Graphene, one-atom-thick planar sheet of carbon atoms densely packed in a honeycomb crystal lattice, has grabbed appreciable attention due to its exceptional electronic, mechanical and optical properties. Chemical functionalization schemes are needed to integrate graphene with the different materials required for potential applications. Molecular self-assembly behavior on graphene is a key method to investigate the mechanism of interaction between molecules and graphene and the promising applications related to molecular devices. In this thesis, we report the molecular self-assembly behavior of phenyl-C61-butyric acid methyl ester (PCBM), C60, perylenetetracarboxylic dianhydride (PTCDA) and Gd3N@C80 on flat and rippled graphene 2D material by the experimental methods of scanning tunneling microscope (STM) and atomic force microscope (AFM) and by the theoretical method of density functional theory (DFT). We found that molecules form ordered structures on flat graphene, while they form disordered structure on rippled graphene. For example, PCBM forms bilayer and monolayer structures, C60 and Gd3N@C80 form hexagonal close packed (hcp) structure on flat graphene and PTCDA forms herringbone structure on flat graphene surface. Although C60 and Gd3N@C80 both form hcp structure, C60 forms a highly ordered hcp structure over large areas with little defects and Gd3N@C80 forms hcp structure only over small areas with many defects. These differences of structure that forms on flat graphene is mainly due to the molecule-molecule interactions and the shape of the molecules. We find that the spherical C60 molecules form a quasi-hexagonal close packed (hcp) structure, while the planar PTCDA molecules form a disordered herringbone structure. From DFT calculations, we found that molecules are more effected by the morphology of rippled graphene than the molecule-molecule interaction, while the molecule-molecule interaction plays a main role during the formation process on flat graphene. The results of this study clearly illustrate significant differences in C60 and PTCDA molecular packing on rippled graphene surfaces. / Doctor of Philosophy / As the first physical isolated two-dimensional (2D) material, graphene has attracted exceptional scientific attention. Due to its impressive properties including high carrier density, flexibility and transparency, graphene has numerous potential applications, such as solar cell, sensors and electronics. 2D molecular self-assembly is an area that focuses on organization and interaction between self-assembly behaviors of molecules on surface. Graphene is an excellent substrate for the study of molecular self-assembly behavior, and study of molecular study is very important for graphene due to potential applications of molecules on graphene. In this thesis, we present investigations of the molecular self-assembly of PCBM, C60, PTCDA and Gd3N@C80 on graphene substrate. First, we report the two types of bilayer PCBM configuration on HOPG with a step height of 1.68 nm and 1.23 nm, as well as two types of monolayer PCBM configuration with a step height of 0.7 nm and 0.88 nm, respectively. On graphene, PCBM forms one type of PCBM bilayer with a step height of 1.37 nm and one type of PCBM monolayer with a step height of 0.87 nm. By building and analyzing the models of PCBM bilayers and monolayers, we believe the main differences between two configurations of PCBM bilayer and monolayer is the tilt angle between PCBM and HOPG, which makes type I configuration the higher molecule density and binding energy. Secondly, we report the investigation of self-assembly behaviors of C60 and PTCDA on flat graphene and rippled graphene by experimental scanning tunneling microscope (STM) and theoretical density functional theory (DFT). On flat graphene, C60 forms hexagon close pack (hcp) structure, while PTCDA forms herringbone structure. On rippled graphene, C60 forms quasi-hcp structure while PTCDA forms disordered herringbone structure. By DFT calculation, we study the effect of graphene curvature on spherical C60 and planar PTCDA. Finally, we report a STM study of a monolayer of Gd3N@C80 on graphene substrate. Gd3N@C80 forms hcp structure in a small domain with a step height of 0.88 nm and lattice constant of 1.15 nm. According to our DFT calculation, for the optimal organization of Gd3N@C80 and graphene, the gap between Gd3N@C80 and graphene is 3.3 Å and the binding energy is 0.95 eV. Besides, the distance between Gd3N@C80 and Gd3N@C80 is 3.5 Å and the binding energy is 0.32 eV.

Single-walled carbon nanotubes (SWNTs) exhibit extraordinary properties: mechanical, thermal, optical and, possibly the most interesting, electrical. These all-carbon cylindrical structures can be metallic or semi-conducting depending on their precise structure. They have the potential to allow faster transistor switching speeds and smaller, more closely-packed interconnects in microelectronics. However, such applications are hindered by the difficulties of positioning the correct type of SWNT in a spatially precise location and orientation. In addition, greater understanding of the fundamental limits of SWNTs, such as the limit of current density, is needed for optimum operation in applications. The primary aim of this project was to increase the understanding of current density limitation by using in situ plasmonically enhanced Raman spectroscopy during electrical transport. The use of plasmonic metal nanostructures to enhance the Raman scattering should allow the acquisition of informative spectra from SWNTs away from their intrinsic resonance conditions. To achieve this aim, SWNTs must be integrated with plasmonic metal structures as well as electrical connections. This thesis presents two approaches for the integration of SWNTs with other nanometre-scaled features, in particular plasmonic nanoparticles. Fountain pen nanolithography uses a hollow nanopipette in place of the probe tip in an atomic force microscope (AFM), through which material can be delivered to a spatially precise position on a surface. Aqueous SWNT dispersion was delivered to chemically-functionalised silicon in this way, through pulled quartz pipettes with aperture diameters of 50 nm, 100 nm and 150 nm. The heights, widths and continuity of lines drawn on the surface by the nanopipette depended on the size, setpoint and lateral speed of the tip. A small bias voltage applied between the SWNT dispersion inside the pipette and the substrate allowed the deposition to be switched on or off depending on the polarity of the voltage, through the action of electroosmotic effects within the quartz capillary. The quality and density of the SWNT dispersion was found to be important for successful deposition to occur, since too low a concentration results in the lines deposited from the pipette being only surfactant but too high a concentration of bundles would quickly block the small tip of the pipette. Polarised Raman spectroscopy on SWNT deposited by fountain pen nanolithography showed that they had a high level of alignment parallel to the direction in which the pipette moved. Spherical gold nanoparticles with plasmonic properties suitable for enhancing Raman scattering were dropped onto samples containing individual SWNTs supported on a Si/SiO2 surface. Nanomanipulation with an atomic force microscope was used to push the gold nanoparticles onto the SWNTs. Raman spectra measured with and without the gold particles showed that the gold nanoparticles gave local enhancement factors of 24 for a single 150 nm nanoshell and 130 for a small cluster of 150 nm nanoshells. Polarised Raman studies on the cluster showed that the angle dependence deviated significantly from that expected of a bare SWNT. Electrical transport experiments with in situ plasmonically enhanced Raman spectroscopy may be performed on samples prepared from the methods described here. Such experiments would increase understanding of the electrical properties of SWNTs and how they relate to the vibrational and optical properties.

This thesis discusses the design and construction of a Near-field Scanning Optical Microscope (NSOM). Basic principles of operation, the characteristics of the hardware components, and the control software are discussed. A unique method of controlling the position of the probe is developed, and scans of a diffraction grating are presented. We show the influence that the surface topology and reflectivity and the interference of direct and reflected light have on the images. A second design of the instrument, for use in a vacuum chamber and with a flexure stage for lateral motion, is accomplished. / text

The purpose of this study was to evaluate the effect of a final rinse of ethanol on depth of sealer penetration in teeth obturated with Gutta Percha (GP)/AH-Plus® (Dentsply, De Trey GmbH, Konstanz, Germany) or Resilon/Epiphany® SE™ (Pentron Clinical Technologies, LLC, Wallingford, CT). Extracted human anterior teeth (n= 32) were shaped to size 30, 0.06 taper using nickel-titanium rotary files and subjected to an identical irrigation protocol. Specimens were randomly divided into eight groups according to final irrigating solution (saline, 70%, 95%, or 100% ethanol) and obturation material (GP/AH-Plus® or Resilon/EpiphanyÒ SE™). A 2mm thick slice was obtained by sectioning each obturated root at 3mm and 5mm from the anatomic apex. Specimens were cleared and assessed using scanning electron microscopy (SEM). Sealer penetration was observed at different magnifications when using GP/AH Plus® across all final rinse concentrations. Among Resilon/Epiphany® SE™ groups, no sealer penetration was evident under SEM. Conclusions: 1) GP/AH-Plus® showed evidence of sealer penetration, however, Resilon/Epiphany® SE™ did not show evidence of sealer penetration at both the dentin and sealer interface. 2) A final rinse with any concentration of ethanol prior to obturation does not improve sealer penetration with GP/AH-Plus® groups. 3). Resilon/Epiphany® SE™ bond can be dislodged at either the interface of sealer and Resilon or dentin and Resilon.

We examine the design of a high speed atomic force microscope using an optical pickup from a commercially available compact disc/digital versatile disc drive. An investigation of the commercial optical pickup is done with the goal of determining how it can be used for dimensional measurements on nanometer scale. An evaluation of noise sources, imaging capabilities, and functionality is performed.

Working with nanoscale optics requires methods and equipment designed for the purpose. This thesis describes the development of techniques and a system for performing highly localised spectroscopy. The system consists of a nanonics multiview 2000 scanning near-field optical microscope, a grating spectrometer and a photonic crystal fibre supercontinuum light source. Discussion of the microscope includes its modes of operation and development of software to collect and analyse data. In order to demonstrate the setup, an example of localised spectroscopy is presented in the form of an investigation of hollow core photonic crystal fibre. Taking spectra of the components of the cladding of these fibres makes it possible to investigate the origins of bandgap guidance. A core focus of nanoscale optics is the interaction of light with metal structures. This field is called plasmonics. Fabrication of structures is presented and requires special facilities and processes. These processes are both time consuming and expensive, both factors that emphasise the need for prior modelling. Forward difference time domain modelling of a proposed structure comprising of a concentrically arranged ring and disk is explored using home written code and a commercial package called CST Microwave Studio. The investigation of this concentric design through modelling shows a very highly localised field enhancement which can be engineered to have a narrow spectral resonance in the near infrared. The interaction of the two components which govern this resonance is explained using a theory called plasmon hybridization. Once the optical behaviour of small metal objects is understood they can be used in other ways. An example of this is shown in Porous Silicon. As a material it provides an excellent template for formation of metal nano-particles. Embedded in a high surface area network of silicon these particles can be used as very effcient catalysts.

Time-lapse fluorescence live cell imaging has been widely used to study various dynamic processes in cell biology. As the initial step of image analysis, it is important to localize and segment cell edges with higher accuracy. However, fluorescence live-cell images usually have issues such as low contrast, noises, uneven illumination in comparison to immunofluorescence images. Deep convolutional neural networks, which learn features directly from training images, have successfully been applied in natural image analysis problems. However, the limited amount of training samples prevents their routine application in fluorescence live-cell image analysis. In this thesis, by exploiting the temporal coherence in time-lapse movies together with VGG-16 [1] pre-trained model, we demonstrate that we can train a deep neural network using a limited number of image frames to segment the entire time-lapse movies. We propose a novel framework, vU-net, which combines the advantages of VGG-16 [1] in feature extraction and U-net [2] in feature reconstruction. Moreover, we design an auxiliary convolutional block at the end of the architecture to enhance edge detection. We evaluate our framework using dice coefficient and the distance between the predicted edge and the ground truth on high-resolution image datasets of an adhesion marker, paxillin, acquired by a Total Internal Reflection Fluorescence (TIRF) microscope. Our results demonstrate that, on difficult datasets: (i) The testing dice coefficient of vU-net is 3.2% higher than U-net with the same amount of training images. (ii) vU-net can achieve the best prediction results of U-net with one third of training images needed by U-net. (iii) vU-net produces more robust prediction than U-net. Therefore, vU-net can be more practically applied to challenging live cell movies than U-net since it requires a small size of training sets and achieved accurate segmentation.

Orientador: José Manoel Balthazar / Banca: Átila Madureira Bueno / Banca: Bento Rodrigues de pontes Junior / Resumo: Desde 1986, quando Binnig et al (1986) criaram o microscópio de força atômica (AFM), esse aparelho se tornou um dos mais importantes microscópios de varredura (SPM), sendo usado para análise de DNA, nanotubos, etc. (Rützel et al, 2006). O AFM tem como componente principal uma microviga, com uma ponteira em uma das extremidades, que vibra próximo de sua frequencia de ressonância para mandar sinais a um fotodetector que traduz esse sinal e gera as imagens da superfície da amostra. O modo de operação tapping é o mais usado, e o comportamento caótico é muito comum nesse modo de operação, por esse motivo, AFM se tornou um assunto muito importante no mundo científico. Nesse trabalho, a microviga é modelada com o uso das equações de Bernoulli, as interações entre ela e a amostra são modeladas usando o potencial de Lennard Jones. Simulações numéricas detectam movimento caótico no sistema, a necessidade de estabilizá-lo nos leva a usar os seguintes métodos: Método do Balanço Harmônico, sincronização de Sistemas Não Lineares, Método das Equações de Estado Dependentes de Riccati (SDRE), Método de Realimentação de Sinal Atrasado. Por fim, a aplicação dos métodos se mostra eficiente, com pequeno erro e fácil implementação / Abstract: Since 1986, when Binnig et al (1986) created the atomic force microscope (AFM), this unit became one of the most important scanning probe microscopes (SPM) being used for DNA analysis, nano tubes, etc. (Rutzel et al, 2006). The AFM has as a main component, a micro cantilever, with a tip at its free end, which vibrates near its resonance frequency to send signals to a photo detector that translates the signal and generates images of the sample surface. The tapping mod of operation is the most widely used and chaotic behavior is very common in this mode, therefore, AFM has become a very interesting subject in the scientific world. In this work, the micro cantilever is modeled using Bernoulli's equation and the interactions between the tip and the sample are modeled using the Lennard Jones potential. Numerical simulations detect chaotic motion in the system and the need to stabilize it leads us to use the following methods, Harmonic Balance Method; Synchronization of Nonlinear Systems; the State Dependent Riccati Equation control method (SDRE); the Method of Feedback Delay. Finally, the application of the methods proved to be effective, with small error and easy implementation / Mestre

We apply the techniques of digital holography to obtain microscopic 3D images of biological cells. Both the amplitude and phase images are obtained from a single hologram, with approximately 1 [mu]m lateral and 10 nm longitudinal resolution. The results are combined with previous experiments resolving the 2 [pi] ambiguity to produce continuous phase images of the samples.

On sait que les sols et les populations bactériennes indigènes ont une influence sur la mobilité des métaux, donc sur leur toxicité. Cette étude a été menée sur des sols uranifères et contrôles collectés dans le Limousin (régions naturellement riches en uranium ). une analyse physico-chimique et minéralogique des échantillons de sol a été réalisée. La structure des communautés bactériennes a été étudiée par électrophorèse en gradient de dénaturant (DGGE). La structure des communautés est remarquablement stable dans les sols uranifères, ce qui indique que l'uranium exerce une forte pression de sélection. D'autre part, une collection de bactéries cultivables à été réalisée à partir des sols, puis criblée pour la résistance à l'uranium, dans le but d'étudier les interactions entre bactéries et uranium. Des observations en Microscopie Électronique à Balayage ont mis en évidence différents mécanismes de chélation de l'uranium à la surface cellulaire / It is well known that soils play a key role in controlling the mobility of toxic metals and this property is greatly influenced by indigeous bacterial communities. This study has been conducted on radioactive and controls soils, collected in natural uraniferous areas (Limousin). A physico-chemical and mineralogical analysis of soils samples was carried out.The structure of bacterial communities was etimated by Denaturing Gradient Gel Electrophoresis (DGGE). The community structure is remarkably more stable in the uranium-rich soils than in the control ones, indicating that uranium exerts a high selection from the soils was constructed and screened for uranium resistance in order to study basteria-uranium interactions. Scanning electron microscopy revealed that a phylogenetically diverse set of uranium-resistant species ware able to chelate uranium at the cell surface.