Expressing synchronization using traditional lock based primitives has been found to be both error-prone and restrictive. Hence there has been considerable research work to develop scalable and programmer-friendly alternatives to lock-based synchronization. Atomic sections have been proposed as a programming idiom for expressing synchronization at a higher-level of abstraction than locks. One way of supporting atomic sections in software is to rely on an underlying Software Transactional Memory (STM) implementation. While STM offers the promise of being a programming paradigm which is less error-prone and more programmer friendly compared to traditional lock-based synchronization, it also needs to be competitive in performance in order for it to be adopted in mainstream software. Unfortunately STMs do not meet the performance goals and are known to incur excessive performance overheads. Prior work by other researchers and our performance analysis of STM applications show that conflicts and the resulting aborts are a major performance bottleneck for STM applications. Second we find that, supporting fine-grained optimistic concurrency can have significant impact on the cache behavior of applications running on STM and hence can adversely affect STM performance. Our systematic quantitative analysis of the cache behavior of STM applications as well as prior work on qualitative analysis of STM overheads show that cache overheads constitute a major performance bottleneck for STM applications. Hence in this thesis, we focus on addressing these two major STM performance bottlenecks. Current STM implementations are typically application unaware in that they do not analyze the application and use that knowledge to improve the application performance on STM. Closer integration of transactions with the programming languages opens up the possibility of using the compiler to analyze STM applications and using that knowledge to perform application code transformations to improve the application performance on STM automatically and in a manner transparent to the programmer. This motivated us to address the two major STM performance bottlenecks namely poor cache performance and performance penalty due to aborts, by compiler transformations. In order to pinpoint the cache bottlenecks of STM, we perform a detailed experimental evaluation of the cache behavior of STM applications and quantify the impact of the different STM factors on the cache misses experienced by the applications. We propose a set of compiler transformations targeted to address the cache performance bottlenecks identified by our analysis. Next we turn our attention to compiler analysis and transformations targeted at reducing the performance overheads due to transactional aborts, effectively utilizing the compiler’s knowledge of the data access patterns of the application. Since not all applications are designed with optimistic concurrency in mind, real world applications typically contain certain atomic sections which are not amenable to STM’s optimistic concurrency control and hence suffer from excessive transactional abort overheads. We propose two compiler techniques for handling such atomic sections. Another major cause of transactional conflicts leading to unnecessary aborts is the uniform granularity access tracking scheme employed by STM implementations. Using a single uniform access tracking granularity leads to poor lock assignment by STM. We propose techniques which use compiler’s knowledge of an application to improve the application unaware lock assignment made by the STM. Last as transactional abort overheads impact STM performance adversely, we propose a compiler-based approach to reduce the transactional abort overheads by reconciling certain kinds of transactions instead of aborting them and then performing a complete re-execution. We show that our combined set of compiler transformations are effective in improving the performance of a set of STAMP benchmarks by reducing the execution time by 7.48% to 54.82%, aborts by 8.98% to 56.61% and the average D-cache miss latency by up to 33.51%.

Since its discovery, the one atom thick material graphene has been at the centre of growing interest in two-dimensional materials. Due to its exceptional properties, graphene is a rich topic to explore by physicists, chemists, engineers and materials scientists. In addition to its use in the fundamental research, graphene is also a promising candidate for future electronics, photonics and energy storage devices.The project presented in this thesis was carried out to explore the structure of suspended graphene in particular in order to probe the metal-graphene interaction via Transmission Electron Microscopy, as most graphene applications require interfacing with metals. As the work was based on free standing graphene, graphene layers obtained by mechanical cleavage or growth on a substrate were transferred onto TEM-grids. Therefore, fabrication, suspended sample preparation and identification of graphene layers were first discussed for a better understanding of how to obtain high quality graphene, as this was essential for the rest of the project.Structural, topographic and chemical analysis of pristine suspended graphene layers were investigated in detail via Transmission Electron Microscopy and Scanning Tunnelling Microscopy. The latter technique was also employed for graphene on a substrate along with establishing annealing conditions for residue free graphene.Metal deposited suspended graphene layers were then investigated in the electron microscopes. Different metal behaviours were observed on the graphene surfaces for the same amount of metal evaporation. Generally, metals interact only weakly with graphene as they are not observed on clean (residue free) parts and are mainly clustered. On the other hand, graphene etching has been observed in the presence of metals. The etching was initiated with graphene vacancy formation as a result of the interaction between metal and carbon atoms on clean graphene. Once a vacancy was created, a hole quickly formed and eventually the graphene layers were destroyed. However, those holes created by metals were healed spontaneously either by non-hexagonal or perfect hexagonal rings. The possible etching and healing mechanisms of the suspended graphene were also discussed.

Organic semiconductor interfaces are promising materials for use in next-generation electronic and optoelectronic devices. Current models for metal-organic interfacial electronic structure and dynamics are inadequate for strongly hybridized systems. This work aims to address this issue by identifying the factors most important for understanding chemisorbed interfaces with an eye towards tuning the interfacial properties. Here, I present the results of my research on chemisorbed interfaces formed between thin-films of phthalocyanine molecules grown on monocrystalline Cu(110). Using atomically-resolved nanoscale imaging in combination with surface-sensitive photoemission techniques, I show that single-molecule level interactions control the structural and electronic properties of the interface. I then demonstrate that surface modifications aimed at controlling interfacial interactions are an effective way to tailor the physical and electronic structure of the interface. This dissertation details a systematic investigation of the effect of molecular and surface functionalization on interfacial interactions. To understand the role of molecular structure, two types of phthalocyanine (Pc) molecules are studied: non-planar, dipolar molecules (TiOPc), and planar, non-polar molecules (H2Pc and CuPc). Multiple adsorption configurations for TiOPc lead to configuration-dependent self-assembly, Kondo screening, and electronic energy-level alignment. To understand the role of surface structure, the Cu(110) surface is textured and passivated by oxygen chemisorption prior to molecular deposition, which gives control over thin-film growth and interfacial electronic structure in H2Pc and CuPc films. Overall, the work presented here demonstrates a method for understanding interfacial electronic structure of strongly hybridized interfaces, an important first step towards developing more robust models for metal-organic interfaces, and reliable, predictive tuning of interfacial properties.

The diploma thesis addresses the design and experimental verification of cooling system and low temperature part of UHV - STM working in temperature range of 20K - 300K. Due to the demand of variable temperature, the flow cooling system with cryogenic (~5 K) helium (He) is used. Two variants of the low temperature part of the microscope are studied. First the version with cooling only the sample holder, and second with cooling of the whole STM. Designed cooling system consists of He flow cryostat allowing to connect it to the Dewar vessel with liquid helium (LHe) using a low-loss transfer line. The cryostat consists of He inlet and outlet, heat exchangers and copper strains (i.e. braids) for the thermal connection of the sample holder/STM and radiation shield around the STM with the heat exchangers. The thesis describes the design of the flow cryostat and its initial tests in the designed vacuum chamber. Heat flow through a spot contact is also discussed to estimate thermal conductance of insulation supports based on thermal resistance of spherical contacts. The thesis was elaborated in collaboration with the Institute of Scientific Instruments of the ASCR, v.v.i.

In this work I briefly described the basic ideas of density functional theory (DFT) for calculations of an electronic structure of molecules, solids and surfaces. I also summarized the fundamentals of DFT based Fireball code that was used for calculations of the atomic and electronic structures of several models. Fur- ther I described theory of scanning tunnelling microscopy (STM) and mentioned some approaches of simulating STM maps by means of results of DFT calcula- tions. The studied models were reconstructions of a Si (111) surface, namely the 7×7, 2×1 Pandey chain and reconstructions with periodicity √ 3 × √ 3, where finding proper atomic structure, fitting to a new experimental observations, was required. I compared energetic favourableness of the reconstructions. I also stud- ied an adsorption of benzene on 7×7. I have analysed the atomic and electronic structure of all the models and made STM simulations using STM code. I com- pared the results with experimental STM maps in literature and with results of the STM experiments made by RNDr. Pavel Kocán, Ph.D. (reconstruction v √ 3 × √ 3) and by Prof. Alastair McLean (benzene on 7×7). Probable model of observed metastable reconstruction √ 3 × √ 3 was found. The proof that benzene chamisorbate in so called di-σ-bridge position was also made. 1

This thesis is focused on adsorption of phthalocyanines on tin and indium passivated silicon Si(111) surfaces with the √3 × √3 reconstruction at room temperature. Scanning tunneling microscopy was used for obtaining atomically resolved surface images. Molecules on these surfaces predominantly adsorb on Si-substitutional defects. Local density of states (LDOS) of strongly adsorbed molecules was obtained by scanning tunneling spectroscopy. The origin of fuzzy imaging of molecules sitting on Si-substitutional double defects was probed. Voltage dependence of mean lifetime of two observed states, between which the "fuzzy" molecule is switching, was measured by analysis of tunneling current fluctuations. We discussed the influence of external parameters on the switching between the two states. We attribute the fuzzy behaviour of the molecule and resulting tunneling current fluctuations to the motion of the molecule in a double-well potential and propose two most likely kinds of the motion which most closely agree with the obtained data.

El procesamiento de información usado hoy en día, está basado fundamentalmente en la industria de los semiconductores. Los imanes moleculares están siendo estudiados actualmente como una gran alternativa o complemento a la electrónica de semiconductores por sus grandes aplicaciones en el desarrollo de los sistemas electrónicos, informáticos y en el campo de la biomedicina entre otros, debido a su fácil miniaturización y posibilidad de formar puertas lógicas de tamaños inferiores a 10 nanómetros. Los imanes moleculares presentan un ordenamiento magnético a nivel molecular en vez de los sistemas tridimensionales comunes. Estas moléculas pueden presentar ordenamiento magnético permanente o histéresis. El fenómeno del magnetismo se debe a estados fundamentales de alto spin y a la anisotropía magnética de estos compuestos. En nuestro caso, vamos a estudiar diversas moléculas consideradas imanes moleculares como son los Polioxometalatos (POMs), ftalocianinas de doble capa [1], anillos de Cr7Ni[2] y así como las distintas moléculas que puedan ser sintetizadas durante el tiempo que dure este proyecto y presenten características de imán molecular. Cabe destacar la importancia que adquieren los POMs, debido a la estrecha colaboración existente con el grupo de Eugenio Coronado de la Universidad de Valencia, tomando importancia las moléculas sintetizadas en el Instituto de Ciencia Molecular ICMOL. Además, también existen colaboraciones con químicos españoles como son Jaume Veciana y Tomás Torres. Los POMs son óxidos metálicos formados por condensación de compuestos de coordinación, que forman estructuras estables perfectamente de- --finidas. Los POMs adquieren propiedades de imanes moleculares, cuando el átomo central que se encuentra en el centro de la estructura es un elemento de transición, más concretamente un lantánido como son el Disprosio y el Erbio [3]. Los POMs formados con estos metales presentan un bloqueo de su momento magnético a temperaturas cercanas a los 5K, por lo que son unas moléculas candidatas para el desarrollo de la computación cuántica. Por otra parte, en este momento existe un gran interés en el desarrollo de dispositivos electrónicos basados en el carbono, más concretamente en las láminas de grafeno, debido a sus posibilidades de sustituir al silicio en algunos elementos de los dispositivos electrónicos. El grafeno es una estructura laminar plana, de un átomo de grosor, compuesta por átomos de carbono densamente empaquetados en una red cristalina en forma de panal de abeja mediante enlaces covalentes que se formarían a partir de la superposición de los híbridos sp² de los carbonos enlazados. El grafeno presenta una gran cantidad de nuevas propiedades electrónicas y magnéticas [4][5][6], que están siendo estudiadas actualmente, y que han cobrado gran relevancia en campo del desarrollo de la electrónica actual. Por esto, estamos muy interesados en el estudio de propiedades magnéticas y electrónicas en superficies de grafito-grafeno, modificadas mecánicamente, para sus posibles aplicaciones en sistemas electrónicos a escala nanométrica Presentamos un breve resumen de cada uno de los capítulos que componen este manuscrito. El capítulo 2 explica todas aquellas técnicas y conceptos básicos que hemos tenido que usar, aprender y fabricar para el desarrollo de esta Tesis doctoral, como son la comprensión de las barreras túnel y sus distintos mecanismos y modelos para su entendimiento teórico y no solo experimental. Además describimos como hemos fabricado el microscopio de efecto túnel con el que se han desarrollado la mayoría del trabajo aquí presentado. También describimos las distintas técnicas espectrométricas y electroquímicas con las cuales se han analizado los datos obtenidos con el STM. Además explicamos las técnicas criogénicas y de alto vacío aplicadas al set up experimental durante el desarrollo de las medidas. Con la conclusión de este trabajo, habremos demostrado la gran versatilidad y capacidad que tiene el STM. No solo como microscopio para obtener imágenes con resolución subatómica sino como su capacidad como espectrómetro y manipulación a escala nanométrica. Los avances experimentales en el campo de la nanoelectrónica molecular, obtenidos a la conclusión de este manuscrito, ayudaran en un futuro al desarrollo de nuevos dispositivos electrónicos más potentes que los actualmente conocidos, haciendo posible el cumplimiento de la Ley de Moore previamente mencionada al inicio de esta introducción.

Hemos desarrollado una técnica de medida a 4 puntas en corriente alterna con amplificadores “Lock-in” sincronizados para obtener la variación local de la capacidad entre los electrodos de un STM simultáneamente con su conductancia. La resolución a la que llegamos es de fF. En la evolución de la capacidad frente a la distancia entre electrodos distinguimos tres regiones: capacidad clásica (cuando los electrodos están más alejados), capacidad cuántica (en un régimen intermedio) y fuga de capacidad (cuando se entra en corriente túnel). En el régimen de emisión de campo, en los puntos en los que la energía incidente del electrón coincide con la de los niveles discretos por encima de la barrera túnel, se produce fuga de capacidad. Por otro lado, hemos encontrado evidencias de una emergencia del momento magnético en cadenas atómicas de platino que conectan dos electrodos hechos del mismo metal. El ajuste de las anomalías a voltaje de polarización cero a la función Kondo-Fano muestra un apantallamiento del momento magnético por parte de los electrones de conducción. Por último, hemos estudiado el papel que juegan los electrones f en el transporte electrónico en nanocontactos. Para ello, hemos realizado medidas de corriente eléctrica en nanocontactos hechos de gadolinio y europio, respectivamente. Hemos visto que para ambos metales la conductancia del contacto de un átomo está por debajo del cuanto de conductancia. Asimismo, los histogramas de conductancia en Gd son reproducibles lo cual contrasta con el caso de Eu en el que dichos histogramas no muestran tal reproducibilidad. Atribuimos este último hecho al desorden magnético que los electrones f causan en el metal. Con el ajuste de las anomalías a voltaje cero a la función Kondo-Fano hemos visto que los electrones f son apantallados por los de conducción. Además, en el caso de Gd vemos que presenta dos temperaturas Kondo, debidas al apantallamiento de los electrones d y f, respectivamente.

Novel two dimensional nanoscale materials like graphene and metal dichalcogenides (MX2) have attracted the attention of the scientific community, due to their rich physics and wide range of potential applications. It has been shown that novel graphene based transparent conductors and radiofrequency transistors are competitive with the existing technologies. Graphene’s properties are influenced sensitively by adsorbates and substrates. As such not surprisingly, physical properties of graphene are found to have a large variability, which cannot be controlled at the synthesis level, reducing the utility of graphene. As a part of my doctorate dissertation, I have developed atomic hydrogen as a novel technique to count the scatterers responsible for limiting the carrier mobility of graphene field effect transistors on silicon oxide (SiO2) and identified that charged impurities to be the most dominant scatterer. This result enables systematic reduction of the detrimental variability in device performance of graphene. Such sensitivity to substrates also gives an opportunity for engineering device properties of graphene using substrate interaction and atomic scale vacancies. Stacking graphene on hexagonal boron-nitride (h-BN) gives rise to nanoscale periodic potential, which influences its electronic graphene. Using state-of-the-art atomic-resolution scanning probe microscope, I correlated the observed transport properties to the substrate induced extrinsic potentials. Finally in efforts to exploit graphene’s sensitivity to discover new sensor technologies, I have explored noncovalent functionalization of graphene using peptides. Molybdenum disulfide (MoS2) exhibits thickness dependent bandgap. Transistors fabricated from single layer MoS2 have shown a high on/off ratio. It is expected that ad-atom engineering can be used to induce on demand a metal-semiconductor transition in MoS2. In this direction, I have iii explored controlled/reversible fluorination and hydrogenation of monolayer MoS2 to potentially derive a full range of integrated circuit technology. The in-depth characterization of the samples is carried out by Raman/photoluminescence spectroscopy and scanning tunneling microscopy

No description available.