Inside standard and hyperspectral low-dose CT variational imaging: parameter identification and material decomposition

The main contribution of this thesis is the proposal of novel strategies for the selection of parameters arising in variational models employed for the solution of inverse problems with data corrupted by Poisson noise. In light of the importance of using a significantly small dose of X-rays in Computed Tomography (CT), and its need of using advanced techniques to reconstruct the objects due to the high level of noise in the data, we will focus on parameter selection principles especially for low photon-counts, i.e. low dose Computed Tomography. For completeness, since such strategies can be adopted for various scenarios where the noise in the data typically follows a Poisson distribution, we will show their performance for other applications such as photography, astronomical and microscopy imaging. More specifically, in the first part of the thesis we will focus on low dose CT data corrupted only by Poisson noise by extending automatic selection strategies designed for Gaussian noise and improving the few existing ones for Poisson. The new approaches will show to outperform the state-of-the-art competitors especially in the low-counting regime. Moreover, we will propose to extend the best performing strategy to the hard task of multi-parameter selection showing promising results. Finally, in the last part of the thesis, we will introduce the problem of material decomposition for hyperspectral CT, which data encodes information of how different materials in the target attenuate X-rays in different ways according to the specific energy. We will conduct a preliminary comparative study to obtain accurate material decomposition starting from few noisy projection data.

Integrated analysis and design of optimization and up-scaling of inductively coupled plasma synthesis of nanoparticles

This study is focused on radio-frequency inductively coupled thermal plasma (ICP) synthesis of nanoparticles, combining experimental and modelling approaches towards process optimization and industrial scale-up, in the framework of the FP7-NMP SIMBA European project (Scaling-up of ICP technology for continuous production of Metallic nanopowders for Battery Applications). First the state of the art of nanoparticle production through conventional and plasma routes is summarized, then results for the characterization of the plasma source and on the investigation of the nanoparticle synthesis phenomenon, aiming at highlighting fundamental process parameters while adopting a design oriented modelling approach, are presented. In particular, an energy balance of the torch and of the reaction chamber, employing a calorimetric method, is presented, while results for three- and two-dimensional modelling of an ICP system are compared with calorimetric and enthalpy probe measurements to validate the temperature field predicted by the model and used to characterize the ICP system under powder-free conditions. Moreover, results from the modeling of critical phases of ICP synthesis process, such as precursor evaporation, vapour conversion in nanoparticles and nanoparticle growth, are presented, with the aim of providing useful insights both for the design and optimization of the process and on the underlying physical phenomena. Indeed, precursor evaporation, one of the phases holding the highest impact on industrial feasibility of the process, is discussed; by employing models to describe particle trajectories and thermal histories, adapted from the ones originally developed for other plasma technologies or applications, such as DC non-transferred arc torches and powder spherodization, the evaporation of micro-sized Si solid precursor in a laboratory scale ICP system is investigated. Finally, a discussion on the role of thermo-fluid dynamic fields on nano-particle formation is presented, as well as a study on the effect of the reaction chamber geometry on produced nanoparticle characteristics and process yield.

Homoleptic and Heteroleptic Metal Carbonyl Clusters and Nanoclusters

This Thesis aims at presenting the general results achieved during my PhD, that was focused on the study and characterisation of new homoleptic and heteroleptic metal carbonyl clusters. From a dimensional point of view, the nuclearity of such species ranges from 2 to 44 metal atoms. Lower nuclearity compounds may be viewed as polymetallic complexes, whereas higher nuclearity species can reach the nanocluster size, by resembling to ultrasmall nanoparticles (USNPs). Initially, my research was focused on the investigation of small MCCs stabilised by N-Heterocyclic carbene (NHCs) ligands. At this regard, a general strategy for the synthesis of mono-anionic [Fe(CO)4(MNHC)]− and neutral Fe(CO)4(MNHC)2, Co(CO)4(MNHC) (M = Cu, Ag, Au; NHC = IMes, IPr) species has been developed. Furthermore, during this investigation, neutral trimetallic Fe(CO)4(MNHC)(M’NHC) (M, M’ = Cu, Ag, Au; M ≠ M'; NHC = IPr) and neutral heteroleptic Fe(CO)4(MNHC)(MNHC’) (M = Au; NHC = IMes, IPr) compounds have been isolated. Thermal treatment turned out to be an efficient method for the growth of the dimension of MCCs. Indeed, species of the type [M3Fe3(CO)12]3– and [M4Fe4(CO)16]4– (M = Ag, Au) as well as larger clusters were formed during the thermal treatment of the new Fe-M (M = Ag, Cu, Au) carbonyl compounds. These species inspired the investigation of promising reaction paths for the synthesis of Fe-M (M = Ag, Cu, Au) carbonyl compounds devoid of ancillary ligands and alloy MCCs, such as the heterometallic [MxM’5-xFe4(CO)16]3− (M, M' = Cu, Ag, Au; M ≠ M'; x = 0-5) carbonyl clusters. The second part of this Thesis regards high nuclearity MCCs. In particular, new strategies for the growth of platinum carbonyl clusters involving, for instance, the employment of bidentate phosphines are described, as well as the syntheses and the thermal decomposition of new Ni-M (Pd, Pt) carbonyl clusters.

Sviluppo di metodi innovativi per la sintesi di nanoparticelle di interesse industriale

Among various nanoparticles, noble metal nanoparticles have attracted considerable attention due to their optical, catalytic and conducting properties. This work has been focused on the development of an innovative method of synthesis for the preparation of metal nanosuspensions of Au, Ag, Cu, in order to achieve stable sols, showing suitable features to allow an industrial scale up of the processes. The research was developed in collaboration with a company interested in the large scale production of the studied nanosuspensions. In order to develop a commercial process, high solid concentration, long time colloidal stability and particle size control, are required. Two synthesis routes, differing by the used solvents, have been implemented: polyol based and water based synthesis. In order to achieve a process intensification the microwave heating has been applied. As a result, colloidal nanosuspensions with suitable dimensions, good optical properties, very high solid content and good stability, have been synthesized by simple and environmental friendly methods. Particularly, due to some interesting results an optimized synthesis process has been patented. Both water and polyol based synthesis, developed in the presence of a reducing agent and of a chelating polymer, allowed to obtain particle size-control and colloidal stability by tuning the different parameters. Furthermore, it has been verified that microwave device, due to its rapid and homogeneous heating, provides some advantages over conventional method. In order to optimize the final suspensions properties, for each synthesis it has been studied the effect of different parameters (temperature, time, precursors concentrations, etc) and throughout a specific optimization action a right control on nucleation and growth processes has been achieved. The achieved nanoparticles were confirmed by XRD analysis to be the desired metal phases, even at the lowest synthesis temperatures. The particles showed a diameter, measured by STEM and dynamic light scattering technique (DLS), ranging from 10 to 60 nm. Surface plasmon resonance (SPR) was monitored by UV-VIS spectroscopy confirming its dependence by nanoparticles size and shape. Moreover the reaction yield has been assessed by ICP analysis performed on the unreacted metal cations. Finally, thermal conductivity and antibacterial activity characterizations of copper and silver sols respectively are now ongoing in order to check their application as nanofluid in heat transfer processes and as antibacterial agent.

Decomposition and reformulation of integer linear programming problems

This thesis deals with an investigation of Decomposition and Reformulation to solve Integer Linear Programming Problems. This method is often a very successful approach computationally, producing high-quality solutions for well-structured combinatorial optimization problems like vehicle routing, cutting stock, p-median and generalized assignment . However, until now the method has always been tailored to the specific problem under investigation. The principal innovation of this thesis is to develop a new framework able to apply this concept to a generic MIP problem. The new approach is thus capable of auto-decomposition and autoreformulation of the input problem applicable as a resolving black box algorithm and works as a complement and alternative to the normal resolving techniques. The idea of Decomposing and Reformulating (usually called in literature Dantzig and Wolfe Decomposition DWD) is, given a MIP, to convexify one (or more) subset(s) of constraints (slaves) and working on the partially convexified polyhedron(s) obtained. For a given MIP several decompositions can be defined depending from what sets of constraints we want to convexify. In this thesis we mainly reformulate MIPs using two sets of variables: the original variables and the extended variables (representing the exponential extreme points). The master constraints consist of the original constraints not included in any slaves plus the convexity constraint(s) and the linking constraints(ensuring that each original variable can be viewed as linear combination of extreme points of the slaves). The solution procedure consists of iteratively solving the reformulated MIP (master) and checking (pricing) if a variable of reduced costs exists, and in which case adding it to the master and solving it again (columns generation), or otherwise stopping the procedure. The advantage of using DWD is that the reformulated relaxation gives bounds stronger than the original LP relaxation, in addition it can be incorporated in a Branch and bound scheme (Branch and Price) in order to solve the problem to optimality. If the computational time for the pricing problem is reasonable this leads in practice to a stronger speed up in the solution time, specially when the convex hull of the slaves is easy to compute, usually because of its special structure.

Silica-Supported Gold Nanoparticles: Synthesis, Characterization and Reactivity

The main aim of this work was the synthesis and applications of functionalized-silica-supported gold nanoparticles. The silica-anchored functionalities employed, e.g. amine, alkynyl carbamate and sulfide moieties, possess a notable affinity with gold, so that they could be able to capture the gold precursor, to spontaneously reduce it (possibly at room temperature), and to stabilize the resulting gold nanoparticles. These new materials, potentially suitable for heterogeneous catalysis applications, could represent a breakthrough among the “green” synthesis of supported gold nanoparticles, since they would circumvent the addition of extra reducing agent and stabilizers, also allowing concomitant absorption of the active catalyst particles on the support immediately after spontaneous formation of gold nanoparticles. In chapter 4 of this thesis is also presented the work developed during a seven-months Marco Polo fellowship stay at the University of Lille (France), regarding nanoparticles nucleation and growth inside a microfluidic system and the study of the corresponding mechanism by in situ XANES spectroscopy. Finally, studies regarding the reparation and reactivity of gold decorated nanodiamonds are also described. Various methods of characterization have been used, such as ultraviolet-visible spectroscopy (UV-Vis), Transmission Electron Microscopy (TEM), Dynamic Light Scattering (DLS), X-ray Fluorescence (XRF), Field Emission Gun Scanning Electron Microscopy (SEM-FEG), X-ray Photoionization (XPS), X ray Absorption Spectroscopy (XAS).

Monitoring and Prediction of Thermal Emergencies in High Performance Computing Systems

Modern scientific discoveries are driven by an unsatisfiable demand for computational resources. High-Performance Computing (HPC) systems are an aggregation of computing power to deliver considerably higher performance than one typical desktop computer can provide, to solve large problems in science, engineering, or business. An HPC room in the datacenter is a complex controlled environment that hosts thousands of computing nodes that consume electrical power in the range of megawatts, which gets completely transformed into heat. Although a datacenter contains sophisticated cooling systems, our studies indicate quantitative evidence of thermal bottlenecks in real-life production workload, showing the presence of significant spatial and temporal thermal and power heterogeneity. Therefore minor thermal issues/anomalies can potentially start a chain of events that leads to an unbalance between the amount of heat generated by the computing nodes and the heat removed by the cooling system originating thermal hazards. Although thermal anomalies are rare events, anomaly detection/prediction in time is vital to avoid IT and facility equipment damage and outage of the datacenter, with severe societal and business losses. For this reason, automated approaches to detect thermal anomalies in datacenters have considerable potential. This thesis analyzed and characterized the power and thermal characteristics of a Tier0 datacenter (CINECA) during production and under abnormal thermal conditions. Then, a Deep Learning (DL)-powered thermal hazard prediction framework is proposed. The proposed models are validated against real thermal hazard events reported for the studied HPC cluster while in production. This thesis is the first empirical study of thermal anomaly detection and prediction techniques of a real large-scale HPC system to the best of my knowledge. For this thesis, I used a large-scale dataset, monitoring data of tens of thousands of sensors for around 24 months with a data collection rate of around 20 seconds.