Pour concevoir avec succès l'instrumentation nécessaire aux nouvelles technologies de fabrication avec une précision nanométrique, la méthodologie de conception doit prendre en compte de nombreux sujets différents liés à la chimie, à la physique, à la mécanique, à l'électronique et à l'automatisation, travaillant ensemble pour atteindre l'objectif souhaité. Dans cette thèse, cette méthodologie de conception a été mise en œuvre avec un grand nombre d’outils et d’approches permettant d’optimiser avec succès une méthode de nanofabrication appelée dépôt par couche atomique spatiale (SALD) afin de déposer des couches minces d’un matériau potentiellement utile en tant que composant du dispositifs à énergie solaire non-silicium, séparateurs d’eau photoélectrochimiques et composants électroniques transparents à couche mince, entre autres: oxyde cuivreux (Cu2O).En ce qui concerne la technologie de fabrication et la conception mécatronique, SALD est une technique de fabrication prometteuse qui permet la fabrication de films minces avec une précision nanométrique et avec la capacité de contrôler leurs propriétés mécaniques, électriques et cristallographiques. De plus, l'approche SALD utilisée dans cette thèse et dans le Laboratoire des matériaux et du génie physique (LMGP) fonctionne à l'air libre (sans chambre de dépôt) et constitue donc potentiellement une approche compatible avec l'industrie pour les films minces homogènes de grande surface fabrication avec un débit élevé. De plus, SALD peut être utilisé dans des conditions qui le rendent compatible avec les substrats flexibles et avec les approches de rouleau à rouleau (R2R). Enfin, SALD offre une flexibilité sur le processus de dépôt afin qu’il puisse être ajusté pour obtenir différentes propriétés sur les films fabriqués avec un minimum de modification de l’instrumentation.À l'aide de simulations CFD (Computational Fluid Dynamics), les phénomènes de la mécanique des fluides qui se produisent pendant le processus de dépôt dans le système SALD ont été analysés pour différentes configurations du réacteur. L'influence sur les propriétés du film a été étudiée et une validation avec des dépôts expérimentaux a été effectuée. Ensuite, en utilisant les connaissances et les directives obtenues avec les simulations CFD, et afin de réduire le coût et la complexité de la modification de certains composants mécaniques du système, un flux de travail comprenant la conception assistée par ordinateur (CAO) et la fabrication additive (également appelé impression 3D) impression) a été mis en place au LMGP pour la fabrication de l’un des composants principaux du système SALD à LMGP: la tête de dépôt. Ici, c'est la première fois qu'une telle technique de fabrication innovante est appliquée aux processus de nanofabrication en couches minces, offrant de nombreuses applications potentielles dans le domaine. Dans cette thèse, un tel flux de travail est présenté et expliqué, ainsi que les directives apprises et les limitations découvertes également présentées.Enfin, couches minces de Cu2O ont été déposé avec succès avec la méthode SALD. Le Cu2O est l’un des rares matériaux aux propriétés électroniques prometteuses en tant que semi-conducteur transparent de type p. Ici, les films de Cu2O fabriqués utilisant le système SALD à LMGP sont rapportés et leur conductivité de type p et leur cristallographie sont analysées.Les résultats de ces travaux fournissent des directives initiales pour la conception industrielle d’un système de fabrication à haut débit basé sur la technologie SALD, dans lequel la conception de ses composants est optimisée pour chaque matériau souhaité. Cette approche de conception rend également ce travail utile pour augmenter la quantité de matériaux compatibles avec le SALD, ainsi que pour développer davantage la méthodologie SALD dans des processus de fabrication innovants de matériaux et de dispositifs. / To successfully design the instrumentation needed for new manufacturing technologies with nanoscale precision, the design methodology must take into account many different topics related to chemistry, physics, mechanics, electronics and automation, working together to achieve the desired goal. In this thesis, this design methodology has been implemented with a large number of tools and approaches to successfully optimize a nanofabrication method called spatial atomic layer deposition (SALD) in order to deposit thin films. a potentially useful material as a component of non-silicon solar energy devices, photoelectrochemical water separators and transparent thin-film electronic components, among others: cuprous oxide (Cu2O).With respect to manufacturing technology and mechatronics design, SALD is a promising manufacturing technique that enables the fabrication of thin films with nanoscale precision and the ability to control their mechanical, electrical and crystallographic properties. In addition, the SALD approach used in this thesis and in the Laboratoire des Matèriaux et du Génie Physique(LMGP) works in the open air (without a repository) and is therefore potentially an industry-compatible approach to film Thin homogeneous high-area manufacturing with high throughput. In addition, SALD can be used under conditions that make it compatible with flexible substrates and roll-to-roll approaches (R2R). Finally, SALD offers flexibility on the deposit process so that it can be adjusted to obtain different properties on films manufactured with a minimum of instrumentation modification.Using CFD (Computational Fluid Dynamics) simulations, the fluid mechanics phenomena that occur during the deposition process in the SALD system were analyzed for different reactor configurations. The influence on the properties of the film was studied and a validation with experimental deposits was carried out. Then, using the knowledge and guidance obtained with CFD simulations, and to reduce the cost and complexity of modifying certain mechanical components of the system, a workflow that includes computer-aided design (CAD) and manufacturing additive (also called 3D printing) printing) was set up at the LMGP for the manufacture of one of the main components of the LMGP SALD system: the deposit head. Here, it is the first time that such an innovative manufacturing technique has been applied to thin-film nanofabrication processes, offering many potential applications in the field. In this thesis, such a workflow is presented and explained, along with learned guidelines and discovered limitations also presented.Finally, thin layers of Cu2O have been successfully deposited with the SALD method. Cu2O is one of the few materials with promising electronic properties as a p-type transparent semiconductor. Here, Cu2O films made using the LMGP SALD system are reported and their p-type conductivity and crystallography are analyzed.The results of this work provide initial guidance for the industrial design of a high throughput manufacturing system based on SALD technology optimized for each desired material. This design approach also makes this work useful for increasing the amount of SALD compatible materials, as well as for further developing the SALD methodology in innovative materials and device manufacturing processes.

No description available.

No description available.

In this dissertation, I developed a Pd seeding method in association with electroless plating, to successfully metallize both lambda DNA and DNA origami templates on different surfaces. On mica surfaces, this method offered a fast, simple process, and the ability to obtain a relatively high yield of metallized DNA nanostructures. When using lambda DNA as the templates, I studied the effect of Pd(II) activation time on the seed height and density, and an optimal activation time between 10 and 30 min was obtained. Based on the Pd seeds formed on DNA, as well as a Pd electroless plating solution, continuous Pd nanowires that had an average diameter of ~28 nm were formed with good selectivity on lambda DNA. The selected Pd activation time was also applied to metallize "T"-shape DNA origami, and Au coated branched nanostructures with a length between 200-250 nm, and wire diameters of ~40 nm were also fabricated. In addition, I found that the addition of Mg2+ ion into the reducing agent and electroless plating solution could benefit the surface retention of Pd seeded DNA and Au plated DNA structures. This work indicated that DNA molecules were promising templates to fabricate metal nanostructures; moreover, the formation of Au metallized branched nanostructures showed progress towards nanodevice fabrication using DNA origami. Silicon surfaces were also used as the substrates for DNA metallization. More complex circular circuit DNA origami templates were used. To obtain high enough seed density, multiple Pd seeding steps were applied which showed good selectivity and the seeded DNA origami remained on the surface after seeding steps. I used distribution analysis of seed height to study the effect of seeding steps on both average height and the uniformity of the Pd seeds. Four-repeated palladium seedings were confirmed to be optimal by the AFM images, seed height distribution analysis, and Au electroless plating results. Both Au and Cu metallized circular circuit design DNA origami were successfully obtained with high yield and good selectivity. The structures were maintained well after metallization, and the average diameters of Au and Cu samples were ~32 nm and 40 nm, respectively. Electrical conductivity measurements were done on these Au and Cu samples, both of which showed ohmic behavior. This is the first work to demonstrate the conductivity of Cu metallized DNA templates. In addition, the resistivities were calculated based on the measured resistance and the size of the metallized structures. My work shows promising progress with metallized DNA and DNA origami templates. The resulting metal nanostructures may find use as conducting interconnects for nanoscale objects as well as in surface enhanced Raman scattering analysis.

The 3-satisfiability Problem (3-SAT) is a demanding combinatorial problem that is of central importance among the nondeterministic polynomial (NP) complete problems, with applications in circuit design, artificial intelligence, and logistics. Even with optimized algorithms, the solution space that needs to be explored grows exponentially with the increasing size of 3-SAT instances. Thus, large 3-SAT instances require excessive amounts of energy to solve with serial electronic computers. Network-based biocomputation (NBC) is a parallel computation approach with drastically reduced energy consumption. NBC uses biomolecular motors to propel cytoskeletal filaments through nanofabricated networks that encode mathematical problems. By stochastically exploring possible paths through the networks, the cytoskeletal filaments find possible solutions. However, to date, no NBC algorithm for 3-SAT has been available. Herein, an algorithm that converts 3-SAT into an NBC-compatible network format is reported and four small 3-SAT instances (with up to three variables and five clauses) using the actin–myosin biomolecular motor system are experimentally solved. Because practical polynomial conversions to 3-SAT exist for many important NP complete problems, the result opens the door to enable NBC to solve small instances of a wide range of problems.

III-V semiconductors are compounds made of elements from groups III and V of the periodic table. Most of these materials exhibit a direct bandgap, which makes them suitable for light emission and detection. Furthermore, ternary and quaternary III-V semiconductors offer some freedom in adjusting their material compositions, which also allows one to modify their bandgap energies, refractive indices, and other optical properties. This quality makes such materials suitable for the monolithic integration of laser sources with passive optical devices and detectors on a single chip. For example, such integration is used in indium phosphide (InP) technology for large-scale photonic integration in optical communication networks. Commercial integrated photonic circuits' functionality can be augmented by the implementation of nonlinear optical devices, enabling all-optical signal processing, frequency conversion, and on-chip sources of quantum light. This doctoral thesis focuses on design, fabrication, and testing of passive optical components based on III-V semiconductors. We explored various fabrication approaches for III-V nonlinear photonic devices. Among the III-V semiconductor platforms used in nonlinear photonics, we focused on AlGaAs as the most studied nonlinear optical platform, and InP and its quaternary derivatives as the most commercially developed platform. The fabrication processes for III-V photonic devices usually require the deposition of silica and chromium layers, and then three etch steps to etch the chromium, silica, and, finally, the III-V layer. In the thesis, we demonstrate a process which allows one to eliminate the chromium deposition and the associated etch step, thereby reducing the process complexity. We implemented this newly developed hard-mask process for etching numerous AlGaAs and InP photonic devices. This work was not only an important contribution to the University of Ottawa's cleanroom facility. The shared recipe can be used to recreate etch recipes for silica using soft masks like ZEP520a, PMMA, etc., at other similar university and research facilities around the world. The silica mask created using this process was later used to fabricate InP/InGaAsP-based half-core-etched and nanowire waveguides, which were used to perform the first reliable measurement of the nonlinear refractive index coefficient n₂ of InGaAsP/InP waveguides. We explored improved fabrication processes for AlGaAs waveguides, photonic crystals, and ring resonators. InP-based integrated optical devices are relatively difficult to fabricate because the etch byproducts are only volatile at elevated temperatures. Using a silica mask, we developed a very smooth etching process for InP waveguides with aspect ratios greater than 1:10. Suspended waveguide structures, where the guiding layer is surrounded by the air, are of great interest as they can exhibit large refractive index contrast for superior compactness and for achieving high intensity at low optical powers. We demonstrated fabrication process flows for creating suspended air-bridge structures in a 500-nm AlGaAs slab, which can be used in mid-IR sensing applications. The processes developed as part of this project cover a wide range of AlGaAs passive photonic devices such as waveguides, photonic crystals and ring resonators. Additionally, we demonstrated plasma etching selectivity improvements for AlGaAs etching using only a soft ZEP mask and were able to achieve a selectivity of 1:2.9. All these developments can be beneficial to other researchers working on III-V photonic devices. We also completed the first theoretical study of third-harmonic generation in dispersion-engineered AlGaAs suspended photonic crystal waveguide. Most importantly, we introduced a reliable and efficient method for modelling higher-order modes in photonic crystal waveguides that is less computationally intensive and far more accurate compared to the 3D FDTD method. We also experimentally demonstrate guided modes lying above the light line in AlGaAs photonic crystal waveguides. In one of the addition projects, we experimentally demonstrate third-harmonic generation (THG) in Silicon Nitride waveguides. In summary, this thesis presents details of the design and testing of different passive nonlinear III-V semiconductor photonic devices. In addition, this thesis presents the fabrication processes which can be used to reliably and repeatably fabricate photonic devices in these materials.

Résumé : Le transistor monoélectronique (SET) est un dispositif nanoélectronique très attractif à cause de son ultra-basse consommation d’énergie et sa forte densité d’intégration, mais il n’a pas les capacités suffisantes pour pouvoir remplacer complètement la technologie CMOS. Cependant, la combinaison de la technologie SET avec celle du CMOS est une voie intéressante puisqu’elle permet de profiter des forces de chacune, afin d’obtenir des circuits avec des fonctionnalités additionnelles et uniques. Cette thèse porte sur l’intégration 3D monolithique de nanodispositifs dans le back-end-of-line (BEOL) d’une puce CMOS. Cette approche permet d’obtenir des circuits hybrides et de donner une valeur ajoutée aux puces CMOS actuelles sans altérer le procédé de fabrication du niveau des transistors MOS. L’étude se base sur le procédé nanodamascène classique développé à l’UdeS qui a permis la fabrication de dispositifs nanoélectroniques sur un substrat de SiO2. Ce document présente les travaux réalisés sur l’optimisation du procédé de fabrication nanodamascène, afin de le rendre compatible avec le BEOL de circuits CMOS. Des procédés de gravure plasma adaptés à la fabrication de nanostructures métalliques et diélectriques sont ainsi développés. Le nouveau procédé nanodamascène inverse a permis de fabriquer des jonctions MIM et des SET métalliques sur une couche de SiO2. Les caractérisations électriques de MIM et de SET formés avec des jonctions TiN/Al2O3 ont permis de démontrer la présence de pièges dans les jonctions et la fonctionnalité d’un SET à basse température (1,5 K). Le transfert de ce procédé sur CMOS et le procédé d’interconnexions verticales sont aussi développés par la suite. Finalement, un circuit 3D composé d’un nanofil de titane connecté verticalement à un transistor MOS est réalisé et caractérisé avec succès. Les résultats obtenus lors de cette thèse permettent de valider la possibilité de co-intégrer verticalement des dispositifs nanoélectroniques avec une technologie CMOS, en utilisant un procédé de fabrication compatible. / Abstract : The single electron transistor (SET) is a nanoelectronic device very attractive due to its ultra-low power consumption and its high integration density, but he is not capable of completely replace CMOS technology. Nevertheless, the hybridization of these two technologies is an interesting approach since it combines the advantages of both technologies, in order to obtain circuits with new and unique functionalities. This thesis deals with the 3D monolithic integration of nanodevices in the back-end-ofline (BEOL) of a CMOS chip. This approach gives the opportunity to build hybrid circuits and to add value to CMOS chips without fundamentally changing the process fabrication of MOS transistors. This study is based on the nanodamascene process developed at UdeS, which is used to fabricate nanoelectronic devices on a SiO2 layer. This document presents the work done on the nanodamascene process optimization, in order to make it compatible with the BEOL of CMOS circuits. The development of plasma etching processes has been required to fabricate metallic and dielectric nanostructures useful to the fabrication of nanodevices. MIM junctions and metallic SET have been fabricated with the new reverse nanodamascene process on a SiO2 substrate. Electrical characterizations of MIM devices and SET formed with TiN/Al2O3 junctions have shown trap sites in the dielectric and a functional SET at low temperature (1.5 K). The transfer process on CMOS substrate and the vertical interconnection process have also been developed. Finally, a 3D circuit consisting of a titanium nanowire connected to a MOS transistor is fabricated and is functional. The results obtained during this thesis prove that the co-integration of nanoelectronic devices in the BEOL of a CMOS chip is possible, using a compatible process.

<p>The fabrication of zone plate lenses that are used for focusing X-rays relies on nanofabrication techniques such as e-beam lithography, reactive ion etching, and electroplating. The circular grating-like zone plate pattern can have a smallest half-period, a so-called zone width, of down to 20 nm while it also needs to have a height that is 5 to 10 times the zone width to have good diffraction efficiency. This high aspect ratio structuring is a very challenging field of nanofabrication.</p><p>This diploma project has focused on improving the process step of fabricating the electroplating mold by cryo-cooling the polymer during the reactive ion etching with O₂. The low temperature causes passivation of the sidewalls of the mold during etching which results in a more ideal rectangular profile of the high aspect ratio plating mold.</p><p>By etching at -100 °C, structures with highly vertical sidewalls and no undercut were realized. The experiments showed that there is a tradeoff between the anisotropy of the zone profile and the formation rate of polymer residue, so-called RIE grass. Through a proper choice of process parameters the grass could be completely removed without introducing any undercut.</p> / QC 20100414

Thin film copper, indium, gallium, selenide (CIGS) solar cells are promising in the field of photovoltaic technology. To reduce material and fabrication cost, as well as increasing electrical properties of the cell, research is ongoing towards ultra-thin film solar cells (absorption layer thickness less than 500 nm). Ultra-thin CIGS solar cells has shown a decrease in interface recombination and improved optical properties when adding a rear contact passivation layer of aluminium oxide. In this work, the process of creating sub-micrometer features of a passivation layer using conventional optical lithography is investigated. To specify, the objective was to optimize the development conditions in the optical lithography process when fabricating equidistant line contacts in aluminium oxide with 800 nm feature size. It was found that line contacts with smaller feature sizes require longer development time, than line contacts with larger feature sizes. The experiments conducted showed that the pre-set development and exposure conditions used by the NOA group are not optimized for 800 nm or smaller line contacts. Further, for the optical lithography process, silicon substrates are not comparable with substrates of soda lime glass coated with molybdenum. Slight underdevelopment of a sample, showed line contacts smaller than the resolution of the laser used in the exposure – suggesting an alternative method of processing small line contacts with optical lithography.

Les solutions colloïdales constituées de nanoparticules en solution sont une famille d'objets aux propriétés optiques uniques. Leur utilisation comme élément de base à la fabrication de composants optiques sublongueur d'onde pourrait permettre la naissance de nouvelles applications en particulier dans le domaine de l'optique intégrée et de la détection biologique. La manipulation de ces particules reste toutefois un défi en raison de leur taille et de leur dispersion aléatoire dans un milieu liquide. Dans ce contexte, nous avons réalisé des nouveaux composants optiques grâce au développement de techniques de fabrication basées sur la méthode d'assemblage capillaire assisté par convection. Deux types de structures ont été réalisés puis évalués en terme de comporte- ment optique : les dimères métalliques d'Au et les microsphères diélectriques de polystyrène assemblées en chaînes ou en réseaux. Pour les dimères, une étude fondamentale a été effectuée sur les phénomènes plasmoniques régissant les propriétés optiques de ces objets. Leur potentiel en tant que détecteur ultrasensible SERS et nanoantenne à boîtes quantiques a ensuite été approfondi. Pour les microsphères, une étude sur la propagation et la diffusion des modes de galerie présents dans ces objets a tout d'abord été réalisée dans le but d'en faire des candidats pour la détection ultrasensible. Les propriétés de guidage de la lumière dans des assemblages en chaîne ont ensuite été traitées. Afin de compléter ce travail un dernier composant optique a été développé en complément des guides et capteurs colloïdaux déjà réalisés. Il s'agit d'une nouvelle génération d'émetteurs localisés conçus pour un usage large et versatile et qu'il est possible de définir comme microsource de lumière blanche.