University of Southern California Dissertations and Theses

Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

(USC Thesis Other)

Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source by Malak Khojasteh A dissertation presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In partial fulfillment of the requirements for the degree DOCTOR OF PHILOSOPHY (CHEMICAL ENGINEERING) August 2019 Copyright 2019 Malak Khojasteh 2 To my family 3 A c k n o w l e d g m e n t s First and foremost, I would like to express my sincere gratitude to my advisor Prof. Vitaly Kresin. It has been my honor to be his Ph.D. student. I could not have imagined having a better advisor and mentor for my Ph.D. study. He taught me, both consciously and unconsciously, how to be a good mentor and scientist. I appreciate all his contributions of time, ideas, and funding to make my Ph.D. experience productive and stimulating. He always appreciated my new ideas and supported me to implement them while helping me to stay on the right path. The joy and enthusiasm he has for his research and generally for science, and his patience and tolerance during tough times, were contagious and motivational for me during my Ph.D. pursuit. I am truly thankful for the excellent example he has provided as a great scientist and as an exceptional human being. Besides my advisor, I would like to thank the rest of my dissertation committee: Prof. Andrea Armani, and Prof. Steve Nutt, for serving as my committee members in both the qualifying exam and the final examination dissertation defense. Beyond that, I sincerely appreciate their support in my tough times to be able to pursue my Ph.D. I am thankful to Prof. Jahan Dawlaty for collaborating in part of my research study and giving me access to his laboratory and research facilities, as well as serving as a committee member in my qualifying exam. I am thankful to Prof. Noah Malmstadt for serving as a committee member in my qualifying exam. 4 I would like to thank Dr. Patrick Malenfant and Dr. Jacques Lefebvre, of the National Research Council of Canada, for their extensive discussions and collaboration in part of my research study through providing me with a thin film of single-wall carbon nanotubes. I would like to thank Dr. Sahand Pirbadian and Dr. Shuxing Li for extensive discussions and assistance with AFM; Dr. Matthew Mecklenburg for TEM imaging and extensive discussions; Dr. Andrew Clough for valuable help with XPS; and Bethany Seckman for DRS help. I would like to thank all my group members who assisted me, specifically Dr. Avik Halder for his invaluable help and advice for getting the deposition source started with me; Dr. Nicholas Guggemos and Dr. Daniel Merthe for helpful discussions and assistance; Mathew Orr and Akash Shah for their assistance with laboratory work. I am thankful to the USC Machine Shop personnel for their expert technical support, specifically Donald Wiggins, Seth Robert Wieman, Mike Cowan; and the engineers and staff of Mantis Deposition Ltd. for their advice. I would like to thank the USC department staff, specifically MaryBeth, Christina, Lisa and Andy. They have always been kind, supportive, and responsive whenever I reached out to them by email or in person. I am deeply grateful and indebted to my parents for their endless love and support. They have always believed in me and provided me their best, and continuously encouraged me to follow my passion. They are the best parents anyone could imagine having. 5 I am also so grateful to my brother, who has always been there for me to support me in my whole educational life, especially in my Ph.D. journey, and helped me to get through the tough times and overcome many challenges. I am truly obliged to all his support and love. 6 Table of Contents Acknowledgments ……………………………………………………………….3 List of figures …… ……………………………………………………………...10 Abstract ………… ………………………………………………………………13 1 Introduction ........................................................................................................ 16 1.1 What are clusters? ........................................................................................ 16 1.2 Production methods of clusters .................................................................... 17 Types of cluster sources ........................................................................... 19 How are clusters formed in cluster sources?............................................ 22 1.2.2.1 Cluster formation in a magnetron gas aggregation source ............... 22 1.3 Summary ...................................................................................................... 23 2 Detailed characterization of the experimental setup for the production of a size- selected metal nanocluster beam .............................................................................. 26 2.1 Introduction .................................................................................................. 26 2.2 Experimental setup....................................................................................... 28 2.3 Investigating the influence of source parameters on the nanocluster size distribution ............................................................................................................... 31 7 Argon and helium supply ......................................................................... 32 Magnetron power and aggregation length ............................................... 37 2.4 Quadrupole mass filter ................................................................................. 41 2.5 Cluster beam characterization ...................................................................... 45 Ratio of charged to neutral nanoclusters .................................................. 46 2.6 Kinetic energy measurement........................................................................ 49 Retarding field analyzer principle and design ......................................... 49 Cluster velocities ...................................................................................... 52 2.6.2.1 Ratio of negatively to positively charged nanoclusters ................... 55 2.7 Deposition .................................................................................................... 56 Cluster impact energy .............................................................................. 56 Characterizing nanocluster thin films with AFM .................................... 58 2.7.2.1 Sample preparation for AFM imaging ............................................. 59 2.8 Conclusion ................................................................................................... 63 3 Electrocatalysis application of size-selected MnO nanoparticle thin films in the process of water splitting ........................................................................................... 65 3.1 Introduction .................................................................................................. 65 3.2 What is water splitting and what are its challenges? ................................... 67 3.3 Experiment and characterization .................................................................. 73 8 Production and deposition of nanoparticles ............................................. 73 AFM characterization .............................................................................. 76 XPS characterization ................................................................................ 77 3.3.3.1 XPS result ........................................................................................ 78 UV-Vis spectrum ..................................................................................... 81 3.3.4.1 Bandgap measurement ..................................................................... 81 Electrochemical data ................................................................................ 83 3.4 Results and discussion ................................................................................. 85 3.5 Conclusion ................................................................................................... 88 4 A new technique to decorate SWCNTs with metal nanoclusters .................. 90 4.1 Introduction .................................................................................................. 90 4.2 Experiment section....................................................................................... 95 SWCNT samples preparation .................................................................. 95 Production and deposition of Aluminum nanoclusters ............................ 98 TEM characterization............................................................................. 100 4.2.3.1 TEM results .................................................................................... 101 4.3 Discussion .................................................................................................. 105 4.4 Conclusion ................................................................................................. 108 9 5 References ......................................................................................................... 110 6 Appendix A: AutoCad design ......................................................................... 127 6.1 Part A: Substrate holder design.................................................................. 127 6.2 Part B: QCM mounting flange ................................................................... 133 6.3 Part C: Heater holder and TEM grid holder............................................... 136 10 List of Figures Figure 1.1: Schematic of a cluster..................................................................................... 16 Figure 1.2: Sputtering cluster sources ............................................................................... 24 Figure 1.3: Aggregation sources ....................................................................................... 25 Figure 2.1: Apparatus schematic....................................................................................... 30 Figure 2.2: Effect of helium flow rate on the size of Mn nanoclusters. ........................... 33 Figure 2.3: Position of the peak of the Mn nanoparticle size distribution vs. Ar and He flow rates. .................................................................................................................................. 34 Figure 2.4: a) Peak vs. Power and Helium flow rate, b) Peak vs. Power and Argon flow rate..................................................................................................................................... 39 Figure 2.5: a) Peak vs. Aggregation length and Argon flow rate, b) Peak vs. Aggregation length and Helium flow rate ............................................................................................. 40 Figure 2.6: Schematic of quadrupole mass filter .............................................................. 44 Figure 2.7: The setup to measure the ratio of charged to neutral nanoclusters ................ 48 Figure 2.8: Retarding field analyzer (RFA) equipped with Faraday cup ......................... 50 Figure 2.9: Velocity distributions of Mn nanoparticles emitted by the source, as determined via retarding potential measurements. .............................................................................. 53 Figure 2.10: Molecular dynamics simulations of the impact of individual Molybdenum clusters (Mo1043) and the morphology of the cluster-assembled films created on a Mo (100) surface as a function of the cluster energy. ....................................................................... 57 Figure 2.11: Schematic of Atomic Force Microscopy (AFM) ......................................... 58 11 Figure 2.12: Substrate holder from the front and top views ............................................. 59 Figure 2.13: AFM images of 3nm Copper nanoclusters on HOPG in different coverages ........................................................................................................................................... 61 Figure 2.14: AFM result ................................................................................................... 62 Figure 3.1: Basic scheme of a water electrolysis system .................................................. 68 Figure 3.2: Structure of the Mn4CaO5 cluster ................................................................... 70 Figure 3.3: Overall size range of Mn nanoparticle ions in the beam. ............................... 74 Figure 3.4: The principle of XPS ...................................................................................... 78 Figure 3.5: XPS spectra .................................................................................................... 80 Figure 3.6: Bandgap of the 6 nm diameter MnO nanoparticles. ....................................... 83 Figure 3.7: Electrochemical data measurement ................................................................ 84 Figure 3.8: Tafel plot ........................................................................................................ 84 Figure 3.9: Stability measurement .................................................................................... 85 Figure 4.1: Visualization of a single wall carbon nanotube.............................................. 91 Figure 4.2: Functionalization possibilities for SWNTs .................................................... 94 Figure 4.3: AFM images of drop casted single wall carbon nanotube solution on Si/SiO2 substrate. ........................................................................................................................... 97 Figure 4.4: Heater setup .................................................................................................... 98 Figure 4.5: Transmission electron microscope ............................................................... 103 Figure 4.6: TEM image of suspended SWCNT on a holey carbon TEM grid before bakeout the sample. ...................................................................................................................... 104 12 Figure 4.7: TEM images of individual suspended SWCNT samples on holey SiN TEM grids after bakeout the samples at 400C and for10 hrs. .................................................. 104 Figure 4.8: TEM images of deposited Aluminum nanoclusters on SWCNTs on a holey SiN TEM grid. ........................................................................................................................ 105 Figure 6.1: Substrate holder design. ............................................................................... 127 Figure 6.2: QCM mounting flange.................................................................................. 133 Figure 6.3: Heater setup and TEM grid holder mounting flange. ................................... 136 13 A b s t r a c t The pursuit of nanoscale-based materials in which the size and composition of the constituent building blocks can be accurately manipulated is an especially promising direction of inquiry. The study of size-selected nanoclusters focuses on exactly this target: by tracing the evolution of nanoscale metal particle properties atom by atom, one can observe, tune, and identify the optimal system with maximum precision. A productive pathway to building structures based on size optimized nanoclusters is cluster beam deposition, which is a new and rapidly developing tool. This technique allows flexible tuning of the relevant parameters and a wide choice of working materials. In this study, a dc magnetron sputtering gas aggregation source equipped with a quadrupole mass filter is employed for producing robust fluxes of metal nanocluster ions, filtering them by size, and soft landing them on a substrate. The performance of this cluster source was characterized in detail by investigating the role of a range of overlapping operating source parameters in nanocluster production. A detailed map of the influence of each parameter on the average nanocluster size was established. In this way, it is possible to identify the main contribution of each parameter to the physical processes taking place within the source. These assignments were supplemented by nanocluster beam characterization, including measurements of the abundance ratios of charged to neutral and negative to positive nanoclusters, and of the size dependence of nanocluster velocity. Finally, the morphology and size of deposited nanoclusters were characterized by atomic force microscopy (see Chapter2). 14 In the subsequent parts of this thesis, we address two subjects. The first is concerned with the size dependence of the chemical activity of metal nanoclusters, and the second is an investigation of the interaction between metal nanoclusters and single wall carbon nanotubes (SWCNTs). The size dependence of the chemical activity of metal nanoclusters was approached through a study of the application of size-selected MnO nanocluster thin films in electrocatalytic water splitting. Different sizes of Manganese nanoclusters (below 10 nm) were produced in the nanocluster source and soft-landed onto conducting electrodes. The mass loading of these catalytic particles was kept constant and corresponded to sub- monolayer coverage. Measurements of the water oxidation threshold revealed that the onset potential decreases significantly with decreasing particle size. The ability of such a sub-monolayer film to lower the reaction threshold signifies that an inherent change takes place in the charge transfer energetics. This suggests that the key role is played by intrinsic size effects, i.e., by changes in the electronic properties and surface fields of the nanoparticles with decreasing size. We anticipate that this work will serve to bridge the knowledge gap between bulk thick film electrocatalysts and natural photosynthetic molecular-cluster complexes (see Chapter3). The investigation of interactions between metal nanoclusters and SWCNTs addressed the goal of decorating SWCNTs with size optimized metal nanoclusters to assemble novel nanoscale structures. This goal requires special fabrication techniques in order to keep the physical and chemical properties of metal nanoclusters and SWCNTs pristine. 15 The novelty of our work is signified by the deposition of pure (ligand-free) size- selected metal nanoclusters onto a naked suspended SWCNT, which is not possible with other physical or chemical approaches. A set of procedures was developed to prepare high- quality electron microscope grids with opening bridged by SWCNTs for use as high- quality nanoparticle deposition substrates, and imaging of the obtained samples was performed. This marks a significant step toward the realization of nanoparticle-nanowire assemblies employing SWCNT transport measurements as a probe of nanocluster electronic structure (see Chapter4). . 16 1 Introduction 1.1 What are clusters? Research on clusters has developed from atomic and molecular science by emphasizing studies of aggregates of a countable number of atoms and molecules (see Figure 1.1). Such aggregations have been found to display properties distinct both from those of atoms and molecules and from the bulk materials. Cluster geometries, electronic structures, phase changes, and intrinsic chemical and physical properties all evolve with increasing cluster size (Johnston, 2002). Clusters can be classified by their sizes (small, medium and large sized clusters), by their types of atoms they are composed of (homogeneous: made of one type of particles Figure 1.1: Schematic of a cluster Cluster, as an intermediate between individual atom or molecule and bulk material, is an aggregate of a countable number of particles (i.e. atoms or molecules), greater than 2 and less than 10 6 , having distinct properties of induvial particles and bulk materials. 17 or heterogeneous: made of more than one type of particles) and by the nature of their bonding. They can also be neutral or charged. Medium and large-sized clusters show a systematic relationship between the number of particles in the cluster and their properties, while the small sized clusters do not follow the systematic trend, and their properties change sharply. Large sized clusters typically are in the size of a few nanometers to 50 nm; therefore, they can be categorized as nanoscale materials. Moreover, depending on the type of clusters, the force among the particles in clusters can be ordered as strong or weak bonds. (Huberland 1994) The investigated clusters in this research study are metallic, homogeneous and in the range of 1 to 10 nm; therefore, according to cluster classification, they are categorized as the large-sized clusters and nanoscale materials; so, in this thesis, they are referred to as either nanoclusters or nanoparticles. 1.2 Production methods of clusters Clusters are produced by two main routes: chemical routes, also known as wet chemical synthesis and physical routes. Chemical routes, as the most widely used technique, offer a decent control over size, shape, and chemical composition. Their setups are simple and cost-effective; therefore, chemical synthesis which is materializing mostly in the liquid phase is energy efficient and easy to scale up as it does not need an expensive source of energy or high pressures. However, generally, chemicals are toxic, explosive and corrosive which cause significant risk to occupational health and environments. Moreover, synthesized clusters 18 need further processing and purification to remove added stabilizers or surfactants to prevent cluster agglomerations and to control cluster sizes; therefore, the purity of synthesized clusters and consequently their chemical and physical properties will be affected (Su and Chang 2018). Although it is comprehensively accepted that a unique approach for producing clusters with all-inclusive controlled properties does not exist, the general essentials that one should seek to fabricate clusters are the control on size distribution, structure and chemical reactivity. Therefore, alternatively, physical routes of the cluster synthesis used as a major technique to study the properties of clusters which mainly have been developed in the gas phase processes and typically they are generated in cluster sources. The cluster sources need a complicated setup, expensive source of energy and other supplies; however, the environment that cluster beams are formed is a vacuum; therefore, it is a contamination-free process, and it can yield clusters with high purity (ligand or surfactant-free). Moreover, the clusters can be generated with precise control on the size up to a few atoms. After cluster generation, they can be size selected with a high degree of accuracy through in-line mass filters. Additionally, the morphology of landed clusters and the quality of the thin film can be tuned by controlling the kinetic energy of clusters (Haberland et al. 1994). Today, in the state-of-the-art cluster source, the environment of the formation of cluster beams can be chosen reactive or non-reactive owing to the source adaptable operating parameters, enabling the manipulation of the inner structure of clusters and the synthesis of the single-component or multicomponent clusters (Palmer et al. 2018). 19 Consequently, an advanced cluster source technique enables us to monitor many steps of the cluster formation and the cluster assembling processes, which is not possible in the chemical routes. Thus far, various types of cluster sources have been developed to produce a wide range of cluster types. Cluster sources can be generally categorized into three main groups: vaporization methods, continuous sources, and pulsed sources, which are discussed in detail in the literature (Milani and Iannotta 1999). The subject of this thesis is the fabrication of metal nanoclusters by a dc magnetron sputtering gas aggregation source, classified as a continuous source. Therefore, the discussion in the next section will not be exhaustive and will simply review some sources with a generally related design in order to provide a better understanding of the performance of the cluster source used in this research. Types of cluster sources Sputtering means to take off atoms from the target surface via high energy inert gas ions bombardment. Ion Sputtering Gun Source, shown in Figure 1.2 (a), produces small, singly ionized clusters through sputtering. The bombardment energies for heavier inert gases are in the range of 10-30 keV and currents of approximately 10mA; therefore, clusters are generated hot, and cluster cooling involves evaporation. The ion sputtering gun source is suitable to operate in a continuous beam with high stability; however, the beam intensity drops off rapidly (Johnston, 2002). 20 Magnetron Sputtering Source, shown in Figure 1.2 (b), is the modified version of the sputtering source. The magnetron head consists of the grounded cup known as an anode and a target known as a cathode. The plasma is ignited in argon by applying dc or ac potential on the target. The discharge produces sputtered atomic vapors and electrons. The electrons are used to ionize the inert gas and sustain the discharge, and they are repelled by the target. Since the number of ejected electrons per ion collision is low, two arrays of magnets behind the target confine the magnetic field close to the target to trap the electrons in the helical path. Indeed, Lorentz force arising from the electromagnetic fields and to inter-particle collisions, maintains the plasma close to the target. Therefore, the continuous discharge current obtained. This makes a ring-shaped glow, accompanied by a circular region of erosion on the target (Huttle 2017). Comparing to non-magnetron sputtering, high discharged current in magnetron sputtering happens in the lower operating pressures and lower operating voltages; therefore, the ionization efficiency increases. Gas Aggregation Source, shown in Figure 1.3(a), reported as the first type of metal cluster sources, has the mechanism of the cluster growth analogous to the cloud and smoke formation. Metal vapors are produced by an evaporation process through in situ positionings of the crucible assembly within the gas flow; then atomic vapors are led into cold inert gas (Helium or Argon) cooled down by colliding to the cold wall of the outer layer of either LN2 or water jacket around the source where they become supersaturated and clusters aggregate. The produced cluster beam is continuous, but the beam intensity is 21 low due to subsonic expansion, and cluster sizes depend on the residence time of clusters in the aggregation zone and physical dimension of the source (Johnston 2001; Binns 2001). DC Magnetron Sputtering Gas Aggregation Source, shown in Figure 1.3(b), was invented by Haberland’s group in 1992 in Freiburg, as the combination of a magnetron sputtering source and a gas aggregation source equipped with LN2 cooling jacket, with the advantages of producing continuous strong fluxes of metal clusters with the large fraction of ionized ones. It also covers a wide range of cluster sizes consist of a couple of atoms to 10 ^6 atoms by sputtering discharge. These advantages provide an opportunity to characterize the size and velocity distribution of clusters in the cluster beam through in- line mass spectrometry devices, resulting in tuning the quality of thin films and having various industrial applications (Haberland et al.1992,1994; Huttle 2017). Nowadays, the source got quite popular in a wide range of disciplines, such as sputtering multiple magnetron heads or utilizing the reactive gases to make multicomponent clusters, such as core-shell, alloy, and Janus clusters, as well as applying radio frequency power sputtering to synthesize clusters of insulating materials (Huttle 2017). All fabricated nanoclusters in this research study are produced by a source called Nanogen 50, sold by Mantis Deposition Ltd., (Oxfordshire, UK). The detailed performance of Nanogen50 will be discussed in the next chapter. 22 How are clusters formed in cluster sources? Commonly, in all cluster sources, cluster generation involves three main steps: vaporization; making of atoms or molecules in the gas phase, nucleation; condensation of atoms or molecules to make initial embryos (nucleus), and finally growth; by adding the number of atoms or molecules to the initial nucleus or by coalescence of small clusters to make the larger ones. Depending on the nature and conditions of a source, different size distributions of clusters are generated (Johnston 2001). 1.2.2.1 Cluster formation in a magnetron sputtering gas aggregation source In a magnetron sputtering gas aggregation source, supersaturated hot metal vapor atoms are formed by the sputtering discharge of a metal target with buffer gas ions. The nucleation process occurs through the quenching of hot metal vapor atoms in a flowing stream of a cool buffer gas, where the gas stream cools down by colliding to the internal wall of LN2 jacket. The initial thermal energy of the buffer gas is much lower than that two vapor atoms; therefore, through three body collisions between two vapor atoms and cold buffer gas the excess energy of vapor atoms is removed, and the initial dimer is formed. Indeed, the thermal energy of the buffer gas stream is getting so high that the other two atoms find themselves bound (Smirnov 2010). The nucleation mechanism is shown in equation 1.1, where M is a metal vapor atom, and A is a buffer gas (Argon gas is used in our case), and M2 is an initial dimer. M + M + A (KE1) → M2 + A (KE2 > KE1) 1.1 23 Afterward, the initially formed clusters (Mn) act as a condensation nucleus for further cluster growth through two body collisions either through accretion another hot vapor atom (M) or coagulation with a smaller cluster (Mm). The cluster growth mechanism is shown in the following: Mn + M → Mn+1 1.2 Mn + Mm → Mn+m 1.3 1.3 Summary A cluster is a bridge between an individual atom or molecule and bulk material. Clusters can be categorized by size, the type of atoms they are composed of or by the nature of their atomic bonding. Cluster science has opened a new pathway to studying and making new types of materials, new chemical reactions, and newly industrialized nanotechnologies. While clusters can be made by two main routes, chemical and physical, they are mainly studied by physical routes and cluster sources have been used as a major technique to generate clusters. Various types of cluster sources have been developed to produce a wide range of clusters. The subject of this thesis is the fabrication and deposition of metal nanoclusters with a dc magnetron sputtering gas aggregation source as a state-of-the-art technique for thin film deposition. All nanoclusters in this research study are produced by a source called Nanogen 50, sold by Mantis Deposition Ltd., (Oxfordshire, UK). The detailed performance of Nanogen50 will be discussed in the next chapter. 24 In the present work, we studied metal clusters in the range of 1 to 10 nm classified as large-sized clusters; therefore, in the rest of this manuscript, clusters are referred to as either nanoclusters or nanoparticles. Figure 1.2: Sputtering cluster sources a) Ion gun sputtering source; it is using an ion gun to accelerate ions (Ar+, Kr+, Xe+) onto the target. The heavier inert gases (Kr and Xe) are generally used as sputtering ions with bombardment energies in the range 10–30 keV and currents of approximately 10 mA. b) Magnetron sputtering source; it is using a magnetically confined plasma as an ion source. Charge production and surface erosion are maximal in the region where the magnetic field is parallel to the surface. Sputter head must be water cooled, depicted as coolant in and out in the picture. 25 Figure 1.3: Aggregation sources a) Gas aggregation source; this source is also known as smoke source. Generated metal vapor atoms are introduced into a cold inert quench gas (He or Ar at a pressure of 50–500 Pa), where the vapor becomes supersaturated and clusters aggregate, analogous to cloud and smoke formation. (Binns 2001) b) DC magnetron sputtering gas aggregation source; this source is a combination of a magnetron sputter discharge source and gas aggregation source equipped with LN 2 cooling jacket. 26 2 Detailed characterization of the experimental setup for the production of a size-selected metal nanocluster beam 2.1 Introduction To explore precisely how the properties and functionality of nanoscale particles depend on the number of constituent atoms, it is essential to have tools which enable full control of particle size, purity, and shape. Consequently, surface deposition of size-selected metal nanoclusters has gained popularity for its ability to tune the particle size and composition over a wide range (Milani and Iannotta 1999; Meiwes-Broer 2000; Binns 2001; Wegner et al. 2006; Vajda and White 2015). A powerful tool for generating beams of neutral and charged nanoclusters covering a range of sizes and materials is the sputtering/aggregation source, also sometimes referred to as the “terminated gas condensation” source (Haberland et al. 1992, 1994; Hutte 2017). It is based on the quenching of atomic vapor produced by magnetron sputtering of the material of interest. The vapor becomes supersaturated due to collisions with the surrounding inert gas atoms which are cooled by the cold walls of the aggregation zone, condenses into nanoclusters, and is carried out of the condensation chamber by a continuous flow of gas. This device has been adopted by many research groups and has evolved from a purely home-built instrument to a commercial thin-film deposition product. 27 Understanding the efficiency of cluster formation in a source of this type is obviously a nontrivial problem because it involves the interplay between multiple processes, including (1) sputtering of atoms and ions, (2) emergence of condensation nuclei, (3) supersaturation and particle growth, (4) transport to the exit aperture and diffusion to the aggregation chamber walls, (5) expansion through the aperture into the process vacuum chamber, accompanied by the formation of the nanoparticle beam and termination of growth. Importantly, as nanoparticles move with the gas through the source towards the exit aperture, their local environment continuously changes, adding a degree of nonequilibrium dynamics to the growth process. Not surprisingly, therefore, the yield and size distribution of the resulting nanoparticle beam are functions of multiple interrelated operational parameters: source geometry, gas flow rates, discharge power and configuration, aggregation residence time, etc. Thus, to enhance particle production and to steer its size distribution towards the desired range it is valuable to have both empirical and conceptual insights into the effect of these parameters on the cluster formation process. Many papers have examined the effect of operating conditions on the size, morphology, and kinetic energy of nanoclusters (examples include Hihara and Sumiyama 1998; Morel et al. 2003; Pratontep et al. 2005; Das et al. 2009; Quesnel et al. 2010; Ayesh et al. 2010; Gracia-Pinilla et al. 2010; Nielsen et al. 2010; Ganeva et al. 2012; Luo et al. 2012; Ayesh et al. 2013; Bray et al. 2015; Fischer et al. 2015; Kusior et al. 2016; Zhao et al. 2016; Rudd et al. 2017), but each typically looked only at a subset of source parameters. Consequently, a comprehensive multidimensional characterization has not yet been presented. 28 In this chapter, we describe a systematic study of the influence of the parameters of our source on the production of metal nanoclusters, using manganese. Four independently controlled variables (argon and helium flow rates, discharge power, and aggregation length) were varied over a set of discrete levels (corresponding to a total of 720 four- dimensional grid points), and the effect of each combination on the cluster size distribution can be traced and visualized with the help of contour plots. Such a map over the permutations and interplay of independent factors is sometimes referred to as a “factorial design” experiment. It allows us to consider and assign the key roles played by the individual parameters listed above, for example, the distinct contributions of argon and helium gases to the processes of nanocluster formation and transport within the source. These assignments were supplemented by characterizing nanocluster ion beams including, the ratio of charged to neutral nanocluster measurements, velocity measurements, and the ratio of negatively to positively ionized nanocluster measurements. Finally, the morphology and the size of deposited nanoclusters were characterized by AFM technique. 2.2 Experimental setup Figure 2.1 shows the scheme of our experimental setup for the production of size- selected nanoclusters. The experimental setup consists of three main parts: magnetron gas aggregation source, quadrupole mass filter and deposition chamber. The source is Nanogen-50 from Mantis Deposition Ltd. As mentioned above, nanoparticles are produced by magnetron (dc) sputtering followed by condensation within the environment of a cold inert gas. The magnetron block is equipped with “magnet type 29 A” whose most useful feature, per company specifications, is that it produces almost exclusively ionized clusters (Mantis Deposition 2017) making it possible to filter and manipulate the entire beam by electric fields (see section 2.5.1). Clusters are generated from 99.95% Mn targets (ACI Alloys) of 2-inch diameter and 0.125-inch thickness, bonded onto a copper backing plate. The magnetron head is mounted on a linear translator, enabling the aggregation length (the distance between the target and the exit aperture) to be varied over a range of 10 cm. Argon and helium gases (both 99.999% purity) are introduced into the source region behind the magnetron head, with flow rates regulated by Alicat MC series mass flow controllers. The flow rate dependence of cluster production will be described below. Argon is used as the plasma discharge medium, and the roles of argon and helium in the nucleation and clustering process are further discussed below. The outer jacket of the source chamber is maintained full of liquid nitrogen with the help of a funnel filling system and a liquid level controller. The gas carries the nanoclusters out of a 5-mm aperture at the source exit, where particle growth is terminated. The resulting directed beam passes through a 6-mm skimmer followed by a high-range/high throughput quadrupole mass filter (Mantis MesoQ, see also (Baker et al. 1997)) with a manufacturer stated size resolution of ~2%. The standard mass range of the filter is from 350 amu to ~10 6 amu, but its performance can be extended somewhat to either side of the standard range. A grid mounted at the quadrupole exit samples the ion flux and an electrometer, included in the mass spectrometer instrumentation package, measures the current corresponding to the selected cluster size. 30 Figure 2.1: Apparatus schematic In the magnetron gas aggregation source sputtered metal atoms enter the condensation zone where they undergo collisions with the inert gas and quickly thermalize. Nanocluster ions form and grow as the mixture moves through the source toward the exit aperture. The ions are filtered by a quadrupole mass analyzer equipped with an ion flux measurement grid and enter the deposition chamber. 31 The resolution of the mass filter is selected by setting the U/V ratio (i.e., the ratio of the dc and ac amplitudes of the quadrupole’s rod voltages) between 0.001 and 0.168; the rf frequency is then adjusted automatically by the MesoQ power supply and its control software. For the data reported below the U/V ratio was kept at 0.02 (see section 2.4 for more details). Upon passing through the quadrupole, the size-selected nanoclusters find themselves in the main deposition chamber (base pressure 10 -6 Pa). Here their mass deposition rate, as a function of size, can be measured using a quartz crystal film thickness oscillator (McVac Manufacturing) and monitor (Inficon XTC). Also, the ratio of charged to neutral nanoclusters produced by the source can be measured using a setup comprised of a quartz crystal oscillator/film thickness monitor (QCM) and deflector plates arrangement positioned in the deposition chamber (see section 2.5.1). The ion current impinging on the deposition surface can be measured by means of a picoammeter (Keithley 6487). A Faraday cup arrangement can be positioned downstream from the quadrupole exit in order to measure the cluster ions’ kinetic energies (see sections 2.6, 2.6.1, 2.6.2). 2.3 Investigating the influence of source parameters on the nanocluster size distribution As described above, the magnetron sputtering, and cluster formation processes involve the interplay of many source parameters. In our work, the four main factors are the 32 magnetron discharge power P, the aggregation length L, and the Ar and He gas flow rates QAr and QHe. The cluster beam distribution was traced over 5×3×4×12 set levels of these variables, respectively, for a total of 720 outcome data points. Such a map enables us to examine both individual effects of the source parameters on the cluster beam distribution as well as, importantly, possible correlations which cannot be detected from separate one- way analyses. Argon and helium supply In dc magnetron sputtering, a high negative voltage is applied to the target, accelerating Ar + ions to sputter material off the target (in our case, Mn) surface. Strong magnets positioned behind the target create a specially shaped magnetic field designed to lengthen electron paths in front of the target and intensify the plasma. In our exploration of the parameter space, we first let in only argon gas to determine the dc power needed to produce a stable flux of Mn nanoclusters as detected by the quadrupole mass filter. This process was performed gradually in order to prevent target thermal shock with possible resulting cracking or debonding from the backing plate. Once the discharge is established, the Ar flow rate can be increased further, and then He admixed gradually. In this way, the variation of cluster sizes as a function of both gas flow rates can be mapped out for a given discharge power and condensation length. We found that the size distribution is quite reproducible for each set of operating parameters. Initially, as the supply of pure argon is increased both the flux and the average size of the cluster ions grow until finally a stable log-normal-type shape of the distribution 33 becomes established. At this point, the helium supply is turned on, and the response of the nanoparticle beam to increasing helium flow is illustrated in Figure 2.2; the overall intensity rises, reaches a maximum, and then starts to decrease, while the average particle size shifts to smaller sizes. At the same time, the width of the beam distribution becomes narrower. Figure 2.2: Effect of helium flow rate on the size of Mn nanoclusters. Effect of He flow rate on the size distribution of Mn nanoclusters. All other source parameters are kept constant: aggregation length 9 cm, discharge power 21.8 W, Ar flow rate 150 sccm. Figure 2.3 puts the influence of both gases into perspective by simultaneously plotting the effect of Ar and He flows on the peak of the cluster beam distribution. With the helium supply fixed, increasing the argon flow has only a moderate influence on the average 34 particle size. However, (as already illustrated in Figure 2.2) an increase in the helium flow shifts the beam distribution toward lower sizes very significantly. How can one interpret these different (indeed, opposite) trends? It is evident that helium and argon perform distinct functions within the cluster source. Their roles and influences can be rationalized as follows. As extensively described in the literature, the formation of nanoclusters Mn out of atomic vapor is initiated by nucleation and sustained by supersaturation and growth (see, e.g., Kappes and Leutwyler 1988; Haberland 1994; Pauly 2000; Smirnov 2000; Hutte 2017). The initial step is the formation of a bound dimer M2 which requires a three-body collision for stabilization: M+M+Ar → M2+Ar. It is well known that the heavier noble gas Figure 2.3: Position of the peak of the Mn nanoparticle size distribution vs. Ar and He flow rates. The flow rates were measured at intervals of 20 sccm, and the values in-between interpolated. The discharge power was 22 W and the aggregation length was 9 cm. 35 atoms are efficient at removing the dimer’s binding energy, and helium is not nearly as effective at enabling nucleation. This is also why heavier carrier gases are better at promoting clustering in supersonic expansion sources (Kappes and Leutwyler 1988). The dimers then serve as condensation nuclei for further growth if the vapor is maintained in a state of supersaturation. In this process, clusters grow by sequential condensation as additional atoms arrive at their surface one by one (with further collisions with noble gas atoms helpful in cooling the cluster seeds by removing the additional condensation energy). At higher nucleation densities, cluster-cluster collisions also can result in the appearance of larger particles. Particles which reach the so-called “critical size” will continue coagulating towards the condensed phase, therefore if a population of finite-sized clusters is desired then the condensation process must be interrupted. In the present source, this comes about as the gas flow carries the atomic vapor through the aperture and out of the condensation zone. Now we can formulate the separate roles of the two noble gases supplied to the source. While the distinction obviously is not sharp, it enables useful qualitative interpretation and guidance. As just stated, the size of nanoclusters in the beam is to a large degree controlled not by a hypothetical equilibrium distribution, but by the fact that the growth process is interrupted by the transit of the clusters out of the source (hence the aforementioned label “terminated gas condensation source”). The stage at which particle condensation is interrupted, and therefore the maximum size that they are able to attain, is defined by the residence time in the growth region, i.e., by the speed at which the metal vapor/inert gas 36 mixture is swept from the sputtering area to the source exit aperture. It is this transport which appears to be mostly affected by the amount of helium flow into the source, in such a way that the average cluster size goes down as the helium supply increases. In what way can the rate of helium gas supply influence the time a nanoparticle spends inside the source? It might be supposed that a higher mass flow¸ Q, translates into a greater speed of the gas column inside the source, vgas, pulling the particles along and reducing their residence/growth time. However, this is mainly not the case. Indeed, in equilibrium, the gas mass flow through the source is Q= ρgasvgasA, where ρgas is the gas mass density in the column and A is its effective cross section. At the same time, Q must equal the mass flow through the nozzle aperture into the vacuum chamber, which is proportional to the stagnation pressure in the plane of the nozzle (Miller 1988; Pauly 2000) and thereby to ρgas: Q Pgas  ρgas. Comparing these two expressions, both of which involve ρgas but only one involves vgas, we conclude that raising the inlet gas flow rate should mainly affect the pressure and density of the gas column inside the source but not its velocity. Therefore, the likely reason for the reduction in average cluster size with greater He density inside the source is that the clusters become more effectively entrapped in the gas streamlines. This derives both (i) from the higher number of cluster collisions with the gas atoms in the column drifting to the exit aperture, and (ii) from the fact that the rate of cluster diffusion toward the surrounding walls decreases inversely with the diffusion coefficient and therefore inversely with the gas density (Smirnov 2000; Shyjumon et al. 2006). As a result, the growing particles have a greater tendency to persist on their direct trajectories, their residence time decreases, and the growth is terminated sooner. 37 The fact that smaller cluster sizes are congruent with entrapment in the gas also is supported by velocity measurements on particles emerging from the source aperture, as described in section 2.6. An increased supply of Ar contributes to cluster transport as well; however, it also performs the essential functions of (1) enabling the sputtering process, thereby feeding atoms and atomic ions into the vapor, and (2) facilitating the appearance of condensation nuclei. Hence the argon density strongly affects the overall intensity of the nanoparticle beam, but its roles in supplying metal vapor for coagulation and in promoting cluster drift toward the source exit appear to balance each other out. As a result, the Ar flow rate does not have a sharp influence on the size distribution of the formed particles. An alternative interpretation of the principal role of helium in a magnetron source was put forward by Pratontep et al. (2005). They suggested that the helium gas is itself involved in cluster formation, in such a way that with increased He flow there is a rise in the nucleation of small seeds which results in more but smaller clusters. However, since argon is even more efficient in enabling the formation of small nuclei, under this scenario one might expect a stronger shift towards small sizes not just with He, but also with increasing Ar flow, which is not observed. Magnetron power and aggregation length The dc sputtering discharge power strongly affects nanocluster production. In principle, the stronger the discharge, the greater the supply of raw cluster material into the vapor; however, one also has to be cognizant of heat load limitations on the target as well 38 as of discharge stability and plasma charging dynamics. In Figure 2.4(a), we examine the influence of power and helium flow rate on the peak of the beam size distribution. We see that in this representation, the helium supply again plays the most influential role. Figure 2.4(b) plots the variation of the peak of the beam size distribution under the influence of discharge power and argon flow rate in the absence of helium gas. Note that the size range variation is significantly narrower than in the presence of He. A comparison of Figure 2.4 (a) and (b) reaffirms that He plays the dominant role in shifting the nanocluster distribution toward smaller sizes. Analogous conclusions are drawn from varying the aggregation length L. Figure 2.5(a) is a plot of the joint influence of L and QAr at zero helium flow rate on the peak size of the nanocluster distribution. We see that the aggregation length plays the main role in changing the size, while there is little sensitivity to argon flow. In contrast Figure 2.5 (b), which follows the joint influence of L and QHe, demonstrates a strong effect along both axes. The marked decrease in average particle size either with increasing He flow rate or with decreasing aggregation length confirms that the source residence time is the most sensitive parameter in determining the extent of particle condensation in the cold, strongly supersaturated, sputtered metal vapor environment, and that the dominant role of helium is in setting the transport time through the aggregation tube, as discussed above. 39 Figure 2.4: a) Peak vs. Power and Helium flow rate, b) Peak vs. Power and Argon flow rate (a) Position of the peak of the Mn nanoparticle size distribution vs. magnetron discharge power and He flow rate. The flow rate was measured at intervals of 20 sccm, and the discharge powers and corresponding discharge currents were P=7.3 W, 11.3 W, 15.8 W, 21.8 W, 35 W and I= 35 mA, 55 mA, 75 mA, 100 mA, and 150 mA, respectively; the values in-between are interpolated. The argon flow rate was 190 sccm and the aggregation length was 9 cm. (b) The peak of the Mn nanocluster size distribution vs. magnetron discharge power and Ar flow rate. The flow rate intervals, and powers, discharge currents and the aggregation length were the same as in (a). No helium flow was present for this plot. 40 Figure 2.5: a) Peak vs. Aggregation length and Argon flow rate, b) Peak vs. Aggregation length and Helium flow rate (a) The peak of the Mn nanocluster size distribution vs. the aggregation length and Ar flow rate. The flow rate was measured at intervals of 20 sccm, and the investigated lengths were 5, 7, and 9 cm. For this plot, no helium flow was present, and the discharge power was 7.3 W. (b) The peak of the Mn nanocluster size distribution vs. the aggregation length and He flow rate. The flow rate intervals, the investigated aggregation lengths, and the power were the same as in (a), and the Ar flow rate was 210 sccm. 41 2.4 Quadrupole mass filter The history of quadrupole mass filter invention goes back to 1953 when Paul and Steinwedel started studying on quadrupole analyzer and ion traps (Paul et al. 1953, 1958), and after that, it was fully developed into the available market instrument by Shoulders, Finningan, and Story (Finningan 1994). As it is shown in Figure 2.6 (a, b), a typical quadrupole mass filter is made of four symmetrically identical rods which are set parallel to each other. The opposing face rods are connected together electrically as a pair, and simultaneously the same (ac) and (dc) voltages are applied to each pair, but with opposite sign (Hoffmann and Stroobant 2007). The Quadrupole mass filter operates based on the stability of the trajectories in oscillating electric fields to separate ions based on their mass to charge ratios (m/z), and its operation is relatively independent of the energy of incoming ions. As depicted in Figure 2.6, the ions move along the z-axis in the space between the rods. They feel the motion force into an oscillating path in the x-direction and y-direction induced by the electric field; If the amplitude of the trajectory oscillation is smaller than the central field radius r0, the ions reach the detector; if the amplitude exceeds this the ions will discharge on the rods or the surrounding surfaces and will not pass through the filter (Milani and Iannotta 1999). The equations of the motion, namely Paul equations, are: 𝑑 2 𝑥 𝑑𝑡 2 + 2q (U + V cos ωt) x m𝑟 0 2 = 0 𝑑 2 𝑦 𝑑𝑡 2 – 2q (U + V cos ωt) y m𝑟 0 2 = 0 42 𝑑 2 𝑧 𝑑𝑡 2 = 0 Mathematically the differential equations of motion in the x and y-directions can be modeled with the help of the Mathieu equation, which is explained in detail in the literature (Dawson 1976). The stable trajectory solutions occur only for certain regions of A and Q. A = 8qU/ mω 2 ro 2 and Q = 4qV / mω 2 ro 2 A and Q are dimensionless stability parameters. U and V are (dc) voltage and (ac) amplitude, respectively. ro is the field radius, 𝜔 = 2𝜋𝑓 is the angular frequency, q=z.e is the charge of a particle (where z is the charge number and e is the elementary charge), and m is the ion mass. In the present work, a commercially available quadrupole mass filter (called Meso- Q mass filter, Mantis Deposition Ltd.) was utilized to measure the mass spectra of the produced nanocluster ions. As depicted in Figure 2.1, the Meso-Q mass filter is placed in line with Nanogen50 cluster source, and it can be set to scan or filter in units of particle mass (amu) or particle diameters (in units of nm). The mass range of the filter is 350-10 6 amu (equivalent to nanoparticles of 1 nm to 10 nm diameter size). The Meso-Q mass filter can be operated either from the front panel of the Meso-Q power supply or from the PC, but it is intended and recommended to be operated primarily from a PC using the supplied software. To acquire mass spectra by using the Meso-Q software, the user has access to choose nanoparticles desired scan range, either in terms of particle diameter or particle mass, and the value of U/V ratio which allows a tradeoff 43 between resolution and flux. The default value for U/V ratio is 0.02, but it can be chosen in any range between 0.001 to 0.168 (practically, it is unlikely that any significant nanoparticle flux passes through the Meso-Q for U/V values above 0.15) and then the software automatically sets U and V voltages by scanning the frequency. The software sets the voltages for the scan differently depending on the chosen initial value of particle size in the scan range. To acquire selective mass filtering by using the front panel controller, the user can change the ac voltage, dc voltage and frequency by ac amplitude (V) dial, dc amplitude (U) dial and frequency (𝑓 ) dial respectively, and then the mass is set automatically by the front panel controller. The ac voltage field will accept values between 0 to 125 volts; however, for the values below 20V, the Meso-Q will not filter reliably because the ac voltage is too low. It is also recommended that in front panel control mode, set the ac voltage (V) to the maximum of 125 volts, then adjust the frequency (The frequency field range is between 1 Hz to 100kHz), to select the desired mass to pass through the filter. The selected mass is shown on the front panel display. By pressing the frequency dial, we can change between current display to the frequency display. To increase the resolution of the Meso-Q, the user can increase the U/V ratio (dc to ac amplitude) by using the dc amplitude dial to increase the dc voltages, for instance, setting for the maximum resolution (U/V=0.168) the U and V values would be 20V dc and 125V ac voltages respectively. A grid as the detector at the exit of the Meso-Q samples the ion flux at the selected mass, and the resultant current is measured and transferred to the controller (displayed on the front of the power supply) by the current amplifier unit. The unit is fitted internally 44 with two 9-volt batteries. (There are a grounded grid and a grounded ring respectively before and after the 18-volt biased detector grid). Figure 2.6: Schematic of quadrupole mass filter a) The trajectories of ions in the quadrupole mass filter; M1 and M3 are ions with unstable trajectories; however, M2 is an ion with a stable trajectory that can pass through the quadrupole. b) The schematic of quadrupole electronic connections; as it is depicted rods are electrically connected as the pair, and equal but opposite sign potential applied to the pairs. U is (dc) voltage and V is the amplitude of the radio frequency (ac) voltage, and 𝜔 = 2𝜋𝑓 is the angular frequency. The distance between the rod tips is 2r o (r o is the field radius). c) Meso-Q mass filter; the quadrupole mass filter made by Mantis Deposition Ltd. and used in our experiment. In the Meso-Q controller, analog to digital convertor (ADC) has specific digits to relay the current from the current amplifier unit. As the user changes the Gain option on a c 45 the software, which is in the range of 0.125nA to 128nA, the current range, as well as the integration time, will change. For the gain range of 0.125nA to 128 nA, the integration time is changing in the range of 409 ms to 2 ms. Therefore, this current value (ion flux) is used as a relative measure of the number of incoming nanoparticles of the selected mass. By scanning across the mass range and measuring the collected current, it is possible to obtain a distribution of nanoparticles according to their size. 2.5 Cluster beam characterization Thus far, we explained the process of the nanocluster growth in the cluster source, the effect of the source operating parameters on the nanocluster size distribution and how a specific size of nanoclusters can be selected from nanocluster ion beams using the quadrupole mass filter. However, to customize nanomaterial synthesis, it is important to go further and characterize the size-selected cluster beam, as well as to study the surface science of preliminary interaction of cluster- surface, film growth and final structure of the film with the help of characterization techniques such as AFM, TEM, XPS, etc. (Millani, and Innotta 1999). Therefore, it is essential to measure the ratio of charged to neutral size-selected nanoclusters that reach to the deposition chamber (Section 2.5.1) and to measure the nanocluster velocity (sections 2.6.1, 2.6.2) as well as to investigate the morphology of deposited nanoclusters and the cluster-surface interaction by controlling the impact energy of landed nanoclusters on a substrate (section 2.7). 46 These measurements and characterization techniques pair the knowledge of solid- state physics and cluster physics together to make the comprehensive understanding in the fabrication and manipulation of thin films made by cluster deposition experiment (Millani, and Innotta 1999). Ratio of charged to neutral nanoclusters The ratio of charged to neutral nanoclusters produced by the cluster source for magnet-type A and magnet type B provided by Mantis Deposition Ltd. was estimated using a homemade setup shown in Figure 2.7 (a) comprised of deflector plates and a quartz crystal oscillator /film thickness monitor arrangement. The deflector plates were mounted on a linear motion MDC feedthrough with a micrometer drive to position them to the center of the nanocluster beam where the mass selected nanocluster beam enters the deposition chamber. The electrostatic potentials that applied to the plates either deflect or reject the ionized nanoclusters and eventually just the neutral nanoclusters will reach the quartz crystal film thickness oscillator (QCM) used to determine the mass of the deposited film by measuring the shift in the oscillation frequency of the quartz membrane. The QCM mounted on the homemade extension piece attached a stepper motor driven linear feedthrough (shown in Figure 2.7(c)) is positioned perpendicular to the beam, right after the deflection plate in the middle of the opening distance. A copper sputtering target with 99.99% purity is used. The ratio of charged to neutral nanoclusters was estimated via measuring the rate of deposition of all clusters (neutral and ionized clusters) by applying no electrostatic 47 potential to the deflector plates and measuring the rate of deposition of neutral clusters by applying the electrostatic potential to the deflector plates. Moreover, this ratio was estimated in different combinations such as for the full scan range and narrow scan ranges in different parts of the log-normal nanocluster beam distribution. The result showed that the ratio of ionized to neutral nanoclusters for magnet- type B is 2:1 (up to ~35% neutral and ~65% ionized clusters) influenced by the power and particle size. And for the magnet type A is ~ 100 % ionized nanoclusters. Consequently, for the copper target, the estimated ratio of ionized to neutral nanoclusters produced by type A and B magnets is indeed close to the company’s claim. In future work, it would be advisable to test this ratio for other target materials. 48 Figure 2.7: The setup to measure the ratio of charged to neutral nanoclusters a) Schematic of the deflector plates and QCM setup. The cluster ion beam passes between the deflector plates and is either deflected or rejected by their electrostatic potential so that only the neutral nanoclusters will reach to the quartz crystal film thickness monitor (QCM) which estimates their rate of deposition. b) Deflector plates and mounting flange. c) QCM and mounting flange: 1) Stepper motor 2) Linear motion micrometer, 3) Electrical connection 4) homemade designed extension length 5) QCM; the QCM setup is mounted from the top flange of deposition chamber and positioned perpendicular to the beam right after the deflector plates in the middle of the gap. 49 2.6 Kinetic energy measurement Retarding field analyzer principle and design The retarding field analyzer (RFA) is an electrostatic device for measuring the kinetic energy distribution and the energy spread of charged particles (Steckelmacher 1973). The kinetic energy is estimated from the maximum height of the potential barriers that can be overcome by the charged particle beam as detected by the collector. The simplest RFA is based on a parallel plate geometry comprising the entrance aperture and the collector, typically a Faraday cup together with its own entrance aperture designed to collect the secondary emission electrons (Simpson 1961). In fact, charged particles are transmitted through an aperture and are analyzed by retarding in the electric field established through bias potentials applied to the grids. The entrance slit must be wide enough to permit adequate flux transmission, and small enough that electrostatic sheath established around the slit edges be large enough to bridge the aperture width. Figure 2.8 shows the homemade retarding field analyzer setup equipped with a Faraday cup used to measure the kinetic energy of size-selected Manganese nanoclusters produced by the magnet type A. The setup can be mounted onto a linear motion feedthrough with the beam axis perpendicular to the RFA entrance opening. 50 The kinetic energy of size-selected charged Manganese nanoclusters (both positive and negative ions) which pass through the quadrupole and find themselves in the deposition chamber is analyzed through the following arrangement: First, constant positive potential Figure 2.8: Retarding field analyzer (RFA) equipped with Faraday cup a) Schematic of RFA; the ion beam, which includes positive and negative charges, is filtered for positive ions by the first grid, and then negative retarding potentials (with 20-volt intervals) are applied to the second grid (“retarding grid”) to slow down the negative ions and the transmitted current is collected on the grounded potential collector. The latter is equipped with a Faraday cup to capture any secondary emitted electrons. The grid electrodes have a circular aperture opening of ~1 cm diameter covered by a 75% transmission wire mesh to create a uniform electrostatic potential on the edges and across the apertures. b) RFA equipped with a Faraday cup and mounting flange. 51 is applied to the first grid which repels all the positive Manganese ions and accelerates the negative ions to the first grid, and then the negative Manganese ions are transmitted through the grid. After having passed the first grid, the negative ions are decelerated and filtered depending on their kinetic energies by the retarding field between the first and second grids (also called retarding grid); therefore, just the negative ions that their kinetic energy is larger than this potential barrier will reach to the collector, and the current is collected with a picoammeter (Keithley 6487). The measured collector current-voltage characteristic is referred to as the I-V characteristic. The collector is equipped with a Faraday cup to trap secondary electrons ejected by the incident ions off the collector surface, and both collector and Faraday cup are kept at the ground potential during current measurement. The ion velocity distribution function 𝑓 (𝜐 ) can be shown (Bohm and Perrin 1992) to be proportional to the first derivative of the collected current with respect to the potential (V) applied to the retarding grid. In the parallel plate geometry RFA, 𝑓 (𝜐 ) is the one- dimensional velocity distribution because all ionized particle coordinated axially in the line of the entrance plane of the analyzer. One finds (Bohm and Perrin 1992); | 𝑑𝐼 (𝑉 ) 𝑑𝑉 | = 𝑒 2 𝑀 𝑓 (𝜐 ) where 𝐼 (𝑉 ) is collected current, 𝑉 is retarding potential, 𝑀 is ion mass, 𝑒 is elementary charge, and the velocity is 𝜐 = √| 2𝑒𝑉 𝑀 | 52 Cluster velocities A measurement of cluster beam velocities was performed to examine the degree to which nanoparticles are susceptible to following the source gas streamlines. During operation, the pressure of the argon and helium mixture inside the source is in the range of 10-100 Pa (Haberland et al. 1992; Hutte 2017). The source walls are cooled by liquid nitrogen, but the stagnation temperature at the exit aperture is expected to be higher. The corresponding mean free path l of the gas atoms lies in the range of ~0.05 mm – 1.5 mm (Haynes 2016). This corresponds to Knudsen numbers Kn=l/d~0.01–0.3, where d=5 mm is the diameter of the exit aperture, placing the expansion in the intermediate to mildly supersonic continuum regime (Hutzler et al. 2012). In this range, atoms and small molecules approach the regime of being fully accelerated by a buffer gas expansion (Hutzler et al. 2012); however, the “velocity slip” phenomenon (Milani and Iannotta 1999) also becomes more and more pronounced as the mass of the diluted species increases. In the context of the present work, it is suggested that increased gas flow promotes the transport of nanoclusters through the interior of the aggregation volume. This decreases the time available for condensation and reduces the average cluster size in the outgoing beam. Therefore, one also would expect that for a given amount of gas flow and a given size distribution, the smaller clusters exit the nozzle with velocities closer to those of the helium and argon atoms, while the larger ones (which are prone to follow the streamlines less efficiently and therefore to spend longer within the condensation region) would exhibit a significantly larger velocity slip. 53 The velocities of negative cluster ions were measured by the retarding potential technique, using a Faraday cup with two grids (the detail is in section 2.6.1), one to repel positively charged clusters and the other to apply a slowing voltage V. The current I of the ion beam was measured by the picoammeter and its kinetic energy distribution was determined by differentiating the I(V) curve and fitting the result with a Gaussian function. A complementary measurement of the ion energies utilizing a quadrupole beam deflector resulted in very close values. Details of these experimental arrangements will be described elsewhere. The velocity distributions of Mn nanoparticles of 2 nm and 9 nm diameter are shown in Figure 2.9. The gas flows for the two cases corresponded to molar fraction ratios within the source of X Ar/XHe=0.7 and X Ar/XHe=3, respectively. The average velocity of the smaller nanoparticles is ~600 m/s while that of the larger ones is much lower, ~140 m/s. Figure 2.9: Velocity distributions of Mn nanoparticles emitted by the source, as determined via retarding potential measurements. (a) Particle diameter 2 nm (aggregation length L=9 cm, Ar and He flow rates Q Ar=150 sccm and Q He=210 sccm, discharge power P=15 W). (b) Particle diameter 9 nm (L=9 cm, Q Ar=150 sccm, Q He=50 sccm, P=15 W). 54 Allowing for assumptions of an ideal gas, isentropic, and compressible flow of argon and helium gas mixture in a free jet expansion, heat conduction and viscous flow get neglected. The energy equation (first law of thermodynamics) would be h +V 2 /2=ho where ho is total or stagnation enthalpy per unit mass and considered constant along any streamline. As the gas expands and cools, enthalpy decreases and the mean velocity increases. For an ideal gas dh=Cp. dt, the Cp is constant and equal to (γ/γ-1) 𝑅 𝑊 , where R is universal gas constant and 𝑊 is molar molecular mass. From the first law, the energy balance relating velocity to temperature would be V 2 =2(ho -h) = 2 ∫ Cp dT 𝑇𝑜 𝑇 and taking into account that the gas cools in the expansion; hence T<<To ;therefore, 𝑉 ∞ = √ 2𝐶 𝑝 𝑇 0 = √2𝑇 0 ((γ/γ − 1)) 𝑘 𝐵 𝑀 So, the maximum or terminal velocity of an ideal gas for γ=5/3 would be 𝑉 ∞ =√ 5𝑘 𝐵 𝑇𝑜 𝑀 , kB =N. R, where N is Avogadro number and kB is Boltzmann constant. Therefore, It is convenient that the terminal velocity for a gas mixture in the continuum expansion regime can be approximated(Cattolica et al. 1979; Miller 1988) by the use of an average mass ii M X M =  , so that for monatomic gases one has ( ) 1/2 0 5/ tB k T M = v (for a purely effusive expansion the forward beam velocity is ≈15% lower (Pauly 2000; Hutzler et al. 2012)). The aforementioned velocity of 2 nm particles is in sensible agreement with the value vt≈650 m/s obtained for a stagnation temperature T0=200 K (as expected, this is somewhat higher than at the source jacket, see above), but the corresponding value for the 9 nm particle source parameters would be ≈500 m/s which 55 is significantly greater than the measured velocity. This implies that the larger nanoclusters display a significant velocity slip. Other groups (Ayesh et al. 2007; Polonskyi et al. 2012; Ganeva et al. 2013) have reported analogously low velocities and evidence of strong velocity slip for the heavier nanoclusters produced by magnetron aggregation sources. These observations support the picture of a more efficient transport of smaller nanoclusters by the gas flowing through the source. 2.6.2.1 Ratio of negatively to positively charged nanoclusters In section 2.5.1 we described our determination of the ratio of charged to neutral nanoclusters produced by the Mantis magnet type A. We found that this magnet produces an almost fully ionized beam. Here, using our retarding field analyzer arrangement, we were able to estimate the ratio of negative to positive ions for the magnet type A during the course of velocity distribution measurements. We found that the ratio of negative (N - ) to positive (N + ) Manganese nanoclusters in the size range between 1nm and 9nm is mostly close to N - /N + =2:1 except for one size where this ratio appears - closer to- 1:1. Although, Mantis claims that magnet type A produces more than 80 % negatively ionized nanoclusters, in reality this percentage can be influenced by various parameters such as experimental conditions, source parameters, target material, and the mass scan size range selected from the log-normal cluster beam distribution. 56 2.7 Deposition The mass selected nanoclusters that see themselves in the deposition chamber are conducted toward the desired substrate to make thin films. Many properties of thin films made by nanoclusters are particularly sensitive to nanocluster morphologies. Moreover, varying the impact energy of landed nanoclusters has a significant effect on preserving or deforming nanocluster morphologies and consequently the quality of thin films. Cluster impact energy Generally, kinetic energy regimes of the landed clusters are categorized into three main regimes: low energy, medium energy, and high energy regimes (Binns2001). In low energy regime with an impact energy of ~0.1 eV/atom cluster, also called as the soft-landing regime, clusters preserve mostly their morphology, physical, and chemical properties as the free clusters when they are landed on the substrate. A porous film is formed in a way that randomly arranged clusters on the surface diffusing to their adjacent clusters to make islands, and by the time the surface is saturated by the density of islands, they will join to make the bigger ones till the continuous film is formed (See Figure 2.10, part a, d). In medium energy regime with an impact energy of 1-10 eV/atom cluster, the clusters are not shattered, but their morphology is changed on impact and pinned into the arrival spot and may make some defects on the surface, and finally, they make the granular films (See Figure 2.10, part b, e). 57 In high energy regime with impact energy of >10 eV/atom cluster, the clusters are shattered or fragmented, and the surface can be destroyed in several layers deep; therefore, the initial morphology and properties of free clusters will completely change after they are landed on the surface, and the film made by them is pretty dense, and it can be used in coating processes (See Figure 2.10, part c, f). There are different techniques to investigate the morphology and properties of thin films such as AFM, XPS, TEM, SEM, etc. In the present work, we mainly used AFM technique to determine the size and morphology of deposited nanoclusters. Figure 2.10: Molecular dynamics simulations of the impact of individual Molybdenum clusters (Mo1043) and the morphology of the cluster-assembled films created on a Mo (100) surface as a function of the cluster energy. Mo 1043 cluster with (a) 0.l eV, (b) 1eV, (c) 10eV kinetic energy per atom impinging on a Mo (100) surface (Haberland et al. 1995). Morphology of a film formed by Mo 1043 cluster with (d) 0.l eV, (e) 1eV, (f) l0eV kinetic energy per atom (Haberland et al. 1995). 58 Characterizing nanocluster thin films with AFM For the first time in 1986 Binning et al. demonstrated the idea of atomic force microscopy (AFM) as an ultra-small probe tip at the end of a cantilever. (Binning et al. 1986) The contour of the surfaces is mapped by approaching the tip over the surface and measuring the forces (such as van der Waals, electrical, magnetic, thermal) between the probe and the sample surface. As the cantilever moves over the surface to map surface topography, the reflection of the laser beam from the back of the cantilever to the center of the detector is displaced as shown in Figure 2.11. Then a feedback loop uses the laser deflection to control the force and tip position (Haugstad 2012). Different operating modes of AFM (contact mode and tapping mode) can be applied to the samples in the air or fluid. Figure 2.11: Schematic of Atomic Force Microscopy (AFM) As the cantilever moves over the surface to map surface topography, the reflection of the laser beam from the back of the cantilever to the center of the detector is displaced. Then a feedback loop uses the laser deflection to control the force and tip position. 59 In the contact mode, cantilever tip is dragged over the sample maintaining the deflection (repulsive force) at a constant order. However, in the tapping mode, the cantilever is oscillated at or slightly below its resonance frequency with constant amplitude oscillation, the tip lightly taps on the sample surface during scanning and has minimum interaction with the sample; therefore, the nanoclusters are not displaced, and their morphology is kept intact (Zhong et al. 1993). 2.7.2.1 Sample preparation for AFM imaging In the present work in order to prepare the samples for AFM imaging, the desired substrates are mounted on the homemade substrate holder protected with collimator shield (shown in Figure 2.12). The setup can be mounted onto linear motion feedthrough with the beam axis perpendicular to the collimator shield. In each deposition trial, just one substrate is moved to the collimator opening and exposed to the beam, and the other substrates are protected from contamination by the shield. Figure 2.12: Substrate holder from the front and top views 60 The samples are prepared by depositing the size-selected metal nanocluster in low energy regime on the substrates, and then they are imaged using AFM in the air (Veeco Innova operated in tapping mode, using silicon probes with the aluminum reflex coating, Tap 150Al-G, Budget Sensors), to determine the size and morphology of deposited nanoclusters. Figure 2.13 depicts the AFM topography images of 3nm Copper nanoclusters in different coverages. The depositions were carried out on highly oriented pyrolytic graphite (HOPG) that is an ideal substrate for these studies since it can easily be cleaned by a freshly cleaved sample. The surface atomically has over large regions separated by steps. It is observed in part (b) of Figure 2.13, nanoclusters formed islands on the HOPG terraces, and along the terrace edges, they shaped wire-like nanocluster chains. Those clusters which are closer to edges move/diffuse toward that and form chains alongside edges. Clusters landing on the surface far from the edges meet each other and organize into islands. To verify the size selectivity of quadrupole mass filter, samples of the size selected soft-landed Mn nanoclusters on Si/SiO2 substrates were prepared and imaged by AFM. Figure 2.14 shows nearly spherical shaped particles, indicating that they do not shatter or deform to a significant degree. Histograms of height profiles clearly confirm the prevalence of selected nanoparticle diameters. 61 Figure 2.13: AFM images of 3nm Copper nanoclusters on HOPG in different coverages 3nm copper clusters with different coverages were deposited on HOPG substrate. As depicted the growth process depends on the amount of coverage of nanoclusters on the surface. a) It shows less than 10% coverage; Individual nanoclusters were landed on HOPG far a way of each other, and there is no trace of making islands. b) It shows about 30% coverage; There is the trace of islands on terraces and wire-like nanocluster chains arrangement along terrace edges for those clusters that were close to the edges and moved toward that and make the chain shape arrangement. c) It shows more than 80 % coverage; There is almost the continuous film of nanoclusters. 62 Figure 2.14: AFM result AFM tapping-mode images and histograms of height distributions of size-selected Mn nanoparticles deposited on Si/SiO 2 substrates with the mass filter set to a diameter of a) 4 nm, b) 6 nm, c) 8nm. AFM tip convolution makes the images appear much wider than actual nanoparticle size, however profile height measurements confirm the good size selectivity of the deposition setup. (The insets show the profiles of sample particles which are marked in the image panels.) 63 2.8 Conclusion I have presented a detailed study of the influence of the main operating parameters of a magnetron/condensation nanocluster source on the particle size. Specifically, I investigated how the peak size of the nanocluster ensemble responds to changes in the argon and helium gas supply flow rates, in the discharge power, and in the aggregation length. A benefit of such a cross-correlation study is that it allows one to classify the main physical role played by each of the variables. The sputtering power supplied to the discharge, and the argon flow are the crucial parameters for nanocluster production. The discharge supplies the metal vapor for building the nanoparticles, while argon is not only responsible for the sputtering process but also is the dominant player in three-body collisions that provide the condensation nuclei triggering further growth. Once the discharge and nucleation processes are stabilized, the next dominant factor is the source residence time, i.e., the length of time over which aggregation of the cryogenically cooled highly supersaturated metal vapor is allowed to proceed. If not terminated, it would result in the formation of large “smoke” particles both by addition of individual atoms and by binary cluster-cluster collisions (Pfau et al. 1982; Zimmermann et al. 1994). Hence for obtaining a population of sufficiently small nanoclusters, it is essential to sweep the aggregating medium out of the source at an adequately fast rate. This is the main role of the helium supply. It is much less efficient than argon at promoting nucleation and aggregation, but an increase in the helium flow raises the pressure and density of the 64 gas column inside the source, resulting in stronger entrapment of nanoparticles within the gas streamlines. This reduces their residence time and enhances the population of smaller particles in the beam. A measurement of the kinetic energies of nanocluster ions exiting the source supports the preferential entrapment of smaller nanoclusters by the gas flow: 2 nm particles followed the terminal velocity of the gas expansion, while 9 nm ones displayed a significant velocity slip. The two variables, the helium supply rate and the aggregation length (controlled by shifting the magnetron head with respect to the source exit aperture) have the dominant influence on the average nanocluster size in the outgoing beam. The conclusions guided by systematic studies of source operation are useful for optimizing source performance and are fruitful in untangling specific physical processes taking place within the dynamic sputtering/condensation source environment. It would be possible and interesting to gain further insight by exploring the above variables over a still wider range of values, as well as by adding new ones, for example other types of noble gases, variable source wall temperature, precise control of internal source pressure, etc., and by position- and time-resolved spectroscopy of the contents of the source interior. Finally, the size and morphology of deposited nanoclusters were characterized by atomic force microscopy (AFM). The AFM images verify the nanoclusters are nearly spherical shaped particles indicating that they are soft landed. Histograms of height profiles show the close correspondence between the selected and imaged nanoparticle sizes and confirms the accuracy of the mass filter. 65 3 Electrocatalysis application of size-selected MnO nanoparticle thin films in the process of water splitting 3.1 Introduction Since the early days of cluster science, unique physical and chemical properties of clusters have been divulged by changing cluster sizes. Therefore, scientists ponder the size- dependence of the catalytic activity of metal clusters to modify the selectivity and reactivity of reactions. Accordingly, it has been proposed to occur new and different geometric arrangement combined with changing cluster surface energies and the number of isomers through changing the size of clusters. Likewise, as the cluster sizes change, the evolution of the electronic structure has been envisaged as another important factor to influence the chemical and catalytic properties of clusters. Now a day, experimental studies of nanocatalysis conducted by either physical methods or chemical methods (also known as wet chemistry) along with theoretical and computational studies have demonstrated unique catalytic activity properties of small clusters just by changing their sizes (Halder et al. 2018, Heiz and Landman 2007). For instance, in the wet chemical synthesis processes, the breakthrough research results have highlighted the effect of metal nanocluster sizes (preferably below 10 nm) on their catalytic activity (Wang and Astruc 2017) and pointed out the significance of new synthesis methods to control the size and shape of nanocatalysts (Zhou et al. 2011), such as 66 manipulating the kinetics involved in the early stage of the nanoparticle nucleation and growth processes (Peng et al. 2015). Despite the remarkable progress made in wet chemical synthesis, experimentally the full control over the size, shape, and purity (ligand-free) of nanocatalysts is still the great subject to challenge due to competition and collaboration growth kinetics and thermodynamics during chemical synthesis. Alternatively, in physical methods, recent developments in cluster sources, mostly developed in the gas phase processes and joined with an appropriate mass filter, make promising to have size selectivity from the nanocluster beam size distribution. Therefore, the vast studies of size-selected free nanoclusters and their thin films (in the range of high porosity to the dense films) are feasible by cluster source technique due to control the nanocluster sizes. Moreover, this technique enables controlling the kinetic energy of the size-selected nanoclusters throughout their deposition on surfaces. The deposition process takes place under high vacuum conditions, and no organic ligands or surfactant are involved during nanocluster productions. Therefore, along with controlling the size and kinetic energy of nanoclusters in the gas phase, purity can be well preserved. In conjunction with experimental studies, theoretical and computational studies are also applied in understanding the kinetics and reaction mechanism to elucidate how to manipulate the structure of nanocatalysts to optimize their catalytic activity and selectivity. (Greeley et al. 2007, Heiz, and Landman, 2007, Barron et al. 2017). Indeed, with such advanced experimental, theoretical and computational techniques in the production of nanomaterials, nanocatalysis is the rapid growing field to bring in new 67 types of catalysts to the market with the advantage of having the qualities better than the conventional ones such as showing higher energy efficiency, reducing chemical wastes as well as being cost-effective and abundant. In this chapter specifically, we study the electrocatalysis application of size- selected MnO nanocluster thin films in the process of water splitting. 3.2 What is water splitting and what are its challenges? Water splitting using electricity named water electrolysis is the process that a water molecule breaks into its elements, hydrogen, and oxygen. A basic electrolysis cell consists of an anode, a cathode, an alkaline electrolyte solution, and a power supply, as depicted in Figure 3.1. The main process involves two half-reactions, namely, Oxygen Evolution Reaction (OER) at the anode and Hydrogen Evolution Reaction (HER) at the cathode (Santos et al. 2013). Cathode: 2H + + 2e - → H2 (1) Anode: 2OH - → ½ O2 + H2O + 2e - (2) The overall chemical reaction of water electrocatalysis: H2O → H2 + ½ O2 It requires the minimum 237.2 kJ/mol input energy. (3) 68 Although hydrogen as the clean source of energy is produced by hydrogen evolution reaction (HER) shown in equation (1) (at 25 o C and 1atm), it depends on the oxygen evolution reaction (OER) shown in equation (2) to provide the electron required to reduce proton (H + ) and produce H2. However, oxygen evolution reaction (OER) itself is a strongly uphill reaction with a large overpotential to generate measurable current densities. Moreover, commercial electrolyzers typically operate at a cell voltage of 1.8 – 2.0 V which is much higher than the theoretical minimum value of 1.23 V, attributed to the complex electron and ion transfer processes that make slow the kinetic reactions and consequently poor energy efficiency. (Zeng and Zhang 2010) Figure 3.1: Basic scheme of a water electrolysis system Water electrolysis unit consists of an anode and a cathode connected through an external power supply and immersed in a conducting electrolyte. A direct current (DC) is applied to the unit; the electrons flow from the positive terminal of DC power supply to the cathode, where they are consumed by hydrogen ions. (Santos et al. 2013). 69 Therefore, an electrocatalyst requires to be applied in this process to decrease this large overpotential, and that should meet two main expectations: one of them is that being highly active; in a way that with decreasing the overpotential, preferably close to the theoretical minimum value, enables producing a large amount of current. The other one is to have long term stability. To date, Pt group metals are known the best catalyst for efficient HER reactions, and Ir and Ru based materials have demonstrated the highest OER activity; however, they are either rare materials that compel the high cost or they are toxic materials that some environmental restrictions prohibit their usage. Consequently, it makes the urge to investigate and develop the earth abundant, cost-effective, nontoxic and highly active catalyst for both OER and HER. (Wang et al. 2016) Delve into nature, most of the oxygen in the atmosphere is generated by plants, algae, and cyanobacteria by the photoinduced oxidation of water to oxygen. Pirson in 1937observed that plants and algae which do not have Manganese in their growth medium lost the ability to evolve oxygen and adding of Manganese element to the growth medium will make the water oxidation within 30 minutes. Further research in biological chemistry proves that in nature, the water oxidation is catalyzed in the photosystem II (PSII) with the Mn4CaO5 cluster known as the oxygen-evolving complex (OEC) in chloroplasts (shown in Figure 3.2), and aerobic respiration releases oxygen from water oxidation in mitochondria found in all aerobic photosynthetic plants (Yachandra et al. 1996). 70 Therefore, the unique role of Manganese as an earth-abundant transitional metal cluster in photosynthesis aroused questions about the relationship between Mn - biogeochemical cycling and the evolution of this key component of photosynthesis. Since then, emulating photosynthetic water oxidation in higher plants with inorganic metal oxides has been a long-standing challenge in electrochemistry (McEvoy and Brudvig 2006; Nocera 2012). The process has a special appeal because of its relevance to solar-to-fuel light harvesting, in which a photo-voltage is used to oxidize water and to drive uphill reduction reactions, thereby storing the light energy in chemical bonds. As we mentioned above, plants achieve water oxidation using the photoexcitation collected by chlorophylls and funneled into OEC. The mechanistic details of water oxidation by OEC have been studied for decades (Ferreira et al. 2004; Nelson and Ben-Shem 2004; Yano et Figure 3.2: Structure of the Mn4CaO5 cluster Structure of the Mn 4CaO 5 cluster consists of five metals and five oxygen atoms. Three manganese, one calcium and four oxygen atoms form a cubane-like structure in which the calcium and manganese atoms occupy four corners and the oxygen atoms occupy the other four. The bond lengths between the oxygens and the calcium in the cubane are generally in the range of 2.4–2.5 Å, and those between the oxygens and Manganese are in the range of 1.8–2.1 Å (Umena et al. 2011). 71 al. 2006; Kärkäs et al. 2014) and still remain at the forefront of research (Khan et al. 2015; Barber 2017; Vinyard and Brudvig 2017). Inspired by the natural OEC complex, it seems reasonable to study the water oxidation problem in an artificial material using manganese oxide minerals as electrocatalysts or photocatalysts. Several forms of manganese oxides have been studied for this purpose (Gorlin and Jaramillo 2010; Najafpour et al. 2010; Takashima et al. 2012; Zaharieva et al. 2012), however not as extensively as other first row transition metal oxide films (Harriman et al. 1988; Artero et al. 2011; Sivula et al. 2011; Singh and Spiccia 2013). Many forms of MnOx minerals are predominantly poor conductors and are also studied in the context of materials for capacitors (Wei et al. 2011; Li et al. 2012). Using them in bulk form as thick electrocatalytic films on an electrode demands that the material function as a catalyst on the surface and as a conductor within the bulk of the film at the same time. This dual requirement is very different from the natural photosynthesis, in which the OEC is a nanocluster catalyst and is not required to conduct charge over large distances. For this reason, it is productive to resort to studying MnOx nanoparticles, so that the catalytic properties are decoupled from transport properties as much as possible. Manganese oxide nanoparticles produced by solution-processed methods have been used for water oxidation (Robinson et al. 2013; Guo et al. 2014; Ramírez et al. 2014; Wiechen and Spiccia 2014; Kuo et al. 2015; Li et al. 2016). However, there are challenges associated with these methods: the associated organic ligands can pose problems, and the particle sizes and the total amount of catalyst loading on the electrodes may not be precisely controlled. Furthermore, they are often restricted to producing nanoparticles larger than 10 72 nm. To synthesize bare nanoparticles without using any solvents, ligands, or surfactants, gas-phase production techniques have been developed (Grammatikopoulos et al. 2016). Nanocatalyst particles for water splitting applications have been produced by using metal vapor condensation sources based on direct-current (dc), see, for example (Srivastava et al. 2014; Stranak et al. 2015; Paoli et al. 2016; Chen et al. 2017; Fan et al. 2017; Lin et al. 2017; McInnes et al. 2017; Srivastava et al. 2017), reactive dc (Patel et al. 2016), and radio- frequency (rf) (El Koura et al. 2016; Kim et al. 2016) magnetron sputtering, as well as by using other gas-phase sources such as ion sputtering (Masudy-Panah et al. 2016; Sreedhara et al. 2017) and laser vaporization (Kwon et al. 2013). To the best of our knowledge, though, these techniques have not been employed to prepare bare MnOx nanoparticle films for water oxidation catalysis. (Water deprotonation by small manganese oxide clusters in a molecular beam was studied by Lang et al. 2016). In this work, we report the production of electrocatalytic films by low-energy deposition of manganese oxide nanoparticles with sizes smaller than 10 nm on a conducting substrate under well-defined flux conditions. The films were used for water oxidation in a conventional three-electrode cell. We show evidence that the size of nanoparticles is correlated with the overpotential for water oxidation, with clear improvement in catalytic activity when the sizes are made smaller. Two aspects of our approach make it possible to reveal the effect of size on catalytic activity. The first is fine control over particle size, which is achieved by using a deposition system that combines a nanoparticle source and a mass filter. The second aspect is a careful determination of the total amount of catalyst material deposited, or catalyst loading, which 73 is important for the comparison of different samples. As described below, catalyst loading in our work has picogram resolution in the total quantity of mass deposited over an area of ∼1 cm 2 . Such precision in the control of catalyst loading exceeds most conventional approaches based on measuring the thickness of spin-cast films. Using these capabilities, we studied the catalytic activity of nanoparticles with diameters of 4 nm, 6 nm, and 8 nm. These particles are larger than the natural OEC which has dimensions of about 0.5 nm × 0.25 nm × 0.25 nm (Yano et al. 2006), and therefore they can serve to bridge the gap between bulk film electrocatalysts and the natural molecular-scale OEC-type systems. 3.3 Experiment and characterization Production and deposition of nanoparticles Size-selected nanoparticle deposition is performed in the system shown in Figure 2.1, chapter 2. Figure 3.3 shows the overall size range of Mn nanoparticle ions in the beam; vertical lines mark the sizes selected for deposition in this work. As it is shown in Figure 2.1, the source produces nanoparticle ions by magnetron sputtering of atoms from a metal target followed by their condensation within a flowing cold inert gas (Huttel 2017). The Nanogen-50 (Mantis Deposition, Ltd.) is used in the present setup. It generates metal vapor by dc sputtering of 2-inch diameter metal targets, here 99.95% Mn (ACI Alloys). The magnetron block is equipped with the manufacturer’s “magnet type A” whose benefit lies in producing almost exclusively ionized particles (see section 2.5). This makes it possible to filter and manipulate the entire beam by electric 74 fields and to avoid the deposition of randomly-sized neutral species. Argon gas (99.999% purity) is introduced into the source region behind the magnetron head, and the magnetron itself is mounted on a linear translator. The interplay between the source parameters has been discussed in detail previously in chapter 2. In the present case, the flow rate of Ar was 190 sccm, the distance between the magnetron and the beam exit aperture was 9 cm, and the discharge power was 10 W. The sputtered metal atoms enter the condensation zone where they undergo collisions with the cold inert gas (the source jacket is filled with liquid nitrogen) and nanoparticle growth takes place as the mixture moves towards the 5 mm diameter exit aperture. The resulting directed beam passes through a 6 mm skimmer followed by a quadrupole mass filter. The standard Figure 3.3: Overall size range of Mn nanoparticle ions in the beam. The overall size range of Mn nanoparticle ions in the beam. The vertical lines mark the sizes selected for deposition in this work. Source operating parameters: argon flow rate was 190Sccm, aggregation length was 9 cm, and the discharged power was 10 W. 75 mass range of the Mantis MesoQ unit is from ~300 amu to ~10 6 amu, but its performance can be extended somewhat to either side of this range. The size-selected nanoparticle ions then find themselves in the main deposition chamber (base pressure 10 -8 Torr). After collimation by a 1 cm × 1 cm mask, the beam lands on room-temperature glass slides covered with fluorine-doped tin oxide (FTO) conducting film. Prior to loading into the substrate holder, the slides are sonicated for five minutes in acetone and methanol, washed with deionized water, and dried with nitrogen gas. The conducting substrates are connected to ground potential, and the impinging nanoparticle ions become neutralized. The beam velocities are in the range of 200 m/s (Ayesh et al. 2007; Ganeva et al. 2013; Khojasteh and Kresin 2017) which corresponds to less than 0.1 eV kinetic energy per Mn atom. Being significantly below the atomic binding energy, this is generally regarded as the “soft-landing” regime, which avoids significant fragmentation and deformation of the nanoparticle (Popok et al. 2011). The described setup allows for the deposition of pure, size-selected nanoparticles with a size resolution of ±2% (Baker et al. 1997). This type of size control is difficult to achieve by solution-processed metal oxide nanoparticle synthesis. For the electrochemical measurements, care was taken to produce films with the same total amount of deposited metal: 1.7 μg of 4 nm, 6 nm, and 8 nm diameter nanoparticles over the 1 cm 2 FTO slide sample area. The deposition rate, which can be converted into the mass loading, was measured using both a quartz crystal film thickness monitor and a Keithley 6487 picoammeter. Both tools yielded consistent results. The substrate ion current ranged from 0.065 nA for the 8 nm diameter particles to 0.1 nA for 76 the 6 nm and 4 nm ones. The corresponding deposition times, adjusted to ensure equal mass loading, were 176 min, 55 min, and 36 min for the 4 nm, 6 nm, and 8 nm particle samples, respectively. Knowing the cross-sectional area of one nanoparticle (assumed to be round, as supported by the atomic force microscope (AFM) images shown in Figure 2.14 (chapter 2)) and the total number that is deposited, we can estimate the covered surface area. It is found to correspond to the sub-monolayer regime: approximately 80%, 60% and 45% coverage by the 4 nm, 6 nm, and 8 nm particles, respectively. The origin of this scaling is as follows: Since the mass of a single nanoparticle of diameter d is proportional to its volume, the total number of nanoparticles required for a fixed mass loading is proportional to 1/d 3 . On the other hand, the surface projection, or shadow, of each such nanoparticle covers an effective surface area proportional to d 2 . Hence if the diameter of the deposited nanoparticles is varied while keeping their total mass constant, the surface area covered by the film will vary as 1/d. The percentages listed above for the actual samples differ only slightly (~10%) from this scaling due to rounding- off variations in the experimental deposition times. AFM characterization To verify the nanoparticle sizes, we prepared samples of size-selected soft-landed Mn particles on Si/SiO2 substrates and imaged them using an AFM (Veeco Innova operated in tapping mode using silicon probes with an aluminum reflex coating, Tap 150Al-G, BudgetSensors). The AFM images in Figure 2.14 (chapter2) show nearly spherical shaped 77 particles, indicating that they do not shatter or deform to a significant degree. Histograms of height profiles clearly confirm the prevalence of selected nanoparticle diameters. XPS characterization X-ray Photoelectron Spectroscopy (XPS) is a surface analytical technique to characterize the chemical properties of materials. It is used to precisely quantify the elemental composition, including different electronic and chemical states of the material. Surface analysis by XPS is carried out by irradiating monoenergetic soft X-ray beam on the sample. Usually, Mg or Al Kα X-rays are used with energies 1.25 keV and 1.49 keV respectively. Except for helium and hydrogen, all other elements can be detected by XPS. The penetration depth of these X-rays is very limited. It can make depth analysis between top 2 atomic layers up to 20 atomic layers, and it is the least destructive technique between other spectroscopy techniques. To obtain depth chemical analysis of the multilayer coatings (above 10 nm), ion beam etching is employed in combination to XPS where the energetic ions beam can etch the surface of the multilayer coating and simultaneously analyzed by XPS. The basic principle of XPS is shown in Figure 3.4. An X-ray photon with energy ℎ𝑣 hits the surface of the material and knocks out an electron. Therefore, an electron with binding energy Eb with respect to the Fermi level is emitted from the atom with a kinetic energy Ek due to photoelectric effect by leaving a core hole behind. The binding energy Eb of the electron can be calculated by 𝐸 𝑏 = ℎ𝑣 – 𝐸 𝑘 – 𝜙 78 Here, ℎ is Planck constant, 𝑣 is the frequency of the radiation, 𝜙 is the work function of the spectrometer, and Ek is the kinetic energy of the emitted electron. The obtained spectrum by the analyzers (for each photoelectron) consists of the characteristic peaks of each element with respect to their kinetic energies. From the binding energy and intensity of a photoelectron peak, the elemental identity, chemical state, and quantity of a detected element can be performed. Additional details are available in the literature (Moulder 1992). 3.3.3.1 XPS result Once the samples are removed from the vacuum chamber, the Mn nanoparticles oxidize. To ascertain their composition, an x-ray photoelectron spectroscopy (XPS) measurement was performed on 6 nm particles deposited on highly oriented pyrolytic Figure 3.4: The principle of XPS 𝝓 79 graphite (HOPG) in a 19 nm (i.e., 3.3 monolayers) thick film. HOPG was chosen for this measurement to avoid any substrate contamination of the XPS lines of interest. The measurement was done in a Kratos Axis Ultra spectrometer at room temperature. An Al anode operating at 60 W generated an x-ray beam that was focused at different spots on the HOPG surface with multiple measurements at each spot. The spectra were referenced to a C1s binding energy of 284.8 eV in all cases. To quantify the elemental concentrations, the peak areas were determined after background subtraction. The spectra of adventitious adsorbates were estimated by performing XPS measurements on an HOPG surface without Mn nanoparticles. The XPS data were fitted using Casa XPS software. In the XPS spectra of manganese oxide, the Mn 3s and 2p regions enable straightforward identification of the oxidation state (Di Castro and Polzonetti 1989). The Splitting ΔE of the Mn 3s line shown in Figure 3.5(a) and the satellite feature in the Mn 2p3/2 region highlighted in Figure 3.5(b) are unequivocal signatures of the MnO stoichiometry, therefore excluding other types of manganese oxide such as Mn 2O3 or MnO2. 80 Figure 3.5: XPS spectra XPS spectra of a) 3s and b) 2p regions of nanoparticles deposited on HOPG. The peak positions and the 2p satellite feature identify the material as MnO. 81 UV-Vis spectrum UV-Vis is a simple method to insight into the electronic and atomic structure of the materials. It is based on measuring how the materials interact with light as the percentage of reflection, transmission, or absorption materials as a function of photon energy. It is possible to perform different modalities of measurement on different types of samples, such as solid films, liquid solutions, nanostructured surfaces, powders, etc. Transmittance, diffuse transmittance, specular reflectance, and diffuse reflectance are examples of those modalities. In this study, the choice of modality was based on the features of our samples (Soares 2014). 3.3.4.1 Bandgap measurement It is well known that the band gap in semiconductor nanocrystals is increased compared to the bulk value as a result of size quantization, see, e.g., (Kittel 2005). It is, therefore, to be expected that our MnO nanoparticles also exhibit an increased bandgap. Unfortunately, the minute optical density of their submonolayer films precludes a straightforward analysis using ultraviolet-visible (UV-Vis) spectrophotometry, especially considering that the FTO slides have an absorption edge at essentially the same wavelength (Guan et al. 2012) as the MnO bandgap of 3.6 eV or 344 nm (Madelung et al. 2000). Therefore, in order to derive an estimate of the gap, we obtained a diffuse reflectance spectrum (DRS) of the same thick 6 nm sample on HOPG as was employed for the XPS measurement described above. Diffuse reflectance is used for samples like powder, some ceramics, and nanostructured films. In this modality, the reflected light from the surfaces 82 with the roughness scattered in different directions and needs to be collected by specially designed mirrors or integrating spheres (Soares 2014). In this study, the measurement was performed using a PerkinElmer Lambda 950 UV/Vis spectrophotometer. It has been reported in the literature, see, e.g., (López and Gómez 2012) that energy gaps of nanoparticle films can be estimated using the Kubelka-Munk function F( hν). Here ν is the light frequency, and F is obtained from the measured reflectance R(hν) as (Hecht 1976) F=(1-R) 2 /2R It is approximately proportional to the extinction coefficient, and thereby to the nanoparticle absorption coefficient α ( h ν ) . Furthermore, for semiconductors, the optical bandgap can be determined using the Tauc extrapolation (Tauc 1966, 1968) of the rising edge of the plot of ( α h ν) 1/n versus hν, where the value of n depends on the nature of the transition. For bulk semiconductors with a direct bandgap n = ½, while it was recently argued (Feng et al. 2015) that n=1 is a more satisfactory value for semiconductor nanocrystals. Figure 3.6 shows fits corresponding to both exponents. Despite the approximate nature of the DRS extrapolation, the data support the conclusion that the investigated nanoparticles possess a larger bandgap than bulk MnO. 83 Figure 3.6: Bandgap of the 6 nm diameter MnO nanoparticles. The bandgap of the 6 nm diameter MnO nanoparticles estimated from extrapolations of the edge of the Kubelka-Munk function obtained from diffuse reflectance data (see text for details). Both fits indicate a bandgap larger than that of the bulk material, indicated by the green arrow. Electrochemical data A Gamry Reference 3000 potentiostat and a three-electrode cell were used with an Ag/AgCl reference electrode and a graphite rod counter electrode. The electrolyte solution was 0.1 M KOH. The cell was purged with nitrogen gas prior to each scan and then was closed to the ambient air. Linear sweep voltammetry experiments were carried out at 10 mV/s scan rate, between 0 and 2 V. The potentials shown in Figure 3.7 are referenced to the normal hydrogen electrode (NHE) by adding +197 mV to the applied potentials. By measuring the electrodes surface area in cm 2 , we calculate the current density by simply dividing the current values by the surface area. The electrochemical stability of the nanoparticle-coated anodes was also tested. As illustrated in Figure 3.9, even the submonolayer film of the smallest, 4 nm, nanoparticles 84 exhibited no degradation over the scale of at least hours. The other samples behaved analogously. Figure 3.7: Electrochemical data measurement The electrochemical water oxidation current versus potential (Normal Hydrogen Electrode) for three electrodes covered with 8, 6, and 4 nm nanoparticles of MnO at the equal total amount of material. The data shows enhanced activity (lower onset potential) with decreasing particle size. Figure 3.8: Tafel plot Tafel plot representation of the data. The slopes in the region of the current onset are 175, 200, and 250 mV/decade for the 4, 6, and 8 nm nanoparticle films, respectively. 85 3.4 Results and discussion The MnO-covered FTO electrodes were used as the anode for water oxidation in a three-electrode cell as described above. Figure 3.7 shows the current density as a function of applied potential for the three different MnO nanoparticle sizes. The onset potential, defined as the voltage corresponding to a current density of 1 mA/cm 2 , is 1.2 V, 0.9 V, and 0.8 V for the 8 nm, 6 nm, and 4 nm nanoparticles, respectively. The respective currents due to water oxidation at 2.0 V are 6 mA, 9 mA, and 11 mA. Since the mass loading of nanoparticles for all three scenarios was identical, we attribute the enhancement of catalytic activity to the decrease in nanoparticle size. This is the main finding of our work. Figure 3.9: Stability measurement Water oxidation current of the electrode containing 4 nm nanoparticles, measured near the reaction threshold over a one-hour period. Films of the larger nanoparticles behaved analogously. 86 The influence of particle size and composition on catalytic activity is well known (Heiz and Landman 2007; Najafpour et al. 2015; Tyo and Vajda 2015; Vajda and White 2015; Wang and Astruc 2017). For example, Jin et al. (2015) synthesized MnO nanocrystals ranging in size from 10 nm to 80 nm and reported that the smallest particles showed the lowest onset potential for water oxidation. In a further study, the same authors (Jin et al. 2017) investigated water oxidation by 300 nm thick films of 10 nm MnO nanoparticles. They proposed that the electrokinetic steps in the nanocatalysts were distinct from their bulk counterparts. Slopes fitted to the Tafel plot representation of our data in Figure 3.8 are somewhat larger than the values of ~70-120 mV/dec by Jin et al. (2015) for their MnxOy nanoparticles; however, the latter were fitted to a steeper part of the current- voltage curve. In point of fact, the nonlinear shape of the plot underscores that the electrochemical reaction rate in the present case is not controlled by a single-electron charge transfer process (Kear and Walsh 2005). In light of the above studies, our results follow the general trend that smaller particles are better catalysts than larger particles and bulk. However, the previous studies all concentrated on nanoparticles of 10 nm size or larger and often did not maintain a constant catalyst mass when comparing different particle sizes. In the present work, in contrast, we have focused on a systematic comparison of substantially smaller particles at constant mass loading. In addition, as mentioned above, the nanoparticles in this work formed sub-monolayer films on the substrate; hence, concerns about porosity, diffusion, and electrolyte transport limitations are not relevant, unlike in cases when thick films of 87 nanoparticles are studied. Therefore, our observations reveal inherent differences in the catalytic properties of MnO nanoparticles. A variety of nanometer size effects have been found in electrocatalytic materials which can be divided into two types (Li and Frenkel 2017). First, are the ‘geometric size effects’ which derive from the increased surface-to-volume ratio. The smaller nanoparticles have larger surface-to-volume ratios and therefore higher densities of catalytically active sites. Second are the size effects that perturb the intrinsic electronic and lattice structure of the particles. A prime example is a change in the band gap of semiconductor nanocrystals with size, as addressed in the preceding section. Another illustrative example is that Co(II) oxide nanoparticles smaller than 8 nm, when dropped into water, spontaneously evolve hydrogen, while larger particles do not (Navrotsky et al. 2010). Having a larger number of active surface sites can multiply the count of charge transfer reactions proportionally, and with them the electrode current. However, the key observation that the smaller nanoparticles initiate catalysis at a significantly lower onset potential signifies that an inherent change takes place in the charge transfer energetics itself. This, in turn, implies that size effects in the electronic structure, as indeed commonly observed in nanoparticles, are likely to be responsible for the observed onset potential variation. At its core, the catalytic reaction is related to the fact that the reactant-surface interaction opens up additional quantum-mechanical transition channels which vastly enhance the reactant-product Franck-Condon factor (Kresin and Lester 1992). Therefore, an increase in surface electric fields characteristic of nanocatalysts (Zhu et al. 2017) may 88 underlie the observed lowering of the onset potential. Indeed, analogous to the discussion of the diffusion voltage in semiconductor junctions (Ibach and Lüth 2009) a dipole layer forming at the surface-electrolyte interface will give rise to a potential difference across the junction whose magnitude grows with the energy bandgap of the material. Consequently, the fact that the bandgap rises with decreasing nanoparticle size is accompanied by a strengthening of the surface dipole field, which enhances the nanocatalyst’s efficacy. A direct investigation of such surface fields is difficult by electrochemical methods alone and lies beyond the scope of this study, but a future careful characterization of the electric fields at nanocatalyst surfaces by a combination of theoretical and spectroscopic methods would be highly informative. 3.5 Conclusion Sub-monolayer films of size-selected manganese nanoparticles, ranging from 4 nm to 8 nm in diameter, were produced on FTO substrates by low energy beam deposition. After they oxidized in air, the films’ electrochemical catalytic activities for the water- splitting reaction were investigated. It was found that the onset potential decreases with decreasing particle size. To the best of our knowledge, this is the first systematic study of the size effect in water oxidation for pure (ligand-and surfactant-free) sub-10 nm MnO nanocatalysts. The ability of individual size-selected Mn-based nanoparticles to lower the water oxidation threshold illustrates the potency of nanocatalysts in improving the efficiency of important chemical reactions, highlights their interesting placement between bulk film 89 electrocatalysts and the natural photosynthetic complexes, and raises interesting questions about the mechanism of their catalytic action and productive ways of optimizing it. Since a reduction in nanoparticle size directly affects the reaction energetics, it may be concluded that the effect derives from intrinsic electronic properties of the particles: a change in the band gap and the appearance of strong electric fields at the nanoparticle surface. It may be possible to investigate the electronic energy level changes by means of optical and tunneling spectroscopy, while the presence of surface fields can be studied by nonlinear spectroscopic techniques. In addition, it can be beneficial to take advantage of the ability of aggregation sources to regulate the nanoparticle composition and structure. For example, inspired by the photosynthetic OEC complex, in future work, it will be interesting to produce mixed Mn/Ca nanoparticle catalysts with different stoichiometries. Furthermore, it has been demonstrated that by varying the source nucleation and growth parameters it is possible to alter the morphology of the generated nanoparticles (Johnson et al. 2015; Zhao et al. 2016; Krishnan et al. 2017) which offers an additional path to tuning the functionality of deposited nanocatalyst films. 90 4 A new technique to decorate SWCNTs with metal nanoclusters 4.1 Introduction Single wall carbon nanotubes (SWCNTs) are well known for their intrinsic geometry. They are strong and stiff, like graphene, owing to the SP 2 carbon bonds that form strong covalent bonds and honeycomb lattice structure. They possess exceptional thermal and electrical conductivity and hold strong chemical stability (Sahu et al. 2018; Hersam 2008; Novoselov et al. 2004; Wallace 1947; Cao et al. 2005). As it is shown in Figure 4.1, a SWCNT can be visualized as a seamless cylinder of a rolled-up graphene sheet along one of its 2D lattice vectors (known as chiral vector C) and described as C=na1+ma2 where a1, and a2 are basis vectors of the graphene lattice and the pair of integers (n,m) are called chiral index or just chirality. While the chirality determines the structure of carbon nanotubes (denoted as metallic or semiconducting types), the diameter of carbon nanotubes (denoted as a single wall or multi-wall carbon nanotube) affects their bandgap energies (Sahu et al. 2018; Hirsch 2002; White and Mintmire 2005). Single wall carbon nanotubes have numerous applications that have been reviewed comprehensively in the literatures, including but not limited to quantum wires (Baughman et al. 2002;Schü nemann et al. 2011) super conduction transitions (Rao et al. 1997), 91 microelectronics (Ionescu and Riel 2011), electron field emission ( Riggs et al. 2000), energy storage (Dai et al. 2012), biosensors ( Köhler et al. 2008), photo- acoustic imaging (Balasubramanian et al. 2012; Luke et al. 2012), composite materials (Gojny et al. 2004), and super capacitors ( Raicopol et al. 2013). And recently, they have been used in pharmaceutical applications as well (He et al. 2013; Allaedini 2016). The transport measurement of an individual SWCNT has been a subject of fundamental research interest for a long time even prior to its synthesis. After the discovery of CNTs in 1991 by Iijima (Iijima 1991), there have been reported tons of transport measurements. All measurements have shown that CNTs, as the most powerful 1D probes, have exceptional electronic properties (Tans et al. 1997; Bockrath et al. 1997; Tans et al. 1998; Martel 1998). These properties arise from its diameter (single wall or multi-wall carbon nanotubes), helicity in the arrangement of the carbon atoms in hexagonal arrays on their surface honeycomb lattice (Allaedini 2016; Rao et al. 1997) and defect on the Figure 4.1: Visualization of a single wall carbon nanotube (Left) It is shown 2D graphene sheet with the lattice vector a 1, a 2 and the chiral vector C = n a 1+m a 2. The pair of integers (n,m) called as chiral index or chirality. (Right) It is shown a graphene sheet rolled up along one of its 2D lattice vectors and make a seamless cylinder called carbon nanotube. (Hirsch 2002) 92 structure of carbon nanotubes. However, selective growths of a specified size, chirality, and placement are still the subject of ongoing researches. One of the important questions is whether the structure and properties of SWCNTs are affected by external doping (coating) and other means (Voggu et al. 2008). This question arises a great deal of interest to integrate the metal nanoclusters with SWCNTs to hybridize both material properties as a quasi-one-dimensional self-assembled metallic or superconducting nanowires by coating the sidewalls of nanotubes with size selected metal nanoclusters (Zhang et al. 2000; Bezryadin et al. 2000; Georgakilas et al. 2007; Maiti et al. 1989, Allain et al. 2012; Nguyen and Zhao 2014). Through this integration, the electronic communication happens between metal nanoclusters and SWCNTs, owing to charge transfer, affecting the electronic structure of SWCNTs and resulting the great potential applications in various technologies such as nanoelectronics devices, sensors, catalysis, energy storage, and biotechnologies (Bezryadin et al. 2000; Nguyen and Zhao 2014; Serp and Castillejos 2010; Barberio et al. 2015; Bardotti et al. 2014; Subrahmanyam et al. 2010). However, the essential fact to develop future technologies is to find reliable and reproducible synthetic techniques. To date, this integration has been achieved through two main methods, chemical reaction methods, and physical methods. In chemical reaction method (also known as the wet chemistry), numerous techniques have been developed to increase the adhesion efficiency of metal nanoclusters on SWCNTs, via functionalizing the nanotubes categorized as A) defect-group functionalization, B) covalent sidewall functionalization, C) noncovalent exohedral functionalization with surfactants, D) noncovalent exohedral 93 functionalization with polymers, and E) endohedral functionalization with, for example, C60 (see Figure 4.2) (Hirsch 2002; Georgakilas 2007; Wilgoose et al. 2006; Planeix et al. 1994). Making defect on the surface of SWCNTs by adding the terminal group or functionalizing the surface has been known a desirable strategy for catalysis application (Mudimela et al. 2014) in order to increase the adhesion of metal nanoclusters to the surface of SWCNTs. However, regardless of simplicity and efficacy of chemical reaction methods, they get involved in multistage chemical reactions that allow the impurities (or contamination) incorporated into either carbon nanotubes or nanoclusters; therefore, it would be challenging to control the purity and composition of affixed nanoclusters to carbon nanotubes. Moreover, making defect on the sidewall of SWCNTs might strongly affect the density of states, and consequently sacrifice the unique property of a naked, defect-free surface of a SWCNT specifically in the electronic device applications, where a defect-free, (noncovalent functionalization) suspended single-walled carbon nanotube is important for probing the intrinsic electrical properties of nanotubes (Cao et al. 2005). In the last decade, physical methods are used to decorate carbon nanotubes with nanoclusters by means of forming electrostatic force directed assembly atoms deposited on nanotubes via physical vaporization (Chen and Lu 2006), and direct thermal or electron beam evaporation (Scarselli et al. 2012). The weak bond between metal atoms and carbon atoms happen through fast diffusivity resulting coalescence of adatoms to form continuous or discontinuous coating layers on the side wall of carbon nanotubes (Zhang and Dai 2000). Moreover, it has been shown that the sputtered atoms produced by rf magnetron sputtering 94 source (Penza et al. 2009; Yang et al. 2012), by dc magnetron sputtering and high power impulse magnetron sputtering (HiPIMS) can coagulate into particles on the sidewall of bundled SWCNTs or MWCNTs due to low resistivity and strong diffusivity ( Mudimela et al. 2014; Wang et al. 2019; Muratore et al. 2013; Hussaina et al. 2017; Fedotov et al. 2013; Alexeeva and Fateev 2016; Yoshii et al. 2016). Lately, it has been presented a couple of gas phase techniques to decorate the MWCNTs with nanoclusters. Their results were compared with the previous gas phase techniques and reported the different coating morphology for the same materials if they are deposited atomically on nanotubes (Barberio et al. 2015; Yang et al. 2012). Figure 4.2: Functionalization possibilities for SWNTs A) defect‐group functionalization, B) covalent sidewall functionalization, C) noncovalent exohedral functionalization with surfactants, D) noncovalent exohedral functionalization with polymers, and E) endohedral functionalization with, for example, C 60. For methods B–E, the tubes are drawn in idealized fashion, but defects are found in real situations (Hirsch 2002). 95 Moreover, it appears to control the size and purity (ligand-free) of nanoclusters are extremely necessary to explore the unique properties of their integration with naked (without functionalization) carbon nanotubes. However, it has remained experimentally uncharted because of the absence of an appropriate technique. In this work, we approached this requirement by using a cluster source technique known as a dc magnetron sputtering gas aggregation source equipped with a quadrupole mass filter. This technique (explained in detail in chapter 2) allows flexible tuning of relevant parameters and a wide choice of working materials. The process was designed for producing robust fluxes of metal nanocluster ions in the vacuum (ligand and surfactant free), filtering them by size and soft-landing them on the sidewall of SWCNTs. Therefore, it is possible to investigate the hybrid morphology and interaction of size-selected metal nanoclusters (here is Aluminum) on a suspended naked SWCNT, without any further need for post-treatment on either nanoclusters or side wall of carbon nanotubes. 4.2 Experiment section SWCNT samples preparation A thin film of semiconducting SWCNTs, was obtained by the National Research Council of Canada. The SWCNTs used in this thin film were fabricated by a standard electric arc discharged reactor. 1.2 mg of the film was added to 10 mL of DriSolv Toluene Anhydrous and bath sonicated for 30 minutes to make 0.12 mg/ml homogeneous suspension of SWCNTs solution. AFM imaging was performed from drop-casted carbon 96 nanotube solution on Si/SiO2 substrate to do a preliminary evaluation of the de-bundled carbon nanotubes, the degree of their cleanliness and their surface coverage in order to tune the amount of drop-casting solution on TEM grids (See Figure 4.3). Then, 10μL pipette was used to drop-cast just one drop of the solution on TEM grids, and spin coated by spin coater (VTC-100, MTI Corporation) for 10 sec in 50 rpm, and 60 sec in100 rpm (2 times) on holey SiN TEM grids, and 10sec in 50rpm, and 60 sec in 300 rpm (one time) on holey carbon TEM grids. The speed of spin coater was tuned to dry the surface of TEM grid as well as to align the direction of SWCNTs on TEM grids. Moreover, the spin coater was operated in lower speed for SiN TEM grid to keep the TEM grid membrane intact. Figure 4.6 shows the first TEM imaging of the drop-casted SWCNTs solution on a holey carbon TEM grid. The result proves the existence of a carbonaceous coating around the SWCNTs attributed to the polymeric surfactant used to unbundle the SWCNTs in the solution of SWCNTs in Toluene. Therefore, a homemade heater setup made of OFHC copper with a maximum capacity of loading three heater elements in parallel (shown in Figure 4.4) was designed and built to remove the carbonaceous layer through bakeout the SWCNTs samples in the deposition chamber. After spin coating, the TEM grids were loaded on a homemade TEM grid holder and mounted on the heater setup. Then they all are assembled on an axial micrometer and loaded into the deposition chamber. An autotune temperature controller (Omega-CN 9000A Series) and two thermocouples (ARi-KMTXL- 040U-10) were arranged to control the temperature at 400C. In practice, just two heater elements (Part No. E3A50-E12H, 3- inch-long, 120 Volts max, 300 W max) were loaded. A variac variable transmitter was 97 applied to supply ac current with a lower voltage than regular outlet voltage (~ 40V) to the controller to operate in the safety zone; therefore, each heater element used ~ 80Watt power. The thermocouples were connected in two different spots of the heater to monitor the uniformity of the temperature endways the heater. The SWCNTs were baked out at 400C for 10 hrs under a high vacuum with base pressure 4E-9 Torr. The bakeout process was performed a day before running the deposition experiment to keep SWCNTs samples as clean as possible in the deposition chamber. Then, TEM imaging was planned exactly the day after the deposition was carried out to decrease the risk of post contamination of SWCNTs or Aluminum nanoclusters. Figure 4.3: AFM images of drop casted single wall carbon nanotube solution on Si/SiO2 substrate. 98 Production and deposition of Aluminum nanoclusters The deposition was carried out using the same experimental setup shown in Figure 2.1 and described in detail in chapter 2. Therefore, in the present case, just the operating parameters are discussed. Here, a 2-inch diameter target with 99.999% Aluminum (ACI alloy) was applied. The magnetron block was equipped with the manufacturer’s “magnet Figure 4.4: Heater setup a) Top view of heater setup; 1) Test substrate 2) Heater holder 3) TEM grid holders 4) Thermocouples 5) Collimator shield b) Front view of heater setup without collimator shield 1 2 3 4 5 a b 99 type A” whose benefit lies in producing almost exclusively ionized particles (see chapter 2 for more detail). Argon and helium gas (99.999% purity) were introduced into the source region behind the magnetron head, and the magnetron itself was mounted on a linear translator. The interplay between the source parameters has been discussed in detail by us previously (see chapter2) (Khojasteh and Kresin 2016, 2017). The flow rate of Ar and He was 200 sccm and 250 sccm respectively: the distance between the magnetron and the beam exit aperture was L~8.5cm, and the discharge power was ~27Watt. After condensation and nucleation of the sputtered atoms and consequently nanoparticles growth, the resulting directed beam that already left the source passes through the quadrupole mass filter. Ultimately, the size-selected nanoparticle ions find themselves in the main deposition chamber (with the base pressure of ~3 e-9 Torr before deposition). In deposition chamber, after collimation by a 1 cm × 1 cm mask, the beam lands on room-temperature on the already bakeout SWCNTs samples. The incident cluster flux is measured by a picoamperemeter, and the samples are imaged by TEM (JEOL 2100F) after transfer to air. The conducting substrates are connected to ground potential, and the impinging nanoparticle ions become neutralized. Being significantly below the atomic binding energy, this is generally regarded as the “soft-landing” regime (discussed in chapter 2 in detail) which avoids significant fragmentation and deformation of nanoparticles (Popok et al. 2011). The described setup allows for the deposition of contamination-free Aluminum nanoclusters (Baker et al. 1997) with size control which is difficult to achieve by solution-processed metal nanoparticle synthesis. 100 TEM characterization Transmission electron microscope (TEM) is a powerful tool for material science. It has the same basic principles as the light microscope but using electrons instead of light, and thus replacing optical lenses with electromagnetic ones. In fact, TEM uses high energy electrons (up to 300KV accelerating voltage) accelerated to near speed of light. The electron beam behaves like a waveform with wavelength about a million times shorter than light waves; therefore, the optimal resolution for TEM images is in many orders of magnitude better than that from a light microscope. Thus, TEM can reveal the finest details of internal structure in some cases as small as individual atoms. In fact, TEM uses transmitted electrons projected as a two-dimensional image, that gets magnified by further electron optics to produce an image. A high energy beam of electrons is shone through a very thin sample, and the interactions between the electrons and atoms can be used to observe features such as the crystal structure and, dislocations and grain boundaries. Chemical analysis can also be performed by TEM. High resolution can be used to analyze the quality, shape, size, and density of quantum wells, wires, and dots. The principle setup is shown in Figure 4.5a. Imaging is performed by focusing a beam of electrons from the electron gun into a small, thin, coherent beam via using the condenser lens. This beam is restricted by the condenser aperture, which excludes high angle electrons. The beam then strikes the specimen and parts of it are transmitted depending upon the thickness and electron transparency of the specimen. This transmitted 101 portion is focused by the objective lens into an image on a phosphor screen or charge coupled device (CCD) camera. Optional objective apertures can be used to enhance the contrast by blocking out high-angle diffracted electrons. The image then passed down the column through the intermediate and projector lenses, is enlarged all the way. The image strikes the phosphor screen and light is generated, allowing the user to see the image. The darker areas of the image represent those areas of the sample that fewer electrons are transmitted through while the lighter areas of the image represent those areas of the sample that more electrons were transmitted through. Figure 4.5(b) shows a simple sketch of the path of a beam of electrons in a TEM from just above the specimen and down the column to the phosphor screen. As the electrons pass through the sample, they are scattered by the electrostatic potential setup by the constituent elements in the specimen. After passing through the specimen, they pass through the electromagnetic objective lens which focuses all the electrons scattered from one point of the specimen into one point in the image plane. Also, shown in Figure 4.5(b) is a dotted line where the electrons scattered in the same direction by the sample are collected into a single point. This is the back focal plane of the objective lens and is where the diffraction pattern is formed (Williams and Carter 2016; Hirsch 1965). 4.2.3.1 TEM results In this work, the TEM imaging was performed using JEOL 2100F available in Core Center of Excellence in Nano Imaging (CNI) at USC. 102 The first TEM imaging was performed after drop casting and spin coating of SWCNT solution on a holy carbon TEM grid. Figure 4.6 shows the existence of a carbonaceous coating around an individual SWCNT resulting polymeric surfactant applied to unbundle the SWCNTs in toluene solution. However, the carbonaceous coating is required to be removed by bakeout the samples. The TEM results of the SWCNT samples after bakeout process in the deposition chamber and under high vacuum are depicted in Figure 4.7. It verifies that the carbon coating around CNTs was removed through the bakeout process and SWCNTs are entirely clean. Figure 4.7 (a, b) shows an individually suspended naked SWCNT, and Figure 4.7(c) shows a bundle of two suspended naked SWCNTs. The diameter of a SWCNT after the bakeout process was estimated between 1.3-1.7 nm by Image J software. Figure 4.8 demonstrates the deposited Aluminum nanoclusters on SWCNTs. Figure 4.8 (a) shows the chain of the Aluminum nanoclusters on a SWCNT (shown by a red circle) and Figure 4.8 (b) shows a sample of deposited Aluminum nanocluster on an individually suspended SWCNT (shown by the red arrow). It also shows a naked suspended SWCNT that proves that after bakeout the samples in the deposition chamber and performing nanocluster deposition, SWCNTs are still clean. (shown by a green arrow). 103 b a Figure 4.5: Transmission electron microscope (a) General layout of a TEM describing the path of electron beam in a TEM. (b) A ray diagram of the diffraction mechanism in TEM; An electron beam passes through a thin section part of a material and electrons are scattered. A sophisticated system of electromagnetic lenses focuses the scattered electron into an image or a diffraction pattern or a nanoanalytical spectrum depending on the mode of operation (Hirsch 1965; Williams 2016). 104 Figure 4.7: TEM images of individual suspended SWCNT samples on holey SiN TEM grids after bakeout the samples at 400C and for10 hrs. SWCNTs samples on SiN TEM grids were baked out at 400Cand for 10 hrs in deposition chamber with the base pressure ~ E-9 Torr. The TEM images show that the carbonaceous coating around SWCNTs (shown in Figure 4.6) was totally removed and what is observed is just suspended clean SWCNTs. a) and b) are TEM images of the individually suspended SWCNTs and c) is a TEM image of a twin suspended SWCNT, and there is no trace of any carbon material around them. Figure 4.6: TEM image of suspended SWCNT on a holey carbon TEM grid before bakeout the sample. The TEM image shows the SWCNTs are coated by a carbon layer. When the beam hits them, the layer of carbon grows (shown by a red arrow) which is related to the polymeric surfactant used to unbundle the SWCNTs in the SWCNTs-Toluene solution. 105 4.3 Discussion Preparing the TEM samples of individually suspended SWCNTs is not a trivial task. Indeed, by itself, finding a right recipe to make a homogeneous suspension of debundeled SWCNTs solution is really challenging. In my case, the debundeling process of SWCNTs in the SWCNTs-Toluene solution via using the polymeric surfactant adds the carbonaceous coating layer around the debundled SWCNTs. Therefore, it is urged to perform an engineering design to bake out them under high vacuum chamber (here at base pressure e-9 Torr) in order to have appropriate sample to explore the direct interaction of Figure 4.8: TEM images of deposited Aluminum nanoclusters on SWCNTs on a holey SiN TEM grid. a) TEM image shows deposited Aluminum nanoclusters on SWCNTs on a holey SiN TEM grid; oval red circle shows the chain of Aluminum nanoclusters aligned on the side walls of SWCNTs. b) TEM image shows the deposited Aluminum nanocluster on the suspended SWCNTs (shown by a red arrow). It also shows a naked suspended SWCNT that proves that after bakingout the samples in the deposition chamber and nanocluster deposition, SWCNTs are still clean. (shown by a green arrow). 106 nanoclusters on a naked SWCNT. In fact, if the CNTs have carbon overcoating on their sidewalls, this could affect the outcome of metal cluster deposition (Zhang et al. 2000). Moreover, the study of size-selected nanoclusters by tracing the evolution of a nanoscale metal particle properties atom by atom, tune and identify the optimal system with maximum precision needs a complicated state of the art technique to be able to manipulate the size and composition of the constituent building blocks. (de Heer et al. 1993; Khanna and Jena 1995). Additionally, hybridizing of size-selected metal nanoclusters with SWCNT to elucidate their integrated electronic properties as a quasi-one-dimensional super current nonohmic wire, requires the special fabrication arrangements. Experimentally so far most of the integration of metal-carbon nanotubes has been conducted by wet chemistry; however, owing to the restrictions in this method such as undergoing multiple reactions and its consequences, it seems this method is unfeasible to decorate a naked (functionalization free) suspended SWCNT with intact (contamination and ligand-free) metal nanoclusters (Georgakilas et al. 2007). In the gas phase methods, previous studies have been performed to form metal wires through coating the suspended SWCNTs with metals or metal alloys (Zhang et al. 2000; Bezryadin 2000). However, in all those trials a full coverage of thin layers of atomic- scale material builds up on the carbon nanotubes which cannot be controlled in terms of the amount of coverage and the size of the nanoclusters formed after the deposition of adatoms. 107 Recently, a couple of trials of the deposition of small metal nanoclusters on multiwall carbon nanotubes have been reported (Barberio et al. 2015; Bardotti et al. 2014). In those cases, the nucleation and growth of deposited nanoclusters on the sidewall of carbon nanotubes follow a different path of the nucleation and growth of the nanoclusters made of sputtered adatoms. For example, Barberio et al. 2015 applied laser ablation in solution to deposit different metal nanoclusters on a network of entangled tube bundles of MWCNTs buckypaper, and they found different coating behavior in comparison with atomic deposition, specifically for aluminum nanoclusters. Moreover, Bardotti et al. 2014 showed the soft-landed deposition of bimetallic clusters (FePt) on suspended multiwall carbon nanotube by laser vaporization method and observed that deposited clusters form islands on the sidewalls of MWCNTs attributed into their diffusion on the CNT surface. However, the same cluster materials deposited on an amorphous carbon layer substrate remain isolated and randomly distributed on the surface. Consequently, they all emphasize on different morphological interactions between the directly deposited clusters on the side wall of carbon nanotubes in comparison with either depositing the same cluster materials on the other substrate or the same cluster materials that are formed after aggregation and coalescence of deposited adatoms on the sidewall of carbon nanotubes Moreover, to the best of my knowledge there is no report of the direct deposition of metal clusters on individually suspended naked SWCNTs in the literature; therefore, the novelty of our work is highlighted by depositing the ligand-free Aluminum 108 nanoclusters on the naked ( free of any covalent functionalizing group) suspended SWCNT via applying a productive pathway to build structures based on size-optimized nanoclusters with a dc magnetron sputtering gas aggregation source , through filtering the nanoclusters by size and soft landing them on SWCNTs as well as tuning the amount of nanocluster coverage on the sidewall of SWCNTs. Therefore, this technique enables us to investigate systematically the morphology and interaction of deposited size selected soft-landed metal nanoclusters on the sidewall of SWCNTs from sub-monolayer to full coating, which is not possible with other techniques. Indeed, I believe that this is a great step to evaluate the surface science interaction of integrating individually suspended naked SWCNTs with the size selected metal nanoclusters before proceeding to make a device of their hybridization and conducting I- V measurement to explore the charge transfer between nanoclusters and carbon nanotubes. 4.4 Conclusion Contamination-free size selected Aluminum nanoclusters ranging from 3–7 nm in diameter were successfully deposited on suspended naked SWCNT by a dc magnetron sputtering gas aggregation source. The TEM images show the deposited nanocluster on SWCNT sidewalls (shown in Figure 4.8) proving the capability of this technique to have versatile controls on the deposition of already made metal nanoclusters which is not easily achievable with other chemical or physical methods. 109 Moreover, keeping the physical and chemical properties of both SWCNTs and nanoclusters pristine during their integration and generation nanoclusters via a cluster beam deposition as the productive new technique, with the capability to produce structures based on size-optimized nanoclusters, opens the new avenue to utilize SWCNTs as supercurrent carriers. Indeed, through this technique, one is able to couple the sidewall of SWCNT to a chain of deposited identical nanoclusters with quantized shell structure which are connected by a tunneling barrier via means of the proximity effect (Heersche et al. 2007; Feigel’man et al. 2008), and finally perform transport measurements to detect conductivity changes which are affected in the nanotube channel. 110 5 References Alexeeva O K and Fateeva V N (2016) Application of the magnetron sputtering for nanostructured electrocatalysts synthesis. Int. J. Hydrogen Energy 41, 3373-3386 Allaedini G, Tasirin1 S M, Aminayi P, Yaakob Z, Meor Talib M Z (2016) Carbon nanotubes via different catalysts and the important factors that affect their production: A review on catalyst preferences, Int J Nano Dimens. 7, 186-200 Allain Ad, Han Z, and Bouchiat V (2012) Electrical control of the superconducting to insulating transition in graphene–metal hybrids. Nat. Mater. 11, 590-594 Artero V, Chavarot-Kerlidou M, Fontecave M (2011) Splitting water with cobalt. Angew. Chem. Int. Ed. 50 7238-7266 Ayesh A I, Ahmed H A, Awwad F, Abu-Eishah S I, Mahmood S T (2013) Mechanisms of Ti nanocluster formation by inert gas condensation. J. Mater. Res. 28,2622-2628 Ayesh A I, Lassesson A, Brown S A, Dunbar A D F, Kaufmann M, Partridge J G, Reichel R, Lith J V (2007) Experimental and simulational study of the operation for a high transmission mass filter. Rev. Sci. Instrum. 78 053906 Ayesh AI, Qamhieh N, Ghamlouche H, Thaker S, El-Shaer M (2010) Fabrication of size- selected Pd nanoclusters using a magnetron plasma sputtering source. J. Appl. Phys. 107, 034317 Balasubramanian K, Kurkina T, Ahmad A, Burghard M and Kern K (2012) Tuning the functional interface of carbon nanotubes by electrochemistry: Toward nanoscale chemical sensors and biosensors. J. Mater. Res. 27, 391-402. Baker S H, Thornton S C, Keen A M, Preston T I, Norris C, Edmonds K W, Binns C (1997) The construction of a gas aggregation source for the preparation of mass-selected ultra-small metal particles. Rev. Sci. Instrum. 68 1853-1857 Barber J (2017) A mechanism for water splitting and oxygen production in photosynthesis. Nat. Plants 3 17041 111 Barberio M, Stranges F and Xu F (2015) Coating geometry of Ag, Ti, Co, Ni, and Al nanoparticles on carbon nanotubes, Appl. Surf. Sci 334, 174-179 Bardotti L, Tournus F, Delagrange R, Benoit J-M, Pierre-Louis O and Dupuis V (2014) Behavior of size-selected iron–platinum clusters soft landed on carbon nanotubes. Appl. Surf. Sci. 301, 564-567 Barron H, Opletal G, Tilley R and Barnarda A S (2017) Predicting the role of seed morphology in the evolution of anisotropic nanocatalysts. Nanoscale, 9, 1502–1510 Baughman R H, Zakhidov A A, and de Heer W A (2002) Carbon nanotubes: The route toward applications. Science 297, 787-792 Bezryadin A, Lau C N and Tinkham M (2000) Quantum suppression of superconductivity in the ultrathin nanowire. Nature 404, 971-974 Binns C (2001) Nanoclusters deposited on surfaces. Surf. Sci. Rep. 44,1-49 Binnig G, Quate C F, and Gerber Ch (1986) Atomic force microscope. Phys. Rev. Lett. 56, 930 Bockrath M, Cobden D H, McEuen P L, Chopra N G, Zettl A, Thess A and Smalley R E (1997) Single-Electron Transport in Ropes of Carbon Nanotubes. Science 275, 1922-1925 Bohm C and Perrin J (1992) Retarding-field analyzer for measurements of ion energy distributions and secondary electron emission coefficients in low-pressure radio frequency discharges. Rev. Sci. Instrum. 64, 31- 44 Bray KR, Jiao CQ, DeCerbo JN (2014) Influence of carrier gas on the nucleation and growth of Nb nanoclusters formed through plasma gas condensation. J. Vac. Sci. Technol., B 32, 031805 Cao J, Wang Q and Dai H (2005) Electron transport in very clean, as grown suspended carbon nanotubes. Nat. Mater. 4, 754-749 Cattolica RJ, Gallagher RJ, Anderson JB, Talbot L (1979) Aerodynamic separation of gases by velocity slip in free jet expansions. AIAA J. Air Transp. 17, 344-355 Chen J and Lu G (2006) Controlled decoration of carbon nanotubes with nanoparticles. Nanotechnology 17, 2891–2894 112 Chen G, Zhou W, Guan D, Sunarso J, Zhu Y, Hu X, Zhang W, Shao Z (2017) Two orders of magnitude enhancement in oxygen evolution reactivity on amorphous Ba0.5Sr0.5Co0.8Fe0.2O3-σ nanofilms with tunable oxidation. Sci. Adv. 3 e1603206 Dai L, Chang D W, Baek J B and Lu W (2012) Carbon nanomaterials for advanced energy conversion and storage. Small, 8, 1130-1166. Das SC, Majumdar A, Shripathi T, Hippler R (2009) Development of metal nanocluster ion source based on DC magnetron plasma sputtering at room temperature. Rev. Sci. Instrum. 80, 095103 Dawson P H (1976) Quadrupole Mass Spectrometry and its Applications. Elsevier, Amsterdam De Heer W A (1993) The physics of simple metal clusters: Experimental aspects and simple Models. Rev. Mod. Phys. 65, 611 Di Castro V, Polzonetti G 1989 XPS study of MnO oxidation. J. Electron. Spectrosc. Relat. Phenom. 48 117-123 Dutka MV, Turkin AA, Vainchtein DI, De Hosson JThM (2015) On the formation of copper nanoparticles in nanocluster aggregation source. J. Vac. Sci. Technol., A 33, 031509 El Koura Z, Cazzanelli M, Bazzanella N, Patel N, Fernandes R, Arnaoutakis G E, Gakamsky A Dick A, Quaranta A, Miotello A (2016) synthesis and characterization of Cu and N codoped RF-sputtered TiO2 films: Photoluminescence dynamics of charge carriers relevant for water splitting. J. Phys. Chem. C 120 12042-12050 Fan R, Mao J, Yin Z, Jie J, Dong W, Fang L, Zheng F, Shen M (2017) Efficient and stable silicon photocathodes coated with vertically standing nano-MOS2 films for solar hydrogen production. ACS Appl. Mater. Interfaces 9 6123-6129 Fedotov A A, Grigoriev S A, Lyutikova E K, Millet P and Fateev V N (2013) Characterization of carbon-supported platinum nanoparticles synthesized using magnetron sputtering for application in PEM electrochemical systems, Int. J. Hydrogen Energy 38, 426 – 430 113 Feigel’man M V, Skvortsov M A and Tikhonov K S (2008) Proximity-induced superconductivity in graphene. JETP Lett. 88, 747–751 Feng Y, Lin S, Huang S, Shrestha S, Conibeer G (2015) Can Tauc plot extrapolation be used for direct-band-gap semiconductor nanocrystals? J. Appl. Phys. 117, 125701 Ferreira K N, Iverson T M, Maghlaoui K, Barber J, Iwata S 2004 Architecture of the photosynthetic oxygen-evolving center. Science 303 1831-1838 Finningan R E (1994) Quadrupole mass spectrometers. Anal. Chem. 66, 969–975 Fischer A, Kruk R, Hahn H (2015) A versatile apparatus for the fine-tuned synthesis of cluster-based materials. Rev. Sci. Instrum. 86, 023304 Geneva M, Peter T, Bornholdt S, Kersten H, Strunskus T, Zaporojtchenko V, Faupel F, Hippler R (2012) Mass spectrometric investigations of nano-size cluster ions produced by high-pressure magnetron sputtering. Contrib. Plasma. Phys. 52, 881- 889 Geneva M, Pipa A V, Smirnov B M, Kashtanov P V, Hippler R (2013) Velocity distribution of mass selected nano-size clusters. Plasma Sources Sci. Technol. 22, 045011 Georgakilas V, Gournis D, Tzitzios V, Pasquato L, Guldi D M and Prato M (2007) Decorating carbon nanotubes with metal or semiconductor nanoparticles. J. Mater.Chem. 17, 2679–2694 Gojny F, Wichmann M, Köpke U, Fiedler B, and Schulte K (2004) Carbon nanotube- reinforced epoxy-composites: Enhanced stiffness and fracture toughness at low nanotube content. Compos. Sci. Technol. 64, 2363-2371. Gorlin Y, Jaramillo T F (2010) A bifunctional nonprecious metal catalyst for oxygen reduction and water oxidation. J. Am. Chem. Soc. 132 13612-13614 Gracia-Pinilla M, Vidaurri G S, Pérez-Tijerina E (2010) Deposition of size-selected Cu nanoparticles by inert gas condensation. Nanoscale Res. Lett. 5,180–188 Grammatikopoulos P, Steinhauer S, Vernieres J, Singh, V, Sowwan M (2016) Nanoparticle design by gas-phase synthesis. Adv. Phys. X 1 81-100 Greeley J, Rossmeisl J, Hellman A, Norskov J K (2007) Theoretical trends in particle size effects for the oxygen reduction reaction. Z. Phys. Chem. 221 1209-1220 114 Guan J, Zhang J, Yu T, Xue G, Yu X, Tang Z, Wei Y, Yang J, Li Z, Zou Z (2012) Interfacial modification of photoelectrode in ZnO-based dye-sensitized solar cells and its efficiency improvement mechanism. RSC Adv. 2, 7708-7713 Guo C X, Chen S, Lu X (2014) Ethylenediamine-mediated synthesis of Mn3O4 nano- octahedrons and their performance as electrocatalysts for the oxygen evolution reaction. Nanoscale 6 10896-10901 Haberland H (1994) Experimental methods. In: Haberland H (ed) Clusters of atoms and molecules: Theory, experiment, and clusters of atoms. Springer, Berlin, pp 207-232 Haberland H, Karrais M, Mall M, Thurner Y (1992) Thin films from energetic cluster impact: A feasibility study. J. Vac. Sci. Technol., A 10, 3266-3271 Haberland H, Mall M, Moseler M, Qiang Y, Reiners T, Thurner Y (1994) Filling of micron- sized contact holes with copper by energetic cluster impact. J. Vac. Sci. Technol., A 12, 2925-2930 Haberland H, Inspov Z, Moseler M (1995) Molecular-dynamics simulation of thin- film growth by energetic cluster impact. Phys. Rev. B: Condens. Matter Mater. Phys. 51, 11061-11067 Halder A, Curtiss L A, Fortunelli A, Vajda S (2018) Perspective: Size-selected clusters for catalysis and electrochemistry. J. Chem. Phys. 148, 110901-110915 Harriman A, Pickering I J, Thomas J M, Christensen P A (1988) Metal oxides as heterogeneous catalysts for oxygen evolution under photo chemical conditions. J. Chem. Soc., Faraday Trans. 1 84 2795-2806 Haugstad G (2012) Atomic Force Microscopy: Understanding Basic Modes and Advanced Applications. John Wiley & Sons, USA Haynes W (ed) (2016) CRC handbook of chemistry and physics, 97th edn. CRC Press, Boca Raton He H, Pham-Huy L A, Dramou P, Xiao D, Zuo P, and PhamHuy C (2013) Carbon nanotubes: Applications in pharmacy and medicine. BioMed. Res. Int. 4, 145-156 Heersche H B, Jarillo-Herrero P, Oostinga J B, Vandersypen L and Morpurgo A F (2007) Bipolar supercurrent in graphene. Nature 446, 56–59 115 Hecht H G (1976) The interpretation of diffuse reflectance spectra. J. Res. Nat. Bur. Stand. Sec. A: Phys. Ch. 80A 567-583 Heiz U, Landman U (eds) (2007) Nanocatalysis. Berlin, Springer Hersam M C (2008) Progress towards monodisperse single-walled carbon nanotubes. Nat. Nanotechnol. 3, 387-394 Hihara T, Sumiyama K (1998) Formation and size control of a Ni cluster by plasma gas condensation. J. Appl. Phys. 84, 5270-5276 Hirsch A (2002) Functionalization of Single-Walled Carbon Nanotubes. Angew. Chem. Int. Ed. 41, 1853-1859 Hirsch P (1965) Electron microscopy of thin crystals, London, Butterworths. Hoffmann E, Stroobant V (2007) Mass Spectrometry: Principles and Applications, 3rd ed., John Wiely & Sons, England Hussaina S, Eriksona H, Kongia N, Merisalu M, Ritslaid P, Sammelselga V and Tammeveski K (2017) Heat-treatment effects on the ORR activity of Pt nanoparticles deposited on multi-walled carbon nanotubes using magnetron sputtering technique. Int. J. Hydrogen Energy 42, 5958-5970 Huttel Y (ed) (2017) Gas-phase synthesis of nanoparticles. Wiley-VCH, Weinheim Hutzler NR, Lu H-I, Doyle JM (2012) The buffer gas beam: An intense, cold, and slow source for atoms and molecules. Chem. Rev. 112, 4803−4827 Ibach H, Lüth H (2009) Solid-State Physics. Berlin, Springer, Sec. 12.6 Iijima S (1991) Helical microtubules of graphitic carbon. Nature 354, 56-58 Ionescu A M and Riel H (2011) Tunnel field-effect transistors as energy-efficient electronic switches. Nature 479, 329337. Jin K, Chu A, Park J, Jeong D, Jerng S E, Sim U, Jeong H-Y, Lee CW, Park Y-S, Yang K D, Pradhan G K, Kim D, Sung N-E, Kim S H, Nam K T (2015) Partially oxidized sub-10 nm MnO nanocrystals with high activity for water oxidation catalysis. Sci. Rep. 5 10279 Jin K, Seo H, Hayashi T, Balamurugan M, Jeong D, Go Y K, Hong J S, Cho K H, Kakizaki H, Bonnet-Mercier N, Kim M G, Kim S H, Nakamura R, Nam K T (2017) 116 Mechanistic investigation of water oxidation catalyzed by uniform, assembled MnO nanoparticles. J. Am. Chem. Soc. 139 2277-2285. Johnson G E, Colby R, Laskin J (2015) Soft landing of bare nanoparticles with controlled size, composition, and morphology. Nanoscale 7 3491-3503 Johnston RL (2002) Atomic and molecular clusters. Taylor & Francis, London Kappes M, Leutwyler S (1988) Molecular beams of clusters. In: Scoles G (ed) Atomic and molecular beam methods, vol l. Oxford University Press, New York, pp 380-415 Kärkäs M D, Verho O, Johnston E V, Åkermark B (2014) Artificial photosynthesis: Molecular systems for catalytic water oxidation. Chem. Rev. 114 11863-12001 Kear G, Walsh F C (2005) The characteristics of a true Tafel slope. Corros. Mater. 30 S1- S4 Kim J H, Kaneko H, Minegishi T, Kubota J, Domen K, Lee J S (2016) Overall photoelectrochemical water splitting using tandem cell under simulated sunlight. ChemSusChem 9 61-66 Kittel C (2005) Introduction to Solid State Physics. Hoboken, Wiley Khan S, Yang K R, Ertem M Z, Batista V S, Brudvig G W (2015) Mechanism of manganese-catalyzed oxygen evolution from experimental and theoretical analyses of 18 O kinetic isotope effects. ACS Catal. 5 7104-7113 Khanna S and Jena P (1995) Atomic clusters: Building blocks for a class of solids, Phys. Rev. B: Condens. Matter Mater. Phys. 51, 13705 Khojasteh M, Kresin V V (2016) Formation of manganese nanoclusters in a sputtering/aggregation source and the roles of individual operating parameters. Proc. SPIE 10174,1017407 Khojasteh M, Kresin V V (2017) Influence of source parameters on the growth of metal nanoparticles by sputter-gas-aggregation. Appl. Nanosci. 7 875-883 Köhler A R, Som C, Helland A and Gottschalk F (2008) Studying the potential release of carbon nanotubes throughout the application life cycle. J. Cleaner Prod.16, 927-37 Kresin V Z, Lester Jr W A (1992) Quantum-mechanical model of heterogeneous catalysis. Chem. Phys. Lett. 197 1-6 117 Krishnan G, de Graaf S, ter Brink G H, Persson POÅ, Kooi B J, Palasantzas G (2017) Strategies to initiate and control the nucleation behavior of bimetallic nanoparticles. Nanoscale 9 8149-8156. Kusior A, Kollbek K, Kowalski K, Borysiewicz M, Wojcie T (2016) Sn and Cu oxide nanoparticles deposited on TiO2 nanoflower 3D substrates by inert gas condensation technique. Appl. Surf. Sci. 380, 193–202 Kuo C-H, Mosa I M, Thanneeru S, Sharma V, Zhang L, Biswas S, Aindow M, Alpay S P, Rusling J F, Suib S L, He J (2015) Facet-dependent catalytic activity of MnO electrocatalysts for oxygen reduction and oxygen evolution reactions. Chem. Commun. 51 5951-5954 Kwon G, Ferguson G A, Heard C J, Tyo E C, Yin C, DeBartolo J, Seifert S, Winans R E, Kropf A J, Greeley J, Johnston R L, Curtiss L A, Pellin M J, Vajda S (2013) Size- dependent subnanometer Pd cluster (Pd4, Pd6, and Pd17) water oxidation electrocatalysis. ACS Nano 7 5808-5817 Lang S M, Bernhardt T M, Kiawi D M, Bakker J M, Barnett N R, Landman U (2016) Cluster size and composition dependent water deprotonation by free manganese oxide clusters. Phys. Chem. Chem. Phys. 18 15727-15737. Li Q, Wang Z-L, Li G-R, Guo R, Ding L-X, Tong Y-X (2012) Design and synthesis of MnO2/Mn/MnO2 sandwich-structured nanotube arrays with a high supercapacitive performance for electrochemical energy storage. Nano Lett. 12 3803-3807 Li X, Hao X, Abudula A, Guan G (2016) Nanostructured catalysts for electrochemical water splitting: current state and prospects. J. Mater. Chem. A 4 11973-12000 Li Y, Frenkel A I (2017) Metal nanocatalysts. In: Iwasawa Y, Asakura K, Tada M (eds) XAFS Techniques for Catalysts, Nanomaterials, and Surfaces. Cham, Springer Chapter19, pp 273-298 Lin J, Zhang X, Zhou L, Li S, Qin G (2017) Pt-doped α-Fe2O3 photoanodes prepared by a magnetron sputtering method for photoelectrochemical water splitting. Mater. Res. Bull. 91 214-219 118 López R, Gómez R (2012) Band-gap energy estimation from diffuse reflectance measurements on sol-gel and commercial TiO2: A comparative study. J. Sol-Gel Sci. Technol. 61 1–7 Luke G P, Yeager D and Emelianov S Y (2012) Biomedical applications of photoacoustic imaging with exogenous contrast agents. Ann. Biomed. Eng. 40, 422-437 Luo Z, Woodward W H, Smith J C, Castleman Jr AW (2012) Growth kinetics of Al clusters in the gas phase produced by a magnetron-sputtering source. Int. J. Mass Spectrom. 309, 176-181 Madelung O, Rössler U, Schulz M (eds.) (2000) MnO: band structure, energy gap. In: Landolt-Börnstein - Group III Condensed Matter 41D (Non-Tetrahedrally Bonded Binary Compounds II). Springer, Berlin Maiti H S, Datta S and Basu R N (1989) High‐Tc superconductor coating on metal substrates by an electrophoretic technique. J. Am. Ceram. Soc.72 ,1733-35 Mantis Deposition Ltd. Application Note App-001. Accessed July(2017) www.mantisdeposition.com/fileadmin/user_upload/images/appnotes/app_001.pdf. Martel R, Schmidt T, Shea H R, Hertel T, and Avouris Ph (1998) Single- and multi-wall carbon nanotube field-effect transistors. Appl. Phys. Lett. 73, 2447-2449 Masudy-Panah S, Moakhar R S, Chua C S, Tan H R, Wong T I, Chi D, Dalapati G K (2016) Nanocrystal engineering of sputter-grown CuO photocathode for visible-light- driven electrochemical water splitting. ACS Appl. Mater. Interfaces 8 1206-1213 Meiwes-Broer K-H (ed.) (2000) Metal clusters at surfaces. Springer, Berlin McEvoy J P, Brudvig G W (2006) Water-splitting chemistry of photosystem II. Chem. Rev. 106 4455-4483 McInnes A, Plant S R, Ornelas I M, Palmer R E, Wijayantha K G U (2017) Enhanced photoelectrochemical water splitting using oxidized mass-selected Ti nanoclusters on metal oxide photoelectrodes. Sustainable Energy Fuels 1 336-344 Milani P, Iannotta S (1999) Cluster beam synthesis of nanostructured materials. Springer, Berlin 119 Miller DR (1988) Free jet sources. In: Scoles G (ed) Atomic and molecular beam methods, vol l. Oxford University Press, New York, pp 14-53 Morel R, Brenac A, Bayle-Guillemaud P (2003) Growth and properties of cobalt clusters made by sputtering gas-aggregation. Eur. Phys. J. D 24, 287-290 Moulder J F, Stickle W F, Sobol P E, Bomben K D (1992) Handbook of X-ray photoelectron spectroscopy: A reference book of standard spectra for identification and interpretation of XPS data. Physical Electronics Division, Perkin-Elmer Corporation Mudimela P R, Scardamaglia M, Gonzalez-Leon O, Reckinger N, Snyders R, Liobet E, Bittencourt C and Colomer J-F (2014) Gas sensing with gold- decorated vertically aligned carbon nanotubes. Beilstein J. Nanotechnol. 5, 910–918 Muratore C, Reed A N, Bultman J E, Ganguli S, and Voevodin A A (2013) Nanoparticle decoration of carbon nanotubes by sputtering. Carbon 57, 274-81 Najafpour M M, Ehrenberg T, Wiechen M, Kurz P (2010) Calcium manganese (III) oxides (CaMn2O4⋅x H2O) as biomimetic oxygen-evolving catalysts. Angew. Chem. Int. Ed. 49 2233-2237 Najafpour M M, Zarei Ghobadi M, Haghighi B, Tomo T, Shen J, Allakhverdiev S I (2015) Comparison of nano-sized Mn oxides with the Mn cluster of photosystems II as catalysts for water oxidation. Biochim. Biophys. Acta, Bioenerg. 1847 294-306 Navrotsky A, Ma C, Lilova K, Birkner N (2010) nanophase transition metal oxides show large thermodynamically driven shifts in oxidation-reduction equilibria. Science 330 199-201 Nelson N, Ben-Shem A (2004) The complex architecture of oxygenic photosynthesis. Nat. Rev. Mol. Cell Biol. 5 971-982 Nguyen KT and Zhao Y (2014) Integrated graphene/nanoparticle hybrids for biological and electronic applications. Nanoscale,6, 6245 –6266 Nielsen R M, Murphy S, Strebel C, Johansson M, Chorkendorff I, Nielsen J H (2010) The morphology of mass selected ruthenium nanoparticles from a magnetron-sputter gas-aggregation source. J. Nanopart. Res. 12, 1249–1262 120 Nocera D G (2012) The artificial leaf. Acc Chem Res 45 767-776 Novoselov K S, Geim A K, Morozov S V, Jiang D, Zhang Y, Dubonos S V, Grigorieva I V and Firsov A A (2004) Electric Field Effect in Atomically Thin Carbon Films. Science 306, 666–669 Palmer E R, Cai R, Vernieres J (2018) Synthesis without Solvents: The cluster (nanoparticle) beam route to catalysts and sensors. Acc. Chem. Res. 51, 2296-2304 Paoli E A, Masini F, Frydendal R, Deiana D, Malacrida P, Hansen T W, Chorkendorff I, Stephens I E L (2016) Fine-tuning the activity of oxygen evolution catalysts: The effect of oxidation pre-treatment on size-selected Ru nanoparticles. Catal. Today 262 57-64 Patel M, Kim H-S, Patel D B, Kim J (2016) CuO photocathode-embedded semitransparent photoelectrochemical cell. J. Mater. Res. 31 3205-3213 Paul W and Steinwedel H (1953) Ein neues massenspektrometer ohne magnetfeld. Z. Naturforsch., A: Phys. Sci.8, 448–450 Paul W, Reinhard H P and von Zahn U (1958) The electric mass filter as a mass spectrometer and isotop separator. Z. Phys. 152, 143-182 Pauly H (2000) Atom, molecule, and cluster beams. Springer, Berlin Peng H C, Park J, Zhang L and Xia Y (2015) Toward a Quantitative Understanding of Symmetry Reduction Involved in the Seed-Mediated Growth of Pd Nanocrystals. J. Am. Chem. Soc.137, 6643–6652 Penza M, Rossi R, Alvisi M, Signore M A, Cassano G, Dimaio D, Pentassuglia R, Piscopiello E, Serra E and Falconieri M (2009) Characterization of metal-modified and vertically-aligned carbon nanotube films for functionally enhanced gas sensor applications. Thin Solid Films 517, 6211–6216 Pfau P, Sattler K, Muhlbach J, Pflaum R, Recknagel E (1982) Influence of condensation parameters on the size distribution of metal clusters. J. Phys. F: Met. Phys. 12, 2131-2139 121 Planeix J M, Coustel N, Coq B, Brotons V, Kumbhar P S, Dutartre R, Geneste P, Bernier P and Ajayan P M (1994) Application of carbon nanotubes as supports in heterogeneous catalysis. J. Am. Chem. Soc. 116, 7935-36 Polonskyi O, Solar P, Kylian O, Drabik M, Artemenko A, Kousal J, Hanus J, Pesicka J, Matolinova I, Kolibalova E, Slavinska D, Biederman H (2012) Nanocomposite metal/plasma polymer films prepared by means of gas aggregation cluster source. Thin Solid Films 520, 4155-4162 Popok V N, Barke I, Campbell E E B, Meiwes-Broer K-H (2011) Cluster-surface interaction: From soft landing to implantation. Surf. Sci. Rep. 66 347-377 Pratontep S, Carroll S J, Xirouchaki C, Streun M, Palmer RE (2005) Size-selected cluster beam source based on radio frequency magnetron plasma sputtering and gas condensation. Rev. Sci. Instrum. 76, 045103 Quesnel E, Pauliac-Vaujour E, Muffato V (2010) Modeling metallic nanoparticle synthesis in a magnetron-based nanocluster source by gas condensation of a sputtered vapor. J. Appl. Phys. 107, 054309 Raicopol M, Pruna A and Pilan L (2013) Supercapacitance of single-walled carbon nanotubes- polypyrrole composites. J. Mater. Chem. 1, 258-261 Ramírez A, Hillebrand P, Stellmach D, May M M, Bogdanoff P, Fiechter S (2014) Evaluation of MnOx, Mn2O3, and Mn3O4 electrodeposited films for the oxygen evolution reaction of water. J. Phys. Chem. C 118 14073-14081 Rao A, Eklund P, Bandow S, Thess A and Smalley R E (1997) Evidence for charge transfer in doped carbon nanotube bundles from Raman scattering. Nature 388, 257-259. Riggs J E, Guo Z, Carroll D L and Sun Y-P (2000) Strong luminescence of solubilized carbon nanotubes. J. Am. Chem. Soc. 122, 5879-5880. Robinson D M, Go Y B, Mui M, Gardner G, Zhang Z, Mastrogiovanni D, Garfunkel E, Li J, Greenblatt M, Dismuke G C (2013) Photochemical water oxidation by crystalline polymorphs of manganese oxides: Structural requirements for catalysis. J. Am. Chem. Soc. 135 3494-3501 122 Rudd R, Obrusnik A, Zikan P, Pratt R, Hall C, Murphy P, Evans D, Charrault E (2017) Manipulation of cluster formation through gas-wall boundary conditions in large area cluster sources. Surf. Coat. Technol. 314, 125-130 Sahu R P, Abdalla AM, Abdel Fattah A R, Ghosh S and Puri I K (2018) Synthesis, characterization, and applications of carbon nanotubes functionalized with magnetic nanoparticles. In: Balasubramanian G (ed), Advances in nanomaterials, fundamentals, properties and applications, Springer, Switzerland, pp 37-57 Santos D M F, Sequeira C A C and Figueiredo J L (2013) Hydrogen production by alkaline water electrolysis. Quim. Nova 36 1176-1193 Scarselli M, Camilli L, Castrucci P, Nanni F, Del Gobbo S, Gautro E, Frant S and Crescenzi D (2012) In situ formation of noble metal nanoparticles on multiwalled carbon nanotubes and its implication in metal–nanotube interactions. Carbon 50, 875-884 Schü nemann C, Schä ffel F, Bachmatiuk A, Queitsch U, Sparing M, Rellinghaus B, and Rü mmeli M H (2011) Catalyst poisoning by amorphous carbon during carbon nanotube growth: Fact or fiction? ACS Nano 5, 8928-8934. Serp P and Castillejos E (2010) Catalysis in Carbon Nanotubes. ChemCatChem 2, 41–44 Shyjumon I, Gopinadhan M, Helm C A, Smirnov B M, Hippler R (2006) Deposition of titanium/titanium oxide clusters produced by magnetron sputtering. Thin Solid Films 500, 41-51 Simpson J A (1961) Design of retarding field energy analyzers. Rev. Sci. Instrum. 32, 1283- 1293 Singh A, Spiccia L (2013) Water oxidation catalysts based on abundant 1st-row transition metals. Coord. Chem. Rev. 257 2607-2622 Sivula K, Le Formal F, Grätzel M (2011) Solar water splitting: Progress using hematite (α- Fe2O3) photoelectrodes. ChemSusChem 4 432-449 Smirnov BM (2000) Clusters and small particles. Springer, New York Smirnov BM (2010) Nanoclusters and microparticles in gases and vapors. Walter de Gruyter, Germany 123 Soares J A N T (2014) Introduction to optical characterization of materials: In Sardela M (ed.) Practical materials characterization. Springer, New York, pp 43-92 Sreedhar A, Sreekanth T V M, Kwon J H, Yi J, Sohn Y, Gwag J S (2017) Ag nanoparticles decorated ion-beam-assisted TiO2 thin films for tuning the water splitting activity from UV to visible light harvesting. Ceram. Int. 43 12814-12821 Srivastava S, Thomas J P, Rahman M A, Abd-Ellah M, Mohapatra M, Pradhan D, Heinig N F, Leung K T (2014) Size-selected TiO2 nanocluster catalysts for efficient photoelectrochemical water splitting. ACS Nano 8 11891-11898 Srivastava S, Thomas J P, Heinig N, Abd-Ellah M, Rahman M A, Leung K T (2017) Photoelectrochemical water splitting on ultrasmall defect-rich TaOx nanoclusters enhanced by size-selected Pt nanocluster promoters. Nanoscale 9 14395-14404 Steckelmacher W (1973) Energy analysers for charged particle beams. J. Phys. E: Sci. Instrum. 6 1061-1071 Stranak V, Drache S, Wulff H, Hubicka Z, Tichy M, Kruth A, Helm C A, Hippler R (2015) Oxidation behavior of Cu nanoparticles embedded into semiconductive TiO2 matrix. Thin Solid Films 589 864-871 Su SS, Chang I (2018) Review of Production Routes of Nanomaterials. In: Barbazon D, Pellicer E, Zivic F, Sort J, Baro M D, Grujovic N, Choy K-L (eds), Commercialization of nanotechnologies - A case study approach. Springer, Switzerland AG, pp 15-31 Subrahmanyam K S, Manna A K, Pati S K and Rao C N R (2010) A study of graphene decorated with metal nanoparticles. Chem. Phys. Lett 497, 70–75 Takashima T, Hashimoto K, Nakamura R (2012) Mechanisms of pH-dependent activity for water oxidation to molecular oxygen by MnO2 electrocatalysts. J. Am. Chem. Soc. 134 1519-1527 Tans S J, Verschueren A R M and Dekker C (1998) Room-temperature transistor based on a single carbon nanotube, Nature 393, 49-52 Tans S J, Devoret M H, Dait H, Thess A, Smalley R E, Geerligs L J and Dekker C (1997) Individual single-wall carbon nanotubes as quantum wires. Nature 386,474–477 124 Tauc J (1968) Optical properties and electronic structure of amorphous Ge and Si. Mater. Res. Bull. 3 37-46 Tauc J, Grigorovici R, Vancu A (1966) Optical properties and electronic structure of amorphous Gemanium. Phys. Status Solidi B 15 627-637 Tyo E C, Vajda S (2015) Catalysis by clusters with precise numbers of atoms. Nat. Nanotechnol. 10 577-588 Umena Y, Kawakami K, Shen J-R and Kamiya N (2011) Crystal structure of oxygen- evolving photosystem II at a resolution of 1.9 Ȧ. Nature 473 55-60 Vajda S, White M G (2015) Catalysis applications of size-selected cluster deposition. ACS Catal. 5 7152–7176 Vinyard D J, Brudvig G W (2017) Progress toward a molecular mechanism of water oxidation in photosystem II. Annu. Rev. Phys. Chem. 68 101-116 Voggu R, Pal S, Pati S K and Rao C N R (2008) Semiconductor to metal transition in SWNTs caused by interaction with gold and platinum nanoparticles. J. Phys.: Condens. Matter 20, 215211-6 Wallace P R (1947) The Band Theory of Graphite. Phys. Rev. 71, 622–634 Wang D, Astruc D (2017) The recent development of efficient earth-abundant transition- metal nanocatalysts. Chem. Soc. Rev. 46 816-854 Wang G, Sun C, Caia Y, Ma Y, Ali Syed J, Wang H, Cao Z and Meng X (2019) Improvement of the interface and electrical properties in carbon nanotube_nanocrystalline copper composite films. Mater. Chem. Phys 223, 374- 379 Wang J, Cui W, Liu Q, Xing Z, Asiri, A M, and Sun X (2016) Recent progress in cobalt- based hetrogeneous catalysts for electrochemical water splitting. Adv. Mater. 28, 215-230 Wegner K, Piseri P, Tafreshi H V, Milani P (2006) Cluster beam deposition: A tool for nanoscale science and technology. J. Phys. D: Appl. Phys. 39, R439-459 Wei W, Cui X, Chen W, Ivey D G (2011) Manganese oxide-based materials as electrochemical supercapacitor electrodes. Chem. Soc. Rev. 40 1697-1721 125 White C T and Mintmire J W (2005) Fundamental properties of single-wall carbon nanotubes. J. Phys. Chem. B, 109, 52-56 Wiechen M, Spiccia L (2014) Manganese oxides as efficient water oxidation catalysts. ChemCatChem 6 439-441 Wilgoose G G, Banks C E and Compton R G (2006) Metal nanoparticles and related materials supported on carbon nanotubes: methods and applications, Small, 2, 182– 193 Williams D B and Carter C B (Eds.) (2016) Transmission electron microscopy, Springer, Switzerland Yachandra V K, Sauer K, Klein M P (1996) Manganese cluster in photosynthesis: where plants oxidize water to dioxygen. Chem. Rev. 96, 2927−2950 Yang M, Kim H C and Hong S-H (2012) DMMP gas sensing behavior of ZnO-coated single-wall carbon nanotube network sensors, Mater. Lett 89, 312-315 Yano J, Kern J, Sauer K, Latimer M J, Pushkar Y, Biesiadka J, Loll B, Saenger W, Messinger J, Zouni A, Yachandra V K (2006) Where water is oxidized to dioxygen: Structure of the photosynthetic Mn4Ca cluster. Science 314 821-825 Yoshii K, Yamaji K, Tsuda T, Matsumoto H, Sato T, Izumi R, Torimotoc T and Kuwabata S(2016) Highly durable Pt nanoparticle-supported carbon catalysts for the oxygen reduction reaction tailored by using an ionic liquid thin layer. J. Mater. Chem. A. 4, 12152–12157 Zaharieva I, Chernev P, Risch M, Klingan K, Kohlhoff M, Fischer A, Dau H (2012) Electrosynthesis, functional, and structural characterization of water-oxidizing manganese oxide. Energy Environ. Sci. 5 7081-7089 Zeng K, Zhang D (2010) Recent progress in alkaline water electrolysis for hydrogen production and applications. Prog. Energy Combust. Sci. 36 307–326 Zhang Y and Dai H (2000) Formation of metal nanowires on suspended single-walled carbon nanotubes. Appl. Phys. Lett.77, 3015-3017 126 Zhang Y, Franklin N W, Chen R J and Dai H (2000) Metal coating on suspended carbon nanotubes and its implication to metal-tube interaction. Chem. Phys. Lett. 331, 35- 41 Zhao J, Baibuz E, Vernieres J, Grammatikopoulos P, Jansson V, Negal M, Steinhauer S, Sowwan M, Kuronen A, Nordlundt K, Djurabekova F (2016) Formation mechanism of Fe nanocubes by magnetron sputtering inert gas condensation. ACS Nano 10, 4684−4694 Zhong Q, Innlss D, Kjoller K and Ehngs V B (1993) Fractured polymer/silica fiber surface studied by tapping mode atomic force microscopy. Surf. Sci. Lett. 290, L688-L692 Zhou Z Y, Tian N, Li J T, Broadwell I and Sun S G (2011) Nanomaterials of high surface energy with exceptional properties in catalysis and energy storage, Chem. Soc. Rev. 40, 4167–4185 Zhu J, Pang S, Dittrich T, Gao Y, Nie W, Cui J, Chen R, An H, Fan F, Li C (2017) Visualizing the nanocatalyst aligned electric fields on single photocatalyst particles. Nano Lett. 17 6735-6741 Zimmermann U, Malinowski N, Näher U, Frank S, Martin T P (1994) Producing and detecting very large clusters. Z. Phys. D: At. Mol. Clusters 31, 85-93 127 6 Appendix A: AutoCad design 6.1 Part A: Substrate holder design Figure 6.1: Substrate holder design. Different parts are labeled by numbers on the figure; part (1) long bracket; part (2) bracket holder; part (3) disk, part (4) frame. Based on the geometry of substrate, two different geometries for the opening part of the frame were designed. The substrate is held between disk and frame; part (5) short bracket to hold collimator/Shield. 128 Part A1: Long bracket Side *This part was designed for loading either 3 substrates (blue color) or 4 substrates (red color) based on needs. Front 129 Part A2: Bracket holder 130 Part A3: Disk 131 Part A4: Frame (A and B) Frame A: Circular opening Frame B: Square opening 132 Part A5: Short bracket to hold collimator/Shield 133 6.2 Part B: QCM mounting flange Figure 6.2: QCM mounting flange Different parts are labeled by numbers on the figure; part (1) stepper motor; part (2) linear motion micrometer; part (3) flange; part (4) QCM holder; part (5) QCM. The details of AutoCAD designs of part (3), and (4) are on pages 134, 135. 134 Part B3: QCM mounting flange 135 Part B4: QCM holder 136 6.3 Part C: Heater holder and TEM grid holder Figure 6.3: Heater setup and TEM grid holder mounting flange. Different parts are labeled by numbers on the figure; part (1) heater holder and part (2) TEM grid holder. The details of their AutoCAD design are on pages 137 and 138. 137 Part C1: Heater holder 138 Part C2: TEM grid holder

Abstract (if available)

Abstract The pursuit of nanoscale-based materials in which the size and composition of the constituent building blocks can be accurately manipulated is an especially promising direction of inquiry. The study of size-selected nanoclusters focuses on exactly this target: by tracing the evolution of nanoscale metal particle properties atom by atom, one can observe, tune, and identify the optimal system with maximum precision. A productive pathway to building structures based on size optimized nanoclusters is cluster beam deposition, which is a new and rapidly developing tool. This technique allows flexible tuning of the relevant parameters and a wide choice of working materials. ❧ In this study, a dc magnetron sputtering gas aggregation source equipped with a quadrupole mass filter is employed for producing robust fluxes of metal nanocluster ions, filtering them by size, and soft landing them on a substrate. The performance of this cluster source was characterized in detail by investigating the role of a range of overlapping operating source parameters in nanocluster production. A detailed map of the inﬂuence of each parameter on the average nanocluster size was established. In this way, it is possible to identify the main contribution of each parameter to the physical processes taking place within the source. These assignments were supplemented by nanocluster beam characterization, including measurements of the abundance ratios of charged to neutral and negative to positive nanoclusters, and of the size dependence of nanocluster velocity. Finally, the morphology and size of deposited nanoclusters were characterized by atomic force microscopy (see Chapter 2). ❧ In the subsequent parts of this thesis, we address two subjects. The first is concerned with the size dependence of the chemical activity of metal nanoclusters, and the second is an investigation of the interaction between metal nanoclusters and single-wall carbon nanotubes (SWCNTs). ❧ The size dependence of the chemical activity of metal nanoclusters was approached through a study of the application of size-selected MnO nanocluster thin films in electrocatalytic water splitting. Different sizes of Manganese nanoclusters (below 10 nm) were produced in the nanocluster source and soft-landed onto conducting electrodes. The mass loading of these catalytic particles was kept constant and corresponded to sub-monolayer coverage. Measurements of the water oxidation threshold revealed that the onset potential decreases significantly with decreasing particle size. The ability of such a sub-monolayer film to lower the reaction threshold signifies that an inherent change takes place in the charge transfer energetics. This suggests that the key role is played by intrinsic size effects, i.e., by changes in the electronic properties and surface fields of the nanoparticles with decreasing size. We anticipate that this work will serve to bridge the knowledge gap between bulk thick film electrocatalysts and natural photosynthetic molecular-cluster complexes (see Chapter 3). ❧ The investigation of interactions between metal nanoclusters and SWCNTs addressed the goal of decorating SWCNTs with size optimized metal nanoclusters to assemble novel nanoscale structures. This goal requires special fabrication techniques in order to keep the physical and chemical properties of metal nanoclusters and SWCNTs pristine. ❧ The novelty of our work is signified by the deposition of pure (ligand-free) size-selected metal nanoclusters onto a naked suspended SWCNT, which is not possible with other physical or chemical approaches. A set of procedures was developed to prepare high-quality electron microscope grids with opening bridged by SWCNTs for use as high-quality nanoparticle deposition substrates, and imaging of the obtained samples was performed. This marks a significant step toward the realization of nanoparticle-nanowire assemblies employing SWCNT transport measurements as a probe of nanocluster electronic structure (see Chapter 4).

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The adsorption and selective deposition of molecular and nanocluster ions on carbon based devices

PDF

Design and characterization of metal and semiconducting nanostructures and nanodevices

PDF

Temperature-dependent photoionization and electron pairing in metal nanoclusters

PDF

Size-resolved particulate matter (PM) in urban areas: toxico-chemical characteristics, sources, trends and health implications

PDF

Fabrication and study of organic and inorganic optoelectronics using a vapor phase deposition (VPD)

PDF

Plasmonic enhancement of catalysis and solar energy conversion

PDF

Semiconducting metal oxide nanostructures for scalable sensing applications

PDF

Controlled synthesis, characterization and applications of carbon nanotubes

PDF

Characterization, process analysis, and recycling of a benzoxazine-epoxy resin for structural composites

PDF

Characterization of black carbon: from source to evolution of physical and optical properties in the atmosphere

PDF

Chemical vapor deposition of graphene and two-dimensional materials: synthesis, characterization, and applications

PDF

Printed electronics based on carbon nanotubes and two-dimensional transition metal dichalcogenides

PDF

The modification of catalysts and their supports for use in various fuel cells

PDF

Development of a novel heterogeneous flow reactor: soot formation and nanoparticle catalysis

PDF

Fabrication and application of plasmonic nanostructures: a story of nano-fingers

PDF

Metallic syntactic foams synthesis, characterization and mechnical properties

PDF

Oxidative potential of urban atmospheric particles: spatiotemporal trends and associations with source-specific chemical components

PDF

3D printing of polymeric parts using Selective Separation Shaping (SSS)

PDF

From cables to biofilms: electronic and electrochemical characterization of electroactive microbial systems

PDF

Material and process development and optimization for efficient manufacturing of polymer composites

Asset Metadata

Creator Khojasteh, Malak (author)

Core Title Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Chemical Engineering

Publication Date 07/26/2019

Defense Date 06/10/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag decoration of single-wall carbon nanotubes,gas aggregation source,magnetron sputtering,mass spectrometry,nanocluster thin films,nanoclusters,nanoparticle catalysts,nanoparticle-nanowire assemblies,nanoparticles,OAI-PMH Harvest,size-selective deposition,water oxidation

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Armani, Andrea (committee chair), Kresin, Vitaly (committee member), Nutt, Steven (committee member)

Creator Email khojaste@usc.edu,malak_khojasteh@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-193612

Unique identifier UC11662791

Identifier etd-KhojastehM-7638.pdf (filename),usctheses-c89-193612 (legacy record id)

Legacy Identifier etd-KhojastehM-7638.pdf

Dmrecord 193612

Document Type Dissertation

Format application/pdf (imt)

Rights Khojasteh, Malak

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA