The Synthesis of Transition Metal Silicide and Germanide Nanostructures Within the Solution- Based Solvent Vapour Growth System

The Synthesis of Transition Metal Silicide and Germanide Nanostructures within the Solution- based Solvent Vapour Growth System

Department of Chemical Sciences and Bernal Institute, University of Limerick

Martin Sheehan

Supervisor: Prof. Kevin M. Ryan

Submitted to the University of Limerick for the Degree of Doctor of Philosophy June 2017 Table of Contents

Acknowledgements ...... i Declaration ...... ii Abstract ...... iii List of Publications ...... v

Chapter 1: Introduction ...... 1

1.1 Background ...... 1 1.2 Formation of Silicon and Germanium based Nanowires (NWs) ...... 2 1.2.1 Elemental Silicon and Germanium NWs ...... 2 1.2.1.1 NW Growth Mechanisms ...... 2 1.2.1.2 Approaches to Silicon and Germanium NW Growth ...... 8 1.2.1.2.1 Chemical Vapour Deposition ...... 8 1.2.1.2.2 Supercritical Fluid-based Synthesis ...... 9 1.2.1.2.3 Solvent Vapour Growth System ...... 10 1.2.2 Formation of Transition Metal Silicide and Germanide NWs ...... 12 1.2.2.1 Silicidation/Germanidation of preformed NWs ...... 12 1.2.2.2 Silicon/Germanium delivery to metal film growth substrates...... 16 1.2.2.2.1 Synthesis of Nickel Silicide NWs ...... 16 1.2.2.2.2 Synthesis of Nickel Germanide NWs ...... 23 1.2.2.2.3 Synthesis of Copper Silicide NWs ...... 24 1.2.2.3 Transition metal delivery to Silicon/Germanium growth substrates ..... 28 1.2.2.4 Simultaneous metal and Silicon/Germanium delivery to growth substrates ...... 31 1.3 Potential Applications of Transition Metal Silicides and Germanides ...... 38 1.3.1 Nanoelectronics ...... 38 1.3.2 Nanoscale Field Emitters ...... 39 1.3.3 Lithium-ion Batteries ...... 40 1.4 References ...... 42

Chapter 2: Experimental Procedures and Characterisation Techniques ...... 60 2.1 General Synthetic Strategy ...... 60 2.2 Characterisation Techniques ...... 62 2.2.1 Scanning Electron Microscopy (SEM) ...... 62 2.2.2 Transmission Electron Microscopy (TEM)...... 65 2.2.2.1 Electron Diffraction ...... 67 2.2.2.2 Scanning Transmission Electron Microscopy (STEM) ...... 68 2.2.3 Energy Dispersive X-ray Spectroscopy (EDS) ...... 70 2.2.4 X-ray Diffraction (XRD) ...... 71 2.2.5 Electrochemical Characterisation ...... 73 2.2.5.1 Galvanostatic Charge and Discharge Cycling ...... 73 2.2.5.2 Calculation of C-rate ...... 75 2.3 References ...... 76

Chapter 3: The Selective Synthesis of Nickel Germanide Nanowires and Nickel Germanide Seeded Germanium Nanowires within a Solvent Vapour Growth System ...... 79 3.1 Abstract ...... 79 3.2 Introduction ...... 79 3.3 Experimental Section ...... 82 3.4 Results and Discussion ...... 84 3.5 Conclusion ...... 102 3.6 References ...... 103

Chapter 4: The Effect of Enhanced Precursor Decomposition on the Synthesis of Copper Silicide Nanostructures within a Solvent Vapour Growth System ...... 110 4.1 Abstract ...... 110 4.2 Introduction ...... 110 4.3 Experimental Section ...... 112 4.4 Results and Discussion ...... 114

4.5 Conclusion ...... 133 4.6 References ...... 134

Chapter 5: Synthesis of Nickel Silicide Nanowires and Axial Heterostructure NixSi/Si Nanowires within a Solution-based System ...... 138 5.1 Abstract ...... 138 5.2 Introduction ...... 138 5.3 Experimental Section ...... 141 5.4 Results and Discussion ...... 143 5.5 Conclusion ...... 156 5.6 References ...... 157

Chapter 6: Amorphous Silicon coated Copper Silicide Nanowire Arrays for Lithium-ion Battery Anodes ...... 163 6.1 Abstract ...... 163 6.2 Introduction ...... 163 6.3 Experimental Section ...... 166 6.4 Results and Discussion ...... 168 6.5 Conclusion ...... 182 6.6 References ...... 183

Chapter 7: Conclusions and Recommendations for Future Work ...... 188 7.1 Conclusions ...... 188 7.2 Recommendations for Future Work ...... 191 7.2.1 FIB sectioning to study NW Formation ...... 191 7.2.2 Alternative Si/Ge Precursors ...... 192

Appendix A: XRD Reference Patterns ...... 193

Acknowledgements

I would first like to acknowledge my supervisor, Prof. Kevin M. Ryan, for his guidance and support over the last number of years. I am eternally grateful for this opportunity he has given to me. It has been a genuine pleasure to have worked in his research group.

I would also like to acknowledge my fellow group members, both past and present. I am very lucky to have made such truly great friends over the last number of years. Claudia and Grace: Thank you both so much for all the help and the laughs and the brownies…. OH GOD, THE BROWNIES… You have no idea how grateful I am to have made such great friends… Hugh and Tadhg: Thank you both for all the help… I now know how a battery works thanks to your excellent tuition! Killian: I’ll miss the stupidity…and the foosball…. OH GOD, THE

FOOSBALL…. Barry better be looking after you! Pai: I have to name my first born after you. Shalini: you have to name your first born after me. Sarah and Ger:

I wish youze the best of the luck… see I can speak ‘Dublin’… Barry: Thank you for introducing me to McDonald’s breakfast….. Yeah…. Thanks for that….

I would also like to thank the staff of the MSSI/Bernal, who made my time there so special. Brid: Thanks for all the help and the gossip! Yina/Paula/Sergei: Thank you for all the help with the TEM/SEM.

Lastly, I would like to thank my family for all their support, in particular my parents, Con and Patricia, this would genuinely not have been possible without their support. Also, thanks for letting me steal the car for the last few months…

Declaration

I declare that the work described in this thesis is entirely the result of my own work carried out in the University of Limerick, except for the assistance mentioned in the acknowledgements section and the collaborative work for publications.

This thesis has not previously been submitted in whole or in part for the award of a degree at this or any other university.

......

Martin Sheehan

Martin Sheehan,

Department of Chemical Sciences and Bernal Institute,

University of Limerick,

Ireland.

Abstract

This thesis describes the use of the solution-based solvent vapour growth (SVG) system in the synthesis of transition metal germanide and silicide nanostructures, as well as the use of transition metal silicide nanowire arrays as high surface area current collectors for lithium ion battery applications.

The growth of nickel monogermanide (NiGe) nanowires directly from Ni foil, within the SVG system, is presented in Chapter 3. The evolution of nanowire growth from the Ni substrate was studied and it was found that the Ge precursor thermally decomposes to form a rough nickel germanide layer on the Ni substrate, from which NiGe nanowire growth subsequently occurs. Investigations into the use of thin Ni films as the growth substrates were conducted and showed that when thin nickel films were used, germanium nanowire growth occurred through the established vapor-solid-solid growth mechanism.

Chapter 4 focuses on the effect of Si precursor decomposition on the formation of copper silicide nanostructures on Cu. By increasing the reactivity of the Si precursor used, through the addition of a reducing agent, the reaction temperature required for the synthesis of copper silicide (Cu15Si4) nanowires was decreased by over 100 °C to 350 °C. At elevated temperatures (≥ 400 °C), it was observed that the enhanced reactivity of the silicon precursor resulted in the formation of unique nanostructures of the more Si rich copper silicide phase, Cu3Si.

In Chapter 5, the development of a novel one-step synthetic approach to the synthesis of nickel silicide and axial heterostructure nickel silicide-silicon (NixSi-

Si) nanowires, within the SVG system, is described. Growth of these nanowires was achieved by growing gold seeded silicon nanowires on nickel substrates using

iii the established solution-liquid-solid mechanism. The high temperature used for silicon nanowire growth allowed nickel to diffuse into the silicon nanowire from the substrate and for in situ silicidation of the silicon nanowire to occur.

In Chapter 6, the use of copper silicide (Cu15Si4) nanowire arrays as high surface area current collectors for silicon lithium ion battery anode applications is outlined. Amorphous Si films were deposited onto pre-formed Cu15Si4 nanowire arrays and planar copper substrates prior to electrochemical testing.

Electrochemical testing revealed markedly improved capacity retention values for the Cu15Si4 nanowire arrays compared to those seen for amorphous Si deposited on planar Cu substrates.

List of Publications

Sheehan, M., Guo, Y., Flynn, G., Geaney, H., Ryan, K.M. ‘The selective synthesis of nickel germanide nanowires and nickel germanide seeded germanium nanowires within a solvent vapour growth system’. CrystEngComm, 2017, 19, 2072-2078.

Flynn, G; Palaniappan, K., Sheehan, M., Kennedy, T., Ryan, K.M. ‘Solution Synthesis of Lead Seeded Germanium Nanowires and Branched Nanowire Networks and their Application as Li-Ion Battery Anodes’. Accepted, May 2017.

Stokes, K., Geaney, H., Flynn, G., Sheehan, M., Kennedy, T., Ryan, K.M. ‘Synthesis of Si1-xGex Alloy Nanowires with Tunable Compositions for Optimized Lithium Ion Battery Anodes’. Awaiting Submission

Chapter 1: Introduction

1.1 Background

Transition metal silicides and germanides are a class of intermetallic compounds, consisting of a transition metal and Si or Ge, respectively.1 The rich phase diagrams between transition metals and Si/Ge allow for the formation of a wide range of stoichiometries, varying from metal-rich to semiconductor-rich. This tunability enables the properties of transition metal silicides and germanides to be manipulated, by controlling the composition.2 For instance, NiSi, NiGe, PtGe,

3-5 CoSi2 and TiSi2 behave like metals and display low electrical resistivities. On the other hand, transition metal silicides, such as Mn1.8Si, FeSi2 and CrSi2, have semiconducting properties.6-9 Due to this variety of properties, transition metal silicides have been utilised in a number of different applications, including heating elements, thermoelectric conversion and as high temperature coatings.10

In the 1980’s, transition metal silicides started to attract interest from the microelectronics industry, due to their low resistivities, stable crystal structures

11 and their compatibility with Si. In particular, TiSi2, CoSi2 and NiSi, have been used as metal contacts and as gates in microelectronics.12-14

With the challenging scaling demands imposed by the microelectronic industry, and inspired by Moore’s law, a great deal of research has focused on the synthesis and study of one-dimensional (1D) nanowires (NWs).15-17 While the majority of efforts so far have focused on semiconductor NWs, more complex compound

NWs have also been investigated (e.g III-V,II-VI)18,19, with transition metal silicide and germanide NWs representing an attractive material set due to their potential use in a number of technological applications. In addition, these NWs

1 also represent a versatile platform for examining key material phenomena, such as metal diffusion in semiconductors, the ability to form targeted stoichiometries in complex material systems and investigations into unusual 1D growth mechanisms.

As a result, diverse synthetic protocols have been developed for NW growth.20-22

1.2 Formation of Silicon and Germanium based Nanowires (NWs)

In this section, an overview of the synthesis of 1D Si and Ge based NWs will be given, with a particular focus on bottom-up synthetic approaches to complement the scope of research work and NW synthetic protocols which were employed in this thesis.

1.2.1 Elemental Silicon and Germanium NWs

1.2.1.1 NW Growth Mechanisms

The synthesis of Si and Ge NWs is typically achieved through the vapour-liquid- solid (VLS) growth mechanism, which was proposed by Wagner and Ellis in

1964.23 In the VLS growth mechanism, shown schematically in Figure 1.1 for the

Au-Si system, a metal particle is used to seed NW growth. Initially, Si precursor vapour is thermally decomposed to form Si monomers, which subsequently alloy with the metal seed (above the eutectic temperature of the Au-Si system) to create a liquid Au–Si eutectic alloy. As this liquid alloy becomes supersaturated with more Si, solid Si extrudes from the liquid alloy and this forms an anisotropic Si

NW.

Figure 1.1. Schematic depiction of the VLS growth mechanism.24

Since the initial report by Wagner and Ellis, Au has been the primary metal used to seed Si and Ge NWs. 25,26 However, alternative catalytic seeds have attracted considerable research interest, due to incompatibility of Au with semiconductor processing.27-29 As a result of these studies, Si and Ge NW growth has been reported for the following catalytic seeds: Ag,30 Al,31,32 Bi,33 Co,34 Cu,35,36 Fe,37

Ga,38 In,39 Mn,40 Ni,41,42 Pt,43 Sn,44 and Ti.45

However, NW growth can occur via different growth mechanisms, due to differences in the respective binary phase diagrams of the metals with Si or Ge.

Thus, seed materials are typically classified into three different types, as outlined in the periodic table shown in Figure 1.2.

Figure 1.2. Periodic table showing the nature of the elements used for NW growth.46

Type-A seeds (i.e. Au, Ag and Al) have a high % of Si and Ge in their eutectic alloys. In the case of the Au-Si system, the amount of Si in the Au-Si alloy needs to be greater than 19%, before NW growth can be initiated. In comparison, Type-

B seeds (i.e. In, Sn and Zn) have a low % of Si and Ge (< 1%) in their eutectic alloys, meaning that the liquid droplet is readily supersaturated. Type-C seeds (i.e.

Cu, Ni, and Fe), are quite different to Type-A and Type-B seeds, as these metals are known to form transition metal silicides and germanides. Figure 1.3 shows the complex phase diagram for the Cu-Si system. At room temperature, there are three stable copper silicide phases (Cu3Si, Cu15Si4 and Cu5Si). For type-C seeds, the temperatures required for VLS growth are much higher. In the case of a Cu seed, a temperature greater than its lowest eutectic point (802 °C) is needed to seed Si NWs via the VLS mechanism.

However, sub-eutectic NW growth is possible through the slightly different vapour-solid-solid growth (VSS) mechanism.25,26,47 In the VSS mechanism, the Si or Ge monomers typically alloy with the metal to form a solid silicide or germanide particle, instead of a liquid alloy. This particle will continue to alloy with Si and Ge monomers until it forms the most Si or Ge rich phase in the phase

42 diagram at that temperature (η’’Cu3Si at 450 °C in Figure 1.3). At this point, this silicide/germanide particle becomes supersaturated with the Si or Ge monomers and eventually a Si or Ge NW will extrude from the particle.

In the VSS mechanism, the crystal orientation of the seed particle in respect to the

NW also plays an important role in how the NW nucleates and grows. In the case of Cu3Ge seeded Ge NWs, it was found that there was an epitaxial relationship between the {200} planes of the orthorhombic Cu3Ge seed and the {220} planes of the Ge NW, with NW growth predominantly taking place in the <110> direction as a result.48 The authors of this report attributed this relationship to the similar d-spacings of the (200) and (220) planes of Cu3Ge and Ge, respectively.

Similar epitaxial relationships between the seed particle were also observed in

Cu3Si seeded Si NWs, where it was noted that the Cu3Si/Si interface again played an important role in the growth direction of the Si NW.36 Thombare et al. also showed that for NiGe seeded Ge NWs, the interface between the seed and the NW played a role in the morphology of the NWs, where the well-defined interface between the seed and the NW for thin NWs (<25 nm) led to the formation of straight <110> oriented NWs.49 However for larger diameter Ge NWs, Thombare et al. found that the lattice mismatch between the NiGe seed and Ge NW led to excessive twinning in the Ge NWs which in turn led to torturous NW growth. In another report by Barth et al, it was shown that defects can intentionally be

5 transferred from the solid seed to the Ge NW using this orientational relationship between a solid seed and the nucleated NW.49

The VSS has some potential advantages over the VLS system, due to its use of a solid seed, such as: (i) reduced growth temperatures, (ii) uniform diameter distributions, (iii) better control of growth orientation, (iv) increased purity and

(v) more abrupt heterostructure interfaces. For these reasons, VSS growth using

Type-C seeds has attracted considerable research interest.47

Figure 1.3. Cu-Si phase diagram.50

Another less common growth mechanism used in the synthesis of NWs is the vapour-solid (VS) mechanism.51-53 Typically, this mechanism is used for the

54 55 56 synthesis of metal and metal oxide NWs, such as Sn, SnO2, and ZnO. Unlike the VLS and VSS growth mechanisms, previously discussed, the VS growth mechanism is quite different in that no discrete particle seeds are required for NW growth. Instead, NW growth is achieved by controlling the supersaturation profile of the vapour phase reactants and the subsequent condensation onto a growth

6 substrate. In the case of VS grown Sn NWs, Sn vapour was condensed onto a Si substrate and crystallised. At a low supersaturation value, the condensation and crystallisation of the Sn was limited leading to a competition between the rates of growth of the different crystallographic planes after a certain critical size nucleus had formed. Crystallographic planes with higher surface free energies are more unstable leading to a faster a growth rate. In the case of Sn NWs, the {100} planes possess the highest surface free energy per unit area, resulting in a fastest growth rate along the {100} planes leading to the formation of [100] oriented Sn anisotropic NWs.54 At higher supersaturation values, the competition between growth rates of different crystallographic planes was not as extreme and growth took place along the [200] and [020] of Sn nanosquares. In other systems, it has been observed that higher levels of supersaturation of the vapour phase reactants can even lead to the formation of bulk morphologies.57,58 Thus, control over the supersaturation profile, is quite important to ensure NW formation when utilising the VS mechanism for NW growth.

1.2.1.2 Approaches to Silicon and Germanium NW growth

1.2.1.2.1 Chemical Vapour Deposition

One of the most common ways to grow Si or Ge NWs through the VLS or VSS mechanisms is by chemical vapour deposition (CVD).46 This technique involves

59 the flow of gaseous precursors, such as silane (SiH4), silicon tetrachloride

60 61,62 (SiCl4) and germane (GeH4), into a chamber containing a growth substrate coated with metal seeds. Figure 1.4 shows a schematic representation of a CVD setup used for the synthesis of Si NWs.

Figure 1.4. a) Schematic representation of CVD setup used for Si NW growth. b) Si NW array grown using CVD.25

Si and Ge NWs can be grown with high levels of control in this system because the precursor supply can be accurately controlled by controlling reaction conditions, such as flow rate and reaction temperature. Complex Si and Ge NW structures, such as axial and radial heterostructure NWs can also be formed using

CVD.62-64 In addition, doping of NWs is possible by including gaseous dopant

65,66 precursors, such as diborane (B2H6) and phosphine (PH3). This level of control makes CVD a very attractive approach for the synthesis of NWs.

However, the product yields in CVD are quite low, which somewhat limits its use for applications that require large quantities of NWs, such as lithium-ion battery

8 applications. Moreover, the high cost of the equipment involved can make CVD processes less appealing for NW syntheses.46

1.2.1.2.2 Supercritical Fluid-based Synthesis

Alternative approaches for NW growth were developed to increase yields, reduce costs, to alleviate the dependence on hazardous precursors such as SiH4 and GeH4 and to ultimately overcome the limitations of CVD processes. In particular, solution-based growth systems have attracted significant attention.

The first solution-based approach was developed by Korgel and co-workers and utilised supercritical fluids (SCFs) to reach the reaction temperatures required for

Si and Ge precursor decomposition.67 In a typical reaction setup, a mixture of the seed nanoparticles with the Si (or Ge) precursor in a fixed ratio was made up in a suitable solvent (i.e. benzene, hexane and toluene).68-71 This mixture was consequently fed into a pressurised reactor, as shown in Figure 1.5. The precursors (diphenylsilane) decomposed and NW growth proceeded through supercritical fluid-liquid-solid growth (SFLS) or supercritical fluid-solid-solid

(SFSS) growth mechanisms depending on the seed particle.42,72

This setup can be slightly altered to grow NWs directly from a substrate, by placing the substrate inside the reactor. Yuan et al. also used this approach to synthesise Si NWs through the SFLS and SFSS mechanisms from a number of different metals, including Ag, Al, Pb, Cu, Ni and Ti.73 Despite the success of the

SCF approach in the synthesis of Si and Ge NWs, the use of high pressures and flammable solvents makes the SCF system potentially very hazardous.

Figure 1.5. a) Schematic of SCF NW growth system. b) TEM image of a Si NW formed in the SCF system.24

1.2.1.2.3 Solvent Vapour Growth System

Alternative solution-based approaches have used high boiling point solvents

(HBSs), such as squalane, squalene or dotriacontane, to reach the reaction temperatures required for Si and Ge NW growth.74-76

The solvent vapour growth (SVG) system was pioneered by Ryan and co-workers and utilises the vapour phase of the HBS as a reaction medium for NW growth.77

In the SVG system, the HBS is placed in a long-neck, round-bottomed flask and is connected to a water condenser, as shown schematically in Figure 1.6. A three- zone furnace is then used to heat the HBS beyond its boiling point, to establish a vapour medium in the round-bottomed flask. The Si or Ge precursor, typically either phenylsilane (PS) or diphenylgermane (DPG), is then injected into the

77,78 system. The precursor decomposes to form SiH4 or GeH4 in situ, which in turn decomposes to provide monomers for NW growth through the VLS or VSS mechanisms.

Figure 1.6. Schematic of the SVG system for Sn seeded Si and Ge NW growth79

The SVG system was initially used for the ‘self-seeded’ synthesis of Ge NWs, but its use has since been extended to allow for the growth of Si and Ge NWs from

Type-B catalytic seeds (i.e. In and Sn).78-81

The SVG system has also been used for the synthesis of Ge NWs, directly grown from Cu foil.48 In this report, the Ge precursor (DPG) decomposed to form a copper germanide (Cu3Ge) layer on the substrate of the Cu foil. As the reaction proceeded, this layer became supersaturated with Ge monomers, thus resulting in the growth of Ge NWs through the VSS growth mechanism.

1.2.2 Formation of Transition Metal Silicide and Germanide NWs

In this section, an overview of the typical methods used for the synthesis of free- standing transition metal silicide and germanide NWs are outlined. In comparison to the synthesis of elemental Si and Ge NWs, the synthesis of transition metal silicide and germanide NWs is much more complex.1,20,22 This is due to the absence of simple phase diagrams, which allow for growth of NWs through the

VLS growth mechanism, and the ability of transition metals to act as seeds for Si or Ge NW growth through the VSS growth mechanism. Nevertheless, there have been numerous reports on the synthesis of transition metal silicide and germanide

NWs. These reports can be divided into four different approaches:

1. Silicidation/Germanidation of preformed NWs.

2. Si/Ge delivery to a transition metal growth substrate.

3. Transition metal delivery to a Si/Ge growth substrate.

4. Simultaneous delivery of the transition metal and Si/Ge to a growth

substrate.

1.2.2.1 Silicidation/Germanidation of preformed NWs

One of the first approaches employed for the synthesis of both transition metal silicide and germanide NWs was the silicidation/germanidation of preformed

Si/Ge NWs. In this approach, the Si/Ge NW template undergoes a solid state reaction with the transition metal at elevated temperatures to form a transition metal silicide or germanide NW. This approach was first used by Lieber and co- workers for the synthesis of NiSi NWs.82

Figure 1.7: a) Preparation of NiSi NWs. Si NWs (blue) (1) of uniform diameter are coated with Ni metal (green) (2) to a total thickness comparable to the Si NW diameter, reacted at 550 °C to form NiSi nanowires (brown) (3), and then etched to remove any excess Ni metal (4). b) TEM image of an individual NiSi NW after silicidation, with the electron diffraction pattern shown in the inset.82

In this report, Au seeded Si NWs were coated with a thermally evaporated, thin Ni film, as shown schematically in Figure 1.7. Subsequent annealing of the Ni-coated

Si NWs caused the Ni to react with the Si NW backbone, which resulted in the formation of NiSi NWs. Excess Ni was then removed from the surface of the NW by using a Ni selective wet etch.

Lieber and co-workers expanded on this approach to synthesise multi-segment axial NiSi-Si heterostructure NWs by incorporating a lithographical process into their procedure (Figure 1.8), which allowed for the evaporation of Ni onto specific segments of the Si NW and for selective silicidation of the Si NW backbone.

TEM analysis of the NiSi-Si heterostructure NWs showed that the interface between the NiSi and Si segments was atomically abrupt.

Figure 1.8: a) Fabrication of NiSi/Si NW heterostructures. Si NWs (blue) (1) dispersed on a substrate are coated with photoresist (grey) (2) and lithographically patterned, selectively coated with Ni metal (green) (3) to a total thickness comparable to the Si NW diameter, and reacted at 550 °C to form NiSi NWs (4). b) TEM image of a NiSi/Si heterostructure NW (scale bar is 1 µm), with NiSi segments highlighted by arrows. A high-resolution TEM image of the junction between NiSi and Si is shown in the inset (scale bar is 5 nm).82

Since this initial report, similar methodologies have been used by a number of groups to synthesise a wide variety of transition metal silicide and germanide

83 84 85,86,87 88,89 90 91 92 NWs, such as NiSi, NiSi2, MnSi, , PtSi, FeSi2, Cu3Ge, Cu3Si and

CoSi.93 In particular, many of these reports have focused on the lateral diffusion approach for the synthesis of axial heterostructure NWs, as shown for Cu3Ge in

Figure 1.9.91 In this approach, lithographically defined pads of the transition metal are placed on the NW. By controlling the anneal time, partial silicidation

(heterostructure NWs) or complete silicidation/germanidation of the NWs can occur.

Figure 1.9: (a) SEM image of the 1 μm long intrinsic Ge-NW with Cu contact pads before the germanidation process. (b) SEM snapshot of Cu3Ge/Ge/Cu3Ge heterostructures after 2 min annealing with a remaining length of the unreacted Ge-NW segment of about 180 nm. (c) SEM snapshot after 4 min annealing. The unreacted Ge-NW length is about 15 nm.91

Recently, in situ TEM observations have been used to determine how the silicidation or germanidation process takes place in transition metal silicide and germanide NWs.20,21 In one report, Lu and co-workers observed the formation of

NiSi NWs by in situ TEM, by annealing a dispersion of Si and Ni NWs that were dropcast onto a suitable TEM grid in high concentrations to ensure contact between a Si NW and a Ni NW.83 Interestingly, the NiSi segments nucleated at the ends of the NW and proceeded back along the NW. The authors attributed this unique nucleation to the point contact between the Ni NW and the Si NW.

Numerous other reports have since stressed the importance of the contact between the metal and the Si NW and the role that this plays on transition metal silicide

NW formation.94-96 Despite the success of the silicidation/germanidation approach in the synthesis of transition metal silicide and germanide NWs and heterostructure NWs, the throughput of this approach is low, which limits its overall applicability.

1.2.2.2 Silicon/Germanium delivery to metal film growth substrates

The delivery of Si or Ge to a transition metal film growth substrate is an alternative approach to the synthesis of transition metal silicide and germanide

20 NWs. In this approach, gaseous based precursors, such as SiH4 or GeH4, are decomposed on transition metal films to form transition metal silicide or germanide NWs.97,98

1.2.2.2.1 Synthesis of Nickel Silicide NWs

The report by Yan et al. was one of the first reports into the synthesis of nickel silicide NWs using this methodology.99 In this report, Ni substrates were placed into a CVD chamber, where SiH4 was delivered to the Ni substrates at a reaction temperature of 500 °C for 60 min. Figure 1.10 shows an SEM image of Ni substrate which was extensively covered with NWs that were several microns in length and had uniform diameters of between 30 – 80 nm. Interestingly, no catalytic seeds were seen at the ends of the nickel silicide NWs, which led the authors to exclude the conventional VLS and VSS growth mechanisms as possible mechanisms for NW formation. Through TEM imaging, electron diffraction and energy dispersive X-ray spectroscopy measurements, the NW composition was determined to be Ni2Si.

Figure 1.10. SEM image of NWs grown on a Ni substrate.99

By modifying this CVD reaction setup, a number of groups have reported on the formation of nickel silicide NWs with different stoichiometries.100 For example,

Lee et al. reported the synthesis of carbon coated NiSi NWs in a CVD setup at a

101,102 reaction temperature of 800 °C. Kim et al. synthesised Ni3Si2 NWs by

102 decomposing SiH4 on Ni films at 350 °C using a plasma-enhanced CVD setup.

In a separate report by Liu et al., Ni2Si NWs were synthesised using SiH4 decomposition on Ni foil in a CVD setup at 650 °C.103 Figure 1.11 shows SEM and TEM images of the NWs formed in this report.

Figure 1.11. a) Top-view and b) tilted SEM images of NW growth from Ni foil. c) Low magnification TEM image of NWs, with a high magnification image shown in d).103

Interestingly, the NWs appeared to be composed of two parts, a root and a tip. In the TEM image in Figure 1.11c, it can be seen that the NW root has quite a broad diameter and a rough surface. On the contrary, the NW tip is quite thin (10 – 50 nm) and grows directly from the root. Thus, the authors suggested that NW growth took place in four sequential processes, which are shown schematically in

Figure 1.12.

103 Figure 1.12. Proposed mechanism for Ni2Si NW growth from Ni foil.

Initially, the SiH4 reacts with the Ni foil to form Ni2Si nanocrystals

(Figure 1.12a). As the reaction proceeds, these Ni2Si nanocrystals continue to grow one-dimensionally to form the NW root, due to the continued decomposition of SiH4 and the out diffusion of Ni from the foil (Figure 1.12b). The authors attributed this anisotropic growth to low supersaturation levels of Si in the vapour phase, which limited the number of nucleation sites. When the rate of incorporation of Ni, through diffusion, equalled the rate of decomposition of SiH4, thin untapered Ni2Si NWs were formed (Figure 1.12c) and NW growth continued until the SiH4 supply was stopped (Figure 1.12d). Similar root-tip structures were

104 seen by Jiang et al. in their report on the synthesis of Ni3Si2 NWs on Ni foam.

In the report by Kim et al. on NiSi NW formation, Ni thin films (60 – 80 nm in thickness) were thermally evaporated onto support substrates, which were then loaded into a CVD setup, where the growth temperatures and the partial pressures

105 of SiH4 were varied from 250 – 600 °C and 10 – 100 Torr, respectively. In this

19 report, it was found that reaction temperature had a significant effect on NW formation. For example, at 250 °C, the decomposition of SiH4 and the Ni diffusion were not sufficient to completely convert the Ni layer to nickel silicide. At the higher temperature of 600 °C, rough nickel silicide films were formed due to the excessive decomposition of SiH4 and Ni diffusion. However, at 400 °C, the decomposition of SiH4 and diffusion of Ni from the Ni layer was found to be optimal, which allowed for the growth of NiSi NWs. The partial pressure of SiH4 was also found to effect NW formation, with an increased partial pressure leading to a change in the NW morphology and the formation of dendritic NWs. The authors attributed this effect to the supersaturation levels of the gaseous Si phase.

At lower partial pressures, the supersaturation of Si in the vapour phase is low, favouring the formation of 1D morphologies because of the limited nucleation sites. As the partial pressures increased, the supersaturation levels increased which favoured the formation of the dendritic NW like structures, as shown in

Figure 1.13.105

105 Figure 1.13. NiSi NWs formed at different SiH4 partial pressures

In an additional study carried out by the same group, the NiOx overlayers on the

106 Ni films were also found to play a role in NW formation. When the NiOx layer was reduced by annealing the substrates in H2 prior to the SiH4 CVD process, a nickel silicide film formed on the substrate. The authors investigated the role of the NiOx overlayers by incorporating an oxidation step into their growth protocol.

Optimal NW growth was observed when the oxide layer was between 20 – 30 nm in thickness. If the oxide layer was much thicker, the density of NWs formed on the substrate was decreased. This led the authors to hypothesise that NiOx acted as a diffusion barrier for Ni from the Ni underlayer, which in turn limited nucleation and favoured 1D growth.

Recently, NixSi NW growth has been shown to occur on Ni films reduced by

107 annealing in the presence of H2. In this study, the Ni substrates were annealed in H2 prior to a CVD process. When SiH4 was delivered to the Ni substrate, it resulted in the formation of nickel silicide thin films on the Ni foil. However, the authors were able to synthesise NixSi NW arrays by the introduction of an additional carrier gas such as H2 or Ar.

In some reports of Si delivery to a transition metal foil, Si powders have been used in place of SiH4 gas as the Si source. For example, in the synthesis of

Ni31Si12 NWs, Si powder was placed in an alumina boat upstream of a Ni foil substrate.108 By heating the Si powder to 1,100 °C at low pressures, a Si vapour was produced, which reacted with the Ni substrate downstream at a reaction temperature of 750 °C to form a Ni31Si12 film. From this film, epitaxially grown

Ni31Si12 NWs were formed. The setup was later modified by Lee et al. to grow

109 NiSi2 NWs from Ni foil. In this particular report, NiCl2 powder was placed in front of the Ni foil to enhance the deposition of Si atoms and to suppress the supply of the Ni atoms. The authors attributed the enhanced deposition of Si and the Ni suppression to the formation of SiCl4 and Cl2 gases in situ. Despite the high temperatures used in this method (i.e. 1,100 °C) and the long reaction times (>

109 180 min), this is the only report of NiSi2 NWs grown from Ni foil to-date.

Other reports have used sputtered Si as the Si source for nickel silicide NW formation.110,111 In these reports, Si was sputtered onto Ni films which are held at a substrate temperature of 575 °C. The sputtered Si, subsequently, reacted with the Ni film to form a rough NiSi film from which NiSi NWs formed as the reaction was allowed to proceed.

1.2.2.2.2 Synthesis of Nickel Germanide NWs

Yan et al. reported the first synthesis of transition metal germanide NWs using Ge delivery to a transition metal foil.97 In this report, Ni foam substrates were placed in a CVD chamber and were reacted with GeH4 gas for 120 min to form NiGe

NWs. Similar to the reports on the synthesis of nickel silicide NWs, the authors of this report found that the reaction temperature and the chamber pressure had to be carefully controlled to allow NW formation. Figure 1.14 shows SEM images of the Ni substrates after reactions at different conditions.

Figure 1.14. SEM images of Ni substrates after different GeH4 CVD processes.

At a reaction temperature and pressure of 300 °C and 15 Torr, sparse NW growth could be seen on the substrate. Interestingly, increasing the temperature to 350 °C, to enhance GeH4 decomposition, resulted in the formation of Ge nanocones on the sample. The authors postulated that the formation of these Ge nanocones was due to nonspecific sidewall deposition. To suppress sidewall deposition, the pressure was lowered to 4 Torr with the reaction temperature set at 350 °C. At these

23 conditions, the density of NiGe NWs across the substrate was increased. Further increasing of the reaction temperature to 400 °C led to the formation of nickel germanide thin films. TEM analysis of the NWs confirmed the crystal structure to be NiGe. Interestingly, X-ray diffraction (XRD) analysis of the NW covered substrate indicated the presence of the Ni5Ge3 phase also. The authors proposed that the Ge reacted to form nickel germanide films initially on the Ni foam and as the growth proceeded, NWs nucleated from this germanide layer. Again, no catalytic seeds were seen at the end of the NWs which allowed the authors to exclude the VLS and VSS growth mechanisms. Interestingly, the NWs could be transformed to the more nickel-rich germanide phase, ɛ-Ni5Ge3, by annealing the

NiGe NW substrates at 500 °C for 120 min in a He atmosphere.

1.2.2.2.3 Synthesis of Copper Silicide NWs

Until recently, the Si/Ge delivery approach for transition metal silicide and germanide NW synthesis had only been reported for silicides and germanides of

Ni. The limitation of this approach to Ni was suggested to be due to the fast diffusivity of Ni in Si and Ge.20

Geaney et al. were able to extend the Si delivery approach to the synthesis of

112 copper silicide (Cu15Si4) NWs from bulk Cu foil. In this report, Cu foil was placed into the solution based SVG system. The reaction temperature was ramped to 460 °C, which was above the boiling point of the HBS squalane establishing a vapour reflux. The organometallic precursor, PS, was then injected into the SVG system where it decomposed to form SiH4 gas in situ. Figure 1.15 shows SEM images of the NW covered substrate after a reaction time of 30 min. The NWs

24 were less than 10 µm in length and had a tight diameter distribution of approximately 110 nm. TEM analysis of the NWs confirmed the structure to be

Cu15Si4.

112 Figure 1.15. SEM images of the Cu15Si4 NW arrays grown directly on Cu foil.

To understand the formation of the Cu15Si4 NWs, Geaney et al. performed a reaction time study. Initially, the PS reacts with the Cu foil to form a crystallite layer on the surface of the Cu foil. As the reaction proceeds, the growth changes from crystallite formation to 1-dimensional NW growth. The authors related this change in crystal habit to the supersaturation levels of Si in the vapour phase.

When the PS was injected initially, the Si supersaturation levels were high and bulk silicide formation was preferred. As the reaction proceeded, the Si supersaturation levels decreased due to silicide formation. At these low supersaturation levels, NW growth from this crystallite layer was favoured.

Interestingly, when the authors limited the Cu source by using thermally evaporated Cu films on support substrates, Cu3Si seeded Si NW growth occurred.

In this case, the Cu source was exhausted in initial silicide formation and subsequent Si saturation led to Si NW growth through the VSS growth mechanism.

Yuan et al. also reported the synthesis of copper silicide NWs from Cu foil using a supercritical fluid based setup.113 In this report, PS was decomposed in supercritical benzene at reaction temperatures between 420 – 475 °C to form copper silicide NWs in high density across the Cu foil. XRD and TEM analysis of the NWs confirmed the structure to be Cu3Si. Interestingly, in an earlier report by

Yuan et al., Cu foil was used as a seed for Si NWs.73 In this report, PS was decomposed in supercritical benzene at the higher reaction temperature of 550 °C.

These reports highlight the importance of controlling Si precursor decomposition to tune whether transition metal silicide NWs or transition metal seeded Si NWs form.

Recently, an alternative approach to the synthesis of copper silicide nanostructures (NSs) has been reported within a CVD based setup.114 In this report, Cu or Cu5Si substrates were placed in a CVD chamber and exposed to different silane based gases with different carrier gases and pressures at a fixed reaction temperature of 500 °C. In addition to the synthesis of Cu83Si17 and Cu3Si

NWs, the authors were able to form a variety of unique Cu3Si NSs by varying the reaction conditions. Figure 1.16 shows an image of Cu3Si nanoplatelets (NPs) synthesised at a pressure of 200 Pa using a mixture of SiH4 and air in a 20:1 ratio.

114 Figure 1.16. SEM image of Cu3Si NPs formed in CVD process.

Again in this report, the SiH4 pressures played an important role on the morphology of the structures formed. At a pressure of 1300 Pa, Cu83Si17 NWs and

Cu3Si NPs formed. When the pressure was increased to 1600 Pa, Cu3Si NPs were exclusively formed. At the even higher pressure of 3600 Pa, Cu83Si17 microcrystals formed. This delivery of Si/Ge to a transition metal approach appears to be limited to the fast diffusing transition metals, Ni and Cu (the diffusivities of Cu and Ni in Si is 2 orders of magnitude greater than Fe, Mn and

Ti). Nevertheless, it is an attractive approach for the synthesis of transition metal silicide and germanide NWs in high yield at relatively low temperatures.115

1.2.2.3 Transition metal delivery to Silicon/Germanium growth substrates

Another approach that has been used to synthesise transition metal silicide and germanide NWs is the delivery of the transition metal to a Si/Ge growth substrate.

In this approach, a transition metal vapour can be used as the transition metal source for NW growth. Chueh and co-workers first utilised this approach to synthesise TaSi2 NWs, by annealing silicon substrates with FeSi2 or NiSi thin film coatings in a Ta ambient.116,117 Jin and co-workers were then able to expand on this approach to synthesise manganese silicide NWs. In this report, Mn powder was placed into an alumina boat along with a suspended piece of Si wafer. As the reaction temperature was increased, a Mn vapour formed which reacted with the

Si wafer to form manganese silicide NWs between 5-50 μm long with diameters ranging from 50 nm to 500 nm.118,119 By progressively increasing the reaction temperature, the composition of the NWs could be changed from MnSi to Mn5Si3 to Mn3Si. Despite these initial reports, the use of metal vapours for transition metal silicide and germanide NW synthesis has not attracted much attention. This is likely due to the very high temperatures (> 900 °C) and long reaction times (16 h) needed.

Perhaps one of the most successful methods of metal delivery to a Si wafer for transition metal silicide NW growth is through the use of anhydrous transition metal halides as the transition metal source.1,20 In this approach the halide precursor is typically vaporised in a tube furnace and is carried downstream by a carrier gas such as Ar or N2, to the Si wafer, where it subsequently reacts to form the transition metal silicide or germanide NWs, as shown schematically in

Figure 1.17. This approach has been adopted by a number of groups to a high degree of success.

Figure 1.17. Experimental setup used for transition metal halide based synthesis of transition metal silicide nanowires.120

Ouyang and co-workers first utilised this approach to synthesise FeSi NWs on Si

121 wafer using FeCl3 powder, as the transition metal halide. In this report, a vapour of FeCl3 or Fe2Cl6 was produced between 180 – 260 °C which then reacted with a

Si wafer downstream at 1,100 °C to form FeSi NWs, which were tens of microns in length with diameters between 5 – 100 nm (Figure 1.18). Interestingly, uncontrolled NW branching was observed in this report. Following this initial report by Ouyang et al., a number of other groups used this approach to synthesise

122 123 124 125 other transition metal silicide NWs, such as TiSi2, Ti5Si3, Fe5Si3, CoSi,

126 and CrSi2. The use of transition metal halides has also shown promise in the synthesis of ternary transition metal silicide NWs such as Fe1-xCoxSi NWs and transition metal germanide NWs such as FeGe NWs.127,128

Figure 1.18: a) SEM image of FeSi NWs as grown on a Si substrate. Inset: SEM image of NW branching. Both scale bars are 500 nm. b) XRD pattern of as grown FeSi NWs.121

Transition metal silicide and germanide NW growth through transition metal halide delivery to a Si substrate, likely takes place through the VS growth mechanism.121 In the original report into the synthesis of FeSi NWs by Ouyang et al., no catalytic seeds were used or were present at the tips of the NWs. Also it was noted that when the vapour pressure of the halide precursor was increased, by increasing the reaction temperature, large crystals of FeSi were seen. From these observations, the authors concluded that growth likely took place through the VS mechanism. Similarly, Kim and co-workers also observed the presence of CrSi2 microcrystals at high vapour pressures in their report into the synthesis of CrSi2

NWs, leading the authors to suggest the VS growth mechanism.126

Despite the high temperatures required using this synthetic protocol, the versatility of transition metal silicide and germanide NW synthesis by halide delivery has made it one of the most popular approaches for transition metal silicide NW synthesis.

1.2.2.4 Simultaneous metal and Silicon/Germanium delivery to growth substrates

Recently, the simultaneous delivery of the transition metal and Si/Ge sources to a growth substrate to synthesise transition metal silicide and germanide NWs has attracted some interest as a means to try to control the composition and phase of

20 the NWs. Figure 1.19 shows a schematic for the synthesis of CrSi2 NWs detailing a chemical vapour transport (CVT) setup typically used to form transition metal silicide NWs by simultaneous delivery of the metal and Si to the growth substrate. In the CVT based approach, a source material and a transport

20 agent react to form gas phase precursors in situ. In the example of CrSi2 NWs,

CrSi2 powder is placed in an alumina boat in the centre of the furnace and an alumina boat containing I2 was placed upstream slightly outside the heating zone

129 of the furnace. Ni(NO3)2 was dispersed onto Si wafers, with a 2 nm surface oxide layer grown. These Si wafer pieces were then used as the NW growth substrates downstream.

Figure 1.19. Schematic of CrSi2 NW synthesis using the chemical vapour transport approach.129

As the temperature is ramped to the reaction temperature (900 °C), the I2 sublimes and is carried downstream by the Ar flow to the CrSi2. At this high temperature, the I2 reacts with the CrSi2 according to the following reversible reaction:

CrSi2 (s) + 5I2 (g) ↔ CrI2 (g) + 2SiI4 (g)

The resulting gas-phase chromium iodide (CrI2) and silicon iodide (SiI4) precursors are carried downstream to the growth substrates which are held at a lower temperature. Here, the reverse reaction takes place and CrI2 reacts with SiI4 to form CrSi2 NWs and CrSi2 microcrystals on the substrates as shown in the

SEM images in Figure 1.20. Interestingly, in the absence of Ni(NO3)2, CrSi2 microcrystals exclusively formed. The authors postulated that the trace amount of

Ni from the Ni(NO3)2 on the growth substrate provided nucleation sites for NW formation. However, Ni was never detected in the NWs. It was also noted that the thickness of the oxide layer on the Si wafer affected the morphology of CrSi2 structures grown. When a 100 nm silicon oxide layer was used, CrSi2 microcrystals formed. However, when the native oxide was removed by HF treatment, epitaxially grown CrSi2 crystals were formed.

Figure 1.20. Representative SEM images of CrSi2 NWs found on substrates at various magnifications.129

Despite these difficulties, Jin and co-workers were able to show the versatility of

130,131 this approach, extending its use to the synthesis of δ-Ni2Si and β-Ni3Si NWs where NW growth likely takes place through the VS growth mechanism. In the case of the β-Ni3Si NWs, TiSi2 powder was used as the source material. It appears that the Ni was supplied from the Ni(NO3)2 with the Si being supplied from the

SiI4 that was generated by the reaction of I2 with the TiSi2 powder.

Kim and co-workers established a CVT process based on a modified halide delivery approach for the synthesis of transition metal silicide and germanide

NWs. Figure 1.21 shows the setup used for the growth of CoSi, Co2Si and Co3Si

NWs.132

132 Figure 1.21. CVT setup used for the synthesis of Co2Si and Co3Si NWs.

In this approach, CoCl2 is placed in an alumina boat upstream of the growth zone.

In the growth zone, two pieces of sapphire are placed on a Si wafer. A CoCl2 vapour is formed by heating the CoCl2 powder. This vapour is carried downstream and reacts with the Si to form CoSi NWs on the wafer through the following reaction:

2CoCl2 (g) + 3Si (s) ↔ 2CoSi (s) + SiCl4 (g)

The SiCl4 produced from this reaction can react with the CoCl2 vapour to grow

NWs on the sapphire substrates through the following reactions:

6CoCl2 (g) + 3SiCl4 (g) ↔ 3Co2Si (s) + 12Cl2 (g)

6CoCl2 (g) + 2SiCl4 (g) ↔ 2Co3Si (s) + 10Cl2 (g)

The reactions are controlled by the SiCl4 concentrations, which can be adjusted by controlling the temperature of the substrates. At lower temperatures, the quantity of SiCl4 produced by the first reaction is low so the formation of Co3Si NWs on the sapphire substrate is preferred. At higher temperatures, Co2Si NW growth on the sapphire substrate dominates due to the higher concentration of SiCl4. Using a

133 134 similar methodology, Kim et al. were able to synthesise Fe5Si3, Fe3Ge,

134 135 136 Fe1.3Ge, CoGe, and Co5Ge7 NWs. In the case of the transition metal germanide NWs, a Ge/C powder mixture was used as the Ge source.

Alternatively, transition metal silicide NWs can be synthesised by the simultaneous delivery of the transition metal and Si to the growth substrate through the use of single source precursors (SSPs) within a CVD setup. Figure

1.22 shows a schematic representation of this approach.

Figure 1.22. CVD setup used for the synthesis of transition metal silicide NWs through the use of SSPs.20 In this setup, a Si wafer with a thin oxide coating is placed downstream. The metal carbonyl-silyl organometallic SSP is placed upstream in an alumina boat. In the synthesis of FeSi NWs (shown schematically in Figure 1.23a) the SSP, trans-

137 Fe(CO)4(SiCl3)2 was used. The SSP vaporised upstream, at a temperature of

150 °C, before being carried downstream, where it decomposed to form FeSi

NWs on the Si wafer, at a reaction temperature of 750 ° C, to form FeSi NWs between 10 - 80 nm in diameter and tens of microns in length, as seen in

Figure 1.23b-d. Jin and co-workers were able to extend this approach to the

138-140 synthesis of CoSi, Mn19Si33 and Fe1-xCoSi NWs.

Figure 1.23. a) Schematic of SSP CVD synthesis of FeSi NWs. b), c) and d) show SEM images of the FeSi NWs.

Similar to the CVT setup developed by Jin et al. discussed previously, it was found that NW formation using the SSP CVD approach was also sensitive to the

36 thickness of the oxide coating on the Si wafer. When the oxide coating was too thick (~100 nm), no NW growth occurred and a FeSi thin film was formed on bare Si wafer. Despite not offering an explanation for the dependence of NW growth on the thickness of the oxide, the authors were able to use this phenomenon to pattern wafers to grow FeSi NWs in selected areas.141

TiSi2 NWs have also been synthesised through a CVD setup. However, here a

142 dual source setup was used instead of a SSP. TiCl4 and SiH4 were co-flowed with H2 gas to a Si substrate, where TiSi2 NWs, as well as other TiSi2 NSs, were formed. Despite the possible benefit of controlling stoichiometries with this dual source CVD setup, this approach has not yet been extended to other NW systems.20

1.3 Potential Applications of Transition Metal Silicides and

Germanides

In this section, an overview of some of the potential applications of transition metal silicide and germanide NWs is provided.

1.3.1 Nanoelectronics

In the area of nanoelectronics, transition metal silicide and germanide NWs are envisioned to play an important role as contacts and as interconnects.21,22 This is due to the low electrical resistivities displayed by many transition metal silicide and germanide NWs.82,97,100 In addition to the low resistivities, the failure current densities (>107 A/cm2) reported for transition metal silicide and germanide NWs are comparable to the failure current densities seen for lithographically defined nanoscale metal lines.48 However, the implementation of free standing transition metal silicide or germanide NWs into a device architecture is challenging.143

Recently, many groups have used the silicidation approach, discussed in section

1.2.2.1, to fabricate single NW transistor devices using axial heterostructure NWs composed of silicide/silicon/silicide segments.21,88 This approach to device fabrication is quite attractive as the length of the Si segments is controlled by the silicide formation and not restricted by a lithographical process.91 Figure 1.24 shows a schematic of a single NW field-effect transistor composed of

144 NiSix/Si/NiSix segments. While the authors were able to form an ultrashort Si channel transistor, of approximately 17 nm in length, using this approach, the performance of this device was somewhat limited by the presence of Schottky

144 barriers between the Si and NiSix segments.

Figure 1.24. a) Schematic of a single NiSi2/Si/NiSi2 NW device with TEM images of the device shown in b) and c).144

1.3.2 Nanoscale Field Emitters

Another potential application for transition metal silicide and germanide NWs is as nanoscale field emitters. In the field emission phenomenon, electrons are emitted from the surface of a NW under the action of an electric field. The relationship between the emission current and the applied electric field can be described by the Fowler-Nordheim equation:

3 퐽 퐴훽2 퐵휑2 ln ( ) − ln⁡( ) = − 퐸2 휑 훽퐸

Where J is the field emission current density, E is the applied electric field strength, β is the field enhancement factor, φ is the work function of the emitter and A and B are constants. Devices based on the field emission phenomenon are useful in a wide range of applications such as flat panel displays.

Due to their 1-dimensional structure, sharp tips and high surface to volume ratio,

NWs have attracted a great deal of research interest as potential field emitters. In particular, transition metal silicides and germanides NWs have been well studied as nanoscale field emitters.22,103 Many of these studies report low turn-on fields

2 103 (for a current density of 10 µA/cm ) between 3 – 4 V/µm. In the case of Co5Ge7

NWs grown on graphite substrates, the turn-on field was as low as 1.6 V/µm.136

Similar results were seen for Cu3Si NWs grown directly on Cu foil, with a turn-on field as low as 1.16 V/µm.113

1.3.3 Lithium-ion Batteries

The use of transition metal silicide and germanide NW arrays for lithium-ion battery (LIB) applications has also been investigated. Elemental Si and Ge are attractive alternatives to the conventional graphitic C anode in LIBs, due to their high theoretical capacities. Si has a theoretical capacity of 3579 mA h/g, which is almost 10 times greater than the theoretical capacity of C (372 mA h/g).145

However, when bulk Si anodes are used, a huge volume expansion (~ 400%) associated with alloying with Li causes pulverisation of the active material and detachment from the current collector, which results in irreversible capacity loss.

Numerous ways have been investigated to circumvent the detrimental effects of this volume expansion. An attractive strategy involves modifying the surface roughness of the current collector to promote the adhesion of the active material.

Recently, the use of transition metal silicide NW arrays, directly grown from their corresponding pure transition metal foil, as high surface area current collectors have been shown to help the adhesion of the active material to the current collector. In one report, Li et al. deposited an amorphous Si (a-Si) coating onto

146 Ni3Si2 NW arrays as shown in Figure 1.25. Galvanostatic cycling of these coated NW arrays is shown in Figure 1.26.146 In comparison to a-Si thin film coatings on planar current collectors, the Ni3Si2 NW arrays showed good capacity retention, of roughly 70% after 50 cycles. Pribat et al. used a similar approach to synthesize alumina coated NiSix/a-Si NWs that displayed a stable capacity of

147 1200 mA h/g for over 1200 cycles. NiCo2O4 coatings on NiSix NW arrays have also been used for LIB applications.148

Figure 1.25. a) and b) show SEM images of a-Si coatings on Ni3Si2 NW arrays with TEM and DF-STEM images of coated NWs shown in c) and d).146

Figure 1.26. Galvanostatic cycling of a-Si coatings on different current collectors.146

1.4 References

(1) Yoon, H.-N.; Yoo, Y.-D.; Seo, K.-Y.; In, J.-H.; Kim, B.-S. Synthesis and

Applications of Noble Metal and Metal Silicide and Germanide 1-Dimensional

Nanostructures. Bull. Korean Chem. Soc. 2012, 33, 2830.

(2) Murarka, S. Transition metal silicides. Annual Review of Materials

Science 1983, 13, 117.

(3) Zhang, S.-L.; Smith, U. Self-aligned silicides for Ohmic contacts in complementary metal–oxide–semiconductor technology: TiSi2, CoSi2, and NiSi.

Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 2004,

22, 1361.

(4) Maszara, W. Fully silicided metal gates for high-performance CMOS technology: a review. Journal of the Electrochemical Society 2005, 152, G550.

(5) Chen, L. J.; Engineers, I. o. E. Silicide Technology for Integrated Circuits;

Institution of Engineering and Technology, 2004.

(6) Huang, Q.; Tan, A.; Tan, J.; Goh, I.; Dong, Z.; Ong, C.; Osipowicz, T.

Effect of exposure to air on sheet resistance and phase composition of Co silicides annealed at a low temperature of 470° C. Applied Physics A: Materials Science &

Processing 2003, 76, 439.

(7) Eizenberg, M.; Tu, K. Formation and Schottky behavior of manganese silicides on n‐type silicon. journal of Applied Physics 1982, 53, 6885.

(8) Borisenko, V.; Berlin: 2000.

(9) Senthilarasu, S.; Sathyamoorthy, R.; Lalitha, S. Synthesis and characterization of β-FeSi2 grown by thermal annealing of Fe/Si bilayers for photovoltaic applications. Solar Energy Materials and Solar Cells 2004, 82, 299.

(10) Reader, A.; Van Ommen, A.; Weijs, P.; Wolters, R.; Oostra, D. Transition metal silicides in silicon technology. Reports on Progress in Physics 1993, 56,

1397.

(11) Chen, L. Metal silicides: An integral part of microelectronics. Synthesis

2004.

(12) Taur, Y.; Sun, J.-C.; Moy, D.; Wang, L.; Davari, B.; Klepner, S. P.; Ting,

C.-Y. Source—Drain contact resistance in CMOS with self-aligned TiSi 2. IEEE

Transactions on Electron Devices 1987, 34, 575.

(13) Lauwers, A.; Schreutelkamp, R.; Brijs, B.; Bender, H.; Maex, K.

Technological aspects of epitaxial CoSi2 layers for CMOS. Applied surface science 1993, 73, 19.

(14) Iwai, H.; Ohguro, T.; Ohmi, S.-i. NiSi salicide technology for scaled

CMOS. Microelectronic Engineering 2002, 60, 157.

(15) Alivisatos, A. P. Semiconductor Clusters, Nanocrystals, and Quantum

Dots. Science 1996, 271, 933.

(16) Yang, P. The chemistry and physics of semiconductor nanowires. MRS bulletin 2005, 30, 85.

(17) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim,

F.; Yan, H. One‐dimensional nanostructures: synthesis, characterization, and applications. Advanced materials 2003, 15, 353.

(18) Barrelet, C. J.; Wu, Y.; Bell, D. C.; Lieber, C. M. Synthesis of CdS and

ZnS nanowires using single-source molecular precursors. Journal of the American

Chemical Society 2003, 125, 11498.

(19) Persson, A. I.; Larsson, M. W.; Stenström, S.; Ohlsson, B. J.; Samuelson,

L.; Wallenberg, L. R. Solid-phase diffusion mechanism for GaAs nanowire growth. Nature materials 2004, 3, 677.

(20) Schmitt, A. L.; Higgins, J. M.; Szczech, J. R.; Jin, S. Synthesis and applications of metal silicide nanowires. Journal of Materials Chemistry 2010, 20,

223.

(21) Lin, Y.-C.; Chen, Y.; Huang, Y. The growth and applications of silicides for nanoscale devices. Nanoscale 2012, 4, 1412.

(22) Chen, L. J.; Wu, W.-W.; Hsu, H.; Chen, S.; Chueh, Y.; Chou, L.-J.; Lu,

K.; Tu, K. Metal Silicide Nanowires. ECS Transactions 2007, 11, 3.

(23) Wagner, R.; Ellis, W. Vapor‐liquid‐solid mechanism of single crystal growth. Applied Physics Letters 1964, 4, 89.

(24) Cahoon, J. 2014; Vol. 2017.

(25) Schmidt, V.; Wittemann, J.; Gosele, U. Growth, thermodynamics, and electrical properties of silicon nanowires. Chemical reviews 2010, 110, 361.

(26) Schmidt, V.; Wittemann, J. V.; Senz, S.; Gösele, U. Silicon nanowires: a review on aspects of their growth and their electrical properties. Advanced

Materials 2009, 21, 2681.

(27) Shchetinin, A.; NEBOLS'IN, V.; Kozenkov, O.; Tatarenkov, A.; Dunaev,

A.; Novokreshchenova, E. DISTRIBUTION COEFFICIENT OF THE

INITIATING IMPURITY AND ITS INFLUENCE ON THE ELECTRICAL

RESISITIVITY OF FILAMENTARY SILICON CRYSTALS. Inorganic materials 1991, 27, 1137.

(28) Allen, J. E.; Hemesath, E. R.; Perea, D. E.; Lensch-Falk, J. L.; Li, Z.; Yin,

F.; Gass, M. H.; Wang, P.; Bleloch, A. L.; Palmer, R. E. High-resolution detection of Au catalyst atoms in Si nanowires. Nature nanotechnology 2008, 3, 168.

(29) Perea, D. E.; Allen, J. E.; May, S. J.; Wessels, B. W.; Seidman, D. N.;

Lauhon, L. J. Three-dimensional nanoscale composition mapping of semiconductor nanowires. Nano letters 2006, 6, 181.

(30) Nebol'Sin, V.; Shchetinin, A. Role of surface energy in the vapor–liquid– solid growth of silicon. Inorganic materials 2003, 39, 899.

(31) Wang, Y.; Schmidt, V.; Senz, S.; Gösele, U. Epitaxial growth of silicon nanowires using an aluminium catalyst. Nature nanotechnology 2006, 1, 186.

(32) Whang, S.-J.; Lee, S.; Chi, D.-Z.; Yang, W.-F.; Cho, B.-J.; Liew, Y.-F.;

Kwong, D.-L. B-doping of vapour–liquid–solid grown Au-catalysed and Al- catalysed Si nanowires: effects of B2H6 gas during Si nanowire growth and B- doping by a post-synthesis in situ plasma process. Nanotechnology 2007, 18,

275302.

(33) Miyamoto, Y.; Hirata, M. Role of agents in filamentary growth of amorphous silicon. Japanese Journal of Applied Physics 1976, 15, 1159.

(34) Tuan, H. Y.; Lee, D. C.; Korgel, B. A. Nanocrystal‐Mediated

Crystallization of Silicon and Germanium Nanowires in Organic Solvents: The

Role of Catalysis and Solid‐Phase Seeding. Angewandte Chemie International

Edition 2006, 45, 5184.

(35) Yao, Y.; Fan, S. Si nanowires synthesized with Cu catalyst. Materials

Letters 2007, 61, 177.

(36) Wen, C.-Y.; Reuter, M.; Tersoff, J.; Stach, E.; Ross, F. Structure, growth kinetics, and ledge flow during vapor− solid− solid growth of copper-catalyzed silicon nanowires. Nano letters 2009, 10, 514.

(37) J. Barsotti Jr, R.; Fischer, J. E.; Lee, C.; Mahmood, J.; Adu, C.; Eklund, P.

C. Imaging, structural, and chemical analysis of silicon nanowires. Applied physics letters 2002, 81, 2866.

(38) Hetzel, M.; Lugstein, A.; Zeiner, C.; Wójcik, T.; Pongratz, P.; Bertagnolli,

E. Ultra-fast vapour–liquid–solid synthesis of Si nanowires using ion-beam implanted gallium as catalyst. Nanotechnology 2011, 22, 395601.

(39) Iacopi, F.; Richard, O.; Eichhammer, Y.; Bender, H.; Vereecken, P.; De

Gendt, S.; Heyns, M. Size-dependent characteristics of indium-seeded Si nanowire growth. Electrochemical and Solid-State Letters 2008, 11, K98.

(40) Lensch-Falk, J. L.; Hemesath, E. R.; Lopez, F. J.; Lauhon, L. J. Vapor−

Solid− Solid Synthesis of Ge Nanowires from Vapor-Phase-Deposited Manganese

Germanide Seeds. Journal of the American Chemical Society 2007, 129, 10670.

(41) Barth, S.; Koleśnik, M. M.; Donegan, K.; Krstic, V.; Holmes, J. D.

Diameter-controlled solid-phase seeding of germanium nanowires: structural characterization and electrical transport properties. Chemistry of Materials 2011,

23, 3335.

(42) Tuan, H.-Y.; Lee, D. C.; Hanrath, T.; Korgel, B. A. Catalytic solid-phase seeding of silicon nanowires by nickel nanocrystals in organic solvents. Nano letters 2005, 5, 681.

(43) Baron, T.; Gordon, M.; Dhalluin, F.; Ternon, C.; Ferret, P.; Gentile, P.

Solid catalyzed PtSi growth of Si nanowires and electrical characterization. Appl.

Phys. Lett 2006, 89, 233111.

(44) Chockla, A. M.; Klavetter, K. C.; Mullins, C. B.; Korgel, B. A. Tin-seeded silicon nanowires for high capacity Li-ion batteries. Chemistry of Materials 2012,

24, 3738.

(45) Sharma, S.; Kamins, T.; Williams, R. S. Diameter control of Ti-catalyzed silicon nanowires. Journal of crystal growth 2004, 267, 613.

(46) Geaney, H.; Mullane, E.; Ryan, K. M. Solution phase synthesis of silicon and germanium nanowires. Journal of Materials Chemistry C 2013, 1, 4996.

(47) Lensch-Falk, J. L.; Hemesath, E. R.; Perea, D. E.; Lauhon, L. J.

Alternative catalysts for VSS growth of silicon and germanium nanowires.

Journal of Materials Chemistry 2009, 19, 849.

(48) Geaney, H.; Dickinson, C.; Barrett, C. A.; Ryan, K. M. High density germanium nanowire growth directly from copper foil by self induced solid seeding. 2011.

(49) Barth, S.; Boland, J. J.; Holmes, J. D. Defect transfer from nanoparticles to nanowires. Nano letters 2011, 11, 1550.

(50) Olesinski, R.; Abbaschian, G. The Cu− Si (Copper-Silicon) system.

Journal of Phase Equilibria 1986, 7, 170.

(51) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim,

F.; Yan, H. One-Dimensional Nanostructures: Synthesis, Characterization, and

Applications. Adv. Mater. 2003, 15, 353.

(52) Barth, S.; Hernandez-Ramirez, F.; Holmes, J. D.; Romano-Rodriguez, A.

Synthesis and applications of one-dimensional semiconductors. Progress in

Materials Science 2010, 55, 563.

(53) Rao, C.; Deepak, F.; Gundiah, G.; Govindaraj, A. Inorganic nanowires.

Progress in Solid State Chemistry 2003, 31, 5.

(54) Hsu, Y.-J.; Lu, S.-Y. Vapor− Solid Growth of Sn Nanowires: Growth

Mechanism and Superconductivity. The Journal of Physical Chemistry B 2005,

109, 4398.

(55) Dai, Z.; Pan, Z.; Wang, Z. Ultra-long single crystalline nanoribbons of tin oxide. Solid State Communications 2001, 118, 351.

(56) Lyu, S. C.; Zhang, Y.; Lee, C. J.; Ruh, H.; Lee, H. J. Low-Temperature

Growth of ZnO Nanowire Array by a Simple Physical Vapor-Deposition Method.

Chemistry of Materials 2003, 15, 3294.

(57) Shen, G.; Chen, D.; Chen, P.-C.; Zhou, C. Vapor−Solid Growth of One-

Dimensional Layer-Structured Gallium Sulfide Nanostructures. ACS Nano 2009,

3, 1115.

(58) Lei, W.; Liu, D.; Zhang, J.; Zhu, P.; Cui, Q.; Zou, G. Direct Synthesis,

Growth Mechanism, and Optical Properties of 3D AlN Nanostructures with

Urchin Shapes. Crystal Growth & Design 2009, 9, 1489.

(59) Schmidt, V.; Senz, S.; Gösele, U. Diameter-dependent growth direction of epitaxial silicon nanowires. Nano letters 2005, 5, 931.

(60) Garnett, E. C.; Liang, W.; Yang, P. Growth and Electrical Characteristics of Platinum‐Nanoparticle‐Catalyzed Silicon Nanowires. Advanced Materials

2007, 19, 2946.

(61) Adhikari, H.; Marshall, A. F.; Chidsey, C. E.; McIntyre, P. C. Germanium nanowire epitaxy: Shape and orientation control. Nano letters 2006, 6, 318.

(62) Wang, D.; Qian, F.; Yang, C.; Zhong, Z.; Lieber, C. M. Rational growth of branched and hyperbranched nanowire structures. Nano Letters 2004, 4, 871.

(63) Lieber, C. M. In Nanoelectronics Conference (INEC), 2010 3rd

International; IEEE: 2010, p 5.

(64) Xiang, J.; Lu, W.; Hu, Y.; Wu, Y.; Yan, H.; Lieber, C. M. Ge/Si nanowire heterostructures as high-performance field-effect transistors. Nature 2006, 441,

489.

(65) Schmid, H.; Björk, M. T.; Knoch, J.; Karg, S.; Riel, H.; Riess, W. Doping limits of grown in situ doped silicon nanowires using phosphine. Nano letters

2008, 9, 173.

(66) Lew, K.-K.; Pan, L.; Bogart, T. E.; Dilts, S. M.; Dickey, E. C.; Redwing, J.

M.; Wang, Y.; Cabassi, M.; Mayer, T. S.; Novak, S. W. Structural and electrical properties of trimethylboron-doped silicon nanowires. Applied Physics Letters

2004, 85, 3101.

(67) Holmes, J. D.; Johnston, K. P.; Doty, R. C.; Korgel, B. A. Control of thickness and orientation of solution-grown silicon nanowires. Science 2000, 287,

1471.

(68) Hanrath, T.; Korgel, B. A. Supercritical fluid–liquid–solid (SFLS) synthesis of Si and Ge nanowires seeded by colloidal metal nanocrystals.

Advanced Materials 2003, 15, 437.

(69) Hanrath, T.; Korgel, B. A. Nucleation and growth of germanium nanowires seeded by organic monolayer-coated gold nanocrystals. Journal of the

American Chemical Society 2002, 124, 1424.

(70) Hanrath, T.; Korgel, B. A. Chemical surface passivation of Ge nanowires.

Journal of the American Chemical Society 2004, 126, 15466.

(71) Shah, P. S.; Hanrath, T.; Johnston, K. P.; Korgel, B. A.; ACS Publications:

2004.

(72) Tuan, H.-Y.; Lee, D. C.; Hanrath, T.; Korgel, B. A. Germanium nanowire synthesis: An example of solid-phase seeded growth with nickel nanocrystals.

Chemistry of materials 2005, 17, 5705.

(73) Yuan, F.-W.; Yang, H.-J.; Tuan, H.-Y. Seeded silicon nanowire growth catalyzed by commercially available bulk metals: broad selection of metal catalysts, superior field emission performance, and versatile nanowire/metal architectures. Journal of Materials Chemistry 2011, 21, 13793.

(74) Heitsch, A. T.; Fanfair, D. D.; Tuan, H.-Y.; Korgel, B. A. solution− liquid− solid (SLS) growth of silicon nanowires. Journal of the American

Chemical Society 2008, 130, 5436.

(75) Zaitseva, N.; Dai, Z. R.; Grant, C. D.; Harper, J.; Saw, C. Germanium nanocrystals synthesized in high-boiling-point organic solvents. Chemistry of

Materials 2007, 19, 5174.

(76) Lu, X.; Fanfair, D. D.; Johnston, K. P.; Korgel, B. A. High yield solution− liquid− solid synthesis of germanium nanowires. Journal of the American

Chemical Society 2005, 127, 15718.

(77) Barrett, C. A.; Geaney, H.; Gunning, R. D.; Laffir, F. R.; Ryan, K. M.

Perpendicular growth of catalyst-free germanium nanowire arrays. Chemical

Communications 2011, 47, 3843.

(78) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High density growth of indium seeded silicon nanowires in the vapor phase of a high boiling point solvent. Chemistry of Materials 2012, 24,

2204.

(79) Mullane, E.; Kennedy, T.; Geaney, H.; Dickinson, C.; Ryan, K. M.

Synthesis of tin catalyzed silicon and germanium nanowires in a solvent–vapor

50 system and optimization of the seed/nanowire interface for dual lithium cycling.

Chemistry of Materials 2013, 25, 1816.

(80) Mullane, E.; Geaney, H.; Ryan, K. Synthesis of silicon–germanium axial nanowire heterostructures in a solvent vapor growth system using indium and tin catalysts. Physical Chemistry Chemical Physics 2015, 17, 6919.

(81) Geaney, H.; Mullane, E.; Ramasse, Q. M.; Ryan, K. M. Atomically abrupt silicon–germanium axial heterostructure nanowires synthesized in a solvent vapor growth system. Nano letters 2013, 13, 1675.

(82) Wu, Y.; Xiang, J.; Yang, C.; Lu, W.; Lieber, C. M. Single-crystal metallic nanowires and metal/semiconductor nanowire heterostructures. Nature 2004, 430,

61.

(83) Lu, K.-C.; Wu, W.-W.; Wu, H.-W.; Tanner, C. M.; Chang, J. P.; Chen, L.

J.; Tu, K. In situ control of atomic-scale Si layer with huge strain in the nanoheterostructure NiSi/Si/NiSi through point contact reaction. Nano letters

2007, 7, 2389.

(84) Weber, W. M.; Geelhaar, L.; Graham, A. P.; Unger, E.; Duesberg, G. S.;

Liebau, M.; Pamler, W.; Chèze, C.; Riechert, H.; Lugli, P. Silicon-nanowire transistors with intruded nickel-silicide contacts. Nano letters 2006, 6, 2660.

(85) Lin, Y.-C.; Chen, Y.; Shailos, A.; Huang, Y. Detection of spin polarized carrier in silicon nanowire with single crystal MnSi as magnetic contacts. Nano letters 2010, 10, 2281.

(86) Tang, J.; Wang, C.-Y.; Chang, L.-T.; Fan, Y.; Nie, T.; Chan, M.; Jiang,

W.; Chen, Y.-T.; Yang, H.-J.; Tuan, H.-Y. Electrical spin injection and detection in Mn5Ge3/Ge/Mn5Ge3 nanowire transistors. Nano letters 2013, 13, 4036.

(87) Tang, J.; Wang, C.-Y.; Hung, M.-H.; Jiang, X.; Chang, L.-T.; He, L.; Liu,

P.-H.; Yang, H.-J.; Tuan, H.-Y.; Chen, L.-J. Ferromagnetic germanide in Ge nanowire transistors for spintronics application. ACS nano 2012, 6, 5710.

(88) Lin, Y.-C.; Lu, K.-C.; Wu, W.-W.; Bai, J.; Chen, L. J.; Tu, K.; Huang, Y.

Single crystalline PtSi nanowires, PtSi/Si/PtSi nanowire heterostructures, and nanodevices. Nano letters 2008, 8, 913.

(89) Liu, B.; Wang, Y.; Dilts, S.; Mayer, T. S.; Mohney, S. E. Silicidation of silicon nanowires by platinum. Nano letters 2007, 7, 818.

(90) Yamamoto, K.; Kohno, H.; Takeda, S.; Ichikawa, S. Fabrication of iron silicide nanowires from nanowire templates. Applied physics letters 2006, 89,

083107.

(91) Burchhart, T.; Lugstein, A.; Hyun, Y.; Hochleitner, G.; Bertagnolli, E.

Atomic scale alignment of copper-germanide contacts for Ge nanowire metal oxide field effect transistors. Nano letters 2009, 9, 3739.

(92) Chiu, C.-H.; Huang, C.-W.; Chen, J.-Y.; Huang, Y.-T.; Hu, J.-C.; Chen,

L.-T.; Hsin, C.-L.; Wu, W.-W. Copper silicide/silicon nanowire heterostructures: in situ TEM observation of growth behaviors and electron transport properties.

Nanoscale 2013, 5, 5086.

(93) Chou, Y.-C.; Wu, W.-W.; Cheng, S.-L.; Yoo, B.-Y.; Myung, N.; Chen, L.

J.; Tu, K. In-situ TEM observation of repeating events of nucleation in epitaxial growth of nano CoSi2 in nanowires of Si. Nano Letters 2008, 8, 2194.

(94) Yaish, Y.; Katsman, A.; Cohen, G.; Beregovsky, M. Kinetics of nickel silicide growth in silicon nanowires: From linear to square root growth. Journal of

Applied Physics 2011, 109, 094303.

(95) Lin, Y.-C.; Chen, Y.; Xu, D.; Huang, Y. Growth of nickel silicides in Si and Si/SiOx core/shell nanowires. Nano letters 2010, 10, 4721.

(96) Lu, K.-C.; Wu, W.-W.; Ouyang, H.; Lin, Y.-C.; Huang, Y.; Wang, C.-W.;

Wu, Z.-W.; Huang, C.-W.; Chen, L. J.; Tu, K.-N. The influence of surface oxide on the growth of metal/semiconductor nanowires. Nano letters 2011, 11, 2753.

(97) Yan, C.; Higgins, J. M.; Faber, M. S.; Lee, P. S.; Jin, S. Spontaneous growth and phase transformation of highly conductive nickel germanide nanowires. ACS nano 2011, 5, 5006.

(98) Decker, C.; Solanki, R.; Freeouf, J.; Carruthers, J.; Evans, D. Directed growth of nickel silicide nanowires. Applied physics letters 2004, 84, 1389.

(99) Yan, X.; Yuan, H.; Wang, J.; Liu, D.; Zhou, Z.; Gao, Y.; Song, L.; Liu, L.;

Zhou, W.; Wang, G. Synthesis and characterization of a large amount of branched

Ni2Si nanowires. Applied Physics A 2004, 79, 1853.

(100) Fan, X.; Zhang, H.; Du, N.; Yang, D. Phase-controlled synthesis of nickel silicide nanostructures. Materials Research Bulletin 2012, 47, 3797.

(101) Lee, K. S.; Mo, Y. H.; Nahm, K. S.; Shim, H. W.; Suh, E. K.; Kim, J. R.;

Kim, J. J. Anomalous growth and characterization of carbon-coated nickel silicide nanowires. Chemical physics letters 2004, 384, 215.

(102) Kim, J.; Shin, D. H.; Lee, E.-S.; Han, C.-S.; Park, Y. C. Electrical characteristics of single and doubly connected Ni silicide nanowire grown by plasma-enhanced chemical vapor deposition. Applied Physics Letters 2007, 90,

253103.

(103) Liu, Z.; Zhang, H.; Wang, L.; Yang, D. Controlling the growth and field emission properties of silicide nanowire arrays by direct silicification of Ni foil.

Nanotechnology 2008, 19, 375602.

(104) Jiang, Y.; Li, Z.; Li, B.; Zhang, J.; Niu, C. Ni 3 Si 2 nanowires grown in situ on Ni foam for high-performance supercapacitors. Journal of Power Sources

2016, 320, 13.

(105) Kim, C. J.; Kang, K.; Woo, Y. S.; Ryu, K. G.; Moon, H.; Kim, J. M.;

Zang, D. S.; Jo, M. H. Spontaneous chemical vapor growth of NiSi nanowires and their metallic properties. Advanced Materials 2007, 19, 3637.

(106) Kang, K.; Kim, S.-K.; Kim, C.-J.; Jo, M.-H. The role of NiO x overlayers on spontaneous growth of NiSi x nanowires from Ni seed layers. Nano letters

2008, 8, 431.

(107) Dasgupta, N. P.; Xu, S.; Jung, H. J.; Iancu, A.; Fasching, R.; Sinclair, R.;

Prinz, F. B. Nickel Silicide Nanowire Arrays for Anti‐Reflective Electrodes in

Photovoltaics. Advanced Functional Materials 2012, 22, 3650.

(108) Lee, C.-Y.; Lu, M.-P.; Liao, K.-F.; Wu, W.-W.; Chen, L.-J. Vertically well-aligned epitaxial Ni 31 Si 12 nanowire arrays with excellent field emission properties. Applied physics letters 2008, 93, 113109.

(109) Lee, C.-Y.; Lu, M.-P.; Liao, K.-F.; Lee, W.-F.; Huang, C.-T.; Chen, S.-Y.;

Chen, L.-J. Free-standing single-crystal NiSi2 nanowires with excellent electrical transport and field emission properties. The Journal of Physical Chemistry C

2009, 113, 2286.

(110) Kim, J.; Lee, E.-S.; Han, C.-S.; Kang, Y.; Kim, D.; Anderson, W. A.

Observation of Ni silicide formations and field emission properties of Ni silicide nanowires. Microelectronic Engineering 2008, 85, 1709.

(111) Kim, J.; Anderson, W. A. Spontaneous nickel monosilicide nanowire formation by metal induced growth. Thin Solid Films 2005, 483, 60.

(112) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chemistry of Materials 2012, 24, 4319.

(113) Yuan, F.-W.; Wang, C.-Y.; Li, G.-A.; Chang, S.-H.; Chu, L.-W.; Chen, L.-

J.; Tuan, H.-Y. Solution-phase synthesis of single-crystal Cu 3 Si nanowire arrays on diverse substrates with dual functions as high-performance field emitters and efficient anti-reflective layers. Nanoscale 2013, 5, 9875.

(114) Klementová, M.; Krabáč, L.; Brázda, P.; Palatinus, L.; Dřínek, V. Cu-Si nanoobjects prepared by CVD on Cu/Cu 5 Si-substrates using various precursors

(SiH 4, EtSiH 3, BuSiH 3) with added H 2 or air. Journal of Crystal Growth 2017,

465, 6.

(115) Weber, E. R. Transition metals in silicon. Applied Physics A: Materials

Science & Processing 1983, 30, 1.

(116) Chueh, Y.; Chou, L.; Cheng, S.; Chen, L.; Tsai, C.; Hsu, C.; Kung, S.

Synthesis and characterization of metallic TaSi 2 nanowires. Applied Physics

Letters 2005, 87, 223113.

(117) Chueh, Y.-L.; Ko, M.-T.; Chou, L.-J.; Chen, L.-J.; Wu, C.-S.; Chen, C.-D.

TaSi2 nanowires: a potential field emitter and interconnect. Nano letters 2006, 6,

1637.

(118) Girard, S. N.; Chen, X.; Meng, F.; Pokhrel, A.; Zhou, J.; Shi, L.; Jin, S.

Thermoelectric Properties of Undoped High Purity Higher Manganese Silicides

Grown by Chemical Vapor Transport. Chemistry of Materials 2014, 26, 5097.

(119) Pokhrel, A.; Degregorio, Z. P.; Higgins, J. M.; Girard, S. N.; Jin, S. Vapor

Phase Conversion Synthesis of Higher Manganese Silicide (MnSi1.75) Nanowire

Arrays for Thermoelectric Applications. Chemistry of Materials 2013, 25, 632.

(120) Seo, K.; Varadwaj, K. S. K.; Cha, D.; In, J.; Kim, J.; Park, J.; Kim, B.

Synthesis and Electrical Properties of Single Crystalline CrSi2 Nanowires. J.

Phys. Chem. C 2007, 111, 9072.

(121) Ouyang, L.; Thrall, E. S.; Deshmukh, M. M.; Park, H. Vapor‐Phase

Synthesis and Characterization of ϵ‐FeSi Nanowires. Advanced Materials 2006,

18, 1437.

(122) Chang, C.-M.; Chang, Y.-C.; Chung, Y.-A.; Lee, C.-Y.; Chen, L.-J.

Synthesis and properties of the low resistivity TiSi2 nanowires grown with TiF4 precursor. The Journal of Physical Chemistry C 2009, 113, 17720.

(123) Lin, H.-K.; Tzeng, Y.-F.; Wang, C.-H.; Tai, N.-H.; Lin, I.-N.; Lee, C.-Y.;

Chiu, H.-T. Ti5Si3 nanowire and its field emission property. Chemistry of

Materials 2008, 20, 2429.

(124) Varadwaj, K. S.; Seo, K.; In, J.; Mohanty, P.; Park, J.; Kim, B. Phase- controlled growth of metastable Fe5Si3 nanowires by a vapor transport method.

Journal of the American Chemical Society 2007, 129, 8594.

(125) Seo, K.; Varadwaj, K.; Mohanty, P.; Lee, S.; Jo, Y.; Jung, M.-H.; Kim, J.;

Kim, B. Magnetic properties of single-crystalline CoSi nanowires. Nano letters

2007, 7, 1240.

(126) Seo, K.; Varadwaj, K. S.; Cha, D.; In, J.; Kim, J.; Park, J.; Kim, B.

Synthesis and electrical properties of single crystalline CrSi2 nanowires. The

Journal of Physical Chemistry C 2007, 111, 9072.

(127) In, J.; Varadwaj, K. S.; Seo, K.; Lee, S.; Jo, Y.; Jung, M.-H.; Kim, J.; Kim,

B. Single-Crystalline Ferromagnetic Fe1-x Co x Si Nanowires. The Journal of

Physical Chemistry C 2008, 112, 4748.

(128) Stolt, M. J.; Li, Z.-A.; Phillips, B.; Song, D.; Mathur, N.; Dunin-

Borkowski, R. E.; Jin, S. Selective Chemical Vapor Deposition Growth of Cubic

FeGe Nanowires That Support Stabilized Magnetic Skyrmions. Nano Letters

2016.

(129) Szczech, J. R.; Schmitt, A. L.; Bierman, M. J.; Jin, S. Single-crystal semiconducting chromium disilicide nanowires synthesized via chemical vapor transport. Chemistry of materials 2007, 19, 3238.

(130) Song, Y.; Schmitt, A. L.; Jin, S. Ultralong single-crystal metallic Ni2Si nanowires with low resistivity. Nano letters 2007, 7, 965.

(131) Song, Y.; Jin, S. Synthesis and properties of single-crystal β 3-Ni 3 Si nanowires. Applied physics letters 2007, 90, 173122.

(132) Seo, K.; Lee, S.; Yoon, H.; In, J.; Varadwaj, K. S.; Jo, Y.; Jung, M.-H.;

Kim, J.; Kim, B. Composition-tuned Co n Si nanowires: location-selective simultaneous growth along temperature gradient. Acs Nano 2009, 3, 1145.

(133) Seo, K.; Lee, S.; Jo, Y.; Jung, M.-H.; Kim, J.; Churchill, D. G.; Kim, B.

Room Temperature Ferromagnetism in Single-Crystalline Fe5Si3 Nanowires. The

Journal of Physical Chemistry C 2009, 113, 6902.

(134) Yoon, H.; Kang, T.; Lee, J. M.; Kim, S.-i.; Seo, K.; Kim, J.; Park, W. I.;

Kim, B. Epitaxially Integrating Ferromagnetic Fe1. 3Ge Nanowire Arrays on

Few-Layer Graphene. The Journal of Physical Chemistry Letters 2011, 2, 956.

(135) Yoon, H.; Kim, S.-i.; Lee, S.; In, J.; Kim, J.; Ryoo, H.; Noh, J.-H.; Ahn, J.-

P.; Jo, Y.; Choo, J. Three-dimensionally kinked high-conducting CoGe nanowire growth induced by rotational twinning. Journal of Materials Chemistry C 2013, 1,

6259.

(136) Yoon, H.; Seo, K.; Bagkar, N.; In, J.; Park, J.; Kim, J.; Kim, B. Vertical epitaxial Co5Ge7 nanowire and nanobelt arrays on a thin graphitic layer for flexible field emission displays. Advanced Materials 2009, 21, 4979.

(137) Schmitt, A. L.; Bierman, M. J.; Schmeisser, D.; Himpsel, F. J.; Jin, S.

Synthesis and properties of single-crystal FeSi nanowires. Nano letters 2006, 6,

1617.

(138) Schmitt, A. L.; Zhu, L.; Schmeiβer, D.; Himpsel, F. J.; Jin, S. Metallic single-crystal CoSi nanowires via chemical vapor deposition of single-source precursor. The Journal of Physical Chemistry B 2006, 110, 18142.

(139) Higgins, J. M.; Schmitt, A. L.; Guzei, I. A.; Jin, S. Higher manganese silicide nanowires of nowotny chimney ladder phase. Journal of the American

Chemical Society 2008, 130, 16086.

(140) Schmitt, A. L.; Higgins, J. M.; Jin, S. Chemical Synthesis and

Magnetotransport of Magnetic Semiconducting Fe1− x Co x Si Alloy Nanowires.

Nano letters 2008, 8, 810.

(141) Schmitt, A. L.; Jin, S. Selective patterned growth of silicide nanowires without the use of metal catalysts. Chemistry of materials 2007, 19, 126.

(142) Zhou, S.; Liu, X.; Lin, Y.; Wang, D. Spontaneous Growth of Highly

Conductive Two‐Dimensional Single‐Crystalline TiSi2 Nanonets. Angewandte

Chemie International Edition 2008, 47, 7681.

(143) Dong, L.; Bush, J.; Chirayos, V.; Solanki, R.; Jiao, J.; Ono, Y.; Conley, J.

F.; Ulrich, B. D. Dielectrophoretically controlled fabrication of single-crystal nickel silicide nanowire interconnects. Nano Letters 2005, 5, 2112.

(144) Tang, W.; Dayeh, S. A.; Picraux, S. T.; Huang, J. Y.; Tu, K.-N. Ultrashort

Channel Silicon Nanowire Transistors with Nickel Silicide Source/Drain

Contacts. Nano Letters 2012, 12, 3979.

(145) Kennedy, T.; Brandon, M.; Ryan, K. M. Advances in the Application of

Silicon and Germanium Nanowires for High‐Performance Lithium‐Ion Batteries.

Advanced Materials 2016.

(146) Li, F.; Yue, H.; Wang, P.; Yang, Z.; Wang, D.; Liu, D.; Qiao, L.; He, D.

Synthesis of core–shell architectures of silicon coated on controllable grown Ni- silicide nanostructures and their lithium-ion battery application. CrystEngComm

2013, 15, 7298.

(147) Nguyen, H. T.; Zamfir, M. R.; Duong, L. D.; Lee, Y. H.; Bondavalli, P.;

Pribat, D. Alumina-coated silicon-based nanowire arrays for high quality Li-ion battery anodes. Journal of Materials Chemistry 2012, 22, 24618.

(148) Zhang, Q.; Chen, H.; Wang, J.; Xu, D.; Li, X.; Yang, Y.; Zhang, K.

Growth of Hierarchical 3D Mesoporous NiSix/NiCo2O4 Core/Shell

Heterostructures on Nickel Foam for Lithium‐Ion Batteries. ChemSusChem 2014,

7, 2325.

Chapter 2: Experimental Procedures and Characterisation

Techniques

2.1 General Synthetic Strategy

In this work, the solution-based Solvent Vapour Growth (SVG) system was used to synthesise transition metal silicide and transition metal germanide nanostructures (NSs). The SVG system, developed by Barrett et al., uses the vapour phase of a high boiling point solvent (HBS) as a reaction medium for the thermal decomposition of organometallic precursors.1 A great deal of success in the synthesis of Si and Ge NWs has been achieved using this approach, using a number of different catalytic seeds such as In,2 Sn3 and Cu.4 Recently, this SVG

5 system has been adapted to synthesise Cu15Si4 NWs.

Figure 2.1 shows a schematic representation of the SVG system. Typically, a growth substrate is placed into a long neck, round-bottomed flask, with the HBS, squalane. In the case of the synthesis of transition metal silicide and germanide

NWs within the SVG system, the substrates that are typically used are transition metal foils. However, thermally evaporated transition metal films onto support substrates (i.e. glass, silicon wafer, stainless steel) could also be used. The flask is connected to a Schlenk line via a water condenser and is placed within a three- zone furnace. Initially, a degas step is applied to remove any moisture from the system. This is achieved by applying a vacuum, of at least 100 mTorr, at a temperature of 125 °C for 60 min. Following this degas step, the flask is heated above the boiling point of the solvent, under an Argon atmosphere, to achieve a vapour reflux. The organometallic precursor, phenylsilane (PS) or

60 diphenylgermane (DPG), is then injected into the flask through a septum cap. PS and DPG thermally decompose through the phenyl redistribution mechanism to form silane (SiH4) and germane (GeH4) gas, respectively, which reacts with the substrate to form 1D Si and Ge based NSs.6,7 Typically higher reaction temperatures and longer reaction times are required for PS in comparison to DPG.

This is due to lower thermal stability of DPG.3

Figure 2.1. Schematic of the Solvent Vapor Growth (SVG) system

2.2 Characterisation Techniques

2.2.1 Scanning Electron Microscopy (SEM)

SEM is an important analytical technique for the characterisation of various NSs and is an indispensable tool for revealing information about the surface topography, chemical composition and crystalline structures of NSs. As the name suggests, SEM works by scanning a high energy electron beam (up to 30 kV) across the surface of the sample. The interaction of the electron beam with the sample produces a number of signals through electron-sample interactions, which can be used to produce an image of the sample when detected. The signals include secondary electrons, backscattered electrons, diffracted backscattered electrons, photons and heat. The most common SEM imaging mode is through the detection of loosely bound electrons, which are emitted due to ionization of the sample by the primary electron beam. These secondary electrons are typically emitted from the surface of the sample and are used to produce an image of the sample, providing morphological information. Figure 2.2 shows a schematic diagram outlining the different parts of the microscope, including the electron gun, an evacuated column with multiple lenses for focusing the beam and the sample chamber with varying detectors around the sample stage.

Figure 2.2. Schematic diagram of an SEM instrument.8

Typically, the electron gun generates the high energy beam of electrons by thermionic emission. This beam passes down the column and is focused by a series of lenses and apertures. At the bottom of the column, the scan coil rasters the electron beam across the sample and the signals produced are detected to generate an image of the sample. SEM images of areas as small as 200 nm in width with a resolution in the order of 1 nm can be obtained, by optimising certain parameters, such as the accelerating voltage, aperture size and the working distance. Figure 2.3 shows SEM images at various magnifications of

CdSe NW arrays, which were synthesised by the selective vaporisation of a

CdSe crystal.9

Figure 2.3. Low magnification SEM images of CdSe NW arrays are shown in a) and c), with higher magnification images of the highlighted areas shown in b) and d).9

One of the main advantages of SEM is that minimal sample preparation is required for conductive materials, thus allowing for facile imaging for NSs grown directly from a bulk substrate. In addition to the imaging capabilities, SEM can also be used to analyse the chemical composition of NSs through energy dispersive x-ray spectroscopy (EDS).

2.2.2 Transmission Electron Microscopy (TEM)

While SEM is useful for the morphological analysis of NSs, TEM is required to further examine the crystallinity of NSs. TEM is a very important instrument in nanomaterial analysis, which is capable of investigating the morphology, crystal structure (via electron diffraction) and chemical composition (through EDS) of a wide range of NSs. Figure 2.4 shows a schematic representation of a TEM.

Figure 2.4. Schematic representation of a TEM instrument.10

Similar to SEM, TEM uses an electron gun, an evacuated column and several lenses to focus the beam. However, in the TEM, the sample is placed in between the condenser and objective lenses. This is because the TEM generates an image by subjecting the sample to an electron beam and using a portion of those

65 electrons (that transmit through the sample) to produce an image on a fluorescent viewing screen at the bottom of the column. As the beam is transmitted through the sample, interactions between the beam and the sample occur. These interactions reduce the number of electrons that can be projected onto the fluorescent screen, thus reducing the brightness of the image that is produced. The number of electrons that are transmitted through the sample decreases with sample thickness, meaning that TEM samples are required to be quite thin (~200 nm) for

TEM analysis.

While these interactions do limit the ability of the TEM to image thicker samples, these electron-sample interactions are not disadvantageous, as important information about crystal structure and chemical composition can be ascertained from the signals produced. In modern TEMs, a number of analytical techniques, such as electron diffraction, scanning transmission electron microscopy (STEM) and energy dispersive x-ray spectroscopy (EDS), can be carried out, which allow for an in-depth characterisation of NSs.

2.2.2.1 Electron Diffraction

Electron diffraction is an important technique used to analyse the crystal structure of materials. As the electron beam is transmitted through the sample, a portion of the electrons are scattered. These scattered electrons interfere and generate a diffraction pattern, due to wave-particle duality. By measuring the distance from the centre spot to the other various spots, a fully indexed pattern of the crystal structure can be determined. The generated pattern is in reciprocal space, so a large lattice will have a correspondingly small distance in the diffraction pattern.

The angles between spots in the diffraction pattern correspond to angles between various crystal planes for the material, thus allowing for complete crystallographic determination. Fast Fourier Transforms (FFTs) are another approach for gaining information about the crystallinity of a sample and can be indexed in much the same way as diffraction patterns.

Figure 2.5a shows a TEM image of a ZnO NW, with the corresponding electron diffraction pattern shown in Figure 2.5b.11 The TEM image and corresponding diffraction pattern were taken on a zone axis, which means that the beam direction is parallel to a certain crystallographic direction. By tilting the sample to different zone axes, different diffraction patterns can be produced.

Figure 2.5. a) TEM image of ZnO NW with corresponding electron diffraction pattern shown in b).11

2.2.2.2 Scanning Transmission Electron Microscopy (STEM)

In addition to TEM, modern TEM suites often come with scanning transmission electron microscopy (STEM) capabilities. STEM works on the same principle as

SEM, in that the electron beam is focused to a narrow spot and is raster-scanned across the sample, which produces signals that are collected to generate a digital image of the sample. However, unlike SEM, a thin sample is required so that transmission of the electrons through the sample is possible.

One of the main advantages of STEM is that multiple detectors can be used simultaneously to produce a complete image of the sample. The typical detectors used in STEM are the bright field (BF) and dark field (DF) detectors. The BF detector includes the transmitted beam, which means that images are produced by electrons which pass straight through the sample or have only been scattered at low angles, and this leads to holes in the sample appearing bright. On the contrary, the DF detector excludes the transmitted beam, which results in the holes appearing dark in the resultant image. In high angle annular dark field

(HAADF) imaging, the images are formed by very high angle, incoherently scattered electrons, which results in the high sensitivity to variations in the atomic number of the sample (Z-contrast). This is particularly evident in Figure 2.6, where a Ge-Si NW heterostructure is imaged in both BF-STEM mode (Figure

2.6a) and HAADF-STEM mode (Figure 2.6b).12

In conjunction with BF-STEM and DF-STEM, chemical analysis can also be conducted through electron energy loss spectroscopy (EELS). In EELS, the change in the kinetic energy of the electrons is measured, as the beam transmits

68 through the sample. As the energy lost by the electrons can be attributed to the ionisation of the inner shells of atoms, EELS can be used to identify the elements present in a sample. These signals can be collected from multiple points to produce elemental maps of the sample, as shown for a Ge-Si NW heterostructure in Figure 2.6c and Figure 2.6d, respectively.

Figure 2.6. a) BF-STEM image of Ge-Si NW heterostructure. b) HAADF-STEM image of highlighted area in a). c) HAADF-STEM image of area used for EELS spectrum analysis, with EELS maps shown in d). e) Graph generated of the normalized intensity across the Ge–Si interface, with the direction given by the yellow arrow in (c).12

2.2.3 Energy Dispersive X-Ray Spectroscopy (EDS)

EDS is an X-ray technique that can be used to identify the chemical composition of materials, by analysing the characteristic X-rays emitted by the sample after its interaction with the electron beam. When high energy electrons are focused on a sample, an electron from the inner shell of an atom of the sample can be excited to a higher energy level or it can ejected from the atom, leaving behind a vacancy.

This vacancy is then filled by electrons from a higher state, and an x-ray is emitted to balance the energy difference between the two electrons states.

The wavelength of this X-ray is characteristic to the atoms in the sample. By measuring the wavelength of the emitted X-rays, it is possible to determine the chemical composition of a sample. In TEM and SEM modes, a spectrum can be collected from an area to determine the elements present in the resultant NSs. In

STEM, line scans and EDS maps can be used to investigate the chemical composition at any point in the sample. In Figure 2.7a, a DF-STEM image of a Ge

NW with Si NW branches is shown.13 The corresponding EDS maps for Ge and

Si overlaid in Figure 2.7b clearly show the segregation between the Ge NW and the Si NW branches.

Figure 2.7. a) DF-STEM image of Ge NW with Si NW branches. b) Corresponding EDS elemental maps for Si and Ge.13

2.2.4 X-Ray Diffraction (XRD)

XRD is an analytical technique used to identify and determine information about a crystalline sample. When a beam of X-rays is scanned through a range of angles across a crystalline sample, a diffraction pattern is produced. This diffraction pattern is due to X-rays, having wavelengths which are similar to the distance between the crystal planes in the crystalline sample, and the constructive interference between the X-rays scattered by atoms in these planes. For constructive interference, experimental conditions must satisfy Bragg’s law, which relates the wavelength of the X-ray source (λ) to the angle of incidence (θ) and the interplanar spacing of the crystal (d), where (n) is the order of reflection and is a whole number for constructive interference:

풏흀 = ퟐ풅 퐬퐢퐧 휽

As different crystals have different interplanar spacings, by scanning through a range of angles, a diffraction pattern which is unique for that crystal phase can be obtained. Figure 2.8 shows a typical setup for XRD, which consists of three basic elements: an X-ray tube, a sample holder and a detector.14 The X-rays are generated in the cathode ray tube by heating a filament to produce electrons.

Typically, a Cu source is used to produce X-rays, which are subsequently filtered to make a monochromatic source with a wavelength of 0.154 nm.

Figure 2.8. Schematic illustration the typical arrangement of the three components of an XRD.14

These X-rays are then directed onto the sample, which is placed on the sample stage. The intensities of the diffracted rays are measured using the detector, which moves around the sample recording the number of diffracted rays over a series of angles (2θ). Figure 2.9 shows a typical XRD pattern for Au seeded Si NWs, in which the different peaks can either be assigned to Au or Si.15 XRD reference patterns used in this thesis are provided in Appendix A.

Figure 2.9. XRD pattern for Au seeded Si NWs.15

2.2.5 Electrochemical Characterisation

2.2.5.1 Galvanostatic Charge and Discharge Cycling

Galvanostatic charge/discharge cycling is an electrochemical technique, where a constant current is applied to an electrochemical system while the potential is monitored. The technique is often also referred to as chronopotentiometry. In Li- ion battery research, galvanostatic experiments are particularly useful and are performed to determine certain electrochemical properties of an experimental electrode material, such as specific capacity and volumetric capacity. The cycle life of an electrode material can also be determined through continuous galvanostatic charging and discharging while monitoring the capacity fade.

In a typical galvanostatic cycling experiment of a Li-alloying electrode such as Si, a Li-ion half-cell (consisting of a Si working electrode and a Li counter and reference electrode) is connected to a galvanostat and a constant negative current is applied. The cathodic current causes Li+ ions to shuttle across from the counter electrode to the working electrode, where Li alloys with Si to form LixSi. This lithiation (charging) process causes the potential of the cell to drop and when it reaches a lower cut-off point (usually 0.01V for Si) a positive current is then applied. This causes delithiation (discharging) of the LixSi alloy to form amorphous Si as the Li+ ions shuttle back across to the counter electrode. A cycle is complete once the potential reaches the upper cut-off point (usually between 1.0

– 1.5V for Si electrodes). The entire process is then repeated for a given number of cycles. The potential is monitored during the experiment and voltage profiles are plotted versus either time or specific capacity on the x-axis, as in Figure

2.10a.16 The potential ranges where lithiation and delithiation occur during

73 cycling can be identified through analysis of these voltage profiles. Long sloping plateaus in the graphs are indicative of these potential ranges, as they are due to the co-existence of two phases within the material. Differential capacity plots illustrate these two-phase potential ranges more obviously, with peaks corresponding to plateaus in the voltage profile (Figure 2.10b).16

The specific capacity of material is an important property, as it represents the specific energy of the electrode per mass of the active material (mAh/g). As is evident from its unit, the specific capacity is derived the current applied, the time taken per charge (or discharge) and the mass of the active material. Capacity is usually presented per cycle, as in Figure 2.10c.16 The rate capability of an electrode can also be determined through galvanostatic cycling experiments. In a standard rate capability experiment, the applied charge and discharge current is increased every few cycles while monitoring the change in the specific capacity, as in Figure 2.10d.16 In the graph, the coulombic efficiency (C.E.) of each cycle is also presented, which is defined as the ratio of the discharge capacity to the charge capacity. A high C.E. is desirable, as it indicates that no deleterious side reactions are occurring during cycling.

Figure 2.10. (a) Voltage proﬁles of Ge-NW at a 0.5 C rate, (b) Differential capacity plots of the 1st and 2nd cycles of a Sn seeded Ge NW electrode. The peaks indicate the potential range where lithiation and delithiation of the Sn seed and Ge NW occurs, (c) Cycle performance of charge capacity of Ge-NWs at a rate of 0.5 C between 0 and 1.5 V. d) Rate capability and coulombic efficiency of a Ge NW electrode16

2.2.5.2 Calculation of C-rate

The charge and discharge current used in galvanostatic experiments is often expressed as a C-rate, where 1C represents the necessary current required to fully charge or discharge the active material in one hour. It is calculated using the following equation:

(Mass of Active Material)(Theoretical Specific Capacity) 1C rate = 1 hour

In general in the field of Li-ion battery research, galvanostatic cycling tests are performed at either divisors or multiples of a 1C rate.

E.g: 5C = (5) x (1C rate); C/10 = (0.1) x (1C rate) etc.

2.3 References

(1) Barrett, C. A.; Geaney, H.; Gunning, R. D.; Laffir, F. R.; Ryan, K. M.

Perpendicular growth of catalyst-free germanium nanowire arrays. Chemical communications 2011, 47, 3843.

(2) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High Density Growth of Indium seeded Silicon Nanowires in the

Vapor phase of a High Boiling Point Solvent. Chem. Mater. 2012, 24, 2204.

(3) Mullane, E.; Kennedy, T.; Geaney, H.; Dickinson, C.; Ryan, K. M.

Synthesis of Tin Catalyzed Silicon and Germanium Nanowires in a Solvent–

Vapor System and Optimization of the Seed/Nanowire Interface for Dual Lithium

Cycling. Chem. Mater. 2013, 25, 1816.

(4) Geaney, H.; Dickinson, C.; Barrett, C. A.; Ryan, K. M. High Density

Germanium Nanowire Growth Directly from Copper Foil by Self-Induced Solid

Seeding. Chemistry of Materials 2011, 23, 4838.

(5) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chem. Mater. 2012, 24, 4319.

(6) Lu, X.; Harris, J. T.; Villarreal, J. n. E.; Chockla, A. M.; Korgel, B. A.

Enhanced nickel-seeded synthesis of germanium nanowires. Chemistry of

Materials 2013, 25, 2172.

(7) Lee, D. C.; Hanrath, T.; Korgel, B. A. The Role of Precursor‐

Decomposition Kinetics in Silicon‐Nanowire Synthesis in Organic Solvents.

Angewandte Chemie International Edition 2005, 44, 3573.

(8) Zhou, W.; Wang, Z. L. Scanning microscopy for nanotechnology: techniques and applications; Springer science & business media, 2007.

(9) Huang, X.; Yu, Y.; Jones, T.; Fan, H.; Wang, L.; Xia, J.; Wang, Z. J.;

Shao, L. D.; Meng, X. M.; Willinger, M. G. In Situ Formation of

Crystallographically Oriented Semiconductor Nanowire Arrays via Selective

Vaporization for Optoelectronic Applications. Advanced Materials 2016, 28,

7603.

(10) Facility, A. M. a. M. R. 2013; Vol. 2017.

(11) Amirifar, N.; Lardé, R.; Talbot, E.; Pareige, P.; Rigutti, L.; Mancini, L.;

Houard, J.; Castro, C.; Sallet, V.; Zehani, E. Quantitative analysis of doped/undoped ZnO nanomaterials using laser assisted atom probe tomography:

Influence of the analysis parameters. Journal of Applied Physics 2015, 118,

215703.

(12) Flynn, G.; Ramasse, Q. M.; Ryan, K. M. Solvent Vapor Growth of Axial

Heterostructure Nanowires with Multiple Alternating Segments of Silicon and

Germanium. Nano Lett. 2016, 16, 374.

(13) Kennedy, T.; Bezuidenhout, M.; Palaniappan, K.; Stokes, K.; Brandon,

M.; Ryan, K. M. Nanowire heterostructures comprising germanium stems and silicon branches as high-capacity li-ion anodes with tunable rate capability. ACS nano 2015, 9, 7456.

(14) Chemwiki.ucdavis.edu; Vol. 2017.

(15) Kirkham, M.; Wang, Z. L.; Snyder, R. L. Tracking the catalyzed growth process of nanowires by in situ x-ray diffraction. Journal of Applied Physics 2010,

108, 014304.

(16) Kennedy, T.; Mullane, E.; Geaney, H.; Osiak, M.; O’Dwyer, C.; Ryan, K.

M. High-performance germanium nanowire-based lithium-ion battery anodes extending over 1000 cycles through in situ formation of a continuous porous network. Nano letters 2014, 14, 716.

Chapter 3: The Selective Synthesis of Nickel Germanide

Nanowires and Nickel Germanide Seeded Germanium Nanowires

within a Solvent Vapour Growth System

3.1 Abstract

We report the formation of NiGe nanowires by the thermal decomposition of diphenylgermane, in the presence of a bulk Ni foil, in a solvent vapour growth system. The reaction occurs by the initial formation of a NiGe layer on the foil which progresses to the growth of nanowire occlusions of the same phase, typically 40 nm in diameter. Switching the substrate from bulk Ni foil to an evaporated layer of Ni results in the growth of NiGe seeded germanium nanowires, in high yield. The nanowires were characterized using high resolution scanning electron microscopy, x-ray diffraction, and high resolution transmission electron microscopy.

3.2 Introduction

Transition metal silicide and germanide nanowires (NWs) are finding a number of applications in areas including microelectronics, photovoltaics and thermoelectrics.1-3 Their inherent compatibility with group IV materials makes these materials particularly promising for microelectronic applications such as local interconnects, where high conductivity combined with stable crystal structures is desirable.4 Nickel germanide (NiGe), in particular, has been highlighted as a highly promising material for interconnect and contact applications due to its low resistivity and its ability to form an ohmic contact to p- type Ge.5-7

Typically, transition metal silicide and germanide NWs have been synthesized indirectly by the silicidation/germanidation of preformed Si or Ge NWs.1 Using this approach, a transition metal is evaporated onto preformed Group IV NWs and annealed in forming gas, to produce a variety of transition metal silicide and

8 9 10 11 12 germanide NWs, such as; NiSi , Cu3Si , Cu3Ge , MnSi and Mn5Ge3 . This approach has also been used to synthesize Ni2Ge/Ge/Ni2Ge heterostructure NWs by annealing Ni contacts attached to preformed Ge NWs.13 Typically this is a low throughput approach that is used to study silicide/germanide formation as well as to create axial heterostructures. The delivery of a transition metal halide to either a

Ge or Si source at elevated temperatures, often as high as 900 °C, has also been identified as a way to grow transition metal silicide and germanide NWs in high

14 15 16 17 18 yield. To date, NWs of FeSi , CoSi , CrSi2 , Fe1.3Ge and Co5Ge7 have been synthesized using this approach.

Another alternative approach used to synthesize silicide/germanide NWs is the delivery of Si or Ge monomers to their corresponding metal substrate. This approach is particularly advantageous as it typically uses low temperatures (≤500

°C) and also adapts existing setups, which have been used for group IV NW growth, to grow metal silicide and germanide NWs. This protocol has been used

19 20 21 to grow NWs of nickel silicide (NiSi , Ni2Si and Ni3Si2 ) by silane delivery to

Ni substrates within a Chemical Vapour Deposition (CVD) setup. NiGe NWs were also synthesized in a CVD setup by germane delivery to Ni foam.6 Typically in this approach, the Si or Ge vapour reacts with the metal substrate to form a silicide/germanide film from which the silicide/germanide NW growth nucleates.

Recently, the low-cost solvent vapour growth (SVG) system previously used to synthesize Si and Ge NWs from a variety of catalytic layers such as In22 and Sn23,

24 25 26 as well as more complex NWs such as Si-Ge , Si–Ge–Si1-xGex and Si-Ge-Si

27 heterostructures, was used to synthesize Cu15Si4 NWs. In that report, it was also found that Cu3Si seeded Si NWs were grown when the growth substrate comprised a thin film of Cu rather than bulk Cu. Growth of Si NWs from a Cu film is not unusual as transition metal films and foils are well known to seed

Group IV NWs through vapour-solid-solid (VSS) and supercritical fluid-solid- solid (SFSS) mechanisms.28-33 This approach allows for the selective growth of transition metal seeded Group IV NWs or transition metal silicide/germanide

NWs by carefully controlling reaction parameters such as metal film thickness.

Herein, we extend this approach to Ni as the transition metal and have similarly obtained both NiGe NWs and NiGe seeded Ge NWs using this SVG approach with either bulk Ni foil or a thin-film of Ni as the growth substrate. This report is, to the best of our knowledge, the first report of NiGe NWs within a solution based system. Using bulk Ni foil as the substrate, a nickel germanide layer initially forms, from which NiGe NW growth subsequently occurs with continued delivery of the Ge precursor. When the substrates are changed to Ni films (thermally evaporated onto support substrates) it was found that nickel germanide (NiGe) seeded Ge NWs are formed. Of significant interest in the growth of these nanowires comes from the fact that growth occurs through the VSS mechanism, with the solid NiGe seed limiting the incorporation of metal atoms within the NW, a consequence which is typical of VLS growth.34 Importantly, the reaction is

81 completely selective to the substrate type with no simultaneous occurrence of both nanowire types in a single sample.

3.3 Experimental Section

Growth Substrate Preparation:

Bulk Ni foil was purchased from Goodfellows with a 0.25 mm thickness and

99.98% purity. The Ni was cleaned with concentrated nitric acid (69% v/v) and rinsed repeatedly with deionized water and then dried before introduction into the reactor setup.

Thin Ni films were prepared by evaporating 99.995% Ni (Kurt J Lesker company) in a MB-EcoVap Mbraun integrated thermal evaporator on to support substrates

(either Si wafer or stainless steel). For a typical reaction, a 5 nm thin film was used. The substrates were stored in an Ar glovebox prior to reactions, and contact with O2 was minimized.

Reaction Setup:

Reactions were carried out in a 100 ml custom-made Pyrex, round bottomed flasks containing 7 ml of Squalane (99% Aldrich). The growth substrate was placed vertically in to the round bottomed flask which was attached to a Schlenk line setup via a water condenser. This was then ramped to a temperature of 125 °C using a three zone furnace. A vacuum, of at least 100 mTorr, was applied for 60 min to remove moisture from the system. Following this, the system was purged with Ar. The flask was then ramped to the reaction temperature. Reactions were conducted at 440 °C. Upon reaching the reaction temperature, 0.25 mL DPG (>

95% Fluorochem) was injected through a septum cap. In a typical reaction, the reaction was allowed to proceed for 45 min. To terminate the reaction, the furnace

82 was turned off and the setup was allowed to cool to room temperature before extracting the NW coated substrates. After the reaction, the NW covered substrates were removed from the reaction flask. The substrates were rinsed with toluene to remove residual high boiling point solvent (HBS) and dried under a N2 line prior to characterization.

Analysis:

Scanning electron microscopy (SEM) analysis was performed on a Hitachi SU-70 system operating between 3 and 20 kV. The NW coated substrates were untreated prior to SEM analysis. For transmission electron microscopy (TEM) analysis, the

NWs were removed from the growth substrates through the use of a sonic bath.

TEM analysis was conducted using a 200 kV JEOL JEM-2100F field emission microscope equipped with a Gatan Ultrascan CCD camera and EDAX Genesis

EDS detector. X-ray diffraction (XRD) analysis was conducted using a

PANalytical X’Pert PRO MRD instrument with a Cu−Kα radiation source (λ=

1.5418 Å) and an X’celerator detector.

3.4 Results and Discussion

NiGe NW growth

Figure 3.1. a) Schematic detailing the growth of NiGe NWs in the SVG system. b) SEM image showing growth of NWs directly from substrate. c) High magnification SEM image of NiGe showing tapered morphology.

Figure 3.1a shows a schematic of the SVG system where a piece of Ni foil is placed vertically into a long-necked round-bottomed flask and the Ge precursor,

DPG, is introduced by injection at the set reaction temperature. Initially, the DPG is thermally decomposed through the established phenyl redistribution pathway to provide germane gas to the Ni substrate for NW growth.35,36 After a typical reaction, the shiny Ni substrate turned a dull brown/purple color (See Figure 2).

Figure 3.2. Photograph of Ni substrate before and after reaction.

SEM analysis, revealing that NWs extrude from a rough layer on the substrate, is shown in Figure 3.1b. In the higher magnification SEM image shown in Figure

3.1c, it can be seen that the NWs grow directly from the rough germanide layer and can extend to greater than 5 µm in length. No catalytic seeds are evident at the tips of the NWs, which suggests that NW growth does not occur through the VLS or VSS growth mechanisms. Interestingly, the NiGe NWs appear to show a degree of tapering from the root of the NW to the tip, which is apparent in Figure

3.1c. The root of a typical NW has a broad base in the order of approximately 200 nm, whereas the tips of the NWs possess a narrower diameter of approximately 40 nm. It is likely that the germane reflux initially reacts with the Ni foil to form a rough nickel germanide film on the substrate. The continued supply to this layer, of both Ni from the underlying substrate and Ge from the vapour phase, then results in the formation of single crystalline NWs from specific nucleation sites on the rough layer. Increasing the amount of the Ge precursor injected did not result

85 in a noticeable difference in NW yields, suggesting that NW formation may only occur at specific nucleation sites (Figure 3.3).

Figure 3.3. a) and b) show SEM images of a NiGe NW synthesis with 0.5 ml DPG precursor used. No increase in NW yield was observed despite the increase of Ge precursor used.

Figure 3.4. a) Low magnification TEM image of NiGe NW. b) High magnification image of area highlighted in a) with inset FFT indexed for NiGe. c) DF-STEM EDS multi-point analysis of individual NiGe NW with table of corresponding atomic percentage inset (Scale bar = 1 µm), d) Low magnification TEM image of NiGe NW with longitudinal defect. e) High magnification TEM image of area highlighted in d). The inset selected area electron diffraction pattern (viewed down the [131̅] zone axis) is indexed for the orthorhombic NiGe phase.

Figure 3.4a shows a low magnification TEM image of a single NiGe NW, approximately 7 µm in length. A high-resolution transmission electron microscopy (HRTEM) image taken from the marked area in (a) is shown in (b) with a FFT inset indexed for orthorhombic NiGe. From this image, it can be seen that the NW possesses a 3 nm thin amorphous oxide coating (confirmed by EDS line scan in Figure 3.5). Figure 3.4c shows a low magnification DF-STEM image of a NiGe NW approximately 6 µm in length. EDS multi-point analysis was carried out at the highlighted points of this NW and the elemental composition of the NW is listed in the inset table. The stoichiometry of the NW was determined to be approximately 1:1 Ni to Ge and the NW was compositionally homogenous along its length. Figure 3.4d shows a low magnification TEM image of a NW on which a selected area electron diffraction (SAED) study was carried out. A higher magnification image is shown in Figure 3.4e, with an electron diffraction pattern indexed for orthorhombic NiGe (space group 푃푛푚푎; a = 5.381 Å, b = 3.428 Å, c

= 5.811 Å) along the [131̅] zone axis.

Figure 3.5. DF-STEM image with overlapping EDS line showing the presence of oxygen in amorphous region (Scale bar = 20 nm).

It is worth noting that the NWs do display crystallographic defects which are particularly prevalent at their roots (Figure 3.6). The high concentration of defects at the root of the NWs may suggest that defects play a role in NW formation from the underlying layer as suggested by others previously.37 These defects appear to be different to defects previously seen for pure Ge NWs38 and are more reminiscent of threading dislocations often seen in the GaN system.39,40

Figure 3.6. TEM images displaying high defect concentration at NW base; (a) bright field TEM image of NW base with low magnification image and corresponding electron diffraction pattern inset, (b), (c) and (d) show dark-field TEM images relating to the highlighted spots in the electron diffraction pattern shown inset in (a).

Figure 3.7. a) SEM image of Ni substrate after a 20 min reaction time. b) XRD pattern taken from a Ni substrate after 20 min reaction time with the marked reflections corresponding to NiGe, Ni5Ge3 and Ni. c) SEM image of a Ni substrate after 30 minute reaction time. d) XRD pattern of NW covered substrate after 30 min reaction time.

To study the formation of the NiGe NWs from the Ni substrate, SEM and XRD analysis of substrates after set reaction times was conducted. After 20 min, no

NW growth was evident on the Ni foil; however, a rough germanide layer had formed on the substrate as seen in Figure 3.7a. The XRD pattern obtained from the Ni substrate after a reaction time of 20 min is shown in Figure 3.7b. Two nickel germanide phases, namely NiGe (ICDD: 98-005-2964) and Ni5Ge3

(ICDD: 98-006-2505), were evident. The presence of multiple phases has been seen previously in other systems and is a strong indication that metal diffusion from the bulk substrate occurs and is important in the formation of transition

91 metal germanide NWs.6,27 After a 30 min reaction time (Figure 3.7c), short NWs were observed growing from the germanide layer. Interestingly, the NWs appear to be clustered together and display a degree of tapering even at this stage of NW growth. Similar NW growth morphologies have previously been observed in the synthesis of NiSi NWs using a CVD approach.41 In a previous study by Jin et al., it was observed that high Ge supersaturation levels favor the formation of bulk film morphologies.6 As the supersaturation levels are decreased, 1-D morphologies are favored.42,43 In the context of the single injection system employed here, the rough nickel germanide layer forms initially, when the Ge supersaturation level is high. As the supersaturation level decreases, NW growth becomes favorable resulting in the growth of the tapered NWs. XRD analysis of the substrate after a 30 min reaction time is shown in Figure 3.3d. Again, the presence of the NiGe phase as well as the nickel rich phase Ni5Ge3 was confirmed.

Figure 3.8. a) SEM image of a Ni substrate after 45 min reaction time. b) XRD pattern of NW covered substrate after 45 minute reaction time with the marked reflections corresponding to NiGe, Ni5Ge3 and Ni. c) SEM image of a Ni substrate after 60 min reaction time. d) XRD pattern of NW covered substrate after 60 minute reaction time with the marked reflections corresponding to NiGe, Ni2Ge, Ni5Ge3 and Ni.

By increasing the reaction time to 45 min, the length of the NWs could be increased. XRD analysis of the Ni substrate after 45 min again confirmed the presence of the NiGe and Ni5Ge3 phases (Figure 3.8b). Interestingly, a noticeable change was detected in the XRD pattern obtained when the reaction time was extended to 60 min, as seen in Figure 3.8d. The characteristic peak for the NiGe phase at 34.6° (highlighted with dotted lines in Figure 3.7 and Figure 3.8) is no longer evident, suggesting that the majority of the germanide present is in the form of the more Ni rich phases (Ni5Ge3, Ni2Ge). It appears that as the Ge supply depletes over time, the continuous diffusion of Ni from the bulk foil transforms the NiGe phase to more Ni rich phases. Despite this transformation in the nickel

93 germanide layer, analysis of the NWs confirmed the phase to be NiGe even after a

120 min reaction time (Figure 3.9).

Figure 3.9. DF-STEM image of NiGe NW formed after 120 min reaction with EDS spot analysis

Ge NW growth

Figure 3.10. a) Low magnification SEM image showing high density of Ge NW growth across growth substrate. b) Low magnification SEM image of highlighted area in the dense mesh of Ge NWs with higher magnification SEM image of the individual NW in c). d) Schematic detailing the growth of NiGe seeded Ge NWs in the SVG system through the VSS growth mechanism.

To investigate the effect on NW formation of limiting the Ni source, experiments were repeated using thin Ni films rather than the bulk Ni foil. Following a typical reaction, using a thin film of Ni (5 nm Ni) thermally evaporated onto a Si support substrate in the SVG system, a dark brown/purple colored layer is formed, which easily flakes from the growth substrate (Figure 3.11). SEM analysis of the substrate after a typical reaction, Fig 10b, confirms the growth of NWs across the substrate in a very high yield. The NWs have an average diameter of 104 nm with a standard deviation of 22 nm and are typically less than 5 µm in length. As can

95 be seen from the higher magnification image in Figure 3.10c, many of the NWs exhibit a torturous morphology and the catalytic seeds can be seen at the tips of the NWs. The growth of Ge NWs, well below the lowest melting eutectic, suggests that the NW growth occurs through the VSS growth mechanism, as described in the schematic in Figure 3.10d.

Figure 3.11. Photograph of NW coated substrate after reaction. Mats of NWs can easily be removed from substrate using tweezers.

When a 100 nm Ni layer was used instead, it was also observed that Ge NWs were grown (Figure 3.12). XRD analysis of this 100 nm thin film shown in

Figure 3.13 confirmed the presence of both NiGe and Ge (ICDD: 01-071-4636).

This suggests that a much thicker layer is needed for NiGe NW formation.

Figure 3.12. a) Low magnification SEM image showing NiGe seeded Ge growth across from a 100 nm Ni thin film, flaking from the substrate can be seen b) Low magnification SEM image of highlighted area in a). c) Higher magnification image of the dense mesh of Ge NWs.

Figure 3.13. XRD pattern of NiGe seeded Ge NWs using a 100 nm thick Ni film.

Figure 3.14. a) Low magnification TEM image of a <110> grown Ge NW . b) A high magnification TEM image of the highlighted area in a) with indexed FFT inset. c) High magnification TEM image of NiGe seed with low magnification image and indexed FFT inset. d) DF-STEM image of Ge NW with overlapping EDS line scan.

The NWs were further analyzed using TEM and DF-STEM EDS. Figure 3.14a shows a low magnification TEM image of a nickel germanide seeded Ge NW with a higher magnification image of the highlighted area in Figure 3.14b. The inset FFT in Figure 3.14b is indexed for diamond cubic Ge (space group 퐹푑3̅푚, a= 5.6576 Å) with a <110> growth direction. It is worth noting that many of the

NWs are torturous in nature (Figure 3.15). This torturous nature appears to be the result of the NW diameter and twinning and stacking faults as described in

99 previous studies of Ge NW growth.38,44 Figure 3.14c shows a high magnification

TEM image of the nickel germanide seed with low magnification image and FFT indexed for orthorhombic NiGe (space group 푃푛푚푎; a = 5.381 Å, b = 3.428 Å, c

= 5.811 Å) inset. EDS analysis of a number of seeds also confirmed a 1:1 Ni to

Ge ratio for the NiGe seed. Interestingly, previous reports on the synthesis of Ge

NWs using nickel germanide seeds identified, the germanium rich phase, NiGe2 as the catalytic seeds.31,45 The overlaid EDS line profile of the NW in the DF-

STEM image shown in Figure 3.14d again confirms that the catalytic seed is nickel germanide which is a strong indication of Ge NW growth through the VSS growth mechanism. It is proposed that the germane flux initially reacts with the Ni layer to form nickel germanide. As the Ni source is limited due to the use of a thin

Ni film, there is no continuous supply of Ni to diffuse into the growing NW. This results in the formation of small particles which then act as seeds for elemental Ge

NW growth.

100

Figure 3.15. a) Low magnification TEM image of NiGe seeded Ge NW. b) High magnification TEM image of area 1 in a) with indexed FFT inset. Double spots in FFT indicate the presence of twinning. c) High magnification image of area 2 in a). A high concentration of defects can be seen as the NW begins to kink.

101

3.5 Conclusion

In this work, we have detailed the growth of nickel germanide (NiGe) NWs using a facile wet chemical growth system. With bulk Ni foil as the growth substrate, it was possible to synthesize NiGe NWs, a material highly sought in the semiconductor industry, by the liquid injection of a germanium precursor onto the substrate placed in the vapour of a high boiling point solvent. We have shown that diffusion of Ni from the underlying substrate plays an important role in the formation of NiGe NWs. This Ge delivery approach to the synthesis of transition metal germanide NWs provides a low cost, single step alternative approach to the direct germanidation of Ge NWs, which is typically used.

By limiting the nickel source, through the use of thermally evaporated Ni thin films, NiGe seeded Ge NWs could also be grown in high yield within the SVG system. The versatility to switch growth between the formation of metal germanide nanowires and metal germanide seeded germanium wires by changing the Ni source is very attractive, as both these materials are desirable in device applications. NiGe NWs are attractive as interconnects and contacts in microelectronics, due to their low resistivities, whereas NiGe seeded Ge NWs boast the advantage of VSS seeding, namely possessing an interface between the metal germanide seeds and resultant NWs, that allows optimal charge transfer.

The extension of this protocol from Cu to Ni is an important demonstrator for the general applicability of this approach, with the potential for a broad range of germanide and silicide nanostructures to be produced across the transition metal spectrum.

102

3.6 References

(1) Schmitt, A. L.; Higgins, J. M.; Szczech, J. R.; Jin, S. Synthesis and applications of metal silicide nanowires. J. Mater. Chem. 2010, 20, 223.

(2) Yoon, H.-N.; Yoo, Y.-D.; Seo, K.-Y.; In, J.-H.; Kim, B.-S. Synthesis and

Applications of Noble Metal and Metal Silicide and Germanide 1-Dimensional

Nanostructures. Bull. Korean Chem. Soc. 2012, 33, 2830.

(3) Yuan, F. W.; Wang, C. Y.; Li, G. A.; Chang, S. H.; Chu, L. W.; Chen, L.

J.; Tuan, H. Y. Solution-phase synthesis of single-crystal Cu3Si nanowire arrays on diverse substrates with dual functions as high-performance field emitters and efficient anti-reflective layers. Nanoscale 2013, 5, 9875.

(4) Lin, Y. C.; Chen, Y.; Huang, Y. The growth and applications of silicides for nanoscale devices. Nanoscale 2012, 4, 1412.

(5) Peter, A. P.; Opsomer, K.; Adelmann, C.; Schaekers, M.; Meersschaut, J.;

Richard, O.; Vaesen, I.; Moussa, A.; Franquet, A.; Zsolt, T.; Van Elshocht, S.

Ultrathin NiGe films prepared via catalytic solid-vapor reaction of Ni with

GeH(4). ACS Appl. Mater. Interfaces 2013, 5, 9605.

(6) Yan, C.; Higgins, J. M.; Faber, M. S.; Lee, P. S.; Jin, S. Spontaneous growth and phase transformation of highly conductive nickel germanide nanowires. ACS nano 2011, 5, 5006.

(7) Tang, J.; Wang, C.-Y.; Xiu, F.; Lang, M.; Chu, L.-W.; Tsai, C.-J.; Chueh,

Y.-L.; Chen, L.-J.; Wang, K. L. Oxide-Confined Formation of Germanium

Nanowire Heterostructures for High-Performance Transistors. ACS nano 2011, 5,

6008.

103

(8) Wu, Y.; Xiang, J.; Yang, C.; Lu, W.; Lieber, C. M. Single-crystal metallic nanowires and metal/semiconductor nanowire heterostructures. Nature 2004, 430,

61.

(9) Chiu, C. H.; Huang, C. W.; Chen, J. Y.; Huang, Y. T.; Hu, J. C.; Chen, L.

T.; Hsin, C. L.; Wu, W. W. Copper silicide/silicon nanowire heterostructures: in situ TEM observation of growth behaviors and electron transport properties.

Nanoscale 2013, 5, 5086.

(10) Burchhart, T.; Lugstein, A.; Hyun, Y. J.; Hochleitner, G.; Bertagnolli, E.

Atomic scale alignment of copper-germanide contacts for ge nanowire metal oxide field effect transistors. Nano Lett. 2009, 9, 3739.

(11) Lin, Y. C.; Chen, Y.; Shailos, A.; Huang, Y. Detection of spin polarized carrier in silicon nanowire with single crystal MnSi as magnetic contacts. Nano

Lett. 2010, 10, 2281.

(12) Tang, J.; Wang, C. Y.; Jiang, W.; Chang, L. T.; Fan, Y.; Chan, M.; Wu,

C.; Hung, M. H.; Liu, P. H.; Yang, H. J.; Tuan, H. Y.; Chen, L. J.; Wang, K. L.

Electrical probing of magnetic phase transition and domain wall motion in single- crystalline Mn(5)Ge(3) nanowire. Nano Lett. 2012, 12, 6372.

(13) Tang, J.; Wang, C. Y.; Xiu, F.; Hong, A. J.; Chen, S.; Wang, M.; Zeng, C.;

Yang, H. J.; Tuan, H. Y.; Tsai, C. J.; Chen, L. J.; Wang, K. L. Single-crystalline

Ni2Ge/Ge/Ni2Ge nanowire heterostructure transistors. Nanotechnology 2010, 21,

505704.

(14) Ouyang, L.; Thrall, E. S.; Deshmukh, M. M.; Park, H. Vapor-Phase

Synthesis and Characterization of ε-FeSi Nanowires. Adv. Mater. 2006, 18, 1437.

104

(15) Seo, K.; Varadwaj, K. S.; Mohanty, P.; Lee, S.; Jo, Y.; Jung, M. H.; Kim,

J.; Kim, B. Magnetic properties of single-crystalline CoSi nanowires. Nano Lett.

2007, 7, 1240.

(16) Seo, K.; Varadwaj, K. S. K.; Cha, D.; In, J.; Kim, J.; Park, J.; Kim, B.

Synthesis and Electrical Properties of Single Crystalline CrSi2 Nanowires. J.

Phys. Chem. C 2007, 111, 9072.

(17) Yoon, H.; Lee, A. T.; Choi, E. A.; Seo, K.; Bagkar, N.; Cho, J.; Jo, Y.;

Chang, K. J.; Kim, B. Structure-induced ferromagnetic stabilization in free- standing hexagonal Fe(1.3)Ge nanowires. J. Am. Chem. Soc 2010, 132, 17447.

(18) Yoon, H.; Seo, K.; Bagkar, N.; In, J.; Park, J.; Kim, J.; Kim, B. Vertical epitaxial co5 ge7 nanowire and nanobelt arrays on a thin graphitic layer for flexible field emission displays. Adv. Mater. 2009, 21, 4979.

(19) Decker, C. A.; Solanki, R.; Freeouf, J. L.; Carruthers, J. R.; Evans, D. R.

Directed growth of nickel silicide nanowires. Appl. Phys. Lett. 2004, 84, 1389.

(20) Yan, X. Q.; Yuan, H. J.; Wang, J. X.; Liu, D. F.; Zhou, Z. P.; Gao, Y.;

Song, L.; Liu, L. F.; Zhou, W. Y.; Wang, G.; Xie, S. S. Synthesis and characterization of a large amount of branched Ni2Si nanowires. App. Phys. A

2004, 79.

(21) Kim, J.; Shin, D. H.; Lee, E.-S.; Han, C.-S.; Park, Y. C. Electrical characteristics of single and doubly connected Ni silicide nanowire grown by plasma-enhanced chemical vapor deposition. Appl. Phys. Lett. 2007, 90, 253103.

(22) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High Density Growth of Indium seeded Silicon Nanowires in the

Vapor phase of a High Boiling Point Solvent. Chem. Mater. 2012, 24, 2204.

105

(23) Mullane, E.; Kennedy, T.; Geaney, H.; Dickinson, C.; Ryan, K. M.

Synthesis of Tin Catalyzed Silicon and Germanium Nanowires in a Solvent–

Vapor System and Optimization of the Seed/Nanowire Interface for Dual Lithium

Cycling. Chem. Mater. 2013, 25, 1816.

(24) Geaney, H.; Mullane, E.; Ramasse, Q. M.; Ryan, K. M. Atomically abrupt silicon-germanium axial heterostructure nanowires synthesized in a solvent vapor growth system. Nano Lett. 2013, 13, 1675.

(25) Mullane, E.; Geaney, H.; Ryan, K. M. Synthesis of silicon-germanium axial nanowire heterostructures in a solvent vapor growth system using indium and tin catalysts. Phys. Chem. Chem. Phys. 2015, 17, 6919.

(26) Flynn, G.; Ramasse, Q. M.; Ryan, K. M. Solvent Vapor Growth of Axial

Heterostructure Nanowires with Multiple Alternating Segments of Silicon and

Germanium. Nano Lett. 2016, 16, 374.

(27) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chem. Mater. 2012, 24, 4319.

(28) Richards, B. T.; Gaskey, B.; Levin, B. D. A.; Whitham, K.; Muller, D.;

Hanrath, T. Direct growth of germanium and silicon nanowires on metal films. J.

Mater. Chem. C 2014, 2, 1869.

(29) Yuan, F.-W.; Yang, H.-J.; Tuan, H.-Y. Seeded silicon nanowire growth catalyzed by commercially available bulk metals: broad selection of metal catalysts, superior field emission performance, and versatile nanowire/metal architectures. J. Mater. Chem. 2011, 21, 13793.

106

(30) Mullane, E.; Geaney, H.; Ryan, K. M. Size controlled growth of germanium nanorods and nanowires by solution pyrolysis directly on a substrate.

Chem. Commun. 2012, 48, 5446.

(31) Barth, S.; Koleśnik, M. M.; Donegan, K.; Krstić, V.; Holmes, J.D.

Diameter-Controlled Solid-Phase Seeding of Germanium Nanowires: Structural

Characterization and Electrical Transport Properties. Chem. Mater. 2011, 23,

3335.

(32) Bezuidenhout, M.; Kennedy, T.; Belochapkine, S.; Guo, Y.; Mullane, E.;

Kiely, P. A.; Ryan, K. M. High density and patternable growth of silicon, germanium and alloyed SiGe nanowires by a rapid anneal protocol. J. Mater.

Chem. C 2015, 3, 7455.

(33) Barrett, C. A.; Gunning, R. D.; Hantschel, T.; Arstila, K.; O'Sullivan, C.;

Geaney, H.; Ryan, K. M. Metal surface nucleated supercritical fluid–solid–solid growth of Si and Ge/SiOx core–shell nanowires. J. Mater. Chem. 2010, 20, 135.

(34) Ke, Y.; Weng, X.; Redwing, J. M.; Eichfeld, C. M.; Swisher, T. R.;

Mohney, S. E.; Habib, Y. M. Fabrication and Electrical Properties of Si

Nanowires Synthesized by Al Catalyzed Vapor−Liquid−Solid Growth. Nano Lett.

2009, 9, 4494.

(35) Lu, X.; Harris, J. T.; Villarreal, J. E.; Chockla, A. M.; Korgel, B. A.

Enhanced Nickel-Seeded Synthesis of Germanium Nanowires. Chem. Mater.

2013, 25, 2172.

(36) Lee, D. C.; Hanrath, T.; Korgel, B. A. The role of precursor- decomposition kinetics in silicon-nanowire synthesis in organic solvents. Angew.

Chem. Int. Edit. 2005, 44, 3573.

107

(37) Bierman, M. J.; Lau, Y. K. A.; Kvit, A. V.; Schmitt, A. L.; Jin, S.

Dislocation-Driven Nanowire Growth and Eshelby Twist. Science 2008, 320,

1060.

(38) Geaney, H.; Dickinson, C.; Weng, W.; Kiely, C. J.; Barrett, C. A.;

Gunning, R. D.; Ryan, K. M. Role of Defects and Growth Directions in the

Formation of Periodically Twinned and Kinked Unseeded Germanium

Nanowires. Cryst. Growth Des. 2011, 11, 3266.

(39) Moram, M. A.; Ghedia, C. S.; Rao, D. V. S.; Barnard, J. S.; Zhang, Y.;

Kappers, M. J.; Humphreys, C. J. On the origin of threading dislocations in GaN films. J. Appl. Phys. 2009, 106, 073513.

(40) Hersee, S. D.; Rishinaramangalam, A. K.; Fairchild, M. N.; Zhang, L.;

Varangis, P. Threading defect elimination in GaN nanowires. J. Mater. Res. 2011,

26, 2293.

(41) Kim, C. J.; Kang, K.; Woo, Y. S.; Ryu, K. G.; Moon, H.; Kim, J. M.;

Zang, D. S.; Jo, M. H. Spontaneous Chemical Vapor Growth of NiSi Nanowires and Their Metallic Properties. Adv. Mater. 2007, 19, 3637.

(42) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim,

F.; Yan, H. One-Dimensional Nanostructures: Synthesis, Characterization, and

Applications. Adv. Mater. 2003, 15, 353.

(43) Guiton, B. S.; Gu, Q.; Prieto, A. L.; Gudiksen, M. S.; Park, H. Single-

Crystalline Vanadium Dioxide Nanowires with Rectangular Cross Sections. J.

Am. Chem. Soc 2005, 127, 498.

(44) Thombare, S. V.; Marshall, A. F.; McIntyre, P. C. Size effects in vapor- solid-solid Ge nanowire growth with a Ni-based catalyst. J. Appl. Phys. 2012,

112, 054325.

108

(45) Tuan, H.-Y.; Lee, D. C.; Hanrath, T.; Korgel, B. A. Germanium Nanowire

Synthesis: An Example of Solid-Phase Seeded Growth with Nickel Nanocrystals.

Chem. Mater. 2005, 17, 5705.

109

Chapter 4: The effect of enhanced precursor decomposition on the

synthesis of Copper Silicide nanostructures within a Solvent

Vapour Growth System

4.1 Abstract

Here, the effect of enhanced silicon precursor decomposition on the synthesis of copper silicide nanostructures within a solvent vapour growth system was investigated. By adding the reducing agent lithium borohydride, it was possible to enhance silicon precursor decomposition and grow copper silicide (Cu15Si4) nanowires at a reaction temperature of 350 °C. Additionally, it was found that the phase and morphology of the resultant NSs could be tuned by adding the reducing agent at temperatures greater than 400 °C, culminating in the formation of Cu3Si nanostructures. The various nanostructures were extensively characterized using high resolution scanning electron microscopy, high resolution transmission electron microscopy and electron diffraction.

4.2 Introduction

Transition metal silicides have attracted a great deal of research interest in recent years, due to their potential in a variety of different fields, such as; microelectronics, photovoltaics, and energy storage.1-4 In particular, transition metal silicides, such as copper silicide, have been investigated as diffusion barriers for Cu interconnects in the microelectronic industry.5,6 As a result of this interest, significant attention has focused on the development of synthetic approaches for transition metal silicide nanostructures (NSs).7 One of the most

110 successful approaches for the growth of these NSs has been the delivery of Si to the transition metal foil, typically through a chemical vapour deposition (CVD) setup, with silane acting as the Si precursor.8 This approach has been used to synthesise nickel silicide nanowires (NWs) of a number of different compositions,

9 10,11 12,13 including; NiSi, Ni2Si, and Ni3Si2 by varying reaction parameters such as the Si precursor, partial pressures and reaction temperatures. Similar work has recently been carried out in the synthesis of copper silicide NSs using a CVD setup where the authors were able to tune the morphology and composition of the

NS formed by varying the pressures and precursors used.14 This ability to switch between different phases is of great importance for device integration of these materials as various properties (e.g. conductivity and magnetic behaviour) of the transition metal silicide NSs can be varied as a function of chemical composition and crystal phase.15,16

Recently, the Si delivery approach to transition metal silicide NW synthesis has been extended to solution based approaches by Yuan et al, who adapted a supercritical fluid based setup, which is typically used for Si NW synthesis, to synthesise copper silicide (Cu3Si) NWs from Cu metal films by the thermal decomposition of phenylsilane in supercritical benzene.17 The solvent vapour growth (SVG) system is another solution based approach, which is typically used for Si and Ge NW synthesis, that has been adapted for the synthesis of copper

18-20 silicide (Cu15Si4) NWs. In that report, the organometallic precursor, phenylsilane was injected into the system to react with bulk Cu foil at 460 ºC. To date, the ability to switch between different phases of transition metal silicides has not been demonstrated within these solution-based approaches.

111

Recently, we have shown that the addition of the reducing agent LiBH4, can enhance the decomposition of phenylsilane, allowing the growth of Si NWs at remarkably low temperatures (< 250 °C).21 Here, we demonstrate that by altering the reactivity of the phenylsilane precursor, through the addition of the reducing agent, LiBH4, and by varying the reaction temperature, copper silicide NSs of different compositions and morphologies can be synthesised within the SVG system.

4.3 Experimental Section

Reaction setup:

Bulk Cu foil was purchased from Goodfellows with a 0.25 mm thickness and

99.9 % purity. The Cu was cleaned with nitric acid (10 % v/v) and rinsed repeatedly with deionized water and then dried before introduction into the reactor setup.

Reactions were carried out in custom-made Pyrex, round bottomed flasks containing 7 ml of squalane (99 % Aldrich). The Cu substrate was placed vertically into the round bottomed flask which was attached to a Schlenk line setup via a water condenser. This was then ramped to a temperature of 125 °C using a three zone furnace. A vacuum of at least 100 mTorr was applied for 60 min to remove moisture from the system. Following this, the system was purged with Ar. The flask was then ramped to the reaction temperature. Reactions were conducted between 300 and 450 °C. Upon reaching the reaction temperature, 0.75 mL phenylsilane (98 % Fluorochem) was injected through a septum cap.

112

For experiments investigating the effect of the LiBH4 addition, 0.10 ml of a LiBH4 solution (2.0 M in THF Aldrich):squalane mixture (1:4 v/v) was injected into the

SVG system, after the initial injection of phenylsilane (< 30 sec between injections).

In a typical synthesis, the reaction was allowed to proceed for 45 min. To terminate the reaction, the furnace was turned off and the setup was allowed to cool to room temperature before extracting the NS coated substrates. After the reaction, the NS covered substrates were removed from the reaction flask. The substrates were rinsed with toluene to remove residual high boiling point solvent

(HBS) and dried under a N2 line prior to characterization.

Material Characterisation:

Scanning electron microscopy (SEM) analysis was performed on a Hitachi SU-70 system operating between 3 and 20 kV. The NS coated substrates were untreated prior to SEM analysis. For transmission electron microscopy (TEM) and dark field scanning transmission electron microscopy (DF-STEM) analysis, the NS were removed from the growth substrates through the use of a sonic bath. TEM analysis was conducted using a 200 kV JEOL JEM-2100F field emission microscope equipped with a Gatan Ultrascan CCD camera and EDAX Genesis

EDS detector. X-ray diffraction (XRD) analysis was conducted using a

PANalytical X’Pert PRO MRD instrument with a Cu−Kα radiation source

(λ = 1.5418 Å) and an X’celerator detector.

113

4.4 Results and Discussion

To establish the effect of the growth temperature on the synthesis of copper silicide NSs, a number of reactions were carried out between 300 and 450 ºC. In this set of experiments, squalane and Cu foil were placed into the SVG system.

Following a degas step, the flask was ramped up to a set reaction temperature.

Phenylsilane was then injected into the SVG system and allowed to react for 45 min.

Figure 4.1a shows an SEM image of a substrate following a reaction carried out at

300 °C. The substrate, shown inset, still had a characteristic Cu colour following the experiment, suggesting no copper silicide formation and this was confirmed through the SEM image which showed no NS growth. Figure 4.1b shows an SEM image of the growth substrate after a 45 min reaction at 350 °C where again, no

NS growth was apparent on the substrate. However, the substrate turned from the characteristic Cu colour to a more grey shade, as seen in Figure 4.1b inset. This grey colour suggests that phenylsilane has thermally decomposed to form copper silicide. At a reaction temperature of 400 ºC, very short NWs can be seen across the substrate, as shown in Figure 4.1c. In the high magnification SEM image shown in Figure 4.1c inset, this initial NW growth from the copper silicide layer can be more clearly observed. These NWs are approximately 100 nm in diameter and typically less than 1 µm in length. Figure 4.1d shows the NW growth across the substrate as the reaction temperature is increased to 450 °C. As seen in Figure

4.1d, the NWs possess a tight diameter distribution with an average diameter of approximately 100 nm and are greater than 5 µm in length.

114

Figure 4.1. SEM images of copper substrates post synthesis: a) 300 °C with image of Cu substrate inset, b) 350 °C with image of substrate inset, c) 400 °C with high magnification image of initial NW growth inset and d) 450 °C with high magnification image of NWs inset.

Figure 4.2 shows XRD patterns of the Cu substrates after reacting with phenylsilane at the different temperatures. The patterns obtained for the reaction temperatures of 350 °C, 400 °C and 450 °C are quite similar, showing reflections for Cu0.83Si0.17 (ICDD: 04-014-4308), Cu15Si4 (ICDD: 04-014-4307) and Cu

(ICDD: 01-070-3039). As expected, the XRD pattern for the substrate from the reaction at 300 °C did not indicate the presence of any copper silicide phases.

Previous studies into the degradation of phenylsilane, for the production of Si

NWs, found that the reaction temperature played an important role in whether

NW growth occurred.22 Phenylsilane undergoes a disproportionation reaction into silane and diphenylsilane at elevated temperatures.23,24 The produced silane then

115 decomposes into silicon monomers at approximately 350 °C. This inability of silane to break down at the lower temperatures here explains the absence of copper silicide formation at 300 °C

Figure 4.2. XRD patterns of copper substrates, at different reaction temperatures, post synthesis with line patterns for Cu0.83Si0.17, Cu15Si4 and Cu shown for reference.

Figure 4.3a shows a low magnification TEM image of a copper silicide NW synthesised at a growth temperature of 400 °C with a high magnification TEM image of the NW, with lattice fringing, shown in Figure 4.3b. The Fast Fourier

Transform (FFT) pattern shown inset in Figure 4.3b is indexed for cubic Cu15Si4

(space group 퐼4̅3푑; a = 9.7040 Å). Figure 4.3c shows a DF-STEM image of a

Cu15Si4 NW grown at 400 °C with an EDS line scan of the highlighted area shown inset which confirms the presence of both Cu and Si. TEM analysis of the copper silicide NWs produced at 450 °C was also conducted. Figure 4.3d shows a low magnification TEM image with a high magnification image of the highlighted area shown in Figure 4.3e. The NWs grown at 450 °C have similar diameters but

116 are much longer compared to the NWs grown at 400 °C. This is likely due to a slower growth rate due to limited phenylsilane decomposition at lower temperatures. The FFT pattern inset in Figure 4.3e was again indexed for the

Cu15Si4 phase. Figure 4.3f shows a DF-STEM image of the NW with an overlaid

EDS line scan shown inset which confirms the presence of Cu and Si.

117

Figure 4.3. TEM images of copper silicide NWs at different temperatures: a) Low magnification TEM image of NW synthesised at 400 °C with corresponding high magnification TEM image shown in b). The FFT inset in b) is indexed for Cu15Si4. c) DF-STEM image with overlaid EDS line scan inset showing presence of both Cu and Si. d) Low magnification TEM image of NW synthesised at 450 °C with corresponding high magnification TEM image shown in e). The FFT inset in e) is indexed for Cu15Si4. f) DF-STEM image with overlaid EDS line scan inset showing presence of both Cu and Si.

118

To investigate the effect of the use of LiBH4 with phenylsilane on the formation of copper silicide NSs, a number of reactions were carried out using LiBH4 between

300 and 450 ºC. Figure 4.4 shows SEM images of the Cu substrates when LiBH4 was used at the reaction temperatures of 300 °C and 350 °C.

After the reaction at 300 °C, the substrate exhibited the characteristic grey colour which is indicative of the formation of copper silicide (Figure 4.4a inset). SEM analysis showed a layer like structure across the substrate, however no NS formation was observed, as shown in Figure 4.4a. At a reaction temperature of

350 °C, short NW growth was seen across the substrate as presented in Figure

4.4b. In the SEM image inset in Figure 4.4b, it can be seen that these NWs are similar to the Cu15Si4 NWs, seen previously in Figure 4.1. XRD analysis of the substrates is presented in Figure 4.5. The pattern for the substrate after a reaction temperature of 350 °C, again, shows reflections for the Cu0.83Si0.17 and Cu15Si4 phases. Interestingly, the pattern for the reaction temperature of 300 °C now shows reflections for Cu0.83Si0.17 and Cu15Si4. It appears that the use of the reducing agent enhances the decomposition of phenylsilane at these low temperatures resulting in the formation of the copper silicide phases. The formation of NWs at 350 °C is particularly interesting as typically much higher temperatures are needed for transition metal silicide NW synthesis.

119

Figure 4.4. SEM images of copper substrates post synthesis at different reaction temperatures with LiBH4: a) 300 °C with image of Cu substrate inset, b) NW growth at 350 °C with high magnification inset.

120

Figure 4.5. XRD patterns of substrates, at 300 °C and 350°C, post synthesis with line patterns for Cu0.83Si0.17, Cu15Si4 and Cu shown for reference.

To further examine the NWs formed at 350 °C when LiBH4 was used, TEM analysis was carried out. In Figure 4.6a, a low magnification TEM image of a NW grown at 350 °C using LiBH4 to enhance Si precursor decomposition is shown.

The NW is approximately 150 nm in diameter and 2 µm in length. In the higher magnification TEM image in Figure 4.6b, it is possible to see clear lattice fringing. An electron diffraction pattern, of this area, indexed for Cu15Si4 (space group 퐼4̅3푑; a = 9.7040 Å) along the [11̅0] zone axis is shown in Figure 4.6c. A

DF-STEM image of a similar NW is shown in Figure 4.6d. The EDS line scan overlaid confirms the presence of both Si and Cu in the NWs.

121

Figure 4.6. TEM images of a copper silicide NW synthesised at 350 °C in the presence of the reducing: a) Low magnification TEM image of NW with corresponding high magnification TEM image shown in b). The electron diffraction pattern shown in c) is indexed for Cu15Si4. d) DF-STEM image with overlaid EDS line scan inset showing presence of both Cu and Si.

Figure 4.7a shows a low magnification SEM image of a Cu substrate after a reaction at 400 °C, with LiBH4. Interestingly, NW growth seemed only to occur at the highlighted areas in Figure 4.7a. A higher magnification SEM image of the one of these areas is shown in Figure 4.7b. The NWs here appear to grow perpendicularly from microcrystals, which may suggest an epitaxial relationship between the microcrystals and the NWs. The formation of NWs from

122 microcrystals is not unusual for transition metal silicide NWs. In the synthesis of

Cu15Si4 NWs, it was seen that NWs grew from copper silicide crystallites in defined angles. However, the geometries here are significantly different. In the higher magnification SEM images shown in Figure 4.7c and Figure 4.7d, it can be seen that the morphology of the NWs is quite different to the straight NWs seen in

Figure 4.1. The surfaces of the NWs have undulations and irregular morphologies, as can be seen in Figure 4.7d. In addition to the formation of NWs, unique NSs, composed of broad caps and narrow stems, were formed in some areas, as shown in Figure 4.7e. These NSs somewhat resemble the previously seen, ZnO, CoSi and

TiSi NSs.25-27 In Figure 4.7f, it can be seen that the cap size varies between 500 nm and 2 µm. It is worth noting that in the previous report into the synthesis of

Cu3Si NWs by Tuan et al, similar Cu3Si NSs with caps were also observed, however the caps of NSs here are more defined.17

123

Figure 4.7. SEM images of copper substrate post synthesis with enhanced precursor decomposition at a reaction temperature of 400 °C. a) Low magnification SEM image showing areas where NS formation occurred. b) High magnification image showing growth of NWs from microcrystals with higher magnification images shown in c) and d). e) and f) show SEM images of unique NS growth which were also seen.

Figure 4.8a shows a low magnification TEM image of an individual NW, grown at 400 °C in the presence of the reducing agent. Here, the undulated nature of the

NW is clearly noticeable. Figure 4.8b shows a high magnification TEM image of

124 the tip of the NW showing lattice fringing. To identify the crystal structure of the

NWs, electron diffraction patterns were taken from multiple zone axes. Figure

4.8c and Figure 4.8d show 2 diffraction patterns indexed for hexagonal η-Cu3Si

(space group 푃3̅푚1; a = 4.0575 Å, c = 7.3288 Å). Due to the low temperatures used in this NW synthesis, it is likely the NWs are the low temperature polymorph, η''-Cu3Si, however, the plane spacings of the η phase can be used for

28-30 the structural identification of η″-Cu3Si, since they have similar structures. As the LiBH4 enhances the decomposition of phenylsilane, it is not surprising that a more Si rich copper silicide phase, Cu3Si, is formed at the higher reaction temperature. Interestingly, in the diffraction patterns in Figure 4.8c and Figure

4.8d, additional reflections can be seen (circled in red). The structure of Cu3Si is quite complex and satellite reflections have previously been seen in a number of studies of Cu3Si NWs due to the presence of long period anti-phase domains in the crystal structure.31-33 Using the diffraction patterns in Figure 4.8c and Figure

4.8d, a growth direction of [0001] can be assigned for the Cu3Si NWs.

As seen in Figure 4.8e, the NW growth can be quite defective. To check the crystal structure of the NW below the defect, an electron diffraction pattern was taken and again indexed for Cu3Si. A DF-STEM image of a Cu3Si NW is shown in Figure 4.8f with an EDS line scan overlaid confirming the presence of both Cu and Si throughout the NW. Due to the narrow window in atomic percentages between Cu15Si4 (21 atomic % Si) and Cu3Si (25 atomic % Si), EDS measurements here were not used to quantify the composition of the NWs.

125

Figure 4.8. TEM analysis of copper silicide NWs synthesised at 400 °C with LiBH4: a) Low magnification TEM image of NW with corresponding high magnification TEM image shown in b). The electron diffraction patterns shown in c) and d) are indexed for Cu3Si. d) TEM image highlighting the defective NW growth with electron diffraction pattern inset confirming Cu3Si structure. f) DF- STEM image with overlaid EDS line scan inset showing presence of both Cu and Si.

126

To examine the crystal structure of the NSs formed at 400 °C, a TEM study including EDS and electron diffraction was carried out. Figure 4.9a shows a low magnification TEM image of a copper silicide NS. Similar to the NWs, the surface of the NS is quite undulated. A high magnification TEM image of the stem of the NS is shown in Figure 4.9b confirming the crystallinity of the sample with the corresponding electron diffraction pattern taken from the area in Figure

4.9c. Again, this diffraction pattern was indexed for hexagonal η-Cu3Si.

A low magnification DF-STEM image of the Cu3Si NS is shown in Figure 4.9d with EDS line scan overlaid confirming the presence of both Cu and Si in the NS.

To examine whether the NS caps were crystalline, TEM images of the caps were taken at different magnifications. In Figure 4.9e, a low magnification TEM image of the cap of a NS is shown. Due to the thickness of the caps of the NSs (~ 500 nm), it is quite difficult to obtain lattice resolved images. Nevertheless, in the high magnification TEM image in Figure 4.9f, lattice fringing can be seen. The inset

FFT confirms the crystallinity of the NS cap.

127

Figure 4.9. TEM analysis of copper silicide NSs synthesised at 400 °C with LiBH4: a) Low magnification TEM image of NS with corresponding high magnification TEM image shown in b). The electron diffraction pattern shown in c) is indexed for Cu3Si. d) DF-STEM image with overlaid EDS line scan inset showing the presence of both Cu and Si. e) Low magnification TEM image of a NS cap with corresponding high magnification TEM image shown in f). The FFT pattern in f) confirms the crystallinity of the NS cap.

128

At 450 °C, similar NW and NS formations were observed in low yield. However, as can be seen in the SEM image in Figure 4.10a, areas of amorphous material, that charge when imaged, were also present. Again, Cu3Si NWs growing from copper silicide microcrystals are observed Figure 4.10b with Cu3Si NSs also present in some areas, as highlighted in Figure 4.10c. TEM analysis was again conducted to examine the crystal structure of the NSs. In Figure 4.10d, a high magnification TEM image with a low magnification image inset of a Cu3Si NW is shown. The corresponding electron diffraction pattern, also inset, is again indexed for η-Cu3Si.

XRD analysis of the substrates was conducted and showed reflections for

Cu0.83Si0.17 and Cu15Si4. No reflections for Cu3Si were observed in these patterns.

This is likely due to the low yield of NSs across the substrate.

129

Figure 4.10. Analysis of the NS synthesised with enhanced precursor decomposition at a reaction temperature of 450 °C: a) Low magnification SEM image showing area where NS formation occurred and the presence of amorphous material. b) High magnification image showing growth of NWs from microcrystals. c) SEM image of unique NS growth in low yield. d) High magnification TEM image of NW with low magnification TEM image shown and electron diffraction pattern indexed for Cu3Si inset.

To further examine the temperature effect on the nucleation of the Cu3Si NSs and

NWs, a reaction was carried out at 400 °C for a shorter reaction time of 15 minutes. In the SEM image of the substrate shown in Figure 4.11a, it can be seen that the caps of the NSs have already formed. Interestingly, in the highlighted regions in Figure 4.11a and the higher magnification SEM image shown in Figure

4.11b, the elongation of some of these structures can be seen. Figure 4.11c and

Figure 4.11d show the initial stages of Cu3Si NW growth from a microcrystal. It is likely that a Cu3Si microcrystal forms initially and from this, subsequent NW growth occurs. It appears that NW growth is dependent on the initial formation of

130 such microcrystals. This formation of NWs from microcrystals has previously been seen in a number of different systems. Here, we propose the VS mechanism

14 as the growth mechanism. When the LiBH4 is introduced into the system, it appears to make the phenylsilane more reactive. As a result of this increased reactivity, it is likely that the, more Si rich, Cu3Si phase forms initially, forming either microcrystals or NS caps. This appears to occur only in certain areas, as seen in Figure 4.7a. This could be due to droplets of the LiBH4 sticking to the Cu substrate when injected into the SVG system. As the reaction proceeds, the Si concentrations in the vapour phase decrease. This decrease in vapour supersaturation causes a change in growth habit resulting in the unique morphologies seen in this report. Specifically for the Cu3Si NWs, the growth appears to be epitaxially related to the microcrystal. Similar observations were seen in VS grown ZnO NWs.25

131

Figure 4.11. Analysis of the NSs synthesised with enhanced precursor decomposition at a reaction temperature of 400 °C for 15 minutes: a) Low magnification SEM image of the initial stages of NS formation with highlighted regions showing the elongation of NS caps. b) High magnification SEM image of NS formation. c) Low magnification SEM image of Cu3Si NW growth from a microcrystal with a higher magnification image of the microcrystal shown in d).

132

4.5 Conclusion

In conclusion, the effects of temperature and Si precursor reactivity on the formation of copper silicide NWs and NSs within the solvent vapour growth system were studied. It was found that the Si precursor, phenylsilane, has limited decomposition at the low temperature of 300 °C and no copper silicide phases were seen on the Cu substrate. By increasing the temperature to 350 °C, the phenylsilane could be decomposed and this resulted in the formation of a copper silicide layer across the substrate. Further increasing the temperature resulted in the formation of Cu15Si4 NWs.

When the reducing agent, LiBH4, was added, the decomposition of phenylsilane was enhanced and resulted in the formation of a copper silicide layer on the substrate at 300 °C and Cu15Si4 NWs at 350 °C. Further increasing to the reaction temperature to 400 °C; led to the formation of Cu3Si NWs and unique Cu3Si NSs.

This report illustrates the ability to form a variety of copper silicide NSs for the first time in a solution based approach by tuning precursor reactivity and reaction temperature.

133

4.6 References

(1) Schmitt, A. L.; Jin, S. Selective patterned growth of silicide nanowires without the use of metal catalysts. Chemistry of materials 2007, 19, 126.

(2) Lin, Y. C.; Chen, Y.; Huang, Y. The growth and applications of silicides for nanoscale devices. Nanoscale 2012, 4, 1412.

(3) Yoon, H.-N.; Yoo, Y.-D.; Seo, K.-Y.; In, J.-H.; Kim, B.-S. Synthesis and

Applications of Noble Metal and Metal Silicide and Germanide 1-Dimensional

Nanostructures. Bull. Korean Chem. Soc. 2012, 33, 2830.

(4) Reader, A.; Van Ommen, A.; Weijs, P.; Wolters, R.; Oostra, D. Transition metal silicides in silicon technology. Reports on Progress in Physics 1993, 56,

1397.

(5) Hymes, S.; Kumar, K.; Murarka, S.; Ding, P.; Wang, W.; Lanford, W.

Thermal stability of copper silicide passivation layers in copper-based multilevel interconnects. Journal of applied physics 1998, 83, 4507.

(6) Parajuli, O.; Kumar, N.; Kipp, D.; Hahm, J.-i. Carbon nanotube cantilevers on self-aligned copper silicide nanobeams. Applied physics letters 2007, 90,

173107.

(7) Chen, L. J.; Wu, W.-W.; Hsu, H.; Chen, S.; Chueh, Y.; Chou, L.-J.; Lu,

K.; Tu, K. Metal Silicide Nanowires. ECS Transactions 2007, 11, 3.

(8) Decker, C. A.; Solanki, R.; Freeouf, J. L.; Carruthers, J. R.; Evans, D. R.

Directed growth of nickel silicide nanowires. Appl. Phys. Lett. 2004, 84, 1389.

(9) Kabir, M. S.; Morjan, R. E.; Nerushev, O. A.; Lundgren, P.; Bengtsson, S.;

Enokson, P.; Campbell, E. E. B. Plasma-enhanced chemical vapour deposition growth of carbon nanotubes on different metal underlayers. Nanotechnology

2005, 16, 458.

134

(10) Liu, Z.; Zhang, H.; Wang, L.; Yang, D. Controlling the growth and field emission properties of silicide nanowire arrays by direct silicification of Ni foil.

Nanotechnology 2008, 19, 375602.

(11) Yan, X. Q.; Yuan, H. J.; Wang, J. X.; Liu, D. F.; Zhou, Z. P.; Gao, Y.;

Song, L.; Liu, L. F.; Zhou, W. Y.; Wang, G.; Xie, S. S. Synthesis and characterization of a large amount of branched Ni2Si nanowires. App. Phys. A

2004, 79.

(12) Jiang, Y.; Li, Z.; Li, B.; Zhang, J.; Niu, C. Ni 3 Si 2 nanowires grown in situ on Ni foam for high-performance supercapacitors. Journal of Power Sources

2016, 320, 13.

(13) Fan, X.; Zhang, H.; Du, N.; Yang, D. Phase-controlled synthesis of nickel silicide nanostructures. Materials Research Bulletin 2012, 47, 3797.

(14) Klementová, M.; Krabáč, L.; Brázda, P.; Palatinus, L.; Dřínek, V. Cu-Si nanoobjects prepared by CVD on Cu/Cu 5 Si-substrates using various precursors

(SiH 4, EtSiH 3, BuSiH 3) with added H 2 or air. Journal of Crystal Growth 2017,

465, 6.

(15) Stolt, M. J.; Li, Z.-A.; Phillips, B.; Song, D.; Mathur, N.; Dunin-

Borkowski, R. E.; Jin, S. Selective Chemical Vapor Deposition Growth of Cubic

FeGe Nanowires That Support Stabilized Magnetic Skyrmions. Nano letters 2016.

(16) Yan, C.; Higgins, J. M.; Faber, M. S.; Lee, P. S.; Jin, S. Spontaneous growth and phase transformation of highly conductive nickel germanide nanowires. ACS nano 2011, 5, 5006.

(17) Yuan, F. W.; Wang, C. Y.; Li, G. A.; Chang, S. H.; Chu, L. W.; Chen, L.

J.; Tuan, H. Y. Solution-phase synthesis of single-crystal Cu3Si nanowire arrays

135 on diverse substrates with dual functions as high-performance field emitters and efficient anti-reflective layers. Nanoscale 2013, 5, 9875.

(18) Geaney, H.; Dickinson, C.; Barrett, C. A.; Ryan, K. M. High density germanium nanowire growth directly from copper foil by self induced solid seeding. 2011.

(19) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High Density Growth of Indium seeded Silicon Nanowires in the

Vapor phase of a High Boiling Point Solvent. Chem. Mater. 2012, 24, 2204.

(20) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chem. Mater. 2012, 24, 4319.

(21) Flynn, G., University of Limerick, 2016.

(22) Lee, D. C.; Hanrath, T.; Korgel, B. A. The Role of Precursor‐

Decomposition Kinetics in Silicon‐Nanowire Synthesis in Organic Solvents.

Angewandte Chemie International Edition 2005, 44, 3573.

(23) Tuan, H.-Y.; Lee, D. C.; Hanrath, T.; Korgel, B. A. Catalytic solid-phase seeding of silicon nanowires by nickel nanocrystals in organic solvents. Nano letters 2005, 5, 681.

(24) Hanrath, T.; Korgel, B. A. Supercritical fluid–liquid–solid (SFLS) synthesis of Si and Ge nanowires seeded by colloidal metal nanocrystals.

Advanced Materials 2003, 15, 437.

(25) Umar, A.; Rahman, M. M.; Kim, S. H.; Hahn, Y.-B. Zinc oxide nanonail based chemical sensor for hydrazine detection. Chemical communications 2008,

166.

136

(26) Tsai, C.-I.; Yeh, P.-H.; Wang, C.-Y.; Wu, H.-W.; Chen, U.-S.; Lu, M.-Y.;

Wu, W.-W.; Chen, L.-J.; Wang, Z.-L. Cobalt Silicide Nanostructures: Synthesis,

Electron Transport, and Field Emission Properties. Crystal Growth & Design

2009, 9, 4514.

(27) Du, J.; Du, P.; Hao, P.; Huang, Y.; Ren, Z.; Han, G.; Weng, W.; Zhao, G.

Growth mechanism of TiSi nanopins on Ti5Si3 by atmospheric pressure chemical vapor deposition. The Journal of Physical Chemistry C 2007, 111, 10814.

(28) Jung, S. J.; Lutz, T.; Bell, A. P.; McCarthy, E. K.; Boland, J. J. Free- standing, single-crystal Cu3Si nanowires. Crystal Growth & Design 2012, 12,

3076.

(29) Chiu, C.-H.; Huang, C.-W.; Chen, J.-Y.; Huang, Y.-T.; Hu, J.-C.; Chen,

L.-T.; Hsin, C.-L.; Wu, W.-W. Copper silicide/silicon nanowire heterostructures: in situ TEM observation of growth behaviors and electron transport properties.

Nanoscale 2013, 5, 5086.

(30) Xu, K.; He, Y.; Ben, L.; Li, H.; Huang, X. Enhanced electrochemical performance of Si–Cu–Ti thin films by surface covered with Cu3Si nanowires.

Journal of Power Sources 2015, 281, 455.

(31) Zhang, Z.; Wong, L. M.; Ong, H. G.; Wang, X. J.; Wang, J. L.; Wang, S.

J.; Chen, H.; Wu, T. Self-assembled shape-and orientation-controlled synthesis of nanoscale Cu3Si triangles, squares, and wires. Nano letters 2008, 8, 3205.

(32) Li, S.; Cai, H.; Gan, C. L.; Guo, J.; Dong, Z.; Ma, J. Controlled synthesis of copper-silicide nanostructures. Crystal Growth & Design 2010, 10, 2983.

(33) Wen, C.-Y.; Spaepen, F. In situ electron microscopy of the phases of

Cu3Si. Philosophical Magazine 2007, 87, 5581.

137

Chapter 5: Synthesis of Nickel Silicide Nanowires and Axial

Heterostructure NixSi/Si Nanowires within a

Solution-based System

5.1 Abstract

Herein, we report a novel approach to synthesise both nickel silicide nanowires and axial heterostructure nanowires composed of nickel silicide (NixSi) and Si segments in solution. This was achieved by using a thin film of Au, evaporated onto a Ni substrate, to initiate Si NW growth on injection of phenylsilane. Nickel silicide then formed by the in situ silicidation of the Si nanowires through Ni diffusion from the growth substrate. The interfacial abruptness and chemical composition of the heterostructure nanowires was analysed through transmission electron microscopy, electron diffraction, energy dispersive x-ray spectroscopy, aberration corrected scanning transmission electron microscopy and atomically resolved electron energy loss spectroscopy.

5.2 Introduction

Transition metal silicide nanowires (NWs) are an interesting material set which are finding a number of applications in areas including microelectronics, photovoltaics and thermoelectrics.1-4 The inherent compatibility with Si makes these materials particularly promising for microelectronic applications, such as multilevel interconnects, where their high conductivities combined with stable crystal structures are desirable.5-7 As a result of the potential of these materials in a number of different applications, many groups have focused on developing

138 synthetic protocols for transition metal silicides NWs.1,8 In particular, much of the research on transition metal silicides has focused on the synthesis of NWs of the nickel silicides, which have been identified as potential next generation interconnects, due to their low resistivities and the ability of some phases (i.e.

NiSi) to form ohmic contacts to p-doped and n-doped Si.9-13

One of the most successful approaches to synthesise transition metal silicide NWs is by the direct silicidation of preformed Si NWs.8,10 In this approach, Si NWs are first grown through the vapour-liquid-solid (VLS) mechanism. These Si NWs are then coated with the transition metal of choice before being subsequently annealed to form transition metal silicide NWs. By incorporating lithographical processes into this methodology, this conversion can be localised to selected regions of the wire resulting in axial heterostructure NWs composed of transition metal silicide and Si segments.14-16 Recently, many groups have fabricated single

NW transistor devices using these heterostructure NWs.17-19 However, a difficulty with this approach is that a lot of processing is required to fabricate heterostructure NWs making this approach have a very low throughput.20 Due to this limitation, alternative approaches to synthesise transition metal silicide NWs in high yield have been sought after.

In particular, the Si delivery to a transition metal film approach has attracted much interest. Typically, the delivery of Si monomers to a transition metal results in the formation of Si NWs.21-23 However, by carefully controlling the reaction conditions transition metal silicide NWs can be formed instead. This makes this Si delivery protocol particularly advantageous as existing setups, for Si NW growth, can easily be adapted to grow transition metal silicide NWs. To date, NWs of NiSi,24,25

139

26 12,27 28 Ni2Si, Ni3Si2, and Cu0.83Si0.17, have been synthesised by silane (SiH4) delivery within a chemical vapour deposition (CVD) setup. Recently, the solution based solvent vapour growth (SVG) system, typically used for Si and Ge NW growth,29-

33 was adapted to grow Cu15Si4 NWs from bulk Cu foil, using phenylsilane as the

34 Si precursor. Tuan and co-workers were also able to synthesise Cu3Si NWs, by thermally decomposing phenylsilane in a supercritical fluid based system.35

However, despite the success of this Si delivery protocol in the synthesis of transition metal silicide NWs, the fabrication of axial heterostructure transition metal silicide/silicon NWs using this approach has remained elusive.

Here, we describe the development of a novel hybrid Si delivery/silicidation approach to fabricate both nickel silicide NWs and axial heterostructure NWs composed of nickel silicide (NixSi) and Si segments within the solution based system SVG system. In this approach, a thin film of Au is evaporated onto a Ni substrate, to initiate NW growth through the solution-liquid-solid (SLS) mechanism. As a consequence of the high temperatures required for Si NW growth, Ni can diffuse out from the substrate into the as grown NWs forming nickel silicide in the NW via a solid state reaction. Using this approach, NixSi/Si heterostructure NWs and complete nickel silicide NWs can be formed in the same system without the need for complex lithographical processes.

140

5.3 Experimental Section

Growth Substrate Preparation:

Bulk Ni foil was purchased from Goodfellows with a 0.25 mm thickness and

99.98% purity. The Ni was cleaned with concentrated nitric acid (69% v/v) and rinsed repeatedly with deionized water and then dried.

A thin film of Au was then thermally evaporated onto the Ni foil using a MB-

EcoVap Mbraun integrated thermal evaporator. For typical reactions, a 5 nm thin film was used. The substrates were stored in an Ar glovebox prior to reactions, and contact with O2 was minimized.

Reaction Setup:

Reactions were carried out in a 100 ml custom-made Pyrex, round bottomed flasks containing 8 ml of squalane (99% Aldrich). The growth substrate was placed vertically in to the round bottomed flask which was attached to a Schlenk line setup via a water condenser. This was then ramped to a temperature of 125 °C using a three zone furnace. A vacuum, of at least 100 mTorr, was applied for

60 min to remove moisture from the system. Following this, the system was purged with Ar. The flask was then ramped to the reaction temperature. Reactions were conducted at 465 °C. Upon reaching the reaction temperature, 0.75 ml of phenylsilane (98 % Fluorochem) was injected through a septum cap. In a typical reaction, the reaction was allowed to proceed for 60 min. To terminate the reaction, the furnace was turned off and the setup was allowed to cool to room temperature before extracting the NW coated substrates. After the reaction, the

NW covered substrate was removed from the reaction flask. The substrate was rinsed with toluene and methanol to remove residual high boiling point solvent

(HBS) and dried under a N2 line prior to characterization.

141

Analysis:

NWs were removed from the growth substrates through the use of a sonic bath and were dropcast onto Cu TEM grids. TEM analysis was conducted using a

200 kV JEOL JEM-2100F field emission microscope equipped with a Gatan

Ultrascan CCD camera and EDAX Genesis energy dispersive x-ray spectroscopy

(EDS) detector. Aberration corrected scanning transmission electron microscopy

(STEM) work was carried out on a Nion UltraSTEM100 microscope operated at

100 keV primary beam energy. In the conditions used for the experiments, the microscope forms a 0.8 Å probe with a convergence semiangle of 31 mrad. The high angle annular dark field (HAADF) detector semiangular range was calibrated as 82−185 mrad for Z contrast imaging. A Gatan Enfina spectrometer was used to acquire electron energy loss spectra (EELS). Although the native energy spread of the beam delivered by the cold field emission emitter of the microscope is 0.32 eV, the spectrometer was set up so both Ni L2,3 and Si L2,3 edges could be recorded simultaneously, resulting in an energy resolution (estimated by the full width at half-maximum of the zero loss peak) of 1.5 eV. All EELS data were processed using principal component analysis to minimize the influence of noise.

The collection semiangle was 37 mrad for all data presented here. For compositional analysis, Si L2,3, Ni L2,3 , O K, and C K edges were integrated over an 80 eV window after the edge onsets, following the removal of the background using a decaying power law function.

142

5.4 Results and Discussion

Figure 5.1. Schematic of NixSi/Si NW growth using SVG system.

In Figure 5.1, a schematic illustrating the synthetic protocol used for the formation of NixSi/Si NWs is shown. No NW growth was observed using Ni foil without any seed layer, as seen in Figure 5.2a. EDS analysis of the substrates, shown in

Figure 5.2b, indicated only trace amounts of Si present. These trace amounts of Si are likely due to residue from the Si precursor on the Ni foil. The lack of any NW growth from bulk Ni foil is likely due to the slow rate of nickel silicide formation when phenylsilane is used as the Si precursor.36 To offset the inability of Ni to initially form nickel silicide, a thin layer of Au was thermally evaporated onto the

Ni substrate to initiate NW growth. In the SEM image of the Au coated Ni substrate, after a reaction, shown in Figure 5.2c, dense NW growth can be seen.

Interestingly, while limited NW growth occurred in the vapour phase, the NW growth below the HBS liquid line was particularly dense suggesting that NW

143 growth predominately took place through the SLS growth mechanism. While an exact explanation for the preference of NW growth through the SLS mechanism is not known, it is plausible that the preference for SLS growth is due to higher concentrations of Si monomers in the solution compared to the vapour. In Figure

5.2d, an EDS spectrum of the area shown in Figure 5.2c, confirms the presence of

Au, Ni and Si.

Figure 5.2. a) SEM image of bare Ni substrate after Si reaction with corresponding EDS spectrum shown in b). c) SEM image of Au coated Ni substrate after Si reaction showing dense NW growth, with the corresponding EDS spectrum confirming presence of Si and Au.

To analyse the NWs further, DF-STEM studies of the NWs were carried out. In

Figure 5.3a, a low magnification DF-STEM image of a group of NWs is shown, where there is an obvious contrast difference between some of the NWs. The overlaid EDS maps of this area, shown in Figure 5.3b, show that this contrast

144 difference is due to the incorporation of Ni into some of the NWs. Interestingly,

Ni does not incorporate into all of the NWs and the rate of Ni incorporation varies among the NWs. In the highlighted area in Figure 5.3a, it can be seen that Ni has partially incorporated into a Si NW to form a NixSi segment. This varied incorporation of Ni into the Si NWs could be due how well the Si NW is contacted to the Ni substrate. In NWs that are contacted well to the Ni substrate,

Ni, which is known as a fast diffusion metal in Si, could readily diffuse from the substrate into the NW to form nickel silicide via a solid state reaction. If the contact between the Si NW and the Ni substrate is poor, the rate of silicidation could be much slower. Previous studies into the formation of transition metal silicide NWs by the silicidation process have also stated the importance of the nature of the contact between the transition metal and the Si NW.8,15,37 In Figure

5.3c, a higher magnification DF-STEM image of a heterostructure NW is shown with the corresponding overlaid EDS maps in Figure 3d. Here, it can be seen that

Ni is confined to the root of the NW and that there is an abrupt interface between the NixSi and Si segments.

145

Figure 5. 3. a) Low magnification DF-STEM image of NWs with corresponding EDS elemental mapping in b). c) DF-STEM image of heterostructure NW showing Au seed, Si and NixSi segments. The corresponding EDS elemental maps are shown in d).

To investigate the abruptness of the interface between the NixSi and the Si NW segments, aberration corrected STEM analysis coupled with atomic resolved

EELS analysis of the NWs was carried out. In Figure 5.4a, a low magnification

HAADF-STEM image of a NixSi/Si NW with a diameter of approximately 10 nm is shown. Interestingly in the highlighted region in Figure 5.4a, there is a slight contrast difference at the interfacial region. Nevertheless, in the higher magnification HAADF-STEM image shown in Figure 5.4b, the exact interface between the NixSi and Si segments appears to be quite abrupt. The inset FFT of the highlighted region in Figure 5.4b is indexed for cubic Si (space group 퐹푑3̅푚;

146 a = 5.4305 Å) and shows twin defects along the <111> direction. It is plausible that the silicidation process is affected by the presence of these defects, as defects could offer a diffusion pathway for Ni in the NW.33,38-40 In the case of the NW shown in Figure 5.4b, the <111> oriented twin defects are orthogonal to the NW growth direction, allowing a channel for Ni to diffuse through. This defect enabled silicidation process could also offer an explanation for the different rates of silicidation in the NWs observed in this study. For instance, in defect free

NWs, the diffusion rate of Ni into the Si NW could be lower due to the absence of these defects (diffusion channels), leading to slower growth rate of the silicide.

Alternatively defects could run across the NW and impede Ni diffusion through the NW.41 However, no such observations of these lateral defects impeding silicidation formation have yet been made in our studies. It is worth noting that the growth direction of the Si NWs has previously been suggested to play a role in the silicidation formation.8,42,43 The coupling of NW orientations and the presence of defects could offer an alternative explanation for the varying silicidation rates present in this study. HAADF-STEM images of the NixSi segment of the NW are shown in Figure 5.4c and Figure 5.4d with the inset FFT in Figure 5.4c indexed for orthorhombic Ni2Si (space group Pbnm; a = 7.0649 Å, b = 5.0012 Å, c =

3.7307 Å).

147

Figure 5.4. a) Low magnification HAADF-STEM images of heterostructure NW. b) High magnification HAADF-STEM image of the NW interface with the FFT inset of the highlighted area indexed for Si. c) High magnification TEM image of nickel silicide segment with inset FFT indexed for Ni2Si. d) Higher magnification HAADF-STEM image of c) showing lattice fringing.

Figure 5.5a, shows the spectrum region for EELS analysis of the NW.

Interestingly, in the corresponding quantitative EELS line scan shown in Figure

5.5b, an intermediate Ni-poor silicide phase, of approximately 10 nm in length, between the Ni2Si and Si segments can be seen. Initially, the relative composition of Ni and Si corresponds to the Ni2Si phase. However, just before the Si segment of the NW, the Ni concentration decreases. At this intermediate segment, the relative composition corresponds to NiSi. After this intermediate NiSi segment, the Ni concentration dramatically decreases to zero in the pure Si segment as shown in the line scan and EELS map for nickel in Figure 5.5c. In the overlaid

148 coloured EELS map for Ni/Si/O in Figure 5.5c, this interfacial region is quite clearly visible. The presence of an intermediate silicide segment has previously been seen in transition metal silicide NWs that were fabricated through the conventional silicidation process.18,44 In the case of nickel silicide NWs formed by conventional silicidation of Si NWs, it was observed that the NiSi2 phase would initially propagate through the Si NW.18 With continued diffusion from the Ni source, nickel enrichment occurred along the NW forming a Ni2Si secondary phase. Here, in this report, it is likely that a nickel silicide phase (such as NiSi or

NiSi2) initially grows along the NW, which is eventually transformed to Ni2Si as

Ni continues to diffuse from the substrate into the NW.

Figure 5.5. a) HAADF image of heterostructure NW with spectrum region for EELS analysis highlighted. b) Relative composition EELS profile of NW. c) EELS maps for O, C, Si and Ni corresponding to the HAADF image spectrum region.

149

In Figure 5.6, the dramatic drop-off in Ni concentration at the interface between the intermediate NiSi and Si segments is further analysed. As seen in the high magnification HAADF-STEM image in Figure 5.6a, the transition from NiSi to between the Si and segments is visibly sharp. In the Si segment of the NW, the Si planes are lattice resolved, with the longitudinal twin defect clearly visible.

However, the lattice fringing from the NiSi segment is not as obvious. This suggests that this intermediate NiSi segment could be slightly disordered. In the

EELS maps in Figure 5.6b, the sharpness of the interface is confirmed with the pure Si planes stopping within one atomic layer spacing.

Figure 5.6. a) High magnification HAADF image of heterostructure NW interface with spectrum region for EELS analysis highlighted.b) EELS maps for O, C, Si and Ni corresponding to the HAADF image spectrum region.

150

Figure 5.7a shows a HAADF-STEM image of another heterostructure NW with a diameter of approximately 10 nm that was characterised using aberration corrected STEM with atomic resolved EELS capabilities. Again, it was seen that there was an intermediate NixSi segment in this NW of approximately 10 nm in length through the use of an EELS analysis in Figure 5.7b. Interestingly, the relative ratio of Ni to Si in the initial NixSi segment is slightly less in this NW (i.e. x < 2). It is known that quantification analysis using EELS has a roughly 10%

45 margin of error, so the initial NW segment is likely be Ni2Si. However, the possibility of the initial segment being a different nickel silicide phase cannot be dismissed.

Figure 5.7. a) HAADF image of heterostructure NW with spectrum region for EELS analysis highlighted. b) Relative composition EELS profile of NW. c) Overlaid EELS maps for O, Si and Ni corresponding to the HAADF image spectrum region.

151

The presence of this intermediate silicide region was also seen in heterostructure

NixSi/Si NWs with larger diameters. In Figure 5.8a, a low magnification TEM image of the interfacial region in a NW of approximately 40 nm in diameter is shown. The initial Ni2Si segment of this NW is slightly torturous; likely a result of the strain associated with the volume expansion in the silicidation process. In

Figure 5.8b, a higher magnification TEM image of the Ni2Si segment of the NW with the inset electron diffraction pattern indexed for orthorhombic Ni2Si. In

Figures 8c and 8d, higher magnification TEM images of the intermediate silicide segment are show that the interface of this intermediate silicide segment with the pure Si segment is quite defective and disordered. Interestingly in Figure 5.8d, it can be seen that the silicide growth front through the NW is not uniform. This could be due to the presence of longitudinal defects which offer channels for Ni diffusion through the NW as previously stated. Using EDX spot analysis in DF-

STEM shown in Figure 5.8e, the relative composition of this intermediate silicide region was determined to be 38% Ni and 62% Si, which would suggest that this intermediate phase was NiSi2. However, FFTs taken from this intermediate region could not be indexed for any nickel silicide phase. Figure 5.8f shows EDS elemental maps for Au, Ni and Si of this NW, showing that Ni is confined to the

NixSi segment of the NW.

152

Figure 5.8. a) Low magnification TEM image of heterostructure NW b) TEM image of top region of NW tilted to zone axis with electron diffraction pattern inset indexed for Ni2Si. c) TEM image of interfacial region with higher magnification TEM image shown in d). e) DF-STEM image of NW with EDS elemental maps in f).

153

In addition to axial heterostructure NWs, complete nickel silicide NWs were also observed. Figure 5.9a shows a low magnification TEM image of a fully silicided

NW also including the Au seed with a higher magnification TEM image of the highlighted area in Figure 5.9b. The electron diffraction patterns inset in Figure

5.9b are indexed for orthorhombic Ni2Si. Through DF-STEM imaging and EDS analysis of the NWs, shown in Figure 5.9c and 5.9d, it appears that the composition of Ni and Si is consistent along the length of the NW, suggesting the

NW is completely Ni2Si. It is worth noting that continued Ni diffusion from the substrate causing a further phase change in the NWs to Ni31Si12, as seen previously, is plausible.18,44 However, no such phase change was observed in any of the NWs analysed in this study.

154

Figure 5.9. a) Low magnification TEM image of nickel silicide NW with high magnification TEM image of the highlighted region shown in b).c) DF-STEM image of NW with EDS line scan overlaid. d) DF-STEM image of NW with EDS line scan overlaid.

155

5.5 Conclusion

In this report, a novel approach to the synthesis of nickel silicide NWs and axial heterostructure NixSi/Si NWs within a solution based system is outlined. Initially, the Au layer seeds Si NW growth on the Ni substrate from which nickel silicide segments are then formed by the in situ silicidation of the Si NWs through Ni diffusion from the growth substrate. The length of the nickel silicide segments varied depending on the nature of the contact between the as grown Si NWs and the Ni substrates, with both nickel silicide NWs and heterostructure NixSi/Si NWs being seen in the same reaction. Studies into the interfacial abruptness between the NixSi and Si segments in the heterostructure NWs were conducted using aberration corrected STEM with atomic resolved EELS capabilities. Interestingly, an intermediate NixSi segment was observed in these NWs. Nevertheless, the interface between this intermediate NixSi segment and pure Si segment was found to be atomically abrupt. Analysis of the fully silicided nickel silicide NWs was conducted and the primary crystal structure of the formed NWs was found to be orthorhombic Ni2Si. This report highlights the potential to synthesise both

NixSi/Si NWs and fully silicided nickel silicide NWs within a solution based technique without the need for lithographical processing. Further studies will focus on improving control of the length of the NixSi segments, by improving the adhesion of Au to Ni substrate and by altering the reaction times.

156

5.6 References

(1) Schmitt, A. L.; Jin, S. Selective patterned growth of silicide nanowires without the use of metal catalysts. Chemistry of materials 2007, 19, 126.

(2) Chen, L. J.; Wu, W.-W.; Hsu, H.; Chen, S.; Chueh, Y.; Chou, L.-J.; Lu,

K.; Tu, K. Metal Silicide Nanowires. ECS Transactions 2007, 11, 3.

(3) Reader, A.; Van Ommen, A.; Weijs, P.; Wolters, R.; Oostra, D. Transition metal silicides in silicon technology. Reports on Progress in Physics 1993, 56,

1397.

(4) Borisenko, V.; Berlin: 2000.

(5) Chen, L. J.; Engineers, I. o. E. Silicide Technology for Integrated Circuits;

Institution of Engineering and Technology, 2004.

(6) Chen, L. Metal silicides: An integral part of microelectronics. Synthesis

2004.

(7) Maszara, W. Fully silicided metal gates for high-performance CMOS technology: a review. Journal of the Electrochemical Society 2005, 152, G550.

(8) Lin, Y.-C.; Chen, Y.; Huang, Y. The growth and applications of silicides for nanoscale devices. Nanoscale 2012, 4, 1412.

(9) Decker, C. A.; Solanki, R.; Freeouf, J. L.; Carruthers, J. R.; Evans, D. R.

Directed growth of nickel silicide nanowires. Appl. Phys. Lett. 2004, 84, 1389.

(10) Wu, Y.; Xiang, J.; Yang, C.; Lu, W.; Lieber, C. M. Single-crystal metallic nanowires and metal/semiconductor nanowire heterostructures. Nature 2004, 430,

61.

(11) Song, Y.; Jin, S. Synthesis and properties of single-crystal β 3-Ni 3 Si nanowires. Applied physics letters 2007, 90, 173122.

157

(12) Jiang, Y.; Li, Z.; Li, B.; Zhang, J.; Niu, C. Ni 3 Si 2 nanowires grown in situ on Ni foam for high-performance supercapacitors. Journal of Power Sources

2016, 320, 13.

(13) Song, Y.; Schmitt, A. L.; Jin, S. Ultralong single-crystal metallic Ni2Si nanowires with low resistivity. Nano letters 2007, 7, 965.

(14) Chiu, C.-H.; Huang, C.-W.; Chen, J.-Y.; Huang, Y.-T.; Hu, J.-C.; Chen,

L.-T.; Hsin, C.-L.; Wu, W.-W. Copper silicide/silicon nanowire heterostructures: in situ TEM observation of growth behaviors and electron transport properties.

Nanoscale 2013, 5, 5086.

(15) Chou, Y.-C.; Wu, W.-W.; Cheng, S.-L.; Yoo, B.-Y.; Myung, N.; Chen, L.

J.; Tu, K. In-situ TEM observation of repeating events of nucleation in epitaxial growth of nano CoSi2 in nanowires of Si. Nano Letters 2008, 8, 2194.

(16) Lu, K.-C.; Wu, W.-W.; Wu, H.-W.; Tanner, C. M.; Chang, J. P.; Chen, L.

J.; Tu, K. In situ control of atomic-scale Si layer with huge strain in the nanoheterostructure NiSi/Si/NiSi through point contact reaction. Nano letters

2007, 7, 2389.

(17) Lin, Y.-C.; Lu, K.-C.; Wu, W.-W.; Bai, J.; Chen, L. J.; Tu, K.; Huang, Y.

Single crystalline PtSi nanowires, PtSi/Si/PtSi nanowire heterostructures, and nanodevices. Nano letters 2008, 8, 913.

(18) Tang, W.; Dayeh, S. A.; Picraux, S. T.; Huang, J. Y.; Tu, K.-N. Ultrashort

Channel Silicon Nanowire Transistors with Nickel Silicide Source/Drain

Contacts. Nano letters 2012, 12, 3979.

(19) Lin, Y.-C.; Chen, Y.; Shailos, A.; Huang, Y. Detection of spin polarized carrier in silicon nanowire with single crystal MnSi as magnetic contacts. Nano letters 2010, 10, 2281.

158

(20) Schmitt, A. L.; Higgins, J. M.; Szczech, J. R.; Jin, S. Synthesis and applications of metal silicide nanowires. J. Mater. Chem. 2010, 20, 223.

(21) Tuan, H. Y.; Lee, D. C.; Korgel, B. A. Nanocrystal‐Mediated

Crystallization of Silicon and Germanium Nanowires in Organic Solvents: The

Role of Catalysis and Solid‐Phase Seeding. Angewandte Chemie International

Edition 2006, 45, 5184.

(22) Yuan, F.-W.; Yang, H.-J.; Tuan, H.-Y. Seeded silicon nanowire growth catalyzed by commercially available bulk metals: broad selection of metal catalysts, superior field emission performance, and versatile nanowire/metal architectures. J. Mater. Chem. 2011, 21, 13793.

(23) Wen, C.-Y.; Reuter, M.; Tersoff, J.; Stach, E.; Ross, F. Structure, growth kinetics, and ledge flow during vapor− solid− solid growth of copper-catalyzed silicon nanowires. Nano letters 2009, 10, 514.

(24) Decker, C.; Solanki, R.; Freeouf, J.; Carruthers, J.; Evans, D. Directed growth of nickel silicide nanowires. Applied physics letters 2004, 84, 1389.

(25) Kim, C. J.; Kang, K.; Woo, Y. S.; Ryu, K. G.; Moon, H.; Kim, J. M.;

Zang, D. S.; Jo, M. H. Spontaneous chemical vapor growth of NiSi nanowires and their metallic properties. Advanced Materials 2007, 19, 3637.

(26) Liu, Z.; Zhang, H.; Wang, L.; Yang, D. Controlling the growth and field emission properties of silicide nanowire arrays by direct silicification of Ni foil.

Nanotechnology 2008, 19, 375602.

(27) Fan, X.; Zhang, H.; Du, N.; Yang, D. Phase-controlled synthesis of nickel silicide nanostructures. Materials Research Bulletin 2012, 47, 3797.

(28) Klementová, M.; Krabáč, L.; Brázda, P.; Palatinus, L.; Dřínek, V. Cu-Si nanoobjects prepared by CVD on Cu/Cu 5 Si-substrates using various precursors

159

(SiH 4, EtSiH 3, BuSiH 3) with added H 2 or air. Journal of Crystal Growth 2017,

465, 6.

(29) Sheehan, M.; Guo, Y.; Flynn, G.; Geaney, H.; Ryan, K. M. The selective synthesis of nickel germanide nanowires and nickel germanide seeded germanium nanowires within a solvent vapour growth system. CrystEngComm 2017, 19,

2072.

(30) Mullane, E.; Kennedy, T.; Geaney, H.; Dickinson, C.; Ryan, K. M.

Synthesis of Tin Catalyzed Silicon and Germanium Nanowires in a Solvent–

Vapor System and Optimization of the Seed/Nanowire Interface for Dual Lithium

Cycling. Chem. Mater. 2013, 25, 1816.

(31) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High density growth of indium seeded silicon nanowires in the vapor phase of a high boiling point solvent. Chemistry of Materials 2012, 24,

2204.

(32) Flynn, G.; Ramasse, Q. M.; Ryan, K. M. Solvent Vapor Growth of Axial

Heterostructure Nanowires with Multiple Alternating Segments of Silicon and

Germanium. Nano Lett. 2016, 16, 374.

(33) Geaney, H.; Mullane, E.; Ramasse, Q. M.; Ryan, K. M. Atomically abrupt silicon-germanium axial heterostructure nanowires synthesized in a solvent vapor growth system. Nano Lett. 2013, 13, 1675.

(34) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chem. Mater. 2012, 24, 4319.

(35) Yuan, F. W.; Wang, C. Y.; Li, G. A.; Chang, S. H.; Chu, L. W.; Chen, L.

J.; Tuan, H. Y. Solution-phase synthesis of single-crystal Cu3Si nanowire arrays

160 on diverse substrates with dual functions as high-performance field emitters and efficient anti-reflective layers. Nanoscale 2013, 5, 9875.

(36) Richards, B. T.; Gaskey, B.; Levin, B. D. A.; Whitham, K.; Muller, D.;

Hanrath, T. Direct growth of germanium and silicon nanowires on metal films. J.

Mater. Chem. C 2014, 2, 1869.

(37) Lu, K.-C.; Wu, W.-W.; Ouyang, H.; Lin, Y.-C.; Huang, Y.; Wang, C.-W.;

Wu, Z.-W.; Huang, C.-W.; Chen, L. J.; Tu, K.-N. The influence of surface oxide on the growth of metal/semiconductor nanowires. Nano letters 2011, 11, 2753.

(38) Barth, S.; Boland, J. J.; Holmes, J. D. Defect transfer from nanoparticles to nanowires. Nano letters 2011, 11, 1550.

(39) Nabbefeld, T.; Wiethoff, C.; Heringdorf, F.-J. M. z.; Hoegen, M. H.-v.

Silver induced faceting of Si(112). Applied Physics Letters 2010, 97, 041905.

(40) Biswas, S.; Singha, A.; Morris, M. A.; Holmes, J. D. Inherent control of growth, morphology, and defect formation in germanium nanowires. Nano letters

2012, 12, 5654.

(41) Barth, S.; Koleśnik, M. M.; Donegan, K.; Krstic, V.; Holmes, J. D.

Diameter-controlled solid-phase seeding of germanium nanowires: structural characterization and electrical transport properties. Chemistry of Materials 2011,

23, 3335.

(42) Dellas, N.; Liu, B.; Eichfeld, S.; Eichfeld, C.; Mayer, T.; Mohney, S.

Orientation dependence of nickel silicide formation in contacts to silicon nanowires. Journal of applied physics 2009, 105, 094309.

(43) Woodruff, S. M.; Dellas, N. S.; Liu, B. Z.; Eichfeld, S. M.; Mayer, T. S.;

Redwing, J. M.; Mohney, S. E. Nickel and nickel silicide Schottky barrier contacts to n-type silicon nanowires. Journal of Vacuum Science & Technology B:

161

Microelectronics and Nanometer Structures Processing, Measurement, and

Phenomena 2008, 26, 1592.

(44) Lin, Y.-C.; Chen, Y.; Xu, D.; Huang, Y. Growth of nickel silicides in Si and Si/SiOx core/shell nanowires. Nano letters 2010, 10, 4721.

(45) Leapman, R. In Transmission Electron Energy Loss Spectrometry in

Materials Science and The EELS Atlas; Wiley-VCH Verlag GmbH & Co. KGaA:

2005, p 49.

162

Chapter 6: Amorphous Silicon coated Copper Silicide Nanowire

Arrays for Lithium-ion Battery Anodes

6.1 Abstract

Cu15Si4/amorphous Si core/shell nanowires were synthesized by deposition of amorphous Si coatings onto preformed Cu15Si4 nanowires. Electrochemical testing of the core/shell nanowires revealed markedly improved capacity retention values (90% after 100 cycles at a C/5 rate) compared to those seen for amorphous

Si deposited on planar Cu substrates. This improvement in performance can be attributed to the Cu15Si4 nanowires acting as a conductive backbone which limits pulverization and delamination of the amorphous Si during lithiation/delithiation.

Additionally, the NWs serve to anchor amorphous-Si present on the remainder of the electrode, preventing delamination noted for planar samples. Post-mortem analysis of the electrodes was conducted using scanning and transmission electron microscopy to investigate changes in morphology occurring during cycling, revealing that the Cu15Si4 retained crystallinity even after extended cycling. This report highlights the benefit of using conductive, electrochemically inactive nanowire backbones for improved performance of highly promising Li-ion battery anode materials.

6.2 Introduction

With the ever-increasing power demands of mobile technologies, extensive research has focused on the development of improved materials for next- generation Li-ion batteries (LIBs).1,2 Within this research, significant focus has

163 centred on the development of alternative anode materials in place of graphitic carbon, which is the current material of choice in commercial LIBs, due to the relatively low theoretical capacity of 372 mA h/g. Recently, alternative materials such as TiO2, NiCo2O4 and SnO2 have been used as alternatives to graphitic carbon in LIB anodes.3-5 Si and Ge are particularly attractive alternative materials to graphitic carbon due to their high theoretical capacities (3579 mA h/g and

1384 mA h/g for Si and Ge, respectively).6-8 However, despite the much higher capacities of both Si and Ge, there are difficulties in implementing these as alternative anode materials. The Li alloying processes for both Si and Ge lead to very large volume expansions (~ 400 % for Si) which can lead to irreversible capacity losses via detachment of the active material from the current collector and active material pulverisation.9 Numerous approaches have been developed to integrate Si and Ge as anode materials for Li-ion applications. In particular, the use of nanoscale morphologies, such as nanoparticles (NPs),10,11 nanowires

(NWs),12-14 and thin-films,15,16 has been shown to somewhat alleviate the detrimental effects associated with volume expansion during lithiation. Recently,

Cui and co-workers have used a pomegranate like structure composed of Si NPs enclosed in a conductive carbon shell to combat the detrimental effects of volume expansion and pulverisation during cycling.17

In addition to using these nano-morphologies, it has been found that the surface roughness of the current collector plays an important role in the performance of the active material in LIB anodes.18,19 This has led to a number of reports attempting to modify the current collector (e.g. by chemically etching, mechanical roughening or texturing approaches) to enhance the adhesion of the active material.19-21 One approach to modify the current collector is to use core/shell NW

164 architectures, with a conductive core (which does not participate in lithiation/delithiation) attached directly to the current collector with an electrochemically active shell (e.g. amorphous Si (a-Si)). Wang et al. used this approach to design a novel anode that was able to retain 90% of its original discharge capacity after 100 cycles utilising a carbon nanotube core.22 Similarly, copper/a-Si core/shell nanocable arrays have also recently been prepared, displaying impressive capacity retention.23

Transition metal silicide NWs directly grown from their corresponding pure transition metal foil are particularly suited for this core-shell NW architecture approach due to their relative simple synthesis and their inherently high conductivities.24,25 He and co-workers used a chemical vapor deposition (CVD) setup to synthesis Ni3Si2/a-Si core/shell NWs using a two-step approach that displayed a good capacity retention of 71% after 50 cycles.26 Pribat et al. used a similar approach to synthesize alumina coated NiSix/a-Si NWs that displayed a stable capacity of 1200 mA h/g for over 1200 cycles.27 Despite the widespread use of Cu as a current collector material in Li-ion applications, we are not aware of any reports detailing the use of Cu silicide NWs as a core material for core/shell

NW anodes.

In this study, the use of Cu15Si4/a-Si core/shell NW arrays for LIB applications was investigated. The electrochemical performance of the coated substrates was analysed and showed remarkably high capacity retention of approximately 90% after 100 cycles (1879 mA h/g). Analysis of the electrodes using Scanning

Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) before and after cycling showed transformation of the a-Si into a stable mesh-like structure anchored to the substrate by the presence of the Cu15Si4 NW cores.

165

6.3 Experimental Section

Substrate preparation:

Bulk Cu foil was purchased from Goodfellow Ltd. with a 0.25 mm thickness and

99.9% purity. The Cu was cleaned with nitric acid (10% v/v) and rinsed repeatedly with deionized water and then dried before introduction into the reaction setup.

Growth of Cu15Si4 NW arrays:

Reactions were carried out in custom-made Pyrex, round bottomed flasks containing 7 ml of Squalane (99% Aldrich). The growth substrate was placed vertically in the round bottomed flask which was attached to a Schlenk line setup via a water condenser. This was then ramped to a temperature of 125 °C using a three zone furnace. A vacuum, of at least 100 mTorr, was applied for 60 min to remove moisture from the system. Following this, the system was purged with Ar.

The flask was then ramped to the reaction temperature. Reactions were conducted at 465 °C. Upon reaching the reaction temperature, 0.75 ml phenylsilane ((PS),

98% Fluorochem) was injected through a septum cap. In a typical reaction, NW growth was allowed to proceed for 30 min. To terminate the reaction, the furnace was turned off and the setup was allowed to cool to room temperature before extracting the NW coated substrates. After the reaction, the NW covered substrates were removed from the reaction flask. The substrates were rinsed with toluene to remove residual high boiling point solvent (HBS) and dried under a N2 line prior to characterization.

166

Deposition of a-Si layers:

2 Deposition of known masses of a-Si (~0.04 mg/cm ) onto Cu15Si4 and Cu substrates was carried out by Meyer Burger Technology, using a commercially sensitive low temperature CVD setup.

Material Characterisation:

NWs were removed from the growth substrates through the use of a sonic bath.

TEM analysis was conducted using a 200 kV JEOL JEM-2100F field emission microscope equipped with a Gatan Ultrascan CCD camera and EDAX Genesis

EDS detector.

Electrochemical Characterisation:

The electrochemical performance of the substrates was evaluated by assembling two electrode Swagelok type cells in an Ar-filled glovebox. The cells consisted of the a-Si coated Cu15Si4 NWs substrates as the working electrode, Li foil as the counter and reference electrodes, a porous polypropylene Celgard separator, and an electrolyte solution of 1 M LiPF6 in ethylene carbonate/dimethyl carbonate

(EC/DMC) (1:1 v/v) + 3 wt % vinylene carbonate (VC). The cycling measurements were carried out galvanostatically using a Biologic MPG-2 multichannel potentiostat in the potential range of 0.01−1.0 V versus Li/Li+. For comparison, the electrochemical performance of a-Si coated Cu substrates was also investigated.

167

6.4 Results and Discussion

Figure 6.1: Schematic detailing growth of Cu15Si4/a-Si NWs using a two-step process.

Figure 6.1 shows the two step process used for the synthesis of Cu15Si4/a-Si core shell NWs. Initially Cu15Si4 NWs were grown by the delivery of PS to Cu foil in the solvent vapour growth (SVG) system, typically used for Si and Ge NW

28-31 growth. PS decomposes through phenyl redistribution to produce SiH4 gas which reacts with the Cu foil to initially form a rough copper silicide crystallite layer.31,32 As the reaction proceeds, the crystal habit changes from bulk crystallites to NWs. The Cu15Si4 NW arrays were then coated with a-Si using a CVD setup.

168

Figure 6.2: Analysis of Cu15Si4 NWs: a) low magnification SEM image of Cu15Si4 NW arrays. b) High magnification SEM image of Cu15Si4 NWs. c) Low magnification TEM image of Cu15Si4 NW with higher magnifications of highlighted area in d). e) SAED pattern of d) indexed for Cu15Si4. f) DF-STEM image of NW with corresponding EDS line scan inset.

169

Figure 6.2a shows a low magnification SEM image of the dense growth of Cu15Si4

NWs after a typical NW synthesis. It can be seen that the NWs can grow to approximately 5 µm in length with no tapering observed along the length of the

NWs. The higher magnification SEM image shown in Figure 6.2b reveals a tight diameter distribution with an average diameter of approximately 110 nm. The growth of NWs from the underlying crystallite layer is also evident.

Figure 6.2c shows a low magnification TEM image of an individual Cu15Si4 NW with a diameter of 94 nm. A high-resolution transmission electron microscopy

(HRTEM) image taken from the marked area is shown in Figure 6.2d. The corresponding electron diffraction pattern shown in Figure 6.2e is indexed for the cubic Cu15Si4 (space group I4̅3d; a = 9.7040 Å) along the [111] zone axis. Figure

6.2f shows a DF-STEM image of a Cu15Si4 NW with an EDS line profile of the

NW inset confirming the presence of both Cu and Si along the length of the NW.

Analysis of the NWs after the deposition of a-Si was carried out to investigate the morphology of the core/shell NW structures. Figure 6.3a shows a low magnification SEM image of a substrate after deposition, where the underlying

NWs can still clearly be seen. In the high magnification SEM image in Figure

6.3b, it can be seen that the diameter of the NW has increased due to the deposition. Figure 6.3c shows a high magnification TEM image of a NW after the coating process with a low magnification inset. Due to the relatively thin coating here, approximately 10 nm, it is possible to see the lattice fringing of the crystalline Cu15Si4 NW core, indicating that they retain crystallinity after the CVD process. A low magnification TEM image of a NW with a much thicker coating is shown in Figure 6.3d, with a corresponding DF-STEM image in Figure 6.3e. The

170 coating is approximately 100 nm in thickness on both sides of the NW shown. It is worth noting that the coating process coats the surfaces which are most exposed, leading to slightly uneven coating for some of the NWs. Figure 6.3f shows an

EDS line profile across the NW showing that the coating is composed of Si and that the Cu signal is confined to the NW.

Figure 6.3: Analysis of a-Si/Cu15Si4 NWs: a) low magnification SEM image of NW array. b) High magnification SEM image of NWs displaying increased diameter size. c) High magnification TEM image of a-Si coated Cu15Si4 NW d). TEM image showing a-Si coated Cu15Si4 NW with corresponding DF- STEM image and EDS line profile in e) and f) respectively.

171

To test the performance of the Cu15Si4/a-Si core shell NW arrays as LIB anodes, samples were analysed by galvanostatic cycling in a two electrode Swagelok cell

+ cycled in the voltage range of 0.01 V to 1.00 V versus Li/Li in a 1 M LiPF6 in

EC/DMC (1:1 v/v) + 3 wt % VC electrolyte. Figure 6.4a shows the first charge and discharge profiles for a Cu15Si4/a-Si core shell NW array and an as-grown

Cu15Si4 array (with no coating). The initial charge capacity for the Cu15Si4/a-Si is much higher than for the as-grown Cu15Si4 array due to the presence of the a-Si coating. More importantly, the first discharge capacity is greatly enhanced for the

Cu15Si4/a-Si core shell NWs suggesting that the small charge capacity seen for the

Cu15Si4 NWs is irreversible. This point is further emphasized by examination of the differential capacity plots of the 1st cycles (Figure 6.4b). The Li alloying process for the Cu15Si4/a-Si NWs to form a-LixSi alloy (0.3 – 0.18 V) is clearly evident due a peak in ~ 0.2 V and corresponds to the plateau seen in the 1st charge cycle in Figure 6.4a. However, for the uncoated Cu15Si4 sample, this Li alloying process is almost negligible. A small peak is present in Figure 6.3b which may represent the alloying of Li and Si. It should be noted that the Cu15Si4 NWs have

31 an amorphous SiO2 coating after synthesis. SiO2 is known to be a Li active material which could explain the small sharp peak at ~ 0.2V in Figure 6.4b.33,34

The lack of any noticeable plateaus or peaks on the first discharge for the as- grown Cu15Si4 arrays suggests that the contribution of the Cu15Si4 NWs to subsequent cycles is negligible.

172

Figure 6.4: Analysis of Cu15Si4 NWs vs. Cu15Si4/a-Si NWs as LIB anodes. a) Voltage profiles of 1st cycle for both coated and uncoated substrates. b) Differential capacity plots for the first cycle of both coated and uncoated substrates.

Having determined that the underlying Cu15Si4 NW array does not contribute significantly to the specific capacity for the Cu15Si4/a-Si core shell NW arrays, extended charge/discharge cycling was investigated. Figure 6.5a shows specific voltage profiles for the Cu15Si4/a-Si NW electrodes cycled for 50 cycles with characteristic plateaus for the lithiation (~ 0.25 V and ~ 0.1 V) and delithiation

(~0.3 V and ~ 0.5 V) of Si. The long sloping profile of the first cycle also indicates electrolyte decomposition and the formation of the solid electrolyte

173 interface (SEI) layer. After the first cycle, the stability of the anode is remarkable as evident from the similarity of the voltage profiles and negligible capacity fade.

Figure 6.5b displays the corresponding differential capacity plots for the voltage profiles presented in Figure 6.5a. The cathodic branch has two obvious peaks at

0.3-0.2 V and 0.1-0.05 V with each peak corresponding to the formation of a more

Li rich LixSi phase. Two obvious peaks are present in the anodic branch at 0.25-

0.3 V and 0.4-0.5 V corresponding to LixSi delithiation processes.

Figure 6.5: Analysis of a-Si/Cu15Si4 LIB anodes. a) Voltage profiles of cycles 1, 2, 5, 10, 20 and 50. The profiles show characteristic plateaus corresponding to the lithiation and delithiation of Si. b) Differential capacity plots of cycles 1, 2, 5, 10, 20 and 50.

174

Figure 6.6: Rate capability testing of the a-Si/Cu15Si4 NW electrodes cycled in the voltage range of 0.01 - 1.0 V at C/5, C/2, C, 2C, 5C and then back to C/5 for extended cycling.

The rate capability of the Cu15Si4/a-Si NW electrodes was determined by charging and discharging the material for 5 cycles at rates of C/5, C/2, C, 2C, and 5C

(Figure 6.6). For the various rates, the average discharge capacity observed over the 5 cycles was 2071, 1978, 1817, 1465 and 860 mA h/g, respectively. After the rate capability testing, the electrode was again cycled at C/5. Impressively, the electrode retained much of its initial capacity and after a total of 50 cycles, only a

2.5% loss in the discharge capacities was observed.

To account for the stable performances of the Cu15Si4/a-Si NW electrodes, SEM analysis of the substrates was carried out after 100 cycles at a C/5 rate. As can be seen in the SEM image in Figure 6.7, the morphology of the underlying NW array is still evident. This provides further evidence that the Cu15Si4 NWs do not

175 actively participate in lithiation/delithiation during cycling, as it has been previously shown that Li-ion active NWs lose both their crystallinity and morphology during cycling. Interestingly, the a-Si appears to have formed a mesh-like network on the surface of the electrode with Cu15Si4 NWs penetrating this network as can be seen in Figure 6.7b-d. This is reminiscent of the formation of a stable network seen when crystalline Ge NWs and similar analogues are used as binder free anodes.12,14 The restructuring of the active material, into this mesh, is most likely due to the pore formation upon delithiation and lithium assisted electrochemical welding.35 During this restructuring, it appears that the NWs offer anchor points for the active material which are not present in planar copper current collectors. The presence of these interpenetrating Cu15Si4 NWs within the mesh appears to help the cycling of the a-Si by acting as an electrochemically inactive (yet electrically conductive) network which aids in anchoring the active material to the electrode. It is also clear that there has been significant expansion of the active material when contrasting the SEM image in Figure 6.7d with the as- prepared Cu15Si4/a-Si core shell NW arrays in Figure 6.3a. The free space between NWs in the array acts as a buffer zone into which the active material can expand upon lithiation while still retaining electrical contact with the current collector substrate.

176

Figure 6.7: SEM images of the a-Si/Cu15Si4 NW electrode after 100 cycles at a C/5 rate. Low magnification image of substrates showing a-Si mesh with straight NWs penetrating through it are shown in a) and b). High magnification images of NWs encompassed in a-Si mesh are shown in c) and d).

To further emphasize the beneficial nature of the Cu15Si4 NW arrays as anchoring points for the active a-Si, examination of planar a-Si coated Cu substrates was also carried out. In Figure 6.8, the electrochemical performance of planar a-Si coated Cu is compared with Cu15Si4/a-Si NWs at a C/5 rate. Whilst the a-Si coated Cu electrodes display high initial capacities (up to approximately 20 cycles), it can be clearly seen that capacity retention rapidly decays after this point, with the electrode retaining only 28% of its original discharge capacity after

100 cycles. In comparison, the Cu15Si4/a-Si NW electrodes have lower capacities initially (2087 mA h/g initial discharge capacity) but show a remarkable stability with discharge capacities of 1879 mA h/g after 100 cycles, which is 90% of the

177 original discharge capacity. The Cu15Si4/a-Si NW arrays also display an average

Coulombic efficiency of 98.6%.

Figure 6.8: Galvanostatic charge-discharge cycling and coulombic efficiency of a-Si coated samples. The electrodes were cycled at a C/5 rate in the potential range of 0.01 – 1.00 V.

Figure 6.9a shows a photograph of a typical a-Si coated planar Cu anode before and after cycling and displays a purple hue across the surface before Li insertion.

In contrast, after cycling the substrate presents a characteristic Cu colour suggesting significant delamination of the active material and loss of contact with the Cu current collector. In the tilted SEM images of the post-cycling anode shown in Figure 6.9b and Figure 6.9c, the a-Si on the rough Cu substrate can be seen. In a similar morphology to the cycled a-Si/Cu15Si4 NW electrodes, the a-Si again formed a mesh like structure. EDS elemental mapping of the SEM image shown in Figure 6.9d, confirmed that the mesh is composed of Si (Figure 6.9f) while the localization of Si signal to certain regions illustrates the incomplete

178 surface coverage on the Cu current collector (Figure 6.9e) due to substantial active material losses during cycling.

Figure 6.9: Analysis of a-Si on Cu electrode after 100 cycles at a C/5 rate. a) Photograph of substrates before and after cycling. The post-cycling substrate has a noticeable Cu colour which suggests the detachment of the a-Si during cycling. b) Tilted SEM image of post-cycling substrate highlighting Si mesh on Cu substrate c) Tilted SEM image of substrate showing loose attachment of a- Si to Cu. d) SEM image of the area analysed by EDS with elemental maps for Cu and Si shown in d) and e), respectively.

179

In addition to SEM analysis, TEM analysis of the a-Si coated Cu15Si4 NWs after cycling was carried out to examine the crystallinity of the NWs. Figure 6.10a shows a mesh containing both a-Si and Cu15Si4 NWs. In the higher magnification

TEM image inset in Figure 6.10a, it can be seen that the NWs maintain their morphology with only a slight surface roughening noted. This surface roughening could be due to the cycling of the thin silicon oxide coating on the NWs or the strain associated with the volume expansion during the lithiation of the a-Si. In

Figure 6.10b, it is also possible to see the lattice fringing of the Cu15Si4 NW core.

The electron diffraction pattern shown in Figure 6.10d is indexed for cubic

Cu15Si4 which shows that the NWs do not undergo any transformation during cycling. This retention of crystallinity is again a strong indication that the Cu15Si4

NWs do not contribute to the capacity of electrodes despite their key role in improving the electrochemical performance. A DF-STEM image of the post- cycling mesh is shown in Figure 6.10c with corresponding EDS elemental maps overlaid in Figure 6.10d, confirming that the NWs are composed of both Cu

(green) and Si (magenta) and that the network is primarily composed of Si. EDS line scans of the NWs shown in Figure 6.10e and Figure 6.10f again show the presence of Cu and Si within the NWs.

180

Figure 6.10: TEM Analysis of Cu15Si4/a-Si NW electrode after 100 cycles at a C/5 rate. a) Low magnification image showing the mesh comprised of amorphous Si and Cu15Si4 NWs with higher magnification image shown inset of 2 NWs b) High magnification TEM image showing lattice fringing of NW highlighted in inset a) with corresponding ED inset indexed for Cu15Si4. c) DF- STEM image showing the mesh comprised of amorphous Si and Cu15Si4 NWs. d) EDS elemental maps for Cu (green) and Si (magenta) overlaid. e) DF-STEM image of 2 NWs with corresponding EDS line scans shown in f).

181

6.5 Conclusion

In conclusion, we have detailed the growth of Cu15Si4/a-Si core/shell NWs through a two-step process. Initially, Cu15Si NWs were grown directly on Cu foil in the SVG system before the deposition of a-Si using a low-temperature CVD setup. The electrochemical performance of the coated substrates was analysed and showed excellent capacity retention of approximately 90% after 100 cycles. Even after cycling at higher C rates, the Cu15Si4/a-Si NW electrodes displayed good capacity retention. SEM and TEM analysis of the electrodes confirmed that this improved capacity retention was due to the Cu15Si4 NWs providing anchor points for the a-Si to the current collector with a marked improvement compared to a-Si on planar Cu current collectors. This report illustrates those electrochemically inactive transition metal silicides NWs can be an excellent core material for high performance core/shell Li-ion battery anode materials.

182

6.6 References

(1) Kennedy, T.; Brandon, M.; Ryan, K. M. Advances in the Application of

Silicon and Germanium Nanowires for High-Performance Lithium-Ion Batteries.

Advanced Materials 2016, 28, 5696.

(2) Su, X.; Wu, Q.; Li, J.; Xiao, X.; Lott, A.; Lu, W.; Sheldon, B. W.; Wu, J.

Silicon-Based Nanomaterials for Lithium-Ion Batteries: A Review. Advanced

Energy Materials 2014, 4, n/a.

(3) Park, M. S.; Wang, G. X.; Kang, Y. M.; Wexler, D.; Dou, S. X.; Liu, H. K.

Preparation and Electrochemical Properties of SnO2 Nanowires for Application in

Lithium‐Ion Batteries. Angewandte Chemie 2007, 119, 764.

(4) Zhang, Q.; Chen, H.; Wang, J.; Xu, D.; Li, X.; Yang, Y.; Zhang, K.

Growth of Hierarchical 3D Mesoporous NiSix/NiCo2O4 Core/Shell

Heterostructures on Nickel Foam for Lithium‐Ion Batteries. ChemSusChem 2014,

7, 2325.

(5) McNulty, D.; Carroll, E.; O'Dwyer, C. Rutile TiO2 Inverse Opal Anodes for Li‐Ion Batteries with Long Cycle Life, High‐Rate Capability, and High

Structural Stability. Advanced Energy Materials 2017.

(6) Li, J.; Dahn, J. R. An In Situ X-Ray Diffraction Study of the Reaction of

Li with Crystalline Si. Journal of The Electrochemical Society 2007, 154, A156.

(7) Baggetto, L.; Notten, P. H. L. Lithium-Ion (De)Insertion Reaction of

Germanium Thin-Film Electrodes: An Electrochemical and In Situ XRD Study.

Journal of The Electrochemical Society 2009, 156, A169.

(8) Zhang, W.-J. A review of the electrochemical performance of alloy anodes for lithium-ion batteries. Journal of Power Sources 2011, 196, 13.

183

(9) Chan, C. K.; Peng, H.; Liu, G.; McIlwrath, K.; Zhang, X. F.; Huggins, R.

A.; Cui, Y. High-performance lithium battery anodes using silicon nanowires. Nat

Nano 2008, 3, 31.

(10) Ryu, J. H.; Kim, J. W.; Sung, Y.-E.; Oh, S. M. Failure Modes of Silicon

Powder Negative Electrode in Lithium Secondary Batteries. Electrochemical and

Solid-State Letters 2004, 7, A306.

(11) Kim, H.; Seo, M.; Park, M.-H.; Cho, J. A Critical Size of Silicon Nano-

Anodes for Lithium Rechargeable Batteries. Angewandte Chemie International

Edition 2010, 49, 2146.

(12) Kennedy, T.; Mullane, E.; Geaney, H.; Osiak, M.; O’Dwyer, C.; Ryan, K.

M. High-Performance Germanium Nanowire-Based Lithium-Ion Battery Anodes

Extending over 1000 Cycles Through in Situ Formation of a Continuous Porous

Network. Nano letters 2014, 14, 716.

(13) Mullane, E.; Kennedy, T.; Geaney, H.; Ryan, K. M. A Rapid, Solvent-Free

Protocol for the Synthesis of Germanium Nanowire Lithium-Ion Anodes with a

Long Cycle Life and High Rate Capability. ACS Applied Materials & Interfaces

2014, 6, 18800.

(14) Kennedy, T.; Bezuidenhout, M.; Palaniappan, K.; Stokes, K.; Brandon,

M.; Ryan, K. M. Nanowire Heterostructures Comprising Germanium Stems and

Silicon Branches as High-Capacity Li-Ion Anodes with Tunable Rate Capability.

ACS Nano 2015, 9, 7456.

(15) Bourderau, S.; Brousse, T.; Schleich, D. M. Amorphous silicon as a possible anode material for Li-ion batteries. Journal of Power Sources 1999, 81–

82, 233.

184

(16) Graetz, J.; Ahn, C. C.; Yazami, R.; Fultz, B. Nanocrystalline and Thin

Film Germanium Electrodes with High Lithium Capacity and High Rate

Capabilities. Journal of The Electrochemical Society 2004, 151, A698.

(17) Liu, N.; Lu, Z.; Zhao, J.; McDowell, M. T.; Lee, H.-W.; Zhao, W.; Cui, Y.

A pomegranate-inspired nanoscale design for large-volume-change lithium battery anodes. Nature nanotechnology 2014, 9, 187.

(18) Lee, K.-L.; Jung, J.-Y.; Lee, S.-W.; Moon, H.-S.; Park, J.-W.

Electrochemical characteristics of a-Si thin film anode for Li-ion rechargeable batteries. Journal of Power Sources 2004, 129, 270.

(19) Takamura, T.; Uehara, M.; Suzuki, J.; Sekine, K.; Tamura, K. High capacity and long cycle life silicon anode for Li-ion battery. Journal of Power

Sources 2006, 158, 1401.

(20) Uehara, M.; Suzuki, J.; Tamura, K.; Sekine, K.; Takamura, T. Thick vacuum deposited silicon films suitable for the anode of Li-ion battery. Journal of

Power Sources 2005, 146, 441.

(21) Zhang, Y. Q.; Xia, X. H.; Wang, X. L.; Mai, Y. J.; Shi, S. J.; Tang, Y. Y.;

Gu, C. G.; Tu, J. P. Three-dimensional porous nano-Ni supported silicon composite film for high-performance lithium-ion batteries. Journal of Power

Sources 2012, 213, 106.

(22) Fan, Y.; Zhang, Q.; Xiao, Q.; Wang, X.; Huang, K. High performance lithium ion battery anodes based on carbon nanotube–silicon core–shell nanowires with controlled morphology. Carbon 2013, 59, 264.

(23) Cao, F.-F.; Deng, J.-W.; Xin, S.; Ji, H.-X.; Schmidt, O. G.; Wan, L.-J.;

Guo, Y.-G. Cu-Si Nanocable Arrays as High-Rate Anode Materials for Lithium-

Ion Batteries. Advanced Materials 2011, 23, 4415.

185

(24) Schmitt, A. L.; Higgins, J. M.; Szczech, J. R.; Jin, S. Synthesis and applications of metal silicide nanowires. Journal of Materials Chemistry 2010, 20,

223.

(25) Wu, Y.; Xiang, J.; Yang, C.; Lu, W.; Lieber, C. M. Single-crystal metallic nanowires and metal/semiconductor nanowire heterostructures. Nature 2004, 430,

61.

(26) Li, F.; Yue, H.; Wang, P.; Yang, Z.; Wang, D.; Liu, D.; Qiao, L.; He, D.

Synthesis of core-shell architectures of silicon coated on controllable grown Ni- silicide nanostructures and their lithium-ion battery application. CrystEngComm

2013, 15, 7298.

(27) Nguyen, H. T.; Zamfir, M. R.; Duong, L. D.; Lee, Y. H.; Bondavalli, P.;

Pribat, D. Alumina-coated silicon-based nanowire arrays for high quality Li-ion battery anodes. Journal of Materials Chemistry 2012, 22, 24618.

(28) Flynn, G.; Ramasse, Q. M.; Ryan, K. M. Solvent Vapor Growth of Axial

Heterostructure Nanowires with Multiple Alternating Segments of Silicon and

Germanium. Nano Letters 2016, 16, 374.

(29) Geaney, H.; Kennedy, T.; Dickinson, C.; Mullane, E.; Singh, A.; Laffir,

F.; Ryan, K. M. High Density Growth of Indium seeded Silicon Nanowires in the

Vapor phase of a High Boiling Point Solvent. Chemistry of Materials 2012, 24,

2204.

(30) Geaney, H.; Dickinson, C.; Barrett, C. A.; Ryan, K. M. High Density

Germanium Nanowire Growth Directly from Copper Foil by Self-Induced Solid

Seeding. Chemistry of Materials 2011, 23, 4838.

186

(31) Geaney, H.; Dickinson, C.; O’Dwyer, C.; Mullane, E.; Singh, A.; Ryan, K.

M. Growth of Crystalline Copper Silicide Nanowires in High Yield within a High

Boiling Point Solvent System. Chemistry of Materials 2012, 24, 4319.

(32) Lee, D. C.; Hanrath, T.; Korgel, B. A. The role of precursor- decomposition kinetics in silicon-nanowire synthesis in organic solvents. Angew

Chem Int Ed Engl 2005, 44, 3573.

(33) Xun, S.; Song, X.; Wang, L.; Grass, M. E.; Liu, Z.; Battaglia, V. S.; Liu,

G. The Effects of Native Oxide Surface Layer on the Electrochemical

Performance of Si Nanoparticle-Based Electrodes. Journal of The

Electrochemical Society 2011, 158, A1260.

(34) Guo, B.; Shu, J.; Wang, Z.; Yang, H.; Shi, L.; Liu, Y.; Chen, L.

Electrochemical reduction of nano-SiO2 in hard carbon as anode material for lithium ion batteries. Electrochemistry Communications 2008, 10, 1876.

(35) Karki, K.; Epstein, E.; Cho, J.-H.; Jia, Z.; Li, T.; Picraux, S. T.; Wang, C.;

Cumings, J. Lithium-Assisted Electrochemical Welding in Silicon Nanowire

Battery Electrodes. Nano letters 2012, 12, 1392.

187

Chapter 7: Conclusions and Recommendations for Future Work

7.1 Conclusions

Transition metal germanides and silicides are an interesting set of materials that are attracting considerable research interest, due to their potential applications in a wide variety of different areas. This interest has led to a great deal of research, particularly focusing on the development of synthetic approaches for transition metal germanide and silicide nanowires (NWs). However, many of the established synthesis approaches to-date have suffered from low throughputs, as they have focused on the transformation of Ge or Si NWs, or have required very high temperatures (often greater than 900 °C). In this thesis, the use of the low temperature, solution-based solvent vapour growth (SVG) system was investigated, which could serve as an alternative approach to the synthesis of transition metal germanide and silicide NWs.

In Chapter 3, the use of the SVG system to synthesise nickel germanide (NiGe)

NWs from bulk Ni foil was described. In this study, a Ge precursor was injected into the SVG system, containing a Ni growth substrate, at a reaction temperature of 440 °C. The Ge precursor initially reacted with the Ni substrate to form a rough nickel germanide layer across the foil, and NiGe NWs subsequently nucleated from the nickel germanide layer as the reaction progressed. Investigations into the effect of limiting the Ni film thickness were conducted, in which it was revealed that Ge NW growth occurred through the vapour-solid-solid (VSS) growth mechanism when Ni thin films were used.

188

In Chapter 4, an investigation into the effect of reaction temperature and the enhanced Si precursor reactivity on the formation of copper silicide NWs, within the SVG system, was described. By altering the reaction temperature, it was found that NW growth on the Cu substrate was highly dependent on the thermal decomposition of the Si precursor, phenylsilane. In particular, when the reducing agent, lithium borohyrdide (LiBH4), was injected into the SVG system, phenylsilane could be made more reactive and this enhanced reactivity allowed for the attainment of copper silicide (Cu15Si4) NWs at 350 °C, which was over

100 °C lower than previously reported. In addition, when LiBH4 was used in conjunction with phenylsilane at reaction temperatures greater than 400 °C, unique copper silicide nanostructures (NSs) were achieved. Through detailed transmission electron microscopy and electron diffraction investigations, these

NSs were found to crystallise in the Si-rich copper silicide phase, Cu3Si.

Chapter 5 describes a novel, one-step approach to the synthesis of both nickel silicide and axial heterostructure NixSi/Si NWs within the SVG system. In this approach, a thin film of Au was evaporated onto Ni foil, which initiated Si NW growth on the Ni substrate through the established solution-liquid-solid (SLS) mechanism. The Au-coated Ni foil was then introduced into the SVG system. The high temperatures used for Si NW growth allowed for the simultaneous diffusion of Ni into the Si NWs from the substrate and for in situ silicidation of the Si NWs to occur. Using this approach, it was possible to synthesise nickel silicide NWs and axial heterostructure NixSi/Si NWs for the first time using the solution-based,

SVG system.

189

In Chapter 6, the use of Cu15Si4 NW arrays as high surface area current collectors for lithium ion battery applications was outlined. After synthesising the Cu15Si14

NW arrays using the SVG system, an amorphous Si coating was deposited onto the NW arrays using a low temperature, chemical vapour deposition (CVD) setup.

The performance of the coated NW arrays in advanced lithium ion battery anode applications was evaluated, and was compared to amorphous Si coatings on planar

Cu substrates. In comparison to the planar Cu substrates, the capacity retention was remarkably improved, displaying capacity retention of 90% after 100 cycles compared to the capacity retention of 28% after 100 cycles for the planar substrates. Through post mortem analysis of the different substrates, it was found that the improved capacity retention of the Cu15Si4 NW arrays was due to the ability of the Cu15Si4 NWs to penetrate the amorphous Si mesh and to anchor the active Si material to the current collector.

190

7.2 Recommendations for Future Work

The synthesis of transition metal silicide and germanide NWs by the delivery of

Si or Ge to a transition metal substrate has only been shown for Ni and Cu substrates to-date, which may be due to the fast diffusion rates displayed by these metals in Si. However, it may be possible to synthesise other transition metal silicide and germanide NWs by using elevated reaction temperatures (> 500 °C).

In particular, the use of elevated temperatures would likely increase the diffusion rate of the transition metal, which appears to play a critical role in NW formation.

In the SVG system, the maximum temperature that can be used is approximately

480 °C. Above this temperature, thermal degradation of the high boiling point solvent (i.e. squalane) takes place. Nevertheless, a number of follow on studies into the synthesis of transition metal silicide and germanide NWs within the SVG system are still possible.

7.2.1 FIB sectioning to study NW formation

As discussed in Chapter 3, the Ge precursor initially reacts to form a rough nickel germanide layer on the surface of the Ni foil before NW growth. To further study the mechanism behind NW growth, a focussed ion beam (FIB) cut-out of the sample could be analysed using transmission electron microscopy (TEM). Such a study would provide important information and insights into how NW growth proceeds in this system. The use of a FIB cut-out to examine NW formation could also be extended to examine the formation of the Cu15Si4 and Cu3Si NWs, as discussed in Chapter 4. In particular, further examination of the epitaxial NW growth from the copper silicide microcrystals in Chapter 4 would be interesting.

191

7.2.2 Alternative Si/Ge precursors

As mentioned in Chapter 5, the reactivity of the Si or Ge precursor plays an important role on the formation of NWs. An investigation into alternative Si and

Ge precursors would provide interesting insights into the synthesis of transition metal silicide and germanide NWs within the SVG system. For instance, trisilane

(Si3H8) is a much more reactive Si precursor than phenylsilane, thus the feasibility of this precursor for NW growth from Cu or Ni foil would be a new avenue to investigate. In addition, triphenylgermane (TPG) is a less reactive Ge precursor than diphenylgermane, which could potentially be used to synthesise Cu3Ge NWs from Cu foil and to uncover new routes to the formation of a variety of Cu3Ge nanostructures.

192

Appendix A: XRD reference patterns

NiGe

Name and formula

Reference code: 98-005-2964

Compound name: Nickel Germanide Chemical name: Nickel Germanide Common name: Nickel Germanide ICSD name: Nickel Germanide

Chemical formula: Ge1Ni1 Second chemical formula: NiGe

Crystallographic parameters

Crystal system: Orthorhombic Space group: P n m a Space group number: 62 a (Å): 5.3810 b (Å): 3.4280 c (Å): 5.8110 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Calculated density (g/cm^3): 8.13 Volume of cell (10^6 pm^3): 107.19 Z: 4.00

RIR: 3.14

Subfiles and quality

Subfiles: ICSD Pattern Inorganic Quality: Calculated (C)

Comments

ICSD collection code: 76623 Creation Date: 01/01/1970 Modification Date: 01/01/1970 Calculated Pattern Original Remarks: The structure has been assigned a PDF number: Structures: MnP Original ICSD space group: PBNM ICSD Collection Code: 76623

193

Original ICSD space group: PBNM

The coordinates are those given in the paper but the atomic distances

do not agree with those calculated during testing.The coordinates are

probably correct.

At least one temperature factor missing in the paper.

No R value given in the paper

X-ray diffraction (powder) Structure type: MnP

The structure has been assigned a PDF number: 7-297

Structure type: MnP

Recording date: 7/16/2001

ANX formula: AX

Z: 4

Calculated density: 8.13

Pearson code: oP8

Wyckoff code: c2

PDF code: 00-007-0297

TRANS b,c,a

REMARK Transformed from setting P b n m.

Publ. title: Neue Phasen vom Mn P (B-31)-Typ

References

Primary reference: Schubert, K. Pfisterer, H., Naturwissenschaften, 37, 112, (1950)

Peak list

No. h k l d [A] 2Theta[deg] I [%] 1 1 0 1 3.94822 22.501 0.3 2 0 1 1 2.95254 30.246 6.7 3 0 0 2 2.90550 30.748 0.3 4 2 0 0 2.69050 33.274 0.3

194

5 1 1 1 2.58849 34.625 100.0 6 1 0 2 2.55661 35.071 49.8 7 2 0 1 2.44150 36.782 6.1 8 2 1 0 2.11647 42.687 23.5 9 1 1 2 2.04941 44.155 55.0 10 2 1 1 1.98867 45.579 82.7 11 2 0 2 1.97411 45.934 28.8 12 1 0 3 1.82252 50.005 12.1 13 0 2 0 1.71400 53.413 58.7 14 2 1 2 1.71072 53.523 8.8 15 0 1 3 1.68640 54.358 33.9 16 1 1 3 1.60922 57.198 0.0 17 1 2 1 1.57224 58.673 7.7 18 2 0 3 1.57199 58.683 8.2 19 3 1 1 1.53296 60.330 3.7 20 3 0 2 1.52626 60.623 1.1 21 0 2 2 1.47627 62.905 0.0 22 0 0 4 1.45275 64.043 1.1 23 2 2 0 1.44558 64.398 0.0 24 2 1 3 1.42891 65.242 3.4 25 1 2 2 1.42366 65.513 13.8 26 2 2 1 1.40283 66.611 7.9 27 1 0 4 1.40253 66.627 7.8 28 3 1 2 1.39431 67.072 1.8 29 4 0 0 1.34525 69.864 2.2 30 3 0 3 1.31607 71.649 3.3 31 4 0 1 1.31059 71.995 2.6 32 1 1 4 1.29809 72.799 2.1 33 2 2 2 1.29424 73.050 11.2 34 2 0 4 1.27831 74.112 1.1 35 4 1 0 1.25228 75.921 5.3 36 1 2 3 1.24858 76.186 5.3 37 3 1 3 1.22864 77.652 0.0 38 4 1 1 1.22417 77.988 4.9 39 4 0 2 1.22075 78.249 1.6 40 3 2 1 1.21194 78.928 12.6 41 2 1 4 1.19774 80.051 1.4 42 2 2 3 1.15852 83.349 5.4 43 4 1 2 1.15001 84.107 3.1 44 3 2 2 1.13985 85.031 0.8 45 1 0 5 1.13601 85.387 4.6 46 3 0 4 1.12892 86.053 0.8 47 0 3 1 1.12120 86.792 0.1 48 0 2 4 1.10823 88.066 0.9 49 4 0 3 1.10492 88.398 3.9 50 0 1 5 1.10066 88.830 2.1 51 1 3 1 1.09762 89.142 5.4

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 GE1 Ge 0.58300 0.18800 0.25000 0.5000 1.0000 4* 2 NI1 Ni 0.19000 0.00500 0.25000 0.5000 1.0000 4*

195

Stick Pattern

196

Ni5Ge3

Name and formula

Reference code: 98-006-2505

Compound name: Nickel Germanide (5/3) Chemical name: Nickel Germanide (5/3) Common name: Nickel Germanide (5/3) ICSD name: Nickel Germanide (5/3)

Chemical formula: Ge3Ni5 Second chemical formula: Ni5Ge3

Crystallographic parameters

Crystal system: Monoclinic Space group: C 1 2 1 Space group number: 5 a (Å): 10.6610 b (Å): 6.7370 c (Å): 6.2640 Alpha (°): 90.0000 Beta (°): 120.1430 Gamma (°): 90.0000

Calculated density (g/cm^3): 8.73 Volume of cell (10^6 pm^3): 389.06 Z: 4.00

RIR: 3.20

Subfiles and quality

Subfiles: ICSD Pattern Inorganic Quality: Calculated (C)

Comments

ICSD collection code: 77958 Creation Date: 01/01/1970 Modification Date: 01/01/1970 Calculated Pattern Original Remarks: The structure has been assigned a PDF number: Original ICSD space group: C121 ICSD Collection Code: 77958

Original ICSD space group: C121

Cell in I121 setting: a'=9.264, beta'=95.64; stable up

197

to 663 K,

quenched in water from 793 K

The coordinates are those given in the paper but the atomic distances

do not agree with those calculated during testing.The coordinates are

probably correct.

The structure has been assigned a PDF number: 24- 249

Recording date: 4/1/2003

Modification date: 10/1/2006

ANX formula: A5X3

Z: 4

Calculated density: 8.73

R value: 0.18

Pearson code: mS32

Wyckoff code: c7 b a

PDF code: 00-024-0249

TRANS a-2c,b,c origin 0 .03900 0

Publ. title: Zur Struktur der Mischung Nickel- Germanium

References

Primary reference: Schubert, K. Goedecke, T. Ellner, M., Journal of the Less-Common Metals, 24, 23, (1971)

Peak list

No. h k l d [A] 2Theta[deg] I [%] 1 1 1 0 5.43946 16.282 7.5 2 0 0 1 5.41695 16.351 3.8 3 2 0 -1 4.94350 17.929 0.1 4 2 0 0 4.60968 19.239 2.5 5 1 1 -1 4.57548 19.384 2.2 6 1 1 1 3.37124 26.417 0.2 7 0 2 0 3.36850 26.438 0.2

198

8 3 1 -1 3.13623 28.436 0.7 9 2 0 -2 3.11684 28.617 1.6 10 2 0 1 2.87053 31.132 1.5 11 0 2 -1 2.86053 31.244 9.4 12 3 1 0 2.79597 31.984 10.7 13 2 2 -1 2.78369 32.129 8.8 14 1 1 -2 2.77463 32.237 11.9 15 2 2 0 2.71973 32.906 5.4 16 0 0 2 2.70847 33.046 5.2 17 4 0 -1 2.65484 33.734 0.1 18 3 1 -2 2.63913 33.941 1.9 19 4 0 -2 2.47175 36.316 11.0 20 4 0 0 2.30484 39.049 1.3 21 2 2 -2 2.28774 39.353 0.5 22 2 2 1 2.18483 41.289 3.7 23 1 3 0 2.18187 41.347 5.4 24 1 1 2 2.18044 41.376 5.3 25 3 1 1 2.12081 42.595 1.6 26 1 3 -1 2.11275 42.765 4.1 27 0 2 -2 2.11078 42.807 4.0 28 4 2 -1 2.08509 43.361 2.8 29 2 0 -3 2.07117 43.667 0.4 30 5 1 -1 2.00124 45.276 40.8 31 5 1 -2 1.99451 45.438 27.8 32 4 2 -2 1.99280 45.479 42.2 33 4 0 -3 1.98562 45.653 37.6 34 2 0 2 1.94692 46.613 85.6 35 1 3 1 1.94533 46.653 100.0 36 4 2 0 1.90218 47.777 2.7 37 3 3 -1 1.89685 47.919 1.8 38 1 1 -3 1.88921 48.125 1.6 39 4 0 1 1.81739 50.156 0.0 40 3 3 0 1.81315 50.281 0.1 41 1 3 -2 1.80729 50.455 0.4 42 0 0 3 1.80565 50.505 0.3 43 5 1 0 1.77846 51.332 2.0 44 6 0 -2 1.77181 51.539 0.4 45 3 3 -2 1.76822 51.652 1.5 46 5 1 -3 1.76440 51.772 1.3 47 6 0 -1 1.72269 53.122 0.1 48 4 2 -3 1.71055 53.529 1.0 49 2 2 2 1.68562 54.385 0.3 50 0 4 0 1.68425 54.433 0.2 51 6 0 -3 1.64783 55.739 0.0 52 3 1 2 1.61186 57.096 0.8 53 1 3 2 1.60831 57.234 0.6 54 4 2 1 1.59945 57.580 1.6 55 2 4 -1 1.59426 57.785 2.2 56 0 2 -3 1.59143 57.898 1.5 57 3 3 1 1.58393 58.198 0.1 58 2 4 0 1.58196 58.278 0.2 59 1 1 3 1.57948 58.378 0.4 60 6 2 -2 1.56812 58.842 0.5 61 4 0 -4 1.55842 59.245 0.4 62 6 0 0 1.53656 60.174 2.1 63 6 2 -1 1.53376 60.295 2.0 64 5 3 -1 1.53222 60.362 3.6 65 5 3 -2 1.52919 60.494 0.7 66 3 3 -3 1.52516 60.671 3.7 67 2 0 -4 1.52243 60.791 3.0

199

68 3 1 -4 1.52169 60.824 2.6 69 5 1 1 1.48940 62.288 0.4 70 7 1 -2 1.48528 62.480 1.5 71 2 4 -2 1.48175 62.645 0.8 72 6 2 -3 1.48021 62.718 1.7 73 5 1 -4 1.47566 62.933 1.5 74 2 4 1 1.45266 64.047 0.7 75 2 0 3 1.45182 64.089 0.7 76 7 1 -3 1.43900 64.729 0.1 77 4 0 2 1.43526 64.918 0.4 78 6 0 -4 1.43419 64.973 0.2 79 0 4 -2 1.43027 65.173 0.8 80 7 1 -1 1.42591 65.397 0.9 81 5 3 0 1.42505 65.441 0.8 82 4 4 -1 1.42219 65.589 0.1 83 5 3 -3 1.41779 65.819 0.6 84 1 1 -4 1.41616 65.904 0.5 85 4 2 -4 1.41439 65.997 0.9 86 6 2 0 1.39798 66.872 7.3 87 4 4 -2 1.39184 67.206 7.5 88 2 2 -4 1.38731 67.455 7.1 89 4 4 0 1.35986 69.007 0.3 90 0 0 4 1.35424 69.334 0.0 91 3 3 2 1.33493 70.485 0.7 92 2 2 3 1.33326 70.586 1.8 93 8 0 -2 1.32742 70.943 0.1 94 4 2 2 1.32040 71.378 1.1 95 6 2 -4 1.31956 71.430 0.7 96 1 5 -1 1.31699 71.591 1.4 97 8 0 -3 1.31643 71.626 1.4 98 6 0 1 1.31501 71.716 0.2 99 7 1 -4 1.31226 71.889 0.1 100 2 4 -3 1.30673 72.241 0.1 101 7 1 0 1.29258 73.159 0.2 102 4 4 -3 1.28442 73.700 0.0 103 3 3 -4 1.28234 73.840 0.3 104 3 1 3 1.27451 74.370 0.4 105 2 4 2 1.27377 74.421 0.4 106 1 5 1 1.27332 74.451 0.3 107 8 0 -1 1.26373 75.113 0.2 108 5 3 1 1.26284 75.175 0.4 109 7 3 -2 1.26032 75.352 0.3 110 3 5 -1 1.25943 75.414 0.7 111 0 2 -4 1.25650 75.621 0.6 112 5 3 -4 1.25443 75.768 0.0 113 4 0 -5 1.25194 75.945 0.0 114 5 1 2 1.23694 77.034 2.2 115 8 0 -4 1.23588 77.113 6.8 116 4 4 1 1.23534 77.152 4.9 117 8 2 -2 1.23499 77.178 3.1 118 3 5 0 1.23400 77.251 0.6 119 1 5 -2 1.23215 77.389 1.2 120 0 4 3 1.23163 77.428 1.5 121 1 1 4 1.23137 77.447 1.5 122 8 2 -3 1.22612 77.841 0.7 123 5 1 -5 1.22592 77.856 0.8 124 6 2 1 1.22497 77.928 1.0 125 7 3 -1 1.22344 78.044 0.2 126 6 4 -2 1.22073 78.251 0.2 127 6 0 -5 1.21967 78.331 0.3

200

128 3 5 -2 1.21955 78.340 0.4 129 1 3 -4 1.21726 78.516 0.6 130 3 1 -5 1.21424 78.749 0.1 131 6 4 -1 1.20431 79.527 0.0 132 2 0 -5 1.19699 80.111 0.0 133 8 2 -1 1.18321 81.238 0.6 134 6 4 -3 1.17786 81.685 0.6 135 4 2 -5 1.17351 82.053 0.6 136 4 0 3 1.16541 82.748 0.1 137 9 1 -3 1.16346 82.917 0.4 138 1 5 2 1.16317 82.942 0.5 139 8 2 -4 1.16025 83.197 0.0 140 7 1 -5 1.15729 83.458 0.1 141 9 1 -2 1.15521 83.642 0.3 142 3 5 1 1.15384 83.763 0.5 143 8 0 0 1.15242 83.890 0.1 144 2 0 4 1.15224 83.907 0.1 145 7 3 -4 1.14937 84.164 0.2 146 6 2 -5 1.14681 84.395 0.2 147 4 4 -4 1.14387 84.663 0.1 148 7 1 1 1.13712 85.284 2.2 149 7 3 0 1.13608 85.380 1.2 150 6 4 0 1.13514 85.468 2.4 151 5 5 -1 1.13339 85.632 2.9 152 5 5 -2 1.13216 85.747 0.3 153 3 5 -3 1.13052 85.901 3.2 154 2 4 -4 1.12941 86.006 2.9 155 1 1 -5 1.12850 86.092 2.8 156 2 2 -5 1.12790 86.150 1.5 157 3 3 3 1.12375 86.546 4.9 158 0 6 0 1.12283 86.634 3.4 159 9 1 -4 1.12062 86.847 0.0 160 8 0 -5 1.11765 87.136 0.0 161 6 0 2 1.11721 87.179 0.1 162 1 5 -3 1.11182 87.709 0.0 163 4 2 3 1.10136 88.759 0.1 164 2 4 3 1.09966 88.933 0.5 165 0 6 -1 1.09946 88.953 0.5 166 9 1 -1 1.09890 89.010 0.3 167 5 3 2 1.09775 89.129 0.0 168 2 6 -1 1.09494 89.418 0.1 169 1 3 4 1.09385 89.532 0.1 170 4 4 2 1.09241 89.681 0.1 171 6 4 -4 1.09194 89.731 0.1 172 2 6 0 1.09094 89.835 0.3 173 8 2 0 1.09037 89.894 0.6 174 2 2 4 1.09022 89.911 0.6 175 5 3 -5 1.09002 89.932 0.5

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 GE1 Ge 0.16700 0.35200 0.49300 1.2100 1.0000 4c 2 GE2 Ge 0.00000 0.82700 0.50000 1.1400 1.0000 2b 3 GE3 Ge 0.35200 0.98300 0.99600 1.9100 1.0000 4c

201

4 GE4 Ge 0.00000 0.03900 0.00000 0.6800 1.0000 2a 5 NI1 Ni 0.32300 0.32200 0.00900 1.3900 1.0000 4c 6 NI2 Ni 0.15700 0.99300 0.49600 3.3900 1.0000 4c 7 NI3 Ni 0.77100 0.17900 0.25500 2.2400 1.0000 4c 8 NI4 Ni 0.41600 0.15700 0.25900 3.0700 1.0000 4c 9 NI5 Ni 0.07000 0.17000 0.25200 1.2800 1.0000 4c

Stick Pattern

202

Ni2Ge

Name and formula

Reference code: 98-006-2926

Compound name: Nickel Germanide (2/1) Chemical name: Nickel Germanide (2/1) Common name: Nickel Germanide (2/1) ICSD name: Nickel Germanide (2/1)

Chemical formula: Ge1Ni2 Second chemical formula: Ni2Ge

Crystallographic parameters

Crystal system: Orthorhombic Space group: P n m a Space group number: 62 a (Å): 5.1130 b (Å): 3.8300 c (Å): 7.2640 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Calculated density (g/cm^3): 8.87 Volume of cell (10^6 pm^3): 142.25 Z: 4.00

RIR: 2.67

Subfiles and quality

Subfiles: ICSD Pattern Inorganic Quality: Calculated (C)

Comments

ICSD collection code: 53743 Creation Date: 01/01/1970 Modification Date: 01/01/1970 Calculated Pattern Original Remarks: The structure has been assigned a PDF number: Structures: Co2Si Original ICSD space group: PNMA ICSD Collection Code: 53743

Original ICSD space group: PNMA

The coordinates are those given in the paper but the

203

atomic distances

do not agree with those calculated during testing.The coordinates are

probably correct.

At least one temperature factor missing in the paper.

No R value given in the paper

X-ray diffraction (powder) Structure type: Co2Si

The structure has been assigned a PDF number: 24- 452

Structure type: Co2Si

Recording date: 4/1/2003

ANX formula: NO2

Z: 4

Calculated density: 8.87

Pearson code: oP12

Wyckoff code: c3

PDF code: 00-024-0452

TRANS Origin 0 1/2 1/2

Publ. title: Zur Struktur der Mischung Nickel- Germanium

References

Primary reference: Schubert, K. Goedecke, T. Ellner, M., Journal of the Less-Common Metals, 24, 23, (1971)

Peak list

No. h k l d [A] 2Theta[deg] I [%] 1 1 0 1 4.18109 21.233 0.0 2 0 0 2 3.63200 24.489 0.2 3 0 1 1 3.38792 26.284 3.4 4 1 0 2 2.96099 30.158 10.8 5 1 1 1 2.82420 31.656 2.7 6 2 0 0 2.55650 35.073 12.3 7 2 0 1 2.41151 37.256 0.2

204

8 1 1 2 2.34256 38.395 18.0 9 1 0 3 2.18835 41.219 17.5 10 2 1 0 2.12632 42.479 41.4 11 2 0 2 2.09055 43.242 28.9 12 0 1 3 2.04663 44.219 97.5 13 2 1 1 2.04069 44.354 100.0 14 0 2 0 1.91500 47.437 62.0 15 1 1 3 1.90007 47.833 33.2 16 2 1 2 1.83499 49.642 1.1 17 0 0 4 1.81600 50.197 2.9 18 2 0 3 1.75798 51.975 15.0 19 1 2 1 1.74107 52.518 0.0 20 1 0 4 1.71127 53.505 0.6 21 0 2 2 1.69396 54.096 0.0 22 3 0 1 1.65927 55.322 3.1 23 1 2 2 1.60801 57.245 3.7 24 2 1 3 1.59772 57.649 12.1 25 1 1 4 1.56240 59.079 2.5 26 3 0 2 1.54290 59.901 9.9 27 2 2 0 1.53268 60.342 4.9 28 3 1 1 1.52253 60.787 3.2 29 2 2 1 1.49966 61.814 0.1 30 2 0 4 1.48049 62.705 10.5 31 1 2 3 1.44112 64.622 7.5 32 3 1 2 1.43114 65.128 0.6 33 2 2 2 1.41210 66.118 12.4 34 1 0 5 1.39748 66.900 2.7 35 3 0 3 1.39370 67.105 0.1 36 2 1 4 1.38091 67.810 0.1 37 0 1 5 1.35836 69.094 0.2 38 0 2 4 1.31772 71.546 1.8 39 1 1 5 1.31282 71.854 2.7 40 3 1 3 1.30968 72.053 0.3 41 2 2 3 1.29504 72.998 9.0 42 4 0 0 1.27825 74.116 2.4 43 1 2 4 1.27602 74.267 0.3 44 2 0 5 1.26309 75.157 0.6 45 4 0 1 1.25891 75.451 5.5 46 0 3 1 1.25739 75.558 0.8 47 3 2 1 1.25402 75.797 1.8 48 3 0 4 1.24275 76.608 0.3 49 1 3 1 1.22101 78.228 0.3 50 4 1 0 1.21250 78.884 1.5 51 0 0 6 1.21067 79.027 1.2 52 4 0 2 1.20576 79.413 0.4 53 3 2 2 1.20146 79.753 7.9 54 2 1 5 1.19955 79.906 0.1 55 4 1 1 1.19596 80.194 0.0 56 3 1 4 1.18208 81.332 6.9 57 1 0 6 1.17809 81.666 7.5 58 1 3 2 1.17234 82.152 2.8 59 2 2 4 1.17128 82.243 8.9 60 4 1 2 1.15011 84.098 6.5 61 2 3 0 1.14217 84.818 4.0 62 4 0 3 1.13040 85.912 2.2 63 0 3 3 1.12931 86.016 13.6 64 1 2 5 1.12886 86.058 14.0 65 2 3 1 1.12831 86.111 13.5 66 3 2 3 1.12686 86.248 1.5 67 1 1 6 1.12603 86.328 5.7

205

68 3 0 5 1.10563 88.327 1.2 69 1 3 3 1.10273 88.620 4.1 70 2 0 6 1.09418 89.498 1.5 71 2 3 2 1.08956 89.980 0.2

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 GE1 Ge 0.27300 0.25000 0.38700 0.5000 1.0000 4c 2 NI1 Ni 0.96200 0.25000 0.79500 0.5000 1.0000 4c 3 NI2 Ni 0.32400 0.25000 0.06100 0.5000 1.0000 4c

Stick Pattern

206

Name and formula

Reference code: 01-071-4653

Mineral name: Nickel, syn Compound name: Nickel PDF index name: Nickel

Empirical formula: Ni Chemical formula: Ni

Crystallographic parameters

Crystal system: Cubic Space group: Fm-3m Space group number: 225 a (Å): 3.5400 b (Å): 3.5400 c (Å): 3.5400 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Volume of cell (10^6 pm^3): 44.36 Z: 4.00

RIR: 7.46

Status, subfiles and quality

Status: Alternate Pattern Subfiles: Alloy, metal or intermetalic Common Phase Forensic Hydrogen Storage Materials Inorganic Mineral Quality: Indexed (I)

Comments

ANX: N ICSD collection code: 53807 Creation Date: 01/09/2005 Modification Date: 01/09/2015 Cross-References: ICDD:04-010-6148, ICSD:53807 ANX: N Analysis: Ni1 Formula from original source: Ni

207

ICSD Collection Code: 53807 Calculated Pattern Original Remarks: SB 1, 68. Same data from 2nd ref. (Hull & Davey). Cell from 3rd ref (Bohlin): 3.53, from 4th ref (McKeehean): 3.517, m.p. 1726 K Sample Source or Locality: Specimen unknown Minor Warning: No e.s.d reported/abstracted on the cell dimension. No R factors reported/abstracted Wyckoff Sequence: a (FM3-M) Unit Cell Data Source: Powder Diffraction.

References

Primary reference: Calculated from ICSD using POWD-12++ Structure: Hull, A.W., Phys. Rev., 21, 402, (1923)

Peak list

No. h k l d [Å] 2Theta[deg] I [%] 1 1 1 1 2.04382 44.283 100.0 2 2 0 0 1.77000 51.596 42.1 3 2 2 0 1.25158 75.971 16.6 4 3 1 1 1.06735 92.389 14.9 5 2 2 2 1.02191 97.838 4.0 6 4 0 0 0.88500 121.009 1.7 7 3 3 1 0.81213 143.060 5.4

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 NI1 Ni 0.00000 0.00000 0.00000 0.5000 1.0000 4a

Stick Pattern

208

Name and formula

Reference code: 01-071-4636

Compound name: Germanium PDF index name: Germanium

Empirical formula: Ge Chemical formula: Ge

Crystallographic parameters

Crystal system: Cubic Space group: Fd-3m Space group number: 227 a (Å): 5.6400 b (Å): 5.6400 c (Å): 5.6400 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Volume of cell (10^6 pm^3): 179.41 Z: 8.00

RIR: 10.07

Status, subfiles and quality

Status: Alternate Pattern Subfiles: Alloy, metal or intermetalic Common Phase Inorganic Quality: Indexed (I)

Comments

ANX: N ICSD collection code: 53788 Creation Date: 01/09/2005 Modification Date: 01/09/2015 Cross-References: ICDD:04-016-2396, ICSD:53788 ANX: N Analysis: Ge1 Formula from original source: Ge ICSD Collection Code: 53788 Melting Point: M.p. 1210.6 K Calculated Pattern Original Remarks: Cell from 2nd ref- (Kolkmeijer): 5.62, from 3rd ref. (Goldschmidt): 5.658. SB 1, 54. Minor Warning: No

209

e.s.d reported/abstracted on the cell dimension. No R factors reported/abstracted. Wyckoff Sequence: a (FD3-MS). Unit Cell Data Source: Powder Diffraction.

References

Primary reference: Calculated from ICSD using POWD-12++ Structure: Hull, A.W., Skr. Nor. Vidensk.-Akad., Kl. 1: Mat.- Naturvidensk. Kl., 1927, 1, (1927)

Peak list

No. h k l d [Å] 2Theta[deg] I [%] 1 1 1 1 3.25626 27.367 100.0 2 2 2 0 1.99404 45.449 64.6 3 3 1 1 1.70052 53.870 35.4 4 2 2 2 1.62813 56.474 0.1 5 4 0 0 1.41000 66.229 8.1 6 3 3 1 1.29390 73.072 10.9 7 4 2 2 1.15126 83.994 12.8 8 5 1 1 1.08542 90.417 6.6 9 4 4 0 0.99702 101.176 3.5 10 5 3 1 0.95333 107.803 5.9 11 6 2 0 0.89176 119.491 4.9 12 5 3 3 0.86009 127.171 2.2 13 4 4 4 0.81406 142.254 1.3

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 GE1 Ge 0.12500 0.12500 0.12500 0.5000 1.0000 8a

Stick Pattern

210

Name and formula

Reference code: 01-070-3039

Mineral name: Copper, syn Compound name: Copper PDF index name: Copper

Empirical formula: Cu Chemical formula: Cu

Crystallographic parameters

Crystal system: Cubic Space group: Fm-3m Space group number: 225 a (Å): 3.6130 b (Å): 3.6130 c (Å): 3.6130 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Volume of cell (10^6 pm^3): 47.16 Z: 4.00

RIR: 8.85

Status, subfiles and quality

Status: Alternate Pattern Subfiles: Alloy, metal or intermetalic Common Phase Forensic Inorganic Mineral Quality: Indexed (I)

Comments

ANX: N ICSD collection code: 53247 Creation Date: 01/09/2004 Modification Date: 01/09/2015 Cross-References: ICDD:04-009-2090, ICSD:53247 ANX: N Analysis: Cu1 Formula from original source: Cu ICSD Collection Code: 53247

211

Calculated Pattern Original Remarks: Cell from 2nd ref (Kantola & Tokola): 3.61443 at 296 K, 3.6247 at 449 K, 3.6321 at 550 K, 3.6357 at 607 K, 3.6483 at 779 K Cell at 777 K: 3.644, at 1076 K: 3.667, at 1343 K: 3.692, m.p. 1356.6 K. Temperature of Data Collection: 293 K. Minor Warning: No e.s.d reported/abstracted on the cell dimension. No R factors reported/abstracted. Wyckoff Sequence: a (FM3-M). Unit Cell Data Source: Powder Diffraction.

References

Primary reference: Calculated from ICSD using POWD-12++ Structure: Suh, I.-K., Ohta, H., Waseda, Y., Ann. Acad. Sci. Fenn., Ser. A6, 223, 1, (1967)

Peak list

No. h k l d [Å] 2Theta[deg] I [%] 1 1 1 1 2.08597 43.342 100.0 2 2 0 0 1.80650 50.479 43.0 3 2 2 0 1.27739 74.174 17.5 4 3 1 1 1.08936 90.001 16.1 5 2 2 2 1.04298 95.217 4.4 6 4 0 0 0.90325 117.037 1.9 7 3 3 1 0.82888 136.660 5.9 8 4 2 0 0.80789 144.906 5.6

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 CU1 Cu 0.00000 0.00000 0.00000 0.1790 1.0000 4a

Stick Pattern

212

Cu15Si4

Name and formula

Reference code: 04-014-4307

Compound name: Copper Silicon PDF index name: Copper Silicon

Empirical formula: Cu15Si4 Chemical formula: Cu15Si4

Crystallographic parameters

Crystal system: Cubic Space group: I-43d Space group number: 220 a (Å): 9.7180 b (Å): 9.7180 c (Å): 9.7180 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Volume of cell (10^6 pm^3): 917.76 Z: 4.00

RIR: 5.72

Subfiles and quality

Subfiles: Alloy, metal or intermetalic Inorganic Quality: Indexed (I)

Comments

Creation Date: 01/09/2011 Cross-References: ICDD:04-014-4307, LPF:1216439 LPF Collection Code: 1216439 Sample Preparation: Compound Preparation: arc-melted, quenched. CRUCIBLE: water-cooled copper boat. ATMOSPHERE: purified argon. Sample annealed at 773 K, quenched. Minor Warning: No R factors reported/abstracted. No e.s.d reported/abstracted on the cell dimension. Unit Cell Data Source: Powder Diffraction.

References

Primary reference: Calculated from LPF using POWD-12++

213

Structure: Mattern N., Seyrich R., Wilde L., Baehtz C., Knapp M., Acker J., J. Alloys Compd., 429, 211,215, (2007)

Peak list

No. h k l d [Å] 2Theta[deg] I [%] 1 2 1 1 3.96736 22.391 0.1 2 2 2 0 3.43583 25.911 3.2 3 3 1 0 3.07310 29.033 3.7 4 3 2 1 2.59724 34.505 24.0 5 4 0 0 2.42950 36.971 3.7 6 4 2 0 2.17301 41.524 0.1 7 3 3 2 2.07188 43.652 100.0 8 4 2 2 1.98368 45.700 16.1 9 5 1 0 1.90586 47.679 56.6 10 5 2 1 1.77426 51.463 2.2 11 4 4 0 1.71792 53.281 0.8 12 5 3 0 1.66662 55.057 2.0 13 6 1 1 1.57647 58.500 5.8 14 6 2 0 1.53655 60.174 0.1 15 5 4 1 1.49952 61.821 3.4 16 6 3 1 1.43284 65.041 0.8 17 4 4 4 1.40267 66.620 0.2 18 7 1 0 1.37433 68.179 0.6 19 6 4 0 1.34764 69.723 6.2 20 5 5 2 1.32245 71.250 5.4 21 6 4 2 1.29862 72.764 1.0 22 7 3 0 1.27604 74.265 0.3 23 6 5 1 1.23419 77.237 12.1 24 8 0 0 1.21475 78.710 0.4 25 7 4 1 1.19620 80.175 0.1 26 8 2 0 1.17848 81.633 1.2 27 6 5 3 1.16152 83.086 1.4 28 8 2 2 1.14528 84.534 3.8 29 8 3 1 1.12969 85.980 9.9 30 7 5 2 1.10035 88.862 0.6 31 8 4 0 1.08651 90.302 0.2 32 9 1 0 1.07317 91.743 0.3 33 8 4 2 1.06032 93.184 3.3 34 9 2 1 1.04792 94.627 2.4 35 6 6 4 1.03594 96.074 1.9 36 9 3 0 1.02437 97.523 2.3 37 9 3 2 1.00234 100.439 0.5 38 8 4 4 0.99184 101.907 0.8 39 8 5 3 0.98167 103.383 0.5 40 8 6 0 0.97180 104.868 0.3 41 10 1 1 0.96223 106.363 1.1 42 10 2 0 0.95293 107.870 0.5 43 9 5 0 0.94390 109.389 0.1 44 9 5 2 0.92658 112.473 1.3 45 8 7 1 0.91017 115.627 0.6 46 10 4 0 0.90229 117.235 1.0 47 9 6 1 0.89461 118.867 0.9 48 10 4 2 0.88713 120.525 0.1 49 11 1 0 0.87983 122.211 0.1 50 11 2 1 0.86575 125.684 1.8 51 8 8 0 0.85896 127.477 0.1 52 11 3 0 0.85233 129.314 0.5 53 11 3 2 0.83951 133.144 2.3

214

54 10 6 0 0.83331 135.151 0.1 55 11 4 1 0.82725 137.231 2.1 56 10 6 2 0.82132 139.396 1.1 57 9 6 5 0.81552 141.662 0.1 58 12 0 0 0.80983 144.047 0.4 59 9 7 4 0.80427 146.577 3.1 60 12 2 0 0.79882 149.289 0.1

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 CU1 Cu 0.12000 0.16000 0.96000 0.5000 1.0000 48e 2 CU2 Cu 0.37500 0.00000 0.25000 0.5000 1.0000 12a 3 SI1 Si 0.95800 0.95800 0.95800 0.5000 1.0000 16c

Stick Pattern

215

Cu0.83Si0.17

Name and formula

Reference code: 04-014-4308

Compound name: Copper Silicon PDF index name: Copper Silicon

Empirical formula: Cu0.83Si0.17 Chemical formula: Cu0.83Si0.17

Crystallographic parameters

Crystal system: Cubic Space group: P4132 Space group number: 213 a (Å): 6.2220 b (Å): 6.2220 c (Å): 6.2220 Alpha (°): 90.0000 Beta (°): 90.0000 Gamma (°): 90.0000

Volume of cell (10^6 pm^3): 240.87 Z: 20.00

RIR: 5.93

Subfiles and quality

Subfiles: Alloy, metal or intermetalic Inorganic Quality: Indexed (I)

Comments

Creation Date: 01/09/2011 Cross-References: ICDD:04-014-4308, LPF:1216440 LPF Collection Code: 1216440 Sample Preparation: Compound Preparation: arc-melted, quenched. CRUCIBLE: water-cooled copper boat. ATMOSPHERE: purified argon. Minor Warning: No R factors reported/abstracted. No e.s.d reported/abstracted on the cell dimension. LPF Editor Comment: editor assigned an approximate value to the Cu/Si ratio of sites Cu,Si1 and Cu,Si2 based on the nominal composition. Unit Cell Data Source: Powder Diffraction.

216

References

Primary reference: Calculated from LPF using POWD-12++ Structure: Mattern N., Seyrich R., Wilde L., Baehtz C., Knapp M., Acker J., J. Alloys Compd., 429, 211,215, (2007)

Peak list

No. h k l d [Å] 2Theta[deg] I [%] 1 1 1 0 4.39962 20.167 1.8 2 1 1 1 3.59227 24.764 2.1 3 2 1 0 2.78256 32.142 5.2 4 2 1 1 2.54012 35.306 0.8 5 2 2 0 2.19981 40.995 1.0 6 2 2 1 2.07400 43.605 100.0 7 3 1 0 1.96757 46.095 54.7 8 3 1 1 1.87600 48.486 23.1 9 2 2 2 1.79614 50.791 0.1 10 3 2 0 1.72567 53.023 0.5 11 3 2 1 1.66290 55.191 7.0 12 4 0 0 1.55550 59.367 0.1 13 4 1 0 1.50906 61.388 0.8 14 4 1 1 1.46654 63.370 4.8 15 3 3 1 1.42742 65.319 0.2 16 4 2 0 1.39128 67.237 4.4 17 4 2 1 1.35775 69.129 0.2 18 3 3 2 1.32653 70.998 0.6 19 4 2 2 1.27006 74.675 1.1 20 4 3 0 1.24440 76.488 1.1 21 5 1 0 1.22023 78.288 14.9 22 5 1 1 1.19742 80.077 2.6 23 5 2 0 1.15540 83.625 11.4 24 5 2 1 1.13598 85.389 2.5 25 4 4 0 1.09990 88.908 0.2 26 5 2 2 1.08311 90.664 0.1 27 4 3 3 1.06706 92.421 0.8 28 5 3 1 1.05171 94.180 2.6 29 4 4 2 1.03700 95.944 3.4 30 6 1 0 1.02289 97.713 2.2 31 6 1 1 1.00934 99.489 2.1 32 6 2 0 0.98378 103.071 0.4 33 6 2 1 0.97171 104.881 1.0 34 5 4 1 0.96008 106.707 0.3 35 5 3 3 0.94885 108.550 0.7 36 6 2 2 0.93800 110.413 0.1 37 6 3 0 0.92752 112.299 1.5 38 6 3 1 0.91738 114.210 0.4 39 4 4 4 0.89807 118.125 0.1 40 6 3 2 0.88886 120.136 1.7 41 5 5 0 0.87992 122.188 1.2 42 5 5 1 0.87125 124.289 0.1 43 6 4 0 0.86284 126.442 0.1 44 6 4 1 0.85466 128.658 1.1 45 7 2 1 0.84671 130.944 1.2 46 6 4 2 0.83145 135.777 1.2 47 5 4 4 0.82412 138.356 1.2 48 7 3 0 0.81699 141.072 0.3 49 7 3 1 0.81003 143.959 2.6

217

Structure

No. Name Elem. X Y Z Biso sof Wyck. 1 CU1 Cu 0.12500 0.20600 0.45600 0.5000 0.8330 12d 2 SI1 Si 0.12500 0.20600 0.45600 0.5000 0.1670 12d 3 CU2 Cu 0.06100 0.06100 0.06100 0.5000 0.8330 8c 4 SI2 Si 0.06100 0.06100 0.06100 0.5000 0.1670 8c

Stick Pattern

218