21 |
A combined experimental and theoretical approach towards the understanding of transport in one-dimensional molecular nanostructuresGrimm, Daniel 09 July 2008 (has links)
This thesis comprises detailed experimental and theoretical investigations of the transport properties of one-dimensional nanostructures. Most of the work is dedicated to the exploration of the fascinating effects occurring in single wall carbon nanotubes (SWCNT). These particular nanostructures gained an overwhelming interest in the past two decades due to its outstanding electronic and mechanical features. We have investigated the properties of a novel family of carbon nanostructures, named here as Y-shaped rings. The studies show that they present very interesting quantum interference effects. A high structural stability under tensile strain and elevated temperatures is observed. Within the semi-classical potential adopted, the critical strain values of structure rupture lie in the same range of their pristine SWCNT counterparts. This is directly verified by the first observations of these ring-like structures in a transmission electron microscopy. A merging process of asymmetric into symmetric rings is investigated in-situ under electron beam irradiation at high temperatures. The electronic properties of these systems are theoretically studied using Monte Carlo simulations and environment dependent tight-binding calculations. From our results, we address the possibility of double-slit like interference processes of counter-propagating electron waves in the ring-like structures. The nature of well defined, sharp peaks in the density of states are determined as the discrete eigenenergies of the central loop part. Furthermore, the formation and dispersion of standing waves inside the ring is shown to originate from the quantum-dot like confinement of each branch between the leads. The obtained dispersion relation is shown to be the same occurring in purely one-dimensional quantum dots of similar geometries. Furthermore, Fabry-Perot-like interferences are observed. We established at the IFW a bottom-up processing route to fabricate nanotube based electronic devices. The SWCNTs are grown by chemical vapor deposition and we present a detailed study of the different approaches to obtain individual nanotubes suitable for a successful integration into electronic devices. Wet-chemistry and ultra-thin films as well as ferritin were employed as catalyst particles in the growth of SWCNT samples. By adjusting the optimized process parameters, we can control the obtained yield from thick nanotube forests down to just a couple of free-standing individual SWCNTs. The nanotubes are localized, contacted by standard e-beam lithography and characterized at ambient- as well as liquid helium temperatures. We usually obtain quite transparent contacts and the devices exhibit metallic or a mixed metallic/semiconducting behavior. The well-known memory effect upon gate voltage sweeping as well as single electron tunneling in the Coulomb blockade regime are addressed.
|
22 |
Epitaxy and characterization of SiGeC layers grown by reduced pressure chemical vapor depositionHållstedt, Julius January 2004 (has links)
Heteroepitaxial SiGeC layers have attracted immenseattention as a material for high frequency devices duringrecent years. The unique properties of integrating carbon inSiGe are the additional freedom for strain and bandgapengineering as well as allowing more aggressive device designdue to the potential for increased thermal budget duringprocessing. This work presents different issues on epitaxialgrowth, defect density, dopant incorporation and electricalproperties of SiGeC epitaxial layers, intended for variousdevice applications. Non-selective and selective epitaxial growth of Si1-x-yGexCy(0≤x≤30, ≤y≤0.02) layershave been optimized by using high-resolution x-ray reciprocallattice mapping. The incorporation of carbon into the SiGematrix was shown to be strongly sensitive to the growthparameters. As a consequence, a much smaller epitaxial processwindow compared to SiGe epitaxy was obtained. Differentsolutions to decrease the substrate pattern dependency (loadingeffect) of SiGeC growth have also been proposed. The key pointin these methods is based on reduction of surface migration ofthe adsorbed species on the oxide. In non-selective epitaxy,this was achieved by introducing a thin silicon polycrystallineseed layer on the oxide. The thickness of this seed layer had acrucial role on both the global and local loading effect, andon the epitaxial quality. Meanwhile, in selective epitaxy,polycrystalline stripes introduced around the oxide openingsact as migration barriers and reduce the loading effecteffectively. Chemical mechanical polishing (CMP) was performedto remove the polycrystalline stripes on the oxide. Incorporation and electrical properties of boron-doped Si1-x-yGexCylayers (x=0.23 and 0.28 with y=0 and 0.005) with aboron concentration in the range of 3x1018-1x1021atoms/cm3 have also been investigated. In SiGeClayers, the active boron concentration was obtained from thestrain compensation. It was also found that the boron atomshave a tendency to locate at substitutional sites morepreferentially compared to carbon. These findings led to anestimation of the Hall scattering factor of the SiGeC layers,which showed good agreement with theoretical calculations. Keywords:Silicon germanium carbon (SiGeC), Epitaxy,Chemical vapor deposition (CVD), Loading effect, Highresolution x-ray diffraction (HRXRD), Hall measurements, Atomicforce microscopy (AFM).
|
23 |
Synthesis of Graphene - Carbon Nanotube Hybrid StructuresPaul, Aniruddha January 2021 (has links)
Graphene and Carbon nanotubes (CNTs) have been researched for more than a decade due to their extraordinary properties and advantages towards applications like electronics, structural re-enforcements, thermal management and energy storage. Graphene-CNT hybrid structures have been predicted to further enhance the exceptional properties and overcome some of the shortcomings of the individual materials. Advantages of a structure consisting of vertically aligned carbon nanotubes (VACNTs) covalently bonded with graphene layers have been predicted to be especially favourable for applications like TIM, supercapacitors and battery electrodes. This project investigates two growth mechanisms for obtaining Graphene-VACNT structures using scalable processes. Shortcomings of previously done research on similar structures like graphene transfer and bad CNT alignment is solved. A novel growth mechanism is also investigated to set the foundation for research into a new approach to grow Graphene-CNT hybrid structures in the future. Chemical Vapor Deposition (CVD) was the method used to grow the graphene and CNT structures. The characterization was done using optical microscopy, Scanning electron microscopy (SEM) and Raman spectroscopy. / Grafen- och kolnanorör (CNT) har forskats i mer än ett decennium på grund av deras extraordinära egenskaper och fördelar gentemot applikationer som elektronik, strukturförstärkning, termisk hantering och energilagring. Grafen-CNT hybridstrukturer har förutspåtts ytterligare förbättra de exceptionella egenskaperna och övervinna några av bristerna i de enskilda materialen. Fördelar med en struktur som består av vertikalt inriktade kolnanorör (VACNT) som är kovalent bundna med grafenskikt har förutspåtts vara särskilt fördelaktiga för applikationer som TIM, superkondensatorer och batterielektroder. Detta projekt undersöker två tillväxtmekanismer för att erhålla Graphene-VACNT-strukturer med hjälp av skalbara processer. Brister i tidigare utförd forskning om liknande strukturer som grafenöverföring och dålig CNT-anpassning är lösta. En ny tillväxtmekanism undersöks också för att lägga grunden för forskning om ett nytt tillvägagångssätt för att växa Graphene- CNT hybridstrukturer i framtiden. Chemical Vapor Deposition (CVD) var metoden som användes för att odla grafen- och CNT-strukturerna. Karakteriseringen gjordes med optisk mikroskopi, Scanning electron microscopy (SEM) och Raman-spektroskopi.
|
24 |
Interfaces dans les matériaux céramiques multicouchesThibaud, Simon 22 December 2010 (has links)
L’augmentation du nombre d’interfaces dans une matrice céramique permet d’améliorer sa ténacité. L’étude de la structure feuilletée de la nacre a démontré que cette ténacité pouvait être accrue par la présence de pontages entre les couches. Dans la première partie, le modèle de décohésion proposé par Pompidou et al. a été utilisé pour choisir un bicouche dont l’interface est naturellement favorable aux décohésions. Compte tenu du contexte de l’étude, cette analyse a permis de choisir le couple SiC/pyC comme bi-couche de base pour l’étude des interfaces. Par la suite, des matrices multicouches modèles (SiC/pyC)n (SiC, carbure de silicium issu du mélange CH3SiCl3/H2 – pyC, pyrocarbone à partir du propane) ont été élaborées par dépôt chimique en phase vapeur (CVD). Deux voies de pontage ont été abordées. La première met en œuvre une discontinuité entre les couches : les conditions d’élaboration ont été optimisées de façon à contrôler la croissance de couches minces massives et le développement de particules de surface (submicroniques) faisant office de pontage. La deuxième est basée sur un gradient de composition entre les couches de SiC grâce au développement d’une couche de SiC riche en co-dépôt de carbone, une interphase mixte est créée. Le pontage est assuré par la présence simultanée dans les couches à gradient de composition de grains de SiC et d’une phase carbonée. Les propriétés physico-chimiques et structurales des différents éléments des matrices ont été analysées et les différents comportements des fissures dans chacune des matrices ont été observés à la suite d’essais mécaniques. / The improvement of ceramic matrix toughness may be achieved through the presence of interfaces. Moreover, studies on a mother of pearl structure have shown the usefulness of mineral bridges between the layers. On the first part of this work, the Pompidou model was used for the selection of a bi-layered ceramic with an interface which is naturally favorable to crack deflection. SiC/pyC was taken as basic material for the interfaces study. Then, multilayered ceramic matrices (SiC/pyC)n (silicon carbide from CH3SiCl3/H2 mixture – pyC from propane) were fabricated using chemical vapor deposition (CVD). In the study, two bypass ways were proposed. On the one hand, a physical discontinuity exists between the different layers: elaboration parameters were optimized in order to develop both bulk layers and submicronic surface particles, acting as ceramic bypass. On the other hand, composition gradient films were developed between each SiC layers: by realizing carbon rich SiC layers, a mixed interphase was created. The presence of both SiC grains and carbon phases ensures the bypass structure. Physico-chemical and structural properties of multilayered ceramic matrices were analyzed and the crack propagation in each of them was observed following mechanical tests.
|
25 |
Quantum Chemical Feasibility Study of Methylamines as Nitrogen Precursors in Chemical Vapor DepositionRönnby, Karl January 2015 (has links)
The possibility of using methylamines instead of ammonia as a nitrogen precursor for the CVD of nitrides is studied using quantum chemical computations of reaction energies: reaction electronic energy (Δ𝑟𝐸𝑒𝑙𝑒𝑐) reaction enthalpy (Δ𝑟𝐻) and reaction free energy (Δ𝑟𝐺). The reaction energies were calculated for three types of reactions: Uni- and bimolecular decomposition to more reactive nitrogen species, adduct forming with trimethylgallium (TMG) and trimethylaluminum (TMA) followed by a release of methane or ethane and surface adsorption to gallium nitride for both the unreacted ammonia or methylamines or the decomposition products. The calculations for the reaction entropy and free energy were made at both STP and CVD conditions (300°C-1300°C and 50 mbar). The ab inito Gaussian 4 (G4) theory were used for the calculations of the decomposition and adduct reactions while the surface adsorptions were calculated using the Density Functional Theory method B3LYP. From the reactions energies it can be concluded that the decomposition was facilitated by the increasing number of methyl groups on the nitrogen. The adducts with mono- and dimethylamine were more favorable than ammonia and trimethylamine. 𝑁𝐻2 was found to be most readily to adsorb to 𝐺𝑎𝑁 while the undecomposed ammonia and methylamines was not willingly to adsorb.
|
26 |
Development Of A Tin Oxide Based Thermoelectric Gas Sensor For Volatile Organic CompoundsAnuradha, S 01 1900 (has links)
Today there is a great deal of interest in the development of gas sensors for applications like air pollution monitoring, indoor environment control, detection of harmful gases in mines etc. Based on different sensing principles, a large variety of sensors such as semiconductor gas sensors, thermoelectric gas sensors, optical sensors and thermal conductivity sensors have been developed.
The present thesis reports a detailed account of a novel method followed for the design and development of a thermoelectric gas sensor for sensing of Volatile Organic
Compounds.
Thermoelectric effect is one of the highly reliable and important working principles that is widely being put into practical applications. The thermoelectric property of
semiconducting tin oxide film has been utilized in the sensor that has been developed. The thermoelectric property of semiconducting tin oxide film has been utilized in the sensor. The deposition parameters for sputtering of tin oxide film have been optimized to obtain a high seebeck coefficient. A test set-up to characterize the deposited films for their thermoelectric property has been designed and developed.
A novel method of increasing the seebeck coefficient of tin oxide films has been
successfully implemented. Thin films of chromium, copper and silver were used for this purpose. Deposition of the semiconducting oxide on strips of metal films has led to a noticeable increase in the seebeck coefficient of the oxide film without significantly affecting its thermal conductivity.
The next part of our work involved development of a gas sensor using this thermoelectric film. These sensors were further tested for their response to volatile organic compounds. The sensor showed significant sensitivity to the test gases at relatively low temperatures.
In addition to this, the developed sensor is also selective to acetone gas.
|
27 |
Ανάπτυξη μεθόδων παραγωγής νανοσωλήνων άνθρακα μέσω χημικής απόθεσης από ατμόΚουράβελου, Αικατερίνη 14 December 2009 (has links)
Στόχος της διδακτορικής αυτής διατριβής ήταν η ανάπτυξη μιας μεθόδου παραγωγής νανοσωλήνων άνθρακα η οποία στηρίζεται στη χημική απόθεση ατμών, χρησιμοποιώντας ως πηγή του άνθρακα ενώσεις σε υγρή μορφή, όπως οι αλκοόλες. Επιπρόσθετα μελετήθηκαν διάφορες παράμετροι της πειραματικής διαδικασίας (πηγή άνθρακα, θερμοκρασία απόθεσης, είδος και συγκέντρωση μετάλλου και υποστρώματος, παρουσία υδρογόνου κ.ά), τόσο ως προς την επίδρασή τους στο ρυθμό εξέλιξης της διεργασίας, όσο και ως προς το είδος των παραγόμενων προϊόντων, με σκοπό τη στοχευμένη παραγωγή νανοσωλήνων άνθρακα με συγκεκριμένες ιδιότητες.
Η κύρια πειραματική διάταξη αποτελούνταν από έναν θερμοβαρομετρικό αντιδραστήρα, ο οποίος επέτρεπε τη συνεχή μέτρηση των μεταβολών του βάρους του δείγματος σε συνάρτηση με το χρόνο, ενώ και φασματογράφος μάζας ήταν συνδεδεμένος στην έξοδο του αντιδραστήρα για να μελετηθεί η αέρια φάση των αντιδράσεων. Τα προϊόντα προκειμένου να πιστοποιηθούν ως προς το είδος των νανοσωλήνων που παρήχθησαν, χαρακτηρίστηκαν με τη βοήθεια ηλεκτρονικής μικροσκοπίας σάρωσης (SEM) και διερχόμενης δέσμης (TEM), καθώς και με φασματοσκοπία Raman και θερμοσταθμική ανάλυση (TGA).
Τα αποτελέσματα των πειραμάτων οδήγησαν στο συμπεράσμα πως οι ατμοί της αιθανόλης είναι καλύτερη πηγή άνθρακα σε σύγκριση με της μεθανόλης, οδηγώντας μάλιστα στη παραγωγή μίγματος πολυφλοιϊκών και μονοφλοιϊκών νανοσωλήνων άνθρακα, με καθαρότητες που ξεπερνούσαν το 90%. Επιτακτική αποδείχθηκε η παρουσία του μετάλλου, το οποίο και αποτελεί το κέντρο πυρημοποιήσης για την ανάπτυξη των νανοσωλήνων, ενώ καθοριστική είναι και η χρήση υποστρώματος προκειμένου ο άνθρακας να αποτεθεί με τη μορφή αυτή. Επιπρόσθετα, η παρουσία του υδρογόνου αύξησε σημαντικά το ποσοστό του άνθρακα που αποτέθηκε οδηγώντας μάλιστα στο σχηματισμό μεταλλικών μονοφλοικών νανοσωλήνων άνθρακα πολύ μικρής διαμέτρου, η οποία υπολογίστηκε ίση με 0.45nm. / The main goal of this research was the development of a new method for the production of carbon nanotubes, based on chemical vapor deposition (CVD), which employs a liquid carbon source. In addition, a detailed investigation of the effect of several parameters (carbon source, deposition temperature, kind and metal concentration and support, hydrogen addition e.t.c.) on both the process and the final carbon product was carried out.
For this purpose, a CVD experimental apparatus was developed, which uses vapors of liquid precursors and allows the continuously recording of sample weight changes in correlation with time. In some cases, a mass spectrometer was used as a way to determine the kind of processes that take place in the gas phase during carbon deposition. The solid product was characterized using scanning electron microscopy (SEM), transmission electron microscopy (TEM), Raman spectroscopy and thermogravimetric analysis (TGA).
The experimental results led to the conclusion that ethanol vapors are preferable because they lead to higher yield of both multi-wall and single-wall carbon nanotubes. Also, it was proved that the presence of a metal catalyst and support is necessary, because the first one is the active site of carbon nanotubes formation, and the second leads to the deposition of carbon in this form. Additionally, it was proved that the use of hydrogen in the gas mixture of the process is very important, as a way to reduce catalyst, leading to the formation of metalic single-wall carbon nanotubes of very small diameter (0.45nm).
|
28 |
Mechanical Behavior Study of Microporous Assemblies of Carbon Nanotube and GrapheneReddy, Siva Kumar C January 2015 (has links) (PDF)
Carbon nanotubes (CNT) and graphene have been one of the noticeable research areas in science and technology. In recent years, the assembly of these carbon nanostructures is one of the most interesting topic to the scientific world due to its variety of applications from nano to macroscale. These bulk nanostructures to be applicable in shock absorbers, batteries, sensors, photodetectors, actuators, solar cells, fuel cells etc.
The present work is motivated to study the detailed compressive behavior of three dimensional cellular assemblies of CNT and graphene. The CNT foams are synthesized by chemical vapor deposition method. It is interesting to study the compressive behavior of CNT foam in the presence external magnetic field applied perpendicular to CNT axis. The peak stress and energy absorption capability of CNT foam enhances by four and nearly two times in the presence of magnetic field as compared to the absence of the magnetic field. In the absence of magnetic field the deformation of CNT foam is obtained elastic, plateau and densification regions. Further CNT foam is loaded with iron oxide nanoparticles of diameter is ~ 40nm on the surface and detailed study of the compressive behavior of the foam by varying iron nanoparticles concentration. The peak stress and energy absorption capability of CNT foam initially decreases with increasing the intensity of the magnetic field, further increases the intensity of the magnetic field the maximum stress and energy absorption capability increases which is due to magnetic CNT and particles align in the direction of the magnetic field.
CNT surfaces were further modified by fluid of different viscosities. The mechanical behavior of CNT foam filled with fluids of varying viscosities like 100%, 95% and 90% glycerol and silicone oil are 612, 237, 109 and 279 mPa-s respectively. The mechanical behavior of CNT foam depends on both the intensity of magnetic field and fluid viscosity. The non linear relation between peak stress of CNT and magnetic field intensity is σp(B, η) = σ0 ± α(B-B0) where σ0 is the peak stress at B = B0 , η is the fluid viscosity, parameter α depends on properties of the MR fluid and B0 is an optimum magnetic field for which peak stress is maximum or minimum depending on the fluid viscosity.
Graphene is assembled into a three dimensional structure called graphene foam. The graphene foam is infiltrated with polymer and study the detailed compressive behavior of graphene foam and graphene foam/PDMS at different strains of 20, 40, 60 and 70%. The maximum stress and energy absorption capability of graphene foam/PDMS is six times higher than the graphene foam. Also the graphene foam/PDMS is highly stable and reversible for 100 cycles at strains of 30 and 50%. The mechanical behavior of CNT, graphene foam, CNT/PDMS and graphene foam/PDMS is compared. Among all the foams, graphene foam/PDMS has shown the highest elastic modulus as compared to other foams. This behavior can be attributed to the wrinkles formation during the growth of graphene and a coupling between PDMS and interfacial interactions of graphene foam. Therefore it suggests potential applications for dampers, cushions and electronic packaging.
Furthermore, the interaction between nanoparticles and polymer in a novel architecture composed of PDMS and iron oxide nanoparticles is studied. The load bearing capacity of uniform composites enhanced by addition of nanoparticles, reaching to a maximum to 1.5 times of the PDMS upon addition of 5wt.% of nanoparticles, and then gradually decreased to 1/6th of PDMS upon addition of 20wt.% of nanoparticles. On the other hand, the load bearing capacity of architectured composites at high strains (≥40%) monotonically increased with addition of nanoparticles in the pillars.
|
29 |
Etude du procédé de CVD en lit fluidisé en vue de revêtir des particules denses pour applications nucléaires / Study of the fluidized bed chemical vapor deposition process on very dense powder for nuclear applicationsVanni, Florence 21 September 2015 (has links)
Cette thèse s’inscrit dans le cadre du développement d’un combustible nucléaire faiblement enrichi pour les réacteurs de recherche, constitué de particules d’uranium-molybdène mélangées à une matrice d’aluminium. Dans certaines conditions sous irradiations, les particules d’U(Mo) interagissent avec la matrice d’aluminium, provoquant un gonflement rédhibitoire de la plaque combustible. Pour inhiber ce phénomène, une solution consiste à déposer, à la surface des particules d’U(Mo), une fine couche de silicium, pour créer un effet barrière. Cette thèse a concerné l’étude du procédé de dépôt chimique à partir d’une phase vapeur (CVD) en lit fluidisé à partir de silane pour déposer le silicium sur la poudre d’U(Mo), qui a une densité exceptionnelle de 17,5. Pour atteindre cet objectif, deux axes d’études ont été traités au cours de la thèse : l’étude et l’optimisation de la mise en fluidisation d’une poudre aussi dense, puis celles du procédé de dépôt de silicium. Pour le premier axe, une campagne d’essais a été réalisée sur poudre simulante de tungstène dans différentes colonnes de fluidisation en verre et en acier avec des diamètres internes compris entre 2 et 5 cm, à température ambiante et à haute température (650°C), proche de celle des dépôts. Cette campagne a permis d’identifier des phénomènes d’effets de bord au sein du lit fluidisé, pouvant conduire à des dépôts hétérogènes ou à des prises en masse. Des dimensions de colonnes de fluidisation et des conditions opératoires permettant une fluidisation satisfaisante de la poudre ont pu être identifiées, ouvrant la voie à l’étude du dépôt de silicium. Plusieurs campagnes d’essais de dépôt sur poudre simulante, puis sur poudre U(Mo), ont ensuite été menées dans le cadre du second axe d’étude. L’influence de la température du lit, de la fraction molaire d’entrée en silane dilué dans l’argon, et du débit total de fluidisation, a été étudiée pour différents diamètres de réacteur et pour diverses masses de poudre. Des analyses de caractérisation morphologique et structurale (MEB, DRX...) ont révélé un dépôt de silicium uniforme sur toute la poudre et autour de chaque grain, majoritairement cristallisé et dont l’épaisseur atteint les objectifs visés. Des recommandations précises ont ainsi pu être émises pour optimiser les caractéristiques du dépôt de silicium sur la poudre combustible U(Mo) par le procédé de CVD en lit fluidisé. / This thesis is part of the development of low-enriched nuclear fuel, for the Materials Test Reactors (MTRs), constituted of uranium-molybdenum particles mixed with an aluminum matrix. Under certain conditions under irradiations, the U(Mo) particles interact with the aluminum matrix, causing unacceptable swelling of the fuel plate. To inhibit this phenomenon, one solution consists in depositing on the surface of the U(Mo) particles, a thin silicon layer to create a barrier effect. This thesis has concerned the study of the fluidized bed chemical vapor deposition (CVD) process to deposit silicon from silane, on the U(Mo) powder, which has an exceptional density of 17,500 kg/m3. To achieve this goal, two axes were treated during the thesis: the study and the optimization of the fluidization of a so dense powder, and then those of the silicon deposition process. For the first axis, a series of tests was performed on a surrogate tungsten powder in different columns made of glass and made of steel with internal diameters ranging from 2 to 5 cm, at room temperature and at high temperature (650°C) close to that of the deposits. These experiments helped to identify wall effects phenomena within the fluidized bed, which can lead to heterogeneous deposits or particles agglomeration. Some dimensions of the fluidization columns and operating conditions allowing a satisfactory fluidization of the powder were identified, paving the way for the study of silicon deposition. Several campaigns of deposition experiments on the surrogate powder and then on the U(Mo) powder were carried out in the second axis of the study. The influence of the bed temperature, the inlet molar fraction of silane diluted in argon, and the total gas flow of fluidization, was examined for different diameters of reactor and for various masses of powder. Morphological and structural characterization analyses (SEM, XRD…) revealed a uniform silicon deposition on all the powder and around each particle, mostly crystallized and whose thickness reached the objectives. Specific recommendations were proposed to optimize the characteristics of the silicon deposit on the U(Mo) powder by the fluidized bed CVD process.
|
30 |
Studies On CVD And ALD Of Thin Films Of Substituted And Composite Metal Oxides, Including Potential High-k DielectricsGairola, Anshita 09 1900 (has links) (PDF)
The work carried out as a part of this thesis has been focussed on understanding different aspects of the chemical vapor deposition process namely, ALD / MOCVD. A large part of the thesis is aimed at solving the problem of a single-source precursor for the MOCVD process to obtain substituted metal oxide thin films. For a chemical vapor deposition technique, it is important to understand the requisite salient features of precursor for deposition of thin films. For this purpose, not only is the structural characterization of the chemical precursor is required but also an in-depth thermal analysis of the precursor to know its vapor pressure. Vapor pressure of a metalorganic complex is one of the important properties to evaluate the applicability of a metalorganic complex as a MOCV/ALD precursor. The thesis discusses a novel approach to use thermal analysis as a tool to gauge the viability of substituted metal “single source” precursor for MOCVD/ALD. The other half deals with material characterization of thin films grown by an ALD process using hydrogen and Ti(OiPr)2(tbob)2 as precursors. The films were further studied for their potential application as high-k dielectric in DRAM applications.
The first chapter is an overview of topics that are relevant to the work carried out in this thesis. The chapter focuses on the description of techniques used for thin film deposition. A detailed review of CVD-type techniques (ALD/ MOCVD) is then given. Chapter1 reviews the various process parameters involved in ALD,i.e. film growth(specifically as a function of the reactant pulse length, the nature of the chemical reactant/precursor and that of the metal precursor, and purge length) and growth temperature. Following the discussion of ALD, CVD and its growth kinetics are also discussed. Chapter 1 then outlines a holistic understanding of precursors, followed the differences in requirement for using them in ALD and MOCVD. Further, an introduction to the titanium oxide (Stoichiometric titanium dioxide and various Magneli phases) system, its phase diagram, oxide properties and their applications is given. Chapter 1 concludes by delineating the scope of the work carried out which is presented in the thesis.
The second chapter deals with the synthesis of a series of substituted metal “single source” precursors to be used for MOCVD of substituted metal oxides thin films. The precursor complexes were of the type AlxCr1-x (acac)3 where 0<x<1. The complexes were synthesized using the novel approach of co-synthesis and were characterized by various spectroscopic techniques. Single crystal X-ray diffraction at low temperature was carried out to understand the substitution of metal in the complex crystallographically.
The substituted metal complexes synthesized and characterized in chapter 2 were further evaluated for their viability as single source precursors for MOCVD application, using thermo-gravimetry as discussed in chapter 3. Vapor pressure of these complexes was determined by using the Langmuir equation, while the enthalpies of submission and evaporation were calculated using the Clausius-Clapeyron equation. One of the composition of the series of substituted metal complexes, viz., Al0.9Cr0.1(acac)3, was employed on MOCVD reactor as precursor to obtain thin films on three substrates, Si(100), fused silica, and polycrystalline x- alumina, simultaneously. The resultant thin films were characterized using XRD, electron microscopy, FTIR, EDS, X-ray mapping, and UV-vis spectroscopy.
Chapter 4 deals with the growth of titanium oxide thin films using ALD. The metal precursor used was Ti(OiPr)2(tbob)2 and the reactant gas was hydrogen. Hydrogen, a reducing gas, was deliberately used to obtain the reduced defect oxide phases of titanium, commonly called Magneli phases. The growth rate of films grown on p-Si(100) was studied with respect to the substrate temperature, vaporizer temperature, pulse duration of metal precursor and pulse duration of the reactive gas. Also, the concept of complementarity of a reaction and self-limiting behavior in a true ALD process was illustrated. The deposition conditions such as substrate temperature and reactive gas flows have been varied to optimize the phase content and the morphology of the films. The films grown were characterized to determine the various phases of titanium oxide present using XRD, TEM, FTIR spectroscopy, Raman spectroscopy, and UV-vis spectroscopy. The presence of carbon was revealed by Raman spectroscopy. By using these characterization techniques, it was concluded that the film grown is a composite made of stiochiometric TiOx matrix embedded with crystallites of (reduced) Magneli phases.
Chapter 5 deals with the electrical properties of the composite thin films grown in chapter 4. the films behave as percolative capacitor which could be used for application as novel high-k dielectric material for DRAM. The effect of change in flow rates of reactive gas (H2) on the dielectric constant (k) and leakage current of the film were studied. It was found that phase composition of the film plays an important role in tuning the dielectric properties of the film was also studied. The effect of thickness of the film also studied on the dielectric properties of the film. The trend observed was correlated to the morphology of the film as a function of its thickness and the grain growth mechanism as observed from high resolution scanning electron microscopy. Further, the effect of change in substrate temperature, metal precursor pulse length, and of the metal used as top electrode, on C-V and I-V characteristics were studied. It was interesting to see that the presence of the more conductingTi5O9 (than Ti3O5) enhances the dielectric constant, which is a requisite for a high-k material for DRAM application. On the other hand, the presence of Ti5O9 also increased the leakage current in the film, which was not desirable.
It therefore suggested itself that an optimum embedment of Ti5O9 in the composite helps in enhancing the dielectric constant, while maintaining a low leakage current. Under optimum conditions, a dielectric constant of 210 at 1MHz was measured with a leakage current of 17 nA. The effect of the presence of carbon in the film was studied using Raman Spectroscopy, and it was found that a high leakage was associated with films having greater carbon content. In this chapter, electrical properties of composite thin films were also compared with those of stoichiometric titanium dioxide (a known dielectric). Further, a multilayer sandwich structure was proposed, such that it had a 53 mm thick stoichiometric TiO2 layer followed by 336nm thick composite film and again a 53nm thick stoichiometric titanium dioxide layer. The dielectric characteristics of this structure were found to be better than those of either of the other two.viz., stoichiometric titanium dioxide film or the composite thin film of titanium oxide.
|
Page generated in 0.1144 seconds