Ce travail de thèse est consacré à l’étude de la réactivité du spirosilane de Martin, une molécule présentant des propriétés intéressantes notamment en présence de bases de Lewis, qui a trouvé des applications diverses comme la détection d’ions fluorures. En ce qui nous concerne, nous avons choisi d’étudier l’interaction de ce dérivé du silicium avec des bases de Lewis neutres fortes tels que les Carbènes N-Hétérocycliques (NHC) qui ont été beaucoup utilisés pour stabiliser les espèces de basse valence des éléments du bloc p. Alors que les NHC sont connus pour former avec les chlorosilanes des adduits pentacoordinés stables, aucun exemple avec des silanes non halogénés n’avaient été décrits avant notre étude. Nous avons montré qu’en fonction de l’encombrement stérique des NHC étudiés, ils forment avec le spirosilane de Martin des adduits de Lewis normaux et anormaux stables. Les propriétés « Paires de Lewis Frustrées » (FLP) du spirosilane avec des NHC encombrés ont été examinées ainsi que l’accès à des nouveaux ligands NHC anioniques portant un motif siliconate. / This thesis work is focused on the reactivity of Martin’s Spirosilane, a molecule that displays some interesting properties in particular with different Lewis base, founding some interesting application such us fluoride sensor. In our study, we have chosen N-Heterocyclic Carbene (NHC) because they are well known for stabilising low-valence states of p-block elements or for disclosing new reactivities. Besides, NHC are known to form relatively stable adducts with tetravalent halosilanes and also to stabilise silicon(0) species through potassium graphite reductions, but, to the best of our knowledge, no pentacoordinated NHC-adducts with a non-halogenated silane partner has been synthesised to date.The first part of my PhD was focused on the update of Martin’s spirosilane synthesis due to some problem of reproducibility with the known procedure. Once obtained the product, it was begun the investigation of the interaction with different NHC carbene that afforded the corresponding adducts that were fully characterised. The different adducts were then studied as potential Frustrated Lewis Pair and as precursors of anionic-type ligands for the metal's coordination.

In today’s dynamic environment, where product demands are highly volatile and unstable, the ability to design and operate manufacturing facilities that are robust with respect to uncertainty and variability is becoming increasingly important to the success of any manufacturing firm in order to operate effectively in such an environment. Hence manufacturing facilities must be able to exhibit high levels of robustness and stability in order to deal with changing market demands. In general, Facility Layout Problem (FLP) is concerned with the allocation of the departments or machines in a facility with an objective to minimize the total material handling cost (MHC) of moving the required materials between pairs of departments. Most FLP approaches assume the flow between departments is deterministic, certain and constant over the entire time planning horizon. Changes in product demand and product mix in a dynamic environment invalidate these assumptions. Therefore there is a need for stochastic FLP approaches that aim to assess the impact of uncertainty and accommodate any possible changes in future product demands.This research focuses on stochastic FLP with an objective to present a methodology in the form of a framework that allows the layout designer to incorporate uncertainty in product demands into the design of a facility. In order to accomplish this objective, a measure of impact of this uncertainty is required. Two solution methods for single and multi period stochastic FLPs are presented to quantify the impact of product demand uncertainty to facility layout designs in terms of robustness (MHC) and variability (standard deviation). In the first method, a hybrid (simulation) approach which considers the development of a simulation model and integration of this model with the VIPPLANOPT 2006 algorithm is presented. In the second method, mathematical formulations of analytic robust and stable indices are developed along with the use of VIPPLANOPT for solution procedure. Several case studies are developed along with numerical examples and case studies from the literature are used to demonstrate the proposed methodology and the application of the two methods to address different aspects of stochastic FLP both analytically and via the simulation method. Through experimentation, the proposed framework with solution approaches has proven to be effective in evaluating the robustness and stability of facility layout designs with practical assumptions such as deletion and expansion of departments in a stochastic environment and in applying the analysis results of the analytic and simulation indices to reduce the impact of errors and make better decisions

No description available.

Submitted by Secretaria Pós de Produção (tpp@vm.uff.br) on 2017-07-27T19:12:43Z No. of bitstreams: 1 D2016 - Frederico Galaxe Paes.pdf: 5062594 bytes, checksum: 6141e589af7945cfd88fe0b9b3d62443 (MD5) / Made available in DSpace on 2017-07-27T19:12:43Z (GMT). No. of bitstreams: 1 D2016 - Frederico Galaxe Paes.pdf: 5062594 bytes, checksum: 6141e589af7945cfd88fe0b9b3d62443 (MD5) / Este trabalho aborda os seguintes problemas: Problema Quadrático de Alocação (PQA), Problema de Leiaute de Facilidades com Áreas Diferentes (PLFAD) e o Problema Job Shop Sem Espera (PJSSE). O PQA é um clássico problema de otimização combinatória, cujo objetivo é minimizar a soma das distâncias entre pares de locais distintos, ponderadas pelos fluxos entre as facilidades neles alocadas. O objetivo desta parte do trabalho é investigar técnicas heurísticas da literatura com base num conjunto de instâncias de referência do PQA. Os experimentos relatadosenvolveramAlgoritmosMeméticos(AM),técnicasdediversidadeadaptativa,algoritmos ILS (Iterated Local Search), busca locais 2-exchange e cadeia de ejeção (Ejection Chain). Doze algoritmos foram testados em 37 instâncias de referência obtidas da QAPLIB levando à escolha da combinação de técnicas mais adequada ao problema. A partir das observações obtidas do estudo anterior, decidiu-se abordar o PLFAD, de natureza semelhante ao PQA. No PLFAD, o objetivo é dimensionar e localizar facilidades retangulares em um espaço ilimitado e contínuo, sem sobreposição, de modo a minimizar a soma das distâncias entre facilidades ponderada pelos fluxos de manuseio de material. Porém, a pesquisa mostrou que devido a estrutura amarrada apresentada pelas soluções do PLFAD, métodos tradicionais de busca local tornam o problema caro computacionalmente, principalmente pelo tratamento da inviabilidade, devido a sobreposição. Duas abordagens algorítmicas são então introduzidas para tratar o problema: um Algoritmo Genético (GA) básico e um GA combinado com uma estratégia de decomposiçãoviadesconstruçãoereconstruçãoparcialdasolução. Paradecomporeficientemente o problema, uma estrutura especial é imposta às soluções impedindo que as facilidade cruzem os eixos X ou Y. Embora esta restrição possa deteriorar o valor da melhor solução encontrada, ela também aumenta muito a capacidade de busca do método em problemas de médio e grande porte. Comomostradopelosexperimentos,oalgoritmoresultanteproduzsoluçõesdealtaqualidadepara doisgruposdeinstânciasclássicasdaliteratura,melhorando6das8melhoressoluçõesconhecidas do primeiro grupo e todas as instâncias de médio e grande porte do segundo grupo. Para algumas das maiores instâncias do segundo grupo, com 90 ou 100 facilidades, a melhora média das soluções ficou em torno de6%ou7%quando comparado aos algoritmos anteriores, com menor tempo de CPU. Para tais instâncias, métodos exatos atuais são impraticáveis. Finalmente é apresentado o PJSSE, escolhido devido às suas soluções apresentarem uma natureza semelhante àquelas do PLFAD. Uma algoritmo baseado em GA, cuja construção da solução é efetuada por um algoritmo guloso eficiente, é proposto para resolver instâncias de referência da literatura obtendo resultados promissores e com menor tempo computacional comparado com abordagens anteriores, principalmente em instâncias de grande porte. / This work address the following problems: Quadratic Assignment Problem (QAP), Unequal Area Facility Layout Problem (UA-FLP), and the Job Shop Problem No-Wait (JSPNW). The QAP is a classic combinatorial optimization problem, which aims to minimize the sum of distances between pairs of different locations, weighted by flows between facilities allocated in them. The objective of this part of the work is to investigate heuristic techniques of the literature based on a benchmark datasets of the QAP. We perform experiments with Memetic Algorithms (MA), adaptive diversity techniques, Iterated Local Search (ILS) algorithms, local searches 2−exchange andEjectionChains. Twelvealgorithmshavebeentestedin37benchmarkdatasets obtained from QAPLIB thus enabling to identify a combination of more suitable techiques for the problem. Based on the observations of the previous study, we decided to address the UA-FLP, of similar nature to QAP. The UA-FLP, aims to dimension and locate rectangular facilities in an unlimited floor space, without overlap, while minimizing the sum of distances among facilities weighted by “material-handling"flows. However, the research has shown that due to the tight structure of good UA-FLP solutions, traditional methods of local search make the problem expensive computationally, mainly by infeasibility treatment due to overlap. We introduce two algorithmic approaches to address this problem: a simple Genetic Algorithm (GA), and a GA combined with a decomposition strategy via partial solution deconstructions and reconstructions. To efficiently decompose the problem, we impose a solution structure where no facility should cross the X or Y axis. Although this restriction can possibly deteriorate the value of the best achievable solution, it also greatly enhances the search capabilities of the method on medium and large problems. As highlighted by our experiments, the resulting algorithm produces solutions of high quality for the two classic datasets of the literature, improving 6 out of the 8 best known solutions from the first set and all medium- and large-scale instances from the second set. For some of the largest instances of the second set, with 90 or 100 facilities, the average solution improvement goes as high as 6% or 7% when compared to previous algorithms, in less CPU time. For such instances, current exact methods are impracticable. Finally is presented the PJSSE, chosen because of its solutions present a nature similar to those of PLFAD. An algorithm based on GA, where the construction of the solution is made by an greedy eficient algorithm, is proposed to solve benchmark instances of the literature. Promising results have been achieved in less CPU-time than previous approaches, especially for larger scale instances.

No description available.

Commonly, when it is desirable to replace an essential gene with an allelic series of mutated genes, or genes with altered expression patterns, the complementing constructs are introduced into heterozygous plants, followed by the selection of homozygous null segregants. To overcome this laborious and time-consuming step, the newly developed two-component system utilizes a site-specific recombinase to excise a wild-type copy of the gene of interest from transformed tissues. In the first component (the first vector), a wild-type version of the gene is placed between target sequences recognized by FLP recombinase from the yeast 2 μm plasmid. This construct is transformed into a plant heterozygous for a null mutation at the endogenous locus, and progeny plants carrying the excisable complementing gene and segregating homozygous knockout at the endogenous locus are selected. The second component (the second vector) carries the experimental gene along with the FLP gene. When this construct is introduced, FLP recombinase excises the complementing gene, leaving the experimental gene as the only functional copy. The FLP gene is driven by an egg apparatus specific enhancer (EASE) to ensure excision of the complementing cDNA in the egg cell and zygote following floral-dip transformation. The utility of this system is being tested using various experimental derivatives of the essential sucrose-proton symporter, AtSUC2, which is required for photoassimilate transport.

<p>The <em>Sinorhizobium meliloti</em> genome consists of 6204 predicted protein-coding regions of which approximately 2000 are proteins of unknown function (PUFs). To identify functions of <em>S. meliloti</em> PUFs, we employed the FRT/Flp recombination system to delete large gene clusters and then screened for phenotypes. Large-scale deletions have been mainly used to define minimal gene sets that contain only those genes that are essential and sufficient to sustain a functioning cell. To adapt FRT/Flp for use in <em>S. meliloti</em>, we used an already constructed pTH1522-derived integration gene library of the <em>S. meliloti</em> genome (pTH1522 carries a single FRT site). A second FRT site was inserted at defined locations in the genome through integration of a second plasmid (pTH1937) that also carries a single FRT site. Here we outline how this Flp/FRT system was used to delete defined regions and hence generate multiple gene knock-out mutants. This system was used to delete 32 and 56 defined regions from the 1340 Kb pSymA and 1678 Kb pSymB megaplasmid, respectively. The structures of the resulting megaplasmid deletion mutants were confirmed by PCR analysis. Carbohydrate and nitrogen utilization phenotypes were associated with the deletion of specific regions. Deleting large, regions of the genome helped us to identify phenotypes such as inability to grow on minimal media with fucose, maltotriose, maltitol, trehalose, palatinose, lactulose and galactosamine as sole carbon source. For several FRT-flanked regions, few or no recombinants were recovered which suggested the presence of essential genes. Through this strategy, two essential genes <em>tRNA^arg </em>and<em> engA</em> located on the pSymB and three toxin/antitoxin-like systems, <em>sma0471</em>/<em>sma0473</em>, <em>sma2105</em> and <em>sma2230</em>/<em>sma2231</em> on pSymA megaplasmid were identified.</p> / Doctor of Philosophy (PhD)

Computational chemistry is a branch of modern chemistry that utilizes the computers to solve chemical problems. The fundamental of computational chemistry is Schrödinger equation. To solve the equation, researchers developed many methods based on BornOppenheimer Approximation, such as Hartree-Fock method and DFT method, etc. Computational chemistry is now widely used on reaction mechanism study and new chemical designing. In the first project described in Chapter 3, we designed phosphine oxide modified Ag3, Au3 and Cu3 nanocluster catalysts with DFT method. We found that these catalysts were able to catalyze the activation of H2 by cleaving the H-H bond asymmetrically. The activated catalyst-2H complex can be further used as reducing agent to hydrogenate CO molecule to afford HCHO. The mechanism study of these catalysts showed that the electron transfer from electron-rich metal clusters to O atom on the phosphine oxide ligand is the major driving force for H2 activation. In addition, different substituent groups on phosphine oxide ligand were tested. Both H affinity of metal and the substituent groups on ligand can both affect the activation energy. Another project described in Chapter 4 is the modelling of catalyst with DFT. We chose borane/NHC frustrated Lewis pair (FLP) catalyzed methane activation reaction as example to establish a relationship between activation energy and catalysts’ physical properties. After performing simulation, we further proved the well-accepted theory that the electron transfer is the main driving force of catalysis. Furthermore, we were able to establish a linear relationship for each borane between activation energy and the geometrical mean value of HOMO/LUMO energy gap (ΔEMO). Based on that, we introduced the formation energy of borane/NHC complex (ΔEF) and successfully established a generalized relationship between Ea and geometrical mean value of ΔEMO and ΔEF. This model can be used to predict reactivity of catalysts.

“Family language policy” (FLP) is the accepted term for the field of study of the explicit planning and practices concerning language within a family unit in a home. Previous research has shown that FLP aids in the bilingual acquisition of a child (DeCapua & Wintergerst 2009; Kayam & Hirsch 2014; King, Fogle, & Logan‐Terry 2008; Li 1999; Oh; Schwartz 2008). However, there has been little research providing answers to whether FLP has a direct influence on language maintenance in adulthood, especially whether they acquire and maintain a native or native-like accent in both languages. The purpose of this study is to determine if any and to what degree FLP influences the bilingual accent acquisition of Spanish/English heritage speakers in the United States. This is a qualitative case study performed through sociolinguistic interviews of three families containing now adult simultaneous bilinguals who learned Spanish and English throughout childhood. After obtaining information of each family’s FLP, each participant (n = 9) was asked to provide a speech sample in both English and Spanish (the heritage language). These samples were then rated by native speakers of English and Spanish respectively. Results suggest that the level of perceived foreign accent of the heritage language may be influenced by certain factors included in an individual FLP, as well as the speaker’s language confidence and individual differences including language aptitude.

Protocols for the FLP-mediated transformation of ethers are presented. Distinct reaction pathways involving either C–O or C–H bond activation occur depending on the application of oxophilic B(C6F5)3 or hydridophilic tritylium ions as the Lewis acid.