Low-Temperature Solution-Phase Synthesis of Chalcogenide and Carbide Materials

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Rick Albert Lionel Morasse

Graduate Program in Chemistry

The Ohio State University

2018

Dissertation Committee:

Dr. Joshua E. Goldberger, Advisor

Dr. L. Robert Baker

Dr. James V. Coe Copyrighted by

Rick Albert Lionel Morasse

2018

Abstract

One-dimensional TiS2 phases tend to form at high ligand-to-metal ratios and relatively lower temperatures, while the parent two-dimensional lattices form at higher temperature.

The organic ligand structure dictates the temperature window at which a dimensionally reduced phase can be accessed. Although a small change in ligand structure, such as from ethylenediamine, en, to propylenediamine, pn, will significantly influence the stability of these phases, it will only subtly change the electronic structure. By developing a systematic understanding of the effects of various factors during the synthesis, we provide a pathway to rationally create new dimensionally reduced materials.

A second project in this dissertation focuses on the development of a solution-phase route towards germanium carbide materials. We hypothesized metal germanium carbide materials could be created via the transmetallation of precursors that

ii contain four C-B(OR)2 bonds with germanium-halogen (Ge-X) bond-containing precursors to form networks containing Ge-C bonds. While this chemistry is an essential step in many well-known organic reaction pathways, it has not been explored for the synthesis of germanium carbide materials in part due to the lack of commercial availability and the complicated synthesis of the C[B(OR)2]4 precursors. To these ends, we established the synthesis of the tetrasubstituted cyclic boronic ester, tetrakis (1,3 propanediolatoboronate) methane, which we denote as C(Bpg)4. We have adapted and improved upon the previously reported route, which utilizes precursors that are not commercially available, and have been able to synthesize this material on the 20 g scale with an overall ~50% yield.

Upon establishing the large scale synthesis of C(Bpg)4 precursors, we explored whether GeC can be created. GeC has attracted considerable theoretical interest, yet no such phase currently exists. In these experiments we describe the reactions between

C(Bpg)4 with many different germanium precursors across multiple different solvent and temperature conditions. These reactions and subsequent characterizations reveal that an analogous precursor, HC(Bpg)3, generates an amorphous GeCH phase while the C(Bpg)4 precursor does not immediately react and rather forms an oxidized Ge phase. The X-ray diffraction (XRD) of these materials showed no long-range crystallinity, and synchrotron

X-ray Pair-distribution function (PDF) identified the presence of Ge-C bonds at 1.86 Å in the amorphous GeCH phase in addition to Ge-O bonds at 1.74 Å in the GeO2 phase. The

Raman analysis showed no crystalline modes in the GeCH phase until annealing above

500 oC, at which point graphite and crystalline germanium modes appear. The lack of

iii direct reactivity between C(Bpg)4 with tetrasubstituted Ge precursors merits the future exploration of base-activation procedures.

In summary, the low-temperature solution-phase syntheses of TiS2(en) and

C(Bpg)4 and the progress towards carbides and small-molecule carbide analogues represent progress towards the wider use of solution-phase synthetic methods in order to generate advanced materials.

This document is dedicated to my family.

Acknowledgments

I would sincerely like to thank Dr. Joshua Goldberger for being my advisor during my graduate studies. I am thankful for the countless time he has invested into my professional career through exam, presentation, and dissertation preparation. I am grateful for this opportunity to perform exciting research, and am especially thankful for all the growth and learning that have come from addressing these challenging scientific endeavors.

I would also like to thank all of my examination committee members over the years. I am thankful for your dedication to my professional career and for your warm collegiality within the department. Thank you to Dr. Coe and Dr. Wu for serving on my first year oral exam, to Dr. Badu-Tawiah, Dr. Baker, and Dr. Coe for serving on my candidacy committee, and Dr. Baker and Dr. Coe for serving on my PhD committee.

Next, I would like to thank everyone in the OSU Department of Chemistry &

Biochemistry involved with General Chemistry. During my time as a GTA, I have grown personally and professionally thanks to the guidance of the Lab Supervisors, Dr. Tatz, Dr.

Zellmer, and Dr. Moga. Additionally, I am thankful for all of the support from Nate

Williams and Tyler Weaver over the years of teaching.

I am also grateful to the many people who have helped me with research, especially Dr. Ken Kuno, Dr. Jay Giblin, Dr. Ewan Hamilton, Dr. Judith Gallucci, Dr.

Tanya Whitmer, Dr. Lisa Alexander, and Dr. Nicole Karn.

I would also like to thank all the current and former members of the Goldberger

Group for their years of advice, support, and many laughs. I am especially grateful to Dr.

Tianyang Li and Zachary Baum for their many years of collaboration. They have helped me with countless experiments. Additionally, I would like to thank Dr. Maxx Arguilla and Nick Cultrara for timely help when running measurements. Most importantly, I would like to thank Maxx and Nick for the many culinary adventures.

I could not have completed all of these years of study without the love, support, prayer, and encouragement of my friends and family. I am thankful for the friendships formed at the University of Notre Dame and the St. Thomas More Newman Center at

OSU. Special thanks to Sean and Laura Puscas and Chris Schreyer. I would also like to thank my loving and patient girlfriend, Rachel, for all of her support during graduate school. Finally, thanks to my entire family, grandparents, parents, Pat and Paul, and brother, Mark.

vii

Vita

June 2008 ...... Lakeshore Catholic High School

May 2012 ...... B.S., Chemistry, University of Notre Dame

2012 to present ...... Graduate Teaching Associate, Department

of Chemistry & Biochemistry, The Ohio

State University

Publications

Morasse, R. A. L., Li, T., Baum, Z., Goldberger, J. E., “Rational Synthesis of

Dimensionally Reduced TiS2 Phases” Chem. Mater., 2014, 26, 4776–4780.

Fields of Study

Major Field: Chemistry

viii

Table of Contents

Abstract ...... ii

Acknowledgments...... vi

Vita ...... viii

Publications ...... viii

Fields of Study ...... viii

Table of Contents ...... ix

List of Tables ...... xii

List of Reactions ...... xiii

List of Figures ...... xiv

Chapter 1 Introduction ...... 1

1.1 Conventional Solid-State Synthesis Methods ...... 2

1.2 Solution-Phase Synthesis Methods ...... 4

1.3 Overview of Dimensional Reduction ...... 9

1.4 TiS2 and Derivatives...... 15

1.4.1 Structure and Applications of TiS2 ...... 15 ix

1.4.2 Dimensional Reduction of TiS2 ...... 15

1.4.3 Solution-Phase Synthesis of TiS2 Derivatives ...... 18

1.5 Carbide Materials ...... 19

1.5.1 Applications of Carbides ...... 19

1.5.2 Applications of Germanium Carbide ...... 24

1.5.3 Synthesis of Carbides ...... 25

1.5.4 Synthesis of Germanium Carbide ...... 28

1.6 Chapter Outlines ...... 33

Chapter 2 Dimensionally Reduced TiS2 Phases ...... 47

2.1 Introduction ...... 47

2.2 Experimental Section ...... 49

2.2.1 Preparation of TiS2(en) and TiS2(pn) ...... 49

2.2.2 Physical Characterizations ...... 52

2.2.3 Density Functional Theory Calculations ...... 52

2.3 Crystal Structures of TiS2(en) and TiS2(pn) ...... 53

2.4 Synthetic Phase Diagrams of the Titanium Sulfide Systems ...... 57

2.5 Electronic Band Structures of the 1-D TiS2 Analogues ...... 67

2.6 Conclusion ...... 73

Chapter 3 Synthesis of C(Bpg)4 and HC(Bpg)3 Precursors ...... 86

3.1 Introduction ...... 86

3.2 Improved C(Bpg)4 Synthesis Protocol ...... 92

3.3 Characterization of C(Bpg)4 and HC(Bpg)3 ...... 104

3.4 Results and Discussion ...... 105

3.5 Conclusion ...... 115

Chapter 4 Solution-Phase Synthesis towards GeC and GeCH ...... 128

4.1 Introduction ...... 128

4.2 Synthesis...... 133

4.3 Characterization ...... 139

4.4 Results and Discussion ...... 147

4.4.1 Physical Characterization of GeCH Product ...... 147

4.4.2 Physical Characterization of Oxidized Germanium Product ...... 161

4.5 Conclusion ...... 170

Chapter 5 Conclusions and Future Outlook ...... 184

References ...... 189

List of Tables

Table 1 Unit Cell Parameters of TiS2(en) and TiS2(pn)...... 56

Table 2 Representative Bond Lengths in TiS2(en) and TiS2(pn)...... 56

1 11 Table 3 Measured H and B NMR ppm shifts of C-B precursor molecules in CDCl3 solvent...... 106

Table 4 XPS atomic percentages of elements in scans of the amorphous GeCH product before and after Ar etching of the surface...... 158

Table 5 XPS atomic percentages of elements in scans of amorphous germanium oxygen product before and after Ar etching of the oxidized Ge product. This confirms the presence of the GeO2 product and adventitious armorphous carbon which is etched away from the surface...... 169

xii

List of Reactions

Reaction 1 TiCl4 and S powder in en to form TiS2(en) ...... 50

Reaction 2 Formation of dimethoxy boron chloride ...... 95

Reaction 3 Formation of tetrakis dimethoxy boryl methane ...... 95

Reaction 4 Formation of tetrakis(1,3 propanediolatoboronate) ...... 95

Reaction 5 Formation of dimethoxy boron chloride ...... 96

Reaction 6 Formation of tetrakis dimethoxy boryl methane ...... 96

Reaction 7 Formation of tetrakis(1,3 propanediolatoboronate) ...... 96

Reaction 8 Proposed reaction of Ge(NMe2)4 with C(Bpg)4 to produce GeC ...... 132

Reaction 9 Proposed reaction of Ge(NMe2)4 with HC(Bpg)3 to produce GeCH ...... 132

xiii

List of Figures

Figure 1 Solution-phase synthetic reaction apparatuses: stirring reagents in three-neck flasks at refluxing temperatures while under Ar on Schlenk line...... 7

Figure 2 Images showing the variety of nanomaterials easily accessible via solution- phase, multi-step injection synthetic methods. Left: straight and branched CdSe nanowires. Right: CdSe nanosheets...... 8

Figure 3 Principle of dimensional reduction illustrated on a framework of corner-sharing octahedral, M atoms black and X atoms white. Reaction of MX3 with AaX incorporates additional X atoms into the M-X framework, inserting into M-X-M bridges to reduce the connectedness and dimensionality of the framework. Reproduced from Tulsky, E. G.;

Long, J. R. Dimensional Reduction: A Practical Formalism for Manipulating Solid

Structures. Chem. Mater. 2001, 13, 1149...... 12

Figure 4 (a) Dimensionally reduced zigzag and linear 1-D structures derived from the

CdI2 structure type. (b) Left and middle: 1-D zigzag edge-sharing chain fragments highlighted in blue in the parent TiS2 lattice; right: 1-D zigzag chain of TiS2(en). Atoms

Ti, blue; S, yellow; C, black; N, green. Reproduced from Li, T.; Goldberger, J. E.

Atomic-Scale Derivatives of Solid-State Materials. Chem. Mater. 2015, 27 (10),

3549-3559...... 13

xiv

(up) and ZnS(ba) lattices (down) (ba = butylamine). Atoms: Zn, gray and red; S, yellow;

C, black; N, blue. Reproduced from Li, T.; Goldberger, J. E. Atomic-Scale Derivatives of Solid-State Materials. Chem. Mater. 2015, 27 (10), 3549–3559...... 14

Figure 6 (a) Crystal structures of TiS2(en) (left) and LiTiS2(en) (right) projected down the c axis from neutron diffraction refinements. Ti octahedra are shaded in red. (b, c) X- ray diffraction patterns of TiS2(en) (black), Li0.66TiS2(en) (red), and LiTiS2(en). (d)

Changes of unit cell parameter (a, black; c, red) with respect to Li/Ti ratio. (e) Crystal structures showing two adjacent 1-D chains along c axis with intercalated Li atoms positions determined from neutron diffraction refinements. Atom colors: Ti, red; S, yellow; N, green; C, black. H atoms have been omitted for clarity. Reproduced from Li,

T.; Liu, Y.-H.; Chitara, B.; Goldberger, J. E. Li Intercalation into 1D TiS2(en) Chains. J.

Am. Chem. Soc. 2014, 136, 2986...... 17

Figure 7 Silicon carbide is a refractory material that is cast as tools, dies, and parts for many fields. Reproduced from Saint Gobain Ceramics and Plastics, Inc...... 22

Figure 8 Catalytic reactions –including hydrogenolysis, hydrogenation, isomerization, and dehydration – that are catalyzed by carbidic and oxycarbidic molybdenum carbide formulations. Transition metals such as molybdenum and tungsten alter their structure by lattice expansion and concurrent incorporation of heteroatoms (C, O, and N) into interstitial sites, typically the octahedral sites of a face-centered cubic (fcc) or hexagonal close-packed

(hcp) lattice. Reproduced from Sullivan, M. M.; Chen, C.-J.; Bhan, A. Catalytic

Deoxygenation on Transition Metal Carbide Catalysts. Catal. Sci. Technol. 2016, 6, 602-

616...... 23

Figure 9 Solution-phase synthetic schemes showing the stepwise formation of metal- carbon bonds from the Matteson research group.40 Reproduced from Matteson, D. S.

Methanetetraboronic and Methanetriboronic Esters as Synthetic Intermediates. Synthesis,

1975, 147–158...... 31

Figure 10 Solution phase gold carbon cluster synthesized by the Schmidbaur group.43

Reproduced from Scherbaum, F.; Grohmann, A.; Huber, B.; Krüger, C.; Schmidbauer, H.

,,Aurophilie" als Konsequenz Relativistischer Effekte: Das Hexakis

2+ (triphenylphosphanaurio) methan- Dikation [(Ph3PAu)6C] . Angew. Chem. 1988, 100,

1602–1604...... 32

Figure 11 The solution-phase synthesis of TiS2(en): S powder is loaded into 2-piece glass reactor and evacuated on Schlenk line. Once backfilled with argon, ethylenediamine is injected to dissolve and reduce the sulfur. The solution is red in colour. Finally, TiCl4 is injected and the jet-black TiS2(en) product is formed. The slurry is heated for various times, and transferred to a Parr reactor for extended reaction times...... 51

Figure 12 Crystal structures of (a) TiS2(en) and (b) TiS2(pn) projected down the c-axis. Ti octahedra are colored blue. (c) TiS2(en) and (d) TiS2(pn) chains along the c-axis. (e) en- intercalated TiS2 layers. Atom colors: Ti, blue; S, yellow; N, green; C, black. H atoms are omitted for clarity...... 55

xvi

Figure 13 (a) Synthetic phase space diagram of the TiS2(en) system: red, intercalated

TiS2; orange, mixed phases; green, TiS2(en); blue, amorphous. (b) Representative powder

X-ray diffraction patterns of each phase...... 62

Figure 14 (a) Powder X-ray diffraction patterns of products obtained by varying the en: Ti ratio: (i) 30:1, (ii) 20:1, and (iii) 10:1. (b) Powder X-ray diffraction patterns of products obtained by varying the temperature: (i, 220; ii, 180; iii, 140; iv; 100 °C). en: Ti ratio of 20:1...... 63

Figure 15 (a) Synthetic phase space diagram of the TiS2(pn) system: red, intercalated

TiS2; orange, mixed phases; green, TiS2(pn); blue, amorphous. (b) Representative powder

X-ray diffraction patterns of each phase...... 64

Figure 16 2-D intercalated TiS2 XRD patterns before (black, bottom) and after (red, top) annealing with an additional 20 equivalents of en at 180 ˚C for 1 day...... 65

Figure 17 1-D TiS2(en) XRD patterns before (black, bottom) and after (red, top) annealing with an additional equivalent of en at 180 ˚C for 1 day...... 66

Figure 18 DRA of 1-D TiS2(en), 1-D TiS2(pn), and 2-D intercalated TiS2...... 69

Figure 19 (a) Diffuse reflectance spectra of TiS2(en) and TiS2(pn). (b) Calculated band structure of TiS2(en). (c) Partial density of states of Ti 3d (black) and S 3p (red) for

TiS2(en). (d) Calculated band structure of TiS2(pn). (e) Partial density of states of Ti 3d

(black) and S 3p (red) for TiS2(pn)...... 70

Figure 20 Partial density of states of (a) TiS2(en) and (b) TiS2(pn) showing Ti, S, and N contributions...... 71

xvii

Figure 21 Brillouin zone and k-space pathway for TiS2(en) represented in the reduced rhombohedral unit cell...... 72

Figure 22 Diboryl-, triboryl, and tetraborylmethanes which have been synthesized by the

Matteson group. Reproduced from Matteson, D. S. Methanetetraboronic and

Methanetriboronic Esters as Synthetic Intermediates. Synthesis, 1975, 147–158...... 90

Figure 23 Aldehyde and ketone condensation products formed from the tris(trimethylenedioxyboryl)methide ion synthesized by the Matteson group. Reproduced from Matteson, D. S. Methanetetraboronic and Methanetriboronic Esters as Synthetic

Intermediates. Synthesis, 1975, 147–158...... 91

Figure 24 Exploded view of Swagelok ® fitting, O-ring, gas inlet hose assembly for delivering neat BCl3 gas from tank (Sigma Aldrich) to condensation flask...... 97

Figure 25 Vacuum filtration of C[B(OMe)2]4 product after passing it through inverted fritted filter funnel assembly...... 98

Figure 26 Transesterification of C[B(OMe)2]4 to C(Bpg)4 stirring at RT for 1 h...... 99

Figure 27 Transesterification of C[B(OMe)2]4 to C(Bpg)4 with stirring stopped in order to show C(Bpg)4 precipitate...... 100

Figure 28 Filtration of C(Bpg)4 transesterified product from THF solvent...... 101

Figure 29 Filtration of C(Bpg)4 transesterified product from THF solvent...... 102

Figure 30 C(Bpg)4 transesterified final product purified, dried, and stored in Ar-filled glovebox...... 103

1 Figure 31 H NMR of C(Bpg)4 showing OCH2 triplet at 3.9, and CH2 quintet at 1.8 ppm

...... 107

xviii

11 Figure 32 B NMR of C(Bpg)4 showing RB(OR’)2 of C-B-O2 at 30.5 ppm...... 108

1 Figure 33 H NMR of trimethyl borate reagent showing OCH3 singlet at 3.47 ppm. .... 109

11 Figure 34 B NMR of trimethyl borate reagent showing B(OR)3 peak at 18.47 ppm. . 110

1 Figure 35 H NMR of C[B(OMe)2]4 precursor showing OCH3 singlet at 3.51 ppm...... 111

11 Figure 36 B NMR of C[B(OMe)2]4 precursor showing C-B-O2 of R(BOR’)2 at 31 ppm and excess trimethyl borate at 18.49 ppm...... 112

Figure 37 Thermogravimetric analysis of C(Bpg)4 sample from room temperature to

1000 °C...... 113

Figure 38 DSC analysis of C(Bpg)4 sample from room temperature to 325°C...... 114

Figure 39 Reaction of GeC target products. Stirring under argon at refluxing temperatures, covered with foil to mitigate light exposure to GeI4 precursor...... 136

Figure 40 Solution-phase synthesis of GeCH material...... 137

Figure 41 Annealing target GeC product in alumina crucible under flowing argon in tube furnace...... 138

Figure 42 Filling 19 mm polyimide tubes with sample, top plugged with glass wool before sealing tube ends with epoxy...... 143

Figure 43 Samples and standards sealed in polyimide tubes for PDF analysis, CeO2 and empty tube standards at left...... 144

Figure 44 Samples and standards sealed in polyimide tubes and loaded into autosampler stage for PDF analysis...... 145

Figure 45 PDF analysis setup: beam inlet, right; samples loaded in autosampler stage, middle; variable tracked length 2-D area detector, left...... 146

xix

Figure 46 Powder XRD of amorphous germanium carbon product. Amorphous carbon broadness between 20-25° 2θ; aluminum sample holder peaks at 38 and 44° 2θ...... 150

Figure 47 Representative IR spectra of dried, crude GeCH material (bottom), and acetonitrile-rinsed and dried material (top). Ge-C bonding below 600 cm-1...... 151

Figure 48 Raman spectra of representative GeCH samples. Bottom trace, Al background slide. Second from bottom, green trace, shows no peaks in unannealed samples. Middle trace, orange, shows no Raman-active modes in 225 °C annealed sample. Second from top, red trace, shows no Raman-active modes in 400 °C annealed sample. Top, black trace, shows the transformation into phase-separated elements Ge and C at elevated

-1 temperature of 550 °C. The Ge E2 vibration appears at 300 cm and the graphite D and

G modes appear at 1350 and 1600 cm-1, respectively...... 152

Figure 49 TGA of sample of target GeCH product showing very little mass loss during phase separation near 500 °C followed by the formation and volatilization of germanium nitride in the N2 environment of the TGA...... 153

Figure 50 DSC showing melting and recrystallizing transitions during the phase separation of germanium and carbon from as-grown GeCH product...... 154

Figure 51 PDF of target GeCH products as synthesized (beige, bottom) and 400 °C annealed (black, top), showing a Ge-C bond length of 1.86 Å and a Ge-Ge bond length of

2.96 Å...... 155

Figure 52 XPS spectra of Ge 3d post etch region of GeCH sample. The scan reveals two

Ge states: one elemental oxidized form and one carbide form...... 156

Figure 53 XPS spectra of C 1s post etch of GeCH sample. The scan reveals a carbide state at 283 eV and amorphous carbon as the shoulder at 284 eV...... 157

Figure 54 Overlay of the XPS of Ge 3d region of GeO2 material (black, top, 32.5 eV peak, and the amorphous GeCH material (red, bottom, 27.5 eV to 31 eV peaks)...... 159

Figure 55 Overlay of C 1s XPS spectra showing the adventitious carbon at 284 eV in the

GeO2 product (black, top) while the GeCH sample displays a shifted C 1s peak corresponding to the carbide species at 283 eV (red, bottom)...... 160

-1 Figure 56 IR spectrum of GeO2 product illustrating Ge-O stretching at 716 cm ...... 163

Figure 57 Representative TGA of several oxidized germanium samples synthesized with different precursors across different solvents, spectrum showing weight % of 0-100.

These oxidized germanium products do not volatilize below 1000 °C...... 164

Figure 58 Representative DSC of oxidized germanium product formed upon reaction of

C(Bpg)4 and Ge(NMe2)4 showing no phase separation below the 550 °C temperature limit of Al DSC pan...... 165

Figure 59 PDF of GeC target material as-synthesized (blue, bottom) and annealed to

400 °C, (orange, top). This displays a Ge-O bond at 1.74 Å and a Ge-Ge second-nearest-neighbor peak at 3.23 Å...... 166

Figure 60 XPS of Ge of target GeC oxidized material post etch. The post-etch scan reveals one oxidized Ge state at 32.5 eV corresponding to GeO2...... 167

Figure 61 XPS of C 1s of target GeC material post etch. The post-etch scan reveals one

C state at 284 eV corresponding to adventitious amorphous carbon. No carbide C species is present at 283 eV...... 168

xxi

Chapter 1 Introduction

The wealth of diversity of solid-state inorganic chemistry has afforded many technological advances throughout history. Irrespective of elementary or advanced synthetic and processing methods, the variety in properties of inorganic elements has led to fascinating new materials. From ancient times through the industrial revolution, mining has provided several elemental building blocks for metal compounds, electronic materials, and chemical manufacturing. The increased use of metal led to automation and mass production in textile making and engines, for example. More recently, the development of silicon-based materials has changed the electronics industry forever.

While theoretical studies often predict the existence of novel materials, and novel devices demand novel material properties, realization of these materials relies on their syntheses. In order to create advanced hybrid materials with tailored properties, alternative synthetic methods to traditional solid-state syntheses need to be developed.

This dissertation focuses on two major projects: establishing the synthesis of dimensionally reduced derivatives of TiS2 and establishing a solution-phase pathway towards carbide materials, using GeC as a proof-of-concept material. In this chapter, we will first review the major principles of and differences between conventional solid-state synthesis methods and solution-phase methods. Secondly, we will introduce

1 dimensionally reduced materials and our motivation for the exploration of dimensionally reduced TiS2 derivatives. Finally, we will discuss the applications and syntheses of carbide phases and, in particular, germanium carbide materials, which motivate our desire to create a solution-phase synthesis route.

1.1 Conventional Solid-State Synthesis Methods

Traditional solid-state synthesis methods entail the intimate mixing and annealing of two different solid-state materials together to create new phases. These pathways depend on atomic diffusion, and therefore, very high temperatures are necessary: the interstitial and substitutional lattice diffusion mechanisms are most active near the melting point of an element, necessitating high temperatures to generate new phases.

High temperature and long growth-time methods mean that the products that are formed are the most thermodynamically stable phases. There are very few ways to attain kinetic control, or form metastable products using conventional solid-state methods.

In a typical solid-state reaction, precursors are first finely ground in an intimate mixture because the homogeneity of products is dependent on diffusion overcoming the concentration gradient between precursor boundaries. Precursors should be ground with a mortar and pestle or pulverized in a ball mill to achieve intimate mixing and small grain-boundary sizes. Additionally, pelletizing the powders helps ensure contact between reactants. Next, it is necessary to anneal the precursors for some time in an appropriate boat (metal or ceramic) or sealed quartz tube under the desired atmosphere. Numerous

2 annealing methods include heating in a reactive gas environment, applying high pressure to the reactant mixture, using precursor combustion reactions, or subjecting the reactants to a microwave environment. In conventional solid-state syntheses, generally, still, the only way to accelerate diffusion, and therefore reaction, is to increase temperature.

Another reaction type commonly employed is solid-state metathesis. Commonly, metal halide or oxide precursors reach high temperatures in a self-sustaining reaction because of the extremely favorable energetic driving forces associated with forming stable salt byproducts. One such example is how MoS2 can be synthesized from MoCl5 and Na2S at a self-initiated temperature of 1050 °C in under one second due to the extreme driving force for the formation of NaCl.1,2 This strategy allows the temperature of initiation to be room temperature, but high temperatures are achieved due to the extreme exothermic nature of the reaction. These temperatures can be lower than traditional solid-state methods, but may not always produce high-purity products as binary melt reactions.

Thin-film growth is also a very common route to grow materials on different substrates. These methods include chemical vapour deposition, laser ablation, sputtering, and molecular beam epitaxy. Chemical vapour deposition involves heating a reactive substrate in a reactive flowing gas and depositing a film of product. This method can also be initiated with photoactive chemicals, plasma assistance, or by using volatile metal-organic precursors. Taking place under vacuum atmosphere, laser ablation involves pulsing a reactive substrate with a laser beam and volatilizing it to the vapour phase in order to deposit a new condensed phase as a film on a targeted substrate. This

3 method can be extended to different systems using sputtering. Here, ion bombardment via plasma is used instead of laser pulses in order to generate reactive species from the initial substrates. These advanced methods have successfully been used to grow thin films of many different phases, but these methods also rely on vacuum environments and very high temperatures and often slowly producing very few layers of films.

1.2 Solution-Phase Synthesis Methods

While traditional solid-state synthesis methods have successfully generated many novel phases, solution-phase syntheses offer alternative routes to creating new materials.

Solution syntheses offer scalability, ease of setup, ability to access metastable phases, improved reaction kinetics, and compatibility with organic precursors. By incorporating organic molecules, for example, these methods generate new materials inaccessible via high-temperature methods. Therefore, novel families of solid-state materials have been created using solution phase methods, such as nanomaterials, dimensional reduction materials, hybrid inorganic-organic materials, and intercalation compounds. These solution-phase chimie-douce methods are also amenable with sonochemical-, microwave-, or pressure-assisted strategies.

Wet chemistry methods provide flexibility in precursor choice, solvent, product morphology, stepwise reagent addition for core-shell materials, purification, scalability, and processing (Figure 1). Additionally, new solution-phase methods such as sol-gel synthesis, hydrothermal synthesis, liquid thin-film growth, liquid-phase epitaxy, solution

4 casting, and nanoparticle hot-injection synthesis have all led to the creation of novel materials with novel properties. Solution-processable small-molecule organics, quantum dots, and perovskite structures have been implemented into next-generation, multijunction, solution-processed, and flexible solar cells,3–6 and quantum dots, nanowires, and liquid crystals have been commercialized in displays, for example. In these products, bandgap tuning is achieved by morphology control in solution growth and processing. For example, common solution-phase nanomaterial syntheses can now produce nanocrystals, straight and branched nanowires, and single- and multi-layer nanosheets, as shown in Figure 2.

Likewise, sol-gel synthesis provides a broad method to create ceramic materials by preparation of a sol, gelation of the sol, and removal of the solvent. The precursors may be particles or polymers in colloidal solutions with ligands, for example, and after some time, they transition to solid-phase products. Subsequently, the solvent is removed, and the amorphous material densifies and is annealed to create a crystalline product. The sol-gel method is amenable to many precursors and substrates, and large morphologies enable crystallization and gelation at lower temperatures than solid-state methods.7

Similar to solution-phase nanomaterial syntheses, sol-gel processing also enables the generation of heterogeneous embedded nanocrystals, for example.8

Sol-gel methods are closely related to hydrothermal and solvothermal methods.

Using water or a different solvent, respectively, these solution-phase methods employ temperature gradients and high-pressure inside of inert autoclave growth chambers in order to generate crystalline materials. These methods produce materials at temperatures

5 lower than the melting point of their precursors and allow for the generation of materials whose vapour pressure would be prohibitively high to synthesize via melt syntheses.

Sol-gel methods can also be adapted to thin-film production by introducing a substrate to the sol. Similar reaction apparatuses are also employed for sonication- and microwave-assisted synthetic strategies. Sonochemical methods rely on the generation of vapour cavities inside of solvent liquid via the application of ultrasonic frequencies.

Cavitation formation, growth, and collapse lead to locally high temperatures and pressures that can cause or enhance nearby chemical reactions. Analogously, microwave-assisted syntheses subject the reactions to microwave radiation. This adaptation enables selective reactant heating. Additionally, these synthetic methods provide outcomes not easily reproducible by traditional convection or conduction, such as rapid heating rates, local heating, and temperature gradients. These alternative methods have been used to synthesize a wide variety of solid-state products.9

In the future, bottom-up solution-phase methods of generating metal cluster products would allow more novel materials to be commercialized via solution casting and

3-D printing. This would enable precisely controlled syntheses of new materials with novel properties to be carried out on an industrial scale with on-demand capabilities.

While 3-D printers with plastics and polymer printing capabilities have already been widely deployed as benchtop units, 3-D printing of metals currently occurs in vacuum or using laser and electron beams, melting elemental powders and recasting them into the desired form factor.

Figure 1 Solution-phase synthetic reaction apparatuses: stirring reagents in three-neck flasks at refluxing temperatures while under Ar on Schlenk line.

Figure 2 Images showing the variety of nanomaterials easily accessible via solution- phase, multi-step injection synthetic methods. Left: straight and branched CdSe nanowires. Right: CdSe nanosheets.

1.3 Overview of Dimensional Reduction

While many basic and revolutionary advances have resulted from the discovery of elements, the combining of elements to form compounds multiplies their applications.

Recently, nanoscience has illuminated the fact that materials’ properties change based on their size and morphology when confined to length scales below 100 nm.10 Nanoscale materials lie in a physical size regime between bulk materials and single molecules and atoms; nanoscience explores why substances differ on the macro and microscopic scale.

A bulk material generally has constant physical properties, while nanoparticles have size-dependent properties. For example, the optical absorption and melting point of materials differ depending on their size.11 Also, most heterogeneous catalysts employ nanomaterials because nanostructures have a high surface-to-volume ratio.12

Additionally, both the absorption and photoluminescence of semiconducting nanostructures change due to quantum confinement effects when the diameter of nanowires and nanocrystals approaches the Bohr radius of the electron-hole pair.

Similarly, 2-D layered nanosheets are effectively potential wells with discrete energies caused by the planar confinement of the charge carriers.

Similar to property changes due to quantum confinement in nanoscience, the formalism of dimensional reduction suggests that disrupting the framework connectivity of an inorganic crystal structure along one or two dimensions with the addition of a salt or an organic terminating ligand drastically changes the physical and electronic properties

9 of the material.13 Due to the wide range of bonding mechanisms of inorganic compounds, inorganic synthesis is far less predictive than organic chemistry, for example. However, the principles of isostructural geometries, coupled with the addition of specific inorganic salts, make it possible to envision a structure with connectivity truncated in one or two dimensions, yet retaining a portion of its original metal-anion connectivity. For example, it is possible to reduce a 3-D M-X binary solid to 2-, 1-, and

O-D structures by injecting additional X atoms into the structure and disrupting M-X-M bridges to reduce the extended connectedness of the original framework,13 as shown in

Figure 3.

Both the metal coordination geometry and polyhedron connectivity

(corner-sharing, edge-sharing, or face-sharing) are predicted to be the same in the dimensionally reduced material as in the original parent material. Additionally, some measure of predictability is afforded because the formal oxidation states, electronegativities, and radii of the metal and anion are unchanged by directional termination of the structure. Dimensional reduction via organic ligand termination further broadens the synthetic possibilities to tune the electronic properties of traditionally 3-D solid-state materials. Multiple different structures can be derived from a lattice simply by differing the organic ligand coordination, as shown in Figure 4 and

Figure 5.14 By altering the framework connectivity along a certain direction of an inorganic lattice via insertion of an ionic salt or coordination by an organic ligand, a novel material with unique properties is formed.

Materials sculpted from three dimensions to two and one dimensions display entirely different properties, such as increased band-edge absorption, decreased thermal conductivity, and tunable bandgaps resulting from a merger of the excellent electromagnetic properties of the inorganic framework and the flexible, electronically insulating behavior of the organic ligand. For example, creating a 2-D layer of

ZnTe(en)1/2 from the 3-D wurtzite parent lattice of ZnTe blue shifts the optical bandgap from 2.1 eV to 3.5 eV.15 Bandgap shifts of up to 2 eV are observed in CdS, ZnS, and

ZnSe systems comprised of zero, one, and two atomic layers of inorganic chalcogenide and organic ethylenediamine spacer ligand.16 Additionally, terminating FeSe with en forms tetrahedral FeSe2 chains within a new material, Fe3Se4(en)2, possessing altered magnetic properties.17 Another family of perovskite-like metal-oxides exhibits layer- and ligand-dependent band gap, thermal conductivity, and dielectric constant tunability, for example.18 Overall, the formalism of dimensional reduction provides a synthetic route to alter the properties of materials. This strategy will be employed to study the property changes between 2-D and 1-D structures of TiS2, as described below.

Figure 4 (a) Dimensionally reduced zigzag and linear 1-D structures derived from the CdI2 structure type. (b) Left and middle: 1-D zigzag edge-sharing chain fragments highlighted in blue in the parent TiS2 lattice; right: 1-D zigzag chain of TiS2(en). Atoms Ti, blue; S, yellow; C, black; N, green. Reproduced from Li, T.; Goldberger, J. E. Atomic-Scale Derivatives of Solid-State Materials. Chem. Mater. 2015, 27 (10), 3549-3559.

Figure 5 (a) Left: the germanium crystal structure with one (111) plane highlighted in red; right: one layer of the GeH lattice. Atoms: Ge, blue and red; H, black. (b) Left: ZnS lattice (wurtzite) with two (110) planes of tetrahedra highlighted in red; right: Zn2S2(ba) (up) and ZnS(ba) lattices (down) (ba = butylamine). Atoms: Zn, gray and red; S, yellow; C, black; N, blue. Reproduced from Li, T.; Goldberger, J. E. Atomic- Scale Derivatives of Solid-State Materials. Chem. Mater. 2015, 27 (10), 3549–3559.

1.4 TiS2 and Derivatives

1.4.1 Structure and Applications of TiS2

Titanium disulfide is a commonly used industrial lubricant because it is a 2-D layered van der Waals (vdW) material. The 2-D layered transition metal dichalcogenides are used alongside graphite as dry lubricants because of the weak vdW forces between the layers of the material. Additionally, TiS2 has been investigated since the 1970s as a lithium-ion battery material because it is amenable to intercalation.19 In addition to Li

19,20 intercalation, TiS2 is also readily impregnated by Hg at room temperature. It was a very promising battery material for both its intercalation capacity and its high electrical conductivity. Currently, it is very easily synthesized via a wide variety of methods, and it is also amenable to intercalation between its vdW layers with organic molecules of varying sizes.

1.4.2 Dimensional Reduction of TiS2

Though the properties of 2-D layered TiS2 are well studied, it is not well understood how changing this 2-D lattice motif to a 1-D chain of atomic size will affect the properties of the material. To this end, the Goldberger research group has created the novel TiS2(en) phase in which coordinating the Ti-S framework with organic

15 ethylenediamine (en) changes the 2-D material to a 1-D atomic-scale material and

21 changes the 0.3 eV indirect bandgap TiS2 material to a 1.7 eV direct bandgap material.

Using the principles of dimensional reduction in solution-phase methods allows for the incorporation of organic ethylenediamine into the lattice at low temperatures, thus coordinating the Ti frame and totally changing the electronic properties of the 2-D parent lattice. Therefore, the solution-phase dimensional reduction of 2-D TiS2 has served as a model system for subsequent syntheses. This system has been extended to analogous chalcogenides in order to track the crystal structure and bandgap changes in TiSe2 and

22 TiSeO phases. Additionally, this TiS2(en) phase retains its ability to reversibly

23 intercalate Li, similar to the original 2-D Li ion battery material, TiS2, as shown in

Figure 6. Finally, the principles of dimensional reduction can be applied to analogous transition metals in order to generate novel phases in solution capped by organic ligands.24 It is for these reasons that developing rational synthetic pathways to generating novel dimensionally reduced materials must be explored. By understanding the ligand type, ligand-to-metal ratio, and temperature used in these solution-phase syntheses, we will be able to create novel phases with properties different from their original lattice framework.

Figure 6 (a) Crystal structures of TiS2(en) (left) and LiTiS2(en) (right) projected down the c-axis from neutron diffraction refinements. Ti octahedra are shaded in red. (b, c) X- ray diffraction patterns of TiS2(en) (black), Li0.66TiS2(en) (red), and LiTiS2(en). (d) Changes of unit cell parameter (a, black; c, red) with respect to Li/Ti ratio. (e) Crystal structures showing two adjacent 1-D chains along c axis with intercalated Li atoms positions determined from neutron diffraction refinements. Atom colors: Ti, red; S, yellow; N, green; C, black. H atoms have been omitted for clarity. Reproduced from Li, T.; Liu, Y.-H.; Chitara, B.; Goldberger, J. E. Li Intercalation into 1D TiS2(en) Chains. J. Am. Chem. Soc. 2014, 136, 2986.

1.4.3 Solution-Phase Synthesis of TiS2 Derivatives

The investigations described in this thesis use low-temperature solution-phase synthesis. Methods such as Molecular Beam Epitaxy (MBE), and Metalorganic

Chemical Vapor Deposition (MOCVD) are some of the many ways to produce solid-state materials, but solution synthesis is a low cost/low temperature alternative. Solution synthesis is not limited by substrate size, substrate uniformity, or the need for expensive heating ovens, as are the above methods. Solution chemistry also enables the control of surface chemistry during reaction. By using organic molecules as surface coordinating or non-coordinating ligands, the growth kinetics and morphologies of materials can be controlled, the resulting materials are protected from oxidation, and structures can be dispersed in common organic solvents for further processing and device integration. The motivation for this study is to develop dimensional reduction pathways and develop strategies to control material growth morphology and ligand functionalization. While excellent control has previously been demonstrated, with solution syntheses producing quantum dots ranging from 2-6 nm in diameter and nanowires with diameters ranging from 5-50 nm, the controlled growth of atomic-scale materials are not yet so developed.

To this end, low-temperature solution-phase synthesis is used to rationally create atomic-scale derivatives of solid-state materials. For these reasons, TiS2 is studied as a model system to apply dimensional reduction principles from solid-state to solution-

18 phase syntheses in order to incorporate organic ligands into inorganic lattices and generate novel phases with novel properties.

1.5 Carbide Materials

1.5.1 Applications of Carbides

Carbides have become especially valuable materials with the rapid industrialization over the past several hundred years. The variety of bonding capabilities of carbides has spurred their use across a wide range of applications, foremost of which are as refractory materials, robust materials, and machine cutting tools, as shown in

Figure 7. Refractory materials are specifically defined as non-metallic materials that maintain their strength properties above 1000 °F, according to ASTM International.

Covalent carbides are refractory because of the incredible bonding between similarly sized lattices and their similar electronegativity. Interstitial carbides find their high- strength properties from favourable intermolecular forces generated by interstitial packing of carbon between close-packed metal atoms. In this regard, carbides are used extensively alongside metal oxides in industrial metallurgy and as temperature-resistant coatings and components. For example, SiC is extremely refractory with a melting point of 2830 °C and is used as an abrasive tool and in compound armour assemblies, automobile parts, electronic components, heating elements, and as a high-temperature synthetic template for graphene and other materials. Similarly, tungsten carbide, WC, is

19 extremely hard, refractory, and, with a melting point of 2870 °C, it is used in abrasive cutting tools, components of ammunition, hole-boring assemblies in mining operations, jewelry, and surgical tools.

The 1973 discovery of the catalytic activity of tungsten carbide25 widened the uses of carbides dramatically. Tungsten carbide has similar catalytic activity to expensive platinum catalysts in hydrodesulfurization and hydrocarbon reformation reactions,25,26,27–29 though it has proven very difficult to synthesize and isolate bulk and nanoparticle single phases of WC and W2C without catalytically inactive tungsten and carbon impurities. Establishing a route towards synthesizing and isolating single-phase products of WC vs. W2C, for example, is necessary in order to fully exploit their catalytic activity in fuel cells and their sulfur resistance in the hydrodesulfurization and reformation of feedstock in petroleum refineries.

In addition to the broad catalytic activity of carbides (see Figure 8), interstitial carbides have excellent electronic properties. Interstitial carbides are very similar electronically to metals due to their excellent thermal and electrical conductivity. A wide range of bonding, from metallic to covalent and ionic, is exhibited in interstitial carbides, and therefore, these materials exhibit favourable strength properties similar to ceramics, but also possess the electronic properties of metals. These cermets, the combination of a ceramic and a metal, are highly desirable for cutting tool, jet engine, defense, and electronic research initiatives.30 The desirable metallic (high strength and thermal conductivity) and plastic-like (ductility, deformation) properties are leveraged with the

20 desirable ceramic properties such as high melting point and chemical stability in order to created advanced materials.

Figure 7 Silicon carbide is a refractory material that is cast as tools, dies, and parts for many fields. Reproduced from Saint Gobain Ceramics and Plastics, Inc.

1.5.2 Applications of Germanium Carbide

Materials with high thermal conductivity are useful as heat sink and heat transfer materials and are therefore highly desirable components of advanced integrated circuits, laser assemblies, thermoelectrics, transistors, and LEDs. Metallic heat sinks such as aluminum and copper (229 W/m•K and 385 W/ m•K) and their related industrial alloys rely on scattering from delocalized electrons but, while inexpensive, have limited application in advanced electronic assemblies. Alternatively, diamond derives its excellent thermal conductivity (2290 W/m•K) from highly ordered covalent bonds and small lattice vibrations. In such high-quality crystalline materials such as diamond, low phonon-phonon scattering imparts a 5-10 x improvement in thermal conductivity compared to metals. Therefore, diamond is a highly sought-after material for use in advanced microelectronics and in laser applications. However, in order to provide sufficient thermal cooling of specific substrates in semiconductor and microchip applications, more complex composite materials are also being developed. For example, alloys such as AlSiC, (200 W/m•K ) and Dymalloy (Cu, Ag, Diamond, 420 W/m•K) have thermal expansion coefficients similar to certain electronic and ceramic materials and are therefore used in integrated circuits, GaAs microchip substrates, and ceramic aircraft components. To this end, carbides are promising materials for advanced thermal management applications. Overall, solid-state carbide materials occupy a unique niche in

24 the materials industry because they are incredibly refractory, chemically robust, catalytically active, and electronically and thermally conducting ceramic materials.

GeC would provide superior alternatives to current heat sink materials used in aerospace and electronic applications. Graphitic germanium powders are also used as robust anti- wear coatings.33 In order to deploy a novel GeC phase for a wide variety of applications its synthesis must be scalable, processable, and amenable to current industrial device substrates.

1.5.3 Synthesis of Carbides

Though there are many uses for carbide materials, their synthesis remains nontrivial. For example, the most refractory binary carbide known is HfC (with a melting point of 3900 °C), but it has not been as widely used as SiC or WC because it is difficult to synthesize in certain stoichiometries. The difficult-to-synthesize binary HfC has partially prompted research on ternary carbides in an attempt to alloy carbides substoichiometrically and reproducibly in order to harness electronic, magnetic, and physical properties of multiple materials.

While many carbide phases can be produced using the traditional solid-state methods described in the introduction above, carbide materials are also produced via

25 arc-melting of metal precursors and carbon at local temperatures of over 1500 °C.29

Similarly, solid-state metathesis reactions have also been developed for carbides wherein solid precursors are intimately mixed so that exothermic, self-propagating reactions generate certain phases in seconds.34,35 Additionally, molten salt flux syntheses have been used to produced carbides wherein liquid salts act as solvents.36 Finally, layer-by-layer crystalline growth may be obtained by molecular beam epitaxy wherein precisely determined vapor fluxes of multiple atoms or molecules are concurrently directed at a target substrate.

Alternatively, the development of solution-phase carbide syntheses would eliminate the need for laser heating sources, would not limit products to already known metal powders, and would enable the generation of lattice-matched carbides amenable to commercialized semiconductor and ceramic substrates. While traditional solid-state methods generate a myriad of carbide products with varying degrees of crystallinity and structural variations, solution-phase methods offer morphological and bond functionalization control and scalability that is amenable to current and next-generation solution printing and casting technologies. Furthermore, solution-phase syntheses offer the opportunity to create ternary carbide compounds in a stepwise fashion that is not controllable in classic solid-state reactions. In order to prepare monodisperse crystalline nanoparticles and small-molecule analogues of metal carbides for use as catalysts, anti-wear coatings, and electronically and thermally conducting materials, a solution- phase route is necessary to enable surface templating and morphological control.

Furthermore, specific metal-carbon bond creation can be achieved using specific

26 precursors, such as metal-organic and carbon-boron precursors. Traditional solid-state growth methods of carbides rely on high temperatures in order to break extremely stable

C-H bonds of hydrocarbon precursors to generate C4- sources. These high operating temperatures would decompose all solvents and surfactants necessary to creating morphologically tunable nanosized small-molecule carbido and dimensionally reduced materials. Establishing low-temperature solution-phase synthetic routes for the creation of carbide materials would benefit solution casting processes, thin-film spin-coating methods, and 3-D printing systems. Additionally, low-temperature, turnkey syntheses allow for the methodical study of the effect of nanoparticle size, surface facet, and morphology on the mechanical, catalytic, and optoelectronic properties of carbides.

Lastly, the rational synthesis of carbide materials could be extended to allow the growth of new classes of metastable carbide materials with unique properties.

Currently, there are very few reports of low-temperature solution-phase carbide growth. Prevalent methods outline nanoparticle syntheses that involve decomposing metal-organic precursors at several hundred degrees in specific surfactant solvents. In

1993, a report outlined the formation of nanoparticle group-6 metal carbide powders by chemical reduction of metal chlorides in tetrahydrofuran and subsequent annealing of the powders at 500 °C.37 Before that, a research program from the Matteson group in the

1960s outlined the low-temperature syntheses of several metal-carbido molecules using carbon-boron precursors. Using base-activated cyclic boronic esters reacted with triphenyl metal halides, for example, tetrametallomethanes containing one, two, or three group 14 metal-carbon bonds were created.38 These low-temperature solution-phase

27 syntheses generated novel metal-carbido molecules that were stable up to 200 °C. The

Matteson group has successfully metallated borylmethanes to various degrees (mono- to tri-substitution) using tin, lead, and mercury precursors 39,40 as shown in Figure 9. These syntheses were inspired by earlier reports of reactions between carbon tetrahalides and organochlorosilanes.41 Analogous work (see Figure 10) was carried out in the 1980s and

1990s by Schmidbaur showing that gold can form both tetrahedral and octahedral compounds with carbon ligands.42,43 These studies serve as the synthetic inspiration for our solution-phase routes towards germanium carbide.

1.5.4 Synthesis of Germanium Carbide

Germanium carbide has been identified as a target material from the solution- phase reaction of carbon-boron precursors and liquid metal-amine precursors.

Germanium carbide does not exist and is not amenable to solid-state syntheses because germanium and carbon phase separate above 500 °C.31 However, it is predicted to have a very high thermal conductivity, rivaling that of diamond.32 The creation of GeC would provide superior alternatives to current heat-sink materials used in aerospace and electronic applications. Additionally, graphitic germanium powders are used as robust anti-wear coatings.33 In order to deploy such novel phases for a wide variety of applications, it is necessary that their syntheses be scalable, processable, and amenable to current industrial device substrates. Taking inspiration from previous reports from

Matteson and Schmidbaur,39,42,43 Figure 9 and Figure 10 show it is possible to create C-B

28 precursors and subsequently react them with group 14 precursors in order to create systematic metal-carbon bonding in small-molecule carbide analogues. Creating C-B precursors that provide C4- ions in solution is imperative to generating analogous Ge-C materials at low temperatures via solution-phase routes.

Low-temperature solution-phase routes provide the environment to produce Ge-C bonds in a stepwise fashion from a variety of precursors. These products are also scalable and amenable to current device architectures as opposed to high-temperature graphitic germanium phases synthesized using traditional solid-state methods. Previously reported solid-state syntheses of Ge-C materials typically detail the formation of Ge products with less than 10% carbon integration, along with excess graphitic carbon coatings, which change the physical, electronic,44 and catalytic properties of the materials.45 Mild solution-phase routes will allow for the methodical study of Ge-C bond formation and will allow for more stoichiometric amounts of carbon to be incorporated into the lattice. Using bottom-up solution phase methods instead of top-down solid-state methods will also mitigate the formation of graphitic carbon coatings deposited by solid-state, high-temperature methods. The use of C-B precursors that provide C4- anions in solution will serve as a general precursor towards the creation of group 14 and transition-metal carbides. The large-scale synthesis of cyclic boronic esters will offer a reactive precursor that combines with commercially available Ge4+ ion sources in the form of tetrakis dimethyl amino germanes, and this solution platform can easily be a model system for the generation of other carbides via low-temperature solution-phase syntheses using C4- and M4+ precursors. GeC is predicted to have a high thermal

29 conductivity, approaching that of diamond, however, a stoichiometric GeC material has yet to be created. Thus far, CVD growth of GeC typically produces films whose stoichiometry has been characterized as Ge1-xCx, where x<0.1. Similarly, UHV CVD synthesis methods have generated germylmethane thin films with carbon content of

1-7%.46 Analogously, thin films of hydrogenated carbonaceous germanium with stoichiometries of Ge.9C.1:H have been generated by sputtering a Ge target in argon, hydrogen, and methane atmospheres.47 The goal of this work is to produce a stoichiometric GeC phase using carbon-boron precursors via low-temperature solution-phase synthetic routes.

Figure 9 Solution-phase synthetic schemes showing the stepwise formation of metal- carbon bonds from the Matteson research group.40 Reproduced from Matteson, D. S. Methanetetraboronic and Methanetriboronic Esters as Synthetic Intermediates. Synthesis, 1975, 147–158.

Figure 10 Solution-phase gold carbon cluster synthesized by the Schmidbaur group.43 Reproduced from Scherbaum, F.; Grohmann, A.; Huber, B.; Krüger, C.; Schmidbauer, H. ,,Aurophilie" als Konsequenz Relativistischer Effekte: Das Hexakis 2+ (triphenylphosphanaurio) methan- Dikation [(Ph3PAu)6C] . Angew. Chem. 1988, 100, 1602–1604.

1.6 Chapter Outlines

The introduction in Chapter 1 is followed by Chapter 2, which outlines the synthesis and characterization of atomic-scale derivatives of the titanium disulfide system. The crystal structures of hybrid organic/inorganic 1-D materials, TiS2(en) and

TiS2(pn), are presented. The synthetic conditions necessary to generate these novel products are summarized. Additionally, temperature and ligand ratio were explored over a wide range in order to construct a synthetic phase-space diagram of each 1-D hybrid material. Finally, the electronic band structures of these analogues were generated and were subsequently compared with the original 2-D titanium disulfide lattice.

In Chapter 3, the synthesis of carbon-boron molecules is outlined. Again, a comprehensive outline of all necessary synthetic conditions is presented. Additionally, seminal synthetic methods of methanetetraboronic esters and cyclic boronic esters will be presented along with several necessary updates to previous syntheses that increase yield and reproducibility. Finally, the characterization of these tetraboronic and triboronic esters will be presented. These products will ultimately serve as synthons for the hybrid germanium carbide materials that are presented in Chapter 4.

Chapter 4 outlines the synthesis and characterization of amorphous germanium-carbon phases. Synthetic conditions including C(Bpg)4 or HC(Bpg)3 precursor source, Ge precursor source, and solvent, purification, and subsequent

33 annealing steps will be presented. Next, a full suite of characterization methods are used in order to analyze the amorphous GeCH product and the oxidized GeC target product.

Finally, the conclusion of this dissertation and the outlook of 1-D hybrid and carbide materials are found in Chapter 5. The proposed synthesis of small-molecule carbide analogues generated via base-activated C-B precursors is presented in Chapter 5.

These syntheses would involve the reaction of base-activated C-B precursors with trimethyl germanium halides in order to form Ge-C bond-containing small molecules.

References

(1) Bonneau, P. R.; Jarvis, R. F.; Kaner, R. B. Rapid Solid-State Synthesis of

Materials from Molybdenum Disulphide to Refractories. Nature 1991, 349 (6309),

510–512.

(2) Gillan, E. G.; Kaner, R. B. Synthesis of Refractory Ceramics via Rapid Metathesis

Reactions between Solid-State Precursors. Chem. Mater. 1996, 8 (2), 333–343.

(3) Yuan, M.; Liu, M.; Sargent, E. H. Colloidal Quantum Dot Solids for Solution-

Processed Solar Cells. Nat. Energy 2016, 1 (February), 16016.

(4) Nie, W.; Tsai, H.; Asadpour, R.; Blancon, J.-C.; Neukirch, A. J.; Gupta, G.;

Crochet, J. J.; Chhowalla, M.; Tretiak, S.; Alam, M. A.; Wang, H.-L.; Mohite, A.

D. High-Efficiency Solution-Processed Perovskite Solar Cells with Millimeter-

Scale Grains. Science (80-. ). 2015, 347 (6221), 522–525.

(5) Sun, Y.; Welch, G. C.; Leong, W. L.; Takacs, C. J.; Bazan, G. C.; Heeger, A. J.

Solution-Processed Small-Molecule Solar Cells with 6.7% Efficiency. Nat. Mater.

2011, 11 (1), 44–48.

(6) Docampo, P.; Ball, J. M.; Darwich, M.; Eperon, G. E.; Snaith, H. J. Efficient

Organometal Trihalide Perovskite Planar-Heterojunction Solar Cells on Flexible

Polymer Substrates. Nat. Commun. 2013, 4, 2761.

(7) Schnepp, Z.; Wimbush, S. C.; Antonietti, M.; Giordano, C. Synthesis of Highly

Magnetic Iron Carbide Nanoparticles via a Biopolymer Route. Chem. Mater. 2010,

22 (18), 5340–5344.

(8) Hoffman, M.; Veinot, J. G. C. Understanding the Formation of Elemental

Germanium by Thermolysis of Sol-Gel Derived Organogermanium Oxide

Polymers. Chem. Mater. 2012, 24, 1283–1291.

(9) Kitchen, H. J.; Vallance, S. R.; Kennedy, J. L.; Tapia-Ruiz, N.; Carassiti, L.;

Harrison, A.; Whittaker, A. G.; Drysdale, T. D.; Kingman, S. W.; Gregory, D. H.

Modern Microwave Methods in Solid-State Inorganic Materials Chemistry: From

Fundamentals to Manufacturing. Chem. Rev. 2013, 114 (2), 1170–1206.

(10) Kuno, M. An Overview of Solution-Based Semiconductor Nanowires: Synthesis

and Optical Studies. Phys. Chem. Chem. Phys. 2008, 10, 620–639.

(11) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Synthesis and Characterization of

Monodisperse Nanocrystals and Close-Packed Nanocrystal Assemblies. Annu.

Rev. Mater. Sci. 2000, 30, 545.

(12) Li, Q. Anisotropic Nanomaterials Preparation, Properties, and Applications, 1st

ed.; Li, Q., Ed.; Springer, 2015.

(13) Tulsky, E. G.; Long, J. R. Dimensional Reduction: A Practical Formalism for

Manipulating Solid Structures. Chem. Mater. 2001, 13, 1149.

(14) Li, T.; Goldberger, J. E. Atomic-Scale Derivatives of Solid-State Materials. Chem.

Mater. 2015, 27 (10), 3549–3559.

(15) Huang, X. Y.; Li, J.; Fu, H. X. The First Covalent Organic-Inorganic Networks of

Hybrid Chalcogenides: Structures That May Lead to a New Type of Quantum

Wells. J. Am. Chem. Soc. 2000, 122, 8789.

(16) Huang, X.; Li, J. From Single to Multiple Atomic Layers: A Unique Approach to

the Systematic Tuning of Structures and Properties of Inorganic-Organic Hybrid

Nanostructured Semiconductors. J. Am. Chem. Soc. 2007, 129 (11), 3157–3162.

(17) Pak, C.; Kamali, S.; Pham, J.; Lee, K.; Greenfield, J. T.; Kovnir, K. Chemical

Excision of Tetrahedral FeSe2 Chains from the Superconductor FeSe: Synthesis,

Crystal Structure, and Magnetism of Fe3Se4(en)2. J. Am. Chem. Soc. 2013, 135,

19111.

(18) Zhang, X.; Hejazi, M.; Thiagarajan, S. J.; Woerner, W. R.; Banerjee, D.; Emge, T.

J.; Xu, W.; Teat, S. J.; Gong, Q.; Safari, A.; Yang, R.; Parise, J. B.; Li, J. From 1D

Chain to 3D Network: A New Family of Inorganic − Organic Hybrid

Semiconductors MO3(L)x (M = Mo, W; L = Organic Linker) Built on Perovskite-

like Structure Modules. J. Am. Chem. Soc. 2013, 135 (L), 17401–17407.

(19) Moreau, P.; Ouvrard, G.; Gressier, P.; Ganal, P.; Rouxel, J. Electronic Structures

and Charge Transfer in Lithium and Mercury Intercalated Titanium Disulfides. J.

Phys. Chem. Solids 1996, 57 (6–8), 1117–1122.

(20) Ong, E. W.; McKelvy, M. J.; Ouvrard, G.; Glaunsinger, W. S. Mercury

Intercalates of Titanium Disulfide: Novel Intercalation Compounds. Chem. Mater.

1992, 4 (1), 14–17.

(21) Liu, Y. H.; Porter, S. H.; Goldberger, J. E. Dimensional Reduction of a Layered

Metal Chalcogenide into a 1D Near-IR Direct Band Gap Semiconductor. J. Am.

Chem. Soc. 2012, 134, 5044.

(22) Li, T.; Liu, Y.-H.; Porter, S.; Goldberger, J. E. Dimensionally Reduced One-

Dimensional Chains of TiSe2. Chem. Mater. 2013, 25, 1477.

(23) Li, T.; Liu, Y.-H.; Chitara, B.; Goldberger, J. E. Li Intercalation into 1D TiS2(en)

Chains. J. Am. Chem. Soc. 2014, 136, 2986.

(24) Li, T.; Baum, Z. J.; Goldberger, J. E. A Vanadium Chalcogenide Dicubane. Eur. J.

Inorg. Chem. 2016, 2016 (1), 28–32.

(25) Levy, R. B.; Boudart, M. Platinum-like Behavior of Tungsten Carbide in Surface

Catalysis. Science 1973, 181 (4099), 547–549.

(26) Hwu, H. H.; Chen, J. G. Surface Chemistry of Transition Metal Carbides. Chem.

Rev. 2005, 105, 185–212.

(27) Hara, Y.; Minami, N.; Itagaki, H. Synthesis and Characterization of High-Surface

Area Tungsten Carbides and Application to Electrocatalytic Hydrogen Oxidation.

Appl. Catal. A Gen. 2007, 323, 86–93.

(28) Chen, W.-F.; Muckerman, J. T.; Fujita, E. Recent Developments in Transition

Metal Carbides and Nitrides as Hydrogen Evolution Electrocatalysts. Chem.

Commun. 2013, 49 (49), 8896–8909.

(29) Oyama, T. The Chemistry of Transition Metal Carbides and Nitrides; 1996.

(30) Ettmayer, P.; Kolaska, H.; Lengauer, W.; Dreyer, K. Ti(C,N) Cermets - Metallurgy

and Properties. Int. J. Refract. Met. Hard Mater. 1995, 13 (6), 343–351.

(31) Taylor, A.; Doyle, N. J. The Ge-C System at High Pressure. Scr. Metall. 1967, I,

161–162.

(32) Lindsay, L.; Broido, D. A.; Reinecke, T. L. Phonon-Isotope Scattering and

Thermal Conductivity in Materials with a Large Isotope Effect: A First-Principles

Study. Phys. Rev. B - Condens. Matter Mater. Phys. 2013, 88 (14), 1–9.

(33) Jacobsohn, L. G., Freire, F. L., Mariotto, G. Investigation on the Chemical,

Structural and Mechanical Properties Carbon-Germanium Films Deposited by Dc-

Magnetron Sputtering. Diam. Relat. Mater. 1998, 7, 440–443.

(34) Nartowski, A. M.; Parkin, I. P.; Craven, A. J.; MacKenzie, M. Rapid, Solid-State

Metathesis Routes to Metal Carbides. Adv. Mater. 1998, 10 (10), 805–808.

(35) Nartowski, A. M.; Parkin, I. P.; Mackenzie, M.; Craven, A. J. Solid State

Metathesis : Synthesis of Metal Carbides from Metal Oxides. J. Mater. Chem.

2001, 11, 3116–3119.

(36) Liu, X.; Fechler, N.; Antonietti, M. Salt Melt Synthesis of Ceramics,

Semiconductors and Carbon Nanostructures. Chem. Soc. Rev. 2013, 42 (21), 8237–

8265.

(37) Zeng, D.; Hampden-Smith, M. J. Synthesis and Characterization of Nanophase

Group 6 Metal (M) and Metal Carbide (M2C) Powders by Chemical Reduction

Methods. Chem. 1993, 5 (M), 681–689.

(38) Matteson, D. S.; Wilcsek, J. Tetrametallomethanes Containing One, Two, or Three

Group IV Metal Atoms and Boron. J. Organomet. Chem. 1973, 57, 231–242.

(39) Larson, Gerald L., Matteson, D. D. Tin and Lead Derivatives of

Tetrakis(dimethoxyboryl)methane. J. Organomet. Chem. 1973, 57, 225–230.

(40) Matteson, D. S. Methanetetraboronic and Methanetriboronic Esters as Synthetic

Intermediates. Synthesis (Stuttg). 1975, 147–158.

(41) Merker, R. L.; Scott, M. J. Preparation and Properties of Poly(tetramethyl-P-

Silphenylene-Siloxane). J. Polym. Sci. Part A 1964, 2 (1), 15–29.

(42) Schmidbaur, H.; Steigelmann, O. The First Complexes of Tetragoldmethane CAu4.

Z. Naturforsch. 1992, 47b (1992), 1721–1724.

(43) Scherbaum, F.; Grohmann, A.; Huber, B.; Krüger, C.; Schmidbauer, H.

,,Aurophilie" als Konsequenz Relativistischer Effekte: Das

2+ Hexakis(triphenylphosphanaurio)methan- Dikation [(Ph3PAu)6C] . Angew. Chem.

1988, 100, 1602–1604.

(44) Kazimierski, P.; Jóźwiak. Transition from Amorphous Semiconductor to

Amorphous Insulator in Hydrogenated Carbon-Germanium Films Investigated by

Raman Spectroscopy. J. Non. Cryst. Solids 2009, 355 (4–5), 280–286.

(45) Mariotto, G.; Vinegoni, C.; Jacobsohn, L. .; Freire, F. . Raman Spectroscopy and

Scanning Electron Microscopy Investigation of Annealed Amorphous Carbon–

germanium Films Deposited by D.c. Magnetron Sputtering. Diam. Relat. Mater.

1999, 8 (March), 668–672.

(46) Todd, M.; McMurran, J.; Kouvetakis, J.; Smith, D. J. Chemical Synthesis of

Metastable Germanium-Carbon Alloys Grown Heteroepitaxially on (100) Si.

Chem. Mater. 1996, 8 (10), 2491–2498.

(47) Drusedau, T. P.; Annen, A.; Schroder, B.; Freistedt, H. Vibrational, Optical and

Electronic Properties of the Hydrogenated Amorphous Germanium-Carbon Alloy

System. Philos. Mag. Part B 1994, 69 (1), 1–20.

(48) Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley, R. E. C60:

Buckminsterfullerene. Nature 1985, 318, 162.

(49) Iijima, S. Helical Mictrotubules of Graphitic Carbon. Nature 1991, 354, 56.

(50) Novoselov, K. S.; Geim, A. K.; Morozov, S. V; Jiang, D.; Zhang, Y.; Dubonos, S.

V; Grigorieva, I. V; Firsov, A. A. Electric Field Effect in Atomically Thin Carbon

Films. Science 2004, 306, 666.

(51) Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H. Chemically Derived, Ultrasmooth

Graphene Nanoribbon Semiconductors. Science 2008, 319, 1229.

(52) Kagan, C. R.; Mitzi, D. B.; Dimitrakopoulos, C. D. Organic-Inorganic Hybrid

Materials as Semiconducting Channels in Thin-Film Field-Effect Transistors.

Science 1999, 286, 945.

(53) Mitzi, D. B. Solution Processing of Chalcogenide Semiconductors via

Dimensional Reduction. Adv. Mater. 2009, 21, 3141.

(54) Morris, C. D.; Chung, I.; Park, S.; Harrison, C. M.; Clark, D. J.; Jang, J. I.;

Kanatzidis, M. G. Molecular Germanium Selenophosphate Salts: Phase-Change

Properties and Strong Second Harmonic Generation. J. Am. Chem. Soc. 2012, 134,

20733.

(55) Long, J. R.; McCarty, L. S.; Holm, R. H. A Solid-State Route to Molecular

2+ Clusters: Access to the Solution Chemistry of [Re6Q8] (Q = S, Se) Core-

Containing Clusters via Dimensional Reduction. J. Am. Chem. Soc. 1996, 118,

4603.

(56) Zhang, X.; Hejazi, M.; Thiagarajan, S. J.; Woerner, W. R.; Banerjee, D.; Emge, T.

J.; Xu, W.; Teat, S. J.; Gong, Q.; Safari, A.; Yang, R.; Parise, J. B.; Li, J. From 1D

Chain to 3D Network: A New Family of Inorganic-Organic Hybrid

Semiconductors MO3(L)x (M = Mo, W; L = Organic Linker) Built on Perovskite-

like Structure Modules. J. Am. Chem. Soc. 2013, 135, 17401.

(57) Huang, X.; Li, J. From Single to Multiple Atomic Layers: A Unique Approach to

the Systematic Tuning of Structures and Properties of Inorganic-Organic Hybrid

Nanostructured Semiconductors. J. Am. Chem. Soc. 2007, 129, 3157.

(58) Roushan, M.; Zhang, X.; Li, J. Solution-Processable White-Light-Emitting Hybrid

Semiconductor Bulk Materials with High Photoluminescence Quantum Efficiency.

Angew. Chem., Int. Ed. 2012, 51, 436.

(59) Peng, X. G.; Wickham, J.; Alivisatos, A. P. Kinetics of II-VI and III-V Colloidal

Semiconductor Nanocrystal Growth: “Focusing” of Size Distributions. J. Am.

Chem. Soc. 1998, 120, 5343.

(60) Burda, C.; Chen, X. B.; Narayanan, R.; El-Sayed, M. A. Chemistry and Properties

of Nanocrystals of Different Shapes. Chem. Rev. 2005, 105, 1025.

(61) Yin, Y.; Alivisatos, A. P. Colloidal Nanocrystal Synthesis and the Organic-

Inorganic Interface. Nature 2005, 437, 664.

(62) Wang, X.; Zhuang, J.; Peng, Q.; Li, Y. D. A General Strategy for Nanocrystal

Synthesis. Nature 2005, 437, 121.

(63) Plashnitsa, V. V; Vietmeyer, F.; Petchsang, N.; Tongying, P.; Kosel, T. H.; Kuno,

M. Synthetic Strategy and Structural and Optical Characterization of Thin Highly

Crystalline Titanium Disulfide Nanosheets. J. Phys. Chem. Lett. 2012, 3, 1554.

(64) Blaha, P.; Schwarz, K.; Madsen, G. K. H.; Kvasnicka, D.; Luitz, J.; Schwars, K.

WIEN2K: Full Potential-Linearized Augmented Plane Waves and Local Orbital

Programs for Calculating Crystal Properties; 2001.

(65) Tran, F.; Blaha, P. Accurate Band Gaps of Semiconductors and Insulators with a

Semilocal Exchange-Correlation Potential. Phys. Rev. Lett. 2009, 102, 226401.

(66) Koller, D.; Tran, F.; Blaha, P. Merits and Limits of the Modified Becke-Johnson

Exchange Potential. Phys. Rev. B 2011, 83, 195134.

(67) Camargo-Martinez, J. A.; Baquero, R. Performance of the Modified Becke-

Johnson Potential for Semiconductors. Phys. Rev. B 2012, 86, 195106.

(68) Boller, H.; Blaha, H. Intercalation Phases of Titanium Disulfide, 2H-Niobium

Disulfide, and 1T-Tantalum Disulfide with Ethylenediamine and

Trimethylenediamine: A Crystal, Chemical and Thermogravimetric Study. J. Solid

State Chem. 1982, 45, 119.

(69) Ogata, H.; Fujimori, H.; Miyajima, S.; Kobashi, K.; Chiba, T.; Taylor, R. E.; Endo,

K. Modification of Electronic Structure by Charge Transfer and Molecular Motion

in Titanium Disulfide-Ethylenediamine Intercalation Compound. J. Phys. Chem.

Solids 1997, 58, 701.

(70) Friend, R. H.; Yoffe, A. D. Electronic Properties of Intercalation Complexes of the

Transition Metal Dichalcogenides. Adv. Phys. 1987, 36, 1.

(71) Thomson, J. W.; Nagashima, K.; Macdonald, P. M.; Ozin, G. A. From Sulfur-

Amine Solutions to Metal Sulfide Nanocrystals: Peering into the Oleylamine-

Sulfur Black Box. J. Am. Chem. Soc. 2011, 133, 5036.

(72) Chen, C. H.; Fabian, W.; Brown, F. C.; Woo, K. C.; Davies, B.; DeLong, B.;

Thompson, A. H. Angle-Resolved Photoemission Studies of the Band Structure of

TiSe2 and TiS2. Phys. Rev. B 1980, 21, 615.

(73) Contemporary Boron Chemistry. In Tenth International Conference on Boron

Chemistry; Davidson, M., Hughes, A. K., Marder, T. B., Wade, K., Eds.; The

Royal Society of Chemistry, 1999; p 538.

(74) Siebert, W. Advances in Boron Chemistry; The Royal Society of Chemistry, 1996.

(75) Abu Ali, H.; Dembitsky, V. M.; Srebnik, M. Contemporary Aspects of Boron:

Chemistry and Biological Applications; Elsevier, 2005.

(76) Hall, D. G. Boronic Acids; Wiley-VCH, 2005.

(77) Science of Synthesis: Houben-Weyl Methods of Molecular Transformations; D, K.,

S, M. D., Eds.; Thieme, 2000.

(78) Castle, B. R.; Matteson, S. D. A Methanetetraboronic Ester. J. Am. Chem. Soc.

1968, 90, 2194.

(79) Furue, Masaoki, Matteson, D. S. Observations on the Reactivity of

Tris(tetramethylethylenedioxyboryl)methide Ion. J. Organomet. Chem. 1974, 69,

63–67.

(80) Pierson, H. O. Handbook of Refractory Carbides and Nitrides; Noyes Publications,

1996.

(81) Schwan, J.; Ulrich, S.; Batori, V.; Ehrhardt, H.; Silva, S. R. P. Raman

Spectroscopy on Amorphous Carbon Films. J. Appl. Phys. 1996, 80 (1), 440–447.

(82) Hu, C. Q.; Zheng, B.; Zhu, J. Q.; Han, J. C.; Zheng, W. T.; Guo, L. F. Increasing

sp3 Hybridized Carbon Atoms in Germanium Carbide Films By Increasing the

Argon Ion Energy and Germanium Content. J. Phys. D. Appl. Phys. 2010, 43 (13),

135103.

(83) Properties of Silicon Germanium and SiGe: Carbon; E, K., K, L., Eds.; IEE, 2000.

(84) Gazicki, M.; Szymanowski, H.; Tyczkowski, J.; Malinovsky, L.; Schalko, J.;

Fallmann, W. Chemical Bonding in Thin Ge/C Films Deposited from

Tetraethylgermanium in an R.f. Glow Discharge-an FTIR Study. Thin Solid Films

1995, 256 (1–2), 31–38.

(85) Jamali, H.; Mozafarinia, R.; Eshaghi, A. Evaluation of Chemical and Structural

Properties of Germanium-Carbon Coatings Deposited by Plasma Enhanced

Chemical Vapor Deposition. J. Alloys Compd. 2015, No. 646, 360–367.

(86) Weber, W. H.; Yang, B. K.; Krishnamurthy, M. The Ge-C Local Mode in

Epitaxial GeC and Ge-Rich GeSiC Alloys. Appl. Phys. Lett. 1998, 73 (5), 626–

628.

(87) White, S. B.; McKenzie, D. R. An Electron Diffraction Study of Amorphous

Hydrogenated Germanium-Carbon Thin Films. J. Appl. Phys. 1990, 68 (7), 3194–

3197.

(88) Vilcarromero, J.; Marques, F. C. XPS Study of the Chemical Bonding in

Hydrogenated Amorphous Germanium-Carbon Alloys. Appl. Phys. a-Materials

Sci. Process. 2000, 70 (5), 581–585.

(89) Han, J.; Jiang, C.; Zhu, J. Non-Hydrogenated Amorphous Germanium Carbide

with Adjustable Microstructure and Properties: A Potential Anti-Reflection and

Protective Coating for Infrared Windows. Surf. Interface Anal. 2013, 45 (3), 685–

690.

(90) Gupta, N.; Veettil, B. P.; Xia, H.; Karuturi, S. K.; Conibeer, G.; Shrestha, S.

Synthesis of Nano-Crystalline Germanium Carbide Using Radio Frequency

Magnetron Sputtering. Thin Solid Films 2015, 592, 162–166.

(91) Jackson, S. T.; Nuzzo, R. G. Determining Hybridization Differences for

Amorphous-Carbon From the XPS C 1s Envelope. Appl. Surf. Sci. 1995, 90 (2),

195–203.

(92) Mérel, P.; Tabbal, M.; Chaker, M.; Moisa, S.; Margot, J. Direct Evaluation of the

sp3 Content in Diamond-like Carbon Films by XPS. Appl. Surf. Sci. 1998, 136 (1–

2), 105–110.

(93) Lippincott, E. R.; Vanvalkenburg, A.; Weir, C. E.; Bunting, E. N. Infrared Studies

on Polymorphs of Silicon Dioxide and Germanium Dioxide. J. Res. Natl. Bur.

Stand. (1934). 1958, 61 (1), 61.

(94) Micoulaut, M.; Cormier, L.; Henderson, G. S. The Structure of Amorphous,

Crystalline and Liquid GeO2. J. Phys. Condens. Matter 2006, 18 (45), R753–R784.

(95) Peralta, J.; Gutiérrez, G.; Rogan, J. Structural and Vibrational Properties of

Amorphous GeO2 : A Molecular Dynamics Study. J. Phys. Condens. Matter 2008,

20 (14), 145215.

(96) Davis, R. A.; Hagelee, L. A.; Matteson, D. S. A Bromomethanetriboromc Ester. J.

Organomet. Chem. 1974, 69, 45–51.

(97) Matteson, D. S.; Hagelee, L. A. Characterization of a Boron-Stabilized Carbanion,

Tris (Trimethylenedioxyboryl)methid Ion, and Its Reactions with Carbonyl

Compounds. J. Organomet. Chem. 1975, 93, 21–32.

Chapter 2 Dimensionally Reduced TiS2 Phases

2.1 Introduction

1Similar to how carbon can be sculpted into low-dimensional allotropes such as fullerenes,48 nanotubes,49 graphene,50 and graphene nanoribbons,51 there is an emerging body of work suggesting that the framework connectivity of atoms for any crystalline solid can be ligand-terminated along speciﬁc axes to create stable, crystalline van der

Waals solids comprised of single- or few-atom thick fragments.52,53,54,13 Indeed, dimensionally reduced solids have been discovered from the perovskite lattice,55,56 zinc blende/wurtzite lattice,57 two-dimensional (2-D) metal chalcogenide lattice,21,22 and layered iron selenide lattice.17 These dimensionally reduced phases are thermodynamically stable because, for a set of metal cations and anions with specific oxidation states, there is some inherent stability for a specific type of polyhedral shape and connectivity.13 For example, in ZnS and numerous dimensionally reduced ZnS derivatives, Zn exhibits tetrahedral coordination and corner- sharing connectivity partly because the Zn2+ and S2- oxidation states, radius ratio, and electronegativity difference

1 Reproduced with permission from Morasse, R. A. L., Li, T., Baum, Z. J., Goldberger, J. E., Rational Synthesis of Dimensionally Reduced TiS2 Phases. Chem. Mater., 2014, 26, 4776-4780. Copyright 2014 American Chemical Society. 47 remain unchanged.57 Dimensionally reduced systems have optical, electronic, and magnetic behavior diﬀerent from those of the parent lattice,21,17 which makes them potentially interesting in phosphors and battery electrodes.58,23 These properties can be further tuned by changing the dimensionality, changing the coordinating ligand, and intercalating metals into the van der Waals gap.

In contrast to the solution-phase growth of colloidal semiconducting nanocrystals, which has experienced more than 30 years of scientiﬁc exploration,59,11,60,61,62,63 the synthesis of dimensionally reduced solids is still in its infancy. While one can envision the existence of diﬀerent ligand-terminated, dimensionally reduced structures from a given lattice framework,13 there are very few formalized strategies for synthesizing these phases. The most common approaches deliver molecular precursors of the metal and anion in the appropriate oxidation states in the presence of organic terminating ligands.21,15 Here a large parameter space exists, including ligand choice, ligand-to-metal ratio, and reaction temperature. It is unclear how these parameters can be tuned to access the dimensionally reduced phases in favor of the parent material.

Here we develop a systematic understanding of the relative eﬀect of the ligand-to-metal ratio, reaction temperature, and ligand choice on the synthesis of dimensionally reduced TiS2(en) (en = ethylenediamine) as a model system. We demonstrate that TiS2(en) is preferred over the 2-D layered metal dichalcogenide

(LMDC) framework at a high ligand-to-metal ratio across a wide range of temperatures.

The organic ligand plays a profound role in determining the thermodynamic stability of the structure. Changing the ligand from en to propylenediamine (pn) signiﬁcantly

48 narrows the temperature window in which the dimensionally reduced product is favored.

This change from en to pn, however, only slightly perturbs the electronic structure and increases the band gap from 1.71 to 1.75 eV.

2.2 Experimental Section

2.2.1 Preparation of TiS2(en) and TiS2(pn)

The chemicals used in these syntheses are as listed. Titanium tetrachloride

(TiCl4, 99.9 %, Aldrich), ethylenediamine (en, 99 %, Acros), and propylenediamine (pn,

99 %, Acros) were purchased and used without further puriﬁcation. Elemental sulfur (S,

Mallinckrodt Chemicals) was evacuated in a Schlenk line for 3 h at room temperature prior to synthesis. Dried iso- propanol (i-PrOH, 99.5 %, Aldrich) and chlorobenzene

(C6H5Cl, 99.8 %, Aldrich) were prepared by a standard distillation procedure in the presence of calcium hydride (CaH2, 95 %, Aldrich). The dried solvents were purged with

N2 for 1 h prior to puriﬁcation.

The synthesis of TiS2(en) and TiS2(pn) is as follows. All synthetic procedures were conducted in dried N2 using standard Schlenk line techniques or in a glovebox. In a typical reaction of TiS2(en), 0.3 g of sulfur powder was dissolved in 10 mL of en, which quickly turned into a dark red solution; 0.5 mL of TiCl4 was diluted in 2 mL of dried chlorobenzene, and was immediately injected into the sulfur solution (see Reaction 1 and

Figure 11). The solution was transferred to a 20 mL Teﬂon cup in a Parr reactor in a dry

N2 atmosphere. The reactor was sealed and annealed in a vacuum oven under evacuation at 220 °C for 5 days. TiS2(pn) was synthesized following the same procedure, but propylenediamine was used instead of ethylenediamine. In both the TiS2(en) and

TiS2(pn) systems, the organic ligand-to-Ti ratio and annealing temperatures were varied to obtain diﬀerent phases and construct the synthetic phase space diagrams.

2 S + excess en + TiCl4 → TiS2(en) + 2 en•2HCl

Reaction 1 TiCl4 and S powder in en to form TiS2(en)

The purification of TiS2(en) and TiS2(pn) is as follows. The crude, gel-like products were transferred from the Parr reactor to a centrifuge tube in a glovebag filled with dry N2. Approximately 30 mL of dried i-PrOH was added to recrystallize the product and wash out the salt byproduct (en·2HCl or pn·2HCl). The upper supernatant was removed after centrifugation at 2000 rpm for 10 min using a Thermo Fisher CL2 centrifuge. This procedure was repeated at least five times. The purified final product was evacuated in a Schlenk line for at least 3 h to remove residual solvent and then stored in a glovebox.

2.2.2 Physical Characterizations

X-ray diﬀraction patterns were collected on powder samples of TiS2(en) and

TiS2(pn) using a Rigaku Miniﬂex powder diﬀractometer (30 kV, 15 mA, Cu X-ray tube,

293 K). Diﬀraction data used in the Rietveld reﬁnement were collected using a Bruker

D8 powder diffractometer (40 kV, 50 mA, sealed Cu X-ray tube). The powder sample was sealed in a 0.8 mm outer diameter glass capillary tube for measurement. Rietveld analysis was performed using the GSAS software package. The structure was determined without refining the H atom positions. Diffuse reflectance absorption spectra were recorded using a Lambda 950 spectrophotometer (PerkinElmer) with a 60 mm integrating sphere. Spectra were collected with 1 nm resolution.

2.2.3 Density Functional Theory Calculations

Electronic band structure calculations were performed using density functional theory as implemented in the WIEN2k package.64 H positions in each rhombohedral lattice were estimated using structural information from Rietveld reﬁnement of the powder X-ray diﬀraction data. The structures were then optimized with WIEN2k’s

“MSR1a” mixing scheme, which simultaneously converges charge densities and atomic positions. Electronic band diagrams were generated by self-consistent calculation of the

Kohn−Sham orbital energies on a 6 × 6 × 6 Monkhorst pack in the irreducible wedge of

52 the Brillouin zone. The modiﬁed Becke−Johnson exchange-correlation potential was selected on the basis of its previously reported accuracy for determining semiconductor band gaps.65,66,67 Each calculation was iterated until the total energy converged within

0.0001 Ry and the charge converged within 0.00005 eV.

2.3 Crystal Structures of TiS2(en) and TiS2(pn)

At least two diﬀerent phases of TiS2 amine derivatives exist, including ones with

2-D68,69 and one-dimensional (1-D)21 octahedral connectivity (Figure 12 a, c, e). It has been well documented that organic amines can be intercalated into the 2-D TiS2 lattice

68,69 (Figure 12 e). Depending on the amount of intercalant present, intercalation of TiS2 with ethylenediamine has generated crystalline products with a 3R′68 and 1T69 polytype.

70 The short chain (C1−C4) amine molecules lie parallel with the TiS2 layer.

In addition, we have recently created TiS2(en), a single-octahedron-thick 1-D derivative (Figure 12 a, c). This structure consists of edge-sharing Ti octahedra that run parallel to the c-axis. Every Ti atom is coordinated by 4 S atoms. Neighboring octahedra are connected by two bridging S atoms. The Ti atom is also coordinated to a cis- binding bidentate ligand, en. These chains are held together by van der Waals forces and are packed into a hexagonal arrangement. This phase adopts an R3̅c unit cell with 3-fold helicity, and the unit cell parameters are listed in Table 1. Here we demonstrate that a similar dimensionally reduced 1-D phase can be created with pn (Figure 12 b, d). This

TiS2(pn) phase also crystallizes into an R3̅c unit cell with an expanded a-axis and a

53 slightly contracted c-axis. The average Ti−S bond length is 2.360 Å, which is slightly greater than the length of 2.336 Å in TiS2(en) (Table 2). Each TiS2 product phase (2-D intercalated or 1-D hybrid) is stable in air for approximately one month.

Figure 12 Crystal structures of (a) TiS2(en) and (b) TiS2(pn) projected down the c-axis. Ti octahedra are colored blue. (c) TiS2(en) and (d) TiS2(pn) chains along the c-axis. (e) en-intercalated TiS2 layers. Atom colors: Ti, blue; S, yellow; N, green; C, black. H atoms are omitted for clarity.

Unit Cell Parameter a (Å) c (Å)

TiS2(en) 18.5964(15) 9.015(5)

TiS2(pn) 20.3062(11) 8.909(5)

Table 1 Unit Cell Parameters of TiS2(en) and TiS2(pn).

Ti-S1 (Å) Ti-S2 (Å) Average Ti-S (Å) Ti-N (Å)

TiS2(en) 2.308 2.364 2.336 2.277

TiS2(pn) 2.359 2.360 2.360 2.280

Table 2 Representative Bond Lengths in TiS2(en) and TiS2(pn).

2.4 Synthetic Phase Diagrams of the Titanium Sulfide Systems

Considering that both 2-D and 1-D TiS2 frameworks exist, we wanted to develop a roadmap to synthetically access each phase by using en as the coordinating ligand. In

4+ 2− this approach, we delivered soluble molecular precursors of Ti via TiCl4 and S via a combination of S and organic diamine. It is well established that amines act as sacrificial agents to reduce the S to S2−.71 Considering the Ti: 2S stoichiometry, 2 equivalents of a sacrificial organic amine are necessary to fully reduce the S. The excess organic diamine directly chelates the titanium, which inhibits the formation of the original 2-D framework and thus creates the dimensionally reduced solid. We varied the temperature and ligand-to-metal ratio of the reaction and identified the product phases by using powder X- ray diffraction (XRD). The resulting synthetic phase-space diagram is illustrated in

Figure 13 a. Representative powder X-ray diﬀraction patterns of each product phase are found in Figure 13 b. For a wide range of temperatures (80 to 220 °C) and at a high en: Ti ratio, the 1-D TiS2(en) dimensionally reduced product phase is favored. The

TiS2(en) hybrid phase is essentially phase pure between 80 and 220 °C for ligand-to- metal ratios ranging from 30:1 to <10:1. The 30:1 en:Ti 1D hybrid product was truncated from Figure 13 a to emphasize the low-ratio region of phase space. At 220 °C, the boundary between the 1-D and intercalated phases occurs near a ligand-to-metal ratio of

6.5:1. This mixed-phase region, as seen in the orange XRD pattern of Figure 13 b, is comprised of both the 1-D TiS2(en) phase and the 2-D intercalated TiS2 phase. The ﬁrst

57 peak at 9.59° 2θ is the (110) reﬂection of 1-D TiS2(en) and is asymmetrical due to a weak shoulder at 9.21° 2θ that corresponds to the (003) reﬂection of the 2-D intercalated

68 TiS2. In addition, the weak, broad reﬂection centered at 18.49° 2θ corresponds to the

(006) reﬂection of the 2-D intercalated TiS2. At 220 °C, with a ligand-to-metal ratio of

<5, the pure 2-D intercalated TiS2 phase is produced. At 120 °C, the 1-D TiS2(en) phase is stable at ligand-to-metal ratios of 30:1 to 3:1, and an amorphous phase emerges when the ratio is reduced further. The boundary between the 2-D intercalated and amorphous phases occurs at 155 °C and does not pass through the 1-D TiS2(en) phase.

As pure 1-D TiS2(en) exists in a large temperature and ligand-to-metal ratio phase space, we explored how the crystallinity changed as a function of these parameters.

Figure 14 a depicts the XRD spectra when the sample was synthesized at 220 °C while the ligand-to-metal ratio was decreased from 30:1 to 10:1. At this temperature, the full width at half maximum (FWHM) of the (110) peak varies from 0.229° 2θ to 0.315° 2θ.

Figure 14 b depicts the XRD spectra when the sample is synthesized at a ligand-to-metal ratio of 20:1 while the temperature is decreased from 220 to 100 °C. A highly crystalline product can be obtained despite a change in synthesis temperature of >100 °C. FWHM values of 0.206, 0.229, 0.195, and 0.200° 2θ were calculated for the ﬁrst peak (110) of

XRD spectra at 220, 180, 140, and 100 °C, respectively. Thus, from the Debye-Scherrer equation, the crystalline domain size of the TiS2(en) product is relatively uniform across several temperatures.

As 1-D TiS2(en) can be created over a wide range of synthetic conditions, another question we addressed is whether similar structures comprised of diﬀerent ligands are

58 synthetically accessible over an equally wide parameter space. Indeed, we were able to create a similar 1-D TiS2 structure using pn as the organic ligand. An R3̅c unit cell was again observed with an expansion in the a-axis and slight contraction in the c-axis

(Table 1). The reaction temperature and pn:Ti ratio were varied to construct a synthetic phase space diagram (Figure 15 a). Again, we observed four distinctive regions representing diﬀerent phases: 2D intercalated TiS2, a mixture of 2D intercalated TiS2 and

1-D TiS2(pn), pure 1-D TiS2(pn), and the amorphous product (Figure 15 b). At high reaction temperatures, the intercalated product was predominant and stable for a wide temperature range. As the temperature was decreased to 140 °C, a mixture of the 2-D and 1-D phases was obtained. The pure 1-D TiS2(pn) phase can be isolated only between

120 and 130 °C. Below 120 °C, no crystalline phase can be identiﬁed. Varying the pn:Ti ratio from 30:1 to 15:1 does not change the phase that is formed at a speciﬁc temperature.

Despite the structural similarity of 1-D TiS2(pn) and TiS2(en), the synthetic conditions for obtaining TiS2(pn) are limited to a very narrow temperature window. This suggests that even though the diﬀerence in the organic ligand is small, it plays a profound role in determining the stability of the dimensionally reduced product.

Here we choose to denote Figure 13 a and Figure 15 a as “synthetic phase space diagrams” instead of phase diagrams. A true phase diagram implies that diﬀerent phases can be reversibly interconverted by changing the temperature and ligand-to-metal ratio of the system. To probe whether these phases are at kinetically trapped states or at thermodynamic equilibrium, we performed two control experiments to try to interconvert the phases. First, annealing a sample of pure 2-D intercalated TiS2 with an additional

20 equivalents of en at 180 °C for 1 day did indeed partially convert it to hybrid 1-D

TiS2(en) and decrease the peak intensity of the 2-D starting material, as seen in the powder XRD pattern of Figure 16.

In the second control experiment, 1 equivalent of en was added to 1-D TiS2(en) and annealed at 180 °C for 1 day. Despite that fact that it exists in this region of the synthetic phase space diagram, 2-D intercalated TiS2 was not formed, as shown in the

XRD patterns in Figure 17. However, the minimal amount of en solvent in this experiment likely means that transformation into a crystalline, 2-D intercalated TiS2 phase would take an incredibly long time. Therefore, although these phases can be partially interconverted between each other by modifying the temperature and ligand-to-metal ratio, these synthetic phase space diagrams are not true thermodynamic phase diagrams.

Together, both of these synthetic phase space diagrams illustrate numerous principles about the relative thermodynamics of a dimensionally reduced material and the original framework. At high temperatures, the original 2-D TiS2 framework is thermodynamically preferred partly because of the increased entropy associated with the loss of a ligand. This eﬀect is most pronounced in the TiS2(pn) system but is also apparent in the TiS2(en) system when the en:Ti ratio is small (<5:1). In this case, the coordinating environment created by en cannot inhibit the formation of the parent lattice

TiS2; therefore, the 2-D intercalated phase is formed. Furthermore, in the low-temperature region for pn and the low-equivalent−low-temperature region for en, the temperature is not high enough to overcome the kinetically trapped amorphous product.

Finally, these experiments show that the nature of the organic ligand plays a signiﬁcant role in determining the relative stability of the dimensionally reduced product. The 1-D

TiS2(en) product can be readily synthesized across a temperature range much larger than that of the 1-D TiS2(pn) product.

Figure 13 (a) Synthetic phase space diagram of the TiS2(en) system: red, intercalated TiS2; orange, mixed phases; green, TiS2(en); blue, amorphous. (b) Representative powder X-ray diffraction patterns of each phase.

Figure 15 (a) Synthetic phase space diagram of the TiS2(pn) system: red, intercalated TiS2; orange, mixed phases; green, TiS2(pn); blue, amorphous. (b) Representative powder X-ray diffraction patterns of each phase.

Figure 16 2-D intercalated TiS2 XRD patterns before (black, bottom) and after (red, top) annealing with an additional 20 equivalents of en at 180 ˚C for 1 day.

Figure 17 1-D TiS2(en) XRD patterns before (black, bottom) and after (red, top) annealing with an additional equivalent of en at 180 ˚C for 1 day.

2.5 Electronic Band Structures of the 1-D TiS2 Analogues

The optical properties of TiS2(en) and TiS2(pn) were investigated using diﬀuse reﬂectance absorption (DRA) measurements (Figure 18). Both of these jet-black, 1-D materials exhibit broad absorption in the visible light range with an onset at near-infrared

(NIR) energies (Figure 19 a). The band gap energies of TiS2(en) and TiS2(pn) were determined to be 1.71 and 1.75 eV, respectively, by linearly ﬁtting the absorption edge of the plot of the Kubelka−Munk function [F(R) = (1 − R)2/2R, where R is the measured reﬂectance]. These values correspond to a 1.4 eV increase in band gap energy compared

72 to that of 2-D bulk TiS2, which has an indirect band gap of 0.3 eV. DRA of 2-D intercalated TiS2 was also measured (Figure 18).

This slight increase in band gap energy from TiS2(en) to TiS2(pn) indicates that changing the diamine ligands does not strongly inﬂuence the electronic structure of the material, Figure 20. This is conﬁrmed by simulation of the band structures of both materials using the TB-mBJ exchange-correlation potential as implemented in WIEN2k.

The band structure calculations for each trigonal lattice were performed using the reduced rhombohedral unit cell along the RHL2 Brillouin zone pathway, Figure 21. TiS2(en) and

TiS2(pn). are calculated to have direct band gaps of 1.81 and 1.94 eV at Γ, respectively

(Figure 19 b, d). These values are in excellent agreement (6−11 % larger) with the observed experimental results. This is much more accurate than previous simulations of

21 TiS2(en) employing CASTEP (GGA-WC) which underestimate the band gap by 35 % .

These ﬁndings are consistent with previous reports that the TB-mBJ exchange-correlation potential is considerably more accurate than standard GGA approaches for predicting semiconductor band gaps.67

The partial density of states plots of TiS2(en) and TiS2(pn) (Figure 19 c, e) show that the valence band maximum (VBM) is composed primarily of S 3p orbitals and that the conduction band minimum (CBM) is composed primarily of Ti 3d orbitals. The N 2p orbitals for both materials lie much lower in energy than the VBM (<2.5 eV) and do not strongly contribute to the conduction and valence band edges (Figure 20). Therefore, considering that the polyhedral connectivity of the edge-sharing zigzag inorganic backbone remains unchanged, changing the amine ligand from en to pn has a minimal eﬀect on the band structure. This phenomenon is similar to that found in other previous studies on dimensionally reduced systems using diﬀerent ligands.56,57

Figure 18 DRA of 1-D TiS2(en), 1-D TiS2(pn), and 2-D intercalated TiS2.

Figure 19 (a) Diffuse reflectance spectra of TiS2(en) and TiS2(pn). (b) Calculated band structure of TiS2(en). (c) Partial density of states of Ti 3d (black) and S 3p (red) for TiS2(en). (d) Calculated band structure of TiS2(pn). (e) Partial density of states of Ti 3d (black) and S 3p (red) for TiS2(pn).

Figure 20 Partial density of states of (a) TiS2(en) and (b) TiS2(pn) showing Ti, S, and N contributions.

Figure 21 Brillouin zone and k-space pathway for TiS2(en) represented in the reduced rhombohedral unit cell.

2.6 Conclusion

Using TiS2 as a model system, we have demonstrated that dimensionally reduced phases can be created using solution- phase processes at large ligand-to-metal ratios and lower temperatures. The parent structure is more stable at higher temperatures likely because of the increasing entropy. The temperature window at which the dimensionally reduced phase is produced is ultimately determined by the organic ligand. Even extending the ligand alkyl chain length by a single carbon atom can profoundly narrow the temperature window in which these phases are accessed. However, extending the alkyl chain length does not change the metal-ligand bond characters and therefore will only marginally change the electronic structure of the material. In summary, this systematic understanding of the inﬂuence of ligand structure, reaction temperature, and ligand-to-metal ratio lays the foundation for the rational synthesis of dimensionally reduced phases.