Higher Institute on Territorial Systems for Innovation
PORTO Publications Open Repository TOrinoNot a member yet
91193 research outputs found
Sort by
Physical processes and materials in memristive devices: a theoretical study
No full textMemristors are ideal devices able to switch among different resistive states and to retain the most recent one even if the input voltage is removed. With such a characteristic memristors would be able to mimic the brain functions or to behave as universal memory device. The existence of the memristor was theorized by Leon Chua in 1971 but only in 2008 a first prototype device based on the use of an oxide layer was realized, later addressed as a redox-based resistance (switching) random access memory (RRAM). Since its discovery, scientists have been working to understand the mechanism behind the device switching and to eliminate the intrinsic device failures and poor reproducibility, that differentiate the real RRAM from the ideal memristor and that prevent its application in circuitry. There exist two main memristor designs, the valence change memory cell (VCM) and the electrochemical metallization memory cell (ECM). These devices rely both on the switching between a high resistance state (HRS) and a low resistance state (LRS) through the application of a high voltage. Along the same two electric terminals that led to the switching, the resistance state is measured by means of a low voltage. Aim of thesis is to study the yet unclear RRAM physical processes and switching mechanisms in order to pave the way for overcoming existing device weak points. In particular, in the first part we focused on understanding the mechanism of the most basic realization of a VCM memristor, i.e. the thin-film based one. Afterwards, we explained some controversial experimental results on more advanced ECM memristors, namely nanowire-based devices, revising the applicability of currently established conventional theories and proposing an alternative operational mechanism. Finally, we presented the stability and the electrical properties of a new two-dimensional material, i.e. MoS2, which could be applied in the memristive field. In thin-film based VCM devices, it is argued that the resistance switching occurs due to the motion of crystallographic defects under the effect of an electric field. These defects locally dope the insulating oxide increasing its conductivity. Once a filament of stacked defects connects the two electrodes an abrupt change in current is measured externally. Nevertheless, not much it is known about the atomistic nature of the switching, even less about its link with measurable external quantities such as total current. In this thesis work, we proposed a mixed continuum and Kinetic Monte Carlo (KMC) simulation to take into account both atomic-level properties, like defect diffusivity, and macroscopic quantities like internal temperature and overall device current. Thanks to this combination of computational methods, temperature can dynamically change during the course of the simulation. Results show how the height of defect diffusion barrier influences the switching mechanism. When the barriers are low (≈0.4 eV) the defects move as soon as the voltage is applied and correspondingly the device switching is fast, however at room temperature the resistance state is volatile. By contrast, if the barriers are relatively high (≈1.1 eV) the oxide has first to heat up in order for the defects to have enough energy to overcome the barriers. Therefore the temperature, although not generally controlled or explicitly considered in the experiments, plays a fundamental role in the switching. Such a heating stage provides the required state retention for practical applications. ECM devices are asymmetric structures composed by an electrochemically active electrode like Cu, a thin film that functions as an electrolyte, either a transition metal oxide or a germanium chalcogenide, and an inactive electrode, such as Pt. The switching from HRS to LRS is thought to take place when, due to the high electric field established by the external bias, atoms of the electrochemically active contact dissolve into the insulating oxide or chalcogenide thin film to form a conductive filament with the opposite contact. Recently, nanomaterials, particularly micrometer-long nanowires, have been applied as an insulating layer between the metallic electrodes thanks to the improved endurance and LRS/HRS ratio that they ensure. Surprisingly, in these devices the LRS is achieved even if no continuous filament was detected. By means of Density Functional Theory simulations, we proposed a mechanism valid for Cu/ZnO-NW/Pt devices in which atoms belonging to the electrochemically active electrode (Cu), rather than aggregating into a filament, spread on the surface. Particularly when in form of adatom, the adsorbed Cu dope the NW surface creating a conductive channel. Copper adatoms are easily dragged by electric field once the diffusion barrier is overcome thanks to the polarizing effect of the surface. In absence of an external voltage clustering is hindered by the same barrier. As opposed to this, the atoms from the inactive electrode (Pt), although they are as mobile as the Cu atoms on the NW surface, are extremely hard to extract from the contact therefore they do not participate in the switching process. Finally, in collaboration with Prof. Jeffrey Grossman at Massachusetts Institute of Technology, we focused our attention on one of the materials that lately has been applied in numerous fields including resistive switching memories: MoS2. Due to its phase-dependent conductivity MoS2 could be applied to RRAM once the control on its phases is achieved. To this aim we proposed a new technique for the stabilization of the metastable metallic MoS2 T-phase over the stable semiconducting MoS2 H-phase by alloying with another metal dichalcogenide, SnS2, existing in the T-phase only. A combined Cluster Expansion and DFT approach was exploited to theoretically predict the phase diagram of MoxSn1-xS2 compounds. Our results show that the addition of impurities efficiently lowers the energetic cost of the MoS2 T-phase, and that alloying is an effective way to tune the TMD electronic properties. In a RRAM made with a MoxSn1-xS2 sheet the electric field may alter the local distribution of substitutional atoms so to induce a phase change only in a small portion of the material and alter the overall resistance state. Additionally, the reported intra-phase metal-semiconductor transition occurring for a slightly doped material could be useful for memristive applications
MPPT technique based on improved evaluation of photovoltaic parameters for uniformly irradiated photovoltaic array
No full textMULTI-SOURCE 3D MODELS SUPPORTING ULTRASONIC TEST TO INVESTIGATE AN EGYPTIAN SCULPTURE OF THE ARCHAEOLOGICAL MUSEUM IN BOLOGNA
No full textThe paper presents the workflow and the results of an ultrasonic 3D investigation and a 3D survey application aimed at the assessment of the internal integrity of an ancient sculpture. The work aimed at highlighting the ability of methods devoted to the 3D geometry acquisition of small objects when applied to diagnosis performed by geophysical investigation. In particular, two methods widely applied for small objects modelling are considered and compared, the digital Photogrammetry with the Structure from Motion (SFM) technique and hand-held 3D scanners. The study concludes with the aim to enhance the final graphical representation of the tomographic results and to subject the obtained results to a quantitative analysis. The survey is applied to the Egyptian naophorous statue of Amenmes and Reshpu, which dates to the reign of Ramses II (1279-1213 BC) or later and is now preserved in the Civic Archaeological Museum in Bologna. In order to evaluate the internal persistency of fractures and visible damages, a 3D Ultrasonic Tomographic Imaging (UTI) test has been performed and a multi-sensor survey (image and range based) was conducted, in order to evaluate the locations of the source and receiver points as accurate as possible The presented test allowed to evaluate the material characteristics, its porosity and degradation state, which particularly affect the lower part of the statue. More in general, the project demonstrated how solution coming from the field of 3D modelling of Cultural Heritage allow the application of 3D ultrasonic tomography also on objects with complex shapes, in addition to the improved representation of the obtained results
Ceria-based nanomaterials as catalysts for CO oxidation and soot combustion: Effect of Zr-Pr doping and structural properties on the catalytic activity
No full text
Electron-photon Interaction and Thermal Disequilibrium Irreversibility
No full textAtoms continuously interact with the photons of the electromagnetic fields in their environment. This electromagnetic interaction is the consequence of the thermal nonequilibrium. It introduces an element of randomness to atomic and molecular motion, which brings to the decreasing of path probability required for microscopic reversibility of evolution. In any atomic electron-photon interaction an energy footprint is given to the atom, and the emitted photon looses energy. The emission of radiation isn't time reversible and this causes the irreversibility in macroscopic systems
Predicting large scale fine grain energy consumption
No full textToday a large volume of energy-related data have been continuously collected. Extracting actionable knowledge from such data is a multi-step process that opens up a variety of interesting and novel research issues across two domains: energy and computer science. The computer science aim is to provide energy scientists with cutting-edge and scalable engines to effectively support them in their daily research activities. This paper presents SPEC, a scalable and distributed predictor of fine grain energy consumption in buildings. SPEC exploits a data stream methodology analysis over a sliding time window to train a prediction model tailored to each building. The building model is then exploited to predict the upcoming energy consumption at a time instant in the near future. SPEC currently integrates the artificial neural networks technique and the random forest regression algorithm. The SPEC methodology exploits the computational advantages of distributed computing frameworks as the current implementation runs on Spark. As a case study, real data of thermal energy consumption collected in a major city have been exploited to preliminarily assess the SPEC accuracy. The initial results are promising and represent a first step towards predicting fine grain energy consumption over a sliding time window