Este trabalho apresenta a fabricação e a caracterização elétrica de contatos Al/Ti/Ni(Pt)Si sobre junções rasas N+P com aproximadamente 0,2 ìm de profundidade, sendo que o monosiliceto de níquel foi formado a partir da estrutura Ni(30nm)/Pt(1,5nm)/Si. O comportamento elétrico dos diodos obtidos no melhor processo foi adequado, com as seguintes médias e desvios padrões: corrente reversa por unidade de área de 33,8nA/cm2 ±12,3 nA/cm2 e corrente reversa por unidade de perímetro de 654pA/cm ±229pA/cm para tensão reversa de -5V, a resistência reversa dos diodos quadrados de 268,9G? ±97,7G? e a resistência reversa dos diodos serpentinas de 35,5G? ±11,5G?, a tensão de início de condução resultou entre 0,55V e 0,56V, a resistência série em condução de 4,7? ±1,3?, fator de idealidade de 1,15 ±0,03, e corrente de saturação de 1,1x10-11A para diodos quadrados (300ìm x 300ìm). O menor valor de resistividade do filme de (Ni(Pt)Si) resultou 25ì?cm e a resistência de folha de 3,13 ?/? foram obtidas após a formação do mono-siliceto de níquel na temperatura de 600 ºC durante 120 segundos. As estruturas Kelvin apresentaram resistividade de contato de 15,0ì?.cm2 ±3,3ì?.cm2 e comportamento ôhmico estável para diversos níveis de corrente. Após uma extensa análise sobre modelagem de contato, foi elaborado um programa computacional desenvolvido em MATLAB, baseado em um método bem conhecido, isto é, uma malha de resistores tridimensional, que analisa os efeitos do fenômeno de concentração das linhas de corrente lateral no contato. Este programa foi aplicado em contatos com siliceto de níquel, onde foram observadas reduções de até 32% na resistividade real do contato. / This work presents the fabrication and electrical characterization of Al/Ti/Ni(Pt)Si contacts having the nickel monosilicide formed from Ni(30nm)/Pt(1.5nm)/Si structure on shallow N+P junctions with about 0.2 ìm of depth. The diodes? electrical behavior achieved at the best process was considered good, with the following average and standard deviations: area diode leakage current of 33.8nA/cm2 ±12.3nA/cm2 and periphery diode leakage current of 654pA/cm ±229pA/cm for reverse voltage of -5V, the square diode reverse resistance of 268.9G? ±97.7G? and serpentine diode reverse resistance of 35.5G? ±11.5G?, forwardbias voltage between 0.55V and 0.56V, forward series resistance of 4.7? ±1.3?, ideality factor of 1.15 ±0.03, and reverse saturation current of 1.1x10-11A for square diodes (300ìm x 300ìm). The lowest film resistivity value (Ni(Pt)Si) of 25ì?cm and sheet resistance of 3.13 ?/? were obtained for the formation of nickel monosilicide under temperature of 600ºC for 120 seconds. The cross-bridge Kelvin resistors presented contact resistivity of 15.0 ì?.cm2 ±3.3 ì?.cm2 and stable ohmic behavior for several electrical current levels. After extensive analysis about contact modeling, a computer program was elaborated in MATLAB, based on a well-known three-dimensional resistor network, which analyses the lateral current crowding effects on contact. This program was applied for contacts with nickel silicide, where a decrease up to 32% at the real contact resistivity was observed.

Cette étude adresse une technologie de diodes laser à semi-conducteur InGaAs/AlGaAs/GaAs émettant à 980 nm en configuration puce nue (CSE-Composant Sur Embase) utilisées pour le pompage optique dans les amplificateurs à fibre dopée Er3+. Il s'agit de composants possédant un tel niveau de maturité technologique que l’évolution des paramètres observés au cours du temps ne présente plus de variations suffisamment significatives pour pouvoir dégager des conclusions exhaustives en termes de fiabilité. La recherche de méthodes alternatives et/ou complémentaires aux méthodes dites « classiques » visant à la compréhension des mécanismes de défaillance et l’identification des signatures indiquant une possible dégradation future des diodes laser, relève aujourd’hui un défi stratégique pour les composants actuels. Dans ce contexte, cette étude propose un ensemble de techniques basées sur la discrimination du fonctionnement des diodes en régime direct et en particulier en régime inverse par spectroscopie électrique. La corrélation des différentes mesures en régime inverse, très peu étudié dans la caractérisation des diodes laser, peut mettre en évidence des comportements atypiques qui révèlent la présence de défauts ponctuels résiduels dans le volume d’une diode car les courants observés sont très faibles. Le régime inverse permet d'offrir des perspectives intéressantes en considérant que ce régime reste de nos jours quasiment inexploré pour les composants optoélectroniques émissifs tels que les diodes laser. / This study addresses InGaAs / AlGaAs / GaAs laser diode emitting at 980 nm in bare chip configuration (COS-Component on Submount) for optical pumping in Er3+ doped fiber amplifiers. These devices have a level of technological maturity that the changes in the parameters observed during aging do not present sufficiently significant variations in order to obtain exhaustive conclusions in terms of reliability. Searching for alternative and/or complementary methods to the so-called « classical » methods aimed to understanding the failure mechanisms and the identification of signatures indicating possible future degradation of the laser diodes, represents today a strategic challenge for the current components. In this context, this study suggests a set of techniques based on the discrimination of the operation mode of the diodes in direct bias and in particular in reverse bias by electrical spectroscopy. The correlation of the different measurements in reverse bias, not more studied in the characterization of laser diodes, can reveal atypical behaviors highlighting presence of residual point defects in the volume of a diode because the currents observed are very weak. The reverse bias makes it possible to offer interesting perspectives considering that, this operation mode remains today almost unexplored for optoelectronic emitting devices such as laser diodes.

Dans cette thèse, les diodes Schottky pour des applications en ondes millimétriques et aux fréquences térahertz sont étudiées. Une méthodologie de conception et d'optimisation est proposée pour améliorer la performance de telles diodes. La conception et les simulations sont effectuées à l'aide d'un programme basé sur un modèle analytique. Les différentes méthodes de calcul de la fréquence de coupure de la diode sont définies, étudiées et classifiées selon les applications potentielles. En utilisant un modèle de diode générique et général, une nouvelle approche pour calculer la fréquence de coupure est suggérée pour les applications de mélangeur / multiplicateur. Cette approche permet d'évaluer la tension seuil avec une précision beaucoup plus grande et proche de la réalité. En outre, la conception d’une diode Schottky en tenant compte dès le départ l’application visée (détecteur direct, mélangeur ou multiplicateur) est étudiée. Cette thèse montre que l'ingénierie de la structure épitaxiale a un impact important lorsque l’on utilise une conception de diode basée sur l’application finale comme proposée. Un procédé de microfabrication a été entièrement développé et caractérisé. Une méthode de planarisation unique est introduite pour permettre de connecter la diode par des ponts à air en minimisant les effets parasites. Afin d'éviter une coûteuse lithographie par faisceau électronique, une anode en forme de T est produite en utilisant une technique de photolithographie. Ce procédé est fiable et répétitif, est de faible coût et offre une grande souplesse en matière de conception en plus de répondre au besoin d‘une production de masse, pour laquelle la lithographie par faisceau d’électrons n’est guère possible. Le procédé final nécessite simplement deux étapes de métallisation, nombre minimal possible que nous avons atteint. En raison des exigences de recuit du contact ohmique, il est impossible d’avoir moins de deux étapes de métallisation. Le processus de planarisation proposé repose sur l'utilisation de différents taux de gravure plasma de deux résines couramment utilisées. Pour les travaux réalisés dans cette thèse, une épitaxie GaAs HBT disponible au sein du laboratoire a été utilisée. Les résultats de caractérisation de diodes réalisés dérivés des mesures DC et RF sont rapportés et comparés avec les résultats de la simulation. Les résultats de mesure montrent une réduction significative de la capacité parasite de la diode à moins de 20% de sa capacité totale. Par conséquent, le procédé de conception et de fabrication de ce travail peut fournir des diodes qui peuvent fonctionner au-delà du térahertz avec des dimensions pour l’anode plus grandes que les diodes trouvées dans la littérature et qui peuvent donc être fabriquées uniquement par des techniques de photolithographie optique. / Abstract: In this thesis, Schottky diodes for millimeter waves and terahertz application are scrutinized. A design and optimization methodology is proposed to improve the diode performance. Design and simulations are performed by using an analytical model based code. Diode cut-off frequency calculation methods are studied and classified for different applications. Considering general diode equivalent circuit model, a new approach for calculating the cut-off frequency is suggested for mixer/multiplier applications. This approach provides cut-off much closer to its practical value. Also, the diode design based on its application, direct detector and mixer/multiplier, is studied. It is shown that the epitaxial structure engineering has impact on diode application based design. For diode realization a microfabrication process is developed. Unique planarization method is introduced which provides necessary substruction for the airbridges. In order to avoid expensive e-beam lithography, a T-shaped anode is produced by employing photolithography technique. This process is repeatable, reliable, low cost, gives high flexibility in design terms, and suitable for mass production. The final process merely requires two metallization steps which is minimum possible number due to annealing requirement of ohmic contact. The proposed planarization process is based on using different plasma etching rates of two common resists. In the diode fabrication an available GaAs HBT epitaxial wafer is used. The realized diode characterization results derived from DC and RF measurements are reported and compared with the simulation results. The measurement results showed significant reduction in parasitic capacitance of the diode to under twenty percent of its total capacitance. Therefore, the design and fabrication method of this work can provide diodes to operate over one terahertz with larger anode area (that can be produced by photolithography techniques).

Kyoto University (京都大学) / 0048 / 新制・論文博士 / 博士(工学) / 乙第12695号 / 論工博第4084号 / 新制||工||1555(附属図書館) / 29947 / (主査）教授 平尾 一之, 教授 田中 勝久, 教授 三浦 清貴 / 学位規則第4条第2項該当

Deep ultraviolet (UV) light sources are useful in a number of applications that include sterilization, medical diagnostics, as well as chemical and biological identification. However, state-of-the-art deep UV light-emitting diodes and lasers made from semiconductors still suffer from low external quantum efficiency and low output powers. These limitations make them costly and ineffective in a wide range of applications. Deep UV sources such as lasers that currently exist are prohibitively bulky, complicated, and expensive. This is typically because they are constituted of an assemblage of two to three other lasers in tandem to facilitate sequential harmonic generation that ultimately results in the desired deep UV wavelength. For semiconductor-based deep UV sources, the most challenging difficulty has been finding ways to optimally dope the (Al,Ga)N/GaN heterostructures essential for UV-C light sources. It has proven to be very difficult to achieve high free carrier concentrations and low resistivities in high-aluminum-containing III-nitrides. As a result, p-type doped aluminum-free III-nitrides are employed as the p-type contact layers in UV light-emitting diode structures. However, because of impedance-mismatch issues, light extraction from the device and consequently the overall external quantum efficiency is drastically reduced. This problem is compounded with high losses and low gain when one tries to make UV nitride lasers. In this thesis, we provide a robust and reproducible approach to resolving most of these challenges. By using a liquid-metal-enabled growth mode in a plasma-assisted molecular beam epitaxy process, we show that highly-doped aluminum containing III-nitride films can be achieved. This growth mode is driven by kinetics. Using this approach, we have been able to achieve extremely high p-type and n-type doping in (Al,Ga)N films with high aluminum content. By incorporating a very high density of Mg atoms in (Al,Ga)N films, we have been able to show, by temperature-dependent photoluminescence, that the activation energy of the acceptors is substantially lower, thus allowing a higher hole concentration than usual to be available for conduction. It is believed that the lower activation energy is a result of an impurity band tail induced by the high Mg concentration. The successful p-type doping of high aluminum-content (Al,Ga)N has allowed us to demonstrate operation of deep ultraviolet LEDs emitting at 274 nm. This achievement paves the way for making lasers that emit in the UV-C region of the spectrum. In this thesis, we performed preliminary work on using our structures to make UV-C lasers based on photonic crystal nanocavity structures. The nanocavity laser structures show that the threshold optical pumping power necessary to reach lasing is much lower than in conventional edge-emitting lasers. Furthermore, the photonic crystal nanocavity structure has a small mode volume and does not need mirrors for optical feedback. These advantages significantly reduce material loss and eliminate mirror loss. This structure therefore potentially opens the door to achieving efficient and compact lasers in the UV-C region of the spectrum.

The resonant tunneling diode (RTD) is the fastest electron device to-date in terms of its ability to generate continuous-wave terahertz frequency at room temperature, owing to its unique characteristic of negative differential resistance (NDR). In this work, a lattice-matched In0.53Ga0.47As (on InP) is used as the cladding layer, while a highly-compressive strained In0.8Ga0.2As is sandwiched between two tensile-strained pseudomorphic AlAs barriers to form the active double barrier quantum well RTD structure grown by Molecular Beam Epitaxy. The ultimate aim of this work was to integrate an optimised RTD into an oscillator circuit to enable a 100 GHz (W-band) MMIC RTD oscillator. One of the key challenges in this work was to improve the DC performance of the RTD, through extensive material and structural characterisations. Growing nano-scale epitaxial layers require a high degree of controllability with mono-layer precision. The dependencies of the NDR components, such as the peak current density, peak voltage and peak-to-valley current ratio (PVCR) towards variations in structural thickness were studied systematically. Through this work, it is found that the peak current density is strongly affected by monolayer variation in barrier thickness. The effect of quantum well thickness variation towards peak current density is relatively weaker. Interestingly, variation in spacer layer thickness has very little influence towards the magnitude of the peak current density. The fabrication of the RTD using a conventional i-line optical lithography created its own challenge. The process capability to reduce mesa active area down to sub-micrometer level to reduce device’s geometrical capacitance for high frequency, THz applications has been made feasible in this work. The conventional i-line optical lithography was combined with a newly developed tri-layer soft reflow technique using solvent vapour resulted in sub-micrometer RTDs. The DC characterisation of the fabricated RTDs showed excellent device scalability, indicating a robust processing. This novel sub-micron processing technique with high throughput and repeatability is a very promising low cost technique. A collaborative effort between the University of Manchester and Glasgow paved the way towards the realisation of an integrated W-band RTD MMIC oscillator. The circuit-combining topology was designed by the High Frequency Electronics Group in Glasgow while the mask-layout and oscillator fabrication took place in Manchester. An active RTD from sample XMBE#301 with peak current density of 1.4 x 105 A/cm2 and PVCR of 4.5 was integrated into a 100 GHz MMIC oscillator to successfully produce a measured frequency of 109 GHz with an un-optimised 5.5 μW output power at room temperature (mesa area = 4x4 μm2).

This diploma thesis deals with optical and optoelectronic components. The first part describes three selected optical and three optoelectronic components. Optical components include power divider, isolator and circulator. The optoelectronic ones are laser diode, photodetector and modulator. Basic measurement parameters were defined for each component. In the experimental part, four components from different manufacturers were measured. Selected components are power dividers, isolators, circulators and a laser diode. Subsequently, the measured parameters were compared with the catalog values.

The purpose of this thesis is implementation of RGB LED driver using circuit ON Semiconductor NCV7430. The main objective was to design a circuit solution for temperature compensation of the driver. The thesis aims to describe the driver and its functions and to eliminate thermal effects caused by heating the circuit by surroundings. It discusses the circuit thermal stabilization in the temperature range of -40 to 80 °C to the RGB diode lights to a constant color. The thesis presents various possibilities of the stabilization and method of evaluating the obtained parameters of applications. Next, the thesis solves the design of equipment for implementation of light effects. The equipment presents options and features of the RGB driver.

The aim of this work is primarily to validate the advanced techniques for treatment planning and dosimetric verification for modern megavoltage x-ray radiotherapy. With the advent of modern radiotherapy techniques, there is a great need for assuring quality of the radiation dose distributions generated by the advanced intensity modulated treatments (IMRT/VMAT). This is typically accomplished by the assessment of the treatment plan quality at the planning stage and then verification of the dose distributions through measurements on the phantoms or independent dose calculations prior to the actual delivery of these plans to patients. The major focus of this work is to clinically evaluate the modern 2D and 3D dose verification techniques. The measurement-based dosimetry systems investigated were ArcCHECK/3DVH and SRS MapCHECK. AcrCHECK/3DVH system uses the measurement-guided dose reconstruction algorithm to correct the predicted dose in the patient dataset. The system was intended for VMAT/IMRT QA. SRS MapCHECK was investigated for SRS treatments. The independent dose calculation system was DoseCHECK which employed a GPU-accelerated convolution-superposition of algorithm for 3D dose reconstruction on the patient dataset. Next, a hybrid dose verification system (PerFRACTION) was evaluated, which takes input from both the treatment planning system and the linac EPID and produces a measurement-guided 3D dose distribution for comparison with the plan. This system was investigated for potential QA applications to a modern, efficient SRS technique, involving simultaneously treating multiple targets with a single isocenter. The performance of all dosimetry systems was validated against well-characterized independent dosimeters, such as ion chamber, film and scintillator detectors, or 3D arrays (Delta4), using stringent dose comparison criteria to test their limits for the intended clinical applications. For the initial plan quality evaluation of a novel tool (Feasibility DVH) was investigated. This tool a priori estimates best achievable dose volume histograms for a specific patient, based on the basic physics properties of the megavoltage x-rays, thus helping the planners to guide their efforts. All studied dosimetry systems showed an excellent agreement of the average gamma (a mathematical combination of DD and DTA) passing rates >98% for most of the plans. The 3% DD/2mm DTA criteria were used for extracranial plans and 3%/1mm for intracranial SRS plans. As dictated by the logic of the application, the comparisons were made against TPS calculations, a bi-planar array, or film measurements. Similarly the average percent point dose errors <2% were observed against the ion chambers or film. In the rare instances when the deviations were larger, intuitive explanations were provided, based on either the physics of the plans or inhomogeneous patient anatomy and resulting algorithm limitations. Feasibility DVH was shown to reliably predict the best possible organ sparing for clinical head-and-neck VMAT plans. Overall the investigated dosimetry systems were found reliable and feasible for their intended clinical use.

Digital silicon photomultipliers (dSiPMs) and single-photon avalanche diode (SPAD) imagers are optical sensing systems formed from the integration of time-to-digital converters (TDCs) with arrays of highly sensitive photodetectors known as SPADs. TDCs are high-performance mixed-signal circuits capable of timestamping events with picosecond level resolution. The digital operation mechanisms of SPADs allow for their outputs to be sent to TDCs, where the timestamps of individual photon detections are recorded. In recent years, time-resolved SPAD-based sensors have been a heavily studied topic due to their exceptional performance potential in biomedical imaging applications, including time-of-flight (ToF) positron emission tomography (PET), fluorescence lifetime imaging microscopy (FLIM), and diffuse optical tomography (DOT). This work targets the optimization of these sensors in low-cost standard complementary metal-oxide-semiconductor (CMOS) processes. Firstly, this thesis provides a detailed review of the work accomplished in CMOS TDCs and their integration in SPAD-based sensors. Next, a feedback time amplification TDC was designed and tested in the TSMC 65 nm process that can achieve < 5 ps timing resolution in a very compact area of 0.016 mm2. The design is then described for a multi-time-gated array of p+/n-well SPADs that aims to mitigate SPAD dark noise while providing high-speed imaging by applying shifted gate windows simultaneously to an array of SPADs. The p+/n-well SPAD is first characterized in a passive quench configuration where it demonstrated a maximum dark count rate of 44.9 kHz, 18.1% peak PDP at 420 nm, and 0.82 ns timing jitter at a 0.7 V excess bias. While the current multi-time-gated prototype is not fully functional, the measurement results for individual pixels of the multi-time-gated array showed a 3.25 ns median gate window with a 2.2x 10-4 dark count probability for a 0.7 V excess bias, with 440 ps timing resolution and ~1 LSBrms timing jitter. Based on the results, limitations of the current design and sources for future improvement are then discussed in detail. / Thesis / Master of Applied Science (MASc) / Medical imaging plays a key role in the diagnosis of diseases like cancer, and as such, the optimized performance of medical imaging systems is a large area of research. Recently, highly sensitive photodetectors known as single-photon avalanche diodes (SPADs) were integrated with high-performance timing circuits known as time-to-digital converters (TDCs) to form digital silicon photomultipliers (dSiPMs) and SPAD imagers. DSiPMs and SPAD imagers are capable of timestamping the detection of individual photons with a very high level of accuracy in order to generate biomedical images. This thesis focuses on the design and measurement of these sensors using standard fabrication processes with the aim of working towards high-performance medical imaging sensors at a low cost. Firstly, we review the results achieved in TDCs and SPAD-based sensors within the recent literature. Following that, we present the design and performance results of a custom-designed TDC that aims to achieve state-of-the-art performance within a small area in order to maintain low-cost and optimal integration with SPADs. Next, the design is described for an array of custom time-gated SPADs with integrated TDCs. Finally, the SPAD is characterized in two different configurations to identify sources of improvement for future design iterations.