Gates, Chem. Rev., 1996, 96, 1519-1532.

35. H. Pedersen, S. Leone, O. Kordina, A. Henry, S. Nishizawa, Y. Koshka and E. Janzen, Chem. Rev., 2012, 112, 2434-2453.

36. K. J. Huettinger, Chem. Vap. Deposition, 1998, 4, 151-158.

37. P. S. Patil, Mater, Chem. Phys., 1999, 59, 185-198.

38. N. Al-Dulaimi, E. A. Lewis, N. Savjani, P. D. McNaughter, S. J. Haigh, M. A. Malik, D. J. Lewis and P. O'Brien, Journal of Materials Chemistry C, 2017, 5, 9044-9052.

39. S. Khalid, M. A. Malik, D. J. Lewis, P. Kevin, E. Ahmed, Y. Khan and P. O'Brien, J. Mater. Chem. C, 2015, 3, 12068-12076.

40. J.-Y. Bae, J. Park, H. Y. Kim, H.-S. Kim and J.-S. Park, ACS Appl. Mater. Interfaces, 2015, 7, 12074-12079.

41. X. Zhu, T. Kawaharamura, A. Z. Stieg, C. Biswas, L. Li, Z. Ma, M. A. Zurbuchen, Q. Pei and K. L. Wang, Nano Lett., 2015, 15, 4948-4954.

42. R. W. Schwartz, Chem. Mater., 1997, 9, 2325-2340.

43. L. Fei, M. Naeemi, G. Zou and H. Luo, Chem. Rec., 2013, 13, 85-101.

44. J. Qiu, Y. C. Wu, Y. C. Wang, M. H. Engelhard, L. McElwee-White and W. D. Wei, Journal of the American Chemical Society, 2013, 135, 38-41.

45. G. V. Hartland, Chem. Rev. (Washington, DC, U. S.), 2011, 111, 3858- 3887.

46. X. Chen, Y. Chen, M. Yan and M. Qiu, ACS Nano, 2012, 6, 2550-2557.

47. C. Wang and S. Erhan, J. Am. Oil Chem. Soc., 1999, 76, 1211-1216.

48. J. Qiu and W. D. Wei, The Journal of Physical Chemistry C, 2014, 118, 20735-20749.

49. G. V. Hartland, Chem. Rev., 2011, 111, 3858-3887.

50. J. Qiu, Y.-C. Wu, Y.-C. Wang, M. H. Engelhard, L. McElwee-White and W. D. Wei, J. Am. Chem. Soc., 2013, 135, 38-41.

51. G. Chen and P. Hui, Appl. Phys. Lett., 1999, 74, 2942-2944.

52. G. Li, X. Chen, L. Huang, J. Wang, W. Hu and W. Lu, Phys. B (Amsterdam, Neth.), 2012, 407, 2223-2228.

169

53. E. Bailo and V. Deckert, Chem. Soc. Rev., 2008, 37, 921-930.

54. C. L. Muhich, J. Qiu, A. M. Holder, Y.-C. Wu, A. W. Weimer, W. D. Wei, L. McElwee-White and C. B. Musgrave, ACS Appl. Mater. Interfaces, 2015, 7, 13384-13394.

55. M. M. B. Holl, P. F. Seidler, S. P. Kowalczyk and F. R. Mcfeely, Inorg. Chem., 1994, 33, 510-517.

56. C. L. Muhich, J. Qiu, A. M. Holder, Y.-C. Wu, A. W. Weimer, W. D. Wei, L. McElwee-White and C. B. Musgrave, ACS Appl. Mater. Interfaces, 2015, 7, 13384-13394.

57. J. Du, Z. Chen, S. Ye, B. J. Wiley and T. J. Meyer, Angew. Chem. Int. Ed., 2015, 54, 2073-2078.

58. J. Zhao, P. D. Tran, Y. Chen, J. S. C. Loo, J. Barber and Z. J. Xu, ACS Catal., 2015, 5, 4115-4120.

59. Y. Tan, J. Gu, W. Xu, Z. Chen, D. Liu, Q. Liu and D. Zhang, ACS Appl. Mater. Interfaces, 2013, 5, 9878-9882.

60. H.-G. Im, S.-H. Jung, J. Jin, D. Lee, J. Lee, D. Lee, J.-Y. Lee, I.-D. Kim and B.-S. Bae, ACS Nano, 2014, 8, 10973-10979.

61. J. Li, T. E. Seidel and J. W. Mayer, MRS Bull., 1994, 19, 15-18.

62. P. Doppelt, Coord. Chem. Rev., 1998, 178-180, 1785-1809.

63. S. Lee, J. Y. Kim, T.-W. Lee, W.-K. Kim, B.-S. Kim, J. H. Park, J.-S. Bae, Y. C. Cho, J. Kim, M.-W. Oh, C. S. Hwang and S.-Y. Jeong, Sci. Rep., 2014, 4, 6230.

64. C. Y. Mak, MRS Bulletin, 1994, 19, 55-62.

65. W. D. Bancroft, J. Phys. Chem., 1904, 9, 277-296.

66. P. Doppelt, Coord. Chem. Rev., 1998, 178-180, 1785-1809.

67. Z. Liu, Y. Yang, J. Liang, Z. Hu, S. Li, S. Peng and Y. Qian, J. Phys. Chem. B, 2003, 107, 12658-12661.

68. Y. Zhao, J.-J. Zhu, J.-M. Hong, N. Bian and H.-Y. Chen, Eur. J. Inorg. Chem., 2004, 2004, 4072-4080.

69. M. Epifani, G. De, A. Licciulli and L. Vasanelli, J. Mater. Chem., 2001, 11, 3326-3332.

170

70. M. J. Hampden-Smith and T. T. Kodas, in The Chemistry of Metal CVD, Wiley-VCH Verlag GmbH, 2007, 239-302.

71. G. L. Griffin and A. W. Maverick, in The Chemistry of Metal CVD, Wiley-VCH Verlag GmbH, 2007, 175-238.

72. P. Doppelt and T. H. Baum, MRS Bulletin, 1994, 19, 41-48.

73. V. N. Vertoprakhov and S. A. Krupoder, Russ. Chem. Rev., 2000, 69, 1057-1082.

74. J. A. Bertrand and R. I. Kaplan, Inorg. Chem., 1966, 5, 489-491.

75. D.-H. Kim, R. H. Wentorf and W. N. Gill, J. Electrochem. Soc., 1993, 140, 3267-3272.

76. D.-H. Kim, R. H. Wentorf and W. N. Gill, J. Electrochem. Soc., 1993, 140, 3273-3279.

77. W. G. Lai, Y. Xie and G. L. Griffin, J. Electrochem. Soc., 1991, 138, 3499-3504.

78. D. Temple and A. Reisman, J. Electrochem. Soc., 1989, 136, 3525-3529.

79. K.-H. Park and W. J. Marshall, J. Am. Chem. Soc., 2005, 127, 9330-9331.

80. M.-J. Mouche, J.-L. Mermet, M. Romand and M. Charbonnier, Thin Solid Films, 1995, 262, 1-6.

81. E. Lay, Y.-H. Song, Y.-C. Chiu, Y.-M. Lin, Y. Chi, A. J. Carty, S.-M. Peng and G.-H. Lee, Inorg. Chem., 2005, 44, 7226-7233.

82. M. B. Naik, S. K. Lakshmanan, R. H. Wentorf, R. R. Reeves and W. N. Gill, J. Cryst. Growth, 1998, 193, 133-147.

83. K. M. Chi, H. K. Shin, M. J. Hampden-Smith, E. N. Duesler and T. T. Kodas, Polyhedron, 1991, 10, 2293-2299.

84. T. H. Baum and C. E. Larson, J. Electrochem. Soc., 1993, 140, 154-159.

85. H. K. Shin, M. J. Hampden-Smith, E. N. Duesler and T. T. Kodas, Can. J. Chem., 1992, 70, 2954-2966.

86. T. Kruck and C. Terfloth, Chem. Ber., 1993, 126, 1101-1106.

87. P. D. Tran, A. Allavena-Valette, F. Kamous and P. Doppelt, Polyhedron, 2009, 28, 4091-4095.

88. P. D. Tran and P. Doppelt, Surf. Coat. Tech., 2007, 201, 9066-9070.

171

89. M. Joulaud, C. Angekort, P. Doppelt, T. Mourier and D. Mayer, Microelectron. Eng., 2002, 64, 107-115.

90. D. S. Laitar, C. J. N. Mathison, W. M. Davis and J. P. Sadighi, Inorg. Chem., 2003, 42, 7354-7356.

91. F. A. Cotton and T. J. Marks, J. Am. Chem. Soc., 1970, 92, 5114-5117.

92. K. M. Chi, H. K. Shin, M. J. Hampden-Smith, T. T. Kodas and E. N. Duesler, Inorg. Chem., 1991, 30, 4293-4294.

93. T. H. Baum and C. E. Larson, Chem. Mater., 1992, 4, 365-369.

94. R. Nast, R. Mohr and C. Schultze, Chem. Ber., 1963, 96, 2127-2131.

95. J. A. T. Norman, B. A. Muratore, P. N. Dyer, D. A. Roberts and A. K. Hochberg, J. Phys. IV, 1991, 1, C2/271-C272/278.

96. E. S. Choi, S. K. Park, H. K. Shin and H. H. Lee, Appl. Phys. Lett., 1996, 68, 1017-1019.

97. K.-K. Choi and S.-W. Rhee, Thin Solid Films, 2002, 409, 147-152.

98. G. S. Girolami, P. M. Jeffries and L. H. Dubois, J. Am. Chem. Soc., 1993, 115, 1015-1024.

99. A. Jain, K. M. Chi, T. T. Kodas and M. J. Hampdensmith, J. Electrochem. Soc., 1993, 140, 1434-1439.

100. G. Buono-Core, K. Iwai, Y. L. Chow, T. Koyanagi, A. Kaji and J. Hayami, Can. J. Chem., 1979, 57, 8-16.

101. T. Y. Chen, J. Vaissermann, E. Ruiz, J. P. Sénateur and P. Doppelt, Chem. Mater., 2001, 13, 3993-4004.

102. H. Hiura and H. Takahashi, J. Phys. Chem., 1992, 96, 8909-8915.

103. J. Qiu, N. E. Richey, J. S. DuChene, Y. Zhai, Y. Zhang, L. McElwee- White and W. D. Wei, The Journal of Physical Chemistry C, 2016, 120, 20775- 20780.

104. M. Besora and A. Lledos, Struct. Bonding, 2008, 130, 149-202.

105. T. J. Marks and J. R. Kolb, Chem. Rev., 1977, 77, 263-293.

106. T. N. Sorrell and P. S. Pearlman, J. Org. Chem., 1980, 45, 3449-3451.

107. A. Zuttel, P. Wenger, S. Rentsch, P. Sudan, P. Mauron and C. Emmenegger, J. Power Sources, 2003, 118, 1-7.

172

108. J. L. Mallek, 2014, Volatile metal borohydride complexes: Synthesis and characterization of new chemical vapor deposition precursors, Chemistry, University of Illinois Urbana-Champaine.

109. Z. Lodziana, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 144108/144101-144108/144112.

110. T. J. Klingen, Inorg. Chem., 1964, 3, 1058-1059.

111. S. J. Lippard and P. S. Welcker, Inorg. Chem., 1972, 11, 6-11.

112. S. J. Lippard and K. M. Melmed, Inorg. Chem., 1969, 8, 2755-2762.

113. S. J. Lippard and D. A. Ucko, Inorg. Chem., 1968, 7, 1051-1056.

114. S. J. Lippard and K. M. Melmed, Inorg. Chem., 1967, 6, 2223-2228.

115. D. A. Barda, in Encyclopedia of Reagents for Organic Synthesis, John Wiley & Sons, Ltd, 2001.

116. J. M. Davidson, Chem. Ind., 1964, 2021-2022.

117. H. J. Gysling, Inorg. Synth., 1979, 19, 92-97.

118. F. A. Cotton and D. M. L. Goodgame, J. Chem. Soc., 1960, 5267-5269.

119. G. G. Messmer and G. J. Palenik, Inorg. Chem., 1969, 8, 2750-2754.

120. C. C. Addison, N. Logan, S. C. Wallwork and C. D. Garner, Q. Rev., Chem. Soc., 1971, 25, 289-322.

121. C. C. Addison and B. M. Gatehouse, J. Chem. Soc., 1960, 613-616.

122. K. M. Melmed, T.-I. Li, J. J. Mayerle and S. J. Lippard, J. Amer. Chem. Soc., 1974, 96, 69-75.

123. K. F. Purcell and R. S. Drago, J. Am. Chem. Soc., 1966, 88, 919-924.

124. A. J. Carty, W. H. Anderson, G. J. Palenik and G. Schreiber, Can. J. Chem., 1971, 49, 761-766.

125. J. C. Bommer and K. W. Morse, Inorg. Chem., 1980, 19, 587-593.

126. M. Mathew, G. J. Palenik and A. J. Carty, Can. J. Chem., 1971, 49, 4119- 4121.

127. C. A. Tolman, Chem. Rev., 1977, 77, 313-348.

173

128. A. O. Govorov, W. Zhang, T. Skeini, H. Richardson, J. Lee and N. A. Kotov, Nanoscale Res. Lett., 2006, 1, 84-90.

129. M. J. Williamson, R. M. Tromp, P. M. Vereecken, R. Hull and F. M. Ross, Nat. Mater., 2003, 2, 532-536.

130. F. Bocquet, C. Maurel, J.-M. Roussel, M. Abel, M. Koudia and L. Porte, Phys. Rev. B, 2005, 71, 075405.

131. M. Nilsing, P. Persson and L. Ojamäe, Chem Phys Lett, 2005, 415, 375- 380.

132. A. Michota and J. Bukowska, J. Raman Spectrosc., 2003, 34, 21-25.

133. L. Jiang, T. You, P. Yin, Y. Shang, D. Zhang, L. Guo and S. Yang, Nanoscale, 2013, 5, 2784-2789.

134. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666-669.

135. A. K. Geim, Science, 2009, 324, 1530-1534.

136. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183-191.

137. X. Du, I. Skachko, A. Barker and E. Y. Andrei, Nat. Nanotechnol., 2008, 3, 491-495.

138. A. A. Balandin, Nat. Mater., 2011, 10, 569-581.

139. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385-388.

140. G. R. Bhimanapati, Z. Lin, V. Meunier, Y. Jung, J. Cha, S. Das, D. Xiao, Y. Son, M. S. Strano, V. R. Cooper, L. Liang, S. G. Louie, E. Ringe, W. Zhou, S. S. Kim, R. R. Naik, B. G. Sumpter, H. Terrones, F. Xia, Y. Wang, J. Zhu, D. Akinwande, N. Alem, J. A. Schuller, R. E. Schaak, M. Terrones and J. A. Robinson, ACS Nano, 2015, 9, 11509-11539.

141. M. Chhowalla, Z. Liu and H. Zhang, Chem. Soc. Rev., 2015, 44, 2584- 2586.

142. M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263-275.

143. S. Manzeli, D. Ovchinnikov, D. Pasquier, O. V. Yazyev and A. Kis, Nat. Rev. Mater., 2017, 2, 17033.

144. P. Miro, M. Audiffred and T. Heine, Chem. Soc. Rev., 2014, 43, 6537- 6554.

174

145. J. A. Wilson and A. D. Yoffe, Advan. Phys., 1969, 18, 193-335.

146. L. Rapoport, N. Fleischer and R. Tenne, J. Mater. Chem., 2005, 15, 1782- 1788.

147. M. S. Whittingham, Prog. Solid State Chem., 1978, 12, 41-99.

148. Y. Liu, M. Zhu and D. Chen, J. Mater. Chem. A, 2015, 3, 11857-11862.

149. P. T. Dirlam, J. Park, A. G. Simmonds, K. Domanik, C. B. Arrington, J. L. Schaefer, V. P. Oleshko, T. S. Kleine, K. Char, R. S. Glass, C. L. Soles, C. Kim, N. Pinna, Y.-E. Sung and J. Pyun, ACS Appl. Mater. Interfaces, 2016, 8, 13437-13448.

150. R. J. Toh, Z. Sofer, J. Luxa, D. Sedmidubsky and M. Pumera, Chem. Commun., 2017, 53, 3054-3057.

151. X. J. Chua, J. Luxa, A. Y. S. Eng, S. M. Tan, Z. Sofer and M. Pumera, ACS Catal., 2016, 6, 5724-5734.

152. Z. Gholamvand, D. McAteer, A. Harvey, C. Backes and J. N. Coleman, Chem. Mater., 2016, 28, 2641-2651.

153. Y. Sun, S. Gao, F. Lei and Y. Xie, Chem. Soc. Rev., 2015, 44, 623-636.

154. H. Schmidt, F. Giustiniano and G. Eda, Chem. Soc. Rev., 2015, 44, 7715- 7736.

155. Y. Chen, C. Tan, H. Zhang and L. Wang, Chem. Soc. Rev., 2015, 44, 2681-2701.

156. A. Kuc, N. Zibouche and T. Heine, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 245213/245211-245213/245214.

157. H. R. Gutiérrez, N. Perea-López, A. L. Elías, A. Berkdemir, B. Wang, R. Lv, F. López-Urías, V. H. Crespi, H. Terrones and M. Terrones, Nano Lett., 2013, 13, 3447-3454.

158. O. Lopez-Sanchez, E. Alarcon Llado, V. Koman, A. Fontcuberta i. Morral, A. Radenovic and A. Kis, ACS Nano, 2014, 8, 3042-3048.

159. K. F. Mak and J. Shan, Nat. Photonics, 2016, 10, 216-226.

160. A. A. Al-Hilli and B. L. Evans, J. Cryst. Growth, 1972, 15, 93-101.

161. J. Mercier, J. Cryst. Growth, 1982, 56, 235-244.

162. H. Zhang and A. S. Hock, Inorg. Chem., 2017, 56, 106-109.

163. W. K. Hofmann, J. Mater. Sci., 1988, 23, 3981-3986.

175

164. C. J. Carmalt, I. P. Parkin and E. S. Peters, Polyhedron, 2003, 22, 1499- 1505.

165. J. Cheon, J. E. Gozum and G. S. Girolami, Chem. Mater., 1997, 9, 1847- 1853.

166. J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. De, R. J. Smith, I. V. Shvets, S. K. Arora, G. Stanton, H.-Y. Kim, K. Lee, G. T. Kim, G. S. Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan, J. C. Grunlan, G. Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins, E. M. Grieveson, K. Theuwissen, D. W. McComb, P. D. Nellist and V. Nicolosi, Science, 2011, 331, 568-571.

167. K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V. Khotkevich, S. V. Morozov and A. K. Geim, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10451- 10453.

168. Y.-H. Lee, X.-Q. Zhang, W. Zhang, M.-T. Chang, C.-T. Lin, K.-D. Chang, Y.-C. Yu, J. T.-W. Wang, C.-S. Chang, L.-J. Li and T.-W. Lin, Adv. Mater., 2012, 24, 2320-2325.

169. Y. Shi, H. Li and L.-J. Li, Chem. Soc. Rev., 2015, 44, 2744-2756.

170. S. M. Eichfeld, L. Hossain, Y.-C. Lin, A. F. Piasecki, B. Kupp, A. G. Birdwell, R. A. Burke, N. Lu, X. Peng, J. Li, A. Azcatl, S. McDonnell, R. M. Wallace, M. J. Kim, T. S. Mayer, J. M. Redwing and J. A. Robinson, ACS Nano, 2015, 9, 2080-2087.

171. K. C. Kwon, C. Kim, Q. V. Le, S. Gim, J.-M. Jeon, J. Y. Ham, J.-L. Lee, H. W. Jang and S. Y. Kim, ACS Nano, 2015, 9, 4146-4155.

172. A. L. Hector, W. Levason, G. Reid, S. D. Reid and M. Webster, Chem. Mater., 2008, 20, 5100-5106.

173. S. L. Benjamin, Y.-P. Chang, C. Gurnani, A. L. Hector, M. Huggon, W. Levason and G. Reid, Dalton Trans., 2014, 43, 16640-16648.

174. S. D. Reid, A. L. Hector, W. Levason, G. Reid, B. J. Waller and M. Webster, Dalton Trans., 2007, DOI: 10.1039/b708809d, 4769-4777.

175. A. L. Hector, M. Jura, W. Levason, S. D. Reid and G. Reid, New J. Chem., 2009, 33, 641-645.

176. T. Ouyang, K. P. Loh, H. Zhang, J. J. Vittal, M. Vetrichelvan, W. Chen, X. Gao and A. T. S. Wee, J. Phys. Chem. B, 2004, 108, 17537-17545.

177. A. Adeogun, M. Afzaal and P. O'Brien, Chem. Vap. Deposition, 2006, 12, 597-599.

176

178. N. E. Richey, C. Haines, J. L. Tami and L. McElwee-White, Chem. Commun., 2017, 53, 7728-7731.

179. H. Kim, R. O. Bonsu, D. C. Bock, N. C. Ou, R. Y. Korotkov, L. McElwee-White and T. J. Anderson, ECS J. Solid State Sci. Technol., 2016, 5, Q3095-Q3105.

180. R. O. Bonsu, H. Kim, C. O'Donohue, R. Y. Korotkov, K. A. Abboud, T. J. Anderson and L. McElwee-White, Inorg. Chem., 2015, 54, 7536-7547.

181. A. N. MacInnes, M. B. Power and A. R. Barron, Chem. Mater., 1992, 4, 11-14.

182. S.-G. Shyu, J.-S. Wu, C.-C. Wu, S.-H. Chuang and K.-M. Chi, Inorg. Chem. Acta, 2002, 334, 276-282.

183. D. A. Brown, W. K. Glass, H. J. Toma and W. E. Waghorne, J. Chem. Soc., Dalton Trans., 1987, 2531-2536.

184. J. N. Smith and T. M. Brown, Inorg. Nucl. Chem. Lett., 1970, 6, 441-443.

185. W. H. Pan, T. R. Halbert, L. L. Hutchings and E. I. Stiefel, J. Chem. Soc., Chem. Commun., 1985, 927-929.

186. M. Chunggaze, M. Azad Malik and P. O'Brien, J. Mater. Chem., 1999, 9, 2433-2437.

187. S. A. Cohen and E. I. Stiefel, Inorg. Chem., 1985, 24, 4657-4662.

188. J. Qian, Z. Peng, P. Wang and X. Fu, ACS Appl. Mater. Interfaces, 2016, 8, 16876-16884.

189. T. A. J. Loh and D. H. C. Chua, J. Phys. Chem. C, 2015, 119, 27496- 27504.

190. E. I. Stiefel and K. D. Karlin, Dithiolene Chemistry: Synthesis, Properties, and Applications, Wiley, 2004.

191. R. Eisenberg and H. B. Gray, Inorg. Chem., 2011, 50, 9741-9751.

192. R. Kato, Chem. Rev., 2004, 104, 5319-5346.

193. A. J. Clough, J. M. Skelton, C. A. Downes, A. A. de la Rosa, J. W. Yoo, A. Walsh, B. C. Melot and S. C. Marinescu, J. Am. Chem. Soc., 2017, 139, 10863- 10867.

194. M. K. Chan, S. Mukund, A. Kletzin, M. W. W. Adams and D. C. Rees, Science (Washington, D. C.), 1995, 267, 1463-1469.

177

195. K. V. Rajagopalan and J. L. Johnson, J. Biol. Chem., 1992, 267, 10199- 10202.

196. K.-M. Sung and R. H. Holm, Inorg. Chem., 2000, 39, 1275-1281.

197. C. A. Goddard and R. H. Holm, Inorg. Chem., 1999, 38, 5389-5398.

198. P. Chandrasekaran and J. P. Donahue, Org. Synth., 2009, 86, 333-343.

199. P. Chandrasekaran, K. Arumugam, U. Jayarathne, L. M. Perez, J. T. Mague and J. P. Donahue, Inorg. Chem., 2009, 48, 2103-2113.

200. A. C. Ghosh, J. K. Reinhardt, M. K. Kindermann and C. Schulzke, Chem. Commun. (Cambridge, U. K.), 2014, 50, 10102-10104.

201. G. N. Schrauzer and V. P. Mayweg, J. Am. Chem. Soc., 1965, 87, 1483- 1489.

202. P. Chandrasekaran, K. Arumugam, U. Jayarathne, L. M. Pérez, J. T. Mague and J. P. Donahue, Inorg. Chem., 2009, 48, 2103-2113.

203. Y. Rong, Y. Fan, A. L. Koh, A. W. Robertson, K. He, S. Wang, H. Tan, R. Sinclair and J. H. Warner, Nanoscale, 2014, 6, 12096-12103.

204. S. Wang, Y. Rong, Y. Fan, M. Pacios, H. Bhaskaran, K. He and J. H. Warner, Chem. Mater., 2014, 26, 6371-6379.

205. Y. Rong, Y. Fan, A. Leen Koh, A. W. Robertson, K. He, S. Wang, H. Tan, R. Sinclair and J. H. Warner, Nanoscale, 2014, 6, 12096-12103.

206. K. Zhang, B. Jariwala, J. Li, N. C. Briggs, B. Wang, D. Ruzmetov, R. A. Burke, J. O. Lerach, T. G. Ivanov, M. Haque, R. M. Feenstra and J. A. Robinson, Nanoscale, 2018, 10, 336-341.

207. Z.-Q. Xu, Y. Zhang, S. Lin, C. Zheng, Y. L. Zhong, X. Xia, Z. Li, P. J. Sophia, M. S. Fuhrer, Y.-B. Cheng and Q. Bao, ACS Nano, 2015, 9, 6178-6187.

208. H. Jeong, S. Bang, H. M. Oh, H. J. Jeong, S.-J. An, G. H. Han, H. Kim, K. K. Kim, J. C. Park, Y. H. Lee, G. Lerondel and M. S. Jeong, ACS Nano, 2015, 9, 10032-10038.

209. D. Cao, T. Shen, P. Liang, X. Chen and H. Shu, The Journal of Physical Chemistry C, 2015, 119, 4294-4301.

210. Y. Sheng, H. Tan, X. Wang and J. H. Warner, Chem. Mater., 2017, 29, 4904-4911.

211. Z. Ali, N. E. Richey, D. C. Bock, K. A. Abboud, J. Akhtar, M. Sher and L. McElwee-White, Dalton Trans., 2018, 47, 2719-2726.

178

212. M. M. Nolan, A. J. Touchton, N. E. Richey, I. Ghiviriga, J. R. Rocca, K. A. Abboud and L. McElwee-White, Eur. J. Inorg. Chem., 2018, 2018, 46-53.

213. Z. Ali, N. E. Richey, D. C. Bock, K. A. Abboud, J. Akhtar, M. Sher and L. McElwee-White, Dalton Trans., 2018, 47, 2719-2726.

214. M.-R. Gao, Y.-F. Xu, J. Jiang and S.-H. Yu, Chem. Soc. Rev., 2013, 42, 2986-3017.

215. N. P. Dasgupta, X. Meng, J. W. Elam and A. B. F. Martinson, Acc. Chem. Res., 2015, 48, 341-348.

216. P. O'Brien and R. Nomura, J. Mater. Chem., 1995, 5, 1761-1773.

217. D. Merki, S. Fierro, H. Vrubel and X. Hu, Chem. Sci., 2011, 2, 1262-1267.

218. R. Jin, Q. Zhai and Q. Wang, Chem. Eur. J., 2017, 23, 14056-14063.

219. M. A. Malik, M. Afzaal and P. O'Brien, Chem. Rev., 2010, 110, 4417- 4446.

220. X. Hou and K. L. Choy, Chem. Vap. Deposition, 2006, 12, 583-596.

221. L. McElwee-White, Dalton Trans., 2006, DOI: 10.1039/b611848h, 5327- 5333.

222. C. J. Carmalt, S. A. O'Neill, I. P. Parkin and E. S. Peters, J. Mater. Chem., 2004, 14, 830-834.

223. S. Schneider, Y. Yang and T. J. Marks, Chem. Mater., 2005, 17, 4286- 4288.

224. S. Mlowe, D. J. Lewis, M. A. Malik, J. Raftery, E. B. Mubofu, P. O'Brien and N. Revaprasadu, Dalton Trans., 2016, 45, 2647-2655.

225. P. Kevin, D. J. Lewis, J. Raftery, M. Azad Malik and P. O'Brien, J. Cryst. Growth, 2015, 415, 93-99.

226. M. Akhtar, N. Revaprasadu, M. A. Malik and J. Raftery, Mater. Sci. Semicond. Process., 2015, 30, 368-375.

227. M. A. Buckingham, A. L. Catherall, M. S. Hill, A. L. Johnson and J. D. Parish, Cryst. Growth Des., 2017, 17, 907-912.

228. N. Alam, M. S. Hill, G. Kociok-Kohn, M. Zeller, M. Mazhar and K. C. Molloy, Chem. Mater., 2008, 20, 6157-6162.

229. E. R. T. Tiekink and I. Haiduc, in Prog. Inorg. Chem., John Wiley & Sons, Inc., 2005, 127-319.

179

230. K. Ramasamy, M. A. Malik, M. Helliwell, J. Raftery and P. O'Brien, Chem. Mater., 2011, 23, 1471-1481.

231. K. Ramasamy, M. A. Malik, P. O'Brien and J. Raftery, Dalton Trans., 2010, 39, 1460-1463.

232. L. P. Mgabi, B. S. Dladla, M. A. Malik, S. S. Garje, J. Akhtar and N. Revaprasadu, Thin Solid Films, 2014, 564, 51-57.

233. J. B. Biswal, S. S. Garje, J. Nuwad and C. G. S. Pillai, J. Solid State Chem., 2013, 204, 348-355.

234. N. H. Huy and U. Abram, Inorg. Chem., 2007, 46, 5310-5319.

235. N. Selvakumaran, S. W. Ng, E. R. T. Tiekink and R. Karvembu, Inorg. Chim. Acta, 2011, 376, 278-284.

236. F. M. Emen and N. Kuelcue, J. Therm. Anal. Calorim., 2012, 109, 1321- 1331.

237. N. Ozpozan, H. Arslan, T. Ozpozan, M. Merdivan and N. Kulcu, J. Therm. Anal. Calorim., 2000, 61, 955-965.

238. S. Saeed, K. S. Ahmed, N. Rashid, M. A. Malik, P. O’Brien, M. Akhtar, R. Hussain and W.-T. Wong, Polyhedron, 2015, 85, 267-274.

239. A. R. Katritzky, N. Kirichenko, B. V. Rogovoy, J. Kister and H. Tao, Synthesis, 2004, 1799-1805.

240. A. Mohamadou, I. Déchamps-Olivier and J.-P. Barbier, Polyhedron, 1994, 13, 1363-1370.

241. Y. S. Won, Y. S. Kim, T. J. Anderson, L. L. Reitfort, I. Ghiviriga and L. McElwee-White, J. Am. Chem. Soc., 2006, 128, 13781-13788.

242. Y. S. Won, Y. S. Kim, T. J. Anderson and L. McElwee-White, Chem. Mater., 2008, 20, 7246-7251.

243. B. H. Lee and K. Yong, J. Vac. Sci. Technol., B, 2004, 22, 2375-2379.

244. H. M. Ajmera, A. T. Heitsch, T. J. Anderson, C. B. Wilder, L. L. Reitfort, L. McElwee-White and D. P. Norton, J. Vac. Sci. Technol. B, 2008, 26, 1800-1807.

245. C. B. Wilder, L. L. Reitfort, K. A. Abboud and L. McElwee-White, Inorg. Chem., 2006, 45, 263-268.

246. O. J. Bchir, K. M. Green, H. M. Ajmera, E. A. Zapp, T. J. Anderson, B. C. Brooks, L. L. Reitfort, D. H. Powell, K. A. Abboud and L. McElwee-White, J. Am. Chem. Soc., 2005, 127, 7825-7833.

180

247. K. R. McClain, C. O’Donohue, A. Koley, R. O. Bonsu, K. A. Abboud, J. C. Revelli, T. J. Anderson and L. McElwee-White, J. Am. Chem. Soc., 2014, 136, 1650-1662.

248. A. Koley, C. T. O’Donohue, M. M. Nolan, K. R. McClain, R. O. Bonsu, R. Y. Korotkov, T. Anderson and L. McElwee-White, Chem. Mater., 2015, 27, 8326-8336.

249. J. P. Coyle, W. H. Monillas, G. P. A. Yap and S. T. Barry, Inorg. Chem., 2008, 47, 683-689.

250. S. E. Potts, C. J. Carmalt, C. S. Blackman, F. Abou-Chahine, D. Pugh and H. O. Davies, Organometallics, 2009, 28, 1838-1844.

251. S. T. Barry, Coord. Chem. Rev., 2013, 257, 3192-3201.

252. C. J. Carmalt, A. C. Newport, S. A. O'Neill, I. P. Parkin, A. J. P. White and D. J. Williams, Inorg. Chem., 2005, 44, 615-619.

253. A. Baunemann, D. Bekermann, T. B. Thiede, H. Parala, M. Winter, C. Gemel and R. A. Fischer, Dalton Trans., 2008, 3715-3722.

254. C. T. O’Donohue, K. R. McClain, A. Koley, J. C. Revelli, L. McElwee- White and T. J. Anderson, ECS J. Solid State Sci. Technol., 2015, 4, N3180-N3187.

255. O. J. Bchir, S. W. Johnston, A. C. Cuadra, T. J. Anderson, C. G. Ortiz, B. C. Brooks, D. H. Powell and L. McElwee-White, J. Cryst. Growth, 2003, 249, 262- 274.

256. NIST, X-ray Photoelectron Spectroscopy Database, Version 4.1.

257. M. J. Powell, D. B. Potter, R. L. Wilson, J. A. Darr, I. P. Parkin and C. J. Carmalt, Materials & Design, 2017, 129, 116-124.

258. H. Kim, R. O. Bonsu, C. O’Donohue, R. Y. Korotkov, L. McElwee-White and T. J. Anderson, ACS Appl. Mater. Interfaces, 2015, 7, 2660-2667.

259. R. O. Bonsu, H. Kim, C. O'Donohue, R. Y. Korotkov, K. R. McClain, K. A. Abboud, A. A. Ellsworth, A. V. Walker, T. J. Anderson and L. McElwee-White, Dalton Trans., 2014, 43, 9226-9233.

260. D. C. Bock, N. C. Ou, R. O. Bonsu, C. T. Anghel, X. Su and L. McElwee- White, Solid State Ionics, 2018, 315, 77-84.

261. R. O. Bonsu, D. C. Bock, H. Kim, R. Y. Korotkov, K. A. Abboud, T. J. Anderson and L. McElwee-White, Dalton Trans., 2016, 45, 10897-10908.

262. R. J. Colton and J. W. Rabalais, Inorg. Chem., 1976, 15, 236-238.

181

263. D. Barreca, G. Carta, A. Gasparotto, G. Rossetto, E. Tondello and P. Zanella, Surf. Sci. Spectra, 2003, 8, 258-267.

264. R. J. Colton, A. M. Guzman and J. W. Rabalais, J. Appl. Phys., 1978, 49, 409-416.

265. P. G. Egan, D. G. Holan and K. W. Morse, Inorg. Chem., 1984, 23, 2203- 2204.

266. M. Grace, H. Beall and C. H. Bushweller, J. Chem. Soc. D, 1970, 701.

267. G. N. Schrauzer and V. P. Mayweg, J. Am. Chem. Soc., 1965, 87, 1483- 1489.

268. D. P. Tate, W. R. Knipple and J. M. Augl, Inorg. Chem., 1962, 1, 433-434.

269. I. B. Douglass and F. B. Dains, J. Am. Chem. Soc., 1934, 56, 719-721.

270. H. Hartmann, L. Beyer and E. Hoyer, J. Prakt. Chem., 1978, 320, 647- 650.

271. N. Gunasekaran, P. Ramesh, M. N. G. Ponnuswamy and R. Karvembu, Dalton Trans., 2011, 40, 12519-12526.

272. N. Selvakumaran, A. Pratheepkumar, S. W. Ng, E. R. T. Tiekink and R. Karvembu, Inorg. Chim. Acta, 2013, 404, 82-87.

273. F.Y. Xie, L. Gong, X. Liu, Y.T. Tao, W.H. Zhang, S.H. Chen, H. Meng, J. Chen, J. Electron. Spectrosc. Relat. Phenom., 2012, 185, 112-118.

182

BIOGRAPHICAL SKETCH

Nathaniel was born in Louisville, Kentucky, but didnʹt stay long enough to remember it.

Before he was one year old he had move to Tucson, Arizona and would stay there until moving to Florida for graduate school in 2013. He attended Catalina Magnet High School for their flight school magnet program, in the hopes of one day becoming a pilot. From freshmen to junior year he studied various aspect of flight, in his senior year he was chosen as one of ten students to be able to start flying and obtain a private pilot’s license. While interesting in its own regards, he soon learned that flying was not as intellectually stimulating as he had hoped. Fortunately, in his senior year he was also taking his first chemistry course and fell in love with the subject and continued to pursue it through at the University of Arizona.

He began attending the University of Arizona in the fall of 2009 and spent the first two years going through his basic course. In the fall of 2011 he joined the research group of Jeffrey

Pyun and started to work under the guidance of Larry Hill on colloidal nanoparticle polymers.

This research sparked interest in inorganic and materials chemistry, which lead him to pursue a graduate degree in inorganic chemistry at the University of Florida.

In the summer of 2013, he moved across the country to join the McElwee-White research group, prior to the beginning of the fall semester. During his time in the McElwee-White group he worked on the synthesis of precursors for various chemical deposition techniques and was one of the first two group members to begin performing in-house depositions. His favorite time in graduate school has been mentoring younger graduate students, undergraduates, and exchange students inside the research group.

183