Dissertations / Theses on the topic 'Matrix interference'

The literature relevant to the generation of volatile hydrides for analytical atomic spectroscopy has been reviewed, with particular reference to atomic absorption spectrometry (AAS). This reveals some conflicting information concerning the nature of various interference effects and strategies to overcome them. The use of flow injection (FI) procedures has been demonstrated by several research groups, to be beneficial. A review of the literature concerning the application of FI techniques to AAS shows that there is a sustained interest in the use of such a combination for analytical purposes. In particular, an interest in the on-line coupling of chemical pretreatment of samples is evident. Atomic absorption spectrometry has a limited working range and requires frequent calibration, consequently, there is a need for a rapid, precise on-line dilution procedure. The potential of FI systems with wide bore manifold tubing for on-line dilution was assessed and found to be limited by variations in dispersion coefficient arising from differences in specific gravities between the sample and carrier fluids. This could be overcome only by the use of unrealistically high flow rates. The use of FI procedures for the generation of volatile hydrides of selenium and arsenic was investigated. Optimization studies of system parameters, including the atomization step, were undertaken which demonstrated the benefits in applying FI in hydride generation atomic absorption spectrometry (HGAAS). Analytical methods were devised and evaluated for the determination of Se in copper metal and As in nickel alloy. These procedures involved the use of an on-line matrix removal step in which potentially interfering matrix elements were retained on a strong cation exchange resin (Dowex 50W). The manifold was designed so that the FI value acted as the interface between the matrix isolation stage and the vapour generation stage, a strategy which allowed independent optimization of each stage. Location of the ion exchange resin in the sample loop of a six-port rotary valve allowed the resin to be regenerated easily and rapidly, with a throughput capability of the order of 50 h⁻¹ and permit the proposed full automation of the whole analytical procedure. In the determination of As in nickel alloy a novel stopped-flow pre-reduction step was developed to permit AsIII quantification, therefore, achieve optimum sensitivity. The two systems permitted limits of detection for Se and As of 2.1 and 3.9 ng ml⁻¹ respectively. Direct comparisons were made with existing matrix isolation systems to emphasise the benefits of system design.

Le déferlement mondial des services de télécommunications, impose, aux réseaux qui les supportent, d’augmenter de plus en plus leurs capacités afin de subvenir aux besoins de ses utilisateurs dont le nombre ne cesse de croître. Le spectre radio, ressource de base pour les communications sans fils, ne suit malheureusement pas cette croissance. Même si des marges réutilisables sont disponibles, leur accès est limité par des politiques strictes de gestion du spectre radio-fréquentiel. Pour remédier à cette situation, les organismes régulateurs des télécommunications se dirigent vers un paradigme de gestion plus flexible, en autorisant de nouvelles méthodes basées sur l’accès dynamique au spectre (DSA - dynamic spectrum access) et les radio cognitives (CR - cognitive radio). Dans ce travail doctoral, est proposée une nouvelle technique pour traiter la problématique de la disponibilité du spectre radio-fréquentiel. Appelée multiplexage fréquentiel par sous-espace de Vandermonde (VFDM -Vandermonde-subspace frequency division multiplexing), elle permet à deux technologies d’accès radio (RATs - radio access technologies) de fonctionner côte-a-côte dans un environnement de petites cellules CR. Ceci se fait en partageant la bande radio tout en protégeant des interférences les systèmes pour lesquels le spectre radio avait été originellement réservé. VFDM transmet les données pré-codées dans le noyau du canal interférant entre l’émetteur opportuniste et le récepteur originel en utilisant la sélectivité en fréquence des canaux et duplexage temporel (TDD - time division duplexing). Le travail de cette thèse propose une approche exhaustive du développement de la technologie VFDM, en allant des bases théoriques jusqu’à la démonstration de faisabilité. Ainsi, les bases théoriques proposées ont été en premier lieu analysées. Puis, en partant de ces bases théoriques, VFDM a été graduellement développé vers une chaîne émetteur-récepteur complète. Des résultats significatifs sont apparus à mi-chemin dans la phase de développement, comme par exemple, l’établissement de stratégies de précodage optimales ou la mise en évidence d‘aspects critiques lors de l’implémentation. Sous certaines conditions, VFDM permet aux réseaux secondaires opportunistes d’être utilisés en même temps que le réseau originel, aux seules contraintes de la connaissance des canaux et de l’accroissement de la complexité du système. Au travers des résultats obtenus en simulation, il a été démontré que des taux de transfert de données significatifs peuvent être atteints, et ce, malgré que les capacités de VFDM soient toujours limitées par la taille du sous-espace de Vandermonde du canal interfèrent primaire-secondaire. Finalement, la chaîne émetteur-récepteur développée démontre la faisabilité de cette méthode<br>The global deployment of PCS is pushing for more and more network capacity to accommodate an exponentially growing user base. Radio spectrum, the basic resource in radio communications, unfortunately does not follow this growth. Even though the current spectrum usage leaves margins for re-use, it is limited by the current fixed spectrum management policies. To remediate this issue, spectrum regulators are switching to a flexible management paradigm, leveraging new DSA schemes based on CR. In this PhD work, a novel CR-DSA technology is introduced to address spectrum scarcity problem. Called VFDM, it allows two RAT to operate side-by-side in a small-cell CR setting, sharing the band while protecting the legacy system from interference. VFDM transmits data pre-coded on the null-space of the interfering cross channel (channel from the opportunistic transmitter to the legacy receiver), assuming frequency selectivity and TDD communications. This PhD work proposes a rather exhaustive approach to the development of VFDM: to go from the theoretical basis up to a proof-of-concept development. Initially the theoretical background, basis of the technique itself was introduced and analyzed. Then, VFDM was gradually developed from a pure mathematical concept up to a full transceiver. During this development, several important mid-way results were developed, such as the multi-user strategy for pre-coding and critical implementation aspects. VFDM, under certain constraints, has been shown to allow a secondary opportunistic network to successfully be installed along with a legacy primary one at merely the cost of channel knowledge and added complexity. By means of numerical examples, it has been shown that significative rates can be attained, even though VFDM's performance is constrained by the size of the Vandermonde-subspace of the interfering channel between the secondary transmitter and primary receiver. Finally, a working VFDM transceiver implementation is shown, providing a proof-of-concept of the technique

<p>The traditional way of increasing capacity in a wirelesscommunication system has been cell splitting and fixedchannel-allocation based on prediction tools. However, theplanning complexity increases rapidly with the number of cellsand the method is not suitable for the large temporal andspatial traffic variations expected in the future. A lot ofresearch has therefore been performed regarding adaptivechannel allocation, where a channel can be used anywhere aslong as the signal-to-interference ratio (SIR) is acceptable. Acommon opinion is that these solutions must be decentralizedsince a centralized one would be overly complex.</p><p>In this thesis, we study the locally centralized<i>bunch concept</i>for radio resource management (RRM) in aManhattan environment and show that it can give a very highcapacity both for outdoor users and for indoor users covered byoutdoor base stations. We show how measurement limitations anderrors affect the performance and wepropose methods to handlethese problems, e.g. averaging of measured values, robustchannel selection algorithms, and increased SIR margins. Wealso study the computational and signaling complexities andshow that they can be reduced by splitting large bunches, usingsparse matrix calculations, and by using a simplified admissionalgorithm. However, a reduction of the complexity often means areduction of the system capacity.</p><p>The measurements needed for RRM can also be used to find amobile terminal's geographical position. We propose and studysome simple yet accurate methods for this purpose. We alsostudy if position information can enhance RRM as is oftensuggested in the literature. In the studied scenario, thisinformation seems to be of limited use. One possible use is toestimate the mobile user's speed, to assist handover decisions.Another use is to find the location of user hotspots in anarea, which is beneficial for system planning.</p><p>Our results show that the bunch concept is a promisingcandidate for radio resource management in future wirelesssystems. We believe that the complexity is manageable and themain price we have to pay for high capacity is frequentreallocation of connections.</p><p><b>Keywords:</b>bunch concept, radio resource management,network-assisted resource management, base station selection,dynamic channel allocation, DCA, channel selection,least-interfered, interference avoidance, interferenceaveraging, handover, power control, path-loss measurements,signal strength, link-gain matrix, TD-CDMA, UTRA TDD, Manhattanscenario, microcells, mobile positioning, position accuracy,trilateration, triangulation, speed estimation</p>

2014 - 2015<br>In questa Tesi di Dottorato, io e il mio supervisor, il Prof. Roberto De Luca, abbiamo analizzato alcuni particolari dispositivi superconduttori, le giunzioni Josephson e gli SQUIDs (Superconducting Quantum Interference Devices), da un punto di vista semi-classico e quantistico. Con la collaborazione di alcuni professori russi, i Professori Larisa Zherikhina e Andrej Tshovrebov del Lebedev Institute of RAS (Russian Academy of Sciences), a Mosca, Russia and il Prof. George Izmailov del Moscow Aviation Institute (MAI), anche a Mosca. Applicazioni di questi dispositivi (in particolare degli SQUIDs) come rivelatori di Materia Oscura sono state considerate. Noi dapprima descriviamo la nostra attività teorica sulle giunzioni Josephson e gli SQUIDs, e poi sottolineiamo il ruolo degli SQUID come rivelatori di Materia Oscura. Iniziamo con l’analisi microscopica di una catena lineare di N superconduttori, per cui abbiamo considerato interazioni solo tra primi vicini. Nel particolare caso di N = 2 superconduttori accoppiati, così che essi formano una giunzione Josephson, abbiamo ottenuto le stesse relazioni caratterizzanti il modello di Feynman, che descrive, da un punto di vista quantistico, questo sistema. I risultati confermano la validità del modello semiclassico di Ohta, che rappresenta l’estensione del modello di Feynman a una giunzione Josephson connessa a una sorgente di f.e.m. Abbiamo poi analizzato le proprietà teoriche delle giunzioni Josephson a doppia barriera e a tripla barriera (DBJJ e TBJJ rispettivamente). Per la DBJJ, che è un sistema superconduttivo a tre elettrodi, in cui l’elettrodo intermedio è considerato come un puro sistema quantistico, abbiamo ipotizzato accoppiamenti non omogenei tra gli elettrodi superconduttivi 1-2 e 2-3. La costante di accoppiamento tra gli elettrodi 1-3 si prende piccola in confronto alle precedenti. Per la TBJJ, un sistema superconduttivo a quattro elettrodi, in cui gli elettrodi più interni sono trattati come un puro sistema quantistico, le costanti di accoppiamento tra i layers 1-2, 2-3 e 3-4 sono diverse, così anche in questo caso non abbiamo accoppiamenti omogenei, e le costanti di accoppiamento tra 1-3 e 2-4 sono considerate minori di quelle precedenti. Per scopo di semplicità, prendiamo la differenza di fase superconduttiva degli elettrodi più interni 2 e 3 uguali a zero. Sotto queste ipotesi, usando il modello semi-classico di Ohta, abbiamo ottenuto la relazione corrente-fase (CPR) per questi sistemi. Si nota che questa relazione è diversa da quella sinusoidale, che caratterizza la semplice giunzione Josephson (SJJ), ed è in buon accordo sia con i risultati teorici ottenuti da Brinkmann, basati su un approccio microscopico, e anche con i risultati sperimentali trovati da Nevirkovets et al. Gli ultimi risultati sono basati sull’osservazione dei passi di Shapiro, e l’analisi delle loro ampiezze in funzione della d.d.p. applicata... [a cura dell'autore]<br>XIV n.s.

Ce travail de thèse consiste à visualiser sur un détecteur sensible en position les oscillations spatiales des électrons lents (~ meV) émis par photoionisation au seuil en présence d’un champ électrique extérieur. La figure d’interférence obtenue représente quantiquement le module carré de la fonction d’onde électronique. Ce travail fondamental nous permet d’avoir accès à la dynamique électronique quelques µm autour de l’atome et donc de mettre en évidence plusieurs mécanismes quantiques (champ coulombien, interaction électron/électron..) se déroulant à l’échelle atomique. Malgré la présence d’un cœur électronique quoique limité dans Li, nous avons réussi, expérimentalement et pour la première fois, à visualiser la fonction d’onde associée aux états Stark quasi-discrets couplés au continuum d’ionisation. En outre, à l’aide des simulations quantiques de propagation du paquet d’ondes, basées sur la méthode de « Split-operator », nous avons réalisé une étude complète sur les atomes H, Li et Cs tout en dévoilant les effets significatifs des résonances Stark. Un très bon accord, sur et hors résonances, a été obtenu entre les résultats simulés et les résultats expérimentaux. Par ailleurs, nous avons développé un modèle analytique généralisable permettant de comprendre profondément le fonctionnement d’un spectromètre de VMI. Ce modèle repose sur l’approximation paraxiale, il est basé sur un calcul d’optique matricielle en faisant une analogie entre la trajectoire électronique et le rayon lumineux. Un excellent accord a été obtenu entre les prédictions du modèle et les résultats expérimentaux<br>This work of thesis aims to visualize, on a position sensitive detector, the spatial oscillations of slow electrons (~meV) emitted by a threshold photoionization in the presence of an external electric field. The interference figure obtained represents the square magnitude of electronic wavefunction. This fundamental work allows us to have access to the electronic dynamics and thus to highlight several quantum mechanisms that occur at the atomic scale (field Coulomb, electron/electron interaction..). Despite the presence an electronic core in Li atom, we have succeeded, experimentally and for the first time, to visualize the wave function associated with the quasi-discrete Stark states coupled to the ionization continuum. Besides, using simulations of wave packet propagation, based on the "Split-operator” method, we have conducted a comprehensive study of the H, Li and Cs atoms while revealing the significant effects of the Stark resonances. A very good agreement, on and off resonances, was obtained between simulated and experimental results. In addition, we have developed a generalized analytical model to understand deeply the function of VMI spectrometer. This model is based on the paraxial approximation; it is based on matrix optics calculation by making an analogy between the electronic trajectory and the light beam. An excellent agreement was obtained between the model predictions and the experimental results

The aim of this work is to investigate the role of optical resonances supported in different layouts of solar cells integrating metallic plasmonic gratings. A simplified ideal model was first developed in order to clarify the role and the potentialities that all the optical resonances involved in such structures have in remodulating light absorption. In a further analysis, a global optimization of both geometrical parameters of the grating (period, thickness, slit width) as well as the dielectric environment has been performed, considering different solar cell layouts, in order to obtain a gain in absorption over the widest bandwidth of solar spectrum. Following the simulation and optimization results, a nanofabrication process for the integration of metal nanostructures on top of large area c-Si solar cells has been designed. A new Laser Interference Lithography system, useful for the fabrication of large area plasmonic structures has been designed, constructed and tested. Several processes have been designed and performed and new materials have been also developed (e.g. Hybrid Organic-Inorganic solgel sinusoidal gratings). System performances are still growing toward the control of the full set of geometric parameters of the structures that can be fabricated. After nanofabrication, performances of solar cells integrating plasmonic crystals have been verified by electro-optic characterizations. Polarization resolved Specular Reflectance measurements of patterned solar cell samples were performed and results validate the model. Figures of merit of devices integrating plasmonic crystals, current density (JV) as well as External Quantum Efficiency (EQE), have been also measured. JV characteristics show a significant improvement in conversion efficiency for cells integrating the Ag nanostructures compared to flat reference cells. Such improvement is mainly due to the enhancement in short circuit current due to the light trapping effect provided by the plasmonic crystals. The EQE spectra of cells with gratings show an enhancement in near infra-red response for TM polarization, as expected from optical simulations, and also an unexpected measured improved absorption in the visible. The latter is a further indication of effectiveness of plasmonic nanostructures in light harvesting.<br>Negli ultimi anni, la comunità scientifica ha largamente riconosciuto la possibilità di impiegare i fenomeni legati alle risonanze plasmoniche nei metalli, sfruttandone le particolari caratteristiche per il miglioramento delle performance di dispositivi optoelettronici. Tali soluzioni sono spesso ampiamente vantaggiose, ad esempio nel campo della sensoristica, in termini di miglioramento dell’efficienza nella rivelazione di specie chimiche e biologiche. Un altro campo di crescente attenzione è l’applicazione delle plasmonica al fotovoltaico. In tal caso, la capacità dei reticoli plasmonici di rimodulare lo spettro della radiazione solare incidente consente di migliorare l’efficienza di conversione energetica. In questo contesto si colloca questo lavoro, incentrato nel corso di questi tre anni, nello studio, nella realizzazione e nella caratterizzazione di reticoli plasmonici con l’intento di integrarli in dispositivi più complessi, come le celle fotovoltaiche a silicio cristallino. In primo luogo, si è proceduto sviluppando un modello semplificato di reticolo metallico, posto al di sopra di uno strato semi-infinito di silicio, al fine di chiarire il ruolo e le potenzialità delle risonanze ottiche che si generano quando il dispositivo viene irradiato dall’alto da una sorgente monocromatica polarizzata. Successivamente, è stata condotta una ulteriore analisi di ottimizzazione globale dei parametri geometrici e dei materiali su diverse configurazioni di celle solari, al fine di individuare le caratteristiche tecniche migliori che garantissero un guadagno in assorbimento, integrato sulla più ampia porzione di spettro solare. Le fasi di simulazione e ottimizzazione sono state infine seguite dal design di un processo di nanofabbricazione che permettesse l’integrazione delle nanostrutture metalliche ottimizzate su celle solari a silicio cristallino. Parte del lavoro è stato mirato alla progettazione e realizzazione di un sistema di litografia per la nanofabbricazione di reticoli plasmonici su grandi aree, chiamato Litografia Interferenziale che, dai primi mesi del 2010 è operativo presso l’Istituto IOM – CNR di Trieste. Mediante l’utilizzo di questa tecnica sono stati messi a punto diversi processi di nanofabbricazione, tra i quali quello per la realizzazione di reticoli con profilo sinusoidale e la loro integrazione in un biosensore per la rivelazione di DNA e lo studio delle potenzialità in termini litografici di materiali di tipo solgel ibridi organici\inorganici. Il processo per la realizzazione dei reticoli sulle celle solari è stato messo a punto e i campioni realizzati sono stati analizzati mediante misure morfologiche (SEM, AFM) e ottiche. Infine, le celle solari ottenute sono state caratterizzate mediante le misure delle principali figure di merito, tra cui le caratteristiche di densità di corrente-tensione (JV) e l’efficienza quantica esterna (EQE). Dalle misure JV è stato riscontrato un significativo aumento di efficienza per le celle integranti i reticoli plasmonici, in confronto a quelle convenzionali usate come riferimento, principalmente dovuto all’aumento della densità di corrente di corto circuito. Gli spettri di EQE mostrano un aumento nella regione del vicino infrarosso, per luce incidente polarizzata in modo tale da eccitare anche risonanze plasmoniche, come atteso dai risultati del modello sviluppato. I risultati raggiunti e presentati in questa tesi dimostrano l’effettiva possibilità dell’uso di nanostrutture metalliche per il rimodulazione e altresì il controllo del fenomeno di assorbimento della luce all’interno delle celle fotovoltaiche a silicio cristallino. I risultati soddisfano ampiamente le ipotesi in quanto aprono una nuova frontiera per l’applicazione anche ad altri tipologie di celle quali ad esempio quelle a film sottile o organiche. In tali dispositivi, infatti, la riduzione del materiale assorbente, al fine di massimizzare la raccolta della carica e di ridurre i costi, rende di primaria importanza la ricerca di queste nuove soluzioni per il miglioramento dell’assorbimento della luce. La direzione verso cui si stanno concentrando enormi sforzi di ricerca negli ultimi anni è quella della realizzazione di dispositivi “spessi dal punto di vista della luce incidente e sottili per la raccolta della carica generata”.

L'utilizzo di eterostrutture a semiconduttore nella realizzazione di dispositivi per la quantum information ha come principali vantaggi la scalabilità e la possibilità di essere integrati nei circuiti elettrici tradizionali. In particolare, è stata dimostrata teoricamente la possibilità di realizzare un set universale di porte logiche quantistiche attraverso le architetture che sfruttano flying-qubit. Nei nostri studi, ci concentriamo sulla loro implementazione attraverso interferometri elettronici basati su canali edge nel regime di Hall quantistico, e ne studiamo numericamente la realizzabilità e capacità di calcolo. Gli stati di edge sono canali conduttivi e chirali che corrono lungo il bordo di un 2DEG confinato e caratterizzati da una notevole lunghezza di coerenza, il cui percorso può essere controllato tramite gate metallici che modulano localmente il coefficiente di svuotamento del 2DEG. Oltre all'utilizzo di correnti elettroniche, è attualmente possibile iniettare in tali canali pacchetti elettronici a singola carica. In particolare, ne è stata recentemente proposta la iniezione attraverso quantum dot pumps. Contrariamente ad altri tipi di eccitazione, questi pacchetti sono caratterizzati da un'energia molto più elevata del livello di Fermi e la loro forma gaussiana garantisce un maggiore controllo sulla loro dinamica. Dopo aver illustrato i modelli numerici e le realizzazioni sperimentali dei pacchetti a singolo elettrone maggiormente utilizzati, ovvero i levitoni e le quasi-particelle di Landau, presenteremo il nostro approccio dinamico per la simulazione dei pacchetti di stati di edge. Per simularne l'evoluzione in una geometria 2D arbitraria, abbiamo sviluppato un risolutore parallelo della equazione di Schroedinger basato sul metodo Split-Step Fourier. I risultati ottenuti nel nostro schema dinamico sono confrontati con calcoli di supporto basati su geometrie efficaci 1D e sul metodo della matrice di scattering. Infine, abbiamo descritto il costo computazionale del metodo Split-Step Fourier per un sistema a due particelle in una geometria 2D, che richiede la distribuzione della funzione d'onda a 4 dimensioni sui nodi di macchine HPC nel paradigma MPI. Il metodo Split-Step è inizialmente applicato alla simulazione d'interferenza a singolo elettrone in una geometria scalabile dell'interferometro Mach-Zehnder (MZI) a doppio filling factor. Includendo la forma esatta degli stati di edge generati dallo specifico design del dispositivo, abbiamo progettato un beam splitter elettronico che garantisce una più ampia visibilità rispetto a un tradizionale quantum point contact. Successivamente, abbiamo dimostrato la validità del nostro approccio numerico tramite lo studio dell'antibunching a due elettroni in un esperimento Hong-Ou-Mandel nel regime di Hall quantistico intero. Ivi, abbiamo caratterizzato l'origine della probabilità di bunching non nulla, analizzando l'interazione tra la dispersione energetica dei pacchetti d'onda interferenti e la esatta geometria del partitore elettronico. Grazie alla computazione esatta della funzione d'onda 4D, abbiamo incluso l'interazione Coulombiana tra i nostri pacchetti d'onda fortemente localizzati, e introdotto l'effetto dello screening per osservare la transizione da un interferometro dominato dalla interazione di scambio a uno controllato dalla repulsione intraelettronica. Infine, abbiamo presentato un'implementazione a stato solido di uno sfasatore di fase condizionale basato sugli stati di edge, che si realizza concatenando in parallelo due MZI multicanale. Qui, l'esatta simulazione dello scattering di elettroni sotto l'azione del potenziale Coulombiano in 2D ci consente di analizzare l'interazione tra l'azione selettiva della repulsione elettronica, che funge da entangler, e la geometria su vasta scala del dispositivo.<br>Quantum information processing devices based on semiconductor heterostructures have the potential to be scalable and easier to be integrated in traditional electronic circuitry than systems based on different platforms. In particular, architectures exploiting the flying-qubit paradigm proved theoretically to allow the implementation of a universal set of quantum gates. We focus on coupled electron interferometers based on edge channels in the Integer Quantum Hall regime for the physical implementation of flying qubits and gates, and study numerically the feasibility of such devices and their quantum computing capabilities. Hall edge states are chiral conductive channels running along the border of a confined 2DEG, with a remarkably long coherence length. Their path can be controlled by means of metallic gates that locally deplete the 2DEG in order to engineer electron interferometers. In addition to the use of interfering currents, electron quantum optics is nowadays realized also by injecting single-electron excitations. Specifically, an injection protocol based on non-adiabatic quantum dot pumps has been recently proposed. In contrast to other types of excitations, these wave packets are characterized by an energy much larger than the Fermi level, and their Gaussian shape ensures a robust control of the wave packet dynamics. After reviewing the theoretical modelling and some experimental realizations of the two most common single-electron excitations, i.e. levitons and Landau quasi-particles, we present our numerical approach for the dynamical simulation of wave packets in edge states. To simulate their evolution in a realistic 2D geometry, we developed an in-house a parallel solver of the time-depedent Schroedinger equation based on the Split-Step Fourier algorithm. Our findings in the dynamic framework are compared to simplified analytical models based on effective 1D geometries and the scattering matrix method. We highlight the numerical challenges in the application of the Split-Step Fourier method for a two-particle system in a 2D geometry, which involves data distribution of the ensuing 4D wavefunction on multi-node HPC architectures with the MPI paradigm. Our time-dependent method has been initially applied to simulate self-electron interference in a scalable geometry of the electron Mach-Zehnder interferometer (MZI) at bulk filling factor two. By accounting for the exact shape of the edge states generated by the full-scale design of the interferometer, we engineer an electron beam splitter that ensures a higher visibility with respect to a traditional quantum point contact. Moreover, we prove the validity of our numerical method by studying two-electron antibunching in an Hall-driven Hong-Ou-Mandel interferometer. Here, we characterize the origin of the non-zero bunching probability - observed experimentally - by studying the interplay between the energy broadening of the two interfering wave packets and the exact geometry of the electron beam splitter. Thanks to the computation of the non separable 4D wavefunction, we include exactly Coulomb interaction between our strongly-localized wave packets, and introduce the effect of screening to observe the transition from an exchange-driven to a Coulomb-driven interferometer in the operating regime of interest. Finally, we propose a solid-state implementation of a conditional phase shifter based on edge states, engineered by concatenating in parallel two multichannel MZIs. Here, the exact simulation of Coulomb-driven electron scattering in 2D allows us to analyze the interplay between the selective action of electron repulsion, that acts as the entangler, and the full-scale geometry of the device.

Nel presente elaborato è riportato uno studio sulle proprietà ottiche di meta-materiali nano-strutturati. In particolare, mediante l'uso di uno spettrofotometro, si è osservato il fenomeno di risonanza plasmonica di superficie di nanofili metallici in matrice di allumina. I nanofili esaminati sono stati cresciuti mediante elettrodeposizione all'interno di una matrice di allumina porosa, creata con specifici parametri geometrici attraverso un processo di anodizzazione. Sempre mediante uno spettrofotometro, attraverso cui sono stati acquisiti gli spettri di assorbimento e riflessione dei materiali, è stato possibile osservare la variazione della risonanza plasmonica al cambiare dei parametri del fascio di luce, come l'angolo di incidenza o l'asse di polarizzazione, oppure delle caratteristiche del campione stesso, come l'altezza dei nanofili. È stata osservata anche l'influenza che i nanofili hanno sul fenomeno delle frange di interferenza da lamina sottile. confrontando gli spessori ricavati dagli spettri di riflessione di nanofili a diversa altezza con quello dell'allumina vuota.

Atrial fibrillation is a common heart arrhythmia which is characterized by a missing or irregular contraction of the atria. The disease is a risk factor for other more serious diseases and the total medical costs in society are extensive. Therefore it would be beneficial to improve and optimize the prevention and detection of the disease.   Pulse palpation and heart auscultation can facilitate the detection of atrial fibrillation clinically, but the diagnosis is generally confirmed by an ECG examination. Today there are several algorithms that detect atrial fibrillation by analysing an ECG. A common method is to study the heart rate variability (HRV) and by different types of statistical calculations find episodes of atrial fibrillation which deviates from normal sinus rhythm.   Two algorithms for detection of atrial fibrillation have been evaluated in Matlab. One is based on the coefficient of variation and the other uses a logistic regression model. Training and testing of the algorithms were done with data from the Physionet MIT database. Several steps of signal processing were used to remove different types of noise and artefacts before the data could be used.   When testing the algorithms, the CV algorithm performed with a sensitivity of 91,38%, a specificity of 93,93% and accuracy of 92,92%, and the results of the logistic regression algorithm was a sensitivity of 97,23%, specificity of 93,79% and accuracy of 95,39%. The logistic regression algorithm performed better and was chosen for implementation in Java, where it achieved a sensitivity of 97,31%, specificity of 93,47% and accuracy of 95,25%.<br>Förmaksflimmer är en vanlig hjärtrytmrubbning som kännetecknas av en avsaknad eller oregelbunden kontraktion av förmaken. Sjukdomen är en riskfaktor för andra allvarligare sjukdomar och de totala kostnaderna för samhället är betydande. Det skulle därför vara fördelaktigt att effektivisera och förbättra prevention samt diagnostisering av förmaksflimmer.   Kliniskt diagnostiseras förmaksflimmer med hjälp av till exempel pulspalpation och auskultation av hjärtat, men diagnosen brukar fastställas med en EKG-undersökning. Det finns idag flertalet algoritmer för att detektera arytmin genom att analysera ett EKG. En av de vanligaste metoderna är att undersöka variabiliteten av hjärtrytmen (HRV) och utföra olika sorters statistiska beräkningar som kan upptäcka episoder av förmaksflimmer som avviker från en normal sinusrytm.   I detta projekt har två metoder för att detektera förmaksflimmer utvärderats i Matlab, en baseras på beräkningar av variationskoefficienten och den andra använder sig av logistisk regression. EKG som kommer från databasen Physionet MIT används för att träna och testa modeller av algoritmerna. Innan EKG-signalen kan användas måste den behandlas för att ta bort olika typer av brus och artefakter.   Vid test av algoritmen med variationskoefficienten blev resultatet en sensitivitet på 91,38%, en specificitet på 93,93% och en noggrannhet på 92,92%. För logistisk regression blev sensitiviteten 97,23%, specificiteten 93,79% och noggrannheten 95,39%. Algoritmen med logistisk regression presterade bättre och valdes därför för att implementeras i Java, där uppnåddes en sensitivitet på 91,31%, en specificitet på 93,47% och en noggrannhet på 95,25%.

Le réseau mobile d’Orange France comprend plus de 100 000 antennes 2G, 3G et 4G sur plusieurs bandes de fréquences sans compter les nombreuses femto-cells fournies aux clients pour résoudre les problèmes de couverture. Ces chiffres ne feront que s’accroître pour répondre à la demande sans cesse croissante des clients pour les données mobiles. Cela illustre le défi énorme que rencontrent les opérateurs de téléphonie mobile en général à savoir gérer un réseau aussi complexe tout en limitant les coûts d’opération pour rester compétitifs. Cette thèse s’attache à utiliser le concept SON (réseaux auto-organisants) pour réduire cette complexité en automatisant les tâches répétitives ou complexes. Plus spécifiquement, nous proposons des algorithmes d’optimisation automatique pour des scénarios liés à la densification par les small cells ou les antennes actives. Nous abordons les problèmes classiques d’équilibrage de charge mais avec un lien backhaul à capacité limitée et de coordination d’interférence que ce soit dans le domaine temporel (notamment avec le eICIC) ou le domaine fréquentiel. Nous proposons aussi des algorithmes d’activation optimale de certaines fonctionnalités lorsque cette activation n’est pas toujours bénéfique. Pour la formulation mathématique et la résolution de tous ces algorithmes, nous nous appuyons sur les résultats de l’approximation stochastique et de l’optimisation convexe. Nous proposons aussi une méthodologie systématique pour la coordination de multiples fonctionnalités SON qui seraient exécutées en parallèle. Cette méthodologie est basée sur les jeux concaves et l’optimisation convexe avec comme contraintes des inégalités matricielles linéaires<br>The mobile network of Orange in France comprises more than 100 000 2G, 3G and 4G antennas with severalfrequency bands, not to mention many femto-cells for deep-indoor coverage. These numbers will continue toincrease in order to address the customers’ exponentially increasing need for mobile data. This is an illustrationof the challenge faced by the mobile operators for operating such a complex network with low OperationalExpenditures (OPEX) in order to stay competitive. This thesis is about leveraging the Self-Organizing Network(SON) concept to reduce this complexity by automating repetitive or complex tasks. We specifically proposeautomatic optimization algorithms for scenarios related to network densification using either small cells orActive Antenna Systems (AASs) used for Vertical Sectorization (VeSn), Virtual Sectorization (ViSn) and multilevelbeamforming. Problems such as load balancing with limited-capacity backhaul and interference coordination eitherin time-domain (eICIC) or in frequency-domain are tackled. We also propose optimal activation algorithms forVeSn and ViSn when their activation is not always beneficial. We make use of results from stochastic approximationand convex optimization for the mathematical formulation of the problems and their solutions. We also proposea generic methodology for the coordination of multiple SON algorithms running in parallel using results fromconcave game theory and Linear Matrix Inequality (LMI)-constrained optimization

This thesis explores various economic environments where the structure of social interactions across individuals determines outcomes. In the first chapter, I study mutual insurance arrangements restricted on a social network. I test the network-based sharing rules on data from Bolivian communities, and I argue that this framework provides a reinterpretation of the standard risk sharing results, predicting household heterogeneity in response to income shocks. In the second paper, I study individual and collective behavior in coordination games where information is dispersed through a network. I show how changes in the distribution of connectivities in the population affect the types of coordination in equilibrium as well as the probability of success. In the third chapter, I explore a framework of learning and turnover in the labor market. I show that positive assortative matching (PAM) extends beyond the stable environment of Eeckhout & Weng (2010) to a situation of residual uncertainty that exhibits periods of unlearning. I also extend this setting to allow for career concerns and I show that PAM can only be sustained under strong assumptions.<br>Esta tesis explora diversos entornos económicos en los que la estructura de las interacciones sociales entre los individuos determina los distintos resultados. En el primer capítulo, se estudia acuerdos de seguro mutuo restringidos en una red social. Utilizo datos de comunidades bolivianas para medir las predicciones teóricas y encuentro que los intercambios observados entre los hogares coinciden con la regla de reparto basada en la red obtenida por la teoría. Sostengo que este marco ofrece una reinterpretación de los resultados estándar de distribución de riesgos, prediciendo heterogeneidad entre los hogares en respuesta a los shocks de ingresos. En el segundo artículo, estudio el comportamiento individual y colectivo en juegos de coordinación, donde la información se dispersa a través de una red. Demuestro cómo los cambios en la distribución de las conectividades de la población afectan a los tipos de coordinación en equilibrio, así como la probabilidad de éxito. En el tercer capítulo, analizo un marco de aprendizaje y cambio de personal en el mercado de trabajo. Muestro que emparejamiento selectivo positivo (PAM) se extiende más allá del entorno estable de Eeckhout y Weng (2010) a una situación de incertidumbre residual que exhibe períodos de des-aprendizaje. También extiendo esta configuración para permitir elementos de career concerns y muestro que el equilibrio de PAM sólo puede sostenerse bajo fuertes supuestos.

Yes<br>The quasi-orthogonal space time block coding (QO-STBC) over orthogonal frequency division multiplexing (OFDM) is investigated. Traditionally, QO-STBC does not achieve full diversity since the detection matrix of QO-STBC scheme is not a diagonal matrix. In STBC, the decoding matrix is a diagonal matrix which enables linear decoding whereas the decoding matrix in traditional QO-STBC does not enable linear decoding. In this paper it is shown that there are some interfering terms in terms of non-diagonal elements that result from the decoding process which limit the linear decoding. As a result, interference from the application of the QO-STBC decoding matrix depletes the performance of the scheme such that full diversity is not attained. A method of eliminating this interference in QO-STBC is investigated by nulling the interfering terms towards full diversity for an OFDM system. It was found that the interference reduction technique permits circa 2dB BER performance gain in QO-STBC. The theoretical and simulation results are presented, for both traditional QO-STBC and interference-free QO-STBC applying OFDM

Angiotensin II signals cardiovascular disease through metalloproteinases including MMP-2, MMP-7 and ADAM-17/TACE. We hypothesized that these metalloproteinases regulate each other at the transcriptional level. Further, MMP-2, being a major gelatinase in cardiac and vascular tissue, could mediate angiotensin II-induced cardiovascular disease. We studied the development of hypertension (by tail cuff plethysmography), cardiac hypertrophy (by M-mode echocardiography and qRT-PCR analysis of hypertrophy marker genes) and fibrosis (by collagen staining and qRT-PCR analysis of fibrosis marker genes) in mice receiving angiotensin II. Angiotensin II induced hypertension, cardiac hypertrophy and fibrosis which correlated with an upregulation of MMP-2. Downregulation of MMP-2 by pharmacological inhibition and RNA interference attenuated hypertension but not cardiac hypertrophy or fibrosis. Downregulation of MMP-7 or ADAM-17/TACE by RNA interference attenuated angiotensin II-induced MMP-2 upregulation as well as hypertension, cardiac hypertrophy and fibrosis. We conclude that MMP-2 selectively mediates angiotensin II-induced hypertension under the transcriptional control of MMP-7 and ADAM-17/TACE.

碩士<br>輔仁大學<br>電子工程學系<br>97<br>This thesis analysis the performance of an equalization/diversity-combiner operating in mobile radio environment with quadrature amplitude modulation (QAM) transmission, frequency-selective Rayleigh fading, and co-channel interference (CCI). The finite-length MMSE linear equalization/diversity-combiner has been shown to be an effective receiver structure for eliminating intersymbol interference (ISI) and cyclostationary CCI. This receiver structure with finite-length equalization uses the linear scheme and the decision feedback (DFE) scheme according to the minimum mean-square-error (MMSE) criterion. The simulation reveals that the symbol-spaced DFE, unlike operating in AWGN channel, cannot eliminate cyclostationary CCI efficiently and is even worse than symbol-spaced linear equalizer case. Since the accurate error probability estimation is time consuming, we use the Gaussian Quadrature Rule (GQR) method which is fast and accurate in evaluating average error probability.

碩士<br>國立中央大學<br>電機工程學系<br>103<br>Interference alignment (IA) is a technique to eliminate the interference in wireless communication system. Combined with coordinated multi-point (CoMP), this method could improve the system sum rate performance for cell edge user in LTE-A system. In this thesis, a new IA algorithm for uplink coordinated multi-point (CoMP) is proposed. The unselfish per-user signal-to-leakage-and-noise ratio (SLNR) criterion is used to design the precoder. On the other hand, the design of decoder adopts the selfish algorithm, per-user signal-to-interference-and-noise-ratio (SINR). The combination of both method does not need the channel reciprocity assumption and thus is suitable to operate in the case of different user transmission power. Through iterative procedure, we show that the per-user-based criterion which keeps user data streams orthogonal can suppress interference effectively and achieve higher sum rate than the conventional IA algorithms, such as minimum weighted leakage interference (min leakage), maximum per-stream SINR algorithms, and the hybrid IA in the multi-user CoMP joint reception scenarios. Eigenvalue decomposition (EVD) plays a key role for our proposed algorithm as well as the conventional algorithm, min leagake method. In our proposed method, EVD is needed to compute the signal subspaces. On the contrary, min leakage method needs the calculation of interference subspace. The decomposed matrix of both IA methods is always rank-deficient. A new eigen-solver based on QR decomposition (QRD) with shift is presented. Hessenberg reduction is implemented in the first stage to reduce the computation complexity. The proposed method could find the correct eigenpairs for both full rank and rank deficient matrix. The architecture of proposed method is shown to be more flexible for any matrix size and has less complexity than the existing method. The proposed eigen-solver could save up to 92% hardware complexity than the conventional EVD to find the nullspacefor hermitian symmetric matrix when the rank of the matrix is quite small . This method is reliable to be implemented due to its equal performance compared to MATLAB “eig” function.

碩士<br>國立成功大學<br>環境醫學研究所<br>93<br>Organophosphorous pesticides (OPPs), mainly constituting phosphoric ester, have been extensively used in pest control in agricultural activities owing to their ready decomposition and rapid efficacy. Dialkylphosphates (DAPs) concentrations in urine have been considered as biological exposure markers for the exposure to OPPs. Conventional sample preparation of solvent extraction has been blamed for its considerable consumption of organic solvents, time and manpower demanding. Therefore, Strong Anion Exchange (SAX) technique is becoming a promising alternative. The recoveries of DAPs in water using SAX technique were about 76-96%, but those in urine were most less than 50%. The abrupt decrease in recovery for urine samples could be due to the anion matrix effect of urine. The purposes of this study were to identify the target anions, esp., chloride and phosphate, causing matrix effects and to develop a method to diminish matrix interference during sample preparation incorporated with SAX technique to determine urinary DAPs. The water solutions of sodium chloride (ranging from 0.001–10,000 µg/mL) and sodium phosphate (ranging from 0.0025–2,500 µg/mL) were used to evaluate the target anion influencing SAX extraction. Silver hydroxide, developed by the combination of silver nitrate and sodium hydroxide, was used as a matrix retardant. We found phosphate ion cause very mild effects on SAX recoveries (the recoveries for DETP, DEDTP, DMTP, and DMDTP all lay between 74 and 94%) except for DEP (recovery decreased to only 14%). On the other hand, chloride ion revealed a significant effect on DAPs recoveries; i.e., the recoveries for DEP, DETP, and DETP were down to 0%, 6%, and 8%, respectively. When silver hydroxide as matrix retardant was added in the water solution of sodium chloride at the concentration 3-fold of chloride, the recoveries for DEP, DETP, and DMTP could reach their maximum of 14%, 98%, and 106%, respectively, at the cost of the recovery decreases of DEDTP and DMDTP to 12% and 16%, respectively. We concluded that two separate procedures incorporated with SAX technique for the determination of DAPs in urine: silver hydroxide could be added as a matrix retardant for DEP, DETP, and DMTP; and no any matrix retardant required for DEDTP and DMDTP.

Yes<br>A quasi-orthogonal space time block coding (QO-STBC) scheme that exploits Hadamard matrix properties is studied and evaluated. At first, an analytical solution is derived as an extension of some earlier proposed QO-STBC scheme based on Hadamard matrices, called diagonalized Hadamard space-time block coding (DHSBTC). It explores the ability of Hadamard matrices that can translate into amplitude gains for a multi-antenna system, such as the QO-STBC system, to eliminate some off-diagonal (interference) terms that limit the system performance towards full diversity. This property is used in diagonalizing the decoding matrix of the QOSTBC system without such interfering elements. Results obtained quite agree with the analytical solution and also reflect the full diversity advantage of the proposed QO-STBC system design scheme. Secondly, the study is extended over an interference-free QO-STBC multi-antenna scheme, which does not include the interfering terms in the decoding matrix. Then, following the Hadamard matrix property advantages, the gain obtained (for example, in 4x1 QO-STBC scheme) in this study showed 4-times louder amplitude (gain) than the interference-free QOSTBC and much louder than earlier DHSTBC for which the new approach is compared with.

abstract: Both two-way relays (TWR) and full-duplex (FD) radios are spectrally efficient, and their integration shows great potential to further improve the spectral efficiency, which offers a solution to the fifth generation wireless systems. High quality channel state information (CSI) are the key components for the implementation and the performance of the FD TWR system, making channel estimation in FD TWRs crucial. The impact of channel estimation on spectral efficiency in half-duplex multiple-input-multiple-output (MIMO) TWR systems is investigated. The trade-off between training and data energy is proposed. In the case that two sources are symmetric in power and number of antennas, a closed-form for the optimal ratio of data energy to total energy is derived. It can be shown that the achievable rate is a monotonically increasing function of the data length. The asymmetric case is discussed as well. Efficient and accurate training schemes for FD TWRs are essential for profiting from the inherent spectrally efficient structures of both FD and TWRs. A novel one-block training scheme with a maximum likelihood (ML) estimator is proposed to estimate the channels between the nodes and the residual self-interference (RSI) channel simultaneously. Baseline training schemes are also considered to compare with the one-block scheme. The Cramer-Rao bounds (CRBs) of the training schemes are derived and analyzed by using the asymptotic properties of Toeplitz matrices. The benefit of estimating the RSI channel is shown analytically in terms of Fisher information. To obtain fundamental and analytic results of how the RSI affects the spectral efficiency, one-way FD relay systems are studied. Optimal training design and ML channel estimation are proposed to estimate the RSI channel. The CRBs are derived and analyzed in closed-form so that the optimal training sequence can be found via minimizing the CRB. Extensions of the training scheme to frequency-selective channels and multiple relays are also presented. Simultaneously sensing and transmission in an FD cognitive radio system with MIMO is considered. The trade-off between the transmission rate and the detection accuracy is characterized by the sum-rate of the primary and the secondary users. Different beamforming and combining schemes are proposed and compared.<br>Dissertation/Thesis<br>Doctoral Dissertation Electrical Engineering 2018

Digestion of proteins in the midgut of lepidopteran larvae relies on different types of peptidases, among the trypsins and chymotrypsins. In this work four chymotrypsinlike peptidases (MsCTLP1–4) were identified from the larval midgut of M. sexta, which are distantly related to another chymotrypsin (MsCT), a previously described peptidase present in the larval midgut of M. sexta. MsCTLP1–4 fit perfectly into a novel subgroup of insect CTLPs by sequence similarity and by the replacement of GP by SA in the highly conserved GDSGGP motif. Examination of MsCTLP expression in different tissues showed that most of the peptidases were predominantly expressed in the anterior and median midgut, while some were found in the Malpighian tubules. Expression analysis of MsCTLPs at different physiological states revealed that the mRNA amounts did not differ considerably in feeding and starving larvae except for MsCTLP2, whose mRNA dropped significantly upon starvation. During molting, however, the mRNA amounts of all MsCTLPs dropped significantly. Immunological determination of MsCTLP1 amounts showed that the mature peptidase was only detectable in the gut lumen of feeding and re-fed larvae, but not in that of starving or molting larvae, suggesting that MsCTLP1 secretion is suspended during starvation or molt. Differential regulation of transcript levels as well as their partial expression in Malpighian tubules might point to a role, which is distinct from digestion for at least some MsCTLPs. In line with this assumption, MsCTLP1 was shown to interact with the chitin synthase 2 (MsCHS2), necessary for chitin synthesis in the course of peritrophic matrix formation in the midgut of M. sexta. The occurrence of this interaction in vivo is supported by colocalization and co-immunoprecipitation. The data suggest that chitin synthesis is controlled by an intestinal proteolytic signaling cascade linking chitin synthase activity to the nutritional state of the larvae. As MsCTLP1 appears to be involved in such signaling cascades, other midgut peptidases could have other targets and may therefore regulate different activities. To gain more insight into the functions of CTLPs, the gene family encoding these peptidases in the genome of the red flour beetle, T. castaneum, was analyzed. Using an extended search pattern, 14 TcCTLP genes were identified that encode peptidases with S1 specificity pocket residues typically found in chymotrypsin-like enzymes. Analysis of the expression patterns of seven TcCTLP genes at various developmental stages revealed that some TcCTLP genes were exclusively expressed in feeding larval and adult stages (TcCTLP-5A/B, TcCTLP-6A). Others were also detected in non-feeding embryonic (TcCTLP-5C, TcCTLP-6D) and pupal stages (TcCTLP-5C, TcCTLP- 6C/D/E). TcCTLP genes were expressed predominantly in the midgut where they presumably function in digestion. However, TcCTLP-5C and TcCTLP-6C also showed considerable expression in the carcass. The latter two genes might therefore encode peptidases that act as molting fluid enzymes. To test this hypothesis, western blots were performed using protein extracts from larval exuviae. The extracts reacted with antibodies to TcCTLP-5C and TcCTLP-6C suggesting that the corresponding peptidases are secreted into the molting fluid. Finally, systemic RNAi experiments were performed. While injections of dsRNAs to TcCTLP-5A/B and TcCTLP-6A/D/E into penultimate larvae did not affect growth or development, injection of dsRNA for TcCTLP-5C and TcCTLP-6C resulted in severe molting defects. Recombinant expressed TcCTLP-5C2 was moreover activated by trypsin and was able to hydrolyze AAPF, hence making TcCTLP-5C the first described chymotrypsin-like peptidase ever to be involved in molting.

Methods for the determination of trace elements in interfering matrices were developed. The methods involve two strategies for the reduction of interferences: the use of new pretreatment methods for use with existing instrumentation, and the development of new instrumentation. A comparison was made of inductively coupled plasma-mass spectrometry (ICP-MS) detection limits and interference reduction for arsenic by hydride generation, flow injection, continuous aspiration, and mixed gas methods. Practical considerations in implementing these common techniques, as well as in the general use and maintenance of an ELAN 5000 ICP-MS instrument are presented for the benefit of new users. Two approaches for the elimination of interferences in the determination of arsenic in gold by ICP-MS. The main challenges in this analysis were the arsenic signal suppression by gold and the polyatomic interference due to chloride. A method was developed for the batch mode reductive precipitation removal of interfering gold matrix from the sample prior to hydride generation and detection of arsenic. This method has great potential value as the basis for other simple methods for the removal of transition metals that interfere with hydride generation. An on-line anion exchange method for the removal of interfering gold prior to hydride generation was subsequently developed to improve the speed of analysis. The batch-mode reduction method for the removal of gold resulted in a better detection limit than the anion exchange method. However, the automated anion exchange gold removal method required less bench chemistry, was faster, consumed less reagents, and created less mess than the off-line method. On-line solid phase extraction methods are a valuable and increasingly popular method for the removal of matrix interferences. The automation of sample pretreatment as well as the small sample size required when the technique is combined with flow injection make it ideal for use in the analysis of hazardous materials. The severe signal suppression of uranium on lithium, beryllium, magnesium, and aluminum was overcome by the on-line flow injection-solid phase extraction removal of uranium using a column packed with Tru-Spec resin. The column was regenerated on-line using ammonium oxalate. The development of an internal combustion engine as a novel source for atomic emission spectrometry was begun. This preliminary stage of development focused on the identification of important experimental parameters to be considered in future work. A one cylinder four stroke engine was modified to allow optical access through the cylinder head. Combustion radiation was monitored with a photomultiplier tube and the signal was stored using LabView software. It is hoped that this work will be the basis for the development of an inexpensive, portable, rugged instrument that will accept a variety of fuels.

Thesis (MTech. degree in Chemistry)--Tshwane University of Technology, 2008.<br>This work is aimed at experimental verification of the theory about the advantages of the two-step sample vapour release in a fast-heated ballast furnace. The term “ballast” was introduced earlier in electrothermal atomic absorption spectrometry, as an alternative to a platform to describe a compact body of refractive material loosely located on the bottom of a tube furnace atomiser. The thermal behaviour of the ballast furnace is similar to that of the platform, but without restriction created by the platform area. Compared with the flat or concave platform, a compact ballast of similar mass to the platform should have less impact on gas temperature because of the smaller surface area. The theoretical predictions concerning atomisation efficiency in the fast-heated ballast furnace were examined by the determination of metals in organic and inorganic matrices using a Quantum Z.ETA atomic absorption spectrometer. The instrument provided fast heating of the tube atomizer, 10 K ms-1. It is shown that in the employed ballast furnace the vapour released into the gas phase occurs after interim condensation on the ballast. For the samples of tetraethyllead, base oil and aqueous solutions of various metals, analytical signals are observed after stabilisation of tube temperature, independent of volatility of the analyte and level of temperature setting. For those samples, a high gas phase temperature provides complete recovery of the analyte without involvement of chemical modifiers and the reduction of spectral interferences from chloride matrices.