Exploring the Chemical Reactivity of Group 14 and Group 15 Molecular Precursors Himashi Purnima Andaraarachchi Iowa State University

Home , Germane, Nickel compounds, Tin selenide

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2018 Exploring the chemical reactivity of group 14 and group 15 molecular precursors Himashi Purnima Andaraarachchi Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Chemistry Commons

Recommended Citation Andaraarachchi, Himashi Purnima, "Exploring the chemical reactivity of group 14 and group 15 molecular precursors" (2018). Graduate Theses and Dissertations. 17139. https://lib.dr.iastate.edu/etd/17139

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Exploring the chemical reactivity of group 14 and group 15 molecular precursors

Himashi P. Andaraarachchi

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Chemistry

Program of Study Committee: Javier Vela, Major Professor Wenyu Huang Gordie Miller Emily Smith Keith Woo

The student author, whose presentation of the scholarship herein was approved by the program of study committee, is solely responsible for the content of this dissertation. The Graduate College will ensure this dissertation is globally accessible and will not permit alterations after a degree is conferred.

Iowa State University

Ames, Iowa

2018

DEDICATION

To Amma and Thaththa iii

TABLE OF CONTENTS

Page

NOMENCLATURE ...... v

ACKNOWLEDGMENTS ...... vi

ABSTRACT ...... viii

CHAPTER 1. INTRODUCTION ...... 1 Colloidal nanocrystals and their applications ...... 1 Controlling nanocrystalline phase, composition, and morphology...... 3 Molecular programming at nanoscale ...... 4 Thesis organization ...... 7 References...... 8

CHAPTER 2. PHASE-PROGRAMMED NANOFABRICATION: EFFECT OF ORGANOPHOSPHITE PRECURSOR REACTIVITY ON THE EVOLUTION OF NICKEL AND NICKEL PHOSPHIDE NANOCRYSTALS ...... 12 Abstract ...... 12 Introduction ...... 13 Experimental section ...... 15 Results and discussion ...... 18 Conclusions ...... 30 Acknowledgements ...... 33 References...... 34 Appendix of supporting information ...... 43

CHAPTER 3. EVOLUTION OF NICKEL PHOSPHIDE NANOCATALYSTS DURING PHENYLACETYLENE HYDROGENATION ...... 55 Abstract ...... 55 Introduction ...... 56 Experimental section ...... 57 Results and discussion ...... 59 Conclusions ...... 69 Acknowledgements ...... 70 References...... 71 Appendix of supporting information ...... 74

CHAPTER 4. GESN HETEROSTRUCTURES: FROM THEORY TO EXPERIMENT ...... 79 Abstract ...... 80 Introduction ...... 80 Experimental section ...... 82 Results and discussion ...... 85 Conclusions ...... 94 Acknowledgements ...... 95 References...... 95 Appendix of supporting information ...... 101

CHAPTER 5. CONCLUSIONS AND OUTLOOK ...... 104 v

NOMENCLATURE

TEM Transmission Electron Microscopy

PXRD Powder X-Ray Diffraction

EDX Energy Dispersive X-Ray Spectroscopy

XPS X-Ray Photoelectron Spectroscopy

ACKNOWLEDGMENTS

I am very thankful for having a great mentor and friends who were always willing to share their knowledge and advice with me throughout my five years of graduate school at Iowa State University. I would like to thank my advisor, Javier Vela for his guidance, encouragement, patience, and support during the past four years. He has been a great mentor and taught me a lot about science and research. Thank you for showing me how to be a good researcher, how to improve critical thinking and soft skills, and giving me an opportunity to present my research at national conferences.

Next, I would like to thank my committee members, Professors Wenyu Huang,

Gordie Miller, Emily Smith, and Keith Woo for their input and guidance throughout this journey. I would like to thank Prof. Smith and Prof. Stanley for giving me the opportunity to collaborate with them. I enjoyed working with their group members and learnt a lot from them.

Last five years have been wonderful, thanks to my great colleagues in Vela group.

I owe a big thank to Long Men for being the greatest lab partner any one could have asked for. A special thanks to Yijun Guo and Elham Tavasoli for showing me ropes in doing experiments, Michelle Thompson for her kindness, encouraging words, and English lessons, Sam Alvarado, Chia-Cheng Lin, Malinda Reichert, and Feng Zhu for all the great advice. Many thanks to Josie Del Pilar for the guidance in writing prelim proposal, Bryan

Rosales and Arthur White for very insightful discussions, Carena Daniels, Alan Gonzales and Lin Wei for being great lab mates.

In addition, I would like to thank Supipi akki and Rajeeva ayya for being a family away from home, Sagarika akki and Gayan ayya for their support, Uma akki, Chamari akii, vii and Roshan ayya for their support and valuable input to the research, and Kasuni and Dilini for being greatest friends to share many important moments of graduate life together.

Many thanks to my grandparents, immediate family, my cousins, and in-laws for their support and encouragement. More importantly, I want to thank my parents and my sister for their love, support, endless enthusiasm and encouragement throughout all these years. Finally, I want to thank my husband Ravindu for being my side at every step of this journey and for all the love, encouragement, patience and unselfish support. I truly couldn’t have done this without them.

viii

ABSTRACT

Colloidal nanocrystals with their fascinating properties have found many opportunities in a diverse range of applications including optoelectronic and electronic devices, catalysis, and biomedical applications. Performance of nanomaterials in these applications are directly related to the well-defined properties, phase, composition, size, and morphology, of the nanocrystals. Strategies exist to synthesize colloidal nanocrystals with well-defined properties tailored to desired applications by optimization of several parameters such as concentration and nature of precursors and surfactants, temperature, additives, and multi-step procedures, which can be time consuming and expensive. Using a conceptually simple ‘molecular programming’ bottom up approach, this thesis describes an efficient approach for the phase and composition controlled synthesis of pnictide nanomaterials and group IV Ge-Sn heterostructures. Phase and composition control is achieved through fine-tuning chemical reactivity of group 14 and group 15 molecular precursors, while keeping other reaction conditions constant.

Using a family of organophosphites (P(OR)3) with tunable reactivates as phosphorus precursors, we demonstrated that different organophosphite precursors selectively yield nickel phosphides (Ni12P5 and Ni2P) or metallic nickel and that these phases evolve over the time through separate mechanistic pathways. As a direct result of our study, we built a reactivity scale for organophosphite precursors presenting the ease of these precursors react with nickel precursors to form nickel phosphide nanocrystals. We believe that the organophosphite family is a great addition to the synthetic tool box of pnictides. ix

Next, fostering our efforts to a controllable synthesis of more complex architectures, we synthesized Ge-Sn heterostructures with different compositions using

’ Ge-Sn molecular precursors (R3GeSnR 3). Sn/Ge core/shell structures with varied shell thickness was observed by varying the Ge-Sn bond strengths of precursors that is tunable through the nature of substituents. This thesis also reports the synthesis of metastable b-Ge in ambient pressure conditions, which otherwise existed and reported in high pressure conditions. These group 14 molecular precursors open up new directions for heterostructure synthesis with unique morphologies that offer interesting properties.

As an attempt to discover how different properties of nanocrystals affect potential applications, we explored how different morphologies of nickel phosphide nanocrystals would affect the catalytic properties of the alkyne hydrogenation reaction. Compared to the bimodal (hollow and solid) distribution of Ni2P nanocatalysts, hollow Ni2P nanoparticles are more catalytically active towards the phenylacetylene hydrogenation. These hollow

Ni2P catalysts seems to be robust with the ability of recycled up to eight times without losing its catalytic activity while preserving their structural integrity.

CHAPTER 1.

INTRODUCTION

General introduction

This thesis describes the application of molecular programming approaches to the synthesis of transition metal pnictide nanocrystals and group 14 heterostructures through group

14 and group 15 molecular precursor reactivity. This work details the progress towards phase controlled nickel phosphide nanocrystal formation, with an emphasis on the pnictide precursor reactivity and how it relates to the phase, morphology, and size of the final products.

Furthermore, a systematic study was carried out to explore how different morphologies of nickel phosphide nanocrystals affect alkyne hydrogenation catalytic reactivity. In addition, this thesis presents work towards the synthesis of Ge-Sn heterostructures, with an emphasis on the

Ge-Sn single source precursor reactivity and how it relates to the composition and morphology of the product.

Colloidal nanocrystals and their applications

Colloidal nanocrystals are small inorganic crystals of metals, semiconductors, and magnetic materials comprising hundreds to a few thousands atoms with particle dimensions ranging from 1-100 nm. The ability to tailor the nanoscale dimensional regime causes the electronic structure, optical, magnetic, and catalytic properties of these materials to be vastly differed from the bulk counterpart.1,2,3 Instances include superparamgnetism of magnetic nanocrystals,4 surface plasmon resonances in metallic nanocrystals,5 and band gap engineering of semiconductor nanocrystals.6,7 Their superior and fascinating properties along with inherent robustness creates many opportunities in optoelectronic and electronic device application,8,9 catalysis,10 and biomedical applications.11,12 2

Formation of thermodynamically stable and easy to handle colloidal nanocrystals with desired properties preferably fit the requisites of solution processed electronic and optoelectronic devices. In addition, the ability of solution grown nanocrystals to form ordered nanostructures through self-assembly known as super lattices or nanocrystal solids make them great candidates for electronics.13 Quantum tunable optical properties of semiconductor nanocrystals leads to wide range of applications including photovoltaics,14 light emitting devices,15 luminescent tags,16 and lasers.17 Harnessing superior fluorescent and magnetic properties, nanocrystals can be used as fluorescence probes in bio-imaging,16 as an efficient diagnostic tool in magnetic resonance imaging,18 and as magnetic separation of biological targets.19 Furthermore, excellent catalytic properties of nanocrystals stem from nanoscale as well as composition,20 surface environment,21 and surface properties. They are employed in many reactions including chemo selective oxidations and reductions, asymmetric hydrogenations, coupling reactions, C–H activations, and oxidative aminations.22-24

Nanocatalysts with carefully tuned surfaces and interfaces are being designed with controllable properties to meet the requirements of enhanced reactivity, selectivity, and recyclability.

These wide ranges of applications benefitted from the controlled synthesis of colloidal nanocrystals with well-defined attributes.25 Even though great strides have been made towards synthesizing nanocrystals with uniform size, well-defined morphology, and unique composition, alteration of various synthetic conditions and optimization of number of reaction parameters are required to achieve the desired properties.

Controlling nanocrystalline phase, composition and morphology

Optimization of devices, catalytic properties, and other application containing colloidal nanocrystals require the utilization of particles with specific and well-defined sizes, composition, and morphologies. A critical aspect as well as a major challenge of bottom-up nanoparticle synthesis is to achieve precise and predictable control of these properties.

Phase and composition control of colloidal nanocrystals are inherently important as different phases of the same material offer different properties, a-tin is a semiconductor while b-tin is a metal, and different compositions provide significantly distinctive properties.1,2

Phase and composition control of colloidal nanocrystals have been realized through alteration of many reaction parameters. In the case of nickel phosphides, phase and composition were tuned by Ni to P precursor ratio,26 counter anions of metal precursors,27 different surfactants

28 29 30 31 and concentration, reaction solvents, time, and temperature. Composition of Co2-xNixP was controlled using the varying amounts of Co and Ni precursors.32 Nickel sulfide phase and composition control was achieved through temperature, concentration of precursors, and additives.33,34

Colloidal nanocrystals with different morphologies offer unique properties. Inorganic nanocrystals existed as hollow or solid particles in 0D (spherical, dodecahedral, tetrahedral, and cubic), 1D (nanotubes, nanoneedles, nanorods or nanowires), 2D (nano sheets, nano plates, and belts), and 3D (nano urchins, nano flowers, nanostars, and other complex morphologies) structures.35 Control of morphology is governed by many factors including thea concentration of precursors, pH effect, solvent, temperature, seeds and templates, and the nature and concentration of surfactants and additives.36-39 4

Despite the above general parameters listed for the controllable synthesis of colloidal nanocrystals with well-defined phases, compositions, and morphologies, colloidal synthesis is complicated and easily affected by various factors resulting in unpredictable outcomes for the final products. Thus, instead of relying on a mix of apparently irreproducible or hard to reproduce parameters, alternative route for more precise control of colloidal nanocrystal properties are required.

Molecular programming at nanoscale

In order to obviate the time consuming process of optimizing several reactions for colloidal nanocrystal synthesis with controlled properties, we and a few other groups have introduced a simple but an efficient and powerful approach to precisely control the outcome of the final products through fine tuning the chemical and electronic structure of the molecular precursors.40-44

According to Hammond’s postulate,45 transition state energy of the nucleation step, which is the rate determining step of nanocrystal formation, is closer to the molecular precursors used in the synthesis than to the final nanocrystalline products. This infers that the transition state of nanocrystal formation closely resembles the initial molecular precursors more than the nanocrystalline product. This forms the basis of our group’s work on molecular programming in which the chemical reactivity of molecular precursors can be tuned by different substituents that allows us to control the relative rates of nucleation and consequently controls the size, morphology, composition, and phase of different colloidal nanocrystals.

Utilizing this conceptually simple ‘bottom up’ synthesis approach, formation rates of nanocrystalline phases can be fine-tuned and ultimately controlling the final result of nanocrystalline products with well-defined properties. Another advantage of our molecular 5 programming approach is that we are only altering the molecular precursor while keeping all the other reaction conditions constant. This approach simply eliminates optimization of numerous reaction parameters, which can be time-consuming and expensive.

Using a tunable library of phosphine chalcogenide precursors (R3PE, where R= alkyl or aryl group and E=S or Se), we demonstrated that we are able to control the length-to- diameter aspect ratio of CdS and CdSe nanorods. As the electron donating ability of P substituents increases, the formal charge of phosphorus center stabilizes, thus increasing the strength of the P=E bond. A stronger P=E bond energy decreases the precursor reactivity resulting in an increase in nanorod length and aspect ratio. As the nanorod diameter is similar for all the molecular precursors, we assume that they proceed through a seed mediated growth mechanism. Upon hot injection of phosphine chalcogenide precursors, more reactive precursors such as diphenylpropyl sulfide form more nuclei compared to the less reactive precursors. Following this fast nucleation, less reactive chalcogenide precursors left in solution continues to grow epitaxially forming nanorods in higher aspect ratios.

In addition, the Vela group exhibited shape-programmed nanocrystal formation through a series of tunable disubstituted chalcogenide precursors (R-E-E-R, where R= alkyl or 6 aryl group, E= S or Se). We observed different morphologies, spheres, rods, or tetrapods, depending on the chalcogenide precursor used. For example, dichalcogenide precursors with weak C-E bonds are very reactive with fast nucleation, leading to the formation of large CdS spherical nanocrystals. However, precursors with intermediate strength C-E bonds lead to the formation of small CdS quantum dots. Dichalcogenide precursors with stronger C-E bonds react very slowly and produce anisotropic CdS nanostructures (tetrapods). While larger CdS spheres originates from the nucleation seeds that are available from a rapid increase of chalcogenide radicals (R-S•), dichalcogenide precursors with intermediate C-E bond strength and E-E bond strength are found to hinder the reaction kinetics and produce smaller CdS quantum dots due to the efficient surface passivation from the in-situ generated thiol radicals.

Furthermore, these thiol radicals can increase the life time of the small CdS nuclei long enough to allow for the slow and selective epitaxial growth of the anisotropic CdS nanocrystals resulting in a growth of tetrapods.

Fostering our efforts in molecular programming approach, this thesis reports the synthesis of pnictide nanocrystals using a tunable family of organophosphite (P(OR)3) 7 precursors and synthesis of complex architectures, Ge-Sn heterostructures, using tunable family of Ge-Sn molecular precursors (R3GeSnR3’) through fine-tuning the P-O and Ge-Sn bond strengths. In addition, as an attempt to explore how nanocrystal properties would affect in potential applications, we conducted a systematic study on how morphology and size of nickel phosphide nanocrystals affect the hydrogenation catalytic properties.

Thesis Organization

Chapter 2 of this thesis describes a facile synthesis of nickel phosphide and metallic nickel nanocrystals via a tunable family of organophosphite precursors. This chapter demonstrates that the nature of the phosphite precursor has a dramatic effect on the phase, and morphology of the resulting nanocrystals. Experimental results revealed that the evolution of nickel phosphide and nickel nanocrystals occur via two separate mechanistic pathways. Under the reaction conditions we studied, a relative phosphite precursor reactivity scale was generated according to the ease of forming nickel phosphide nanocrystals. Overall, the rate of formation of nickel phosphide nanocrystals increases in the order P(OMe)3 < P(OEt)3 < P(OnBu)3 <

P(OCH2tBu)3 < P(OiPr)3 < P(OPh)3.

Chapter 3 of this thesis discusses the investigation of how morphology and size of the nickel phosphide nanocrystals affect the catalytic properties of alkyne hydrogenation reaction.

A systematic study was conducted on phenylacetylene hydrogenation reaction with two morphologies of Ni2P nanocatalysts derived from organophosphite precursors. Our results revealed that small hollow Ni2P nanocrystals are more catalytically active than a bimodal mixture of Ni2P nanocrystals (both hollow and solid). Catalytic activity increases over the reaction cycles for both morphologies and these nanocatalysts can be recycled up to 8 times without losing their catalytic activity. 8

Chapter 4 of this thesis describes our attempt to expand our molecular programming approach to synthesize complex nano architectures; Ge-Sn heterostructures. This chapter demonstrates that Ge-Sn single source precursors with varied Ge-Sn bonds affect the

’ composition of Ge-Sn heterostructures. These tunable molecular precursors (R3GeSnR 3, R= alkyl or aryl) yields a Sn/Ge core-shell like structure with varying shell thicknesses. A single source precursor with the weakest Ge-Sn bond yield Sn/Ge core/shell with the thickest shell and the precursor with the strongest Ge-Sn bond yields b-Sn nanocrystals. Our observations revealed formation of b-Ge in ambient pressure conditions, which was previously reported to form under high pressure conditions. Chapter 5 summarizes the previous chapters and presents the future outlook of this work.

References

1 Talapin, D. V.; Lee, J.-S.; Kovalenko, M. V.; Shevchenko, E. V. Prospects of Colloidal Nanocrystals for Electronic and Optoelectronic Applications. Chem. Rev. 2010, 110, 389–458.

2 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, 545–610.

3 Smith, A. M.; Nie, S. Semiconductor Nanocrystals: Structure, Properties, and Band Gap Engineering. Acc. Chem.Res. 2010, 43, 190–200.

4 Link, S.; El-Sayed, M. A. Spectral Properties and Relaxation Dynamics of Surface Plasmon Electronic Oscillations in Gold and Silver Nanodots and Nanorods. J. Phys. Chem. B 1999, 8410–8426.

5 Amendola, V.; Pilot, R.; Frasconi, M.; Maragò, O. M.; Iatì, M. A. Surface Plasmon Resonance in Gold Nanoparticles: A Review. J. Phys. Condens. Matter 2017, 29, 203002.

6 Alivisatos, A. P. Semiconductor Clusters, Nanocrystals, and Quantum Dots. Science 1996, 271, 933–937.

7 Kamat, P. V. Quantum Dot Solar Cells. Semiconductor Nanocrystals as Light Harvesters. J. Phys. Chem. C 2008, 112, 18737–18753.

8 Pietryga, J. M.; Park, Y. S.; Lim, J. H.; Fidler, A. F.; Bae, W. K.; Brovelli, S.; Klimov, V. I. Spectroscopic and Device Aspects of Nanocrystal Quantum Dots. Chem. Rev. 2016, 116, 10513–10622.

9 Jang, E.; Jun, S.; Jang, H.; Lim, J.; Kim, B.; Kim, Y. White-Light-Emitting Diodes with Quantum Dot Color Converters for Display Backlights Adv. Mater. 2010, 22, 3076–3080.

10 Wu, Y.; Wang, D.; Li, Y. Nanocrystals from Solutions: Catalysts. Chem. Soc. Rev. 2014, 43, 2112–2124.

11 Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Semiconductor Nanocrystals as Fluorescent Biological Labels. Science 1998, 281, 2013–2016.

12 Jun, Y. W.; Seo, J. W.; Cheon, J. Nanoscaling Laws of Magnetic Nanoparticles and Their Applicabilities in Biomedical Sciences. Acc. Chem. Res. 2008, 41, 179–189.

13 Boles, M. A.; Engel, M.; Talapin, D. V. Self-Assembly of Colloidal Nanocrystals: From Intricate Structures to Functional Materials. Chem. Rev. 2016, 116, 11220–11289.

14 Chistyakov, A. A.; Zvaigzne, M. A.; Nikitenko, V. R.; Tameev, A. R.; Martynov, I. L.; Prezhdo, O. V. Optoelectronic Properties of Semiconductor Quantum Dot Solids for Photovoltaic Applications. J. Phys. Chem. Lett. 2017, 8, 4129–4139.

15 Klimov, V. I.; Mikhailovsky, A. A.; Xu, S.; Malko, A.; Hollingsworth, J. A.; Leatherdale, C. A.; Eisler, H. J.; Bawendi, M. G. Optical gain and stimulated emission in nanocrystal quantum dot. Science 2000, 290, 314–317.

16 Michalet, X.; Pinaud, F.; Lacoste, T. D.; Dahan, M.; Bruchez, M. P.; Alivisatos, A. P.; Weiss, S. Properties of Fluorescent Semiconductor Nanocrystals and their Application to Biological Labeling. Single Mol. 2001, 4, 261–276.

17 Klimov, V. I.; Ivanov, S. A.; Nanda, J.; Achermann, M.; Bezel, I.; McGuire, J. A.; Piryatinski, A. Single Exciton Optical Gain in Semiconductor Nanocrystals. Nature 2007, 447, 441–446.

18 Jun, Y. W.; Seo, J. W.; Cheon, A. Nanoscaling Laws of Magnetic Nanoparticles and Their Applicabilities in Biomedical Sciences. Acc. Chem. Res. 2008, 41, 179–189.

19 Jain, P. K.; Huang, X.; El-Sayed, I. H.; El-Sayad, M. A. Review of Some Interesting Surface Plasmon Resonance-Enhanced Properties of Noble Metal Nanoparticles and Their Applications to Biosystems. Plasmonics 2007, 2, 107–118.

20 Valden, M.; Lai, X.; Goodman, D. W. Onset of Catalytic Activity of Gold Clusters on Titania with the Appearance of Nonmetallic Properties. Science 1998, 281, 1647−1650.

21 Wang, X.; Peng, Q.; Li, Y. Interface-Mediated Growth of Monodispersed Nanostructures.Acc. Chem. Res.2007, 40, 635–643.

22 Ying, J. Y. Design and Synthesis of Nanostructured Catalysts. Chem. Eng. Sci. 2006, 61, 1540–1548.

23 Ying, J. Y. Nanostructural Tailoring: Opportunities for Molecular Engineering in Catalysis. AIChE J. 2000, 46, 1902–1906.

24 Chng, L. L.; Erathodiyil, N.; Ying, J. Y. Nanostructured Catalysts for Organic Transformations. Acc. Chem. Res. 2013, 46, 1825–1837.

25 Kovalenko, M. V.; Manna, L.; Cabot, A.; Hens, Z.; Talapin, D. V.; Kagan, C. R.; Klimov, V. I.; Rogach, A. L.; Reiss, P.; Milliron, D. J.;Guyot-Sionnnest, P.; Konstantatos, G.; Parak, W. J.; Hyeon, T.; Korgel, B. A.;Murray, C. B.; Heiss, W. Prospects of Nanoscience with Nanocrystals. ACS Nano 2015, 9, 1012–1057.

26 Wang, A.; Qin, M.; Guan, J.; Wang, L.; Guo, H.; Li, X.; Wang, Y.; Prins, R.; Hu, Y. The Synthesis of Metal Phosphides: Reduction of Oxide Precursors in a Hydrogen Plasma. Angew. Chem., Int. Ed. 2008, 47, 6052–6054.

27 Deng, Y.; Zhou, Y.; Yao, Y.; Wang, J. Facile Synthesis of Nanosized Nickel Phosphides with Controllable Phase and Morphology. New J. Chem.2013, 37, 4083–4088.

28 Maneeprakorn, W.; Nguyen, C. Q.; Malik, M. A.; O’Brien, P.; Raftery, J. Synthesis of the Nickel Selenophosphinates [Ni(Se2PR2)2] (R = iPr, tBu and Ph) and Their Use as Single Source Precursors for the Deposition of Nickel Phosphide or Nickel Selenide Nanoparticles. J. Dalton Trans.2009, 2103–2108.

29 Mi, K.; Ni, Y.; Hong, J. Solvent-Controlled Syntheses of Ni12P5 and Ni2P Nanocrystals and Photocatalytic Property Comparison. J. Phys. Chem. Solids 2011, 72, 1452–1456.

30 Chang, J.; Ning, Y. H.; Wu, S. L.; Zhang, S. F. Phase-Controllable Synthesis of Nickel Phosphides Using a New Ligand Myristic Acid Adv. Mater. Res. 2013, 679, 23–26.

31 Muthuswamy, E.; Savithra, G. H. L.; Brock, S. L. Synthetic Levers Enabling Independent Control of Phase, Size, and Morphology in Nickel Phosphide Nanoparticles. ACS Nano 2011, 5, 2402–2411.

32 Liyanage, D. R.; Danforth, S. J.; Liu, Y.; Bussell, M. E.; Brock, S. L. Simultaneous Control of Composition, Size, and Morphology in Discrete Ni2-xCoxP Nanoparticles. Chem. Mater. 2015, 27, 4349– 4357.

33 Sun, C.; Ma, M.; Yang, J.; Zhang, Y.; Chen, P.; Huang, W.; Dong, X. Phase-Controlled Synthesis of α-NiS Nanoparticles Confined in Carbon Nanorods for High Performance Supercapacitors. Sci. Rep. 2014, 4, 1–6. 11

34 Roffey, A.; Hollingsworth, N.; Islam, H-U.; Mercy, M.; Sankar, G.; Catlow, C. R. A.; Hogarth, G.; de Leeuw, N. H. Phase Control during the Synthesis of Nickel Sulfide Nanoparticles from Dithiocarbamate Precursors. Nanoscale 2016, 8, 11067–11075.

35 Ruditskiy, A.; Peng, H.-C.; Xia, Y. Shape-Controlled Metal Nanocrystals for Heterogeneous Catalysis. Annu. Rev. Chem. Biomol. Eng. 2016, 7, 327–348.

36 Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Shape-Controlled Synthesis of Metal Nanocrystals: Simple Chemistry Meets Complex Physics? Angew. Chem., Int. Ed. 2009, 48, 60–103.

37 Xia, Y.; Xia, X.; Peng, H.-C. Shape-Controlled Synthesis of Colloidal Metal Nanocrystals: Thermodynamicversus Kinetic Products. J. Am. Chem. Soc. 2015, 137, 7947–7966.

38 Ortiz, N.; Skrabalak, S. E. On the Dual Roles of Ligands in the Synthesis of Colloidal Metal Nanostructures. Langmuir 2014, 30, 6649– 6659.

39 Vasquez, Y.; Henkes, A.; Bauer, J. C.; Schaak, R. E. Nanocrystal Conversion Chemistry: A Unified and Materials-General Strategy for the Template-Based Synthesis of Nanocrystalline Solids. J. Solid State Chem.2008, 181, 1509– 1523.

40 Guo, Y.; Alvarado, S. R.; Barclay, J. D.; Vela, J. Shape-Programmed Nanofabrication: Understanding the Reactivity of Dichalcogenide Precursors ACS Nano 2013, 7, 3616–3626.

41 Ruberu, T. P. A.; Albright, H. R.; Callis, B.; Ward, B.; Cisneros, J.; Fan, H.-J.; Vela, J. Molecular Control of the Nanoscale: Effect of Phosphine Chalcogenide Reactivity on CdS- CdSe Nanocrystal Composition and Morphology. ACS Nano 2012, 6, 5348–5359.

42 Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La Plante, I.; Owen, J. S. A Tunable Library of Substituted Thiourea Precursors to Metal Sulfide Nanocrystals. Science 2015, 348, 1226–1230.

43 Vela, J. Molecular Chemistry to the Fore: New Insights into the Fascinating World of Photoactive Colloidal Semiconductor Nanocrystals. J. Phys. Chem. Lett. 2013, 4, 653–668.

44 Alvarado, S. R.; Shortt, I.; Fan, H.-J.; Vela, J. Assessing Phosphine-Chalcogen Bond Energetics from Calculations. Organometallics 2015, 34, 4023–4031.

45 Hammond, G. S. A Correlation of Reaction Rates. J. Am. Chem. Soc. 1955, 77, 334–338.

CHAPTER 2.

PHASE-PROGRAMMED NANOFABRICATION: EFFECT OF

ORGANOPHOSPHITE PRECURSOR REACTIVITY ON THE EVOLUTION OF

NICKEL AND NICKEL PHOSPHIDE NANOCRYSTALS

Reprinted with permission from The Chemistry of Materials, 2015, 27, 8021–8023

American Chemical Society

Himashi P. Andaraarachchi, Michelle J. Thompson, Miles, A. White, Hua-Jun Fan,

Javier Vela

Abstract

A better understanding of the chemistry of molecular precursors is useful in achieving more predictable and reproducible nanocrystal preparations. Recently, an efficient approach was introduced that consists of fine-tuning the chemical reactivity of the synthetic molecular precursors used, while keeping all other reaction conditions constant. Using nickel phosphides as a research platform, we have studied how the chemical structure and reactivity of a family of commercially available organophosphite precursors (P(OR)3, R = alkyl or aryl) alter the preparation of metallic and metal phosphide nanocrystals. Organophosphites are a versatile addition to the pnictide synthetic toolbox, nicely complementing other available precursors such as elemental phosphorus or trioctylphosphine (TOP). Experimental and computational data show that different organophosphite precursors selectively yield Ni, Ni12P5, and Ni2P and that these phases evolve over time through separate mechanistic pathways. Based on our observations, we propose that nickel phosphide formation requires organophosphite coordination to a nickel precursor, followed by intramolecular rearrangement. We also propose 13 that metallic nickel formation involves outer sphere reduction by uncoordinated organophosphite. These two independent pathways are supported by the fact that preformed

Ni nanocrystals do not react with some of the most reactive phosphide-forming organophosphites, failing to evolve into nickel phosphide nanocrystals. Overall, the rate at which organophosphites react with nickel(II) chloride or acetate to form nickel phosphides increases in the order P(OMe)3 < P(OEt)3 < P(OnBu)3 < P(OCH2tBu)3 < P(OiPr)3 < P(OPh)3.

Some organophosphites, such as P(OMe)3 or P(OiPr)3, transiently form zerovalent, metallic nickel, while this is the only persistent product observed with the bulky organophosphite P(O-

2,4-tBu2C6H3)3. We expect that these results will alleviate the need for time-consuming testing and random optimization of several different reaction conditions, thus enabling a faster development of these and similar pnictide nanomaterials for practical applications.

Introduction

A better understanding of the chemistry of molecular precursors can be useful in achieving more predictable and reproducible nanocrystal preparations, and thus in producing more desirable nanocrystalline properties. Recently, we and others demonstrated an efficient approach to manipulate the outcome of nanocrystal preparations that consists of replacing the synthetic molecular precursors used while keeping all other conditions constant.1-7 This simple but powerful approach requires investigating the effect of chemical group substitution on the relative rates of decomposition of the molecular precursors, which directly impact their relative rates of nanocrystal nucleation and growth. The direct results of these types of studies are working scales of chemical reactivities for families of closely related molecular precursors (for example, phosphine chalcogenides,2 disubstituted dichalcogenides,1 thioureas3), each one of which obviates the need for much more time-consuming testing and optimization of several 14 different reaction conditions or of unrelated precursors at random. Here, we expand and generalize this chemical reactivity approach to the controllable synthesis of metal phosphides as an entry into the more general field of pnictide (M–V, M = metal) nanomaterials.

The most common phosphorus precursors used to synthesize nanocrystalline metal phosphides are triphenylphosphine (PPh3), trioctylphosphine (TOP) and its oxide (TOPO), red and white phosphorus (P4), phosphates (R3PO4, R = alkyl or metal cation), and single source

8-24 precursors such as [Ni(Se2PR2)2]. Other less explored precursors include hypophosphites

25-34 35 (MH2PO2, M = metal or ammonium cation) and the organophosphite P(OEt)3. Also known as phosphite esters or simply phosphites (P(OR)3, R = alkyl or aryl), organophosphites are particularly appealing as synthetic precursors because they are highly reactive and potentially tunable with group (R) substitution, while also being readily commercially available and fairly inexpensive. This contrasts with other available alternatives which, while synthetically also useful, are unsupported and not tunable (P4) or require relatively high reaction temperatures in excess of >320–340 °C (TOP).13-16 In this work, we explore how the structure and reactivity of a significantly expanded family of easily accessible organophosphites affects the synthesis of nickel phosphide nanocrystals.

Nanostructured nickel phosphides garnered a lot of recent interest because of their unique optoelectronic properties and applications.36 Nanocrystalline nickel phosphides are catalytically active in hydrodesulfurization (HDS) and hydrodenitrification (HDN) reactions,37 and in the hydrogen evolution reaction (HER), where they benefit from low overpotentials and some of the largest cathodic densities among nonprecious metal catalysts.38-

48 Nanocrystalline nickel phosphides are also of interest as electrodes for lithium ion batteries.49 Several methods exist for the synthesis of nickel phosphide nanocrystals (mostly 15

50 Ni2P and Ni12P5), including chemical vapor deposition, solution phase (colloidal) synthesis,16,51-70 solvothermal synthesis,71-77 electrosynthesis,78 hydrothermal synthesis,79 solid-state synthesis,80 microwave synthesis,81 and temperature-programmed phosphate reduction.82 These methods yield nickel phosphide nanocrystals with various morphologies.83 Their exact phase and composition obtained is affected by the Ni to P ratio,84,85 counteranions,86 voltage (in case of electrosynthesis),87 the identity and concentration of surfactants,88 the reaction solvent(s),89 time, and temperature.90 Some methods call for fine- tuning and optimization of multiple factors in order to achieve satisfactory levels of phase and composition control.91-95 Organophosphites offer a distinctive system where the control of phase composition can be achieved under a single set of directly comparable (identical) reaction conditions and linked to the chemical structure, bonding, and reactivity of closely related molecular precursors. Here, we demonstrate that organophosphite precursor reactivity controllably affects the composition and phase evolution of nickel (Ni) and nickel phosphide

(Ni2P and Ni12P5) nanocrystals.

Experimental Section

Materials. Nickel(II) chloride hexahydrate (NiCl2·6H2O, 99.9%), trimethyl phosphite

(P(OMe)3, 97%), triethyl phosphite (P(OEt)3, 98%), tri-n-butyl phosphite (P(OnBu)3, 94%), trineopentyl phosphite (P(OCH2Bu)3, 90%), tri-isopropyl phosphite (P(OiPr)3, 94%), triphenyl phosphite (P(OPh)3, 97%), and tris(2,4-di-t-butylphenyl) phosphite (P(O-2,4-tBu2C6H3)3,

98%) were purchased from Strem; 1-octadecene (ODE, 90%) and oleylamine (80–90%) from

Acros; and nickel(II) acetate (Ni(OAc)2, 98%) from Aldrich. All compounds were used as received and handled under an inert (dry-N2 or Ar) atmosphere inside a glovebox or with a

Schlenk line. 16

Synthesis: Phosphite Addition Solution. Inside a glovebox filled with dry N2, the phosphite precursor [0.40 mmol: 0.05 mL of P(OMe)3, 0.07 mL of P(OEt)3, 0.11 mL

P(OnBu)3, 117 mg of P(OCH2Bu)3, 0.10 mL of P(OiPr)3, 0.11 mL of P(OPh)3, or 259 mg of

P(O-2,4-tBu2C6H3)3] was thoroughly dissolved in ODE (1.00 g, 1.27 mL).

Nanocrystal Synthesis: General Procedure. Inside a three-neck flask, NiCl2·6H2O (26 mg, 0.10 mmol) or Ni(OAc)2 (18 mg, 0.10 mmol), oleylamine (270 mg, 1.00 mmol, 0.33 mL), and ODE (5.00 g, 6.34 mL) were degassed under a vacuum at 80 °C for 1 h, refilled with Ar, and heated to 275 °C. After 5 min, the organophosphite addition solution (above) was quickly injected and kept at this temperature while stirring. Aliquots were taken out after 1, 10, and 30 min reactions. The mixture was allowed to cool to room temperature, and nanocrystalline products were isolated by twice washing with toluene and centrifugation at 4900 rpm for 5 min. Thermal analysis (TGA/DSC) experiments showed that typical yields of nanocrystalline phases (without organics) ranged between 44 and 58% (see Supporting Information).

Control Experiments. Ni nanocrystals were synthesized with P(O-2,4-tBuC6H3)3 using the general synthetic procedure above. The nanocrystals were isolated and purified thrice by washing with toluene and centrifugation at 4900 rpm for 5 min, before drying under a dynamic vacuum. The general synthetic procedure was then repeated replacing Ni nanocrystals for the nickel precursor.

Characterization: Optical Characterization. Absorption spectra were measured with a photodiode array Agilent 8453 UV–Vis spectrophotometer.

Structural Characterization. Powder X-ray Diffraction (XRD) was measured using Cu

Kα radiation on a Scintag XDS-2000 diffractometer. Transmission Electron Microscopy

(TEM) was conducted on carbon-coated copper grids using a FEI Technai G2 F20 field 17 emission scanning transmission electron microscope (STEM) at 200 kV (point-to-point resolution <0.25 nm, line to line resolution <0.10 nm). Particle dimensions were measured manually and/or with ImageJ for >50–100 particles. Averages are reported ± one standard deviation.

Computations: Organophosphite Precursors. We modeled homolytic and heterolytic

P–O and C–O bond cleavage in organophosphites by optimizing the geometries of whole molecule and separate molecular fragments under similar methods and basis sets. The general process is illustrated as AB → A + B, where the cleavage energy (assumed to correspond to the bond strength) was calculated as ΔEA + ΔEB – ΔEAB. The multiplicities of the different molecular fragments were monitored for spin contamination through their S2 values.

Homolysis involves triplet multiplicity, while heterolysis involves singlet multiplicity. The calculated total energy (ΔE), electronic energy with zero-point energy correction (ΔEZPE), change in enthalpy (ΔH), and change in Gibbs free energy (ΔG) were then corrected to 298.15

K and 1 atm (gas phase). We also calculated the Gibbs free energy (ΔG) with the nonpolar solvent of cyclohexane and the polar solvent water by carrying out a single point energy correction with the aforementioned optimized geometries. The solvated Gibbs free energies

(ΔG) follow the same patterns of those reported in the gas phase (see Supporting Information)

All calculations were carried out using the Gaussian 03 package96 running on the CenterOS based Linux cluster at the Department of Chemistry, Prairie View A&M University. The Tao–

Perdew–Staroverov–Scuseria (TPSS) method97 implemented in Gaussian 03 was used for all geometry optimizations, solvation modeling, and frequency calculations. As a new generation of density functional, TPSS matches or even exceeds in accuracy almost all prior functionals, including the most popular functional-B3LYP with hybrid exchange functionals.98 For 18 example, TPSS recognizes relatively weak interactions (such as agostic interactions), while

B3LYP significantly underestimates them.99 Because hydrogen atoms in the modeling system do not play significant roles in our study, the 6-311G(d) basis set100, 101 was used for all elements in the modeling system. Not applying polarization functions on the H’s far from the phosphorus center atom does not significantly degrade computational precision and accuracy, while significantly accelerating the calculations.102 All structures were fully optimized and frequency analyses performed to ensure minima were achieved, with zero imaginary vibrational frequencies derived from vibrational frequency analysis. Thermodynamic functions, including enthalpies, entropies, and free energies, were calculated at 298.15 K and

1 atm. The Polarizable Continuum Model (PCM) using the integral equation formalism variant

(IEFPCM) was applied to compute aqueous solvation Gibbs free energies for all compounds

(see Supporting Information available).

Results and Discussion

General Observations. The basic reaction we studied consists of injecting an organophosphite precursor into a solution containing a nickel(II) source (chloride or acetate) and oleylamine in 1-octadecene (ODE) at 275 °C (Scheme 1). We specifically chose these parameters so that we could directly compare the molecular precursors under conditions where they all reacted (see Experimental Section for details). Aliquots were taken from the mixture after 1, 10, and 30 min reactions and the contents characterized by optical and, after precipitation, structural methods. We repeated this procedure multiple times for each of several commercially available organophosphites bearing aliphatic or aromatic substituents, including trimethyl phosphite (P(OMe)3), triethyl phosphite (P(OEt)3), tri-n-butyl phosphite (P(OnBu)3), trineopentyl phosphite (P(OCH2tBu)3), triisopropyl phosphite (P(OiPr)3), triphenyl phosphite 19

(P(OPh)3), and tris(2,4-ditert-butylphenyl) phosphite (P(O-2,4-tBu2C6H3)3. Figure 1 shows representative results from our precursor screening, including powder X-ray diffraction (XRD) and transmission electron microscopy (TEM) data for selected precursors (additional data are available from the Supporting Information). All results, summarized in Table 1, are fully reproducible, as the same products are observed every time that a given set of experimental conditions was repeated (for example, see Figure S1).

Scheme 1. Synthetic Conditions Used to Study Organophosphite Precursor Reactivity

Our experimental observations show that different organophosphite precursors consistently lead to the formation of metallic nickel or nickel phosphides or both and that these phases evolve over time (Figure 2). It is clear that the chemical structure and reactivity of the specific molecular organophosphite precursor used exert a significant influence on the ease

(rate) of formation as well as on the selectivity toward different nickel-containing nanocrystals.

The results of these reactions are independent of the exact nickel(II) precursor used; for example, NiCl2 and Ni(OAc)2produce the same product(s) after similar reaction times (Figure

S2). This could be important for applications, as halides can poison the catalytic activity of nickel phosphides.103 20

Figure 1. Powder XRD patterns and representative TEM images (after 30 min reaction) of nanocrystals obtained by reacting nickel(II) chloride in oleylamine and octadecene at 275 °C with an organophosphite precursor: P(OMe)3 (a), P(OCH2tBu)3 (b), P(OPh)3 (c), or P(O-2,4- tBu2C6H3)3 (d). Reference PDF numbers: NiO, 47-1049; Ni, 4-850; Ni12P5, 22-1190; and Ni2P, 3-953. 21

Table 1. Organophosphite Precursor Reactivity: Time Evolution and Selected Properties of Nanocrystalline Products.a Precursor t = 1 min t = 10 min t = 30 min Phase(s)b Phase(s)b Phase(s)b XRD TEM TEM Size (nm)d Shape Phosphide-forming precursors c c b d d P(OMe)3 A A (60%) + Ni Ni2P 39 49±9 Hollo (40%) w c e P(OEt)3 A Ni12P5 (50%) + Ni2P 36 37±9 Solid A (50%)b c P(OnBu)3 A Ni12P5 (85%) + Ni2P 30 51±8 Hollo Ni2P (15%) w d d P(OCH2tBu)3 - Ni12P5 Ni12P5 15 55±8 Hollo (96%) + w e Ni2P (4%) e P(OiPr)3 Ni Ni2P Ni2P 19 26±5 Solid e P(OPh)3 Ni12P5 Ni2P Ni2P 14 17±3 Solid (90%) + Ni2P (10%) Non-phosphide-forming precursor P(O-2,4- Ni Ni Ni 17d 55±1 Solide d tBu2C6H3)3 2 a Conditions: [Ni]T = 12.5 mM, [oleylamine]T = 125 mM, ODE solvent, 275 ºC (see Experimental for details). bEstimated from the relative heights of the most intense XRD peaks. c“A” is characterized by a diffraction peak at ca. 47, but remains unidentified (see text). dSize differences are due to the type of measurement technique. XRD measures single crystalline domains, while TEM measures overall grain size. e“Solid” refers to dense e (not hollow) particles (particles without a noticeable void by TEM). Becomes 100% Ni2P after 90 min reaction.

Evolution of Nickel Phosphides. Under the conditions studied (Scheme 1 above), the organophosphite precursors P(OMe)3, P(OEt)3, P(OnBu)3, P(OCH2tBu)3, P(OiPr)3, and

P(OPh)3 lead to nanocrystalline Ni12P5 or Ni2P or both. In the cases where it is observed, formation of the nickel-rich tetragonal Ni12P5 phase precedes the formation of the hexagonal

Ni2P phase, the latter being the final product after 30 min at 275 °C in most of these cases. The subsequent phase transformation from Ni12P5 to Ni2P is significantly slower in the case of the

P(OCH2tBu)3 precursor (complete after 90 min), possibly because it is relatively bulky, with a 22

Tolman cone angle θ of 180° (Table 2).104-106 Generally, primary aliphatic organophosphites such as P(OMe)3 and P(OCH2tBu)3 are the slowest to react with the nickel(II) precursor

(Figures 1a,b and 2), whereas the aromatic phosphite P(OPh)3 is the fastest to react with the nickel(II) precursor (Figures 1c and 2). At early reaction times (1–10 min), some primary alkyl phosphites, P(OMe)3, P(OEt)3, and P(OnBu)3, form a minor, transient crystalline impurity

(“A”) that is characterized by a broad X-ray diffraction at ca. 2θ = 47°. Because it roughly matches one of the main diffractions of either Ni12P5 or Ni2P, we speculate this peak may correspond to poorly diffracting nickel phosphide nuclei, perhaps with significant preferred orientation (Figure 1a). However, we are at present unable to unambiguously characterize this phase.

Figure 2. Time evolution and approximate distribution of nanocrystalline phases produced by reacting nickel(II) chloride with different organophosphite precursors (P(OR)3) in oleylamine and octadecene at 275 °C. “A”, characterized by an XRD peak at ca. 47 degrees, remains unidentified (see text).

Table 2. Selected electronic and structural parameters of organophosphite precursors. Tolman cone Bond Bond heterolysisb anglea homolysisb Precursor θ (degrees) P–O (kcal/mol) P+|O- (kcal/mol) P-|O+ (kcal/mol) P(OMe)3 107 63.16 193.69 290.48 P(OEt)3 109 62.69 186.38 275.69 P(OnBu)3 109 62.49 181.27 259.04 P(OCH2tBu)3 180 62.97 174.24 248.86 P(OiPr)3 130 61.21 172.07 263.48 P(OPh)3 128 46.43 149.51 218.25 P(O-2,4- 192 36.34 125.75 174.91 tBu2C6H3)3 aTaken from references 104, 105, and 106. bGibbs free energies, ΔG, calculated in Gaussian03 using DFT and the Tao-Perdew-Staroverov-Scuseria (TPSS) method (see Experimental section under Computations for details).

Extensive TEM analysis of the 30 min reaction products showed a strong correlation between organophosphite precursor reactivity and the size and morphology of nickel phosphide nanocrystals. The most reactive phosphide-forming precursors such as P(OPh)3 lead to dense (nearly solid or void-free) and relatively small (ca. 17–37 nm) Ni2P nanocrystals

(Figure 1c and Table 1). In contrast, the least reactive phosphide-forming precursors such as

P(OMe)3 yield hollow and relatively larger (ca. 49–55 nm) Ni2P nanocrystals (Figure 1a and Table 1). The presence and formation of such nanocrystal voids following a phase transformation is widely attributed to the Kirkendall effect.15,107-117 During the conversion from

Ni12P5 to Ni2P, nickel ions diffuse outward and phosphide ions diffuse inward, migrating toward and away from the reactive nanocrystal surface, respectively. Highly reactive precursors such as P(OPh)3 are very efficient at generating phosphide ions, which can then quickly diffuse inward toward the nanocrystal interior, thus counterbalancing the outward diffusion of nickel ions and resulting in dense (solid or void-free) Ni2P nanocrystals. In contrast, less reactive precursors such as P(OMe)3 are not as efficient sources of phosphide 24 ions; outward diffusion of nickel ions dominates in this case, resulting in the formation of hollow Ni2P nanocrystals.

Evolution of Metallic Nickel. Some of the organophosphite precursors we screened produce face-centered cubic (fcc) nickel (Ni) nanocrystals. P(OiPr)3 and P(OMe)3 form this phase transiently, with Ni quickly and completely disappearing after 1 and 10 min reaction, respectively (Table 1). Other organophosphites such as P(OEt)3, P(OnBu)3, P(OCH2tBu)3, or

P(OPh)3 could also be forming Ni transiently, although this was not observed here, likely because we mostly sampled reactions at specific times (1, 10, and 30 min and longer). In the past, the transient formation and disappearance of Ni were used to propose that this was an intermediate phase en route to the formation of nickel phosphides such as Ni12P5 and Ni2P

(Figure 1d and Table 1).15, 56,108 However, in this study, we consistently found that at least one precursor, P(O-2,4-tBu2C6H3)3, yields Ni nanocrystals that never evolve to a nickel phosphide phase, not even after the addition of a different organophosphite precursor (see detailed discussion below).

Mechanistic Considerations: Formation of Nickel Phosphides. We have considered up to five potential mechanisms for the initial decomposition of molecular organophosphite precursors leading to nanocrystalline nickel phosphides (NixPy = Ni12P5 or Ni2P or both; Scheme 2).118,119 Mechanism a requires phosphite coordination to the nickel precursor, followed by an intramolecular rearrangement where the P–O and Ni–X bonds break in a concerted fashion (X = chloride, acetate, or oleylamine derived amide; Scheme 2a).

Mechanism b involves homolytic P–O bond cleavage of the free phosphite, followed by reaction with the nickel precursor (Scheme 2b). Mechanisms c and d involve heterolytic P–O bond cleavage in the free phosphite, followed by reaction with the nickel precursor (Scheme 2c 25

120 and d, respectively). Mechanism e involves β-hydride elimination (βHE) to produce PH3, which can act as a phosphorus source to generate nickel phosphides (Scheme 2e).

Scheme 2. Five possible mechanisms to account for the formation of nickel phosphide phases (NixPy) in the presence of organophosphites.

(a) R R, R' = Alkyl, aryl (RO)2P O X = Amide, halide, acetate

NixPy + ROX Ni X X

(b) R NiX2 Ni P (RO)2P O (OR)2P + OR x y

(d) R R (OR)2P O ROX ROX (RO)2P O + Ni X (OR)2P Ni NixPy Ni X X X X

(e) R'CH O 2 R'CH O P(R'H2CO)2 NiX2 (R'H CO) PH PH3 Ni P O H 2 2 x y βHE C H R'

As evidenced by the outcome of reactions using P(OPh)3 (Table 1 and Figure 1c), formation of nickel phosphides under our experimental conditions does not require (is not contingent upon) the presence of β-hydrogens; therefore, we can rule out mechanism e. To address the likelihood that the other mechanisms (a–d) could be involved in our reactions, we performed simple Density Functional Theory (DFT) calculations of the bonds surrounding the reactive P(O−)3 unit. Selected results from these calculations are shown in Table 2 (additional data are available in the Supporting Information). The calculated P–O bond dissociation 26

(homolysis) energies or “BDEs” are only ca. 36–63 kcal/mol (Table 2). For comparison, we estimate that each methylene C–H bond in P(OCH2tBu)3 has a BDE of 90.72 kcal/mol. In contrast, P–O bond heterolysis energies are much higher, with polarization favoring the movement of electrons toward the more electronegative O atom. Heterolytic P–O bond cleavage is significantly more favorable when phosphonium-alkoxide ion pairs (P+|O–) are produced (125–193 kcal/mol) than when phosphide-oxenium121 ion pairs (P–|O+) are produced

(174–290 kcal/mol, Table 2). Thus, while they may be important in other systems (different phosphide or metal precursors),120 pathways involving P–O bond heterolysis are clearly too energy intensive and demanding to be viable in our reactions, which strongly argues against mechanism c and, in particular, against mechanism d.

An important point when considering the homolytic mechanism b, as well as the heterolytic mechanisms c and d, is that these do not require precoordination of the phosphite precursor to the nickel center. If b or c or d were operative, one would predict (and should fully expect) that a bulky phosphite would react just as easily as a nonbulky one, at a rate that is simply commensurate with its relative P–O bond energy. Critically, this is not observed experimentally. The most sterically encumbered (bulkiest) phosphite that we studied, P(O-2,4- tBu2C6H3)3, with a large Tolman cone angle of θ = 192° (Table 2), fails to produce any detectable crystalline nickel phosphide, despite the fact that it has the smallest homolytic and heterolytic P–O energies and, thus, the weakest P–O bond (Table 1). This observation alone allows us to rule out mechanisms b, c, and d and strongly suggests that phosphite coordination is a necessary prerequisite for nickel phosphide formation. Therefore, among the five phosphide-forming mechanisms considered in Scheme 2, we conclude that only mechanism a is consistent with all of our data. 27

Mechanistic Considerations: Formation of Metallic Nickel. We next considered the possibility (see above) that a separate pathway may be responsible for the formation of metallic

Ni nanocrystals. We concretely evaluated two possible mechanisms (Scheme 3). Mechanism i involves coordination of the phosphite to nickel(II), forming a five-coordinate intermediate that reductively eliminates to produce a Ni(0) phosphite complex,122-125 which then decomposes into zerovalent (metallic) fcc Ni particles (Scheme 3i). This inner sphere reaction and its required intermediates are well-known, having ample precedent in the organometallic literature.126,127 Alternatively, mechanism ii involves direct reduction of nickel(II) to Ni(0) by the free, uncoordinated phosphite (Scheme 3ii). This outer sphere, electron transfer mechanism is supported by the observation that organophosphites can indeed act as reducing agents.128

While the small phosphite P(OMe)3 (θ = 107°, Table 2) and intermediate size phosphite

P(OiPr)3(θ = 130°) transiently produce Ni, it is by far the most sterically encumbered phosphite, P(O-2,4-tBu2C6H3)3 (θ = 192°), that is the most active in producing crystalline fcc

Ni. Ni particles were the only product observed, and no phosphide phases were present at any point during the reaction with this precursor. Because this bulkiest phosphite P(O-2,4- tBu2C6H3)3 is the most reactive toward the formation of crystalline Ni, we can rule out mechanism i. Thus, only mechanism ii, outer sphere reduction, is consistent with our data.

Assuming that the reducing abilities of the different phosphites are comparable, this suggests that the outer-sphere electron transfer reduction of Ni(II) is a slower process overall compared to the formation of nickel phosphides, only occurring when the binding of a sterically hindered organophosphite to Ni(II) becomes highly unfavorable.

Scheme 3. Two possible mechanisms for Ni formation in the presence of organophosphites.

(a)

(RO) P P(OR) X X 3 3 X2 (RO)3P P(OR)3 Ni Ni (RO)3P Ni P(OR)3 Ni RE X X P(OR)3 P(OR)3

(b) e-- transfer L = Amine, phosphite LnNiX2 + P(OR)3 Ni X = Amide, halide

Decoupling Nickel and Nickel Phosphide Formation. As noted above, we have considered two separate mechanistic hypotheses for the evolution of nickel and nickel phosphide nanophases. Our results call into question whether nanocrystalline Ni can really serve as an intermediate during the formation of nickel phosphides, as was reported previously in the literature (albeit, this was for different sets of precursors and reaction conditions compared to those used here).15,67,108 To address this question, we first synthesized Ni nanocrystals as described above using P(O-2,4-tBu2C6H3)3 (Figure 1d). We then subjected these preformed Ni nanocrystals to some of our most active phosphide-forming organophosphites. As shown in Figure 3, the reaction of isolated, purified Ni nanocrystals with

P(OPh)3 and oleylamine in 1-octadecene (ODE) at 275 °C (Scheme 4) does not result in any observable crystalline nickel phosphides. In fact, under these conditions, the Ni nanocrystals simply decompose, forming intractable, amorphous product(s) that is (are) silent by powder

XRD. It is important to purify the Ni nanocrystals before running this control reaction, as otherwise some unreacted nickel(II) chloride remains, leading to the very slow formation of some nanocrystalline Ni12P and Ni2P (Figure 3). In summary, we conclude that, under the 29 reaction conditions used here, preformed Ni nanocrystals are not a competent intermediate toward the formation of either Ni12P5 or Ni2P.

Scheme 4.

275 °C, 10 min, oleylamine NiCl2 + P(O-2,4-tBu2C6H3)3 Ni octadecene / Ar

275 °C, 30 min, oleylamine Ni + P(OPh)3 Amorphous product octadecene / Ar

Figure 3. Powder XRD patterns of solids obtained after reacting preformed Ni nanocrystals with P(OPh)3 in oleylamine and octadecene at 275 °C for 30 min. Isolation and purification of Ni is necessary in order to remove unreacted nickel(II) chloride precursor. Without this purification or “washing” step, the unreacted nickel(II) precuror reacts with the added organophosphite to form a small amount of Ni2P. Reference PDF numbers: NiO, 47-1049; Ni, 4-850; Ni12P5, 22-1190; and Ni2P, 3-953.

Building a chemical reactivity scale for organophosphite precursors. Having studied the general reactivity and decomposition mechanisms of multiple organophosphite precursors 30 toward nickel(II) chloride (or acetate), we are able to build a chemical reactivity scale for the formation of nickel phosphide nanophases under the conditions we studied. Based on our experimental observations, the ease or rate at which organophosphites react with nickel(II) to form nickel phosphides (Ni12P5 or Ni2P or both) increases in the order: P(OMe)3 < P(OEt)3 <

P(OnBu)3 < P(OCH2tBu)3 < P(OiPr)3 < P(OPh)3 (Figure 4a). Some organophosphites, such as

P(OMe)3 and P(OiPr)3 (and likely, other organophosphites too, see above) transiently form zerovalent, metallic nickel (Ni) (Figure 4b). Under the conditions studied, the P(O-2,4- tBu2C6H3)3 precursor is either unable or kinetically incompetent of forming nickel phosphides, but forms Ni nanocrystals that persist over time (Figure 4b).

(a) O O O O O P P P O O O O O O O P O O O O O P P P O O O O

Did not observe Ni12P5 or Ni2P Increasing precursor reactivity

(b)

O O O P P O O O O O O P O P O O P O O O O O P P O O O O Could transiently form Ni Transiently form Ni Forms Ni

Figure 4. Ability of different commercially available organophosphite precursors to form nickel phosphides (a) and metallic nickel (b).

Conclusions

A critical aspect of the synthesis and application of nanostructured materials is to precisely control their phase, composition, and, consequently, properties. Using a powerful 31 chemical reactivity approach that is well established for metal chalcogenides, we have studied how the structure and reactivity of a family of commercially available organophosphite precursors (P(OR)3, R = alkyl or aryl) affect the evolution of nickel and nickel phosphide nanocrystals. Our observations show that different organophosphite precursors selectively yield nickel phosphide (Ni12P5, Ni2P) or nickel (Ni) nanophases and that these evolve over time through well-defined and separate mechanistic pathways.

In agreement with prior literature reports, we find that the formation of a nickel-rich, kinetic tetragonal Ni12P5 phase precedes the formation of the final, thermodynamically preferred hexagonal Ni2P phase. In the specific case of phosphide-forming organophosphites, very reactive precursors such as P(OPh)3 form small, nearly void-free (dense) nanocrystals, while less reactive precursors such as P(OMe)3 form large, hollow nanocrystals due to the

Kirkendall effect. Some organophosphites such as P(OiPr)3 and P(OMe)3 yield fcc-Ni transiently, while at least one organophosphite, P(O-2,4-tBu2C6H3)3, more persistently yields

Ni nanocrystals that never evolve into a nickel phosphide phase.

In the specific case of nickel phosphide forming reactions, it is important to note that the observations and discussions in this paper actually deal with two reactions and processes that occur simultaneously, namely the initial decomposition and reaction of organophosphite and nickel(II) precursors to form Ni12P5 nanocrystals, and the phase transformation of Ni12P5 to

Ni2P nanocrystals. Because these are two fundamentally different and separate processes, it is very likely that precursor electronic and steric effects may impact their rates to varying degrees and different extents. For example, both of the primary alkyl phosphites P(OnBu)3 (θ = 109°) and P(OCH2tBu)3 (θ = 180°) react with the nickel(II) precursor to form Ni12P5 at comparable rates, but conversion of Ni12P5 into Ni2P is significantly slower in the case of the bulkier 32

P(OCH2tBu)3precursor. Clearly, and on the basis of the relatively larger cone angles (θ) of the latter, sterics play a much more prominent role in the reaction of organophosphite with

Ni12P5 than with the initial Ni(II) precursor. Similarly, during the conversion of Ni12P5 to Ni2P and its associated Kirkendall-type void formation or “hollowing,” some precursors such as

P(OMe)3, P(OnBu)3, and P(OCH2tBu)3 lead to hollow particles, whereas others such as

P(OEt)3 and P(OiPr)3 do not (these tend to give more solid Ni2P particles). As explained above, these trends must be a consequence of a close interplay and tight competition between the precursor decomposition and ion diffusion rates (phosphide moving inward, nickel moving outward) that are specific to this second, phase transformation step.

We have carefully considered five and two separate mechanisms for the initial reaction and decomposition of organophosphites and nickel(II) precursors to form nickel phosphide and nickel(0) nanocrystals, respectively. A mechanism requiring organophosphite coordination to the nickel(II) precursor, followed by an intramolecular rearrangement where the P–O and Ni–

X bonds break in a concerted fashion (X = chloride, acetate, or oleylamine-derived amide) is consistent with our observations about the formation of nickel phosphides. Another mechanism, involving direct, outer sphere reduction of nickel(II) to Ni(0) by free, uncoordinated organophosphite is consistent with our data on the formation of Ni. Control experiments show that preformed Ni nanocrystals are not competent intermediates in the formation of either Ni12P5 or Ni2P under the reaction conditions we used, because they fail to react with some of the most reactive phosphide-forming organophosphites.

Organophosphites are a nice addition to the synthetic toolbox available to pnictide chemists. Other useful phosphide precursors include elemental phosphorus, trioctylphosphine

(TOP), and its oxide (TOPO). Unlike elemental phosphorus, which is unsupported, 33 organophosphites are amenable to chemical reactivity fine-tuning via R group substitution with a wide range of aliphatic or aromatic substituents. Unlike TOP or TOPO, which require relatively high reaction temperatures in excess of >320–340 °C, organophosphites can be made more or less reactive simply by altering their chemical structure, bonding, and reactivity.

As a direct result of our study, we have built a chemical reactivity scale for organophosphite precursors. The ease or rate at which organophosphites react with nickel(II) dichloride to form nickel organophosphites increases in the order P(OMe)3 < P(OEt)3 <

P(OnBu)3 < P(OCH2tBu)3 < P(OiPr)3 < P(OPh)3. At least two organophosphites, P(OMe)3 and

P(OiPr)3, also form nickel nanocrystals transiently, while P(O-2,4-tBu2C6H3)3 only forms nickel nanocrystals, and these persist over time. Other available methods call for fine-tuning and optimizing several reaction conditions in order to achieve satisfactory levels of synthetic control over the phase and composition of metal phosphides. In contrast, we envision our results and approach will enable a faster, more systematic development of these and similar pnictide nanomaterials for practical applications.

Acknowledgements

J.V. gratefully acknowledges the National Science Foundation for funding of this work through the Division of Materials Research, Solid State and Materials Chemistry program

(NSF-DMR-1309510). H.-J.F. thanks the Department of Chemistry at Prairie View A&M

University for release time, and partial financial support from a 2014 Summer Research mini- grant (115103-00011), the U.S. Department of Energy, NNSA (#DE-NA0002630 and DE-

NA0001861), and the Welch Foundation (#L0002). The authors thank Gordie Miller, Irmi

Schewe-Miller, Brandi Cossairt, and Yuemei Zhang for comments and valuable discussions.

References

1 Guo, Y.; Alvarado, S. R.; Barclay, J. D.; Vela, J. Shape-Programmed Nanofabrication: Understanding the Reactivity of Dichalcogenide Precursors. ACS Nano 2013, 7, 3616–3626. 2 Ruberu, T. P. A.; Albright, H. R.; Callis, B.; Ward, B.; Cisneros, J.; Fan, H.-J.; Vela, J. Molecular Control of the Nanoscale: Effect of Phosphine Chalcogenide Reactivity on CdS- CdSe Nanocrystal Composition and Morphology. ACS Nano 2012, 6, 5348–5359. 3 Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La Plante, I.; Owen, J. S. A Tunable Library of Substituted Thiourea Precursors to Metal Sulfide Nanocrystals. Science 2015, 348, 1226–1230. 4 Vela, J. Molecular Chemistry to the Fore: New Insights into the Fascinating World of Photoactive Colloidal Semiconductor Nanocrystals. J. Phys. Chem. Lett. 2013, 4, 653–668. 5 Alvarado, S. R.; Shortt, I.; Fan, H.-J.; Vela, J. Assessing Phosphine-Chalcogen Bond Energetics from Calculations. Organometallics 2015, 34, 4023–4031. 6 Ortiz, N.; Skrabalak, S. E. On the Dual Roles of Ligands in the Synthesis of Colloidal Metal Nanostructures. Langmuir 2014, 21, 6649–6659. 7 McElwee-White, L.; Koller, J.; Kim, D.; Anderson, T. J. Mechanism-Based Design of Precursors for MOCVD. ECS Trans. 2009, 25, 161−171. 8 Carenco, S.; Portehault, D.; Boissière, C.; Mézailles, N.; Sanchez, C. Nanoscaled Metal Borides and Phosphides: Recent Developments and Perspectives. Chem. Rev. 2013, 113, 7981−8065. 9 Carenco, S.; Portehault, D.; Boissière, C.; Mézailles, N.; Sanchez, C. 25th Anniversary Article: Exploring Nanoscaled Matter from Speciation to Phase Diagrams: Metal Phosphide Nanoparticles as a Case of Study. Adv. Mater. 2013, 26, 371−390. 10 Brock, S. L.; Senevirathne, K. Recent Developments in Synthetic Approaches to Transition Metal Phosphide Nanoparticles for Magnetic and Catalytic Application. J. Solid State Chem. 2008, 181, 1552−1559. 11 Brock, S. L.; Perera, S. C.; Stamm, K. L. Chemical Routes for Production of Transition- Metal Phosphides on the Nanoscale: Implications for Advanced Magnetic and Catalytic Materials. Chem. Eur. J. 2004, 10, 3364–3371. 12 Vasquez, Y.; Henkes, A.; Bauer, J. C.; Schaak. R. E. Nanocrystal Conversion Chemistry: A Unified and Materials-General Strategy for the Template-Based Synthesis of Nanocrystalline Solids. J. Solid State Chem. 2008, 181, 1509–1523. 13 Mobarok, M. H.; Luber, E. J.; Bernard, G. M.; Peng, L.; Wasylishen, R. E.; Buriak, J. M., Phase-Pure Crystalline Zinc Phosphide Nanoparticles: Synthetic Approaches and Characterization. Chem. Mater. 2014, 26, 1925–1935. 13 Luber, E. J.; Mobarok, M. H.; Buriak, J. M., Solution-Processed Zinc Phosphide (α-Zn3P2) Colloidal Nanocrystals for Thin Film Photovoltaic Applications. ACS Nano 2013, 7, 8136– 8146. 35

14 Henkes, A. E.; Vasquez, Y.; Schaak, R. E., Converting Metals into Phosphides: A General Strategy for the Synthesis of Metal Phosphide Nanocrystals. J. Am. Chem. Soc. 2007, 129, 1896–1897. 15 Henkes, A. E.; Schaak, R. E., Trioctylphosphine: A General Phosphorus Source for the Low- Temperature Conversion of Metals into Metal Phosphides. Chem. Mater. 2007, 19, 4234– 4242. 16 Lu, A.-H.; Salabas, E. L.; Schüth, F. Magnetic Nanoparticles: Synthesis, Protection, Functionalization, and Application. Angew. Chem. Int. Ed. 2007, 46, 1222–1244. 17 Braunstein, P. Functional Ligands and Complexes for New Structures, Homogeneous Catalysts and Nanomaterials. J. Organomet. Chem. 2004, 689, 3953-3967. 18 Carenco, S.; Boissière, C.; Mézailles, N.; Sanchez, C. Metal phosphides: A Revival at the Nanoscale. Actual Chimique, 2012, 362, 22–28. 19 Kharisov, B. I.; Kharissova, O. V.; Yacaman, M. J. Nanostructures with Animal-like Shapes. Ind. Eng. Chem. Res. 2010, 49, 8289–8309. 20 Pickett, N. L.; O’Brien, P. Syntheses of Semiconductor Nanoparticles Using Single- Molecular Precursors. Chem. Rec. 2001, 1, 467–479. 21 Masala, O.; Seshadri, R. Synthesis Routes for Large Volumes of Nanoparticles. Annu. Rev. Mater. Res. 2004, 34, 41–81. 22 Vilas G. Pol, V. G.; Pol, S. V.; Gedanken, A. Dry Autoclaving for the Nanofabrication of Sulfides, Selenides, Borides, Phosphides, Nitrides, Carbides, and Oxides. Adv. Mater. 2011, 23, 1179–1190. 23 Shen, G. Z.; Chen, D. One-Dimensional Nanostructures and Devices of II-V Group Semiconductors. Nanoscale Res. Lett. 2009, 4, 779–788. 24 Shi, G.; Shen, J. New Synthesis Method for Nickel Phosphide Nanoparticles: Solid Phase Reaction of Nickel Cations with Hypophosphites. J. Mater. Chem. 2009, 19, 2295–2297. 25 Zhao, Y.; Zhao, Y.; Feng, H.; Shen, J. Synthesis of Nickel Phosphide Nanoparticles in a Eutectic Mixture for Hydrotreating Reactions. J. Mater. Chem. 2011, 21, 8137–8145. 26 Cecilia, J. A.; Infantes-Molina, A.; Rodríguez-Castellón, E.; Jiménez-López, A. A Novel Method for Preparing an Active Nickel Phosphide Catalyst for HDS of Dibenzothiophene. J. Catal. 2009, 263, 4–15. 27 Song, H.; Dai, M.; Song, H. L.; Wan, X.; Xu X. W. A Novel Synthesis of Ni2P/MCM-41 Catalysts by Reducing a Precursor of Ammonium Hypophosphite and Nickel Chloride at Low Temperature. Appl. Catal., A. 2013, 462, 247–255. 28 Guan, Q.; Li, W. The Synthesis and Evaluation of Highly Active Ni2P–MoS2 Catalysts using the Decomposition of Hypophosphites. Catal. Sci. Technol. 2012, 2, 2356–2360. 29 Song, Li.; Zhang, S.; A Versatile Route to Synthesizing Bulk and Supported Nickel Phosphides by Thermal Treatment of a Mechanical Mixing of Nickel Chloride and Sodium Hypophosphite. Powder Technol. 2011, 208, 713–716. 30 Guan, Q.; Li, W. A Novel Synthetic Approach to Synthesizing Bulk and Supported Metal Phosphides. J. Catal. 2010, 271, 413–415. 36

31 Guan, Q.; Li, W.; Zhang, M.; Tao, K. Alternative Synthesis of Bulk and Supported Nickel Phosphide from the Thermal Decomposition of Hypophosphites. J. Catal. 2009, 263, 1–3. 32 Infantes-Molina, A.; Cecilia, J. A.; Pawelec, B.; Fierro, J. L. G. Rodríguez-Castellón, E.; Jiménez-López, A. Ni2P and CoP Catalysts Prepared from Phosphite-type Precursors for HDS–HDN Competitive Reactions. Appl. Catal. A-Gen. 2010, 390, 253–263. 33 Ni, Y.; Tao, A.; Hu, G.; Cao, X.; Wei, X.; Yang, Z. Synthesis, Characterization and Properties of Hollow Nickel Phosphide Nanospheres. Nanotechnology 2006, 17, 5013–5018. 34 Jun, S. E.; Jang, E. J.; Lim, J. E. (2013). US Patent No. 8,545,736. Washington, DC: U.S. Patent and Trademark Office. 35 Prins, R.; Bussell, M. E. Metal Phosphides: Preparation, Characterization and Catalytic Reactivity. Catal. Lett. 2012, 142, 1413−1436. 36 Alexander, A.-M.; Hargreaves, J. S. J. Alternative Catalytic Materials: Carbides, Nitrides, Phosphides and Amorphous Boron Alloys. Chem. Soc. Rev. 2010, 39, 4388–4401. 37 Morales-Guio, C. G.; Stern, L.-A.; Hu, X. Nanostructured Hydrotreating Catalysts for Electrochemical Hydrogen Evolution. Chem. Soc. Rev. 2014, 43, 6555–6569. 38 Popczun, E. J.; McKone, J. R.; Read, C. G.; Biacchi, A. J.; Wiltrout, A. M.; Lewis, N. S.; Schaak, R. E. Nanostructured Nickel Phosphide as an Electrocatalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2013, 135, 9267−9270. 39 Tian, T.; Ai, L.; Jiang, J. Metal–Organic Framework-Derived Nickel Phosphides as Efficient Electrocatalysts toward Sustainable Hydrogen Generation from Water Splitting. RSC Adv. 2015, 5, 10290-10295. 40 Bai, Y.; Zhang, H.; Li, X.; Liu, L.; Xu, H.; Qiu, H.; Wang, Y.; Novel Peapod-like Ni2P Nanoparticles with Improved Electrochemical Properties for Hydrogen Evolution and Lithium Storage. Nanoscale 2015, 7, 1446–1453. 41 Lu, A.; Chen, Y.; Li, H.; Dowd, A.; Cortie, M. B.; Xie, Q.; Guo, H.; Qi, Q.; Peng, D.-L. Magnetic Metal Phosphide Nanorods as Effective Hydrogen-Evolution Electrocatalysts. Int. J. Hydrogen Energy 2014, 39, 18919–18928. 42 Kucernak, A. R. J.; Sundaram, V. N. N. Nickel phosphide: The Effect of Phosphorus Content on Hydrogen Evolution Activity and Corrosion Resistance in Acidic Medium. J. Mater. Chem. A 2014, 2, 17435–17445. 43 Jiang, P.; Liu, Q.; Sun, X. NiP2 Nanosheet Arrays Supported on Carbon Cloth: An Efficient 3D Hydrogen Evolution Cathode in Both Acidic and Alkaline Solutions. Nanoscale 2014, 6, 13440–13445. 44 Huang, Z.; Chen, Z.; Chen, Z.; Lv, C.; Meng, H.; Zhang, C. Ni12P5 Nanoparticles as an Efficient Catalyst for Hydrogen Generation via Electrolysis and Photoelectrolysis. ACS Nano 2014, 8, 8121–8129. 45 Pu, Z.; Liu, Q.; Tang, C.; Asiri, A. M.; Sun, X. Ni2P Nanoparticle Films Supported on a Ti Plate as an Efficient Hydrogen Evolution Cathode. Nanoscale 2014, 6, 11031–11034. 37

46 Cao, S.; Chen, Y.; Wang, C.-J.; Hea, P.; Fu, W.-F. Highly Efficient Photocatalytic Hydrogen Evolution by Nickel Phosphide Nanoparticles from Aqueous Solution. Chem. Commun. 2014, 50, 10427–10429. 47 Feng, L.; Vrubel, H.; Bensimonb, M.; Hu, X. Easily-prepared Dinickel Phosphide (Ni2P) Nanoparticles as an Efficient and Robust Electrocatalyst for Hydrogen Evolution. Phys. Chem. Chem. Phys. 2014, 16, 5917–5921. 48 Monconduit, L. Short Review on the Nanoarchitectured Pnictides Electrodes for Li Ionery. Nanosci. Nanotechnol. Lett. 2012, 4, 118–123. 49 Paneerselvam, A.; Malik, M. A.; Afzaal, M.; O’Brien, P.; Helliwell, M. The Chemical Vapor Deposition of Nickel Phosphide or Selenide Thin Films from a Single Precursor. J. Am. Chem. Soc. 2008, 130, 2420–2431. 50 Senevirathne, K.; Burns, A. W.; Bussell, M. E.; Brock, S. L. Synthesis and Characterization of Discrete Nickel Phosphide Nanoparticles: Effect of Surface Ligation Chemistry on Catalytic Hydrodesulfurization of Thiophene. Adv. Funct. Mater. 2007, 17, 3933–3939. 51 Zheng, X.; Yuan, S.; Tian, Z.; Yin, S.; He, J.; Liu, K.; Liu, L. Nickel/Nickel Phosphide Core- Shell Structured Nanoparticles: Synthesis, Chemical, and Magnetic Architecture. Chem. Mater. 2009, 21, 4839–4845. 52 Seo, H.; Cho, K.; Lee, Y. Formation Mechanisms of Ni2P Nanocrystals using XANES and EXAFS Spectroscopy. Mater. Sci. Eng., B. 2011, 176, 132–140. 53 Zafiropoulou, I.; Papagelis, K.; Boukos, N.; Siokou, A.; Niarchos, D.; Tzitzios, V. J. Chemical Synthesis and Self-Assembly of Hollow Ni/Ni2P Hybrid Nanospheres. Phys. Chem. C 2010, 114, 7582–7585. 54 Lu, Y.; Tu, J.; Xiong, Q.; Zhang, H.; Gu, C.; Wang, X.; Mao, S.X. Large-Scale Synthesis of Porous Ni2P Nanosheets for Lithium Secondary Batteries. CrystEngComm. 2012, 14, 8633– 8641. 54 Carenco, S.; Hu, Y.; Florea, I.; Ersen, O.; Boissière, C.; Mézailles, N.; Sanchez, C. Metal- Dependent Interplay between Crystallization and Phosphorus Diffusion during the Synthesis of Metal Phosphide Nanoparticles. Chem. Mater. 2012, 24, 4134−4145. 55 Sarac, M. F.; Wu, W.; Tracy, J. B. Control of Branching in Ni3C1-x Nanoparticles and Their

Conversion into Ni12P5 Nanoparticles. Chem. Mater. 2014, 26, 3057−3064. 56 Moreau, L. M.; Ha, D.; Bealing, C. R.; Zhang, H.; Hennig, R. G.; Robinson, R.D. Unintended Phosphorus Doping of Nickel Nanoparticles during Synthesis with TOP: A Discovery through Structural Analysis. Nano Lett. 2012, 12, 4530−4539. 57 Heng, X.; Yuan, S.; Tian, Z.; Yin, S.; He, J.; Liu, K.; Liu, L. One- Pot Synthesis of Hollow Nickel Phosphide Nanoparticles with Tunable Void Sizes Using Triphenylphosphine. Mater. Lett. 2009, 63, 2283–2285. 58 Muthuswamy, E.; Brock, S., L. Oxidation Does Not (Always) Kill Reactivity of Transition Metals: Solution-Phase Conversion of Nanoscale Transition Metal Oxides to Phosphides and Sulfides. J. Am. Chem. Soc. 2010, 132, 15849–15851. 38

59 Carenco, S.; Le Goff, X. F.; Shi, J.; Robian, L.; Ersen, O.; Boissière, C.; Sanchez, C.; Mézailles, N. Magnetic Core_Shell Nanoparticles from Nanoscale-Induced Phase Segregation. Chem. Mater. 2011, 23, 2270–2277. 60 Chen, Y.; She, H.; Luo, X.; Yue, G.-H.; Peng, D.-L. Solution-Phase Synthesis of Nickel Phosphide Single-Crystalline Nanowires. J. Cryst. Growth 2009, 311, 1229–1235. 61 Park, J.; Koo, B.; Yoon, K.Y.; Hwang, Y.; Kang, M.; Park, J.; Hyeon, T. Generalized Synthesis of Metal Phosphide Nanorods via Thermal Decomposition of Continuously Delivered Metal-Phosphine Complexes Using a Syringe Pump. J. Am. Chem. Soc. 2005, 127, 8433-8440. 62 Huang, Z. P.; Chen, Z. B.; Chen, Z. Z.; Lv. C. C.; Meng, H.; Zhang, C. C. Ni12P5 Nanoparticles as an Efficient Catalyst for Hydrogen Generation via Electrolysis and Photoelectrolysis. ACS Nano 2014, 8, 8121–8129. 63 Savithra, G. H. L.; Muthuswamy, E.; Bowker, R. H.; Carillo, B. A.; Bussel, M. E.; Brock, S. L. Rational Design of Nickel Phosphide Hydrodesulfurization Catalysts: Controlling Particle Size and Preventing Sintering. Chem. Mater. 2013, 25, 825–833. 64 Singh, B.; Ho, C.-L.; Tseng, Y.-C.; Lo, C.-T. Controlled Synthesis and Magnetic Properties of Nickel Phosphide and Bimetallic Iron–Nickel Phosphide Nanorods. J. Nanopart. Res. 2012, 14, 706–718. 65 Carenco, S.; Resa, I.; Le Goff, X.; Le Floch, P.; Mézailles, N. White Phosphorus as Single Source of ‘‘P’’ in the Synthesis of Nickel Phosphide. Chem. Commun. 2008, 2568–2570. 66 Mandel, K.; Dillon, F.; Koos, A. A.; Aslam, Z.; Jurkschat, K.; Cullen, F.; Crossley, A.; Bishop, H.; Moh, K.; Cavelius, C.; Arztb, E.; Grobert, N. Facile, Fast, and Inexpensive Synthesis of Monodisperse Amorphous Nickel-Phosphide Nanoparticles of Predefined Size. Chem. Commun. 2011, 47, 4108–4110. 67 Carenco, S.; Surcin, C.; Morcrette, M.; Larcher, D.; Mézailles, N.; Boissière, C.; Sanchez, C. Improving the Li-Electrochemical Properties of Monodisperse Ni2P Nanoparticles by Self- Generated Carbon Coating. Chem. Mater. 2012, 24, 688–697. 68 Hitihami-Mudiyanselage, H.; Senevirathne, K.; Brock, S. L. Bottom-Up Assembly of Ni2P Nanoparticles into Three-Dimensional Architectures: An Alternative Mechanism for Phosphide Gelation. Chem. Mater. 2014, 26, 6251–6256. 69 Wang, X.; Wan, F.; Gao, Y.; Liu, J.; Jiang, K. J. Synthesis of High-quality Ni2P Hollow Sphere via a Template-Free Surfactant-Assisted Solvothermal Route. Cryst. Growth 2008, 310, 2569–2574. 70 Su, H.; Xie, Y.; Li, B.; Liu, X. M.; Qian, Y. T. A Simple, Convenient, Mild Solvothermal Route to Nanocrystalline Cu3P and Ni2P. Solid State Ionics 1999, 122, 157–160. 71 Lü, B.; Bai, Y. J.; Feng, X.; Zhao, Y. R.; Yang, J.; Chi, J. R. Solvo-Thermal Synthesis of Crystalline Dinickel Phosphide. J. Cryst. Growth 2004, 260, 115–117. 72 Liu, S.; Liu, X.; Xu, L.; Qian, Y.; Ma, X. Controlled Synthesis and Characterization of Nickel Phosphide Nanocrystal. J. Cryst. Growth 2007, 304, 430–434. 39

73 Xie, Y.; Su, H. L.; Qian, X. F.; Liu, X. M.; Qian, Y. T. A Mild One-Step Solvothermal Route to Metal Phosphides (Metal=Co, Ni, Cu). J. Solid State Chem. 2000, 149, 88−91. 74 Barry, B. M.; Gillan, E. G. Low-Temperature Solvothermal Synthesis of Phosphorus-Rich Transition-Metal Phosphides. Chem. Mater. 2008, 20, 2618–2620. 75 Liu. J.; Chen, X.; Shao, M.; An, C.; Yu, W.; Qian, Y. Surfactant-aided Solvothermal Synthesis of Dinickel Phosphide Nanocrystallites using Red Phosphorus as Starting Materials. J. Cryst. Growth 2003, 252, 297–301. 76 Budnikova, Y.; Tazeev, D.; Trofimov, B.; Sinyashin, O. Electrosynthesis of Nickel Phosphides on the Basis of White Phosphorus. Electrochem. Commun. 2004, 6, 700–702. 77 Liu, Z.; Huang, X.; Zhu, Z.; Dai, J. A Simple Mild Hydrothermal Route for the Synthesis of Nickel Phosphide Powders. Ceram. Int. 2010, 36, 1155–1158. 78 Barry, B. M.; Gillan, E. G. A General and Flexible Synthesis of Transition-Metal Polyphosphides via PCl3 Elimination. Chem. Mater. 2009, 21, 4454–4461. 79 Hu, X.; Yu, J. C. High-Yield Synthesis of Nickel and Nickel Phosphide Nanowires via Microwave-Assisted Processes. Chem. Mater. 2008, 20, 6743–6749. 80 Landau, M.V.; Herskowitz, M.; Hoffman, T.; Fuks, D.; Liverts, E.; Vingurt D.; Froumin, N. Ultradeep Hydrodesulfurization and Adsorptive Desulfurization of Diesel Fuel on Metal-Rich Nickel Phosphides. Ind. Eng. Chem. Res. 2009, 48, 5239–5249. 81 She, H.; Chen, Y.; Luo, X.; Yue, G.-H.; Peng, D.-L. Preparation of Anisotropic Transition Metal Phosphide Nanocrystals: The Case of Nickel Phosphide Nanoplatelets, Nanorods, and Nanowires. J. Nanosci. Nanotechnol. 2010, 10, 5175-5182. 81 Wang, A.; Qin, M.; Guan, J.; Wang, L.; Guo, H.; Li, X.; Wang, Y.; Prins, R.; Hu, Y. The Synthesis of Metal Phosphides: Reduction of Oxide Precursors in a Hydrogen Plasma. Angew. Chem., Int. Ed. 2008, 47, 6052–6054. 82 Pan, Y.; Liu, Y.; Zhao, J.; Yang, K.; Liang, J.; Liu, D.; Hu, W.; Liu, D.; Liu, Y.; Liu, C. Monodispersed Nickel Phosphide Nanocrystals with Different Phases: Synthesis, Characterization and Electrocatalytic Properties for Hydrogen Evolution. J. Mater. Chem. A. 2015, 3, 1656–1665. 83 Deng, Y.; Zhou, Y.; Yao, Y.; Wang, J. Facile Synthesis of Nanosized Nickel Phosphides with Controllable Phase and Morphology. New J. Chem. 2013, 37, 4083–4088. 84 Guan, J.; Wang, Y.; Qin, M.; Yang, Y.; Li, X.; Wang, A. Synthesis of Transition-Metal Phosphides from Oxidic Precursors by Reduction in Hydrogen Plasma. J. Solid State Chem. 2009, 182, 1550–1555. 85 Maneeprakorn, W.; Nguyen, C. Q.; Malik, M. A.; O’Brien, P.; Raftery, J. Synthesis of the Nickel Selenophosphinates [Ni(Se2PR2)2] (R = iPr, tBu and Ph) and Their Use as Single Source Precursors for the Deposition of Nickel Phosphide or Nickel Selenide Nanoparticles . J. Dalton Trans. 2009, 2103–2108. 86 Mi, K.; Ni, Y.; Hong, J. Solvent-Controlled Syntheses of Ni12P5 and Ni2P Nanocrystals and Photocatalytic Property Comparison. J. Phys. Chem. Solids 2011, 72, 1452–1456. 40

87 Chang, J.; Ning, Y. H.; Wu, S. L.; Zhang, S. F. Phase-Controllable Synthesis of Nickel Phosphides Using a New Ligand Myristic Acid. Adv. Mat. Res. 2013, 679, 23–26. 88 Moreau, L.; Ha, D.; Zhang, H.; Hovden, R.; Muller, D.A.; Robinson, R. D. Defining Crystalline/Amorphous Phases of Nanoparticles through X-ray Absorption Spectroscopy and X-ray Diffraction: The Case of Nickel Phosphide. Chem. Mater. 2013, 25, 2394−2403. 89 Pan, Y.; Liu, Y.; Liu, C.; An Efficient Method for the Synthesis of Nickel Phosphide Nanocrystals via Thermal Decomposition of Single-source Precursors. RSC Adv. 2015, 5, 11952–11959. 90 Muthuswamy, E.; Galbokka, H. L. S.; Brock, S. L. Synthetic Levers Enabling Independent Control of Phase, Size, and Morphology in Nickel Phosphide Nanoparticles. ACS Nano 2011, 5, 2402–2411. 91 Wang, J.; Yang, Q.; Zhang, Z.; Sun, S. Phase-Controlled Synthesis of Transition-Metal Phosphide Nanowires by Ullmann-Type Reactions. Chem. Eur. J. 2010, 16, 7916–7924. 92 Ni, Y.; Jina, L.; Hong, J. M. Phase-Controllable Synthesis of Nanosized Nickel Phosphides and Comparison of Photocatalytic Degradation Ability. Nanoscale, 2011, 3, 196–200. 93 Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; et al. Gaussian 03, revision D.01; Gaussian Inc.: Wallingford, CT, 2004. 94 Tao, J. M.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Climbing the Density Functional Ladder: Nonempirical Meta-generalized Gradient Approximation Designed for Molecules and Solids. Phys. Rev. Lett. 2003, 91, 12129–12137. 95 Staroverov, V. N.; Scuseria, G. E.; Tao, J. M.; Perdew, J. P. Comparative Assessment of a New Nonempirical Density Functional: Molecules and Hydrogen-bonded Complexes. J. Chem. Phys. 2003, 119, 12129–12137. 96 Fan, Y. B.; Bacon, S.; Kolade, A.; Wasiu, A.; Fan, H.-J. A DFT Study on the Mechanism of Wolff Rearrangement in a Fivemember Iridacycle, International Conference on Computational Science, 2010, 1, 2659–2667. 97 Hehre, W. J.; Ditchfie.R; Pople, J. A. Self-Consistent Molecular-Orbital Methods .12. Further Extensions of Gaussian-Type Basis Sets for Use in Molecular-Orbital Studies of Organic-Molecules. J. Chem. Phys. 1972, 56, 2257–2261. 98 Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. Self-Consistent Molecular-Orbital Methods .20. Basis Set for Correlated Wave-Functions. J. Chem. Phys. 1980, 72, 650–654. 99 Fan, Y. B.; Hall, M. B., The Competition between Allene and Butadiene in the Carbon- hydrogen Bond Activation Initiated by a Tungsten Allyl Complex: A DFT Study. Organometallics 2005, 24, 3827–3835. 100 Guan, Q.; Li, W.; Zhang, M.; Tao, K. Alternative Synthesis of Bulk and Supported Nickel Phosphide from the Thermal Decomposition of Hypophosphites. J. Catal. 2009, 263, 1–3. 101 Tolman, C. A. Steric Effects of Phosphorus Ligands in Organometallic Chemistry and Homogeneous Catalysis. Chem. Rev. 1977, 77, 313–348. 41

102 Fernandez, A. L.; Reyes, C.; Procka, A.; Gieringa, W. P. The Stereoelectronic Parameters of Phosphites. The Quantitative Analysis of Ligand Effects (QALE). J. Chem. Soc., Perkin Trans. 2000, 2, 1033–1041. 103 Janse van Rensburg, J. M.; Roodt, A.; Muller, A. Carbonyl(8-hydroxyquinolinato)[tris(2,4- di-tert-butylphenyl)phosphite] Rhodium(I) Acetone Hemisolvate Acta Cryst. E 2006, 62, M2978–M2980. 104 Anderson, D. A.; Tracy, J. B. Nanoparticle Conversion Chemistry: Kirkendall Effect, Galvanic Exchange, and Anion Exchange. Nanoscale 2014, 6, 12195–12216. 105 Chiang, R.-K.; Chiang, R.-T. Formation of Hollow Ni2P Nanoparticles Based on the Nanoscale Kirkendall Effect. Inorg. Chem. 2007, 46, 369–371. 106 Yin, Y.; Rioux, R. M.; Erdonmez, C. K.; Hughes, S.; Somorjai, G. A.; Alivisatos, A. P. Formation of Hollow Nanocrystals Through the Nanoscale Kirkendall Effect. Science 2004, 304, 711–714. 107 Wang, W.; Dahl, M.; Yin, Y. Hollow Nanocrystals through the Nanoscale Kirkendall Effect. Chem. Mater. 2013, 25, 1179–1189. 108 Cabot, A.; Ibáñez, M.; Guardia, P.; Alivisatos, A. P. Reaction Regimes of Hollow Particles by the Kirkendall Effect. J. Am. Chem. Soc. 2009, 131, 11326–11328. 109 Fan, H. J.; Knez, M.; Scholz, R.; Hesse, D.; Nielsch, K.; Zacharias, M.; Gösele, U. Influence of Surface Diffusion on the Formation of Hollow Nanostructures Induced by the Kirkendall Effect: The Basic Concept. Nano Lett. 2007, 7, 993–997. 110 Zhang, Q.; Wang, W.; Goebl, J.; Yin, Y. Self-Templated Synthesis of Hollow Nanostructures. Nano Today 2009, 4, 494–507. 111 Fan, H. J.; Gösele, U.; Zacharias, M. Formation of Nanotubes and Hollow Nanoparticles Based on Kirkendall and Diffusion Processes: A Review. Small 2007, 3, 1660–1671. 112 Fan, H. J.; Knez, M.; Scholz, R.; Nielsch, K.; Pippel, E.; Hesse, D.; Zacharias, M.;, Gösele U. Monocrystalline Spinel Nanotube Fabrication Based on the Kirkendall Effect. Nat. Mater. 2006, 5, 627–631. 113 Smigelskas, A. D.; Kirkendall, E. O. Zinc Diffusion in Alpha-Brass. Trans. Am. Inst. Mining Metall. Eng. 1947, 171, 130–142. 114 Railsback, J. G.; Johnston-Peck, A. C.; Wang, J.; Tracy, J. B. Size- Dependent Nanoscale Kirkendall Effect During the Oxidation of Nickel Nanoparticles. ACS Nano 2010, 4, 1913−1920. 115 Buhro, W. E. Metallo-Organic Routes to Phosphide Semiconductors. Polyhedron 1994, 13, 1131–1148. 116 Allen, P. M.; Walker, J. M.; Bawendi, M. G.; Mechanistic Insights into the Formation of InP Quantum Dots. Angew. Chem. Int. Ed. 2010, 49, 760–762. 117 Gary, C. D.; Glassy, B. A.; Cossairt, B. Investigation of Indium Phosphide Quantum Dot Nucleation and Growth Utilizing Triarylsilylphosphine Precursors. Chem. Mater. 2014, 26, 1734–1744. 42

118 Hanway, P. J.; Xue, J.; Bhattacharjee, U.; Milot, M. J.; Ruixue, Z.; Phillips, D. L.; Winter, A. H. Direct Detection and Reactivity of the Short-Lived Phenyloxenium Ion. J. Am. Chem. Soc. 2013, 135, 9078–9082. 119 Kampmann, S. S.; Sobolev, A. N.; Koutsantonis, G. A.; Stewart, S. G. Stable Nickel(0) Phosphites as Catalysts for C–N Cross-Coupling Reactions. Adv. Synth. Catal. 2014, 356, 1967–1973. 120 Meichel, E.; Stein, T.; Kralik, J.; Rheinwald, G.; Lang, H. Synthese von [(h5- C5H5)2Ti(Cl)(CºCSiMe3)]Ni(CO) und dessen Reaktionsverhalten Gegenüber Phosphiten: die Festkörperstruktur von (CO)2Ni[P(OC6H4CH3-2)3]2 J. Organomet. Chem. 2002, 649, 191–198. 121 Rosenthal, U.; Pulst, S.; Arndt, P.; Baumann, W.; Tillack, A.; Kempe, R. Zum Ligandeinflub in neuen Ni(0)-Komplexen disubstituierter Butadiine. Z. Naturforsch 1995, 50b, 368–376. 122 Guggenberger, L. J. Crystal Structures of (Acrylonitrile)bis(tri-o-tolyl phosphite)nickel and (Ethylene)bis(tri-o-tolyl phosphite)nickel. Inorg. Chem. 1973, 12, 499–508. 123 McKinney, R. J.; Roe, D. C. The Mechanism of Nickel-Catalyzed Ethylene Hydrocyanation. Reductive elimination by an Associative Process. J. Am. Chem. Soc. 1986, 108, 5167–5173. 124 Gillie, A.; Stille, J. K. Mechanisms of 1,l -Reductive Elimination from Palladium. J. Am. Chem. Soc. 1980, 102, 4933–4941. 125 Gardner, J. N.; Carlon, F. E.; Gnoj, O. One-Step Procedure for the Preparation of Tertiary α-Ketols from the Corresponding Ketones. J. Org. Chem. 1968, 33, 3294–3297.

Appendix of Supporting Information

Figure S1. Reproducibility. Powder XRD diffraction patterns of solids isolated after three separate repeat runs with either P(O-2,4-tBu2C6H4)3 (a) or P(OMe)3 (b) and nickel(II) chloride (all reactions were performed in oleylamine and octadecene at 275 °C).

Figure S2. Nickel precursor generality. Powder XRD diffraction patterns of solids isolated after similar runs with either nickel(II) chloride (a) or nickel(II) acetate (halide-free conditions) (b) and similar organophosphites (all reactions were performed in oleylamine and octadecene at 275 °C for 30 min). 44

Figure S3. Powder XRD patterns, solution phase optical density (absorption and scattering) spectra, and representative TEM data (after 30 min reaction) of nanocrystalline solids obtained by reacting nickel(II) chloride with the phosphite precursor P(OEt)3 in oleylamine and octadecene at 275 °C.

Figure S4. Powder XRD patterns, solution phase optical density (absorption and scattering) spectra, and representative TEM data (after 30 min reaction) of nanocrystalline solids obtained by reacting nickel(II) chloride with the phosphite precursor P(OiPr)3 in oleylamine and octadecene at 275 °C.

Figure S5. Powder XRD patterns, solution phase optical density (absorption and scattering) spectra, and representative TEM data (after 30 min reaction) of nanocrystalline solids obtained by reacting nickel(II) chloride with the phosphite precursor P(OnBu)3 in oleylamine and octadecene at 275 °C.

Figure S6. Solution phase optical density (absorption and scattering) of nanocrystals obtained from reacting nickel(II) chloride and the phosphite precursors P(OMe)3 (a), P(OCH2tBu)3 (b), P(OPh)3 (c), or P(O-2,4-tBu2C6H3)3 (d) in oleylamine and octadecene at 275 °C. These optical spectra did not show any obvious characteristic features or systematic variations that could be attributed to the evolution of different phases or particle morphologies, which prompted us to pursue density of states (DOS) calculations (see below).

Electronic structure of Ni2P. During our study, we attempted to analyze solution phase optical spectra (see above), but could not find a straightforward set of changes or patterns during the course of the synthetic reactions. To start probing this question, we calculated the electron structure of the final phosphide product, Ni2P using the Vienna ab initio simulation package (VASP).S1,S2 Projected augmented-wave (PAW) pseudopotentials were used with a cutoff energy of 500 eV and a convergence energy of 1x10-6 eV.S3 A conjugated algorithm was applied for the structural optimization with a 15 x 15 x 15 Monkhorst-pack k-point grid.S4 Total energy was calculated using the tetrahedron method with BlöchlS5 corrections applied.

Exchange and correlation were treated by Perdew Burke-Enzerhoff (PBE).S6

As shown in Figure S7 (below), the Density of States (DOS) curve for perfect, stoichiometric Ni2P has a local maximum at the Fermi level, indicating that this compound should be metallic according to traditional band theory. Interestingly, Ni2P was reported to be a semiconductor with an experimental band gap of 1.0 eV.S7,S8 A rationale for both the apparent semiconducting nature of this compound and the lack of predictable changes in the optical spectra of our Ni2P-containing solutions could be the presence of variable amounts of defects.

- This rationale is supported by the presence of a pseudogap at 2.6 e per unit cell below the Ni2P

Fermi level, the local minima of which corresponds to the transition from bonding to antibonding states. The structure may therefore be prone to losing electrons by way of Ni or P vacancies, thus moving the Fermi level to the DOS minimum located 0.7 eV below the bulk

Fermi level, and alleviating the population of antibonding states.

Figure S7. Density of State (DOS) curve for Ni2P. Total DOS is depicted in black. Partial DOS (pDOS) are given in blue and purple for nickel and phosphorus, respectively. Dashed guidelines are provided to indicate the Fermi level (black) and minimum in the pseudogap (red).

Table S1. Calculated Bond Dissociation (Homolysis and Heterolysis) Energies (Gibbs free energies, ΔG, see Experimental section, under Computations) for Phosphite Precursors in the Gas Phase. Precursor P–O C–O P+|O- P-|O+ PO-|R+ PO+|R- homolysis homolysis heterolysis heterolysis heterolysis heterolysis (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) P(OMe)3 63.16 40.63 193.69 290.48 225.66 228.04

P(OEt)3 62.69 39.43 186.38 275.69 184.80 223.45

P(OnBu)3 62.49 34.52 181.27 259.04 154.75 213.90

P(OCH2tBu)3 62.97 42.81 174.24 248.86 136.54 210.65

P(OiPr)3 61.21 31.34 172.07 263.48 157.90 207.19

P(OPh)3 46.43 54.39 149.51 218.25 177.01 210.50

P(O-2,4- 36.34 45.55 125.75 174.91 141.51 182.31 tBu2C6H3)3

Table S2. Calculated Bond Dissociation (Homolysis and Heterolysis) Energies (Gibbs free energies, ΔG, see Experimental section, under Computations) for Phosphite Precursors in Cyclohexane. Precursor P–O C–O P+|O- P-|O+ PO-|R+ PO+|R- homolysis homolysis heterolysis heterolysis heterolysis heterolysis (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) P(OMe)3 65.35 42.62 134.97 228.24 161.00 165.40

P(OEt)3 65.12 41.40 131.85 218.27 125.97 166.79

P(OnBu)3 65.07 36.63 129.83 206.99 74.45 161.90

P(OCH2tBu)3 62.36 41.65 125.95 198.92 87.45 161.18

P(OiPr)3 63.56 37.46 120.62 209.69 75.58 155.56

P(OPh)3 49.29 57.54 103.33 171.04 128.74 161.84

P(O-2,4- 35.23 44.25 87.62 141.22 102.72 143.58 tBu2C6H3)3

Table S3. Calculated Bond Dissociation (Homolysis and Heterolysis) Energies (Gibbs free energies, ΔG, see Experimental section, under Computations) for Phosphite Precursors in Water. Precursor P–O C–O P+|O- P-|O+ PO-|R+ PO+|R- homolysis homolysis heterolysis heterolysis heterolysis heterolysis (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) P(OMe)3 59.62 35.90 78.35 165.66 97.21 103.62

P(OEt)3 59.01 34.65 77.69 159.34 66.75 109.87

P(OnBu)3 59.00 29.89 77.06 151.97 74.21 107.39

P(OCH2tBu)3 61.46 39.94 76.16 145.69 37.01 108.96

P(OiPr)3 57.52 30.83 69.40 153.89 77.76 104.22

P(OPh)3 39.75 48.51 56.73 122.85 79.56 111.10

P(O-2,4- 33.56 42.31 46.44 99.12 60.55 100.95 tBu2C6H3)3

Yield calculation for reaction with P(OPh)3:

2NiCl + P(OPh) Ni P + byproducts 2 3 2 0.1 mmol 0.05 mmol

Theoretical yield of Ni2P (without volatiles) = 7.41 mg

Mass of Ni2P + volatiles (organics, water) content = 6.90 mg

Volatiles content from TGA analysis = 2.62 mg (~38% of mass loss)

Experimental yield of Ni2P = 4.28 mg

Percentage yield = 58%

Figure S8. TGA/DSC curve of Ni2P nanocrystals obtained from reacting nickel(II) chloride with P(OPh)3 in oleylamine and octadecene at 275 °C for 30 min.

Yield calculation for reaction with P(OEt)3:

Ni P + byproducts 2NiCl2 + P(OEt)3 2 0.1 mmol 0.05mmol

Theoretical yield of Ni2P (without volatiles) = 7.41 mg

Mass of Ni2P + volatiles (organics, water) content = 3.86 mg

Volatiles content from TGA analysis = 0.58 mg (~15% of mass loss)

Experimental yield of Ni2P = 3.28mg

Percentage yield = 44%

Figure S9. TGA/DSC curve of Ni2P nanocrystals obtained from reacting nickel(II) chloride with P(OEt)3 in oleylamine and octadecene at 275 °C for 30 min.

Appendix References

S1 Kresse, G.; Furthmüller, J. Efficiency of Ab Initio Total Energy Calculations for Metals and Semiconductors using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6, 15–50. S2 Kresse, G. Efficient Iterative Schemes for Ab Initio Total-Energy Calculations using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54, 11169–11186. S3 Kresse, G. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B 1999, 59, 1758–1775. S4 Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B 1976, 13, 5188–5192. S5 Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953–17979. S6 Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865–3868. S7 Sharon, M.; Tamizhmani, G.; Levy-Clement, C.; Rioux, J. Study of Electrochemical and Photoelectrochemical Properties of Nickel Phosphide Semiconductors. Solar Cells 1989, 26, 303–312. S8 Panneerselvam, A.; Malik, M. A.; Afzaal, M.; O’Brien, P.; Helliwell, M. The Chemical Vapor Deposition of Nickel Phosphide or Selenide Thin Films from a Single Precursor. J. Am. Chem. Soc. 2008, 130, 2420–2421.

CHAPTER 3.

EVOLUTION OF NICKEL PHOSPHIDE NANOCATALYSTS DURING

PHENYLACETYLENE HYDROGENATION

Himashi P. Andaraarachchi, Avipsa Ghosh, Long Men, Kevin Vickerman, Levi, M. Stanley,

Javier Vela

Manuscript in Preparation

Abstract

Nickel phosphides with various stoichiometries are predicted to be potential catalysts for hydrogen evolution reactions, hydrodesulfurization, and hydrodeoxygenation reactions.

While several studies provide a foundation for the catalytic potential of Ni2P in alkene and alkyne hydrogenation reactions, we have conducted a systematic study on how size and morphology of Ni2P nanocrystals would affect the catalytic activity of phenylacetylene hydrogenation. We observed that hollow, small Ni2P nanoparticles are more catalytically active than bimodal (a mixture of hollow and solid) Ni2P particles. We showed that catalytic activity increases over the reaction cycles with increase of yield and selectivity for both morphologies. Our studies revealed that Ni2P nanocrystals can be recycled up to eight times without losing their catalytic activity. We probed the structural integrity of these nanocatalysts after recycling via powder X-ray diffraction, TEM, and XPS. Morphology and size are observed to be not significantly altered after recycling, which demonstrates that nanocrystalline Ni2P particles are a robust hydrogenation catalyst. XPS analyses revealed the presence of acidic oxidized Ni and P species on recycled Ni2P catalyst surface could be account for enhance catalytic activity. 56

Introduction

Transitions metal phosphides have gained extensive research interest due to their superior catalytic activity in hydrotreating reactions and hydrogen generation reactions.1-5

Nickel phosphides, mainly Ni2P, are proven to be stable and active catalysts for hydrodesulfurization, hydrodeoxygenation, and hydrodenitrification reactions.6-10

Furthermore, Ni2P nanostructures with a high accessible surface area and a high density of exposed (001) facets are promising earth abundant HER catalysts stable in both acidic and basic conditions with low over potential values.6-10 Their HER catalytic activity is one of the

17 highest among non-precious metals. In addition to extensively studied Ni2P, nickel rich

18 Ni12P5 display catalytic activity in oxygen evolution reactions (OER). Thus, nickel phosphides can serve as a dual catalyst for electrochemical water splitting reaction where Ni2P

18 act as a HER catalyst and Ni12P5 act as an OER catalyst.

Ni2P nanocatalysts have also been studied in the context of selective hydrogenation of alkynes. Well-defined, 25 nm Ni2P nanoparticles acted as colloidal catalysts for the chemoselective hydrogenation of terminal and internal alkynes.19 The activity of various alumina-supported nickel phosphides (Ni12P5, Ni2P, and NiP2) with particle sizes ranging from

20 5 to 15 nm toward selective hydrogenation of phenylacetylene was evaluated. Ni2P/Al2O3 exhibited higher selectivity for styrene formation (up to 88%) compared to Ni12P5/Al2O3

(48%), NiP2/Al2O3 (66%), and Ni/Al2O3 (1%) catalysts. X-ray absorption fine structure

(EXAFS) and in situ IR measurements revealed that the incorporation of P increases the Ni−Ni bond distance, which imposes a key influence on the adsorption state of alkene intermediates.

These studies further demonstrated that electron transfer occurs from Ni to P, leading to positively charged Ni, reduced desorption energy for the alkene, and improved reaction 57

selectivity. While these studies provide a foundation for the catalytic potential of Ni2P in alkene and alkyne hydrogenation reactions, further studies are necessary to establish the fundamental principles governing the activity of Ni2P in such reactions.

Several methods exist for the synthesis of Ni2P nanocrystals with different morphologies including chemical vapor deposition, solution phase (colloidal) synthesis, solvothermal synthesis, electrosynthesis, hydrothermal synthesis, solid-state synthesis, microwave synthesis, and temperature-programmed phosphate reduction.21-30 We synthesized phase pure colloidal Ni2P nanocrystals using triphenyl phosphite as the

31 phosphorus precursor and Ni(OAc)2 and NiCl2 as nickel precursors.

Herein, we report a systematic study detailing the effects of size, morphology and preparation method of Ni2P nanocrystals on its catalytic activity toward hydrogenation of phenylacetylene. We further discuss the effects of recycling on the catalyst surface through characterization of recycled Ni2P nanocrystals via powder X-ray diffraction (XRD), transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS) analyses.

Experimental Section

Materials. Nickel(II) chloride hexahydrate (NiCl2•6H2O, 99.9%) and triphenylphosphite (P(OPh)3, 97%), were purchased from Strem; 1-octadecene (ODE, 90%) and oleylamine (80–90%) from Acros; nickel(II) acetate (Ni(OAc)2, 98%), phenylacetylene

(98%), styrene (99%), ethylbenzene (99%), dodecane (99%), 1,4-dioxane (99.9%) and 1- propanol (99.5%) from Aldrich. All compounds were used as received and handled under an inert (dry-N2 or Ar) atmosphere inside a glove box or with an Schlenk line.

31 Synthesis. Ni2P was synthesized by a slightly modified literature procedure. Inside a three-neck round-bottom flask, NiCl2•6H2O (0.127 g, 0.50 mmol) or Ni(OAc)2 (90 mg, 0.50 mmol), oleylamine (1.32 g, 5.00 mmol, 1.65 mL), and ODE (5.00 g, 6.34 mL) were degassed under vacuum at 80 ºC for 1 h, refilled with dry Ar, and heated to 275 ºC. After 5 min, a solution of P(OPh)3 (2.0 mmol, 0.55 ml) in ODE (1.00 g, 1.27 ml) was quickly injected. After

30 min at 275 ºC, the mixture was allowed to cool to room temperature (R.T., 21 °C), and the nanocrystalline products isolated by twice washing with toluene and centrifugation at 4900 rpm for 5 min.

Phenylacetylene hydrogenation with Ni2P nanocrystals. General procedure. In an

N2-filled drybox, Ni2P nanocrystals (13.4 mg, 0.0900 mmol, 0.300 equiv), phenylacetylene

(30.7 mg, 0.300 mmol, 1.00 equiv), a magnetic stirring bar, and 1,4-dioxane (1.0 mL) were added to a 20 mL vial. The vial was sealed with a PTFE septum and removed from the glovebox. The septum was pierced with a 20-gauge needle and quickly placed in a pressure reactor. The pressure reactor was sealed and pressurized with H2 gas at 40 bar. The pressure reactor was placed in a pre-heated oil bath at 100 °C and the reaction mixture was stirred at the same temperature for 14 h. The pressure reactor was removed from the oil bath after 14 h of stirring and cooled to R.T. The H2 gas was then released and the vial removed from the reactor.

Dodecane (22.6 µL, 0.300 mmol) was added to the mixture as an internal standard. 1,4-dioxane

(3.0 mL) was added to the mixture and this stirred at for 5 min. The mixture was centrifuged for 15 min at 5000 rpm and the supernatant liquid was injected for gas chromatographic (GC) analysis. The conversion of phenylacetylene and yields of styrene and ethylbenzene were calculated from the relative area of dodecane and analyte using a GC calibration curve. 59

Catalyst recycling. After conducting the procedure above, the solid catalyst was further washed twice with 3.0 mL solvent and isolated by centrifugation at 5000 rpm for 15 min. After removing the supernatant, nanocrystalline solids were used as a catalyst for the next reaction.

Material characterization. Optical characterization. Absorption spectra were measured with a photodiode array Agilent 8453 UV–Vis spectrophotometer. Structural

Characterization. Powder X-ray Diffraction (XRD) was measured using Cu Kα radiation on a

Rigaku Ultima U4 diffractometer. Transmission Electron Microscopy (TEM) was conducted on carbon-coated copper grids using a FEI Tecnai G2 F20 field emission scanning transmission electron microscope (STEM) at 200 kV (point-to-point resolution <0.25 nm, line to line resolution < 0.10 nm). Particle dimensions were measured manually and/or with ImageJ for

>50–100 particles. Averages are reported ± one standard deviation. X-ray photoelectron spectroscopy (XPS) measurements were performed using a Kratos Amicus/ESCA 3400 instrument. The sample was irradiated with 240 W non-monochromated Mg Kα x-rays, and photoelectrons emitted at 0°C from the surface were analyzed using a DuPont-type analyzer.

The pass energy was set at 75 eV. CasaXPS was used to process raw data files. The binding energy of C 1s at 284.6 eV was used as a reference.

Results and Discussion

Synthesis and Characterization of Crystalline Ni2P Catalysts. Ni2P nanocrystals were synthesized by injecting a P(OPh)3 into a solution containing a nickel precursor (NiCl2 or

Ni(OAc)2) and oleylamine in 1-octadecene (ODE) at 275 °C (Scheme 1). We structurally characterized these nanocrystals using powder X-ray diffraction (XRD) and transmission electron microscopy (TEM). The diffraction peaks in the powder XRD patterns of the resulting nanocrystals from both nickel sources were corresponded to the hexagonal structure of Ni2P 60

(Figure 1) indicating the formation of phase pure crystalline Ni2P. The diffraction peaks at

40.74, 44.74, 47.32, and 54.18o can be indexed to the (111), (021), (210), and (002) reflections.

Scherrer analysis revealed single crystalline domains with an average size of 17 ± 4 nm.

However, TEM images revealed that while NiCl2 produces a mixture of solid and hollow Ni2P particles, Ni(OAc)2 produces only hollow Ni2P crystals (Figure 2 and Table 1).

Scheme 1. General synthesis for the Ni2P nanocatalysts.

275 °C, 30 min NiX + P(OPh) Ni P (unbalanced) 2 3 octadecene, 2 oleylamine Hollow (X = OAc) Solid & hollow (X = Cl)

a Table 1. Synthesis of Ni2P Hydrogenation Nanocatalysts.

Precursors XRD Size TEM Size TEM (nm)b (nm) Morphologyc

NiCl2 + 17 ± 4 14 ± 3 Solid P(OPh)3 58 ± 5 Hollow

Ni(OAc)2 + 17 ± 2 29 ± 6 Hollow P(OPh)3

aAccording to Scheme 1. bAverage size of single crystalline domains from Scherrer equation. cWhen more than one observed.

Figure 1. Powder XRD patterns and TEM images of (A) as synthesized Ni2P nanocrystals from

NiCl2 and (B) as synthesized Ni2P nanocrystals from Ni(OAc)2. Decoding Morphology of Ni2P Hydrogenation Catalysts. The formation of hollow

32-37 Ni2P nanocrystals are due to the nanoscale Kirkendall effect. Faster inward diffusion of

2+ phosphide anions, compared to the slower Ni outward diffusion causes hollow Ni2P nanocrystals.28,29 The net effect was that all the Ni2+ would move away from the center and get converted into Ni2P, leaving a hollow core behind. TEM images revealed hollow nanocrystals with larger voids (51 ± 5 nm) from NiCl2 and comparably smaller voids (25 ± 1 nm) from

38 Ni(OAc)2 precursor. As observed by certain systems, the bimodal distribution might be due to the formation of cracks in the thin shell of Ni2P particles that ultimately resulting smaller dense particles along with large hollow particles (Figure 2).

Figure 2. Schematic representation of the formation of hollow nanoparticles due to the nanoscale Kirkendall effect.

Ni2P-Catalyzed Hydrogenation. To evaluate the catalytic activity of Ni2P nanocrystals, we studied phenylacetylene hydrogenations at 40 bar of H2 and 100 °C for 14 h in 1,4-dioxane in the presence of 30 mol% Ni2P synthesized from both nickel(II) precursors (Scheme 2). The initial hydrogenation in 1,4-dioxane with bimodal Ni2P catalysts —henceforth referred to as

Cycle 1, see below—formed 23% styrene and 5% ethylbenzene, a total yield of 28% with a selectivity for styrene over ethylbenzene of 4.6:1 (Table 2). Under the same conditions only hollow Ni2P catalyzed hydrogenation formed 55% styrene and 9% ethylbenzene with a total yield of 64% with a selectivity for styrene over ethylbenzene. Hollow Ni2P particles exhibits higher catalytic activity than bimodal Ni2P particles under these reaction conditions.

Scheme 2. Selective phenylacetylene hydrogenation with Ni2P nanocatalysts.

Ni2P (30 mol %) H2 (40 bar) + dioxane (0.3 M) 100 °C, 14 h

a Table 2. Selective phenylacetylene hydrogenation with Ni2P nanocatalysts.

Run Catalyst Cycle Conversion Total Yield % Selectivity (C6H5CHCH2

to C6H5CH2CH3)

(%) (C6H5CHCH2 +

C6H5CH2CH3)

1 Bimodal Ni2P (solid & 1 40 28 4.6:1

hollow, from NiCl2) 2 99 69 9.0:1

3 99 67 2.3:1

2 Ni2P (hollow only, 1 99 64 6.1:1

from Ni(OAc)2) 2 99 90 1:44

3 99 87 0:1

4 99 86 0:1

5 99 85 0:1

6 99 83 0:1

7 99 83 0:1

8 99 89 0:1

a 30 mol% of catalyst at 40 bar of H2 and 100 °C for 14h in dioxane.

Ni2P Catalyst Recycling. To test if the Ni2P nanocrystals used in this initial test could be recycled, we isolated the Ni2P solids from the reaction mixture by centrifugation and used them for additional hydrogenation reactions under identical conditions. With bimodal Ni2P nanocatalysts, Cycle 2 formed 63% styrene and 7% yield ethylbenzene, an increase in the total yield to 69% and an increase in the selectivity towards styrene to 9:1 (Table 2 and Figure 3).

These observations prompted us to further isolate and recycle the same Ni2P solids.

Interestingly, cycle 3 formed 47% styrene and 20% ethylbenzene, a 67% total yield similar to that in the previous cycle, although the selectivity toward the formation of styrene was reduced to 2.3:1. However with hollow only Ni2P nanocatalysts, Cycle 2 formed 2% styrene and 98% ethylbenzene, an increase in the total yield to 90% and an increase in the selectivity towards ethylbenzene to 1:44. We recycled same hollow Ni2P solid up to eight times without losing its catalytic activity (Table 2 and Figure 4). We also carried out this reaction in 1-propanol under similar conditions and results were summarized in Supporting Information. While Bimodal

Ni2P catalyzed hydrogenation in 1-propanol demonstrated comparable catalytic activity over recycling, under the same reaction conditions hollow Ni2P catalyzed hydrogenation carried out in 1-propanol demonstrated a continual drop of total yield of products over recycling (See

Supporting Information). 65

Figure 3. Conversion, total yield, and yield of styrene and ethylbenzene with bimodal Ni2P nanocatalysts in dioxane.

Figure 4. Conversion, total yield, and yield of styrene and ethylbenzene with hollow Ni2P nanocatalysts.

Structural Integrity. We performed XRD and TEM analyses of the isolated Ni2P solids to probe the integrity of the catalyst after each cycle of recycling and are summarized in Figure

5 and supporting information. Hollow Ni2P solids were recycled up to 8 times and XRD patterns after each cycle up to cycle 7 correspond to crystalline Ni2P. However, after cycle 8

XRD pattern did not exhibit any features that indicates loss of crystallinity (Figure 5). TEM 66 images revealed hollow nanoparticles without any significant difference in total diameter of the particles or diameter of the shell of the particles (Figure 6 and SI).

Figure 6. Powder XRD patterns of (A) Ni2P from Ni(OAc)2 (B) Ni2P after cycle 1 (C) Ni2P after cycle 2 (D) Ni2P after cycle 3 (E) Ni2P after cycle 4 in (F) Ni2P after cycle 8.

Figure 7. TEM images of (A) Ni2P from Ni(OAc)2 (B) Ni2P after cycle 1 (C) Ni2P after cycle 2 (D) Ni2P after cycle 3 (E) Ni2P after cycle 4 in (F) Ni2P after cycle 8. Average Ni2P nanocrystal diameters were determined by TEM to be: (A) 29 ± 6 nm, (B) 27 ± 3 nm, (C) 28 ± 4 nm, (D) 22 ± 4 nm, (E) 28 ± 5 nm, (F) 28 ± 3 nm. 67

Surface Chemistry. We characterized the chemical states of Ni and P in the as synthesized and isolated Ni2P nanocrystals after few recycles by XPS. Ni2P nanoparticles derived from NiCl2 and Ni(OAc)2 shows a peak around 853.4 eV for the Ni 2p3/2 energy level,

δ+ 39-41 corresponding to Ni in Ni2P nanocrystals (Figure 8 and Supporting Information). This binding energy (853.4 eV) is very close to the binding energy of zero valent Ni metal, 852.6

42-44 δ+ eV, which indicates that Ni2P has a very small positive charge (Ni ). Furthermore, two peaks can be observed in the P 2p3/2 energy region at 129.8 eV and 133.4 eV. The 129.8 eV is lower than the binding energy of elemental phosphorus (130.3 eV),45 which indicates the presence of anionic P species (Pδ−). In addition, 133.4 eV corresponds to the oxidized P species

3- - 39 (PO4 and HPO3H ) formed on the surface of the nanocrystals.

Isolated Ni2P nanocrystals after the hydrogenation of phenylacetylene exhibits extra peaks in the Ni 2p that indicate different Ni chemical states on the Ni2P surface. After the reaction, three peaks observed at 853.4, 856.9, and 863.1 eV correspond to Niδ+, oxidized Ni

18 species, and satellite peak of Ni 2p3/2 respectively. After the reactions, P XPS exhibits two peaks at ~130 eV and ~133.5-134.3 eV that correspond to Pδ− and oxidized P species respectively. Furthermore, we observed that oxidized Ni and P species peaks emerge and become more prominent and Niδ+ and Pδ− peaks become inconspicuous after recycles. This might be due to the air exposure of samples during the recycling and characterization process.

These observations are also consistent with the other system that we studied with bimodal Ni2P catalysts in 1,4-dioxane. (See Supporting Information). It is observed that superior hydrogenation catalytic activity is related to the synergistic effect between Ni2P nanocrystals and their acidic properties. The acidic sites will adsorb the reactants during the hydrogenation reaction, and the dissociated hydrogen species can then migrate to nearby acidic sites, which 68

46 could result in an improvement of the catalytic activity. Recycled Ni2P exhibit superior catalytic activity than pristine Ni2P and XPS spectra shows existence of oxidized Ni and P species on the recycled catalyst surface. Previous reports indicate that oxidized Ni2+ are accountable for Lewis acidity and oxidized P species in the form of hydrogen phosphite could

47 be associated with Brønsted acid sites. Thus, higher catalytic activity of recycled Ni2P could be explained by the enhanced acidic sites on the nanocrystal surface.

Figure 8. XPS of A) Ni2P (from Ni(OAc)2) B) Ni2P after trial 1 C) Ni2P after trial 2 D) Ni2P after trial 3 E) Ni2P after trial 4 F) Ni2P after trial 8 in 1,4-dioxane.

Decoupling the Cl- effect. Chloride ions are considered as catalytic poison for hydrotreating reactions. To probe whether there is any impact on hydrogenation reactions, we

- analyzed the chloride ion content of Ni2P catalysts by ICPMS from both nickel sources. Cl content from as synthesized and purified Ni2P catalysts from NiCl2 contain 11 mol% and Ni2P from Ni(OAc)2 is below the detection limit. We conducted a control experiment of hydrogenation of phenylacetylene with Ni2P catalysts synthesized from Ni(OAc)2 spiked with

- 0.2 equivalent of NaCl and KCl. Our observations revealed that Ni2P spiked with Cl ions exhibit similar total yield and selectivity compared to the pristine Ni2P from Ni(OAc)2. NaCl 69

added Ni2P resulted a total yield of 60% and KCl added Ni2P resulted a total yield of 66% as

- compared to the 64% of total yield of pristine Ni2P. Thus, we concluded that presence of Cl ions does not hinder the hydrogenation catalytic activity of Ni2P nanocrystals.

Conclusions

In conclusion, we have conducted a systematic study of the hydrogenation of phenylacetylene catalyzed by Ni2P nanocrystals at 100 °C with 40 bar of H2 in 1,4-dioxane.

NiCl2 produced a bimodal distribution of Ni2P nanocrystals with both hollow and solid morphologies and Ni(OAc)2 produced a single morphology of hollow Ni2P nanocrystals. Both hollow Ni2P particles and a mixture of hollow and solid Ni2P particles exhibit significant catalytic activity towards the hydrogenation of phenylacetylene. However, Ni2P nanoparticles derived from Ni(OAc)2 revealed superior catalytic activity over Ni2P nanoparticles derived from NiCl2. Recycling studies exhibited an increases in catalytic activity over reaction cycles for both catalysts. Hydrogenation of phenylacetylene with hollow Ni2P catalysts, synthesized from Ni(OAc)2, in 1,4-dioxane was recycled up to 8 times without losing its catalytic activity.

However, the same reaction condition with hollow Ni2P in 1-propanol exhibited a gradual drop in total yield of products over recycling. Hydrogenation of phenylacetylene with bimodal Ni2P catalysts can be recycled in both solvents without losing its catalytic activity. Ni2P nanocrystals isolated from each reaction were structurally characterized using XRD, TEM, and XPS.

Nanocrystalline Ni2P was observed with hollow morphology without a significant morphology or size change over recycling. XPS analyses revealed formation of oxidized Ni(II) and P(V) species on the surface after hydrogenation reaction and might be accountable for enhanced catalytic activity due to the increasing acidic sites.

Acknowledgements

J. Vela thanks the US National Science Foundation for a CAREER grant from the

Division of Chemistry, Macromolecular, Supramolecular and Nanochemistry program

(1253058). L. M. Stanley thanks the US National Science Foundation for a CAREER grant from the Division of Chemistry, Chemical Synthesis program (1353819).

References

1 Prins, R.; Bussell, M. E. Metal Phosphides: Preparation, Characterization and Catalytic Reactivity. Catal. Lett. 2012, 142, 1413−1436.

2 Alexander, A.-M.; Hargreaves, J. S. J. Alternative Catalytic Materials: Carbides, Nitrides, Phosphides and Amorphous Boron Alloys. Chem. Soc. Rev. 2010, 39, 4388–4401.

3 Morales-Guio, C. G.; Stern, L.-A.; Hu, X. Nanostructured Hydrotreating Catalysts for Electrochemical Hydrogen Evolution. Chem. Soc. Rev. 2014, 43, 6555–6569.

4 Carenco, S.; Portehault, D.; Boissière, C.; Mézailles, N.; Sanchez, C. Nanoscaled Metal Borides and Phosphides: Recent Developments and Perspectives. Chem. Rev. 2013, 113, 7981−8065.

5 Callejas, J. F.; Read, C. G.; Roske, C. W.; Lewis, N. S.; Schaak, R. E. Synthesis, Characterization, and Properties of Metal Phosphide Catalysts for the Hydrogen-Evolution Reaction. Chem. mater. 2016, 17, 6017−6044.

6 Brock, S. L.; Senevirathne, K. Recent Developments in Synthetic Approaches to Transition Metal Phosphide Nanoparticles for Magnetic and Catalytic Application. J. Solid State Chem. 2008, 181, 1552−1559.

7 Carenco, S.; Boissière, C.; Mézailles, N.; Sanchez, C. Metal phosphides: A Revival at the Nanoscale. Actual Chimique, 2012, 362, 22–28.

8 Wang, H. M.; Male, J.; Wang, Y. Recent Advances in Hydrotreating of Pyrolysis Bio-Oil and Its Oxygen-Containing Model Compounds. ACS Catal. 2013, 3, 1047–1070.

9 Shamanaev, I. V.; Deily, I. V.; Aleksandrov, P. V.; Gerasimov, E. U.; Pakharukova, V. P.; Kodnev, E. G.; Ayupov, A. B.; Andreev, A. S.; Lapina, O. B.; Bukhtiyarova, G. A. Effect of Precursor on the Catalytic Properties of Ni2P/SiO2 in Methyl Palmitate Hydrodeoxygenation. RSC. Adv. 2016, 6, 30372-30383.

10 Zhao, H. Y.; Li, D.; Bui, P.; Oyama, S. T. Hydrodeoxygenation of Guaiacol as Model Compound for Pyrolysis Oil on Transition Metal Phosphide Hydroprocessing Catalysts. Appl. Catal. A 2011, 391, 305–310.

11 Zou, X.; Zhang, Y. Noble Metal-Free Hydrogen Evolution Catalysts for Water Splitting. Chem. Soc. Rev. 2015, 44, 5148–5180.

12 Faber, M. S.; Jin, S. Earth-Abundant Inorganic Electrocatalysts and Their Nanostructures for Energy Conversion Applications. Energy Environ. Sci. 2014, 7, 3519–3542.

13 Roger, I.; Shipman, M. A.; Symes, M. D. Earth-Abundant Catalysts for Electrochemical and Photoelectrochemical Water Splitting. NatRevChem 2017, 1, 0003.

14 Popczun, E. J.; McKone, J. R.; Read, C. G.; Biacchi, A. J.; Wiltrout, A. M.; Lewis, N. S.; Schaak, R. E. Nanostructured Nickel Phosphide as an Electrocatalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2013, 135, 9267−9270.

15 Cao, S.; Chen, Y.; Wang, C.-J.; Hea, P.; Fu, W.-F. Highly Efficient Photocatalytic Hydrogen Evolution by Nickel Phosphide Nanoparticles from Aqueous Solution. Chem. Commun. 2014, 50, 10427–10429. 16 Feng, L.; Vrubel, H.; Bensimonb, M.; Hu, X. Easily-prepared Dinickel Phosphide (Ni2P) Nanoparticles as an Efficient and Robust Electrocatalyst for Hydrogen Evolution. Phys. Chem. Chem. Phys. 2014, 16, 5917–5921.

17 Wang, J.; Xu, F. Haiyan, J.; Yiqing, C.; Wanh, Y. Non-Noble Metal-based Carbon Composites in Hydrogen Evolution Reaction: Fundamentals to Applications. Adv. Mater. 2017, 1605838.

18 Menezes, W. P.; Indra, A.; Das, C.; Walter, C.; Göbel, C.; Gutkin, G.; Schmeiβer, V.; Driess, M. Uncovering the Nature of Active Species of Nickel Phosphide Catalysts in High- Performance Electrochemical Overall Water Splitting. ACS. Catal. 2017, 7, 103–109.

19 Carenco, S.; Leyva-Pérez, A.; Concepción, P.; Boissière, C.; Mézailles, N.; Sanchez, C.; Corma, A. Nickel Phosphide Nanocatalysts for the Chemoselective Hydrogenation of Alkynes. Nano Today 2012, 7, 21–28.

20 Chen, Y.; Li, C.; Zhou, J.; Zhang, S.; Rao, D.; He, S.; Wei, M.; Evans, D. G.; Duan, X. Metal Phosphides Derives from Hydrotalcite Precursors toward the Selective Hydrogenation of Phenylacetylene. ACS Catal. 2015, 5, 5756-5765.

21 Carenco, S.; Portehault, D.; Boissière, C.; Mézailles, N.; Sanchez, C. 25th Anniversary Article: Exploring Nanoscaled Matter from Speciation to Phase Diagrams: Metal Phosphide Nanoparticles as a Case of Study. Adv. Mater. 2013, 26, 371−390.

22 Brock, S. L.; Perera, S. C.; Stamm, K. L. Chemical Routes for Production of Transition- Metal Phosphides on the Nanoscale: Implications for Advanced Magnetic and Catalytic Materials. Chem. Eur. J. 2004, 10, 3364–3371.

23 Vasquez, Y.; Henkes, A.; Bauer, J. C.; Schaak. R. E. Nanocrystal Conversion Chemistry: A Unified and Materials-General Strategy for the Template-Based Synthesis of Nanocrystalline Solids. J. Solid State Chem. 2008, 181, 1509–1523.

24 Vilas G. Pol, V. G.; Pol, S. V.; Gedanken, A. Dry Autoclaving for the Nanofabrication of Sulfides, Selenides, Borides, Phosphides, Nitrides, Carbides, and Oxides. Adv. Mater. 2011, 23, 1179–1190.

25 Prieto, G.; Tüysüz, H.; Duyckaerts, N.; Knossalla, J.; Wang, G.-H.; Schüth, F. Hollow Nano- and Microstructures as Catalysts. Chem. Rev. 2016, 116, 14056−14119.

26 Munnik, P.; De Jongh, P. E.; De Jong, K. P. Recent Developments in the Synthesis of Supported Catalysts. Chem. Rev. 2015, 115, 6687–6718.

27 Wang, X.; Feng, J.; Bai, E.; Yin, Y. Synthesis, Properties, and Applications of Hollow Micro-/Nanostructures. Chem. Rev. 2016, 116, 10983−11060.

28 Chiang, R.-K.; Chiang, R.-T. Formation of Hollow Ni2P Nanoparticles Based on the Nanoscale Kirkendall Effect. Inorg. Chem. 2007, 46, 369–371.

29 Henkes, A.E.; Vasquez. Y.; Schaak, R.E. Converting Metals into Phosphides: A General Strategy for the Synthesis of Metal Phosphide Nanocrystals. J. Am. Chem. Soc. 2007, 129, 1896−1897.

30 Pickett, N. L.; O’Brien, P. Syntheses of Semiconductor Nanoparticles Using Single- Molecular Precursors. Chem. Rec. 2001, 1, 467–479.

31 Andaraarachchi, H. P.; Thompson, M. J.; White, M. A.; Fan, H.-J.; Vela, J. Phase- Programmed Nanofabrication: Effect of Organophosphite Precursors on Evolution of Nickel Phosphide and Nickel Nanocrystals. Chem. Mater. 2015, 27, 8021-8031.

32 Fan, H. J.; Gösele, U.; Zacharias, M. Formation of Nanotubes and Hollow Nanoparticles Based on Kirkendall and Diffusion Processes: A Review. Small 2007, 3, 1660–1671.

33 Zhang, Q.; Wang, W.; Goebl, J.; Yin, Y. Self-Templated Synthesis of Hollow Nanostructures. Nano Today 2009, 4, 494–507.

34 Anderson, D. A.; Tracy, J. B. Nanoparticle Conversion Chemistry: Kirkendall Effect, Galvanic Exchange, and Anion Exchange. Nanoscale 2014, 6, 12195–12216.

35 Wang, W.; Dahl, M.; Yin, Y. Hollow Nanocrystals through the Nanoscale Kirkendall Effect. Chem. mater. 2012, 25, 1179−1189. 73

36 Heng, X.; Yuan, S.; Tian, Z.; Yin, S.; He, J.; Liu, K.; Liu, L. One- Pot Synthesis of Hollow Nickel Phosphide Nanoparticles with Tunable Void Sizes Using Triphenylphosphine. Mater. Lett. 2009, 63, 2283–2285.

37 Yin, Y.; Rioux, R. M.; Erdonmez, C. K.; Hughes, S.; Somorjai, G. A.; Alivisatos, A. P. Formation of Hollow Nanocrystals Through the Nanoscale Kirkendall Effect. Science 2004, 304, 711–714.

38 Wang, W.; Dahl, M.; Yin, Y. Hollow Nanocrystals through the Nanoscale Kirkendall Effect Chem. Mater. 2013, 25, 1179–1189.

39 Blanchard, P. E. R.; Grosvenor, A. P.; Cavell, R.G.; Mar, A. X-ray Photoelectron and Absorption Spectroscopy of Metal-rich Phosphides M2P and M3P (M = Cr-Ni). Chem. Mater. 2008, 20, 7081-7088.

40 Kanama, D.; Oyama, S. T.; Otani, S.; Cox, D. F. Ni2P (0001) by XPS. Surf. Sci. Spectra 2001, 8, 220–224.

41 Kanama, D.; Oyama, S. T.; Otani, S.; Cox, D. F. Photoemission and LEED characterization of Ni2P (0001). Surf. Sci. 2004, 552, 8–16. 42 Nesbitt, H. W.; Legrand, D.; Bancroft, G. M.; Interpretation of Ni2p XPS Spectra of Ni Conductors and Ni Insulators. Phys Chem Min 2000, 27, 357–366.

43 Grosvenor, A. P.; Biesinger, M. C.; C. Smart, R., St.; McIntyre, N. S. New Interpretations of XPS Spectra of Nickel metal and Oxides. Surf. Sci. 2006, 600, 1771–1779.

44 Matienzo, L. J.; Yin, L. I.; Grim, S. O.; Swartz, W. E., Jr. X-ray Photoelectron Spectroscopy of Nickel Compounds. Inorg. Chem. 1973, 12, 2762–2769.

45 Moffat, T. P.; Latanision, R. M.; Ruf, R. R. An X-Ray Photoelectron Spectroscopy Study of Chromium-Mettaloid Alloys–III. Electrochim. Acta 1995, 40, 1723–1734.

46 Pimerzin, A. A.; Nikulshin, P. A.; Mozhaev, A. V.; Pimerzin, A. A.; Lyashenko, A. I. Investigation of Spillover Effect in Hydrotreating Catalysts based on Co2Mo10−Heteropolyanion and Cobalt Sulphide Species. Applied Catalysis B: Environmental 2015, 168, 396–407.

47 Jiménez-Gómez, C. P.; Cecilia, J. A.; Ramón Moreno-Tost, R.; Maireles-Torres, P. Nickel Phosphide/Silica Catalysts for the Gas-Phase Hydrogenation of Furfural to High–Added– Value Chemicals. ChemCatChem 2017, 9, 2881–2889.

Appendix of Supporting Information

a Table S1. Selective phenylacetylene hydrogenation with Ni2P nanocatalysts.

Run Catalyst Cycle Conversion Total Yield % Selectivity (C6H5CHCH2 to C6H5CH2CH3)

(%) (C6H5CHCH2 +

C6H5CH2CH3)

1 Bimodal Ni2P (solid & 1 99 38 8.5:1

hollow, from NiCl2) 2 99 71 1.1:1

3 99 72 0:1

2 Ni2P (hollow only, 1 99 58 3.8:1

from Ni(OAc)2) 2 99 96 3.8:1

3 99 77 0:1

4 99 68 0:1

5 99 68 0:1

6 99 63 0:1

7 99 55 0:1

8 99 59 0:1

a 30 mol% of catalyst at 40 bar of H2 and 100 °C for 14h in 1-propanol.

Figure S1. Powder XRD patterns and TEM images of (A) Ni2P from NiCl2 (B) Ni2P after cycle 3 in dioxane (C) Ni2P after cycle 3 in 1-propanol.

Figure S2. XPS spectra of (A) Ni2P from NiCl2 (B) Ni2P after cycle 3 in dioxane (C) Ni2P after cycle 3 in 1-propanol.

Figure S3. Powder XRD patterns and TEM images of (A) Ni2P from Ni(OAc)2 (B) Ni2P after cycle 1 (C) Ni2P after cycle 2 (D) Ni2P after cycle 3 (E) Ni2P after cycle 4 in 1- propanol.

Figure S4. XPS spectra of (A) Ni2P from Ni(OAc)2 (B) Ni2P after cycle 1 (C) Ni2P after cycle 2 (D) Ni2P after cycle 3 (E) Ni2P after cycle 4 (F) Ni2P after cycle 8 in 1-propanol. 79

CHAPTER 4.

GE-SN HETEROSTRUCTURES: FROM THEORY TO EXPERIMENT

Himashi P. Andaraarachchi, Arthur White, Bryan A. Rosales, Long Men, Javier Vela

Manuscript in preparation

Abstract

Synthesis of complex nanostructures with precisely controlled properties is an important aspect for their many applications. However, the main challenge of the synthesis of

Ge-Sn heterostructures stems from the disparity in atomic radii of two elements and their existence in different thermodynamically stable crystal structures. To avert synthetic challenges, we explored Ge-Sn single source precursors with tunable substituents (R3GeSnR’3) that already contain a preformed Ge-Sn bond as potential precursors for heterostructure synthesis. We chose three different molecular precursors with varying Ge-Sn bond strengths to explore how they would affect the properties of Ge-Sn heterostructures. Solution phase thermal decomposition of these precursors yield b-Sn/b-Ge core/shell like structures with varying Ge content % and shell thicknesses. We observed the b-Ge phase at ambient pressure conditions that otherwise typically exists at high pressure conditions (above 10 GPa). We hypothesize that tetragonal b-Sn nuclei might act as seeds for the tetragonal b-Ge phase.

Metallic b-Sn/b-Ge heterostructures with controlled properties could be potentially harnessed for Li ion batteries and other electronic applications. Overall, the Ge-Sn single source precursor family will provide a facile approach to synthesize Ge-Sn heterostructures with varying morphologies and compositions through fine-tuning the Ge-Sn bond strength.

Introduction

Group IV elements (Si, Ge, and Sn) are considered as relatively earth abundant and low toxic materials with their wide range of applications in photovoltaic devices, optoelectronic devices, and bio-imaging.1-3 Compared with their single element analogues, Group IV heterostructures (SiGe, GeSn, and SiGeSn) exhibit improved electron hole mobility and enhanced light emission and serve as a promising materials for field effect transistors,4 photodiodes,5,6 lasers,7-9 and light emitting diodes.10-13

Elemental germanium (Ge) is a group IV semiconductor that has a small indirect band gap (0.661 eV) and a large Bohr radius (24 nm).2 These attributes provide for a wide range of tunable emission energies through size tunable quantum confinement. However, the indirect energy band gap makes it difficult to achieve efficient light harvesting and emission. Several theoretical studies have concluded that a transition to a direct band gap material is possible through either lattice strain, quantum confinement, or doping with tin in values of ~ 6 to > 20 atom % depending on the study.14-16 While all of these options are potentially feasible, pursuing the avenue of tin incorporation is beneficial for two primary reasons. The first is to gain additional information about the impact of precursor bond dissociation energies on nanocrystal size and morphology. Secondly, tin incorporation is expected to greatly enhance the use of these nanocrystals for all suggested applications.

However, synthesis of GeSn heterostructures/alloys with controllable properties seems specifically challenging due to the large disparity in atomic radii of two elements and the existence of thermodynamically stable different crystal structures. Germanium and tin have atomic radii of 122 pm and 140 pm, respectively. Thus, the difference in atomic radii between the solute and solvent atoms is 14.8 %. Based on Hume-Rothery rules, a difference in radii of 81

≤ 15 % is expected to form solid solutions.17 Furthermore, the thermodynamically stable crystal structure for germanium and tin is also different where germanium prefers the cubic alpha structure while tin prefers the tetragonal beta structure. These obstacles can be seen experimentally by consulting the Ge-Sn phase diagram which shows a solid solution only up to < 2 % Sn.18

Several syntheses of nanocrystalline GeSn alloys and heterostructures have been reported through different methods including colloidal synthesis,19-22 microwave assisted synthesis,23-25 chemical vapor deposition,26-28 and gas phase laser plasma methods.29

Nanoscale materials have been synthesized up to as high as 42% of tin incorporation.19 Despite these advances, a facile method of synthesizing GeSn alloys and heterostructures with controlled properties has yet to be realized. The development of efficient single source precursors with a pre-formed Ge-Sn bond with tunable substituent groups have the potential to yield larger control of the synthesis with new morphologies while controlling the size and the composition.

Typically many reaction conditions have to be optimized to synthesize nanocrystals with predicted outcomes, which can be very time consuming. Recently, we and a few other groups introduced a simple but efficient approach to control the morphology, size, and phase of nanocrystals through alteration of precursor binding energies while keeping all other reaction conditions constant.30-33 In these studies, nanocrystals were synthesized by the reaction of a metal salt with either pnictide or chalcogenide precursors of differing structure.

The resulting size, morphology, and phase of the nanocrystals were observed to be strongly impacted by the choice of the main group molecular precursor. 82

While the impact of precursor reactivity on nanocrystal formation has been widely studied in the past, a comprehensive study of the effects of single-source precursors has not been completed. However, these type of single-source precursors has been employed to synthesize nanocrystalline materials with complex morphologies.34-37 Additionally, group IV chalcogenides of GeS and SnS have been synthesized in a host of different morphologies depending on the single-source precursor employed.38-42 Furthermore, nanorods, nanotubes, nanocubes, nanostars, etc., have been synthesized through the thermal decomposition of various single-source precursors in the presence of a capping ligand.

In this work, we demonstrate how the structure and reactivity of Ge-Sn single source precursors with tunable substituents controllably affects the morphology and composition of

GeSn heterostructures.

Experimental Section

Materials and Precursors. Chlorotrimethylsilane (TMSCl, 99%), lithium (Li, granular, 99%), potassium tert-butoxide (Me3COK, 99%), n-butyllithium (n-BuLi, 1.6 M in hexanes), triphenyl(tripehnylstannyl)germane (Ph3GeSnPh3) and oleylamine (70%) were purchased from Aldrich; triphenylgermanium chloride (Ph3GeCl, 99%), trimethyltin chloride

(Me3SnCl, 95%), germanium(IV) chloride (GeCl4, 99.99%), 18-crown-6 (99.9%), and triphenyltinchloride (Ph3SnCl, 95%) from Strem; and chloroform-d (CDCl3), and benzene- d6(C6) from Cambridge Isotopes Laboratories. All compounds were used as received and handled under an inert (dry-N2 or Ar) atmosphere inside a glovebox or with a Schlenk line.

Water and oxygen were removed from toluene and tetrahydrofuran (THF) solvents using an

43 IT PureSolv system. Ph3GeSnMe3 was synthesized using a modified literature procedure.

Briefly, Ph3GeCl (1.76 g, 5.08 mmol) and Li (0.14g) were mixed in dry THF (40 mL) and 83 stirred for 12 h at room temperature (RT). The resulting solution was added dropwise to a solution of Me3SnCl (1.00 g, 5.08 mmol) in THF (15 mL) pre-equilibrated at –25 °C. After 0.5 h stirring at this temperature, the mixture was slowly warmed to RT and stirred overnight.

Solvent was removed on a rotary evaporator. The crude product was dissolved in hexane and

LiCl removed by gravity filtration, followed by the removal of hexane on a rotary evaporator:

1 1.12 g (52%). H NMR (600 MHz, CDCl3): 0.32 (s, 9H, SnMe3); 7.39, 7.47 (15H, m, GePh3).

13 1 C{ H} NMR (150 MHz, CDCl3): -9.57 (SnMe3); 128.42, 128.60, 135.31, 139.07 (GePh3).

119 Sn NMR (224 MHz, CDCl3): -89.74. Absorption (hexane) lmax, = 238 nm. PLmax (hexane)

1 = 317 (lex = 270 nm). Eox (vs. Ag/AgCl) = 1906 mV. Ph3GeSnPh3. H NMR (600 MHz,

13 1 CDCl3): 7.28–7.53 (30H, m, SnPh3, GePh3). C{ H} NMR (150 MHz, CDCl3): 135.51,

119 137.61, 137.88, 139.20, 128.40, 128.62, 128.83, 129.03 (SnPh3, GePh3). Sn NMR (224

MHz, CDCl3): -162.58. Absorption (hexane) lmax = 244 nm. PLmax (hexane) = 312 (lex = 270

44 nm). Eox (vs. Ag/AgCl) = 1959 mV. TMS3GeSnPh3. TMS3GeSnPh3 (TMS = SiMe3) was

45 synthesized from TMS4Ge according to literature procedures.

Precursor Characterization. Solution Phase Thermolysis. Inside a three-neck flask, the heterobimetallic precursor (0.10 mmol; 31 mg of Ph3GeSnMe3, or 64 mg of TMS3GeSnPh, or 65 mg of Ph3GeSnPh3) and oleylamine (2.5 mL, 7.6 mmol) were degassed under dynamic vacuum at 80 °C for 1 h and refilled with Ar. 1.6 M n-BuLi (0.13 mL, 0.2 mmol) was quickly added via a Luer lock syringe. The temperature was increased to 225 °C, and the mixture stirred for 15 min. After cooling to RT, solids were isolated by twice dissolving in toluene (5 mL) and centrifugation at 4900 rpm for 5 min. NMR Spectroscopy. 1H, 13C, and 119Sn and NMR spectra were collected on a Bruker Avance III-600 spectrometer. 1H and 13C NMR chemical shifts (δ)

119 are reported in ppm relative to residual protiated solvent in CDCl3 (7.26 ppm). Sn NMR 84

chemical shifts (δ) are reported in ppm relative to SnCl2 (-352 ppm). Optical Spectroscopy.

Absorption spectra were measured with a photodiode array Agilent 8453 UV–Vis spectrophotometer. Steady-state photoluminescence (PL) spectra were recorded using a

Horiba-Jobin Yvon Nanolog scanning spectrofluorometer equipped with a liquid nitrogen- cooled InGaAs photodiode array. Cyclic voltammetry (CV). CV measurements were performed using a Pine WaveNow potentiostat, a 2-mm diameter platinum working electrode, a platinum counter electrode, an Ag/AgCl/KCl aqueous reference electrode, and a sweep rate of 100 mV/s. All measurements were performed in a glovebox under nitrogen atmosphere.

Samples solutions contained 2 mM heterobimetallic complex, 0.1 M (n-Bu)4NBF4 as the supporting electrolyte, and dry methylene chloride as the solvent. The potential of the ferrocene couple (Fc/Fc+) vs. our reference electrode under these conditions is 0.59 V. Our reference electrode was calibrated against the ferrocene couple (Fc/Fc+, + 0.59 V).

Material Characterization. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). TGA/DSC was performed on a Netzsch STA449 F1 Jupiter® instrument coupled to a Quadrupole Mass Spectrometer (QMS) and Bruker Tensor 27 FT-IR spectrometer. Samples (~5-10 mg) in alumina crucibles were heated from 40 °C to 600 °C using Ar as the carrier gas and 10 °C /min as the ramp rate. Powder X-ray Diffraction (XRD).

XRD was measured using Cu Kα radiation on a Rigaku Ultima diffractometer. Transmission

Electron Microscopy (TEM). TEM was conducted on carbon-coated copper grids using a FEI

Titan Themis 300 Cubed probe aberration scanning transmission electron microscope (STEM) at 200 kV. Particle composition was determined by energy dispersive spectrometry (EDX).

More than 300 particle dimensions were measured manually using ImageJ. Averages are reported ± one standard deviation. 85

Single-Source Precursor Molecular Calculations: To investigate the bond strengths of either the Ge-Sn or ligand bond, GAMESS software was utilized.46,47 First, a geometry optimization of the parent precursor was performed and a Hessian analysis of the results was done to ensure a minimum in the potential energy surface. Next, radical fragments were generated by breaking the bond of interest. These radical fragments were allowed to optimize geometry and an energy was obtained. The difference in energy between the sum of the radical fragments and the energy of the parent precursor was attributed to the bond dissociation energy

(BDE). For all BDE calculations, density functional theory (DFT) with the B3LYP function and a 6-311G(d,p) basis set was used (larger Sn and Ge used 3-21G(d) basis sets). A scheme of this method is given in Figure 1 for the example compound Me3GeSnPh3. Arthur White has performed the molecular calculations.

Results and Discussion

Exploration of Ge-Sn Single Source Precursors. Single source precursors offer the advantage of assembling the required components into a single molecule enabling desired stoichiometry with an already preformed bond. Probing the availability of heterobimetallic single source precursors, we employed a series of compounds with a single Ge-Sn bond with

43,44,48-52 tunable substituents (R3GeSnR’3) reported in the prior literature. These types of precursors are particularly appealing because the strength of Ge-Sn bond can be simply manipulated by the nature of the substituents (electron donating and/or electron withdrawing), thus affecting the final product morphology and composition. To develop a better understanding of bonding properties of different precursors, we computationally studied seven precursors that were reported in the prior literature using GAMESS software.46,47 Figure 1 shows a schematic of computational method employed using Me3GeSnPh3 as a representative 86 example. The bond of interest is measured by generating fragments from the homolysis of that interaction. For example, to measure the bond dissociation energy of Sn-C in this compound: a Me3GeSnPh2 and Ph radical energies are summed and subtracted from the parent compound’s energy.

Figure 1. Schematic of computational method employed using Me3GeSnPh3 as a representative example.

Figure 2 summarizes the calculated bond dissociation energies (BDE) of Ge-Sn, Ge-R, and Sn-R bonds. In general, the BDE of Ge-R and Sn-R is considerably higher than Ge-Sn.

The nature of the Ge-Sn bond strength appears to be related to the electron donating and withdrawing character of the substituent ligands.

Figure 2. Calculated bond dissociation energies for Ge-Sn, Ge-R, and Sn-R for all screened single-source precursors.

Using the progression from R3GeSnPh3 (R = H, Me, Ph, TMS), it is apparent that the bond strength of Ge-Sn increases as the R group becomes more electron donating. As such,

TMS3GeSnPh3 is the precursor with the highest calulated BDE. Furthermore, a more electron donating group (such as Me) bonded to Sn results in a substantially weaker Ge-Sn bond. With these factors in mind, we selected the two bond strength extremes, Ph3GeSnMe3 and

TMS3GeSnPh3, for Ge-Sn to examine their effect on the synthesis of GeSn alloy or heterostructure nanocrystals as well as using the commercially available Ph3GeSnPh3.

Furthermore, an investigation of the frontier orbitals of the selected precursors provides a basis for how to attempt these reactions (Figure 3).

The HOMO orbitals for both single-source precursors display a considerable amount of Ge-Sn σ-bonding. Therefore, a reaction scheme utilizing oxidation should likely be avoided because oxidation will directly weaken the Ge-Sn bond that needs to be retained for formation 88

of Ge-Sn nanocrystals. Contrarily, decomposing Ph3GeSnMe3, TMS3GeSnPh3, and

Ph3GeSnPh3 in the presence of a reducing agent will populate the Ge-R and Sn-R antibonding

states.

Figure 3. Frontier orbitals for Ph3GeSnMe3, TMS3GeSnPh3, and Ph3GeSnPh3 displaying the Ge-Sn σ-bonding in the HOMO and presence of antibonding Ge-R and Sn-R states in the LUMO.

43 Synthesis and Characterization of Single Source Precursors. Ph3GeSnMe3 and

45 (SiMe3)3GeSnPh3 were prepared by the reaction of a lithium or potassium germyl salt with alkyl tin halide in dry THF or toluene and isolated as white solids. Ph3GeSnPh3 was purchased.

All precursors were structurally characterized using 1H, 13C, and 119Sn NMR spectroscopy, optical properties were investigated using UV/Visible absorption and fluorescence emission, and reduction potentials were determined using cyclic voltammetry measurements (See

Supporting Information) and all data are in accord with those described earlier. Furthermore, new characterization data that reported in this work are summarized in Table 1. The 119Sn NMR shift of TMS3GeSnPh3 has been previously determined and matches with our observation.

Table 1. 119Sn NMR, electrochemical, and optical data for single source precursors

119 a b c c d precursor Sn NMR Eox (V) absorption max PL max (nm) (ppm) (nm)

Ph3GeSnMe3 -89.74 1.906 237 316

Ph3GeSnPh3 -162.58 1.959 243 311

TMS3GeSnPh3 -111.45 1.883 241 307

a b c Referenced against SnCl2 (d-352 ppm). Ag/AgCl/KCl as reference electrode. Hexane. d Excitation wavelength, 270 nm.

Decomposition of Precursors. Bulk solid decompositions of the precursors were followed by thermogravimetric analysis (TGA). Precursors were heated to 600 °C in the presence of a reductant, 1 molar equivalent of Li, under Ar atmosphere (Figure 4). Ph3GeSnPh3 undergoes single step decomposition to yield crystalline b-Sn particles while Ph3GeSnPh3 and

(SiMe3)3GeSnPh3 undergo multistep decomposition to yield b-Sn particles as per shown by powder XRD patterns. 32.44%, 46.58%, and 44.55% of masses were left after the thermal analysis of Ph3GeSnMe3, Ph3GeSnPh3, and (SiMe3)3GeSnPh3 samples respectively and these data agree with the percentage of the sum of Ge, Sn, and Li molar masses in each precursor.

All the DSC curves contain peaks at 180 °C that indicate an endothermic process of metallic

Li melting. The peaks around 250 – 300 °C indicate exothermic processes which might be probable decomposition of precursors. Ph3GeSnMe3 melts at 83-86 °C exhibiting an endothermic peak in DSC curve. Ph3GeSnPh3 melts around 284-286 °C, which might be buried in exothermic decomposition process. However, melting of (SiMe3)3GeSnPh3 (158-160 °C) 90 isn’t visible in the DSC curve. We have also studied the thermal decomposition of precursors without any reductant that exhibit similar results (Figure S1).

Figure 4. (a) TGA and DSC traces of precursors with Li and (b) PXRD patterns after TGA experiments with Li. Si powder was used as an internal standard to detect any peak shift which is denoted by asteriks.

Solution Phase Thermolysis. Solution phase decomposition reactions were carried out in a high boiling point solvent, oleylamine. Briefly, neat n-BuLi was injected at 80 °C to a degassed mixture of precursor and oleylamine, heated up to 225 °C and kept at 15 min (see

Scheme 1). Powder XRD patterns exhibit that all precursors decomposed to produce crystalline b-Sn particles (Figure 5a). Si was used as an internal standard. Morphology of b-Sn particles has been studied using scanning transmission electron microscopy (STEM). STEM images reveal that Ph3GeSnMe3, Ph3GeSnMe3, and (SiMe3)3GeSnPh3 decomposes to spherical b-Sn particles with different sizes (Figure 5b, 5c, and 5d). However, absence of any crystalline elemental Ge or Ge oxide species observed through powder XRD prompt us to conduct in- depth elemental composition and structural analysis through STEM-EDX mapping.

Scheme 1. General synthetic procedure for solution phase decomposition

225 °C, 15 min R GeSnR’ + n-BuLi β-Sn 3 3 Oleylamine

R = Ph,TMS R’ = Ph,Me

Figure 5. (a) Powder XRD patterns and representative STEM images and size distribution histograms of solution phase thermolysis products of (b) Ph3GeSnMe3 (c) Ph3GeSnPh3 (d) TMS3GeSnPh3.

Composition Analysis. Sn/Ge core/shell structures and isolated Sn particles were observed through HAADF-STEM for Ph3GeSnMe3, Ph3GeSnPh3, and TMS3GeSnPh3 respectively, with the elemental composition revealed by STEM-EDX mapping, which is summarized in Table 2. 93

These results revealed that two precursors, Ph3GeSnMe3 and Ph3GeSnPh3, decompose to produce heterostructures or Sn/Ge core/shell particles with the elemental composition of 92

± 1% Sn and 8 ± 1 % Ge and 97 ± 1% Sn and 3 ± 1 % Ge respectively, in contrast to only b-

Sn, which is depicted in powder XRD patterns. Ph3GeSnMe3 forms a relatively thicker Ge shell compared to the thin Ge shell from Ph3GeSnPh3 (Figure 6a and S4).

Table 2. Size, composition, and morphology of solution phase thermolysis products precursor TEM size (nm) elemental composition (%) structure

Ph3GeSnMe3 43 ± 8 92 ± 1 Sn, 8 ± 1 Ge Sn core with Ge shell

Ph3GeSnPh3 31 ± 5 97 ± 1 Sn, 3 ± 1 Ge Sn core with Ge shell

TMS3GeSnPh3 59 ± 9 99.7 Sn Sn particles

While fast Fourier transform (FFT) of the core image confirms that d spacing of core,

2.958 Å, matches with the (020) plane of b-Sn, d spacing of shell, 2.6 Å best matches with

2.56 Å (020) plane b-Sn Ge phase (Figure 6b, 6c). Semiconductor cubic diamond Ge can be converted to metallic b-Sn Ge upon compression at 10-12 GPa and this Ge allotrope has been so far observed in the range of 10 GPa- 66 GPa.53-57 We were unable to find synthesis of b- Ge at ambient pressure conditions in the literature and these heterostructures might be the first instances of b-Sn Ge at ambient pressure. Our hypothesis was that tetragonal b-Sn nuclei might act as a catalyst for the synthesis of a tetragonal b-Ge phase excluding the synthesis of the thermodynamically stable diamond cubic Ge phase. On the contrary, TMS3GeSnPh3 fails to 94

synthesize a core/shell structure and decomposes to produce b-Sn particles with a size distribution of 59 ± 9 nm as shown in Figure S5.

Figure 6. (a) EDX elemental mapping of Sn/Ge core/shell nanocrystals from solution phase thermolysis of Ph3GeSnMe3. HAADF-STEM images of (b) Sn core with a fast Fourier transform onset, (c) Ge shell with a fast Fourier transform onset.

Conclusions

Synthesis of complex nanostructures with precisely controlled properties is an important aspect for their many applications. To obviate the challenges involved with the synthesis of Ge-Sn heterostructures, we chose single source precursors with a preformed Ge- 95

Sn bond with tunable substituents (R3GeSnR’3). Single source precursors already contain Ge-

Sn molecules in close proximity, which is otherwise challenging due to their large atomic radii difference and their different thermodynamically stable crystal structures. Moreover, depending on the nature of the substituents it is able to fine tune the Ge-Sn, Ge-R, and Sn-R bond strengths and consequently control the size, composition, and morphology of the Ge-Sn heterostructures. Our observations show that solution phase thermolysis of these precursors under reduced reaction conditions yields core/shell structure of b-Sn/b-Ge nano heterostructures with varying shell thicknesses. We observed the b-Ge allotrope in ambient pressure conditions, which typically exists at high pressure conditions (10 Gpa – 66 Gpa) and we postulated that tetragonal b-Sn nuclei might serve as a catalyst for the synthesis of tetragonal b-Ge in ambient pressure conditions. These metallic core/shell heterostructures with unique morphologies will offer novel properties and have potential applications in electronics.

Acknowledgements

J. Vela thanks the US National Science Foundation for a CAREER grant from the

Division of Chemistry, Macromolecular, Supramolecular and Nanochemistry program

(1253058).

References

1 Cloutier, S. G.; Kossyrev, P. A.; Xu, J. Optical Gain and Stimulated Emission in Periodic Nanopatterned Crystalline Silicon. Nat. Mater. 2005, 4, 887.

2 Vaughn, D. D., II; Schaak, R. E. Synthesis, Properties and Applications of Colloidal Germanium and Germanium-Based Nanomaterials. Chem. Soc. Rev. 2013, 42, 2861–2879.

3 Prabakar, S.; Shiohara, A.; Hanada, S.; Fujioka, K.; Yamamoto, K.; Tilley, R. D. Size Controlled Synthesis of Germanium Nanocrystals by Hydride Reducing Agents and Their Biological Applications Chem. Mater. 2010, 22, 482–486.

4 Wirths, S.; Stange, D.; Pampillón, M.-A.; Tiedemann, A. T.; Mussler, G.; Fox, A.; Breuer, U.; Baert, B.; San Andrés, E.; Nguyen, N.D.; Hartmann, J.-M.; Ikonic, Z.; Mantl, S.; 96

Buca, D. High- K Gate Stacks on Low Bandgap Tensile Strained Ge and GeSn Alloys for Field-Effect Transistors. ACS Appl. Mater. Interfaces 2015, 7, 62–67.

5 Xu, C.; Beeler, R. T.; Grzybowski, G.; Chizmeshya, A. V. G.; Menendez, J.; Kouvetakis, J. Molecular Synthesis of High-Performance Near-IR Photodetectors with Independently Tunable Structural and Optical Properties Based on Si–Ge–Sn. J. Am. Chem. Soc. 2012, 134, 20756–20767.

6 Senaratne, C.; Wallace, P.; Gallagher, J.; Sims, J.; Kouvetakis, J.; Menendez, J. Direct Gap Ge1−y Sny Alloys: Fabrication and Design of Mid–IR Photodiodes. J. Appl. Phys. 2016, 120, 025701.

7 Stange, D.; Wirths, S.; Geiger, R.; Schulte-Braucks, C.; Marzban, B.; von den Driesch, N.; Mussler, G.; Zabel, T.; Stoica, T.; Hartmann, J.-M.; Mantl, S.; Ikonic, Z.; Grützmacher, D.; Sigg, H.; Witzens, J.; Buca, D. Optically Pumped GeSn Microdisk Lasers on Si. ACS Photonics 2016, 3, 1279–1285.

8 Wirths, S.; Geiger, R.; von den Driesch, N.; Mussler, G.; Stoica, T.; Mantl, S.; Ikonic, Z.; Luysberg, M.; Chiussi, S.; Hartmann, J. M.; Sigg, H.; Faist, J.; Buca, D.; Grützmacher, D. Lasing in Direct-Bandgap GeSn Alloy Grown on Si. Nat. Photonics 2015, 9, 88–92.

9 Homewood, K. P.; Lourenço, M. A. Optoelectronics: The Rise of the GeSn Laser. Nat. Photonics 2015, 9, 78–79.

10 von den Driesch, N.; Stange, D.; Wirths, S.; Rainko, D.; Povstugar, I.; Savenko, A.; Breuer, U.; Geiger, R.; Sigg, H.; Ikonic, Z.; Hartmann, J. M.; Grützmacher, D.; Mantl, S.; Buca, D. SiGeSn Ternaries for Efficient Group IV Heterostructure Light Emitters. Small 2017, 13, 1603321.

11 Beeler, R. T.; Gallagher, J. D.; Xu, C.; Jiang, L.; Senaratne, C.; Smith, D. J.; Menéndez, J.; Chizmeshya, A. V. G.; Kouvetakis, J. Band Gap-Engineered Group-IV Optoelectronic Semiconductors, Photodiodes and Prototype Photovoltaic Devices. ECS J. Solid State Sci. Technol. 2013, 2, Q172–Q177.

12 Soref, R. Group IV Photonics: Enabling 2 µm Communications. Nat. Photonics 2015, 9, 358–359.

13 Jiang, L.; Xu, C.; Gallagher, J. D.; Favaro, R.; Aoki, T.; Menéndez, J.; Kouvetakis, J. Development of Light Emitting Group IV Ternary Alloys on Si Platforms for Long Wavelength Optoelectronic Applications. Chem. Mater. 2014, 26, 2522–2531.

14 Gupta, S.; Magyari-Köpe, B.; Nishi, Y.; Saraswat, K. C. Achieving Direct Band Gap in Germanium through Integration of Sn Alloying and External Strain. J. Appl. Phys. 2013, 113, 073707– 1–073707–7. 97

15 Gallagher, J.; Senaratne, C.; Kouvetakis, J.; Menendez, J. Compositional Dependence of the Bowing Parameter for the Direct and Indirect Band Gaps in Ge1–ySny Alloys. Appl. Phys. Lett. 2014, 105, 142102–1–142102–5.

16 Senaratne, C. L.; Gallagher, J. D.; Xu, C.; Sims, P.; Menendez, J.; Kouvetakis, J. Doping of Direct Gap Ge1-ySny Alloys to Attain Electroluminescence and Enhanced Photoluminescence. ECS Trans. 2015, 69, 157–164.

17 Hume-Rothery, W. J. Inst. Met. 1926, 295–361.

18 ASM Alloy Phase Diagram Database, ASM International: Materials Park, OH, 2016.

19 Esteves, R. J. A.; Ho, M. Q.; Arachchige, I. U. Nanocrystalline Group IV Alloy Semiconductors: Synthesis and Characterization of Ge1–xSnx Quantum Dots for Tunable Bandgaps. Chem. Mater. 2015, 27, 1559–1568.

20 Ramasamy, K.; Kotula, P. G.; Fidler, A. F.; Brumbach, M. T.; Pietryga, J. M.; Ivanov, S. A. SnxGe1–x Alloy Nanocrystals: A First Step toward Solution-Processed Group IV Photovoltaics. Chem. Mater. 2015, 27, 4640–4649.

21 Esteves, R. J. A.; Hafiz, S.; Demchenko, D. O.; Özgür, Ü.; Arachchige, I. U. Ultra-Small Ge1-xSnx Quantum Dots with Visible Photoluminescence. Chem. Commun. 2016, 52, 11665– 11668.

22 Boote, B. W.; Men, L.; Andaraarachchi, H. P.; Bhattacharjee, U.; Petrich, J. W.; Vela, J.; Smith, E. A. Germanium–Tin/Cadmium Sulfide Core/Shell Nanocrystals with Enhanced Near- Infrared Photoluminescence. Chem. Mater. 2017, 29, 6012–6021.

23 Barth, S.; Seifner, M. S.; Bernardi, J. Microwave-Assisted Solution–Liquid–Solid Growth of Ge1-xSnx Nanowires with High Tin Content. Chem. Commun. 2015, 51, 12282–12285.

24 Seifner, M. S.; Biegger, F.; Lugstein, A.; Bernardi, J.; Barth, S. Microwave-Assisted Ge1- xSnx Nanowire Synthesis: Precursor Species and Growth Regimes. Chem. Mater. 2015, 27, 6125–6130.

25 Seifner, M. S.; Hernandez, S.; Bernardi, J.; Romano-Rodriguez, A.; Barth, S. Pushing the Composition Limit of Anisotropic Ge1–xSnx Nanostructures and Determination of Their Thermal Stability. Chem. Mater. 2017, 29, 9802–9813.

26 Meng, A. C.; Fenrich, C. S.; Braun, M. R.; McVittie, J. P.; Marshall, A. F.; Harris, J. S.; McIntyre, P. C. Core/Shell Germanium/Germanium-Tin Nanowires Exhibiting Room Temperature Direct-and Indirect-Gap Photoluminescence. Nano Lett. 2016, 16, 7521–7529.

27 Křenek, T.; Bezdička, P.; Murafa, N.; Šubrt, J.; Pola, J. Laser CVD of Nanodisperse Ge–Sn Alloys Obtained by Dielectric Breakdown of SnH4/GeH4 Mixtures. Eur. J. Inorg. Chem. 2009, 48, 1464–1467. 98

28 Assali, S.; Dijkstra, A.; Li, A.; Koelling, S.; Verheijen, M. A.; Gagliano, L.; Von Den Driesch, N.; Buca, D.; Koenraad, P. M.; Haverkort, J. E. M.; Bakkers, E. P. A. M. Growth and Optical Properties of Direct Band Gap Ge/Ge0.87Sn0.13 Core/Shell Nanowire Arrays. Nano Lett. 2017, 17, 1538–1544.

29 Cho, Y. J.; Kim, C. H.; Im, H. S.; Myung, Y.; Kim, H. S.; Back, S. H.; Lim, Y. R.; Jung, C. S.; Jang, D. M.; Park, J.; Lim, S. H.; Cha, E. H.; Bae, K. Y.; Song, M. S.; Cho, W. I. Germanium–Tin Alloy Nanocrystals for High-Performance Lithium Ion Batteries. Phys. Chem. Chem. Phys. 2013, 15, 11691–11695.

30 Guo, Y.; Alvarado, S. R.; Barclay, J. D.; Vela, J. Shape-Programmed Nanofabrication: Understanding the Reactivity of Dichalcogenide Precursors. ACS Nano 2013, 7, 3616–3626.

31 Ruberu, T. P. A.; Albright, H. R.; Callis, B.; Ward, B.; Cisneros, J.; Fan, H.-J.; Vela, J. Molecular Control of the Nanoscale: Effect of Phosphine Chalcogenide Reactivity on CdS- CdSe Nanocrystal Composition and Morphology. ACS Nano 2012, 6, 5348–5359.

32 Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La Plante, I.; Owen, J. S. A Tunable Library of Substituted Thiourea Precursors to Metal Sulfide Nanocrystals. Science 2015, 348, 1226–1230.

33 Andaraarachchi, H. P.; Thompson, M. J.; White, M. A.; Fan, H.-J.; Vela, J. Phase- Programmed Nanofabrication: Effect of Organophosphite Precursors on Evolution of Nickel Phosphide and Nickel Nanocrystals. Chem. Mater. 2015, 27, 8021-8031.

34 Trentler, T. J.; Hickman, K. M.; Goel, S. C.; Viano, A. M.; Gibbons, P. C.; Buhro, W. E. Solution–Liquid–Solid Growth of Crystalline III–V Semiconductors—An Analogy to Vapor–Liquid–Solid Growth. Science 1995, 270, 1791–1794.

35 Wooten, A.; Werder, D.; Williams, D.; Casson, J.; Hollingsworth, J. A. Solution-Liquid- Solid Growth of Ternary Cu-In-Se Semiconductor Nanowires from Multiple- and Single- Source Precursors. J. Am. Chem. Soc. 2009, 131, 16177–16188.

36 McElwee-White, L.; Koller, J.; Kim, D.; Anderson, T.J. Mechanism-Based Design of Precursors for MOCVD. ECS Transactions 2009, 25, 161-171.

37 McElwee-White, L. Design of Precursors for the CVD of Inorganic Thin Films. Dalton Trans. 2006, 0, 5327–5333.

38 Knisley, T. J.; Kalutarage, L. C.; Winter, C. H. Precursors and Chemistry for the Atomic Layer Deposition of Metallic First Row Transition Metal Films. Coord. Chem. Rev. 2013, 257, 3222–3231.

39 Zhang, Y. J.; Lu, J.; Shen, S. L.; Xu, H. R.; Wang, Q. B. Ultra Large Single Crystal SnS Rectangular Nanosheets. Chem. Commun. 2011, 47, 5226–5228.

40 Sharma, R. K.; Kedarnath, G.; Wadawale, A.; Betty, C. A.; Vishwanadh, B.; Jain, V. K. Diorganotin(IV) 2-pyridyl Selenolates: Synthesis, Structures and Their Utility as Molecular 99

Precursors for the Preparation of Tin Selenide Nanocrystals and Thin Films. Dalton. Trans. 2012, 41, 12129–12138.

41 Tyagi, A.; Kedarnath, G.; Wadawale, A.; Jain, V. K.; Kumar, M.; Vishwanadh, B. Diorganotin(IV) 2-pyridyl and 2-pyrimidyl thiolates: Synthesis, Structures and Their Utility as Molecular Precursors for the Preparation of Tin Sulfide Nanosheets. RSC Adv. 2015, 5, 62882– 62890.

42 Shah, A. Y.; Kedarnath, G.; Tyagi, A.; Betty, C. A.; Jain, V. K.; Vishwanadh, B. Germanium Xanthates: Versatile Precursors for Photo Responsive Germanium Sulfide Nanostructures. ChemistrySelect 2017, 2, 4598–4604.

43 Pannell, K. H.; Parkanyi, L.; Sharma, H.; Cervantes-Lee, F. The Germanium-Tin Bond. Structures of the Isomeric Complexes Me3E-E'Ph3 (E = Ge, E' = Sn; E = Sn, E' = Ge). Inorg. Chem. 1992, 31, 522–524.

44 Zaitsev, K. V.; Lermontova, E. K.; Churakov, A. V.; Tafeenko, V. A.; Tarasevich, B. N.; Poleshchuk, O. K.; Kharcheva, A. V.; Magdesieva, T. V.; Nikitin, O. M.; Zaitseva, G. S.; Karlov, S. S. Compounds of Group 14 Elements with an Element–Element (E= Si, Ge, Sn) Bond: Effect of the Nature of the Element Atom. Organometallics 2015, 34, 2765–2774.

45 Brooke, A. G.; Abdesaken, F.; Söllradl, H. J. Synthesis of Some Tris(trimethylsilyl)germyl Compounds. J. Org. Chem. 1986, 299, 9–13.

46 Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. General Atomic and Molecular Electronic Structure System. J. Comput. Chem. 1993, 14, 1347–1363.

47 Gordon, M. S.; Schmidt, M. W. Advances in Electronic Structure Theory: GAMESS A Decade Later. In Theory and Applications of Computational Chemistry: The First Forty Years; Dykstra, C. E., Frenking, G., Kim, K. S., Scuseria, G. E., Eds.; Elsevier: Amsterdam, 2005; pp 1167–1189.

48 Parkanyi, L.; Kalman, A.; Pannell, K. H.; Sharma, H. K. New Refinement of the Me,E-E’Ph, (E = Ge, E’ = Sn; E = Sn, E’ = Ge) Isomeric Compounds and the Crystal Structure of their Solid Solution. J. Organomet. Chem. 1994, 484, 153−159.

49 Tice, J. B.; Chizmeshya, A. V. G.; Groy, T. L.; Kouvetakis, J. Synthesis and Fundamental Properties of Stable Ph3SnSiH3 and Ph3SnGeH3 Hydrides: Model Compounds for the Design of Si−Ge−Sn Photonic Alloys. Inorg. Chem. 2009, 48, 6314–6320.

50 Baumgartner, J.; Fischer, R.; Fischer, J.; Wallner, A.; Marschner, C.; Flörke, U. Structural Aspects of Trimethylsilylated Branched Group 14 Compounds. Organometallics 2005, 24, 6450–6457.

51 Gilman, H.; Gerow, C. W. Analogs of Hexaphenylethane. VI. Triphenyl-germyltriphenyltin. J. Am. Chem. Soc. 1957, 22, 334–336. 100

52 The Cambridge Structural Database, version 5.38; Cambridge Crystallographic Data Centre: Cambridge, UK, 2017.

53 Jamieson, C. J. Crystal Structures at High Pressures of Metallic Modifications of Silicon and Germanium. Science 1963, 139, 762–764.

54 Olijnyk, H.; Sikka, S. K.; Holzapfel, W. B. Structural Phase Transitions in Si and Ge under Pressures up to 50 GPa. Phys. Lett. A 1984, 103, 137–140.

55 Cui, H. B.; Graf, D.; Brooks, J. S.; Kobayashi, H. Pressure-Dependent Metallic and Superconducting Phases in a Germanium Artificial Metal. Phys. Rev. Lett. 2009, 102, 237001.

56 Menoni, C. S.; Hu, J. Z.; Spain, I. L. Germanium at High Pressures. Phys. Rev. B 1986, 34, 362–368.

57 Bates, C. H.; Dachille, F.; Roy, R. High-Pressure Transitions of Germanium and a New High-Pressure Form of Germanium. Science 1965, 147, 860–862.

101

Appendix of Supporting Information

Figure S1. Optical properties, (a) absorption and (b) PL, of single source precursors; Ph3GeSnMe3, Ph3GeSnPh3 and TMS3GeSnPh3 in hexane.

Figure S2. CV curves of a 2 mM solutions of precursors, Ph3GeSnMe3, Ph3GeSnPh3, and TMS3GeSnPh3.

102

Figure S3. (a) TGA and DSC traces of solid precursors (b) PXRD patterns after TGA experiment

103

Figure S4. EDX elemental mapping of Sn/Ge core/shell particles from solution phase thermolysis of Ph3GeSnPh3.

Figure S5. EDX elemental mapping of Sn particles from solution phase thermolysis of TMS3GeSnPh3.

104

CHAPTER 5.

CONCLUSIONS AND OUTLOOK

This thesis demonstrates some new directions that molecular chemistry can offer for colloidal nanocrystal synthesis and how different properties of these nanocrystals affect their catalytic properties. As described in Chapter 2, applying a simple bottom up molecular programming approach, tunable library of organophosphite precursors selectively yield Ni12P5,

Ni2P, and Ni nanocrystals over time with different morphologies. Nickel rich Ni12P5 precedes the formation of Ni2P nanocrystals, while some precursors transiently form metallic Ni nanocrystals. We believe that organophosphites are a valuable addition to the phosphide synthetic tool box complementing other phosphorous precursors. Chapter 3 discusses how different morphologies of phosphite derived Ni2P nanocrystals affect the catalytic activity of alkyne hydrogenation reaction. Hollow nanocrystals demonstrate higher catalytic activity towards phenylacetylene hydrogenation that bimodal (a mixture of hollow and solid) Ni2P nanocrystals. These particles can be used up to eight times without losing their catalytic activity significantly while preserving structural integrity. As an effective emerging hydrogenation catalyst, nanocrystalline nickel phosphide should be extensively studied with their different stoichiometric phases and unique morphologies to achieve better catalytic activity. Chapter 4 applies the same molecular programming approach to synthesize more complex nanostructures, Ge-Sn heterostructures. With the aid of computations, we used Ge-Sn single

’ source precursors (R3GeSnR 3) to fine-tune the Ge-Sn bond strength and consequently fine- tune the final composition of Ge-Sn heterostructures. Thermal decomposition of Ge-Sn single molecular precursors with different Ge-Sn bond strengths yield Sn/Ge core/shell nanostructures with varying shell thickness. Tunable Ge-Sn single source molecular precursors 105 will offer new pathways to synthesize Ge-Sn alloys and heterostructures with well-defined compositions and morphologies. These unique structures will offer intriguing properties and essentially many desirable properties to electronic devices and other applications.

This work will offer new tunable family of precursors for pnictide nanocrystals synthesis with well-defined phases, sizes, and morphologies. While our work exhibited feasible synthesis of nickel phosphide nanocrystals with different phases, this tunable precursor family can be simply utilized to synthesize other transition metal phosphide nanostructures and fine-tune the different phases, compositions, and morphologies of the metal phosphide nanocrystals. This will obviate the random optimization of several reaction parameters, which can be very time consuming. Using our reactivity scale of organophosphites,

Liu et al.1 demonstrated that triphenylphosphite, the most reactive organophosphite precursor, can be used as a versatile phosphorus precursor for the scalable and cost effective synthesis of transition metal phosphides including Ni2P, Co2P, MoP, Fe2P and Cu3P. At the same time, tunable single source precursors offer the advantage of an already pre-formed bond and precise control of alloy and heterostructure synthesis with predictable properties. This work reports the synthesis of Ge-Sn heterostructures via tunable Ge-Sn single source precursors. This approach can be extended towards the synthesis other bimetallic alloys and heterostructures with intriguing morphologies and phases. Moreover, there are many studies of organometallic complexes bearing two different elements (e.g. Lantern complexes) with different substituents that could be potentially used as molecular precursors for predictable colloidal synthesis of intermetallic compounds.

Many factors govern the superior catalytic activity of nanocrystals; large surface area, synthetic method, morphology, surface properties, and composition. Our study reveals that the 106

morphology of Ni2P nanoparticles effects the catalytic activity of phenylacetylene hydrogenation. While hollow nanocrystals exhibit better catalytic activity, this can be coupled with surface properties to enhance the catalytic activity of other reactions including the alkyne hydrogenation reaction.

The work presented in this thesis will improve our ability to develop more predictable syntheses of pnictide nanocrystals and Ge-Sn heterostructures with well-defined phases, compositions, and morphologies for catalysis, photovoltaics, and electronic devices.

References

1Liu, J.; Meyns, M.; Zhang, T.; Arbiol, J.; Cabot, A.; Shavel, A. Triphenyl Phosphite as the Phosphorus Source for the Scalable and Cost-Effective Production of Transition Metal Phosphides. Chem. Mater. 2018, 30, 1799–1807.