Magic clusters on group IV surfaces 1

Magic clusters have been found to exist in a specific size and structure consisting of a “magic” number of atoms exhibiting closed shell structures leading to enhanced local stability [2

Chapter 1: Introduction

1.1 Introduction

Silicon (Si) and Silicon Carbide (SiC) are studied as the primary substrate materials in this work As the incumbent raw material in the micro-electronics industry, the study of Si surfaces remains important, especially with the growing interest in the use

of mono-dispersed particles in the fabrication of substrate supported nanostructures Si(111) is the substrate of choice because of the presence of faulted/unfaulted halves of the (7x7) reconstruction which possess available dangling bonds which in turn can function as adsorption sites in promoting the self assembly of ordered nano particles SiC

on the other hand is fast gaining recognition as a substrate material for electronic device applications in areas where common silicon technology finds limitations Due to its unique properties such as wide band gap, high thermal conductivity and high breakdown field strength, SiC based semiconductors electronic devices are presently being developed for use in high-temperature, high power and high radiation conditions under which conventional semiconductors cannot adequately perform [1-4] In particular, we focus on the 6H-SiC(0001) substrate which exhibits a wide range of surface reconstructions and is also one of the more commonly used substrates [5-6]

In this work, we focus on the formation of low dimensional structures on these substrates as the electronics industry has always been driven by the pursuit of faster and more efficient devices via reduction in device dimensions Examples of typical low dimensional structures used in electronic applications include Quantum Dot Light Emission, Single Electron Transistors and Quantum Computing Arrays [7-8] In particular, the advancement of silicon quantum dot nanostructures in solar photovoltaics

is of significant interest due to the urgency in developing clean energy alternatives [9-10] Amorphous silicon is currently used to fabricate solar cells, but has always been plagued

by material degradation as prolonged illumination introduces defects which decrease the solar cell efficiency [11-12] In this respect, nano-crystalline silicon is much more stable and has a broader absorption range in the visible part of the solar radiation [13-14] Furthermore the optical properties could be tuned by affecting the band gap via changing the size of the Si nano-crystals [15] Hence the challenge is to be able to manufacture semiconductor structures with greatly reduced sizes (in the regime of nanometers) on Si and SiC surfaces controllably and reproducibly

As the properties derived from these nano-structures are directly dependent on feature size, shape, distribution as well as interface sharpness, it would therefore be important to be able to re-produce these attributes with a high degree of uniformity However current top down methodologies such as optical lithography have encountered limitations in feature size reduction due to radiation wavelength [16-17] while bottom up techniques such as quantum dot formation via strain induced MBE thin film growth and SAMs (self-assembled mono-layers) have struggled to generate particles with narrow size

distribution [18-21] The key question therefore is whether we can deliberately prepare and assemble specific sizes of materials such as metals/semiconductors on supporting Si and SiC surfaces

It is well known that small atomic clusters exhibit unique size-dependent physical and chemical properties that differ from those of bulk materials [1] and maybe utilized to form the desired nano-structure architecture Hence there have been much experimental and theoretical interest on atomic clusters, however most of the work was focused on freestanding clusters [22] which do not address the issues in the formation of substrate supported nano-structures Interestingly the solution may lie in a unique phase of atomic clusters termed as “magic clusters” Magic clusters have been found to exist in a specific size and structure consisting of a “magic” number of atoms exhibiting closed shell structures leading to enhanced local stability [23] This in turn leads to a high degree of uniformity in physical, chemical and electronic properties being retained by magic cluster species Thus, the ability to fabricate substrate supported magic clusters, would address the challenges related to fabrication of atomic structures and presents a progressive platform for device miniaturization

In fact, a considerable number of work [24-29] on magic clusters of various material systems have been reported in a bid to create well ordered mono-dispersed nano-structures for potential applications in micro-electronics, magnetic data storage and atomically precise manufacturing [30-32] These studies range from homogeneous metallic systems (e.g Ag/Ag(100) and Pt/Pt(110)[28-29]) to semiconducting systems (e.g

Si/Si(111) and Si/SiC [33-36]) as well as heterogeneous systems comprising of mixed metal/semi-conducting materials (e.g In/Si(001) and Ga, In, Ag, Mn, Pb, Co-Si/Si(111) [37-41]) However there is still a bereft of studies concerning the origin and formation, atomic structure, adsorption and diffusion characteristics, physical and electronic properties of magic clusters We therefore contribute to this area by studying the physical and electronic properties of Si and binary magic clusters in terms of their size distribution, chemical information, formation and transport behaviour on Si surfaces such as Si(111) and SiC(0001) substrates as outlined in the following 3 Research Objectives

1.2 Research Objectives

(I) Si Magic Clusters on 6H-SiC(0001)

One of the primary issues is to be able to prepare well ordered substrate surfaces

on which we can grow magic clusters Hence we first study SiC surfaces in order to be able to produce high quality surface templates repeatedly 6H-SiC(0001) exhibits several surface reconstructions ranging from (3x3), (√3x√3), (5x5), (2√3x2√13), (6x6) and (6√3x6√3)R30° as shown by various techniques [5-6] In this work, we report new STM and XPS data that revealed the formation of an ordered (6x6) arrangement of Si clusters (diameter~14.3±0.5Å) [42] These clusters possessed a uniform shape and size, which were attributed to Si magic clusters due to the Si-rich nature of the surface This was surprising as the occurrence of magic clusters on SiC surfaces have not been reported before We showed that these clusters were derived from the ejection of the Si-tetra cluster unit of the initial (3x3) reconstruction when the surface is heated to elevated temperatures The result of this study is interesting, as it implies that the (6x6) structure was likely to be due to a self assembly of Si magic clusters rather than a reconstruction The formation and ordering of the (6x6) structure also appears to be mediated by clusters rather than by adatoms

Despite the myriad of surface reconstructions observed on the SiC surface, there has been very little coherent work discussing the evolution of these structures and especially involving the occurrence of Si magic clusters In particular, the role of silicon adatoms or magic clusters in terms of their atomic re-arrangement, bond breaking and

formation within the top few layers beginning with the SiC(0001)-(3x3) phase have been neglected Therefore objective (I) of this PhD program focuses on 3 aspects of magic cluster phenomenon on the SiC(0001) surface;

1) The first aim is to probe the origin and formation of Si magic clusters on the SiC(0001) substrate surface We intend to proceed with this work by annealing the SiC(0001) surface progressively beyond 800°C when the (3x3) surface is formed, to study the (3x3) phase transition to Si-rich (6x6) clusters with in-situ STM and XPS While the STM will provide real space evidence of the micro-structural regime, the XPS will be able to provide information on binding energy shifts as well as changes in peak areas as the surface is annealed This will allow us to address key issues in the formation

6H-of these clusters, such as whether this process occurs via the nucleation 6H-of Si adatoms or re-arrangement of the (3x3) surface or the ejection and agglomeration of whole tetra-clusters

2) The observation of these Si magic clusters would suggest a preference for Si atoms to exist as clusters with a specific magic number of atoms with similar atomic structure which promotes stability on the SiC(0001) surface The second intention is to utilize in-situ STM, with variation in tunneling biasing, to obtain real space information and consequently elucidate the structure of these magic clusters By analyzing physical cluster dimensions such as size and geometry derived as a function of tunneling voltage,

we intend to estimate the “magic” number of atoms in each cluster through a statistical study of cluster size distribution This will lead to the elucidation of the magic cluster structure using high resolution images and we will eventually propose a model of the

cluster through the co-relation of STM data with the (3x3) tetra-cluster model as reference

3) The third aim is to investigate the role these clusters play in the facilitating of phase transformations as the 6H-SiC(0001) surface is annealed By co-relating STM data on structural evolution with corresponding XPS spectra, we intend to trace the Si and C signal trends as the surface is progressively annealed from 800°C to 1200°C By doing so

we would be able to probe the chemical information of different surface structures at various temperatures and study the transition of a Si-rich surface consisting of predominantly Si-Si bonding to a C-rich one dominated by Si-C bonding Together with high resolution STM, we will also be able to analyze the different structures arising from the surface evolution as well as the self assembly process of Si magic clusters on 6H-SiC(0001) surface, by analyzing the final adsorption sites as well as direction and separation of ordering In particular, there has been much controversy surrounding the high temperature surface phases of SiC(0001) when it is heated beyond 1000°C The existence of a (6√3x6√3)R30° surface structure upon heating to ~1200°C [43-44] have been reported by several diffraction studies, which also attributed the structure to an incommensurate graphite overlayer [45-46] However, recent XPS studies suggested that the (6√3x6√3) R30° reconstruction could be attributed to carbon in a Si deficient environment and did not yet amount to surface graphitization, which only appeared after heating beyond 1200°C [47] Despite exhaustive studies on this structure, real space STM evidence of the (6√3x6√3) R30° reconstruction is still not obvious [48] Hence, in an effort to resolve the surface structures observed at higher temperatures, we intend to

investigate the surface structural evolution further by heating the 6H-SiC(0001) surface beyond 1000oC We will study again with STM, how the clusters participate in surface re-ordering and evolution leading towards the formation of high temperature surface structures

In order to explain the structural transformation observed with the occurrence of magic clusters, we will first analyze the geometry and atomic structure of the surface at various temperatures using high resolution STM and propose a model to describe the

observations We will use J Schardt et al’s (3x3) atomic model as a basis for discussion

with our STM observations [49] Specifically, we will use the tetra-clusters as basic-units

to propose structural models, to account for the observation of the clusters and to map the microstructure evolution arising from the (3x3) phase to the self organized (6x6) cluster arrangement and subsequent high temperature phases We will also determine if each structural phase observed can be accounted for by moving or removing M x 4 Si atoms (where M=number of clusters) This will complete objective (I)

(II) Si Magic Clusters on Si(111)

It is interesting to note that the size and shape of the Si magic clusters on SiC(0001) is similar to that of Si clusters observed on Si(111) surface [42] While the existence of Si magic clusters on the Si(111)-(7x7) surface have already been reported [33-35], there has been no work addressing the origin and formation of these clusters Although the literature suggest a preference for Si atoms to exist as clusters with a magic number of atoms with similar atomic structures on both surfaces, there have been intriguing lack of evidence of magic cluster ordering on Si(111) unlike SiC(0001) and how clusters influence surface structural changes Even fewer reports carry experimental studies addressing the fine structure and magic atom number of these substrate supported clusters on Si(111) Therefore objective (II) of this PhD program will address 3 key aspects of Si magic cluster work on the Si(111) surface;

6H-1) While the occurrence of Si magic clusters on Si(116H-1) have been recently reported 35], it is still unknown how these clusters are formed and the conditions of these clusters’ origin However we observe cluster-like particles of a uniform size occurring during the cooling transition from the high temperature “1x1” disordered phase to the more stable (7x7) reconstruction This observation is important as it suggests that these particles could be Si magic clusters which mediate structural transformation and function

[33-as vehicles for m[33-ass transport across surfaces Hence the role of Si magic clusters h[33-as important implications on the nucleation and growth processes of epitaxial Si surfaces

Therefore, our first aim in objective (II) is to use in-situ STM to determine if the particles are magic clusters and address the formation and origin of these clusters on Si(111) as well as the role it plays in facilitating the ordering process of (7x7) from

“1x1” However as the clusters are metastable and diffuse very quickly, we intend to create clusters by fast quenching the Si(111) surface from high temperatures of > 800°C, through means of kinetically limiting the diffusion of clusters Having formed domains of ordered (7x7) and “1x1”, we would be able to probe the gradual formation of ordered (7x7) from “1x1” using high resolution STM in real time This may be achieved via slow annealing at slightly elevated temperatures of ~ 400°C in order to slow down the kinetic processes In this way, we will able to trace the initial stages of Si adatoms undergoing self re-arrangement into magic clusters as a function of annealing time This would form the basis of our study of substrate induced formation of magic clusters during the phase transformation

2) Since Si magic clusters appear to form spontaneously from surface Si adatoms, in the second part of this work, we will try to grow Si magic clusters from Si adatoms deposited

on Si(111) Unlike previous work, where typical cluster-based evaporating sources are used [22, 50-51], we avoid depositing clusters of different sizes by using a Si solid source evaporator instead As the growth of these clusters appear to arise naturally as a tendency

of atoms on the substrate surfaces to self-assemble under specific preparation conditions [27, 26], we will use STM to study the formation of Si magic clusters from adatoms, as the surface is being annealed We will also probe if spatial ordering of Si magic clusters can be achieved on the Si(111) template using the available dangling bonds on the